i

K E TO G E N I C D I E T
A N D M E TA B O L I C T H E R A P I E S
ii
iii

K E TO G E N I C D I E T A N D
M E TA B O L I C T H E R A P I E S

Expanded Roles in Health and Disease

EDITED BY

S U S A N A . M A S I N O,   P H D
Vernon Roosa Professor of Applied Science
Professor of Psychology and Neuroscience
Trinity College
Hartford, CT

SECTION EDITORS

D E T L E V B O I S O N,   P H D
D O M I N I C P. D ’ AG O S T I N O,   P H D
E R I C H . KO S S O F F,   M D
J O N G M . R H O,   M D

3
iv

Oxford University Press is a department of the University of Oxford. It furthers

the University’s objective of excellence in research, scholarship, and education
by publishing worldwide. Oxford is a registered trade mark of Oxford University Press
in the UK and certain other countries.

Published in the United States of America by Oxford University Press

198 Madison Avenue, New York, NY 10016, United States of America.

© Oxford University Press 2017

All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by license, or under terms agreed with the appropriate reproduction
rights organization. Inquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above.

You must not circulate this work in any other form

and you must impose this same condition on any acquirer.

Library of Congress Cataloging-in-Publication Data

Names: Masino, Susan, editor.
Title: Ketogenic diet and metabolic therapies : expanded roles in health and disease /
edited by Susan A. Masino.
Description: Oxford ; New York : Oxford University Press, [2017] | Includes bibliographical
references and index.
Identifiers: LCCN 2016019577 | ISBN 9780190497996 (alk. paper)
Subjects: | MESH: Ketogenic Diet | Metabolism—physiology
Classification: LCC RM237.73 | NLM WB 427 | DDC 613.2/83—dc23
LC record available at https://lccn.loc.gov/2016019577

This material is not intended to be, and should not be considered, a substitute for medical or
other professional advice. Treatment for the conditions described in this material is highly dependent
on the individual circumstances. And, while this material is designed to offer accurate information
with respect to the subject matter covered and to be current as of the time it was written, research and
knowledge about medical and health issues is constantly evolving and dose schedules for medications are
being revised continually, with new side effects recognized and accounted for regularly. Readers must
therefore always check the product information and clinical procedures with the most up-to-date published
product information and data sheets provided by the manufacturers and the most recent codes of conduct
and safety regulation. The publisher and the authors make no representations or warranties to readers, express
or implied, as to the accuracy or completeness of this material. Without limiting the foregoing, the publisher
and the authors make no representations or warranties as to the accuracy or efficacy of the drug dosages mentioned
in the material. The authors and the publisher do not accept, and expressly disclaim, any responsibility for
any liability, loss or risk that may be claimed or incurred as a consequence of the use and/or application of any
of the contents of this material.

1 3 5 7 9 8 6 4 2
Printed by Sheridan Books, Inc., United States of America
v
CONTENTS
Preface  ix
Contributors  xi
SECTION I: Ketogenic Diet for Epilepsy
in the Clinic
Eric H. Kossoff, MD, Section Editor
1. Overview: Ketogenic Diets
and Pediatric Epilepsy: An Update  3
Eric H. Kossoff, md
2. “Alternative” Ketogenic Diets  5
Elizabeth Neal, RD, MSc, PhD
3. Dietary Therapy in Adults: History,
Demand, and Results  16
Emily L. Johnson, md and
Mackenzie C. Cervenka, md
4. How Do You Implement the Diet?  26
A. G. Christina Bergqvist, md
5. Glut1 Deficiency and
the Ketogenic Diets  35
Joerg Klepper, MD, PhD
6. Ketogenic Diet in Established
Epilepsy Indications  40
Ann M. Bergin, MB, ScM, MRCP(UK)
7. Ketogenic Diet for Other Epilepsies  50
David T. Hsieh, MD and
Elizabeth A. Thiele, MD, PhD
8. The Ketogenic Diet
and Related Therapies in “Novel”
Situations: Idiopathic Generalized
Epilepsy Syndromes  56
Sudha Kilaru Kessler, MD, MSCE
9. Ketogenic Diet in Status Epilepticus  60
Rima Nabbout, MD, PhD
10. Preventing Side Effects and Diet
Discontinuation  66
Cherie L. Herren, MD and
Rana R. Said, md
SECTION II: Ketogenic Diet:
Emerging Clinical Applications
and Future Potential
Jong M. Rho, MD, Section Editor
11. Overview: Expanded Uses
of Ketogenic Therapies  77
Jong M. Rho, md
12. Metabolism-Based Treatments
to Counter Cancer: Scientific Rationale  79
Thomas N. Seyfried, PhD and
Laura M. Shelton, PhD
13. Ketogenic Diet as Adjunctive Therapy
for Malignant Brain Cancer  88
Eric C. Woolf, PhD and
Adrienne C. Scheck, PhD
14. Metabolic Therapy for Autism
Spectrum Disorder
and Comorbidities  101
Ning Cheng, PhD, Susan A. Masino, PhD,
and Jong M. Rho, md
15. Glucose and Ketone Metabolism in
the Aging Brain: Implications for
Therapeutic Strategies to Delay the
Progression of Alzheimer’s Disease  113
Stephen C. Cunnane, PhD,
Alexandre Courchesne-Loyer, msc,
Valerie St-Pierre, bsc,
Camille Vandenberghe, BSC,
Etienne Croteau, PhD, and
Christian-Alexandre Castellano, PhD
vi
vi Contents
16. Ketogenic Diet and Ketones for
the Treatment of Traumatic Brain
and Spinal Cord Injury  133
Femke Streijger, PhD,
Ward T. Plunet, PhD, and
Wolfram Tetzlaff, MD, Dr. Med, PhD
17. Anti-Inflammatory Effects of
a Ketogenic Diet: Implications for
New Indications  147
Nina Dupuis, PhD and
Stéphane Auvin, MD, PhD
18. Dietary Therapy for Neurological
Disorders: Focus on Amyotrophic
Lateral Sclerosis, Parkinson’s Disease,
Mood Disorders, and Migraine  156
Carl E. Stafstrom, MD, PhD
SECTION III: Ketogenic Diet in
the Laboratory
Detlev Boison, PhD, Section Editor
19. Overview of Ketogenic Diet in the
Laboratory: Progress on Models
and Mechanisms  165
Detlev Boison, PhD
20. Ketogenic Diet and PPARgamma  167
Timothy A. Simeone, PhD
21. Ketogenic Diet in a Hippocampal
Slice: Models and Mechanisms  186
Masahito Kawamura JR., MD, PhD
22. Metabolic Therapy and Pain  196
David N. Ruskin, PhD
23. Ketogenic Diet, Adenosine,
Epigenetics, and Antiepileptogenesis  209
Theresa A. Lusardi, PhD and
Detlev Boison, PhD
24. Ketogenic Diet, Aging, and
Neurodegeneration  216
Kui Xu, MD, PhD,
Joseph C. LaManna, PhD, and
Michelle A. Puchowicz, PhD
25. Endocrine and Reproductive Effects
of Ketogenic Diets  227
Jacob P. Harney, PhD,
Kathryn Gudsnuk, ms, Ami Patel, md,
Anantha R. Vellipuram, md,
Sathyajit Bandaru, ms, and
David Butler, PHD
26. Alzheimer’s Disease: Causes
and Treatment  241
Richard L. Veech, MD, PhD, DPhil and
M. Todd King
27. Mitigation of Damage
from Reactive Oxygen Species
and Ionizing Radiation by
Ketone Body Esters  254
William Curtis, Martin Kemper, PhD,
Alexandra Miller, PhD,
Robert Pawlosky, PHD, M. Todd King, and
Richard L. Veech, MD, PhD, DPhil
28. Metabolic Seizure Resistance via BAD
and KATP Channels  271
Juan Ramón Martínez-François, PhD,
Nika N. Danial, PhD, and
Gary Yellen, PhD
29. Lactate Dehydrogenase: A Novel
Metabolic Target  281
Nagisa Sada, PhD and
Tsuyoshi Inoue, PhD
30. Effects of the Ketogenic Diet on
the Blood-Brain Barrier  289
Manoj Banjara, PhD and
Damir Janigro, PhD
SECTION IV: Ketone-Based Metabolism:
General Health and Metabolic Alternatives
Dominic P. D’Agostino, PhD,
Section Editor
31. Overview of Ketone-Based
Metabolism: General Health and
Metabolic Alternatives  307
Dominic P. D’Agostino, PhD
32. Ketone Supplementation for Health
and Disease  310
Angela M. Poff, PhD,
Shannon L. Kesl, PhD, and
Dominic P. D’Agostino, PhD
33. Identifying the Molecular Mechanism
of the Medium Chain Triglyceride
(Ketogenic) Diet  328
Matthew C. Walker, FRCP, PhD and
Robin S.B. Williams, PhD
34. Triheptanoin in Epilepsy and Beyond  336
Karin Borges, PhD
vii
35. Amino Acids in the Treatment of
Neurological Disorders  346
Adam L. Hartman, md
36. 2-Deoxyglucose: Metabolic Control
of Seizures through Inhibition
of Glycolysis  353
Carl E. Stafstrom, MD, PhD and
Thomas P. Sutula, MD, PhD
37. Ketogenic Diets as Highly Effective
Treatments for Diabetes Mellitus
and Obesity  362
Eric C. Westman, MD, MHS,
Emily Maguire, MSc, and
William S. Yancy Jr., MD, MHS
Contents vii
38. Keto-Adaptation in Health
and Fitness  376
Parker Hyde, CSCS, CISSN,
Vincent J. Miller, MS, and
Jeff S. Volek, PhD, RD
39. Advancing the Awareness and
Application of Ketogenic Therapies
Globally: The Charlie Foundation
and Matthew’s Friends  386
Beth Zupec-Kania, RD, CD,
Jim Abrahams,
Emma Williams, MBE, and
Susan A. Masino, PhD
Index  397
viii
ix
P R E FA C E
Metabolism is a fundamental cellular process, and
metabolic dysfunction is associated with disease.
The ketogenic diet is a metabolic therapy first pub-
lished in 1921 as an effective treatment for seizures
in both children and adults, and it has been pre-
scribed to a subset of patients with epilepsy ever
since. Today there are many drugs available to
control epileptic seizures, yet this metabolic ther-
apy can stop seizures even when all medications
fail: for some patients a ketogenic diet is superior
to all known drug treatments. The ketogenic diet
was developed nearly 100 years ago because it had
been observed—for centuries—that fasting would
stop seizures. Adhering to a medically prescribed
and carefully formulated high-fat ketogenic diet
can maintain the ketone-based metabolism used
during fasting.
Metabolic therapy targets the most fundamen-
tal aspect of cell function: cell energy. Targeting cell
function or dysfunction metabolically is concep-
tually distinct from treating a disease specifically
and pharmacologically. While a pharmacological
approach has dominated drug development, and
can be effective for some symptoms and condi-
tions, it is also more likely to produce off-target
side effects and less likely to produce lasting
changes. In contrast, supporting cell energy and
promoting metabolic homeostasis can improve
overall health and may offer long-term benefits in
preventing or modifying disease.
Recent basic and translational research has
provided new insight into mechanisms as well
as evidence that metabolic therapy with a keto-
genic diet can treat diverse conditions beyond
epilepsy. New research has also provided evi-
dence that alternatives which can substitute for
or complement the diet—and potentially aug-
ment its efficacy—may be close at hand. Evidence
is also mounting that ketogenic diets can reverse
chronic health conditions and provide general
health benefits beyond treating any particular dis-
ease. Understanding key mechanisms underlying
the success of metabolic therapy is of the high-
est biomedical significance: it is anticipated these
mechanisms will apply to provide breakthroughs
for multiple common, chronic, and poorly treated
disorders. Similarly, a comprehensive understand-
ing of the range and type of acute and chronic con-
ditions that metabolic therapies can prevent, delay,
or reverse is of urgent clinical importance.
Here we provide a fresh view on the promise
of using the biochemistry of metabolism to treat
disease and promote health by compiling the lat-
est research and perspectives of leading experts on
ketogenic diets and metabolic therapies. This vol-
ume is an up-to-date and comprehensive resource
organized into four key subsections spearheaded
by leaders in each area: the latest clinical research
for treatment of epilepsy (Eric Kossoff, MD),
emerging clinical applications (Jong Rho, MD),
laboratory research into key mechanisms (Detlev
Boison, PhD), and diverse metabolic therapies
to treat disease and improve health (Dominic
D’Agostino, PhD). The last chapter is devoted to
two key organizations:  the Charlie Foundation,
established in 1994 in the United States, and
Matthew’s Friends, established in 2004 in the
United Kingdom. In the last two decades growth
of research in the ketogenic diet field has been
exponential, and the Charlie Foundation played
an enormously important role in raising aware-
ness and spearheading its resurgence in the clinic
and the laboratory. Ongoing efforts of the Charlie
Foundation have been furthered and multiplied by
Matthew’s Friends, and together these foundations
are devoted to research, education, outreach, and
applications of ketogenic therapies throughout
the world.
x
x Preface
My personal path to a research program on the
ketogenic diet was unusual:  it arose organically
from a basic science hypothesis on the regulation
of adenosine. Adenosine is present throughout
the body and the central nervous system and is a
powerful neuromodulator and bioenergetic regu-
lator of network homeostasis. Like the ketogenic
diet, adenosine links metabolism and brain activ-
ity and has been proven to have powerful anti-
seizure, neuroprotective, and disease-modifying
benefits. Years of basic research on adenosine led
me unexpectedly to the most important and excit-
ing work of my career thus far and connected me
with a motivated and collaborative global com-
munity of researchers, clinicians, patients, and
advocates. The ketogenic diet has been proven to
cure devastating cases of epilepsy, and we know
that unlocking its key mechanisms—whatever
they may be—will be a major biomedical break-
through. Together we look forward to the 100th
anniversary of the ketogenic diet in 2021 with
optimism that metabolic therapies will offer new,
safe, and effective options to promote health and
cure disease.
Susan A. Masino, PhD
Hartford, CT
xi
C O N T R I BU TO R S
Jim Abrahams
Charlie Foundation for Ketogenic Therapies
Santa Monica, CA
Stéphane Auvin, MD, PhD
Service de Neurologie Pédiatrique
Hôpital Universitaire Robert-Debré
Paris, France
Sathyajit Bandaru, MS
Neuroscience Graduate Program
University of Hartford
West Hartford, CT
Manoj Banjara, PhD
Departments of Biomedical Engineering
and Cerebrovascular Research
Cleveland Clinic Foundation
Cleveland, OH
Ann M. Bergin, MB, ScM, MRCP(UK)
Assistant Professor of Neurology
Harvard Medical School
Children’s Hospital of Boston
Boston, MA
A. G. Christina Bergqvist, MD
Medical Director, Dietary Treatment Program
Associate Professor of Neurology and Pediatrics
Children’s Hospital of Philadelphia
Perelman School of Medicine at
the University of Pennsylvania
Philadelphia, PA
Detlev Boison, PhD
Robert Stone Dow Chair and Director
of Neurobiology
Director of Basic and Translational Research
Legacy Research Institute
Portland, OR
Karin Borges, PhD
Department of Pharmacology
School of Biomedical Sciences
The University of Queensland
St. Lucia, Australia
David Butler, PhD
Neuroscience Graduate Program
University of Hartford
West Hartford, CT
Christian-Alexandre Castellano, PhD
Research Center on Aging
Departments of Medicine and
Pharmacology and Physiology
University of Sherbrooke
Sherbrooke, Québec, Canada
Mackenzie C. Cervenka, MD
Director, Adult Epilepsy Diet Center
Medical Director, Epilepsy Monitoring Unit
Assistant Professor of Neurology
Johns Hopkins Hospital
Baltimore, MD
Ning Cheng, PhD
Department of Pediatrics
Alberta Children’s Hospital
Research Institute
Cumming School of Medicine
University of Calgary
Calgary, Canada
Alexandre Courchesne-Loyer, MSc
Research Center on Aging
Departments of Medicine and
Pharmacology and Physiology
University of Sherbrooke
Sherbrooke, Québec, Canada
xii
xii Contributors
Etienne Croteau, PhD
Research Center on Aging
Departments of Medicine and Pharmacology
and Physiology
Université de Sherbrooke
Sherbrooke, Québec, Canada
Stephen C. Cunnane, PhD
Research Center on Aging
Departments of Medicine and Pharmacology
and Physiology
University of Sherbrooke
Sherbrooke, Québec, Canada
William Curtis
Guest Worker, Lab of Metabolic Control
National Institute of Alcohol Abuse
and Alcoholism
Rockville, MD
Dominic P. D’Agostino, PhD
Associate Professor
Department of Molecular Pharmacology
and Physiology
Morsani College of Medicine
University of South Florida
Tampa, FL
Nika N. Danial, PhD
Department of Cancer Biology
Dana-Farber Cancer Institute
Department of Cell Biology
Harvard Medical School
Boston, MA
Nina Dupuis, PhD
Université Paris Diderot
Sorbonne Paris Cité
Paris, France
Kathryn Gudsnuk, MS
Neuroscience Graduate Program
University of Hartford
West Hartford, CT
Jacob P. Harney, PhD
School of Mathematics and Sciences
Reinhardt University
Waleska, GA
Neuroscience Graduate Program
University of Hartford
West Hartford, CT
Adam L. Hartman, MD
Associate Professor of Neurology and Pediatrics
Johns Hopkins Medicine
Baltimore, MD
Cherie L. Herren, MD
Medical Director, OU Children’s Physicians
Ketogenic Diet Program
Department of Neurology
University of Oklahoma Health Science Center
Oklahoma City, OK
David T. Hsieh, MD
Division of Child Neurology
Department of Pediatrics
San Antonio Military Medical Center
Fort Sam Houston, TX
Parker Hyde, CSCS, CISSN
Department of Human Sciences
The Ohio State University
Columbus, OH
Tsuyoshi Inoue, PhD
Department of Biophysical Chemistry
Graduate School of Medicine
Dentistry and Pharmaceutical Sciences
Okayama University
Okayama, Japan
Damir Janigro, PhD
Flocel, Inc.
Cleveland, OH
Emily L. Johnson, MD
Assistant Professor of Neurology
Department of Neurology
Johns Hopkins Hospital
Baltimore, MD
Masahito Kawamura Jr., MD, PhD
Department of Pharmacology
Jikei University School of Medicine
Tokyo, Japan
Martin Kemper, PhD
Lab of Metabolic Control
National Institute of Alcohol Abuse
and Alcoholism
National Institute of Health
Rockville, MD
xiii

Contributors xiii

Shannon L. Kesl, PhD Alexandra Miller, PhD

Department of Molecular Pharmacology Armed Forces Radiobiology
and Physiology Research Institute
Morsani College of Medicine Uniformed Services University of
University of South Florida the Health Sciences
Tampa, FL Bethesda, MD

Sudha Kilaru Kessler, MD, MSCE Vincent J. Miller, MS

Assistant Professor of Neurology and Pediatrics Department of Human Sciences
Children’s Hospital of Philadelphia The Ohio State University
Perelman School of Medicine at the University Columbus, OH
of Pennsylvania
Philadelphia, PA Rima Nabbout, MD, PhD
Reference Centre for Rare Epilepsies
M. Todd King Department of Pediatric Neurology
Lab of Metabolic Control Necker Enfants Malades Hospital
National Institute of Alcohol Abuse Assistance Publique Hopitaux de Paris
and Alcoholism Paris Descartes University
National Institute of Health Paris, France
Rockville, MD
Elizabeth Neal, RD, MSc, PhD
Joerg Klepper, MD, PhD Specialist Ketogenic Dietitian
Chief Physician Matthews Friends Clinics
Children’s Hospital Aschaffenburg Lingfield, UK
Aschaffenburg, Germany Honorary Research Associate
Department of Neuroscience
Eric H. Kossoff, MD University College London-Institute
Professor, Neurology and Pediatrics of Child Health
Medical Director, Ketogenic Diet Center London, UK
Director, Pediatric Neurology Residency Program
Johns Hopkins Hospital Ami Patel, MD
Baltimore, MD Neuroscience Graduate Program
University of Hartford
Joseph C. LaManna, PhD West Hartford, CT
Departments of Physiology and Biophysics
Case Western Reserve University
Robert Pawlosky, PhD
Cleveland, OH
Lab of Metabolic Control
National Institute of Alcohol Abuse
Theresa A. Lusardi, PhD
and Alcoholism
Robert Stone Dow Neurobiology Laboratories
National Institute of Health
Legacy Research Institute
Rockville, MD
Portland, OR

Emily Maguire, MSc Ward T. Plunet, PhD

Low Carb Genesis International Collaboration
Edinburgh, Scotland, UK on Repair Discoveries (ICORD)
University of British Columbia
Juan Ramón Martínez-François, PhD Blusson Spinal Cord Centre
Department of Neurobiology Vancouver, Canada
Harvard Medical School
Boston, MA Angela M. Poff, PhD
Department of Molecular Pharmacology
Susan A. Masino, PhD and Physiology
Vernon Roosa Professor of Applied Science Morsani College of Medicine
Professor of Psychology and Neuroscience University of South Florida
Trinity College Tampa, FL
Hartford, CT
xiv

xiv Contributors

Michelle A. Puchowicz, PhD Timothy A. Simeone, PhD

Department of Nutrition Creighton University School of Medicine
Case Western Reserve University Department of Pharmacology
Cleveland, OH Omaha, NE

Jong M. Rho, MD Valerie St-Pierre, BSc

Professor of Pediatrics and Clinical Neurosciences Research Center on Aging
Dr. Robert Haslam Chair in Pediatric Neurology Departments of Medicine and Pharmacology
Cumming School of Medicine, University and Physiology
of Calgary Université de Sherbrooke
Calgary, Canada Sherbrooke, Québec, Canada

David N. Ruskin, PhD Carl E. Stafstrom, MD, PhD

Department of Psychology and Neuroscience Professor of Neurology and Pediatrics
Program Lederer Chair in Pediatric Epilepsy
Trinity College Johns Hopkins School of Medicine
Hartford, CT Director, Division of Pediatric Neurology
Johns Hopkins Hospital
Nagisa Sada, PhD Baltimore, MD
Department of Biophysical Chemistry
Graduate School of Medicine Femke Streijger, PhD
Dentistry and Pharmaceutical Sciences International Collaboration on Repair Discoveries
Okayama University (ICORD)
Okayama, Japan Blusson Spinal Cord Centre
University of British Columbia
Rana R. Said, MD Vancouver, Canada
Director, Pediatric Neurology
Residency Program Thomas P. Sutula, MD, PhD
Department of Pediatrics, Neurology, and Department of Neurology
Neurotherapeutics School of Medicine and Public Health
UT Southwestern Medical Center University of Wisconsin
University of Texas Southwestern Madison, WI
Dallas, TX
Wolfram Tetzlaff, MD, Dr. Med, PhD
Adrienne C. Scheck, PhD Department of Zoology
Neuro-Oncology Research International Collaboration on Repair Discoveries
Barrow Brain Tumor Research Center (ICORD)
Barrow Neurological Institute® dba St. Joseph’s Blusson Spinal Cord Centre
Hospital and Medical Center University of British Columbia
Phoenix, AZ Vancouver, Canada
School of Life Sciences
Arizona State University Elizabeth A. Thiele, MD, PhD
Tempe, AZ Pediatric Epilepsy Program
Department of Neurology
Thomas N. Seyfried, PhD Massachusetts General Hospital
Biology Department Boston, MA
Boston College
Boston, MA Camille Vandenberghe, BSc
Research Center on Aging
Laura M. Shelton, PhD Departments of Medicine and Pharmacology and
Scientific Project Coordinator Physiology
Human Metabolome Technologies America University of Sherbrooke
Boston, MA Sherbrooke, Québec, Canada
xv
Richard L. Veech, MD, PhD, DPhil
Lab of Metabolic Control
National Institute of Alcohol Abuse and
Alcoholism
National Institute of Health
Rockville, MD
Anantha R. Vellipuram, MD
Neuroscience Graduate Program
University of Hartford
West Hartford, CT
Jeff S. Volek, PhD, RD
Department of Human Sciences
The Ohio State University
Columbus, OH
Matthew C. Walker, FRCP, PhD
Department of Clinical and Experimental
Epilepsy
Institute of Neurology
University College London
London, UK
Eric C. Westman, MD, MHS
Division of General Internal Medicine
Department of Medicine
Duke University Medical Center
Durham, NC
Emma Williams, MBE
Matthew’s Friends
Lingfield, UK
Robin S.B. Williams, PhD
Centre for Biomedical Sciences
School of Biological Sciences
Royal Holloway University of London
Egham, UK
Contributors xv
Eric C. Woolf, PhD
Neuro-Oncology Research
Barrow Brain Tumor Research Center
Barrow Neurological Institute® dba St. Joseph’s
Hospital and Medical Center
Phoenix, AZ
School of Life Sciences
Arizona State University
Tempe, AZ
Kui Xu, MD, PhD
Departments of Physiology
and Biophysics
Case Western Reserve University
Cleveland, OH
William S. Yancy Jr., MD, MHS
Division of General Internal Medicine
Department of Medicine
Duke University Medical Center
Center for Health Services Research in
Primary Care
Durham Veterans Affairs Medical Center
Duke Diet and Fitness Center
Duke University Health System
Durham, NC
Gary Yellen, PhD
Professor of Neurobiology
Harvard Medical School
Boston, MA
Beth Zupec-Kania, RD, CD
Ketogenic Therapies LLC
Elm Grove, WI
xvi
1

SECTION I
Ketogenic Diet for Epilepsy
in the Clinic

E R I C H . K O S S O F F, M D , S E C T I O N E D I T O R
2
3
1
Overview: Ketogenic Diets and Pediatric Epilepsy
An Update
E R I C H . K O S S O F F, M D
A s it approaches its 100-year anniversary, the
ketogenic diet (KD) is reaching an interest
level not previously seen. Originally published in
1921 by Dr. Russell Wilder at the Mayo Clinic, its
creation came at a time in which there were few
other options for epilepsy (Wilder, 1921). The KD
was widely used for the next several decades in
both children and adults, with approximately 50%
of patients reporting at least a 50% reduction in
seizures in multiple studies. The advent of phe-
nytoin and other modern pharmaceutical antisei-
zure drugs in the 1940s and afterward relegated
the KD to “alternative” medicine and it was largely
ignored by epilepsy specialists. For many decades
it was used only as a last resort in children with
intractable epilepsy; only very select institutions
were still implementing it sporadically.
In 1993, one such refractory case prompted
renewed interest in dietary therapies. Hollywood
producer Jim Abrahams brought his 2-year old
son Charlie to Johns Hopkins Hospital, where
Charlie experienced rapid seizure control within
days after starting the KD. Abraham created the
Charlie Foundation in 1994, which revitalized
research efforts, and produced First Do No Harm, a
TV movie starring Meryl Streep, which promoted
the KD. In 1998, the first multicenter prospective
study of the KD in children with refractory epi-
lepsy demonstrated that more than half of patients
had a greater than 50% reduction in seizure fre-
quency after 6 months (Vining et al., 1998).
In the now 20+ years since the formation of
the Charlie Foundation, dietary therapies have
experienced a rapid resurgence in research and
use. The majority of countries have implemented
KDs, and more than 100 research articles are pub-
lished yearly (Kossoff & McGrogan, 2005). Several
randomized controlled clinical trials, crossover
studies, and prospective studies have confirmed
a response rate of approximately 50% in children
with refractory epilepsy. In 2009, Dr.  Freeman
and colleagues performed the first blinded study
of the KD by having all participants consume the
ketogenic diet plus a daily supplement of either
saccharin (treatment group) or glucose (to pre-
vent ketosis; control group) (Freeman et al., 2009).
They found a trend toward improved seizure fre-
quency in the saccharin group, though the effect
did not reach statistical significance, possibly due
to complex actions of the KD that were not pre-
vented with ingestion of glucose once a day. Neal
and colleagues randomized patients to no change
in standard medical management or addition of
the KD; they found that patients with refractory
epilepsy who were randomized to receive the KD
were more likely to have a 50% decrease in sei-
zure frequency than the control group (Neal et al.,
2008). Another study by Sharma et  al. in 2013,
using a similar study design to Dr.  Neal’s 2008
trial, found the modified Atkins diet to be effective
in a randomized controlled study as well (Sharma
et al., 2013). In light of the accumulating evidence
to support the efficacy of KDs, the International
Ketogenic Diet Study Group, a panel of 26 neu-
rologists and dietitians, recommended that dietary
therapies be strongly considered in patients of
any age who had failed two to three medications
(Kossoff et al., 2009).
Beyond the formal prospective studies which
have proven efficacy, perhaps an even more impor-
tant factor that has led to the resurgence of dietary
therapies has been a combination of flexibility in
implementation and recognition of true indica-
tions for its use (Kossoff et al., 2009). Treating the
appropriate patients (sooner rather than later) as
well as considering alternative diets and methods
of starting this treatment have led to widespread
availability, willingness of patients and neurolo-
gists to consider it in their treatment algorithm,
and better (and safer) outcomes. In this section,
4
4 section I: Ketogenic Diet for Epilepsy in the Clinic
“Ketogenic Diet for Epilepsy in the Clinic,” these
factors are discussed in more detail.
First, Dr. Neal highlights that now there are not
one but four types of KD now available, each with
excellent reported efficacy:  the classic ketogenic
diet (KD), the medium chain triglyceride (MCT)
diet, the low glycemic index treatment (LGIT),
and the modified Atkins diet (MAD) (chapter 2).
The latter two diets have certainly been responsible
for the acceptance of dietary therapies by adults,
which is discussed by Drs. Cervenka and Johnson
in chapter 3 (see Cervenka et al., 2013). Flexibility
during the initiation week of the classic KD has
also revolutionized approaches to the diet by
many epilepsy centers as outlined by Dr. Bergqvist
(chapter 4; Bergqvist et al., 2005).
Second, pediatric epilepsy experts discuss
the indications for dietary therapy in pediatric
patients. Approximately 20  years ago, there was
little to no ability to predict which child would be
a KD responder. That has radically changed due to
research and large cohort studies. The most famous
indication, GLUT1 (glucose-1 transporter) defi-
ciency syndrome, uses the KD as its primary,
gold-standard therapy, and Dr. Klepper has been
involved in much of the research on this condition
and its response to the KD (chapter  5; Klepper,
2012). Drs. Bergin, Hsieh, and Thiele then discuss
some of the other well-known epilepsy syndromes
and genetic indications for dietary therapy such
as infantile spasms, myoclonic-astatic epilepsy,
Dravet syndrome, Rett syndrome, tuberous scle-
rosis complex, and more (Bergin, chapter 6; Hsieh
and Thiele, chapter  7). In chapters  8 and 9 Drs.
Kessler and Nabbout highlight the more recent,
“novel” indications such as absence epilepsy, juve-
nile myoclonic epilepsy, status epilepticus, and
others that have attracted investigators in the last
few years (Nabbout et al., 2010).
Lastly, Drs. Herren and Said conclude this
Section with a review of the latest research on
how to identify and treat the adverse effects inher-
ent in dietary therapy as well as how to eventually
discontinue treatment when clinically indicated
(chapter  10). This important chapter shows how
clinical researchers are attempting to make the diet
safer for those who require it, especially long-term.
We hope you enjoy reading this Section and gain
understanding of just how far the clinical use of
dietary therapy has come in such a short time.
REFERENCES
Bergqvist, A.G., Schall, J.I., Gallagher, P.R., Cnaan, A.,
and Stallings, V. A., (2005). Fasting versus grad-
ual initiation of the ketogenic diet: a prospective,
randomized clinical trial of efficacy. Epilepsia 46,
1810–1819.
Cervenka, M.C., Henry, B., Nathan, J., Wood, S., and
Volek, J.S. (2013). Worldwide dietary therapies for
adults with epilepsy and other disorders. J Child
Neurol 28, 1034–1040.
Freeman, J.M., Vining, E.P., Kossoff, E.H., Pyzik, P.L.,
Ye, X., and Goodman, S.N. (2009). A blinded,
crossover study of the efficacy of the ketogenic
diet. Epilepsia 50, 322–325.
Klepper, J. (2012). GLUT1 deficiency syndrome in
clinical practice. Epilepsy Res 100, 272–277.
Kossoff, E.H., and McGrogan, J.R. (2005). Worldwide
use of the ketogenic diet. Epilepsia 46, 280–289.
Kossoff, E.H., Zupec-Kania, B.A., Amark, P.E.,
Ballaban-Gil, K.R., Christina Bergqvist, A.G.,
Blackford, R., Buchhalter, J.R., Caraballo, R.H.,
Helen Cross, J., Dahlin, M. G., et al. (2009). Optimal
clinical management of children receiving the keto-
genic diet:  recommendations of the International
Ketogenic Diet Study Group. Epilepsia 50, 304–317.
Nabbout, R., Mazzuca, M., Hubert, P., Peudennier, S.,
Allaire, C., Flurin, V., Aberastury, M., Silva, W.,
and Dulac, O. (2010). Efficacy of ketogenic diet
in severe refractory status epilepticus initiating
fever induced refractory epileptic encephalopa-
thy in school age children (FIRES). Epilepsia 51,
2033–2037.
Neal, E.G., Chaffe, H., Schwartz, R.H., Lawson, M.S.,
Edwards, N., Fitzsimmons, G., Whitney, A., and
Cross, J.H. (2008). The ketogenic diet for the treat-
ment of childhood epilepsy:  a randomised con-
trolled trial. Lancet Neurol 7, 500–506.
Sharma, S., Sankhyan, N., Gulati, S. & Agarwala, A.
(2013). Use of the modified Atkins diet for treat-
ment of refractory childhood epilepsy: a random-
ized controlled trial. Epilepsia 54, 481–486.
Vining, E.P., Freeman, J.M., Ballaban-Gil, K., Camfield,
C.S., Camfield, P.R., Holmes, G.L., Shinnar, S.,
Shuman, R., Trevathan, E., and Wheless, J.W.
(1998). A multicenter study of the efficacy of the
ketogenic diet. Arch Neurol 55, 1433–1437.
Wilder, R.M. (1921). The effects of ketonemia on the
course of epilepsy. Mayo Clin Proc 2, 307–308.
5
2
“Alternative” Ketogenic Diets
ELIZABETH NEAL, RD, MSC, PHD
INTRODUCTION
As the classical ketogenic diet fast approaches
a centennial anniversary, the wider ketogenic
landscape has expanded considerably both in
application and implementation. Although still
extensively used today, this traditional dietary
therapy has been the basis for development of
alternative ketogenic protocols. One ketogenic diet
incorporating medium chain fatty acids is used for
many children and adolescents, who benefit from
the generous carbohydrate allowance facilitated by
the increased ketogenic potential of medium chain
triglycerides. More recently, two less restrictive
dietary approaches have been developed: the low
glycemic index treatment and the modified Atkins
diet. These are now being used worldwide as the
advantages of a more liberal ketogenic diet are rec-
ognized, especially in adults and older children,
supported by an increasing body of scientific data.
This chapter explores the background and evi-
dence for use of these alternative ketogenic diets.
THE MEDIUM CHAIN
T R I G LY C E R I D E
K E TO G E N I C   D I E T
The predominant fatty acids in the human diet
contain 12 or more carbon atoms and originate
from animal and plant sources of long chain tri-
glycerides (LCT), which can be saturated, mono-
unsaturated, or polyunsaturated. The shorter chain
length medium chain fatty acids (6 to 12 carbon
atoms) originate from medium chain triglycer-
ides (MCT), whose main constituents are octa-
noic (C8) and decanoic (C10) fatty acids. Dietary
sources are limited: mainly coconut and palm
kernel oils. Medium chain triglycerides have dis-
tinct physical and metabolic differences from LCT
with a more efficient digestion, absorption, and
mitochondrial transport process facilitating faster
metabolism to acetyl CoA. Hepatic ketone body
production is primarily determined by the rate of
acetyl CoA generation, which led to suggestions by
Dr. Huttenlocher and colleagues that a ketogenic
diet (KD) replacing LCT with MCT would induce
higher ketosis and allow inclusion of significantly
more carbohydrate and protein, improving palat-
ability and acceptance. After an initial trial of a
KD providing 60% of total dietary energy from
MCT in 12 children and adolescents with epi-
lepsy (Huttenlocher et al., 1971), further results
were reported from 18 patients aged 1.5–18 years,
of whom 16 had over 50% seizure reduction
(Huttenlocher 1976).
Interest in the MCT diet continued with further
studies reported from the United States (Trauner
et  al., 1985), the United Kingdom (Sills et  al.,
1986), and Taiwan (Mak et  al., 1999). A  dietary
modification with less MCT (30% energy) was
also suggested in response to concerns about gas-
trointestinal side effects of MCT given in large
doses (Schwartz et al., 1989). In 2008, researchers
based at Great Ormond Street Hospital in London
published a trial of classical and MCT KDs in
intractable childhood epilepsy in which children
aged 2–16 years were randomized to receive a diet
either immediately or after a 3-month delay with
no additional treatment changes (control group).
After 3 months, seizure frequency was significantly
lower in the 54 children in the diet group com-
pared with the 49 controls (Neal et al., 2008a). Of
the children who were randomized, 125 received
dietary treatment at some stage (61 classical and
64 MCT diets). Comparing the two diet groups
using an intention to treat analysis found no sig-
nificant difference between the two diets; 29% of
the MCT group had over 50% seizure reduction at
3 months (Neal et al., 2009). Tolerability or with-
drawals were also not significantly different at 3
and 6 months, with no evidence that the MCT diet
caused more gastrointestinal problems; indeed a
history of vomiting was significantly higher in the
classical KD children at 12 months. In this trial, the
MCT diet was initiated at a starting dose of 40%–
50% energy from the MCT supplement, aiming to
6

6 section I: Ketogenic Diet for Epilepsy in the Clinic

BOX 2.1
MEDIUM CHAIN TRIGLYCERIDE DIETARY PROTOCOL

• Starting dose of 40%–50% energy MCT given as prescribed supplement of oil or emulsion
divided between all meals and snacks (MCT % can be increased as needed and tolerated
during dietary fine-tuning)
• Protein—10% energy, increase to ≥ 12% if low energy needs to ensure meeting protein
requirements
• Carbohydrate—15%–18% energy (may be lower in older children)
• Remaining 20%–30% energy LCT (from foods)
• Food choice lists or electronic calculation of recipes, all food weighed
• Stepwise increase to starting MCT dose over 1–2 weeks during which rest of diet can be
implemented as above, although may need extra LCT to maintain total energy intake if
slower MCT introduction required
• Full vitamin and mineral supplementation
• Carbohydrate-free medications where possible

Source: (Neal, 2012)

provide the optimal balance between gastrointes-
tinal tolerance and good ketosis. However many
children and adolescents will need a higher dose
to achieve optimal seizure control. Christiana Liu
reports that in her extensive experience of using
the MCT diet in Canada, MCT at 40% to greater
than 70% energy can be well tolerated without side
effects (Liu and Wang, 2013). Prospective follow-
up of 48 children and adolescents aged 1–18 years
on mostly (79%) the MCT diet has recently been
reported from Holland. Responder rates were
lower in this study, only 17% achieving over 50%

Fat - 90% Fat - 60%

Protein - 6% Protein - 30%
Carbohydrate - 4% Carbohydrate - 10%

Classical KD at a 4 : 1 ratio Low glycemic index treatment

Fat - 73% (30%-60% MCT)

Fat - 65%
Protein - 10%
Protein - 30%
Carbohydrate - 17%
Carbohydrate - 5%

Medium chain triglyceride KD Modified Atkins diet

FIGURE 2.1 Ketogenic diet therapies: a comparison of dietary energy contribution from macronutrients.
7
Chapter 2: “Alternative” Ketogenic Diets
seizure reduction after 3  months, increasing to
23% after 6 months (Lambrechts et al., 2015).
The MCT diet is implemented using commer-
cially available products of MCT oil or emulsion
(Liquigen, Nutricia, 50% MCT; Betaquik, Vitaflo,
20% MCT), which are supplied in some countries
on medical prescription. The remaining energy is
provided from carbohydrate, protein, and LCT.
Calculation of this diet is not based on the keto-
genic ratio but instead looks at the percentage of
dietary energy provided by macronutrients (Box 2.1).
Total energy intake is controlled as with the classi-
cal KD, although it will theoretically depend on the
figure applied for energy content of MCT, which
is lower than LCT (Ranhotra et  al., 1995); this is
not always reflected in the conversion factors listed
on the products or used for dietary calculation.
The MCT diet is strictly prescribed and all food
weighed, often using food choice lists to develop
meal plans. It is the most generous in carbohy-
drate of all ketogenic therapies (see Figure 2.1),
and many children and adolescents benefit from the
flexibility this offers. Medium chain triglycerides
should be included in all meals and snacks, and
compliance is improved by encouraging creative
incorporation into recipes and ketogenic drinks.
The dose of MCT should be slowly built up over
the first week or two of treatment according to tol-
erance, ketosis, and seizure control (see Box 2.1).
Recent data indicate there may be additional
specific efficacy benefits of medium chain fatty
acids. Chang et al. (2013) found C10 significantly
outperformed valproate acid in both in vitro and
in vivo models of seizure control. Neuronal cell-
line data from Hughes et al. (2014) suggested that
C10 might increase mitochondrial number, medi-
ated via activation of the PPARγ (peroxisome
proliferator-activated) receptor and its target genes
involved in mitochondrial biogenesis.
THE MODIFIED
AT K I N S   D I E T
In 2003 Dr.  Kossoff and colleagues at Johns
Hopkins Hospital in Baltimore published a brief
communication to report their use of the Atkins’s
diet in six patients with epilepsy aged between 7
and 52  years; three had over 90% seizure reduc-
tion, of whom two became seizure free (Kossoff
et al., 2003). The Atkins diet, designed in the 1970s
as a weight loss treatment, restricts carbohydrates
and encourages fat in a similar way to the classical
KD but allows free protein. It was suggested that
this could be the basis of a less restrictive keto-
genic therapy for epilepsy, the goal being seizure
control rather than weight loss. The team at Johns
7
Hopkins went on to trial this modified Atkins diet
(MAD) in 20 children: 13 achieved over 50% sei-
zure reduction after 6 months, including four who
became seizure free (Kossoff et  al., 2006). In a
further study in 30 adults, seizures were reduced
by over 50% in 10 patients after 6 months on the
MAD (Kossoff et al., 2008a). The real advantage of
this diet is that it allows free protein and calories,
so can be easier to implement and comply with
than the classical KD. Although approximating
a ketogenic ratio of 1:1 (see Figure 2.1), the only
macronutrient strictly controlled on the MAD is
carbohydrate. A  randomized crossover compari-
son of daily carbohydrate limits in children sug-
gested a lower intake (10 g vs. 20 g) during the
initial 3 months of the MAD was associated with
significantly higher likelihood of over 50% seizure
reduction at 3 months, after which time carbohy-
drate could be increased (Kossoff et al., 2007).
The MAD has led the way in a shift of approach
to implementation of ketogenic therapy. The
emphasis had been previously on absolute preci-
sion in calculation and accuracy in food weigh-
ing, albeit with very successful outcomes in many
who followed the strict KD, but with compliance
problems in others. As practitioners of the diet,
we initially viewed what appeared to be such a
liberal MAD protocol with caution. Now, over
10 years on, this is being increasingly adopted as
an alternative ketogenic therapy, especially suited
to adolescents, adults, and those unable to com-
ply with the stricter classical KD. It is used with
success worldwide in children and adults (see
Table 2.1, studies included if five or more subjects),
and has potential for use in resource-poor coun-
tries with more limited dietetic support (Kossoff
et al., 2008b).
A review in 2012 combined data from published
MAD studies to examine responder rates. After
3  months of treatment, 54% of 105 children (six
studies, both retrospective and prospective) and
34% of 56 adults (three studies, all prospective) had
greater than 50% seizure reduction. Prospective
data was available on 82 children (four studies), of
whom 52% had greater than 50% seizure reduc-
tion after 3  months (Auvin, 2012). The following
year, Dr.  Kossoff and colleagues comprehensively
reviewed 10 years of the MAD with similar findings.
Their review included all published primary stud-
ies in which the MAD was used as the first dietary
treatment and also case reports of single patients.
Of a combined total of 342 children, 53% had over
50% seizure reduction, with 15% achieving seizure
freedom; in a combined total of 92 adults these
figures were lower at 30% and 3%, respectively
8

TABLE 2.1 WORLDWIDE USE OF THE MODIFIED ATKINS DIET

Country Study Dietary prescription—daily

carbohydrate allowance*

Children and adolescents

Argentina Vaccarezza et al., 2014 (Retrospective, n = 9) 10% energy

Denmark Weber et al., 2009 (Prospective, n = 15) 10% energy, restricted further to 10 g
Miranda et al., 2011 (Prospective, n = 33) after 1–2 weeks if poor seizure control
(Weber)
10 g for 1st 3 months (Miranda)
Egypt El-Rashidy et al., 2013 (Prospective, n = 15) 10 g for 1st month then can increase by 5-g
increments up to 10% energy
France Porta et al., 2009 (Prospective, n = 10) 10 g for 1st month then can increase by 5-g
increments up to 10% energy
India Sharma et al., 2012 (Prospective, n = 15) 10 g
Sharma et al., 2013 (Randomized controlled
trial, n = 50 in diet group)
Sharma et al., 2015 (Retrospective, n = 25)
Iran Tonekaboni et al., 2010 (Prospective, n = 51) 10 g
Ghazavi et al., 2014 (Retrospective, n = 20)
Japan Ito et al., 2011 (Retrospective, n = 6) 10 g initially (one child needed 30 g–20 g
Kumada et al., 2012 (Prospective, n = 10) step down start, Kumada et al.)
Korea Kang et al., 2007 (Prospective, n = 14) 10 g for 1st month then can increase by
Kim et al., 2012 (Retrospective, n = 20) 5-g increments up to 10% energy
USA Kossoff et al., 2003 (Retrospective, n = 6) 10 g for 1–2 months then can increase by
Kossoff et al., 2006 (Prospective, n = 20) 5-g increments up to 20 g
Kossoff et al., 2007 (Prospective, n = 20) (20 g initially for five Sturge-Weber
Kossoff et al., 2010a (Prospective, n = 5) syndrome children in 2010 paper)
Kossoff et al., 2011a (Prospective, n = 30)
Groomes et al., 2011 (Retrospective, n = 13)
Adults
Belgium Carrette et al., 2008 (Prospective, n = 8) 20 g
Canada Smith et al., 2011 (Prospective, n = 18) 20 g
USA Kossoff et al., 2008a (Prospective, n = 30) 15 g
Cervenka et al., 2012 (Prospective, n = 25) 20 g
Kossoff et al., 2013a (Retrospective, n = 8) 20 g

* all studies allowed free calories except Kim et al., 2012, who report using 75% of recommended daily intakes

BOX 2.2
MODIFIED ATKINS DIET PROTOCOL

Carbohydrate for first month: 10 g daily for children, 10–15 g daily for adolescents and 20 g
daily for adults (does not include fiber but does include sugar alcohols)
Encourage high-fat foods, eat with each meal/snack
Free protein
Free calories but need to avoid excess weight gain
Full vitamin and mineral supplementation
Carbohydrate-free medications
Ketocal formula can be used as daily supplement for first month

Source: (Kossoff et al., 2011b)

9

Chapter 2: “Alternative” Ketogenic Diets 9

(Kossoff et al., 2013b). This review included results
from Dr. Sharma and colleagues in India, who pub-
lished the first randomized trial of the MAD in 102
children with intractable epilepsy aged 2–14 years.
Using a delayed diet start control group in a design
similar to that of Neal et  al. (2008a), seizure fre-
quency after 3  months was significantly lower in
the 50 diet children compared with the 52 controls
(Sharma et al., 2013). One study has looked at long-
term outcomes of children on the MAD: over 50%
seizure reduction was maintained in 55% of the 54
who continued treatment for over 6 months (Chen
and Kossoff, 2012).
Most centers prescribe the diet as recom-
mended by Dr.  Kossoff and his team at Johns
Hopkins (Box 2.2). A  dietary variant adopted
by some UK centers and described by Magrath
et al. (2012) is frequently and more appropriately
termed the modified ketogenic diet; this is distin-
guished from the MAD by a more generous initial
carbohydrate allowance (up to 30 g with further
reduction during fine-tuning depending on sei-
zure control) and a prescribed fat intake using
food choice lists. There is no published data on this
type of modified diet.
L O W G LY C E M I C I N D E X
T R E AT M E N T
The glycemic index (GI) is a measurement based
on the blood glucose response to carbohydrate-
containing foods (Jenkins et al., 1981); this will be
influenced by their rate of digestion and absorp-
tion, slower absorbed foods having a lower GI rat-
ing. Glycemic index values are compared with a
standard reference value of glucose. Other variables
influencing the GI of a food include fiber content,
cooking methods, processing, ripeness, and com-
bination of different macronutrients within a meal.
Whole grains and other high-fiber foods will lower
GI, as will the addition of fat or protein. Most veg-
etables and many fruits are low in GI.
Blood glucose levels tend to be fairly stable
while on the low-carbohydrate KD, and this obser-
vation led to the suggestion that a diet based on
only low GI carbohydrate (<50) choices could
maintain this glucose stability and facilitate a more
liberal type of ketogenic regime (see Figure 2.1).
This alternative low GI treatment (LGIT) (Box 2.3)
was first tested by researchers from Boston: a pre-
liminary retrospective review of 20 patients found
half to have greater than 90% seizure reduction
(Pfeifer & Thiele, 2005). Results were updated in
2009 with a larger review of 76 patients aged 1.5–
22 years, half of whom had over 50% seizure reduc-
tion at 3 months, increasing to 66% by 12 months
(Muzykewicz et al., 2009). Interestingly, carbohy-
drate intake ranged from 15 to 150 g daily (mean
53 g at 3 months); some individuals needed a more
restricted amount for seizure control, whereas
others were able to relax the carbohydrate without
adverse effect. The same group has successfully
used LGIT in 15 children with tuberous sclero-
sis (Larson et  al., 2012)  and a small group of six
children with Angelman syndrome studied pro-
spectively (Thibert et  al., 2012). Again, reported
carbohydrate intake was very varied, between 30
and 137 g at the 4-month follow-up.
The LGIT is now used in many centers world-
wide. Results in children and adolescents have
been reported from Italy (retrospective data on

BOX 2.3
LOW GLYCEMIC INDEX TREATMENT DIETARY
PROTOCOL

Dietary goals set for carbohydrate, protein, and fat intake based on a target energy prescription
Carbohydrate prescribed at 10% energy (approximately 40–60 g daily depending on baseline
intake); this figure includes dietary fiber. Only carbohydrates with glycemic index < 50
allowed
Approximately 30% energy protein
Approximately 60% energy LCT
Carbohydrate should be evenly distributed over the day and always eaten with some fat and/
or protein
Foods not weighed but based on household portion sizes

Source: (Pfeifer, 2012)
10

10 section I: Ketogenic Diet for Epilepsy in the Clinic

15 patients: over half had over 50% seizure reduc-
tion after a mean treatment duration of 14 months;
Coppola et al., 2011) and Iran (prospective study
of 42 patients: over 50% seizure reduction in 74%
after 1 month; Karimzadeh et al., 2014). An inter-
esting case study reports a 13-year-old Japanese
girl who was able to maintain a 50-g carbohy-
drate daily LGIT choosing from specially designed
menus including unpolished rice and Natto (fer-
mented soybeans) (Kumada et al., 2013).
A R E “A LT E R N AT I V E ”
DIETS AS EFFECTIVE
AS THE CLAS SICAL
K E TO G E N I C   D I E T ?
With the range of ketogenic therapies available, the
initial clinic assessment consultation will include
consideration of which diet to choose for an indi-
vidual patient. A key question for those embarking
on ketogenic therapy is which diet could work best
to treat the seizures. Will an alternative protocol
with potential compliance and tolerance advan-
tages still be as effective as the strict classical KD?
The scientific literature has also discussed this
question and certainly review of the many MAD
trials shows results comparable to those for the
KD (Kossoff et al., 2013b), including longer-term
follow-up data (Chen and Kossoff, 2012). Direct
comparison trials of different ketogenic therapies
are limited. In the trial of Neal et  al. previously
discussed, there were no significant differences
in the mean percentage of baseline seizures or
numbers, with over 50% or 90% seizure reduc-
tion between the classical and MCT diet groups
after 3, 6, and 12 months (Neal et al., 2009). The
authors concluded classical and MCT KDs were
comparable in efficacy and tolerability and both
had their place in the treatment of childhood epi-
lepsy. In view of the similar design between this
trial and the more recent MAD randomized trial
already mentioned (Sharma et  al., 2013), results
of the two studies are provided together in Table
2.2. Although we cannot directly compare data,
seizure improvement was lower in the KD trial,
which may in part reflect differences in baseline
patient characteristics, as a similar difference was
seen in the control group.
A number of studies have tried to compare
classical KD and the MAD. A retrospective review
of children on classical KD or MAD found sig-
nificantly more KD children had greater than 50%
seizure reduction at 3 months, but not at 6 months
(Porta et  al., 2009). A  prospective evaluation of
children on the MAD compared with a previously
treated KD group found an initial trend toward
greater KD efficacy disappeared when data were
age-adjusted (Miranda et  al., 2011). Data from
Iran on 40 children of whom half were prescribed
a classical KD and the others a MAD showed no
significant difference in numbers achieving 50%
seizure reduction after 1–3  months (Ghazavi
et al., 2014).

TABLE 2.2 RESULTS OF RANDOMIZED CONTROLLED TRIALS ON

THE KETOGENIC DIET AND THE MODIFIED ATKINS DIET

Sharma et al., 2013 (MAD) Neal et al., 2008a (classical and MCT KDs)

Subjects
Number enrolled 102 (2–14 years) 145 (2–16 years)
in study (ages)
Inclusion criteria Minimum 7 weekly seizures, tried at least 3 Minimum 7 weekly seizures, tried at least 2
anticonvulsant medications, no compli- anticonvulsant medications, no previous
ance issues or medical contraindications ketogenic diet, no medical contraindications
Results
Mean % 59% diet group (44–75) 63% diet group (50–74)
baseline seizure 95% control group (82–109) 137% control group (105–169)
frequency (CI)
>90% seizure 30% diet 7% diet
reduction 8% controls 0% controls
>50% seizure 52% diet 38% diet
reduction 12% controls 6% controls
Most frequent Constipation (46%) Constipation (33%)
side effect
11
Chapter 2: “Alternative” Ketogenic Diets
However other data suggest there may be
benefits of a stricter diet. In a retrospectively ana-
lyzed multicenter group of 27 children switched
from the MAD to classical KD, additional seizure
reduction was reported in 10, of whom five with
Doose syndrome (myoclonic astatic epilepsy)
became seizure-free; the authors identified the KD
as a “higher dose” of ketogenic therapy than the
MAD (Kossoff et al., 2010b). In a small pilot study,
researchers in Egypt randomly assigned 40 young
children (aged 1–3 years) to either classical KD fed
by a liquid formula, MAD, or no diet treatment.
After 3 months the classical KD group showed sig-
nificantly reduced seizure frequency and severity
compared with the MAD group; at 6 months the
reduction in frequency was still significant, but not
that in severity (El-Rashidy et al., 2013).
A randomized trial comparing efficay, saftey
and tolerability of classical KD and MAD in 104
children aged 1–18  years (51 KD, 53 MAD) has
recently been published. Mean percentage of base-
line seizures was lower in the KD group at 3 months
(38.6% KD, 47.9% MAD) and 6  months (33.8%
KD, 44.6% MAD), although differences were not
statistically significant. In infants aged 1–2  years
the mean and median baseline seizure frequen-
cies were much lower in the classical KD group
with seizure freedom after 3 months achieved in 9
(53%) of 17 KD patients compared to 4 (20%) of 20
MAD patients. The MAD showed tolerability and
side effect advantages. The authors concluded that
while the MAD may be suitable for many children,
the classical KD should always be first choice in
those under 2 years (Kim et al., 2016).
Although one review of studies on dietary
treatment in adults with refractory epilepsy con-
cluded that both classical KD and MAD were
equally effective and tolerated (Klein et al., 2014),
this was challenged by a recent meta-analysis of 12
studies on ketogenic therapy in adults: 7 classical
KD, 5 MAD, and one MCT. A total of 270 patients
were evaluated with a combined efficacy rate of
52% for classical KD and 34% for MAD and an
odds ratio for therapeutic success of classical KD
relative to MAD of 2.04, a significant difference
between the two types of diet (Ye et al., 2015). This
analysis examined compliance, which was also sig-
nificantly different between the two diets, at 38%
classical KD and 56% MAD, suggesting that while
the classical KD may be more effective in adults, it
is not as well tolerated.
There is some evidence from the above stud-
ies that a stricter diet might be more efficacious.
This could be particularly important at the out-
set of treatment, a hypothesis supported by two
11
randomized trials examining the use of different
prescriptions within a particular ketogenic therapy.
A lower carbohydrate MAD initiation (10 g vs. 20
g) was associated with improved efficacy outcome
(Kossoff et al., 2007), as was a higher classical KD
ketogenic ratio (4:1 vs. 3:1) (Seo et  al., 2007); in
both these studies the benefits of a stricter diet at
the outset were maintained even after increasing
carbohydrate intake later in the course of treatment.
Although it is now recognized that ketosis may
not be directly linked to seizure control, any cor-
relations between the two seem limited to the first
3  months of dietary treatment (Neal et  al., 2009).
Further support is provided by results of a study of
30 children at Johns Hopkins who were given the
4:1 ratio classical KD supplement Ketocal daily dur-
ing the first month of the MAD; this had benefit
on seizure control when responder rate outcomes
were compared to published results of the MAD
alone (Kossoff et  al., 2011a). This practice is now
frequently implemented at their institution (see
Box 2.2), where the importance of a strict initiation
period during the first month has been recognized
(Kossoff et al., 2013b). It could perhaps be argued
that key questions to ask when embarking on keto-
genic therapy are not only “which dietary protocol
to choose?” but also “how we are going to manage
the initiation period for optimal seizure outcomes?”
SO WHICH DIET
TO   C H O O S E ?
The question of which diet to use for an individ-
ual will also take into account age, lifestyle, food
preferences, and feeding method. As illustrated
in Figure  2.1, all three of the alternative dietary
protocols have less fat than the classical KD as
a proportion of the overall dietary energy. The
MCT diet is the most generous in carbohydrate,
but consideration must be given to the need to
incorporate the MCT supplement into all meals
and snacks and the use of a strict prescription with
food weighing. However this can work well for
those who find it too difficult to follow a stricter
carbohydrate restriction and need a more struc-
tured dietary prescription. Adolescents and adults
usually prefer the flexibility of the MAD with no
food weighing and free protein and calories, and
to maintain ongoing compliance this option would
routinely be recommended for these age groups.
An alternative would be LGIT if unable to adhere
to the stricter MAD carbohydrate restriction. The
classical KD is recommended for all ages if using a
feeding tube and also for infants under 2 years. In
older children aged between 2 and 12 years it may
12
12 section I: Ketogenic Diet for Epilepsy in the Clinic
be sensible to start with a stricter KD in light of the
efficacy data previously discussed; this can then be
changed to the MAD at a later date if children and
parents are happy to switch (Kossoff et al., 2013b).
The classical KD will be preferable for children
who have poor appetites and need small meals, the
nutritionally at risk, and families who require close
supervision on meal planning and the control of
their child’s energy intake. In other cases it is clear
that both the child and the family would not be
able to comply with the high-fat classical KD and
the MAD or MCT diet are better options. In view
of its reduced demands on time for training and
supervision, the MAD would be first choice in cen-
ters with fewer dietitians. This potential in devel-
oping countries has been identified (Kossoff et al.,
2008b), including the possibility of e-mail-based
MAD management in adults (Cervenka et  al.,
2012). However both the MAD and LGIT require
patients or families to design their own meals from
the food choices given; in some situations such as
a residential multicarer setting this may be more
difficult and require greater dietetic input.
The stricter classical KD has been recom-
mended in epilepsy syndromes where rapid
improvement is needed, such as infantile spasms
or status epilepticus (Auvin, 2012)  and in Glut 1
deficiency syndrome and Doose syndrome
(Miranda et  al., 2012), although the MAD has
been used successfully in both spasms (Sharma
et al., 2012) and Glut 1 deficiency syndrome (Ito
et al., 2011). Following observed benefits in Doose
syndrome children changed from the MAD to
classical KD, this switch should be considered in
this group if not seizure-free after 6–12 months on
the MAD (Kossoff et al., 2010b).
PRACTICAL
C O N S I D E R AT I O N S :
I N I T I AT I O N A N D
F O L L O W- U P O F
“A LT E R N AT I V E ”   D I E T S
Medical contraindications to ketogenic therapy
are detailed in recommendations for clinical KD
management (Kossoff et  al., 2009). These will
apply to both classical and alternative KDs and
should be excluded prior to initiation with any
necessary baseline biochemistry. Although some
centers do hospitalize children on initiation of
the MCT diet, as with the classical KD it is now
generally accepted that this can be started at home
without any prior fasting period. The daily intake
of MCT will be stepwise, increased as tolerated
(Box 2.1). The MAD and LGIT are started on an
out-patient basis. All dietary protocols require full
vitamin and mineral supplementation to ensure
requirements of micronutrients are met while
on a restricted diet. Adequate training from the
ketogenic team must be given prior to initiation of
any ketogenic therapy to ensure patients and car-
ers understand the dietary prescription and how
to manage the practicalities of ketogenic therapy
at home, including strategies for illness and acute
situations.
All ketogenic therapies must be carefully mon-
itored to ensure they can be implemented safely.
Home monitoring will include regular weight
checks; this is also important on the MAD and
LGIT, where the calorie content is not strictly pre-
scribed. Traditionally, ketones would be checked
up to twice daily at home while on the KD, as good
ketosis was thought to have been key to the success
of treatment. This premise has been challenged
with alternative diets where ketones are often
much lower. Regular home testing of urine or
blood ketones is recommended on the MCT diet,
but levels are not as high as those seen on the clas-
sical KD (Neal et al., 2009). After the first month
of the MAD, with a relaxing of dietary restrictions,
ketones will usually be much lower than those seen
on the KD, especially in older children and adults.
Although Kossoff et al. (2011b) recommend only
a weekly check after this point, others suggest con-
sistently high ketones above 3  mmol/L (blood
β-hydroxybutyrate) in children on the MAD could
be important for maintaining efficacy (Kang et al.,
2007). Ketones on the LGIT will usually be very
low and undetectable in some cases; no correlation
was seen between ketosis and seizures on this diet
(Muzykewicz et al., 2009).
Clinic monitoring at regular follow-up visits
will include full laboratory studies, growth assess-
ment, review of seizures, tolerance, and other
benefits or adverse events. Side effects have been
reported with the MCT diet, primarily gastrointes-
tinal and usually managed with dietary manipu-
lation (Neal et  al., 2009; Liu and Wang, 2013).
Growth faltering has been reported in children on
MCT diets, similar to that seen in children on a
classical KD despite the significantly higher pro-
tein intake on the MCT diet (Neal et al., 2008b).
Gastrointestinal symptoms including constipation
have also been reported as a side effect of the MAD
(Kang et  al., 2007; Auvin, 2012; Chen & Kossoff,
2012; Sharma et  al., 2013). Raised lipid levels on
the MAD have been reported in children (Kang
et  al., 2007)  and adults (Cervenka et  al., 2014);
these were transient and normalized within a year
of treatment. No significant side effects have been
reported in LGIT studies, although increases in
13
Chapter 2: “Alternative” Ketogenic Diets
blood urea nitrogen were detected in about a third
of patients (Muzykewicz et al., 2009; Karimzadeh
et al., 2014).
Fine-tuning of ketogenic therapy will aim to
alleviate side effects where possible, and to opti-
mize seizure outcomes. Adjustments to prescrip-
tions and micronutrient supplementation will also
be needed as a child grows older. The MCT dose
may be increased or decreased on the MCT diet,
as can carbohydrate intake, aiming to maximize
benefit. As an addition to the MAD or modified
ketogenic diet, MCT has also been used to give a
boost to ketosis and seizure control and aid compli-
ance by facilitating increased carbohydrate allow-
ance. This practice is evidence of a more flexible
approach to ketogenic therapy, designing an indi-
vidualized treatment based primarily on specific
dietary and lifestyle requirements rather than on
a rigid diet protocol that is offered by a particular
hospital center. This may primarily use one type
of diet, but alternatively may use different aspects
of some, or indeed all, of the ketogenic therapies.
Well-defined dietary parameters are needed if
conducting research studies on a specific diet, but
anecdotal reports suggest more dietitians are tend-
ing toward this “patient-tailored” prescription of
ketogenic therapy in clinical practice (Miranda
et al., 2012).
CONCLUSION
As worldwide use of the KD continues to grow,
it is clear that the alternative dietary protocols
described in this chapter have an important place
within the treatments we can offer to children and
adults with intractable seizures. The MAD in par-
ticular has emerged as a therapy with great poten-
tial for treating not only children with epilepsy
but also adults and those in countries with more
limited resources. Further research will enable us
to optimize protocols for clinical implementa-
tion to ensure the best possible outcome for those
embarking on dietary treatment of epilepsy.
REFERENCES
Auvin, S. (2012). Should we routinely use modified
Atkins diet instead of regular ketogenic diet to
treat children with epilepsy? Seizure 21, 237–240.
Carrette, E., Vonck, K., De Herdt, V., DeWaele, I.,
Raedt, R., Goossens, L., Van Zandijcke, M.,
Wadman, W., Thadani, V., and Boon P. (2008).
A pilot trial with modified Atkins’ diet in adult
patients with refractory epilepsy. Clin Neurol
Neurosurg 110, 797–803.
Cervenka, M.C., Terao, N.N., Bosarge, J.L., Henry,
B.J., Klees, A.A., Morrison, P.F., and Kossoff,
13
E.H. (2012). E-mail management of the modified
Atkins diet for adults with epilepsy is feasible and
effective. Epilepsia 53, 728–732.
Cervenka, M.C., Patton, K., Eloyan, A., Henry, B., and
Kossoff, E.H. (2014). The impact of the modified
Atkins diet on lipid profiles in adults with epilepsy.
Nutr Neurosci [Epub ahead of print].
Chang, P., Terbach, N., Plant, N., Chen, P.E., Walker,
M.C., and Williams, R.S. (2013). Seizure control
by ketogenic diet-associated medium chain fatty
acids. Neuropharmacology 69, 105–114.
Chen, W., and Kossoff, E.H. (2012). Long-term follow-
up of children treated with the modified Atkins
diet. J Child Neurol 27, 754–758.
Coppola, G., D’Aniello, A., Messana, T., Di Pasquale,
F., della Corte, R., Pascotto, A., and Verrotti, A.
(2011). Low glycemic index diet in children and
young adults with refractory epilepsy: first Italian
experience. Seizure 20, 526–528.
El-Rashidy, O.F., Nassar, M.F., Abdel-Hamid, I.A.,
Shatla, R.H., Abdel-Hamid, M.H., Gabr, S.S.,
Mohamed, S.G., El-Sayed, W.S., and Shaaban, S.Y.
(2013). Modified Atkins diet vs classic ketogenic
formula in intractable epilepsy. Acta Neurol Scand
128, 402–408.
Ghazavi, A., Tonekaboni, S.H., Karimzadeh, P.,
Nikibakhsh, A.A., Khajeh, A., and Fayyazi, A.
(2014). The ketogenic and Atkins diets effect on
intractable epilepsy:  a comparison. Iran J Child
Neurol 8, 12–17.
Groomes, L.B., Pyzik, P.L., Turner, Z., Dorward, J.L.,
Goode, V.H., and Kossoff, E.H. (2011). Do patients
with absence epilepsy respond to ketogenic diets? J
Child Neurol 26, 160–165.
Hughes, S.D., Kanabus, M., Anderson, G., Hargreaves,
I.P., Rutherford, T., O’Donnell, M., Cross, J.H.,
Rahman, S., Eaton, S., and Heales, S.J. (2014).
The ketogenic diet component decanoic acid
increases mitochondrial citrate synthase and com-
plex I activity in neuronal cells. J Neurochem 129,
426–433.
Huttenlocher, P.R., Wilbourne, A.J., and Sigmore, J.M.
(1971). Medium chain triglycerides as a therapy
for intractable childhood epilepsy. Neurology 1,
1097–1103.
Huttenlocher, P.R. (1976). Ketonaemia and sei-
zures: metabolic and anticonvulsant effects of two
ketogenic diets in childhood epilepsy. Pediatr Res
10, 536–540.
Ito, Y., Oguni, H., Ito, S., Oguni, M., and Osawa, M.
(2011). A modified Atkins diet is promising as a
treatment for glucose transporter type 1 deficiency
syndrome. Dev Med Child Neurol 53, 658–663.
Jenkins, D.J., Wolever, T.M., Taylor, R.H., Barker,
H., Fielden, H., Baldwin, J.M., Bowling, A.C.,
Newman, H.C., Jenkins, A.L., and Goff, D.V.
(1981). Glycemic index of foods:  a physiological
basis for carbohydrate exchange. Am J Clin Nutr
34, 362–366.
14
14 section I: Ketogenic Diet for Epilepsy in the Clinic
Kang, H.C., Lee, H.S., You, S.J., Kang, du C., Ko, T.S.,
and Kim, H.D. (2007). Use of a modified Atkins
diet in intractable childhood epilepsy. Epilepsia 48,
182–186.
Karimzadeh, P., Sedighi, M., Beheshti, M., Azargashb,
E., Ghofrani, M., and Abdollahe-Gorgi, F. (2014).
Low glycemic index treatment in pediatric refrac-
tory epilepsy: the first Middle East report. Seizure
23, 570–572.
Kim, Y.M., Vaidya, V.V., Khusainov, T., Kim, H.D.,
Kim, S.H., Lee, E.J., Lee, Y.M., Lee, J.S., and Kang,
H.C. (2012). Various indications for a modified
Atkins diet in intractable childhood epilepsy.
Brain Dev 34, 570–575
Kim, J.A.,Yoon, J.R, Lee, E.J., Lee, J.S., Kim, J.T., Kim,
H.D. and Kang, H.C. (2016). Efficacy of the classic
ketogenic and the modified Atkins diets in refrac-
tory childhood epilepsy. Epilepsia 57(1), 51–58
Klein, P., Tyrlikova, I., and Mathews, G.C. (2014).
Dietary treatment in adults with refractory epi-
lepsy: a review. Neurology 83, 1978–1985.
Kossoff, E.H., Krauss, G.L., McGrogan, J.R., and
Freeman, J.M. (2003). Efficacy of the Atkins diet
as therapy for intractable epilepsy. Neurology 61,
1789–1791.
Kossoff, E.H., McGrogan, J.R., Bluml, R.M., Pillas, D.J.,
Rubenstein, J.E., and Vining, E.P. (2006). A modi-
fied Atkins diet is effective for the treatment of
intractable pediatric epilepsy. Epilepsia 47, 421–424.
Kossoff, E.H., Turner, Z., Bluml, R.M., Pyzik, P.L.,
and Vining, E.P. (2007). A randomized, crossover
comparison of daily carbohydrate limits using the
modified Atkins diet. Epilepsy Behav 10, 432–436.
Kossoff, E.H., Rowley, H., Sinha, S.R., and Vining,
E.P. (2008a). A prospective study of the modi-
fied Atkins diet for intractable epilepsy in adults.
Epilepsia 49, 316–319.
Kossoff, E.H., Dorward, J.L., Molinero, M.R., and
Holden, K.R. (2008b). The modified Atkins diet: a
potential treatment for developing countries.
Epilepsia 49, 1646–1647.
Kossoff, E.H., Zupec-Kania, B.A., Amark, P.E.,
Ballaban-Gil, K.R., Bergqvist, A.G., Blackford,
R., Buchhalter, J.R., Caraballo, R.H., Cross,
J.H., Dahlin, M.G., et  al. (2009). Optimal clini-
cal management of children receiving the keto-
genic diet:  recommendations of the International
Ketogenic Diet Study Group. Epilepsia 50, 304–317.
Kossoff, E.H., Borsage, J.L., and Comi, A.M. (2010a).
A pilot study of the modified Atkins diet for Sturge-
Weber syndrome. Epilepsy Res 92, 240–243.
Kossoff, E.H., Bosarge, J.L., Miranda, M.J., Wiemer-
Kruel, A., Kang, H.C., and Kim, H.D. (2010b). Will
seizure control improve by switching from the
modified Atkins diet to the traditional ketogenic
diet? Epilepsia 51, 2496–2499.
Kossoff, E.H., Dorward, J.L., Turner, Z., and Pyzik,
P. (2011a). Prospective study of the modified
Atkins diet in combination with a ketogenic
liquid supplement during the initial month. J
Child Neurol 26, 147–151.
Kossoff, E.H., Freeman, J.M., Turner, Z., and Rubenstein,
J.E. (2011b). Ketogenic Diets:  Treatments for
Epilepsy and other Disorders, 5th edition.
(New York: Demos Medical Publishing).
Kossoff, E.H., Henry, B.J., and Cervenka, M.C. (2013a).
Efficacy of dietary therapy for juvenile myoclonic
epilepsy. Epilepsy Behav 26, 162–164.
Kossoff, E.H., Cervenka, M.C., Henry, B.J., Haney,
C.A., and Turner, Z. (2013b). A decade of the mod-
ified Atkins diet (2003–2013): results, insights, and
future directions. Epilepsy Behav 29, 437–442.
Kumada, T., Miyajima, T., Oda, N., Shimomura, H.,
Saito, K., and Fujii, T. (2012). Efficacy and toler-
ability of modified Atkins diet in Japanese chil-
dren with medication-resistant epilepsy. Brain Dev
34, 32–38.
Kumada, T., Hiejima, I., Nozaki, F., Hayashi, A., and
Fujii, T. (2013). Glycemic index treatment using
Japanese foods in a girl with Lennox-Gastaut syn-
drome. Pediatr Neurol 48, 390–392.
Lambrechts, D.A., de Kinderen, R.J., Vles, H.S., de
Louw, A.J., Aldenkamp, A.P., and Majoie, M.J.
(2015). The MCT-ketogenic diet as a treatment
option in refractory childhood epilepsy:  a pro-
spective study with 2-year follow-up. Epilepsy
Behav 51, 261–266.
Larson, A.M., Pfeifer, H.H., and Thiele, E.A. (2012).
Low glycemic index treatment for epilepsy in
tuberous sclerosis complex. Epilepsy Res 99,
180–182.
Liu, Y.M., and Wang, H.S. (2013). Medium-chain tri-
glyceride ketogenic diet, an effective treatment for
drug-resistant epilepsy and a comparison with
other ketogenic diets. Biomed J 36, 9–15.
Magrath, G., Leung, M.A., and Randall, T. (2012). The
Modified Atkins Diet. In Dietary Treatment of
Epilepsy, E.G. Neal, ed. (Oxford: Wiley-Blackwell),
pp. 89–99.
Mak, S.C., Chi, C.S., and Wan, C.J. (1999). Clinical expe-
rience of ketogenic diet on children with refractory
epilepsy. Acta Paediatr Taiwan 40, 97–100.
Miranda, M.J., Mortensen, M., Povlsen, J.H., Nielsen,
H., and Beniczky, S. (2011). Danish study of a mod-
ified Atkins diet for medically intractable epilepsy
in children: can we achieve the same results as with
the classical ketogenic diet? Seizure 20, 151–155.
Miranda, M.J., Turner, Z., and Magrath, G. (2012).
Alternative diets to the classical ketogenic
diet--can we be more liberal? Epilepsy Res 100,
278–285.
Muzykewicz, D.A., Lyczkowski, D.A., Memon, N.,
Conant, K.D., Pfeifer, H.H., and Thiele, E.A.
(2009). Efficacy, safety, and tolerability of the low
glycemic index treatment in pediatric epilepsy.
Epilepsia 50, 1118–1126.
Neal, E.G., Chaffe, H.M., Schwartz, R., Lawson, M.,
Edwards, N., Fitzsimmons, G., Whitney, A., and
15
Chapter 2: “Alternative” Ketogenic Diets
Cross, J.H. (2008a). The ketogenic diet in the treat-
ment of epilepsy:  a randomised controlled trial.
Lancet Neurol 7, 500–506.
Neal, E.G., Chaffe, H.M., Edwards, N., Lawson, M.,
Schwartz, R., and Cross, J.H. (2008b). Growth of
children on classical and MCT ketogenic diets.
Pediatrics 122, e334–340.
Neal, E.G., Chaffe, H.M., Schwartz, R., Edwards,
N., Fitzsimmons, G., Whitney, A., and Cross,
J.H. (2009). A randomized trial of classical and
medium-chain triglyceride ketogenic diets in
the treatment of childhood epilepsy. Epilepsia
49, 1–9.
Neal, E.G. (2012). The medium chain ketogenic diet.
In Dietary Treatment of Epilepsy, E.G. Neal, ed.
(Oxford: Wiley-Blackwell), pp. 78–88.
Pfeifer, H.H., and Thiele, E. (2005). Low glycemic index
treatment: A liberalized ketogenic diet for treatment
of intractable epilepsy. Neurology 65, 1810–1812.
Pfeifer, H.H. (2012). The low glycaemic index treat-
ment. In Dietary Treatment of Epilepsy, E.G.
Neal, ed. (Oxford: Wiley-Blackwell), pp. 100–108.
Porta, N., Vallée, L., Boutry, E., Fontaine, M., Dessein,
A.F., Joriot, S., Cuisset, J.M., Cuvellier, J.C., and
Auvin, S. (2009). Comparison of seizure reduction
and serum fatty acid levels after receiving the keto-
genic and modified Atkins diet. Seizure 18, 359–364.
Ranhotra, G.S., Gelroth, J.A., and Glaser, B.K. (1995).
Levels of medium-chain triglycerides and their
energy value. Cereal Chem 72, 365–367.
Schwartz, R.H., Eaton, J., Bower, B.D., and Aynsley-
Green A. (1989). Ketogenic diets in the treatment
of epilepsy:  short term clinical effects. Dev Med
Child Neurol 31, 145–151.
Seo, J.H., Lee, Y.M., Lee, J.S., Kang, H.C., and Kim HD.
(2007). Efficacy and tolerability of the ketogenic
diet according to lipid:nonlipid ratios:  compari-
son of 3:1 with 4:1 diet. Epilepsia 48, 801–805.
Sharma, S., Sankhyan, N., Gulati, S., and Agarwala, A.
(2012). Use of the modified Atkins diet in infantile
spasms refractory to first-line treatment. Seizure
21, 45–48.
15
Sharma, S., Sankhyan, N., Gulati, S., and Agarwala, A.
(2013). Use of the modified Atkins diet for treat-
ment of refractory childhood epilepsy: a random-
ized controlled trial. Epilepsia 54, 481–486.
Sharma, S., Jain, P., Gulati, S., Sankhyan, N., and
Agarwala A. (2015). Use of the modified Atkins
diet in Lennox Gastaut syndrome. J Child Neurol
30, 576–579.
Sills, M.A., Forsythe, W.I., Haidukewych, D.,
MacDonald, A., and Robinson, M. (1986). The
medium chain triglyceride diet and intractable
epilepsy. Arch Dis Child 61, 1168–1172.
Smith, M., Politzer, N., MacGarvie, D., McAndrews,
M.P., and Del Campo, M. (2011). Efficacy and tol-
erability of the modified Atkins diet in adults with
pharmacoresistant epilepsy:  a prospective obser-
vational study. Epilepsia 52, 775–780.
Thibert, R.L., Pfeifer, H.H., Larson, A.M., Raby,
A.R., Reynolds, A.A., Morgan, A.K., and Thiele,
E.A. (2012). Low glycemic index treatment for
seizures in Angelman syndrome. Epilepsia 53,
1498–1502.
Tonekaboni, S.H., Mostaghimi, P., Mirmiran, P.,
Abbaskhanian, A., Abdollah Gorji, F., Ghofrani,
M., and Azizi, F. (2010). Efficacy of the Atkins
diet as therapy for intractable epilepsy in children.
Arch Iran Med 13, 492–497.
Trauner, D.A. (1985). Medium chain triglyceride diet
in intractable seizure disorders. Neurology 35,
237–238.
Vaccarezza, M.M., Toma, M.V., Ramos Guevara, J.D.,
Diez, C.G., and Agosta, G.E. (2014). Treatment of
refractory epilepsy with the modified Atkins diet.
Arch Argent Pediatr 112, 348–351.
Weber, S., Mølgaard, C., Taudorf, K., and Uldall, P.
(2009). Modified Atkins diet to children and ado-
lescents with medical intractable epilepsy. Seizure
18, 237–240.
Ye, F., Li, X.J., Jiang, W.L., Sun, H.B., and Liu, J. (2015).
Efficacy of and patient compliance with a keto-
genic diet in adults with intractable epilepsy:  a
meta-analysis. J Clin Neurol 11, 26–31.
16
3
Dietary Therapy in Adults
History, Demand, and Results
E M I LY L . J O H N S O N, M D A N D M AC K E N Z I E C . C E RV E N K A , M D
H I S T O RY
Dietary therapy has been used for the treatment
of epilepsy since antiquity. Hippocrates wrote of
fasting “purifications” as a cure for seizures, and
reported that some of his contemporaries believed
certain foods such as eel and goat to exacerbate
or cause seizures (Hippocrates, c. 400 bc). In the
Roman era, drinking gladiators’ blood was thought
to be a cure for epilepsy (Barborka, 1929).
In modern times, intermittent fasting has
been studied for over 100  years. Drs. Marie and
Guelpa described a cyclical fasting regimen of
4 days of fasting and purges followed by 4 days of
a restricted vegetarian diet. Three-quarters of the
20 patients (adults and adolescents) with epilepsy
that were studied could not adhere to the diet for
more than one cycle. Of the remaining patients,
those who followed the diet had significant benefit
and in some cases had seizure remission; however,
long-term compliance with the diet was limited (in
more than one case, by friends of the patient who
provided foods that were not permitted) and they
concluded that their regimen was too difficult for
most adults to follow (Marie and Guelpa, 1911).
Dr. Geyelin of New York Presbyterian observed
a 10-year-old boy with 4  years of refractory epi-
lepsy become cured after intermittent fasting (four
fasts over 4 months) under the care of Dr. Conklin
of Battle Creek. Dr. Geyelin then treated a 9-year-
old boy with a 3-day fast; his multiple daily seizures
stopped after the 2nd day. Dr. Geyelin went on to
treat patients with intermittent fasting of length-
ening duration (Geyelin, 1921)  and expanded
these treatments to adults as well as children. In
22/26 patients (ages 3–35 years), he observed sei-
zure remission by the 10th day of fasting; 18/26
had marked improvement 1 year following fasting,
and had no further seizures.
R.M. Wilder of the Mayo Clinic, analyzing
Geyelin’s work, was the first to speculate that
the benefit … may be dependent on the keto-
nemia which much result from such fasts,
and that possibly equal good results could be
obtained if a ketonemia were produced by
some other means. The ketone bodies, ace-
toacetic acid and its derivatives, are formed
from fat and protein whenever a dispropor-
tion exists between the amount of fatty acid
and the amount of sugar actually burning in
the tissues…. it is possible to provoke ketosis
by feeding diets which are very rich in fat and
low in carbohydrate. It is proposed, therefore,
to try the effect of such ketogenic diets on a
series of epileptics. (Wilder, 1921)
The Mayo Clinic began treating adults with
epilepsy with this “ketogenic diet” in 1924. C.J.
Barborka wrote that “epileptic patients have an
unusual ability to consume and utilize fat,” and
hypothesized that the benefits of ketosis may be
due to changes in nerve cells, and “decreased irri-
tability of nerves.” General wisdom held that the
acid-base balance contributed to seizures or sei-
zure protection.
Barborka believed that dietary therapy offered
a “ray of hope,” and recognized that while the
diet was difficult, it was far better to try it than to
“merely employ a sedative, and to wait.”
Barborka published several articles on the
Mayo Clinic experience with the ketogenic diet.
He emphasized the need for patient education, and
required patients to spend 2–3 weeks under strict
supervision while learning the diet (Barborka,
1928). The diet was designed to mimic the metab-
olism of a fasting person to produce mild ketosis,
using a method originally developed for diabetics.
The target maintenance diet was calculated to have
sufficient calories to maintain a neutral weight in
adults; carbohydrates were limited to develop and
maintain ketosis. The original diet consisted of six
17
Chapter 3: Dietary Therapy in Adults
phases with varying amounts of carbohydrates
and fat, with a stepwise decrease in the content of
carbohydrates and increase in the amount of fat.
Sample menus reveal an emphasis on heavy cream
(100 cc of 40% cream with each meal), mayon-
naise, and butter (Barborka, 1929). Patients were
educated to test their urine for ketosis.
In 1930, Barborka published a series of 100
adolescent and adult patients (ages 16–51) remain-
ing on diet from 3 months to 5 years. Twelve of the
100 patients achieved complete seizure remission
on the ketogenic diet, and of those, two relaxed
to a less strict diet without food weighing, and
maintained seizure control. Seven patients had
at least a 90% reduction in seizures, and 37 addi-
tional patients experienced significant benefit,
giving a 56% response rate (Barborka, 1930). Of
the 44 patients who had no improvement, 23 had
not achieved ketosis (though some patients with
substantial improvement lacked consistent ketosis
as well).
In addition to seizure control, Barborka
reported an improvement in patients’ cognition,
the “appearance of intelligence, more normal atti-
tude,” and decreased irritability.
Twelve of the 56 women had complete ces-
sation of their menses; the seven women who
restarted a standard diet had resumption of nor-
mal menstrual cycles within a few months. One
woman with a history of menorrhagia had nor-
malization of her menstrual cycle.
Across the Atlantic, Dr. C. Bastible studied 29
institutionalized women with epilepsy in Dublin.
Their diet included low carbohydrate biscuits
made with local Carrigeen moss, “an inexpensive
seaweed found off the shores of Ireland … which
gave excellent results.” Two of the 29 women
became seizure-free. Six of the remaining patients
had a 50%–90% decrease in seizures, and six had
an increase in seizures. Bastible concluded that
“there is a definite hope of improvement or cure”
for adults with epilepsy (Bastible, 1931).
With the introduction of phenytoin in 1938,
which was more straightforward to initiate and to
maintain, the ketogenic diet was used and studied
less for the next 7 decades (Jóźwiak et al., 2011).
DEMAND
Children Transitioning
to Adult Epilepsy Providers
The ketogenic diet for children is widely used and
growing in popularity; in 2013, there were 148 diet
centers for children in North America, of which
half were started since 2000 (Jung et al., 2015). The
17
ketogenic diet is used in over 40 countries world-
wide (Kossoff and McGrogan, 2005). Altogether,
it is estimated that there are thousands of chil-
dren currently on the ketogenic diet. While many
children do not continue dietary treatments into
adulthood, there is a large and growing population
of children who will transition to dietary therapy
as adults.
Not all children who are treated with the keto-
genic diet require transition to adult epilepsy care.
Many children become seizure-free on the keto-
genic diet and successfully wean off of the diet
within 2 years; however, there is a risk of seizure
recurrence with change to a less restrictive diet.
At the Johns Hopkins Ketogenic Diet Center,
Martinez et  al. reviewed 557 children who
started the ketogenic diet between 1993 and 2007
(Martinez et al., 2007). Sixty-six children who were
seizure-free discontinued the diet (after median
2.1  years, range 0.5–8  years). Of those, 20% (13
children) had seizure recurrence up to 5.5  years
after discontinuing the diet (median:  2.4  years;
minimum 0 years). Seven of those patients decided
to restart the ketogenic diet. Risk factors for recur-
rence included an abnormal MRI and EEG with
epileptiform abnormalities. Parents and patients
with a higher risk of recurrence due to MRI and
EEG findings may elect to continue dietary ther-
apy into adulthood.
Children and adolescents with chronic dis-
eases require thoughtful transition from pedi-
atric to adult specialists, generally at age 18;
however, discussions and planning for this tran-
sition must take place much earlier. Kossoff et al.
identified 10 patients who started the ketogenic
diet or the modified Atkins diet as children in
the pediatric epilepsy center, and who remained
on dietary therapy until at least age 18 (the mean
age at initiation was 10.3, range 6–16). These
patients remained on the diet from 4 to 32 years
(mean:  15.5  years). All had good to complete
seizure control (2 with 100% seizure control,
8 with 50%–99% reduction) while on dietary
therapy. Four patients had previously attempted
to reduce the ketogenic diet ratio or increase
carbohydrates, with immediate seizure worsen-
ing. Eight patients transitioned to adult epilepsy
clinics; the oldest patients did so at ages 26 and
43 years (after several years of self-management).
Four patients switched from ketogenic diet to
modified Atkins diet (MAD; 20 grams per day
net carbohydrate limit) with no worsening of
seizures. All remained on anticonvulsants. Six
patients remained on dietary therapy, five at
the Johns Hopkins Adult Epilepsy Diet Center
18
18 section I: Ketogenic Diet for Epilepsy in the Clinic
(AEDC), and maintain good seizure control. At
the AEDC, most patients transition to adult pro-
viders by 21 years (Kossoff et al., 2013c).
Children and adolescents with specific genetic
or mitochondrial conditions represent a popula-
tion that requires adult dietary therapy, as they age
past 18. The ketogenic diet is frequently helpful
for mitochondrial disorders. While mitochondrial
disorders with onset in infancy or early childhood
may be fatal within a few years, those with onset
in later childhood may benefit from the ketogenic
diet and require transitioning to an adult epilepsy
provider familiar with the ketogenic diet (Kossoff
et al., 2014).
Glucose transporter type 1 (GLUT1) deficiency
is a rare genetic condition, caused by impaired
glucose transport into the brain and associated
with an abnormality in the gene SLC2A1. The
optimum treatment for GLUT1 deficiency is the
ketogenic diet, which may be prescribed life-long.
It is not known whether GLUT1 deficiency can
successfully transition to less restrictive forms of
dietary therapy such as the MAD (Kossoff et al.,
2014). While more commonly diagnosed in child-
hood, GLUT1 is also diagnosed in adults. Ninety-
one cases of adults with GLUT1 deficiency have
been described in the literature, and the ketogenic
diet remains a cornerstone of treatment (Leen et
al., 2014).
Juvenile myoclonic epilepsy (JME) is highly
treatment responsive, with 90% of patients achiev-
ing seizure freedom with appropriate antiepileptic
drugs (AEDs). However, the remaining 10% have
medically resistant seizures. Dietary therapy has
been shown to be effective for JME in a small case
series. Eight adolescents and adults (ages 15–44,
mean 24.3) were started on the MAD for treatment
of JME. After 1 month, 7 remained on the MAD;
6 patients (75%) had >50% seizure reduction; after
3  months, 5 had >50% reduction. Two patients
became seizure-free (25%). The mean duration
on diet at the time of publication was 13.2 months
(range, 0.5–40  months). Three patients had
increased seizures during brief periods of non-
compliance, but returned to seizure control when
they reinitiated the diet (Kossoff et al., 2013b). As
JME is a diagnosis requiring lifelong treatment,
patients with medically refractory JME are a large
population of adolescents and adults who could
benefit from dietary therapy.
Refractory Epilepsy
Worldwide, there are 65 million people with epi-
lepsy; 30% are medically refractory (Moshe et al.,
2015), leaving approximately 19.5  million people
in the world with seizures uncontrolled by medica-
tions. Many of these patients are not surgical can-
didates, due to generalized epilepsy (of whom up
to 26% may be refractory), multifocal nature, or
nonresectable locations of ictal onset.
Patients with seizures resistant to two or more
AEDs have a low chance of seizure freedom with
additional drugs added. In a longitudinal study of
1,098 newly diagnosed epilepsy patients followed
2–26  years, 49% of patients were seizure-free on
the first AED prescribed; an additional 13.2%
became seizure-free with the second drug tried,
3.7% with the third AED, and 1% with the fourth;
with successive AEDs added or attempted, the
percent of patients achieving seizure freedom with
each additional AED was less than 1% (Brodie
et al., 2012).
In addition to medically refractory epilepsy,
the desire to avoid additional AED side effects
is a consideration leading patients to pursue the
ketogenic diet. Adverse effects from medica-
tions (along with psychiatric comorbidities such
as depressive symptoms) are the largest predic-
tors of health-related quality of life in patients
with epilepsy, much more strongly predictive of
patients’ perceived quality of life than seizure
frequency. In fact, in a study of 809 adult Italian
patients with pharmacoresistant epilepsy, seizure
frequency and the presence of generalized tonic-
clonic seizures did not significantly affect qual-
ity of life, whereas quality of life declined with
increased medication side effects (Luoni et  al.,
2011). In patients without comorbid depression,
adverse medication effects are the main drivers of
health-related quality of life (Luoni et al., 2011).
The ketogenic diet has the benefit of freedom
from many of the adverse effects that can accom-
pany additional medications, particularly cogni-
tive side effects.
Super-Refractory Status Epilepticus
Patients with status epilepticus (prolonged sei-
zure lasting more than 5 minutes, or recurrent
seizures without return to baseline) are generally
treated with benzodiazepines or other medica-
tions; if seizure activity continues despite treat-
ment with intravenous antiepileptic drugs, the
condition is termed refractory status epilepticus,
and the patient may be placed in a medically
induced coma. If status epilepticus continues after
at least 24 hours of general anesthetic medications,
it is deemed super-refractory status epilepticus
(SRSE), which is associated with high morbidity
and mortality, with up to 61% mortality reported
(Brophy et al., 2012).
19
Chapter 3: Dietary Therapy in Adults
The ketogenic diet has been used in children
for SRSE since 1999 (Baumeister et al., 2004), and
is now used for refractory status of different etiolo-
gies in children (O’Connor et al., 2014).
In 2008, the first report of ketogenic diet
for SRSE in an adult was published in France
(Bodenant et  al., 2008). At the University of
Pennsylvania, two adults in super-refractory status
were then successfully treated with the ketogenic
diet, after 20  days and 101  days of seizures, with
successful medication weaning at 6 and 11  days
following diet initiation, respectively (Wusthoff
et  al., 2010). Thakur et  al. published the largest
series of adults, a series at four medical centers of
10 adult patients (median age, 33  years) treated
with the ketogenic diet for SRSE, of whom 70%
had encephalitis. The diet started after a median
of 21.5 days (range, 2–60) and a median of seven
antiepileptic drugs were tried (range, 5–13). The
SE ceased in all nine patients who achieved ketosis,
in a median of 3 days (Thakur et al., 2014).
Dietary therapy is now used as an adjunct
strategy for SRSE in both children and adults,
and in proposed treatment strategies the recom-
mendation has been made that the ketogenic diet
“should probably be tried in all severe cases of
super-refractory status epilepticus” (Shorvon &
Ferlisi., 2011).
R E S U LT S
Feasibility, Tolerability, and Adherence
While the ketogenic diet has been widely used in
children in modern times, concerns over adults’
ability to tolerate the diet and maintain ketosis has
slowed adoption for use in adults (Swink et  al.,
1997; Barborka, 1930). Modern studies report a
wide range of adherence to the ketogenic diet in
adults, 22%–75% at 3 months (Mosek et al., 2009;
Klein et al., 2010), with significant variation. The
MAD, first published in 2003 as a less restrictive
alternative to the classic ketogenic diet (Kossoff
et al., 2003), has had published adherence rates of
56%–100% at 3 months and 22%–77.8% at 1 year
(Kossoff et al., 2003; Cervenka et al., 2012; Smith
et  al., 2011). These retention rates are somewhat
lower than those seen in add-on drug trials for
new AEDs (75%–80% retention after 12–18 weeks;
Elger et al., 2007; Ben-Menachem et al., 2007).
The initial decision to begin dietary treatment
is not undertaken lightly. In some studies, up to
two-thirds of eligible patients screened decline
to participate, due to concerns about restrictive-
ness or complexity of the diet (Mosek et al., 2009,
18/27 declined; Klein et al., 2010, 23/35 declined).
19
However, many people choose to continue
dietary treatment beyond the initial study periods
requested (Carrette et al., 2008; Klein et al., 2010;
Cervenka et  al., 2012), and some patients have
remained on dietary therapy as long as 32  years
(Kossoff et al., 2013c).
A meta-analysis in 2015 comparing six classic
ketogenic diet studies and five MAD studies con-
cluded that adherence rates are higher in the MAD
(combined compliance rate 56%) than the classic
ketogenic diet (38% adherence) (Ye et al., 2015).
Not surprisingly, when dietary treatment
is effective, patients are motivated to continue
treatment. When patients decide to stop dietary
treatment, the most common reason cited is lack
of efficacy, followed by restrictiveness of the diet
(Kossoff et  al., 2008; Lambrechts et  al., 2012;
Schoeler et al., 2014). Financial reasons have been
cited in a few patients due to higher cost of meats
compared to processed carbohydrates (Smith
et al., 2011).
Efficacy
Adults with pharmacoresistant epilepsy have
response rates (defined as a ≥50% decrease in
seizures) of 33%–54% to the newer antiepileptic
drugs (Mbizvo et al., 2012; Elger et al., 2007; Ben-
Menachem et al., 2007). Rates of seizure freedom
with additional agents are much lower, with each
additional add-on agent after the second pro-
viding a less than 5% chance of seizure freedom
(Brodie et  al., 2012). Dietary therapy compares
favorably with these rates in most published stud-
ies (Payne et al., 2011), especially as the patients
starting dietary treatment are typically the most
refractory patients, with mean prior AEDs tried
ranging from 5.4 to 10.6 (Sirven et  al., 1999;
Kossoff et al., 2003).
The classic ketogenic diet reduces seizures by
≥50% in 22%–55% of patients, using intent-to-
treat analysis (Sirven et al.,1999; Mosek et al., 2009;
Klein et al., 2014; Figure 3.1 and Table 3.1). Many
patients have even higher response rates, with 8%–
27% of patients seeing >90% decrease in seizures
(Sirven et al.,1999; Schoeler et al., 2014) and seizure
freedom in up to 8% (Klein et al., 2010). In a com-
parison of seizure-free months, Klein et al. found
an improvement from 20% of months seizure-free
at baseline to 56.2% of months seizure-free on the
ketogenic diet (Klein et al., 2010).
The MAD has wider variability in pub-
lished response rates, ranging from 12% to 67%
with ≥50% seizure reduction (Smith et  al., 2011;
Kossoff et al., 2008; Kossoff et al., 2013b; Kossoff
et  al., 2013c) and up to 33% of patients with
20

20 section I: Ketogenic Diet for Epilepsy in the Clinic

Response rates to dietary therapy for epilepsy in adults
(number of patients)
100%

90%

80%

70%
Percent of adults

60%
11 9 12 15 29 23 3 8 30 18 3 22 6 3 Percent decrease
50%
in seizures

40% <50%

≥50%
30% ≥90%

20%

10%

0%
2

11

11
99

09

10

14

03

08

08

13

13
01

01

01
20

20
19

20

20

20

20

20

20

20

20
s2

r2

a2
ith

la
en

ek

ei

ff

tte

ff

ff

n
ele
ht

nk
lei

se
so

so

so
N

po
os
rv

re

Sm
ec

er
ve
ho
K

os

os

os
op
M
Si

ar
br

et
K

K
er
Sc

-P
m

C
La

m
am
R
Ketogenic diet Modified Atkins diet
FIGURE 3.1 Studies of the classic ketogenic diet and the modified Atkins diet report varying response rates in adults,
ranging from 12% to 67%.

>90% reduction (Cervenka et  al., 2012; Kossoff larger proportion of ≥50% response rates (Kossoff
et al., 2013b; Kossoff et al., 2013c; Figure 3.1 and et al., 2008), though this has not been detected in
Table 3.1). other studies (Mady et al., 2003, Mosek et al., 2009).
Other dietary therapies such as the medium With regard to seizure type and response
chain triglyceride (MCT) diet and the low glyce- to diet therapies, Nei et  al. detected a trend
mic index treatment (Pfeifer and Thiele, 2005) have toward greater seizure reduction in patients with
not been widely studied in adults. In a series of 11 symptomatic generalized epilepsy, with 64% of
patients on the MCT diet (and four on the classic symptomatic generalized patients having ≥50%
ketogenic diet or a combination of MCT/ketogenic reduction versus 28% of focal epilepsies (Nei et al.,
diet during the study) Lambrechts found that 5/12 2014). In Mady et  al.’s study of 45 adolescents,
patients continued the diet at 1 year, and of those, those with multiple seizure types had a greater
two had a 50%–90% reduction in seizures, while improvement than those with complex partial
the remaining three patients had a <50% reduction or generalized seizure types alone (Mady et  al.,
(Lambrechts et  al., 2012). The mean AEDs used 2003). As discussed previously, high response
decreased slightly, from 2.7 at baseline to 2.2 at the rates and seizure freedom were seen in a small
end of the diet. series of adolescents and adults with JME, with
Disproving the initial speculations that adults 4/6 adults showing >50% decrease in seizures,
could not maintain ketosis, the majority of adults 2/6  >90% decrease, and 1/6 seizure-free (Kossoff
on ketogenic diets have been successful at achiev- et al., 2013b).
ing and maintaining urinary and/or serum ketosis In patients with GLUT1 deficiency, up to 90% of
(range of published rates, 58.3%–87.5%; Sirven patients were seizure-free on the ketogenic or MAD,
et al., 1999; Klein et al., 2010; Mosek et al., 2009); including three adults. One adult had resolution of
levels of ketosis have not been predictive of sei- generalized convulsive seizures, but had persistence
zure improvement (Mosek et al., 2009; Klein et al., of likely nonepileptic events; the other two adults
2010; Nei et al., 2014). were seizure-free (Ramm-Petersen et al., 2013).
Kossoff et al. found a trend toward patients Beyond a reduction in the number of seizures,
with more frequent seizures at baseline having a the severity or duration of seizures reportedly
 21
TABLE 3.1 SUMMARY OF PUBLISHED STUDIES OF THE KETOGENIC
DIET AND MODIFIED ATKINS DIET IN ADULTS

Study first Year Journal Diet Number of >=50% >=90% Retention

author patients improvement improvement

Sirven 1999 Epilepsia Ketogenic 11 55% (6/11) 27% (3/11) 7/11 (63.6%) at 8 months
Coppola 2002 Epilepsy Res Ketogenic 5 (18–23) No age-specific information No age-specific information
Mady 2003 Epilepsia Ketogenic 45 adolescents 13/28 (46%); no age-specific 8/28 (28%); no age-specific 28/45 (62%) at 6 months
(12–19) information information
Groesbeck 2006 Dev Med Child Ketogenic 28 No age-specific 24/28 No age-specific information
Neurol information
Mosek 2009 Seizure Ketogenic 9 2/9 (22%) 2/9 (22%) at 3 months
Klein 2010 Epilepsy Behav Ketogenic 12 5/12 (42%) 2/12 (17%) 9/12 (75%) at 4 months
Lambrechts 2012 Epilepsy Behav KD or MCT 15 2/15 (13%) 5/15 (33%) at 12 months
Nei 2014 Seizure Ketogenic 29 (11–51) 13 (45%) 1/29 (3%) Mean 9 months
Schoeler 2014 Epilepsy Behav Ketogenic 23 9/23 (39%) 2/23 (8%) 9/23 (29%) at 12 months
Kossoff 2003 Neurology MAD 3 (18–52) 1/3 (33%) 1/3 (33%) 3/3 (100%) at 3 months
Carrette 2008 Clin Neurol MAD 8 1 (12%) 3/8 (37.5%) at 6 months
Neurosurg
Kossoff 2008 Epilepsia MAD 30 14/30 (47%) 1/30 (3%) 20/30 (67%) at 3 months
Smith 2011 Epilepsia MAD 18 4/18 (22%) 14/18 (78%) at 6 months
Coppola 2011 Seizure LGID 3 adults 3/3 (100%) 3/3 (100%) at 2 months
Cervenka 2012 Epilepsia MAD 22 6 (27%) 4/22 (18%) 14/22 (64%) at 3 months
Kossoff 2013 Epilepsy Behav MAD 6 adults 4/6 (67%) 2/6 (33%) 5/6 at (83%) 2 months
Ramm-Pettersen 2013 Dev Med Child MAD 3 adults 2/3 (67%) 2/3 (67%)* 1 year (at least)
Neurol

* suspicion of PNES in remaining patient not seizure-free

22
22 section I: Ketogenic Diet for Epilepsy in the Clinic
decreased, or the amount of time to recover from a
seizure shortened in a subset of patients (Schoeler
et al., 2014, Smith et al., 2011).
Some studies have shown that weight loss
predicts diet efficacy, with 67% of patients with
greater than 0.9  kg/m2 decrease in BMI having
>50% seizure reduction, compared with 27% of
patients with <0.9  kg/m2 decrease in BMI, with
p = .03 (Kossoff et al., 2008). However, this is not a
consistent finding in all studies (Smith et al., 2011),
and patients with weight gain can also respond to
the diet (Kossoff et al., 2008).
Beneficial Effects
Cognition and Mood
Dietary treatment often has positive cognitive and
mood effects in studies of adults with epilepsy
(Table 3.2). Many patients report an improve-
ment in cognition and mood, as well as (or even
despite the lack of) improved seizure control; in
fact, some patients with no or <50% improvement
in seizure frequency opt to continue dietary treat-
ment for the cognitive benefits alone (Sirven et al.,
1999; Coppola et  al., 2002). The majority (7/11)
TABLE 3.2 REPORTED BENEFICIAL
AND ADVERSE EFFECTS OF DIETARY
TREATMENT (OTHER THAN SEIZURE
CONTROL)
Additional reported beneficial effects
Improved cognition
Improved mood
Weight loss (when desired)
Increased alertness
Increased energy
Reduced depression
Improved quality of life
Decreased anxiety
Decreased tension
Decreased length or severity of seizures
Reported adverse effects
Nausea or vomiting
Bloating
Constipation
Temporary lipid elevations
Impaired concentration
Menstrual irregularities
Possible growth restriction
Kidney stones
Skeletal fractures
of patients in one study of the ketogenic diet saw
an improvement in mood and cognition, though
2/11 also reported impaired concentration (Sirven
et  al., 1999). Increased alertness and energy are
common findings, seen in 33%–65% of adults and
adolescents on the ketogenic diet (Mady et  al.,
2003; Mosek et al., 2009; Lambrechts et al., 2012;
Schoeler et al., 2014). One study of the MAD that
administered detailed cognitive and depression
questionnaires found reduction in depression
scores and improved concentration in 6/7 patients
on diet for 1 month, and in all three patients com-
pleting 6  months of the study (Carrette et  al.,
2008). Quality of life scores tend to rise rather
than decrease on both the MAD and the keto-
genic diet, though not significantly (Carrette et al.,
2008, Klein et al., 2010; Lambrechts et al., 2012).
Anxiety, tension, and fatigue may all be improved
as well (Lambrechts et al., 2012).
Weight Loss
Weight loss is often a desirable effect of ketogenic
diet therapies, and many patients are able to lose
significant amounts of weight, and may success-
fully move from a clinically “obese” body-mass
index (BMI) to a “normal” or “overweight” BMI.
Weight loss is of particular importance in the
adult population as it is estimated that over one-
third of adults in the United States are obese (BMI
≥ 30 kg/m2; Ogden et al., 2014). Obesity can lead
to type II diabetes, obstructive sleep apnea, and
metabolic syndrome, all of which can be combated
with weight loss. In a series of studies investigating
weight loss with the MAD, mean weight loss was
7 kg over 3 months (Kossoff et al., 2008) and 10 kg
over 6 months (Carrette et al., 2008), including 4 of
the 11 patients who were obese when they started
the diet who were no longer obese at the conclu-
sion (Kossoff et  al., 2008). In a separate study of
the ketogenic diet, mean BMI improved 18%,
from 33.8 (obese) to 27.5 (overweight) in 12 adults
over 4 months, and the majority of overweight or
obese subjects had at least a 10% reduction in BMI
(Klein et al., 2010).
When patients are of average weight or under-
weight when starting diet therapy, total calories
can be adjusted to prevent or reverse weight loss.
Adverse Effects
Gastrointestinal
Gastrointestinal side effects are common, with half
to all patients reporting some degree of nausea,
constipation, bloating, or vomiting at some point
on diet therapy; these generally resolve after the
23
Chapter 3: Dietary Therapy in Adults
first few days or weeks of treatment with the keto-
genic diet (Sirven et al., 1999; Coppola et al., 2002;
Klein et  al., 2010). Rarely, patients are unable to
continue dietary treatment due to intractable nau-
sea or vomiting.
Lipids
Lipids may increase on ketogenic diets and should
be monitored. Sirven et  al. found a significant
increase in total fasting cholesterol at 3 and at
6  months on diet, with an increase of the mean
cholesterol from 208 mg/dL (range, 120–304) to
291 mg/dL (220–395). Triglycerides also increased
at 3  months from mean 190 mg/dL (41–542) to
203 mg/dL (68–417), then plateaued (Sirven et al.,
1999). The extension of this study continued to
show a significant increase in total cholesterol and
in the cholesterol/HDL ratio at the time of diet dis-
continuation after up to 35  months on diet (Nei
et al., 2014). If extreme, lipid changes may prompt
discontinuation of dietary therapy (Mosek et  al.,
2009). However, elevated lipids are not present in
all patients or in all studies, and triglycerides and
LDL may not change (Klein et al., 2010).
Lipids increase on the MAD as well (Carrette
et al., 2008), though end lipid levels in some stud-
ies remained within average cardiovascular risk
ranges (Kossoff et  al., 2008, Smith et  al., 2011).
One study of the MAD found a decrease in tri-
glycerides with dietary treatment over 12 months
(Smith et al., 2011).
Lipids may increase during the initial phase
of the diet, then return to baseline: one study of
37 adults on the MAD for at least 3 months found
that while total cholesterol and LDL had increased
at 3 months, there was no difference from base-
line after 1 year (p = 0.2 and p = 0.5, respectively)
(Cervenka et al., 2014).
If cholesterol elevation is present, this may
be manageable without stopping dietary therapy;
one patient whose LDL doubled after 3  months
continued the MAD, and with carnitine supple-
mentation and the substitution of saturated fats
for polyunsaturated fats saw his cholesterol and
LDL return to normal (Cervenka et  al., 2012).
Carnitine supplementation successfully decreased
elevated triglycerides in three patients as well (Nei
et al., 2014).
Effects on the Menstrual Cycle
Menstrual irregularities and cessation of men-
struation are common in the starvation state.
Given that the ketogenic diet is designed to
mimic starvation, it is not surprising that it
23
can also cause menstrual irregularity. Barborka
reported that 12/56 women had cessation of
their menses during ketogenic diet treatment;
however, in the seven that stopped the diet, nor-
mal menstruation resumed (Barborka, 1930).
In Sirven’s 1999 study, all nine women devel-
oped menstrual irregularities (irregular cycles
or cessation of menses), which resolved on diet
discontinuation (Sirven et  al., 1999). Menstrual
irregularities were also frequent in Mady et  al.’s
2003 study of the ketogenic diet (45% of women).
Menstrual irregularities seem to be much less
common with the MAD: there were no menstrual
irregularities in any of the 19 women in one study
(Kossoff et al., 2003)  and none reported in nine
women in a second study (Smith et  al., 2011),
and they were present in only 1 out of 17 women
in a third (Cervenka et  al., 2012). Lambrechts
found none in two women on the ketogenic diet
and two women on the MCT diet (Lambrechts
et al., 2012).
Other Side Effects
Long-term effects in patients on the ketogenic diet
for 6  years or more (in patients ages 7–23  years)
included decrease in growth rate: at diet initiation,
14/28 (50%) were at or below the 10th percentile
for weight, which increased to 23/28 (82%) at last
follow-up (Groesbeck et al., 2006). Growth restric-
tion was not related to degree of ketosis and is less
of a concern in patients who begin ketogenic diets
as adults. In one study, one-quarter of patients on
ketogenic diets developed kidney stones, with a
median of 2 years after diet onset (Groesbeck et al.,
2006). Subsequent studies have shown that urine
alkalinization with potassium citrate reduces the
risk of kidney stones (Sampath et  al., 2007). Six
patients in the long-term study (21%) had skeletal
fractures, occurring a median of 18 months after
diet initiation (Groesbeck et al., 2006).
Kidney stones have not been reported in other
studies of adults on dietary treatment. One patient
had a jaw fracture related to a seizure and stopped
the diet (Mosek et  al., 2009), but other skeletal
fractures have not been reported.
CONCLUSIONS
Dietary treatment is feasible in adults and often
highly effective, with seizure reduction rates in
medically refractory populations of 33%–67%,
comparable with response rates in children.
A  significant proportion of patients may become
seizure-free. In addition to seizure reduction,
patients may also benefit from improved mood
24
24 section I: Ketogenic Diet for Epilepsy in the Clinic
and cognition, as well as intentional weight loss.
Lipids should be monitored, but in most cases
persistent lipid elevations can be managed with
diet adjustments. Diet adherence remains a major
challenge for adults.
REFERENCES
Barborka, C.J. (1928). Ketogenic diet treatment of epi-
lepsy in adults. JAMA 91(2), 73–78.
Barborka, C.J. (1929). The ketogenic diet and its use.
Med Clin North Am 12(6), 1639–1653.
Barborka, C.J. (1930). Epilepsy in adults:  results of
treatment by ketogenic diet in one hundred cases.
Arch Neurol Psych 23, 904–914.
Bastible, C. (1931). The ketogenic treatment of epi-
lepsy. Irish J Med Sci 6, 506–520.
Baumeister, F.A., Oberhoffer, R., Liebhaber, G.M.,
Kunkel, J., Eberhardt, J., Holthausen, H., and
Peters, J. (2004). Fatal propofol infusion syndrome
in association with ketogenic diet. Neuropediatrics
35, 250–252.
Ben-Menachem, E., Biton, V., Jatuzis, D., Abou-Khalil,
B., Doty, P., and Rudd, G.D. (2007). Efficacy and
safety of oral lacosamide as adjunctive therapy in
adults with partial-onset seizures. Epilepsia 48,
1308–1317.
Bodenant, M., Moreau, C., Sejourne, C., Auvin, S.,
Delval, A., Cuisset, J.M., Derambure, P., Destee, A.,
and Defebvre, L. (2008). Interest of the ketogenic
diet in refractory status epilepticus in an adult. Rev
Neurol (Paris) 164, 194–199.
Brodie, M.J., Barry, S.J., Bamagous, G.A., Norrie,
J.D., and Kwan, P. (2012). Patterns of treatment
response in newly diagnosed epilepsy. Neurology
78, 1548–1554.
Brophy, G.M., Bell, R., Claassen, J., Alldredge, B.,
Bleck, T.P., Glauser, T., LaRoche, S.M., Riviello,
J.J., Jr., Shutter, L., Sperling, M.R., et  al. (2012).
Guidelines for the evaluation and management of
status epilepticus. Neurocrit Care 17(1), 3–23.
Carrette, E., Vonck, K., de Herdt, V., Dewaele, I., Raedt,
R., Goossens, L., Van Zandijcke, M., Wadman, W.,
Thadani, V., and Boon, P. (2008). A pilot trial with
modified Atkins’ diet in adult patients with refrac-
tory epilepsy. Clin Neurol Neurosurg 110, 797–803.
Cervenka, M., Terao, N., Bosarge, J., Henry, B., Klees,
A., Morrison, P., and Kossoff, E. (2012). Email
management of the MAD for adults with epilepsy
is feasible and effective. Epilepsia 53, 728–732.
Cervenka, M.C., Patton, K., Eloyan, A., Henry, B., and
Kossoff, E.H. (2014, Nov 10). The impact of the
modified Atkins diet on lipid profiles in adults with
epilepsy. Nutr Neurosci [Epub ahead of print].
Coppola, G., D’Aniello, A., Messana, T., Di Pasquale,
F., della Corge, R., Pascotto, A., and Verrotti, A.
(2011). Low glycemic index diet in children and
young adults with refractory epilepsy: first Italian
experience. Seizure 20(7), 526–528.
Coppola, G., Veggiotti, P., Cusmai, R., Bertoli, S.,
Cardinali, S., Dionisi-Vici, C., Elia, M., Lipsi,
M.L., Sarnelli, C., Tagliabue., A., et al. (2002). The
ketogenic diet in children, adolescents, and young
adults with refractory epilepsy:  an Italian multi-
centric experience. Epilepsy Res 48, 221–227.
Elger, C., Bialer, M., Cramer, J.A., Maia, J., Almeida, L.,
and Soares-da-Silva P. (2007). Eslicarbazepine ace-
tate:  a double-blind, add-on, placebo-controlled
exploratory trial in adult patients with partial-
onset seizures. Epilepsia 48, 497–504.
Geyelin, H.R. (1921). Fasting as a method of treating
epilepsy. M Rec 99, 1037–1039.
Groesbeck, D.K., Bluml, R.M., and Kossoff, EH.
(2006). Long-term use of the ketogenic diet in the
treatment of epilepsy. Dev Med Child Neurol 48,
978–981.
Guelpa, G., and Marie, A. (1911). La lutte contre
l’epilepsie par la desintoication et par la reducation
alimentaire. Rev de Therap Med-Chir 78, 8–13.
Hippocrates. (400 bc). On the Sacred Disease. From
The Internet Classics Archive, available at: http://
classics.mit.edu/ Hippocrates/ sacred.html.
Accessed July 27, 2015.
Jóźwiak, S., Kossoff, E.H., and Kotulska-Jóźwiak, K.
(2011). Dietary treatment of epilepsy:  rebirth of
an ancient treatment. Neurol Neurochir Pol 45(4),
370–378.
Jung, D.E., Joshi, S.M., and Berg, A.T. (2015). How
do you keto? Survey of North American pediat-
ric ketogenic diet centers. J Child Neurol 30(7),
868–873.
Klein, P., Janousek, J., Barber, A., and Weissberger, R.
(2010). Ketogenic diet treatment in adults with
refractory epilepsy. Epilepsy Behav 19, 575–579.
Klein, P., Tyrlikova, I., and Mathews, G.C. (2014).
Dietary treatment in adults with refractory epi-
lepsy: a review. Neurology 18, 1978–1985.
Kossoff, E.H., Krauss, G.L., McGrogan, J.R., and
Freeman, JM. (2003). Efficacy of the Atkins diet as
therapy for intractable epilepsy. Neurology 61(12),
1789–1791.
Kossoff, E.H., and McGrogan, J.R. (2005). Worldwide
use of the ketogenic diet. Epilepsia 46(2), 280–289.
Kossoff, E.H., Rowley, H., Sinha, S.R., and Vining EP.
(2008). A prospective study of the modified Atkins
diet for intractable epilepsy in adults. Epilepsia
49(2), 316–319.
Kossoff, E.H., Cervenka, M.C., Henry, B.J., Haney,
C.A., and Turner, Z. (2013a). A decade of the
MAD (2003–2013):  Results, insights, and future
directions. Epilepsy Behav 29(3), 437–442.
Kossoff, E.H., Henry, B.J., and Cervenka, M.C. (2013b).
Efficacy of dietary therapy for juvenile myoclonic
epilepsy. Epilepsy Behav 26, 162–164.
Kossoff, E.H., Henry, B.J., and Cervenka, M.C. (2013c).
Transitioning pediatric patients receiving keto-
genic diets for epilepsy into adulthood. Seizure
22(6), 487–489.
25
Chapter 3: Dietary Therapy in Adults
Kossoff, E.H., Veggiotti, P., Genton, P., and Desguerre,
I. (2014). Transition for patients with epilepsy
due to metabolic and mitochondrial disorders.
Epilepsia 55(suppl 3), 37–40.
Lambrechts, D.A., Wielders, L.H., Aldenkamp, A.P.,
Kessels, F.G., de Kinderen, R.J., and Majoie, M.J.
(2012). The ketogenic diet as a treatment option
in adults with chronic refractory epilepsy: efficacy
and tolerability in clinical practice. Epilepsy Behav
23, 310–314.
Leen, W.G., Taher, M., Verbeek, M.M., Kamsteed,
E.J., van de Warrenburg, B.P., and Willemsen,
MA. (2014). GLUT1 deficiency syndrome into
adulthood:  a follow-up study. J Neurol 261(3),
589–599.
Luoni, C., Bisulli, F., Canevini, M.P., De Sarro, G.,
Fattore, C., Galimberti, C.A., Gatti, G., La Neve,
A., Muscas, G., Specchio, L.M., et  al. (2011).
Determinants of health-related quality of life
in pharmacoresistant epilepsy:  results from a
large multicenter study of consecutively enrolled
patients using validated quantitative assessments.
Epilepsia 52, 2181–2191.
Mady, M.A., Kossoff, E.H., McGregor, A.L., Wheless,
J.W., Pyzik, P.L., and Freeman, J.M. (2003). The
ketogenic diet: adolescents can do it, too. Epilepsia
44, 847–851.
Martinez, C.C., Pyzik, P.L., and Kossoff, E.H. (2007).
Discontinuing the ketogenic diet in seizure-free
children:  recurrence and risk factors. Epilepsia
48(1), 187–190.
Mbizvo, G.K., Dixon, P., Hutton, J.L., and Marson,
A.G. (2012). Levetiracetam add-on for drug-
resistant focal epilepsy:  an updated Cochrane
review. Cochrane Database Syst Rev 9.
Mosek, A., Natour, H., Neufeld, M., Shiff, Y., and
Vaisam, N. (2009). Ketogenic diet treatment in
adults with refractory epilepsy: a prospective pilot
study. Seizure 18, 30–33.
Moshe, S.L., Perucca, E., Ryvlin, P., and Tomson,
T. (2015). Epilepsy:  new advances. Lancet 385,
884–898.
Nei, M., Ngo, L., Sirven, J., and Sperling, M. (2014).
Ketogenic diet in adolescents and adults with epi-
lepsy. Seizure 23, 439–442.
O’Connor, S.E., Ream, M.A., Richardson, C., Mikati,
M.A., Trescher, W.H., Byeler, D.L., Sather, J.D.,
Michael, E.H., Urbanik, K.B., Richards, J.L., et  al.
(2014). The ketogenic diet for the treatment of pedi-
atric status epilepticus. Pediatr Neurol 50, 101–103.
Ogden, C.L., Carroll, M.D., Kit, B.K., and Flegal, K.M.
(2014). Prevalence of childhood and adult obe-
sity in the United States, 2011–2012. JAMA 311,
806–814.
25
Payne, N., Cross, J., Sander, J., and Sisodiya, S. (2011).
The ketogenic and related diets in adolescents and
adults—A review. Epilepsia 52, 1941–1948.
Pfeifer, H.H., and Thiele, E.A. (2005). Low-glycemic-
index treatment:  a liberalized ketogenic diet for
treatment of intractable epilepsy. Neurology 65,
1810–1812.
Ramm-Pettersen, A., Nakken, K., Skogseid, I., Randby,
H., Skei, E., Bindoff, L., and Selmer, K.K. (2013).
Good outcome in patients with early dietary treat-
ment of GLUT-1 deficiency syndrome:  results
from a retrospective Norwegian study. Dev Med
Child Neurol 55, 440–447.
Sampath, A., Kossoff, E.H., Furth, S.L., Pyzik, P.L., and
Vining, E.P. (2007). Kidney stones and the keto-
genic diet:  risk factors and prevention. J Child
Neurol 22, 375–378.
Schoeler, N., Wood, S., Aldridge, V., Sander, J., Cross,
J., and Sisodiya, S. (2014). Ketogenic diet therapies
for adults with epilepsy: feasibility and classifica-
tion of response. Epilepsy Behav 37, 33–81.
Shorvon, S., and Ferlisi, M. (2011). The treatment
of super-refractory status epilepticus:  a critical
review of available therapies and a clinical treat-
ment protocol. Brain 134, 2802–2818.
Sirven, J., Whedon, B., Caplan, D., Liporace, J.,
Glosser, D., O’Dwyer, J., and Sperling, M.R.
(1999). The ketogenic diet for intractable epi-
lepsy in adults:  preliminary results. Epilepsia 40,
1721–1726.
Smith, M., Politzer, N., MacGarvie, D., McAndrews,
M., and del Campo, M. (2011). Efficacy and tol-
erability of the MAD in adults with pharmacore-
sistant epilepsy: a prospective observational study.
Epilepsia 52, 775–780.
Swink, T.D., Vining, E.P., and Freeman, J.M. (1997).
The ketogenic diet:  1997. Adv Pediatr 44,
297–329.
Thakur, K.T., Probasco, J. C., Hocker, S.E., Roehl, K.,
Henry, B., Kossoff, E.H., Kaplan, P.W., Geocadin,
R.G., Hartman, A.L., Venkatesan, A., et al (2014).
Ketogenic Diet for adults in super-refractory sta-
tus epilepticus. Neurology 82, 665–670.
Wilder, R.M. (1921). The effect of ketonemia on the
course of epilepsy. Mayo Clin Proc 2, 307–308.
Wusthoff, C.J., Kranick, S.M., Morley, J.F., and
Christina Berggvist, A.G. (2010). The ketogenic
diet in treatment of two adults with prolonged
nonconvulsive status epilepticus. Epilepsia 51,
1083–1085.
Ye, F., Li, X.J., Jian, W.L., Sun, H.B., and Liu, J. (2015).
Efficacy of and patient compliance with a keto-
genic diet in adults with intractable epilepsy:  a
meta-analysis. J Clin Neurol 11, 26–31.
26
4
How Do You Implement the Diet?
A . G . C H R I S T I N A B E R G Q V I S T, M D
INTRODUCTION
The ketogenic diet (KD) has survived and thrived
for almost 100  years as an effective treatment of
intractable epilepsy. For many years the imple-
mentation of the diet remained in large parts
unchanged from its initial conception by Wilder
(Livingston, 1951; Wilder, 1921). Management
was based on our clinical practice, strongly tied
to its history of “fasting, calorie and fluid restric-
tions.” As the use of the ketogenic diet increased
locally in the United States and spread across the
world, and side effects became better understood,
many of the standard practices were questioned
and tested and some were changed or discontin-
ued (Kossoff and McGrogan, 2005). As a result,
alternative implementation of the classic KD and
creation of newer diets have emerged. Most of
the data analyzed consist of retrospective chart
reviews but there is some support from random-
ized trials. In 2008 a statement regarding the man-
agement of the KD was written by 28 authors from
nine countries. This served as a basis from which
future improvements to dietary therapies were
made (Kossoff et  al., 2009). This document, the
International Consensus Statement for Ketogenic
Diet, is referred to as the ICSKD in this chapter.
There are multiple books detailing site-specific
implementation practices of the KD (Freeman
et  al., 2007; Kossoff et  al., 2011b; Snyder, 2006;
Stafstrom et  al., 2008). It is not within the scope
of this chapter to provide the same detail; instead
I  discuss the implementation changes that have
occurred in the 100-year history of the KD.
NEED FOR A
FA M I LY- C E N T E R E D ,
T E A M - BA S E D A P P ROAC H
T O   D I E TA RY T H E R A P I E S
The KD is a prescribed medical diet therapy
requiring supervision from a healthcare team. It
should never be confused with weight loss diets
that individuals can manage safely themselves.
Dietary therapies for epilepsy require periodic
physical examination and laboratory testing to
remain safe, promote general health, and pre-
vent side effects. This is particularly true in chil-
dren, whose nutritional needs are continuously
changing during childhood and adolescence.
Technology has changed much of the interac-
tion between the care provider and the KD team,
and allows for long-distance communication.
Communication with the team via phone, e-mail,
and electronic medical records (such as EPIC)
is necessary when managing the ketogenic diet.
However, it is not a replacement for a follow-up
visit with the KD team.
The ICSKD recommends that patients be
evaluated at least four times in the first year of
treatment and twice a year thereafter. Younger
children, particularly infants, need more frequent
(even monthly) evaluations to keep up with their
nutritional needs.
COMPOSITION OF THE
K E TO G E N I C D I E T   T E A M
The ketogenic diet is best supervised by a team
of experienced healthcare professionals. For a
small program, this comprises, at a minimum, an
epileptologist and a dietitian. For truly compre-
hensive care and for a larger program, this team
is expanded to also include a nurse and social
worker. Each team member contributes unique
knowledge to optimize the care for these chil-
dren and will grow as the number of patients in
the dietary treatment program increases. The lat-
est addition to our team at the Children’s Hospital
of Philadelphia is a chef to assist with creation of
recipes in our in-patient KD kitchen (Fenton et al.,
2014; Groveman et al., 2014). Any institution that
supports dietary therapies must also provide phar-
macy services to find medications with the lowest
carbohydrate content, a very important task, as the
content in medication is not routinely reported by
industry and is subject to frequent change.
27
Chapter 4: How Do You Implement the Diet?
Access to other subspecialties such as gastro-
enterology and nutrition, nephrology, urology,
endocrinology, a feeding team, and bone health
experts is also essential to provide truly com-
prehensive care. The children started on these
diets today rarely have just “treatment resis-
tant epilepsy” but carry many additional diag-
noses:  cerebral palsy, developmental disability,
intellectual disability, feeding difficulties some
with g-tube dependencies, behavioral difficulties,
autism spectrum disorder, and genetic conditions.
Adding the KD to their treatment regimen makes
their care truly complex. Parents are important
partners, and essential to the success of the diet
and a keto program. They can be effective coaches
of other parents, assisting them with nonmedi-
cal information, and also become trained edu-
cators as the keto-community grows to include
school, nursing agencies, and so forth (Chee et al.,
2014; MacCracken and Scalisi, 1999). Creating
a keto community makes it possible to provide
comprehensive care.
P R E D I E T E VA L UAT I O N ( S ) ,
E D U C AT I O N A N D
COUNSELING
For families considering the KD, education
about dietary therapies is provided in various
forms:  reading materials; referrals to Internet
sites; foundations that advocate for the KD such
as the Charlie Foundation for ketogenic therapies
(www.charliefoundation.org), Matthews Friends
(www.matthewsfriends.org), and the Epilepsy
Foundation (www.efa.org); DVDs, and so on.
Some dietary treatment programs offer separate
education classes to assist the families in their deci-
sion to try dietary therapies (e.g., the Children’s
Hospital of Philadelphia [CHOP], www.chop.edu/
treatment/ketogenicdiet)
The KD requires the body to switch from using
carbohydrates as the primary energy source to
using lipids. There are some disorders for which
the use of the KD or fasting can lead to signifi-
cant morbidity, even mortality. These include
inborn errors of metabolism related to carnitine
(mitochondrial transport), beta oxidation defects,
pyruvate carboxylase deficiencies, and porphyria
(which requires a high-carbohydrate diet). As a
group, mitochondrial cytopathies are in their own
category, as some may benefit from KD therapies,
such as pyruvate dehydrogenase deficiency (Di
Pisa et al., 2012; Weber et al., 2001; Wijburg et al.,
1992), while for other cytopathies the KD would
worsen the condition (Bergqvist, 2004; Horvath
et  al., 2008; Kang et  al., 2007). Assistance from
27
metabolic or neuromuscular specialists before
attempting the KD in these disorders is recom-
mended. A  metabolic screen is recommended
for any child for whom the KD is considered. The
ICSKD suggests that this include acylcarnitine
esters, urine organic acids, serum amino acids,
lactate, and pyruvate. For screening labs that are
followed every 3–6 months while on the KD treat-
ment, see the ICSKD.
I N I T I AT I O N
O F   K E TO G E N I C   D I E T
Initiating the KD can be intimidating to the
family, caretaker and KD Team alike. A well pre-
pared admission and a well educated parent is the
best assurance for success. In the following section
I will review the various types of initiation meth-
ods that have been used and discuss both their
benefits and drawbacks.
In-Patient versus Out-Patient Initiation
The classic ketogenic diet created by Wilder, and
advocated by Johns Hopkins University (Freeman
et  al., 2007)  is started in the hospital so that the
patient can be closely observed, monitored, and
treated if needed. Most centers in the United States
will use an in-patient setting to start the KD, as
this makes it possible to advance the KD relatively
quickly and achieve “ketosis” within a few days
(Kossoff et al., 2009). The child is closely observed
by nursing staff and physicians, and interventions
for hypoglycemia, acidosis, dehydration, vomiting,
weight loss, or feeding intolerances can be insti-
tuted in a timely fashion to minimize any com-
plications. Although seizures more often improve
during the admission they can worsen from the
stress of switching metabolic substrate. In this
situation the in-patient setting allows for imme-
diate adjustments in medication and ICU care,
should it be needed. While the family is in the
hospital, many hours of direct teaching—“hands-
on” education—is provided by the KD team. Many
centers admit one child at a time to start the KD.
The KD team is on 24/7 permanent call to initi-
ate the KD. Organization of the admissions to a
monthly basis, using a “small group setting” for
teaching, improves efficiency and frees up the
team’s time to assist with out-patient management.
The Children’s Hospital of Philadelphia now also
provides cooking classes in its keto-kitchen as part
of starting the KD (Groveman et al., 2014). These
classes provide hands-on experience with the KD
food before leaving the hospital, and improve
chances of acceptance and success with the KD
at home. Finally, although inborn errors in fatty
28
28 section I: Ketogenic Diet for Epilepsy in the Clinic
acid oxidation are rare, screening is not infallible.
We have picked up a handful of children with
beta-oxidation defects during admission in the
past 20 years. It is more likely that these children’s
defects would have been missed, and resulted in
significant morbidity had the diet been started as
an out-patient.
Out-patient initiation of the KD can be suc-
cessful, and in some countries it is the standard of
care (Neal et al., 2008; Rizzutti et al., 2007; Vaisleib
et  al., 2004). The out-patient advancement of the
KD is in general slower, often over several weeks
before a “full KD is achieved.” The centers that
use out-patient initiation of the KD must have a
flexible, available staff, and the ability to sched-
ule these patients for frequent out-patient visits
to make sure the diet is proceeding as expected
and that the child is safe. Transition into ketosis
is not directly observed. Any issues after hours
must be directed to the emergency department.
The amount of education provided is limited and
it requires that the family live near the epilepsy
center to minimize time traveling. Benefits with
an out-patient initiation include starting the KD in
the comfort of the child’s own home and a cheaper
overall cost. However, centers that use out-patient
initiation often have a higher dropout rate before
the 3-month mark when effectiveness is typically
determined, perhaps due to the above factors
(Levy et al., 2012).
THE CLAS SIC
K E TO G E N I C   D I E T
The classic ketogenic diet begins with fasting,
and only fluids excluding carbohydrates are
consumed. The duration of fasting varies. In the
initial protocols from the 1920 to the 1930s fast-
ing was commonly extended until 10% of body
weight was lost (Livingston, 1951; Wilder, 1921).
The actual time centers fast their patients has
decreased, but 12–72 hours is often implemented,
or “until ketones are large.” The KD is then started
at the full 90% fat composition and at a third of
the calories, advanced daily, until the full calorie
meal is tolerated.
Fasting
The KD has a strong historical tie to fasting. The
father of medicine, Hippocrates prescribed fast-
ing for his epilepsy patients. It is described in the
Bible (Mark 9:29), and was used in the early 1900s
in a cyclic fashion (for several weeks at a time) as
treatments for patients with epilepsy (Wheless,
2008). The KD was created to mimic the metabolic
changes that occur when we fast: lowering of the
blood glucose and insulin levels, utilization of the
body’s glycogen stores, slowing of the flux through
the glycolytic pathway, and finally utilization of
our fat stores via beta oxidation (Cahill, 1970;
Cahill and Owen, 1970). In the process of breaking
down fat, ketone bodies are produced and trans-
ported into our central nervous system for direct
use in energy production or indirectly affecting a
myriad of metabolic pathways leading to the “mir-
acle of seizure reduction” (Lutas and Yellen, 2013).
Ketone bodies are acidic, therefore, while fast-
ing or while on the KD, the overall “acid load” is
increased. Ketone bodies also suppress our appetite
and it is often difficult to get a lethargic, acidotic,
dehydrated child to eat a 90% fat meal without
vomiting. For children, particularly young ones,
this could become an issue that prolongs admis-
sions and worsens morbidity, and it frankly deters
some families from trying the KD. Many centers
have used a “kinder, gentler, gradual” advance-
ment without fast initiation. Wirrell (Wirrell
et al., 2002), first described their success with this
approach in a retrospective case series of 14 chil-
dren. The Korean group (Kim et al., 2004) stopped
fasting all of their patients. They instead used the
gradual caloric advancement approach of the 4:1
ratio and reported similar success in seizure reduc-
tion at 3 months in 41 patients compared with 81
historical fasting controls. Finally, a randomized
prospective trial compared the classical fasting
KD with a gradual initiation approach (Bergqvist
et al., 2005). In 48 patients using equivalence test-
ing, the gradual approach (1:1, 2:1, 3:1, 4:1, daily
advancement, full calorie KD) was equally effec-
tive at reducing seizures at 3  months, compared
with the classical KD protocol. Both protocols
achieved strong ketosis by the 5th day discharge,
the gradual protocol about 1  day later than the
fasting. Side effects were reduced by about two-
thirds, and interventions were significantly fewer
in the gradual protocol, with less weight loss, mild
and severe hypoglycemia, dehydration, acidosis,
need for bicarbonate and intravenous fluid admin-
istration. Vomiting was not quantified but rather
reported as present or not, and occurred in both
groups. Additional days needed in the hospital
were also reduced in the gradual group compared
with the fasting group, with a dropout rate of 4%
compared with 8% in the fasting group (very low,
overall) (Bergqvist et al., 2005). With this data in
hand many centers stopped the fasting process
for their routine admission and use it only when
speed of achieving ketosis is of the essence, as with
a child in status epilepticus (Cobo et  al., 2015).
The gradual initiation protocol has been modified,
29
Chapter 4: How Do You Implement the Diet?
individualized to fit the child better. The CHOP
protocol has eliminated the 1:1 ratio, instead using
the (2:1, 3:1, 4:1) approach. Many centers do not
advance to a full 4:1 ratio, depending on the sever-
ity of the child’s epilepsy, frequency of seizures,
and the response to lower ratios (Seo et al., 2007).
The ICSKD considers fasting as an option, but no
longer necessary (Kossoff et al., 2009).
Liquid/Formula versus Food
Ketogenic diet formulas have been created for chil-
dren who have gastrostomy tubes, and for infants.
In an attempt to shorten the admission time, to
quicken the acceptance of the high-fat KD, some
centers use a formula or liquid eggnog to initiate
the KD with similar success. Some centers have
reported higher overall seizure reduction with the
formula-fed children, perhaps due to lower chance
of noncompliance (Kossoff et al., 2004). The prob-
lem with not modeling a home situation during
the admission is that families are discharged hav-
ing had no experience or assistance cooking 90%
fat meals that are palatable and have to attempt this
by themselves, be it with the support of websites,
and other keto-coaches once home.
Caloric Restriction
Initiation
Caloric restriction was traditionally part of the KD
initiation, with the calories gradually advanced
during the initiation after the fast (1/3, 2/3, full cal-
orie 90% fat diet as tolerated). Bansal et al. found
better effectiveness at 3  months in 30 children
started on full calorie 4:1 KD without fast, com-
pared with 30 historical controls, but no difference
in side effects or interventions, perhaps related to
the lack of gradual adjustment to the 90% fat diet
in their protocol (Bansal et al., 2014).
Maintenance
Calories are important to a growing child, whose
needs are continuously changing during infancy
and childhood. Calories are controlled on the
KD until the dietitian changes the prescription.
Clinical practice from KD centers tells us if a
child is given excessive calories and gains weight
too quickly, it is associated with less reduction in
seizures and lower ketosis, until ideal body weight
is achieved (Freeman et  al., 2007). This has not
been tested in any trial. The calories are tradi-
tionally determined by a weighed 3-day dietary
record provided by parents before the KD is
started. Comparison of this caloric estimate to age
and gender RDA, estimated or measured resting
29
energy expenditure (REE) and an activity factor.
Children with epilepsy are less active than healthy
children and those who have additional motor dis-
abilities may need even fewer calories (Wong and
Wirrell, 2006). All these factors makes it difficult
to estimate calories for a KD plan.
Two studies have actually measured REE (an
estimate of the basal metabolic rate) in children
treated with the KD. In a short-term 6-month study
the KD did not change REE in 18 children, but
change in the respiratory quotient (RQ) correlated
with seizure reduction (Tagliabue et al., 2012). In
a longer, 15-month, prospective trial of 24 chil-
dren treated with the KD compared with 75 age-
matched controls, linear growth status declined
while weight status and REE were unchanged; REE
remained reduced in children with CP (Groleau
et al., 2014). Further, although the children gained
weight as calories were adjusted, this weight gain
came in the form of a change in body composi-
tion and relative increase in fat mass. That is, “you
become what you eat.” The increased calories did
not prevent the height deceleration seen in these
children, a long-term side effect now well estab-
lished in several prospective studies of the KD
(Bergqvist et al., 2008; Nation et al., 2014; Spulber
et al., 2009; Vining et al., 2002; Williams et al., 2002;
see chapter on side effects). In summary, calorie
restriction is not needed for initiation or mainte-
nance of the KD. Normal growth during the treat-
ment period is encouraged, but the KD appears to
alter body composition and caloric adjustments do
not prevent reduction in height velocity (growth
failure), which maybe IGF-1 mediated.
WHEN DO WE DETERMINE
EFFECTIVENES S OF THE KD?
Almost all efficacy trials related to dietary thera-
pies determine initial effectiveness at 3  months.
This is likely due to comparisons with anticonvul-
sive drug trials, where a 3-month design is stan-
dard practice (Sachdeo, 2007; Schmidt et al., 2014).
Do we really need to stay on the diet that long to
determine whether it works? Can we stop sooner,
or perhaps should we stay on the diet longer? To
answer these questions, Johns Hopkins University
and Children’s Memorial Hospital in a combined
study looked at when seizures began to reduce on
the ketogenic diet (Kossoff et  al., 2008)  in a ret-
rospective analysis of 99 children started on a 4:1
fasting KD or 3–4:1 gradual KD protocol. They
found that the median time to first improvement
was 5 days, with a range of 1–65 days. Reduction
in seizures occurred sooner in the children who
were fasted compared with a gradual protocol, but
30
30 section I: Ketogenic Diet for Epilepsy in the Clinic
long-term effectiveness was not different between
the two initiation protocols, confirming prior stud-
ies. Five subjects did not began to have any change
in their seizures until day 60 or more. Then 4/5 had
>50%–90% reduction in seizures and one became
seizure-free. They concluded that if no change is
seen in seizures frequency by 2  months, the KD
could be discontinued and that there was no need
for the full 3-month trial period. However, the
data can also be interpreted to confirm the use of
a 3-month minimal trial period of the KD. A sig-
nificant improvement including seizure freedom
in a 5% treatment resistant epilepsy cohort is not
unsignificant!
SEIZURE FREEDOM
Seizure freedom is the goal for majority of
patients trying the KD. In the short-term ran-
domized trials it has been reported in 15%–55%
( Martin et al, 2016). For longer-term studies the
data presented is not always a continuous vari-
able, instead often obtained just prior to the
time of interest, thereby artificially inflating the
percentage reported as seizure-free. This makes
it difficult to interpret the data and counsel our
families appropriately. Taub et al. tried to answer
this question in a retrospective chart review of
275 patients started on the KD, of whom 65 chil-
dren (24%) achieved seizure freedom (defined
as no seizure for 28  days) (Taub et  al., 2014).
All children had daily seizures before starting
the KD. The median time to becoming seizure-
free was 1.5  months, 72% became seizure-free
before 3 months, (considered early seizure-free),
while 28% became seizure-free after 3  months,
(considered late seizure-free). The longest time
to becoming seizure-free was 18  months in one
subject. Seizure recurrence (defined as any sei-
zure after having been seizure-free for 28  days)
occurred in 84%, and median time to recur-
rence was 3 months. However, 60% of those who
recurred still had less than one seizure per month.
Timing of discontinuation of the AED did not
affect return of seizures. In general, for those who
became seizure-free and recurred, their seizures
did not return to their prediet frequency. The
chance of remaining seizure-free at 18  months
was only 3%. Families should be counseled that
seizure freedom may occur at some point during
KD treatment, most often in the first few months.
If it happens, recurrence is more common rather
than complete remission. In general, seizures
remain greatly improved and do not return to the
frequency before the KD was started.
R AT I O S :   D O T H E Y M AT T E R ?
As the KD practices were questioned and clini-
cal protocols changed, less restrictive diets
emerged:  the Modified Atkins diet (MAD)
(Kossoff et al., 2003), or unlimited protein diet, as
it is called in Europe, and the low glycemic index
diet (Pfeifer and Thiele, 2005). These diets have
become popular alternatives to the KD because
of their less strict nature, less need for resources,
possibly milder side-effects and claims of “similar”
effectiveness to the KD. Some have even suggested
that the modified diets should replace the formal
KD for the majority of our treatment-resistant
patients (Auvin, 2012; Miranda et al., 2012). What
data do we have to support this claim? Both diets
are lower ratio KD, MAD at most a 2:1 if tightly
controlled and LGIT a 1–1.5:1 ratio.
Animal data to date suggest that higher ratios
are more effective in controlling seizures, although
the ratios used in rodent models (4–6.3:1) are
often much higher than what we would use in
humans (Bough et  al., 1999; Bough et  al., 2000;
Nylen et al., 2005). Some clinical data to compare
ratios have been randomized, but not blinded. Seo
et al. compared a 3:1 with a 4:1 KD in 76 random-
ized patients, and seizure reduction was measured
at 3  months. They found that the 4:1 ratio KD
reduced seizures more effectively and that there
were significantly more seizure-free patients in
this group than the 3:1 group (Seo et al., 2007). At
3  months, they crossed the two protocols for an
additional 3-month extension. Children who were
seizure-free at the 3-month mark on the 4:1 main-
tained their seizure freedom on the lower ratio,
indicating that perhaps a lower ratio can be used
later in the KD treatment without sacrificing effec-
tiveness. El-Rashidy et al. studied 40 patients with
symptomatic treatment-resistant epilepsy ran-
domized to MAD, 4:1 KD formula, and standard
diet continued medical treatment. They found that
the 4:1 formula at 3 and 6 months resulted in better
seizure reduction than the MAD and regular diet
(El-Rashidy et  al., 2013). Raju et  al. randomized
38 children to a 4:1 versus 2.5:1 KD and measured
effectiveness at 3 months. There was no significant
difference in responders or seizure-freedom rate
at 3 months (58% vs. 63% and 26 vs. 21%) (Raju
et al., 2011).
Kossoff et al. looked at MAD versus MAD with
liquid 4:1 supplement in the first month and found
that the liquid supplement (i.e., higher ratio) had
a better reduction in seizures at 3 months (Kossoff
et  al., 2011a). In a multicenter study, 27 patients
from several countries were identified who had
31
Chapter 4: How Do You Implement the Diet?
not responded completely to the MAD and who
were switched to the KD. Ten had further improve-
ment, 11 did not. Children with myoclonic astatic
epilepsy appeared to particularly improve with the
formal KD (Kossoff et al., 2010). In summary, cur-
rent data animal and human-based studies, some
prospective and randomized but none blinded,
indicate that higher ratios and the formal KD are
more effective at reducing seizures than the modi-
fied diets, at least in the short term. It is premature
to routinely discard the KD in favor for the modi-
fied diets. There are situations where the modified
diets are more appropriate (see chapter on modi-
fied diets). When choosing the MAD, families
should be counseled that these less restrictive diets
might come at the expense of less reduction in sei-
zures. A blinded randomized trial with long-term
follow-up to assess retention of seizure control
and seizure-freedom rate is needed to definitively
answer this question.
P R E D I C TO R O F   R E S P O N S E
The KD appears to work particularly well in some
electroclinical epilepsy syndromes. However many
patients do to fit into one of these categories, there-
fore a predictor of response to the dietary therapies
would be very helpful when trying to guide par-
ents about staying on the KD or not. In the mul-
tiple observational studies published there were no
demographic parameters that predicted response
(such as age, gender, seizure type).
The electroencephalogram however, appears
to be a helpful tool. Kessler et al. in a prospective
blinded study of 37 patients on the fast or gradual
KD initiation protocol assessed the routine EEG
at baseline, 1 and 3  months into KD treatment.
The EEGs were analyzed for background slowing,
interictal epileptiform discharges (IED), power
spectrum analysis, and manual spike index. They
found that 70% of the patients became respond-
ers, with 27% seizure-free. Over time the EEG
background slowing improved and there was an
overall increase in beta, but this change did not
predict responder status. Interictal epileptiform
discharges were reduced by 65% by the 1-month
EEG. The 1-month reduction in IED strongly pre-
dicted response status to KD with odds ratio of
4.8 (95% CI 1.1–21.3, p  =  .03). They concluded
that the routine EEG is an excellent predictor of
response and can assist neurologists when coun-
seling and encouraging parents about staying with
the KD in the first 3  months of therapy (Kessler
et  al., 2011). Ebus et  al. confirmed these find-
ings via 24-hour video-EEG in 34 patients on at
31
baseline 6 weeks and 3 months into the KD. They
had a low response rate over all, only 26%, but did
notice a proportional reduction in IED of 30%
during sleep strongly predicted reduction in sei-
zure or response status (Ebus et al., 2014).
D U R AT I O N O F   T R E AT M E N T
The KD was initially created for a 2-year treat-
ment period and a 6- to 12-month wean. However,
there is no definitive treatment period, and it can
be individualized based on the child’s under-
lying condition and response. There are some
electroclinical epilepsy syndromes, such as infan-
tile spasms, where if the patient responds and
becomes seizure-free, the diet treatment period
may be shortened to 6–12 months, particularly if
the EEG has normalized (Kossoff et al., 2002). For
others, the KD has resulted in such improvement
in the child’s overall quality of life, not only reduc-
tion in seizures but also removal of all or most
medications, that the family chooses to stay on the
KD longer (Hallbook et  al., 2007). Most centers
have a few patients who have been on the KD for
>15 years (Kossoff et al., 2007). Often the ratios in
the KD in this group of patients have been weaned
to the lowest ratio that successfully maintains sei-
zure control, but attempts to come off the KD have
resulted in return of seizures. Continued diet use
of dietary therapies has to be weighed against the
risk and complications and long-term side effects
of a high-fat diet (Zupec-Kania and Zupanc, 2008;
see chapter on side effects). Patients who are on the
KD for life, due to their underlying inborn error of
metabolism such as Glut1 syndrome and pyruvate
dehydrogenase disorder require particular care
in management of their diets (see chapter about
Glut1).
REFERENCES
Auvin, S. (2012). Should we routinely use modified
Atkins diet instead of regular ketogenic diet to
treat children with epilepsy? Seizure 21, 237–240.
Bansal, S., Cramp, L., Blalock, D., Zelleke, T.,
Carpenter, J., and Kao, A. (2014). The ketogenic
diet:  initiation at goal calories versus gradual
caloric advancement. Pediatr Neurol 50, 26–30.
Bergqvist, A.G., Schall, J.I., Gallagher, P.R., Cnaan, A.,
and Stallings, V.A. (2005). Fasting versus gradual
initiation of the ketogenic diet:  a prospective,
randomized clinical trial of efficacy. Epilepsia 46,
1810–1819.
Bergqvist, A.G., Schall, J.I., Stallings, V.A., and Zemel,
B.S. (2008). Progressive bone mineral content loss
in children with intractable epilepsy treated with
the ketogenic diet. Am J Clin Nutr 88, 1678–1684.
32
32 section I: Ketogenic Diet for Epilepsy in the Clinic
Bergqvist, A.G.C. (2004). Indications and contraindi-
cations of the ketogenic diet. In Epilepsy and the
Ketogenic Diet, C.E. Stafstrom and J.M. Rho, eds.
(Totowa, NJ: Humana Press), pp. 115–117.
Bough, K.J., Chen, R.S., and Eagles, D.A. (1999). Path
analysis shows that increasing ketogenic ratio, but
not beta-hydroxybutyrate, elevates seizure thresh-
old in the rat. Dev Neurosci 21, 400–406.
Bough, K.J., Yao, S.G., and Eagles, D.A. (2000).
Higher ketogenic diet ratios confer protection
from seizures without neurotoxicity. Epilepsy Res
38, 15–25.
Cahill, G.F., Jr. (1970). Starvation in man. N Engl J
Med 282, 668–675.
Cahill, G.F., Jr., and Owen, O.E. (1970). Body fuels and
starvation. Int Psychiatry Clin 7, 25–36.
Chee, C., Rice, C., Fenton, C., Groveman, S., and
Bergqvist, A.G.C. (2014). School personnel
understanding of ketogenic diet. In 4th Global
Symposium for Dietary Treatments of Epilepsy
and other Neurological Disorders (Matthews
Friends, pg 57, Liverpool, England).
Cobo, N.H., Sankar, R., Murata, K.K., Sewak, S.L.,
Kezele, M.A., and Matsumoto, J.H. (2015). The
ketogenic diet as broad-spectrum treatment for
super-refractory pediatric status epilepticus: chal-
lenges in implementation in the pediatric and
neonatal intensive care units. J Child Neurol 30,
259–266.
Di Pisa, V., Cecconi, I., Gentile, V., Di Pietro, E.,
Marchiani, V., Verrotti, A., and Franzoni, E. (2012).
Case report of pyruvate dehydrogenase deficiency
with unusual increase of fats during ketogenic diet
treatment. J Child Neurol 27, 1593–1596.
Ebus, S.C., Lambrechts, D.A., Herraets, I.J., Majoie,
M.J., de Louw, A.J., Boon, P.J., Aldenkamp, A.P.,
and Arends, J.B. (2014). Can an early 24-hour
EEG predict the response to the ketogenic diet?
A prospective study in 34 children and adults with
refractory epilepsy treated with the ketogenic diet.
Seizure 23, 468–474.
El-Rashidy, O.F., Nassar, M.F., Abdel-Hamid, I.A.,
Shatla, R.H., Abdel-Hamid, M.H., Gabr, S.S.,
Mohamed, S.G., El-Sayed, W.S., and Shaaban, S.Y.
(2013). Modified Atkins diet vs classic ketogenic
formula in intractable epilepsy. Acta Neurol Scand
128, 402–408.
Fenton, C., Groveman, S., Chee, C., Rice, C., and
Bergqvist, A.G.C. (2014). Establishing a keto-
genic teaching kitchen in the acute care setting.
In 4th Global Symposium for Dietary Treatments
of Epilepsy and Other Neurological Disorders,
(Matthews Friends, pg 47, Liverpool, England).
Freeman, J.M., Kossoff, E., and Freeman, M.T.K. (2007).
The Ketogenic Diet: A Treatment for Children and
Others with Epilepsy (New York: Demos Medical
Publishing).
Groleau, V., Schall, J.I., Stallings, V.A., and Bergqvist,
C.A. (2014). Long-term impact of the ketogenic
diet on growth and resting energy expenditure in
children with intractable epilepsy. Dev Med Child
Neurol 56, 898–904.
Groveman, S., Fenton, C., Chee, C., Rice, C., and
Bergqvist, A.G.C. (2014). Use of a ketogenic teach-
ing kitchen in acute care setting. In 4th Global
Symposium for Dietary Treatments of Epilepsy
and Other Neurological Disorders, (Matthews
Friends, pg 47, Liverpool, England).
Hallbook, T., Lundgren, J., and Rosen, I. (2007).
Ketogenic diet improves sleep quality in chil-
dren with therapy-resistant epilepsy. Epilepsia
48, 59–65.
Horvath, R., Gorman, G., and Chinnery, P.F. (2008).
How can we treat mitochondrial encephalomyop-
athies? Approaches to therapy. Neurotherapeutics
5, 558–568.
Kang, H.C., Lee, Y.M., Kim, H.D., Lee, J.S., and Slama,
A. (2007). Safe and effective use of the ketogenic
diet in children with epilepsy and mitochon-
drial respiratory chain complex defects. Epilepsia
48, 82–88.
Kessler, S.K., Gallagher, P.R., Shellhaas, R.A., Clancy,
R.R., and Bergqvist, A.G. (2011). Early EEG
improvement after ketogenic diet initiation.
Epilepsy Res 94, 94–101.
Kim, D.W., Kang, H.C., Park, J.C., and Kim, H.D.
(2004). Benefits of the nonfasting ketogenic diet
compared with the initial fasting ketogenic diet.
Pediatrics 114, 1627–1630.
Kossoff, E.H., Pyzik, P.L., McGrogan, J.R., Vining,
E.P., and Freeman, J.M. (2002). Efficacy of the
ketogenic diet for infantile spasms. Pediatrics 109,
780–783.
Kossoff, E.H., Krauss, G.L., McGrogan, J.R., and
Freeman, J.M. (2003). Efficacy of the Atkins diet
as therapy for intractable epilepsy. Neurology 61,
1789–1791.
Kossoff, E.H., McGrogan, J.R., and Freeman, J.M.
(2004). Benefits of an all-liquid ketogenic diet.
Epilepsia 45, 1163.
Kossoff, E.H., and McGrogan, J.R. (2005). Worldwide
use of the ketogenic diet. Epilepsia 46, 280–289.
Kossoff, E.H., Turner, Z., and Bergey, G.K. (2007).
Home-guided use of the ketogenic diet in a
patient for more than 20 years. Pediatr Neurol 36,
424–425.
Kossoff, E.H., Laux, L.C., Blackford, R., Morrison, P.F.,
Pyzik, P.L., Hamdy, R.M., Turner, Z., and Nordli,
D.R., Jr. (2008). When do seizures usually improve
with the ketogenic diet? Epilepsia 49, 329–333.
Kossoff, E.H., Zupec-Kania, B.A., Amark, P.E.,
Ballaban-Gil, K.R., Christina Bergqvist, A.G.,
Blackford, R., Buchhalter, J.R., Caraballo, R.H.,
Helen Cross, J., Dahlin, M.G., et  al. (2009).
33
Chapter 4: How Do You Implement the Diet?
Optimal clinical management of children receiv-
ing the ketogenic diet:  recommendations of
the International Ketogenic Diet Study Group.
Epilepsia 50, 304–317.
Kossoff, E.H., Bosarge, J.L., Miranda, M.J., Wiemer-
Kruel, A., Kang, H.C., and Kim, H.D. (2010). Will
seizure control improve by switching from the
modified Atkins diet to the traditional ketogenic
diet? Epilepsia 51, 2496–2499.
Kossoff, E.H., Dorward, J.L., Turner, Z., and Pyzik, P.L.
(2011a). Prospective study of the modified Atkins
diet in combination with a ketogenic liquid sup-
plement during the initial month. J Child Neurol
26, 147–151.
Kossoff, E.H., Freeman, J.M., Turner, Z., and Rubenstein,
J.E. (2011b). Ketogenic Diets:  Treatments for
Epilepsy and Other Disorders, 5th edition
(New York: Demos Medical Publishing).
Levy, R.G., Cooper, P.N., and Giri, P. (2012). Ketogenic
diet and other dietary treatments for epilepsy.
Cochrane Database Syst Rev 3, CD001903.
Livingston, S. (1951). The ketogenic diet in the treat-
ment of epilepsy in children. Postgrad Med 10,
333–336.
Lutas, A., and Yellen, G. (2013). The ketogenic
diet: metabolic influences on brain excitability and
epilepsy. Trends Neurosci 36, 32–40.
MacCracken, K.A., and Scalisi, J.C. (1999).
Development and evaluation of a ketogenic diet
program. J Am Diet Assoc 99, 1554–1558.
Martin, K., Jackson, C.F., Levy, R.G., Cooper, P.N.
(2016). Ketogenic diet and other dietary treat-
ments for epilepsy. Cochrane Database Syst Rev 4,
CD001903
Miranda, M.J., Turner, Z., and Magrath, G. (2012).
Alternative diets to the classical ketogenic diet: can
we be more liberal? Epilepsy Res 100, 278–285.
Nation, J., Humphrey, M., MacKay, M., and Boneh, A.
(2014). Linear growth of children on a ketogenic
diet:  does the protein-to-energy ratio matter? J
Child Neurol 29, 1496–1501.
Neal, E.G., Chaffe, H., Schwartz, R.H., Lawson, M.S.,
Edwards, N., Fitzsimmons, G., Whitney, A., and
Cross, J.H. (2008). The ketogenic diet for the treat-
ment of childhood epilepsy:  a randomised con-
trolled trial. Lancet Neurol 7, 500–506.
Nylen, K., Likhodii, S., Abdelmalik, P.A., Clarke, J.,
and Burnham, W.M. (2005). A comparison of the
ability of a 4:1 ketogenic diet and a 6.3:1 ketogenic
diet to elevate seizure thresholds in adult and
young rats. Epilepsia 46, 1198–1204.
Pfeifer, H.H., and Thiele, E.A. (2005). Low-glycemic-
index treatment:  a liberalized ketogenic diet for
treatment of intractable epilepsy. Neurology 65,
1810–1812.
Raju, K.N., Gulati, S., Kabra, M., Agarwala, A.,
Sharma, S., Pandey, R.M., and Kalra, V. (2011).
33
Efficacy of 4:1 (classic) versus 2.5:1 ketogenic ratio
diets in refractory epilepsy in young children:  a
randomized open labeled study. Epilepsy Res 96,
96–100.
Rizzutti, S., Ramos, A.M., Muszkat, M., and Gabbai,
A.A. (2007). Is hospitalization really necessary
during the introduction of the ketogenic diet? J
Child Neurol 22, 33–37.
Sachdeo, R. (2007). Monotherapy clinical trial design.
Neurology 69, S23–27.
Schmidt, D., Friedman, D., and Dichter, M.A. (2014).
Anti-epileptogenic clinical trial designs in epi-
lepsy:  issues and options. Neurotherapeutics 11,
401–411.
Seo, J.H., Lee, Y.M., Lee, J.S., Kang, H.C., and Kim,
H.D. (2007). Efficacy and tolerability of the keto-
genic diet according to lipid:nonlipid ratios: com-
parison of 3:1 with 4:1 diet. Epilepsia 48, 801–805.
Snyder, D. (2006). Keto kids (New  York:  Demos
Medical Publishing).
Spulber, G., Spulber, S., Hagenas, L., Amark, P.,
and Dahlin, M. (2009). Growth dependence on
insulin-like growth factor-1 during the ketogenic
diet. Epilepsia 50, 297–303.
Stafstrom, C.E., Zupec-Kania, B., and Rho, J.M. (2008).
Ketogenic diet and treatments:  introduction/per-
spectives. Epilepsia 49(Suppl 8), 1–2.
Tagliabue, A., Bertoli, S., Trentani, C., Borrelli, P., and
Veggiotti, P. (2012). Effects of the ketogenic diet
on nutritional status, resting energy expenditure,
and substrate oxidation in patients with medically
refractory epilepsy: a 6-month prospective obser-
vational study. Clin Nutr 31, 246–249.
Taub, K.S., Kessler, S.K., and Bergqvist, A.G. (2014).
Risk of seizure recurrence after achieving initial
seizure freedom on the ketogenic diet. Epilepsia
55, 579–583.
Vaisleib, I.I., Buchhalter, J.R., and Zupanc, M.L.
(2004). Ketogenic diet: outpatient initiation, with-
out fluid, or caloric restrictions. Pediatr Neurol
31, 198–202.
Vining, E.P., Pyzik, P., McGrogan, J., Hladky, H.,
Anand, A., Kriegler, S., and Freeman, J.M. (2002).
Growth of children on the ketogenic diet. Dev
Med Child Neurol 44, 796–802.
Weber, T.A., Antognetti, M.R., and Stacpoole, P.W.
(2001). Caveats when considering ketogenic diets
for the treatment of pyruvate dehydrogenase com-
plex deficiency. J Pediatr 138, 390–395.
Wheless, J.W. (2008). History of the ketogenic diet.
Epilepsia 49(Suppl 8), 3–5.
Wijburg, F.A., Barth, P.G., Bindoff, L.A., Birch-Machin,
M.A., van der Blij, J.F., Ruitenbeek, W., Turnbull,
D.M., and Schutgens, R.B. (1992). Leigh syndrome
associated with a deficiency of the pyruvate dehy-
drogenase complex:  results of treatment with a
ketogenic diet. Neuropediatrics 23, 147–152.
34
34 section I: Ketogenic Diet for Epilepsy in the Clinic
Wilder, R.M. (1921). The effect of ketonemia on
the course of epilepsy. Mayo Clin Bulletin 2,
307–308.
Williams, S., Basualdo-Hammond, C., Curtis, R., and
Schuller, R. (2002). Growth retardation in children
with epilepsy on the ketogenic diet: a retrospective
chart review. J Am Diet Assoc 102, 405–407.
Wirrell, E.C., Darwish, H.Z., Williams-Dyjur, C.,
Blackman, M., and Lange, V. (2002). Is a fast
necessary when initiating the ketogenic diet? J
Child Neurol 17, 179–182.
Wong, J., and Wirrell, E. (2006). Physical activity in chil-
dren/teens with epilepsy compared with that in their
siblings without epilepsy. Epilepsia 47, 631–639.
Zupec-Kania, B., and Zupanc, M.L. (2008). Long-term
management of the ketogenic diet: seizure moni-
toring, nutrition, and supplementation. Epilepsia
49(Suppl 8), 23–26.
35
5
Glut1 Deficiency and the Ketogenic Diets
JOERG KLEPPER, MD, PHD
GLUT1 DEFICIENCY
In the fed state the human brain relies entirely
on glucose for energy metabolism. Glucose entry
into the brain is exclusively mediated by the facili-
tated glucose transporter protein Glut1. Impaired
glucose transport into the brain resulting from
Glut1 deficiency (Glut1D, OMIM 606777)  (De
Vivo et al., 1991) will cause a cerebral “energy cri-
sis,” particularly as the developing brain requires
3–4 times more energy than the adult brain.
Clinical features are global developmental delay,
early-onset epilepsy, and a complex movement
order (Klepper and Leiendecker, 2007). As such,
the disease mechanism of Glut1D is intriguingly
straightforward:
1. The Glut1 defect results in low CSF
glucose concentrations, termed
hypoglycorrhachia.
2. Hypoglycorrhachia results in an epileptic
encephalopathy and movement disorder.
3. Glut1D is treatable by means of a ketogenic
diet providing ketones as an alternative fuel
for brain energy metabolism (Figure 5.1).
DIAGNO SIS
The diagnosis of Glut1D rests on (1) low glucose
concentrations in CSF, termed hypoglycorrha-
chia, determined via fasting lumbar puncture, and
(2) mutations in the SLC2A1 gene (De Giorgis and
Veggiotti, 2013). The lumbar puncture should be
performed in a metabolic steady-state, for exam-
ple, following a 4- to 6-hour fast. In Glut1D, CSF
glucose concentrations usually are below 40 mg/
dL. A  CSF/plasma glucose ratio below 0.45 is
helpful to determine hypoglycorrhachia, and CSF
lactate should always be low to normal (for age-
dependent reference values of CSF glucose and
lactate, see Leen et al., 2013b). Patients with milder
phenotypes do have CSF glucose values above 41
mg/dL but never normal (De Vivo and Wang,
2008). Mild or borderline hypoglycorrhachia
associated with missense mutations are often pres-
ent in extended phenotypes and indicate that a
SLC2A1 analysis should be performed along with
a diagnostic lumbar puncture (Weber et al., 2008).
In approximately 85%–90% of cases the condition
can be confirmed by heterozygous mutations in
the SLC2A1 gene. Of note, the absence of SLC2A1
mutations does not exclude the diagnosis (Klepper,
2013). Inheritance is predominantly autosomal
dominant and mutations are mostly de novo, but
individual cases of autosomal recessive trans-
mission have been described (Brockmann et  al.,
2001; Klepper et  al., 2009; Klepper et  al., 2001).
Prenatal testing is available if the disease-causing
mutation has been identified. Neuroimaging is
generally uninformative, but PET scanning has
been reported to show a distinctive metabolic
footprint in brain (Pascual et  al., 2002). Further
diagnostic tools such as glucose uptake assay into
erythrocytes are available on a research basis only
(Klepper et al., 1999).
C L I N I C A L P R E S E N TAT I O N
Classic Glut1D presents with unspecific develop-
mental delay associated with early-onset epilepsy,
a complex movement disorder, and paroxysmal
nonepileptic events. Any combinations of these
features are possible (Figure 5.2) (Pearson et al.,
2013). Severe cases develop secondary micro-
cephaly, reflecting the impairment of the devel-
oping brain. Seizure type may vary from cyanotic
attacks and staring spells in early infancy to
childhood absence epilepsy and myoclonic/gen-
eralized grand-mal seizures in later childhood.
Certain epilepsy syndromes such as early-onset
absence epilepsy (onset < age 4 years) and myo-
clonic astatic epilepsy (Doose syndrome) are
associated with Glut1D (Mullen et al., 2010; Suls
et al., 2008). The complex movement disorder in
Glut1D features ataxic-spastic and gait abnormal-
ities, action limb dystonia, mild chorea, cerebellar
action tremor, myoclonus, and dyspraxia (Pons
36

36 section I: Ketogenic Diet for Epilepsy in the Clinic

Glucose
KDT
1 BBB

Fat

Brain β-Ox.

Acetyl-CoA Liver
2

Ketones

TCA cycle

Energy

FIGURE 5.1 The metabolic concept of Glut1 Deficiency.

➊: the defect of the Glut1 transporter at the blood-brain barrier results in cerebral energy failure
➋: Ketones derived from either body fat (fasting) or nutritional fat (KDT) serve as an alternative fuel to restore cerebral energy.

et al., 2010). In adults, exertion-induced dystonia, K E TO G E N I C D I E T

and stomatin-associated cryohydrocytosis (Flatt THERAPIES IN GLUT1D
et  al., 2011; Weber et  al., 2008), a rare form of Currently ketogenic diet therapy (KDT) is the
hemolytic anemia, have been identified as allelic only treatment for this disorder of brain energy
diseases associated with SLC2A1-mutations. metabolism. The metabolic concept is to mimic

Epilepsy Movement
Infantile seizures Spasticity disorder
Ataxia
Absence seizures Dystonia
Chorea

Cognitive impairment
Classic phenotype Delayed adaptive behavior
Developmental Variable attention
encephalopathy

Cognitive/behavioral
disturbances
FIGURE 5.2 The complex phenotype of Glut1 deficiency (with permission from Pearson et al., 2013).
37
Chapter 5: Glut1 Deficiency and the Ketogenic Diets
the metabolic state of fasting and thus “refuel” the
brain using ketones from the KDT (Figure 5.1).
Anticonvulsant mechanisms described for KDT
in intractable childhood epilepsy (Masino and
Rho, 2012) also apply in Glut1D and contribute to
seizure control. In addition, KDTs provide unspe-
cific positive effects such as improved physical
endurance, improved attention span, and alert-
ness (Lambrechts et  al., 2012; Ramm-Pettersen
et al., 2014) and have a substantial positive impact
on the associated movement disorder (Leen et al.,
2013a).
However, KDTs for Glut1D differ from KDTs
for intractable childhood epilepsy in the following
aspects:
• Ketones in Glut1D provide an alternative
fuel for the brain, restoring the cerebral
energy deficit. The higher the ketosis,
the more energy is provided for the
developing brain. Within this concept the
KDT providing the highest ketosis is best
for the patient.
• Introduction of KDT should be as early
as possible, even in infancy, to meet the
energy demands of the developing brain.
• KDT in Glut1D should be continued
throughout childhood into adolescence
to provide ketones as an alternative
fuel for adequate brain development.
Discontinuation of KDT after
approximately 2 years, as discussed in
intractable childhood epilepsy, is not
recommended in Glut1D patients.
The practical approach to KDT in Glut1D
does not differ substantially from KDT for intrac-
table childhood epilepsy (Klepper, 2012). Clinical
management such as initiation, supplementa-
tion, and maintenance of KDT follow established
guidelines (Kossoff et  al., 2009). In theory, any
KDT providing measurable ketosis may be used
in Glut1D according to age, clinical presenta-
tion, and compliance. In practice, long-chain and
medium-chain KDT and the MAD are established
therapies for Glut1D, whereas the low glycemic
index treatment is not.
Certain aspects are special to KDT in Glut1D.
In contrast to the efficacy of KDT in the treat-
ment of childhood epilepsy the vast majority of
patients with Glut1D show an immediate and
continuing response. This is even considered a
diagnostic tool:  if KDT is ineffective, reconsider
the diagnosis. Long-term adverse effects of KDT
such as growth retardation, prolonged QT interval
37
in ECG, selenium deficiency, renal stones, ele-
vated serum lipids, and atherosclerosis are serious
concerns in Glut1D—long-term data is only just
emerging to assess these questions. Emerging tools
for follow-up screening are dexa-scans for bone
density (Bergqvist et al., 2008) and ultrasound of
the carotid arteries for atherosclerosis (Coppola
et al., 2014; Kapetanakis et al., 2014). On follow-up
into puberty, recommendations to maintain KDT
in Glut1D throughout childhood and adolescence
often are particularly difficult. Achieving compli-
ance after years on KDT often is the most impor-
tant practical problem in this age group.
Novel compounds such as ketone-esters are
currently discussed to supplement or even replace
KDT in childhood epilepsy and metabolic disease.
Tripheptanoin-based metabolic therapy and the
concept of anaplerosis are discussed elsewhere in
this book (see ; Borges, chapter 34). Clinical trials
and experimental data in Glut1-deficient mice may
offer novel approaches toward Glut1D therapy.
OPEN QUESTIONS
Clinical symptoms of Glut1D appear to be age-
dependent. Epilepsy is the prominent clinical
feature in infants and children, but it often stabi-
lizes whereas movement disorders and paroxys-
mal events increase with age and seem to be the
main clinical problem in adolescence and adult-
hood (Alter et al., 2015). Here the impact of KDT
is less clear, although response to classic KDT
and MAD have been described for Glut1D. In a
small subgroup of patients with confirmed Glut1D
symptoms do not respond adequately to KDT. For
reasons yet unclear, epilepsy remains uncontrolled
and movement disorders, in particular paroxysmal
events, increase despite KDT.
The use of KDT in adult Glut1D seems reasonable,
but is by no means an established recommendation—
patients find it very difficult, symptoms such as epi-
lepsy seem to stabilize in adults, and there is simply
not enough data on the benefit and long-term adverse
effects of KDT in this age group (Leen et al., 2014).
The use of KDT in Glut1D allelic diseases such as
paroxysmal exertion-induced dystonia, early-onset
absence epilepsy, or stomatin-deficient cryohydro-
cytosis also remains unclear—originally considered
an exclusively pediatric disorder, Glut1D has now
moved into the focus of adult neurology.
A B B R E V I AT I O N S
Glut1D Glut1 Deficiency
PED Paroxysmal exercise-induced
dystonia
38
38 section I: Ketogenic Diet for Epilepsy in the Clinic
CSF cerebrospinal fluid
KDT ketogenic diet therapy
MAD modified Atkins diet
CONFLICTS OF INTEREST
The author has received speaker honoraria and
travel funds from Nutricia GmbH, SHS, Heilbronn,
Germany, and has received project support from
the parents support group Glut1-Förderverein,
Bremen, Germany.
REFERENCES
Alter, A.S., Engelstad, K., Hinton, V.J., Montes, J.,
Pearson, T.S., Akman, C.I., and De Vivo, D.C.
(2015). Long-term clinical course of Glut1 defi-
ciency syndrome. J Child Neurol 30, 160–169.
Bergqvist, A.G., Schall, J.I., Stallings, V.A., and Zemel,
B.S. (2008). Progressive bone mineral content loss
in children with intractable epilepsy treated with
the ketogenic diet. Am J Clin Nutr 88, 1678–1684.
Brockmann, K., Wang, D., Korenke, C.G., von Moers,
A., Ho, Y.Y., Pascual, J.M., Kuang, K., Yang, H., Ma,
L., Kranz-Eble, P., et al. (2001). Autosomal domi-
nant Glut-1 deficiency syndrome and familial epi-
lepsy. Ann Neurol 50, 476–485.
Coppola, G., Natale, F., Torino, A., Capasso, R.,
D’Aniello, A., Pironti, E., Santoro, E., Calabrò, R.,
and Verrotti, A. (2014). The impact of the keto-
genic diet on arterial morphology and endothelial
function in children and young adults with epi-
lepsy: a case-control study. Seizure 23, 260–265.
De Giorgis, V., and Veggiotti, P. (2013). Glut1 defi-
ciency syndrome 2013:  current state of the art.
Seizure 22, 803–811.
De Vivo, D.C., Trifiletti, R.R., Jacobson, R.I., Ronen,
G.M., Behmand, R.A., and Harik, S.I. (1991).
Defective glucose transport across the blood-brain
barrier as a cause of persistent hypoglycorrhachia,
seizures, and developmental delay. N Engl J Med
325, 703–709.
De Vivo, D.C., and Wang, D. (2008). Glut1 defi-
ciency:  CSF glucose; How low is too low? Rev
Neurol (Paris) 164, 877–880.
Flatt, J.F., Guizouarn, H., Burton, N.M., Borgese,
F., Tomlinson, R.J., Forsyth, R.J., Baldwin, S.A.,
Levinson, B.E., Quittet, P., Aguilar-Martinez, P.,
et al. (2011). Stomatin-deficient cryohydrocytosis
results from mutations in SLC2A1: a novel form of
Glut1 deficiency syndrome. Blood 118, 5267–5277.
Kapetanakis, M., Liuba, P., Odermarsky, M., Lundgren,
J., and Hallböök, T. (2014). Effects of ketogenic
diet on vascular function. Eur J Paediatr Neurol
18, 489–494.
Klepper, J., Garcia-Alvarez, M., O’Driscoll, K.R.,
Parides, M.K., Wang, D., Ho, Y.Y., and De Vivo,
D.C. (1999). Erythrocyte 3-O-methyl-D-glucose
uptake assay for diagnosis of glucose-transporter-
protein syndrome. J Clin Lab Anal 13, 116–121.
Klepper, J., Willemsen, M., Verrips, A., Guertsen, E.,
Herrmann, R., Kutzick, C., Florcken, A., and Voit,
T. (2001). Autosomal dominant transmission of
GLUT1 deficiency. Hum Mol Genet 10, 63–68.
Klepper, J., and Leiendecker, B. (2007). GLUT1 defi-
ciency syndrome -2007 update. Dev Med Child
Neurol 49, 707–716.
Klepper, J., Scheffer, H., Elsaid, M.F., Kamsteeg,
E.J., Leferink, M., and Ben-Omran, T. (2009).
Autosomal recessive inheritance of GLUT1 defi-
ciency syndrome. Neuropediatrics 40, 207–210.
Klepper, J. (2012). Glut1 deficiency syndrome in clini-
cal practice. Epilepsy Res 100, 272–277.
Klepper, J. (2013). Absence of SLC2A1 mutations
does not exclude Glut1 deficiency syndrome.
Neuropediatrics 44, 235–236.
Kossoff, E.H., Zupec-Kania, B.A., Amark, P.E.,
Ballaban-Gil, K.R., Christina Bergqvist, A.G.,
Blackford, R., Buchhalter, J.R., Caraballo, R.H.,
Helen Cross, J., Dahlin, M.G., et al. (2009). Optimal
clinical management of children receiving the keto-
genic diet:  recommendations of the International
Ketogenic Diet Study Group. Epilepsia 50, 304–317.
Lambrechts, D.A., Bovens, M.J., de la Parra, N.M.,
Hendriksen, J.G., Aldenkamp, A.P., and Majoie,
M.J. (2012). Ketogenic diet effects on cognition,
mood, and psychosocial adjustment in children.
Acta Neurol Scand. 127(2), 103–108.
Leen, W.G., Mewasingh, L., Verbeek, M.M., Kamsteeg,
E.J., van de Warrenburg, B.P., and Willemsen,
M.A. (2013a). Movement disorders in GLUT1
deficiency syndrome respond to the modified
Atkins diet. Mov Disord 28, 1439–1442.
Leen, W.G., Wevers, R.A., Kamsteeg, E.J., Scheffer, H.,
Verbeek, M.M., and Willemsen, M.A. (2013b).
Cerebrospinal fluid analysis in the workup of
GLUT1 deficiency syndrome: a systematic review.
JAMA Neurol 70, 1440–1444.
Leen, W.G., Taher, M., Verbeek, M.M., Kamsteeg, E.J.,
van de Warrenburg, B.P., and Willemsen, M.A.
(2014). GLUT1 deficiency syndrome into adult-
hood: a follow-up study. J Neurol 261, 589–599.
Masino, S.A., and Rho, J.M. (2012). Mechanisms of
Ketogenic Diet Action. In Noebels, J.L., Avoli, M.,
Rogawski, M.A., Olsen, R.W., Delgado-Escueta, A.V.,
editors. Jasper's Basic Mechanisms of the Epilepsies
[Internet]. 4th edition. Bethesda (MD):  National
Center for Biotechnology Information (US).
Mullen, S.A., Suls, A., De Jonghe, P., Berkovic, S.F.,
and Scheffer, I.E. (2010). Absence epilepsies with
widely variable onset are a key feature of familial
Glut1 deficiency. Neurology 75, 432–440.
Pascual, J.M., Van Heertum, R.L., Wang, D., Engelstad,
K., and De Vivo, D.C. (2002). Imaging the meta-
bolic footprint of Glut1 deficiency on the brain.
Ann Neurol 52, 458–464.
39
Chapter 5: Glut1 Deficiency and the Ketogenic Diets
Pearson, T.S., Akman, C., Hinton, V.J., Engelstad, K.,
and De Vivo, D.C. (2013). Phenotypic spectrum
of glucose transporter type 1 deficiency syndrome
(Glut1 DS). Curr Neurol Neurosci Rep 13, 342.
Pons, R., Collins, A., Rotstein, M., Engelstad, K., and
De Vivo, D.C. (2010). The spectrum of movement
disorders in Glut-1 deficiency. Mov Disord 25,
275–281.
Ramm-Pettersen, A., Stabell, K.E., Nakken, K.O., and
Selmer, K.K. (2014). Does ketogenic diet improve
cognitive function in patients with GLUT1-DS?
A 6- to 17-month follow-up study. Epilepsy Behav
39, 111–115.
39
Suls, A., Dedeken, P., Goffin, K., Van Esch, H.,
Dupont, P., Cassiman, D., Kempfle, J., Wuttke,
T.V., Weber, Y., Lerche, H., et  al. (2008).
Paroxysmal exercise-induced dyskinesia and
epilepsy is due to mutations in SLC2A1, encod-
ing the glucose transporter Glut1. Brain 131,
183–144.
Weber, Y.G., Storch, A., Wuttke, T.V., Brockmann, K.,
Kempfle, J., Maljevic, S., Margari, L., Kamm, C.,
Schneider, S.A., Huber, S.M., et al. (2008). Glut1
mutations are a cause of paroxysmal exertion-
induced dyskinesias and induce hemolytic ane-
mia by a cation leak. J Clin Invest 118, 2157–2168.
40
6
Ketogenic Diet in Established Epilepsy Indications
A N N M . B E R G I N, M B , S C M , M R C P ( U K )
INTRODUCTION
Arising from biblical anecdote and early 20th-
century explorations into the effect of fasting for
seizure management, the ketogenic diet (KD)
developed alongside medical knowledge of the
biochemistry of fasting/starvation and ketosis
(Wheless, 2014). Falling into relative disuse with
the arrival of phenytoin and subsequent anticon-
vulsant medications, it continued to be used mainly
in pediatrics, and primarily under the auspices of
Dr. John Freeman and colleagues at Johns Hopkins
University. After the development of waves of new
anticonvulsant medications that were associated
with reduction in toxicity but not with greater effi-
cacy in patients with refractory epilepsy, interest in
the KD reemerged. Over the years, multiple case
series, describing cases achieving remarkable and
sustained efficacy in a substantial proportion of
children with the most intractable epilepsies (>400
seizures/month) had been published from the
Johns Hopkins group (Hemingway et  al., 2001).
With the renewed interest, a multicenter group
showed that other centers using the same protocol
could achieve similar results (Vining et al., 1998),
and finally a randomized controlled study was
conducted in London, again supporting the effi-
cacy of this treatment in children with refractory
epilepsies (Neal et al., 2008). This chapter reviews
diet treatment in epilepsies where its use is well
established—refractory nonsurgical epilepsies and
epileptic encephalopathies:  Lennox Gastaut syn-
drome, infantile spasms, myoclonic astatic epi-
lepsy (Doose syndrome), and severe myoclonic
epilepsy of infancy (Dravet syndrome).
R E F R A C T O RY
NONSURGICAL EPILEPSY
This broad group, defined by nonsuitability for
a potentially curative surgery, includes lesional
and nonlesional, focal, multifocal, and secondary
generalized epilepsies. Etiologies include remote
symptomatic injuries (pre-, peri- and postnatal
brain injuries of hypoxic, traumatic, infectious,
hemorrhagic origin), developmental brain mal-
formations, genetic/metabolic disorders causing
epilepsy as their primary problem, or resulting in
symptomatic epilepsy, primary and secondary epi-
leptic encephalopathies and progressive/degenera-
tive epilepsies. In this era of molecular specificity,
it seems counterintuitive to dwell on such a het-
erogeneous group. And indeed, increasing surgi-
cal prowess and advancing ability to define specific
genetic etiologies and offer individualized molecu-
larly based therapies will continue to “chip away”
at this broad group. However it is in precisely such
a heterogeneous group that the ketogenic diet’s
efficacy was first described. And, in epilepsy prac-
tice, patients presenting with refractory epilepsy
constitute exactly this mixed group of patients.
In treatment of epilepsy, serially trialed anti-
epileptic drugs (AEDs), after initial failure, are
known to have diminishing likelihood of achiev-
ing seizure freedom, with only 14% achieving
seizure freedom with a second or third agent
and only 3% seizure-free with a combination of
two drugs (Kwan and Brodie, 2000). Alternative
approaches, such as surgery for patients with
amenable lesions, and neurostimulation and/or
diet manipulation for others, are available in this
setting. “Drug resistant epilepsy may be defined
as failure of adequate trials of two tolerated and
appropriately chosen and used AED schedules
(whether as monotherapies or in combination) to
achieve sustained seizure freedom” (Kwan et  al.,
2010). Once epilepsy is identified as refractory,
nonpharmacologic approaches are appropriately
considered. Among the options, resective epilepsy
surgery offers the possibility of “cure” of epilepsy,
in certain cases the best chance of seizure-free,
and even medication-free survival. Consideration
of surgical candidacy is therefore important at this
point. Reasons for ineligibility for epilepsy surgery
can be manifold—inability to lateralize or localize
seizure focus, presence of multiple foci, high risk
41
Chapter 6: Ketogenic Diet in Established Epilepsy Indications
of injury to eloquent cortex, patient/family prefer-
ence. In this setting, as well as in those for whom
epilepsy surgery has already failed, dietary manip-
ulation is an important treatment option.
For many years, evidence of diet efficacy was
in the realm of “expert opinion,” within a nar-
row community of experts familiar with the
diet. This took the form of retrospective reviews
(Kinsman, 1992; Prasad et al., 1996; Hemingway,
2001) describing diet experience in 58 of the most
severely affected patients over an 18- to 24-month
follow-up period. Studies of this type included eti-
ologies and seizures of all kinds in children of a
broad range of ages, often with very high seizure
frequency. Reported efficacy was surprisingly
good, comparable to industry standards for effi-
cacy of new drugs today. Retrospective nature and
lack of blinding and control groups limited the
interpretability of these studies. Following media
exposure in 1994 in a Dateline NBC review of the
diet and in 1997 after release of the film First Do
No Harm, there was a tidal wave of interest in the
diet among parents of children with refractory
epilepsy, pressuring their providers to provide,
or refer them for a trial of, diet treatment. In this
setting the first prospective studies emerged—a
multicenter study (Vining et al., 1998) and a single
institution report (Freeman et  al., 1998), again
describing diet efficacy in patients with an aver-
age monthly seizure frequency of 230/month and
410/month respectively. The multicenter study in
particular assured the community of pediatric epi-
leptologists that the efficacy of the Johns Hopkins
protocols could be reproduced in other hands. At
this point, in 2000, a review of 11 reports on diet
efficacy, including the two cited prospective stud-
ies, cosponsored by the BlueCross and Blue Shield
association, stated, “it is unlikely that this degree of
benefit can result from a placebo response and/or
spontaneous remission” and therefore concluded,
“the evidence is sufficient to determine that the
ketogenic diet is efficacious in reducing seizure
frequency in children with refractory epilepsy”
(Lefevre and Aronson, 2000). Finally, in 2008,
a randomized, controlled but unblinded study
of diet in a mixed group of children with refrac-
tory epilepsy revealed a statistically significant
improvement in seizure control in a group of chil-
dren treated with ketogenic diet now, compared
with a matched group treated with ketogenic diet
3 months later, improving the quality of evidence
in support of diet efficacy (Neal et al., 2008).
These studies have the great benefit of “gener-
alizability” to the real world of refractory epilepsy.
As might therefore be expected, subsequent studies
41
from around the world in similar case series have
reported roughly similar efficacy, which can be
broadly stated as ~50% of refractory epilepsy
patients starting the diet can expect to be respond-
ers (i.e. have >50% seizure reduction in seizures),
25%–30% will experience >90% seizure reduc-
tion, of whom approximately half will be seizure-
free (Henderson et  al., 2006; Nathan et  al., 2009;
Hallbook et  al., 2015; Kossoff and McGrogan,
2005; Kossoff et al., 2012).
Diet therapy in epilepsy is rigorous, requiring
precision in preparation and careful attention in
administration by caregivers. Formula-fed infants
and tube-fed infants are easiest to initiate and
maintain on the diet, and usually use the classic
ketogenic diet at a ratio sufficient to induce keto-
sis with plasma betahydroxybutyrate ≥ 4 mmol/L
or seizure freedom, whichever comes first. These
children are not troubled by dietary restriction and
poor palatability as older, orally fed children are,
who are are accustomed to making dietary choices
and may refuse unpalatable and unpreferred foods
and drink. Reduced GI motility may worsen
esophageal reflux and constipation in all, result-
ing in discomfort. Diet therapy is not without a
variety of systemic adverse effects occurring at low
frequency (Kang et al., 2004; Kossoff et al., 2009)
but at least comparable in range and significance
to those associated with anticonvulsant medica-
tions, although a significant advantage is the lack
of sedating and/or behavioral adverse effects.
It would be extremely helpful to be able to
predict which patients are most likely to respond
to diet therapy in advance, instead of having to
anticipate ~50% failure to achieve a useful seizure
reduction. Unfortunately, though many studies
and reports attempt to identify predictors, small
study size, lack of control groups, heterogeneous
diagnostic groups, and rarity of specific epilepsy
syndromes have all limited the ability to reliably
identify these factors. Schoeler et al. recently pub-
lished an extensive review of the articles provid-
ing data on effectiveness and putative predictive
factors (Schoeler et al., 2013). Twenty-one factors
were examined. For each factor, the predomi-
nance of evidence in favor of an effect, the relative
strength of evidence against an effect of that factor,
and the number of patients in the cohorts “for” and
“against” that factor were considered. No factor had
strong evidence for a positive or negative response
to diet. Strong evidence for absence of effect is also
lacking, except in the case of gender and intel-
lectual ability, which appear to have no effect on
diet response. There were “mixed” findings (effect
on response is reported in approximately half of
42
42 section I: Ketogenic Diet for Epilepsy in the Clinic
reported cases) for epilepsy cause/syndrome, sei-
zure type, and biochemical markers other than
ketosis and plasma glucose. “Weak evidence” for
an effect on diet response (effect reported in less
than half of reported cases) exists for age of sei-
zure onset, age at diet initiation, time from seizure
onset to diet initiation, seizure frequency, diet
type, ratio, levels of ketosis (betahydroxybutyr-
ate, acetone), EEG parameters, AEDs (individual
or number/combinations), and BMI. Presence or
absence of an effect of blood glucose, genetics, and
imaging findings could not be assessed because of
a limited number of studies reporting and patients
reported. Assessment of predictive factors in this
way is complicated by interactions between factors
that cannot be resolved in small, uncontrolled,
retrospective studies. Hopefully the future will
include the development of consortia of expert
centers devoted to analysis of greater numbers of
patients managed prospectively and in a standard
manner, which will allow dissection of factors pre-
dictive of successful treatment. Better understand-
ing may also illuminate potential mechanisms for
further study and manipulation.
Patients started on KD may have a seizure
response within 14  days, but most will con-
tinue a trial of diet for at least 3 months, assum-
ing no exacerbation of seizures occurs beyond
the initial initiation period (Kossoff et  al., 2009).
Discontinuation after successful treatment is usu-
ally attempted after 2 years, though there is no data
determining this to be the optimal time. Rarely,
diet initiation is associated with persistent exacer-
bation of seizures beyond the initiation itself. Diet
withdrawal is appropriate in this setting. Early diet
discontinuation (<3 months) is usually due to lack
of efficacy. Growing intolerance of dietary restric-
tion can be problematic in young patients able to
make diet choices.
Barriers to provision of ketogenic diets con-
tinue to include access to clinical expertise in
diet treatment, cost of higher grade protein and
high-fat foods in some communities, individual
feeding/dietary preferences, co-morbid medical
complexity, and systemic fragility raising con-
cern for ability to tolerate the stress of dietary
conversion.
In summary, for patients falling into the het-
erogeneous category of “nonsurgical epilepsy,”
dietary therapy with a ketogenic diet should be
considered once drug resistance and ineligibil-
ity for surgery has been established. In general,
patients can expect a better chance of efficacy
than with trial of another AED (Kwan and Brodie,
2000).
Many conditions have been recently estab-
lished for which the ketogenic diet can be particu-
larly beneficial. Several are epilepsy syndromes
with refractory generalized seizures. This chap-
ter covers the most common four conditions, for
which there are sufficient data to recommend the
ketogenic diet as potentially very helpful.
I N FA N T I L E   S PA S M S
Infantile spasms (IS) are an age-dependent sei-
zure type, typically occurring between 6 and
18  months of age, which are clinically charac-
terized by serial body jerks occurring in clus-
ters, often at the interfaces of waking and sleep.
They represent the most common type of epi-
leptic spasm. In infancy, they are often, but not
always, associated with a characteristic EEG pat-
tern known as hypsarrhythmia, a chaotic waking
pattern of asynchronous high-voltage irregular
slowing with superimposed multifocal spikes,
which are more synchronous and discontinu-
ous in sleep. When spasms are associated with
developmental regression and hypsarrhythmia
this triad constitutes West syndrome. Infantile
spasms can be idiopathic or associated with
genetic disorders, brain malformations, or pre-
existing brain injuries. ACTH and vigabatrin are
first-line treatments for this epileptic encepha-
lopathy, which is refractory to initial therapies in
30%–40% overall (Pellock et al., 2010; Lux et al.,
2005). Refractory IS are associated with serious
delay in development in most survivors. Early
and effective treatment of this seizure type is con-
sidered the best chance for normal developmen-
tal outcome. With first-line treatment, ACTH
in most cases, elimination of clinical spasms by
14 days of treatment is the marker of successful
treatment. Resolution of hypsarrhythmia is also
considered a marker of successful treatment of
IS and is usually assessed first at 14  days, with
evaluation for maintenance of improvement in a
further 28 days.
In comparison with other epilepsies, assess-
ment of efficacy of any treatment for infantile
spasms should consider the seizure-free rate to
be the first desirable seizure outcome, with EEG
resolution of underlying hypsarrhythmia of near
equal importance. Complicating the assessment
of efficacy of any treatment of infantile spasms is
the known occurrence of spontaneous remission
of infantile spasms in untreated cases, which
can occur as early as 1 month after onset, and
cumulatively in 10%–15% at 6 months and up to
25% of patients at 1 year (Hrachovy et al., 1991).
Notably, in this retrospective cohort of untreated
43
Chapter 6: Ketogenic Diet in Established Epilepsy Indications
infants ~90% suffered moderate to severe devel-
opmental impairment at follow-up, an average of
80 months later. There is also a significant rate
of relapse of spasms during treatment with first-
line agents, which therefore should also be con-
sidered in assessing dietary treatments for this
condition.
The ketogenic diet is among the treatment
options considered after failure of the first-line
treatments, or if their use is contraindicated for
any reason. In a study of ketogenic diet in infants
by Nordli, 17 of the 32 infants with refractory epi-
lepsy had infantile spasms. Of the 32 infants, 6
achieved seizure freedom all of whom had infantile
spasms, (Nordli et al., 2001). Another 6 patients
with spasms had “worthwhile improvement,” and
as a group, patients with infantile spasms were
more responsive to ketogenic diet than infants
with other seizure types in this study. Following
on this finding, 23 children with IS were studied,
of whom 18 remained on diet at 6 months, 13 of
whom had ≥50% reduction in seizures (3 seizure-
free), and 13/13 remaining on diet at 12 months
had ≥50% reduction (3 seizure-free) (Kossoff et
al., 2002). In 2010, 104 infants with IS at the same
institution, treated with KD after exposure to a
mean of 3.6 AEDs, including steroids or vigaba-
trin in >70% were reported (Hong et al., 2010).
Overall, 64% had a ≥50% reduction in seizures,
and 38 achieved at least 6 months spasm free after
a median of 2.4 months on the diet. Of these, 30
(79%) maintained spasm freedom. In a prospec-
tive case study of 20 patients with epileptic spasms,
among 70% and 72% achieving a >50% reduction
in seizures at 3 and 6 months respectively, 3 infants
achieved and maintained seizure freedom for at
least 6 months (Kayyali et al., 2014). Improvement
in spasms was noted within 1 month of start-
ing diet treatment. EEG improved in 12 (60% of
patients) when assessed at 3–6 months, and in
five of six infants with hypsarrhythmia the pat-
tern resolved and did not recur up to 12 months
after diet initiation. Pires et al. reported 6/17
(35%) seizure-free at 1 month after failing vigaba-
trin and hydrocortisone (Pires et al., 2013). Eleven
of these 17 patients (65%) were seizure-free at 3
months, one after the addition of felbamate. The
addition of felbamate to their regimen brought five
more into the responder (>50% reduction) group.
Dressler, in a retrospective, cross-sectional study
comparing early (<1.5 yrs) with late diet initiation,
included 59 infants with IS in their population of
115 infants. Nineteen (32%) were seizure-free at
3 months (Dressler et al., 2015a). Numis et al. had
9/26 (32%) seizure-free at 1–3 months. They noted
43
usefully an association of spasm response and
“other seizure” response in these patients (p = .02)
(Numis et al., 2011).
Factors favoring diet therapy in IS include
easily available ketogenic formula and lack of
independent choice of diet intake in infants.
Nonetheless, three of 20 caregivers discontinued
the diet; of these one indicated difficulty maintain-
ing the protocol (Kayyali et al., 2014). In an Asian
population, 9 of 43 patients discontinued the diet
due to “unacceptable” complications, and 7 due to
intolerability. An additional 9 (21%) discontinued
due to insufficient efficacy (Eun et  al., 2006). In
contrast, 6/26 (23%) had side effects (Numis et al.,
2011), none serious enough to discontinue the
diet. In this group, 5/26 discontinued the diet by
~1 year due to either lack of efficacy or difficulty
maintaining the diet.
In these young patients, concerns regarding
(Hong et  al., 2010), weight gain and appropri-
ate growth were amenable to adjustment down-
ward in ratio for increased protein intake. Numis
et al. described no significant differences between
responders and nonresponders in terms of weight-
for-height z scores, although the best responders
(>90% seizure reduction) had lower weight-for-
height z scores at all points.
Among the 104 patients described by Hong
et  al., 10/18 in whom diet was used as the first-
line treatment became seizure-free, had normal
EEG within 2  months, and were maintained on
the diet for 6  months, after which the diet was
withdrawn without relapse (Hong et  al., 2010).
The same group reported a retrospective case-
controlled study of diet versus ACTH as initial
therapy for new-onset spasms (Kossoff et  al.,
2008). Eight of 13 (62%) patients on diet therapy
became seizure-free at a median of 6.5  days of
treatment, compared with 18/20 (90%) treated
with high dose ACTH who were spasm free at a
median of 4 days (p = .06). One of the 8 relapsed
(12.5%), versus 6 (33%) of the 18 ACTH respond-
ers (p = .23). Also, EEG normalization was more
likely in the ACTH group (1/11 vs. 9/17, p = .02)
at 1  month, though the EEG normalized in all
eight responders by 5 months of diet treatment.
Adverse effects were significantly less likely in the
diet therapy group (p = .006). There was no dif-
ference in developmental outcome between the
treatment groups at their last examination at a
median of 12 months.
Predicting efficacy of diet has been difficult.
Eun et al. associated underlying etiology with effi-
cacy, cryptogenic cases responding at a higher rate
than symptomatic cases, but no other predictor
44
44 section I: Ketogenic Diet for Epilepsy in the Clinic
emerged from their series (Eun et  al., 2006). No
significant predictive factor of diet efficacy was
identified in the small diet group in the case con-
trol comparison with ACTH. Numis et al. reported
11/18 males and 1/8 females were responders
(p  =  0.04) but cited low sample size and skewed
distribution of etiologies as potential biases
(Numis et al., 2011). This has not been reported in
other studies.
A final area of interest is the possibility of
synergistic effects of combination of KD with
other treatments for IS. A recent study of epilep-
tic encephalopathies (EE) (IS, Lennox Gastaut
syndrome/other generalized epilepsies, CSWS)
examined the efficacy of KD added to steroid
therapy when not controlled by the initial ste-
roid treatment (Ville et al., 2015). Patients were
on a mean of 3.5 mg/kg/day of hydrocortisone
(1–10 mg/kg/day) when starting the diet. Ketosis
was achieved in all but one case with status epi-
lepticus (SE), and no diet adjustment was made
to maintain ketonuria. Plasma ketones were not
measured. There were 23 IS cases in the total
of 42 with EE. Twenty were steroid-resistant,
three steroid-dependent. Overall, 5/27 (18.5%)
steroid-resistant cases and 9/15 (60%) steroid-
dependent cases were considered responders.
The data presented do not allow specific analysis
with respect to IS.
In summary, studies of modest quality indi-
cate that KD is a moderately effective therapy for
refractory IS, and is generally safe and tolerable
in this young population. As a first-line treat-
ment, it has better efficacy for seizure freedom
over time, but is not as quickly effective as tradi-
tional first-line treatments, in particular ACTH.
Patients experiencing improvement do so within
1  month, and definitely by 3  months. There may
be reason to examine the KD in combination with
first-line agents rather than serially, with the goal
of maximizing the chance of seizure freedom.
High-quality studies would be best, and given the
low frequency of infantile spasms would require a
multicenter approach.
L E N N O X G A S TAU T
SYNDROME
This is an epilepsy syndrome characterized by
peak onset between 3 and 5 years of age. It is associ-
ated with a very refractory pleomorphic epilepsy,
tonic seizures being most characteristic, and also
including atypical absences and atonic and myo-
clonic seizures (Arzimanoglou et al., 2009). Focal
seizures and complex partial seizures may also
occur. Seizures may be very frequent, occurring
multiple times per day. Episodes of SE, often pro-
longed, are common. Lennox Gastaut syndrome
(LGS) may occur de novo or in about 20% devel-
ops following West syndrome, and might also
follow other remote symptomatic brain injuries
(e.g., hypoxic ischemic encephalopathy). If there
is no preexisting developmental delay, cognitive
regression occurs and at least moderate intellec-
tual disability is expected, with virtually all indi-
viduals being dependent in adulthood. The EEG
is characterized by slow (1.5–2 Hz) spike and
wave activity, often multifocal spikes, and bursts
of 10-Hz fast activity during sleep. Slow spike and
wave activity may wane over the years. Treatment
with broad-spectrum anticonvulsant drugs is
the mainstay of treatment but is rarely effective
in achieving seizure control. Side effects of leth-
argy and drowsiness are particularly damaging,
as these states are associated with increased sei-
zures. Nonpharmacologic treatments including
surgical approaches (corpus callosotomy, vagus
nerve stimulation) and diet therapies offer non-
sedating adjunctive therapies that, if successful,
may allow reduction of medication burden and
a resultant improvement in seizures and quality
of life.
The early reports of efficacy in the most refrac-
tory epilepsies typically included patients with
this disorder, and this subgroup of patients were
found to respond to diet with a >50% reduction
in seizures within 5 days of diet initiation with 36
hours of fasting and the development of ketosis
(Freeman and Vining, 1999). A blinded crossover
study of 20 LGS patients, initiated on diet and
subsequently, in a blinded manner, given either
60 gm glucose with saccharin to abort diet, or sac-
charin only (allowing continued ketosis) failed to
show a significant difference in outcome, but this
was likely due to the failure to eliminate ketosis
in the control group (Freeman et  al., 2009). In
2012, Lemmon et  al. reported that in the litera-
ture to that date 88/189 (47%) LGS patients had
responded to KD (Lemmon et al 2012). They ret-
rospectively reviewed 71 patients from their insti-
tution, John Hopkins Hospital, of whom 36 (51%)
achieved a >50% reduction in seizures, 16 (23%) a
>90% reduction, and one was seizure-free, using
an intention to treat analysis. The results were
similar at 12  months. Caraballo et  al. reported
similar results in a smaller prospectively studied
group of 20 LGS patients started on diet and fol-
lowed for a minimum of 16  months with reten-
tion of 75% of patients on diet, and 2/20 (15%)
seizure-free and 25% with 50%–99% reduction
in seizures (Caraballo et  al., 2014). Five children
45
Chapter 6: Ketogenic Diet in Established Epilepsy Indications
discontinued diet in this study, three because of
lack of efficacy within 3 months, and two because
of persistent vomiting, one of whom was also
hypoglycemic. Patients with LGS have also been
reported to respond with similar efficacy to the
modified Atkins diet (Sharma et al., 2015), 12/25
were responders and 2 (8%) were seizure-free.
In summary, LGS patients, some of the most
refractory pediatric epilepsy patients, respond to
KD therapies. As a result, medication reduction
or withdrawal provides special benefit to these
patients, possibly reducing medications even fur-
ther and improving quality of life. Diet therapy
should be considered early once diagnosis is clear
and refractoriness established.
D R AV E T S Y N D R O M E
( S E V E R E M YO C L O N I C
E P I L E P S Y O F   I N FA N C Y )
Dravet syndrome (severe myoclonic epilepsy of
infancy; DS SMEI) is a clinical syndrome charac-
terized by initial febrile seizure, often febrile SE,
and/or hemiclonic febrile seizures in the first year
of life, in the setting of initially normal develop-
ment. It occurs in 1:20,000–1:40,000 individuals.
Febrile and afebrile seizures are quickly recurrent
and associated with subsequent developmental
stagnation or regression. The epilepsy is pleo-
morphic (multiple seizure types—myoclonic,
focal, generalized absence, and generalized
motor seizures) and pharmacoresistent. Fever/
hyperthermia continues to provoke seizures.
Subsequent to seizure onset, ataxia, pyramidal
signs and interictal myoclonus are seen. Early
EEGs are normal but later, generalized spike
and polyspike waves and multifocal spikes are
seen. Historically, many children thought to
have pertussis vaccine-related encephalopathy
and long-term cognitive sequelae, were found
to have underlying DS (Berkovic et  al., 2006).
Up to 80% are associated with a heterozygous
mutation in the sodium channel gene SCN1A,
although not all SCN1A mutations result in DS.
Other genetic and/or environmental factors play
an as yet unclear role in evolution of the syn-
drome. Development generally stabilizes, but at
least a moderate degree of intellectual disability
is the rule. The temporal relationship between
seizure onset and developmental impairment
suggests but does not prove a causal relationship
between seizures, SE, and developmental regres-
sion. Nonetheless, vigorous efforts to control and
prevent seizures are appropriate. Problems with
this approach include medication side effects.
The gold standard is combination treatment with
45
stiripentol (STP), valproic acid (VPA), and cloba-
zam (CLB). There is a higher mortality in children
with this syndrome (Genton et al., 2011), due to
a variety of factors (sudden unexpected death in
epilepsy, SE, accident) further motivating efforts
to control convulsive seizures. An Scn1A mutant
mouse model of DS showed decreased latency
to seizure onset after an epileptogenic challenge
compared with wild-type littermates. Feeding the
mutant mouse a ketogenic diet increased latency
to seizures to levels that were not significantly
different from the wild-type littermates (Dutton
et al., 2011).
In 2011, Caraballo and colleagues reported
their experience in 59 DS patients, of whom 24
were treated with a 4:1 ketogenic diet and fol-
lowed for at least 2 years (Caraballo 2011). Sixteen
remained on the diet at 2 years, of whom 2 (16%)
were seizure-free, 10 (62.5%) had 75%–99%
reduction in seizures and another 4 (25%) had
50%–74% reduction in seizures. All five with SE
responded to diet and had no additional SE dur-
ing follow-up. Medication reduction, even in
those without dramatic seizure efficacy, resulted in
improvement in quality of life. Ten of 15 patients
followed by Nabbout et al. achieved a >75% reduc-
tion in seizures (Nabbout et al., 2011). All of these
patients were already receiving triple combination
therapy (VPA, STP, CLB). However, efficacy was
lost in ~50% by 1 year. Behavioral improvements
were noted in all responders.
Laux et  al. followed a subgroup of 20/48 DS
patients on KD at their own center, all of whom
remained on diet at least 6  months, 17 continu-
ing for at least 18 months. They saw 65% achieve a
greater than 50% reduction in seizures and 30% a
>90% reduction. They noted that the KD appeared
to reduce the frequency of all seizure types, though
quantification of myoclonic and absence seizures
was not robust (Laux et al., 2013).
In a retrospective cross-sectional study, 10
of 32 mutation positive patients (31 SCN1A and
1 GABRG mutations) with DS were treated with
KD (Dressler et al., 2015b) for at least 3 months.
Seven of 10 patients were responders at 3 months,
6 at 6 months and 12 months. One child became
seizure-free on KD. A  striking finding in this
group was the lack of SE in the eight patients
who had previously suffered this seizure type. No
child on KD, responder or not, had SE while on
the diet. Given that excess mortality in this syn-
drome is partly related to SE, this is a notable
finding that would bear further study. Early with-
drawal from the KD treatment was due to failure
of efficacy; later withdrawals were due to difficulty
46
46 section I: Ketogenic Diet for Epilepsy in the Clinic
maintaining compliance. Responder status was
not associated with mutation type in this group.
An interesting aspect of the Dressler study is
the comparison with other treatments (Dressler
et al., 2015b). There was no significant difference
in responder rates between the gold standard
treatment, a combination of STP, VPA, and CLB
(89%); bromide (78%); KD (70%); VPA mono-
therapy (48%); TPM (topiramate) (35%); and VNS
(vagus nerve stimulation )(37%). Ketogenic diet
was more effective than LEV (levetiracetam ). The
conclusions are limited by small numbers but
appear worthy of further study.
With respect to positive interaction with other
treatments, 7 of 16 diet responders were on topi-
ramate in Caraballo’s updated group, and 3 of 7 of
Dressler’s responders were also.
In summary, treatment with KD in DS is associ-
ated with a reduction in seizures, and possibly in SE.
In this highly pharmacoresistant epilepsy, polyphar-
macy is often the rule, with the attendant burden of
medication side effects. The goal of optimizing sei-
zure control while minimizing medication adverse
effects may be facilitated by introduction of the
ketogenic diet, often allowing reduction in medica-
tion burden and therefore in cognitive and behav-
ioral side effects. An as yet unanswerable question is
whether early use of the ketogenic diet, prior to sei-
zure onset, with its lack of cognitive side effects and
prominent effect on status epilepticus (SE), might
ameliorate the developmental stagnation/regression
associated with seizure onset in this disorder.
DOO SE SYNDROME
( M YO C L O N I C A S TAT I C
EPILEPSY)
Doose syndrome (myoclonic astatic epilepsy;
MAE) is a genetic generalized epilepsy, initially
described by Doose in 1970 and distinguished
from other causes of myoclonic epilepsy in child-
hood (SMEI, LGS), although the underlying
genetic abnormality is not yet known. It occurs
in previously normal children, with onset, some-
times explosively, possibly as early as the latter half
of the first year of life, but more typically between
2 and 5 years of age. There is a high rate of pre-
ceding febrile seizures, and of a family history of
epilepsy. It is characterized by frequent myoclonic
and myoclonic astatic/atonic seizures; absences,
with or without myoclonia; and generalized tonic-
clonic, clonic, and/or tonic seizures. The occur-
rence of nocturnal tonic seizures is thought by
some to indicate a poor outcome. With extremely
frequent seizures, often absences interspersed with
myoclonias, and nonconvulsive SE an epileptic
encephalopathy can develop. Seizure-related falls
are a major component of associated morbid-
ity, and these children are at risk of injury. Many
need to wear helmets for protection. The epilepsy
is often highly pharmacoresistant, and medication
adverse effects add to the impact of the disorder
on cognitive performance and behavior regula-
tion. Often EEG is associated with generalized
spike and polyspike wave activity, ranging from 2
to 5 Hz, which also underlies the myoclonic sei-
zures. Prominent theta activity is a feature of the
interictal EEG. This disorder, among other refrac-
tory childhood epilepsies, is very responsive to the
ketogenic diets. This has led to exploration of the
relationship of this disorder to GLUT1 deficiency,
a hypothesis that would provide an explanation for
its particular responsiveness to ketosis. However,
only 0%–5% of cases have been shown to be asso-
ciated with this genetic disorder (Larsen et al.,
2015; Mullen et al., 2011). Cognitive outcome is
normal in about 60% of patients in the long term.
Continued active seizures are correlated with vary-
ing degrees of intellectual disability, ranging from
mild to moderate severity.
Using ILAE criteria for diagnosis, Oguni et
al. reported treatment outcome in a group of 81
patients with MAE (Oguni et al., 2002). Seizure
freedom for >2 years was achieved in 55/81 (65%)
by 50 months (+/– 16 mo). Later recurrence of
GTCs occurred in 14% up to 18 years later but
were then easily controlled. Eighteen percent con-
tinued to have refractory epilepsy. Among the
treatments used in this group, KD was used for 26
patients, of whom 15 (58%) became free of myo-
clonic (MS) and atonic (AS) seizures and 9 more
had >50% reduction in these seizures. This efficacy
was better than that of ACTH (n = 22, 36% MS/
AS free, 23% >50% decrease) and ethosuximide
(n = 32, 32% MS/AS free, 32% >50% decrease),
the best of the conventional AEDs used, and than
valproic acid (n = 57, 12% MS/AS free, 28% >50%
decrease). Generalized motor seizures gradually
decreased thereafter in these patients.
Caraballo et  al. reported 30 MAE patients
of whom 11 were treated with diet (Caraballo
et al., 2006). Six of these remained on the diet at
18  months of whom two (18.2%) were seizure-
free, and two each had 75%–99% reduction and
50%–74% reduction. The EEG was improved in
all six, and was normal in the seizure-free patients
at 18 months. Five patients discontinued the diet
within the first 3 months, four for lack of efficacy
and one for persistent vomiting. Those remain-
ing on the diet tolerated it without significant side
effects. The authors remarked the good response
47
Chapter 6: Ketogenic Diet in Established Epilepsy Indications
to the diet and considered that for this disorder, it
should be considered earlier in the course.
Kilaru et al. reported the order of exposure to
various treatments and their efficacy in 23 MAE
patients (Kilaru et  al., 2007). Ketogenic diet was
tried last in 10 patients, and was the most effec-
tive treatment in achieving seizure remission.
It was initiated an average of 17  months (range
2–58 months) after onset, and after an average of
five AED trials. Five (50%) became seizure-free,
three within 1  month, one by 7  months, and the
last by 19  months after diet initiation. Three of
these patients remained seizure-free for more than
6  months. Ethosuximide (25%) and topiramate
(23%) were the next most effective in achieving
remission of seizures. In this group, steroids (pred-
nisone or ACTH) were ineffective. The authors
report the possibility of spontaneous remission in
3 of 14 patients who became seizure-free in this
group. When this is a factor, late-used treatments
may appear more effective than those used early in
the disorder, when spontaneous remission is less
likely. Only 43% of patients in this group were cog-
nitively normal at last follow-up, the rest having
mostly mild disability. Cognitive outcome was not
related to time to seizure freedom.
Perhaps because this epilepsy seems especially
responsive to the ketogenic diet, use of a lower ratio,
as in the modified Atkins diet has been considered
as an option for treatment—with the potential of
seeing efficacy with a less rigorous, more palatable
diet. There is little information available about effi-
cacy in comparison to the classic ketogenic diet.
Simard-Tremblay used the modified Atkins diet in
six and KD in three of nine patients after a median
of four AEDs (Simard-Tremblay et al., 2015). They
reported seven of nine patients achieved seizure
remission, with one patient on KD and one on
the modified Atkins diet failing to respond. One
patient advanced from the modified Atkins diet to
4:1 KD and achieved seizure freedom. When 27
patients were switched from the modified Atkins
diet to KD there was improvement (additional
10% decrease in seizures) in 10 (37%) (Kossoff
et al., 2010). There were nine MAE patients in this
group, of whom seven improved with the change
to KD, five of them achieving seizure freedom.
In summary, MAE appears to be more respon-
sive to ketogenic diets than other epilepsies, with
approximately 50% achieving seizure freedom,
allowing medication withdrawal in many and ulti-
mately diet withdrawal also. Assessment of efficacy
is complicated by a possible but poorly defined
spontaneous remission rate. Diet therapy is being
offered as an earlier option more frequently in
47
light of the above reports. It remains to be seen if
efficacy remains high in this setting.
CONCLUSION
In conclusion, there is general agreement, based
on the literature described in this chapter, that
patients with these four epilepsies (infantile
spasms, Lennox-Gastaut syndrome, Dravet syn-
drome, and myoclonic-astatic epilepsy) may ben-
efit from a trial of diet therapy once their epilepsy
has become refractory to medication. The quality
of the data for individual epilepsy syndromes is
modest, at best. These diagnoses are rare, so that
only multicenter studies are likely to produce more
robust data, comparing diet therapy to other treat-
ments in a controlled manner, and/or establish-
ing the best timing of diet trials. There is also very
little information on the efficacy of diet treatment
as first-line therapy. The difficulty of administra-
tion of diet therapy is likely to limit its acceptance
in this position unless it becomes possible to
induce ketosis without diet restrictions, that is, via
a supplement or pill. Future, multicenter collab-
orative studies will be necessary to examine these
possibilities.
REFERENCES
Arzimanoglou, A., French, J., Blume, W.T., Cross,
J.H., Ernst, J.P., Feucht, M., Genton, P., Guerrini,
R., Kluger, G., Pellock, J.M., et al. (2009). Lennox-
Gastaut syndrome:  a consensus approach on
diagnosis, assessment, management, and trial
methodology. Lancet Neurol 8, 82–93.
Berkovic, S.F., Harkin, L., McMahon, J.M., Pelekanos,
J.T., Zuberi, S.M., Wirrell, E.C., Gill, D.S., Iona,
X., Mulley J.C., and Scheffer, I.E. (2006). De-novo
mutations of the sodium channel gene SCN1A in
alleged vaccine encephalopathy:  a retrospective
study. Lancet Neurol 5, 488–492.
Caraballo, R.H., Cersósimo, R.O., Sakr, D., Cresta, A.,
Escoba, N., and Fejerman, N. (2006). Ketogenic
diet in patients with myoclonic-astatic epilepsy.
Epileptic Disord 8, 151–155.
Caraballo, R.H. (2011). Nonpharmacologic treatments
of Dravet syndrome: focus on the ketogenic diet.
Epilepsia 52(Suppl. 2), 79–82.
Caraballo, R.H., Fortini, S., Fresler, S., Armeno, M.,
Ariela, A., Cresta, A., Mestre, G., and Escobal, N.
(2014). Ketogenic diet in patients with Lennox–
Gastaut syndrome. Seizure 23, 751–755.
Doose, H., Gerken, H., Leonhardt, R., Völzke, E.,
and Völz, C. (1970). Centrencephalic myoclonic-
astatic petit mal: clinical and genetic investigation.
Neuropadiatrie 2, 59–78.
Dressler, A., Trimmel-Schwahofer, P., Reithofer, E.,
Gröppel, G., Mühlebner, A., Samueli S., Grabner,
V., Abraham, K., Benninger, F., and Feucht, M.
48
48 section I: Ketogenic Diet for Epilepsy in the Clinic
(2015a). The ketogenic diet in infants—Advantages
of early use. Epilepsy Res. 116, 53–58.
Dressler, A., Trimmel-Schwahofer, P., Reithofer,
E., Mühlebner A., Gröppel, G., Reiter-Fink, R.,
Benninger, F., Grassl, R., and Feucht, M. (2015b).
Efficacy and tolerability of the ketogenic diet in
Dravet syndrome: comparison with various stan-
dard antiepileptic drug regimen. Epilepsy Res
109, 81–89.
Dutton, S.B., Sawyer, N.T., Kalume, F., Jumbo-Lucioni,
P., Borges, K., Catterall, W.A., and Escayg, A.
(2011). Protective effect of the ketogenic diet in
Scn1a mutant mice. Epilepsia 52, 2050–2056.
Eun, S.H., Kang, H.C., Kim D.W., and Kim, H.D.
(2006). Ketogenic diet for treatment of infantile
spasms. Brain Dev 28, 566–571.
Freeman, J.M., Vining, E.P., Pillas, D.J., Pyzik, P.L.,
Casey, J.C., and Kelly, L.M. (1998). The efficacy
of the ketogenic diet:  a prospective evaluation
of intervention in 150 children. Pediatrics 102,
1358–1363.
Freeman, J.M., and Vining, E.P. (1999). Seizures
decrease rapidly after fasting: preliminary studies
of the ketogenic diet. Arch Pediatr Adolesc Med
153, 946–949.
Freeman, J.M., Vining, E.P., Kossoff, E.H., Pyzik, P.L.,
Ye, X., and Goodman, S.N. (2009). A blinded,
crossover study of the efficacy of the ketogenic
diet. Epilepsia 50, 322–325.
Genton, P., Velizarova, R., and Dravet, C. (2011).
Dravet syndrome:  the long-term outcome.
Epilepsia 52(Suppl 2), 44–49.
Hallbook, T., Sjolander, A., Armak, P., Miranda, M.,
Bjurulf, B., and Dahlin, M. (2015). Effectiveness
of the ketogenic diet used to treat resistant child-
hood epilepsy in Scandinavia. Eur J Paed Neurol
19, 29–36
Hemingway, C., Freeman, J.M., Ribbs, D.J., and Pyzik,
P.L. (2001). The ketogenic diet: a 3-6 year follow up
of 150 children prospectively enrolled. Pediatrics
108, 898–905.
Henderson, C.B., Filloux, F.M., Alder, S.C., Lyon, J.L.,
and Caplin, D.A. (2006). Efficacy of the ketogenic
diet as a treatment option for epilepsy:  meta-
analysis. J Child Neurol 21, 193–198.
Hong, A.M., Turner, Z, Hamdy, R.F., and Kossoff, E.H.
(2010). Infantile spasms treated with the ketogenic
diet:  prospective single-center experience in 104
consecutive infants. Epilepsia 51, 1403–1407.
Hrachovy, R.A., Glaze, D.J., Frost, J.D., Jr. (1991). A
retrospective study of spontaneous remission
and long-term outcome in patients with infantile
spasms. Epilepsia 32, 212–214.
Kang, H.C., Cheung, D.E., Kim, D.W., and Kim, H.D.
(2004). Early and late onset complications in the
ketogenic diet for intractable epilepsy. Epilepsia
45, 1116–1123.
Kayyali, H.R., Gustafson, M., Myers, T., Thompson,
L., Williams, M., and Abdelmoity, A. (2014).
Ketogenic diet efficacy in the treatment of intrac-
table epileptic spasms. Ped Neurol 50, 224–227.
Kilaru, S., and Bergqvist, A.G.C. (2007). Current treat-
ment of myoclonic astatic epilepsy: clinical expe-
rience at the Children’s Hospital of Philadelphia.
Epilepsia 48, 1703–1707.
Kinsman, S.L., Vining, E.P., Quaskey, S.A., Mellits, D.,
and Freeman, J.M. (1992). Efficacy of the keto-
genic diet for intractable seizure disorders: review
of 58 cases. Epilepsia 33, 1132–1136.
Kossoff, E.H., Pyzik, P.L., McGrogan, J.R., Vining, E.P.,
and Freeman, J.M. (2002). Efficacy of the keto-
genic diet for infantile spasms. Pediatrics 109,
780–783.
Kossoff, E.H., and McGrogan, J.R. (2005). Worldwide
use of the ketogenic diet. Epilepsia 46, 280–289.
Kossoff, E.H., Hedderick E.F., Turner, Z., and Freeman,
J.M. (2008). A case-control evaluation of the keto-
genic diet versus ACTH for new-onset infantile
spasms. Epilepsia 49, 1504–1509.
Kossoff, E.H., Zupec-Kania, B.A., Amark, P.E.,
Ballaban-Gil, K.R., Bergqvist, C.A.G., Blackford,
R., Buchhalter, J.R., Caraballo, R.H., Cross,
H.J., Dahlin, M.G., et  al. (2009). Optimal clini-
cal management of children receiving the keto-
genic diet: recommendations of the International
Ketogenic Diet Study Group. Epilepsia 50,
304–317.
Kossoff, E.H., Bosarge, J.L., Miranda, M.J., Wiemer-
Kruel, A., Kang, H.C., and Kim, H.D. (2010). Will
seizure control improve by switching from the
modified Atkins diet to the traditional ketogenic
diet? Epilepsia 51, 2496–2499.
Kossoff, E.H., Caraballo, R.H., du Toit, T., Kim, H.D.,
MacKay, M.T., Nathan, J.K., and Philip, S.G.
(2012). Dietary therapy:  a world wide phenom-
enon. Epilepsy Res 100, 205–209.
Kwan, P., Brodie, M.J. (2000). Early identification of
refractory epilepsy. N Engl J Med 342, 314–319.
Kwan, P., Arzimanoglou, A., Berg, A.T., Brodie, M.J.,
Hauser, A.W., Mathern, G., Moshe, S.L., Perucca,
E., Wiebe, S., and French, J. (2010). Definition of
drug-resistent epilepsy:  consensus proposed by
the ad hoc task force of the ILAE Commission on
Therapeutic Strategies. Epilepsia 52, 1069–1077.
Larsen, J., Johannesen, K.M., Ek, J., Tang, S., Marini,
C., Blichfeldt, S., Kibæk, M., von Spiczak, S.,
Weckhuysen, S., Frangu, M., et al. (2015). The role
of SLC2A1 mutations in myoclonic astatic epilepsy
and absence epilepsy, and the estimated frequency
of GLUT1 deficiency syndrome. Epilepsia *, 1–6.
Laux, L., and Blackford, R. (2013). The ketogenic diet
in Dravet syndrome. J Child Neurol 28, 1041–1044.
Lefevre, F., and Aronson, N. (2000). Ketogenic diet for
the treatment of refractory epilepsy in children: a
systematic review of efficacy. Pediatrics 105, E46.
Lemmon, M.E., Terao, N.N., Ng, Y-T., Reisig, W.,
Rubenstein, J.E., and Kossoff, E.H. (2012).
Efficacy of the ketogenic diet in Lennox–Gastaut
49
Chapter 6: Ketogenic Diet in Established Epilepsy Indications
syndrome:  a retrospective review of one institu-
tion’s experience and summary of the literature.
Dev Med Child Neurol 54, 464–468.
Lux, A.L., Edwards, S.W., Hancock, E., Johnson,
A.L., Kennedy, C.R., Newton, R.W., O’Callaghan,
F.J.K., Verity, C.M., Osborne, J.P., et al. (2005). The
United Kingdom Infantile Spasms Study (UKISS)
comparing hormone treatment with vigabatrin
on developmental and epilepsy outcomes to age
14 months: a multicentre randomised trial. Lancet
Neurol 4, 712–717.
Mullen, S.A., Marini, C., Suls, A., Mei, D., Della
Giustina, E., Buti, D., Arsov, T., Damiano, J.,
Lawrence, K., De Jonghe, P., et al. (2011). Glucose
transporter 1 deficiency as a treatable cause of myo-
clonic astatic epilepsy. Arch Neurol 68, 1152–1155.
Nabbout, R., Copioli, C., Chipaux, M., Chemaly, N.,
Desguerre, I., Dulac, O., and Chiron, C. (2011).
Ketogenic diet  also benefits Dravet syndrome
patients receiving stiripentol:  a prospective pilot
study. Epilepsia 52, e54–e57.
Nathan, J.K., Purandare, A.S., Parek, Z.B., and
Manohar, H.V. (2009). Ketogenic diet in Indian
children with uncontrolled epilepsy. Indian
Pediatrics 46, 669–673.
Neal, E.G., Chaffe, H., Schwartz, R.H., Lawson, M.S.,
Edwards, N., Fitzsimmons, G., Whitney, A., and
Cross, J.H. (2008). The ketogenic diet for the treat-
ment of childhood epilepsy:  a randomized con-
trolled trial. Lancet Neurol 7, 500–506.
Nordli, D.R., Jr., Kuroda, M.M., Carroll, J.,
Koenigsberger, D.Y., Hirsch, L.J., Bruneri, H.J.,
Seidel, W.T., and De Vivo, D.C. (2001). Experience
with the ketogenic diet in infants. Pediatrics 108,
129–133.
Numis, A.L., Yellen, M.B., Chu-Shore, C.J., Pfeifer,
H.H., and Thiele, E.A. (2011). The relationship of
ketosis and growth to the efficacy of the ketogenic
diet in infantile spasms. Epilepsy Res 96, 172–175.
Oguni, H., Tanaka, T., Hayashi, K., Funatsuka, M.,
Sakauchi, M., Shirakawa, S., and Osawa, M.
(2002). Treatment and long-term prognosis of
49
myoclonic-astatic epilepsy of early childhood.
Neuropediatrics 33, 122–132.
Pellock, J.M., Hrachovy, R., Shinnar, S., Baram, T.Z.,
Bettis, D., Dlugos, D., Gaillard, W.D., Gibson, P.A.,
Holmes, G.L., Nordli, D.R., et al. (2010). Infantile
spasms:  a U.S.  consensus report. Epilepsia 51,
2175–2189.
Pires, M.E., Ilea, A., Bourel, E., Bellavoinea, V.,
Merdariua, D., Berquin P., and Auvin, S. (2013).
Ketogenic diet for infantile spasms refractory to
first-line treatments:  AN open prospective study.
Epilepsy Res 105, 189–194.
Prasad, A.N., Stafstrom, C.F., Holmes, GL. Alternative
epilepsy therapies:  the ketogenic diet, immuno-
globulins, and steroids. Epilepsia 19966666;37
Suppl:S81–95.
Schoeler, N.E., Cross, J.H., Sander, J.W., and Sisodiya,
S.M. (2013). Can we predict a favourable response
to ketogenic diet therapies for drug-resistant epi-
lepsy? Epilepsy Res 106, 1–16.
Sharma, S., Jain, P., Gulati, S., Sankhyan, N., and
Agarwala, A. (2015). Use of the modified Atkins
diet in Lennox Gastaut syndrome. J Child Neurol
30, 576–579.
Simard-Tremblay, S., Berry, P., Owens, A., Cook, W.B.,
Sittner, H.R., Mazzanti, M., Huber, J., Warner,
M., Shurtleff, H., and Saneto, R.P. (2015). High-
fat diets and seizure control in myoclonic-astatic
epilepsy:  a single center’s experience. Seizure 25,
184–186.
Ville, D., Chiron C., Laschet, J., and Dulac, O. (2015).
The ketogenic diet can be used successfully in
combination with corticosteroids for epileptic
encephalopathies. Epilepsy Behav 48, 61–65.
Vining, E.P, Freeman, J.M., Ballaban-Gil, K., Camfield,
C.S., Camfield, P.R., Holmes, G.L., Shinnar, S.,
Shuman, R., Trevathan, E., and Wheless, J.W.
(1998). A multicenter study of the efficacy of the
ketogenic diet. Arch Neurol 55, 1433–1437.
Wheless, J.W. (2014). History and origin of the keto-
genic diet. In Epilepsy and the Ketogenic Diet
(Totowa NJ: Humana Press), chapter 2.
50
7
Ketogenic Diet for Other Epilepsies
D AV I D T. H S I E H , M D A N D E L I Z A B E T H A . T H I E L E , M D , P H D
INTRODUCTION
As discussed in previous chapters, the ketogenic
diet (KD) is the treatment of choice for epilepsy
in certain disorders of brain metabolism, in par-
ticular glucose transporter protein 1 (Glut1) defi-
ciency and pyruvate dehydrogenase deficiency
(PDHD). An understanding of the underlying
metabolic pathophysiology of these genetic disor-
ders has been coupled with successful clinical effi-
cacy of the KD in patients with Glut1 deficiency
or PDHD. However, there are also several other
genetic disorders that manifest with epilepsy for
which dietary therapies appear to be especially
effective, but for which the possible underlying
metabolic mechanisms are not as clearly delin-
eated. The International Ketogenic Diet Study
Group has listed several conditions for which
the KD has been reported as being particularly
beneficial (Box 7.1) and could be offered ear-
lier. This list includes Doose syndrome, tuberous
sclerosis complex (TSC), Rett syndrome (RTT),
Dravet syndrome, and infantile spasms (Kossoff
et  al., 2009b). Whether efficacy in these condi-
tions is due in part to the broad-spectrum effi-
cacy of the KD or to specific mechanisms specific
to these conditions is still under investigation.
Furthermore, patient populations who are often
fed with gastrostomy tubes, as in RTT, may be
especially good candidates for the KD in liquid
form, due to the relative ease of administration. In
this chapter, we discuss the use of dietary thera-
pies for the treatment of epilepsy in certain genetic
disorders, including RTT and TSC, as listed by the
International Ketogenic Diet Study Group, and
additionally discuss the use of epilepsy dietary
therapies in patients with Angelman syndrome
(AS) and Sturge-Weber syndrome (SWS).
RETT SYNDROME
Rett syndrome is an X-linked dominant neu-
rogenetic disorder characterized clinically by
developmental regression around 12–18  months
of age, specifically with regression of purpose-
ful hand use and spoken language, development
of a gait abnormality, and hand stereotypies
such as hand wringing (Neul et  al., 2010). Rett
syndrome is associated with mutations in the
methyl-CpG-binding protein 2 (MECP2) gene at
Xq28, which are detected in 95%–97% of typi-
cal RTT patients (Neul et  al., 2010). Epilepsy
occurs in about 70% of patients with RTT, with
a median onset of 3–4  years (Nissenkorn et  al.,
2010; Nissenkorn et al., 2015). Seizure semiology
includes all types, with generalized tonic-clonic
occurring more frequently than complex partial
and secondarily generalized seizures (Cardoza
et  al., 2011). The treatment of epilepsy in RTT
usually involves the use of standard anticonvul-
sant medications, but about a third of patients
remain medication-refractory (Pintaudi et  al.,
2010). Patients with RTT are usually not typical
candidates for epilepsy surgery, thus RTT patients
with medication-refractory epilepsy are often
considered for nonmedication treatments, such as
dietary therapy.
Rett syndrome is listed by the International
Ketogenic Diet Study Group as a condition for
which the KD has been reported as having probable
benefit, defined as having at least two publications
in the literature. In a published case series of seven
girls with clinical RTT and medication-refractory
epilepsy, the KD was initiated in and tolerated by
five of the patients (Haas et al., 1986). Of these five
patients, four of them had a 50% or better decrease
in seizures, including one with a 75% decrease and
one with a 100% decrease. All five of the patients
who were able to tolerate the diet also had reported
slight behavioral improvements. In another pub-
lished case of an 8-year-old girl with genetically
confirmed RTT with medication-refractory sei-
zures, a 70% reduction in seizures was achieved on
the KD (Liebhaber et  al., 2003). This patient was
51
Chapter 7: Ketogenic Diet for Other Epilepsies
BOX 7.1
CONDITIONS LISTED BY THE INTERNATIONAL KETOGENIC DIET STUDY
GROUP AS PROBABLE BENEFIT
Glucose transporter protein 1 deficiency
Pyruvate dehydrogenase deficiency
Doose syndrome
Tuberous sclerosis complex
Rett syndrome
Dravet syndrome
Infantile spasms
Children receiving only formula
able to tolerate the diet for over 4 years, and also
had improvement reported in her behavior on the
KD. Additionally, in a 10-year-old girl with geneti-
cally confirmed RTT and medication-refractory
seizures, the KD significantly decreased seizure
frequency, and this patient was also reported to
concurrently became “more communicative” with
her family members (Giampietro et al., 2006).
It is uncertain whether there is a specific under-
lying metabolic process in patients with RTT that
would provide a physiological benefit to treatment
with the KD. Some researchers have noted that
many patients with RTT have serum markers sug-
gestive of defects in energy metabolism, including
serum lactate or pyruvate elevations (Haas et  al.,
1995)  or cerebrospinal fluid lactate elevations
(Matsuishi et al., 2011). It has been hypothesized
that perhaps the KD alters energy metabolism
in a favorable direction in patients with RTT
(Haas et al., 1986; Mantis et al., 2009), but further
research is needed to better understand the impli-
cations of the KD in patients with RTT.
TUBEROUS SCLERO SIS
COMPLEX
Tuberous sclerosis complex is an autosomal domi-
nant neurocutaneous disorder characterized by
hamartomatous growths in multiple organ sys-
tems, including the skin, central nervous system,
eyes, kidneys, heart, and lungs. Mutations in the
TSC1 and TSC2 genes, encoding for hamartin and
tuberin, are present in over 85% of cases, result-
ing in hamartomatous growths due to unregulated
activation of the mammalian target of rapamycin
(mTOR) pathway (Napolioni et al., 2008). The most
common lesions found in the brain include corti-
cal tubers, subependymal nodules, subependymal
51
giant cell astrocytomas, and radial migration lines.
Epilepsy occurs in up to 85% of TSC patients, and
is a significant factor in neurological morbidity.
Seizures begin within the first year of life in 63%
and prior to 3 years of age in 83% (Chu-Shore et al.,
2010a). The majority of patients have multiple sei-
zure types, with the most common being complex
partial seizures, followed by infantile spasms, gen-
eralized tonic-clonic seizures, and atypical absence
(Chu-Shore et al., 2010a). The treatment of epilepsy
in TSC usually starts with the use of standard anti-
convulsant medications, but TSC is medication-
refractory in 30%–60% of patients (Chu-Shore
et al., 2010a; Vignoli et al., 2013). The TSC patients
with medication-refractory epilepsy are often con-
sidered early for epilepsy surgery candidacy, with
cortical tubers and abnormal surrounding cortex
usually the targets of interest. When surgical candi-
dates are chosen carefully, good outcomes with sei-
zure freedom can be achieved in about 60% (Zhang
et al., 2013). For those patients who are not surgi-
cal candidates, or for whom the family does not
wish to pursue surgical management, alternative
epilepsy treatment options such as dietary therapy
should be considered.
Tuberous sclerosis complex is listed by the
International Ketogenic Diet Study Group as a
condition for which the KD has been reported as
having probable benefit, defined as having at least
two publications in the literature. In a published
case series of 12 children with TSC between the
ages of 8  months and 18  years treated with the
KD, at 6  months, 92% had at least a 50% reduc-
tion in seizures, with 67% having at least a 90%
reduction (Kossoff et  al., 2005). In this series,
five children had at least a 5-month seizure-free
response. The authors recommended the KD in
52
52 section I: Ketogenic Diet for Epilepsy in the Clinic
patients with TSC for whom medications fail
and no clear epileptogenic tuber is identified. In
another case series of three children with TSC
and medication-refractory seizures who were fol-
lowed for 12 months on the KD, two of the patients
became seizure-free, and the third patient had his
drop seizures decreased (Coppola et  al., 2006).
Within a larger series of 71 patients with TSC, six
were started on the KD, and two of the six had a
50% decrease in seizures (Overwater et al., 2015).
When the KD is effective, the possibility
of weaning off the KD can be considered after
2 years. Unfortunately, a diagnosis of TSC is a risk
factor for seizure recurrence in patients weaned off
of the KD after 2 or more years of seizure freedom.
In one published series, all three patients with TSC
who were weaned off of the KD after becoming
seizure-free had a recurrence of seizures, and only
one regained full control with reinitiation of the
KD (Martinez et al., 2007). The authors concluded
that patients with TSC may need to be treated for
longer than 2 years, or possibly indefinitely if sei-
zure control is achieved.
There is also evidence that the low glyce-
mic index treatment (LGIT) diet can be effective
in reducing seizures, and is better tolerated, in
patients with TSC. In a case series of 15 patients
with TSC started on the LGIT, 47% had a 50% or
greater reduction in their seizures (Larson et  al.,
2012). Many of these patients had been unable to
tolerate the KD. In addition there is a report of a
Japanese patient with TSC and Lennox Gastaut
syndrome responding to the LGIT, achieved using
Japanese ethnic foods (Kumada et al., 2013).
It has now been well characterized that the
manifestations in TSC are largely due to dys-
functional regulation of the mTOR pathway. It is
unclear, however, whether there is a similar under-
lying mechanism to explain why patients with
TSC respond well to dietary therapies. There is
some bench evidence that markers of the mTOR
pathway, including pS6 and pAkt, are reduced
in the hippocampus and liver of rats fed the KD
(McDaniel et  al., 2011). However, in a series of
five patients with TSC who were treated with KD,
three of the five patients had progression of known
tumors (renal angiomyolipomas or subependy-
mal giant cell astrocytomas) or had new tumors
develop while being treated with the KD (Chu-
Shore et al., 2010b). Thus, if there is some mTOR
inhibition provided by the KD, it does not appear,
although only in a small sample, to provide the
same level of tumor regression effects in TSC as
better-studied mTOR inhibitors such as sirolimus
and everolimus.
ANGELMAN SYNDROME
Angelman syndrome is a neurogenetic disorder
characterized by cognitive and language impair-
ment, epilepsy, ataxia and tremorous move-
ments, and sleep disturbances, with patients also
tending to have an apparent happy demeanor.
Angelman syndrome is maternally inherited,
with four possible known genetic mechanisms
resulting in reduced expression of the ubiquitin-
protein ligase E3A (UBE3A) gene, most com-
monly from a deletion in the 15q11.2-13.1 region
(70%) (Jiang et  al., 1999). Epilepsy is present in
over 85% of patients with AS (Laan et al., 1997;
Thibert et al., 2009), with the onset of epilepsy in
50% by 1 year of age, and over 75% by 3 years of
age (Valente et al., 2006). The underlying etiology
of epilepsy in AS is not clear, but there are known
differences in epilepsy genotype-phenotypes cor-
relations:  AS patients with maternal deletions
have higher rates of epilepsy (90%) than those
without (55%–75%) (Thibert et  al., 2009), have
an earlier onset, and have a more severe epilepsy
phenotype (Valente et al., 2005; Minassian et al.,
1998). It has been hypothesized that there might
be other factors related in patients with mater-
nal deletions specifically associated within their
15q11-13 deletions, such as the genes coding for
subunits of the γ-aminobutyric acid (GABA), a
receptor complex within this region (Minassian
et  al., 1998). Epilepsy in AS typically manifests
with multiple seizure types, including myoclonic,
atypical absence, generalized tonic-clonic, and
atonic seizures, and AS patients commonly have
frequent or prolonged seizures (Thibert et  al.,
2009). Most patients are medication-refractory,
with up to 77% of patients requiring combination
medication therapy (Laan et  al., 1997; Thibert
et  al., 2009). Patients with AS and epilepsy are
unlikely to be candidates for epilepsy surgery,
thus alternative therapies such as dietary thera-
pies are often considered.
Several examples of successful treatment of
epilepsy in AS with the KD have been published.
In one series of 19 patients with genetically con-
firmed AS and medication-refractory epilepsy, the
KD was described as “effective” for all four patients
it was initiated in (Valente et al., 2006). In a case
report of an infant with AS, the KD produced a
rapid and significant reduction in seizures from
hundreds a day to near complete control, and was
accompanied by increased alertness and smiling
(Stein et al., 2010). In another report, a 5-year-old
with AS became seizure-free within 2 months on
the KD and was able to wean off of two of her three
anticonvulsant medications (Evangeliou et  al.,
53
Chapter 7: Ketogenic Diet for Other Epilepsies
2010). This patient also had an improvement in
her sleep pattern and reduced hyperactivity. In a
large electronic survey sent through the Angelman
Syndrome Foundation to 461 family members
of patients with AS, 11% reported trying dietary
therapy, to include the KD in 8% and LGIT in 2%.
Of those trying the KD, 36% reported that the
KD was the best overall treatment, although only
19% reported still being on the KD (Thibert et al.,
2009). Finally in a series of six children with AS
followed prospectively on the LGIT for 4 months,
five of six of the patients had at least an 80% reduc-
tion in seizures, with five remaining on the LGIT
for over a year (Thibert et al., 2012).
It is also uncertain whether there are underly-
ing mechanisms explaining the efficacy of dietary
therapies in patients with AS. The anticonvulsant
effects of the ketogenic diet are probably multiple
(Bough et  al., 2007). It has been reported that
children on the KD have an increase in GABA,
taurine, serine, and glycine in their cerebrospinal
fluid (Dahlin et al., 2005), thus a possible specific
action partly explaining the efficacy of dietary
therapies in patients could be the increase of
GABA synthesis in the brain. Specifically for AS,
as mentioned, there are genes encoding for the
GABAa receptor within the region of the mater-
nal 15q deletions associated with AS (Minassian
et  al., 1998), thus it has been hypothesized that
increasing GABA levels through the KD may be
an important component in its effectiveness in
patients with AS.
STURGE-WEBER SYNDROME
Sturge-Weber syndrome is a sporadic neurocu-
taneous disorder characterized by capillary mal-
formations in the distribution of the trigeminal
nerve, glaucoma, and cerebral venous malforma-
tions. Sturge-Weber syndrome is associated with
a somatic mutation in the GNAQ gene, which
plays a key role in cell proliferation (Shirley
et  al., 2013). Epilepsy is an especially common
manifestation, occurring in over 85% of patients
(Pascual-Castroviejo et  al., 2008), with onset by
1 year of age in up to 75% (Sujansky et al., 1995;
Kossoff et al., 2009a). Most seizures are focal, but
generalized seizures can also occur (Ewen et  al.,
2007), with seizures commonly occurring in a
pattern of clustering and intense seizures fol-
lowed by prolonged seizure-free periods (Kossoff
et  al., 2009a). Seizure control in SWS is particu-
larly important because of radiographic evidence
of hypoperfusion that may occur ictally in some
patients, which may contribute to progressive cor-
tical atrophy (Aylett et  al., 1999). The treatment
53
of epilepsy in SWS usually starts with the use of
standard anticonvulsant medications. The SWS
patients with medication-refractory epilepsy are
often considered for an epilepsy surgical evalu-
ation to consider options of either a lesionec-
tomy or hemispherectomy (Arzimanoglou et  al.,
2000; Kossoff et al., 2002). Hemispherectomy can
result in seizure freedom in 81% (Kossoff et  al.,
2002), without significant worsening in post-
operative hemiparesis or cognitive functioning
(Arzimanoglou et al., 2000).
The successful implementation of the KD
has been reported in a 4-year-old patient with
SWS, although a recurrence of seizures occurred
4  months into treatment (Petit et  al., 2008).
There is a theoretical risk in SWS with the KD
due to the fasting and fluid restrictions imple-
mented during the initiation of the KD, which
could possibly provoke stroke-like events that
SWS patients are prone to (Kossoff et al., 2010).
Thus, the use of less restrictive dietary thera-
pies has been proposed for patients with SWS.
In 2010, Kossoff et al. published a case series of
the modified Atkins diet (MAD) in five children
with SWS:  three of the patients had at least a
50% reduction in seizures, including one hav-
ing a 90% reduction, and one becoming seizure-
free (Kossoff et al., 2010). Only one patient had
a stroke-like event on the MAD, although this
patient had a prior history of stroke-like events
prior to initiation of the MAD.
CONCLUSIONS
In conclusion, dietary therapies are effective for a
wide spectrum of epilepsy types, and we present
four examples of genetic epilepsy syndromes that
have been reported to respond well to dietary ther-
apies. Whether efficacy in these conditions is due
in part to the broad-spectrum efficacy of the KD or
to specific mechanisms specific to these conditions
is still under investigation. If specific pathways
affected in particular genetic syndromes are better
characterized, this may help us further understand
the underlying pathophysiologic mechanisms con-
tributing to the efficacy of dietary therapies in the
treatment of epilepsy.
DISCLOSURES
The view(s) expressed herein are those of the
author(s) and do not reflect the official policy or
position of Brooke Army Medical Center, the US
Army Medical Department, the US Army Office of
the Surgeon General, the United States Air Force,
the Department of the Army and Department of
Defense, or the US Government.
54
54 section I: Ketogenic Diet for Epilepsy in the Clinic
REFERENCES
Arzimanoglou, A.A., Andermann, F., Aicardi, J.,
Sainte-Rose, C., Beaulieu, M.A., Villemure, J.G.,
Olivier, A., and Rasmussen, T. (2000). Sturge-
Weber syndrome:  indications and results of sur-
gery in 20 patients. Neurology 55, 1472–1479.
Aylett, S.E., Neville, B.G., Cross, J.H., Boyd, S., Chong,
W.K, and Kirkham, F.J. (1999). Sturge-Weber syn-
drome:  cerebral haemodynamics during seizure
activity. Dev Med Child Neurol 41, 480–485.
Bough, K.J., and Rho, J.M. (2007). Anticonvulsant
mechanisms of the ketogenic diet. Epilepsia
48, 43–58.
Cardoza, B., Clarke, A., Wilcox, J., Gibbon, F., Smith,
P.E., Archer, H., Hryniewiecka-Jaworska, A., and
Kerr, M. (2011). Epilepsy in Rett syndrome: asso-
ciation between phenotype and genotype, and
implications for practice. Seizure 20, 646–649.
Chu-Shore, C.J., Major, P., Camposano, S., Muzykewicz,
D., and Thiele, E.A. (2010a). The natural history of
epilepsy in tuberous sclerosis complex. Epilepsia
51, 1236–1241.
Chu-Shore, C.J., and Thiele, E.A. (2010b). Tumor
growth in patients with tuberous sclerosis complex
on the ketogenic diet. Brain Dev 32, 318–322.
Coppola, G., Klepper, J., Ammendola, E., Fiorillo,
M., della Corte, R., Capano, G., and Pascotto, A.
(2006). The effects of the ketogenic diet in refrac-
tory partial seizures with reference to tuberous
sclerosis. Eur J Paediatr Neurol 10, 148–151.
Dahlin, M., Elfving, A., Ungerstedt, U., and Amark, P.
(2005). The ketogenic diet influences the levels of
excitatory and inhibitory amino acids in the CSF
in children with refractory epilepsy. Epilepsy Res
64, 115–125.
Evangeliou, A., Doulioglou, V., Haidopoulou, K.,
Aptouramani, M., Spilioti, M., and Varlamis, G.
(2010). Ketogenic diet in a patient with Angelman
syndrome. Pediatr Int 52, 831–834.
Ewen, J.B., Comi, A.M., and Kossoff, E.H. (2007).
Myoclonic-astatic epilepsy in a child with Sturge-
Weber syndrome. Pediatr Neurol 36, 115–117.
Giampietro, P.F., Schowalter, D.B., Merchant, S.,
Campbell, L.R., Swink, T., and Roa, B.B. (2006).
Widened clinical spectrum of the Q128P MECP2
mutation in Rett syndrome. Childs Nerv Syst 22,
320–324.
Haas, R.H., Rice, M.A., Trauner, D.A., and Merritt,
T.A. (1986). Therapeutic effects of a ketogenic diet
in Rett syndrome. Am J Med Genet 24, 225–246.
Haas, R.H., Light, M., Rice, M., and Barshop, B.A.
(1995). Oxidative metabolism in Rett syn-
drome:  1.  Clinical studies. Neuropediatrics
26, 90–94.
Jiang, Y., Lev-Lehman, E., Bressler, J., Tsai, T.F., and
Beaudet, A.L. (1999). Genetics of Angelman syn-
drome. Am J Hum Genet 65, 1–6.
Kossoff, E.H., Buck, C., and Freeman, J.M. (2002).
Outcomes of 32 hemispherectomies for
Sturge-Weber syndrome worldwide. Neurology
59, 1735–1738.
Kossoff, E.H., Thiele, E.A., Pfeifer, H.H., McGrogan,
J.R., and Freeman, J.M. (2005). Tuberous sclero-
sis complex and the ketogenic diet. Epilepsia 46,
1684–1686.
Kossoff, E.H., Ferenc, L., and Comi, A.M. (2009a). An
infantile-onset, severe, yet sporadic seizure pattern
is common in Sturge-Weber syndrome. Epilepsia
50, 2154–2157.
Kossoff, E.H., Zupec-Kania, B.A., Amark, P.E.,
Ballaban-Gil, K.R., Christina Bergqvist, A.G.,
Blackford, R., Buchhalter, J.R., Caraballo, R.H.,
Helen Cross, J., Dahlin, M.G., et  al. (2009b).
Optimal clinical management of children receiv-
ing the ketogenic diet:  recommendations of
the International Ketogenic Diet Study Group.
Epilepsia 50, 304–317.
Kossoff, E.H., Borsage, J.L., and Comi, A.M. (2010). A
pilot study of the modified Atkins diet for Sturge-
Weber syndrome. Epilepsy Res 92, 240–243.
Kumada, T., Hiejima, I., Nozaki, F., Hayashi, A., and
Fujii, T. (2013). Glycemic index treatment using
Japanese foods in a girl with Lennox-Gastaut syn-
drome. Pediatr Neurol 48, 390–392.
Laan, L.A., Renier, W.O., Arts, W.F., Buntinx, I.M., vd
Burgt, I.J., Stroink, H., Beuten, J., Zwinderman,
K.H., van Dijk, J.G., Brouwer, O.F. (1997).
Evolution of epilepsy and EEG findings in
Angelman syndrome. Epilepsia 38, 195–199.
Larson, A.M., Pfeifer, H.H., and Thiele, E.A. (2012).
Low glycemic index treatment for epilepsy in tuber-
ous sclerosis complex. Epilepsy Res 99, 180–182.
Liebhaber, G.M., Riemann, E., and Baumeister, F.A.
(2003). Ketogenic diet in Rett syndrome. J Child
Neurol 18, 74–75.
Mantis, J.G., Fritz, C.L., Marsh, J., Heinrichs, S.C., and
Seyfried, T.N. (2009). Improvement in motor and
exploratory behavior in Rett syndrome mice with
restricted ketogenic and standard diets. Epilepsy
Behav 15, 133–141.
Martinez, C.C., Pyzik, P.L., and Kossoff, E.H. (2007).
Discontinuing the ketogenic diet in seizure-free
children: recurrence and risk factors. Epilepsia 48,
187–190.
Matsuishi, T., Yamashita, Y., Takahashi, T., and
Nagamitsu, S. (2011). Rett syndrome: the state of
clinical basic research, and future perspectives.
Brain Dev 33, 627–631.
McDaniel, S.S., Rensing, N.R., Thio, L.L., Yamada,
K.A., and Wong, M. (2011). The ketogenic diet
inhibits the mammalian target of rapamycin
(mTOR) pathway. Epilepsia 52, e7–e11.
Minassian, B.A., DeLorey, T.M., Olsen, R.W.,
Philippart, M., Bronstein, Y., Zhang, Q., Guerrini,
R., Van Ness, P., Livet, M.O., Delgado-Escueta,
A.V. (1998). Angelman syndrome:  correlations
between epilepsy phenotypes and genotypes. Ann
Neurol 43, 485–493.
55
Chapter 7: Ketogenic Diet for Other Epilepsies
Napolioni, V., and Curatolo, P. (2008). Genetics and
molecular biology of tuberous sclerosis complex.
Curr Genomics 9, 475–487.
Neul, J.L., Kaufmann, W.E., Glaze, D.G.,
Christodoulou, J., Clarke, A.J., Bahi-Buisson, N.,
Leonard, H., Bailey, M.E., Schanen, N.C., Zapella,
M., et al. (2010). Rett syndrome: revised diagnos-
tic criteria and nomenclature. Ann Neurol 68,
944–950.
Nissenkorn, A., Gak, E., Vecsler, M., Reznki, H.,
Menascu, S., and Ben Zeev, B. (2010). Epilepsy in
Rett syndrome: the experience of a National Rett
Center. Epilepsia 51, 1252–1258.
Nissenkorn, A., Levy-Drummer, R.S., Bondi, O.,
Renieri, A., Villard, L., Mari, F., Mencarelli, M.A.,
Lo Rizzo, C., Meloni, I., Pineda, M., et al. (2015).
Epilepsy in Rett syndrome—lessons from the Rett
networked database. Epilepsia 56, 569–576.
Overwater, I.E., Bindels-de Heus, K., Rietman, A.B.,
Ten Hoopen, L.W., Vergouwe, Y., Moll, H.A., and
de Wit, M.C. (2015). Epilepsy in children with
tuberous sclerosis complex:  chance of remission
and response to antiepileptic drugs. Epilepsia 56,
1239–1245.
Pascual-Castroviejo, I., Pascual-Pascual, S.I.,
Velazquez-Fragua, R., and Viano, J. (2008). Sturge-
Weber syndrome:  study of 55 patients. Can J
Neurol Sci 35, 301–307.
Petit, F., Auvin S., Lamblin, M.D., and Vallee, L. (2008).
Myoclonic astatic seizures in a child with Sturge-
Weber syndrome. Rev Neurol 164, 953–956.
Pintaudi, M., Calevo, M.G., Vignoli, A., Parodi, E.,
Aiello, F., Baglietto, M.G., Hayek, Y., Buini, S.,
Renieri, A., Russo, S., et al. (2010). Epilepsy in Rett
syndrome:  clinical and genetic features. Epilepsy
Behav 19, 296–300.
Shirley, M.D., Tang, H., Gallione, C.J., Baugher, J.D.,
Frelin, L.P., Cohen, B., North, P.E., Marchuk,
D.A., Comi, A.M., and Pevsner, J. (2013). Sturge-
Weber syndrome and port-wine stains caused by
55
somatic mutation in GNAQ. N Engl J Med 368,
1971–1979.
Stein, D., Chetty, M., and Rho, J.M. (2010). A “happy”
toddler presenting with sudden, life-threatening
seizures. Semin Pediatr Neurol 17, 35–38.
Sujansky, E., and Conradi, S. (1995). Sturge-Weber
syndrome: age of onset of seizures and glaucoma
and the prognosis for affected children. J Child
Neurol 10, 49–58.
Thibert, R.L., Conant, K.D., Braun, E.K., Bruno,
P., Said, R.R., Nespeca, M.P, and Thiele, E.A.
(2009). Epilepsy in Angelman syndrome:  a
questionnaire-based assessment of the natural
history and current treatment options. Epilepsia
50, 2369–2376.
Thibert, R.L., Pfeifer, H.H., Larson, A.M., Raby,
A.R., Reynolds, A.A., Morgan, A.K., and Thiele,
E.A. (2012). Low glycemic index treatment for
seizures in Angelman syndrome. Epilepsia 53,
1498–1502.
Valente, K.D., Fridman, C., Varela, M.C., Koiffmann,
C.P., Andrade, J.Q., Grossman, R.M., Kok, F.,
and Marques-Dias, M.J. (2005). Angelman syn-
drome:  uniparental paternal disomy 15 deter-
mines mild epilepsy but has no influence on EEG
patterns. Epilepsy Res 67, 163–168.
Valente, K.D., Koiffmann, C.P., Fridman, C., Varella,
M., Kok, F., Andrade, J.Q., Grossman, R.M, and
Marques-Dias, M.J. (2006). Epilepsy in patients
with Angelman syndrome caused by deletion
of the chromosome 15q11-13. Arch Neurol 63,
122–128.
Vignoli, A., La Briola, F., Turner, K., et  al. (2013).
Epilepsy in TSC:  certain etiology does not mean
certain prognosis. Epilepsia 54, 2134–2142.
Zhang, K., Hu, W.H., Zhang, C., Meng, F.G., Chen,
N., and Zhang, J.G. (2013). Predictors of seizure
freedom after surgical management of tuberous
sclerosis complex: a systematic review and meta-
analysis. Epilepsy Res 105, 377–383.
56
8
The Ketogenic Diet and Related Therapies
in “Novel” Situations
Idiopathic Generalized Epilepsy Syndromes
SUDHA KILARU KES SLER, MD, MSCE
INTRODUCTION
The ketogenic diet (KD) and related dietary thera-
pies, which are characterized by carbohydrate
restriction, are now widely accepted as a treatment
option for epilepsies that are resistant to medica-
tions. Though many studies have demonstrated
the efficacy of the KD in symptomatic generalized
or mixed epilepsies including infantile spasms,
Dravet syndrome, Doose syndrome, and Lennox
Gastaut syndrome, the KD is perhaps less often
considered for idiopathic generalized epilepsies,
which include childhood absence epilepsy (CAE),
juvenile absence epilepsy (JAE), juvenile myo-
clonic epilepsy (JME), and generalized epilepsy
with generalized tonic-clonic seizures (GE-GTC).
This chapter covers these “novel” epilepsy uses.
CHILDHOOD ABSENCE
EPILEPSY
Childhood absence epilepsy (CAE) is the most
common pediatric epilepsy syndrome, accounting
for more than 10% of all cases of epilepsy in chil-
dren (Berg et  al., 2000). Onset occurs between 4
and 8 years of age in an otherwise neurologically
normal child. Typical seizures are staring spells,
lasting only seconds, occurring many times daily,
and characterized on EEG by 3-Hz generalized
spike and wave activity with a normal background.
Childhood absence epilepsy has classically been
considered a benign epilepsy syndrome that
responds to medication, remits with time, and
does not affect neurodevelopmental outcome.
Now it appears that treatment response and remis-
sion rates are not as high as once assumed, and
long-term psychosocial difficulties are common
(Masur et  al., 2013; Nickels; Wirrell et  al., 1997).
Juvenile myoclonic epilepsy presents later in child-
hood, typically between 11 and 17 years of age, and
carries a higher risk of treatment resistance and
lower rate of eventual remission. The occurrence
of generalized tonic-clonic (GTC) seizures is more
common in JAE than in CAE, but in both syn-
dromes, GTC seizures portend a poorer prognosis
for remission.
Ethosuximide is the first-line treatment for
CAE, and valproic acid is also efficacious but has a
greater impact on tests of attention than does ETX
(Glauser et  al., 2010). Lamotrigine is efficacious
in a smaller proportion of patients. Second-line
therapies are broad-spectrum AEDs such as topira-
mate, zonisamide, and clobazam, or conventional
benzodiazepines such as clonazepam. Antiepileptic
drugs for focal mechanisms of action in epilepsy,
including the sodium channel blockers, are avoided
because they might exacerbate seizures with gen-
eralized mechanisms of action. Thus, treatment
choices are limited when first-line drugs fail.
Unfortunately the published literature on keto-
genic diet related therapies for absence epilepsy
is quite limited. One study from Johns Hopkins
specifically evaluated the efficacy of the KD and
the modified Atkins diet (MAD) in children with
treatment-resistant absence epilepsy, and also
included a review of studies in the literature that
included absence epilepsy patients (Groomes et al.,
2011). In the Johns Hopkins cohort, 8 patients
with absence epilepsy were treated with the KD
and 13 patients with absence epilepsy were treated
with the MAD. Of the combined 21 patients, only
2 had a history of GTC seizures in addition to
absence, but 8 had early-onset absence with sei-
zures beginning prior to 4  years of age. Diet ini-
tiation occurred 1 to 12 years after seizure onset,
and the median number of antiepileptic medica-
tions used was 4 (range 2–10). After 3 months of
dietary therapy, 18 children (82%) had a >50%
57
Chapter 8: The Ketogenic Diet and Related Therapies in “Novel” Situations
seizure reduction, and 4 (19%) were seizure-free.
Too few patients (only three) had EEG data before
and during dietary therapy to determine the effect
on spike wave discharges.
The same paper also included a summary of 17
previously published KD studies dating from 1922
to 2008, which included patients with CAE or JAE.
Of the 133 children with absence epilepsy and
documented outcomes (with responses recorded
at 3  days to 3  months), more than two-thirds
reported a greater than 50% seizure reduction and
one-third experienced periods of seizure freedom.
What is not known from this historical review is
the proportion of patients presenting with a CAE-
like syndrome due to glucose transporter type 1
deficiency, which is expected to respond well to
the ketogenic diet.
A recent case report describes a 7-year-old
girl with absence epilepsy who became seizure-
free with the paleolithic ketogenic diet consisting
entirely of meat, offal, fish, egg, and animal fats
without calorie restriction, which she apparently
tolerated well (Clemens et  al., 2013). Of note,
dietary therapy was used as first-line antiepilep-
tic treatment because of parental concerns about
the potential for medication side effects. Also of
note, the child had other neurologic abnormali-
ties (developmental delays, social withdrawal, and
a history of febrile seizures)—and thus was atypi-
cal for classic CAE. Whether the child underwent
diagnostic testing for glucose transporter type
1 deficiency is not reported. After an immedi-
ate and gradual decline in seizure frequency, the
patient became seizure-free at 6 weeks of therapy.
At 3  months, carbohydrates were reintroduced,
in the form of low glycemic index vegetables.
The patient had no evidence of epileptiform dis-
charges on long-term video-EEG monitoring, and
continued to be seizure-free 20 months after diet
initiation.
Interestingly several large, single-center series
of KD effectiveness include no patients with
absence epilepsy (Dressler et  al., 2010; Hallböök
et  al., 2015; Sharma et  al., 2009; Wibisono et  al.,
2015). In a recent large, single-center series from
Australia that included 61 patients initiating
the KD, the 29 patients who were responders at
3  months included two with treatment-resistant
CAE, both of whom continued to be responders at
12 months (Thammongkol et al., 2012).
J U V E N I L E M YO C L O N I C
EPILEPSY
Another genetic generalized epilepsy syndrome
infrequently appearing in ketogenic diet series,
57
JME is a common and highly recognizable epilepsy
syndrome with onset in adolescence or young
adulthood (Dreifuss, 1989). The hallmarks of the
syndrome are myoclonic seizures (brief bilateral
jerks with persevered awareness), often cluster-
ing in the morning and sometimes triggered by
photic stimulation, and 4-Hz or faster generalized
spike and wave or polyspike and wave discharges
on EEG. Many patients initially come to medical
attention not when the myoclonic jerks begin, but
when the first GTC seizure occurs. Absence sei-
zures occur in a subset of patients, and may be a
marker of worse long-term seizure prognosis (Senf
et al., 2013).
Because valproic acid (VPA) is efficacious
in most patients with JME, it has long been con-
sidered the treatment of choice (Dreifuss, 1989).
However, growing concerns about the long-term
side effects and teratogenic potential of VPA have
made it a less desirable choice, particularly in
young women (Tomson et al., 2015; Wheless et al.,
2005). In addition, treatment resistance, even to
VPA, does occur in JME. Thus, dietary therapy
may be considered in this population as well.
Very little published literature has addressed
the use of KD-related therapies in JME. In 2013,
Dr. Kossoff and colleagues described the use of the
MAD in eight JME patients ages 15 to 44 years, six
of whom were female (Kossoff et al., 2013). Except
for one patient who was medication naïve, all oth-
ers were treatment-resistant to four or more AEDs,
and all but three had tried valproate. Of note, the
mean age of onset was 10.5 years, which is atypi-
cally young for JME. Seven of the eight remained
on the diet at 1 month and achieved at least mod-
erate urinary ketosis. After 3  months, five (63%)
reported a greater than 50% drop in seizure fre-
quency, and two became seizure-free. At the time
the paper was published, patients had remained
on the diet 0.5 to 40  months; two patients were
able to reduce medications, but none had stopped
all AEDs.
Just as for CAE, many of the larger, single-
center, long-term follow-up studies of the KD
included no patients with JME. The case series
from Australia discussed previously also included
a single patient with JME, who was a responder
at 3  months (no 12-month follow-up reported).
A  very recent study from Norway prospectively
evaluated 13 adults with idiopathic generalized
epilepsy (IGE), ages 16 to 57  years, initiating the
MAD (Kverneland et al., 2015). The majority, nine,
had JME, while two had CAE, one had Jeavon syn-
drome (absence epilepsy with eyelid myoclonia),
and one had an IGE not otherwise specified. Nine
58
58 section I: Ketogenic Diet for Epilepsy in the Clinic
patients continued the MAD for at least 4 weeks,
and only six reached the 12-week time point. Four
patients achieved a 50% or greater reduction in
seizure frequency—in one, frequent GTC seizures
stopped. Lack of motivation or lack of adherence
to the diet were the reasons most frequently cited
for dropping out of the study.
While the reason for referral for KD treatment
for most IGE patients may be treatment-resistant
seizures, the KD has other potential advantages
over AEDs besides efficacy. Though specific cog-
nitive and behavioral challenges have been noted
(inattention in CAE, impulsivity and other frontal
lobe dysfunction in JME, for example), patients
with IGE syndromes typically have normal intel-
ligence at baseline—thus, cognitive adverse effects
of AEDs may have a substantial impact on func-
tion, particularly for those on polytherapy. Though
rigorous controlled studies are still lacking, the
potentially beneficial effects of the KD on cogni-
tion have been repeatedly noted. Therefore, for
IGE patients who are experiencing cognitive side
effects from one or more AEDs, the KD may be a
desirable alternative, or a strategy for reducing the
number of AEDs.
A key challenge for dietary epilepsy therapy
in patients with IGE is the feasibility and toler-
ability of the diet at a time when food choices are
typically made more autonomously than in early
childhood. In addition, maintaining the mini-
mum protein requirements in an adolescent or
adult in the context of a protein and carbohydrate
restricted, but normal caloric, KD can be difficult.
The MAD, which is high-fat and low carbohydrate,
but less stringent in protein restriction, might be
more feasible and tolerable than a classic KD in
these patients. However, as shown in the study by
Kverneland and colleagues, even adult patients
may-find the MAD difficult to carry out con-
sistently (Kverneland et  al., 2015). There are no
reports of the use of two other KD-related thera-
pies, the low glycemic index treatment (LGIT) and
the MCT oil diet, in IGE syndromes specifically.
The published series on LGIT include patients
with generalized seizure types, but specific syn-
drome information is not reported (Muzykewicz
et al., 2009; Pfeifer and Thiele, 2005).
In conclusion, the KD or MAD are viable
treatment options for patients with medication-
resistant IGE syndromes, and may be preferred
over AEDs causing cognitive adverse effects.
However, feasibility and tolerability continue to be
a major barrier. Further investigations of the long-
term outcomes of IGE treated with dietary treat-
ments are needed.
REFERENCES
Berg, A.T., Shinnar, S., Levy, S.R., Testa, F.M., Smith-
Rapaport, S., and Beckerman, B. (2000). How well
can epilepsy syndromes be identified at diagno-
sis? A reassessment 2 years after initial diagnosis.
Epilepsia 41, 1269–1275.
Clemens, Z., Kelemen, A., Fogarasi, A., and Tóth, C.
(2013). Childhood absence epilepsy successfully
treated with the paleolithic ketogenic diet. Neurol
Ther 2, 71–76.
Dreifuss, F.E. (1989). Juvenile myoclonic epi-
lepsy:  characteristics of a primary generalized
epilepsy. Epilepsia 30(Suppl 4), S1–S7; discussion
S24–S27.
Dressler, A., Stöcklin, B., Reithofer, E., Benninger, F.,
Freilinger, M., Hauser, E., Reiter-Fink, E., Seidl, R.,
Trimmel-Schwahofer, P., and Feucht, M. (2010).
Long-term outcome and tolerability of the keto-
genic diet in drug-resistant childhood epilepsy: the
Austrian experience. Seizure 19, 404–408.
Glauser, T.A., Cnaan, A., Shinnar, S., Hirtz, D.G.,
Dlugos, D., Masur, D., Clark, P.O., Capparelli, E.
V, and Adamson, P.C. (2010). Ethosuximide, val-
proic acid, and lamotrigine in childhood absence
epilepsy. N Engl J Med 362, 790–799.
Groomes, L.B., Pyzik, P.L., Turner, Z., Dorward, J.L.,
Goode, V.H., and Kossoff, E.H. (2011). Do patients
with absence epilepsy respond to ketogenic diets? J
Child Neurol 26, 160–165.
Hallböök, T., Sjölander, A., Åmark, P., Miranda, M.,
Bjurulf, B., and Dahlin, M. (2015). Effectiveness of
the ketogenic diet used to treat resistant childhood
epilepsy in Scandinavia. Eur J Paediatr Neurol
19, 29–36.
Kossoff, E.H., Henry, B.J., and Cervenka, M.C. (2013).
Efficacy of dietary therapy for juvenile myoclonic
epilepsy. Epilepsy Behav 26, 162–164.
Kverneland, M., Selmer, K.K., Nakken, K.O., Iversen,
P.O., and Taubøll, E. (2015). A prospective study of
the modified Atkins diet for adults with idiopathic
generalized epilepsy. Epilepsy Behav 53, 197–201.
Masur, D., Shinnar, S., Cnaan, A., Shinnar, R.C., Clark,
P., Wang, J., Weiss, E.F., Hirtz, D.G., and Glauser,
T.A. (2013). Pretreatment cognitive deficits
and treatment effects on attention in childhood
absence epilepsy. Neurology 81, 1572–1580.
Muzykewicz, D.A., Lyczkowski, D.A., Memon, N.,
Conant, K.D., Pfeifer, H.H., and Thiele, E.A.
(2009). Efficacy, safety, and tolerability of the low
glycemic index treatment in pediatric epilepsy.
Epilepsia 50, 1118–1126.
Nickels, K. Seizure and psychosocial outcomes of
childhood and juvenile onset generalized epilep-
sies: wolf in sheep’s clothing, or well-dressed wolf?
Epilepsy Curr 15, 114–117.
Pfeifer, H.H., and Thiele, E.A. (2005). Low-glycemic-
index treatment:  a liberalized ketogenic diet for
treatment of intractable epilepsy. Neurology 65,
1810–1812.
59
Chapter 8: The Ketogenic Diet and Related Therapies in “Novel” Situations
Senf, P., Schmitz, B., Holtkamp, M., and Janz, D.
(2013). Prognosis of juvenile myoclonic epilepsy
45 years after onset: seizure outcome and predic-
tors. Neurology 81, 2128–2133.
Sharma, S., Gulati, S., Kalra, V., Agarwala, A., and
Kabra, M. (2009). Seizure control and biochemi-
cal profile on the ketogenic diet in young chil-
dren with refractory epilepsy:  Indian experience.
Seizure 18, 446–449.
Thammongkol, S., Vears, D.F., Bicknell-Royle, J.,
Nation, J., Draffin, K., Stewart, K.G., Scheffer, I.E.,
and Mackay, M.T. (2012). Efficacy of the keto-
genic diet: which epilepsies respond? Epilepsia 53,
e55–e59.
Tomson, T., Marson, A., Boon, P., Canevini, M.P.,
Covanis, A., Gaily, E., Kälviäinen, R., and Trinka,
E. (2015). Valproate in the treatment of epilepsy
59
in girls and women of childbearing potential.
Epilepsia 56, 1006–1019.
Wheless, J.W., Clarke, D.F., and Carpenter, D. (2005).
Treatment of pediatric epilepsy:  expert opinion,
2005. J Child Neurol 20(Suppl 1), S1–S56; quiz
S59–S60.
Wibisono, C., Rowe, N., Beavis, E., Kepreotes,
H., Mackie, F.E., Lawson, J.A., and Cardamone,
M. (2015). Ten-year single-center experience
of the ketogenic diet:  factors influencing effi-
cacy, tolerability, and compliance. J Pediatr 166,
1030–1036.e1.
Wirrell, E.C., Camfield, C.S., Camfield, P.R., Dooley,
J.M., Gordon, K.E., and Smith, B. (1997). Long-
term psychosocial outcome in typical absence epi-
lepsy. Sometimes a wolf in sheeps’ clothing. Arch
Pediatr Adolesc Med 151, 152–158.
60
9
Ketogenic Diet in Status Epilepticus
R I M A N A B B O U T, M D , P H D
INTRODUCTION
Status epilepticus (SE) is defined by continuous
epileptic status with alteration of consciousness.
Status epilepticus might be a single prolonged
seizure or recurrent seizures repeated frequently
without recovery of consciousness between the
seizures. The incidence of SE is 3–41/100,000
individuals per year (Chin et  al., 2006, Raspall-
Chaure, 2007, Dham, 2014) and it is the second
most common neurologic emergency in adults
and the first in children. The duration of SE was
long debated, and an operational definition is
proposed for tonic clonic status defining T1 as the
time that emergency treatment should be started
and T2 as the time at which long-term conse-
quences on the brain might be expected (Trinka
et  al., 2015). For instance, T1 is from 5 minutes
and above and T2 is over 30 minutes of duration
(Trinka et al., 2015). 12%–43% of patients with SE
fail to respond to first- and second-line therapy,
and they enter refractory SE (after 2 hours of the
onset of SE). Super-refractory SE (SRSE) is defined
by the persistence of SE over 24 hours (Ferlisi
and Shorvon, 2012). Overall, ~15% of SE cases
admitted to the hospital become super-refractory
(DeLorenzo, 1995; Krumholz et  al., 1999; Ferlisi
and Shorvon, 2012).
In adults, SRSE presents a high rate of mortal-
ity (>60%) (Ferlisi and Shorvon, 2012). Although
the risk of death is low in the pediatric popula-
tion, the risk of subsequent neurological morbid-
ity and cognitive problems is high (Scott, 2009).
The therapeutic intervention aims to reduce SRSE
duration, mortality, and short- and long-term
comorbidities.
Status epilepticus can be tonic clonic, tonic
or myoclonic. It can be associated with acute or
chronic brain disease or can occur in patients with
known epilepsy. It might occur at the onset or in
the course of the disease (Trinka et al., 2015). For
instance,inaugural convulsive SE without previ-
ous known epilepsy occurs in CNS infections
(encephalitis, meningitis), autoimmune epilepsies,
epilepsies with presumed immune etiology such
as FIRES and NORSE, metabolic disorders, and
stroke. SE might occur during the course of epi-
lepsy syndromes; tonic clonic or clonic in Dravet
syndrome, tonic in Lennox Gastaut syndrome.
In myoclonic atonic epilepsy (Doose syndrome),
myoclonic status might occur at onset or during
the first months. Myoclonic nonconvulsive status
is frequent in some genetic syndromes such as
Angelman syndrome, and in mitochondrial dis-
eases (Trinka et al., 2015).
Identifying a possible underlying cause and
treating it is an important step in the treatment of
SE. An etiology-targeted therapy should be insti-
tuted as soon as the etiology is identified (anti-
biotics for CNS infections, antiviral agents for
encephalitis, immune therapy for auto-immune
encephalitis).
Benzodiazepines are the first-line therapy
in SE, with different molecules used and differ-
ent routes of administration; buccal, intra rectal
or intra muscular. If benzodiazepines fail to stop
the SE, phenytoin is usually the second-line drug.
The use of other AEDS with IV formulations is
the usual next step after the failure of phenytoin
(Shorvon and Ferlisi, 2011). Barbiturate anesthesia
is the most common third-line drug and is used in
adults more often than in children. Propofol and
ketamine can also be used (Ferlisi et al., 2015). The
failure rate of anesthesic barbiturate and the fre-
quent recurrence after withdrawal with associated
high mortality and morbidity make it essential to
develop other therapies for medication-resistant
SE.
The KD has reported efficacy in SE (Kossoff
and Nabbout, 2014). The multiple mechanisms of
action of KD make it a good choice for refractory
SE. The inherent combination of these mechanisms
can mimic AED polytherapy, an approach that is
suggested to be a good choice for RSE (Löscher,
2015, Gama et al., 2015, Lusardi et al., 2015).
61

Chapter 9: Ketogenic Diet in Status Epilepticus 61

K E TO G E N I C D I E T etiologies (Table 9.1). This trend is confirmed by

I N   C O N V U L S I V E S TAT U S
EPILEPTICUS
The first report of efficacy of KD in a series of
patients with highly recurrent seizures was in 2009
by Villeneuve et  al. (Villeneuve et  al., 2009). The
authors reported KD response in a retrospective
pediatric series of patients with focal pharmacore-
sistant epilepsies. The efficacy of KD was higher in
patients who presented with SE or recent worsen-
ing of seizure frequency (Villeneuve et al., 2009).
In this first series, efficacy was reported in different
etiologies such as Sturge-Weber syndrome, FIRES,
and cryptogenic focal SE (Villeneuve et al., 2009).
A second paper in 2010 reported an international
series of nine patients with SRSE due to FIRES
with 7/9 responders to KD (Nabbout et al., 2010).
The KD was well tolerated via nasogastric tube in
the pediatric ICU setting. Ketosis was achieved
mostly within 24 hours and SE stopped during the
first days of the diet and in all responders within
the first week (Nabbout et al., 2010).
Other reports of SE due to FIRES or similar
immune entities were further reported in the liter-
ature, including pediatric (Vaccarezza et al., 2012,
Nam et al., 2011, Sort et al., 2013, O’Connor et al.,
2014)  and adult patients (Cervenka et  al., 2011,
Thakhur et  al., 2014). The pathophysiology of
FIRES, based on activation of an inflammatory cas-
cade (Nabbout et al., 2011), makes this syndrome
a possible specific target for KD. The majority of
patients reported had RSE due to inflammatory
the largest pediatric series with FIRES (Nabbout
et  al., 2010)  and in adult series (Thakur et  al.,
2014). In this last series of 10 patients, 4 patients
presented with NORSE, 2 with anti-NMDA
encephalitis, and 1 with LGI1 encephalitis.
Other etiologies, possibly involving inflam-
mation cascades, were reported as a possible
indication for KD in SE; the over representa-
tion of these inflammatory etiologies might
be due to their frequency in SRSE. Patients
with Rasmussen syndrome, an encephalitis
involving the activation of the T cell pathway
were reported as responders in a few pediat-
ric (Villeneuve et al., 2009) and adult patients
(Wusthoff et al., 2010).
SE in mitochondrial diseases as in Alpers dis-
ease with POLG1 mutations (Martikainen et  al.,
2012) and with stroke-like accidents as in MELAS
syndrome (Steriade et  al., 2014)  are good can-
didates for KD to be introduced early at onset
(Desguerres et  al., 2014). Other small series of
patients with focal SE due to structural lesions
were also reported as responders to KD in pediatric
(Villeneuve et al., 2009, Caraballo et al., 2014, Lin
et al., 2015) or adult series (Bodenant et al, 2008).
K E TO G E N I C D I E T
I N   N O N C O N V U L S I V E S TAT U S
EPILEPTICUS
The efficacy of KD is not restricted to convulsive
SE. The efficacy of KD was reported in myoclonic

TABLE 9.1 REPORTS OF EFFICACY OF KETOGENIC DIET (KD) IN SE IN CHILDREN

AND ADULTS. *KD WAS INTRA VENOUS (IV)

Author Population Diet Etiology Time to response

Bodenant (2008) 1 adult KD Partial 7 days

Villeneuve (2009) 5 children KD SWS, encephalitis… 1–10 days
Kumada (2009) 2 children MAD Frontal lobe, heterotopia 5–10 days
Wusthoff (2010) 2 adults KD Rasmussen, head trauma 8–10 days
Nabbout (2010) 9 children KD FIRES 4–6 days
Cervenka (2011) 1 adult KD Idiopathic, autoimmune 7 days
Ismail (2012) 1 child KD FIRES 10 days
Nam (2012) 4 children, 1 adult KD Encephalitis 1–19 days
Vaccarezza (2012) 5 children KD FIRES, partial status ?
Sort (2013) 3 children KD FIRES, HHE, mito 1–6 days
Thakur (2014) 10 adults KD Super refractory SE 1–7 days
(70% “encephalitis”)
O’Connor (2014) 5 children KD Mito, myoc, FIRES? 2–8 days
Caraballo (2015) 10 children (7/10) KD Focal seizures 5–7 days
Lin (2015) 1 child KD* Focal with Gen 2–3 days
Caraballo (2015) 2 children KD Myoclonic status (EE in 1) 3 days
62

62 section I: Ketogenic Diet for Epilepsy in the Clinic

SE of mitochondrial diseases and POLG muta-
tions (Martikainen et  al., 2012, Desguerres et  al.,
2014) and of myoclonic astatic epilepsy, or Doose
syndrome (Kelley et  al., 2010, Caraballo et  al.,
2013,), and in patients with myoclonic SE from
unknown etiologies (Caraballo et  al., 2015). KD
efficacy was also reported in patients with electri-
cal SE during slow sleep (ESES) (Reyes et al., 2015;
Veggiotti et al., 2012).
CHALLENGES
O F   K E TO G E N I C D I E T U S E
I N   S TAT U S E P I L E P T I C U S
The challenges of the KD might raise some con-
cerns that limit its widespread use for SE in ICUs.
Implementation of the diet could be a complex
issue in ICUs, especially in centers lacking KD
teams. The availability of the KD team and the
daily communication between the neurologists,
child neurologists, dietitians, and ICU teams are
strongly recommended, enabling initiation of the
diet under optimal conditions. This daily commu-
nication makes it possible to respect the follow-up
of the diet, thus avoiding any additional glucose
load from fluids and concomitant medications.
Enteral feeding should be privileged, since par-
enteral feeding cannot achieve a high-ratio KD.
Enteral feeding is usually well tolerated in our
experience, especially when initiated with con-
tinuous infusion slowly increasing the feeding rate
to achieve the total caloric intake within 48–72
hours. The major steps for successful implementa-
tion in the ICU are summarized in Figure 9.1.
Another limiting factor for the use of the diet
might be the time lag for efficacy. In the reported
series, ketosis appears within 24 to 72 hours and
seizure reduction within the first week (Table 9.1).
This time lag is difficult to accept in a severe condi-
tion such as SE. Although many studies confirmed
the main role of underlying etiology in the cogni-
tive outcome of SE, the long duration of SE might
also negatively impact the long-term outcome
(Kilbride et al., 2013). However, after the first- and
second-line therapies for SE, treatment alterna-
tives are scarce and the use of an anesthetic agent is
usually the main strategy left (Ferlisi and Shorvon,
2012). Anesthetic agents are potent seizure sup-
pressors and might help to shorten the SE but their
use is based on expert opinion and is not evidence
based. In addition, some concerns have been raised
about possible worsening of the outcome of refrac-
tory SE after the use of anesthetic agents (Sutter
et al., 2014, 2015, Ferlisi et al., 2015). Limiting dex-
trose containg IV fluids early in the course of the SE

• Confirm the indication of KD with the ICU team

• Check for errors of fatty acid metabolism and for severe hypercholesterolemia
• Stop glucose intake after normal glycemia control:
• Stop all fluids with glucose
• Change medication with glucose when possible to glucose-free medications
Pre-diet • Inform the nutrition team

• Monitor serum glucose every 6 hours

• Monitor ketosi
• Monitor serum electrolytes and acidosis
• Control all fluids and medicines to avoid glucose accidental intake
Initiation of
• Initiate continuous enteralfeeding with 1/4-1/3 goal caloric rate and advance as tolerated
the KD

• Continue the control of accidental glucose intake

• Continue for at least 2 weeks if no major side effect
• Taper if no efficacy
follow-up

FIGURE 9.1 Steps to implement the diet in ICU.

63
Chapter 9: Ketogenic Diet in Status Epilepticus
after blood glucose has been controlled, and initiat-
ing the diet as soon as inborn errors of fat metabo-
lism are ruled out might help to shorten the delay
of ketosis and improve the efficacy of KD. Along
these lines, some medications used in the setting
of SE and refractory SE, such as steroids, can delay
ketosis. They should be avoided when medically
unnecessary (Nabbout et al., 2010).
The reports of KD in SE are mainly retro-
spective, reporting small numbers of individuals
treated and rarely referring to patients where the
diet failed. In addition, comedication and chang-
ing doses of comedication with the possible time
to achieve efficacy of the diet might make causal
attribution debatable. Finally, reports lack the
long-term follow-up studies reporting cognitive
and neurological outcome, which is, apart from
mortality, the major endpoint of any treatment of
refractory SE. Indeed, KD reports share the same
weakness with all third-line therapies in refrac-
tory SE where no drug or therapy has achieved a
high level of evidence-base medicine (Ferlisi and
Shorvon, 2012).
The KD is well tolerated with low rates of side
effects in the short and long term in the ICU set-
ting, as detailed in different reports (Table 9.1).
The increasing number of reports worldwide dem-
onstrates its possible implementation in ICUs. Its
efficacy in inflammatory SE or in SE from other
etiologies, convulsive and nonconvulsive, should
make it a therapeutic option in the treatment of
refractory SE. Recently a few data on possible
improvement of cognitive outcome are emerging
in patients with FIRES (Kramer et al., 2011, Singh
et al., 2014).
A prospective, randomized controlled trial is
necessary to validate this treatment option, as for all
third-line therapies for refractory SE. This is impor-
tant for physicians—at least one patient has died
after the KD was stopped following seizure arrest
because “this indication was not considered as good
clinical practice” (Nabbout et  al, 2010). This trial
should evaluate efficacy and tolerance and would
be mandatory for the acceptance of KD by physi-
cians and by health authorities and institutions.
Outcomes should be evaluated in the short term -
aiming for control of SE - and also in the long-term
(at a few months or longer) to evaluate seizure
control as well as cognitive outcomes. Pending the
results of such a trial, KD should be available in
ICUs and be part of the treatment arsenal of refrac-
tory SE, a critical situation where evidence-based
medicine is dramatically lacking to date.
63
REFERENCES
Amer, S., Shah, P., and Kommineni, V. (2015).
Refractory status epilepticus from NMDA receptor
encephalitis successfully treated with an adjunc-
tive ketogenic diet. Ann Indian Acad Neurol 18(2),
256–257.
Bodenant, M., Moreau, C., Sejourné, C., Auvin, S.,
Delval, A., Cuisset, J.M, Derambure, P., Destée, A.,
and Defebvre, L. (2008). [Interest of the ketogenic
diet in a refractory status epilepticus in adults].
Rev Neurol (Paris) 164(2), 194–199.
Caraballo, R.H., Chamorro, N., Darra, F., Fortini, S.,
and Arroyo H. (2013). Epilepsy with myoclonic
atonic seizures:  an electroclinical study of 69
patients. Pediatr Neurol 48(5), 355–362.
Caraballo, R.H., Flesler, S., Armeno, M., Fortini, S.,
Agustinho, A., Mestre, G., Cresta, A., Buompadre,
M.C., and Escobal, N. (2014). Ketogenic diet in
pediatric patients with refractory focal status epi-
lepticus. Epilepsy Res 108(10), 1912–1916.
Caraballo, R.H., Valenzuela, G.R., Armeno, M.,
Fortini, S., Mestre, G., and Cresta A. (2015, Nov
26). The ketogenic diet in two paediatric patients
with refractory myoclonic status epilepticus.
Epileptic Disord [Epub ahead of print]. PubMed
PMID: 26610306.
Chin, R.F., Neville, B.G, Peckham, C., Bedford, H.,
Wade, A., and Scott, R.C; NLSTEPSS Collaborative
Group. (2006). Incidence, cause, and short-term
outcome of convulsive status epilepticus in child-
hood: prospective population-based study. Lancet
368(9531), 222–229.
DeLorenzo, R.J., Pellock, J.M., Towne, A.R., and Boggs,
J.G. (1995). Epidemiology of status epilepticus. J
Clin Neurophysiol 12(4), 316–325.
Desguerre, I., Hully, M., Rio, M., and Nabbout, R.
(2014). Mitochondrial disorders and epilepsy. Rev
Neurol (Paris) 170(5), 375–380.
Dham, B.S., Hunter, K., and Rincon, F. (2014). The
epidemiology of status epilepticus in the United
States. Neurocrit Care 20(3), 476–483.
Ferlisi, M., and Shorvon, S. (2012). The outcome of
therapies in refractory and super-refractory con-
vulsive status epilepticus and recommendations
for therapy. Brain 135(Pt 8), 2314–2328.
Ferlisi, M., Hocker, S., Grade, M., Trinka, E., and
Shorvon, S.; International Steering Committee of
the StEp Audit. (2015). Preliminary results of the
global audit of treatment of refractory status epi-
lepticus. Epilepsy Behav 49, 318–324.
Gama, I.R., Trindade- Filho, E.M., Oliveira, S.L.,
Bueno, N.B., Melo, I.T., Cabral- Junior, C.R.,
Barros, E.M., Galvão, J.A., Pereira, W.S.,
Ferreira, R.C., et al. (2015). Effects of ketogenic
diets on the occurrence of pilocarpine-induced
status epilepticus of rats. Metab Brain Dis
30(1), 93–98.
64
64 section I: Ketogenic Diet for Epilepsy in the Clinic
Kelley, S.A., and Kossoff, E.H. (2010). Doose syndrome
(myoclonic-astatic epilepsy): 40 years of progress.
Dev Med Child Neurol 52(11), 988–993.
Kilbride, R.D., Reynolds, A.S., Szaflarski, J.P., and
Hirsch, L.J. (2013). Clinical outcomes following
prolonged refractory status epilepticus (PRSE).
Neurocrit Care 18(3), 374–385.
Kossoff, E.H., and Nabbout, R. (2013). Use of dietary
therapy for status epilepticus. J Child Neurol 28(8),
1049–1051.
Kramer, U., Chi, C.S., Lin, K.L., Specchio, N., Sahin,
M., Olson, H., Nabbout, R., Kluger, G., Lin, J.J.,
and van Baalen, A. (2011). Febrile infection-
related epilepsy syndrome (FIRES): pathogenesis,
treatment, and outcome: a multicenter study on 77
children. Epilepsia 52(11), 1956–1965.
Krumholz, A. (1999). Epidemiology and evidence for
morbidity of nonconvulsive status epilepticus. J
Clin Neurophysiol 16(4), 314–322; discussion 353.
Review.
Lin, J.J., Lin, K.L., Chan, O.W., Hsia, S.H., and Wang,
H.S.; CHEESE Study Group. (2015). Intravenous
ketogenic diet therapy for treatment of the acute
stage of super-refractory status epilepticus in a
pediatric patient. Pediatr Neurol 52(4), 442–445.
Löscher W. (2015). Single versus combinatorial thera-
pies in status epilepticus: novel data from preclini-
cal models. Epilepsy Behav 49:20–25.
Lusardi, T.A., Akula, K.K., Coffman, S.Q., Ruskin, D.N.,
Masino, S.A., and Boison, D. (2015). Ketogenic
diet prevents epileptogenesis and disease progres-
sion in adult mice and rats. Neuropharmacology
99:500–509.
Martikainen, M.H., Päivärinta, M., Jääskeläinen, S.,
and Majamaa, K. Successful treatment of POLG-
related mitochondrial epilepsy with antiepileptic
drugs and low glycaemic index diet.
Nabbout, R., Mazzuca, M., Hubert, P., Peudennier, S.,
Allaire, C., Flurin, V., Aberastury, M., Silva, W.,
and Dulac, O. (2010). Efficacy of ketogenic diet
in severe refractory status epilepticus initiating
fever induced refractory epileptic encephalopathy
in school age children (FIRES). Epilepsia 51(10),
2033–2037.
Nabbout, R., Vezzani, A., Dulac, O., and Chiron, C.
(2011). Acute encephalopathy with inflammation-
mediated status epilepticus. Lancet Neurol 10(1),
99–108.
Nam, S.H., Lee, B.L., Lee, C.G., Yu, H.J., Joo, E.Y., Lee,
J., and Lee, M. (2011). The role of ketogenic diet
in the treatment of refractory status epilepticus.
Epilepsia 52(11), e181–184.
O’Connor, S.E., Ream, M.A., Richardson, C., Mikati,
M.A., Trescher, W.H., Byler, D.L., Sather, J.D.,
Michael, E.H., Urbanik, K.B., Richards, J.L., et al.
(2014). The ketogenic diet for the treatment of
pediatric status epilepticus. Pediatr Neurol 50(1),
101–103.
Raspall-Chaure, M., Chin, R.F., Neville, B.G.,
Bedford, H., and Scott, R.C. (2007). The epi-
demiology of convulsive status epilepticus
in children:  a critical review. Epilepsia 48(9),
1652–1663.
Reyes, G., Flesler, S., Armeno, M., Fortini, S.,
Ariela, A., Cresta, A., Mestre, G., and Caraballo,
R.H. (2015). Ketogenic diet in patients with
epileptic encephalopathy with electrical status
epilepticus during slow sleep. Epilepsy Res 113,
126–131.
Scott, R.C. (2009). Status epilepticus in the developing
brain:  longterm effects seen in humans. Epilesia
50(Suppl. 12), 32e3.
Shorvon, S., and Ferlisi, M. (2011). The treatment
of super-refractory status epilepticus:  a critical
review of available therapies and a clinical treat-
ment protocol. Brain 134(Pt 10), 2802–2818.
Singh, R.K., Joshi, S.M., Potter, D.M., Leber, S.M.,
Carlson, M.D., and Shellhaas, R.A. (2014).
Cognitive outcomes in febrile infection-related
epilepsy syndrome treated with the ketogenic diet.
Pediatrics 134(5), e1431–1435.
Sort, R., Born, A.P., Pedersen, K.N., Fonsmark, L., and
Uldall, P. (2013). Ketogenic diet in 3 cases of child-
hood refractory status epilepticus. Eur J Paediatr
Neurol 17(6), 531–536.
Steriade, C., Andrade, D.M., Faghfoury, H.,
Tarnopolsky, M.A., and Tai, P. (2014).
Mitochondrial encephalopathy with lactic acidosis
and stroke-like episodes (MELAS) may respond
to adjunctive ketogenic diet. Pediatr Neurol 50(5),
498–502.
Sutter, R., Marsch, S., Fuhr, P., Kaplan, P.W., and
Rüegg, S. (2014). Anesthetic drugs in status epi-
lepticus:  risk or rescue? A  6-year cohort study.
Neurology 82(8), 656–664.
Sutter, R., and Kaplan, P.W. (2015). Can anesthetic
treatment worsen outcome in status epilepticus?
Epilepsy Behav 49, 294–297.
Thakur, K.T., Probasco, J.C., Hocker, S.E., Roehl, K.,
Henry, B., Kossoff, E.H., Kaplan, P.W., Geocadin,
R.G., Hartman, A.L., Venkatesan, A., et al. (2014).
Ketogenic diet for adults in super-refractory status
epilepticus. Neurology 82(8), 665–670.
Trinka, E., Cock, H., Hesdorffer, D., Rossetti, A.O.,
Scheffer, I.E., Shinnar, S., Shorvon, S., and
Lowenstein, D.H. (2015). A definition and classi-
fication of status epilepticus:  Report of the ILAE
Task Force on Classification of Status Epilepticus.
Epilepsia 56, 1515–1523.
Vaccarezza, M., Silva, W., Maxit, C., and Agosta, G.
(2012). [Super-refractory status epilepticus: treat-
ment with ketogenic diet in pediatrics]. Rev
Neurol 55(1), 20–25. Spanish.
Veggiotti, P., Pera, M.C., Teutonico, F., Brazzo, D.,
Balottin, U., and Tassinari, C.A. (2012). Therapy
of encephalopathy with status epilepticus
65

Chapter 9: Ketogenic Diet in Status Epilepticus 65

during sleep (ESES/CSWS syndrome): an update.
Epileptic Disord 14(1), 1–11.
Villeneuve, N., Pinton, F., Bahi-Buisson, N., Dulac, O.,
Chiron, C., and Nabbout, R. (2009). The ketogenic
diet improves recently worsened focal epilepsy.
Dev Med Child Neurol 51(4), 276–281.
Wusthoff, C.J., Kranick, S.M., Morley, J.F., and
Christina Bergqvist, A.G. (2010). The ketogenic
diet in treatment of two adults with prolonged
nonconvulsive status epilepticus. Epilepsia 51(6),
1083–1085.
66
10
Preventing Side Effects and Diet Discontinuation
C H E R I E L . H E R R E N, M D A N D R A NA R . S A I D, M D
INTRODUCTION
In general, the ketogenic diet (KD) is well toler-
ated. On average, 60% of patients remain on the
diet for over 6  months. Those who stop the diet
typically do so due to lack of efficacy rather than
tolerability (Keene et  al., 2006). Adverse effects
of KD therapy can be divided in to common and
serious. They can also be categorized according
to timing during KD therapy (during initiation,
during maintenance, or during discontinuation).
Common side effects include constipation, vom-
iting, acidosis, and vitamin/mineral deficiencies.
More significant side effects are rare, but include
pancreatitis, hepatitis, kidney stones, and cardio-
myopathy. During initiation, transient side effects
can include dehydration, hypoglycemia, acidosis,
vomiting, and refusal to eat. Children need to be
monitored closely during initiation so that these
adverse effects can be avoided and treated. Fasting
and fluid restriction may aggravate these side
effects and are not uniformly implemented in KD
programs. Fluid restriction contributes to dehydra-
tion and metabolic acidosis, which subsequently
can cause vomiting and lethargy. With appropriate
monitoring and supplementation, these adverse
effects can be minimized so the patient can remain
on the diet as long as indicated. In addition, there
may be social issues, including refusal to eat and
managing special occasions and holidays. With
support and resources, most families are able to
overcome these obstacles. When patients come off
dietary therapy, this must be done gradually and
with close supervision, as there may be an increase
in seizures. Patients must be provided support and
direction on how to safely discontinue the diet.
SCREENING
One of the most important steps in avoiding poten-
tial side effects (potentially severe or life threaten-
ing) is proper screening of patients. Patients with
disorders of fat metabolism could present with
severe and potentially fatal complications if started
on the KD, as the diet will require a shift from car-
bohydrate metabolism to fat metabolism for energy.
All patients should have a thorough metabolic
assessment prior to initiation of dietary therapy as
detailed in table 10.2. This screening should include
electrolytes, blood counts, lipids, acylcarnitine
panel, serum amino acids, urine organic acids, liver
and kidney function, uninalysis, and urine calcium
and creatinine (Kossoff et  al., 2009). Patients with
disorders of fat metabolism with impaired fatty acid
transport and oxidation should not be started on
the KD. There are other specific diagnoses for which
the KD should not be used including (but not lim-
ited to) pyruvate carboxylase deficiency, porphyria
and primary carnitine deficiency.
SOCIAL
To avoid complications it is imperative to screen
patients both medically and socially. Patients and
families should be adequately educated about
what to expect while on the diet and the follow-up
that will be needed (typically every 3–6 months).
For those unwilling or unable to comply with the
needed follow-up, careful consideration should be
taken to determine whether the diet is an appro-
priate treatment option. Once patients are started
on the diet, good social support can be beneficial
to families to maintain adherence and can increase
the likelihood that they continue on the diet. This
support often comes from medical staff including
neurologists, dietitians, and nurses who are well
educated on the diet and the demands it places on
families. Engagement and education of teachers
and schools is vital to the success of KD therapy in
school-age children. Reassurance and encourage-
ment can also come from support groups designed
specifically for families and KD-specific functions,
such as Halloween parties. Special occasions tend
to be especially difficult, but families may take
comfort in knowing there are numerous recipes
available for everything from ice cream to birth-
day cake! Some centers also use parent coaches,
67
Chapter 10: Preventing Side Effects and Diet Discontinuation
TABLE 10.1 TIME PATIENT REMAINS ON KETOGENIC DIET (KEENE, 2006)
Author Total sample % Sample at 3 months
DiMario [2002] 48 ? 50
Coppola [2002] 56 75
Maydell [2001] 147 80
Hassan [1999] 52 68
Kankirawatana [2001] 35 62
Kang [2005] 199 88
Vining [1998] 51 88
Freeman [1998] 150 83
Ruthenstein [2005] 13 85
Kossoff [2002] 23 91
who can help families navigate issues such as food
refusal and menu ideas in real life.
VO M I T I N G
Vomiting is one of the more common side effects
seen in patients on the KD and is estimated to
occur in 5% of patients (Keene et al., 2006). This is
most commonly seen during the initiation phase
of the diet. Metabolic acidosis may contribute to
vomiting, and in turn, vomiting may worsen aci-
dosis, creating a worsening cycle of symptoms.
In many cases reversing metabolic acidosis will
alleviate nausea and vomiting. In patients who are
experiencing vomiting, maintaining proper hydra-
tion, either enterally or IV, is an important factor
in breaking this cycle. Initiation is a common time
for patients to experience vomiting related to the
diet itself; however, vomiting can occur for other
reasons (infectious, etc.) at any time while on the
diet. Families and providers should be watchful
for vomiting and have a low threshold to provide
intervention. In those patients who cannot ade-
quately maintain oral hydration due to vomiting or
in those who have increasing lethargy, intravenous
fluids may be required. Monitoring for constipa-
tion is also important, as children with constipa-
tion may be predisposed to vomit.
Families often are concerned about whether to
re-dose antiseizure medications with emesis; we
advise our parents to repeat the dose if vomiting
occurred less than 30 minutes from dose admin-
istration. Management of constipation with poly-
ethylene glycol is well tolerated. Some children
require slower delivery of feeds, either with more
frequent, smaller meals or, if gastrostomy tube fed,
with slower rates of infusion with a feeding pump.
For short-term vomiting around times of illness,
use of Powerade Zero or Propel Zero can provide
hydration and electrolytes, without carbohydrates.
67
% Sample at 6 months % Sample at 12 months
37
38 7
66 48
39 13
57 12
61 27
69 47
71 55
77 50
78 56
In instances of more prolonged feeding intoler-
ance, half strength Pedialyte to provide a source of
calories is usually well tolerated, without signifi-
cant reduction in ketosis.
ACIDO SIS
Metabolic acidosis can be found in patients on
the KD and can present with lethargy and vom-
iting. This will typically present during initia-
tion and be minimized with the use of buffering
agents such as sodium bicarbonate or Polycitra K
(Cypress Pharmaceuticals, Madison, MS, USA).
Practitioners should be aware of medications
that could worsen acidosis, especially common
antiseizure medications, such as topiramate and
zonisamide. In some cases, it may be necessary
to decrease these medications in order to man-
age acidosis. During initiation, acidosis should be
treated with aggressive hydration and use of oral
alkalinizing agents. Rarely would IV alkalization
be indicated. Once the patient is on a stable diet
ratio, with adequate doses of alkalinizing agents,
acidosis becomes less problematic. However, dur-
ing times of illness, with decreased oral intake and/
or vomiting, acidosis can again become problem-
atic and typically can be managed with hydration.
C O N S T I PAT I O N
Given the composition of the KD, an increase
in constipation would not be unexpected.
Constipation is common and occurs in up to 65%
of patients (Wibisono et al., 2015). This can be par-
ticularly problematic in those who had preexisting
constipation, and for those who have impaired
mobility. Increased hydration may help with
symptoms, but many patients will require medica-
tions to manage their constipation. Polyethylene
glycol (Miralax) is a common choice for treatment
as it is not significantly absorbed in the gut and is
68
68 section I: Ketogenic Diet for Epilepsy in the Clinic
safe for patients on the KD. Glycerin suppositories,
increased dietary fiber, medium chain triglyceride
(MCT) oil, or mineral oil could also be considered.
H Y P O G LY C E M I A
Like many other potential side effects, the risk
for hypoglycemia tends to be highest during the
initiation phase of the diet. If patients have severe
hypoglycemia (<30 mg/dL) or symptomatic hypo-
glycemia (<40 mg/dL with lethargy, vomiting, dia-
phoresis, seizures, or shakiness), small amounts
of juice (10–20 mL) or dextrose (D5W infusion
until blood glucose is >60 mg/dL) can be given
to correct the blood sugar. If hypoglycemia is a
persistent problem, a reduction in ratio should
be considered. Once stable on the diet, glucose
tends to be on the low end of normal (50–70 mg/
dL), but is overall very stable. Fasting prior to ini-
tiation of KD therapy may increase the chances of
hypoglycemia (Bergqvist et  al., 2005)  and there-
fore close glucose monitoring in these patients is
recommended.
K I D N E Y   S TO N E S
Renal calculi have been reported in 3%–7% of chil-
dren on the KD (Kielb et al., 2000). These typically
form, on average, after 18 months of KD therapy.
However, there have been reports of development
of urolithiasis after just 1 month on the diet. The
majority of stones are uric acid stones, however,
calcium oxalate, calcium phosphate, and mixed
calcium/uric acid stones are seen.
Approximately 50% of stones are uric acid
stones (Kielb et  al., 2000). This is likely due to
KD lowering urinary pH, which facilitates forma-
tion of uric acid crystals, as the solubility of uric
acid decreases as pH drops. The KD also has been
associated with uric acidemia and increased uri-
nary uric acid, which further predisposes to the
development of uric acid stones. This can be com-
pounded by fluid restriction, which produces a
more acidic urine and decreased urine flow, with
precipitation of urate crystals.
The KD has also been associated with hyper-
calciuria, which can result in calcium crystal for-
mation. Hypocitraturia also occurs due to chronic
metabolic acidosis. Urinary citrate is an inhibitor
of calcium crystal formation; therefore, low uri-
nary levels increase the risk of calcium stone for-
mation. Fluid restriction serves to increase this
risk (Herzberg, 1990).
Children typically present with gross or micro-
scopic hematuria, therefore patients require reg-
ular screening urinalysis. They may also, less
commonly, have gritty urine or pain. There may
be a role for periodically checking serum uric acid
and urinary calcium/creatinine ratios. Renal ultra-
sound is necessary if there is evidence of hema-
turia. Children with family histories of kidney
stones receiving carbonic anhydrase inhibitors
(topiramate, zonisamide, or acetazolamide) are
also at higher risk for stones (Furth et al., 2000 and
Kossoff, 2002). Treatment includes fluid liberali-
zation and urinary alkalization with bicarbonate
(Sampath, 2007). In children with a higher risk for
kidney stone formation, prophylactic treatment
with Polycitra K has been shown to reduce the
incidence of kidney stones in this susceptible pop-
ulation (McNally et al., 2009).
INCREASED INFECTIONS
Several studies have reported rates of increased
infections in up to 2%–4% of children on the KD
(Woody et  al., 1989; Vining et  al., 1998; Summ
et  al., 1996). Woody et  al. assessed neutrophil
function in nine children on the KD, demon-
strating that while ketotic, these patients had
significantly less bacterial phagocytosis and kill-
ing. These effects were reversible, with resolution
when the diet was discontinued (Woody et  al.,
1989). Other conditions associated with ketosis,
such as diabetes mellitus, alcoholism, glycogen
storage disease, protein-calorie malnutrition,
certain carbohydrate-restricted diets, and intra-
lipid infusions, have also reported impairments
in neutrophil function. The exact mechanism is
not well understood but is likely related to serum
metabolites that affect early processes in phagocy-
tosis (Woody et al., 1989). Interestingly, only one
of Woody’s patients experienced serious bacterial
infections.
V I TA M I N D E F I C I E N C I E S
Due to the restrictive nature of the diet (with
limited vegetables and grains), it is known to be
deficient in several vitamins and minerals, espe-
cially B vitamins, calcium, and vitamin D.  The
International Ketogenic Study Group (Kossoff
et  al., 2009)  made it a universal recommenda-
tion that all patients on the KD should be on a
multivitamin with minerals and receive calcium
with vitamin D supplements. In the study, some
did suggest additional supplementation may
be needed for zinc, magnesium, selenium, and
phosphorous, but this was not a universal rec-
ommendation, and may be considered on an as-
needed basis depending on laboratory values. Care
should be taken to select a carbohydrate-free or
69
Chapter 10: Preventing Side Effects and Diet Discontinuation
low-carbohydrate multivitamin such as NanoVM
(Solace Nutrition, Rockville, MD, USA), FruitiVits
(Vitaflo, Alexandria, VA, USA), Centrum (Wyeth,
Madison, NJ, USA), or Bugs Bunny Sugar Free
(Bayer, Morristown, NJ, USA). Vitamin lev-
els should be followed on a routine basis during
follow-up, and early supplementation to avoid
deficiencies is optimal.
E L E C T R O LY T E I M B A L A N C E S
Electrolytes such as calcium, magnesium, sodium,
potassium, and phosphorus are often depleted or
deficient in children on KD therapy. These labo-
ratory values must also be monitored closely and
adequately supplemented. In children on KD for-
mulas, many of these electrolytes may be included
in the formula. However, children on solid diets
will often need additional supplements. These
can be in the form of dietary supplements (such
as Morton’s Lite Salt) or prescription (K Phos
Neutral).
HYPERLIPIDEMIA
Studies show that between 30% and 60% of
patients on the KD develop hypercholesterolemia
(Wibisono et  al., 2015; Nizamuddin et  al., 2008).
Nizamuddin found this was increased from 25%
at preinitiation (baseline). Notably, patients eat-
ing a solid diet had a greater risk for developing
high cholesterol compared with those who con-
sumed a formula-based diet, likely secondary to an
increase in saturated fats in solid foods compared
with the liquid formulation. The dyslipidemia seen
improved spontaneously and without interven-
tion in about half of patients, suggesting patients
are better able to metabolize the fat over time
(Nizamuddin et al., 2008). It is unclear whether the
dyslipidemia seen with the diet has any long-term
cardiovascular or atherosclerotic effects, but given
the temporary use of the diet it appears unlikely
and the benefit of the diet outweighs the potential
risk (Kwitwerovich, 2003). Currently, in children,
there are no universal recommendations for inter-
vention for elevated cholesterol or triglyceride
levels.
PA N C R E AT I T I S
Hyperlipidemia and hypertriglyceridemia are risk
factors for development of pancreatitis, which
are also potential complications of KD therapy.
Concomitant therapy with valproic acid (VPA)
may increase this risk. There has been a case
report of a 9-year-old who died from acute hem-
orrhagic pancreatitis while on KD therapy. The
69
development of pancreatitis while on the KD
would necessitate coming off the diet (Stewart
et al., 2001). Parenteral KD therapy with intralip-
ids also places children at high risk for developing
pancreatitis and should be used judiciously.
H E PAT I T I S
While rare, hepatitis is a potentially significant
side effect of the KD. It has been suggested that
concomitant use of VPA may increase the risk
of hepatitis, however Kang at al. (2004) did not
find a statistically significant increase in hepatitis
for those on VPA, compared with those not tak-
ing this antiseizure medication. Decreased car-
nitine levels may also increase the risk for liver
dysfunction, however the risk appears to be low
(Berry-Kravis et  al., 2001). Nonetheless, rou-
tine monitoring of serum carnitine and carnitine
supplementation is recommended. Regular moni-
toring of liver enzymes is recommended, and in
the event that elevations are noted, hepatic ultra-
sound and comanagement with gastroenterol-
ogy should be considered, including possible diet
discontinuation.
CARDIAC SIDE EFFECTS
Cardiac complications have been reported in
children on the KD, including cardiomyopathy
and prolonged QT interval (Neal, 2012). These
risks may be related to carnitine and selenium
deficiencies associated with the KD. Bergqvist
et  al. (2003) reported a child with cardiomyopa-
thy in association with selenium deficiency and
other children who had lowering of selenium lev-
els while on the KD. In a study of 20 children on
the KD, 15% (three children) had prolongation of
the QTc interval. Of these three children, two had
preinitiation electrocardiograms (EKGs), which
confirmed that the QTc abnormalities developed
while they were on the diet. Two of the patients
had left atrial and left ventricular enlargement,
and one had severe ventricular dilatation and dys-
function, with associated symptoms of heart fail-
ure. Interestingly, all of the patients in this series
had normal selenium levels. It was postulated that
higher beta-hydroxybuturate levels and more sig-
nificant metabolic acidosis may be associated with
the development of cardiac complications.
There are also reports of two children develop-
ing severe dilated cardiomyopathy while on the KD
(Ballaban-Gil et al., 1999). In these cases, the car-
diomyopathy resolved following discontinuation
of the KD. It is suggested that baseline EKGs and
echocardiograms be completed prior to initiation
70
70 section I: Ketogenic Diet for Epilepsy in the Clinic
of the KD and while on the diet, including carnitine
and selenium levels. Supplementation with carni-
tine and selenium should be strongly considered.
G ROW T H
There are conflicting reports of how the KD affects
growth—both linear and weight—in children on
the diet. Williams et  al. (2002) retrospectively
reviewed 21 children treated with the KD for
9.6  months to 2  years, specifically looking at lin-
ear growth velocity. Eighteen children (86%) had
a drop from their original height percentiles while
on the diet. This was independent of mean age,
duration on the diet, protein intake, or calories
consumed per body weight while on the diet. They
concluded that decreased linear growth might be
secondary to poor nutritional status. While these
children received the recommended nutrient
intake for protein, calories were typically restricted
to 75% of recommended daily intake. The authors
postulated that dietary protein was needed to fill
the calorie gap for energy and gluconeogenesis,
resulting in relative protein insufficiency to sup-
port growth (Williams et al., 2002).
Vining et al. prospectively reviewed the growth
of 237 children treated with the KD. They found
that older children grew taller “almost normally”
but younger children fell more than two standard
deviations below the mean in height (Vining et al.,
2002). However, Couch et  al. (1999) retrospec-
tively evaluated the nutritional status and growth
of 21 children on the KD for 6 months. They found
a statistically significant increase in both height
and weight after 6  months on the diet. Six chil-
dren had a decrease in mean percentile standard
weight for height, but no one fell below their pre-
KD height for age percentile after 6 months on the
diet. However, the study was limited by the short
duration of follow-up (Couch et al., 1999).
Children on the KD need to be monitored
closely for adequate protein intake and growth
parameters. Often children with intractable epi-
lepsy become more physically active as seizures
improve, and energy needs must be recalculated.
Protein intake must be adequate and regular labo-
ratory studies to assess protein, albumin, and pre-
albumin are necessary. Lack of weight gain can be
commonly seen; however, inadequate or slowed
growth (which is more problematic in infants) and
inappropriate weight loss occur less frequently.
OSTEOPENIA
Baseline bone mineral content may be low in
patients starting the KD, with only an estimated
54%–63% of patients receiving adequate calcium
and vitamin D through their preinitiation diet
(Bergqvist et  al., 2008). The poor bone health
seen may be secondary to antiseizure medication
(ASM) use, which can interfere with calcium and
vitamin D metabolism and directly impacts the
cells of bone formation and absorption. Despite
adequate supplementation with calcium and vita-
min D, and an overall reduction in ASMs, once
patients were started on the KD there was further
decline in bone mineral content. Chronic acido-
sis may contribute to the decline in bone health,
and buffering agents such as sodium bicarbonate
or sodium citrate/citric acid may be beneficial
in reducing the risk of osteopenia and fractures.
Bone density scans may not be indicated for all
patients, but could be considered in those at
higher risk (younger nonambulatory patients)
or for those who have experienced pathologic
fractures.
FOOD REFUSAL
Tolerability of the KD is the single most important
factor limiting individual acceptance for initiation,
however, 60% of children who start the KD remain
on it at 6 months follow-up (Table 10.1). The most
common reason for discontinuation of the diet is
not food refusal or concerns with palatability of
the diet, but lack of efficacy.
Food refusal is typically related to a limited
repertoire of foods. Families are encouraged to
keep several menus on rotation to minimize food
fatigue. Even changing presentation of meals can
help with tolerability; for example, one family used
a child’s tea service, with its smaller plates to make
meals more enjoyable. Close communication with
the KD registered dietitian is vital as they have a
wealth of suggestions for improved food intake.
Parent KD coaches are also a tremendous resource.
Food refusal may also represent a food intolerance;
changing to a lactose-free formula may be helpful.
If food refusal becomes a significant battle, reduc-
ing the KD ratio or changing to an MCT-based
diet to increase amounts of carbohydrates may
be necessary. At times, changing to other dietary
therapies, such as the modified Atkins diet or low
glycemic index therapy, is warranted.
DISCONTINUING
T H E   K E TO G E N I C   D I E T
There are many reasons why the KD is discontin-
ued, the most common being lack of efficacy after
3–6 months or if seizures worsen. After 2 years of
effective use of the KD, most neurologists would
consider discontinuation. Gradually lowering
the ratio of fat to protein and carbohydrate, then
71

Chapter 10: Preventing Side Effects and Diet Discontinuation 71

TABLE 10.2 RECOMMENDATIONS FOR PRE- KD EVALUATION (KOSSOFF ET AL, 2008)

Counseling
  Discuss seizure reduction, medication, and cognitive expectations
  Identify potential psychosocial barriers to the use of KD
  Review anticonvulsants and other medications for carbohydrate content
  Recommend family read parent-oriented KD information
Nutritional evaluation
  Baseline weight, height, and ideal weight for stature
  Body mass index (BMI) when appropriate
  Nutrition intake history: 3-day food record, food preferences, allergies, aversions, and intolerances
  Establish diet formulation: infant, oral, enteral, or a combination
  Decision on which diet to begin (MCT, classic, modified Atkins, or low glycemic index)
  Calculation of calories, fluid, and ketogenic ratio (or percentage of MCT oil)
  Establish nutritional supplementation products based on dietary reference intake
Laboratory evaluation
  Complete blood count with platelets
  Electrolytes to include serum bicarbonate, total protein, calcium, zinc, selenium, magnesium, and phosphate
  Serum liver and kidney tests (including albumin, AST, ALT, blood urea nitrogen, creatinine)
  Fasting lipid profile
  Serum acylcarnitine profile
  Urinalysis
  Urine calcium and creatinine
  Anticonvulsant drug levels (if applicable)
  Urine organic acids
  Serum amino acids
Ancillary testing (optional)
  Renal ultrasound and nephrology consultation (if a history of kidney stones)
  EEG
  MRI
Cerebrospinal fluid (CSF) (if no clear etiology has been identified)
EKG (echocardiogram) if history of heart disease

relaxing the weighing of ingredients, and finally
adding new carbohydrate foods while keeping
calories constant is typical. This is followed by
relaxation of calorie restriction, then gradual intro-
duction of carbohydrate-containing foods. The
ideal weaning speed of the KD is not clear, specifi-
cally as it pertains to the risk of increased seizures.
Worden et al. (2011) retrospectively reviewed 183
children who discontinued the KD. There was no
significant difference in the incidence of seizures
worsening between immediate (<1 week), quick
(1–6 weeks), or slow (>6 weeks) rates. However,
there was an increased risk of seizures worsening
in those patients specifically with a 50%–99% sei-
zure reduction and those who were receiving more
antiseizure medications (Worden et al., 2011).
Approximately 80% of children will remain
seizure-free when tapered off the KD if they have
completely responded to diet therapy and are
medication free. Factors associated with a higher
risk of recurrence include an epileptiform elec-
troencephalogram (EEG), focal abnormalities on
neuroimaging, and tuberous sclerosis complex.
Approximately 50% of children with recurrence
are difficult to control once again, even with
resumption of the KD (Martinez et al., 2007).
CONCLUSION
While the KD is an effective therapy for refractory
epilepsy and certain metabolic disorders, there are
potential adverse effects. Providers must be aware
of these risks for proper monitoring of children
and counseling families. Like all medical therapies,
potential risks of treatment are weighed against
benefits of the diet.
REFERENCES
Ballaban-Gil, K.R. (1999). Cardiomyopathy associated
with the ketogenic diet. Epilepsia 40(Suppl 7), 129.
72
72 section I: Ketogenic Diet for Epilepsy in the Clinic
Berry-Kravis, E., Booth, G., Sanchez, A.C., and
Woodbury-Kolb, J. (2001). Carnitine levels and
the ketogenic diet. Epilepsia 42, 1445–1451.
Bergqvist, A.G.C, Chee, C.M., Lutchka, L., Rychik, J.,
and Stallings, V.A. (2003). Selenium deficiency
associated with cardiomyopathy:  a complication
of the ketogenic diet. Epilepsia 44, 618–620.
Bergqvist, A.G.C, Schall, J.I., Gallagher, P.R., Cnaan, A.,
Stallings, V.A. (2005). Fasting versus gradual ini-
tiation of the ketogenic diet:  a prospective, ran-
domized clinical trial of efficacy. Epilepsia 46,
1810–1819.
Bergqvist, A.G.C, Schall, J.I., Stallings, V.A., and
Zemel, B.S. (2008). Progressive bone mineral loss
in children with intractable epilepsy treated with
the ketogenic diet. Am J Clin Nutr 88, 1678–1684.
Coppola, G., Veggiotti, P., Cusmai, F. et al. (2002). The
ketogenic diet in children, adolescents and young
adults with refractory epilepsy:  An Italian multi-
centric experience. Epilepsy Res. 48, 221–227.
Couch, S.C., Schwarzman, F., Carroll, J., Koeniqsberger,
D., Nordli, D.R., Deckelbaum, R.J., and DeFelice,
A.R. (1999). Growth and nutritional outcomes of
children treated with the ketogenic diet. J Am Diet
Assoc 99, 1573–1575.
DiMario, F.J.Jr, Holland, J. (2002). The Ketogenic
diet: A review of the experience at the Connecticut
Children’s Medical Center, Pediatr Neurol. 26,
288–292.
Freeman, J., Vining, E., Pillas, D., Pyzik, P., Casey, J.,
Kelly, M. (1998). The efficacy of the ketogenic diet–
1998: A prospective evaluation of the intervention
in 150 children. Pediatrics, 102, 1358–1363.
Furth S.I., Casey J.C., Pyzik P.L., Neu A.M., Docimo,
S.G., Vining, E.P., Freeman, J.M., and Fivush,
B.A. (2000). Risk factors for urolithiasis in chil-
dren on the ketogenic diet. Pediatr Nephrol 15,
125–128.
Hassan, A., Keene, D., Whiting, S., Jacob, P.,
Champagne, J., Humphreys, P. (1999). Ketogenic
diet in the treatment of refractory epilepsy in
childhood. Pediatr Neurol, 21, 548–552.
Herzberg, G.Z., Fivush, B.A., Kinsman S.L., and
Gearhart J.P. (1990). Urolithiasis associated with
the ketogenic diet. J Pediatr 117, 743–745.
Kang, H.C., Chung, D.E., Kim, D.W., and Kim, H.D.
(2004). Early and late-onset complications of the
ketogenic diet for intractable epilepsy. Epilepsia
45, 1116–1123.
Kang, H., Kim, Y., Kim, D., Kim, H. (2005). Efficacy and
safety of the ketogenic diet for intractable child-
hood epilepsy:  Korean Multicentric Experience.
Epilepsia, 46, 272–279.
Kankirawatana, P., Jirapinyo, P., Kankirawatana, S.,
Wongarn, R., Thamasasiri, N. (2001). Ketogenic
diet:  An alternative treatment for refractory
epilepsy of childhood. J Med Assoc Thai, 84,
1027–1032.
Keene, D.L. (2006). A systematic review of the use of
the ketogenic diet in childhood epilepsy. Pediatr
Neurol 35, 1–5.
Kielb, S., Koo, H.P., Bloom, D.A., and Faerber, G.J.
(2000). Nephrolitiasis associated with the keto-
genic diet. J Urol 164, 464–466.
Kossoff, E.H., Pyzik, P.I., Furth, S.L., Hladky, K.D.,
Freeman, J.M., and Vining, E.P.G. (2002). Kidney
stones, carbonic anhydrase inhibitors, and the
ketogenic diet. Epilepsia 43, 1168–1171.
Kossoff, E., Pyzik, P., McGrogan, J., Vining, E.,
Freeman, J. Efficacy of the ketogenic diet for infan-
tile spasms. Pediatrics 109 (2002), pp. 780–783.
Kossoff, E.H., Zupec-Kania, B.A., Amark, P.E.,
Ballaban-Gil, K.R., Bergqvist, A.G., Blackford, R.,
Buchhalter, J.R., Baraballo, R.H., Cross, J.H.,
Dahlin, M.G., et  al. (2009). Optimal clinical
management of children receiving the keto-
genic diet: recommendations of the International
Ketogenic Diet Study Group. Epilepsia 50,
304–317.
Martinez, C.C., Pyzik, P.I., and Kossoff, E.H. (2007).
Discontinuing the ketogenic diet in seizure-free
children: risk factors and recurrence. Epilepsia 48,
187–190.
Maydell, B., Wyllie, E., Akhtar, H. et al. (2001). Efficacy
of the ketogenic diet in focal versus generalized
seizures. Pediatr Neurol 25, 208–212.
McNally, M.A., Pyzik, P.L., Rubenstein, J.E., Hamdy,
R.F., and Kossoff, E.H. (2009). Empiric use of
potassium citrate reduces kidney stone incidence
with ketogenic diet. Pediatrics 124, e300–304.
Neal, E.G., Zupec-Kania, B., and Pfeifer, H.H. (2012).
Carnitine, nutritional supplementation and dis-
continuation of ketogenic diet therapies. Epilepsy
Res 100, 267–271.
Nizamuddin, J., Turner, Z., Rubenstein, J.E., Pyzik,
P.L., and Kossoff, E.H. (2008). Management and
risk factors of dyslipidemia with ketogenic diet.
J Child Neurol 23, 758–761.
Ruthenstein, J., Kossoff, E., Pyzik, P., Vining, E.,
McGrogan, J., Freeman, J. (2005). Experience in
the use of the ketogenic diet as early therapy. J Chil
Neurol, 20, 31–34
Sampath, A., Kossoff, E.H., Furth, P.L., Pyzik, P.L., and
Vining, E.P. (2007). Kidney stones and the keto-
genic diet:  risk factors and prevention. J Child
Neurol 22, 375–378.
Stewart, W.A., Gordon, K., and Camfield, P. (2001).
Acute pancreatitis causing death in a child on the
ketogenic diet. J Child Neurol. 16, 682.
Summ, M., Woch, M.A., and McNeil, T. (1996).
Success and complications of the ketogenic diet for
intractable childhood epilepsy. Epilepsia 37 (Suppl
5), 109.
Vining, E., Freeman, J., Ballaban-Gil, K. et al. (1998).
A multicenter study of the efficacy of the ketogenic
diet. Arch Neurol, 55, pp. 1433–1437.
73
Chapter 10: Preventing Side Effects and Diet Discontinuation
Vining, E.P., Freeman, J.M., Ballaban-Gil,
K.,Camfield, C.S., Camfield, P.R., Holmes, G.L.,
Shinnar, S., Shuman, R., Trevathan, E., Wheless,
J.W. (1998). A multicenter study of the effi-
cacy of the ketogenic diet. Arch Neurology 55,
1433–1437.
Vining, E.P., Pyzik, P., McGrogan, J., Hladky, H.,
Anand, A., Kriegler, S., and Freeman, J.M. (2002).
Growth of children on the ketogenic diet. Dev
Med Child Neurol. 44, 796–802.
Wibisono, C., Rowe, N., Beavis, E., Kepreotes, H.,
Mackie, F.E., Lawson, J.A., and Cardamone,
M. (2015). Ten-year single-center experi-
ence of the ketogenic diet:  factors influencing
73
efficacy, tolerability, and compliance. J Pediatr 166,
1030–1036.
Williams, S., Basualdo-Hammond, C., Curtis, R., and
Schuller, R. (2002). Growth retardation in children
with epilepsy on the ketogenic diet: a retrospective
chart review. J Am Diet Assoc 102, 405–407.
Woody, R.C., Steele, R.W., Knapple, W.L., and
Pilkington, N.S., Jr. (1989). Impaired neutrophil
function in children with seizures treated with the
ketogenic diet. J Pediatr 115, 427–430.
Worden, L.T., Turner, Z., Pyzik, P.L., Rubenstein, J.E.,
and Kossoff, E.H. (2011). Is there an ideal way to
discontinue the ketogenic diet? Epilepsy Res 95,
232–236.
74
75

SECTION II
Ketogenic Diet

Emerging Clinical Applications

and Future Potential

JONG M. RHO, MD, SECTION EDITOR

76
77
11
Overview
Expanded Uses of Ketogenic Therapies
JONG M. RHO, MD
T he ketogenic diet (KD) is a proven therapy
for drug-resistant epilepsy (Vining et al.,
1998; Neal et al., 2008), and while the mecha-
nisms underlying its antiseizure effectiveness
remain unclear (Masino and Rho, 2012), there is
growing experimental evidence for the neuropro-
tective properties of this metabolism-based treat-
ment (Barañano and Hartman, 2008; Maalouf et
al., 2009; Gano et al., 2014). Although definitive
clinical evidence for use outside of epilepsy is
presently lacking, there are nevertheless ongoing
human trials for a few neurological conditions
(www.ClinicalTrials.gov) assessing whether a
ketogenic therapy can ameliorate symptoms and
even the disease processes themselves. Much of
the experimental evidence has been converging
on the notion that the broad efficacy of the KD
is due in major part to normalization of aberrant
energy metabolism (Masino and Rho, 2012). The
concept that multiple neurological disorders are
pathophysiologically linked to energy dysregula-
tion (Pathak et al., 2013) has been strengthened
by numerous studies evaluating brain bioenerget-
ics, and provides a solid rationale for how dietary
treatments that ameliorate underlying dysfunction
can treat or even prevent disease symptomatology.
In this section, “Ketogenic Diet—Emerging
Clinical Applications and Future Potential,” lead-
ing investigators in the field of ketogenic therapies
provide up-to-date reviews summarizing the evi-
dence that the KD may prove useful in the treat-
ment of a number of neurological conditions. Drs.
Thomas Seyfried, Adrienne Scheck, and colleagues
put forward a compelling case for cancer as being
in part a metabolic (and perhaps more specifically,
a mitochondrial) disorder, and that the KD should
be strongly considered as adjunctive treatment
for malignant brain cancer (Seyfried and Shelton,
chapter 12; Woolf and Scheck, chapter 13). These
chapters are followed by a comprehensive review
by Dr.  Ning Cheng and co-authors (chapter  14)
of metabolic treatments for autism spectrum dis-
order, whose prevalence has been rising rapidly
throughout the world. Next, Dr. Stephen Cunnane
and collaborators (chapter  15) provide a unique
translational perspective on brain metabolism—
using novel positron emission tomography tech-
niques to make the case that nutritional ketosis,
achieved through several currently available strate-
gies, can correct the underlying problem of declin-
ing brain fuels during aging. Dr. Wolf Tetzlaff and
coauthors (Streijger et al., chapter 16) then review
the evidence that the KD (and in particular, ketone
bodies) can prevent the metabolic, oxidative, and
inflammatory cascades that follow after spinal
cord injury and traumatic brain injury. The next
chapter, by Drs. Nina Dupuis and Stéphane Auvin
(chapter 17), reviews the mounting evidence that
the KD possesses anti-inflammatory properties,
relevant not only to the pathophysiology of pain
syndromes and multiple sclerosis but also to other
neurological disorders. Finally, Dr. Carl Stafstrom
(chapter 18) covers the current evidence that the
KD may also afford beneficial effects against two
neurodegenerative disorders (amyotrophic lateral
sclerosis and Parkinson’s disease), mood disorders,
and migraine. Throughout all of these apparently
distinct neurological conditions, it should become
clear to the reader that metabolic derangements
are now considered key to the pathophysiology of
these disorders, and that a dietary treatment such
as the KD can potentially benefit patients afflicted
with any one of several neurological diseases. It
is also the hope of the aforementioned authors
that those curious about how altered metabolism
can both induce and mitigate neurodegeneration
and brain dysfunction will be compelled to pur-
sue further laboratory and/or clinical research in
this growing and fascinating area of translational
neurosciences.
78
78 section II: E merging Clinical Applications and Future Potential
REFERENCES
Barañano, K.W., and Hartman, A.L. (2008). The
ketogenic diet: uses in epilepsy and other neuro-
logic illnesses. Curr Treat Options Neurol 10(6),
410–419.
Clinical Trials. NCT02046187, NCT01865162,
NCT02477904, NCT02174016, NCT01915433.
www.ClinicalTrials.gov
Gano, L.B., Patel, M., and Rho, J.M. (2014). Ketogenic
diets, mitochondria, and neurological diseases. J
Lipid Res 55(11), 2211–2228.
Maalouf, M., Rho, J.M., and Mattson, M.P. (2009). The
neuroprotective properties of calorie restriction,
the ketogenic diet, and ketone bodies. Brain Res
Rev 59(2), 293–315.
Masino, S.A., and Rho, J.M. (2012). Mechanisms of keto-
genic diet action. In Jasper’s Basic Mechanisms of
the Epilepsies [Internet], 4th edition, J.L. Noebels,
M. Avoli, M.A. Rogawski, R.W. Olsen, and A.V.
Delgado-Escueta, eds. (Bethesda, MD:  National
Center for Biotechnology Information).
Neal, E.G., Chaffe, H., Schwartz, R.H., Lawson, M.S.,
Edwards, N., Fitzsimmons, G., Whitney, A., and
Cross, J.H. (2008). The ketogenic diet for the treat-
ment of childhood epilepsy:  a randomised con-
trolled trial. Lancet Neurol 7(6), 500–506.
Pathak, D., Berthet, A., and Nakamura, K. (2013).
Energy failure: does it contribute to neurodegen-
eration? Ann Neurol 74(4), 506–516.
Vining, E.P., Freeman, J.M., Ballaban-Gil, K., Camfield,
C.S., Camfield, P.R., Holmes, G.L., Shinnar, S.,
Shuman, R., Trevathan, E., and Wheless, J.W.
(1998). A multicenter study of the efficacy of the
ketogenic diet. Arch Neurol 55(11), 1433–1437.
79
12
Metabolism-Based Treatments to Counter Cancer
Scientific Rationale
T H O M A S N . S E Y F R I E D , P H D A N D L AU R A M . S H E LT O N, P H D
INTRODUCTION
Cancer has long been recognized as a genetic
disease involving mutations in oncogenes and
tumor suppressor genes that reside in the tumor
cell nucleus. The nuclear gene mutations found in
nearly all types of tumors are considered the pri-
mary cause of the cancer’s hallmarks, including
(1)  sustained proliferative signaling, (2)  evasion
of growth suppressors, (3) resistance to cell death,
(4)  replicative immortality, (5)  enhanced angio-
genesis, and (6) activation of invasion and metas-
tasis (Hanahan and Weinberg, 2000, 2011). The
somatic mutations in tumor cells arise randomly
during DNA replication in normal nontumori-
genic stem cells and are thought to be the origin of
cancer (Tomasetti and Vogelstein, 2015).
The somatic mutation theory is the most
widely accepted explanation for the origin of can-
cer, and is the justification for developing targeted
or personalized approaches for treating various
forms of the disease (Hou and Ma, 2014; McLeod,
2013; Seyfried, 2015; Vaux, 2011). Despite numer-
ous inconsistencies associated with the somatic
mutation theory (Baker and Kramer, 2007;
Burgio and Migliore, 2015; Cao et al., 2015; Rous,
1959; Sonnenschein and Soto, 2000; Soto and
Sonnenschein, 2004), this theory is presented as if
it were dogma in most current college textbooks of
genetics, biochemistry, and cell biology, and is the
mainstay of the National Cancer Institute’s position
which states, “Cancer is a genetic disease—that is,
it is caused by changes to genes that control the
way our cells function, especially how they grow
and divide” (http://www.cancer.gov/cancertopics/
what-is-cancer) (Seyfried, 2015). It should be rec-
ognized, however, that there is no place in science
for epistemological dogmatism, as dogma tran-
quilizes creative thinking and blocks consideration
of viable alternatives (Rous, 1959; Sonnenschein
and Soto, 2000).
IS CANCER A GENETIC
DISEASE OR A
M I TO C H O N D R I A L
M E TA B O L I C D I S E A S E ?
As an alternative to the somatic mutation theory,
emerging evidence suggests that cancer is primar-
ily a mitochondrial metabolic disease (Bartesaghi
et al., 2015; Hu et al., 2012; Seyfried, 2015; Seyfried
et al., 2014; Seyfried and Shelton, 2010; Verschoor
et  al., 2013). This concept originated from the
experiments of Otto Warburg (Burk et  al., 1967;
Warburg, 1956a, 1956b). According to Warburg’s
theory, respiratory insufficiency is the origin of can-
cer. All other phenotypes of the disease, including
the somatic mutations and aerobic fermentation,
arise either directly or indirectly from insufficient
respiration (Seyfried, 2012a; Seyfried et al., 2014;
Warburg, 1956a). Warburg’s metabolic theory was
also in line with the concepts of C. D. Darlington
and others showing that cancer is largely a cyto-
plasmic mitochondrial disease (Darlington, 1948;
Seeger, 1959; Seyfried, 2012a, 2015; Seyfried
and Shelton, 2010; Woods and du Buy, 1945).
Advocates of the somatic mutation theory, how-
ever, consider the abnormal energy metabolism of
tumor cells as simply another phenotype that “is
programmed by proliferation-inducing oncogenes
and defective tumor suppressor genes” (Hanahan
and Weinberg, 2011; Seyfried, 2015). The gene
theory of cancer, however, is inconsistent with the
data obtained from the nuclear-cytoplasm transfer
experiments.
A recent review of data from a broad range of
nuclear-cytoplasm transfer experiments provides
the most compelling evidence to date showing
that somatic mutations alone cannot account for
the origin of cancer (Seyfried, 2015). These experi-
ments showed that cytoplasm containing normal
mitochondria suppressed tumorigenesis in cells
with tumor-derived nuclei. In contrast, cytoplasm
80
80 section II: E merging Clinical Applications and Future Potential
containing abnormal mitochondria from tumor
cells enhanced tumorigenesis in cells with normal
nuclei. Viewed in the light of Warburg’s central
theory, the nuclear-cytoplasm transfer experi-
ments show that cancer originates from damage
to the mitochondria in the cytoplasm rather than
from damage to the genome in the nucleus. The
genomic instability seen in tumor cells follows,
rather than precedes, the disturbances in cellular
respiration.
Seyfried et  al. described how the common
pathophysiological mechanism underlying can-
cer, that is, “the oncogenic paradox,” was better
explained when cancer is considered as a mito-
chondrial metabolic disease than when considered
as a genetic disease (Seyfried, 2012c; Seyfried et al.,
2014; Seyfried and Shelton, 2010), and recently
outlined how all roads to the origin and progres-
sion of cancer pass through the mitochondria
(Seyfried et  al., 2014). Hence, novel therapeutic
strategies for management and prevention emerge
when cancer becomes recognized as a mitochon-
drial metabolic disease.
C A N C E R C E L L M E TA B O L I S M
It is well known that lactate fermentation is the
common metabolic malady seen in most if not all
cancer cells (Seyfried et al., 2014; Warburg, 1931,
1956a, 1956b, 1969). Indeed, lactate fermenta-
tion can be detected even in the slowest growing
tumors (Burk and Schade, 1956; Burk et al., 1967).
The greater the lactate production, the faster is
the rate of tumor growth. In addition to express-
ing robust fermentation in hypoxic environments,
cancer cells also ferment lactate in aerobic envi-
ronments. Aerobic fermentation is the antithesis to
the Pasteur effect (suppression of lactate fermenta-
tion in the presence of oxygen) and has become
known as aerobic glycolysis or the Warburg effect
(Racker, 1972). Lactate production is greater in
cancer cells than in normal cells, because the effi-
ciency of oxidative phosphorylation (OxPhos) is
less in cancer cells than in normal cells. Reduced
OxPhos efficiency arises from any type of defect
in the number, structure, or function of mitochon-
dria (Arismendi-Morillo and Castellano-Ramirez,
2008; Elliott et  al., 2012; Feichtinger et  al., 2015;
Feichtinger et  al., 2010; Kiebish et  al., 2008;
Pedersen, 1978; Seyfried et al., 2014; Singh et al.,
2015; Srinivasan et al., 2015). Mitochondrial struc-
ture is linked directly to mitochondrial function
(Lehninger, 1964). No malignant tumor tissue has
yet been found that contains mitochondria of nor-
mal number or structure when evaluated under
electron microscopy.
Although the evidence supporting mitochon-
drial abnormalities and OxPhos insufficiency in
cancer is now overwhelming, some metabolic
studies on cultured tumor cells can be inconsis-
tent with the findings from intact tumor tissue.
The unnatural conditions of the in vitro growth
environment (Crabtree effect) can make some
cancer cells appear to have normal respiration
despite their continued production of lactate. The
information on tumor cell energy metabolism
obtained from in vitro studies, however, is some-
times at odds with the data from in vivo studies.
The results from many in vitro studies suggest
that normal cells and tumor cells have similar
mitochondrial energy metabolism (Cairns, 2015;
Koppenol et al., 2011). If mitochondrial structure
is abnormal in the vast majority of tumor cells seen
in various cancerous tissues, how is it possible that
mitochondria and OxPhos could be normal in cul-
tured cells from the tumors? Seyfried et al. recently
addressed several confounding issues related to
tumor cell energy metabolism that is measured in
cultured cells (Seyfried et al., 2014).
The reduced efficiency of mitochondrial
OxPhos for ATP production in tumor cells is
compensated for with an increased dependency
on substrate level phosphorylation for energy pro-
duction. Substrate level phosphorylation occurs
at steps 7 and 10 in the glycolytic pathway, and at
the succinyl-CoA synthase reaction (Step 5) of the
citric acid cycle. Only those cells that can transi-
tion their energy production from OxPhos to sub-
strate level phosphorylation can become tumor
cells. Cells that cannot make this transition will
die and can never become tumor cells (Warburg,
1956a). As substrate level phosphorylation is less
efficient in producing ATP than is OxPhos, cancer
cells must consume greater amounts of ferment-
able fuels than normal cells in order to maintain
an adequate supply of ATP for viability (Seyfried
and Shelton, 2010).
The increased consumption of glucose is
clearly seen with positron emission tomogra-
phy (PET) scanning using radiolabeled glucose
for detecting cancer in the body (Vander Heiden
et al., 2009). Elevated glucose consumption leads
to an increase in lactate production and acidifica-
tion of the tumor cell microenvironment (Gatenby
and Gillies, 2004). Microenvironment acidifica-
tion increases inflammation and angiogenesis,
which ultimately increases tumor progression.
Succinate, produced through amino acid fermen-
tation (Hochachka et  al., 1975), can also con-
tribute to microenvironment acidification and
inflammation (Tannahill et  al., 2013). Hence, the
81
Chapter 12: Metabolism-Based Treatments to Counter Cancer
majority of cancer characteristics can be linked
directly to mitochondrial dysfunction coupled
with increased glucose consumption for compen-
satory fermentation.
In addition to a reliance on glucose, cancer
cells also depend heavily on glutamine for growth
and survival (DeBerardinis and Cheng, 2010;
Yuneva, 2008). Glutamine is anaplerotic and can
be rapidly metabolized to glutamate and then to
α-ketoglutarate for entry into the TCA cycle. In
addition to serving as a carbon/nitrogen source for
tumor cell growth, glutamine may also play a role in
cancer cell survival and growth through enzymatic
release of ammonia into the microenvironment
(Huang et  al., 2013). Glucose and glutamine act
synergistically for driving rapid tumor cell growth.
Glutamine can produce some ATP under aerobic
conditions and can serve as fuel for lipid synthesis
under hypoxic conditions (Ta and Seyfried, 2015).
In contrast to previous suggestions (DeBerardinis
et al., 2007), we found that only minor amounts of
glutamine are metabolized to lactate under either
normoxia or hypoxia in the VM-M3 invasive glio-
blastoma cells (Ta and Seyfried, 2015). We suggest
that glucose and glutamine can be either oxidized
or fermented depending on the physiological state
of the tumor microenvironment. Our findings
suggest that glutamine targeting can be effective
in managing systemic metastatic cancer (Shelton
et al., 2010b).
Besides glucose and glutamine, other meta-
bolic fuels might also play a role in tumor cell
metabolism. It has been shown recently that in
the presence of electron transport chain (ETC)
abnormalities, aspartate can rescue cell prolifera-
tion without increasing ATP synthesis, suggesting
that the redox reactions of the ETC are required
for survival (Sullivan et al., 2015). In cancer cells,
this acts as a potential target for metabolic thera-
pies because aspartate is central to the de novo
synthesis of purines and pyrimidines and pro-
teins, and for maintaining cellular redox balance.
While aspartate itself is not necessarily used for
ATP synthesis, a number of amino acids such as
serine, glycine, and the branched chain amino
acids can be catabolized to pyruvate, succinate,
and fumarate, all of which participate in the meta-
bolic pathways required for ATP generation and
anaplerosis. Additionally, there is evidence sug-
gesting that in the absence of oxygen, Complex III
of the ETC can run in reverse, effectively reducing
fumarate to succinate, which will accumulate as a
fermentation byproduct (Hochachka et al., 1975).
This will maintain the redox balance and result in
ATP synthesis.
81
Several byproducts of amino acid fermentation
can also accumulate in the tumor microenviron-
ment including acetate, glutamate, alanine, suc-
cinate, and ammonia. Although acetate has been
considered a potential fuel for supporting tumor
cell growth (Comerford et  al., 2014; Hosios and
Vander Heiden, 2014), acetate levels are generally
low in the circulation (Ballard, 1972). Jaworski
et  al. recently provided a comprehensive discus-
sion on the potential role of acetate in tumor
metabolism (Jaworski et  al., 2015). It should be
recognized that with the exception of glucose and
glutamine, none of the other potential fuels for
tumor cell metabolism would be present in suffi-
ciently high quantities to maintain robust tumor
cell growth. Hence, the restriction of glucose and
glutamine becomes of prime importance for tar-
geting tumor cell growth and survival.
CALORIE RESTRICTION
Calorie restriction (CR) with adequate nutrition or
underfeeding has been long known to limit tumor
growth (Hursting et  al., 2010; Kritchevsky, 2001;
Mukherjee et  al., 1999; Tannenbaum, 1942). The
mechanisms by which CR reduces tumor growth
have been described. Calorie restriction targets
several of cancer’s hallmarks, including angio-
genesis, inflammation, edema, proliferation, and
distal invasion (Jiang and Wang, 2013; Mukherjee
et al., 1999; Mulrooney et al., 2011; Shelton et al.,
2010a; Thompson et al., 2004; Woolf et al., 2015).
Calorie restriction lowers the level of circulating
glucose, the prime fermentable fuel for driving
tumor growth (Seyfried et al., 2003). In addition,
CR or therapeutic fasting, elevates the level of
circulating ketone bodies, D-β-hydroxybutyrate
and acetoacetate. Ketone bodies serve as a non-
fermentable super fuel for functional mitochon-
dria that also reduce reactive oxygen species
while increasing the ΔG’ of ATP hydrolysis and
metabolic efficiency (Seyfried et al., 2014; Seyfried
et  al., 2003; Veech, 2004). Based on these obser-
vations, Cahill and Veech characterized ketone
bodies as “Good Medicine” (Cahill and Veech,
2003). In contrast to normal cells, which transi-
tion to ketone bodies for energy when glucose
becomes limiting, tumor cells cannot use ketone
bodies effectively for energy (Fredericks and
Ramsey, 1978; Magee et  al., 1979; Maurer et  al.,
2011; Tisdale, 1984). Ketone bodies can even be
toxic to some tumor cells (Poff et al., 2015; Skinner
et al., 2009). Shimazu et al. recently suggested that
high-dose D-β-hydroxybutyrate could act as a his-
tone deacetylase inhibitor (Shimazu et  al., 2013).
82
82 section II: E merging Clinical Applications and Future Potential
The multiple defects in mitochondrial structure
and function are considered responsible for failure
of tumor cells to effectively use ketone bodies for
energy (Seyfried and Mukherjee, 2005; Seyfried
and Shelton, 2010).
Notwithstanding this growing experimental
literature, the therapeutic benefit of a 40% CR
for managing cancer in rodents would not be the
same as for managing cancer in humans. The basal
metabolic rate in humans is about seven times less
than that of mice (Terpstra, 2001). Consequently,
water-only therapeutic fasting for 2–3 weeks
would be predicted to produce a therapeutic ben-
efit similar to that seen in rodents under 40% CR
(Mahoney et al., 2006). As a ketogenic diet (KD)
can sometimes replicate the physiological condi-
tions of therapeutic fasting (Seyfried et al., 2009),
it will be easier for most cancer patients to use
a KD than to use therapeutic fasting for tumor
management.
T H E K E TO G E N I C   D I E T
The KD is a high-fat and low-carbohydrate diet
that is widely used to reduce refractory epileptic
seizures in children (Freeman and Kossoff, 2010;
Kossoff and Hartman, 2012), and is also effective
in managing symptoms of a wide range of neuro-
logical and neurodegenerative diseases (Seyfried,
2014). The KD is also gaining recognition as an
effective therapy for any cancer that expresses
the Warburg effect, which includes the majority
of metastatic cancers. The anticancer mechanism
of action for the KD is potentially very simple.
The KD lowers the prime fermentable fuel (glu-
cose) needed for driving the Warburg effect, while
elevating blood ketone bodies that require mito-
chondrial function for effective metabolism. As
mitochondrial defects are ultimately responsible
for the reliance of tumor cells on fermentation,
the KD targets the Warburg effect and tumor cell
energy metabolism. The KD can more effectively
reduce glucose and elevate blood ketone bodies
than can CR alone, making the KD potentially
more therapeutic against tumors than CR. There
are no anticancer drugs presently known that can
target tumor cells while enhancing the metabolic
efficiency of normal cells, as can the KD.
It is important to mention, however, that the
unrestricted feeding of KDs can cause weight gain,
insulin insensitivity, and dyslipidemia in some
strains of mice, leading to elevated blood levels
of insulin and glucose (Meidenbauer et al., 2014).
Dyslipidemia from excessive consumption of KD
can occur despite the very low amounts of carbo-
hydrate in the diet. This could account in part for
our failure to detect a therapeutic benefit against
either naturally occurring seizures in EL mice, or
against astrocytoma growth in mice that were fed
KDs in unrestricted amounts (Mantis et al., 2004;
Seyfried et al., 2003; Zhou et al., 2007).
Reduced blood glucose is considered essential
for managing either epileptic seizures or cancer.
As high-fat diets will elevate the appetite sup-
pressor hormone, cholecystokinin, it is possible
that dyslipidemia might not occur in most can-
cer patients on a KD. A  therapeutic KD should
lower glucose, cholesterol, and triglycerides,
while elevating ketone bodies and high-density
lipoprotein (Meidenbauer et al., 2014). The body
weight loss associated with KD is considered
therapeutic, in contrast to the weight loss asso-
ciated with cachexia or toxic cancer therapies,
which is pathological. In contrast to the thera-
peutic shift in blood metabolites associated with
the KD weight loss, cachexic weight loss is asso-
ciated with elevations of blood glucose and insu-
lin insensitivity that can stimulate tumor growth
(Kotler, 2000). High-dose steroids (dexametha-
sone), which elevate blood glucose levels, are
often given to cancer patients undergoing che-
motherapy to reduce vomiting and to improve
weight gain (Seyfried et al., 2015). It is clear from
an understanding of cancer metabolism that that
steroid administration for improved weight gain
would not be in the best interest of the cancer
patient (Wong et al., 2015). Hence, fundamental
metabolic differences exist between therapeutic
weight loss associated with a KD and the patho-
logical weight loss associated with cachexia and
toxic cancer therapies.
The protein and fat composition of the KD dif-
fers from that of Atkins-type diets in having less
protein and more fat than the Atkins diets. This
is important as several amino acids found in pro-
teins can be de-aminated to form pyruvate, which
is then metabolized to form glucose through glu-
coneogenesis (Burt et  al., 1982). The fats in KDs
also contain more saturated medium chain triglyc-
erides than do Atkins diets. Consequently, blood
glucose levels will be lower and ketone body lev-
els will be higher with KDs than with Atkins-type
diets. We recently developed the Glucose/Ketone
Index calculator (GKIC) to assess the potential
therapeutic effects of various low-carbohydrate
and KDs for cancer management (Meidenbauer
et al., 2015). The GKIC is a simple tool that mea-
sures the ratio of blood glucose to blood ketones
and can help monitor the efficacy of metabolic
therapy in preclinical animal models and in clini-
cal trials for malignant brain cancer or for any
83
Chapter 12: Metabolism-Based Treatments to Counter Cancer
cancer that expresses aerobic fermentation. The
GKI can therefore serve as a biomarker to validate
the therapeutic efficacy of various diets in target-
ing cancer cell energy metabolism.
THE FUTURE OF KDS
FOR CANCER MANAGEMENT
AND PREVENTION
It is our view that reduced glucose with therapeu-
tic ketosis, achieved through low-carbohydrate
KDs, can serve as an alternative or complemen-
tary approach to cancer management. Evidence
also suggests that the therapeutic efficacy of the
KD against tumor growth can be enhanced when
combined with certain drugs and procedures
under a “press-pulse” paradigm as we previously
mentioned (Seyfried et  al., 2014). For example,
therapeutic synergy was seen in combining the
calorie restricted KD with the glycolysis inhibitor,
2-deoxyglucose, for management of the syngeneic
CT-2a glioma stem cell tumor (Marsh et al., 2008),
and in combining the KD with hyperbaric oxygen
therapy and ketone esters for the management
of systemic metastasis of the syngeneic VM-M3
tumor (Poff et al., 2013; Poff et al., 2015). Scheck
and colleagues also showed that the KD could act
synergistically with radiation therapy for man-
agement of the GL261 mouse model of glioma
(Abdelwahab et al., 2012).
The therapeutic effects of KDs seen in these
preclinical brain tumor studies have been corrobo-
rated in preclinical studies for several other can-
cer models including neuroblastoma, lung cancer,
prostate cancer, and breast and ovarian cancers
(Allen et  al., 2013; Lv et  al., 2014; Mavropoulos
et  al., 2009; Morscher et  al., 2015; Zhuang et  al.,
2014). These preclinical studies are also motivat-
ing case reports and pilot studies in humans with
brain cancer and other cancers (Champ et  al.,
2013; Champ et  al., 2014; Maroon et  al., 2015;
Rieger et  al., 2014; Schmidt et  al., 2011; Zuccoli
et al., 2010). It is clear from these studies, and from
the original observation of Linda Nebeling and
colleagues (Nebeling et  al., 1995), that treatment
of cancer patients with KDs is generally well toler-
ated, which is consistent with decades of research
obtained from the evaluation of children treated
with KDs for epilepsy management.
In recognizing cancer as a mitochondrial
metabolic disease, we suggested that protect-
ing cellular mitochondria from toxic or meta-
bolic stress could best prevent cancer (Seyfried,
2012b). Calorie-restricted KDs and therapeutic
fasting are excellent ways to reduce oxidative
stress in mitochondria. Metabolism of the ketone
83
body β-D-hydroxybutyrate maintains the coen-
zyme Q couple in an oxidized state thus reducing
the production of damaging ROS (Veech, 2004).
Therapeutic water-only fasting or restricted KDs
reduce tissue inflammation (Mulrooney et  al.,
2011; Youm et al., 2015). Chronic inflammation is
known to produce mitochondrial stress and can-
cer (Bissell and Hines, 2011; Coussens and Werb,
2002; Fosslien, 2008; Kamp et al., 2011). It is our
contention that glucose/ketone ratios (GKI val-
ues) of 1.0 or below could help reduce systemic
inflammation. The GKI might therefore serve as
an effective biomarker, along with C-reactive pro-
tein, for assessing systemic inflammation and for
reducing mitochondrial stress that can prevent
cancer. Further studies will be needed to validate
this concept.
It has not escaped our attention that KDs could
have a significant impact on global budgeting con-
tracts for patient care. It is now recognized that
global budget contracts with quality incentives
encourage changes in practice patterns that can
help reduce spending and improve quality of gen-
eral health (Song et  al., 2014). The KDs are well
positioned to serve as a low-cost, nontoxic alterna-
tive for both the prevention and management of
cancer. In addition to cancer, we recently showed
how therapeutic ketosis could be effective for the
management and prevention of a broad range of
chronic inflammatory diseases including obesity,
type 2 diabetes, cardiovascular disease, epilepsy,
Alzheimer’s disease, Parkinson’ s disease, and trau-
matic brain injury (Seyfried, 2014). Hence, KDs
could have utility for improving health and reduc-
ing costs for many of the most challenging diseases
seen in Western societies.
AC K N OW L E D G M E N T S
This work was supported in part by the NIH,
Grants (HD-39722, NS-108055195, and CA-
102135); and grants from the American Institute
of Cancer Research; the Single Cause, Single Cure
Foundation; the George Yu Foundation; and the
Boston College Research Expense Fund.
REFERENCES
Abdelwahab, M.G., Fenton, K.E., Preul, M.C., Rho,
J.M., Lynch, A., Stafford, P., and Scheck, A.C.
(2012). The ketogenic diet is an effective adjuvant
to radiation therapy for the treatment of malignant
glioma. PloS One 7, e36197.
Allen, B.G., Bhatia, S.K., Buatti, J.M., Brandt, K.E.,
Lindholm, K.E., Button, A.M., Szweda, L.I.,
Smith, B.J., Spitz, D.R., and Fath, M.A. (2013).
Ketogenic diets enhance oxidative stress and
84
84 section II: E merging Clinical Applications and Future Potential
radio-chemo-therapy responses in lung cancer
xenografts. Clin Cancer Res 19, 3905–3913.
Arismendi-Morillo, G.J., and Castellano-Ramirez,
A.V. (2008). Ultrastructural mitochondrial pathol-
ogy in human astrocytic tumors: potentials impli-
cations pro-therapeutics strategies. J Electron
Microsc (Tokyo) 57, 33–39.
Baker, S.G., and Kramer, B.S. (2007). Paradoxes in car-
cinogenesis: new opportunities for research direc-
tions. BMC Cancer 7, 151.
Ballard, F.J. (1972). Supply and utilization of acetate in
mammals. Am J Clin Nutr 25, 773–779.
Bartesaghi, S., Graziano, V., Galavotti, S., Henriquez,
N.V., Betts, J., Saxena, J., Minieri, V.A.D.,
Karlsson, A., Martins, L.M., Capasso, M., et  al.
(2015). Inhibition of oxidative metabolism leads
to p53 genetic inactivation and transformation
in neural stem cells. Proc Nat Acad Sci USA 112,
1059–1064.
Bissell, M.J., and Hines, W.C. (2011). Why don’t we
get more cancer? A proposed role of the microen-
vironment in restraining cancer progression. Nat
Med 17, 320–329.
Burgio, E., and Migliore, L. (2015). Towards a systemic
paradigm in carcinogenesis:  linking epigenetics
and genetics. Mol Biol Rep 42, 777–790.
Burk, D., and Schade, A.L. (1956). On respiratory
impairment in cancer cells. Science 124, 270–272.
Burk, D., Woods, M., and Hunter, J. (1967). On the
significance of glucolysis for cancer growth, with
special reference to Morris rat hepatomas. J Natl
Cancer Inst 38, 839–863.
Burt, M.E., Gorschboth, C.M., and Brennan, M.F.
(1982). A controlled, prospective, randomized
trial evaluating the metabolic effects of enteral and
parenteral nutrition in the cancer patient. Cancer
49, 1092–1105.
Cahill, G.F., Jr., and Veech, R.L. (2003). Ketoacids?
Good medicine? Trans Am Clin Climatol Assoc
114, 149–161; discussion 162–143.
Cairns, R.A. (2015). Drivers of the Warburg pheno-
type. Cancer J 21, 56–61.
Cao, S., Zhang, C., and Xu, Y. (2015). Somatic muta-
tions may not be the primary drivers of cancer for-
mation. Int J Cancer 137, 2762–2765.
Champ, C.E., Mishra, M.V., Showalter, T.N., Ohri, N.,
Dicker, A.P., and Simone, N.L. (2013). Dietary
recommendations during and after cancer treat-
ment:  consistently inconsistent? Nutr Cancer 65,
430–439.
Champ, C.E., Palmer, J.D., Volek, J.S., Werner-Wasik,
M., Andrews, D.W., Evans, J.J., Glass, J., Kim, L.,
and Shi, W. (2014). Targeting metabolism with a
ketogenic diet during the treatment of glioblas-
toma multiforme. J Neurooncol 117, 125–131.
Comerford, S.A., Huang, Z., Du, X., Wang, Y., Cai, L.,
Witkiewicz, A.K., Walters, H., Tantawy, M.N., Fu,
A., Manning, H.C., et  al. (2014). Acetate depen-
dence of tumors. Cell 159, 1591–1602.
Coussens, L.M., and Werb, Z. (2002). Inflammation
and cancer. Nature 420, 860–867.
Darlington, C.D. (1948). The plasmagene theory of the
origin of cancer. Br J Cancer 2, 118–126.
DeBerardinis, R.J., Mancuso, A., Daikhin, E., Nissim,
I., Yudkoff, M., Wehrli, S., and Thompson, C.B.
(2007). Beyond aerobic glycolysis:  transformed
cells can engage in glutamine metabolism that
exceeds the requirement for protein and nucleo-
tide synthesis. Proc Nat Acad Sci USA 104,
19345–19350.
DeBerardinis, R.J., and Cheng, T. (2010). Q’s next: the
diverse functions of glutamine in metabolism, cell
biology and cancer. Oncogene 29, 313–324.
Elliott, R.L., Jiang, X.P., and Head, J.F. (2012).
Mitochondria organelle transplantation:  intro-
duction of normal epithelial mitochondria into
human cancer cells inhibits proliferation and
increases drug sensitivity. Breast Cancer Res Treat
136, 347–354.
Feichtinger, R.G., Zimmermann, F., Mayr, J.A.,
Neureiter, D., Hauser-Kronberger, C., Schilling,
F.H., Jones, N., Sperl, W., and Kofler, B. (2010).
Low aerobic mitochondrial energy metabolism in
poorly- or undifferentiated neuroblastoma. BMC
Cancer 10, 149.
Feichtinger, R.G., Weis, S., Mayr, J.A., Zimmermann,
F.A., Bogner, B., Sperl, W., and Kofler, B. (2015).
Alterations of oxidative phosphorylation in
meningiomas and peripheral nerve sheath tumors.
Neuro Oncol. 8(2):184–94.
Fosslien, E. (2008). Cancer morphogenesis:  role
of mitochondrial failure. Ann Clin Lab Sci 38,
307–329.
Fredericks, M., and Ramsey, R.B. (1978). 3-Oxo acid
coenzyme A  transferase activity in brain and
tumors of the nervous system. J Neurochem 31,
1529–1531.
Freeman, J.M., and Kossoff, E.H. (2010). Ketosis
and the ketogenic diet, 2010:  advances in treat-
ing epilepsy and other disorders. Adv Pediatr 57,
315–329.
Gatenby, R.A., and Gillies, R.J. (2004). Why do can-
cers have high aerobic glycolysis? Nat Rev Cancer
4, 891–899.
Hanahan, D., and Weinberg, R.A. (2000). The hall-
marks of cancer. Cell 100, 57–70.
Hanahan, D., and Weinberg, R.A. (2011). Hallmarks
of cancer: the next generation. Cell 144, 646–674.
Hochachka, P.W., Owen, T.G., Allen, J.F., and Whittow,
G.C. (1975). Multiple end products of anaerobio-
sis in diving vertebrates. Comp Biochem Physiol
B 50, 17–22.
Hosios, A.M., and Vander Heiden, M.G. (2014).
Acetate metabolism in cancer cells. Cancer
Metab 2, 27.
Hou, J.P., and Ma, J. (2014). DawnRank:  discover-
ing personalized driver genes in cancer. Genome
Med 6, 56.
85
Chapter 12: Metabolism-Based Treatments to Counter Cancer
Hu, Y., Lu, W., Chen, G., Wang, P., Chen, Z., Zhou,
Y., Ogasawara, M., Trachootham, D., Feng, L.,
Pelicano, H., et al. (2012). K-ras(G12V) transfor-
mation leads to mitochondrial dysfunction and a
metabolic switch from oxidative phosphorylation
to glycolysis. Cell Res 22, 399–412.
Huang, W., Choi, W., Chen, Y., Zhang, Q., Deng, H.,
He, W., and Shi, Y. (2013). A proposed role for glu-
tamine in cancer cell growth through acid resis-
tance. Cell Res 23, 724–727.
Hursting, S.D., Smith, S.M., Lashinger, L.M., Harvey,
A.E., and Perkins, S.N. (2010). Calories and car-
cinogenesis:  lessons learned from 30  years of
calorie restriction research. Carcinogenesis 31,
83–89.
Jaworski, D.M., Namboodiri, A.M., and Moffett,
J.R. (2015). Acetate as a metabolic and epigen-
etic modifier of cancer therapy. J Cell Biochem.
117(3):574–88.
Jiang, Y.S., and Wang, F.R. (2013). Caloric restriction
reduces edema and prolongs survival in a mouse
glioma model. J Neurooncol 114, 25–32.
Kamp, D.W., Shacter, E., and Weitzman, S.A. (2011).
Chronic inflammation and cancer:  the role of
the mitochondria. Oncology (Williston Park) 25,
400–410, 413.
Kiebish, M.A., Han, X., Cheng, H., Chuang, J.H., and
Seyfried, T.N. (2008). Cardiolipin and electron
transport chain abnormalities in mouse brain
tumor mitochondria: lipidomic evidence support-
ing the Warburg theory of cancer. J Lipid Res 49,
2545–2556.
Koppenol, W.H., Bounds, P.L., and Dang, C.V. (2011).
Otto Warburg’s contributions to current con-
cepts of cancer metabolism. Nat Rev Cancer 11,
325–337.
Kossoff, E.H., and Hartman, A.L. (2012). Ketogenic
diets:  new advances for metabolism-based thera-
pies. Curr Opin Neurol. 25(2):173–8.
Kotler, D.P. (2000). Cachexia. Ann Intern Med 133,
622–634.
Kritchevsky, D. (2001). Caloric restriction and cancer.
J Nutr Sci Vitaminol (Tokyo) 47, 13–19.
Lehninger, A.L. (1964). The Mitochondrion: Molecular
Basis of Structure and Function. (New  York:
Benjamin).
Lv, M., Zhu, X., Wang, H., Wang, F., and Guan, W.
(2014). Roles of caloric restriction, ketogenic diet
and intermittent fasting during initiation, progres-
sion and metastasis of cancer in animal models: a
systematic review and meta-analysis. PloS One 9,
e115147.
Magee, B.A., Potezny, N., Rofe, A.M., and Conyers,
R.A. (1979). The inhibition of malignant cell
growth by ketone bodies. Aust J Exp Biol Med Sci
57, 529–539.
Mahoney, L.B., Denny, C.A., and Seyfried, T.N. (2006).
Caloric restriction in C57BL/6J mice mimics ther-
apeutic fasting in humans. Lipids Health Dis 5, 13.
85
Mantis, J.G., Centeno, N.A., Todorova, M.T.,
McGowan, R., and Seyfried, T.N. (2004).
Management of multifactorial idiopathic epilepsy
in EL mice with caloric restriction and the keto-
genic diet: role of glucose and ketone bodies. Nutr
Metab (Lond) 1, 11.
Maroon, J.C., Seyfried, T.N., Donohue, J.P., and Bost,
J. (2015). The role of metabolic therapy in treating
glioblastoma multiforme. Surg Neurol Int 6, 61.
Marsh, J., Mukherjee, P., and Seyfried, T.N. (2008).
Drug/diet synergy for managing malignant
astrocytoma in mice:  2-deoxy-D-glucose and the
restricted ketogenic diet. Nutr Metab (Lond) 5, 33.
Maurer, G.D., Brucker, D.P., Baehr, O., Harter, P.N.,
Hattingen, E., Walenta, S., Mueller-Klieser, W.,
Steinbach, J.P., and Rieger, J. (2011). Differential
utilization of ketone bodies by neurons and glioma
cell lines: a rationale for ketogenic diet as experi-
mental glioma therapy. BMC Cancer 11, 315.
Mavropoulos, J.C., Buschemeyer, W.C., 3rd, Tewari,
A.K., Rokhfeld, D., Pollak, M., Zhao, Y., Febbo,
P.G., Cohen, P., Hwang, D., et al. (2009). The effects
of varying dietary carbohydrate and fat content
on survival in a murine LNCaP prostate cancer
xenograft model. Cancer Prev Res (Phila Pa) 2,
557–565.
McLeod, H.L. (2013). Cancer pharmacogenom-
ics:  early promise, but concerted effort needed.
Science 339, 1563–1566.
Meidenbauer, J.J., Mukherjee, P., and Seyfried, T.N.
(2015). The glucose ketone index calculator: a sim-
ple tool to monitor therapeutic efficacy for meta-
bolic management of brain cancer. Nutr Metab
(Lond) 12, 12.
Meidenbauer, J.J., Ta, N., and Seyfried, T.N. (2014).
Influence of a ketogenic diet, fish-oil, and calorie
restriction on plasma metabolites and lipids in
C57BL/6J mice. Nutr Metab 11, 23.
Morscher, R.J., Aminzadeh-Gohari, S., Feichtinger,
R.G., Mayr, J.A., Lang, R., Neureiter, D., Sperl, W.,
and Kofler, B. (2015). Inhibition of neuroblastoma
tumor growth by ketogenic diet and/or calorie
restriction in a CD1-Nu mouse model. PloS One
10, e0129802.
Mukherjee, P., Sotnikov, A.V., Mangian, H.J., Zhou,
J.R., Visek, W.J., and Clinton, S.K. (1999). Energy
intake and prostate tumor growth, angiogenesis,
and vascular endothelial growth factor expression.
J Natl Cancer Inst 91, 512–523.
Mulrooney, T.J., Marsh, J., Urits, I., Seyfried, T.N., and
Mukherjee, P. (2011). Influence of caloric restric-
tion on constitutive expression of NF-kappaB in
an experimental mouse astrocytoma. PloS One 6,
e18085.
Nebeling, L.C., Miraldi, F., Shurin, S.B., and Lerner, E.
(1995). Effects of a ketogenic diet on tumor metab-
olism and nutritional status in pediatric oncology
patients:  two case reports. J Am Coll Nutr 14,
202–208.
86
86 section II: E merging Clinical Applications and Future Potential
Pedersen, P.L. (1978). Tumor mitochondria and the
bioenergetics of cancer cells. Prog Exp Tumor Res
22, 190–274.
Poff, A.M., Ari, C., Seyfried, T.N., and D’Agostino, D.P.
(2013). The ketogenic diet and hyperbaric oxygen
therapy prolong survival in mice with systemic
metastatic cancer. PloS One 8, e65522.
Poff, A.M., Ward, N., Seyfried, T.N., Arnold, P.,
and D’Agostino, D.P. (2015). Non-toxic meta-
bolic management of metastatic cancer in VM
mice: Novel combination of ketogenic diet, ketone
supplementation, and hyperbaric oxygen therapy.
PloS One 10, e0127407.
Racker, E. (1972). Bioenergetics and the problem of
tumor growth. Am Sci 60, 56–63.
Rieger, J., Bahr, O., Maurer, G.D., Hattingen, E., Franz,
K., Brucker, D., Walenta, S., Kammerer, U., Coy,
J.F., Weller, M., et al. (2014). ERGO: a pilot study
of ketogenic diet in recurrent glioblastoma. Int J
Oncol 44, 1843–1852.
Rous, P. (1959). Surmise and fact on the nature of can-
cer. Nature 183, 1357–1361.
Schmidt, M., Pfetzer, N., Schwab, M., Strauss, I., and
Kammerer, U. (2011). Effects of a ketogenic diet
on the quality of life in 16 patients with advanced
cancer: A pilot trial. Nutr Metab 8, 54.
Seeger, P.G. (1959). [Cell respiration and cancer]. Das
Deutsche Gesundheitswesen 14, 893–898.
Seyfried, T.N., Sanderson, T.M., El-Abbadi, M.M.,
McGowan, R., and Mukherjee, P. (2003). Role of
glucose and ketone bodies in the metabolic con-
trol of experimental brain cancer. Br J Cancer 89,
1375–1382.
Seyfried, T.N., and Mukherjee, P. (2005). Targeting
energy metabolism in brain cancer:  review and
hypothesis. Nutr Metab (Lond) 2, 30.
Seyfried, T.N., Mantis, J.G., Todorova, M.T., and Greene,
A.E. (2009). Dietary management of epilepsy: role
of glucose and ketone bodies. In Encyclopedia of
Basic Epilepsy Research, Vol. 2, P.A. Schwartzkroin,
ed. (Oxford: Academic Press), pp. 687–693.
Seyfried, T.N., and Shelton, L.M. (2010). Cancer as a
metabolic disease. Nutr Metab (Lond) 7, 7.
Seyfried, T.N. (2012a). Cancer as a Metabolic
Disease:  On the Origin, Management, and
Prevention of Cancer (Hoboken, NJ: Wiley).
Seyfried, T.N. (2012b). Cancer prevention. In Cancer as
a Metabolic Disease: On the Origin, Management,
and Prevention of Cancer (Hoboken, NJ: Wiley),
pp. 375–386.
Seyfried, T.N. (2012c). Mitochondrial respiratory
dysfunction and the extrachromosomal origin of
cancer. In Cancer as a Metabolic Disease: On the
Origin, Management, and Prevention of Cancer
(Hoboken, NJ: Wiley), pp. 253–259.
Seyfried, T.N. (2014). Ketone strong:  emerging evi-
dence for a therapeutic role of ketone bodies in
neurological and neurodegenerative diseases. J
Lipid Res. 55(9):1815–7.
Seyfried, T.N., Flores, R.E., Poff, A.M., and D’Agostino,
D.P. (2014). Cancer as a metabolic disease: impli-
cations for novel therapeutics. Carcinogenesis 35,
515–527.
Seyfried, T.N. (2015). Cancer as a mitochondrial meta-
bolic disease. Front Cell Devel Biol 3, 43.
Seyfried, T.N., Flores, R., Poff, A.M., D’Agostino, D.P.,
and Mukherjee, P. (2015). Metabolic therapy:  a
new paradigm for managing malignant brain can-
cer. Cancer Lett 356, 289–300.
Shelton, L.M., Huysentruyt, L.C., Mukherjee, P., and
Seyfried, T.N. (2010a). Calorie restriction as an
anti-invasive therapy for malignant brain cancer
in the VM mouse. ASN Neuro 2, e00038.
Shelton, L.M., Huysentruyt, L.C., and Seyfried, T.N.
(2010b). Glutamine targeting inhibits systemic
metastasis in the VM-M3 murine tumor model.
Int J Cancer 127, 2478–2485.
Shimazu, T., Hirschey, M.D., Newman, J., He, W.,
Shirakawa, K., Le Moan, N., Grueter, C.A.,
Lim, H., Saunders, L.R., Stevens, R.D., et  al.
(2013). Suppression of oxidative stress by beta-
hydroxybutyrate, an endogenous histone deacety-
lase inhibitor. Science 339, 211–214.
Singh, L., Nag, T.C., and Kashyap, S. (2015).
Ultrastructural changes of mitochondria in
human retinoblastoma:  correlation with tumor
differentiation and invasiveness. Tumour Biol.
37(5):5797–803.
Skinner, R., Trujillo, A., Ma, X., and Beierle, E.A.
(2009). Ketone bodies inhibit the viability of
human neuroblastoma cells. J Pediatr Surg 44,
212–216; discussion 216.
Song, Z., Rose, S., Safran, D.G., Landon, B.E., Day, M.P.,
and Chernew, M.E. (2014). Changes in health care
spending and quality 4 years into global payment.
N Engl J Med 371, 1704–1714.
Sonnenschein, C., and Soto, A.M. (2000). Somatic
mutation theory of carcinogenesis: why it should be
dropped and replaced. Mol Carcinog 29, 205–211.
Soto, A.M., and Sonnenschein, C. (2004). The somatic
mutation theory of cancer: growing problems with
the paradigm? Bioessays 26, 1097–1107.
Srinivasan, S., Guha, M., Dong, D.W., Whelan, K.A.,
Ruthel, G., Uchikado, Y., Natsugoe, S., Nakagawa,
H., and Avadhani, N.G. (2015). Disruption of cyto-
chrome c oxidase function induces the Warburg
effect and metabolic reprogramming. Oncogene.
35(12):1585–95
Sullivan, L.B., Gui, D.Y., Hosios, A.M., Bush, L.N.,
Freinkman, E., and Vander Heiden, M.G. (2015).
Supporting aspartate biosynthesis is an essential
function of respiration in proliferating cells. Cell
162, 552–563.
Ta, N.L., and Seyfried, T.N. (2015). Influence of
serum and hypoxia on incorporation of [C]-
D-glucose or [C]-L-glutamine into lipids and
lactate in murine glioblastoma cells. Lipids.
50(12):1167–84
87
Chapter 12: Metabolism-Based Treatments to Counter Cancer
Tannahill, G.M., Curtis, A.M., Adamik, J., Palsson-
McDermott, E.M., McGettrick, A.F., Goel, G., Frezza,
C., Bernard, N.J., Kelly, B., Foley, N.H., et al. (2013).
Succinate is an inflammatory signal that induces IL-
1beta through HIF-1alpha. Nature 496, 238–242.
Tannenbaum, A. (1942). The genesis and growth of
tumors:  II. Effects of caloric restriction per se.
Cancer Res 2, 460–467.
Terpstra, A.H. (2001). Differences between humans
and mice in efficacy of the body fat lowering effect
of conjugated linoleic acid: role of metabolic rate. J
Nutr 131, 2067–2068.
Thompson, H.J., McGinley, J.N., Spoelstra, N.S., Jiang,
W., Zhu, Z., and Wolfe, P. (2004). Effect of dietary
energy restriction on vascular density during mam-
mary carcinogenesis. Cancer Res 64, 5643–5650.
Tisdale, M.J. (1984). Role of acetoacetyl-CoA syn-
thetase in acetoacetate utilization by tumor cells.
Cancer Biochem Biophys 7, 101–107.
Tomasetti, C., and Vogelstein, B. (2015). Cancer eti-
ology: variation in cancer risk among tissues can
be explained by the number of stem cell divisions.
Science 347, 78–81.
Vander Heiden, M.G., Cantley, L.C., and Thompson,
C.B. (2009). Understanding the Warburg
effect: the metabolic requirements of cell prolifera-
tion. Science 324, 1029–1033.
Vaux, D.L. (2011). In defense of the somatic mutation
theory of cancer. Bioessays 33, 341–343.
Veech, R.L. (2004). The therapeutic implications of
ketone bodies:  the effects of ketone bodies in
pathological conditions:  ketosis, ketogenic diet,
redox states, insulin resistance, and mitochondral
metabolism. Prostaglandins Leukot Essent Fatty
Acids 70, 309–319.
Verschoor, M.L., Ungard, R., Harbottle, A., Jakupciak,
J.P., Parr, R.L., and Singh, G. (2013). Mitochondria
and cancer: past, present, and future. Biomed Res
Int 2013, 612369.
Warburg, O. (1931). The Metabolism of Tumours
(New York: Richard R. Smith).
Warburg, O. (1956a). On the origin of cancer cells.
Science 123, 309–314.
Warburg, O. (1956b). On the respiratory impairment
in cancer cells. Science 124, 269–270.
87
Warburg, O. (1969). Revised Lindau Lectures:  The
prime cause of cancer and prevention—Parts  1  &
2. In Meeting of the Nobel-Laureates D. Burk, ed.
(Lindau, Lake Constance, Germany:  K.Triltsch),
pp. 1–9. http://www.hopeforcancer.com/OxyPlus.htm
Wong, E.T., Lok, E., Gautam, S., and Swanson,
K.D. (2015). Dexamethasone exerts profound
immunologic interference on treatment efficacy
for recurrent glioblastoma. Br J Cancer 113,
232–241.
Woods, M.W., and du Buy, H.G. (1945). Cytoplasmic
diseases and cancer. Science 102, 591–593.
Woolf, E.C., Curley, K.L., Liu, Q., Turner, G.H.,
Charlton, J.A., Preul, M.C., and Scheck, A.C.
(2015). The ketogenic diet  alters the hypoxic
response and affects expression of proteins associ-
ated with angiogenesis, invasive potential and vas-
cular permeability in a mouse glioma model. PloS
One 10, e0130357.
Youm, Y.H., Nguyen, K.Y., Grant, R.W., Goldberg, E.L.,
Bodogai, M., Kim, D., D’Agostino, D., Planavsky,
N., Lupfer, C., Kanneganti, T.D., et al. (2015). The
ketone metabolite beta-hydroxybutyrate blocks
NLRP3 inflammasome-mediated inflammatory
disease. Nat Med 21, 263–269.
Yuneva, M. (2008). Finding an “Achilles’ heel” of can-
cer: the role of glucose and glutamine metabolism
in the survival of transformed cells. Cell Cycle 7,
2083–2089.
Zhou, W., Mukherjee, P., Kiebish, M.A., Markis, W.T.,
Mantis, J.G., and Seyfried, T.N. (2007). The calo-
rically restricted ketogenic diet, an effective alter-
native therapy for malignant brain cancer. Nutr
Metab (Lond) 4, 5.
Zhuang, Y., Chan, D.K., Haugrud, A.B., and
Miskimins, W.K. (2014). Mechanisms by which
low glucose enhances the cytotoxicity of metfor-
min to cancer cells both in vitro and in vivo. PloS
One 9, e108444.
Zuccoli, G., Marcello, N., Pisanello, A., Servadei, F.,
Vaccaro, S., Mukherjee, P., and Seyfried, T.N.
(2010). Metabolic management of glioblastoma
multiforme using standard therapy together with a
restricted ketogenic diet: case report. Nutr Metab
(Lond) 7, 33.
88
13
Ketogenic Diet as Adjunctive Therapy
for Malignant Brain Cancer
E R I C C . WO O L F, P H D A N D A D R I E N N E C . S C H E C K , P H D
INTRODUCTION
Human malignant glioma is a uniformly fatal
disease due in part to the limitations of currently
available treatments, which include surgery, che-
motherapy, and radiation therapy. The average
survival of patients with glioblastoma multiforme
(GBM) is 1.5  years. It is therefore of paramount
importance that new therapeutic strategies for
brain cancer patients be developed, especially
those that can enhance the efficacy of current
treatment options without damaging normal brain
tissue. Advances in our understanding of the biol-
ogy of these tumors have led to an increase in the
number of targeted therapies in preclinical and
clinical trials (Nicholas et  al., 2011; Niyazi et  al.,
2011; Roesler et  al., 2010). While these therapies
may prove somewhat effective, the heterogeneity
of this tumor often precludes the targeted mole-
cules from being found on all cells in the tumor,
thus reducing the efficacy of these treatments. In
contrast, one trait shared by virtually all tumor
cells is altered metabolism.
T U M O R M E TA B O L I S M
Alterations in the metabolism of cancer cells, what
we now call the “Warburg effect” or aerobic glycol-
ysis, was first described by Otto Warburg in 1927
(Warburg et al., 1927). Cancer cells use glycolysis to
provide energy and biomolecules regardless of the
availability of oxygen. This results in the produc-
tion of fewer ATP molecules per molecule of glu-
cose, and thus tumor cells require large amounts of
glucose. This shift toward increased glycolytic flux
in the cytosol and away from the tricarboxylic acid
cycle and oxidative phosphorylation in the mito-
chondria occurs very early in tumorigenesis. This
allows for rapid cell proliferation even under con-
ditions of hypoxia and in the presence of dysfunc-
tional mitochondria. Since Warburg’s discovery,
metabolism has been of interest in the cancer field,
but it often seemed overshadowed by discoveries
of oncogenes, tumor suppressor genes, growth
factor pathways, molecular subtypes of cancers,
and so forth. There is a resurgence of interest in
metabolism as a central theme in cancer, and we
continue to find that metabolic pathways intersect
and often regulate key components of tumor ini-
tiation, progression and therapy response (Wolf
et al., 2010; Nijsten and van Dam, 2009). In fact,
altered metabolism itself has been referred to as
a hallmark of cancer (Cantor and Sabatini, 2012;
Ward and Thompson, 2012), in addition to being
involved in virtually all of the cancer hallmarks
described in the seminal paper by Hanahan and
Weinberg (Hanahan and Weinberg, 2011)  (see
Figure 13.1; Lewis and Abdel-Haleem, 2013).
The term “metabolic remodeling” has been
used to describe metabolic changes that can occur
in cancer cells (Obre and Rossignol, 2015), and
oncogene-associated pathways are now known to
intersect with and alter metabolic pathways. For
example, the tumor suppressor protein p53, which
plays a pivotal role in the cellular responses to
hypoxia, DNA damage, and oncogene activation,
is now known to regulate glycolysis and assist in
maintaining mitochondrial integrity (Puzio-Kuter,
2011). Another important connection between
metabolism and tumor growth is through regula-
tion of c-MYC. Overexpression of c-MYC occurs
in a wide variety of cancers, including gliomas. C-
MYC is a multifunctional transcription factor, and
the list of its target genes include those involved in
both cell proliferation and cell metabolism (Miller
et al., 2012). In addition to stimulating glycolysis,
c-MYC has been found to activate glutaminoly-
sis and lipid synthesis from citrate (Obre and
Rossignol, 2015).
With the advent of molecular analyses, studies
of growth factor pathways seemed to overshadow
the influence of metabolism on cancer growth.
89

Chapter 13: Ketogenic Diet as Adjunctive Therapy for Malignant Brain Cancer 89

mTOR Anabolism Perturbed

Enhanced serine/glycine Oxidative Glycolysis
glycolysis PI3K metabolism phosphorylation
inhibition
Akt
Ras HIF-1
PGM
r a t i ve Der
e
fe met gulate
HK-VDAC Proli ling abo d SCO2
Oxidative association signa lism
phosphorylation p53
inhibition defects
g

En licat
s
op lin

rep
si
ap isab

ha ion
to

nc
D

ed
NO
synthase
Glycosylation
VEGF
Angiogensis

Inflammation
HIF-1

Ang-2 NO
synthase
Perturbed
glycosylation IDH1

b il &
t a io n
it y
Im yste ion

ns utat
2-hydroxy-
s as
m m
ev

un

glutarate

M
Kynurenines i
e

FH
SHD
l
Eva cel
ding be d
g
su rowth Perturesion & Fumarate
Succinate
p pr a dh st a si s
Lactate e s s io n m et a HIF-1
Oxygen Acidification
independence of interstitium Prolyl 4-hydroxylases
Anaerobic
glycolysis Lactate

FIGURE  13.1 An illustration of the interconnectedness of tumor metabolism with Hanahan and Weinberg’s
hallmarks of cancer (Lewis and Abdel-Haleem, 2013).

Overactivation of the stress-responsive PI3K/AKT which is typically found in high-grade gliomas

signaling pathway is typical in many cancers and
often due to activation of growth factor signal-
ing pathways involved in glioma growth, such as
platelet-derived growth factor, epidermal growth
factor, and insulin growth factor. We now know
that these growth factor pathways are intertwined
with metabolic signaling pathways (Dibble and
Cantley, 2015; Courtnay et al., 2015; Roberts and
Miyamoto, 2015; Martini et al., 2014; Iurlaro et al.,
2014). The PI3K/AKT signaling pathway has been
closely linked to metabolism, and under low glu-
cose conditions it results in rapid tumor cell death
(Marie and Shinjo, 2011; Robey and Hay, 2009;
Yang et al., 2009).
Another important “hub” linking metabolism
and cancer is hypoxia-inducible factor 1 (HIF-1).
The expression of HIF-1 is activated by hypoxia,
and other cancers. HIF-1 is a heterodimeric
transcription factor that induces the transcrip-
tion of a variety of genes involved in angiogen-
esis (vascular endothelial growth factor [VEGF]
and other cytokines) in an attempt to improve
tissue perfusion. This results in the formation
of abnormal blood vessels that can increase
inflammation and edema in brain tumors, as
well as induction of the transcription of a vari-
ety of genes that promote invasion, migration,
and tumor growth (Fischer et al., 2002; Fujiwara
et  al., 2007; Hayashi et  al., 2007; Horing et  al.,
2012; Justus et al., 2015; Kaur et al., 2005; Masson
and Ratcliffe, 2014; Mou et al., 2010; Proescholdt
et  al., 2012; Yang et  al., 2012). In addition to
specific actions that relate to the tumor cell’s
response to oxygen availability, HIF-1 interacts
90
90 section II: E merging Clinical Applications and Future Potential
with the PI3K/AKT signaling path to act as a reg-
ulator of cancer metabolism, proliferation, and
glycolysis (Courtnay et  al., 2015; Justus et  al.,
2015; Wei et  al., 2013; Pore et  al., 2006). It also
affects the activation of nuclear factor–kappa B
(NF-κB), a transcriptional activator that is cen-
tral to the regulation of various signal transduc-
tion pathways and to transcriptional activation
events that mediate inflammation, cell prolif-
eration, cell migration, and angiogenesis. HIF-1
may, at least in part, provide the molecular basis
for the Warburg effect by “reprograming” cel-
lular metabolism in response to oxygen avail-
ability (Corbet and Feron, 2015; Courtnay et al.,
2015). HIF-1 also is a central figure in alterations
to the tumor microenvironment, which affects
not only tumor cell growth but also response to
therapy (Azzi et  al., 2013; Bayley and Devilee,
2010; Bruzzese et  al., 2014; Casey et  al., 2015;
Danhier et  al., 2013; Dewhirst, 2009; Hattingen
et al., 2011; Horing et al., 2012; Hu et al., 2012;
Joon et  al., 2004; Justus et  al., 2015; Marie
and Shinjo, 2011; Masson and Ratcliffe, 2014;
Amberger-Murphy, 2009; Metallo et  al., 2011;
Semenza, 2013; Shweiki et  al., 1995; Stegeman
et al., 2014; Wei et al., 2011; Yamada et al., 1999;
Yang et al., 2012).
The molecular background of a tumor cell can
also affect the regulation of the pathways already
described. Loss of function of phosphatase and
tensin homolog (PTEN) or mutation of p53 also
increase HIF-1, as does the accumulation of reac-
tive oxygen species (ROS). ROS are multifaceted
effector molecules involved in numerous cellular
pathways, including those regulating autophagic/
apoptotic responses to genotoxic stress, hypoxia,
and nutrient deprivation. Cancer cells often have
increased levels of ROS (Fruehauf and Meyskens,
2007), and they have been implicated in angio-
genesis induction and tumor growth through the
regulation of VEGF and HIF-1 (Weinberg and
Chandel, 2009).
It is clear that cancer cell metabolism is far
more complex than originally thought. A number
of cancer-associated mutations affect metabolism,
and defects in mitochondria are seen in cancer
that also link metabolism with cancer initiation
and progression. Although some of these interac-
tions are mentioned above, in-depth discussions
of all of the interactions that occur between can-
cer and metabolism are beyond the scope of this
chapter and the reader is referred to a number of
reviews on these subjects (Cantor and Sabatini,
2012; Gatenby and Gillies, 2004; Semenza, 2013;
Vander Heiden et al., 2009; Ward and Thompson,
2012; Robey et al., 2015; Gaude and Frezza, 2014;
Masson and Ratcliffe, 2014). The fact that meta-
bolic dysregulation is seen in virtually all tumor
cells has led to suggestions that a promising thera-
peutic strategy may be to exploit this feature. One
potential way to achieve this goal is through the
use of the therapeutic ketogenic diet (KD) or phys-
iologically similar methods, such as caloric restric-
tion or intermittent fasting.
T H E K E TO G E N I C   D I E T
The KD is more correctly referred to as “metabolic
therapy” rather than a “diet.” This high-fat, low-
carbohydrate diet is used to treat medically refrac-
tory epilepsy (Kim and Rho, 2008; Cross, 2013) in
children and, more recently, in some adults. The
diet is not without side effects; however, these are
typically readily managed when the patient has
appropriate supervision by a multidisciplinary
team (i.e., dietitian, nurse, and physician) skilled
in its use. The KD has been shown to have neu-
roprotective effects, and there are now studies to
determine its efficacy for a number of neurological
disorders, including Alzheimer’s disease, traumatic
brain injury, and amyotrophic lateral sclerosis
(Maalouf et al., 2009; Stafstrom and Rho, 2012).
The KD increases blood ketones and decreases
blood glucose by simulating the physiological
response to fasting, thus leading to high rates of
fatty acid oxidation and an increase in the produc-
tion of acetyl coenzyme A  (acetyl-CoA). When
the amount of acetyl-CoA exceeds the capacity of
the tricarboxylic acid cycle to utilize it, there is an
increase in the production of the ketone bodies β-
hydroxybutyrate (βHB) and acetoacetate (ACA),
which can be used as an energy source in the nor-
mal brain (Cahill and Veech, 2003; Morris, 2005;
Vanitallie and Nufert, 2003; Veech et  al., 2001;
Gasior et al., 2006). Since normal cells readily use
ketones as an alternate energy source, they are
unlikely to be adversely affected by reduced glu-
cose. In contrast, the metabolic alterations found
in cancer cells are generally thought to reduce
their ability to be “flexible” regarding their pri-
mary energy source, and thus they require glucose
(Seyfried et  al., 2011; Maurer et  al., 2011; Zhou
et al., 2007; Seyfried and Mukherjee, 2005; Tisdale
and Brennan, 1983; Fredericks and Ramsey, 1978;
Seyfried, 2012). The reduction in glucose that
results from the KD essentially “starves” the tumor
cells and thus inhibits their growth (Baranano and
Hartman, 2008; Bozzetti and Zupec-Kania, 2015;
Freedland et  al., 2008; Klement, 2013; Maroon
et  al., 2015; Seyfried et  al., 2009; Seyfried et  al.,
2015; Simone et  al., 2013; Wallace et  al., 2010).
91
Chapter 13: Ketogenic Diet as Adjunctive Therapy for Malignant Brain Cancer
Thus, when used as a therapy, the KD can take
advantage of the Warburg effect.
In addition to the effects mediated by glucose
reduction, work in the epilepsy field and more
recent work in cancer research has shown that
the KD can exhibit antitumor effects even in the
absence of glucose reduction. We demonstrated
the effect of adding ketones to media containing
glucose in vitro using the AO2V4 cell line (Scheck
et  al., 2012). This cell line was derived from a
recurrent human glioblastoma and is grown in
Waymouth’s MAB 87/3 media containing 28 mM
glucose and supplemented with 20% fetal calf
serum. When 5 mM βHB plus 5 mM ACA was
added to complete media, cell growth was signifi-
cantly inhibited. When 1,3-bis(2-chloroethyl)-1
nitrosourea (BCNU, carmustine), one of the che-
motherapeutic agents given to this patient prior
to tumor recurrence, was used in addition to
ketones there was additional growth inhibition
compared to either ketones or BCNU alone. More
recent work has shown that the ketones themselves
exert antitumor effects separate from the effects of
reduced blood glucose (Magee et al., 1979; Scheck
et al., 2012; Skinner et al., 2009). The ketone βHB
has recently been shown to inhibit histone deac-
tylases (HDACs), which can result in epigen-
etic suppression of the expression of a variety of
genes. Thus, ketones may provide a link between
metabolism and tumorigenesis, although the pre-
cise nature of these changes is as yet unknown
(Newman and Verdin, 2014a, 2014b; Sassone-
Corsi, 2013; Shimazu et al., 2013). The remainder
of this chapter addresses the utility of increasing
blood ketones and reducing blood glucose for the
treatment of brain tumors.
C U R R E N T S TAT U S
O F   K N OW L E D G E
While the idea that metabolism may be a thera-
peutic target for the treatment of cancer originated
with Otto Warburg, there have been only a limited
number of case reports describing the use of a KD
for cancer therapy. Similarly, preclinical studies of
efficacy and mechanisms of action were limited
until recently.
Preclinical Evidence
The use of metabolic alteration for the therapy of
brain tumors has been championed by Seyfried
and colleagues. They used the VM (Shelton et al.,
2010)  and CT-2A (Marsh et  al., 2008)  mouse
tumor models to show that a KD, especially when
given in restricted amounts, extends survival.
D’Agostino and colleagues have added hyperbaric
91
oxygen and ketone supplementation to demon-
strate reduced tumor cell growth and metastatic
spread in the VM metastatic tumor model (Poff
et  al., 2015; Poff et  al., 2014). Scheck et  al. used
the syngeneic intracranial GL261-luc/albino C57/
Bl6 model to demonstrate that caloric restriction
was not necessary for the antitumor effect of the
KD (Stafford et al., 2010), particularly when a 4:1
fat:carbohydrate plus protein formulation is used
(Scheck et al., 2012).
The KD and similar diets used as a mono-
therapy have a pluripotent effect on the growth
of tumors both in vitro and in vivo. This may
depend, at least in part, on the model system, the
specific metabolic intervention, and the molecu-
lar underpinnings of the tumor itself (Caso et al.,
2013; Freedland et al., 2008; Hao et al., 2015; Kim
et  al., 2012c; Lv et  al., 2014; Mavropoulos et  al.,
2009; Otto et al., 2008; Poff et al., 2015; Poff et al.,
2013; Simone et  al., 2013; Stafford et  al., 2010;
Woolf et  al., 2015). The striking feature of the
work done to date is that alterations in metabo-
lism have a far-reaching effect on tumor cells,
tumors, and the tumor microenvironment. Studies
have shown reductions in growth rate as one
might expect; however, there are also changes in
the formation of ROS and oxidative stress (Allen
et al., 2013; Stafford et al., 2010; Milder and Patel,
2012), angiogenesis (Puchowicz et al., 2008; Woolf
et al., 2015; Seyfried et al., 2015; Zhou et al., 2007),
hypoxia (Maurer et  al., 2011; Poff et  al., 2015;
Woolf et al., 2015), inflammation and peritumoral
edema (Kim et  al., 2012a; Mavropoulos et  al.,
2009; Woolf et al., 2015), metastasis and invasion
(Amann and Hellerbrand, 2009; Gluschnaider
et al., 2014; Hao et al., 2015; Lv et al., 2014; Otto
et al., 2008; Poff et al., 2015), and the expression of
various transcriptional modulators such as NF-κB
(Woolf et al., 2015) and microRNAs (Wang et al.,
2015a; Wang et al., 2015c; Irani and Hussain, 2015;
Bishop and Ferguson, 2015; Wang et  al., 2015b;
Chan et al., 2015).
Ketogenic Diet in Combination
with Standard Therapies
Although evidence suggests that the KD provides
antitumor benefits on its own, perhaps the most
effective use of the KD is in combination with
standard cancer therapies such as radiation and
chemotherapy (Allen et al., 2014). The KD greatly
enhanced survival in a mouse model of malignant
glioma when combined with TMZ when compared
with either treatment alone (Figure 13.2) (Scheck
et  al., 2011). Using a bioluminescent, syngeneic
intracranial model of malignant glioma, the KD
92

92 section II: E merging Clinical Applications and Future Potential

was shown to significantly potentiate the anti- with radiation and other anticancer therapeutics in
tumor effect of radiotherapy. In fact, 9 out of 11 both preclinical and clinical studies (Raffaghello
animals maintained on the KD and treated with et  al., 2008; Lee et  al., 2010; Seyfried et  al., 2012;
radiation had complete and sustained remission of Lee et al., 2012; Safdie et al., 2012; Saleh et al., 2013;
their implanted tumors, even after being switched Champ et  al., 2013; Champ et  al., 2014; Klement
back to a standard rodent diet (Figure 13.2) and Champ, 2014; Poff et  al., 2013; Raffaghello
(Abdelwahab et  al., 2012). Allen and colleagues et al., 2010).
reported similar results when the KD was com- The effectiveness of radiation therapy is due
bined with radiation and chemotherapy in a lung to a number of factors including relative damage
cancer xenograft model (Allen et  al., 2013). That done to tumor cells versus normal tissue and the
is, they found decreased tumor growth rate and ability of normal cells and tumor cells to repair the
increased survival. Calorie restriction and short- damage (Klement and Champ, 2014). Radiation
term fasting have also been found to be synergistic works, in part, by creating ROS through the radiol-
ysis of water. The ROS damage the DNA and other
macromolecules, causing sublethal damage that
(A) can become lethal if not repaired. The potentiation
SD
100 KetoCal® of radiation therapy by the KD or CR seems para-
p value (0.0049) doxical in light of our data demonstrating a reduc-
75
tion in ROS in tumors from animals maintained
50 on a KD (Stafford et al., 2010). However, radiation
effects do not only occur through ROS, and radia-
25 tion can directly damage DNA and other cellu-
lar macromolecules. Furthermore, in addition to
0 ROS, radiation causes the production of reactive
0 10 20 30 40 50 100 150
(B) nitrogen species (RNS), a potential source of mac-
Rad
100 KetoCal® + Rad romolecular damage following radiation (Saenko
p value (<0.0001) et al., 2013). Whether the KD or CR reduces the
75 formation of RNS is as yet unknown. In fact, the
main effect of the KD or CR may not be in alter-
50 ing the amount of radiation-induced damage, but
may in fact be in modulating the ability of tumor
25
and normal cells to repair radiation-induced dam-
0
age (Klement and Champ, 2014; Santivasi and Xia,
0 50 100 150 200 250 2014). Studies have shown that CR can enhance
(C) DNA repair in normal cells (Heydari et al., 2007);
TMZ
100 KetoCal® + TMZ however, this may not be the case in tumor cells,
p value (0.0025)
75 and the differential response of tumor cells and
normal cells to genotoxic stress may be mediated
50 by reduced IGF1 and glucose in the tumor cells.
We and others have shown that insulin-like growth
25 factor is reduced in animals maintained on a KD
(Freedland et al., 2008; Klement and Champ, 2014;
0 Mavropoulos et  al., 2009; Scheck et  al., 2012). In
0 10 20 30 40 50
addition, a number of studies have shown that
FIGURE  13.2 Kaplan-Meier survival plot of animals reduced activation of the PI3K/Akt pathway, acti-
implanted intracranially with GL261-luc2 malignant vation of the adenosine monophosphate-activated
glioma cells and (A) maintained on KetoCal® (KC, the 4:1 protein kinase (AMPK) signaling pathway, and
fat:carbohydrate plus protein formulation of the ketogenic reduction of receptor tyrosine kinase growth fac-
diet) versus standard diet (SD); (B) treated with 2 × 4 Gy tor pathways can all reduce radioresistance in
radiation versus KC plus radiation, and (C) treated with tumor cells (Zhang et al., 2014; Choi et al., 2014;
50 mg/kg temozolomide (TMZ) versus KC plus TMZ. Medova et  al., 2013; Gil Del Alcazar et  al., 2014;
Animals on KC survived significantly longer when treated Munshi and Ramesh, 2013; Li et  al., 2014; Sanli
with KC alone, when KC was combined with radiation, et al., 2014; Danhier et al., 2013; Wang et al., 2013).
and when KC was combined with TMZ (Abdelwahab Finally, ketones and the KD have been shown
et al., 2012; Scheck et al., 2011). to affect the immune system (Rahman et al.,
93
Chapter 13: Ketogenic Diet as Adjunctive Therapy for Malignant Brain Cancer
2014; Husain et al., 2013; Kim et al., 2012b; Youm
et al., 2015), and we have shown that the KD also
reverses tumor-mediated immune suppression
in a mouse model of malignant glioma (Lussier,
Woolf et al. 2016). As radiation-induced tumor
killing is known to expose the immune system to
a greater diversity of tumor antigens, it is possible
that the KD as an adjuvant works to augment the
effect of radiation in part by enhancing immunity
against GBM.
The variety of effects seen when glucose is low-
ered and/or ketones are increased suggests that
this may also potentiate other therapies, including
newer immune- and targeted therapies. Concerns
that potentiation of the antitumor effect of a par-
ticular therapy may also increase its effect on nor-
mal brain are valid. However, we and others have
shown that the gene expression changes seen in
tumor are different from those seen in normal
brain (Stafford et  al., 2010; Chang et  al., 2013;
Maurer et  al., 2011). Further, the KD is known
to have neuroprotective effects (Lund et al., 2009;
Maalouf et  al., 2009; Hartman, 2012; Puchowicz
et al., 2008), and thus it has been postulated that
this may actually help to protect the normal brain
from the deleterious effects of radio- and chemo-
therapy. Taken together, the preclinical data pro-
vides strong support for the clinical use of the KD
or CR as an adjuvant therapy for the treatment of
gliomas and other cancers.
K E TO G E N I C D I E T
IN HUMANS
Studies of glucose utilization in cancer go back
prior to the 1980s, including studies of metabo-
lism and cancer cachexia (Fearon et  al., 1988;
Tisdale et al., 1987). These and other studies sug-
gested that a KD consisting of a high percentage
of medium chain triglycerides (MCT) along with
various supplements resulted in weight gain and
improved nitrogen balance in both animals and
humans. In 1995, Nebeling and colleagues pub-
lished a case report in which they used a similar
KD based on MCT oil to treat two female pediat-
ric patients with advanced stage malignant brain
tumors (Nebeling and Lerner, 1995; Nebeling et al.,
1995). They demonstrated that dietary-induced
ketosis decreased the availability of glucose to
the tumor without causing a decrease in patient
weight or overall nutritional status. Furthermore,
both children had long-term tumor management
(Nebeling et  al., 1995). The second case report
was published in 2010 by Zuccoli and coworkers
(Zuccoli et al., 2010). This patient was a 65-year-
old female with a multicentric glioblastoma. She
93
was put on a 4:1 (ratio of fats:carbohydrate plus
protein) calorie-restricted (600 kcal/day) KD
during radiation and chemotherapy. During this
time, her body weight dropped by 20%, she had
reduced blood glucose, increased urinary ketones,
and, most importantly, no observable brain tumor
detectable by either fluoro-deoxyglucose positron
emission tomography (FDG-PET) or magnetic
resonance imaging (MRI). The tumor recurred 10
weeks after the patient resumed her normal eat-
ing habits, and she succumbed to her disease less
than 2 years after diagnosis. While this patient did
not experience long-term tumor control after ces-
sation of the diet, this report demonstrated that
the diet could be tolerated, even when used in a
calorie-restricted setting.
Results of a phase 1 clinical trial were reported
in 2011 by a German group (Schmidt et al., 2011).
Tolerability of a restricted calorie KD was tested in
16 patients with a variety of advanced (end-stage)
cancers. There were no severe side effects, and 5 of
the 16 patients were able to complete the 3-month
treatment. These five patients had stable disease
while on the diet. Two of the 11 remaining patients
died early following the beginning of the trial, one
was unable to tolerate the diet and dropped out
immediately, two patients dropped out for per-
sonal reasons, one was unable to continue the diet
for more than a month, three had disease progres-
sion within less than 2 months of starting the diet,
and one dropped out to resume chemotherapy.
While this trial demonstrated tolerability and a
favorable side-effect profile, the antitumor efficacy
could not be assessed due to the variety and sever-
ity of disease in the patients. Recently, Schwartz
et  al. reported on two patients with recurrent
GBM treated with a calorie-restricted KD as
monotherapy, and although the diet was tolerated,
both patients showed tumor progression—the first
within 4 weeks and the second within 12 weeks
of beginning the protocol (Schwartz et al., 2015).
More recently, a number of prospective clinical
trials have been initiated which have been sum-
marized in Table 13.1. These trials include studies
of up-front treatment using the KD in addition to
standard radiation and chemotherapy in patients
diagnosed with GBM.
The case reports described above, along with
numerous anecdotal reports, suggest that the KD
may be a promising anticancer therapy. However,
more work is needed to determine how to best
utilize the KD and other metabolic therapies for
the treatment of tumors. Most of the informa-
tion regarding the most effective way to use the
KD comes from the epilepsy literature. Further
94
94 section II: E merging Clinical Applications and Future Potential
research is needed to determine optimum blood
ketone and glucose levels for anticancer effects.
In addition, a variety of KDs are used for seizure
control, and it is not clear whether one or more
of the different formulations will provide the best
results for cancer patients. Finally, while the KD
has a long record of safety in the epilepsy com-
munity, side effects that occur when used in com-
bination with cancer therapies may differ in type
or severity. These data will come from carefully
controlled clinical trials that include input from
healthcare professionals well versed in the use
of the KD. Furthermore, patient enrollment into
clinical trials requires “buy-in” from the medical
community. Physicians must be educated on the
therapeutic benefits of metabolic alteration as an
adjuvant therapy. As with any decision regarding
therapy, the patient’s overall condition, including
nutritional status, must be taken into account. As
suggested by Klement and Champ (2014), can-
cer patients should be comprehensively assessed
for nutritional needs and tolerability of such
interventions.
Concern about patients’ quality of life is
sometimes given as a reason not to employ the
KD. Compliance can be made more difficult by
the use of steroids (prescribed for peritumoral
edema) that often increase hunger and raise
blood glucose levels. To address this, at least one
clinical trial (NCT02046187) includes an analy-
sis of both patient and caregiver quality of life.
Quality of life measurements are being added
to more clinical trials, as the importance of this
has become recognized at the national level
(Dirven et  al., 2014; Boele et  al., 2013; van den
Bent et al., 2011). While some clinicians are con-
cerned compliance will reduce quality of life, the
patients that do remain on the KD often com-
ment that this allows them to participate in their
own therapy. Despite these caveats, the existing
preclinical data suggesting antitumor efficacy
and a synergistic effect with standard therapies
provides a strong impetus to conduct controlled
clinical trials, particularly those that will shed
light on the interactions between the KD and
other therapies.
CONCLUSION
Improvements in the survival and quality of life
for patients with malignant brain tumors require
the implementation of new therapeutic modali-
ties, especially those that increase the efficacy of
current therapies without increasing toxic side
effects. While the rapid accumulation of data
defining the molecular and genetic aberrations
present in these tumors has suggested a host of
targets for the development of new treatments,
targeted therapies tried to date have met with
limited success. This is at least in part due to the
molecular heterogeneity of these tumors that
prevents any one target from being present in all
cells. In contrast, metabolic dysregulation is pres-
ent in virtually all tumor cells and there is rapidly
increasing interest in using metabolic therapies
such as the KD for the treatment of various can-
cers, especially brain tumors. Preclinical data
have demonstrated that the antitumor effects
of the KD and CR are multifaceted, and altera-
tions in energy metabolism can inhibit cancer
cell growth and increase the tumor’s response
to therapy. This provides a strong impetus to
continue work designed to elucidate the mecha-
nisms through which the KD exerts its antican-
cer effects, as well as suggesting the need for the
design of controlled clinical trials that will shed
light on the most effective ways to implement
metabolic therapies in combination with stan-
dard therapies for the treatment of malignant dis-
ease. This is a novel therapeutic paradigm, and we
have only begun to scratch the surface in terms of
its potential.
REFERENCES
Abdelwahab, M.G., Fenton, K.E., Preul, M.C., Rho,
J.M., Lynch, A., Stafford, P., and Scheck, A.C.
(2012). The ketogenic diet is an effective adjuvant
to radiation therapy for the treatment of malignant
glioma. PLoS One 7, e36197.
Allen, B.G., Bhatia, S.K., Buatti, J.M., Brandt, K.E.,
Lindholm, K.E., Button, A.M., Szweda, L.I., Smith,
B., Spitz, D.R., and Fath, M.A. (2013). Ketogenic
diets enhance oxidative stress and radio-chemo-
therapy responses in lung cancer xenografts. Clin
Cancer Res 19, 3905–3913.
Allen, B.G., Bhatia, S.K., Anderson, C.M.,
Eichenberger-Gilmore, J.M., Sibenaller, Z.A.,
Mapuskar, K.A., Schoenfeld, J.D., Buatti, J.M.,
Spitz, D.R., and Fath, M.A. (2014). Ketogenic diets
as an adjuvant cancer therapy:  HISTORY and
potential mechanism. Redox Biology 2C, 963–970.
Amann, T., and Hellerbrand, C. (2009). GLUT1 as a
therapeutic target in hepatocellular carcinoma.
Expert Opin Ther Targets 13, 1411–1427.
Amberger-Murphy, V. (2009). Hypoxia helps glioma
to fight therapy. Curr Cancer Drug Targets 9,
381–390.
Azzi, S., Hebda, J.K., and Gavard, J. (2013). Vascular
permeability and drug delivery in cancers. Front
Oncol 3, 211.
Baranano, K.W., and Hartman, A.L. (2008). The keto-
genic diet:  uses in epilepsy and other neurologic
illnesses. Curr Treat Options Neurol 10, 410–419.
95
Chapter 13: Ketogenic Diet as Adjunctive Therapy for Malignant Brain Cancer
Bayley, J.P., and Devilee, P. (2010). Warburg tumours
and the mechanisms of mitochondrial tumour
suppressor genes: barking up the right tree? Curr
Opin Genet Dev 20, 324–329.
Bishop, K.S., and Ferguson, L.R. (2015). The interac-
tion between epigenetics, nutrition and the devel-
opment of cancer. Nutrients 7, 922–947.
Boele, F.W., Heimans, J.J., Aaronson, N.K., Taphoorn,
M.J., Postma, T.J., Reijneveld, J.C., and Klein, M.
(2013). Health-related quality of life of significant
others of patients with malignant CNS versus non-
CNS tumors:  a comparative study. J Neurooncol
115, 87–94.
Bozzetti, F., and Zupec-Kania, B. (2015). Toward a
cancer-specific diet. Clin Nutr 10.
Bruzzese, L., Fromonot, J., By, Y., Durand-Gorde, J.M.,
Condo, J., Kipson, N., Guieu, R., Fenouillet, E.,
and Ruf, J. (2014). NF-kappaB enhances hypoxia-
driven T-cell immunosuppression via upregula-
tion of adenosine A(2A) receptors. Cell Signal 26,
1060–1067.
Cahill, G.F., Jr., and Veech, R.L. (2003). Ketoacids?
Good medicine? Trans Am Clin Climatol Assoc
114, 149–161.
Cantor, J.R., and Sabatini, D.M. (2012). Cancer cell
metabolism:  one hallmark, many faces. Cancer
Discov 2, 881–898.
Casey, S.C., Amedei, A., Aquilano, K., Azmi, A.S.,
Benencia, F., Bhakta, D., Bilsland, A.E., Boosani,
C.S., Chen, S., Ciriolo, M.R., et al. (2015). Cancer
prevention and therapy through the modulation
of the tumor microenvironment. Semin Cancer
Biol 35, S199–223.
Caso, J., Masko, E.M., Ii, J.A., Poulton, S.H., Dewhirst,
M., Pizzo, S.V., and Freedland, S.J. (2013). The
effect of carbohydrate restriction on prostate can-
cer tumor growth in a castrate mouse xenograft
model. Prostate 73, 449–454.
Champ, C.E., Baserga, R., Mishra, M.V., Jin, L., Sotgia,
F., Lisanti, M.P., Pestell, R.G., Dicker, A.P., and
Simone, N.L. (2013). Nutrient restriction and
radiation therapy for cancer treatment: when less
is more. Oncologist 18, 97–103.
Champ, C.E., Palmer, J.D., Volek, J.S., Werner-Wasik,
M., Andrews, D.W., Evans, J.J., Glass, J., Kim, L.,
and Shi, W. (2014). Targeting metabolism with a
ketogenic diet during the treatment of glioblas-
toma multiforme. J Neurooncol 117, 125–131.
Chan, B., Manley, J., Lee, J., and Singh, S.R. (2015). The
emerging roles of microRNAs in cancer metabo-
lism. Cancer Lett 356, 301–308.
Chang, H.T., Olson, L.K., and Schwartz, K.A. (2013).
Ketolytic and glycolytic enzymatic expression pro-
files in malignant gliomas:  implication for keto-
genic diet therapy. Nutr Metab (Lond) 10, 47.
Choi, E.J., Cho, B.J., Lee, D.J., Hwang, Y.H., Chun,
S.H., Kim, H.H., and Kim, I.A. (2014). Enhanced
cytotoxic effect of radiation and temozolomide in
malignant glioma cells: targeting PI3K-AKT-mTOR
95
signaling, HSP90 and histone deacetylases. BMC
Cancer 14, 17.
Corbet, C., and Feron, O. (2015). Metabolic and mind
shifts:  from glucose to glutamine and acetate
addictions in cancer. Curr Opin Clin Nutr Metab
Care 18, 346–353.
Courtnay, R., Ngo, D.C., Malik, N., Ververis, K.,
Tortorella, S.M., and Karagiannis, T.C. (2015).
Cancer metabolism and the Warburg effect:  the
role of HIF-1 and PI3K. Mol Biol Rep 42, 841–851.
Cross, J.H. (2013). New research with diets and epi-
lepsy. J Child Neurol 28, 970–974.
Danhier, P., De Saedeleer, C.J., Karroum, O., De,
P.G., Porporato, P.E., Jordan, B.F., Gallez, B.,
and Sonveaux, P. (2013). Optimization of tumor
radiotherapy with modulators of cell metabo-
lism:  toward clinical applications. Semin Radiat
Oncol 23, 262–272.
Dewhirst, M.W. (2009). Relationships between cycling
hypoxia, HIF-1, angiogenesis and oxidative stress.
Radiat Res 172, 653–665.
Dibble, C.C., and Cantley, L.C. (2015). Regulation of
mTORC1 by PI3K signaling. Trends Cell Biol 25,
545–555.
Dirven, L., Taphoorn, M.J., Reijneveld, J.C., Blazeby,
J., Jacobs, M., Pusic, A., La, S.E., Stupp, R., Fayers,
P., and Efficace, F. (2014). The level of patient-
reported outcome reporting in randomised con-
trolled trials of brain tumour patients: a systematic
review. Eur J Cancer 50, 2432–2448.
Fearon, K.C., Borland, W., Preston, T., Tisdale, M.J.,
Shenkin, A., and Calman, K.C. (1988). Cancer
cachexia: influence of systemic ketosis on substrate
levels and nitrogen metabolism. Am J Clin Nutr
47, 42–48.
Fischer, S., Wobben, M., Marti, H.H., Renz, D., and
Schaper, W. (2002). Hypoxia-induced hyperper-
meability in brain microvessel endothelial cells
involves VEGF-mediated changes in the expression
of zonula occludens-1. Microvasc Res 63, 70–80.
Fredericks, M., and Ramsey, R.B. (1978). 3-Oxo acid
coenzyme A  transferase activity in brain and
tumors of the nervous system. J. Neurochem. 31,
1529–1531.
Freedland, S.J., Mavropoulos, J., Wang, A., Darshan,
M., Demark-Wahnefried, W., Aronson, W.J.,
Cohen, P., Hwang, D., Peterson, B., Fields, T., et al.
(2008). Carbohydrate restriction, prostate cancer
growth, and the insulin-like growth factor axis.
Prostate 68, 11–19.
Fruehauf, J.P., and Meyskens, F.L., Jr. (2007). Reactive
oxygen species:  a breath of life or death? Clin
Cancer Res 13, 789–794.
Fujiwara, S., Nakagawa, K., Harada, H., Nagato, S.,
Furukawa, K., Teraoka, M., Seno, T., Oka, K., Iwata,
S., and Ohnishi, T. (2007). Silencing hypoxia-
inducible factor-1alpha inhibits cell migration and
invasion under hypoxic environment in malignant
gliomas. Int J Oncol 30, 793–802.
96
96 section II: E merging Clinical Applications and Future Potential
Gasior, M., Rogawski, M.A., and Hartman, A.L.
(2006). Neuroprotective and disease-modifying
effects of the ketogenic diet. Behav Pharmacol 17,
431–439.
Gatenby, R.A., and Gillies, R.J. (2004). Why do can-
cers have high aerobic glycolysis? Nat Rev Cancer
4, 891–899.
Gaude, E., and Frezza, C. (2014). Defects in mito-
chondrial metabolism and cancer. Cancer Metab
2, 10–12.
Gil Del Alcazar, C.R., Hardebeck, M.C., Mukherjee, B.,
Tomimatsu, N., Gao, X., Yan, J., Xie, X.J., Bachoo,
R., Li, L., Habib, A.A., et al. (2014). Inhibition of
DNA double-strand break repair by the dual PI3K/
mTOR inhibitor NVP-BEZ235 as a strategy for
radiosensitization of glioblastoma. Clin Cancer
Res 20, 1235–1248.
Gluschnaider, U., Hertz, R., Ohayon, S., Smeir, E.,
Smets, M., Pikarsky, E., and Bar-Tana, J. (2014).
Long-chain fatty acid analogues suppress breast
tumorigenesis and progression. Cancer Res 74,
6991–7002.
Hanahan, D., and Weinberg, R.A. (2011). Hallmarks
of cancer: the next generation. Cell 144, 646–674.
Hao, G.W., Chen, Y.S., He, D.M., Wang, H.Y., Wu,
G.H., and Zhang, B. (2015). Growth of human
colon cancer cells in nude mice is delayed by keto-
genic diet with or without omega-3 fatty acids and
medium-chain triglycerides. Asian Pac J Cancer
Prev 16, 2061–2068.
Hartman, A.L. (2012). Neuroprotection in metabolism-
based therapy. Epilepsy Res 100, 286–294.
Hattingen, E., Jurcoane, A., Bahr, O., Rieger, J.,
Magerkurth, J., Anti, S., Steinbach, J.P., and
Pilatus, U. (2011). Bevacizumab impairs oxidative
energy metabolism and shows antitumoral effects
in recurrent glioblastomas: a 31P/1H MRSI and
quantitative magnetic resonance imaging study.
Neuro Oncol 13, 1349–1363.
Hayashi, Y., Edwards, N.A., Proescholdt, M.A.,
Oldfield, E.H., and Merrill, M.J. (2007). Regulation
and function of aquaporin-1 in glioma cells.
Neoplasia 9, 777–787.
Heydari, A.R., Unnikrishnan, A., Lucente, L.V.,
and Richardson, A. (2007). Caloric restriction
and genomic stability. Nucleic Acids Res 35,
7485–7496.
Horing, E., Harter, P.N., Seznec, J., Schittenhelm, J.,
Buhring, H.J., Bhattacharyya, S., von Hattingen,
E., Zachskorn, C., Mittelbronn, M., and Naumann,
U. (2012). The “go or grow” potential of gliomas is
linked to the neuropeptide processing enzyme car-
boxypeptidase E and mediated by metabolic stress.
Acta Neuropathol 124, 83–97.
Hu, Y.L., DeLay, M., Jahangiri, A., Molinaro, A.M.,
Rose, S.D., Carbonell, W.S., and Aghi, M.K. (2012).
Hypoxia-induced autophagy promotes tumor cell
survival and adaptation to antiangiogenic treat-
ment in glioblastoma. Cancer Res 72, 1773–1783.
Husain, Z., Huang, Y., Seth, P., and Sukhatme, V.P.
(2013). Tumor-derived lactate modifies antitu-
mor immune response: effect on myeloid-derived
suppressor cells and NK cells. J Immunol 191,
1486–1495.
Irani, S., and Hussain, M.M. (2015). Role of
microRNA-30c in lipid metabolism, adipogenesis,
cardiac remodeling and cancer. Curr Opin Lipidol
26, 139–146.
Iurlaro, R., Leon-Annicchiarico, C.L., and Munoz-
Pinedo, C. (2014). Regulation of cancer metab-
olism by oncogenes and tumor suppressors.
Methods Enzymol 542, 59–80.
Joon, Y.A., Bazar, K.A., and Lee, P.Y. (2004). Tumors
may modulate host immunity partly through
hypoxia-induced sympathetic bias. Med
Hypotheses 63, 352–356.
Justus, C.R., Sanderlin, E.J., and Yang, L.V. (2015).
Molecular connections between cancer cell metab-
olism and the tumor microenvironment. Int J Mol
Sci 16, 11055–11086.
Kaur, B., Khwaja, F.W., Severson, E.A., Matheny, S.L.,
Brat, D.J., and Van Meir, E.G. (2005). Hypoxia and
the hypoxia-inducible-factor pathway in glioma
growth and angiogenesis. Neuro Oncol 7, 134–153.
Kim, D.Y., and Rho, J.M. (2008). The ketogenic diet
and epilepsy. Curr Opin Clin Nutr Metab Care 11,
113–120.
Kim, D.Y., Hao, J., Liu, R., Turner, G., Shi, F.D., and
Rho, J.M. (2012a). Inflammation-mediated mem-
ory dysfunction and effects of a ketogenic diet in
a murine model of multiple sclerosis. PLoS One 7,
e35476.
Kim, D.Y., Hao, J., Liu, R., Turner, G., Shi, F.D., and
Rho, J.M. (2012b). Inflammation-mediated mem-
ory dysfunction and effects of a ketogenic diet in a
murine model of multiple sclerosis. PLoS One 7,
e35476.
Kim, H.S., Masko, E.M., Poulton, S.L., Kennedy, K.M.,
Pizzo, S.V., Dewhirst, M.W., and Freedland, S.J.
(2012c). Carbohydrate restriction and lactate
transporter inhibition in a mouse xenograft model
of human prostate cancer. BJU Int 110, 1062–1069.
Klement, R.J. (2013). Calorie or carbohydrate restric-
tion? The ketogenic diet as another option for sup-
portive cancer treatment. Oncologist 18, 1056.
Klement, R.J., and Champ, C.E. (2014). Calories,
carbohydrates, and cancer therapy with radia-
tion: exploiting the five R’s through dietary manip-
ulation. Cancer Metastasis Rev 33, 217–229.
Lee, C., Safdie, F.M., Raffaghello, L., Wei, M., Madia,
F., Parrella, E., Hwang, D., Cohen, P., Bianchi, G.,
and Longo, V.D. (2010). Reduced levels of IGF-I
mediate differential protection of normal and can-
cer cells in response to fasting and improve che-
motherapeutic index. Cancer Res 70, 1564–1572.
Lee, C., Raffaghello, L., Brandhorst, S., Safdie, F.M.,
Bianchi, G., Martin-Montalvo, A., Pistoia, V., Wei,
M., Hwang, S., Merlino, A., et  al. (2012). Fasting
97
Chapter 13: Ketogenic Diet as Adjunctive Therapy for Malignant Brain Cancer
cycles retard growth of tumors and sensitize a
range of cancer cell types to chemotherapy. Sci
Transl Med 4, 124ra27.
Lewis, N.E., and Abdel-Haleem, A.M. (2013). The evo-
lution of genome-scale models of cancer metabo-
lism. Front Physiol 4, 237.
Li, W., Guo, F., Wang, P., Hong, S., and Zhang, C.
(2014). miR-221/222 confers radioresistance in
glioblastoma cells through activating Akt indepen-
dent of PTEN status. Curr Mol Med 14, 185–195.
Lussier, D. M., et al. (2016). Enhanced immunity in a
mouse model of malignant glioma is mediated by
a therapeutic ketogenic diet. BMC Cancer. May 13:
310.
Lund, T.M., Risa, O., Sonnewald, U., Schousboe, A.,
and Waagepetersen, H.S. (2009). Availability of
neurotransmitter glutamate is diminished when
beta-hydroxybutyrate replaces glucose in cultured
neurons. J Neurochem 110, 80–91.
Lv, M., Zhu, X., Wang, H., Wang, F., and Guan, W.
(2014). Roles of caloric restriction, ketogenic diet
and intermittent fasting during initiation, progres-
sion and metastasis of cancer in animal models: a
systematic review and meta-analysis. PLoS One 9,
e115147.
Maalouf, M., Rho, J.M., and Mattson, M.P. (2009). The
neuroprotective properties of calorie restriction,
the ketogenic diet, and ketone bodies. Brain Res
Rev 59, 293–315.
Magee, B.A., Potezny, N., Rofe, A.M., and Conyers,
R.A. (1979). The inhibition of malignant cell
growth by ketone bodies. Aust J Exp Biol Med Sci
57, 529–539.
Marie, S.K., and Shinjo, S.M. (2011). Metabolism
and brain cancer. Clinics (Sao Paulo). 66(Suppl
1), 33–43.
Maroon, J.C., Seyfried, T.N., Donohue, J.P., and Bost,
J. (2015). The role of metabolic therapy in treat-
ing glioblastoma multiforme. Surg Neurol Int 6:61.
doi:  10.4103/2152-7806.155259. eCollection;%20
15., 61–7806
Marsh, J., Mukherjee, P., and Seyfried, T.N. (2008).
Drug/diet synergy for managing malignant
astrocytoma in mice:  2-deoxy-D-glucose and
the restricted ketogenic diet. Nutr Metab (Lond)
5:33, 33.
Martini, M., De Santis, M.C., Braccini, L., Gulluni, F.,
and Hirsch, E. (2014). PI3K/AKT signaling path-
way and cancer: an updated review. Ann Med 46,
372–383.
Masson, N., and Ratcliffe, P.J. (2014). Hypoxia signal-
ing pathways in cancer metabolism:  the impor-
tance of co-selecting interconnected physiological
pathways. Cancer Metab 2, 3.
Maurer, G.D., Brucker, D.P., Bahr, O., Harter, P.N.,
Hattingen, E., Walenta, S., Mueller-Klieser, W.,
Steinbach, J.P., and Rieger, J. (2011). Differential
utilization of ketone bodies by neurons and
glioma cell lines:  a rationale for ketogenic diet
97
as experimental glioma therapy. BMC Cancer
11:315, 315.
Mavropoulos, J.C., Buschemeyer, W.C., III, Tewari,
A.K., Rokhfeld, D., Pollak, M., Zhao, Y., Febbo,
P.G., Cohen, P., Hwang, D., Devi, G., et al. (2009).
The effects of varying dietary carbohydrate and fat
content on survival in a murine LNCaP prostate
cancer xenograft model. Cancer Prev Res (Phila)
2, 557–565.
Medova, M., Aebersold, D.M., and Zimmer, Y.
(2013). The molecular crosstalk between the MET
receptor tyrosine kinase and the DNA damage
response-biological and clinical aspects. Cancers
(Basel) 6, 1–27.
Metallo, C.M., Gameiro, P.A., Bell, E.L., Mattaini, K.R.,
Yang, J., Hiller, K., Jewell, C.M., Johnson, Z.R.,
Irvine, D.J., Guarente, L., et  al. (2011). Reductive
glutamine metabolism by IDH1 mediates lipogen-
esis under hypoxia. Nature 481, 380–384.
Milder, J., and Patel, M. (2012). Modulation of oxida-
tive stress and mitochondrial function by the keto-
genic diet. Epilepsy Res 100, 295–303.
Miller, D.M., Thomas, S.D., Islam, A., Muench, D., and
Sedoris, K. (2012). c-Myc and cancer metabolism.
Clin Cancer Res 18, 5546–5553.
Morris, A.A.M. (2005). Cerebral ketone body metabo-
lism. J Inherit Metab Dis 28, 109–121.
Mou, K., Chen, M., Mao, Q., Wang, P., Ni, R., Xia, X.,
and Liu, Y. (2010). AQP-4 in peritumoral edema-
tous tissue is correlated with the degree of glioma
and with expression of VEGF and HIF-alpha. J
Neurooncol 100, 375–383.
Munshi, A., and Ramesh, R. (2013). Mitogen-activated
protein kinases and their role in radiation response.
Genes Cancer 4, 401–408.
Nebeling, L.C., and Lerner, E. (1995). Implementing a
ketogenic diet based on medium-chain triglycer-
ide oil in pediatric patients with cancer. J Am Diet
Assoc 95, 693–697.
Nebeling, L.C., Miraldi, F., Shurin, S.B., and Lerner, E.
(1995). Effects of a ketogenic diet on tumor metabo-
lism and nutritional status in pediatric oncology
patients: two case reports. J Am Coll Nutr 14, 202–208.
Newman, J.C., and Verdin, E. (2014a). Beta-
hydroxybutyrate:  much more than a metabolite.
Diabetes Res Clin Pract 106, 173–181.
Newman, J.C., and Verdin, E. (2014b). Ketone bodies
as signaling metabolites. Trends Endocrinol Metab
25, 42–52.
Nicholas, M.K., Lukas, R.V., Chmura, S., Yamini, B.,
Lesniak, M., and Pytel, P. (2011). Molecular het-
erogeneity in glioblastoma: therapeutic opportuni-
ties and challenges. Semin Oncol 38, 243–253.
Nijsten, M.W., and van Dam, G.M. (2009).
Hypothesis: using the Warburg effect against can-
cer by reducing glucose and providing lactate. Med
Hypotheses 73, 48–51.
Niyazi, M., Siefert, A., Schwarz, S.B., Ganswindt,
U., Kreth, F.W., Tonn, J.C., and Belka, C. (2011).
98
98 section II: E merging Clinical Applications and Future Potential
Therapeutic options for recurrent malignant gli-
oma. Radiother Oncol 98, 1–14.
Obre, E., and Rossignol, R. (2015). Emerging concepts
in bioenergetics and cancer research:  metabolic
flexibility, coupling, symbiosis, switch, oxida-
tive tumors, metabolic remodeling, signaling
and bioenergetic therapy. Int J Biochem Cell Biol
59C:167–181. doi: 10.1016/j.biocel.2014.12.008.
Otto, C., Kaemmerer, U., Illert, B., Muehling, B.,
Pfetzer, N., Wittig, R., Voelker, H.U., Thiede, A.,
and Coy, J.F. (2008). Growth of human gastric
cancer cells in nude mice is delayed by a ketogenic
diet supplemented with omega-3 fatty acids and
medium-chain triglycerides. BMC Cancer 8, 122.
Poff, A.M., Ari, C., Seyfried, T.N., and D’Agostino, D.P.
(2013). The ketogenic diet and hyperbaric oxygen
therapy prolong survival in mice with systemic
metastatic cancer. PLoS One 8, e65522.
Poff, A.M., Ari, C., Arnold, P., Seyfried, T.N., and
D’Agostino, D.P. (2014). Ketone supplementation
decreases tumor cell viability and prolongs sur-
vival of mice with metastatic cancer. Int J Cancer
135, 1711–1720.
Poff, A.M., Ward, N., Seyfried, T.N., Arnold, P.,
and D’Agostino, D.P. (2015). Non-toxic meta-
bolic management of metastatic cancer in vm
mice: novel combination of ketogenic diet, ketone
supplementation, and hyperbaric oxygen therapy.
PLoS One 10, e0127407.
Pore, N., Jiang, Z., Shu, H.K., Bernhard, E., Kao, G.D.,
and Maity, A. (2006). Akt1 activation can aug-
ment hypoxia-inducible factor-1alpha expression
by increasing protein translation through a mam-
malian target of rapamycin-independent pathway.
Mol Cancer Res 4, 471–479.
Proescholdt, M.A., Merrill, M.J., Stoerr, E.M.,
Lohmeier, A., Pohl, F., and Brawanski, A. (2012).
Function of carbonic anhydrase IX in glioblastoma
multiforme. Neuro Oncol 14, 1357–1366.
Puchowicz, M.A., Zechel, J.L., Valerio, J., Emancipator,
D.S., Xu, K., Pundik, S., LaManna, J.C., and Lust,
W.D. (2008). Neuroprotection in diet-induced
ketotic rat brain after focal ischemia. J Cereb Blood
Flow Metab 28, 1907–1916.
Puzio-Kuter, A.M. (2011). The role of p53 in metabolic
regulation. Genes Cancer 2, 385–391.
Raffaghello, L., Lee, C., Safdie, F.M., Wei, M., Madia,
F., Bianchi, G., and Longo, V.D. (2008). Starvation-
dependent differential stress resistance protects
normal but not cancer cells against high-dose
chemotherapy. Proc Natl Acad Sci USA 105,
8215–8220.
Raffaghello, L., Safdie, F., Bianchi, G., Dorff, T.,
Fontana, L., and Longo, V.D. (2010). Fasting and
differential chemotherapy protection in patients.
Cell Cycle 9, 4474–4476.
Rahman, M., Muhammad, S., Khan, M.A., Chen,
H., Ridder, D.A., Muller-Fielitz, H., Pokorna, B.,
Vollbrandt, T., Stolting, I., Nadrowitz, R., et  al.
(2014). The beta-hydroxybutyrate receptor HCA2
activates a neuroprotective subset of macrophages.
Nat Commun 5, 3944.
Roberts, D.J., and Miyamoto, S. (2015). Hexokinase II
integrates energy metabolism and cellular protec-
tion:  Akting on mitochondria and TORCing to
autophagy. Cell Death Differ 22, 248–257.
Robey, R.B., and Hay, N. (2009). Is Akt the “Warburg
kinase”?-Akt-energy metabolism interactions and
oncogenesis. Semin Cancer Biol 19, 25–31.
Robey, R.B., Weisz, J., Kuemmerle, N.B., Salzberg,
A.C., Berg, A., Brown, D.G., Kubik, L., Palorini,
R., Al-Mulla, F., Al-Temaimi, R., et  al. (2015).
Metabolic reprogramming and dysregulated
metabolism: cause, consequence and/or enabler of
environmental carcinogenesis? Carcinogenesis 36,
S203–S231.
Roesler, R., Brunetto, A.T., Abujamra, A.L., de Farias,
C.B., Brunetto, A.L., and Schwartsmann, G. (2010).
Current and emerging molecular targets in glioma.
Expert Rev Anticancer Ther 10, 1735–1751.
Saenko, Y., Cieslar-Pobuda, A., Skonieczna, M., and
Rzeszowska-Wolny, J. (2013). Changes of reac-
tive oxygen and nitrogen species and mitochon-
drial functioning in human K562 and HL60 cells
exposed to ionizing radiation. Radiat Res 180,
360–366.
Safdie, F., Brandhorst, S., Wei, M., Wang, W., Lee, C.,
Hwang, S., Conti, P.S., Chen, T.C., and Longo, V.D.
(2012). Fasting enhances the response of glioma to
chemo- and radiotherapy. PLoS One 7, e44603.
Saleh, A.D., Simone, B.A., Palazzo, J., Savage, J.E.,
Sano, Y., Dan, T., Jin, L., Champ, C.E., Zhao, S.,
Lim, M., et al. (2013). Caloric restriction augments
radiation efficacy in breast cancer. Cell Cycle 12,
1955–1963.
Sanli, T., Steinberg, G.R., Singh, G., and Tsakiridis, T.
(2014). AMP-activated protein kinase (AMPK)
beyond metabolism: a novel genomic stress sensor
participating in the DNA damage response path-
way. Cancer Biol Ther 15, 156–169.
Santivasi, W.L., and Xia, F. (2014). Ionizing radia-
tion-induced DNA damage, response, and repair.
Antioxid Redox Signal 21, 251–259.
Sassone-Corsi, P. (2013). Physiology: when metabolism
and epigenetics converge. Science 339, 148–150.
Scheck, A. C., Abdelwahab, M. G., Stafford, P., Kim,
D. Y., Iwai, S., Preul, M. C., and Rho, J. M. (2011).
Mechanistic studies of the ketogenic diet as an
adjuvant therapy for malignant gliomas. Cancer
Res 70[8 Suppl.], 638.
Scheck, A.C., Abdelwahab, M.G., Fenton, K., and
Stafford, P. (2012). The ketogenic diet for the treat-
ment of glioma:  insights from genetic profiling.
Epilepsy Res 100, 327–337.
Schmidt, M., Pfetzer, N., Schwab, M., Strauss, I., and
Kammerer, U. (2011). Effects of a ketogenic diet on
the quality of life in 16 patients with advanced can-
cer: a pilot trial. Nutr Metab (Lond) 8, 54.
99
Chapter 13: Ketogenic Diet as Adjunctive Therapy for Malignant Brain Cancer
Schwartz, K., Chang, H.T., Nikolai, M., Pernicone,
J., Rhee, S., Olson, K., Kurniali, P.C., Hord, N.G.,
and Noel, M. (2015). Treatment of glioma patients
with ketogenic diets:  report of two cases treated
with an IRB-approved energy-restricted ketogenic
diet protocol and review of the literature. Cancer
Metab 3, eCollection 2015.
Semenza, G.L. (2013). HIF-1 mediates metabolic
responses to intratumoral hypoxia and oncogenic
mutations. J Clin Invest 123, 3664–3671.
Seyfried, T.N., and Mukherjee, P. (2005). Targeting
energy metabolism in brain cancer:  review and
hypothesis. Nutr Metab 2, 30–38.
Seyfried, B.T., Kiebish, M., Marsh, J., and Mukherjee,
P. (2009). Targeting energy metabolism in brain
cancer through calorie restriction and the keto-
genic diet. J Cancer Res Ther 5(Suppl 1):S7–15.
Seyfried, T.N., Kiebish, M.A., Marsh, J., Shelton, L.M.,
Huysentruyt, L.C., and Mukherjee, P. (2011).
Metabolic management of brain cancer. Biochim
Biophys Acta 1807, 577–594.
Seyfried, T.N. (2012). Cancer as a Metabolic Disease:
On the Origin, Management and Prevention of
Cancer (Hoboken, NJ: Wiley).
Seyfried, T.N., Marsh, J., Shelton, L.M., Huysentruyt,
L.C., and Mukherjee, P. (2012). Is the restricted
ketogenic diet a viable alternative to the standard
of care for managing malignant brain cancer?
Epilepsy Res 100, 310–326.
Seyfried, T.N., Flores, R., Poff, A.M., D’Agostino, D.P.,
and Mukherjee, P. (2015). Metabolic therapy:  a
new paradigm for managing malignant brain can-
cer. Cancer Lett 356, 289–300.
Shelton, L.M., Huysentruyt, L.C., Mukherjee, P., and
Seyfried, T.N. (2010). Calorie restriction as an
anti-invasive therapy for malignant brain cancer
in the VM mouse. ASN Neuro 2, e00038.
Shimazu, T., Hirschey, M.D., Newman, J., He,
W., Shirakawa, K., Le, M.N., Grueter, C.A.,
Lim, H., Saunders, L.R., Stevens, R.D., et  al.
(2013). Suppression of oxidative stress by beta-
hydroxybutyrate, an endogenous histone deacety-
lase inhibitor. Science 339, 211–214.
Shweiki, D., Neeman, M., Itin, A., and Keshet, E.
(1995). Induction of vascular endothelial growth
factor expression by hypoxia and by glucose defi-
ciency in multicell spheroids:  implications for
tumor angiogenesis. Proc Natl Acad Sci USA 92,
768–772.
Simone, B.A., Champ, C.E., Rosenberg, A.L., Berger,
A.C., Monti, D.A., Dicker, A.P., and Simone, N.L.
(2013). Selectively starving cancer cells through
dietary manipulation: methods and clinical impli-
cations. Future Oncol 9, 959–976.
Skinner, R., Trujillo, A., Ma, X., and Beierle, E.A. (2009).
Ketone bodies inhibit the viability of human neuro-
blastoma cells. J Pediatr Surg 44, 212–216.
Stafford, P., Abdelwahab, M.G., Kim, D.Y., Preul,
M.C., Rho, J.M., and Scheck, A.C. (2010). The
99
ketogenic diet reverses gene expression patterns
and reduces reactive oxygen species levels when
used as an adjuvant therapy for glioma. Nutr
Metab (Lond) 7, 74.
Stafstrom, C.E., and Rho, J.M. (2012). The ketogenic
diet as a treatment paradigm for diverse neuro-
logical disorders. Front Pharmacol 3:59.
Stegeman, H., Span, P.N., Kaanders, J.H., and Bussink,
J. (2014). Improving chemoradiation efficacy
by PI3-K/AKT inhibition. Cancer Treat Rev 40,
1182–1191.
Tisdale, M.J., and Brennan, R.A. (1983). Loss of aceto-
acetate coenzyme A transferase activity in tumours
of peripheral tissues. Br J Cancer 47, 293–297.
Tisdale, M.J., Brennan, R.A., and Fearon, K.C. (1987).
Reduction of weight loss and tumour size in a
cachexia model by a high fat diet. Br J Cancer
56, 39–43.
van den Bent, M.J., Wefel, J.S., Schiff, D., Taphoorn, M.J.,
Jaeckle, K., Junck, L., Armstrong, T., Choucair, A.,
Waldman, A.D., Gorlia, T., et  al. (2011). Response
assessment in neuro-oncology (a report of the
RANO group): assessment of outcome in trials of dif-
fuse low-grade gliomas. Lancet Oncol 12, 583–593.
Vander Heiden, M.G., Cantley, L.C., and Thompson,
C.B. (2009). Understanding the Warburg
effect: the metabolic requirements of cell prolifera-
tion. Science 1029–1033.
Vanitallie, T.B., and Nufert, T.H. (2003).
Ketones: metabolism’s ugly duckling. Nutr Rev 61,
327–341.
Veech, R.L., Chance, B., Kashiwaya, Y., Lardy, H.A.,
and Cahill, G.F., Jr. (2001). Ketone bodies, poten-
tial therapeutic uses. IUBMB Life 51, 241–247.
Wallace, D.C., Fan, W., and Procaccio, V. (2010).
Mitochondrial energetics and therapeutics. Annu
Rev Pathol 5:297–348, 297–348.
Wang, G., Wang, J., Zhao, H., Wang, J., and Tony To
SS (2015a). The role of Myc and let-7a in glioblas-
toma, glucose metabolism and response to ther-
apy. Arch Biochem Biophys 580, 84–92.
Wang, H., Xu, T., Jiang, Y., Yan, Y., Qin, R., and Chen, J.
(2015b). MicroRNAs in human glioblastoma: from
bench to beside. Front Biosci (Landmark Ed) 20,
105–118.
Wang, T., Wang, G., Hao, D., Liu, X., Wang, D., Ning,
N., and Li, X. (2015c). Aberrant regulation of the
LIN28A/LIN28B and let-7 loop in human malig-
nant tumors and its effects on the hallmarks of
cancer. Mol Cancer 14, 125.
Wang, Y., Yuan, J.L., Zhang, Y.T., Ma, J.J., Xu, P., Shi,
C.H., Zhang, W., Li, Y.M., Fu, Q., Zhu, G.F., et al.
(2013). Inhibition of both EGFR and IGF1R
sensitized prostate cancer cells to radiation by
synergistic suppression of DNA homologous
recombination repair. PLoS One 8, e68784.
Warburg, O., Wind, F., and Negelein, E. (1927). The
metabolism of tumors in the body. J Gen Physiol
8, 519–530.
100
100 section II: E merging Clinical Applications and Future Potential
Ward, P.S., and Thompson, C.B. (2012). Metabolic
reprogramming: a cancer hallmark even Warburg
did not anticipate. Cancer Cell 21, 297–308.
Wei, J., Wu, A., Kong, L.Y., Wang, Y., Fuller, G., Fokt,
I., Melillo, G., Priebe, W., and Heimberger, A.B.
(2011). Hypoxia potentiates glioma-mediated
immunosuppression. PLoS One 6, e16195.
Wei, W., Shi, Q., Remacle, F., Qin, L., Shackelford,
D.B., Shin, Y.S., Mischel, P.S., Levine, R.D., and
Heath, J.R. (2013). Hypoxia induces a phase tran-
sition within a kinase signaling network in cancer
cells. Proc Natl Acad Sci USA 110, E1352–E1360.
Weinberg, F., and Chandel, N.S. (2009). Reactive oxy-
gen species-dependent signaling regulates cancer.
Cell Mol Life Sci 66, 3663–3673.
Wolf, A., Agnihotri, S., and Guha, A. (2010). Targeting
metabolic remodeling in glioblastoma multiforme.
Oncotarget 1, 552–562.
Woolf, E.C., Curley, K.L., Liu, Q., Turner, G.H.,
Charlton, J.A., Preul, M.C., and Scheck, A.C.
(2015). The ketogenic diet  alters the hypoxic
response and affects expression of proteins associ-
ated with angiogenesis, invasive potential and vas-
cular permeability in a mouse glioma model. PLoS
One 10, e0130357.
Yamada, M., Tomida, A., Yun, J., Cai, B., Yoshikawa,
H., Taketani, Y., and Tsuruo, T. (1999). Cellular
sensitization to cisplatin and carboplatin with
decreased removal of platinum-DNA adduct
by glucose-regulated stress. Cancer Chemother
Pharmacol 44, 59–64.
Yang, C., Sudderth, J., Dang, T., Bachoo, R.G.,
McDonald, J.G., and DeBerardinis, R.J. (2009).
Glioblastoma cells require glutamate dehydroge-
nase to survive impairments of glucose metabo-
lism or Akt signaling. Cancer Res 69, 7986–7993.
Yang, L., Lin, C., Wang, L., Guo, H., and Wang, X.
(2012). Hypoxia and hypoxia-inducible fac-
tors in glioblastoma multiforme progression
and therapeutic implications. Exp Cell Res 318,
2417–2426.
Youm, Y.H., Nguyen, K.Y., Grant, R.W., Goldberg,
E.L., Bodogai, M., Kim, D., D’Agostino, D.,
Planavsky, N., Lupfer, C., Kanneganti, T.D.,
et  al. (2015). The ketone metabolite beta-
hydroxybutyrate blocks NLRP3 inflammasome-
mediated inflammatory disease. Nat Med 21,
263–269.
Zhang, H., Gu, C., Yu, J., Wang, Z., Yuan, X., Yang,
L., Wang, J., Jia, Y., Liu, J., and Liu, F. (2014).
Radiosensitization of glioma cells by TP53-
induced glycolysis and apoptosis regulator
knockdown is dependent on thioredoxin-1
nuclear translocation. Free Radic Biol Med 69C,
239–248.
Zhou, W., Mukherjee, P., Kiebish, M.A., Markis, W.T.,
Mantis, J.G., and Seyfried, T.N. (2007). The calo-
rically restricted ketogenic diet, an effective alter-
native therapy for malignant brain cancer. Nutr
Metab (Lond) 4, 5.
Zuccoli, G., Marcello, N., Pisanello, A., Servadei, F.,
Vaccaro, S., Mukherjee, P., and Seyfried, T.N.
(2010). Metabolic management of glioblastoma
multiforme using standard therapy together with a
restricted ketogenic diet: case report. Nutr Metab
7, 33–53.
101
14
Metabolic Therapy for Autism Spectrum
Disorder and Comorbidities
N I N G C H E N G, P H D, S U S A N A . M A S I N O, P H D, A N D J O N G M . R H O, M D
OV E RV I E W O F   AU T I S M
S P E C T RU M D I S O R D E R
Autism spectrum disorder (hereafter referred to as
“ASD”) is a group of complex disorders of neurode-
velopment. It is characterized by persistent deficits
in social communication and interaction across
multiple contexts, as well as restricted and repeti-
tive patterns of behavior and interests (DiCicco-
Bloom et al., 2006; Lai et al., 2014; Llaneza et al.,
2010). Further, it is well known that ASD is a mul-
tiorgan disorder that can affect the normal func-
tion of the gastrointestinal, immune, hepatic, and
endocrine systems (Frye et al., 2015; Goines and
Van de Water, 2010; Hsiao, 2013; Mayer et al., 2015;
Patterson, 2011). The term “spectrum” refers to the
wide range of symptoms and levels of impairment
that can occur in individuals with ASD. Severity
of language impairment of the patients, for exam-
ple, can range from deficits in pragmatic use of
language to complete lack of spoken language.
Another level of heterogeneity is the existence of
many comorbidities that include both neurologic
conditions such as epilepsy, sleep impairment,
delays and deficits in motor functions, and psychi-
atric dysfunctions including depression, anxiety,
irritability, attention deficit hyperactivity disorder,
as well as physical health issues such as gastroin-
testinal disturbances. The co-occurrence rate of
one or more non-ASD developmental diagnoses
is as high as 83% (Levy et al., 2010). By contrast,
some people with ASD show extraordinary talents
in math, music, and art.
According to epidemiological studies, ASD
occurs in all racial, ethnic, and socioeconomic
groups, and is highly prevalent. It affects tens of
millions individuals worldwide and costs a fam-
ily approximately $60,000 USD a year on aver-
age. In the United States, the incidence of ASD is
1 in 68 children (1 in 42 boys and 1 in 189 girls)
based on the data released by the Centers for
Disease Control and Prevention in 2014. For con-
troversial reasons, the prevalence of ASD appears
to be on the rise (a 10-fold increase in 40 years),
which may be explained only partly by improved
diagnosis and awareness (Hansen et  al., 2015).
Developmental delay in ASD can be detected as
early as 6  months of age, a critical time for the
development of higher-order social, emotional,
and communications functions (Courchesne et al.,
2007), and increasingly, the importance of early
intervention is being recognized (Orinstein et al.,
2014). However, on average, children identified
with ASD were not diagnosed until after 4 years of
age (CDC, 2014), even though patients can be reli-
ably diagnosed as early as 2 years of age (Kleinman
et al., 2008; Lord et al., 2006).
Because ASD has broad and heterogeneous
manifestations and has been associated with a
plethora of possible etiological factors that are
both genetic and environmental, ASD remains a
clinical syndrome, but not a specific etiologically
defined disorder. The prominent heterogeneity in
autism has made it a challenge to investigate its
neurobiological mechanisms and find interven-
tions for the affected individuals. Currently, only
comorbid manifestations of the disorder can be
alleviated, such as epileptic seizures, psychiatric
disturbances, hyperactivity, sleep disturbances,
and digestive issues, but not the core symptoms
(DiCicco-Bloom et  al., 2006; Lai et  al., 2014;
Llaneza et al., 2010).
Despite much research, there remains limited
knowledge about the pathophysiological mecha-
nisms underlying ASD. It is generally accepted
that genetic susceptibility factors and environmen-
tal influences both contribute to ASD (Chaste and
Leboyer, 2012; Kim and Leventhal, 2015; Sandin
et al., 2014; Tordjman et al., 2014), but many ques-
tions still remain unanswered. Recent efforts using
genome screening and sequencing have identified
102
102 section II: E merging Clinical Applications and Future Potential
rare chromosomal abnormalities and copy num-
ber variations, as well as hundreds of rare gene
mutations associated with autism (Baker and Jeste,
2015; Devlin and Scherer, 2012; Huguet et  al.,
2013; Jeste and Geschwind, 2014). A small num-
ber of these genetic deficits appear highly pene-
trant and sufficient to cause autism by themselves.
However, the majority of the genetic changes asso-
ciated with ASD appear to only increase the risk
to varying degrees. In these cases, environmental
factors influencing early brain development are
thought to play a significant role in the etiology of
the disorder. In other words, in the presence of a
genetic predisposition to autism, a number of pre-
natal, perinatal, and postnatal risk factors appear
to further increase a child’s risk. These include
advanced parental age at time of conception (both
mother and father), maternal illness during preg-
nancy, birth complications, and exposure to tox-
ins and/or drugs during early brain development
(Christensen et  al., 2013; Durkin et  al., 2008;
Gardener et al., 2011; Stromland et al., 1994). It is
important to note that these factors do not cause
ASD by themselves. Rather, they seem to modestly
increase the risks of developing the disorder in
combination with genetic predispositions.
In summary, ASD inflicts a grave burden on
affected individuals, their families, and society as a
whole. Currently, ASD remains a syndrome char-
acterized by behaviors, but not specific etiologies.
Its heterogeneity in terms of both risk factors and
clinical presentations makes it a great challenge to
unravel the pathophysiological mechanisms and
find therapeutic treatments.
M I TO C H O N D R I A L
A N D M E TA B O L I C
D Y S F U N C T I O N I N   AU T I S M
S P E C T RU M D I S O R D E R
Because a combination of diverse factors such as
genetic background and environmental exposure
are thought to contribute to the development of
ASD, it has been hypothesized that these hetero-
geneous influences elicit similar developmental
outcomes by modulating a common central nexus,
a perturbation of which leads to ASD (Berg and
Geschwind, 2012; Geschwind, 2008). Identifying
such a node would provide novel insights into the
development of ASD, while targeting this pathway
could form the basis of therapeutic approaches for
various forms of the disorder. A candidate for this
central node is mitochondrial function, which is
integrated into and key to many cellular pathways.
For example, in addition to the well-known role as
the “powerhouse of the cell,” producing the bulk
of the cellular energy, mitochondria are also criti-
cally involved in cellular metabolism, intracellular
calcium signaling, generation of reactive oxygen
species, and apoptosis (Antico Arciuch et al., 2012;
Murphy, 2009; Palmieri et al., 2010; Rizzuto et al.,
2012; Suen et al., 2008), as well as in the regulation
of innate and adaptive immunity (Weinberg et al.,
2015). Furthermore, mitochondria are known be
affected by many of the same endogenous and
exogenous risk factors of ASD, such as toxins,
drugs, immune activation, and metabolic distur-
bances (Frye and Rossignol, 2011). Thus, under-
standing the role of mitochondrial dysfunction
in ASD may help unify our understanding of this
complex disorder.
Mitochondria play a vital role especially in the
nervous system. The brain has very high energy
demands, as it consumes approximately 20% of the
basal oxygen supply in humans, and hence, calo-
ries consumed by the body, while accounting for
only 2% of the body weight (Raichle and Gusnard,
2002). This is mainly because cells in the nervous
system need a great amount of adenosine triphos-
phate (ATP) to maintain ionic gradients across
cellular membranes, which are essential to gener-
ate membrane excitability for neurotransmission
and plasticity (Harris et  al., 2012). In addition,
mitochondria are involved in various aspects of
neurodevelopment such as the proliferation, dif-
ferentiation, and maturation of neural stem cells;
formation of dendritic arbors and spines; develop-
mental and synaptic plasticity; and the determina-
tion of cell survival and death (Kann and Kovacs,
2007; Kimura and Murakami, 2014; Li et al., 2004;
Mattson et  al., 2008; Xavier et  al., 2015). Thus, it
is not surprising that there are now multiple lines
of evidence from clinical, genetic, and biochemi-
cal studies—in both humans and animal models—
supporting a role for mitochondrial dysfunction in
the etiology of ASD (Dhillon et al., 2011; Frye and
Rossignol, 2011; Haas, 2010; Legido et  al., 2013;
Rossignol and Frye, 2012).
Clinical evidence points to the involvement
of mitochondrial disease and dysfunction in
ASD. The prevalence of mitochondrial disease in
the ASD population is estimated to be about 5%,
500 times higher than that found in the general
population (≈0.01%). The prevalence of abnor-
mal values of biomarkers related to mitochon-
drial function in the ASD population may be even
higher, suggesting that as many as 30% of children
with ASD may experience mitochondrial dys-
function. For example, almost one-third of autis-
tic children had elevations in plasma lactate and/
or the lactate-to-pyruvate ratio, and the levels of
103
Chapter 14: Metabolic Therapy for Autism Spectrum Disorder
many other mitochondrial biomarkers (pyruvate,
carnitine, and ubiquinone) are significantly dif-
ferent between ASD and controls (Rossignol and
Frye, 2012). Conversely, the rate of ASD is higher
among children with mitochondrial disease
(Randolph-Gips and Srinivasan, 2012). Together,
these findings suggest that mitochondrial dys-
function likely represents a significant subclass
of ASD.
Furthermore, common comorbidities of ASD
also suggest the involvement of mitochondrial
dysfunction. One of the most significant comor-
bidities associated with ASD is epilepsy. A number
of studies suggest that epilepsy affects a great pro-
portion of individuals with ASD, with the reported
prevalence of epilepsy in the ASD population
ranging from 5% to 38%, which is much higher
than the 1%–2% prevalence in the general popu-
lation of children (Frye, 2015). In contrast, sei-
zures have been reported to occur in about 35% to
60% of individuals with biochemically confirmed
mitochondrial disease (Rahman, 2012). Thus,
shared symptoms suggest a common etiopathol-
ogy. Similarly, gastrointestinal dysfunction, a fre-
quent comorbidity of ASD (Chaidez et al., 2014), is
also commonly reported in mitochondrial disease
(Frye et al., 2015).
Interestingly, many animal models of ASD
also display mitochondrial dysfunction. These
models include ones based on both susceptibility
genes such as MECP2, UBE3A, and SLC25A12
as well as environmental risk factors includ-
ing maternal immune activation, and exposure
to propionic acid and valproic acid. The cur-
rent evidence linking mitochondrial perturba-
tions to ASD is summarized in Table 14.1. Taken
together, we believe that mitochondria act as a
central nexus responding to and regulating many
domains of cellular biology that have been impli-
cated in ASD.
K E TO G E N I C D I E T A S   A
PROMISING THERAPY
F O R   AU T I S M S P E C T R U M
DISORDER
The ketogenic diet (KD) is a special high-fat, low-
carbohydrate diet, which is a remarkably effective
nonpharmacological treatment for patients with
medically intractable epilepsy (Neal et  al., 2008).
Based on the historical observation that fasting
or starvation can render antiseizure effects, the
KD was designed to reproduce the biochemical
changes seen in these physiological states (Masino
and Rho, 2012). Recently, dietary and metabolic
therapies have been attempted in a wider variety
103
of neurological diseases including Alzheimer’s
disease, Parkinson’s disease, multiple sclerosis,
sleep disorders, and brain cancer (Frye et al., 2013;
Napoli et al., 2014; Stafstrom and Rho, 2012).
There are a number of mechanisms through
which the KD may provide neuroprotective activ-
ity. Two hallmark features of biochemical changes
after the KD treatment are the increase in ketone
body production by the liver through fatty acid
oxidation and a reduction in blood glucose lev-
els (Stafstrom and Rho, 2012). Specific polyun-
saturated fatty acids such as arachidonic acid,
docosahexaenoic acid, and eicosapentaenoic acid
might themselves regulate neuronal membrane
excitability (Voskuyl and Vreugdenhil, 2001),
reduce inflammation (Cullingford, 2008; Jeong et
al., 2011), or decrease the production of reactive
oxygen species (ROS) by mitochondria (Kim do
and Rho, 2008). Additionally, ketone bodies have
been shown to possess neuroprotective properties
through improved bioenergetics. They have been
reported to raise ATP levels and reduce ROS pro-
duction through enhancement of nicotinamide
adenine dinucleotide (NADH) oxidation and
inhibition of mitochondrial permeability tran-
sition (Kim do et al., 2007; Kim do et al., 2015).
Furthermore, the KD has been shown to stimulate
mitochondrial biogenesis (Ahola-Erkkila et al.,
2010; Bough et al., 2006).
The second major biochemical feature of the
KD is the reduction of glycolysis. In addition to
suppressing seizures, it also improves mitochon-
drial function and decreases oxidative stress,
reduces activity of pro-apoptotic factors, and
inhibits inflammatory mediators such as inter-
leukins and tumor necrosis factor alpha (Maalouf
et al., 2009).
Beyond enhancing bioenergetics and mito-
chondrial function, the KD has also been shown
to modulate the metabolism of γ-aminobutyric
acid (GABA) and acetylcholine, two major cen-
tral nervous system neurotransmitters (Napoli
et al., 2014), as well as purines (such as ATP and
adenosine) that have pleiotropic neural modu-
latory roles (Masino et  al., 2010; Masino et  al.,
2009). In addition, the KD has also been reported
to regulate energy-sensing pathways such as
those involving the insulin-like growth factor
and the mammalian target of rapamycin (Napoli
et al., 2014).
As mentioned above, mitochondrial and meta-
bolic dysfunction may play a key role in the patho-
physiology of ASD. Based on research showing
that the KD provides numerous neuroprotective
effects through modulating mitochondrial and
104
TABLE 14.1 STUDIES SHOWING A LINK BET WEEN ASD
Reference
Al-Mosalem et al., 2009
Boccuto et al., 2013
Celestino-Soper
et al., 2012
Chen et al., 2015
Cohen et al., 1976
Correia et al., 2006
Filipek et al., 2004
Frye et al., 2013
Glessner et al., 2009
Goh et al., 2014
Kent et al., 2006
Kim et al., 2011 Multiple
Silverman et al., 2008
Kuwabara et al., 2013
Laszlo et al., 1994
Maestrini et al., 2010
Moreno et al., 1992
Oliveira et al., 2005
Poling et al., 2006
Rose et al., 2012
Rose et al., 2014
Ramoz et al., 2004 Multiple
Segurado et al., 2005
Turunen et al., 2008
Tang et al., 2013
Ahn et al., 2014
Jin et al., 2015
Kriaucionis et al., 2006
Macfabe et al., 2012
AND MITOCHONDRIAL DYSFUNCTION
Cases Evidence of mitochondrial dysfunction
Studies in ASD patients*
30 Increased plasma lactate levels and activity of creatine kinase
87 Decreased tryptophan metabolism in lymphoblastoid cell lines
909 The deficiency of the trimethyllysine hydroxylase epsilon, which encodes the
first enzyme in carnitine biosynthesis, was more frequent in probands from
male-male multiplex ASD families
78 Mitochondrial DNA copy number in peripheral blood cells was elevated
25 Increased serum creatine phosphokinase levels
241 Increased plasma lactate levels and lactate/pyruvate ratio, but not associated
with the variation at the SLC25A12 gene
100 Reduced levels of carnitine and pyruvate, increased levels of alanine and ammo-
nia in serum
213 Abnormal acyl-carnitine panels and glutathione metabolism in blood samples
859 Copy number variations in genes involved in the ubiquitin degradation were
implicated in susceptibility for ASD
75 Lactate doublets detected by brain magnetic resonance spectroscopic imaging
were present at a higher rate
129 The 3243A>G mitochondrial DNA mutation was concluded to be a rare cause
of isolated Asperger syndrome
Polymorphism in a mitochondrial aspartate/glutamate carrier gene (SLC25A12)
families is found to be associated with restricted repetitive behavior in autism
25 higher plasma levels of arginine and taurine, and lower levels of 5-oxoproline
and lactic acid
30 Increased serum lactate and pyruvate levels
127 Gene encodes an inner mitochondrial membrane protease-like protein
(IMMP2L) was implicated in susceptibility for ASD
60 Increased lactate and pyruvate levels
69 20% of ASD patients showed hyperlactacidemia, while 7% were classified with
definite mitochondrial respiratory chain disorder
159 Increase blood aspartate aminotransferase and creatine kinase levels
43 Primary immune cells in the blood have a more oxidized intracellular and
extracellular microenvironment and a deficit in glutathione-mediated redox/
antioxidant capacity
25 Mitochondrial dysfunction observed in a subset of autism lymphoblastoid cell
lines
Polymorphism in a mitochondrial aspartate/glutamate carrier gene (SLC25A12)
families is found to be associated with autism
45 Mitochondrial function and intracellular redox status were compromised in the
pyramidal neurons of the temporal cortex
Studies in animal models
Mitochondrial dysfunction in a rat model of ASD using prenatal VPA exposure
Mecp2, whose mutations cause Rett syndrome, was observed to regulate mito-
chondrial bioenergetics through a glutamine transporter in microglia.
Mitochondrial abnormalities observed in Mecp2-null mouse, a model of Rett
syndrome.
Mitochondrial dysfunction observed in a rat ASD model in which pro-
pionic acid, an enteric bacterial fermentation product, is infused
intracerebroventricularly
(continued)
105

Chapter 14: Metabolic Therapy for Autism Spectrum Disorder 105

TABLE 14.1 CONTINUED

Reference Cases Evidence of mitochondrial dysfunction

Naviaux et al., 2013 Antipurinergic therapy improves the autism-like features in the maternal
Naviaux et al., 2014 immune activation mouse model and the Fragile X mouse model
Naviaux et al., 2015
Park et al., 2014 Dietary therapy with triheptanoin enhanced mitochondrial substrate use and
improved metabolism and behaviors of Mecp2-null mouse model of Rett
syndrome
Sakurai et al., 2010 Loss of a mitochondrial aspartate-glutamate carrier gene results in hypomyelin-
ation. Myelin deficits in slice cultures from KO mice are reversed by admin-
istration of pyruvate
Su et al., 2011 Mitochondrial dysfunction observed in hippocampal neurons of the UBE3A
deficient mouse model for Angelman syndrome
Zhao et al., 2010 ASD-like features observed in neuronal glucose transporter isoform 3 deficient
mice

* Only studies reporting more than 25 subjects are included in this table

metabolic pathways, it is logical to hypothesize
that this diet could prove to be beneficial for autis-
tic individuals.
THE EFFECTS OF THE
K E TO G E N I C D I E T
O N   AU T I S M S P E C T R U M
D I S O R D E R PAT I E N T S
To date, there have been limited trials of treat-
ing autism patients with the KD. Evangeliou and
colleagues carried out a pilot prospective study
on autistic children aged between 4 and 10 years
to investigate the role of the KD, specifically a
modified medium chain triglyceride (MCT) diet
(Evangeliou et  al., 2003). Of the 18 patients who
adhered to the diet, improvement was recorded in
most individuals, based on several parameters and
in accordance with the Childhood Autism Rating
Scale (CARS). Significant (>12 units of decrease
in CARS) and average (>8–12 units of decrease
in CARS) improvement was recorded in two and
eight patients, respectively, while minor (2–8 units
of decrease in CARS) improvement was reported
in the remaining eight patients.
In a separate study, Herbert and Buckley
reported the history of a 12-year-old child with
autism and epilepsy treated with a gluten- and
casein-free KD with fats composed mostly of
medium chain triglycerides (Herbert and Buckley,
2013). The patient showed significant improve-
ment in seizure activity, resolution of morbid
obesity, and improvement in cognitive and behav-
ioral functioning. Over the course of several years
following her initial diagnosis, the child’s CARS
score decreased from 49 to 17, representing a
change from severe autism to a nonautistic state,
and her intelligence quotient increased 70 points.
Finally, Spilioti and colleagues reported the
effects following the KD treatment in a group of
Greek children with ASD aged between 3.5 and
6 years (Spilioti et al., 2013). Of the 6 patients who
successfully implemented the diet, significant and
average improvement was recorded in one and two
patients, respectively, while minor improvement
was reported in the remaining three patients.
In summary, results from these few studies
show that more than 50% of the autism patients
who received the KD treatment showed moderate-
to-significant improvement, while the remaining
displayed minor improvement. These preliminary
studies suggest that the KD could be a promising
therapy for ASD; more and larger clinical studies
are needed.
T R E AT I N G A N I M A L M O D E L S
O F   AU T I S M W I T H   T H E
K E TO G E N I C   D I E T
Experiments using animal models of human dis-
eases have contributed much to our understand-
ing of the mechanisms of disease, and the efficacy
of potential therapeutics in preclinical settings.
Due to the complex etiology of ASD, many ani-
mal models exist, and several of them have been
used to test the effects of the KD. For example,
succinic semialdehyde dehydrogenase (SSADH)
deficiency is a rare autosomal recessive condition
106
106 section II: E merging Clinical Applications and Future Potential
that results in mild-to-moderate mental retar-
dation, disproportionate language dysfunction,
seizures, hypotonia, hyporeflexia, hallucinations,
and autistic behaviors (Pearl et  al., 2003). In an
animal model of SSADH deficiency, the SSADH
knockout mouse, Nylen and colleagues found that
KD treatment normalized electroencephalogram
(EEG) activity. In addition, the decrease in min-
iature inhibitory postsynaptic currents in pyra-
midal cells of CA1 hippocampal slices observed
in the mutant animals was completely restored by
the KD when compared with wild-type controls.
In contrast, there were no significant differences
between the groups in terms of miniature excit-
atory postsynaptic currents. Moreover, GABAA
receptor-associated chloride channel binding
of [35S]tert-butylbicyclophosphorothionate was
restored fully in hippocampus and frontoparietal
cortex, but not in amygdala or thalamus, of KD-
fed mutant mice. Behaviorally, KD-treated mutant
animals experienced significantly fewer seizures
compared with mutant animals fed with the con-
trol diet (Nylen et al., 2008).
Another study examined the effects of restricted
standard and ketogenic diets on a mouse model of
Rett syndrome. Rett syndrome is a neurodevel-
opmental disorder characterized by normal early
growth and development followed by a slowing
of development, impairment of motor functions,
seizure susceptibility, and intellectual disability. In
most cases, it is caused by mutations in the methyl-
CpG-binding protein 2 (MECP2) gene (Amir
et  al., 1999). Children with Rett syndrome often
exhibit autistic-like behaviors in the early stages
(Percy, 2011). By feeding animals with standard
chow in unrestricted or restricted amounts or a
KD in restricted amounts, Mantis and colleagues
found that performance in assays of motor behav-
ior and anxiety was significantly worse in Mecp2
mutant mice compared with wild-type animals,
and restriction of either the standard diet or the
KD improved motor behavior and reduced anxiety
in the mutant animals (Mantis et al., 2009).
More recently, using the BTBR T+tf/J (BTBR)
mouse model of autism, Ruskin and colleagues
investigated the effects of the KD on core behav-
ioral abnormalities that define autism (Ruskin
et al., 2013). The BTBR inbred strain is one of the
most clinically relevant animal models of autism,
mainly because it displays all the core behavioral
features that define the disorder—specifically,
impaired social behaviors and communication as
well as stereotyped behaviors and resistance to
change (Ellegood and Crawley, 2015; McFarlane
et  al., 2008; Meyza et  al., 2013; Moy et  al., 2007;
Ruskin et al., 2013; Smith et al., 2014). This model
was identified in an extensive effort to character-
ize 10 inbred mouse strains in terms of possess-
ing autism-like behaviors (Moy et  al., 2007), and
was later shown to display behavioral phenotypes
relevant to all three core diagnostic symptoms of
ASD (McFarlane et  al., 2008). Additional tests
conducted in multiple independent laboratories
further confirmed that BTBR animals display
prominent deficits in multiple social interaction
and communication assays, and exhibit repetitive
and stereotyped behaviors. Thus, the BTBR strain
has been widely regarded as a consistent and robust
animal model of autism, reflecting more common
forms of ASD (Gaugler et  al., 2014). During the
relatively short time since its discovery as an ASD
model, the BTBR strain has been increasingly used
to study the etiology and potential intervention of
ASD (Llaneza et al., 2010; McFarlane et al., 2008;
Moy et al., 2007; Ruskin et al., 2013). These stud-
ies have identified alterations in BTBR animals
at many levels, including genetic and epigenetic;
synaptic; neuroanatomic and functional connec-
tivity; and immunological—all of which have been
implicated in individuals with ASD (Ellegood and
Crawley, 2015; Llaneza et  al., 2010; Meyza et  al.,
2013; Shpyleva et  al., 2014). To test the effects of
the KD on the BTBR model, Ruskin and colleagues
fed the animals with the diet for 3 weeks and then
performed an array of behavioral assays. They
reported that the BTBR mice showed increased
sociability in a three-chamber test, decreased
self-directed repetitive behavior, and improved
social communication in a food-preference assay
(Ruskin et  al., 2013). In addition, the authors
showed that the behavioral improvements were
probably not related to any antiseizure effect of the
diet, because no spontaneous seizures or abnormal
EEG features were observed in the BTBR animals.
These results strongly suggest that the KD may be
able to improve the core behavioral symptoms of
autism.
As mentioned earlier, both genetic and envi-
ronmental factors contribute to the risks of devel-
oping ASD. An example of environmental risk
factors is exposure to exogenous chemicals such
as valproic acid (VPA) during pregnancy. Valproic
acid is a pharmacological anticonvulsant used in
humans primarily for the treatment of epilepsy
and migraine. Epidemiological studies showed
that maternal use of VPA during pregnancy was
associated with a significantly increased risk of
the offspring being autistic (Bromley et al., 2013;
Christensen et al., 2013). The model of VPA expo-
sure is one of the most frequently studied animal
107
Chapter 14: Metabolic Therapy for Autism Spectrum Disorder
models of ASD (Chomiak et  al., 2013; Roullet
et  al., 2013). This model exhibits many similar
structural and behavioral features of ASD indi-
viduals. For example, it has been shown that VPA
exposure in both rats and mice leads to autistic-
like behavioral impairment including decreased
social interactions, increased repetitive behav-
iors, and deficits in communication, as well as
increased anxiety and increased sensitivity to both
painful and nonpainful stimuli. Overall, the VPA
rodent model possesses construct, face, and pre-
dictive validity for ASD studies. Using this model,
Ahn and colleagues found that the KD treatment
recovered part of the play behavior of juvenile rats
exposed to VPA prenatally, shown by increased
number of play initiations/attacks. However, the
diet did not change the disrupted pattern of play
responses (Ahn et  al., 2014). Interestingly, the
authors also found that prenatal exposure to VPA
altered mitochondrial respiration, and the KD was
able to partially restore it. Specifically, basal and
ATP-linked respiration was significantly increased
in the VPA animals, but was partly normalized by
the KD (Ahn et al., 2014).
FUTURE RESEARCH
DIRECTIONS USING
M E TA B O L I C T H E R A P Y
Although strong evidence exists that mitochon-
drial and metabolic dysfunction may play a piv-
otal role in the complex pathophysiology of ASD,
the exact mechanisms remain undefined. For
example, is mitochondrial dysfunction a cause,
consequence, or epiphenomenon of the disrupted
neurodevelopment observed in ASD? Considering
that ASD is a multisystem disorder, in which tis-
sue, and more specifically, in which cell types is
mitochondrial dysfunction relevant to the dis-
ease state? Can impaired mitochondrial function
define a subtype of the otherwise heterogeneous
patient population? Is the severity of mitochon-
drial and metabolic disturbance correlated with
any behavioral abnormalities or comorbidities?
When mitochondrial function is improved, can it
ameliorate the general features of ASD? Answering
these questions will require the collective efforts of
many basic, translational, and clinical research-
ers, as well as investigators with diverse expertise
in multiorgan dysfunction, metabolism, and both
genetic and environmental risk factors.
Similarly, although the KD has shown prom-
ising results from limited studies in both patients
and animal models, a mechanistic understanding
of its effects in ASD is lacking. Borrowing from the
rich literature on the KD treatment for epilepsy,
107
shifts in energy metabolism and the direct actions
of the ketone bodies on the mitochondria are two
of the promising candidate mechanisms. In addi-
tion, the optimum formulation of the KD needs
to be established, which may be different for ASD
compared with epilepsy, and distinct biomarkers
need to be developed that will accurately track
behavioral symptoms as well as disease ontogeny.
Last but not least, the potential side effects of the
KD should be carefully investigated, especially for
ASD patients who may also suffer from intrinsic
metabolic derangements.
REFERENCES
Ahn, Y., Narous, M., Tobias, R., Rho, J. M., and
Mychasiuk, R. (2014). The ketogenic diet modifies
social and metabolic alterations identified in the
prenatal valproic acid model of autism spectrum
disorder. Dev Neurosci 36, 371–380. doi:10.1159/
000362645
Ahola-Erkkila, S., Carroll, C.J., Peltola-Mjosund,
K., Tulkki, V., Mattila, I., Seppanen-Laakso, T.,
Oresic, M., Tyynismaa, H., and Suomalainen, A.
(2010). Ketogenic diet slows down mitochondrial
myopathy progression in mice. Hum Mol Gen 19,
1974–1984.
Al-Mosalem, O. A., El-Ansary, A., Attas, O., and Al-
Ayadhi, L. (2009). Metabolic biomarkers related
to energy metabolism in Saudi autistic chil-
dren. Clinic Biochem 42, 949–957. doi:10.1016/
j.clinbiochem.2009.04.006
Amir, R.E., Van den Veyver, I.B., Wan, M., Tran, C.Q.,
Francke, U., and Zoghbi, H.Y. (1999). Rett syn-
drome is caused by mutations in X-linked MECP2,
encoding methyl-CpG-binding protein 2. Nat
Genet 23, 185–188.
Antico Arciuch, V.G., Elguero, M.E., Poderoso, J.J.,
and Carreras, M.C. (2012). Mitochondrial regula-
tion of cell cycle and proliferation. Antioxid Redox
Signal 16, 1150–1180.
Baker, E., and Jeste, S.S. (2015). Diagnosis and man-
agement of autism spectrum disorder in the era
of genomics:  rare disorders can pave the way for
targeted treatments. Pediatr Clin North Am 62,
607–618.
Berg, J.M., and Geschwind, D.H. (2012). Autism
genetics:  searching for specificity and conver-
gence. Genome Biol 13, 247.
Boccuto, L., Chen, C. F., Pittman, A. R., Skinner,
C. D., McCartney, H. J., Jones, K., Bochner, B.
R., Stevenson, R. E., Schwartz, C. E. (2013).
Decreased tryptophan metabolism in patients
with autism spectrum disorders. Mol Autism 4,
16. doi:10.1186/2040-2392-4-16
Bough, K.J., Wetherington, J., Hassel, B., Pare, J.F.,
Gawryluk, J.W., Greene, J.G., Shaw, R., Smith,
Y., Geiger, J.D., and Dingledine, R.J. (2006).
Mitochondrial biogenesis in the anticonvulsant
108
108 section II: E merging Clinical Applications and Future Potential
mechanism of the ketogenic diet. Ann Neurol 60,
223–235.
Bromley, R.L., Mawer, G.E., Briggs, M., Cheyne, C.,
Clayton-Smith, J., Garcia-Finana, M., Kneen,
R., Lucas, S.B., Shallcross, R., Baker, G.A., et  al.
(2013). The prevalence of neurodevelopmental
disorders in children prenatally exposed to anti-
epileptic drugs. J Neurol Neurosurg Psychiatry 84,
637–643.
CDC. (2014). Prevalence of autism spectrum disor-
der among children aged 8  years  —autism and
developmental disabilities monitoring network, 11
sites, United States, 2010. Morbidity and Mortality
Weekly Report (MMWR) Surveillance Summaries
63, 1–21.
Celestino-Soper, P. B., Violante, S., Crawford, E. L.,
Luo, R., Lionel, A. C., Delaby, E., Cai, G., Sadikovic,
B., Lee, K., Lo, C., et  al. (2012). A common X-
linked inborn error of carnitine biosynthesis may
be a risk factor for nondysmorphic autism. Proc
Natl Acad Sci U S A 109, 7974–7981. doi:10.1073/
pnas.1120210109
Chaidez, V., Hansen, R.L., and Hertz-Picciotto, I.
(2014). Gastrointestinal problems in children with
autism, developmental delays or typical develop-
ment. J Autism Dev Disord 44, 1117–1127.
Chaste, P., and Leboyer, M. (2012). Autism risk fac-
tors:  genes, environment, and gene-environment
interactions. Dialogues Clin Neurosci. 14,
281–292.
Chen, S., Li, Z., He, Y., Zhang, F., Li, H., Liao, Y.,
Wei, Z., Wan, G., Xiang, X., Hu, M., et al. (2015).
Elevated mitochondrial DNA copy number in
peripheral blood cells is associated with child-
hood autism. BMC Psychiatry 15, 50. doi:10.1186/
s12888-015-0432-y
Chomiak, T., Turner, N., and Hu, B. (2013). What
we have learned about autism spectrum disorder
from valproic acid. Patholog Res Int 2013, 712758.
doi: 10.1155/2013/712758.
Christensen, J., Gronborg, T.K., Sorensen, M.J.,
Schendel, D., Parner, E.T., Pedersen, L.H., and
Vestergaard, M. (2013). Prenatal valproate expo-
sure and risk of autism spectrum disorders and
childhood autism. JAMA 309, 1696–1703.
Cohen, D. J., Johnson, W., Caparulo, B. K., and Young,
J. G. (1976). Creatine phosphokinase levels in
children with severe developmental disturbances.
Arch Gen Psychiatry 33, 683–686
Correia, C., Coutinho, A. M., Diogo, L., Grazina,
M., Marques, C., Miguel, T., Ataide, A., Almeida,
J., Borges, L., Oliveira, C., et  al. (2006). Brief
report: High frequency of biochemical markers for
mitochondrial dysfunction in autism: no associa-
tion with the mitochondrial aspartate/glutamate
carrier SLC25A12 gene. J Autism Dev Disord 36,
1137–1140. doi:10.1007/s10803-006-0138-6
Courchesne, E., Pierce, K., Schumann, C.M., Redcay,
E., Buckwalter, J.A., Kennedy, D.P., and Morgan,
J. (2007). Mapping early brain development in
autism. Neuron 56, 399–413.
Cullingford, T. (2008). Peroxisome proliferator-
activated receptor alpha and the ketogenic diet.
Epilepsia 49(Suppl 8), 70–72.
Devlin, B., and Scherer, S.W. (2012). Genetic archi-
tecture in autism spectrum disorder. Curr Opin
Genetics Dev 22, 229–237.
Dhillon, S., Hellings, J.A., and Butler, M.G. (2011).
Genetics and mitochondrial abnormalities in autism
spectrum disorders:  a review. Curr Genomics 12,
322–332.
DiCicco-Bloom, E., Lord, C., Zwaigenbaum, L.,
Courchesne, E., Dager, S.R., Schmitz, C., Schultz,
R.T., Crawley, J., and Young, L.J. (2006). The devel-
opmental neurobiology of autism spectrum disor-
der. J Neurosci 26, 6897–6906.
Durkin, M.S., Maenner, M.J., Newschaffer, C.J.,
Lee, L.C., Cunniff, C.M., Daniels, J.L., Kirby,
R.S., Leavitt, L., Miller, L., Zahorodny, W., et  al.
(2008). Advanced parental age and the risk of
autism spectrum disorder. Am J Epidemiol 168,
1268–1276.
Ellegood, J., and Crawley, J.N. (2015). Behavioral and
neuroanatomical phenotypes in mouse models of
autism. Neurotherapeutics 12, 521–533.
Evangeliou, A., Vlachonikolis, I., Mihailidou, H.,
Spilioti, M., Skarpalezou, A., Makaronas, N.,
Prokopiou, A., Christodoulou, P., Liapi-Adamidou,
G., Helidonis, E., et  al. (2003). Application of a
ketogenic diet in children with autistic behav-
ior: pilot studyJ Child Neurol 18, 113–118.
Filipek, P.A., Juranek, J., Nguyen, M.T., Cummings,
C., and Gargus, J.J. (2004). Relative carnitine defi-
ciency in autism. J Autism Dev Disord 34, 615–623
Frye, R.E., and Rossignol, D.A. (2011). Mitochondrial
dysfunction can connect the diverse medical
symptoms associated with autism spectrum disor-
ders. Pediatr Res 69, 41R–47R.
Frye, R.E., Melnyk, S., and Macfabe, D.F. (2013a).
Unique acyl-carnitine profiles are potential bio-
markers for acquired mitochondrial disease in
autism spectrum disorder. Transl Psychiatry 3,
e220. doi:10.1038/tp.2012.143
Frye, R.E., Rossignol, D., Casanova, M.F., Brown,
G.L., Martin, V., Edelson, S., Coben, R., Lewine,
J., Slattery, J.C., Lau, C., et  al. (2013b). A review
of traditional and novel treatments for seizures
in autism spectrum disorder:  findings from a
systematic review and expert panel. Front Public
Health 1, 31.
Frye, R.E. (2015). Metabolic and mitochondrial disor-
ders associated with epilepsy in children with autism
spectrum disorder. Epilepsy Behav 47, 147–157.
Frye, R.E., Rose, S., Slattery, J., and MacFabe, D.F.
(2015). Gastrointestinal dysfunction in autism
spectrum disorder:  the role of the mitochondria
and the enteric microbiome. Microb Ecol Health
Dis 26, 27458.
109
Chapter 14: Metabolic Therapy for Autism Spectrum Disorder
Gardener, H., Spiegelman, D., and Buka, S.L. (2011).
Perinatal and neonatal risk factors for autism:  a
comprehensive meta-analysis. Pediatrics 128,
344–355.
Gaugler, T., Klei, L., Sanders, S.J., Bodea, C.A.,
Goldberg, A.P., Lee, A.B., Mahajan, M., Manaa, D.,
Pawitan, Y., Reichert, J., et al. (2014). Most genetic
risk for autism resides with common variation.
Nat Genet 46, 881–885.
Geschwind, D.H. (2008). Autism:  many genes, com-
mon pathways? Cell 135, 391–395.
Glessner, J. T., Wang, K., Cai, G., Korvatska, O., Kim,
C. E., Wood, S., Zhang, H., Estes, A., Brune,
C. W., et  al. (2009). Autism genome-wide copy
number variation reveals ubiquitin and neuro-
nal genes. Nature 459, 569–573. doi:10.1038/
nature07953
Goh, S., Dong, Z., Zhang, Y., DiMauro, S., and Peterson,
B.S. (2014). Mitochondrial dysfunction as a neuro-
biological subtype of autism spectrum disorder: evi-
dence from brain imaging. JAMA Psychiatry 71,
665–671. doi:10.1001/jamapsychiatry.2014.179
Goines, P., and Van de Water, J. (2010). The immune
system’s role in the biology of autism. Curr Opin
Neurol 23, 111–117.
Haas, R.H. (2010). Autism and mitochondrial disease.
Dev Dis Res Rev 16, 144–153.
Hansen, S.N., Schendel, D.E., and Parner, E.T. (2015).
Explaining the increase in the prevalence of autism
spectrum disorders:  the proportion attributable
to changes in reporting practices. JAMA Pediatr
169, 56–62.
Harris, J.J., Jolivet, R., and Attwell, D. (2012). Synaptic
energy use and supply. Neuron 75, 762–777.
Herbert, M.R., and Buckley, J.A. (2013). Autism and
dietary therapy: case report and review of the lit-
erature. J Child Neurol 28, 975–982.
Hsiao, E.Y. (2013). Immune dysregulation in autism
spectrum disorder. International Rev Neurobiol
113, 269–302.
Huguet, G., Ey, E., and Bourgeron, T. (2013). The
genetic landscapes of autism spectrum disorders.
Annu Rev Genomics Hum Genet 14, 191–213.
Jeong, E.A., Jeon, B.T., Shin, H.J., Kim, N., Lee, D.H.,
Kim, H.J., Kang, S.S., Cho, G.J., Choi, W.S., and
Roh, G.S. (2011). Ketogenic diet-induced peroxi-
some proliferator-activated receptor-gamma acti-
vation decreases neuroinflammation in the mouse
hippocampus after kainic acid-induced seizures.
Exp Neurol 232, 195–202.
Jeste, S.S., and Geschwind, D.H. (2014). Disentangling
the heterogeneity of autism spectrum disor-
der through genetic findings. Nat Rev Neurol
10, 74–81.
Jin, L. W., Horiuchi, M., Wulff, H., Liu, X. B.,
Cortopassi, G. A., Erickson, J. D., and Maezawa,
I. (2015). Dysregulation of glutamine trans-
porter SNAT1 in Rett syndrome microglia:  a
mechanism for mitochondrial dysfunction
109
and neurotoxicity. J Neurosci 35, 2516–2529.
doi:10.1523/JNEUROSCI.2778-14.2015
Kann, O., and Kovacs, R. (2007). Mitochondria and
neuronal activity. Am J PhysiolCell Physiol 292,
C641–657.
Kent, L., Lambert, C., Pyle, A., Elliott, H., Wheelwright, S.,
Baron-Cohen, S., and Chinnery, P.F.The mitochon-
drial DNA A3243A>G mutation must be an infre-
quent cause of Asperger syndrome. J. Pediatr 149,
280–281. doi:10.1016/j.jpeds.2005.08.060
Kim do, Y., Davis, L.M., Sullivan, P.G., Maalouf, M.,
Simeone, T.A., van Brederode, J., and Rho, J.M.
(2007). Ketone bodies are protective against oxi-
dative stress in neocortical neurons. J Neurochem
101, 1316–1326.
Kim do, Y., and Rho, J.M. (2008). The ketogenic diet
and epilepsy. Curr Opin Clin Nutr Metab Care 11,
113–120.
Kim do, Y., Simeone, K.A., Simeone, T.A., Pandya, J.D.,
Wilke, J.C., Ahn, Y., Geddes, J.W., Sullivan, P.G.,
and Rho, J.M. (2015). Ketone bodies mediate anti-
seizure effects through mitochondrial permeabil-
ity transition. Ann Neurol 78, 77–87.
Kim, S.J., Silva, R.M., Flores, C.G., Jacob, S., Guter,
S., Valcante, G., Zaytoun, A.M., Cook, E.H.,
and Badner, J.A. (2011). A quantitative associa-
tion study of SLC25A12 and restricted repetitive
behavior traits in autism spectrum disorders. Mol
Autism 2, 8. doi:10.1186/2040-2392-2-8
Kim, Y.S., and Leventhal, B.L. (2015). Genetic epide-
miology and insights into interactive genetic and
environmental effects in autism spectrum disor-
ders. Biol Psychiatry 77, 66–74.
Kimura, T., and Murakami, F. (2014). Evidence that
dendritic mitochondria negatively regulate den-
dritic branching in pyramidal neurons in the neo-
cortex. J. Neurosci 34, 6938–6951.
Kleinman, J.M., Ventola, P.E., Pandey, J., Verbalis,
A.D., Barton, M., Hodgson, S., Green, J., Dumont-
Mathieu, T., Robins, D.L., and Fein, D. (2008).
Diagnostic stability in very young children with
autism spectrum disorders. J Autism Dev Disord
38, 606–615.
Kriaucionis, S., Paterson, A., Curtis, J., Guy, J.,
Macleod, N., and Bird, A. (2006). Gene expression
analysis exposes mitochondrial abnormalities in a
mouse model of Rett syndrome. Mol Cell Biol 26,
5033–5042. doi:10.1128/MCB.01665-05
Kuwabara, H., Yamasue, H., Koike, S., Inoue, H.,
Kawakubo, Y., Kuroda, M., Takano, Y., Iwashiro,
N., Natsubori, T., Aoki, Y., et  al. (2013). Altered
metabolites in the plasma of autism spectrum dis-
order:  a capillary electrophoresis time-of-flight
mass spectroscopy study. PloS One 8, e73814.
doi:10.1371/journal.pone.0073814
Lai, M.C., Lombardo, M.V., and Baron-Cohen, S.
(2014). Autism. Lancet 383, 896–910.
Laszlo, A., Horvath, E., Eck, E., and Fekete, M. (1994).
Serum serotonin, lactate and pyruvate levels in
110
110 section II: E merging Clinical Applications and Future Potential
infantile autistic children. Clinica Chimica Acta;
International J Clin Chem 229, 205–207
Legido, A., Jethva, R., and Goldenthal, M.J. (2013).
Mitochondrial dysfunction in autism. Semin
Pediatr Neurol 20, 163–175.
Levy, S.E., Giarelli, E., Lee, L.C., Schieve, L.A., Kirby, R.S.,
Cunniff, C., Nicholas, J., Reaven, J., and Rice, C.E.
(2010). Autism spectrum disorder and co-occurring
developmental, psychiatric, and medical condi-
tions among children in multiple populations of the
United States. J Dev Behav Pediatr 31, 267–275.
Li, Z., Okamoto, K., Hayashi, Y., and Sheng, M. (2004).
The importance of dendritic mitochondria in the
morphogenesis and plasticity of spines and syn-
apses. Cell 119, 873–887.
Llaneza, D.C., DeLuke, S.V., Batista, M., Crawley,
J.N., Christodulu, K.V., and Frye, C.A. (2010).
Communication, interventions, and scientific
advances in autism: a commentary. Physiol Behav
100, 268–276.
Lord, C., Risi, S., DiLavore, P.S., Shulman, C., Thurm,
A., and Pickles, A. (2006). Autism from 2 to 9 years
of age. Arch Gen Psychiatry 63, 694–701.
Maalouf, M., Rho, J.M., and Mattson, M.P. (2009). The
neuroprotective properties of calorie restriction,
the ketogenic diet, and ketone bodies. Brain Res
Rev 59, 293–315.
Macfabe, D. F. (2012). Short-chain fatty acid fermenta-
tion products of the gut microbiome: implications
in autism spectrum disorders. Microb Ecol Health
Dis 23. doi:10.3402/mehd.v23i0.19260
Maestrini, E., Pagnamenta, A.T., Lamb, J.A., Bacchelli,
E., Sykes, N.H., Sousa, I., Toma, C., Barnby, G.,
Butler, H., Winchester, L., et  al. (2010). High-
density SNP association study and copy number
variation analysis of the AUTS1 and AUTS5 loci
implicate the IMMP2L-DOCK4 gene region in
autism susceptibility. Mol Psychiatry 15, 954–968.
doi:10.1038/mp.2009.34
Mantis, J.G., Fritz, C.L., Marsh, J., Heinrichs, S.C., and
Seyfried, T.N. (2009). Improvement in motor and
exploratory behavior in Rett syndrome mice with
restricted ketogenic and standard diets. Epilepsy
Behav 15, 133–141.
Masino, S.A., Kawamura, M., Wasser, C.D., Pomeroy,
L.T., and Ruskin, D.N. (2009). Adenosine, keto-
genic diet and epilepsy: the emerging therapeutic
relationship between metabolism and brain activ-
ity. Curr Neuropharmacol 7, 257–268.
Masino, S.A., Kawamura, M., Jr., Ruskin, D.N.,
Gawryluk, J., Chen, X., and Geiger, J.D. (2010).
Purines and the Anti-Epileptic Actions of
Ketogenic Diets. Open Neurosci J 4, 58–63.
Masino, S.A., and Rho, J.M. (2012). Mechanisms of
ketogenic diet action. In Jasper’s Basic Mechanisms
of the Epilepsies, J.L. Noebels, M. Avoli, M.A.
Rogawski, R.W. Olsen, and A.V. Delgado-
Escueta, eds. (Bethesda, MD: National Center for
Biotechnology Information), pp. 1001–1022.
Mattson, M.P., Gleichmann, M., and Cheng, A. (2008).
Mitochondria in neuroplasticity and neurological
disorders. Neuron 60, 748–766.
Mayer, E.A., Tillisch, K., and Gupta, A. (2015). Gut/
brain axis and the microbiota. J Clin Inves 125,
926–938.
McFarlane, H.G., Kusek, G.K., Yang, M., Phoenix, J.L.,
Bolivar, V.J., and Crawley, J.N. (2008). Autism-
like behavioral phenotypes in BTBR T+tf/J mice.
Genes, Brain, Behav 7, 152–163.
Meyza, K.Z., Defensor, E.B., Jensen, A.L., Corley, M.J.,
Pearson, B.L., Pobbe, R.L., Bolivar, V.J., Blanchard,
D.C., and Blanchard, R.J. (2013). The BTBR T+
tf/J mouse model for autism spectrum disorders-
in search of biomarkers. Behavioural Brain Res
251, 25–34.
Moreno, H., Borjas, L., Arrieta, A., Saez, L., Prassad,
A., Estevez, J., and Bonilla, E. (1992). [Clinical het-
erogeneity of the autistic syndrome: a study of 60
families]. Invest Clin 33, 13–31.
Moy, S.S., Nadler, J.J., Young, N.B., Perez, A., Holloway,
L.P., Barbaro, R.P., Barbaro, J.R., Wilson, L.M.,
Threadgill, D.W., Lauder, J.M., et al. (2007). Mouse
behavioral tasks relevant to autism: phenotypes of
10 inbred strains. Behav Brain Res 176, 4–20.
Murphy, M.P. (2009). How mitochondria produce
reactive oxygen species. Biochem J 417, 1–13.
Napoli, E., Duenas, N., and Giulivi, C. (2014). Potential
therapeutic use of the ketogenic diet in autism
spectrum disorders. Front Pediatr 2, 69.
Naviaux, J.C., Schuchbauer, M.A., Li, K., Wang, L.,
Risbrough, V.B., Powell, S.B., and Naviaux, R.K.
(2014). Reversal of autism-like behaviors and
metabolism in adult mice with single-dose anti-
purinergic therapy. Transl Psychiatry 4, e400.
doi:10.1038/tp.2014.33.
Naviaux, J.C., Wang, L., Li, K., Bright, A.T., Alaynick,
W.A., Williams, K.R., Powell, S.B., and Naviaux,
R.K. et  al. (2015). Antipurinergic therapy cor-
rects the autism-like features in the Fragile X
(Fmr1 knockout) mouse model. Mol Autism 6, 1.
doi:10.1186/2040-2392-6-1
Naviaux, R.K., Zolkipli, Z., Wang, L., Nakayama,
T., Naviaux, J.C., Le, T.P., Schuchbauer, M.A.,
Rogac, M., Tang, Q., Dugan, L.L., et  al. (2013).
Antipurinergic therapy corrects the autism-like
features in the poly(IC) mouse model. PLoS One
8, e57380. doi:10.1371/journal.pone.0057380
Neal, E.G., Chaffe, H., Schwartz, R.H., Lawson, M.S.,
Edwards, N., Fitzsimmons, G., Whitney, A., and
Cross, J.H. (2008). The ketogenic diet for the treat-
ment of childhood epilepsy:  a randomised con-
trolled trial. Lancet Neurol 7, 500–506.
Nylen, K., Velazquez, J.L., Likhodii, S.S., Cortez, M.A.,
Shen, L., Leshchenko, Y., Adeli, K., Gibson, K.M.,
Burnham, W.M., and Snead, O.C., 3rd (2008). A
ketogenic diet rescues the murine succinic semi-
aldehyde dehydrogenase deficient phenotype. Exp
Neurol 210, 449–457.
111
Chapter 14: Metabolic Therapy for Autism Spectrum Disorder
Oliveira, G., Diogo, L., Grazina, M., Garcia, P., Ataide,
A., Marques, C., Miguel, T., Borges, L., Vicente,
A.M., and Oliveira, C.R. (2005). Mitochondrial
dysfunction in autism spectrum disorders:  a
population-based study. Dev Med Child Neurol
47, 185–189.
Orinstein, A.J., Helt, M., Troyb, E., Tyson, K.E.,
Barton, M.L., Eigsti, I.M., Naigles, L., and Fein,
D.A. (2014). Intervention for optimal outcome in
children and adolescents with a history of autism.
J Dev Behav Pediatr 35, 247–256.
Palmieri, L., Papaleo, V., Porcelli, V., Scarcia, P., Gaita,
L., Sacco, R., Hager, J., Rousseau, F., Curatolo, P.,
Manzi, B., et  al. (2010). Altered calcium homeo-
stasis in autism-spectrum disorders:  evidence
from biochemical and genetic studies of the mito-
chondrial aspartate/glutamate carrier AGC1. Mol
Psychiatry 15, 38–52.
Park, M.J., Aja, S., Li, Q., Degano, A.L., Penati, J.,
Zhuo, J., Roe, C.R., and Ronnett, G.V. (2014).
Anaplerotic triheptanoin diet enhances mitochon-
drial substrate use to remodel the metabolome
and improve lifespan, motor function, and socia-
bility in MeCP2-null mice. PLoS One 9, e109527.
doi:10.1371/journal.pone.0109527
Patterson, P.H. (2011). Maternal infection and
immune involvement in autism. Trends Mol Med
17, 389–394.
Pearl, P.L., Gibson, K.M., Acosta, M.T., Vezina, L.G.,
Theodore, W.H., Rogawski, M.A., Novotny, E.J.,
Gropman, A., Conry, J.A., Berry, G.T., et al. (2003).
Clinical spectrum of succinic semialdehyde dehy-
drogenase deficiency. Neurology 60, 1413–1417.
Percy, A.K. (2011). Rett syndrome:  exploring the
autism link. Archiv Neurol 68, 985–989.
Poling, J. S., Frye, R. E., Shoffner, J., and Zimmerman,
A. W. (2006). Developmental regression and mito-
chondrial dysfunction in a child with autism. J
Child Neurol 21, 170–172
Rahman, S. (2012). Mitochondrial disease and epi-
lepsy. Dev Med Child Neurol 54, 397–406.
Raichle, M.E., and Gusnard, D.A. (2002). Appraising
the brain’s energy budget. Proc Natl Acad Sci USA
99, 10237–10239.
Ramoz, N., Reichert, J.G., Smith, C.J., Silverman, J.M.,
Bespalova, I.N., Davis, K.L., and Buxbaum, J.D.
(2004). Linkage and association of the mitochon-
drial aspartate/glutamate carrier SLC25A12 gene
with autism. Am J Psychiatry 161, 662–669.
Randolph-Gips, M., and Srinivasan, P. (2012).
Modeling autism:  a systems biology approach. J
Clin Bioinform 2, 17.
Rizzuto, R., De Stefani, D., Raffaello, A., and
Mammucari, C. (2012). Mitochondria as sensors
and regulators of calcium signalling. Nat Rev Mol
Cell Biol 13, 566–578.
Rose, S., Melnyk, S., Trusty, T.A., Pavliv, O., Seidel, L.,
Li, J., Nick, T., and James, S.J. (2012). Intracellular
and extracellular redox status and free radical
111
generation in primary immune cells from chil-
dren with autism. Autism Res Treat 2012, 986519.
doi:10.1155/2012/986519
Rose, S., Frye, R.E., Slattery, J., Wynne, R., Tippett,
M., Pavliv, O., Melnyk, S., and James, S.J. (2014).
Oxidative stress induces mitochondrial dysfunc-
tion in a subset of autism lymphoblastoid cell lines
in a well-matched case control cohort. PLoS One
9, e85436. doi:10.1371/journal.pone.0085436
Rossignol, D.A., and Frye, R.E. (2012). Mitochondrial
dysfunction in autism spectrum disorders:  a sys-
tematic review and meta-analysis. Mol Psychiatry
17, 290–314.
Roullet, F.I., Lai, J.K., and Foster, J.A. (2013). In utero
exposure to valproic acid and autism—a current
review of clinical and animal studies. Neurotoxicol
Teratol 36, 47–56.
Ruskin, D.N., Svedova, J., Cote, J.L., Sandau, U., Rho,
J.M., Kawamura, M., Jr., Boison, D., and Masino,
S.A. (2013). Ketogenic diet improves core symp-
toms of autism in BTBR mice. PLoS One 8,
e65021.
Sakurai, T., Ramoz, N., Barreto, M., Gazdoiu, M.,
Takahashi, N., Gertner, M., Dorr, N., Gama
Sosa, M.A., De Gasperi, R., Perez, G., et  al.
(2010). Slc25a12 disruption alters myelination
and neurofilaments:  a model for a hypomyelin-
ation syndrome and childhood neurodevelop-
mental disorders. Biol Psychiatry 67, 887–894.
doi:10.1016/j.biopsych.2009.08.042
Sandin, S., Lichtenstein, P., Kuja-Halkola, R.,
Larsson, H., Hultman, C.M., and Reichenberg,
A. (2014). The familial risk of autism. JAMA 311,
1770–1777.
Segurado, R., Conroy, J., Meally, E., Fitzgerald, M.,
Gill, M., and Gallagher, L. (2005). Confirmation
of association between autism and the mitochon-
drial aspartate/glutamate carrier SLC25A12 gene
on chromosome 2q31. Am J Psychiatry 162, 2182–
2184. doi:10.1176/appi.ajp.162.11.2182
Shpyleva, S., Ivanovsky, S., de Conti, A., Melnyk,
S., Tryndyak, V., Beland, F.A., James, S.J., and
Pogribny, I.P. (2014). Cerebellar oxidative DNA
damage and altered DNA methylation in the
BTBR T+tf/J mouse model of autism and similari-
ties with human post mortem cerebellum. PLoS
One 9, e113712.
Silverman, J.M., Buxbaum, J.D., Ramoz, N.,
Schmeidler, J., Reichenberg, A., Hollander, E.,
Angelo, G., Smith, C.J., and Kryzak, L.A. (2008).
Autism-related routines and rituals associated
with a mitochondrial aspartate/glutamate carrier
SLC25A12 polymorphism. Am J Med Gen:  Part
B, Neuropsychiatric Genetics 147, 408–410.
doi:10.1002/ajmg.b.30614
Smith, J.D., Rho, J.M., Masino, S.A., and Mychasiuk,
R. (2014). Inchworming: a novel motor stereotypy
in the BTBR T+ Itpr3tf/J mouse model of autism. J
Vis Exp. 2014 Jul 5(89). doi: 10.3791/50791.
112
112 section II: E merging Clinical Applications and Future Potential
Spilioti, M., Evangeliou, A.E., Tramma, D.,
Theodoridou, Z., Metaxas, S., Michailidi, E., Bonti,
E., Frysira, H., Haidopoulou, A., Asprangathou, D.,
et al. (2013). Evidence for treatable inborn errors
of metabolism in a cohort of 187 Greek patients
with autism spectrum disorder (ASD). Front Hum
Neurosci 7, 858.
Stafstrom, C.E., and Rho, J.M. (2012). The ketogenic
diet as a treatment paradigm for diverse neuro-
logical disorders. Front Pharmacol 3, 59.
Stromland, K., Nordin, V., Miller, M., Akerstrom, B.,
and Gillberg, C. (1994). Autism in thalidomide
embryopathy: a population study. Dev Med Child
Neurol 36, 351–356.
Su, H., Fan, W., Coskun, P.E., Vesa, J., Gold, J.A.,
Jiang, Y.H., Potluri, P., Procaccio, V., Acab, A.,
Weiss, J.H., et  al. (2011). Mitochondrial dys-
function in CA1 hippocampal neurons of the
UBE3A deficient mouse model for Angelman syn-
drome. Neurosci Lett 487, 129–133. doi:10.1016/
j.neulet.2009.06.079
Suen, D.F., Norris, K.L., and Youle, R.J. (2008).
Mitochondrial dynamics and apoptosis. Genes
Dev 22, 1577–1590.
Tang, G., Gutierrez Rios, P., Kuo, S.H., Akman,
H.O., Rosoklija, G., Tanji, K., Dwork, A., Schon,
E.A., Dimauro, S., Goldman, J., et  al. (2013).
Mitochondrial abnormalities in temporal lobe
of autistic brain. Neurobiol Dis 54, 349–361.
doi:10.1016/j.nbd.2013.01.006
Tordjman, S., Somogyi, E., Coulon, N., Kermarrec, S.,
Cohen, D., Bronsard, G., Bonnot, O., Weismann-
Arcache, C., Botbol, M., Lauth, B., et  al. (2014).
Gene x environment interactions in autism spec-
trum disorders:  role of epigenetic mechanisms.
Front Psychiatry 5, 53.
Turunen, J.A., Rehnstrom, K., Kilpinen, H.,
Kuokkanen, M., Kempas, E., and Ylisaukko-Oja, T.
(2008). Mitochondrial aspartate/glutamate carrier
SLC25A12 gene is associated with autism. Autism
Res 1, 189–192. doi:10.1002/aur.25
Voskuyl, R.A., and Vreugdenhil, M. (2001). Effects
of essential fatty acids on voltage-regulated ionic
channels and seizure thresholds in animals.
In Fatty Acids:  Physiological and Behavioral
Functions, Mostofsky D., Yehuda S., Salem N.Jr .
eds. (Totowa, NJ: Humana Press), pp. 63–78.
Weinberg, S.E., Sena, L.A., and Chandel, N.S. (2015).
Mitochondria in the regulation of innate and
adaptive immunity. Immunity 42, 406–417.
Xavier, J.M., Rodrigues, C.M., and Sola, S. (2015).
Mitochondria: major regulators of neural develop-
ment. Neuroscientist 22(4), 346–358. doi: 10.1177/
1073858415585472.
Zhao, Y., Fung, C., Shin, D., Shin, B.C., Thamotharan,
S., Sankar, R., Ehninger, D., Silva, A., and Devaskar,
S.U. (2010). Neuronal glucose transporter isoform
3 deficient mice demonstrate features of autism
spectrum disorders. Mol Psychiatry 15, 286–299.
doi:10.1038/mp.2009.51
113

15
Glucose and Ketone Metabolism in the Aging Brain
Implications for Therapeutic Strategies to Delay
the Progression of Alzheimer’s Disease

STEPHEN C. CUNNANE, PHD, ALEXANDRE COURCHESNE- LOYER, MSC,

VALERIE ST- PIERRE, BSC, CAMILLE VANDENBERGHE, BSC,
ETIENNE CROTEAU, PHD, AND CHRISTIAN- ALEXANDRE CASTELLANO, PHD

INTRODUCTION brain’s predominant fuel, but, like other organs,

The impact of deteriorating cognitive function in
older age has become a major sociomedical pre-
occupation worldwide. Alzheimer’s disease (AD),
the most common form of impaired cognitive
function in older age, is debilitating and dehu-
manizing, and is expensive to manage. To make
matters worse, there is currently no effective treat-
ment. Despite considerable effort and expense in
the development of antiamyloid medications,
there is no clear sign that this treatment strat-
egy will be effective in the near future (Jack et al.,
2009; Rafii and Aisen, 2009). Some major lifestyle
risk factors for AD are well known and include
type 2 diabetes, hypertension, and sedentarity.
Whether inadequate intake of certain nutrients
such as omega-3 fatty acids and certain B vitamins
also contributes to AD is receiving considerable
attention but remains unclear and controversial
(Douaud et al., 2013; Jerneren et al., 2015; Quinn
et al., 2010). As a general rule, “prevention” strate-
gies based on reducing the impact of lifestyle risk
factors can be implemented and have been shown
in clinical trials to be beneficial in reducing the
progression of not only prodromal AD but also
other lifestyle-related chronic diseases (Brown
et al., 2010; Ngandu et al., 2015; Okonkwo et al.,
2014). We speculate that effective strategies to
delay the progression of AD may have at least
one feature in common—that of improving brain
energy metabolism in older people.
It is well understood that the brain requires
a disproportionately large amount of energy;
whereas the adult brain represents about 2% of
adult body weight, it consumes about 20%–23%
of whole body energy requirements. Glucose is the
the brain has a back-up fuel for occasions when
glucose supply is insufficient, for example, dur-
ing fasting, starvation, strenuous exercise, or mal-
nutrition. While other organs use free fatty acids
directly to replace insufficient availability of glu-
cose, ketones (also known as ketone bodies) are
the only significant alternative fuel to glucose for
the brain. The two ketones that replace glucose
for the brain are beta-hydroxybutyrate and aceto-
acetate. As the decarboxylation product of aceto-
acetate, acetone appears to mainly be excreted on
breath but can also be metabolized through other
pathways.
For over 30 years now it has become increas-
ingly well established that brain glucose uptake
and metabolism are defective in AD, a problem
that is particularly evident in the parietal and tem-
poral cortex (reviewed in Cunnane et  al., 2015;
Cunnane et al., 2011). These results have generally
been interpreted to mean that low brain glucose
uptake in AD is a consequence of neuronal fail-
ure and death. While fewer functioning neurons
would indeed diminish the need for glucose, this
perspective does not account for multiple exam-
ples of conditions in which regional brain glucose
hypometabolism is present presymptomatically in
individuals at elevated risk of AD, that is, before the
clinical (cognitive) onset of the disease. An emerg-
ing literature is discussed here that provides sev-
eral clear examples in which individuals at genetic
or lifestyle risk of AD have impaired brain glucose
presymptomatically. Given that brain function is
acutely dependent on a constant supply of glucose
and oxygen, it is crucial to know whether latent
presymptomatic deterioration in brain glucose
114
114 section II: E merging Clinical Applications and Future Potential
uptake and/or utilization could be contributing to
the development of AD, because this could have
an impact on prevention and treatment strategies.
We have developed a positron emission
tomography (PET) research program using the
ketone tracer, 11C-acetoacetate (AcAc), to bet-
ter understand the relation between brain fuel
uptake and brain function in people with or at
risk of AD. In each individual studied, we use
this dual tracer PET protocol to compare the
brain uptake of 11C-AcAc with that of the glucose
tracer, 18F-fluorodeoxyglucose (FDG). We quan-
tify the magnitude of glucose and ketone uptake
regionally throughout the brain. Clinical studies
are underway to test the potential therapeutic
utility of ketogenic supplements based primarily
on medium chain triglycerides (MCT), with the
dual tracer PET imaging done before and after
the interventions. Here, we review the rationale
for these studies with the objective of highlighting
the main points supporting a strategy that impli-
cates maintenance of brain fuel supply to delay the
onset and/or progression of AD.
A R T E R I OV E N O U S
DIFFERENCE STUDIES
ACRO S S THE BRAIN: THE
F O U N D AT I O N
The PET-FDG protocol was first used to image
brain glucose metabolism in the late 1970s with
reports on AD first being published in the early
1980s (Benson et al., 1983). Since then, PET-FDG
has been a cornerstone of human and animal
studies on brain energy metabolism in aging and
AD (reviewed in Cunnane et  al., 2011). Indeed,
without PET-FDG, it is doubtful that brain glu-
cose hypometabolism in AD would have become
widely studied, because the only option besides
PET, the arteriovenous difference (AVD) method,
is highly invasive and used increasingly less in
research. Nevertheless, it was the AVD method
that first showed that brain oxygen and glucose
uptake were unaffected by healthy aging (Dastur,
1985). This method produced several ground-
breaking reports comparing brain ketone and
glucose uptake that laid the foundation for our
current understanding that ketones are an essen-
tial physiological fuel working daily in tandem
with glucose to assure brain energy requirements
are being met.
The first of these pioneering AVD reports
was by Owen et  al. (1967), who showed that
ketones could supply about two-thirds of the
brain’s fuel requirement when glucose avail-
ability was severely limited during medically
supervised starvation. This experiment was
done in three obese individuals who underwent
total dietary deprivation for 40  days. At that
time, it was suspected that some other fuel was
helping to meet the brain energy requirements
during extreme hypoglycemia because brain
oxygen consumption was not decreased as much
as glucose uptake, but the replacement fuel was
still unknown. The report by Owen et al. (1967)
was the first to demonstrate that ketones were
the unknown replacement fuel for glucose for
the brain, that is, that ketones played a physi-
ologically important role in the adult brain
rather than simply being a pathological marker
of severe insulin deficiency in decompensated
type 1 diabetes. This paper also proposed that
brain ketone uptake was proportional to blood
ketone concentration. Owen et  al. (1967) did
not describe brain ketone uptake under less
extreme conditions, but shortly after their paper
was published, Gottstein et al. (1971) used AVD
to show that the brain also consumed ketones
under the more physiological conditions of a
12- to 16-hour overnight fast. Hence, ketones
contribute to brain energy metabolism on a
daily basis, not just when brain glucose availa-
bility is severely limited during chronic hypo-
glycemia. Drenick et  al. (1972) were the first
to note that the symptoms of severe glucopenia
(as low as 0.5 mM plasma glucose) induced by
controlled insulin infusion could be avoided
by up to 60  days of medically supervised star-
vation. Their study showed that ketones could
account for ≥85% of brain fuel requirements
in obese adults. Drenick et al. (1972) also used
AVD to show that skeletal muscle of the forearm
was not using ketones at all during prolonged
starvation. These starvation studies were clearly
extreme, but they unequivocally demonstrated
the physiological role of ketones particularly if
not uniquely for the brain.
The first AVD report on brain energy metabo-
lism in AD was by Lying-Tunell et al. (1981), who
showed that brain glucose uptake was impaired
by 26% but that brain ketone uptake was still nor-
mal (Table 15.1). This report confirmed earlier
reports using AVD that had demonstrated lower
brain glucose uptake in AD and provided the
first suggestion that the disease did not necessar-
ily adversely affect the brain ketone utilization to
the same extent as glucose, even when the cogni-
tive deficit became severe. Hoyer et al. (1988) also
used the AVD method and demonstrated 45%
lower brain glucose utilization in suspected (i.e.,
early stage) AD. Unlike in more advanced AD,
115
Chapter 15: Glucose and Ketone Metabolism in the Aging Brain
TABLE 15.1 LOW BRAIN GLUCOSE UPTAKE BUT NORMAL BRAIN BETA-
HYDROXYBUT YRATE UPTAKE IN MILD ALZHEIMER’S DISEASE COMPARED
WITH AGE- MATCHED HEALTHY CONTROLS. MEASUREMENTS WERE MADE BY ARTE-
RIOVENOUS DIFFERENCE
Fuel Controls Mild Alzheimer’s
Glucose 24.8 18.7
(µmol/100 g/min) 24.9±7.2 11.6±4.0
Ketones 0.6 0.45
(µmol/100 g/min) 0.14±0.08 0.11±0.06
ns—not significant between Controls and mild Alzheimer’s
they showed that neither blood flow to the brain
nor oxygen uptake was adversely affected early in
AD. Hoyer et al. (1988) surmised from their results
that the problem with brain glucose utilization in
AD probably involved impaired glycolysis, par-
ticularly the conversion of glucose to pyruvate via
pyruvate dehydrogenase (Sorbi et al., 1983). Hoyer
et al. (1992; 1988) were the first to conceptualize
the significance of lower brain glucose utilization
as an early and possibly presymptomatic problem
in AD. However, they did not report brain ketone
uptake.
The second pioneering AVD paper that
addressed brain energy metabolism in AD was
by Ogawa et al. (1996), who confirmed the report
by Lying-Tunell et  al. (1981) showing that AD
adversely affected brain glucose but not brain
ketone uptake (Table 15.1). Unlike Lying-Tunell
et  al. (1981) but similar to Hoyer et  al. (1988),
Ogawa et  al. (1996) found that brain oxygen
uptake was less adversely affected in AD than
brain glucose uptake. These AVD studies raised
the possibility that brain glucose hypometabo-
lism was a relatively early problem in AD, and that
this problem was apparently not seen with brain
ketone utilization.
While AVD is quite invasive, it is also a
robust method that permits quantification of
global brain energy metabolism and is still in use
(Dalsgaard et al., 2004; Ide et al., 2000). The PET-
FDG data were initially expressed as the “cerebral
metabolic rate” (CMR) of glucose. These units
matched those used with AVD, which provided
a reference method against which early PET-
FDG results could be calibrated (Sokoloff et al.,
1977). Both methods converge on the point that
the AD brain is glucose-deprived and at risk of
starvation or exhaustion (Cunnane et  al., 2011;
Hoyer et al., 1988; Mamelak, 2012). However, the
115
p value (versus Control) Reference
<0.01 Lying-Tunell et al., 1981
<0.01 Ogawa et al., 1996
ns Lying-Tunell et al., 1981
ns Ogawa et al., 1996
CMR of glucose (CMRg) is measured less and
less because, ideally, it requires blood sampling
from a catheter inserted into the brachial artery
as well as exhaustive data processing. As a result
of declining quantification of CMRg, the actual
deficit in brain glucose uptake in people at risk
of AD is now only infrequently measured; rather,
PET-FDG data are increasingly expressed using
a statistical parametric map. Statistical paramet-
ric mapping clearly identifies the regions of the
brain with FDG uptake below a certain statistical
threshold, but the actual magnitude of that dif-
ference compared with the controls is generally
unknown (Baker et  al., 2011; Kalpouzos et  al.,
2009). Hence, although the regional nature of
the deficit in brain FDG uptake in AD can be
clearly observed in some detail with PET (which
is impossible with AVD), there are fewer and
fewer quantitative PET-FDG reports, so we have
actually learned very little about the magnitude
of the regional deficit in brain glucose uptake in
AD or how it changes as AD progresses since
the AVD studies and insights of Hoyer et  al.
(1988). Furthermore, with the focus on glucose
as the representative brain fuel using PET-FDG,
until recently, the possibility that brain ketone
uptake could be less affected in AD and hence
possibly part of a treatment strategy has largely
been overlooked. However, knowing the magni-
tude of the deficit in brain glucose uptake in AD
and the extent to which it continues to worsen
without treatment provides critical information
about existing treatments that may fortuitously
improve brain glucose uptake, namely, acetyl-
cholinesterase inhibitors (Nordberg, 2006)  or
NMDA receptor antagonists (Wang et  al.,
2013), or new treatments aiming to bypass or
correct this problem early enough to delay the
progression of AD.
116

116 section II: E merging Clinical Applications and Future Potential

P R E S Y M P T O M AT I C at elevated risk of the disease but in whom cogni-

BRAIN GLUCOSE
H Y P O M E TA B O L I S M :
M U LT I P L E E X A M P L E S
While AVD has largely become a technique of the
past, it helped seed our interest in applying PET-
FDG to the question of causality—is brain glucose
hypometabolism a consequence of AD or part of
the cause or, indeed, both? The stage of AD can
be difficult to determine precisely, which in turn
means that it can be difficult to determine whether
those on the threshold or at the early stages of
the disease have a specific deficit in brain glu-
cose uptake or a broader deficit that also includes
impaired oxygen uptake and lower blood flow to
the brain. Assessing the presence of a possible pre-
symptomatic neurometabolic deficit in a disease
of uncertain and gradual onset like AD therefore
requires a different strategy such as studying those
tive scores are still normal.
Carriers of the presenilin-1 mutation are at
very high, almost certain, risk of developing AD
(Blennow et al., 2006; Fox et al., 1997; Scholl et al.,
2011). Carriers of the apolipoprotein E4 allele and
persons with a family history of AD, especially on
the maternal side, are also at increased risk of AD
(Bu, 2009; Kennedy et  al., 1995; Mosconi et  al.,
2005; Reiman et al., 2004), although their risk is
still relatively low compared with presenilin-1
carriers. Type 2 diabetes ranks among the most
important of the nongenetic or lifestyle risk fac-
tors for AD. In presymptomatic individuals with
these aforementioned conditions in whom the
results on cognitive batteries are still normal,
PET-FDG clearly shows that regional brain glu-
cose hypometabolism is present before the onset
of cognitive deficit (Table 15.2). These examples

TABLE 15.2 PRESYMPTOMATIC BRAIN GLUCOSE HYPOMETAB OLISM IN PERSONS

AT RISK OF ALZHEIMER’S DISEASE (AD)

Group at risk of AD Mean age Brain region Magnitude of brain Reference

(y) affected glucose hypometabo-
lism (% difference from
control)

Young adult carriers of 30 Posterior cingulate –14 to –25 Schöll et al., 2011
Presenilin-1 Parietal cortex
Temporal cortex
Young adult carriers of 31 Parietal cortex –9 to –11 Reiman et al., 2004
Apolipoprotein-E4 Temporal cortex
Posterior cingulate
Prefrontal cortex
Maternal family history 43 Parietal cortex –12 to –21 Mosconi et al.,
of AD Temporal cortex 2006
Hippocampus
Entorhinal cortex
Posterior cingulate
Prediabetic older persons 74 Temporal cortex N/A Baker et al., 2011
Parietal cortex
Posterior cingulate
Precuneus
Prefrontal cortex
Insulin resistant young 25 Frontal cortex –9 to –14 Castellano et al.,
women with polycystic Middle temporal 2015
ovary syndrome cortex
Cognitively healthy older 72 Frontal cortex –10 to –18 Nugent et al., 2015
adults Temporal cortex
Anterior cingulate
Putamen
Thalamus

N/A—not available
117
Chapter 15: Glucose and Ketone Metabolism in the Aging Brain
are a good basis for concluding that presymptom-
atic brain glucose hypometabolism can be pres-
ent and could therefore be contributing to the
development of AD.
Aging is considered to be the most significant
risk factor for AD, so we have extensively stud-
ied brain energy metabolism in cognitively nor-
mal older persons. Using quantitative PET-FDG,
we have shown that in cognitively normal people
aged 72 years, the CMRg in the superior frontal
cortex is about 35±5 instead of the 40±7 mmol/
100 g/min seen in young adults, that is, the glu-
cose uptake deficit in this brain region is of the
order of 12%–13% (Nugent et  al., 2015; Nugent
et al., 2014a; Nugent et al., 2014b). The CMRg did
not differ significantly between young and older
adults in either white or gray matter as a whole
nor did it differ in the temporal or parietal cor-
tex where CMRg decreases in AD. Lower CMRg
that is present mostly if not exclusively in the
frontal cortex of cognitively normal older people
has been reported on several previous occasions
(Cunnane et  al., 2011; Kalpouzos et  al., 2009).
Thus, lower CMRg in the temporal and parietal
cortex in AD is a fundamentally different situa-
tion (but perhaps related to) lower frontal CMRg
during aging. Incidentally, in older people with
normal cognition, cortical thickness and regional
volumes are lower in many more brain regions
than CMRg (Nugent et al., 2015). Thus, although
lower frontal CMRg is clearly presymptomatic in
relation to aging-associated cognitive decline, it
is by no means the only or even the most signifi-
cant change occurring in the brain during aging.
We have also observed a deficit of about 14%
in CMRg in the superior and middle frontal cor-
tex in 24-year-old women with polycystic ovary
syndrome (Castellano et  al., 2015a). Polycystic
ovary syndrome is a multifactorial endocrine
disease involving infertility, hyperandrogenism,
and mild-to-moderate insulin resistance. It was
the mild insulin resistance that was of particu-
lar interest to us because of its association with
increased risk of AD (Baker et  al., 2011; Craft,
2009, 2012; Matsuzaki et  al., 2010; Ronnemaa
et al., 2008; Schrijvers et al., 2010). These women
with polycystic ovary syndrome were of nor-
mal weight and body-mass index, so this condi-
tion represents a possible model of the impact of
mild insulin resistance on brain glucose metabo-
lism that is independent of age and obesity. The
mild insulin resistance in women with polycystic
ovary syndrome was significantly inversely cor-
related to CMRg in several brain regions. These
women also had borderline low-normal scores on
117
a test of working memory, suggesting that insulin
resistance can adversely affect both brain glucose
metabolism and possibly cognitive performance
in young adults (Castellano et al., 2015a). Whether
women with polycystic ovary syndrome are pre-
disposed to a higher risk of cognitive decline as
they age remains to be seen but deserves further
attention.
These examples demonstrate that regional
brain glucose hypometabolism can be present
in those in whom cognition is still normal even
though they are at risk of AD due to old age, or
certain genetic or lifestyle factors associated with
AD. The presymptomatic glucose uptake defi-
cit is commonly but not exclusively in the fron-
tal cortex, and its magnitude is of the order of
12%–15% (Castellano et al., 2015b; Cunnane et al.,
2011; Nugent et al., 2015). These observations run
counter to the widespread perception that brain
glucose hypometabolism is uniquely a conse-
quence of synaptic dysfunction and neuronal fail-
ure or death. Indeed, they suggest that a vicious
cycle can develop in which latent presymptomatic
brain glucose hypometabolism develops and leads
to chronic brain energy deficit, deteriorating neu-
ronal function, further decline in demand for glu-
cose, and further cognitive decline (Figure 15.1;
Cunnane et  al., 2011). The presence of presymp-
tomatic brain glucose hypometabolism does not
necessarily mean that it is a cause of AD, nor that
neuronal function is necessarily normal, nor even
that glucose hypometabolism is the first abnormal-
ity detectable in those at risk of AD. Nevertheless,
knowing that brain glucose hypometabolism can
Latent brain glucose
hypometabolism:
↓ glycolysis
↓ brain glucose uptake
Ketones
Deteriorating brain
function:
↓ ATP generation
Alzheimer’s
↓ neurotransmitters
disease
Cognitive symptoms:
↓ memory
↓ executive function
↓ processing speed
FIGURE 15.1 Schematic representation of latent
brain glucose hypometabolism leading to a vicious cycle
of accelerating metabolic deterioration and neuronal
dysfunction that increases the risk of developing Alzh-
eimer’s disease (from Cunnane et al., 2015).
118
118 section II: E merging Clinical Applications and Future Potential
be present in people at risk of AD before the onset
of measurable clinical (cognitive) deficit has impli-
cations for potential therapeutic strategies (Blass,
2008; Cunnane et al., 2015; Cunnane et al., 2011;
Gibson et al., 2000; Henderson et al., 2009; Veech
et al., 2001).
B R A I N G L U C O S E U P TA K E
O R M E TA B O L I S M O R   B O T H ?
The AVD studies show the magnitude of the
lower glucose “disappearance” into the brain in
AD. However, they do not establish whether the
problem is with brain glucose uptake, that is, its
transport into the brain, or with glycolysis, that is,
glucose metabolism within the brain, or both. The
PET studies clearly show that there is a problem in
AD at the level of glucose (FDG) uptake, that is,
with glucose transport into the brain and its con-
version to glucose-6-phosphate by hexokinase.
However, PET-FDG provides no information
about whether, in addition to defective glucose
transport, the glycolytic steps are also altered. It
may not matter to the brain whether the prob-
lem with glucose hypometabolism is with brain
glucose uptake per se or with glucose metabo-
lism via glycolysis; either way, if uncorrected, the
brain’s main energy source is compromised and it
is at increased risk of chronic energy deficit and
eventual exhaustion. In fact, the energy supply
problem seems to be at the level of both brain
glucose uptake and metabolism. In vitro studies
show that several enzymes in glycolysis are mark-
edly impaired in AD, including phosphofructo-
kinase (Bowen et al., 1979; Iwangoff et al., 1980),
alpha-ketoglutarate dehydrogenase complex
(Gibson et  al., 2000), and pyruvate dehydrogen-
ase (Cunnane et al., 2011; Perry et al., 1980; Sorbi
et al., 1983).
K E TO N E S — T H E B R A I N ’ S
PHYSIOLOGICAL
A LT E R N AT I V E   F U E L
The now classic studies by Owen et al. (1967) and
Drenick et  al. (1972) demonstrated that ketones
are the main reserve fuel for the brain when glu-
cose supply is compromised by starvation. The
human liver can produce ketones at a rate of 100–
150 g/day (Flatt, 1972; Reichard et al., 1974), which
is more than sufficient to account for the brain’s
ketone utilization even during prolonged star-
vation. The energy cost to the liver of producing
ketones is mostly supplied by gluconeogenesis, the
rate of which parallels and may eventually actu-
ally limit ketone production (Flatt, 1972; Garber
et al., 1974).
In adults, ketogenesis is principally from long
chain fatty acids stored in adipose tissue, the release
of which is controlled by insulin. The branched
chain amino acids, isoleucine and leucine, are
also ketogenic (Mitchell et al., 1995). During fast-
ing, blood glucose and insulin decrease, which
releases the inhibition on lipolysis in adipose tis-
sue, thereby allowing plasma free fatty acids to
increase. This increase in plasma free fatty acids
helps meet the need for an alternative fuel to glu-
cose for most tissues with the notable exception of
the brain. The increased supply of fatty acids enter-
ing the liver leads to ketogenesis by condensation
of two acetyl-CoAs, which are present in excess
due to fatty acid beta-oxidation (Figure 15.2).
Implicitly, if insulin controls the ebb and flow
of plasma glucose, free fatty acids, and ketones,
insulin sensitivity becomes an important param-
eter in the process. Indeed, conditions involving
decreased insulin sensitivity, that is, insulin resis-
tance, also impair tissue glucose uptake and are
associated with a vicious cycle of hyperglycemia,
hyperinsulinemia, and an eventual trend toward
type 2 diabetes, which is a major risk factor for
AD (Baker et al., 2011; Craft, 2009). Postprandial
hyperinsulinemia inhibits fatty acid release from
adipose tissue and ketogenesis, which is appro-
priate as long as insulin sensitivity and tissue
glucose uptake are normal, that is, when and insu-
lin returns to normal 3–4 hours postprandially.
However, chronic sedentarity commonly leads
to chronic hyperinsulinemia and insulin resis-
tance, which not only compromise tissue glucose
uptake but also ketogenesis and ketone metabo-
lism (Bickerton et al., 2008; Fukao et al., 2004). In
effect, this puts the aging brain in double jeopardy
because now it is not only getting insufficient glu-
cose but is also getting less of the main alternative
fuel, ketones (Cunnane et al., 2015; Cunnane et al.,
2011; Mamelak, 2012).
K E TO N E K I N E T I C S
AND TRANSPORT
I N   A D U LT   H U M A N S
Elegant kinetic studies show that ketone utilization
essentially matches synthesis in normal weight
adults fasted overnight (Avogaro et  al., 1990;
Balasse and Fery, 1989; Hall et al., 1984). The rapid
utilization of ketones as they are produced during
short-term fasting generally keeps plasma ketones
≤0.3 mM (Table 15.3). Exercise for 30 minutes has
little or no effect on ketone synthesis, utilization or
clearance. After 3–5 days fasting, plasma ketones
rise about 10-fold due mainly to increased synthe-
sis and lower clearance. In obese adults, overnight
119

Chapter 15: Glucose and Ketone Metabolism in the Aging Brain 119

O O
+ 2 Acetyl-CoA
H3C SCoA H3C SCoA

Thiolase

O O
Acetoacetyl-CoA
H3C SCoA

HMG-CoA synthase

O OH O
HMG-CoA
O SCoA
CH3

HMG-CoA lyase

O O
Acetoacetate
O CH3

Nonenzymatic D-beta-hydroxybutyrate
decarboxylation dehydrogenase

O O OH

H3C CH3 O CH3

Acetone D-beta-hydroxybutyrate

FIGURE 15.2 Ketogenesis arising from the condensation of two acetyl-CoAs to form the ketones, acetoacetate, beta-
hydroxybutyrate, and acetone. HMG-CoA: 3-hydroxy-3-methylglutaryl-coenzyme A.

or 3–5 days fasting has little to no effect on these
kinetic parameters, but synthesis may exceed uti-
lization, thereby raising plasma ketones a bit more
than in normal weight adults. In type 1 diabetes,
12 hours of fasting increases plasma ketones about
10-fold more than in nondiabetic adults, but it is
not clear why, as utilization appears to keep pace
with synthesis, both of which are similar to values
seen in nondiabetic adults after 12 hours of fasting
(Table 15.3).
As far as is known, ketones can be trans-
ported into and catabolized by all tissues except
the liver. However, during extended fasting, free
fatty acids compete with ketones and become the
main fuel for some tissues such as skeletal mus-
cle, leaving most if not all of ketone production
available to complement the available glucose in
meeting the energy needs of the brain (Drenick
et  al., 1972; Owen and Reichard, 1971). Ketone
transport into tissues occurs via monocarboxylic
acid transporters, of which there are at least six
subtypes (Simpson et  al., 2007). Several of the
monocarboxylic acid transporters are expressed
in the brain. Monocarboxylic acid transporter
120

120 section II: E merging Clinical Applications and Future Potential

TABLE 15.3 KETONE (BETA- HYDROXYBUT YRATE + ACETOACETATE) KINETICS

IN HUMANS SUMMARIZED FROM A SERIES OF PUBLISHED STUDIES

Condition Fasting Plasma Utilization Synthesis Metabolic Urinary

period ketones (µmol/kg/ (µmol/kg/ clearance excretion
(mM) min) min) (mL/kg/min) (µmol/min)

Healthy adults 12–16 ha,f,g 0.1–0.3 3–5 2–5 18 —

3 dd 2.5 — 10 — 4
Healthy adults 16 hf,j 0.2–0.4 6 6 21 —
+ 30 min exercise 3–5 df,h,j 4-5 20 22 4–6 —
Obesity 12–16 ha,c 0.4–0.7 4 3–8 12 0.4
≥ 3 db,c,e,i 3-7 9 5–25 3 80–240
Type I diabetes 12 ha 2.5 6 6 — —

h—hours
d—days
A
(Hall et al., 1984)
B
(Reichard et al., 1974)
C
(Balasse, 1979)
D
(Garber et al., 1974)
E
(Owen et al., 1967)
F
(Fery and Balasse, 1983)
G
(Avogaro et al., 1990)
H
(Balasse et al., 1978)
I
(Balasse, 1986)
J
(Fery and Balasse, 1986)

expression responds rapidly to hyperketonemia
(Pan et  al., 2001), so, under normal conditions,
brain uptake of ketones is directly proportional
to their plasma concentration over about a 600-
fold range of beta-hydroxybutyrate (0.02–12
mM; Courchesne-Loyer et  al., 2013; Cunnane
et  al., 2015; Cunnane et  al., 2011). However,
as with brain glucose uptake, the transport of
ketones into the brain may not be the limiting
variable in their uptake. Rather, ketones are
“pushed” into the brain in proportion to their
plasma ketone concentration, which usually only
increases when plasma glucose decreases. This
contrasts with glucose, which is “pulled” into
the brain in proportion to its utilization. Hence,
the “push-pull” strategy assures that under con-
ditions in which glucose availability decreases,
ketone synthesis would normally be stimulated
and the brain’s energy supply would be main-
tained (Figure 15.3).

Blood Brain
Glucose:
Metabolic demand
Glucose Glucose

Ketones
Ketones Ketones:
Plasma availability

FIGURE 15.3 The contrasting “push-pull” mechanism of brain fuel supply. Glucose is pulled from the blood into the
brain as a function of the brain’s metabolic demand during neuronal activation. Under normal conditions (excluding
insulin resistance), ketones are pushed from the blood into the brain in direct proportion to their plasma concentration.
121

Chapter 15: Glucose and Ketone Metabolism in the Aging Brain 121

R E G I O NA L B R A I N K E TO N E 2011; Roy et  al., 2012)  studies. This dual tracer

U P TA K E I N   E A R LY
ALZHEIMER’S DISEASE
Positron emission tomography provides a means
of directly quantifying and comparing brain glu-
cose and ketone uptake. Using PET, Blomqvist
et al. (2002; 1995) reported the brain metabolism
of 11C-beta-hydroxybutyrate in humans and con-
firmed the earlier AVD studies showing that the
uptake was directly proportional to plasma beta-
hydroxybutyrate even at very low plasma ketone
concentrations (Cunnane et al., 2011; Lying-Tunell
et  al., 1981; Figure 15.4). We found that synthe-
sizing 11C-AcAc was easier than synthesizing 11C-
beta-hydroxybutyrate (Tremblay et al., 2007), and
since acetoacetate is the ketone that actually enters
the mitochondria and is catabolized to acetyl CoA,
we worked with it as our brain ketone PET tracer.
We developed a PET protocol in which 11C-AcAc is
injected intravenously first, followed by a wash-out
period, and then FDG is injected. We have applied
this protocol in human (Castellano et  al., 2015b;
Nugent et  al., 2014b) and animal (Pifferi et  al.,
technique allows for a quantitative comparison of
brain uptake of glucose and ketones with a delay of
no more than 2 hours between the two tracer infu-
sions, which avoids the potentially greater biologi-
cal variability between scans done on different days.
This dual tracer technique permitted us to com-
pare brain uptake of FDG and 11C-AcAc in early
AD (Castellano et al., 2015b). The aim was three-
fold:  (1)  to confirm the AVD reports of normal
ketone but low brain glucose uptake early in AD
(Ogawa et al., 1996), (2) to assess brain fuel metab-
olism in early AD rather than the more advanced
stages previously reported (Lying-Tunell et  al.,
1981), and (3)  to quantify the regional pattern
of brain uptake of both fuels under postprandial
conditions. We confirmed that global brain FDG
uptake (CMRg) was 14% lower in early AD versus
cognitively normal controls, and that this global
deficit was primarily confined to the parietal cortex,
posterior cingulate, and thalamus. However, nei-
ther 11C-AcAc uptake (CMRa) nor the acetoacetate
uptake constant (Ka) were significantly different

100
100 metabolism as ketones
(log µmol/100 g/min)

10 % of total brain
CMR ketone

1
10

0.1

0.01 1
0.01 0.1 1 10
Plasma ketones
(log mmol/L)
FIGURE 15.4 Direct, linear relation between plasma β-hydroxybutyrate (β-HB), brain ketone uptake (left-hand Y
axis), and percentage contribution to total brain energy requirement (right-hand Y axis) in adults. Two relationships
are shown, one for plasma β-HB versus the rate of brain β-HB uptake (solid line, R2 = 0.97; Y = 1.57X - 0.20; p < .0001),
and the other for plasma acetoacetate versus the rate of brain acetoacetate uptake (Castellano et  al., 2015b) (dotted
line, R2 = 0.83; Y = 3.46X - 0.03; p < .0001): Alzheimer’s disease (AD), ⚬ cognitively healthy age-matched controls.
Units are the same for both ketones—cerebral metabolic rate (CMR; μmol/100 g/min). β-HB data have been combined
from several sources: postprandial state (Blomqvist et al., 1995), (Blomqvist et al., 2002) β-HB infusion, AD
■
and healthy older controls(Lying-Tunell et al., 1981), 40-day fast (Owen et al., 1967), 60-day fast(Drenick et al.,
1972), and AD and healthy older controls (Ogawa et al., 1996; indicated by the arrow). All the β-HB data are from
arteriovenous difference studies except for one report, which used β-HB-PET (Blomqvist et al., 1995). The AcAc data
were obtained using 11C-AcAc PET (Castellano et al., 2015b). Each symbol represents a single individual except when
■
not available in the original publication: Drenick et al. (1972), for which represents the mean of n = 5 participants,
and Ogawa et al. (1996), for which and both represent the mean of n = 7. The relationship between plasma β-HB
and the percent of brain energy consumption supplied by β-HB in adults is as follows: at plasma β-HB values around
0.1 mM, ketone supply >5% of brain energy; at 1 mM β-HB, they supply about 10%–15%; at 5–7 mM β-HB, 50%–65%
and over 7–8 mM β-HB, >75% of brain energy consumption. For a given plasma AcAc concentration, AcAc is taken up
by the brain more rapidly than β-HB, which is why the dotted regression line lies above that of the solid line for β-HB.
122

122 section II: E merging Clinical Applications and Future Potential

Age-matched controls Mild Alzheimer

50 0.6

Brain 11C-AcAc uptake

Brain 18F-FDG uptake
*

(µmol/100 g/min)
(µmol/100 g/min)
40
0.4
30
20
0.2
10
0 0.0

0.08 0.05

Brain 11C-AcAc
0.04
Brain 18F-FDG
uptake constant

uptake constant
0.06
0.03

(min–1)
(min–1)

0.04
0.02
0.02 0.01
0.00 0.00

FIGURE  15.5 Whole-brain 18F-fluorodeoxyglucose (18F-FDG) uptake is 14% lower in mild Alzheimer’s disease
(white bars; n = 10) compared with cognitively normal age-matched controls (black bars; n = 30; * p = .003). In contrast
to FDG, whole-brain 11carbon-acetoacetate (11C-AcAc) uptake is not significantly different in mild Alzheimer’s disease
compared with cognitive normal, age-matched controls (from Castellano et al., 2015b).

in the brain as a whole or in any brain region in (1996). Unlike with FDG, the 11C-AcAc PET tracer
AD versus the age-matched controls (Figure 15.5). is chemically identical to acetoacetate produced by
The kinetics of 11C-AcAc metabolism (or 11C-beta- the body, so it is possible to interpret our results as
hydroxybutyrate metabolism; Blomqvist et  al., showing that both uptake and metabolism of 11C-
1995) suggest a one-compartment model in which AcAc were normal in early AD. Unlike for glucose,
brain utilization essentially matches brain uptake. which generates ATP by both oxidative phospho-
Plasma acetoacetate and CMRa were significantly rylation and aerobic glycolysis, production of ATP
positively correlated, the slope of which did not from ketones is exclusively via the citric acid cycle
differ between controls and early AD. This implies and oxidative phosphorylation. Since we observed
that brain ketone utilization in AD was still propor- that brain 11C-AcAc metabolism was normal in
tional to plasma concentration on the same slope early AD, our results with brain ketone PET there-
as in controls (Figure 15.6). These results confirm fore indirectly support the speculation by Hoyer
those of Lying-Tunell et al. (1981) and Ogawa et al. et  al. (1988) that oxidative phosphorylation may
still be normal early in AD. These observations
provide a rationale for the concept that if provided
1.0
Mild Alzheimer with more ketones, the aging brain would be able to
Age-matched controls use them which, in turn, might help it overcome its
Brain 11C-AcAc uptake

0.8
(µmol/100 g/min)

deficit in glucose uptake and metabolism, thereby

0.6 delaying brain energy exhaustion and decreasing
the risk of cognitive decline and AD.
0.4

0.2
0.0
0 50
Plasma AcAc (µM)
FIGURE  15.6 The direct, linear relationship between
plasma ketone (acetoacetate; AcAc) concentration and
brain ketone uptake is not altered in mild Alzheimer’s
disease (white triangles, dotted line; n  =  10) compared
with cognitively normal age-matched elderly (black
squares, solid line; n = 30) (from Castellano et al., 2015b).
K E TO G E N E S I S
FROM MEDIUM CHAIN
T R I G LY C E R I D E S :   D O S E -
100 150 200 250 R E S P O N S E R E L AT I O N S H I P
In adults on a normal mixed diet, fasting induces
ketosis because long chain fatty acids are beta-
oxidized by the liver to acetyl CoA. Long chain
fatty acids are of 14–22 carbons in length and are
essentially the only fatty acids found in adipose tis-
sue of adults. However, breast-fed infants achieve
mild ketosis in part due to the presence in breast
123

Chapter 15: Glucose and Ketone Metabolism in the Aging Brain 123

1200

Maximal plasma BHB (µM)

1000

800

600

400

200

0
0 10 20 30 40 50 60 70 80
MCT dose (g)
FIGURE  15.7 Direct relationship dose of medium chain triglycerides (MCT) and maximal observed plasma β-
hydroxybutyrate [BHB] (r = 0.95; p < .0001) in young adults unless otherwise specified. MCT were given as a single or
multiple doses*: ( Alzheimer; 10 g MCT; n = 77, ⚬ Alzheimer; 20 g MCT*; n = 77; Henderson et al., 2009), ( mild
cognitive impairment, AD; 40 g MCT; n = 10; Reger et al., 2004), (× type 1 diabetes; 40 g MCT; n = 11; Page et al., 2009),
( 10 g MCT emulsion; n = 10; 20 g MCT emulsion, n = 10, 30 g MCT emulsion; n = 10, 10 g neat MCT; n = 10;
20 g neat MCT; n = 10; 30 g neat MCT; n = 10; Courchesne-Loyer et al., unpublished), ( young adults; 71 g MCT
■
drink; n = 22; older adults 71 g MCT drink; n = 14; Freemantle et al., 2009), (✳ 48 g; n = 7; MCT; Seaton et al., 1986),
and (+ 70 g MCT; n = 14; Pi-Sunyer et al., 1969).

milk of medium chain fatty acids, that is, fatty acids
of 6–12 carbons (Cunnane and Crawford, 2014).
Medium chain fatty acids are mostly absorbed
through the portal vein, hence gaining direct and
more rapid access to the liver than long chain fatty
acids, which are absorbed into the peripheral cir-
culation via the lymph. Medium chain fatty acids
are also beta-oxidized without needing to be acti-
vated by carnitine. The net result is rapid beta-
oxidation and ketogenesis. With rare exceptions,
there is normally no further opportunity to con-
sume medium chain fatty acids from the diet once
breast-feeding is terminated. The exceptions are
coconut oil and palm kernel oil, in which medium
chain fatty acids make up about 15% and 10% of
the fatty acid composition, respectively. The frac-
tion of these oils that contains these medium
chain fatty acids can be concentrated, resulting in
a generic product containing mostly fatty acids of
eight (octanoic or caprylic acid) and 10 carbons
(decanoic or capric acid) called medium chain tri-
glycerides (MCTs). The ratio of these two principal
fatty acid components and their proportion of the
total can vary widely from one MCT product to
another. The MCTs are well known to be ketogenic
(Bach and Babayan, 1982; Freund and Weinsier,
1966). Notwithstanding the generic nature of
MCTs and different study designs to assess their
metabolism, there is a remarkably good correla-
tion in the literature between the maximal plasma
ketone levels achieved on oral doses of MCTs from
10 to 70 g (Figure 15.7). These dose-response
studies show similar results whether a single or
multiple MCT doses are given, probably because
of rapid plasma ketone clearance.
Healthy older people have the same level
of hyperketonemia and 13C-ketone oxidation to
13
C-CO2 after a standard high-fat ketogenic break-
fast containing MCTs as middle aged or young
adults (76  years old versus 50 or 23  years old,
respectively; Freemantle et al., 2009; Figure 15.8).
Indeed, hyperketonemia after 18 hours of fasting
may actually be somewhat higher in the seventh
to eighth decade of life compared with younger
adults (London et al., 1986). Hence, the capacity to
produce and utilize ketones does not appear to be
reduced and may actually be increasing somewhat
during healthy aging. These studies did not distin-
guish between ketone utilization by the brain ver-
sus other organs, but our brain ketone PET results
confirm that the brain has similar ketone uptake
and utilization in healthy older compared with
younger persons (Nugent et al., 2014b).
CLINICAL STUDIES
O F   H Y P E R K E TO N E M I A
IN CONDITIONS
OF COGNITIVE DEFICIT
Several products based on MCTs are now available
on the market or by prescription in the United States
and in Europe. The arrival of these products was
greatly stimulated by reports that MCTs had benefi-
cial effects on cognitive outcomes in mild-moderate
AD after a single dose (Reger et  al., 2004)  or
124
124 section II: E merging Clinical Applications and Future Potential
0.4
Plasma β-HB change vs. young (mM)
0.2
0.0
–0.2
–0.4
0 1
FIGURE  15.8 Plasma β -hydroxybutyrate (β-HB) response after a high-fat meal in a middle-aged group (gray
triangles; 50 years old; n = 18) amd older group (white circles; 76 years old; n = 14) compared with young adults (dotted
line; 23 years old; n = 22) (from Freemantle et al., 2009).
with regular consumption over several months
(Henderson et  al., 2009). Improved cognitive out-
comes after MCTs are also seen in type 1 diabetics
during controlled acute hypoglycemia caused by
insulin infusion (Page et al., 2009). The very high-
fat ketogenic diet has been reported to have a simi-
lar beneficial effect on cognitive and cardiovascular
outcomes in mild cognitive impairment, the pro-
dromal state to AD (Krikorian et al., 2012). These
reports complement the studies showing that auto-
nomic and neurological symptoms of acute severe
experimental hypoglycemia and starvation can be
avoided by ketone infusion (Table 15.4).
Acetoacetate and beta-hydroxybutyrate can
also be directly administered orally or by infusion.
This requires that they be provided as salts or as
esters (Clarke et al., 2012; Hasselbalch et al., 1996;
Plecko et  al., 2002), with the esters being more
practical for long-term use. The safety (Clarke
et al., 2012) of a beta-hydroxybutyrate-monoester
and its anecdotal utility in improving some aspects
of cognitive function in an advanced case of early-
onset Alzheimer’s disease have recently been
reported (Newport et  al., 2015). The results of
these clinical studies are still preliminary, but they
support the hypothesis that brain glucose deficit
contributes to impaired cognition associated with
aging and that this deficit can at least in part be
reversed or bypassed by either a ketogenic supple-
ment containing MCTs, a ketone ester or a very
high-fat ketogenic diet.
ANAPLERO SIS
During prolonged starvation, ketones can sup-
ply upward of 70% of the energy requirement of
2 3 4 5 6
Time (h)
the adult brain (Drenick et al., 1972; Owen et al.,
1967) but glucose is still essential in the process. In
addition to being the main engine to generate ATP,
the citric acid cycle also spins off molecules needed
for brain function including the neurotransmitters
GABA and acetylcholine, a process known as cat-
aplerosis (Owen et  al., 2002). Cataplerosis would
rapidly deplete the citric acid cycle of its inter-
mediates except that, via oxaloacetate, glucose
supplies carbon to replace those intermediates, a
process known as anaplerosis. Hence, glucose is
not only a major fuel but also controls the balance
between anaplerosis and cataplerosis, that is, the
net flow of carbon out of the citric acid cycle via
products other than CO2. Like glucose, ketones
generate ATP via the citric acid cycle and oxida-
tive phosphorylation, but, unlike glucose, ketones
do not contribute to anaplerosis (Brunengraber
and Roe, 2006). In fact, ketones are cataplerotic
because they increase citric acid cycle activity (Roy
et al., 2015) but do not provide carbon to replace
intermediates in the cycle. A metaphor would be
the complementary roles of gasoline and motor oil
in an engine; glucose provides both the gasoline
and the motor oil, so the engine will work effi-
ciently, but ketones only provide the gasoline, so
if forced to run excessively on ketones, the engine
soon burns out. Therefore, as long as sufficient
glucose is available, anaplerosis will be maintained
and citric acid cycle function will continue even
during extreme ketosis.
During short-term starvation, gluconeogen-
esis is directly proportional to ketogenesis (Garber
et al., 1974). Some of that glucose is needed by the
liver (which cannot catabolize ketones; (Fukao
125

Chapter 15: Glucose and Ketone Metabolism in the Aging Brain 125

TABLE 15.4 CLINICAL STUDIES IN WHICH HORMONAL AND COGNITIVE

RESPONSES SUGGEST THAT ACUTE OR CHRONIC KETOGENIC TREATMENTS
MAINTAIN BRAIN FUNCTION BY COMPENSATING FOR HYPOGLYCEMIA

Treatment Reference

Controlled insulin-
induced hypoglycemia
± fasting in obesity (n = 9)
Controlled insulin-
induced hypoglycemia
in healthy adults (n = 6)
Controlled insulin-
induced hypoglycemia
in healthy adults (n = 13)
Controlled insulin-
induced hypoglycemia
in Type 1 diabetes (n = 11)
Mild cognitive
Impairment (n = 23)
Mild-moderate
Alzheimer’s disease
(n = 20)
Mild-moderate
Alzheimer’s disease
(n = 77)
Severe Alzheimer’s
disease (n = 1)
Acute studies
Treatment: 2 h insulin infusion ± 60 d fast Drenick et al.,
Outcomes: ↓ effect of acute severe hypoglycemia (0.5 mM 1972
in one case), including ↓ mental confusion, anxiety, sweat-
ing, tachycardia, blood pressure if fasted for 60 d before
the insulin infusion
Treatment: 4 h i.v. β-HB infusion Amiel et al., 1991
Dose: 30 μmol/min/kg body weight
Outcomes: ↓hormonal response to hypoglycemia
Treatment: 6 h i.v. β-HB infusion Veneman et al.,
Dose: 20 µmol/min/kg body weight 1994
Outcomes: ↓ hormonal and cognitive symptoms of acute
hypoglycemia.
Treatment: single oral MCT Page et al., 2012
Dose: 40 g in three stages (20, 10, 10 g)
Outcomes: ↓ cognitive symptoms of acute hypoglycemia.
Age-associated cognitive decline
Treatment: 6 weeks high-fat ketogenic diet Krikorian et al.,
Outcomes:↑ secondary memory performance 2012
Treatment: Single dose oral of MCT (95% octanoate) Reger et al., 2004
Dose: 40 g
Outcomes: ↑ cognitive score in APOE4(-) patients
Treatment: 90 days oral MCT (95% octanoate) Henderson et al.,
Dose: 20 g/d 2009
Outcomes: ↑ cognitive score in APOE4(-) patients
Treatment: 20 months oral MCT + coconut oil (4:3) Newport et al.,
Dose: 165 mL/day 2015
Outcomes: ↑ mood, affect, self-care, and cognitive and daily
activities

β-HB: beta-hydroxybutyrate

et al., 2004), but part of the resulting glucose pro-
duction is undoubtedly used to sustain anaplero-
sis, especially in the brain. During conditions in
which glucose supply to the brain is more severely
limited, for example, inherited glucose transporter
(Glut1) deficiency (Brunengraber and Roe, 2006;
Mochel et al., 2005; Roe and Mochel, 2006), there
is insufficient glucose to maintain anaplerosis or
energy production in the brain, so providing a
ketogenic supplement may be beneficial. However,
beyond a certain point, supplying more ketones
is futile because they further stimulate the citric
acid cycle (Roy et al., 2015) without contributing
to anaplerosis so, in essence, they burn the cycle
out (Brunengraber and Roe, 2006). Whether the
situation is analogous in AD remains to be seen,
but it is clear that as AD becomes more severe,
brain glucose uptake and/or utilization continue
to deteriorate thereby further compromising both
ATP production and anaplerosis, especially if
ketogenesis is stimulated. There is most likely to
be a trade-off between supplying more ketones to
compensate for the glucose deficit and oversupply-
ing them, thereby further depleting glucose and
risking burning out the citric acid cycle.
B R A I N E N E R G Y S TAT U S ,
N E U R A L P ROT E C T I O N,
A N D M I TO C H O N D R I A L
FUNCTION
Regardless of whether glucose or ketones are being
catabolized, mitochondria are required to pro-
duce the bulk of the ATP being generated. Glucose
can produce some ATP via aerobic glycolysis, a
126
126 section II: E merging Clinical Applications and Future Potential
process occurring outside mitochondria, whereas
ketones are metabolized in an obligatory manner
via oxidative phosphorylation within mitochon-
dria. Although controversial, the enzymes of mito-
chondrial oxidative phosphorylation may actually
continue to function normally, at least early in AD,
but ATP production declines because of lower gly-
colysis to acetyl CoA. If so, there may be increased
reliance on lactate as a fuel and even on gluconeo-
genesis within the brain, with adverse consequences
for amino acid and neurotransmitter production
including acetylcholine (Hoyer et al., 1988).
Mitochondrial damage and increased pro-
duction of reactive oxygen species are widely
considered to coincide in AD and to exacerbate
the disease (Gibson et  al., 2000). Mitochondrial
dysfunction has been proposed to underlie beta-
amyloid accumulation and cognitive deteriora-
tion in AD (Swerdlow et al., 2014; Swerdlow and
Khan, 2004; Yao et  al., 2009). Neural protection
by ketones seems to be related to improved mito-
chondrial function including reduced mitochon-
drial production of reactive oxygen species in
response to glutamate (Maalouf et al., 2009). The
energy status of the brain appears to be increased
in rats on a ketogenic diet based on ATP levels and
brain glucose relative to blood glucose (Cahill,
2006; DeVivo et  al., 1978; Veech et  al., 2001), as
well as indirect measurement of citric acid cycle
activity (Roy et  al., 2015). These changes could
potentially be accompanied by mitochondrial bio-
genesis and improved respiratory function (Bough
et  al., 2006). Others have shown that brain glu-
cose uptake and Glut expression increase in rats
on a ketogenic diet (Pifferi et al., 2011; Puchowicz
et al., 2007; Roy et al., 2012), but this has not been
confirmed in humans; in fact; both PET and AVD
studies suggest that, in human adults, brain glu-
cose uptake decreases as brain ketone availability
increases (Hasselbalch et  al.,  1995, Courchesne-
Loyer et al. 2016).
SAFETY OF MEDIUM CHAIN
T R I G LY C E R I D E S
In doses up to 1 g/kg/d, MCTs have a robust safety
record in all species studies including humans
(Bach and Babayan, 1982; Traul et al., 2000). Thus,
they are not associated with increased risk of
cancer, obesity, or other diseases. However, they
can have secondary side effects involving gastro-
intestinal distress, an issue that can be partially
mitigated by gradual dose titration. The MCTs are
saturated fats and, as such, have long been painted
with the brush of increasing cardiovascular risk.
However, treatment for 30  days with 30 g/d of
MCTs does not adversely affect serum glucose,
insulin, triglycerides, cholesterol, free fatty acids,
body weight, or body-mass index (Courchesne-
Loyer et al., 2013).
E A R LY B R A I N
DEVELOPMENT AND
H U M A N B R A I N E VO L U T I O N
Although outside the scope of this review, it has
become clear over the past 3–4 decades that ketones
are much more important to the developing brain
than to the adult brain. Unlike in adults, ketones
are an essential fuel and brain lipid substrate in
infants because there is insufficient glucose avail-
able for it alone to meet brain energy require-
ments in human neonates (Bougneres et al., 1986;
Robinson and Williamson, 1980; Settergren et al.,
1976). The fatness of human babies makes this
essential role of ketones in neonates all the more
plausible given that long chain fatty acids stored in
adipose tissue are the main substrate for ketogen-
esis. It is unlikely to be a coincidence that ketones
are essential for human brain development and
that humans have fat babies; rather, these two con-
ditions probably coevolved during human evolu-
tion with the former undoubtedly being facilitated
by the latter (Cunnane and Crawford, 2014). The
potential utility of ketogenic supplements to the
aging brain is perhaps analogous to their impor-
tant role in the developing brain.
PERSPECTIVES
We propose that AD is in part exacerbated by a
form of chronic gradual fuel starvation or exhaus-
tion due to brain glucose deficit and that early in the
disease this feature may be amenable to treatments
that moderately but safely raise plasma ketones. The
broadly similar neurological/cognitive benefit of
prolonged fasting, a very high-fat ketogenic diet con-
taining no MCTs, or a regular diet to which MCTs or
ketone esters are added (Table 15.4) suggests that the
improvement in cognition is related to a common
denominator of raised plasma ketones, which bypass
chronically impaired glucose uptake and utilization
by the AD brain. However, this is still speculative;
to date, no study has yet provided a mechanism
by which MCT or ketones improve cognitive out-
comes. We have deliberately not discussed the
beta-amyloid hypothesis of AD, but beta-amyloid
accumulation could arise as a result of (Meier-Ruge
et al., 1994) or be contributing to (Meier-Ruge and
Bertoni-Freddari, 1997)  impaired glycolysis in the
AD brain. It has also not been established whether
certain MCTs (e.g., octanoate vs. decanoate) are
more effective for ketogenesis than others. Octanoic
127

Chapter 15: Glucose and Ketone Metabolism in the Aging Brain 127

TABLE 15.5 ONGOING REGISTERED CLINICAL TRIALS INVOLVING KETOGENIC

TREATMENTS IN MILD COGNITIVE IMPAIRMENT OR ALZHEIMER’S DISEASE

Conditions Treatment Reference #

Mild cognitive impairment Treatment: 6 months oral MCT beverage NCT02551419

Mild cognitive impairment
Mild Alzheimer’s disease
Mild-moderate Alzheimer’s
disease
Design: Double-blind, randomized, placebo-controlled trial
(n = 50/group)
Dose: 30 g/d
Outcome: Improvement in cognitive performance and
change in brain glucose and ketone metabolism
Treatment: 6 months oral MCT oil NCT01669200
Design: Double-blind, randomized controlled trial (n = 25/
group)
Dose: 40 g/d
Outcome: Improvement in cognitive performance
Treatment: 14 days oral MCT NCT01702480
Design: Double-blind, randomized, placebo-controlled trial
(n = 8)
Dose: 10, 20, 30, and 40 g/d
Outcome: Change in CANTAB task performances and P300
EEG measurement
Treatment: ≥ 6 months of AXONA (MCT powder) NCT01538212
Design: retrospective trial (n = 200)
Outcome: Change in living situation, psychiatric condition,
daily living, and cognitive performance

acid can be taken up by the brain (Ebert et al., 2003;
Kuge et  al., 1995), so it may have direct effects on
brain function including but not limited to conver-
sion to ketones by astrocytes (Auestad et al., 1991).
Several clinical trials are underway that combine
either a ketogenic supplement or diet with cognitive
evaluation (Table 15.5) and should produce results
before the end of 2017, which will shed further light
on the potential utility of ketotherapeutics in neuro-
degenerative diseases such as AD.
Attempts to correct or bypass impaired brain
glucose metabolism in AD have also been made
using several other approaches including supple-
ments of glucose (Korol and Gold, 1998), a cocktail
of glucose, malate, and antioxidant (Blass, 2008) or
intranasal insulin (Benedict et  al., 2011; Craft,
2012). The details of those studies are beyond the
scope of this review but they are important because
they show that for 20  years now, the metabolic
defect in the AD brain has been viewed by many
different groups as a logical target for therapeutic
strategies in AD (Cunnane et  al., 2015; Cunnane
et al., 2011; Gibson et al., 2000; Henderson, 2008;
Hertz et al., 2015; Kapogiannis and Mattson, 2011;
Yao et  al., 2011). Whether trying to correct the
problem of impaired glucose uptake-metabolism
and deteriorating mitochondrial function should
be undertaken at the same time as providing a
ketogenic supplement remains to be assessed.
AC K N OW L E D G M E N T S
Christine Brodeur-Dubreuil and Eric Lavallée
provided excellent technical assistance. Financial
support was provided by a Canada Research
Chair and University Research Chair (SCC), CFI,
CIHR, NSERC, FRQS, FQRNT, Sojecci II, The
Alzheimer Association (USA), and the Université
de Sherbrooke. Some MCTs used in our studies
were provided as gifts by ABITEC and Nutricia.
REFERENCES
Auestad, N., Korsak, R.A., Morrow, J.W., and Edmond,
J. (1991). Fatty acid oxidation and ketogenesis by
astrocytes in primary culture. J Neurochem 56,
1376–1386.
Avogaro, A., Nosadini, R., Bier, D.M., Cobelli,
C., Toffolo, G., Doria, A., Valerio, A., and
Christopherson, H. (1990). Ketone body kinet-
ics in vivo using simultaneous administration of
acetoacetate and 3-hydroxybutyrate labelled with
stable isotopes. Acta Diabetol Lat 27, 41–51.
Bach, A.C., and Babayan, V.K. (1982). Medium-
chain triglycerides: an update. Am J Clin Nutr 36,
950–962.
128
128 section II: E merging Clinical Applications and Future Potential
Baker, L.D., Cross, D.J., Minoshima, S., Belongia, D.,
Watson, G.S., and Craft, S. (2011). Insulin resis-
tance and Alzheimer-like reductions in regional
cerebral glucose metabolism for cognitively nor-
mal adults with prediabetes or early type 2 diabe-
tes. Arch Neurol 68, 51–57.
Balasse, E.O. (1979). Kinetics of ketone body metabo-
lism in fasting humans. Metabolism 28, 41–50.
Balasse, E.O. (1986). Importance of ketone bodies in
endogenous fat transport. Clin Nutr 5, 73–80.
Balasse, E.O., Fery, F., and Neef, M.A. (1978). Changes
induced by exercise in rates of turnover and oxi-
dation of ketone bodies in fasting man. J Appl
Physiol Respir Environ Exerc Physiol 44, 5–11.
Balasse, E.O., and Fery, F. (1989). Ketone body produc-
tion and disposal: effects of fasting, diabetes, and
exercise. Diabetes Metab Rev 5, 247–270.
Benedict, C., Frey, W.H., 2nd, Schioth, H.B., Schultes,
B., Born, J., and Hallschmid, M. (2011). Intranasal
insulin as a therapeutic option in the treatment
of cognitive impairments. Exp Gerontol 46,
112–115.
Benson, D.F., Kuhl, D.E., Hawkins, R.A., Phelps, M.E.,
Cummings, J.L., and Tsai, S.Y. (1983). The fluoro-
deoxyglucose 18F scan in Alzheimer’s disease and
multi-infarct dementia. Arch Neurol 40, 711–714.
Bickerton, A.S., Roberts, R., Fielding, B.A., Tornqvist,
H., Blaak, E.E., Wagenmakers, A.J., Gilbert, M.,
Humphreys, S.M., Karpe, F., and Frayn, K.N.
(2008). Adipose tissue fatty acid metabolism in
insulin-resistant men. Diabetologia 51, 1466–1474.
Blass, J.P. (2008). A new approach to treating
Alzheimer’s disease. Ann NY Acad Sci 1147,
122–128.
Blennow, K., de Leon, M.J., and Zetterberg, H. (2006).
Alzheimer’s disease. Lancet 368, 387–403.
Blomqvist, G., Alvarsson, M., Grill, V., Von Heijne, G.,
Ingvar, M., Thorell, J.O., Stone-Elander, S., Widen,
L., and Ekberg, K. (2002). Effect of acute hyperke-
tonemia on the cerebral uptake of ketone bodies
in nondiabetic subjects and IDDM patients. Am J
Physiol Endocrinol Metab 283, E20–28.
Blomqvist, G., Thorell, J.O., Ingvar, M., Grill, V.,
Widen, L., and Stone-Elander, S. (1995). Use of
R-beta-[1-11C]hydroxybutyrate in PET studies
of regional cerebral uptake of ketone bodies in
humans. Am J Physiol 269, E948–959.
Bough, K.J., Wetherington, J., Hassel, B., Pare, J.F.,
Gawryluk, J.W., Greene, J.G., Shaw, R., Smith,
Y., Geiger, J.D., and Dingledine, R.J. (2006).
Mitochondrial biogenesis in the anticonvulsant
mechanism of the ketogenic diet. Ann Neurol 60,
223–235.
Bougneres, P.F., Lemmel, C., Ferre, P., and Bier, D.M.
(1986). Ketone body transport in the human neo-
nate and infant. J Clin Invest 77, 42–48.
Bowen, D.M., White, P., Spillane, J.A., Goodhardt,
M.J., Curzon, G., Iwangoff, P., Meier-Ruge, W.,
and Davison, A.N. (1979). Accelerated ageing or
selective neuronal loss as an important cause of
dementia? Lancet 1, 11–14.
Brown, A.D., McMorris, C.A., Longman, R.S., Leigh,
R., Hill, M.D., Friedenreich, C.M., and Poulin,
M.J. (2010). Effects of cardiorespiratory fitness and
cerebral blood flow on cognitive outcomes in older
women. Neurobiol Aging 31, 2047–2057.
Brunengraber, H., and Roe, C.R. (2006). Anaplerotic
molecules: current and future. J Inherit Metab Dis
29, 327–331.
Bu, G. (2009). Apolipoprotein E and its receptors in
Alzheimer’s disease:  pathways, pathogenesis and
therapy. Nat Rev Neurosci 10, 333–344.
Cahill, G.F., Jr. (2006). Fuel metabolism in starvation.
Annu Rev Nutr 26, 1–22.
Castellano, C.A., Baillargeon, J.P., Nugent, S.,
Tremblay, S., Fortier, M., Imbeault, H., Duval, J.,
and Cunnane, S.C. (2015a). Regional brain glucose
hypometabolism in young women with polycys-
tic ovary syndrome: possible link to mild insulin
resistance. PLOS One, 10(12)
Castellano, C.A., Nugent, S., Paquet, N., Tremblay, S.,
Bocti, C., Lacombe, G., Imbeault, H., Turcotte, E.,
Fulop, T., and Cunnane, S.C. (2015b). Lower brain
18F-fluorodeoxyglucose uptake but normal 11C-
acetoacetate metabolism in mild Alzheimer’s dis-
ease dementia. J Alzheimers Dis 43, 1343–1353.
Clarke, K., Tchabanenko, K., Pawlosky, R., Carter,
E., Todd King, M., Musa-Veloso, K., Ho, M.,
Roberts, A., Robertson, J., Vanitallie, T.B., et  al.
(2012). Kinetics, safety and tolerability of (R)-3-
hydroxybutyl (R)-3-hydroxybutyrate in healthy
adult subjects. Regul Toxicol Pharmacol 63,
401–408.
Courchesne-Loyer, A., Fortier, M., Tremblay-
Mercier, J., Chouinard-Watkins, R., Roy, M.,
Nugent, S., Castellano, C.A., and Cunnane, S.C.
(2013). Stimulation of mild, sustained keto-
nemia by medium-chain triacylglycerols in
healthy humans:  estimated potential contribu-
tion to brain energy metabolism. Nutrition 29,
635–640.
Courchesne-Loyer, A., Croteau, E., Castellano, C.A.,
St-Pierre, V., Hennebelle, M., and Cunnane, S.C.
(2016). Inverse relationship between brain glucose
and ketone metabolism in adults during short-
term moderate dietary ketosis: a dual tracer quan-
titative PET study. J Cereb Blood Flow Metab.
Craft, S. (2009). The role of metabolic disorders in
Alzheimer disease and vascular dementia:  two
roads converged. Arch Neurol 66, 300–305.
Craft, S. (2012). Alzheimer disease: Insulin resistance
and AD—extending the translational path. Nat
Rev Neurol 8, 360–362.
Cunnane, S., Nugent, S., Roy, M., Courchesne-Loyer,
A., Croteau, E., Tremblay, S., Castellano, A., Pifferi,
F., Bocti, C., Paquet, N., et  al. (2011). Brain fuel
metabolism, aging, and Alzheimer’s disease.
Nutrition 27, 3–20.
129
Chapter 15: Glucose and Ketone Metabolism in the Aging Brain
Cunnane, S.C., and Crawford, M.A. (2014). Energetic
and nutritional constraints on infant brain devel-
opment:  implications for brain expansion during
human evolution. J Hum Evol 77, 88–98.
Cunnane, S.C., Courchesne-Loyer, A., St-Pierre, V.,
Vandenberghe, C., Pierrotti, T., Fortier, M.,
Croteau, E., and Castellano, A. (2015). Can ketones
compensate for deteriorating brain glucose uptake
during aging? Implications for the risk and treat-
ment of Alzheimer’s disease. Ann NY Acad Sci
1367, 12–20.
Dalsgaard, M.K., Quistorff, B., Danielsen, E.R.,
Selmer, C., Vogelsang, T., and Secher, N.H. (2004).
A reduced cerebral metabolic ratio in exercise
reflects metabolism and not accumulation of
lactate within the human brain. J Physiol 554,
571–578.
Dastur, D.K. (1985). Cerebral blood flow and metabo-
lism in normal human aging, pathological aging,
and senile dementia. J Cereb Blood Flow Metab
5, 1–9.
DeVivo, D.C., Leckie, M.P., Ferrendelli, J.S., and
McDougal, D.B., Jr. (1978). Chronic ketosis and
cerebral metabolism. Ann Neurol 3, 331–337.
Douaud, G., Refsum, H., de Jager, C.A., Jacoby, R.,
Nichols, T.E., Smith, S.M., and Smith, A.D. (2013).
Preventing Alzheimer’s disease-related gray mat-
ter atrophy by B-vitamin treatment. Proc Natl
Acad Sci USA 110, 9523–9528.
Drenick, E.J., Alvarez, L.C., Tamasi, G.C., and
Brickman, A.S. (1972). Resistance to symptom-
atic insulin reactions after fasting. J Clin Invest 51,
2757–2762.
Ebert, D., Haller, R.G., and Walton, M.E. (2003).
Energy contribution of octanoate to intact rat
brain metabolism measured by 13C nuclear
magnetic resonance spectroscopy. J Neurosci 23,
5928–5935.
Fery, F., and Balasse, E.O. (1983). Ketone body turn-
over during and after exercise in overnight-fasted
and starved humans. Am J Physiol 245, E318–325.
Fery, F., and Balasse, E.O. (1986). Response of ketone
body metabolism to exercise during transition
from postabsorptive to fasted state. Am J Physiol
250, E495–501.
Flatt, J.P. (1972). On the maximal possible rate of keto-
genesis. Diabetes 21, 50–53.
Fox, N.C., Kennedy, A.M., Harvey, R.J., Lantos, P.L.,
Roques, P.K., Collinge, J., Hardy, J., Hutton, M.,
Stevens, J.M., Warrington, E.K., et  al. (1997).
Clinicopathological features of familial Alzheimer’s
disease associated with the M139V mutation in
the presenilin 1 gene: pedigree but not mutation;
specific age at onset provides evidence for a further
genetic factor. Brain 120 (Pt 3), 491–501.
Freemantle, E., Vandal, M., Tremblay Mercier, J.,
Plourde, M., Poirier, J., and Cunnane, S.C. (2009).
Metabolic response to a ketogenic breakfast in the
healthy elderly. J Nutr Health Aging 13, 293–298.
129
Freund, G., and Weinsier, R.L. (1966). Standardized
ketosis in man following medium chain triglycer-
ide ingestion. Metabolism 15, 980–991.
Fukao, T., Lopaschuk, G.D., and Mitchell, G.A. (2004).
Pathways and control of ketone body metabolism: on
the fringe of lipid biochemistry. Prostaglandins
Leukot Essent Fatty Acids 70, 243–251.
Garber, A.J., Menzel, P.H., Boden, G., and Owen, O.E.
(1974). Hepatic ketogenesis and gluconeogenesis
in humans. J Clin Invest 54, 981–989.
Gibson, R.E., Burns, H.D., Hamill, T.G., Eng, W.S.,
Francis, B.E., and Ryan, C. (2000). Non-invasive
radiotracer imaging as a tool for drug develop-
ment. Curr Pharm Des 6, 973–989.
Gottstein, U., Muller, W., Berghoff, W., Gartner, H.,
and Held, K. (1971). [Utilization of non-esterified
fatty acids and ketone nbodies in human brain].
Klin Wochenschr 49, 406–411.
Hall, S.E., Wastney, M.E., Bolton, T.M., Braaten, J.T.,
and Berman, M. (1984). Ketone body kinetics in
humans: the effects of insulin-dependent diabetes,
obesity, and starvation. J Lipid Res 25, 1184–1194.
Hasselbalch, S.G., Knudsen, G.M., Jakobsen, J.,
Hageman, L.P., Holm, S., and Paulson, O.B. (1995).
Blood-brain barrier permeability of glucose and
ketone bodies during short-term starvation in
humans. Am J Physiol 268, E1161–1166.
Hasselbalch, S.G., Madsen, P.L., Hageman, L.P., Olsen,
K.S., Justesen, N., Holm, S., and Paulson, O.B.
(1996). Changes in cerebral blood flow and carbo-
hydrate metabolism during acute hyperketonemia.
Am J Physiol 270, E746–751.
Henderson, S.T. (2008). Ketone bodies as a therapeu-
tic for Alzheimer’s disease. Neurotherapeutics 5,
470–480.
Henderson, S.T., Vogel, J.L., Barr, L.J., Garvin, F.,
Jones, J.J., and Costantini, L.C. (2009). Study of
the ketogenic agent AC-1202 in mild to moderate
Alzheimer’s disease:  a randomized, double-blind,
placebo-controlled, multicenter trial. Nutr Metab
(Lond) 6, 31.
Hertz, L., Chen, Y., and Waagepetersen, H.S. (2015).
Effects of ketone bodies in Alzheimer’s disease in rela-
tion to neural hypometabolism, beta-amyloid toxic-
ity, and astrocyte function. J Neurochem 134, 7–20.
Hoyer, S. (1992). Oxidative energy metabolism in
Alzheimer brain. Studies in early-onset and
late-onset cases. Mol Chem Neuropathol 16,
207–224.
Hoyer, S., Oesterreich, K., and Wagner, O. (1988).
Glucose metabolism as the site of the primary
abnormality in early-onset dementia of Alzheimer
type? J Neurol 235, 143–148.
Ide, K., Schmalbruch, I.K., Quistorff, B., Horn, A.,
and Secher, N.H. (2000). Lactate, glucose and O2
uptake in human brain during recovery from max-
imal exercise. J Physiol 522 Pt 1, 159–164.
Iwangoff, P., Armbruster, R., Enz, A., and Meier-
Ruge, W. (1980). Glycolytic enzymes from human
130
130 section II: E merging Clinical Applications and Future Potential
autoptic brain cortex: normal aged and demented
cases. Mech Ageing Dev 14, 203–209.
Jack, C.R., Jr., Lowe, V.J., Weigand, S.D., Wiste, H.J.,
Senjem, M.L., Knopman, D.S., Shiung, M.M.,
Gunter, J.L., Boeve, B.F., Kemp, B.J., et al. (2009).
Serial PIB and MRI in normal, mild cognitive
impairment and Alzheimer’s disease: implications
for sequence of pathological events in Alzheimer’s
disease. Brain 132, 1355–1365.
Jerneren, F., Elshorbagy, A.K., Oulhaj, A., Smith, S.M.,
Refsum, H., and Smith, A.D. (2015). Brain atrophy
in cognitively impaired elderly:  the importance
of long-chain omega-3 fatty acids and B vitamin
status in a randomized controlled trial. Am J Clin
Nutr 102, 215–221.
Kalpouzos, G., Chetelat, G., Baron, J.C., Landeau, B.,
Mevel, K., Godeau, C., Barre, L., Constans, J.M.,
Viader, F., Eustache, F., et al. (2009). Voxel-based
mapping of brain gray matter volume and glucose
metabolism profiles in normal aging. Neurobiol
Aging 30, 112–124.
Kapogiannis, D., and Mattson, M.P. (2011). Disrupted
energy metabolism and neuronal circuit dysfunc-
tion in cognitive impairment and Alzheimer’s dis-
ease. Lancet Neurol 10, 187–198.
Kennedy, A.M., Frackowiak, R.S., Newman, S.K.,
Bloomfield, P.M., Seaward, J., Roques, P.,
Lewington, G., Cunningham, V.J., and Rossor,
M.N. (1995). Deficits in cerebral glucose metabo-
lism demonstrated by positron emission tomogra-
phy in individuals at risk of familial Alzheimer’s
disease. Neurosci Lett 186, 17–20.
Korol, D.L., and Gold, P.E. (1998). Glucose, memory,
and aging. Am J Clin Nutr 67, 764S–771S.
Krikorian, R., Shidler, M.D., Dangelo, K., Couch, S.C.,
Benoit, S.C., and Clegg, D.J. (2012). Dietary keto-
sis enhances memory in mild cognitive impair-
ment. Neurobiol Aging 33, 425 e419–427.
Kuge, Y., Yajima, K., Kawashima, H., Yamazaki, H.,
Hashimoto, N., and Miyake, Y. (1995). Brain
uptake and metabolism of [1-11C]octanoate in
rats: pharmacokinetic basis for its application as a
radiopharmaceutical for studying brain fatty acid
metabolism. Ann Nucl Med 9, 137–142.
London, E.D., Margolin, R.A., Duara, R., Holloway,
H.W., Robertson-Tchabo, E.A., Cutler, N.R., and
Rapoport, S.I. (1986). Effects of fasting on ketone
body concentrations in healthy men of different
ages. J Gerontol 41, 599–604.
Lying-Tunell, U., Lindblad, B.S., Malmlund, H.O., and
Persson, B. (1981). Cerebral blood flow and met-
abolic rate of oxygen, glucose, lactate, pyruvate,
ketone bodies and amino acids. Acta Neurol Scand
63, 337–350.
Maalouf, M., Rho, J.M., and Mattson, M.P. (2009). The
neuroprotective properties of calorie restriction,
the ketogenic diet, and ketone bodies. Brain Res
Rev 59, 293–315.
Mamelak, M. (2012). Sporadic Alzheimer’s disease: the
starving brain. J Alzheimers Dis 31, 459–474.
Matsuzaki, T., Sasaki, K., Tanizaki, Y., Hata, J., Fujimi,
K., Matsui, Y., Sekita, A., Suzuki, S.O., Kanba,
S., Kiyohara, Y., et  al. (2010). Insulin resistance
is associated with the pathology of Alzheimer
disease:  the Hisayama study. Neurology 75,
764–770.
Meier-Ruge, W., Bertoni-Freddari, C., and Iwangoff,
P. (1994). Changes in brain glucose metabolism as
a key to the pathogenesis of Alzheimer’s disease.
Gerontology 40, 246–252.
Meier-Ruge, W.A., and Bertoni-Freddari, C. (1997).
Pathogenesis of decreased glucose turnover
and oxidative phosphorylation in ischemic and
trauma-induced dementia of the Alzheimer type.
Ann NY Acad Sci 826, 229–241.
Mitchell, G.A., Kassovska-Bratinova, S., Boukaftane,
Y., Robert, M.F., Wang, S.P., Ashmarina, L.,
Lambert, M., Lapierre, P., and Potier, E. (1995).
Medical aspects of ketone body metabolism. Clin
Invest Med 18, 193–216.
Mochel, F., DeLonlay, P., Touati, G., Brunengraber, H.,
Kinman, R.P., Rabier, D., Roe, C.R., and Saudubray,
J.M. (2005). Pyruvate carboxylase deficiency: clin-
ical and biochemical response to anaplerotic diet
therapy. Mol Genet Metab 84, 305–312.
Mosconi, L., Tsui, W.H., De Santi, S., Li, J., Rusinek,
H., Convit, A., Li, Y., Boppana, M., and de Leon,
M.J. (2005). Reduced hippocampal metabolism in
MCI and AD: automated FDG-PET image analy-
sis. Neurology 64, 1860–1867.
Newport, M.T., VanItallie, T.B., Kashiwaya, Y., King,
M.T., and Veech, R.L. (2015). A new way to pro-
duce hyperketonemia: use of ketone ester in a case of
Alzheimer’s disease. Alzheimers Dement 11, 99–103.
Ngandu, T., Lehtisalo, J., Solomon, A., Levalahti,
E., Ahtiluoto, S., Antikainen, R., Backman, L.,
Hanninen, T., Jula, A., Laatikainen, T., et al. (2015).
A 2  year multidomain intervention of diet, exer-
cise, cognitive training, and vascular risk monitor-
ing versus control to prevent cognitive decline in
at-risk elderly people (FINGER):  a randomised
controlled trial. Lancet 385, 2255–2263.
Nordberg, A. (2006). Mechanisms behind the neuro-
protective actions of cholinesterase inhibitors in
Alzheimer disease. Alzheimer Dis Assoc Disord
20, S12–18.
Nugent, S., Castellano, C.A., Goffaux, P., Whittingstall,
K., Lepage, M., Paquet, N., Bocti, C., Fulop, T., and
Cunnane, S.C. (2014a). Glucose hypometabolism
is highly localized but lower cortical thickness and
brain atrophy are widespread in cognitively nor-
mal older adults. Am J Physiol Endocrinol Metab.
Nugent, S., Tremblay, S., Chen, K.W., Ayutyanont, N.,
Roontiva, A., Castellano, C.A., Fortier, M., Roy,
M., Courchesne-Loyer, A., Bocti, C., et al. (2014b).
Brain glucose and acetoacetate metabolism:  a
131
Chapter 15: Glucose and Ketone Metabolism in the Aging Brain
comparison of young and older adults. Neurobiol
Aging 35, 1386–1395.
Nugent, S., Castellano, C.A., Bocti, C., Dionne, I.,
Fulop, T., and Cunnane, S.C. (2015). Relationship
of metabolic and endocrine parameters to brain
glucose metabolism in older adults: do cognitively-
normal older adults have a particular metabolic
phenotype? Biogerontology 17, 241–255.
Ogawa, M., Fukuyama, H., Ouchi, Y., Yamauchi, H.,
and Kimura, J. (1996). Altered energy metabolism
in Alzheimer’s disease. J Neurol Sci 139, 78–82.
Okonkwo, O.C., Schultz, S.A., Oh, J.M., Larson, J.,
Edwards, D., Cook, D., Koscik, R., Gallagher,
C.L., Dowling, N.M., Carlsson, C.M., et al. (2014).
Physical activity attenuates age-related biomarker
alterations in preclinical AD. Neurology 83,
1753–1760.
Owen, O.E., Morgan, A.P., Kemp, H.G., Sullivan,
J.M., Herrera, M.G., and Cahill, G.F., Jr. (1967).
Brain metabolism during fasting. J Clin Invest 46,
1589–1595.
Owen, O.E., and Reichard, G.A., Jr. (1971). Human
forearm metabolism during progressive starva-
tion. J Clin Invest 50, 1536–1545.
Owen, O.E., Kalhan, S.C., and Hanson, R.W. (2002).
The key role of anaplerosis and cataplerosis
for citric acid cycle function. J Biol Chem 277,
30409–30412.
Page, K.A., Williamson, A., Yu, N., McNay, E.C.,
Dzuira, J., McCrimmon, R.J., and Sherwin, R.S.
(2009). Medium-chain fatty acids improve cogni-
tive function in intensively treated type 1 diabetic
patients and support in vitro synaptic transmis-
sion during acute hypoglycemia. Diabetes 58,
1237–1244.
Pan, J.W., Telang, F.W., Lee, J.H., de Graaf, R.A.,
Rothman, D.L., Stein, D.T., and Hetherington, H.P.
(2001). Measurement of beta-hydroxybutyrate
in acute hyperketonemia in human brain. J
Neurochem 79, 539–544.
Perry, E.K., Perry, R.H., Tomlinson, B.E., Blessed, G.,
and Gibson, P.H. (1980). Coenzyme A-acetylating
enzymes in Alzheimer’s disease:  possible cholin-
ergic “compartment” of pyruvate dehydrogenase.
Neurosci Lett 18, 105–110.
Pi-Sunyer, F.X., Hashim, S.A., and Van Itallie, T.B.
(1969). Insulin and ketone responses to ingestion
of medium and long-chain triglycerides in man.
Diabetes 18, 96–100.
Pifferi, F., Tremblay, S., Croteau, E., Fortier, M.,
Tremblay-Mercier, J., Lecomte, R., and Cunnane,
S.C. (2011). Mild experimental ketosis increases
brain uptake of (11)C-acetoacetate and (18)F-
fluorodeoxyglucose:  a dual-tracer PET imaging
study in rats. Nutr Neurosci 14, 51–58.
Plecko, B., Stoeckler-Ipsiroglu, S., Schober, E., Harrer,
G., Mlynarik, V., Gruber, S., Moser, E., Moeslinger,
D., Silgoner, H., and Ipsiroglu, O. (2002). Oral
131
beta-hydroxybutyrate supplementation in two
patients with hyperinsulinemic hypoglyce-
mia: monitoring of beta-hydroxybutyrate levels in
blood and cerebrospinal fluid, and in the brain by
in vivo magnetic resonance spectroscopy. Pediatr
Res 52, 301–306.
Puchowicz, M.A., Xu, K., Sun, X., Ivy, A., Emancipator,
D., and LaManna, J.C. (2007). Diet-induced keto-
sis increases capillary density without altered
blood flow in rat brain. Am J Physiol Endocrinol
Metab 292, E1607–1615.
Quinn, J.F., Raman, R., Thomas, R.G., Yurko-Mauro,
K., Nelson, E.B., Van Dyck, C., Galvin, J.E.,
Emond, J., Jack, C.R., Jr., Weiner, M., et al. (2010).
Docosahexaenoic acid supplementation and cog-
nitive decline in Alzheimer disease: a randomized
trial. JAMA 304, 1903–1911.
Rafii, M.S., and Aisen, P.S. (2009). Recent develop-
ments in Alzheimer’s disease therapeutics. BMC
Med 7, 7.
Reger, M.A., Henderson, S.T., Hale, C., Cholerton, B.,
Baker, L.D., Watson, G.S., Hyde, K., Chapman, D.,
and Craft, S. (2004). Effects of beta-hydroxybutyrate
on cognition in memory-impaired adults. Neurobiol
Aging 25, 311–314.
Reichard, G.A., Jr., Owen, O.E., Haff, A.C., Paul, P., and
Bortz, W.M. (1974). Ketone-body production and
oxidation in fasting obese humans. J Clin Invest
53, 508–515.
Reiman, E.M., Chen, K., Alexander, G.E., Caselli,
R.J., Bandy, D., Osborne, D., Saunders, A.M., and
Hardy, J. (2004). Functional brain abnormali-
ties in young adults at genetic risk for late-onset
Alzheimer’s dementia. Proc Natl Acad Sci USA
101, 284–289.
Robinson, A.M., and Williamson, D.H. (1980).
Physiological roles of ketone bodies as substrates
and signals in mammalian tissues. Physiol Rev 60,
143–187.
Roe, C.R., and Mochel, F. (2006). Anaplerotic diet
therapy in inherited metabolic disease: therapeu-
tic potential. J Inherit Metab Dis 29, 332–340.
Ronnemaa, E., Zethelius, B., Sundelof, J., Sundstrom,
J., Degerman-Gunnarsson, M., Berne, C., Lannfelt,
L., and Kilander, L. (2008). Impaired insulin
secretion increases the risk of Alzheimer disease.
Neurology 71, 1065–1071.
Roy, M., Nugent, S., Tremblay-Mercier, J., Tremblay,
S., Courchesne-Loyer, A., Beaudoin, J.F.,
Tremblay, L., Descoteaux, M., Lecomte, R., and
Cunnane, S.C. (2012). The ketogenic diet increases
brain glucose and ketone uptake in aged rats:  a
dual tracer PET and volumetric MRI study. Brain
Res 1488, 14–23.
Roy, M., Beauvieux, M.C., Naulin, J., El Hamrani,
D., Gallis, J.L., Cunnane, S.C., and Bouzier-Sore,
A.K. (2015). Rapid adaptation of rat brain and
liver metabolism to a ketogenic diet: an integrated
132
132 section II: E merging Clinical Applications and Future Potential
study using (1)H- and (13)C-NMR spectroscopy. J
Cereb Blood Flow Metab 35, 1154–1162.
Scholl, M., Almkvist, O., Bogdanovic, N., Wall, A.,
Langstrom, B., Viitanen, M., and Nordberg, A.
(2011). Time course of glucose metabolism in rela-
tion to cognitive performance and postmortem
neuropathology in Met146Val PSEN1 mutation
carriers. J Alzheimers Dis 24, 495–506.
Schrijvers, E.M., Witteman, J.C., Sijbrands, E.J.,
Hofman, A., Koudstaal, P.J., and Breteler,
M.M. (2010). Insulin metabolism and the risk
of Alzheimer disease:  the Rotterdam Study.
Neurology 75, 1982–1987.
Seaton, T.B., Welle, S.L., Warenko, M.K., and Campbell,
R.G. (1986). Thermic effect of medium-chain and
long-chain triglycerides in man. Am J Clin Nutr
44, 630–634.
Settergren, G., Lindblad, B.S., and Persson, B.
(1976). Cerebral blood flow and exchange of
oxygen, glucose, ketone bodies, lactate, pyruvate
and amino acids in infants. Acta Paediatr Scand
65, 343–353.
Simpson, I.A., Carruthers, A., and Vannucci, S.J.
(2007). Supply and demand in cerebral energy
metabolism:  the role of nutrient transporters. J
Cereb Blood Flow Metab 27, 1766–1791.
Sokoloff, L., Reivich, M., Kennedy, C., Des Rosiers,
M.H., Patlak, C.S., Pettigrew, K.D., Sakurada, O.,
and Shinohara, M. (1977). The [14C]deoxyglu-
cose method for the measurement of local cerebral
glucose utilization: theory, procedure, and normal
values in the conscious and anesthetized albino
rat. J Neurochem 28, 897–916.
Sorbi, S., Bird, E.D., and Blass, J.P. (1983). Decreased
pyruvate dehydrogenase complex activity in
Huntington and Alzheimer brain. Ann Neurol
13, 72–78.
Swerdlow, R.H., Burns, J.M., and Khan, S.M. (2014).
The Alzheimer’s disease mitochondrial cascade
hypothesis:  progress and perspectives. Biochim
Biophys Acta 1842, 1219–1231.
Swerdlow, R.H., and Khan, S.M. (2004). A “mitochon-
drial cascade hypothesis” for sporadic Alzheimer’s
disease. Med Hypotheses 63, 8–20.
Traul, K.A., Driedger, A., Ingle, D.L., and Nakhasi,
D. (2000). Review of the toxicologic properties of
medium-chain triglycerides. Food Chem Toxicol
38, 79–98.
Tremblay, S., Ouellet, R., Rodrigue, S., Langlois, R.,
Benard, F., and Cunnane, S.C. (2007). Automated
synthesis of 11C-acetoacetic acid, a key alter-
nate brain fuel to glucose. Appl Radiat Isot 65,
934–940.
Veech, R.L., Chance, B., Kashiwaya, Y., Lardy, H.A.,
and Cahill, G.F., Jr. (2001). Ketone bodies, poten-
tial therapeutic uses. IUBMB Life 51, 241–247.
Wang, T., Huang, Q., Reiman, E.M., Chen, K., Li, X.,
Li, G., Lin, Z., Li, C., and Xiao, S. (2013). Effects of
memantine on clinical ratings, fluorodeoxyglucose
positron emission tomography measurements, and
cerebrospinal fluid assays in patients with moder-
ate to severe Alzheimer dementia: a 24-week, ran-
domized, clinical trial. J Clin Psychopharmacol 33,
636–642.
Yao, J., Irwin, R.W., Zhao, L., Nilsen, J., Hamilton, R.T.,
and Brinton, R.D. (2009). Mitochondrial bioen-
ergetic deficit precedes Alzheimer’s pathology in
female mouse model of Alzheimer’s disease. Proc
Natl Acad Sci USA 106, 14670–14675.
Yao, J., Rettberg, J.R., Klosinski, L.P., Cadenas, E., and
Brinton, R.D. (2011). Shift in brain metabolism
in late onset Alzheimer’s disease: implications for
biomarkers and therapeutic interventions. Mol
Aspects Med 32, 247–257.
133
16
Ketogenic Diet and Ketones for the Treatment
of Traumatic Brain and Spinal Cord Injury
F E M K E S T R E I J G E R , P H D , WA R D T. P L U N E T, P H D ,
A N D WO L F R A M T E T Z L A F F, M D , D R . M E D , P H D
INTRODUCTION
Every year, over 1,700 Canadians suffer an acute
traumatic spinal cord injury (SCI) (Rick Hansen
Registry, 2006)  (Rick Hansen Registry:  http://
www.rickhanseninstitute.org/work/our- projects-
initiatives/rhscir), and in the United States the
number is around 12,500 (National Spinal Cord
Injury Center). Spinal cord injury has a devastat-
ing impact on the quality of life that goes beyond
the obvious paralysis and includes pain, spasticity,
sexual dysfunction, respiratory problems, bowel
and bladder disorders, skin ulcers, autonomic dys-
regulation, cardiovascular disease and metabolic
disease, leading to a greatly reduced life expec-
tancy (SCIRE, 2105). The economic burden of
traumatic SCI in Canada is enormous and ranges
from $1.47 million CAD for a person with incom-
plete paraplegia and close to $3 million CAD for a
complete tetraplegic individual over the course of
life (when injury is sustained at age 35) (Krueger
et al., 2013).
The acute management of SCI is mainly focused
on surgical stabilization of the spinal column and
surgical decompression of the spinal cord to relieve
extrinsic pressure (Fehlings et  al., 2012; Furlan
et al., 2011), although the benefits from early sur-
gical intervention are challenging to demonstrate
(Furlan et al., 2011). There are currently no neu-
roprotective or neurorestorative therapies avail-
able for SCI patients, and the only drug approved
by the US Food and Drug Administration (FDA),
methylprednisolone, showed only marginal neu-
roprotective effects when administered within 8
hours after injury (Bracken et al., 1998). Due to its
limited efficacy and significant side effects, methyl-
prednisolone is no longer recommended (Evaniew
et  al., 2015). The reality is that the only effective
treatments for people with SCI to date are reha-
bilitation programs together with medical care for
the wide range of secondary complications listed
above (SCIRE, 2105). Rehabilitation interventions
have led to significant improvements in the neu-
rological outcome after chronic motor incomplete
SCI, and more recently promising strides were
made in a few individuals with functionally com-
plete SCI (Harkema et al., 2012a; Harkema et al.,
2012b). However, there is still a large unmet clini-
cal need for treatments to protect the injured spi-
nal cord from secondary damage in the wake of
a primary injury. While little can be done for the
immediate necrosis of the tissue due to mechani-
cal destruction by the primary impact, the injury
triggers a cascade of secondary damage due to vas-
cular disruption, hemorrhage, edema, and isch-
emia, leading to energy depletion and ion pump
failures, depolarization, excitotoxicity, calcium
overload, activation of proteases, free oxygen radi-
cal formation, cell membrane compromise, lipid
peroxidation and protein nitrosylation, inflamma-
tion, and cell death by various mechanisms. Such
secondary damage results in oligodendrocyte
death with demyelination, neuronal losses, and
permanent functional impairments (for reviews
see, Oyinbo, 2011).
The quest for treatments to mitigate second-
ary injury after neurotrauma has led to the pre-
clinical discovery of numerous “neuroprotective”
drug candidates as well as metabolic and dietary
treatments. The latter include a number of health
supplements and dietary regimes that are neu-
roprotective after acute SCI in rodents including
acetyl-L-carnitine (Patel et  al., 2012; Patel et  al.,
2010; Patel et  al., 2014), creatine (Rabchevsky
et  al., 2003), melatonin (Yang et  al., 2015), natu-
ral polyphenols from green tea, olive oil, resvera-
trol, turmeric (for review, see Khalatbary, 2014),
fish oils (King et  al., 2006; Michael-Titus and
Priestley, 2014), and inosine (Kim et al., 2013) as
134
134 section II: E merging Clinical Applications and Future Potential
well as multinutrient combinations of DHA (doc-
osahexaenoic acid), EPA (eicosapentaenoic acid),
choline, phospholipids, and various vitamins mar-
keted as Fortasyn (Pallier et  al., 2015). Many of
these approaches with “naturally occurring” nutri-
ents and over-the-counter health supplements
appear to have anti-inflammatory, anti-apoptotic,
and antioxidant properties by mechanisms that
are incompletely understood, while some also
enhance synaptic plasticity in the injured spi-
nal cord of rodents (Kim et  al., 2013). However,
none of these treatments have yet to become clini-
cal standard through validation in clinical trials.
As a matter of fact, the nutritional guidelines for
acute SCI are merely based on expert opinions and
few very small cohort studies (Class  III/IV evi-
dence) (Consortium-for-Spinal-Cord-Medicine,
2006; Rodriguez et  al., 1997; Thibault-Halman
et al., 2011). Nutrition is generally recommended
within 24 hours using a balanced enteral for-
mula composed of carbohydrates, fat and protein
(Consortium-for-Spinal-Cord-Medicine, 2006;
Thibault-Halman et al., 2011).
Contrasting these clinical guidelines, our lab-
oratory discovered in rats that a form of caloric
restriction known as intermittent fasting (food
withdrawal every other day) improved outcomes
from acute cervical and thoracic SCI by promot-
ing neuroprotection and neuroplasticity (Jeong et
al., 2011; Plunet et al., 2010; Plunet et al., 2008).
These results raise new and interesting questions,
especially given the widely appreciated notion that
intermittent fasting can have a positive impact on
multiple chronic diseases in animal models and
on parameters of metabolic, cardiovascular and
inflammatory diseases in humans (Brandhorst
et al., 2015; Longo and Mattson, 2014; Maalouf
et al., 2009). Despite the growing enthusiasm
for fasting in the research field in recent years,
introducing such a regimen to acutely injured
patients is met with little enthusiasm from clini-
cians, despite the fact that the caloric basal met-
abolic rate and energy expenditure requirements
are drastically reduced after SCI due to paralysis
(Cook et al., 2008; Magnuson et al., 2011)—unlike
in traumatic brain injury (TBI), where they may
be increased (Cook et al. 2008). Fasting affects
multiple pathways, including the generation of
ketone bodies by the liver, most prominently D-
beta-hydroxybutyrate (βHB) and acetoacetate
(AcAc), which have become increasingly known
to be “neuroprotective” through their intensely
studied role in the so-called ketogenic diet (KD)
for epilepsy (for reviews, see Gano et al., 2014;
Gasior et al., 2006; Rho, 2015). KDs are high in
fat, adequate in protein, and very low in carbo-
hydrates, and induce a fasting-like state. KDs
or ketones have been shown to be beneficial in
models of Alzheimer’s disease (Hertz et al., 2015;
Imamura et al., 2006; Kashiwaya et al., 2000; Paoli
et al., 2014), Parkinson’s disease (Cheng et al.,
2009; Kashiwaya et al., 2000; Tieu et al., 2003),
Huntington’s disease (Lim et al., 2011), amyo-
trophic lateral sclerosis (Zhao et al., 2012; Zhao et
al., 2006), stroke (Austin et al., 2012; Gibson et al.,
2012; Puchowicz et al., 2008; Rahman et al., 2014;
Suzuki et al., 2002), multiple sclerosis (Kim do et
al., 2012), and TBI (Appelberg et al., 2009; Davis
et al., 2008a; Hu et al., 2009a; Prins et al., 2005).
Clinical trials of ketogenic regimens in various
forms are ongoing for a wide range of neurolog-
ical and nonneurological disorders. The former
include Parkinson’s disease (NCT013644545),
Tay Sachs disease (NCT01364545), stroke (NCT
01820663), multiple sclerosis (NCT01915433),
mild cognitive impairments (NCT02521818),
childhood epilepsy (NCT02497105), refractory
epileptic status (NCT01796574), adult epilepsy
(NCT01906398), autism (NCT02477904), gli-
oblastoma (NCT01865162) and acute TBI in
children (NCT02174016). Clinically, KDs are
successfully used for drug-resistant epilepsy in
children and for a variety of genetic metabolic dis-
orders, notably Glut1 deficiency (Kossoff, 2011).
In this chapter, we review the neuroprotective
effects of KDs and ketones as they relate to neu-
rotrauma, specifically SCI and TBI; place them
into context with our own data using KDs in
rodents with acute SCI; and discuss the potential
mechanisms. Due to the overlapping pathophysi-
ology of SCI and traumatic TBI we discuss some of
the evidences for possible benefits of KDs in both
conditions concurrently.
K E T O G E N I C D I E T I M P R OV E S
OUTCOME AFTER SPINAL
C O R D   I N J U RY
The previously established neuroprotective effects
of KD/ketone administration and the fact that
our rats treated with intermittent fasting after
SCI showed ketosis every second day, presented
our rationale for a battery of tests of whether
KD after rodent SCI could promote neuropro-
tection and recovery (Streijger et al., 2013). We
developed and applied a unilateral hemicontu-
sion model of the cervical spinal cord for adult
rats (Lee et al., 2012; Streijger et al., 2013). At 4
hours after injury, access to ad lib KD consist-
ing of a fat:protein+carbohydrate ratio of 3:1 was
given and maintained for 12 weeks. Blood βHB
135
Chapter 16: Treatment of Traumatic Brain and Spinal Cord Injury
levels reached 0.7 mmol/L at 24 hours and peaked
around 1.8 mmol/L at 7 days post injury. While
spinal cord injured rats showed a lower food
intake on the first 2 days and lost more weight
acutely after SCI, their weight returned to normal
values within approximately 2 weeks, and both
standard diet (SD) and KD rats gained weight
above the preinjury levels thereafter. We ruled out
that a 48-hour fast acutely after injury was respon-
sible for the improved recoveries seen with KD
(unpublished observation). The KD-fed rats used
the forelimb on their injured side more frequently
than carbohydrate-based SD-fed controls during
vertical exploration of their environment in a cyl-
inder. Importantly, this improvement was main-
tained beyond week 12, when KD was replaced by
SD. The KD-fed rats also displayed a greater active
range of forelimb movement during grooming.
Given that hand function is the most coveted
function in people with cervical SCI (Anderson,
2004), we performed two different reaching tests,
and on both tasks we observed an improvement
in rats fed KD in their ability to reach forward
and grasp a small pellet compared with SD-fed
animals. In addition to scoring reaching success,
which indicates little about the strategy the ani-
mal uses to obtain the pellet, reaching can be bro-
ken down into discrete components of motion in
time-lapse video-analysis (Whishaw and Gorny,
1994). In our experiments, KD improved the
pellet grasp and the subsequent supination that
directs the pellet toward the mouth. Histological
analysis revealed that KD treatment resulted in
smaller gray matter damage and greater protec-
tion of neuronal survival in the vicinity of the
lesion, less infiltration by inflammatory cells, and
increased expression of the glucose transporter
Glut1. An RT-qPCR mRNA analysis did not reveal
differences in BDNF, PDGFβ, VEGF, HIF1α, or
SDF1α—although a trending threefold increase in
SDF1α remains to be revisited.
In agreement with others (Leino et al., 2001),
KD treatment significantly increased monocar-
boxylic acid transporter 1 (MCT1) expression at
2 weeks post injury (Streijger et al., 2013), which
coincided with elevated blood βHB levels indica-
tive of increased cerebral uptake of ketones.
MCTs facilitate the transport of monocarboxylic
acids such as lactate, pyruvate, and ketone bod-
ies across biological membranes (Halestrap and
Wilson, 2012). In the brain, MCT1 is predomi-
nantly expressed by vascular endothelial cells,
astrocytes, and oligodendrocytes (Halestrap
and Wilson, 2012; Pierre et al., 2002; Pierre and
Pellerin, 2005), while MCT2 is mainly neuronal
135
(Pierre et al., 2002). Interestingly, while KD pre-
served tissue following SCI, this was abolished
following administration of the MCT inhibitor α-
cyano-4-hydroxycinnamic acid (4-CIN) beginning
after injury. This suggest that transport of ketone
bodies and/or lactate plays a role in the observed
protection rather than general metabolic effects
like low glucose levels or free fatty acid-mediated
events (Streijger et al., 2013). The interpretation of
this experiment, however, is complicated by the
fact that 4-CIN also inhibits the mitochondrial
pyruvate transporter (McKenna et al., 2001).
Taken together, manipulating the composi-
tion and timing of macronutrient intake can have
profound effects on the outcome from spinal cord
trauma in rodents, which questions our current,
poorly evidenced clinical nutritional guidelines
for SCI. However, our understanding of how KD
and ketone bodies mediate improved recovery
from injury is very incomplete.
K E TO G E N I C D I E T
I M P R OV E S O U T C O M E
A F T E R   T R AU M AT I C
B R A I N   I N J U RY
KD has also been shown to improve functional
outcome following TBI in adolescent (35 days old)
but not in adult rats (75 days old) (Appelberg et
al., 2009). After a controlled cortical impact injury
(CCI) and subsequent feeding of a 7:1 KD imme-
diately after injury, adolescent rats showed signifi-
cant improvements in the time to traverse a beam,
a measure of fine motor coordination and balance,
on postinjury day 3, but not on postinjury days
4–7. Additionally, on post injury day 3 the number
of foot slips was reduced in the 35-day-old KD-fed
CCI rats (Appelberg et al., 2009). In the Morris
Water Maze, 35-day-old KD-fed animals showed
significantly shorter escape latencies compared
with CCI SD-fed animals. Such motor and cogni-
tive improvements were not observed among 75-
day-old animals; this, however, might have been
attributable to increased hyperactivity, which has
been shown by others in adult rats kept on KD
(Murphy and Burnham, 2006; Ziegler et al., 2005).
In a preceding study, Prins and coworkers showed
that pretreatment with KD reduced CCI lesion
volume by 58% in 35-day-old adolescent KD-fed
rats and not in the adult KD-fed rats (Prins et al.,
2005). There were also fewer Fluoro-Jade positive
cells, indicative of better neuronal preservation.
Oddly, very young rats 17-days of age did not show
a reduction in lesion volume under these condi-
tions (Prins et al., 2005). These findings are inde-
pendently supported by Hu and coworkers, who
136
136 section II: E merging Clinical Applications and Future Potential
performed similar KD experiments on 35-day-old
TBI rats using a weight drop model (Hu et al.,
2009a: Hu et al., 2009b). They found a decreased
expression of the pro-apoptotic gene Bax (Bcl-2
Associated X Protein), together with a reduction of
several markers for apoptosis as well as less edema
(Hu et al., 2009a; Hu et al., 2009b). It is noteworthy
to recall that our KD studies in the injured spinal
cord were all performed in adult rats weighing
around 350 g at time of injury (11–13 weeks old).
POS SIBLE MECHANISMS
O F   K E TO G E N I C   D I E T S
Ketones Provide an Alternative Source
of Energy—Anaplerosis
It is well known that a physiologically low intake
of carbohydrates during consumption of KDs or
fasting typically causes hepatic ketogenesis fueled
by the beta-oxidation of fatty acids, resulting in
increased blood levels of the ketone bodies: βHB,
acetoacetate (AcAc), and acetone (Cahill, 2006).
These ketone bodies cross the blood-brain/spinal
cord barrier and enter neuronal and glial cells via
MCTs (Nijland et al., 2014; Pierre and Pellerin,
2005). Here, ketone bodies are converted into ace-
toacetyl-CoA by the enzyme succinyl-CoA:3-CoA
transferase (SCOT) and are broken down into two
molecules of Acetyl-CoA that are subsequently
used in the Krebs cycle, (see, e.g., Fukao et al., 2014;
Kim do and Rho, 2008). Interestingly, brain astro-
cytes are an additional site for ketogenesis (Auestad
et al., 1991; Guzman and Blazquez, 2001). Taken
together, KDs provide an effective source of energy
without the need for glycolysis or mitochondrial
complex-I and the associated free oxygen radical
byproducts. The brain can utilize ketones for up
to 60% of its energy demands (Owen et al., 1967).
Energy supplies are further boosted by an increase
in mitochondrial biogenesis together with an
increase of numerous energy metabolism genes in
rat hippocampi after several weeks of KD (Bough
and Rho, 2007; Bough et al., 2006). This can pro-
vide an important alternative source of energy
when the activity of pyruvate dehydrogenase is
low, as seen after both SCI and TBI (McEwen et
al., 2011; Sharma et al., 2009). Pyruvate dehydro-
genase converts pyruvate to acetyl-CoA, which is
for mitochondrial ATP production. Low activity of
this enzyme could trigger an energy crisis. Hence,
some of the benefits of KD observed after SCI or
TBI (Prins and Matsumoto, 2014) could be due to
this metabolic rescue by ketones.
A TBI triggers a transient increase and subse-
quent depression in glucose uptake accompanied
by decreased glycolysis (Prins and Matsumoto,
2014). Oxidative damage-driven activation of the
DNA repair enzyme poly ADP-ribose polymerase
(PARP) may lead to depletion of NAD+ (Besson
et al., 2003; LaPlaca et al., 1999) that would cause a
decrease in glyceraldehyde phosphate dehydroge-
nase activity. Together with an oxidative damage-
driven decrease in pyruvate dehydrogenase
activity, less glucose can be utilized for ATP gen-
eration in the TCA cycle and energy failure ensues
(Lee et al., 1999; Sharma et al., 2009; Singh et al.,
2006). Exogenously applied βHB is readily taken
up by the injured brain after a CCI, and used to
restore the depleted ATP stores (Prins et al., 2004).
Such uptake is not seen in the uninjured brain,
where energy supplies are largely glucose-derived
(Prins et  al., 2004). However, this depletion of
energy stores after contusion injury occurs more
slowly in adolescent rats than in adult animals
(Deng-Bryant et al., 2011). Administration of βHB
results in a more pronounced increase in βHB lev-
els in the brain of younger rats and also restores
their energy stores more effectively compared
with older rats. This could be explained by the fact
that young rats express higher level of MCTs for
the uptake of ketone bodies into the brain com-
pared with older rats (Prins and Giza, 2006); the
age specific differences in neuroprotection seen
with KD treatments after TBI are likely related. In
addition, young rats show an elevation of ketone
levels within 6 hours, while this takes closer to a
day in adult rats (Prins et al., 2005; Prins and Giza,
2006). This time difference might be critical, since
injury triggers an immediate secondary cascade
and the rescue of neuronal tissue after injury is
time sensitive.
This metabolic concept of KD and ketone
treatment as key players in protection is sup-
ported by the pronounced neuroprotective
effect after SCI and TBI of acetyl-L-carnitine,
which supplies an acetyl group for mitochon-
drial acetyl-CoA synthesis used in the TCA
cycle (Patel et al., 2012; Scafidi et al., 2010). Like
ketones, in the absence of pyruvate, acetyl-L-
carnitine can be used for mitochondrial respira-
tion (Patel et al., 2012; Patel et al., 2010). Direct
proof for the metabolic role of ketone bodies
could be provided with a transgenic mouse line
allowing for a cell (tissue)-specific deletion of
the mitochondrial enzyme succinyl-CoA:3-CoA
transferase (SCOTfl/fl). This will discern the met-
abolic role played by ketones as a biofuel for the
TCA cycle (anaplerosis) from the nonmetabolic
functions of ketones as inhibitors of inflamma-
tion and oxidation (see next section).
137
Chapter 16: Treatment of Traumatic Brain and Spinal Cord Injury
Non-Anaplerotic Effects of Ketogenic Diet
In addition to the aforementioned anaplerotic
effects of ketones as fuel for the TCA cycle, the
beneficial effects of KD may also be due to low-
normal glucose levels, with reduced glycolysis
and/or due to increases in free fatty acids and a
myriad of lipid signaling molecules (for review, see
Rho, 2015). For example, elevated blood glucose
levels are often observed in humans after SCI and
have been shown to correlate with less favorable
recovery (Kobayakawa et  al., 2014). Mice under
hyperglycemic conditions show exacerbation of
inflammation and secondary damage, activation
of NFκB, and inferior functional outcome com-
pared with mice in which glucose levels are kept
normal (Kobayakawa et  al., 2014). Similar con-
siderations apply to TBI in humans, in particular
when hyperglycemia is stress induced (Bosarge
et al., 2015), and there is consensus that blood glu-
cose should be carefully controlled (Badjatia et al.,
2014). However in the case of our KD experiments
in SCI rats, glucose levels were in the normal range
and not significantly different from controls, mak-
ing a low-glucose mediated effect unlikely.
Changes in circulating lipids due to KD may
have significant effects on functional outcome fol-
lowing SCI. Lipids are important constituents of
all mammalian cells and have diverse biological
functions, including as (1) major building material
for neuronal and glial membranes; (2) precursors
for various messenger molecules such as arachi-
donic acid (AA), docosahexaenoic acid (DHA),
ceramide, 1,2-diacylglycerol (DAG), phosphatidic
acid, and lyso-phosphatidic acid; and (3) energy
reservoirs such as triglycerides. Free fatty acids
increase the expression of mitochondrial uncou-
pling proteins (UCPs) via activation of peroxisome
proliferator activating receptors (PPAR) (Debril et
al., 2001; Kiec-Wilk et al., 2005). UCPs regulate the
mitochondrial transmembrane potential and pre-
vent free oxygen radical (ROS) production when
this potential gets too high. The addition to iso-
lated mitochondria of AA, docosapentaenoic acid
(DPA), eicosapentaenoic acid (EPA), palmitoleic
acid, myristic acid, and butyric acid all reduced
the mitochondrial transmembrane potential and
thereby ROS production (Davis et al., 2008b;
Sullivan et al., 2004). Both fasting and KD increase
mitochondrial uncoupling proteins which are
likely responsible for the reduction of mitochon-
drial ROS production (Davis et al., 2008a). Along
the same vein, DHA has been shown on multiple
occasions to be neuroprotective after SCI and TBI
(Michael-Titus, 2007; Michael-Titus and Priestley,
2014), although its mode of action is very broad.
137
It is therefore of significant interest to understand
how different lipid compositions in varying KD
regimens affect their efficacy after neurotrauma.
In addition, it would be interesting to see whether
KD is beneficial for myelination in the aftermath
of injury when lipids and metabolic energy are
in high demand. Demyelination is regularly seen
after SCI due to apoptosis of oligodendrocytes
(Crowe et al., 1997) and remyelination is notori-
ously sluggish in the CNS (Franklin and Ffrench-
Constant, 2008).
In addition, it is becoming increasingly clear
that ketone bodies have direct non-metabolic
actions as ligands for various receptors, transport-
ers, and regulators of enzymes. These likely con-
tribute to the beneficial effects of KDs for a wide
range of neurological disorders that are covered in
other chapters of this book. Of particular interest
in the context of SCI are the anti-excitotoxic, anti-
oxidant, and anti-inflammatory actions of ketone
bodies themselves.
Anti-Excitotoxic
The anti-excitotoxic mechanisms of KD and
ketone bodies have been extensively reviewed on
multiple occasions due to their relevance to epi-
lepsy treatment (Gano et al., 2014; Kim do et al.,
2015; Kossoff and Rho, 2009; Lutas and Yellen,
2013; Masino and Rho, 2012; Rho, 2015). Both TBI
and SCI trigger excitotoxic neuronal damage, and
a variety of drugs blocking neuronal excitation are
neuroprotective after injury (Chen et al., 2014b;
Park et al., 2004). However, clinical translation of
glutamate receptor blockers has been challenging
in contrast to the less specific excitation-damp-
ening drugs like valproic acid, used in epilepsy
(Chen et al. 2014b), or riluzole, a nonspecific
sodium channel blocker used in ALS (Wilson and
Fehlings, 2014). A clinical trial of riluzole for acute
SCI is currently under way (Fehlings et al., 2015).
However, these anti-excitotoxic treatments for SCI
have in common that they need to be adminis-
tered within hours after SCI in order to be effec-
tive. In contrast, the antiepileptogenic effect of KD
requires several weeks of treatment to manifest,
suggesting that prevention of excessive excitation
is not a major player in the neuroprotection pro-
vided by KD in the setting of acute SCI.
Antioxidant
Oxidative stress and damage occurs after SCI
and TBI (for an excellent overview, see Bains and
Hall, 2012). ROS lead to peroxynitrite formation
and lipid peroxidation, leading to perturbation of
138
138 section II: E merging Clinical Applications and Future Potential
many cellular processes including Ca2+ and Na+/
K+ pumps. These failures lead to further Ca2+ over-
load, mitochondrial damage, and secondary ROS
production. In both TBI and SCI, the byprod-
ucts of oxidative damage can be measured within
minutes of injury and both 3-nitrotyrosine and 4-
hydroxynonenal peak between 1 and 3 days. This
implies some urgency when attempting to prevent
secondary injury with an antioxidative treatment.
Indeed, in laboratory animals, most neuroprotec-
tive treatments are effective if given prior to or at
the time of injury but lose efficacy when admin-
istered after a delay of several hours (Kwon et al.,
2011). Similarly, in humans with SCI, methyl-
prednisolone, an antioxidant with glucocorticoid
actions, was only somewhat successful when given
within 8 hours post-SCI (Bracken et al., 1998).
This implies that any diet-induced benefit pro-
vided by the initiation of KD in the acute stage
of injury would likely miss much of the oxidative
damage during the first day of injury, since it takes
16–20 hours for the endogenous ketone levels to
significantly rise.
The antioxidant effects of KDs have been
extensively reviewed (Gano et al., 2014). KD
up-regulates mitochondrial glutathione (GSH)
biosynthesis in the brain (Jarrett et al., 2008),
enhances mitochondrial antioxidant status, and
reduces ROS production (Maalouf et al., 2007),
protecting mtDNA from oxidant-induced dam-
age (Jarrett et al., 2008). The exact mechanism of
this GSH increase is unknown. The GSH is syn-
thetized by glutamate cysteine ligase (GCL) form-
ing gamma-glutamylcycteine, to which glycine
is added by glutathione synthase. The activity of
GCL, which is rate limiting, is increased by KD
(Jarrett et al., 2008) and could be mediated by the
transcription factor Nrf2 that is found in nuclear
fractions of KD-fed rats (Milder et al., 2010). Nrf2
binds to antioxidant response elements (ARE) of
several genes including GCL (Milder and Patel,
2012). It is hypothesized that Nrf2 is inactivated
by forming a complex with Kelch-like ECH-asso-
ciated protein 1 under normal physiological con-
ditions. This complex functions as a redox sensor
and releases Nrf2 in situations of stress and high
ROS levels. It remains to be shown whether a
transient stress or ROS response after initiation
of KD (Milder et al., 2010) is eliciting an increase
in Nrf2 by KD after neurotrauma. Interestingly,
there is an increase in Nrf2-ARE regulated anti-
oxidant genes after TBI, but it does not occur
until 24–48 hours after injury (Miller et al., 2014),
which may be too late to prevent much second-
ary damage. Acceleration of Nrf2-ARE activation
with carnosic acid provides effective protection
when given within 15 minutes of injury, and some
benefits such as reduced cytoskeletal breakdown
are observed with application at 8 hours after
TBI (Miller et al., 2015). It remains to be shown
whether KD or administration of ketone bodies
will be able to elicit such a protective response via
this Nrf2-ARE pathway in the time-sensitive set-
ting of acute neurotrauma.
A recently recognized antioxidant function of
βHB is that of a direct endogenous inhibitor of
class histone deacetylases at concentrations that
occur with KD administration (EC50 between 2.5
and 5 mM) (Shimazu et al., 2013). Kidneys of mice
fasted for 24 hours showed increased acetylation
at the Foxo3a and metallothionin 2 (Mt2) promot-
ers due to HDAC1 inhibition by βHB. Both genes
protect against oxidative stress, whereby Foxo3a
increases the expression of mitochondrial antioxi-
dant manganese superoxide dismutase (MnSOD)
and catalase (Shimazu et al., 2013) (for review,
see Newman and Verdin, 2014). Whether a simi-
lar inhibition of class I HDACs by βHB occurs in
the brain or spinal cord and how this relates to
the known antioxidant effects of KD has yet to be
clarified. What is becoming clear, however, is that
ketones may have profound epigenetic effects that
reach far beyond anaplerosis.
Anti-Inflammatory Effects of Ketones
It is increasingly appreciated that KD has anti-
inflammatory effects in models of multiple scle-
rosis (Kim do et al., 2012), stroke (Rahman et
al., 2014), pain (Ruskin et al., 2009), and other
disorders (Dupuis et al., 2015). While part of this
may be a consequence of reduced excitotoxic-
ity and improved mitochondrial oxidation with
less ROS formation, recent research has revealed
direct effects of βHB as inflammatory regulators.
The βHB inhibits caspase 1 cleavage as part of the
NLRP3 inflammasome complex, which ultimately
dampens increases in IL1β and IL18 and contrib-
utes further to the anti-inflammatory actions of
ketones (Youm et al., 2015). The inflammasome is
a multiprotein complex involved in caspase 1 acti-
vation and cleavage of pro-IL1β and pro-IL18. It
is composed of the apoptosis-associated speck-like
protein containing a caspase recruitment domain
(CASD) termed ASC and a nucleotide oliomeriza-
tion domain (NOD)-like receptor (e.g., NLRP1
or NLRP3). In vitro, inhibition of procaspase 1
cleavage was shown in lipopolysaccharide-primed
macrophages on activation with ATP using con-
centrations of βHB in the 2- to 10-mM range as
they occur during fasting. Importantly, this effect
139
Chapter 16: Treatment of Traumatic Brain and Spinal Cord Injury
was not seen with acetoacetate or the related
butryrate and was specific to βHB (Youm et al.,
2015). The authors tested a battery of other NLRP3
activators, including the fatty acids palmitate,
ceramide, and sphingosine and obtained similar
results (i.e., inhibition by βHB). Priming macro-
phages with Toll-like agonists revealed a potent
inhibition of Pro-IL-1β cleavage by βHB treatment
at similar concentrations. These βHB effects were
specific to the NLRP3 inflammasome, and accord-
ingly, the formation of ASC oligomers, aka specks,
in primed and ATP-activated macrophages was
greatly reduced with βHB addition. In vivo, βHB
reduced peritoneal infiltration by neutrophils and
IL1β-secreting macrophages following i.p. injec-
tion of uric acid crystals in mice.
Mice carrying a Nlrp3 mutation that leads to
NLRP3 inflammasome activation show severe
peritoneal neutrophilia, which is greatly reduced
by feeding of oral ketone esters. Thus, even with
levels in the order of 1mM, βHB is an effective
and specific inhibitor the NLRP3 inflammasome.
Importantly, these actions are independent of the
presence of the GPR109a (HCA2) receptor (see
below), UCP2, and the metabolic function of
βHB in the TCA cycle. SCI leads to a pronounced
increase of the NLRP3 inflammasome, including
the partnering protein ASC (both mainly in acti-
vated microglia) and the well-known activation of
caspase-1 and pro-IL1β cleavage (Zendedel et al.,
2015). Increased ASC expression, caspase-1 acti-
vation, and cleavage of pro-IL1β had previously
been described in a moderate cervical spinal cord
contusion model at the C5 level in rats (de Rivero
Vaccari et al., 2008). However, these authors did
not find the NLRP3 increase but detected NLRP1
in co-immunoprecipitations with ASC with-
out increases in NLRP1 expression. These dif-
ferences might be due to differences in the SCI
models used. Importantly, antibody-mediated
ASC neutralization reduced caspase-1 activation
and cleavage of IL1β and IL18, reducing lesion
volume and improving functional recovery in
multiple forelimb tests (i.e., grip strength, sticker
removal, gait analysis), underscoring the impor-
tance of inflammasome activation in secondary
injury cascades after SCI. Whether KD or oral
ketone esters are mediating their benefits in the
injured spinal cord via this mechanism remains
unclear.
An additional way to directly modify inflam-
mation by ketones is through activation of the
hydroxyl carboxylic acid (HCA2) receptor also
known as GPR109A, PUMA-G or niacin recep-
tor. This receptor has received much attention in
139
the past due to its favorable effects on blood lipid
levels/composition and anti-inflammatory actions
deemed responsible for cardiovascular benefits
(Gille et al., 2008; Lukasova et al., 2011). However,
βHB is the main known endogenous ligand of the
HCA2 receptor at concentrations that are easily
reached by the KD (EC50 0.7 mM) (Taggart et al.,
2005). While HCA2 is widely expressed in adi-
pose tissue, where it regulates lipolysis (Taggart et
al., 2005), it is also found on immune cells such
as neutrophils, macrophages, and microglia but
not in astrocytes or neurons (Kostylina et al.,
2008; Rahman et al., 2014). HCA2 receptor acti-
vation promotes neutrophil apoptosis (Kostylina
et al., 2008), and on macrophages/monocytes it
activates/attracts a neuroprotective Ly-6Lo subset
(Rahman et al., 2014). The latter study is of partic-
ular interest, showing that the beneficial effects of
KD or niacin in a model of stroke by middle cere-
bral artery occlusion (MCAO) in mice were lost
when these mice were lacking the HCA2 receptor
(HCA2–/–). Transplantation of bone marrow from
HCA+/+ wild-type into the HCA–/– mice restored
this rescue effect, demonstrating that monocyte/
macrophage HCA2 expression is required. It was
also necessary since the inversion of this trans-
plantation paradigm (i.e., HCA–/– bone marrow
into wild-type mice) abolished the neuroprotec-
tive effect. This beneficial effect of niacin (and,
by inference, βHB) on stroke outcome requires
Cox1 and hematopoietic prostaglandin D2 synth-
ase (Rahman et al., 2014). The HCA2 receptor is
also activated by dimethylfumarate (DMF), a drug
used clinically for MS. The HCA2 receptors have
been shown to mediate the inhibition of DMF on
neutrophil adhesion and migration in an inflam-
matory mouse model of MS (Chen et al., 2014a).
Interestingly, prostaglandin D2 and its PD1 recep-
tor have been implicated in the mitigation of dam-
age after excitotoxic challenge or stroke (Dore and
Shafique Ahmad, 2015) and to dampen seizures
(Kaushik et al., 2014). Taken together, these data
indicate that βHB at concentrations measured
during fasting or KD treatment can modify the
molecular phenotype of blood monocytes and
macrophages toward an anti-inflammatory (M2-
like) tissue protective response via direct activation
of the HCA2 receptor. Since neutrophil invasion
and M1 macrophage polarization play important
damaging roles after SCI (for review, see Donnelly
and Popovich, 2008; Kigerl et al., 2009; Popovich,
2014), it is conceivable that the protective effect of
the KD is mediated at least in part by these HCA2
receptor triggered mechanisms. This hypothesis
remains to be tested, and in light of an arsenal of
140
140 section II: E merging Clinical Applications and Future Potential
HCA2 agonists developed to treat dyslipidemias
and atherosclerosis, it is becoming increasingly
attractive.
Translational Consideration of Ketogenic
Diet in the Clinical Setting
KD and ketone bodies have powerful anti-excito-
toxic, antiepileptic and anti-inflammatory effects
that go far beyond their classic function as an
energy-providing metabolic boost and provide
neuroprotection by multiple mechanisms. One
translational challenge lies in the timing of KD
administration, since secondary injury cascades
begin within minutes of neurotrauma. As a matter
of fact, most neuroprotective treatments are most
effective when given prior to injury and lose their
effects when administered more than 1–3 hours
after injury (Kwon et al., 2011). However, a dietary
intervention with KD does not increase ketone body
levels until the end of the first day—in particular
in the context of injury stress as it occurs after SCI
(Streijger et al., 2013). In this regard, the neuropro-
tective effects of KD after SCI represent one of the
more promising interventions currently available.
While this may be encouraging, we hypothesize that
in the acute setting of neurotrauma, a faster increase
in ketone bodies would be greatly advantageous. A
carefully monitored direct infusion of βHB for the
first days after trauma in parallel with a KD could be
one avenue. An alternative approach has been envi-
sioned by Richard Veech and colleagues by suggest-
ing the use of oral ketone-ester supplementation
after TBI (Veech et al., 2012). In vivo oral adminis-
tration of ketone esters readily increases βHB levels
to the desired range of 3–4 mmol/L and has been
shown to improve cognition and histopathology in
3xTgAD mice that are a model for Alzheimer’s dis-
ease (Kashiwaya et al., 2013). Whether or not such
supplementation with oral ketone esters after acute
TBI and SCI is best combined with KD or will be
effective by itself as a “diet in a bottle” remains to be
determined.
Our clinical colleagues in Guangzhou (China)
recently conducted a clinical trial of KD in patients
with acute SCI to evaluate its safety and feasibility
(Guo et al., 2014). These investigators introduced
KD treatment in the acute phase after injury in 10
patients with SCI. Comparable to what was seen
in the rodent models of SCI (Streijger et al., 2013),
all patients developed ketone levels above 2 mM
(typically ~ 3 mmol/L) while maintaining normal
glucose levels (9 out of 10). Routine blood tests for
electrolytes and liver and kidney function showed
no changes. This demonstrates that diet-induced
ketosis in patients with SCI is feasible and that the
KD is safe when initiated after SCI in the hospital
setting. Yet, there was some transient gastrointes-
tinal discomfort (diarrhea, nausea, poor appetite,
gastric pain, and abdominal distension) in five
patients and a low enthusiasm for the Chinese KD
formulation. This underscores the translatability of
this dietary approach, but strongly motivates us to
improve this metabolic treatment in order to mini-
mize unpleasant side effects and increase compli-
ance to be able to implement a 2- to 3-month-long
regimen after SCI in a human clinical trial.
For the same reason, more “liberal” diets
have been explored in the epilepsy field with
the hope for similar benefits. These alterna-
tives share restriction in the amount/type of
carbohydrates and vary in the lipid and protein
content. They include a medium chain triglyc-
eride enriched KD diet (MCT-KD), a low glyce-
mic index treatment (LGIT) and the “modified
Atkins diet” (MAD) (Miranda et al., 2012). The
latter alternative has been the most extensively
studied and was in its original form introduced
by Dr. Atkins in the 1970s. However ketosis with
the original Atkins is variable and in order to
mimic the ketosis of classical KD the amount of
carbohydrates are strictly reduced in the MAD to
less than 20 g per day (in adults) while the intake
in protein is not restricted (unlike in KD). While
still ketogenic, this MAD results in low-normal
blood glucose levels due to gluconeogenesis from
proteins. Compared with the 3:1 ratio (3 g fat to
1 g of protein+carbohydrate) in KD, the ratio in
MAD is ~1:1 whereby the carbohydrates provide
<6% of the total calories, protein ~25%–35% and
the lipids ~60%–70% of the total calories; per
gram, fat has more than twice as many calories
as protein or carbohydrates. Thus the MAD is
much easier to implement and better tolerated
by the patients than the 3:1 KD, where <3% of
the calories stem from carbohydrates, ~11%
from protein and ~88% from fat. MAD appears
almost as effective as KD as long as the carbo-
hydrate intake is strictly controlled (Sharma et
al., 2013; for review, see Auvin, 2012; Kossoff,
2011, 2013; Miranda et al., 2012). A meta-anal-
ysis of close to 20 studies indicates that a 50%
rate of seizure reduction with MAD is nearly as
good as the 56% rate with classical KD (Kossoff
et al., 2013). Whether MAD is as effective as KD
in the SCI setting is worth further investigation.
It would certainly make a huge difference for
individuals with neurotrauma and most likely
increase their compliance, if we could trial a
MAD instead of a classical KD, even when con-
ducting a dietary treatment for only 2–3 months,
141
Chapter 16: Treatment of Traumatic Brain and Spinal Cord Injury
which is the presently anticipated treatment
duration after SCI.
We hope to close the “translational gap” from
bench to bedside by collecting the necessary data
to better understand the underlying mechanisms
responsible for the beneficial effects of KDs in the
setting of SCI and by conducting the necessary
preclinical experiments to decide in what exact
form the benefits of the KD and ketone bodies
could be used for clinical trials in humans with
acute SCI and TBI.
AC K N OW L E D G M E N T S
The authors thank Brett Hilton for critically read-
ing this manuscript; he is supported by a Banting
and Best Scholarship. Wolfram Tetzlaff holds the
John and Penny Ryan British Columbia Leadership
Chair in Spinal Cord Injury; the research in his
laboratory is supported by the Canadian Institute
for Health Research, the Multiple Sclerosis Society
of Canada, Wings for Life, the International Spinal
Research Trust, Craig H.  Nielsen Foundation,
and the Rick Hansen Institute and Rick Hansen
Foundation.
REFERENCES
Anderson, K.D. (2004). Targeting recovery:  pri-
orities of the spinal cord-injured population. J
Neurotrauma 21, 1371–1383.
Appelberg, K.S., Hovda, D.A., and Prins, M.L. (2009).
The effects of a ketogenic diet on behavioral out-
come after controlled cortical impact injury in
the juvenile and adult rat. J Neurotrauma 26,
497–506.
Auestad, N., Korsak, R.A., Morrow, J.W., and Edmond,
J. (1991). Fatty acid oxidation and ketogenesis by
astrocytes in primary culture. J Neurochem 56,
1376–1386.
Austin, J.W., Kang, C.E., Baumann, M.D., DiDiodato,
L., Satkunendrarajah, K., Wilson, J.R., Stanisz, G.J.,
Shoichet, M.S., and Fehlings, M.G. (2012). The
effects of intrathecal injection of a hyaluronan-
based hydrogel on inflammation, scarring and
neurobehavioural outcomes in a rat model of
severe spinal cord injury associated with arach-
noiditis. Biomaterials 33, 4555–4564.
Auvin, S. (2012). Should we routinely use modified
Atkins diet instead of regular ketogenic diet to
treat children with epilepsy? Seizure 21, 237–240.
Badjatia, N., Vespa, P., and Participants of the
International Multi-disciplinary Consensus
Conference on Multimodality, M. (2014).
Monitoring nutrition and glucose in acute brain
injury. Neurocrit Care 21(Suppl 2), S159–167.
Bains, M., and Hall, E.D. (2012). Antioxidant therapies
in traumatic brain and spinal cord injury. Biochim
Biophys Acta 1822, 675–684.
141
Besson, V.C., Croci, N., Boulu, R.G., Plotkine, M.,
and Marchand-Verrecchia, C. (2003). Deleterious
poly(ADP-ribose)polymerase-1 pathway activa-
tion in traumatic brain injury in rat. Brain Res
989, 58–66.
Bosarge, P.L., Shoultz, T.H., Griffin, R.L., and Kerby,
J.D. (2015). Stress-induced hyperglycemia is
associated with higher mortality in severe trau-
matic brain injury. J Trauma Acute Care Surg 79,
289–294.
Bough, K.J., and Rho, J.M. (2007). Anticonvulsant
mechanisms of the ketogenic diet. Epilepsia
48, 43–58.
Bough, K.J., Wetherington, J., Hassel, B., Pare, J.F.,
Gawryluk, J.W., Greene, J.G., Shaw, R., Smith,
Y., Geiger, J.D., and Dingledine, R.J. (2006).
Mitochondrial biogenesis in the anticonvulsant
mechanism of the ketogenic diet. Ann Neurol 60,
223–235.
Bracken, M.B., Shepard, M.J., Holford, T.R., Leo-
Summers, L., Aldrich, E.F., Fazl, M., Fehlings,
M.G., Herr, D.L., Hitchon, P.W., Marshall, L.F.,
et  al. (1998). Methylprednisolone or tirilazad
mesylate administration after acute spinal cord
injury:  1-year follow up; results of the third
National Acute Spinal Cord Injury randomized
controlled trial. J Neurosurg 89, 699–706.
Brandhorst, S., Choi, I.Y., Wei, M., Cheng, C.W.,
Sedrakyan, S., Navarrete, G., Dubeau, L., Yap, L.P.,
Park, R., Vinciguerra, M., et al. (2015). A periodic
diet that mimics fasting promotes multi-system
regeneration, enhanced cognitive performance,
and healthspan. Cell Metabol 22, 86–99.
Cahill, G.F., Jr. (2006). Fuel metabolism in starvation.
Annu Rev Nutr 26, 1–22.
Chen, H., Assmann, J.C., Krenz, A., Rahman, M.,
Grimm, M., Karsten, C.M., Kohl, J., Offermanns,
S., Wettschureck, N., and Schwaninger, M.
(2014a). Hydroxycarboxylic acid receptor 2 medi-
ates dimethyl fumarate’s protective effect in EAE. J
Clin Invest 124, 2188–2192.
Chen, S., Wu, H., Klebe, D., Hong, Y., and Zhang, J.
(2014b). Valproic acid: a new candidate of thera-
peutic application for the acute central nervous
system injuries. Neurochem Res 39, 1621–1633.
Cheng, B., Yang, X., An, L., Gao, B., Liu, X., and Liu,
S. (2009). Ketogenic diet protects dopaminergic
neurons against 6-OHDA neurotoxicity via up-
regulating glutathione in a rat model of Parkinson’s
disease. Brain Res 1286, 25–31.
Consortium-for-Spinal-Cord-Medicine. (2006). Early
Acute Management in Adults with Spinal Cord
Injury. http://wwwnxtbookcom/nxtbooks/pva/
earlyacutemanagement/indexphp
Cook, A.M., Peppard, A., and Magnuson, B. (2008).
Nutrition considerations in traumatic brain injury.
Nutr Clin Pract 23, 608–620.
Crowe, M.J., Bresnahan, J.C., Shuman, S.L., Masters,
J.N., and Beattie, M.S. (1997). Apoptosis and
142
142 section II: E merging Clinical Applications and Future Potential
delayed degeneration after spinal cord injury in
rats and monkeys. Nat Med 3, 73–76.
Davis, L.M., Pauly, J.R., Readnower, R.D., Rho, J.M.,
and Sullivan, P.G. (2008a). Fasting is neuroprotec-
tive following traumatic brain injury. Journal of
Neuroscience Research 86, 1812–1822.
Davis, L.M., Rho, J.M., and Sullivan, P.G. (2008b). UCP-
mediated free fatty acid uncoupling of isolated
cortical mitochondria from fasted animals: corre-
lations to dietary modulations. Epilepsia 49(Suppl
8), 117–119.
de Rivero Vaccari, J.P., Lotocki, G., Marcillo, A.E.,
Dietrich, W.D., and Keane, R.W. (2008). A
molecular platform in neurons regulates inflam-
mation after spinal cord injury. J Neurosci 28,
3404–3414.
Debril, M.B., Renaud, J.P., Fajas, L., and Auwerx, J.
(2001). The pleiotropic functions of peroxisome
proliferator-activated receptor gamma. J Mol Med
79, 30–47.
Deng-Bryant, Y., Prins, M.L., Hovda, D.A., and Harris,
N.G. (2011). Ketogenic diet prevents alterations
in brain metabolism in young but not adult rats
after traumatic brain injury. J Neurotrauma 28,
1813–1825.
Donnelly, D.J., and Popovich, P.G. (2008).
Inflammation and its role in neuroprotection, axo-
nal regeneration and functional recovery after spi-
nal cord injury. Exp. Neurol. 209, 378–388.
Dore, S., and Shafique Ahmad, A. (2015). Cytoprotective
role of prostaglandin D2 DP1 receptor against neu-
ronal injury following acute excitotoxicity and cere-
bral ischemia. In Brain Neurotrauma:  Molecular,
Neuropsychological, and Rehabilitation Aspects,
F.H. Kobeissy, ed. (CRC Press/Taylor & Francis,
Boca Raton, FL).
Dupuis, N., Curatolo, N., Benoist, J.F., and Auvin, S.
(2015). Ketogenic diet exhibits anti-inflammatory
properties. Epilepsia 56, e95–98.
Evaniew, N., Belley-Cote, E.P., Fallah, N., Noonan,
V., Rivers, C.S., and Dvorak, M.F. (2015).
Methylprednisolone for the treatment of patients
with acute spinal cord injuries:  A  systematic
review and meta-analysis. J Neurotrauma. Mar
1;33(5):468–81
Fehlings, M.G., Vaccaro, A., Wilson, J.R., Singh, A., D,
W.C., Harrop, J.S., Aarabi, B., Shaffrey, C., Dvorak,
M., Fisher, C., et  al. (2012). Early versus delayed
decompression for traumatic cervical spinal cord
injury:  results of the Surgical Timing in Acute
Spinal Cord Injury Study (STASCIS). PLoS One 7
(2), e32037.
Fehlings, M.G., Nakashima, H., Nagoshi, N., Chow,
D.S., Grossman, R.G., and Kopjar, B. (2015).
Rationale, design and critical end points for
the Riluzole in Acute Spinal Cord Injury Study
(RISCIS): a randomized, double-blinded, placebo-
controlled parallel multi-center trial. Spinal Cord.
2016 Jan;54(1):8–15
Franklin, R.J., and Ffrench-Constant, C. (2008).
Remyelination in the CNS: from biology to ther-
apy. Nat Rev Neurosci 9, 839–855.
Fukao, T., Mitchell, G., Sass, J.O., Hori, T., Orii, K., and
Aoyama, Y. (2014). Ketone body metabolism and
its defects. J. Inherit. Metab. Dis. 37, 541–551.
Furlan, J.C., Noonan, V., Cadotte, D.W., and Fehlings,
M.G. (2011). Timing of decompressive surgery of
spinal cord after traumatic spinal cord injury: an
evidence-based examination of pre-clinical and
clinical studies. J Neurotrauma 28, 1371–1399.
Gano, L.B., Patel, M., and Rho, J.M. (2014). Ketogenic
diets, mitochondria, and neurological diseases. J
Lipid Res 55, 2211–2228.
Gasior, M., Rogawski, M.A., and Hartman, A.L. (2006).
Neuroprotective and disease-modifying effects of
the ketogenic diet. Behav Pharmacol 17, 431–439.
Gibson, C.L., Murphy, A.N., and Murphy, S.P. (2012).
Stroke outcome in the ketogenic state: a systematic
review of the animal data. J Neurochem 123 (Suppl
2), 52–57.
Gille, A., Bodor, E.T., Ahmed, K., and Offermanns,
S. (2008). Nicotinic acid:  pharmacological effects
and mechanisms of action. Annu Rev Pharmacol
Toxicol 48, 79–106.
Guo, C., Zhou, J., Wu, X., Jiang, H., Lu, K., Chen, J.,
Wu, Z., Yu, R., Liu, J., and Zhu, Q. (2014). [A clini-
cal trial of ketogenic diet in patients with acute spi-
nal cord injury:  safety and feasibility]. Nan Fang
Yi Ke Da Xue Xue Bao = J Southern Med Univ 34,
571–575.
Guzman, M., and Blazquez, C. (2001). Is there an
astrocyte-neuron ketone body shuttle? Trends
Endocrinol Metabol: TEM 12, 169–173.
Halestrap, A.P., and Wilson, M.C. (2012). The mono-
carboxylate transporter family:  role and regula-
tion. IUBMB Life 64, 109–119.
Harkema, S., Behrman, A., and Barbeau, H. (2012a).
Evidence-based therapy for recovery of function
after spinal cord injury. Handb Clin Neurol 109,
259–274.
Harkema, S.J., Hillyer, J., Schmidt-Read, M., Ardolino,
E., Sisto, S.A., and Behrman, A.L. (2012b).
Locomotor training: as a treatment of spinal cord
injury and in the progression of neurologic reha-
bilitation. Arch Phys Med Rehabil 93, 1588–1597.
Hertz, L., Chen, Y., and Waagepetersen, H.S. (2015).
Effects of ketone bodies in Alzheimer’s disease
in relation to neural hypometabolism, beta-
amyloid toxicity, and astrocyte function. Journal
of Neurochemistry 134, 7–20.
Hu, Z.G., Wang, H.D., Jin, W., and Yin, H.X. (2009a).
Ketogenic diet reduces cytochrome c release and
cellular apoptosis following traumatic brain injury
in juvenile rats. Ann Clin Lab Sci 39, 76–83.
Hu, Z.G., Wang, H.D., Qiao, L., Yan, W., Tan, Q.F., and
Yin, H.X. (2009b). The protective effect of the keto-
genic diet on traumatic brain injury-induced cell
death in juvenile rats. Brain Injury 23, 459–465.
143
Chapter 16: Treatment of Traumatic Brain and Spinal Cord Injury
Imamura, K., Takeshima, T., Kashiwaya, Y.,
Nakaso, K., and Nakashima, K. (2006). D-beta-
hydroxybutyrate protects dopaminergic SH-SY5Y
cells in a rotenone model of Parkinson’s disease. J
Neurosci Res 84, 1376–1384.
Jarrett, S.G., Milder, J.B., Liang, L.P., and Patel, M.
(2008). The ketogenic diet increases mitochondrial
glutathione levels. J Neurochem 106, 1044–1051.
Jeong, M.A., Plunet, W., Streijger, F., Lee, J.H., Plemel,
J.R., Park, S., Lam, C.K., Liu, J., and Tetzlaff, W.
(2011). Intermittent fasting improves functional
recovery after rat thoracic contusion spinal cord
injury. J Neurotrauma 28, 479–492.
Kashiwaya, Y., Takeshima, T., Mori, N., Nakashima,
K., Clarke, K., and Veech, R.L. (2000). D-beta-
hydroxybutyrate protects neurons in models of
Alzheimer’s and Parkinson’s disease. Proc Natl
Acad Sci USA 97, 5440–5444.
Kashiwaya, Y., Bergman, C., Lee, J.H., Wan, R., King,
M.T., Mughal, M.R., Okun, E., Clarke, K., Mattson,
M.P., and Veech, R.L. (2013). A ketone ester diet
exhibits anxiolytic and cognition-sparing proper-
ties, and lessens amyloid and tau pathologies in a
mouse model of Alzheimer’s disease. Neurobiol
Aging 34, 1530–1539.
Kaushik, M.K., Aritake, K., Kamauchi, S., Hayaishi,
O., Huang, Z.L., Lazarus, M., and Urade, Y.
(2014). Prostaglandin D(2) is crucial for seizure
suppression and postictal sleep. Exp Neurol
253, 82–90.
Khalatbary, A.R. (2014). Natural polyphenols and spi-
nal cord injury. Iran Biomed J 18, 120–129.
Kiec-Wilk, B., Dembinska-Kiec, A., Olszanecka, A.,
Bodzioch, M., and Kawecka-Jaszcz, K. (2005). The
selected pathophysiological aspects of PPARs acti-
vation. J Physiol Pharmacol 56, 149–162.
Kigerl, K.A., Gensel, J.C., Ankeny, D.P., Alexander,
J.K., Donnelly, D.J., and Popovich, P.G. (2009).
Identification of two distinct macrophage subsets
with divergent effects causing either neurotoxicity
or regeneration in the injured mouse spinal cord. J
Neurosci 29, 13435–13444.
Kim, D., Zai, L., Liang, P., Schaffling, C., Ahlborn, D.,
and Benowitz, L.I. (2013). Inosine enhances axon
sprouting and motor recovery after spinal cord
injury. PloS One 8, e81948.
Kim do, Y., and Rho, J.M. (2008). The ketogenic diet
and epilepsy. Curr Opin Clin Nutr Metab Care 11,
113–120.
Kim do, Y., Hao, J., Liu, R., Turner, G., Shi, F.D., and
Rho, J.M. (2012). Inflammation-mediated mem-
ory dysfunction and effects of a ketogenic diet in
a murine model of multiple sclerosis. PloS One 7,
e35476.
Kim do, Y., Simeone, K.A., Simeone, T.A., Pandya, J.D.,
Wilke, J.C., Ahn, Y., Geddes, J.W., Sullivan, P.G.,
and Rho, J.M. (2015). Ketone bodies mediate anti-
seizure effects through mitochondrial permeabil-
ity transition. Ann Neurol 78, 77–87.
143
King, V.R., Huang, W.L., Dyall, S.C., Curran, O.E.,
Priestley, J.V., and Michael-Titus, A.T. (2006).
Omega-3 fatty acids improve recovery, whereas
omega-6 fatty acids worsen outcome, after spi-
nal cord injury in the adult rat. J Neurosci 26,
4672–4680.
Kobayakawa, K., Kumamaru, H., Saiwai, H., Kubota,
K., Ohkawa, Y., Kishimoto, J., Yokota, K., Ideta, R.,
Shiba, K., Tozaki-Saitoh, H., et  al. (2014). Acute
hyperglycemia impairs functional improvement
after spinal cord injury in mice and humans. Sci
Transl Med 6, 256ra137.
Kossoff, E. (2011). Ketogenic Diets:  Treatments
for Epilepsy and Other Disorders, 5th edition
(New York: Demos Health).
Kossoff, E.H. (2013). Nonpharmacological
approaches:  diet and neurostimulation. Handb
Clin Neurol 111, 803–808.
Kossoff, E.H., and Rho, J.M. (2009). Ketogenic
diets:  evidence for short- and long-term efficacy.
Neurotherapeutics 6, 406–414.
Kossoff, E.H., Cervenka, M.C., Henry, B.J., Haney,
C.A., and Turner, Z. (2013). A decade of the modi-
fied Atkins diet (2003–2013): Results, insights, and
future directions. Epilepsy Behav 29, 437–442.
Kostylina, G., Simon, D., Fey, M.F., Yousefi, S., and
Simon, H.U. (2008). Neutrophil apoptosis medi-
ated by nicotinic acid receptors (GPR109A). Cell
Death Differ 15, 134–142.
Krueger, H., Noonan, V.K., Trenaman, L.M., Joshi, P.,
and Rivers, C.S. (2013). The economic burden of
traumatic spinal cord injury in Canada. Chronic
Dis Inj Can 33, 113–122.
Kwon, B.K., Okon, E., Hillyer, J., Mann, C., Baptiste,
D., Weaver, L.C., Fehlings, M.G., and Tetzlaff, W.
(2011). A systematic review of non-invasive phar-
macologic neuroprotective treatments for acute
spinal cord injury. J Neurotrauma 28, 1545–1588.
LaPlaca, M.C., Raghupathi, R., Verma, A., Pieper, A.A.,
Saatman, K.E., Snyder, S.H., and McIntosh, T.K.
(1999). Temporal patterns of poly(ADP-ribose)
polymerase activation in the cortex following
experimental brain injury in the rat. J Neurochem
73, 205–213.
Lee, J.H., Streijger, F., Tigchelaar, S., Maloon, M., Liu,
J., Tetzlaff, W., and Kwon, B.K. (2012). A contu-
sive model of unilateral cervical spinal cord injury
using the infinite horizon impactor. J Vis Exp. Jul
24;(65). pii: 3313
Lee, S.M., Wong, M.D., Samii, A., and Hovda, D.A.
(1999). Evidence for energy failure following irre-
versible traumatic brain injury. Ann N Y Acad Sci
893, 337–340.
Leino, R.L., Gerhart, D.Z., Duelli, R., Enerson, B.E.,
and Drewes, L.R. (2001). Diet-induced ketosis
increases monocarboxylate transporter (MCT1)
levels in rat brain. Neurochem Int 38, 519–527.
Lim, S., Chesser, A.S., Grima, J.C., Rappold, P.M.,
Blum, D., Przedborski, S., and Tieu, K. (2011).
144
144 section II: E merging Clinical Applications and Future Potential
D-beta-hydroxybutyrate is protective in mouse
models of Huntington’s disease. PloS One 6,
e24620.
Longo, V.D., and Mattson, M.P. (2014). Fasting: molec-
ular mechanisms and clinical applications. Cell
Metab 19, 181–192.
Lukasova, M., Malaval, C., Gille, A., Kero, J., and
Offermanns, S. (2011). Nicotinic acid inhibits
progression of atherosclerosis in mice through its
receptor GPR109A expressed by immune cells. J
Clin Invest 121, 1163–1173.
Lutas, A., and Yellen, G. (2013). The ketogenic
diet: metabolic influences on brain excitability and
epilepsy. Trends Neurosci 36, 32–40.
Maalouf, M., Rho, J.M., and Mattson, M.P. (2009). The
neuroprotective properties of calorie restriction,
the ketogenic diet, and ketone bodies. Brain Res
Rev 59, 293–315.
Maalouf, M., Sullivan, P.G., Davis, L., Kim, D.Y., and
Rho, J.M. (2007). Ketones inhibit mitochondrial
production of reactive oxygen species production
following glutamate excitotoxicity by increasing
NADH oxidation. Neuroscience 145, 256–264.
Magnuson, B., Peppard, A., and Auer Flomenhoft, D.
(2011). Hypocaloric considerations in patients
with potentially hypometabolic disease states.
Nutr Clin Pract 26, 253–260.
Masino, S.A., and Rho, J.M. (2012). Mechanisms of
ketogenic diet action. In Jasper’s Basic Mechanisms
of the Epilepsies, J.L. Noebels, M. Avoli, M.A.
Rogawski, R.W. Olsen, and A.V. Delgado-
Escueta, eds. (National Center for Biotechnology
Information (US), Bethesda, MD).
McEwen, M.L., Sullivan, P.G., Rabchevsky, A.G., and
Springer, J.E. (2011). Targeting mitochondrial
function for the treatment of acute spinal cord
injury. Neurotherapeutics 8, 168–179.
McKenna, M.C., Hopkins, I.B., and Carey, A. (2001).
Alpha-cyano-4-hydroxycinnamate decreases both
glucose and lactate metabolism in neurons and
astrocytes:  implications for lactate as an energy
substrate for neurons. J Neurosci Res 66, 747–754.
Michael-Titus, A.T. (2007). Omega-3 fatty acids and
neurological injury. Prostaglandins Leukot Essent
Fatty Acids 77, 295–300.
Michael-Titus, A.T., and Priestley, J.V. (2014).
Omega-3 fatty acids and traumatic neurological
injury:  from neuroprotection to neuroplasticity?
Trends Neurosci 37, 30–38.
Milder, J., and Patel, M. (2012). Modulation of oxida-
tive stress and mitochondrial function by the keto-
genic diet. Epilepsy Res 100, 295–303.
Milder, J.B., Liang, L.P., and Patel, M. (2010). Acute
oxidative stress and systemic Nrf2 activation by
the ketogenic diet. Neurobiol Dis 40, 238–244.
Miller, D.M., Singh, I.N., Wang, J.A., and Hall, E.D.
(2015). Nrf2-ARE activator carnosic acid decreases
mitochondrial dysfunction, oxidative damage
and neuronal cytoskeletal degradation following
traumatic brain injury in mice. Exp Neurol 264,
103–110.
Miller, D.M., Wang, J.A., Buchanan, A.K., and Hall,
E.D. (2014). Temporal and spatial dynamics of
nrf2-antioxidant response elements mediated gene
targets in cortex and hippocampus after controlled
cortical impact traumatic brain injury in mice. J
Neurotrauma 31, 1194–1201.
Miranda, M.J., Turner, Z., and Magrath, G. (2012).
Alternative diets to the classical ketogenic diet: can
we be more liberal? Epilepsy Res 100, 278–285.
Murphy, P., and Burnham, W.M. (2006). The ketogenic
diet causes a reversible decrease in activity level in
Long-Evans rats. Exp. Neurol. 201, 84–89.
Newman, J.C., and Verdin, E. (2014). Ketone bodies
as signaling metabolites. Trends Endocrinol Metab
25, 42–52.
Nijland, P.G., Michailidou, I., Witte, M.E., Mizee, M.R.,
van der Pol, S.M., van Het Hof, B., Reijerkerk, A.,
Pellerin, L., van der Valk, P., de Vries, H.E., et al.
(2014). Cellular distribution of glucose and mono-
carboxylate transporters in human brain white
matter and multiple sclerosis lesions. Glia 62,
1125–1141.
Owen, O.E., Morgan, A.P., Kemp, H.G., Sullivan,
J.M., Herrera, M.G., and Cahill, G.F., Jr. (1967).
Brain metabolism during fasting. J Clin Invest 46,
1589–1595.
Oyinbo, C.A. (2011). Secondary injury mechanisms in
traumatic spinal cord injury: a nugget of this mul-
tiply cascade. Acta Neurobiol Exp 71, 281–299.
Pallier, P.N., Poddighe, L., Zbarsky, V., Kostusiak,
M., Choudhury, R., Hart, T., Burguillos, M.A.,
Musbahi, O., Groenendijk, M., Sijben, J.W., et  al.
(2015). A nutrient combination designed to
enhance synapse formation and function improves
outcome in experimental spinal cord injury.
Neurobiol Dis 82, 504–515.
Paoli, A., Bianco, A., Damiani, E., and Bosco, G. (2014).
Ketogenic diet in neuromuscular and neurodegen-
erative diseases. BioMed Res Int 2014, 474296.
Park, E., Velumian, A.A., and Fehlings, M.G. (2004).
The role of excitotoxicity in secondary mechanisms
of spinal cord injury:  a review with an emphasis
on the implications for white matter degeneration.
Journal of Neurotrauma 21, 754–774.
Patel, S.P., Sullivan, P.G., Lyttle, T.S., and Rabchevsky,
A.G. (2010). Acetyl-L-carnitine ameliorates mito-
chondrial dysfunction following contusion spinal
cord injury. J Neurochem 114, 291–301.
Patel, S.P., Sullivan, P.G., Lyttle, T.S., Magnuson, D.S.,
and Rabchevsky, A.G. (2012). Acetyl-L-carnitine
treatment following spinal cord injury improves
mitochondrial function correlated with remark-
able tissue sparing and functional recovery.
Neuroscience 210, 296–307.
Patel, S.P., Sullivan, P.G., Pandya, J.D., Goldstein, G.A.,
VanRooyen, J.L., Yonutas, H.M., Eldahan, K.C.,
Morehouse, J., Magnuson, D.S., and Rabchevsky,
145
Chapter 16: Treatment of Traumatic Brain and Spinal Cord Injury
A.G. (2014). N-acetylcysteine amide preserves
mitochondrial bioenergetics and improves func-
tional recovery following spinal trauma. Exp
Neurol 257, 95–105.
Pierre, K., Magistretti, P.J., and Pellerin, L. (2002).
MCT2 is a major neuronal monocarboxylate
transporter in the adult mouse brain. J Cereb
Blood Flow Metab 22, 586–595.
Pierre, K., and Pellerin, L. (2005). Monocarboxylate
transporters in the central nervous system:  dis-
tribution, regulation and function. J Neurochem
94, 1–14.
Plunet, W.T., Lam, C.K., Lee, J.H., Liu, J., and Tetzlaff,
W. (2010). Prophylactic dietary restriction may
promote functional recovery and increase lifespan
after spinal cord injury. Ann N Y Acad Sci 1198
Suppl 1, E1–11.
Plunet, W.T., Streijger, F., Lam, C.K., Lee, J.H., Liu, J.,
and Tetzlaff, W. (2008). Dietary restriction started
after spinal cord injury improves functional recov-
ery. Exp Neurol 213, 28–35.
Popovich, P.G. (2014). Neuroimmunology of trau-
matic spinal cord injury: a brief history and over-
view. Exp Neurol 258, 1–4.
Prins, M.L., Fujima, L.S., and Hovda, D.A. (2005).
Age-dependent reduction of cortical contusion
volume by ketones after traumatic brain injury. J
Neurosci Res 82, 413–420.
Prins, M.L., and Giza, C.C. (2006). Induction of mono-
carboxylate transporter 2 expression and ketone
transport following traumatic brain injury in juve-
nile and adult rats. Dev Neurosci 28, 447–456.
Prins, M.L., Lee, S.M., Fujima, L.S., and Hovda, D.A.
(2004). Increased cerebral uptake and oxidation of
exogenous betaHB improves ATP following trau-
matic brain injury in adult rats. J Neurochem 90,
666–672.
Prins, M.L., and Matsumoto, J.H. (2014). The collective
therapeutic potential of cerebral ketone metabolism
in traumatic brain injury. J Lipid Res 55, 2450–2457.
Puchowicz, M.A., Zechel, J.L., Valerio, J., Emancipator,
D.S., Xu, K., Pundik, S., LaManna, J.C., and Lust,
W.D. (2008). Neuroprotection in diet-induced
ketotic rat brain after focal ischemia. J Cereb Blood
Flow Metab 28, 1907–1916.
Rabchevsky, A.G., Sullivan, P.G., Fugaccia, I., and
Scheff, S.W. (2003). Creatine diet supplement for
spinal cord injury: influences on functional recov-
ery and tissue sparing in rats. J Neurotrauma 20,
659–669.
Rahman, M., Muhammad, S., Khan, M.A., Chen,
H., Ridder, D.A., Muller-Fielitz, H., Pokorna, B.,
Vollbrandt, T., Stolting, I., Nadrowitz, R., et  al.
(2014). The beta-hydroxybutyrate receptor HCA2
activates a neuroprotective subset of macrophages.
Nat Commun 5, 3944.
Rho, J.M. (2015). How does the ketogenic diet
induce anti-seizure effects? Neurosci Lett Jul 26.
pii: S0304–3940(15)30054–9
145
Rodriguez, D.J., Benzel, E.C., and Clevenger, F.W.
(1997). The metabolic response to spinal cord
injury. Spinal Cord 35, 599–604.
Ruskin, D.N., Kawamura, M., and Masino, S.A. (2009).
Reduced pain and inflammation in juvenile and
adult rats fed a ketogenic diet. PloS One 4, e8349.
Scafidi, S., Racz, J., Hazelton, J., McKenna, M.C., and
Fiskum, G. (2010). Neuroprotection by acetyl-L-
carnitine after traumatic injury to the immature
rat brain. Dev Neurosci 32, 480–487.
SCIRE. (2105). Spinal Cord Injury Rehabilitation
Evidence. wwwscireprojectcom.
Sharma, P., Benford, B., Li, Z.Z., and Ling, G.S. (2009).
Role of pyruvate dehydrogenase complex in trau-
matic brain injury and measurement of pyruvate
dehydrogenase enzyme by dipstick test. J Emerg
Trauma Shock 2, 67–72.
Sharma, S., Sankhyan, N., Gulati, S., and Agarwala, A.
(2013). Use of the modified Atkins diet for treat-
ment of refractory childhood epilepsy: a random-
ized controlled trial. Epilepsia 54, 481–486.
Shimazu, T., Hirschey, M.D., Newman, J., He, W.,
Shirakawa, K., Le Moan, N., Grueter, C.A.,
Lim, H., Saunders, L.R., Stevens, R.D., et  al.
(2013). Suppression of oxidative stress by beta-
hydroxybutyrate, an endogenous histone deacety-
lase inhibitor. Science 339, 211–214.
Singh, I.N., Sullivan, P.G., Deng, Y., Mbye, L.H., and
Hall, E.D. (2006). Time course of post-traumatic
mitochondrial oxidative damage and dysfunc-
tion in a mouse model of focal traumatic brain
injury: implications for neuroprotective therapy. J
Cereb Blood Flow Metab 26, 1407–1418.
Streijger, F., Plunet, W.T., Lee, J.H., Liu, J., Lam, C.K.,
Park, S., Hilton, B.J., Fransen, B.L., Matheson, K.A.,
Assinck, P., et al. (2013). Ketogenic diet improves
forelimb motor function after spinal cord injury in
rodents. PloS One 8, e78765.
Sullivan, P.G., Rippy, N.A., Dorenbos, K., Concepcion,
R.C., Agarwal, A.K., and Rho, J.M. (2004). The
ketogenic diet increases mitochondrial uncou-
pling protein levels and activity. Ann Neurol 55,
576–580.
Suzuki, M., Suzuki, M., Kitamura, Y., Mori, S., Sato,
K., Dohi, S., Sato, T., Matsuura, A., and Hiraide,
A. (2002). Beta-hydroxybutyrate, a cerebral func-
tion improving agent, protects rat brain against
ischemic damage caused by permanent and tran-
sient focal cerebral ischemia. Jpn J Pharmacol
89, 36–43.
Taggart, A.K., Kero, J., Gan, X., Cai, T.Q., Cheng, K.,
Ippolito, M., Ren, N., Kaplan, R., Wu, K., Wu, T.J.,
et  al. (2005). (D)-beta-Hydroxybutyrate inhibits
adipocyte lipolysis via the nicotinic acid receptor
PUMA-G. J Biol Chem 280, 26649–26652.
Thibault-Halman, G., Casha, S., Singer, S., and Christie,
S. (2011). Acute management of nutritional
demands after spinal cord injury. J Neurotrauma
28, 1497–1507.
146
146 section II: E merging Clinical Applications and Future Potential
Tieu, K., Perier, C., Caspersen, C., Teismann, P., Wu,
D.C., Yan, S.D., Naini, A., Vila, M., Jackson-
Lewis, V., Ramasamy, R., et  al. (2003). D-beta-
hydroxybutyrate rescues mitochondrial respiration
and mitigates features of Parkinson disease. J Clin
Nutr 112, 892–901.
Veech, R.L., Valeri, C.R., and VanItallie, T.B. (2012).
The mitochondrial permeability transition pore
provides a key to the diagnosis and treatment of
traumatic brain injury. IUBMB Life 64, 203–207.
Whishaw, I.Q., and Gorny, B. (1994). Arpeggio and
fractionated digit movements used in prehension
by rats. Behav Brain Res 60, 15–24.
Wilson, J.R., and Fehlings, M.G. (2014). Riluzole for
acute traumatic spinal cord injury:  a promis-
ing neuroprotective treatment strategy. World
Neurosurgery 81, 825–829.
Yang, L., Yao, M., Lan, Y., Mo, W., Sun, Y.L., Wang,
J., Wang, Y.J., and Cui, X.J. (2015). Melatonin
for spinal cord injury in animal models:  a sys-
tematic review and network meta-analysis. J
Neurotrauma.
Youm, Y.H., Nguyen, K.Y., Grant, R.W., Goldberg, E.L.,
Bodogai, M., Kim, D., D’Agostino, D., Planavsky,
N., Lupfer, C., Kanneganti, T.D., et  al. (2015).
The ketone metabolite beta-hydroxybutyrate
blocks NLRP3 inflammasome-mediated inflam-
matory disease. Nat Med 21, 263–269.
Zendedel, A., Johann, S., Mehrabi, S., Joghataei, M.T.,
Hassanzadeh, G., Kipp, M., and Beyer, C. (2015).
Activation and regulation of NLRP3 inflam-
masome by intrathecal application of SDF-1a
in a spinal cord injury model. Mol Neurobiol
Jul;53(5):3063–75
Zhao, W., Varghese, M., Vempati, P., Dzhun, A., Cheng,
A., Wang, J., Lange, D., Bilski, A., Faravelli, I., and
Pasinetti, G.M. (2012). Caprylic triglyceride as a
novel therapeutic approach to effectively improve
the performance and attenuate the symptoms due
to the motor neuron loss in ALS disease. PloS One
7, e49191.
Zhao, Z., Lange, D.J., Voustianiouk, A., MacGrogan,
D., Ho, L., Suh, J., Humala, N., Thiyagarajan, M.,
Wang, J., and Pasinetti, G.M. (2006). A keto-
genic diet as a potential novel therapeutic inter-
vention in amyotrophic lateral sclerosis. BMC
Neurosci 7, 29.
Ziegler, D.R., Gamaro, G.D., Araujo, E., Bassani, M.G.,
Perry, M.L., Dalmaz, C., and Goncalves, C.A.
(2005). Nociception and locomotor activity are
increased in ketogenic diet fed rats. Physiol Behav
84, 421–427.
147
17
Anti-Inflammatory Effects of a Ketogenic Diet
Implications for New Indications
N I N A D U P U I S , P H D A N D S T É P H A N E AU V I N, M D , P H D
INTRODUCTION
The ketogenic diet (KD) is a high-fat, low-
carbohydrate diet that results in ketosis, modula-
tion of glycemia, relative caloric restriction, and
elevations in the levels of certain fatty acids. In
clinical practice, KD is an established treatment
for pharmacoresistant epilepsy, and is increasingly
being explored for some inflammation-induced
epileptic encephalopathies such as fever-induced
refractory epileptic encephalopathy in school-
age children (FIRES) (Nabbout et al., 2011). The
FIRES syndrome begins with severe seizures
evolving into status epilepticus, and is preceded
by fever or common viral infection but without
an identifiable cause (Dupuis and Auvin, 2015).
Whether fever itself or some other factor is the
trigger remains to be determined. The outcome
is very severe with adverse motor and cognitive
consequences as well as medically refractory epi-
lepsy (Kramer et al., 2011). It is hypothesized that
the synergy between inflammation and status epi-
lepticus, the latter resulting partly from inflam-
mation itself and contributing to the induction of
inflammation, generates a vicious cycle that leads
to major neurological consequences (Nabbout
et al., 2011).
While the antiseizure mechanism(s) remain(s)
unclear, new uses for the KD are increasingly
suggested, given its broad efficacy in various
experimental models, including animal models
exhibiting prominent inflammatory changes. Our
aim in this chapter is to review the data support-
ing the anti-inflammatory profile of the KD, and
to link improvements in various animal models of
neurological disease with KD-induced modula-
tion of inflammation. Finally, we discuss the possi-
ble mechanisms underlying the anti-inflammatory
properties of the KD.
K E TO G E N I C D I E T E X H I B I T S
A N T I - I N F L A M M AT O RY
PROPERTIES
The use of the KD in a pain model first led to
the idea that this high-fat, low-carbohydrate diet
might possess anti-inflammatory properties. In
this study, the KD attenuated thermal nociception
and decreased peripheral edema (details below)
(Ruskin et  al., 2009). We recently further estab-
lished the anti-inflammatory effects of the KD
using a rat model of fever (Dupuis et al., 2015). In
Wistar rats, 2 weeks of KD treatment before lipo-
polysaccharide (LPS) i.p. injection reduced the
fever response as well as pro-inflammatory cyto-
kine levels. In this fever model, the body tempera-
ture did not rise similarly in KD-treated animals
compared with the controls. The body tempera-
ture was actually significantly lower in the KD
group (around 37.6°C vs 38.5°C) from 1 h 30 to
4 h after LPS administration (Dupuis et al., 2015).
The fever profile correlated with plasma IL- IL-
1B levels: 1 h after injection 130.8 ± 36.3 pg/mL
in control versus 51.4 ± 9.1 pg/mL in KD (p < .05);
2 h after injection 203.3 ± 65.8 pg/mL in con-
trol versus 171.9 ± 38.8 pg/mL in KD (p < .05).
Four hours after LPS injection, IL-1β and TNF-α
plasma levels were significantly lower in the KD
group compared with the control group, although
no difference was observed before LPS injection.
In contrast, 4 hours after LPS injection, PGE2
plasma levels were not different between the two
groups. Interestingly, 4 hours after LPS injection,
the KD group also showed lower IL1β mRNA lev-
els in the hippocampus compared with controls.
As expected, total fatty acid levels were increased
after 14 days of KD compared with a standard diet.
The levels of the long-chain omega-3 (n-3) PUFAs
eicosapentaenoic acid (EPA; C20:5 n-3) and
148
148 section II: E merging Clinical Applications and Future Potential
docosahexaenoic acid (DHA; C22:6 n-3), which
are precursors of some anti-inflammatory agents,
were decreased in the KD group. In contrast,
arachidonic acid (AA; C20:4 n-6), an n-6 PUFA
that is a precursor for the synthesis of some pro-
inflammatory eicosanoids, was also reduced by
the KD (Dupuis et al., 2015).
The KD further modulated fever by decreas-
ing peripheral inflammation and brain IL-1β
expression. Although we cannot exclude an
effect of ketone bodies and/or caloric restric-
tion in our recent study, we propose that the
decrease of AA may be key factor in the anti-
inflammatory properties of the KD. This sugges-
tion is underscored by our previous report of a
decrease in AA levels in patients responding to
KD (Porta et al., 2009).
Another recent study evaluated the effect of
a KD on fibroblast growth factor 21 (FGF21)
and inflammation (Asrih et  al., 2015). The
authors first established that the KD promoted
weight loss through increased energy expendi-
ture and reduced food intake. The mice treated
with a KD had increased plasma levels of FGF21;
Fgf21 gene expression was also increased in
the liver but decreased in white adipose tissue
(WAT). The FGF21 is produced principally in
the liver and mainly acts on adipocytes. It has
been shown that by targeting adipose tissue,
FGF21 promotes adiponectin production and
secretion; adiponectin subsequently mediates
the systemic effects of FGF21 (Holland et  al.,
2013; Lin et al., 2013). The authors reported an
increase in the expression of pro-inflammatory
cytokines (Tnfα, Il-6) and macrophage accumu-
lation (Emr1, Cd68, Itgam). In addition, a key
mediator of the inflammasome (Nlrp3) was sig-
nificantly increased in the liver of mice treated
with a KD. Interestingly, these inflammatory
markers were decreased in adipose tissue of KD-
fed mice (Asrih et al., 2015).
K E TO G E N I C D I E T
I M P R OV E S T H E   O U T C O M E
OF NEUROLOGICAL
DISEASE ANIMAL MODELS
W I T H   I N VO LV E M E N T
O F   I N F L A M M AT I O N
The KD has been widely studied in experimen-
tal models of seizure and epilepsy, and has also
been explored in other models of neurologi-
cal diseases, such as pain, multiple sclerosis, and
Parkinson’s disease. In these latter conditions,
the KD appears to render broad neuroprotective
effects (Table 17.1).
Pain
Chronic pain persists or progresses over a long
period of time. It seems that some chronic pain
patients suffer from long-term immune-system
activation resulting in continuous release of pro-
inflammatory cytokines. In contrast, there is
evidence that levels of anti-inflammatory cyto-
kines are decreased in patients with widespread
pain (Ueceyler et al., 2006) and neuropathic pain
(Uceyler et  al., 2007). This explains why chronic
pain may arise through an increase in pro-
inflammatory cytokines and the resulting inhibi-
tion of anti-inflammatory cytokines.
In a model of pain induced by complete
Freund’s adjuvant through intraplantar injection,
KD treatment over 3 weeks resulted in a decrease
in thermal nociception. This effect was associated
with a reduction in the peripheral inflammatory
response as measured by paw swelling (volume
measure) and plasma extravasation (measure-
ment of the dye Evans Blue from dissected tissue)
(Ruskin et  al., 2009). In this study, the authors
compared the response of adult and juvenile
(beginning diet at P21) animals. The KD was effec-
tive in both age groups, but both the hypoalgesic
effect and the reduction in plasma extravasation
were more robust in juveniles. However, the pres-
ence of edema is only one indicator of an inflam-
matory response, so these findings only tell part of
the story.
Multiple Sclerosis
Multiple sclerosis (MS) is an autoimmune inflam-
matory disorder of the central nervous system
affecting the white matter. This demyelinating dis-
order results in a wide range of symptoms includ-
ing motor, cognitive, and sometimes psychiatric
problems. The most widely studied animal model
of MS is the experimental autoimmune enceph-
alomyelitis (EAE) murine model induced by the
subcutaneous injection of myelin oligodendrocyte
glycoprotein peptide in complete Freund’s adju-
vant, followed by injection of pertussis toxin. This
model is characterized by CNS neurodegeneration
and synaptic loss, particularly in the hippocampus,
with resultant spatial learning and memory defi-
cits and motor impairment. Spatial learning and
memory is usually quantified by the Morris water
maze test, where mice must repeatedly find a hid-
den platform in clouded water, and motor impair-
ment is quantified by swim speed (latency) and
distance (path length).
In the only such study to date, a KD was given
7 days prior to EAE induction and continued until
motor and cognitive disabilities could be assessed
 149
TABLE 17.1 SUMMARY OF PRECLINICAL STUDIES ESTABLISHING THE ANTI- INFLAMMATORY EFFECT OF KD

Type of diet Species Model Anti-inflammatory effect of KD

Ruskin et al., 2009 6.6:1 KD Rats subcutaneous injection of complete Freund’s adju- Decrease both swelling and plasma extravasation
–3 weeks before Exp. vant into one hindpaw
Xan and Chen, 2010 Made by the lab C57BL/6J Mice MPTP model Decrease of activated microglia (Iba1 staining)
–2 weeks before Exp. Decrease of IL-1β, IL-6, TNFα (ELISA of SN)
Kim do et al., 2012 6.3:1 KD (Bio-Serv C57BL/6 mice Experimental autoimmune encephalomyelitis –2–2.5 fold reduction in CNS-derived CD4+ cells and
F3666 diet) –S.C. myelin oligodendrocyte glycoprotein CD11b+ CD45+ cells (macrophage and microglia
–7 days before Exp. (MOG)35–55 peptide + complete Freund’s –tendency toward increased CD4+
adjuvant (CFA) CD25+ Foxp3+ Treg cells
–I.V. 20 ng of pertussis toxin –Lymph node & CNS reduction in cytokines (IL-1β,
IL-6, TNF-α, IL-12, IL-17) and chemokines (IFN-γ,
MCP-1, MIP-1a, MIP-1b)
Asrih et al., 2015 6.3:1 KD (Bio-Serv C57BL/6 mice Liver and White Adipose Tissue (WAT) Liver: Increase of expression of Tnfα, Il-6, Emr1, Cd68,
F3666 diet) Itgam, Nlrp3
–4 weeks before Exp. WAT: Decrease of expression of Tnfα, Il-6, Emr1, Cd68,
Itgam, Nlrp3
Dupuis et al., 2015 3:1 KD (Ketocal) Wistar rats Fever model. Modulate rise in body temperature
–2 weeks before Exp. 50 μg/kg of LPS (Escherichia coli 055:B5) Blood: reduce IL-1β, TNF-α
Brain: reduce IL-1β mRNA
150
150 section II: E merging Clinical Applications and Future Potential
in comparison with a control group (Kim do et al.,
2012). The KD ameliorated both motor func-
tion and cognitive impairment; spatial learning-
memory function was also significantly improved,
as demonstrated with the Morris water maze and
measurements of hippocampal long-term potenti-
ation. Further, using 7T-MRI imaging techniques,
there was a reduction in the size of periventricu-
lar white matter lesions and restoration of hip-
pocampal volume loss after KD treatment in the
induced EAE group of mice. Importantly, this con-
stellation of beneficial effects was associated with
modulation of systemic inflammation. Specifically,
the KD reduced by over twofold the number of
CD4-positive T cells and CD11b/CD45-positive
macrophages/microglia. The KD-fed EAE mice
also showed a tendency toward increases in CD4+
CD25+ Foxp3+ Treg cells. Furthermore, levels
of several cytokines (IL-1β, IL-6, TNF-α, IL-12,
IL-17) and chemokines (IFN-γ, MCP-1, MIP-
1a, MIP-1b) were substantially lower in the KD-
treated EAE group compared with control EAE
mice. This modulation of cytokine and chemo-
kine levels was observed in both brain and blood
(Kim do et al., 2012). Finally, other than this anti-
inflammatory modulation of lymphocyte prolif-
eration and cytokine/chemokine expression, the
KD decreased overall oxidative stress in the brain
(Kim do et al., 2012).
Parkinson’s Disease
Parkinson’s disease (PD) is a progressive neuro-
degenerative disorder characterized by abnormal
shaking, rigidity, and slowness. The symptoms
arise from the death of dopaminergic neurons
in the substantia nigra (SN). Despite abundant
clinical information about PD, the mechanisms
of neurodegeneration in the SN remain unclear.
Nevertheless, there is some evidence for the role
on inflammation in PD, since activated microglia
have been implicated in the pathogenesis and pro-
gression of PD (Hirsch and Hunot, 2009; Hirsch
et al., 2012).
The murine MPTP (1-methyl-4-phenyl-
1,2,3,6-tetrahydropyridine) model of PD recapitu-
lates the motor dysfunction and neurochemical
changes observed in humans diagnosed with this
disorder. As with PD patients, MPTP-treated mice
are characterized by selective progressive loss of
dopaminergic neurons in the SN and their pro-
jections to the striatum. Ketogenic diet pretreat-
ment has been shown to provide neuroprotective
and anti-inflammatory effects and to alleviate the
motor dysfunction induced by MPTP (Yang and
Cheng, 2010).
In addition to the improvement in motor func-
tion, the KD was found to inhibit degeneration of
tyrosine hydroxylase (TH)-positive neurons and
preserved dopamine levels in the SN (Yang and
Cheng, 2010). This effect was associated with a
reduction in activated microglia as shown by iba-
1 immunostaining. Moreover, the KD decreased
pro-inflammatory cytokine levels (specifically, IL-
1β, IL6, and TNF-α) in the SN of MPTP-treated
animals (Yang and Cheng, 2010).
Most of the experimental data thus far has
shown a correlation between amelioration of dis-
ease symptomatology by the KD and decreased
inflammation (mostly with respect to pro-
inflammatory cytokine levels). However, whether
the beneficial effects of a KD in models of PD are a
direct consequence of decreased inflammation has
yet to be firmly demonstrated.
POS SIBLE MECHANISMS
SUPPORTING ANTI-
I N F L A M M AT O RY
PROPERTIES OF THE
K E TO G E N I C   D I E T
Although the clinical efficacy of the KD against
forms of pharmacoresistant epilepsy is well estab-
lished, the underlying mechanisms are incompletely
understood. There have been a number of intrigu-
ing hypotheses advanced to explain KD action
(Lutas and Yellen, 2013; Masino and Rho, 2010).
However, it is unlikely that a single mechanism can
explain all of the clinical effects of the KD. In the
context of the present chapter, there are also proba-
bly multiple mechanisms that are likely involved in
the anti-inflammatory effect of KD (Figure 17.1).
Ketone Bodies
The ketone bodies beta-hydroxybutyrate (BHB)
and acetoacetate (AcAc) are the products of
fatty acid oxidation in hepatic mitochondria.
Recently, several lines of research support a direct
role for BHB as an anti-inflammatory mediator
(Rahman et  al., 2014; Youm et  al., 2015). Beta-
hydroxybutyrate has been shown to directly inhibit
the NLRP3 (NOD-, LRR-, and pyrin domain-
containing 3)  inflammasome assembly, which is
responsible for caspase-1 activation and the release
of the pro-inflammatory cytokines IL-1β and IL-
18. The NLRP3 functions primarily in cells of the
myeloid lineage. In mouse bone-marrow-derived
macrophages (BMDMs) and human monocytes,
BHB inhibited ATP-induced cleavage of caspase-1
and the processing of the active form of IL-1β in
LPS-primed cells (Youm et  al., 2015). This effect
was independent of fasting-regulated mechanisms
151

Chapter 17: Anti-Inflammatory Effects of a Ketogenic Diet 151

FIGURE 17.1 Current understanding of the possible mechanisms supporting the anti-inflammatory mechanisms of
the ketogenic diet.
AA: arachidonic acid; BHB: beta-hydroxyButyrate; CR: caloric restriction; HCA2: hydroxyl-carboxylic acid receptor 2; Nfkbia: NFκB
inhibitor alpha; NLRP3:  NOD-, LRR- and pyrin domain-containing 3; mTOR:  mammalian target of rapamycin; PPAR:  peroxisome
proliferator-activated receptors; PUFA: polyunsaturated fatty acids; Timp3: tissue inhibitor of metalloproteinases-3

such as reactive oxygen species (ROS) production
and inhibition of glycolysis and autophagy, both of
which are known to regulate the NLRP3 inflam-
masome. The inhibitory effect of BHB appears
specific to NLRP3 because BHB does not inhibit
caspase-1 activation in response to pathogens acti-
vating other inflammasomes—namely, NLRC4
(NLR family, CARD domain containing 4)  and
AIM2 (absent in melanoma 2) (Youm et al., 2015).
Additionally, BHB was shown to inhibit caspase-
1 activation and IL-1β secretion in mouse models
of Muckle-Wells syndrome, a condition associated
with a mutation in NLRP3. Finally, the KD and the
associated elevation in BHB levels protected mice
bearing the missense Nlrp3 mutation that leads to
familial cold auto-inflammatory syndrome (Youm
et al., 2015).
Beta-hydroxybutyrate has also been shown to
be a potent modulator of the HCA2 (hydroxyl-
carboxylic acid 2) receptor, and through this target
exerts neuroprotective effects in a rodent model of
stroke (Rahman et al., 2014). The HCA2 activation
by BHB induces Ly-6CLo monocytes and/or mac-
rophages to deliver a neuroprotective signal to the
brain (Rahman et al., 2014).
Caloric Restriction
The KD is frequently associated with a reduced
calorie intake. More recently, it has been shown
that KD promotes weight loss through increased
energy expenditure and reduced food intake
(Asrih et  al., 2015). Caloric restriction (CR) is
a pragmatic strategy to handle excess energy,
stored in adipose tissues, that might contribute
to systemic inflammation (Ye and Keller, 2010).
Caloric restriction can increase anti-inflammatory
protein gene expression of mediators such as
NFκB inhibitor alpha (Nfkbia), tissue inhibitor of
metalloproteinases-3 (Timp3), and peroxisome
proliferator-activated receptors (PPARs) (Sung
et al., 2004; Swindell, 2009), and conversely inhibit
pro-inflammatory genes such as TNF-α, IL-6,
COX-2, iNOS, VCAM-1, and ICAM-1 (Higami
et al., 2006; Jung et al., 2009).
Despite these data, evidence for the potential
anti-inflammatory mechanisms of CR in humans
is more limited, and most of the studies address-
ing this have been developed in obese patients. In
this particular context, CR appears to reduce pro-
inflammatory molecules (Lee et  al., 2010; Salas-
Salvado et  al., 2006), but these data reflect more
the correction of an abnormal state to the non-
pathologic state than the changes induced by CR
under healthy conditions.
Overall, the underlying mechanisms explain-
ing the anti-inflammatory effects of CR are not
well understood. However, since many nutri-
ents and growth factors are able to activate
152
152 section II: E merging Clinical Applications and Future Potential
mammalian target of rapamycin (mTOR), the
anti-inflammatory effect of CR might be in part
the result of stimulating this pathway. Laboratory
studies have already demonstrated that the KD
inhibits mTOR signaling in the brain and liver of
normal rats (McDaniel et  al., 2011). This inhibi-
tion is thought to be related to a decrease in Akt
signaling in the brain and the liver. McDaniel and
colleagues also showed that the KD prevented
mTOR hyperactivation after kainate-induced
status epilepticus (SE) (McDaniel et  al., 2011).
However, this study did not correlate the modula-
tion of the mTOR by the KD with any SE-induced
pro-inflammatory changes.
Polyunsaturated Fatty Acids
Polyunsaturated fatty acids (PUFAs) are dietary
lipids that contain more than one double bond.
There are two groups of PUFAs: the omega-3 (n-3)
and the omega-6 (n-6) PUFAs. This nomenclature
refers to the position of the double bond relative to
the methyl terminal of the molecule (Taha et  al.,
2010). Diet is an important source of PUFAs.
Dietary n-3 PUFAs are found in flaxseed, in some
nuts, in marine fish, and in marine mammals, such
as seals. In contrast, n-6 PUFAs are found in a vari-
ety of animal products and in vegetable oils, such
as canola and corn oil, and make up the major-
ity of PUFAs in the modern Western diet (Taha
et al., 2010).
With normal dietary intake, n-3 PUFAs can
decrease the production of inflammatory eico-
sanoids, cytokines, and ROS and the expression
of adhesion molecules. Toward these ends, n-3
PUFAs act both directly (by replacing arachi-
donic acid as an eicosanoid substrate and inhibit-
ing arachidonic acid metabolism) and indirectly
(by altering the expression of inflammatory genes
through effects on transcription factor activa-
tion). The n-3 PUFAs, particularly eicosapentae-
noic acid (EPA; 20:5n-3) and docosahexaenoic
acid (DHA, 22:6n-3), also have anti-inflammatory
actions, mediated primarily by their hydroxylated
metabolites, which include resolvins and doco-
sanoids (Hong et  al., 2003). For instance, EPA
and DHA have been reported to incorporate into
human neutrophil cells and to decrease the pro-
duction of the pro-inflammatory prostaglandin
E2 (PGE2) in a dose-dependent manner (Rees
et al., 2006).
Ketogenic diet treatment not only increases
fatty acid levels in the blood but also increases
levels of specific PUFAs that can bind to and acti-
vate peroxisome proliferator-activated receptors
(PPARs). The PPARs have been considered as
potential drug targets for seizure control (Sampath
and Ntambi, 2005). Both PPARα and PPARγ are
activated by PUFAs and their eicosanoid derivates,
as well as by synthetic ligands such as the fibrates
(Kersten et al., 2000). Activation of PPARs stimu-
lates the transcription of the genes involved in
fatty acid oxidation via the formation of heterodi-
meric transcription-factor complexes with the
retinoid-X-receptor (RXR) (Smith, 2002). Three
types of PPARs have been identified—alpha, beta,
and gamma:  PPARα is expressed mainly in liver,
kidney, and heart muscle; PPARβ is expressed in
a number of tissues, such as brain, adipose tissue,
and skin; PPARγ is also expressed in the brain.
Both ALA and LA have a greater binding affinity
for PPARα than EPA or DHA (Lin et  al., 1999).
Synthetic PPAR agonists reduce experimentally
induced inflammation (Cuzzocrea et  al., 2003;
Cuzzocrea et  al., 2004; Lo Verme et  al., 2005).
This effect is the result of the inhibition of pro-
inflammatory pathways involving nuclear fac-
tor κ B, signal transduction and transcription-1,
and nuclear factor of activated T-cells (Blanquart
et al., 2003).
Adenosine Modulation
The KD has been shown to modulate adenosine
levels. Astrocytic adenosine kinase (ADK) is an
enzyme responsible for adenosine phosphoryla-
tion and clearance of adenosine from the extra-
cellular space. Expression of ADK is reduced by
the KD, thereby increasing extracellular levels of
adenosine and the activation of inhibitory ade-
nosine A1 receptor (A1AR) (Masino et  al., 2011).
Importantly, the effect of the KD on adenosine
has been linked to a decrease in electrographic
seizure activity (Masino et  al., 2011). The anti-
inflammatory effects of the KD may be explained
in part by adenosine, as this metabolic mediator
has long been known to induce anti-inflammatory
activity (Kowaluk et  al., 1998; Linden, 2001).
Along these lines, adenosine has been reported
to reduce central and peripheral inflammation.
Using the A1AR(-/-) genetically engineered mice
and a pretreatment with the specific A1AR ago-
nist (2’Me-2-chloro-N6-cyclopentyladenosine) in
wild-type mice, it was shown that A1AR modulates
LPS-induced transmigration of polymorphonu-
clear cells and reduced levels of TNF-α, IL-6, and
CXCL2/3 in the lung (Ngamsri et al., 2010). There
are also data suggesting that A1Rs are up-regulated
in acute forms of neuroinflammation and down-
regulated in chronic forms. Activation of cerebral
A1R act as a brake for the microglial response after
traumatic brain injury (Haselkorn et  al., 2010).
153
Chapter 17: Anti-Inflammatory Effects of a Ketogenic Diet
Both A1ARs and A2ARs appear to be involved in
the inflammatory response, and pharmacological
interventions might mitigate this (Lukashev et al.,
2005; Sitkovsky and Ohta, 2005). However, A2ARs
as well as pharmacological intervention targeting
A2ARs might have bidirectional effects on neu-
roinflammation. Specifically, the local glutamate
level following brain injury is one of the crucial
factors that determines the direction of an A2AR-
mediated effect on neuroinflammation (Dai and
Zhou, 2011).
Reactive Oxygen Species production,
Uncoupling Proteins, Mitochondrial
Membrane Potential
Both BHB and AcAc have been shown to induce
neuroprotective effects. These effects have been
linked to reduction of ROS through enhanced
NADH oxidation and inhibition of mitochon-
drial permeability transition (Kim do et  al.,
2007; Kim do et al., 2015; Maalouf et al., 2007).
Furthermore, the KD has been shown to enhance
mitochondrial biogenesis (Bough et  al., 2006).
The ROS and inflammation are tightly linked
through a strong reciprocal relationship; ROS
generated by inflammatory cells act as inflamma-
tory signaling molecules and participate in the
inflammatory response. Furthermore, ROS play
a key role in the control of nuclear factor kappa
B (NF-kappaB), activator protein-1 (AP-1), and
other transcription factors involved in gene
expression of both inflammatory and immune
mediators. When sustained, oxidative stress can
lead to a chronic inflammatory state and inflam-
mation can produce aggravate ROS production
(Harijith et al., 2014).
The KD can reduce both ROS production
and inflammation. Recently, it has been shown
that ROS can activate the NLRP3 inflamma-
some protein, a molecular platform activated
on signs of cellular “danger” to trigger innate
immune defenses through the maturation of pro-
inflammatory cytokines such as IL-1β and IL-18
(Zhou et al., 2011). Moreover, oxidative stress can
act as a central regulator of HMGB1’s transloca-
tion, release, and activity in inflammation and
cell death. Thus, modulation of HMGB1 can limit
inflammation and reduce tissue damage during
infection as well as sterile inflammation (Yu et al.,
2015). Ketone bodies now appear to have the abil-
ity to target the inflammatory pathways in multi-
ple ways: (1) direct action on NLRP3 as described
previously (Youm et  al., 2015), and (2)  decrease
in ROS production and levels (Kim do et  al.,
2007; Kim do et al., 2015; Maalouf et al., 2007).
153
CLINICAL EVIDENCE AND
F U T U R E I N D I C AT I O N S
There is currently no clinical open label study or
controlled trial that demonstrates that the effects
of the KD are through an anti-inflammatory
mechanism. Notwithstanding this lack of data, it is
intriguing to note once again that the KD is an effec-
tive treatment for a condition associated with both
inflammation and medically refractory seizures—
namely, FIRES (Nabbout et al., 2010; Singh et al.,
2014). At present, whether the KD would indeed
be useful for diverse conditions associated with
neuroinflammation—for example, pain, multiple
sclerosis, and Parkinson’s disease, among others—
remains unclear, but the current evidence point-
ing to broad anti-inflammatory effects of the KD
is compelling, as this metabolism-based treatment
could constitute a readily available therapy for
neurological disorders associated with inflamma-
tion. Detailed preclinical studies in relevant ani-
mal models of neurological disease are no doubt
of paramount importance to uncover the underly-
ing biology of KD action. Ultimately though, well-
designed clinical trials must be implemented to
validate readily available therapeutic approaches
such as the KD.
REFERENCES
Asrih, M., Altirriba, J., Rohner-Jeanrenaud, F., and
Jornayvaz, F.R. (2015). Ketogenic diet impairs
FGF21 signaling and promotes differential inflam-
matory responses in the liver and white adipose
tissue. PLoS One 10.
Blanquart, C., Barbier, O., Fruchart, J.C., Staels, B.,
and Glineur, C. (2003). Peroxisome proliferator-
activated receptors:  regulation of transcriptional
activities and roles in inflammation. J Steroid
Biochem Mol Biol 85, 267–273.
Bough, K.J., Wetherington, J., Hassel, B., Pare, J.F.,
Gawryluk, J.W., Greene, J.G., Shaw, R., Smith,
Y., Geiger, J.D., and Dingledine, R.J. (2006).
Mitochondrial biogenesis in the anticonvulsant
mechanism of the ketogenic diet. Ann Neurol 60,
223–235.
Cuzzocrea, S., Mazzon, E., Dugo, L., Patel, N.S.A.,
Serraino, I., Di Paola, R., Genovese, T., Britti, D.,
De Maio, M., Caputi, A.P., et al. (2003). Reduction
in the evolution of murine type II collagen-
induced arthritis by treatment with rosiglitazone,
a ligand of the peroxisome proliferator-activated
receptor gamma. Arthritis Rheum 48, 3544–3556.
Cuzzocrea, S., Pisano, B., Dugo, L., Ianaro, A., Maffia,
P., Patel, N.S.A., Di Paola, R., Ialenti, A., Genovese,
T., Chatterjee, P.K., et  al. (2004). Rosiglitazone, a
ligand of the peroxisome proliferator-activated
receptor-gamma, reduces acute inflammation. Eur
J Pharmacol 483, 79–93.
154
154 section II: E merging Clinical Applications and Future Potential
Dai, S.-S., and Zhou, Y.-G. (2011). Adenosine 2A
receptor:  a crucial neuromodulator with bidi-
rectional effect in neuroinflammation and brain
injury. Rev Neurosci 22, 231–239.
Dupuis, N., and Auvin, S. (2015). Inflammation and
epilepsy in the developing brain:  clinical and
experimental evidence. CNS Neurosci Ther 21,
141–151.
Dupuis, N., Curatolo, N., Benoist, J.-F., and Auvin, S.
(2015). Ketogenic diet exhibits anti-inflammatory
properties. Epilepsia 56, e95–e98.
Harijith, A., Ebenezer, D.L., and Natarajan, V.
(2014). Reactive oxygen species at the cross-
roads of inflammasome and inflammation. Front
Physiol 5.
Haselkorn, M.L., Shellington, D.K., Jackson, E.K.,
Vagni, V.A., Janesko-Feldman, K., Dubey, R.K.,
Gillespie, D.G., Cheng, D., Bell, M.J., Jenkins,
L.W., et al. (2010). Adenosine A(1) Receptor acti-
vation as a brake on the microglial response after
experimental traumatic brain injury in mice. J
Neurotrauma 27, 901–910.
Higami, Y., Barger, J.L., Page, G.P., Allison, D.B., Smith,
S.R., Prolla, T.A., and Weindruch, R. (2006).
Energy restriction lowers the expression of genes
linked to inflammation, the cytoskeleton, the
extracellular matrix, and angiogenesis in mouse
adipose tissue. J Nutr 136, 343–352.
Hirsch, E.C., and Hunot, S. (2009). Neuroinflammation
in Parkinson’s disease:  a target for neuroprotec-
tion? Lancet Neurol 8, 382–397.
Hirsch, E.C., Vyas, S., and Hunot, S. (2012).
Neuroinflammation in Parkinson’s disease.
Parkinsonism Rel Disord 18 (Suppl 1), S210–212.
Holland, W.L., Adams, A.C., Brozinick, J.T., Bui,
H.H., Miyauchi, Y., Kusminski, C.M., Bauer, S.M.,
Wade, M., Singhal, E., Cheng, C.C., et al. (2013).
An FGF21-adiponectin-ceramide axis controls
energy expenditure and insulin action in mice.
Cell Metabol 17, 790–797.
Hong, S., Gronert, K., Devchand, P.R., Moussignac,
R.L., and Serhan, C.N. (2003). Novel docosatri-
enes and 17S-resolvins generated from docosa-
hexaenoic acid in murine brain, human blood, and
glial cells—autacoids in anti-inflammation. J Biol
Chem 278, 14677–14687.
Jung, K.J., Lee, E.K., Kim, J.Y., Zou, Y., Sung, B., Heo,
H.S., Kim, M.K., Lee, J., Kim, N.D., Yu, B.P., et al.
(2009). Effect of short term calorie restriction on
pro-inflammatory NF-kB and AP-1 in aged rat
kidney. Inflamm Res 58, 143–150.
Kersten, S., Desvergne, B., and Wahli, W. (2000).
Roles of PPARs in health and disease. Nature 405,
421–424.
Kim do, Y., Davis, L.M., Sullivan, P.G., Maalouf, M.,
Simeone, T.A., van Brederode, J., and Rho, J.M.
(2007). Ketone bodies are protective against oxi-
dative stress in neocortical neurons. J Neurochem
101, 1316–1326.
Kim do, Y., Hao, J., Liu, R., Turner, G., Shi, F.D., and
Rho, J.M. (2012). Inflammation-mediated mem-
ory dysfunction and effects of a ketogenic diet in
a murine model of multiple sclerosis. PLoS One 7,
e35476.
Kim do, Y., Simeone, K.A., Simeone, T.A., Pandya, J.D.,
Wilke, J.C., Ahn, Y., Geddes, J.W., Sullivan, P.G.,
and Rho, J.M. (2015). Ketone bodies mediate anti-
seizure effects through mitochondrial permeabil-
ity transition. Ann Neurol 78, 77–87.
Kowaluk, E.A., Bhagwat, S.S., and Jarvis, M.F. (1998).
Adenosine kinase inhibitors. Curr. Pharm. Des. 4,
403–416.
Kramer, U., Chi, C.S., Lin, K.L., Specchio, N., Sahin,
M., Olson, H., Nabbout, R., Kluger, G., Lin, J.J.,
and van Baalen, A. (2011). Febrile infection-
related epilepsy syndrome (FIRES): pathogenesis,
treatment, and outcome A multicenter study on 77
children. Epilepsia 52, 1956–1965.
Lee, I.S., Shin, G., and Choue, R. (2010). A 12-week
regimen of caloric restriction improves levels of
adipokines and pro-inflammatory cytokines in
Korean women with BMIs greater than 23  kg/
m(2). Inflamm Res 59, 399–405.
Lin, Q.O., Ruuska, S.E., Shaw, N.S., Dong, D., and
Noy, N. (1999). Ligand selectivity of the per-
oxisome proliferator-activated receptor alpha.
Biochemistry 38, 185–190.
Lin, Z., Tian, H., Lam, K.S.L., Lin, S., Hoo, R.C.L.,
Konishi, M., Itoh, N., Wang, Y., Bornstein, S.R.,
Xu, A., et  al. (2013). Adiponectin mediates the
metabolic effects of FGF21 on glucose homeosta-
sis and insulin sensitivity in mice. Cell Metabol 17,
779–789.
Linden, J. (2001). Molecular approach to adenosine
receptors:  receptor-mediated mechanisms of tis-
sue protection. Annu Rev Pharmacol Toxicol 41,
775–787.
Lo Verme, J., Fu, J., Astarita, G., La Rana, G., Russo,
R., Calignano, A., and Piomelli, D. (2005). The
nuclear receptor peroxisome proliferator-activated
receptor-alpha mediates the anti-inflammatory
actions of palmitoylethanolamide. Mol. Pharmacol
67, 15–19.
Lukashev, D.E., Ohta, A., and Sitkovsky, M.V. (2005).
Physiological regulation of acute inflammation
by A(2A) adenosine receptor. Drug Dev. Res 64,
172–177.
Lutas, A., and Yellen, G. (2013). The ketogenic
diet: metabolic influences on brain excitability and
epilepsy. Trends Neurosci 36, 32–40.
Maalouf, M., Sullivan, P.G., Davis, L., Kim, D.Y., and
Rho, J.M. (2007). Ketones inhibit mitochondrial
production of reactive oxygen species production
following glutamate excitotoxicity by increasing
NADH oxidation. Neuroscience 145, 256–264.
Masino, S.A., Li, T., Theofilas, P., Sandau, U.S., Ruskin,
D.N., Fredholm, B.B., Geiger, J.D., Aronica, E., and
Boison, D. (2011). A ketogenic diet suppresses
155
Chapter 17: Anti-Inflammatory Effects of a Ketogenic Diet
seizures in mice through adenosine A(1) recep-
tors. J Clin Invest 121, 2679–2683.
Masino, S.A., and Rho, J.M. (2010). Mechanisms of
ketogenic diet action. Epilepsia 51, 85–85.
McDaniel, S.S., Rensing, N.R., Thio, L.L., Yamada,
K.A., and Wong, M. (2011). The ketogenic diet
inhibits the mammalian target of rapamycin
(mTOR) pathway. Epilepsia 52, e7–e11.
Nabbout, R., Mazzuca, M., Hubert, P., Peudennier, S.,
Allaire, C., Flurin, V., Aberastury, M., Silva, W., and
Dulac, O. (2010). Efficacy of ketogenic diet in severe
refractory status epilepticus initiating fever induced
refractory epileptic encephalopathy in school age
children (FIRES). Epilepsia 51, 2033–2037.
Nabbout, R., Vezzani, A., Dulac, O., and Chiron, C.
(2011). Acute encephalopathy with inflammation-
mediated status epilepticus. Lancet Neurol 10,
99–108.
Ngamsri, K.-C., Wagner, R., Vollmer, I., Stark, S., and
Reutershan, J. (2010). Adenosine receptor A1
regulates polymorphonuclear cell trafficking and
microvascular permeability in lipopolysaccharide-
induced lung injury. J Immunol 185, 4374–4384.
Porta, N., Vallee, L., Boutry, E., Fontaine, M., Dessein,
A.F., Joriot, S., Cuisset, J.M., Cuvellier, J.C., and
Auvin, S. (2009). Comparison of seizure reduc-
tion and serum fatty acid levels after receiving the
ketogenic and modified Atkins diet. Seizure Eur J
Epilep 18, 359–364.
Rahman, M., Muhammad, S., Khan, M.A., Chen,
H., Ridder, D.A., Muller-Fielitz, H., Pokorna, B.,
Vollbrandt, T., Stolting, I., Nadrowitz, R., et  al.
(2014). The beta-hydroxybutyrate receptor HCA2
activates a neuroprotective subset of macrophages.
Nat Commun 5, 3944.
Rees, D., Miles, E.A., Banerjee, T., Wells, S.J., Roynette,
C.E., Wahle, K.W.J., and Calder, P.C. (2006). Dose-
related effects of eicosapentaenoic acid on innate
immune function in healthy humans: a comparison
of young and older men. Am J Clin Nutr 83, 331–342.
Ruskin, D.N., Kawamura, M., and Masino, S.A. (2009).
Reduced pain and inflammation in juvenile and
adult rats fed a ketogenic diet. PLoS One 4, e8349.
Salas-Salvado, J., Bullo, M., Garcia-Lorda, P.,
Figueredo, R., Del Castillo, D., Bonada, A., and
Balanza, R. (2006). Subcutaneous adipose tissue
cytokine production is not responsible for the res-
toration of systemic inflammation markers during
weight loss. Int J Obes 30, 1714–1720.
Sampath, H., and Ntambi, J.M. (2005). Polyunsaturated
fatty acid regulation of genes of lipid metabolism.
Annu Rev Nut 25, 317–340.
155
Singh, R.K., Joshi, S.M., Potter, D.M., Leber, S.M.,
Carlson, M.D., and Shellhaas, R.A. (2014).
Cognitive outcomes in febrile infection-related
epilepsy syndrome treated with the ketogenic diet.
Pediatrics 134, e1431–1435.
Sitkovsky, M.V., and Ohta, A. (2005). The “danger”
sensors that STOP the immune response:  the
A2 adenosine receptors? Trend Immunol 26,
299–304.
Smith, S.A. (2002). Peroxisome proliferator-activated
receptors and the regulation of mammalian lipid
metabolism. Biochem Soc Trans 30, 1086–1090.
Sung, B., Park, S., Yu, B.P., and Chung, H.Y. (2004).
Modulation of PPAR in aging, inflammation, and
calorie restriction. J Gerontol A Biol Sci Med Sci
59, 997–1006.
Swindell, W.R. (2009). Genes and gene expres-
sion modules associated with caloric restric-
tion and aging in the laboratory mouse. BMC
Genomics 10.
Taha, A.Y., Burnham, W.M., and Auvin, S. (2010).
Polyunsaturated fatty acids and epilepsy. Epilepsia
51, 1348–1358.
Uceyler, N., Rogausch, J.P., Toyka, K.V., and Sommer,
C. (2007). Differential expression of cytokines in
painful and painless neuropathies. Neurology
69, 42–49.
Ueceyler, N., Valenza, R., Stock, M., Schedel, R.,
Sprotte, G., and Sommer, C. (2006). Reduced lev-
els of antiinflammatory cytokines in patients with
chronic widespread pain. Arthritis Rheum 54,
2656–2664.
Yang, X., and Cheng, B. (2010). Neuroprotective and
anti-inflammatory activities of ketogenic diet on
MPTP-induced neurotoxicity. J Mol Neurosci 42,
145–153.
Ye, J., and Keller, J.N. (2010). Regulation of energy
metabolism by inflammation: a feedback response
in obesity and calorie restriction. Aging-Us 2,
361–368.
Youm, Y.H., Nguyen, K.Y., Grant, R.W., Goldberg,
E.L., Bodogai, M., Kim, D., D’Agostino, D.,
Planavsky, N., Lupfer, C., Kanneganti, T.D.,
et  al. (2015). The ketone metabolite beta-
hydroxybutyrate blocks NLRP3 inflammasome-
mediated inflammatory disease. Nat Med 21,
263–269.
Yu, Y., Tang, D., and Kang, R. (2015). Oxidative stress-
mediated HMGB1 biology. Front Physiol 6.
Zhou, R., Yazdi, A.S., Menu, P., and Tschopp, J. (2011).
A role for mitochondria in NLRP3 inflammasome
activation. Nature 469, 221–225.
156
18
Dietary Therapy for Neurological Disorders
Focus on Amyotrophic Lateral Sclerosis, Parkinson’s Disease,
Mood Disorders, and Migraine
C A R L E . S TA F S T R O M , M D ,   P H D
INTRODUCTION
The ketogenic diet (KD) has established efficacy
for epilepsy and is best known for this role (Neal
et al., 2008; Winesett et al., 2015). Since the 1920s,
the KD has been used for treatment of intractable
epilepsy, often as a last resort. The mechanism
of the KD is complex and not fully elucidated,
but likely involves metabolic adaptations in a
variety of cellular signaling pathways leading to
decreased neuronal excitability and neuroprotec-
tion. As such, it is possible that the KD can be
used therapeutically in other neurological diseases
(Barañano and Hartman, 2008; Paoli et al., 2014;
Stafstrom and Rho, 2012; Gano et al., 2014), and
small preliminary trials have been performed in
brain cancer, Alzheimer disease, autism spectrum
disorder, multiple sclerosis, and pain, as discussed
elsewhere in this volume.
Here, I  review the use of the KD in amyo-
trophic lateral sclerosis (ALS), Parkinson disease,
mood disorders, and migraine. These diverse dis-
orders do not share a pathophysiological basis that
is immediately obvious. Amyotrophic lateral scle-
rosis is a motor neuron disease involving degen-
eration of alpha motor neurons. Parkinson disease
entails excitotoxic degeneration of dopaminergic
neurons of the substantia nigra leading to abnor-
mal motor function and cognition. Bipolar dis-
order and other mood disorders are related to
dysfunction of neurochemical balance in the brain,
especially monoamines. Migraine is a paroxysmal
headache disorder involving abnormal sensitivity
of blood vessels and neurons, especially involving
the trigeminal system.
Despite this phenotypic and mechanistic het-
erogeneity, each disorder involves abnormalities in
cellular energy utilization, implying that a benefi-
cial effect might be possible by manipulating nutri-
ents and metabolic substrates. Specifically, the role
of altered energy metabolism points to altered
mitochondrial function as a common factor in
pathogenesis (Kunz, 2002). Similar therapeutic
efficacy might be achieved through alternative
forms of the KD, including the medium chain tri-
glyceride (MCT) diet, modified Atkins diet, low
glycemic index treatment, or calorie restriction;
these therapies are mentioned only in passing in
this chapter, as their role in nonepileptic disorders
has barely been explored to date.
T H E N E U RO P ROT E C T I V E
ROLE OF THE
K E TO G E N I C   D I E T
Over the past 15  years or so, investigators have
identified numerous mechanisms by which the
KD diet might provide neuroprotection (Hartman,
2012; Rho and Stafstrom, 2012; Lutas and Yellen,
2013; Danial et  al., 2013). This complex patho-
physiology is covered in other chapters of this
volume. Nevertheless, a brief discussion of patho-
physiological mechanisms is warranted here, as a
basis for explaining why diverse neurologic disor-
ders could be amenable to KD treatment.
Two hallmark features of KD treatment are
the increase in ketone body production by the
liver and a reduction in blood glucose levels.
Ketone elevation is largely a consequence of fatty
acid oxidation. In the setting of high fat and low
carbohydrate intake, as in the KD, fatty acids are
oxidized in the liver to produce ketone bodies:
β-hydroxybutyrate and acetoacetate. These ketones
then occasion metabolic adaptations that produce
an antiseizure effect. Ketone bodies, as well as
specific polyunsaturated fatty acids (PUFAs) such
as arachidonic acid, docosahexaenoic acid, and
eicosapentaenoic acid might themselves regulate
neuronal membrane excitability by (1)  directly
blocking voltage-gated sodium and calcium
157
Chapter 18: Dietary Therapy for Neurological Disorders
channels (Vreugdenhil et al., 1996); (2) altering the
synthesis, degradation, or reuptake of excitatory or
inhibitory neurotransmitters (Yudkoff et al., 2001;
Juge et  al., 2010); or (3)  reducing inflammation
through activation of peroxisome proliferator-
activated receptors (Hajjar et al., 2012).
How could altered neuronal excitability and
dysregulated cellular energy metabolism be
linked? One way is via ATP-sensitive potassium
(KATP) channels. β-hydroxybutyrate increases the
probability that KATP channels open, which would
decrease cellular excitability and reduce firing rate
and seizure occurrence (Ma et al., 2007). Another
route is adenosine, which is produced from ATP
and causes seizure suppression through activa-
tion of inhibitory adenosine A1 receptors. The KD
increases ATP and thus adenosine, producing an
inhibitory effect on cellular excitability (Masino
et al., 2009). Ketone bodies have also been shown
to facilitate neuroprotection by raising ATP lev-
els and reducing ROS production (Kim et  al.,
2007)  through enhanced NADH oxidation and
inhibition of mitochondrial permeability tran-
sition (Kim et  al., 2015). Along similar lines of
improved bioenergetics, the KD has been shown
to stimulate mitochondrial biogenesis, resulting in
stabilized synaptic function (Bough et  al., 2006).
Therefore, altered mitochondrial biogenesis or
function represents a mechanistically appropriate
target for the KD in both epilepsy and nonepilep-
tic neurologic disorders (Milder and Patel, 2012;
Pani, 2015).
The second major biochemical feature of the
KD is decreased glycolytic flux. Reduction of gly-
colysis is an essential feature of calorie restriction,
which has been shown to suppress seizures as
well as prolong the lifespan of numerous species,
including primates (Pani, 2015). While the link
between calorie restriction and KD effectiveness
remains controversial, it is clear that both treat-
ments result in reduction of blood glucose, likely
involving reduced glycolytic flux. In this regard,
2-deoxy-D-glucose (2-DG), an analog of glucose
that blocks phosphoglucose isomerase and hence
inhibits glycolysis, has been shown to block epi-
leptogenesis in the rat kindling model by decreas-
ing expression of brain-derived neurotrophic
factor and its principal receptor, tyrosine kinase B
(Garriga-Canut et  al., 2006). 2-DG is also effec-
tive in several acute seizure models (Stafstrom
et  al., 2009). Similarly, fructose-1,6-diphosphate
reduces glycolysis by diverting glucose to the pen-
tose phosphate pathway and is neuroprotective in
multiple seizure models (Lian et  al., 2007; Ding
et al., 2013).
157
Other important mechanisms also contribute
to the neuroprotective consequences of calorie
restriction, including improved mitochondrial
function and decreased oxidative stress (similar
to that seen with ketones and PUFAs), decreased
activity of pro-apoptotic factors, and inhibition
of inflammatory mediators such as interleukins
and tumor necrosis factor alpha (Maalouf et  al.,
2009). Restoring exhausted metabolic substrates
may constitute another novel treatment approach
(Kovac et al., 2013). In that regard, a biochemical
process called anaplerosis aims to replenish tricar-
boxylic acid cycle intermediates that are depleted
during the intense neuronal firing that constitutes
seizure activity (Willis et al., 2010).
In the end, numerous mechanisms are likely
to contribute to the neuroprotective properties
of the KD. While several of these mechanisms
are thought to relate principally to the antiseizure
activity of the KD, the KD also likely contributes
to cellular homeostasis and the prevention of
neuronal dysfunction and injury. An important
caveat, however, is that yet unidentified mecha-
nisms operate in disorders outside of epilepsy, pre-
senting further opportunities for examining the
pleiotropic effects of metabolism-based therapies
at a mechanistic level. Furthermore, an important
consideration applicable to all neurodegenerative
diseases is the timing of intervention that is crucial
for a neuroprotective effect to be enabled by KD
treatment. Neurological disorders in late stages of
progression may have such extreme neuronal dys-
function and death that neuroprotective therapies
may no longer work. All of these potential mecha-
nisms are discussed in greater detail in other chap-
ters of this volume.
EXAMPLES OF NEUROLOGIC
D I S O R D E R S P O T E N T I A L LY
A M E NA B L E TO   T H E
K E TO G E N I C   D I E T
Several neurologic diseases might benefit from
dietary therapy, including amyotrophic lateral
sclerosis, Parkinson disease, mood disorders, and
migraine. These disorders have diverse underlying
pathophysiologies, yet each entails some contribu-
tion of cellular energy dysfunction.
Amyotrophic Lateral Sclerosis
Amyotrophic lateral sclerosis (ALS) is a rapidly
progressive disease due to degeneration of motor
neurons of the cortex and anterior horn of the
spinal cord. As a consequence, voluntary motor
activity gradually deteriorates, leaving the affected
individual profoundly weak despite largely retained
158
158 section II: E merging Clinical Applications and Future Potential
cognitive function. The essential pathophysi-
ological mechanisms that underlie this relentless
disorder are yet to be fully elucidated, and likely
involve oxidative damage, glutamate excitotoxic-
ity, inflammation, and mitochondrial membrane
dysfunction (Vucic et  al., 2014). Similar to other
neurodegenerative disorders, energy-producing
systems likely play a role and mitochondrial dys-
function probably contributes to disease patho-
genesis (Martin, 2011). In this regard, the KD
may be a promising adjunctive treatment for ALS
(Siva, 2006).
Amyotrophic lateral sclerosis is hereditary in
an estimated 10% of cases. Approximately 20%
of patients with familial ALS harbor a mutation
in the gene encoding copper/zinc superoxide dis-
mutase (SOD1). A transgenic mouse model of ALS
has been created in which the gene SOD 1-G93A is
overexpressed, leading to progressive muscle weak-
ness and death resulting from respiratory failure as
in the human disease. Administration of the KD to
these mutant mice led to both histological (higher
motor neuron counts) and functional (preserved
motor function on the rotarod test) improvements
compared with non-KD fed SOD 1-G93A trans-
genic animals (Zhao et al., 2006). However, the KD
did not extend survival time compared with non-
KD fed control mice. Although providing proof-
of-principle for using the KD in ALS, this study
is limited by a number of factors including very
small group sizes and initiation of motor testing
before the onset of ALS-type symptoms.
Mitochondria isolated from brains of SOD 1-
G93A mutant mice had increased levels of ATP
and increased rates of ATP synthesis, as well
as decreased function of complex I  of the elec-
tron transport chain. Administration of the β-
hydroxybutyrate restored complex I  activity in
neurons in which complex I function was blocked
pharmacologically (Zhao et  al., 2006), a result
also seen in mitochondria isolated from brains of
a Parkinson’s disease animal model (Tieu et  al.,
2003). An alternative approach involves caprylic
triglyceride, a medium-chain triglyceride with
antiseizure effects that is metabolized to ketone
bodies (Wlaź et  al., 2012). Caprylic triglycer-
ide, when fed to SOD 1-G93A transgenic mice,
reduced motor neuron loss and preserved motor
performance on the rotarod and other tests (Zhao
et al., 2012). Finally, a recent study examined SOD
1-G93A mutant mice fed a KD variant called the
Deanna Protocol (DP, which provides alternative
fuels in the form of tricarboxylic acid cycle inter-
mediates such as arginine-α-ketoglutarate), or
both the standard KD and the DP together (Ari
et al., 2014). Both the KD and DP diets were asso-
ciated with improved motor scores compared with
control-fed mutant mice, and the KD and KD+DP
diets prolonged life expectancy. These results raise
the possibility that the KD or similar metabolic
therapies, by raising ketone levels, might benefit
patients with ALS. In a broader sense, the role of
calories or simply increasing dietary fat in ALS
patients remains a subject of intense interest, since
dietary approaches are a rather straightforward
way to potentially enhance the quality and dura-
tion of life in this devastating disease (Paganoni
and Wills, 2013).
Parkinson Disease
The primary pathophysiology in Parkinson disease
(PD) is excitotoxic degeneration of dopaminergic
neurons in the substantia nigra pars compacta,
leading to abnormalities of movement, cognition,
and other cortical functions. Seizures are not com-
mon in PD (Szot, 2012). Therefore, how could the
KD benefit patients with PD? A  primary patho-
genic mechanism of cell death in PD is thought to
be a defect in complex I of the electron transport
chain. The KD might bypass this defect and allow
oxidative phosphorylation to proceed. In a small
clinical study, five of seven PD patients treated
with the KD showed improved scores on a stand-
ard PD rating scale (Vanitallie et al., 2005). Sample
sizes were small, however, and a placebo effect
could not be fully excluded. There is also evidence
that calorie restriction alone helps patients with
PD (Srivastava and Haigis, 2011), and this finding
was verified in a primate model (Maswood et al.,
2004); these observations support a critical role of
energy dysregulation in this disorder.
An animal model of PD is produced by 1-
methyl- 4- phenyl- 1,2,3,6- tetrahydropyridine
(MPTP) administration; this toxin causes death of
substantia nigra neurons by inhibiting mitochon-
drial NADH dehydrogenase (Tieu et al., 2003). β-
hydroxybutyrate, elevated during KD treatment,
ameliorated MPTP-induced mitochondrial respi-
ratory chain damage in this model by bypassing
the defect in complex I  activity. Additional sup-
port for the potential benefits of ketone bodies in
PD comes from in vitro experiments. In substantia
nigra neuron precursors, ketone bodies protect
against mitochondrial respiratory chain dysfunc-
tion induced exogenously by complex I  and II
inhibitors and provide anti-inflammatory action
on MPTP-induced neurotoxicity (Kashiwaya et al.,
2000; Yang and Cheng, 2010). As in ALS, the KD
effect in PD appears to require intact complex
I function (Tieu et al., 2003). In another model of
159
Chapter 18: Dietary Therapy for Neurological Disorders
PD induced by 6-hydroxydopamine neurotoxic-
ity in rats, the KD protected nigral dopaminergic
neurons from degeneration (Cheng et al., 2009).
Mood Disorders
Beneficial effects of the KD have been hypoth-
esized in mood disorders, based on the observa-
tion that other antiseizure agents also stabilize
mood (e.g., valproate, carbamazepine, lamotrig-
ine) (El-Mallakh and Paskitti, 2001; Bialer, 2012).
In addition, the KD alters levels of metabolites of
biogenic amines such as dopamine and seroto-
nin, neurotransmitters critical to the pathophysi-
ology of depression (Murphy et al., 2004; Dahlin
et al., 2012; Szot, 2012). The KD increased levels of
endogenous norepinephrine levels in rats, which
had an antiseizure effect (Weinshenker, 2008).
With respect to mood disorders, only anecdotal
clinical studies have been conducted thus far. One
brief KD trial on a single patient with depression was
reportedly ineffective, although urinary ketosis was
not achieved (Yaroslavsky et al., 2002). Two women
with bipolar disorder type II were treated with a KD
for 2–3  years; both subjects tolerated the diet and
demonstrated improved mood (Phelps et al., 2013).
Obviously, large-scale studies are needed.
Depression in rats is studied using the forced
swim test, in which an animal is placed into a
water-filled chamber with smooth walls, from
which it cannot escape. After an initial phase of
frantic swimming, a rat becomes immobile, sim-
ply floating to prevent drowning. Depressed ani-
mals have a shorter latency to the immobile phase,
regarded as a measure of despair. The role of the
KD in depression has been studied using this
model. Compared with control-fed rats, those fed
the KD spent significantly less time in the immo-
bility phase of the test, reflecting an overall reduc-
tion in “behavioral despair” (Murphy et al., 2004).
In recent experiments, mice were exposed to the
KD prenatally (by feeding dams the KD) and then
their behavior was tested in adulthood (Sussman
et  al., 2015). All rats received the standard (non-
KD) diet after birth. Adult KD offspring exhibited
reduced susceptibility to anxiety and depression
and were physically more active than control mice.
The authors concluded that prenatal exposure to a
KD was associated with decreased stress responses
in adulthood. At present, it is not clear whether
there are any practical or translational implications
of these results.
Migraine
Epilepsy and migraine involve paroxysmal excita-
bility changes in the brain, and many of the same
159
pharmacological agents are used to treat both con-
ditions (e.g., topiramate, valproate, lamotrigine)
(Bialer, 2012). Migraine and epilepsy share con-
siderable phenotypic overlap (Rogawski, 2008),
although the intrinsic mechanisms underlying
seizures and migraine attacks differ in some fun-
damental respects. There are theoretical reasons to
consider the KD for chronic migraine. Migraine
involves a complex interplay between genes and
environment, and some of the causative genes
code for proteins involved in energy metabolism
(e.g., ATP1A2 in one form of familial hemiplegic
migraine; Gritz and Radcliffe, 2013). Brain func-
tion is highly dependent on normal mitochondrial
function to produce sufficient energy via oxida-
tive phosphorylation, so disrupted mitochondrial
function has been hypothesized as a causative fac-
tor in migraine (Roos-Araujo et al., 2014).
It might seem unlikely that an individual
with migraine would undertake such a compli-
cated dietary regimen as the KD, but in light of
suboptimal alternatives, the KD is at least worth
considering in some cases, particularly in per-
sons with headaches that are refractory to conven-
tional agents (Strahlman, 2006; Maggioni et  al.,
2011). Interestingly, the first report using the KD
for migraine appeared in 1928, only a few years
after the diet’s first use for epilepsy (Schnabel,
1928). Nine of 28 migraine patients reported
“some improvement,” although the validity of
this clinical study is uncertain and some patients
admitted to poor compliance. In patients with epi-
lepsy, better compliance is often achieved using
the less restrictive modified Atkins diet (MAD)
(Kossoff and Dorward, 2008). A trial of the MAD
was attempted in adolescents with chronic daily
headache (Kossoff et al., 2010). Of eight enrolled
patients, only three patients were able to com-
plete the 3-month trial. Although there was some
improvement in reported quality of life in those
three patients, all of them continued to experience
chronic daily headaches. These findings suggest
that even the MAD is too restrictive for long-term
use in adolescents with headache and any thera-
peutic benefit of ketosis in treating headache is
quite modest.
One large prospective observational study of
the KD in migraine patients has been reported
(DiLorenzo et al., 2015). Ninety-six obese female
patients with migraine self-referred to a dietitian
for weight control. Patients were randomized to a
KD or standard diet for 1 month. The KD group
experienced a decrease in mean baseline head-
aches from 2.9 per month to 0.71 per month; the
total number of headache days was also fewer in
160
160 section II: E merging Clinical Applications and Future Potential
the KD group (0.91 headache days for the KD
group vs. 5.1 days in controls). When the KD was
stopped and the patients returned to a normal diet,
the higher migraine headache frequency resumed.
A case report of a pair of obese twin sisters with
migraines reported an improvement in headache
control on a KD, and decreased migraine attacks
correlated with the level of ketosis (DiLorenzo
et al., 2013).
Cortical spreading depression (CSD), a wave of
depolarization that spreads slowly along the cor-
tical surface (~3 mm/min) and is thought to cor-
relate with migraine aura (Ferrari et al., 2015), is
associated with neurogenic inflammation, form-
ing a target for current antimigraine drugs (Cui
et al., 2014). Laboratory studies have investigated
the effect of the KD on CSD (de Almeida Rabello
Oliveira et al., 2008). It was found that short-term
treatment (10 days) with either the MCT or long-
chain triglyceride forms of the KD resulted in
significant reduction in CSD velocity in rats, but
long-term treatment (7 weeks) with either diet had
no effect on CSD. Although the clinical implica-
tions of these results are uncertain, it is clear that
the KD reduces cortical excitability in a time-
dependent fashion. It is unknown whether the KD
will play a role in headache treatment in the future,
but the diet may provide some clues as to migraine
pathophysiology. For example, mitochondrial dys-
function, leading to decreased ATP production,
has been linked to the occurrence of CSD (Sparaco
et al., 2006).
CONCLUSIONS
This chapter has considered several neurologic
disorders that might be amenable to metabolic
treatment with therapies such as the KD and its
variants. Despite the relative lack of clinical data
supporting the use of KD in these disorders,
preliminary studies support the idea that KD-
induced metabolic shifts may lead to neuropro-
tection. How can a simple dietary change lead
to improvement in a set of disorders with such
a huge span of pathophysiological mechanisms?
A  common theme appears to be alteration in
energy metabolism. So while the mechanisms
through which the KD exerts such effects are
certainly diverse, there may be a finite set of
final common pathways that are shared mecha-
nistically. Ultimately, the details of how such
metabolic factors reduce excitability, diminish
ongoing neurodegeneration, or mitigate func-
tional disability remain unknown. Herein lie rich
opportunities for further investigation in both
the laboratory and the clinic.
REFERENCES
Ari, C., Poff, A.M., Held, H.E., Landon, C.S., Goldhagen,
C.R., Mavromates, N., and D’Agostino, D.P. (2014).
Metabolic therapy with Deanna Protocol supple-
mentation delays disease progression and extends
survival in amyotrophic lateral sclerosis (ALS)
mouse model. PLoS One 9, e103526.
Barañano, K.W., and Hartman, A.L. (2008). The keto-
genic diet:  uses in epilepsy and other neurologic
illnesses. Curr Treat Options Neurol 10, 410–419.
Bialer, M. (2012). Why are antiepileptic drugs used
for nonepileptic conditions? Epilepsia 53 (Suppl
7), 26–33.
Bough, K.J., Wetherington, J., Hassel, B., Pare, J.F.,
Gawryluk, J.W., Greene, J.G., Shaw, R., Smith,
Y., Geiger, J.D., and Dingledine, R.J. (2006).
Mitochondrial biogenesis in the anticonvulsant
mechanism of the ketogenic diet. Ann Neurol 60,
223–235.
Cheng, B., Yang, X., An, L., Gao, B., Liu, X., and Liu,
S. (2009). Ketogenic diet protects dopaminergic
neurons against 6-OHDA neurotoxicity via up-
regulating glutathione in a rat model of Parkinson’s
disease. Brain Res 1286, 25–31.
Cui, Y., Kataoka, Y., and Watanabe, Y. (2014). Role of
cortical spreading depression in the pathophysiol-
ogy of migraine. Neurosci Bull 30, 812–822.
Dahlin, M., Månsson, J.E., and Åmark, P. (2012). CSF
levels of dopamine and serotonin, but not norepi-
nephrine, metabolites are influenced by the keto-
genic diet in children with epilepsy. Epilepsy Res
99, 132–138.
Danial, N.N., Hartman, A.L., Stafstrom, C.E., and
Thio, L.L. (2013). How does the ketogenic diet
work? Four potential mechanisms. J Child Neurol
28, 1027–1033.
de Almeida Rabello Oliveira, M., da Rocha Ataíde, T.,
de Oliveira, S.L., de Melo Lucena, A.L., de Lira,
C.E., Soares, A.A., de Almeida, C.B., and Ximenes-
da-Silva, A. (2008). Effects of short-term and long-
term treatment with medium- and long-chain
triglycerides ketogenic diet on cortical spreading
depression in young rats. Neurosci Lett 434, 66–70.
DiLorenzo, C., Currà, A., Sirianni, G., Coppola, G.,
Bracaglia, M., Cardillo, A., DeNardis, L., and
Pierelli, F. (2013). Diet transiently improves
migraine in two twin sisters: possible role of keto-
genesis? Funct Neurol 28, 305–308.
DiLorenzo, C., Coppola, G., Sirianni, G., DiLorenzo,
G., Bracaglia, M., DiLenola, D., Siracusano, A.,
Rossi, P., and Pierelli, F. (2015). Migraine improve-
ment during short lasting ketogenesis: a proof-of-
concept study. Eur J Neurol 22, 170–177.
Ding, Y., Wang, S., Jiang, Y., Yang, Y., Zhang, M., Guo,
Y., Wang, S., and Ding, M.P. (2013). Fructose-1,6-
diphosphate protects against epileptogenesis by
modifying cation-chloride co-transporters in a
model of amygdaloid-kindling temporal epilepti-
cus. Brain Res 1539, 87–94.
161
Chapter 18: Dietary Therapy for Neurological Disorders
El-Mallakh, R.S., and Paskitti, M.E. (2001). The keto-
genic diet may have mood-stabilizing properties.
Med Hypotheses 57, 724–726.
Ferrari, M.D., Klever, R.R., Terwindt, G.M., Ayata, C.,
and van den Maagdenberg, A.M. (2015). Migraine
pathophysiology: lessons from mouse models and
human genetics. Lancet Neurol 14, 65–80.
Gano, L.B., Patel, M., and Rho, J.M. (2014). Ketogenic
diets, mitochondria, and neurological diseases. J
Lipid Res 55, 2211–2228.
Garriga-Canut, M., Schoenike, B., Qazi, R., Bergendahl,
K., Daley, T.J., Pfender, R.M., Morrison, J.F.,
Ockuly, J., Stafstrom, C., Sutula, T., et  al. (2006).
2-Deoxy-D-glucose reduces epilepsy progres-
sion by NRSF-CtBP-dependent metabolic regula-
tion of chromatin structure. Nature Neurosci 9,
1382–1387.
Gritz, S.M., and Radcliffe, R.A. (2013). Genetic effects
of ATP1A2 in familial hemiplegic migraine type II
and animal models. Hum Genomics 7, 8.
Hajjar, T., Meng, G.Y., Rajion, M.A., Vidyadaran, S.,
Othman, F., Farjam, A.S., Li, T.A., and Ebrahimi,
M. (2012). Omega-3 polyunsaturated fatty acid
improves spatial learning and hippocampal peroxi-
some proliferator activated receptors (PPARα and
PPARγ) gene expression in rats. BMC Neurosci
13, 109.
Hartman, A.L. (2012). Neuroprotection in metabolism-
based therapy. Epilepsy Res 100, 286–294.
Juge, N., Gray, J.A., Omote, H., Miyaji, T., Inoue, T.,
Hara, C., Uneyama, H., Edwards, R.H., Nicoll,
R.A., and Moriyama, Y. (2010). Metabolic con-
trol of vesicular glutamate transport and release.
Neuron 68, 99–112.
Kashiwaya, Y., Takeshima, T., Mori, N., Nakashima,
K., Clarke, K., and Veech, R.L. (2000). D-beta-
hydroxybutyrate protects neurons in models of
Alzheimer’s and Parkinson’s disease. Proc Natl
Acad Sci USA 97, 5440–5444.
Kim, D.Y., Davis, L.M., Sullivan, P.G., Maalouf, M.,
Simeone, T.A., van Brederode, J., and Rho, J.M.
(2007). Ketone bodies are protective against oxi-
dative stress in neocortical neurons. J Neurochem
101, 1316–1326.
Kim, D.Y., Simeone, K.A., Simeone, T.A., Pandya, J.D.,
Wilke, J.C., Ahn, Y., Geddes, J.W., Sullivan, P.G.,
and Rho, J.M. (2015). Ketone bodies mediate anti-
seizure effects through mitochondrial permeabil-
ity transition. Ann Neurol 78, 77–87.
Kossoff, E.H., and Dorward, J.L. (2008). The modified
Atkins diet. Epilepsia 49 (Suppl 8), 37–41.
Kossoff, E.H., Huffman, J., Turner, Z., and Gladstein,
J. (2010). Use of the modified Atkins diet for ado-
lescents with chronic daily headache. Cephalalgia
30, 1014–1016.
Kovac, S., Abramov, A.Y., and Walker, M.C. (2013).
Energy depletion in seizures: anaplerosis as a strat-
egy for future therapies. Neuropharmacology 69,
96–104.
161
Kunz, W.S. (2002). The role of mitochondria in epilep-
togenesis. Curr Opin Neurol 15, 179–184.
Lian, X.-Y., Khan, F.A., and Stringer, J.L. (2007).
Fructose-1,6-biphosphate has anticonvulsant
activity in models of acute seizures in adult rats. J
Neurosci 27, 12007–12011.
Lutas, A., and Yellen, G. (2013). The ketogenic
diet: metabolic influences on brain excitability and
epilepsy. Trends Neurosci 36, 32–40.
Ma, W., Berg, J., and Yellen, G. (2007). Ketogenic diet
metabolites reduce firing in central neurons by open-
ing K(ATP) channels. J Neurosci 27, 3618–3625.
Maalouf, M.A., Rho, J.M., and Mattson, M.P. (2009).
The neuroprotective properties of calorie restric-
tion, the ketogenic diet, and ketone bodies. Brain
Res Rev 59, 293–315.
Maggioni, F., Margoni, M., and Zanchin, G. (2011).
Ketogenic diet in migraine treatment:  a brief but
ancient history. Cephalalgia 31, 1150–1151.
Martin, L.J. (2011). Mitochondrial pathobiology in
ALS. J Bioenerg Biomembr 43, 569–579.
Masino, S.A., Kawamura, M., Jr., Wasser,
C.D., Pomeroy, L.T., and Ruskin, D.N.
(2009). Adenosine, ketogenic diet and epi-
lepsy:  the emerging therapeutic relationship
between metabolism and brain activity. Curr
Neuropharmacol 7, 257–268.
Maswood, N., Young, J., Tilmont, E., Zhang, Z.,
Gash, D.M., Gerhardt, G.A., Grondin, R., Roth,
G.S., Mattison, J., Lane, M.A., et  al. (2004).
Caloric restriction increases neurotrophic
factor levels and attenuates neurochemical
and behavioral deficits in a primate model of
Parkinson’s disease. Proc Natl Acad Sci USA 52,
18171–18176.
Milder, J., and Patel, M. (2012). Modulation of oxida-
tive stress and mitochondrial function by the keto-
genic diet. Epilepsy Res 100, 295–303.
Murphy, P., Likhodii, S., Nylen, K., and Burnham,
W.M. (2004). The antidepressant properties of the
ketogenic diet. Biol Psychiatry 56, 981–983.
Neal, E.G., Chaffe, H., Schwartz, R.H., Lawson, M.S.,
Edwards, N., Fitzsimmons, G., Whitney, A., and
Cross, J.H. (2008). The ketogenic diet for the treat-
ment of childhood epilepsy:  a randomised con-
trolled trial. Lancet Neurol 7, 500–506.
Paganoni, S., and Wills, A.M. (2013). High-fat and
ketogenic diets in amyotrophic lateral sclerosis. J
Child Neurol 28, 989–992.
Pani, G. (2015). Neuroprotective effects of dietary
restriction: Evidence and mechanisms. Semin Cell
Dev Biol 40, 106–114.
Paoli, A., Bianco, A., Damiani, E., and Bosco, G.
(2014). Ketogenic diet in neuromuscular and neu-
rodegenerative diseases. Biomed Res Int 2014,
article number 474296.
Phelps, J.R., Siemers, S.V., and El-Mallakh, R.S. (2013).
The ketogenic diet for type II bipolar disorder.
Neurocase 19, 423–426.
162
162 section II: E merging Clinical Applications and Future Potential
Rho, J.M., and Stafstrom, C.E. (2012). The ketogenic
diet: what has science taught us? Epilepsy Res 100,
210–217.
Rogawski, M.A. (2008). Common pathophysiologic
mechanisms in migraine and epilepsy. Arch
Neurol 65, 709–714.
Roos-Araujo, D., Stuart, S., Lea, R.A., Haupt, L.M., and
Griffiths, L.R. (2014). Epigenetics and migraine;
complex mitochondrial interactions contributing
to disease susceptibility. Gene 543, 1–7.
Schnabel, T.G. (1928). An experience with a ketogenic
diet in migraine. Ann Int Med 2, 314–347.
Siva, N. (2006). Can ketogenic diet slow progression of
ALS? Lancet Neurol 5, 476.
Sparaco, M., Feleppa, M., Lipton, R.B., Rapoport,
A.M., and Bigal, M.E. (2006). Mitochondrial dys-
function and migraine: evidence and hypotheses.
Cephalalgia 26, 361–372.
Srivastava, S. and Haigis, M.C. (2011). Role of sirtuins
and calorie restriction in neuroprotection:  impli-
cations in Alzheimer’s and Parkinson’s diseases.
Curr Pharm Des 17, 3418–3433.
Stafstrom, C.E., Ockuly, J.C., Murphree, L., Valley,
M.T., Roopra, A., and Sutula, T.P. (2009).
Anticonvulsant and antiepileptic actions of 2-
deoxy-D-glucose in epilepsy models. Ann Neurol
65, 435–447.
Stafstrom, C.E., and Rho, J.M. (2012). The ketogenic
diet as a treatment paradigm for diverse neuro-
logical disorders. Front Pharmacol 3, 1–8.
Strahlman, R.S. (2006). Can ketosis help migraine suf-
ferers? A case report. Headache 46, 182.
Sussman, D., Germann, J., and Henkelman, M. (2015).
Gestational ketogenic diet programs brain struc-
ture and susceptibility to depression and anxiety in
the adult mouse offspring. Brain Behav 5, e00300.
Szot, P. (2012). Common factors among Alzheimer’s
disease, Parkinson’s disease, and epilepsy:  pos-
sible role of the noradrenergic nervous system.
Epilepsia 53 (Suppl 1), 61–66.
Tieu, K., Perier, C., Caspersen, C., Teismann, P., Wu,
D.C., Yan, S.D., Naini, A., Vila, M., Jackson-
Lewis, V., Ramasamy, R., et  al. (2003). D-beta-
hydroxybutyrate rescues mitochondrial respiration
and mitigates features of Parkinson disease. J Clin
Invest 112, 892–901.
Vanitallie, T.B., Nonas, C., DiRocco, A., Boyar,
K., Hyams, K., and Heymsfield, S.B. (2005).
Treatment of Parkinson disease with diet-
induced hyperketonemia:  a feasibility study.
Neurology 64, 728–730.
Vreugdenhil, M., Bruehl, C., Voskuyl, R.A., Kang,
J.X., Leaf, A., and Wadman, W.J. (1996).
Polyunsaturated fatty acids modulate sodium and
calcium currents in CA1 neurons. Proc Nat Acad
Sci USA 93, 12559–12563.
Vucic, S., Rothstein, J.D., and Kiernan, M.C. (2014).
Advances in treating amyotrophic lateral scle-
rosis:  insights from pathophysiological studies.
Trends Neurosci 37, 433–442.
Weinshenker, D. (2008). The contribution of norepi-
nephrine and orexigenic neuropeptides to the anti-
convulsant effect of the ketogenic diet. Epilepsia 49
(Suppl 8), 104–107.
Willis, S., Stoll, J., Sweetman, L., and Borges, K. (2010).
Anticonvulsant effects of a triheptanoin diet in two
mouse chronic seizure models. Neurobio Dis 40,
565–572.
Winesett, S.P., Kordecki, S., and Kossoff, E.H.
(2015). The ketogenic diet in pharmacoresistant
childhood epilepsy. Expert Rev Neurother 15,
621–628.
Wlaź, P., Socała, K., Nieoczym, D., Łuszczki, J.J.,
Zarnowska, I., Zarnowski, T., Czuczwar, S.J.,
and Gasior, M. (2012). Anticonvulsant profile of
caprylic acid, a main constituent of the medium-
chain triglyceride (MCT) ketogenic diet, in mice.
Neuropharmacology 62, 1882–1889.
Yang, X., and Cheng, B. (2010). Neuroprotective and
anti-inflammatory activities of ketogenic diet on
MPTP-induced neurotoxicity. J Mol Neurosci 42,
145–153.
Yaroslavsky, Y., Stahl, Z., and Belmaker, R.H.
(2002). Ketogenic diet in bipolar illness. Bipolar
Dis 4, 75.
Yudkoff, M., Daikhin, Y., Nissim, I., Lazarow, A., and
Nissim, I. (2001). Ketogenic diet, amino acid
metabolism, and seizure control. J Neurosci Res
66, 931–940.
Zhao, W., Varghese, M., Dzhun, P.V.A., Cheng, A.,
Wang, J., Lange, D., Bilski, A., Faravelli, I., and
Pasinetti, G.M. (2012). Caprylic triglyceride as a
novel therapeutic approach to effectively improve
the performance and attenuate the symptoms due
to the motor neuron loss in ALS disease. PLoS One
7, e49191.
Zhao, Z., Lange, D.J., Voustianiouk, A., MacGrogan,
D., Ho, L., Suh, J., Humala, N., Thiyagarajan, M.,
Wang, J., and Pasinetti, G.M. (2006). A keto-
genic diet as a potential novel therapeutic inter-
vention in amyotrophic lateral sclerosis. BMC
Neurosci 7, 29.
163

SECTION III
Ketogenic Diet in the Laboratory

D E T L E V B O I S O N, P H D, S E C T I O N E D I T O R
164
165
19
Overview of Ketogenic Diet in the Laboratory
Progress on Models and Mechanisms
D E T L E V B O I S O N, P H D
I ncreased clinical interest in the ketogenic diet
and similar metabolic treatments has naturally
spurred research into mechanisms underlying
their anticonvulsant/antiepileptogenic efficacy;
significant progress and new insights are being
made after many decades of relative obscurity.
This section, “Ketogenic Diet in the Laboratory,”
includes discussions of a number of molecular
targets that are affected by ketogenic diet feeding.
In parallel, several alternative experimental meth-
ods to either elucidate or mobilize ketogenic diet
mechanisms are explored. Based on this labora-
tory research, putative and novel uses for meta-
bolic treatments are proposed.
The first published reports in the medi-
cal literature of the effect of a ketogenic diet
(KD) on epilepsy are two one-page reports by
R.M. Wilder in the daily Mayo Clinic Bulletin.
The first makes a brief theoretical case for use
of a high-fat, low-carbohydrate diet based on
prior successful work with fasting in epilepsy
(Wilder, 1921a). The second (published the very
next day) consists of case descriptions of three
patients who became seizure-free after KD treat-
ment (Wilder, 1921b). Interestingly, given that
the KD became best known as a treatment for
pediatric epilepsy, two of these three patients
were adults (aged 23 and 31). Naturally, from the
beginning there were questions about the anti-
convulsant mechanism of the KD. Wilder’s own
speculation was that elevated ketone bodies were
crucial, given that this metabolic response is in
common between fasting and eating a high-fat,
low-carbohydrate diet (Wilder, 1921a). This is
still seen as the key mechanism by many inves-
tigators, though not without dispute. Other early
postulated mechanisms such as dehydration and
acidosis have fallen by the wayside. In fact, these
latter two are seen now as unnecessary and treat-
able side effects. One main point of this section,
“Ketogenic Diet in the Laboratory,” is to discuss
recent mechanistic studies.
Early in vivo work in rodents showed that KD
feeding was effective against some but not all types
of experimentally induced seizures (Appleton and
DeVivo, 1974; Millichap et  al., 1964; Uhlemann
and Neims, 1972). Later mechanistic work in ani-
mals turned to in vitro preparations (e.g., Stafstrom
et al., 1999; Thio et al., 2000). Here, Dr. Kawamura’s
chapter reviews in vitro methodologies for mod-
eling the KD. These involve direct application of
ketone bodies, modulation of glucose and intracel-
lular ATP, and maintaining the metabolic milieu of
KD-fed animals in acute brain slices.
Several chapters address anticonvulsant or
antiepileptogenic molecular targets for the KD.
Dr. Simeone reviews the literature on peroxisome
proliferator-activated receptor γ (PPARγ), a fatty
acid receptor that is expressed in many brain neu-
rons. He describes the regulation of this nutri-
ent sensor by seizure activity and by KD feeding,
and that pharmacological or genetic inactivation
of PPARγ blocks the anticonvulsant effect of KD
feeding. Drs. Sada and Inoue review the functions
of the astrocyte/neuron lactate shuttle, and then
discuss their work in brain slices, in which it was
shown that lactate reverses the inhibition caused by
ketone bodies. Conversely, ketogenic diet feeding
lowered hippocampal lactate levels. Drs. Lusardi
and Boison describe the epigenetics of epilepsy,
review the augmentation of the neuromodulator
adenosine by KD treatment, and outline the dual
functions of adenosine: anticonvulsant effects via
modulating synaptic activity and antiepileptogenic
effects via modulating DNA methylation.
Several chapters address mechanistic effects
of KD treatment in pathophysiologies other than
epilepsy. Drs. Curtis, Kemper, et  al. review how
reactive oxygen and nitrogen species are involved
in aging and radiation damage. They then discuss
166
166 section III: Ketogenic Diet in the Laboratory
how ketone body-based metabolism generates
NADPH, thereby quenching these species through
reactions driven by the [NADP+]/[NADPH]
couple and also by increasing transcription of cer-
tain antioxidant enzymes. Also in the context of
aging, Drs. Xu, LaManna, and Puchowicz describe
the beneficial effects of KD treatment on neuro-
degenerative processes. They propose that these
effects arise from ketosis-induced stabilization of
hypoxia-inducible factor 1α, a metabolic sensor
and transcription factor, and its target genes. Drs.
Veech and King review the literature on the patho-
physiology of Alzheimer’s disease and present evi-
dence that ketosis-inducing treatments could be
beneficial in circumventing the inhibition of pyru-
vate dehydrogenase found in that disease.
Additional chapters address other physiologi-
cal effects of the KD. Drs. Banjara and Janigro
review the biology of the blood-brain barrier
including the process of ketone body uptake. The
regulation of glucose and ketone body transport-
ers in the barrier and the suggestion of barrier
repair by KD treatment are discussed. Drs. Harney,
Gudsnuk, et  al. describe delays in puberty and
effects on reproduction in female rats fed a strict
KD. Effects of KD feeding on spatial reference
learning in dams and their pups are also presented.
Three chapters deal with novel uses of KDs that
are mostly (or completely) still confined to the lab-
oratory. Such is the case with Drs. Veech and King
and Drs. Curtis, Kemper, et al. proposing KDs for
treating Alzheimer’s disease and ionizing radia-
tion damage, respectively, both discussed previ-
ously. In addition, Dr. Ruskin reviews the evidence
for KD treatment being able to alleviate pain and
inflammation.
Finally, among these chapters in this sec-
tion many also address treatment stategies and
methodologies to invoke the mechanisms of the
KD with nondietary treatments. Drs. Martínez-
François, Danial, and Yellen review the functions
of BCL-2-associated agonist of cell death (BAD),
a protein that regulates glucose metabolism. They
describe how genetic manipulation of BAD to
reduce glucose metabolism results in mice resist-
ant to seizures, an effect involving ATP-sensitive
potassium channels. Drs. Sada and Inoue pro-
pose lactate dehydrogenase blockers to reduce
activity of the astrocyte/neuron lactate shuttle;
Dr.  Simeone proposes agonists of PPARγ; Drs.
Lusardi and Boison propose blockers of adeno-
sine kinase to boost adenosine and decrease DNA
methylation; and Drs. Veech and King and Drs.
Curtis, Kemper, et  al. propose administration of
ketone esters, which are rapidly metabolized to
ketone bodies.
REFERENCES
Appleton, D.B., and DeVivo, D.C. (1974). An ani-
mal model for the ketogenic diet. Epilepsia 15,
211–227.
Millichap, J.G., Jones, J.D., and Rudis, B.P. (1964).
Mechanism of anticonvulsant action of ketogenic
diet. Am J Dis Child 107, 593–604.
Stafstrom, C.E., Wang, C., and Jensen, F.E. (1999).
Electrophysiological observations in hippocam-
pal slices from rats treated with the ketogenic diet.
Dev Neurosci 21, 393–399.
Thio, L.L., Wong, M., and Yamada, K.A. (2000). Ketone
bodies do not directly alter excitatory or inhibitory
hippocampal synaptic transmission. Neurology
54, 325–331.
Uhlemann, E.R., and Neims, A.H. (1972).
Anticonvulsant properties of the ketogenic diet in
mice. J Pharmacol Exp Ther 180, 231–238.
Wilder, R.M. (1921a). The effects of ketonemia on the
course of epilepsy. Mayo Clin Bull 2, 307.
Wilder, R.M. (1921b). High fat diets in epilepsy. Mayo
Clin Bull 2, 308.
167
20
Ketogenic Diet and PPARgamma
TIMOTHY A. SIMEONE, PHD
INTRODUCTION
Historical Overview
A switch from a normal Western diet of low-fat
and high-carbohydrate/protein consumption to
a high-fat and low-carbohydrate/protein keto-
genic diet (KD) necessitates an adjustment in
the metabolic machinery to handle new primary
fuel sources. As detailed elsewhere in this vol-
ume, there are three prominent biochemical con-
sequences of a KD:  decreased glucose, increased
free fatty acids, and increased ketone bodies (i.e.,
beta-hydroxybutyrate, acetoacetate, and acetone).
The lower glucose and higher free fatty acids, of
course, directly result from the formulation of the
diet. The ketone bodies are a partial breakdown
product from using fatty acids for fuel in a low-
glucose environment, and themselves can be used
as an efficient, deliverable fuel source. Ketone bod-
ies skip glycolysis and enter the tricarboxylic acid
cycle at the level of acetyl-CoA.
In order for fatty acids to be used for fuel
and for ketogenesis to occur, a series of enzymes
and proteins must be present and have increased
expression. These include, but are not limited to,
beta-oxidation enzymes such as long chain acetyl-
CoA synthetase I  (ACS1); carnitine palmitoyl-
transferase (CPT), which commits palmitate to
mitochondrial entry; and the ketogenic enzyme
mitochondrial 3-hydroxy-3-methylglutaryl-CoA
synthase (HMGCS2). For animals that are on a
ketogenic diet, this mainly occurs in the periphery
in hepatocytes of the liver, but there is evidence
that astrocytes are capable of local beta-oxidation
and ketogenesis (Auestad et al., 1991; Cullingford
et  al., 2002a; Cullingford et  al., 2002b; Guzmán
and Blázquez, 2004;Takahashi et al., 2014).
There are two immediate implications from
these findings. One is that dramatic dietary
changes in fuel sources somehow coordinate the
expression of large numbers of genes. The second
implication is that there is(are) nutrient sensor(s)
that regulate this shift in gene expression. In hepa-
tocytes, fatty acids are endogenous ligands for
peroxisome proliferator activated receptor alpha
(PPARalpha), a nuclear receptor transcription fac-
tor that regulates genes involved in beta-oxidation
and ketogenesis. Logically, it has been hypoth-
esized that the KD engages PPARalpha in the
brain, providing increased ability to use fatty acids
to produce ATP and ketone bodies, which may be
critically important for KD-mediated neuropro-
tective and antiseizure effects. Indeed, PPARalpha
agonists exert antiseizure effects in acute seizure
models (Porta et al., 2009; Puligheddu et al., 2013).
However, the KD also exhibits anti-inflammatory
and antioxidant properties and promotes mito-
chondrial health, mechanisms with ever-growing
empirical support for their importance in con-
trolling seizures in refractory epilepsy. An alter-
native nutrient-sensing transcription factor that
regulates genes in these pathways, as well as genes
involved in beta-oxidation and ketogenesis, is
PPARgamma. Remarkably, PPARgamma, like the
KD, is under intense investigation for therapeu-
tic potential in Alzheimer’s disease, Parkinson’s
disease, Huntington’s disease, amyotrophic lateral
sclerosis, multiple sclerosis, stroke, and cancer
(reviewed in Heneke and Landreth, 2007; Collino
et  al., 2008; Mandrekar-Colucci and Landreth,
2011; Dineley et al., 2014; Johri et al., 2014; Carta
and Simuni, 2015; Corona and Duchen, 2015;
Shen et  al., 2015). In this chapter, we review
PPARgamma function and modulation and the
evidence for PPARgamma involvement in the KD
mechanism of action.
What Are Peroxisome Proliferator
Activated Receptors?
Peroxisome proliferator activated receptors
(PPARs) are ligand-inducible transcription factors
that belong to the superfamily of nuclear recep-
tors (NR), which comprises 48 transcription fac-
tors including receptors for endogenous steroid
168
168 section III: Ketogenic Diet in the Laboratory
hormones, thyroid hormone, lipophilic vitamins,
and cholesterol metabolites (Burris et  al., 2013).
Functional diversity between the 48 NRs is pro-
vided by distinct preferences for both ligands and
DNA sequences recognized as response elements
within the genome. In most cases the NRs func-
tion as dimers, either homodimers or heterodi-
mers, and bind response elements in the absence
and presence of ligand. The NR binding in the
absence of ligand may have passive or active (i.e.,
constitutive basal transcription or silencing) con-
sequences on target gene regulation (Burris et al.,
2013). As a group, NRs are a significant and impor-
tant therapeutic target for human disease, which is
illustrated by the fact that 10%–15% of US Food
and Drug Administration-approved drugs target
NRs (Overington et al., 2006).
There are three different PPAR isoforms,
PPARalpha (also known as NR1C1), PPARbeta
(also known as delta and NR1C2), and
PPARgamma (also known as NR1C3), which are
encoded by genes located on human chromosomes
22, 6, and 3, respectively (Collino et al., 2008). The
PPARs are involved in several aspects of devel-
opment, such as differentiation of adipose tissue,
brain, placenta, and skin; and the three isoforms
display unique tissue- and temporal-dependent
expression patterns during development in cell
types having ectodermal, mesodermal, and endo-
dermal embryonic origins (Desvergne and Wahli,
1999). Postnatally, the three isoforms exhibit con-
siderable tissue-specific expression. PPARalpha
is highly expressed in tissues that catabolize fatty
acids (e.g., hepatocytes, cardiomyocytes, entero-
cytes, and kidney proximal tubule cells). PPARbeta
is ubiquitously expressed in all cell types with
higher levels in proliferating and differentiating
cells. PPARgamma is strongly expressed in adi-
pose tissue and the immune system. All PPARs
have roles in lipid and glucose metabolism and
homeostasis, inflammation, cell proliferation and
differentiation, and vascular biology (Braissant
et al., 1996; Tontonoz and Spiegelman, 2008); how-
ever, a role in peroxisome proliferation is species-
dependent and it remains unclear whether PPARs
are involved in this function in humans (Schrader
et al., 2015).
Members of the nuclear hormone receptor
superfamily contain four domains (Figure 20.1A).
The N-terminal transactivation domain (activation
function 1, AF1) is involved in ligand-independent
transactivation, interaction with transcriptional
coactivators and influences ligand-binding affin-
ity of the ligand binding domain (LBD). The AF1
sequence has minimal homology between PPAR
isoforms, which underlies the differential biolog-
ical function of the isoforms. The DNA binding
domain (DBD) contains two highly conserved
zinc fingers. There is a hinge region between the
DBD and the C-terminal LBD, which is required
for receptor dimerization. The LBD contains the
ligand-dependent transactivation domain AF2
and is less homologous between the isoforms,
resulting in isoform specificity and sensitivity to
ligands (Glass, 1994; Kersten and Wahli, 2000).
In the nucleus, PPARs form obligate heterodimers
with retinoid X receptors (RXR).
Of the three isoforms PPARgamma is the most
extensively studied. Two PPARgamma isoforms,
PPARgamma1 and PPARgamma2, result from
alternative splicing and differential promoter use
(Tontonoz et  al., 1994). The PPARgamma iso-
forms are identical except that PPARgamma2
contains an additional 28 amino acids at its N-
terminus that convey a 5- to 10-fold more effec-
tive ligand-independent transactivation and
increased ligand binding affinity to the LBD rel-
ative to PPARgamma1 (Werman et al., 1997; Shao
et al., 1998; Castillo et al., 1999; Bugge et al., 2009).
The expression of PPARgamma1 appears ubiqui-
tous. In contrast, PPARgamma2 is restricted to
adipose tissue; however, high-fat diets can induce
the expression of PPARgamma2 (Vidal-Puig et al.,
1996, 1997).
PPARgamma activation is initiated by ligand
binding. The PPARgamma ligand binding pocket
is large (1,300 angstroms3), which allows struc-
tural promiscuity for a wide variety of endogenous
or natural agonists (i.e., unsaturated fatty acids,
eicosanoids, oxidized lipids, nitroalkenes), syn-
thetic agonists (e.g., thiazolidinediones or TZDs),
and synthetic anatagonists (currently there are
no known endogenous antagonists) (Itoh et  al.,
2008; Fong et al., 2010; Kroker and Bruning, 2015;
Sauer, 2015). The nature of PPARgamma’s large
LBD makes it an effective sensor and transducer
of environmental nutritional and inflammatory
states.
Ligand binding induces a conformational
change in the receptor leading to dissociation of
corepressors such as NCoR (nuclear receptor core-
pressor 1) and SMRT (silencing mediator of reti-
noic acid and thyroid hormone receptor), which
normally keep basal levels of PPAR-mediated tran-
scription low. Without corepressor complexes, the
PPARgamma-RXR heterodimer is free to bind spe-
cific recognition sequences called PPAR-response
elements (PPREs) in the promoter regions of target
genes. Once corepressors disengage, coactivators
are recruited and alter chromatin structure and
169

Chapter 20: Ketogenic Diet and PPARgamma 169

PPARγ1
(a) 477aa
N AF-1 AF-2 C
52kDa
PPARγ2
505aa
N AF-1 AF-2 C
Ligand DNA Hinge Ligand
56kDa
Independent Binding Binding
Activation Domain Domain
Domain

(b)

Agonist
Coactivator Complexes
CBP/p300
SRCs
Corepressor Complexes Corepressor Complexes
PGC-1α
PPARγ RXR PPARγ RXR

Coactivator Complexes

PPARγ RXR
Corepressor Complexes
NCoR
SMRT
PPRE

FIGURE 20.1 Illustrations of PPARgamma splice variants and activity. (A) PPARgamma has two splice variants that
have identical primary sequences except that the N-terminal of PPARgamma 2 has an extension of 28 amino acids.
PPARgamma1 and PPARgamma2 have an activation function 1 domain (AF-1) required for ligand-independent
activation, a DNA-binding domain required for sequence-specific binding to genomic DNA at a peroxisome proliferator
response element (PPRE), a Hinge domain required for receptor dimerization, a ligand-binding domain required for
ligand-dependent modulation and an activation function 2 domain (AF-2) within the ligand-binding domain that is
required for ligand-dependent activation, ligand-dependent dimerization, coactivator recruitment, and corepressor
release. The extra 28 amino acids in the AF-1 of PPARgamma2 confers 5- to 10-fold ligand-independent activation and
greater ligand binding affinity relative to PPARgamma1. (B) PPARgamma forms a heterodimer with the retinoic acid
receptor (RXR) and in the absence of agonist it is bound to corepressor complexes (e.g., NCoR and SMRT). Once an
agonist is bound, the corepressors are released, PPARgamma-RXR recruits coactivators (e.g., CBP/p300, SRCs and PGC-
1alpha), and the complex binds to PPREs in the promoter region of target genes leading to initiation of transcription by
RNA polymerase II and general transcription factors.

recruit transcriptional machinery to promote the homeostasis, insulin-sensitization, fluid homeo-

commencement of transcription (Figure  20.1b).
Coactivators of PPARgamma include CREB bind-
ing protein (CBP), SRC1/2/3 (steroid receptor
coactivator), and PPARgamma coactivator 1alpha
(PGC-1alpha) (Auwerx, 1999; Katsouri et al., 2012;
Kroker and Bruning, 2015). The PPARgamma-
RXR-corepressor or -coactivator complexes are
further regulated by several posttranslational
modifications such as phosphorylation, acetyla-
tion, SUMOylation, and ubiquitination, which can
increase or decrease PPARgamma transcriptional
activity (Ahmadian et al., 2013).
PPARgamma controls hundreds of genes, with
many involved in lipid and glucose metabolism and
stasis, anti-inflammation, antioxidant, and mito-
chondrial health. Its best-known role is as a master
regulator of adipogenesis, where it coordinates
gene expression necessary for adipocyte formation
(Auwerx, 1999; Fong et al., 2010; Ahmadian et al.,
2013; Krocker and Bruning, 2015). The insulin-
sensitizing effects of TZDs such as pioglitazone
and rosiglitazone are clinically useful in the treat-
ment of metabolic disorders such as type II diabe-
tes mellitus. However, these full agonists can lead
to weight gain, fluid retention, and bone loss due
to PPARgamma activation in a wide range of tis-
sues (Ahmadian et al., 2013). Several strategies to
reduce side effects are under investigation, ranging
170
170 section III: Ketogenic Diet in the Laboratory
from partial agonism to tissue-specific activation
(Sujii and Evans, 2011).
Many of the PPARgamma regulated genes
overlap with KD-responsive genes. However,
determining which genes PPARgamma regulates
at any one time is not straightforward. An addi-
tional level of complexity is that the gene sets regu-
lated by PPARgamma depend on tissue-dependent
differential expression of types of corepressors and
coactivators, the state of biochemical cascades
that regulate post-translational modifications, and
the nature of the bound agonist (full or partial),
which determines the degree of conformational
change and ability of PPARgamma to interact
with particular corepressors and coactivators
(Ahmadian et  al., 2013; Krocker and Bruning,
2015; Sauer, 2015). For example, the full agonist
TZDs exert beneficial insulin-sensitizing effects
as well as detrimental side effects such as weight
gain, whereas several partial agonists or so-called
selective PPARgamma modulators (SPPARMs)
have been shown to retain antidiabetic effects
while eliminating weight gain via attenuated and
selective gene-regulatory activity in comparison
with full agonists (Carmona et al, 2007; Tan et al.,
2012). Additionally, phosphorylation of Ser273
in the LBD of PPARgamma dysregulates a group
of distinct genes resulting in insulin-insensitivity.
TZDs prevent Ser273 phosphorylation, but their
effectiveness in producing insulin-sensitization in
humans has been found to be inversely related to
the degree of PPARgamma phosphorylation (Choi
et  al., 2010). This implies that PPARgamma acti-
vation may differentially regulate gene sets based
on the state of the tissue (e.g., normal brain vs.
oxidatively stressed and inflamed epileptic brain).
This is an intriguing prospect that warrants further
investigation.
P PA R G A M M A I N
THE BRAIN
The majority of knowledge concerning
PPARgamma has been gained from study of its
function in peripheral tissues and diseases as is
evident in the previous sections. In the brain,
much less is known. PPARgamma has differen-
tial regional expression patterns suggesting spe-
cialized and complex roles in both cellular and
network function. It seems that PPARgamma has
roles in maintaining homeostasis and protection
against injury. In particular, PPARgamma acti-
vation in the brain consistently results in three
general effects: prevention of mitochondrial apop-
tosis signaling pathways, suppression of inflamma-
tory mediators, and limitation of reactive oxygen
species. In this section, we will discuss the location
and function of PPARgamma in the brain and its
potential roles in epilepsy and the KD.
Where Is PPARgamma in the Brain?
PPARgamma is regionally distinct with high
expression in neurons, astrocytes, and microglia in
the cortex and moderate expression in neurons in
the hippocampus (Cullingford et al., 1998; Lu et al.,
2011; Moreno et al., 2004; Sarruf et al., 2009; Zhao
et al., 2009). Moreno et al. (2004) have performed
the most extensive study to date, cataloging the
expression patterns throughout the brain of not
only PPARgamma but also PPARalpha, PPARbeta
and the three RXR isoforms. PPARgamma has
the most restrictive distribution in the CNS. For
example, PPARgamma is absent from neurons in
the olfactory bulb, the perirhinal and entorhinal
cortices, the temporal and occipital neocortices,
the thalamic reticular nucleus, the ventral teg-
mental area, and the cerebellar Purkinjie cells. It
is highly expressed in neurons of the basal gan-
glia, thalamic rhomboid, centromedial and para-
fascicular nuclei, nuclei of the reticular formation,
and cerebellar stellate, basket, and golgi cells.
PPARgamma has moderate to weak expression in
neurons of the hypothalamic nuclei, septohippo-
campal nucleus, and the hippocampal formation.
Oligodendrocytes do not express PPARgamma,
but it is present in some astrocytes (Moreno et al.,
2004). These descriptions are in broad agreement
with studies using neuron-specific PPARgamma
knockout mice (Sarruf et al., 2009; Lu et al., 2011).
Sarruf et  al. (2009) confirmed moderate expres-
sion in the hypothalamus but found expression
in the ventral tegmental area in contrast to the
previous study. They also reported that neuronal
loss of PPARgamma resulted in 90% reduction of
PPARgamma mRNA in brain, indicating that most
PPARgamma in the brain is expressed in neurons
(Sarruf et al., 2009). A subsequent study found that
in normal mice, PPARgamma mRNA was greatest
in cortex, followed by cerebellum, hippocampus,
diencephalon, and hypothalamus (Lu et al., 2011).
Knockout of PPARgamma from neurons reduced
mRNA in all CNS regions except cerebellum.
The reduction was greatest in the hippocampus
and least in cortex (Lu et  al., 2011). If the resid-
ual mRNA is from expression in glia, than we can
infer that hippocampal glia express minimal levels
of PPARgamma.
Only one study has attempted to discern differ-
ential expression of the two PPARgamma isoforms
in the hippocampal CA1 region of mice (Gahring
et  al., 2005). The PPARgamma antibody, which
171
Chapter 20: Ketogenic Diet and PPARgamma
recognizes both isoforms, stained essentially all
cells of the CA1 pyramidal layer and some associ-
ated cells including presumed astrocytes. However,
using an antibody that recognized an epitope
within the extra 28 N-terminal amino acids of
PPARgamma2, it was found that only a small sub-
set of neurons associated with the pyramidal cell
layer were stained, suggesting that PPARgamma1
is the primary isoform in the hippocampus. The
PPARgamma2 positive cells lacked astrocytic
morphology, but did colabel with nicotinic ace-
tylcholine receptors alpha4 and beta4 subunits;
thus, the authors concluded that PPARgamma2
was exclusively expressed in putative interneurons
(Gahring et al., 2005). This intriguing finding war-
rants confirmation and expansion to other brain
regions.
What Is the Function of PPARgamma
in the Brain?
The role of PPARgamma in the brain during nor-
mal conditions remains largely unknown, but
most likely involves minimal activity in order
to maintain lipid and glucose homeostasis.
PPARgamma may also have a role in neuronal
development and neurogenesis. PPARgamma is
highly expressed in mouse embryonic brain and
neural stem cells (NSCs) compared to the rela-
tively low levels expressed in adult mouse brain
(Wada et al., 2006), and may have differential roles
in NSC differentiation during development and
adulthood. Activation of PPARgamma in embry-
onic neural stem cells promotes NSC proliferation
and inhibits NSC differentiation. In contrast, the
ever-interesting compound cannabidiol (CBD)
increased adult rat hippocampal neurogenesis via
interaction with a PPARgamma pathway (Espisito
et al., 2011). PPARgamma has also been implicated
in promoting neurite outgrowth and dendritic
spine density (Brodbeck et  al., 2008; Quintanilla
et al., 2013).
PPARgamma’s primary action appears to be
neuroprotective in environments that involve
inflammation and oxidative stress, such as dur-
ing Alzheimer’s disease, Parkinson’s disease,
Huntington’s disease, amyotrophic lateral sclero-
sis, multiple sclerosis, and stroke. There are mul-
tiple reviews detailing PPARgamma effects in each
of these neurodegenerative disorders, to which
the reader is referred (Heneke and Landreth,
2007; Collino et  al., 2008; Mandrekar-Colucci
and Landreth, 2011; Dineley et  al., 2014; Johri
et al., 2014; Carta and Simuni, 2015; Corona and
Duchen, 2015; Shen et  al., 2015). Here, an over-
view of PPARgamma’s major effects that most
171
likely contribute to neuroprotection is briefly
presented.
PPARgamma has low expression in the brain
and is primarily in neurons, however, in in vitro
and in vivo models of various neurodegenera-
tive disorders PPARgamma expression increases
in neurons, astrocytes, and microglia (Kitamura
et  al., 1999; Diab et  al., 2002; Victor et  al., 2006;
Fong et al., 2010; Wang et al., 2012). It is hypothe-
sized that the inflammatory response and increase
in oxidative stress results in the generation of fatty
acid metabolites such as eicosanoids, oxidized lip-
ids, and nitroalkenes, which are all potent endog-
enous PPARgamma agonists (Figure  20.2a; Fong
et  al., 2010). In a positive feed-forward loop,
PPARgamma increases expression of itself and
coactivators such as PGC-1alpha (Figure  20.2b).
Experimental evidence has proven that this
endogenous neuroprotective mechanism is impor-
tant in limiting damage, because when it is absent
mice experience significantly more brain damage
and oxidative stress in response to an insult such
as middle cerebral artery occlusion (Victor et al.,
2006; Zhao et al., 2009).
The neuroprotective properties of
PPARgamma activation in all of the above-
mentioned neurodegenerative disorders have
consistently been found to involve three general
areas:  prevention of mitochondrial apoptosis
signaling pathways (Figure  20.2c), suppression
of inflammatory mediators (Figure  20.2d), and
limitation of reactive oxygen species (ROS;
Figure  20.2e). Mitochondrial membrane poten-
tial depends on several factors, one important
one being the balanced presence of anti-apoptotic
Bcl-2 and Bcl-xl and pro-apoptotic Bad and Bax.
Upon injury, Bad and Bax translocate to mito-
chondria, bind Bcl-2 and Bcl-xl, and depolarize
mitochondria. Significant depolarization of the
mitochondrial membrane potential can lead to
the release of pro-apoptotic factors such as cyto-
chrome C, caspase 9, and caspase 3 (Youle and
Strasser, 2008). PPARgamma activation increases
neurotrophic alpha-1 transcription, which leads
to activation of Akt signaling pathways, resulting
in increased Bcl-2 expression and phosphoryla-
tion of Bad (p-Bad). PPARgamma activation also
up-regulates 14-3-3epsilon expression, which
sequesters p-Bad and prevents Bad transloca-
tion into mitochondria (Wu et  al., 2009a; Wu
et al., 2009b; Fuenzalida et al., 2007; Thouennon
et al., 2015).
PPARgamma activation suppresses pro-
inflammatory gene expression by inhibiting the
transcription factor NFkappaB. PPARgamma
172

172 section III: Ketogenic Diet in the Laboratory

PUFAs (a)

FA metabolites: Energy Failure:

Eicosanoids ATP
PPARγ Oxidized lipids [Ca2+]i
Nitroalkenes ROS

RXR

Coactivator Complexes
(e)
PPARγ RXR
(c) (d) (c)
SOD/
catalase/ Mito
PPARγ PGC1α 14-3-3ε IκB
GSH/ Biogenesis
PPRE UCP2
(b)

14-3-3ε/P-Bad NFκB ROS ATP

Antioxidant Neuroprotection

Apoptosis Cytokine/
Inflammation

Neuroprotection Anti-Inflammatory

FIGURE 20.2 Schematic illustration depicting hypothetical homeostatic neuroprotection by endogenous PPARgamma

agonists during situations of energy failure experienced during epilepsy, stroke, and other neurodegenerative disorders.
(a)  During extended periods of high oxidative stress, metabolites of fatty acids (FA) are generated that have higher
affinities for PPARgamma than their parent lipids, resulting in an increased activation of PPARgamma. (b) In a positive
feed-forward loop, PPARgamma up-regulates expression of itself and the coactivator PGC-1alpha. Neuroprotection
is achieved through multiple pathways. (c)  PPARgamma decreases apoptosis by increasing the expression of 14-3-
3epsilon, which prevents the translocation of p-Bad into the mitochondria and the release of pro-apoptotic factors
and it increases mitochondrial biogenesis, which provides neurons increased ATP and additional calcium buffering.
(d)  Inflammation is dampened by PPARgamma-mediated inhibition of the pro-inflammatory transcription factor
NFκB and (e) oxidative stress is controlled by up-regulating the expression of antioxidants such as superoxide dismutase
(SOD), catalase, glutathione (GSH), and uncoupling protein 2 (UCP2).

accomplishes this by increasing the expression
of IkappaBalpha and IkappaBbeta, which pre-
vent the nuclear translocation and DNA binding
of NFkappaB (Heneka et al., 2003). PPARgamma
may also inhibit NFkappaB activity by direct
physical interaction, steric inhibition, or cofac-
tor competition (Sauer et  al., 2015). The end
effect is that PPARgamma activation decreases
inflammatory cytokine production (TNF-alpha,
interleukin-6, INFgamma), iNOS and COX2
expression, and astrogliosis and microglial activa-
tion (Combs et al., 2000; Heneka et al., 2000; Lim
et al., 2000 Breidert et al., 2002; Yan et al., 2003;
Dehmer et  al., 2004; Sundararajan et  al., 2005;
Zhao et al., 2005).
Cellular injury and inflammation can increase
the production of ROS, which further causes cel-
lular damage by oxidizing proteins, lipids, and
DNA. This results in altered neuronal signaling,
further neuroinflammation, and the initiation of
apoptosis. PPARgamma activation limits ROS by
increasing the antioxidant genes SOD, catalase,
glutathione, and uncoupling protein 2 (UCP2)
(Chen et  al., 2006; Doonan et  al., 2009; Girnun
et al., 2002; Zhao et al., 2006; Diano et al., 2011).
As mentioned previously, PPARgamma increases
PGC-1alpha. PGC-1alpha also acts as a coactivator
of the oxidative stress-induced transcription factor
Nrf2, which is a master regulator of antioxidant
gene sets as well as mitochondrial biogenesis (St.
Pierre et  al., 2006; Milder et  al., 2010; Clark and
Simon, 2009). Deficiency in PPARgamma is linked
to reduced expression of antioxidants (superoxide
dismutase 1, catalase, glutathione S-transferase,
uncoupling protein-1, UCP1), lipid metabolism
enzymes (lipoprotein lipase), and the transcription
factor liver X receptor-alpha (Victor et  al., 2006;
Zhao et al., 2009).
173
Chapter 20: Ketogenic Diet and PPARgamma
K E TO G E N I C D I E T
R E G U L AT I O N
O F   P PA R G A M M A
PPARgamma in Epilepsy
PPARgamma was first cloned from Xenopus lae-
vis and mouse liver in the early 1990s (Dreyer
et al., 1992; Zhu et al., 1993; Elbrecht et al., 1996).
Since then, thousands of studies have addressed
PPARgamma’s structure/function, regulation,
pharmacology, expression, and importance in dis-
ease primarily in peripheral organs. At the turn
of the 21st century, neuroscientists began inves-
tigating PPARgamma’s therapeutic potential for
neurodegenerative diseases, in which a role for
neuroinflammation and oxidative stress in disease
processes has long been recognized (Kitamura
et al., 1999; Combs et al., 2000) and has been inves-
tigated in hundreds of publications.
Interest in the epilepsy community has had
a late and slow start. The first research study
appeared in 2006, and the total count to date is
a little over a dozen (Okada et al., 2006; Luna-
Medina et al., 2007; Maurois et al., 2008; Yu et
al., 2008; Hong et al., 2008, 2011, 2013; Abdallah,
2010; Han et al., 2011; Jeong et al., 2011; Adabi
Mohazab et al., 2012; Chuang et al., 2012; Hughes
et al., 2014; Boes et al., 2015; Simeone et al., 2016).
These studies span acute seizure models, post–
status epilepticus (SE) models, and kindling and
genetic models of chronic epilepsy and, for the
most part, consistently support beneficial neuro-
protective and antiseizure effects of PPARgamma
agonists in epilepsy.
Comparable to findings in chronic neurode-
generative disorders and acute stroke models, the
expression of brain PPARgamma increases sub-
sequent to SE induced by lithium-pilocarpine,
unilateral intrahippocampal kainic acid, intra-
peritoneal injection of kainic acid, and electri-
cally induced self-sustaining SE (Yu et al., 2008;
Hong et al., 2008; Jeong et al., 2011; Chuang et
al., 2012; Boes et al., 2015). Concomitantly, the
products of the PPARgamma-regulated genes
Pgc-1alpha and Ucp2 are increased after lith-
ium-pilocarpine and intrahippocampal kainic
acid (Han et al., 2011; Chuang et al., 2012). In a
recent study, we examined the nuclear content of
both splice variants, PPARgamma1 and gamma2,
in the brains of wild-type mice and epileptic lit-
termates that lack the Kv1.1 potassium channel
(Kcna1-null). Kcna1-null mice develop severe
spontaneous recurrent seizures (SRS) and are
a model of temporal lobe epilepsy and sudden
unexpected death in epilepsy (SUDEP) (Smart
173
et al., 1998; Wenzel et al., 2007; Fenoglio-Simeone
et al., 2009a; Fenoglio-Simeone et al., 2009b;
Glasscock et al., 2010; Simeone et al., 2013;
Simeone et al., 2014a; Simeone et al., 2014b;
Kim et al., 2015; Simeone et al., 2016). Similar to
previous reports, we found that PPARgamma1
predominated in wild-type brain. In contrast,
in epileptic Kcna1-null brain we found that
PPARgamma2 was the dominant form. The total
nuclear PPARgamma did not change, but the
PPARgamma2/gamma1 ratio increased three-
fold in Kcna1-null brains (Simeone et al., 2016).
Further studies are needed to determine the cause
of this flip in isoform nuclear content. Potential
causes that preferentially enhance nuclear trans-
location of PPARgamma2 over PPARgamma1
may be generation of isoform specific ligands,
altered expression of isoform-specific cofac-
tors, or post-translational modifications. Altern-
atively, transcription, translation, or splicing of
PPARgamma2 could be increased in epilepsy.
Recently, it was found that at the initial stage of
adipogenesis another nuclear receptor, gluco-
corticoid receptor (GR), is transiently recruited
along with the transcription factor C/EBPbeta to
a complex consisting of PBP/MED1/TRAP220
and p300 to enhancer regions of the Pparg2 iso-
form. In response to glucocorticoids, this results
in a transient increase in H3K9 acetylation
and enhances the induction of PPARgamma2,
which becomes the principal driver of adipo-
genesis (Steger et al., 2010). Intriguingly both
C/EBPbeta and glucocorticoids increase in the
brain with seizures (Lu et al., 2013; Engel et al.,
2013; Maguire and Salpekar, 2013). PGC-1alpha
activation of PPARgamma also enhances inter-
actions with p300/CBP (Puigserver et al., 1999).
Whether PPARgamma2 regulates distinct gene
sets is unclear, but it has been shown to up-reg-
ulate catalase expression to a greater degree than
PPARgamma1 and is important in providing pro-
tection against lipotoxicity (Medina-Gomez et
al., 2007a; Medina-Gomez et al., 2007b; Yakunin
et al., 2014).
Assuming that the seizure/injury-induced
changes in PPARgamma expression are part of
an endogenous neuroprotective mechanism that
limits damage, similar to what has been proposed
for other neurodegenerative disorders, then loss of
PPARgamma should exacerbate markers of injury
and possibly seizures. Chuang et al. (2012) found
that pretreatment with bilateral focal injections of
the PPARgamma antagonist GW9662 (150 nl of a 12
mM solution or 1.5-2 mg/kg) reduces UCP2; exac-
erbates intrahippocampal kainic acid SE-induced
174
174 section III: Ketogenic Diet in the Laboratory
increases in ROS, oxidized proteins, mito-
chondrial Bax, cytosolic cytochrome C, and
DNA fragmentation; and decreases mitochon-
drial respiratory complex I (MRCI) activ-
ity. Unfortunately, the authors failed to report
the details of SE, so it is not known whether
PPARgamma antagonism worsened the SE
(Chuang et al., 2012). We administered GW9662
(1 mg/kg/day in the drinking water for 2 weeks)
to Kcna1-null mice and wild-type littermates,
however, contrary to expectations, inhibiting
PPARgamma did not increase SRS in Kcna1-
null mice, nor did it lower seizure threshold in
wild-type mice (Simeone et al., 2016). A possi-
ble explanation for the lack of effect on Kcna1-
null seizures may be that the dose of GW9662
was too low. Supporting this hypothesis, we
found that GW9662 had no effect on the nuclear
PPARgamma2/gamma1 ratio in nonepileptic
wild-type brain. It did reduce the PPARgamma2/
gamma1 ratio in epileptic Kcna1-null brain, but
only from threefold to twofold of wild-type ratio
(Simeone et al., 2016). A greater reduction may
be needed to observe an effect on SRS. We also
found that seizure thresholds of Pparg2-null
mice and neuron specific-PPARgamma knockout
mice were no different than control littermates
(Simeone et al., 2016). From these experimen-
tal results we can draw the tentative conclusions
that PPARgamma itself is somehow involved in
increasing nuclear PPARgamma2 and that the
seizure-induced increase affords some neuropro-
tection against injury, but may not raise the sei-
zure threshold per se.
PPARgamma Agonism in Epilepsy
The PPARgamma agonists pioglitazone and rosi-
glitazone have been clinically useful in the man-
agement of type II diabetes mellitus since 1999,
but as of today there are no published reports,
case reports or otherwise, describing epilepsy
patients with comorbid type II diabetes that have
been prescribed pioglitazone or rosiglitazone.
Nevertheless, experimental evidence from ani-
mal studies supports that PPARgamma agonists
will have antiseizure, neuroprotective, and pos-
sibly disease-modifying effects in epilepsy. Even
though this field of research is small and there
are relatively few studies, there is wide diver-
sity in treatment protocol (agonist, dose, pre/
post-treatment) and endpoints (acute seizures,
SRS, inflammatory markers, cell death, cogni-
tive comorbidities), and it is worth detailing the
similarities and differences to gain a sense of the
robustness of PPARgamma effects.
PPARgamma Agonist Pretreatment Raises
Thresholds in Acute Seizure Models
Abdallah (2010) and Adabi Mohazab et al. (2012)
utilized penelenyltetrazole (PTZ) to induce sei-
zures in mice with two different paradigms.
Abdallah (2010) pretreated mice with pioglitazone
(10 mg/kg; p.o.) 30 minutes prior to an intraperi-
toneal (i.p.) injection of PTZ and measured latency
to stage 4/5 clonic-tonic seizures. All vehicle con-
trols reached stage 4/5, whereas only 20% of the
pioglitazone group reached stage 4/5 and those
that did took twice as long as the control group.
Adabi Mohazab et al. (2012) injected PTZ intrave-
nously and measured the threshold dose required
to induce a whole-body clonic seizure. They found
that a 1-h pretreatment with pioglitazone was inef-
fective, whereas a 4-h pretreatment provided a
dose-dependent (10–80 mg/kg; p.o.) increase in
the seizure threshold dose. Maurois et  al. (2008)
also found that the PPARgamma agonist FMOC-
L-leucine ((N-[9-fluorenylmethoxycarbonyl]-)-
L-leucine; up to 100 mg/kg; i.p.) was not effective
in the 6-Hz seizure test with 1-h pretreatment. We
have found similar effects exposing mice to fluro-
thyl gas and measuring latency to clonus and gen-
eralized tonic-clonic seizures. A 1-h pretreatment
with pioglitazone had inconsistent effects, whereas
a 4-h pretreatment dose-dependently (1–80 mg/
kg; i.p.) increased the latency to generalized tonic-
clonic seizures but had no effect on clonic seizures
(Simeone et al., unpublished observations).
PPARgamma Agonists Raise Thresholds
in Seizure Susceptible and Kindling Models
The EL mice develop susceptibility to stress-
induced tonic-clonic seizures beginning 12 weeks
postnatally. Providing pioglitazone (20 mg/kg/
day) from 5 to 11 weeks postnatally delayed the
age of onset and duration of stress-induced sei-
zures (Okada et  al., 2006). Maurois et  al. (2008)
used a dietary-induced magnesium deficiency-
dependent audiogenic seizure model. Mice were
pretreated with either FOMC-L-leucine or rosigli-
tazone (up to 100 mg/kg; i.p.) 1 h prior to acous-
tic stimuli. FOMC-L-leucine dose-dependently
reduced the number of mice responding to acous-
tic stimuli by 50%, whereas rosiglitazone was inef-
fective. The FOMC-L-leucine effects were blocked
by coadministration of the PPARgamma antago-
nist GW9662 (1 and 2 mg/kg, i.p.) (Maurois et al.,
2008). Finally, Abdallah (2010) kindled mice with
daily i.p. injections of subconvulsive PTZ (40 mg/
kg) until the mice exhibited stage 4/5 clonic-tonic
seizures on three consecutive days. Pretreatment
with pioglitazone (10 mg/kg; p.o.) 30 minutes prior
175
Chapter 20: Ketogenic Diet and PPARgamma
to each PTZ injection resulted in only 30% of mice
reaching stage 4/5 seizures by day 17 (Abdullah,
2010). However, because pioglitazone was admin-
istered before PTZ and most pioglitazone-treated
mice did become kindled as defined by the author,
it is difficult to determine whether pioglitazone
exerted any antiepileptogenic effect in addition to
its anticonvulsant effect.
PPARgamma Agonists Reduce SRS
and Improve Spatial Learning
Multiple studies have used SE models to later
investigate neuropathology during the latent
period and after development of SRS (discussed
in the next section), however, only three studies
have provided information concerning the effect
of PPARgamma agonists on either seizures during
SE and/or the development of SRS (Luna-Medina
et al., 2007; Hong et al., 2013; Boes et al., 2015).
Luna-Medina et al. (2007) pretreated rats with a
thiadiazolidine compound, NP031112 (50 mg/kg;
p.o.), 1 hour before kainic acid injection (10 mg/
kg; i.p.) and monitored the development of behav-
ioral seizures and SE. NP031112 had no effect on
SE. Similarly, Hong et al. (2013) found that rosi-
glitazone had no effect on the development of
lithium-pilocarpine SE. Rats were pretreated with
rosiglitazone (0.1 mg/kg; i.c.v.) 1 hour before lith-
ium chloride injection (127 mg/kg; i.p.). Twenty-
four hours later rats were again pretreated 1 hour
prior to pilocarpine hydrochloride (30 mg/kg; i.p.)
administration. Rats then received repeated injec-
tions of pilocarpine (10 mg/kg; i.p.) every 30 min
until they developed convulsive seizures, which
were scored behaviorally according to the Racine
scale. Rosiglitazone had no effect on SE. The
authors continued daily administration of rosigli-
tazone and examined development of SRS by per-
forming video-EEG recordings beginning 2 weeks
post-SE for 6 weeks. They found that 50% of rosi-
glitazone-treated rats developed SRS compared
with 83% of vehicle-treated rats. Moreover, rosigli-
tazone nearly doubled the latency to SRS develop-
ment and decreased the frequency and duration
of SRS (Hong et al., 2013). An earlier study by
this group demonstrated that the same protocol
improved the SE rats’ spatial learning as assessed
in the Morris water maze (Hong et al., 2011). In
contrast, rosiglitazone (3 mg/kg; i.p.) administered
to rats daily immediately following the end of self-
sustained SE had no effect on the development
and occurrence of SRS; however, the rosiglitazone-
treated rats did have improved spatial learning in
the Morris water maze (Boes et al., 2015). Finally,
we have found that administration of pioglitazone
175
(10 mg/kg; i.p.) for 5 days decreases SRS and sei-
zure severity in epileptic Kcna1-null mice by 75%
(Simeone et al., 2016).
PPARgamma Agonists Are Neuroprotective
post-SE and during SRS
Within hours to days, intrahippocampal kainic
acid-induced SE results in increases in UCP2, ROS,
oxidized proteins, mitochondrial Bax, cytosolic
cytochrome c, DNA fragmentation, and decreases
in MRCI activity in rat hippocampi (Chuang et al.,
2012). Pretreatment with bilateral focal injections
of rosiglitazone (150 nL of a 4 mM solution or
0.7–0.8 mg/kg) further increased UCP2 and pre-
vented all other changes. In fact, MRCI activity
was actually improved compared with the sham
controls (Chuang et al., 2012). In a similar study,
Luna-Medina et  al. (2007) found that intrahip-
pocampal injection of kainic acid results in sig-
nificant cerebral edema, hippocampal cell loss,
astrogliosis, and the induction of TNFalpha in
astrocytes, microglia, and neurons. Coinjection of
NP031112 (5–8 ng/kg) prevented edema, cell loss,
and TNFalpha induction and reduced astrogliosis.
Furthermore, pretreatment with NP031112 (50
mg/kg; p.o.), 1 hour before kainic acid injection
(10 mg/kg; i.p.) also prevented hippocampal neu-
ronal loss as quantified 72 hours post-SE (Luna-
Medina et al., 2007).
In the rat lithium-pilocarpine SE model,
post-SE daily administration of rosiglitazone
has antioxidant and anti-inflammatory effects.
Rosiglitazone increases hippocampal neuronal
survival, decreases ROS and lipid peroxidation,
increases SOD and GSH, and decreases microglia
and astrocytic activation, TNF-alpha, and BDNF
(Yu et  al., 2008; Hong et  al., 2008, 2012, 2013).
Similarly in EL mice and the PTZ-kindling model,
pioglitazone decreased TNF-alpha, IL-1beta, IL-6,
IL-10, and caspase3 (Okada et al., 2006; Abdallah,
2010). However, rosiglitazone did not reduce hip-
pocampal cell death in the self-sustained SE model
(Boes et al., 2015).
Evidence for Ketogenic Diet Activation
of PPARgamma
Although the possibility of a role for PPARgamma
in the mechanism of the KD has been briefly
speculated in review articles for the past several
years (Kobow et al., 2012; Masino and Rho, 2012;
Gano et al., 2014), there have only been two stud-
ies published exploring this possibility, one using
an acute SE model and one using a chronic epi-
lepsy model and acute seizure threshold model
(Jeong et al., 2011; Simeone et al., 2016). Jeong
176
176 section III: Ketogenic Diet in the Laboratory
et al. (2011) fed wild-type mice a KD (4:1 ratio
of fats to carbohydrate+protein) for 3 weeks and
found a threefold increase of PPARgamma pro-
tein in cell lysate from hippocampal tissue com-
pared with standard diet (SD)-fed mice. The
mice were injected with kainic acid (30 mg/kg) to
induce status and tissue collected at 2 and 6 hours
post–kainic acid injection. The KD delayed the
onset of generalized tonic-clonic seizures, but the
overall seizure burden that the mice experienced
during the 2 and 6 hours post–kainic acid was not
reported. In the SD-fed mice, there was a fourfold
and twofold increase of PPARgamma at the 2- and
6-hour time points after injection, respectively. In
contrast, the KD-induced elevated PPARgamma
levels fell to normal diet control levels by 6 hours
post–kainic acid. Hippocampal TNF-alpha fol-
lowed the same pattern as PPARgamma, whereas
the KD blunted the SE-induced increases in NF-
kappaB, microglial CD11b, and COX2 expres-
sion. From the studies mentioned previously
detailing the effect of injury or degeneration
and PPARgamma agonists on the expression of
PPARgamma, one would expect that PPARgamma
would be further increased, not decreased, by SE
in the KD-fed mice. The authors do not address
this discrepancy (Jeong et al., 2011). Nevertheless,
the authors do present evidence that the KD does
have effects on PPARgamma and downstream
markers of inflammation known to be regulated
by PPARgamma.
In a recent study, we aimed to determine
whether the KD changes PPARgamma expres-
sion in the brains of epileptic and normal mice,
and, if so, whether it is necessary for the antisei-
zure effects of the KD. As described earlier, we
have found that in nuclear extracts from wild-
type mouse brain, PPARgamma1 is the domi-
nant isoform, whereas PPARgamma2 is dominant
in epileptic Kcna1-null brain (Simeone et al.,
2016). Treating Kcna1-null mice for 2 weeks with
a KD (6:1 ratio) reduced seizure incidence and
severity by ~75%, as we have reported previ-
ously (Fenoglio-Simeone et al., 2009b; Kim et al.
2015; Simeone et al., 2016). Ketogenic diet treat-
ment increased PPARgamma2 in both genotypes,
resulting in PPARgamma2/gamma1 ratios that
were twofold and sixfold for wild-type and Kcna1-
null brain, respectively, compared with SD-fed
wild-type mice.
To determine whether the effect of the KD
on PPARgamma was necessary for KD anti-
seizure efficacy, we performed pharmacologic
and genetic experiments. Coadministering the
PPARgamma antagonist GW9662 (1 mg/kg/day)
in the drinking water for the 2-week KD treatment
prevented the increase in nuclear PPARgamma2
and prevented KD-mediated seizure reduction in
Kcna1-null mice. To further test the importance of
PPARgamma in the KD mechanism, we obtained
Pparg2-null mice and conditional neuron-specific
PPARgamma knockout mice. Previous studies
have demonstrated that KD treatment raises the
threshold for flurothyl seizures (Rho et al., 1999;
Dutton et al., 2011). Similarly, we found that KD-
treatment effectively raised seizure thresholds of
control mice. However, KD-treatment was unable
to raise the seizure threshold of Pparg2-null mice
and conditional neuron-specific PPARgamma
knockout mice. Furthermore, the pharmaco-
logic and genetic manipulations of PPARgamma
expression/function did not affect the stereo-
typic KD increase of blood beta-hydroxybutyr-
ate or decrease of glucose (Simeone et al., 2016).
Results from this study strongly support that cen-
tral actions of PPARgamma, more specifically
PPARgamma2, play a critical role in the antisei-
zure mechanism of the KD.
What Does the KD provide That Could Be
Activating PPARgamma?
The selective increase of nuclear PPARgamma2
over PPARgamma1 may be due to the additional
30 amino acids in the PPARgamma2 AF1 that
convey a 5- to 10-fold more effective ligand-
independent transactivation and increased
ligand binding affinity to the LBD relative to
PPARgamma1 (Werman et  al., 1997; Shao et  al.,
1998; Castillo et  al., 1999; Bugge et  al., 2009).
PPARgamma2 is the only PPARgamma isoform
regulated at the transcriptional level by nutrition
(Medina-Gomez et al., 2007b). The PPARgamma2
expanded AF1 also confers differential interac-
tion with transcriptional cofactors and post-
translational modifications that would most
likely result in tissue-specific differences in the
regulation of gene sets by the PPARgamma splice
variants. This is evident in the periphery, where
PPARgamma2, but not PPARgamma1, is induced
during high-fat diets and initiates adipogenesis,
increases lipid-buffering, and reduces lipotoxicity
(Medina-Gomez et al., 2007a). Therefore, the most
likely mechanisms for this isoform-specificity of
the KD probably involve either regulation of sig-
naling cascades responsible for post-translational
modifications that contribute to selective nuclear
translocation or transcription of PPARgamma2
and/or providing a PPARgamma ligand, possibly
selective for PPARgamma2.
177
Chapter 20: Ketogenic Diet and PPARgamma
The KD provides plenty of fat, of course, and
unsaturated fatty acids, such as omega-3 and
omega-6 long chain polyunsaturated fatty acids,
are notably increased in blood serum of patients
(Fraser et al., 2003). Importantly, long chain pol-
yunsaturated fatty acids and their metabolites,
eicosanoids, oxidized lipids, and nitroalkenes are
all natural ligands for PPARgamma (Yamamoto
et al., 2005; Fong et al., 2010). Also, it was recently
determined that the saturated fatty acid decanoic
acid (a.k.a. capric acid), a primary constituent of
the medium chain triglyceride ketogenic diet, is a
ligand for PPARgamma at physiologically relevant
concentrations (Malapaka et al., 2012).
In vivo treatment suggests that decanoic acid
is a selective PPARgamma modulator (i.e., par-
tial agonist) as it improves glucose sensitivity
and lipid profiles without weight gain in diabetic
mice (Malapaka et  al., 2012). In vitro experi-
ments in hippocampal slices found that decanoic
acid decreases PTZ-induced epileptiform activ-
ity in a concentration-dependent manner (Chang
et  al., 2013). Furthermore, decanoic acid-induced
increases in citrate synthase, catalase, and MRCI
activity in SH-SY5Y neuronal cultures were inhib-
ited by a PPARgamma antagonist (Hughes et  al.,
2014). Alternatively, Jeong et al. (2012) found that
treatment of cultured HT22 cells (a hippocampal
neuronal cell line) with the ketone body acetoace-
tate (5 mM) increases PPARgamma over a 12-hour
period; however, we have been unable to replicate
this finding in primary hippocampal neuronal cul-
tures, and in fact this concentration of acetoacetate
results in significant neuronal death within 24 and
48 hours (Simeone et  al., unpublished observa-
tions). Therefore, at the moment it seems that the
unsaturated and saturated fatty acids provided by a
ketogenic diet may be the ligands for PPARgamma,
and that PPARgamma may be involved in the anti-
seizure mechanism of the various formulations
of the ketogenic diet regardless of the type of fat
content.
FUNCTIONAL
CONSEQUENCES
O F   P PA R G A M M A
A C T I VAT I O N R E L E VA N T
TO   E P I L E P S Y
PPARgamma agonists and the KD regulate simi-
lar anti-inflammatory, antioxidant and promito-
chondrial pathways. These include, but are not
limited to, up-regulation of IkappaB; inhibition
of NFkappaB; reduction of cytokines such as
IL-1beta, IL-6, and TNF-alpha; up-regulation of
genes encoding mitochondrial enzymes involved
177
in oxidative phosphorylation (e.g., multiple sub-
units of complexes I, II, IV, and V); induction
of mitochondrial biogenesis; and up-regulation
of UCP2, catalase, and glutathione (Masino and
Rho, 2012; Mandrekar-Colucci et al., 2013; Fong
et  al., 2010; Bernardo et  al., 2006; Heneka and
Landreth, 2007; Chuang et al., 2012; Hong et al.,
2008, 2012, 2013; Abdallah, 2010; Adabi Mohazab
et al., 2012; Bough et al., 2006; Miglio et al., 2009;
Sullivan et  al., 2004; Yang and Cheng, 2010; Yu
et al., 2008). All of these have been suggested as
possible disease modifying targets for epilepsy.
Mitochondria are of particular interest in tempo-
ral lobe epilepsy. In human epilepsies and animal
chemoconvulsant seizure models, brain tissue
from seizure foci exhibit mitochondrial reac-
tive oxygen species (ROS) and oxidative damage,
reduced ATP-producing complex I  activity and
diminished antioxidant systems (Sullivan et  al.,
2003; Wallace, 1999; Bruce and Baudry, 1995;
DiMauro et  al., 1999; Kunz et  al., 2000; Mueller
et al., 2001; Sudha et al., 2001; Malthankar-Phatak
et  al., 2006; Vielhaber et  al., 2008; Kudin et  al.,
2009; Ryan et al., 2012; Simeone et al., 2014a; Kim
et al., 2015; Simeone et al., unpublished observa-
tion). Oxidative stress and mitochondrial over-
load likely contribute to the neuronal cell loss
in severe epilepsy associated with sclerosis. In
this circumstance, it is clear that the proposed
actions of PPARgamma would be neuroprotec-
tive and increase the number of surviving cells.
But this illustrates only one outcome of unhealthy
mitochondria and does not address the potential
consequences that chronic mitochondrial dys-
function has on neuronal hyperexcitability and
seizure severity.
Several elegant studies have demonstrated that
mitochondria regulate synaptic transmission via
three properties: (1) production of ATP and (2) ROS
and (3)  sequestration of cytosolic calcium (Lee
et al., 2007; Lee et al., 2012; Harris et al., 2012). Thus,
any perturbation of mitochondrial health will send
ripples of dysregulation across synaptic, neuronal,
and network activity. Stabilizing synaptic mito-
chondria could be another mechanism by which
PPARgamma and the KD reduce seizure activity.
To test this hypothesis, we turned once again to
the Kcna1-null mouse model of epilepsy, because
we have found mitochondrial pathology similar
to reports from human epilepsies (Simeone et al.,
2014a; Kim et al., 2015; Simeone et al., unpublished
observations). Mitochondria isolated from cortical
and hippocampal tissue from epileptic Kcna1-null
mice have reduced ATP-producing MRCI-driven
respiration, increased ROS, and decreased UCP2
178
178 section III: Ketogenic Diet in the Laboratory
(Simeone et al., 2014a). The decrease of UCP2, a
mitochondrial protein with downstream antiox-
idant effects, may contribute to the rise in ROS.
The MRCI inhibition appears to be due to post-
translational inhibition by elevated ROS because
acute addition of antioxidants such as ascorbic
acid restores Kcna1-null MRCI-driven respiration
to wild-type levels, whereas exogenous H2O2 mim-
ics the MRCI dysfunction of Kcna1-null mito-
chondria (Simeone et al., 2014a). In addition, the
mitochondrial membrane potential is depolarized
and mitochondrial calcium sequestration capacity
is reduced (Simeone et al., 2014a; Kim et al., 2015;
Simeone et al., unpublished observation).
Thus, the three properties of mitochondria
that play a role in synaptic transmission (ATP,
ROS, and calcium) are dysfunctional in Kcna1-
null mitochondria and predict widespread synap-
tic and network activity. This is exactly what we
observed using a multielectrode array to record
extracellular potentials from in vitro hippocam-
pal slices from Kcna1-null brains (Simeone et al.,
2013). We found that Kcna1-null hippocampi
generate spontaneous network activity originat-
ing in the CA3 region in the form of sharp waves
(SPWs) and ripples (80–200 Hz bandwidth) with
an increased incidence and duration compared
with slices from wild-type hippocampi. Also
present in Kcna1-null hippocampi were epilepsy-
associated pathologic high-frequency oscillations
in the fast ripple bandwidth (200–600 Hz), which
can be viewed as biomarkers of hyperexcitable
networks. Furthermore, Kcna1-null CA3 has
enhanced coupling of excitatory inputs and popu-
lation spike generation and CA3 principal cells
have reduced spike-timing reliability. Removing
the influence of granule cell mossy fiber path
inputs by microdissecting the Kcna1-null CA3
region mostly rescued the network oscillatory
behavior and improved spike timing. We found
that Kcna1-null mossy fibers are hyperexcit-
able and reduced paired pulse ratios, suggesting
increased neurotransmitter release at these termi-
nals (Simeone et al., 2013). Collectively, these data
indicate enhanced synaptic release in the Kcna1-
null CA3 region reduces spike timing precision of
individual neurons leading to disorganization of
network oscillatory activity and promotion of the
emergence of fast ripples.
This synaptic phenotype is mimicked by acute
application of the MRCI inhibitor rotenone to
wild-type hippocampal slices. Experimental inhi-
bition of MRCI results in increased SPWs, ripples,
and fast ripples (Simeone et  al., 2014a; Simeone
et  al., unpublished observations). Moreover,
rotenone exacerbated provoked seizure-like events
in vitro, supporting a possible role in worsening in
vivo seizures (Simeone et al., 2014a).
We have further demonstrated that in vivo
treatment of Kcna1-null mice with either pio-
glitazone or a KD or a cocktail of ascorbic acid,
alpha-tocopherol, and pyruvate designed to tar-
get mitochondrial health rescues MRCI respira-
tion, restores mossy fiber neurotransmitter release
probabilities, and dampens network hyperexcit-
ability (i.e., reduces incidence of SPWs, ripples,
and fast ripples) (Simeone et  al., 2014a; Simeone
et al., 2014b; Simeone et al., unpublished observa-
tions). Therefore, we speculate that in Kcna1-null
mice one important antiseizure mechanism result-
ing from KD modulation of PPARgamma is res-
toration of MRCI function, possibly by increasing
endogenous antioxidant pathways and decreasing
ROS. Recently, Chuang et  al. (2012) implicated
just such a pathway involved in the neuroprotec-
tive effects of rosiglitazone in the intrahippocam-
pal kainate model of status epilepticus. Specifically,
rosiglitazone increased UCP2, decreased ROS, and
restored MRCI function.
CONCLUDING REMARKS
The development of epilepsy necessarily originates
with a precipitating event (e.g., genetic predisposi-
tion, traumatic brain injury, virus, etc.) that lowers
the seizure threshold or moves the baseline activ-
ity of the brain closer to the seizure threshold so
that previously nonsignificant internal or external
stimuli are capable of synchronizing large net-
works into high-frequency oscillatory activity. If
the resulting seizure phenotype is severe, than a
chronic inflammatory and oxidative state develops
with concomitant mitochondrial dysfunction. The
result is not only death of vulnerable cell types but
also dysregulation of cellular, synaptic, and net-
work excitability, which further lowers the seizure
threshold and exacerbates the seizure phenotype
(Figure 20.3, left).
Experimental evidence indicates that the KD
prevents cell death and hyperexcitability by promot-
ing mitochondrial health and anti-inflammatory
and antioxidant pathways. Limited, but strong,
evidence that we have reviewed here suggests that
the KD regulates PPARgamma to achieve these
effects (Figure 20.3, right). This discovery presents
multiple opportunities for both researchers and
clinicians. The immediate translational poten-
tial resides in the availability of the PPARgamma
agonists pioglitazone and rosiglitazone that are
FDA-approved for therapeutic use in type II dia-
betes mellitus. The TZDs are not perfect and can
179

Chapter 20: Ketogenic Diet and PPARgamma 179

Diet
Precipitating Event
ic
Ketogen
(e.g., genetics, TBI, stroke, virus)

ATP
FFA
[Ca++]i
ROS
SFAs
PUFAs +

MITO Neuronal, Synaptic, Network

PPARγ2
FUNCTION HYPEREXCITABILITY

nction
ndrial Fu
[Ca++]i
Mitocho ROS

SEIZURES
ant
Antioxid

n
ammatio
Anti-Infl
CELL INFLAMMATION
DEATH (e.g., TNF-α)

FIGURE 20.3 Schematic illustration of the proposed mechanisms of action of PPARgamma2 activation underlying
the antiseizure efficacy of the ketogenic diet. (Right) A depiction of the interactions of oxidative stress, inflammation,
and mitochondrial dysfunction that may participate in worsening seizure severity and genesis. Seizure genesis originates
from a precipitating event such as a genetic predisposition, injury, stroke, or virus that results in neuronal, synaptic,
and network hyperexcitability. The ictal events lead to increases in reactive oxygen species (ROS) production and
intracellular calcium, which can overload cellular buffering capacities. Seizures can also initiate the release of pro-
inflammatory cytokines that can further increase ROS and calcium. The ROS and excess calcium cause dysregulation of
mitochondrial function most notably at complex I, resulting in decreased ATP generation, increased ROS production,
mitochondrial membrane depolarization, and further increase in cytosolic calcium as mitochondria lose their capacity
to sequester calcium. Extreme mitochondrial damage will lead to release of pro-apoptotic factors and cell death, which
will increase inflammatory processes. Chronic mitochondrial dysfunction in neurons will dysregulate neuronal,
synaptic, and network excitability and exacerbate the precipitating event-induced hyperexcitability. (Left) The ketogenic
diet increases free fatty acids. Whether this is an increase in polyunsaturated fatty acids such as docahexaenoic acid and
eicosapentaenoic acid or medium chain saturated fatty acids such as decanoic acid depends on the formulation of the
diet. Nevertheless, this increase in fatty acids represents increases in natural agonists for the nuclear transcription factor
PPARgamma. In particular, enrichment of nuclear PPARgamma2 is important for ketogenic diet antiseizure efficacy.
Upregulation of gene sets involved in promoting mitochondrial function, antioxidation, and anti-inflammation may
contribute to not only the KD’s neuroprotective properties but also the KD’s ability to dampen hyperexcitabilty.

have significant side effects. The good news is that
the diabetes academic and industry communi-
ties have active research efforts into developing
PPARgamma partial agonists (SPPARMs) with
improved side-effect profiles. These may also prove
useful in epilepsy. Furthermore, additional tar-
gets with therapeutic potential in epilepsy may be
identified by focusing on the upstream pathways
or ligands leading to PPARgamma activation and
the downstream effectors that result in increasing
the seizure threshold and/or dampening hyperex-
citability and neuroprotection.
The identification of PPARgamma as a poten-
tial mediator of KD efficacy in epilepsy does not
preclude the other mechanisms discussed in this
volume. Some may work synergistically with
PPARgamma, such as ketone bodies, and others
may be downstream effectors of PPARgamma,
such as adenosine.
AC K N OW L E D G M E N T
This work was supported by an award from Citizens
United for Research in Epilepsy Foundation
and the NIH (NS085389). I  thank Dr.  Kristina
180
180 section III: Ketogenic Diet in the Laboratory
A.  Simeone for many insightful discussions and
comments on this manuscript.
REFERENCES
Abdallah, D.M. (2010). Anticonvulsant potential
of the peroxisome proliferator-activated recep-
tor γ agonist pioglitazone in pentylenetetrazole-
induced acute seizures and kindling in mice. Brain
Res 1351, 246–253.
Adabi Mohazab, R., Javadi-Paydar, M., Delfan, B., and
Dehpour, A.R. (2012). Possible involvement of
PPAR-gamma receptor and nitric oxide pathway
in the anticonvulsant effect of acute pioglitazone
on pentylenetetrazole-induced seizures in mice.
Epilepsy Res 101, 28–35.
Ahmadian, M., Suh, J.M., Hah, N., Liddle, C., Atkins,
A.R., Downes, M., and Evans, R.M. (2013). PPARγ
signaling and metabolism: the good, the bad and
the future. Nat Med 19, 557–566.
Auestad, N., Korsak, R.A., Morrow, J.W., and Edmond,
J. (1991) Fatty acid oxidation and ketogenesis by
astrocytes in primary culture. J Neurochem 56,
1376–1386.
Auwerx, J. (1999). PPARgamma, the ultimate thrifty
gene. Diabetologia 42, 1033–1049.
Bernardo, A., and Minghetti, L. (2006). PPAR-
gamma agonists as regulators of microglial acti-
vation and brain inflammation. Curr Pharm Des
12, 93–109.
Boes, K., Russmann, V., Ongerth, T., Licko, T.,
Salvamoser, J.D., Siegl, C., Potschka, H. (2015).
Expression regulation and targeting of the peroxi-
some proliferator-activated receptor γ following
electrically-induced status epilepticus. Neurosci
Lett 604, 151–156.
Bough, K.J., Wetherington, J., Hassel, B., Pare, J.F.,
Gawryluk, J.W., Greene, J.G., Shaw, R., Smith,
Y., Geiger, J.D., and Dingledine, R.J. (2006).
Mitochondrial biogenesis in the anticonvulsant
mechanism of the ketogenic diet. Ann Neurol 60,
223–235.
Braissant, O., Foufelle, F., Scotto, C., Dauça, M.,
and Wahli, W. (1996). Differential expression
of peroxisome proliferator-activated receptors
(PPARs): tissue distribution of PPAR-alpha, -beta,
and -gamma in the adult rat. Endocrinology 137,
354–366.
Breidert, T., Callebert, J., Heneka, M.T., Landreth, G.,
Launay, J.M., and Hirsch, E.C. (2002). Protective
action of the peroxisome proliferator-activated
receptor-gamma agonist pioglitazone in a mouse
model of Parkinson’s disease, J Neurochem 82,
615–624.
Brodbeck, J., Balestra, M. E., Saunders, A. M., Roses,
A. D., Mahley, R. W., and Huang, Y. (2008).
Rosiglitazone increases dendritic spine density and
rescues spine loss caused by apolipoprotein E4 in
primary cortical neurons. PNAS 105, 1343–1346.
Bruce, A.J., Baudry, M. (1995). Oxygen free radicals
in rat limbic structures after kainate-induced sei-
zures. Free Radic Biol Med 18, 993–1002.
Bugge, A., Grontved, L., Aagaard, M.M., Borup, R.,
and Mandrup, S. (2009). The PPARgamma2 A/
B-domain plays a gene-specific role in transactiva-
tion and cofactor recruitment. Mol Endocrinol 23,
794–808.
Burris, T.P., Solt, L.A., Wang, Y, Crumbley, C.,
Banerjee, S., Griffett, K., Lundasen, T., Hughes,
T., and Kojetin, D.J. (2013). Nuclear receptors
and their selective pharmacologic modulators.
Pharmacol Rev 65, 710–778.
Carmona, M.C., Louche, K., Lefebvre, B., Pilon, A.,
Hennuyer, N., Audinot-Bouchez, V., Fievet, C.,
Torpier, G., Formstecher, P., Renard, P., et  al.;
Consortium of the French Ministry of Research
and Technology. (2007). S 26948:  a new specific
peroxisome proliferator activated receptor gamma
modulator with potent antidiabetes and antiath-
erogenic effects. Diabetes 56, 2797–2808.
Carta, A.R., and Simuni, T. (2015). Thiazolidinediones
under preclinical and early clinical development
for the treatment of Parkinson’s disease. Expert
Opin Investig Drugs 24, 219–227.
Castillo, G., Brun, R.P., Rosenfield, J.K., Hauser, S., Park,
C.W., Troy, A.E., Wright, M.E., and Spiegelman,
B.M. (1999). An adipogenic cofactor bound by the
differentiation domain of PPARgamma. EMBO J
18, 3676–3687.
Chang, P., Terbach, N., Plant, N., Chen, P.E., Walker,
M.C., and Williams, R.S. (2013). Seizure control
by ketogenic diet-associated medium chain fatty
acids. Neuropharmacology 69, 105–114.
Chen, S.D., Wu, H.Y., Yang, D.I., Lee, S.Y., Shaw, F.Z.,
Lin, T.K., Liou, C.W., and Chuang, Y.C. (2006).
Effects of rosiglitazone on global ischemia-induced
hippocampal injury and expression of mitochon-
drial uncoupling protein 2. Biochem Biophys Res
Commun 351, 198–203.
Choi, J.H., Banks, A.S., Estall, J.L., Kajimura, S.,
Boström, P., Laznik, D., Ruas, J.L., Chalmers,
M.J., Kamenecka, T.M., Blüher, M., et  al. (2010).
Anti-diabetic drugs inhibit obesity-linked phos-
phorylation of PPARgamma by Cdk5. Nature 466,
451–456.
Chuang, Y.C., Lin, T.K., Huang, H.Y., Chang, W.N., Liou,
C.W., Chen, S.D., Chang, A.Y.W., and Chan, S.H.H.
(2012). Peroxisome proliferator-activated receptors
gamma/mitochondrial uncoupling protein 2 signal-
ing protects against seizure-induced neuronal cell
death in the hippocampus following experimental
status epilepticus. J Neuroinflammation 9, 184.
Clark, J., and Simon, D.K. (2009). Transcribe to sur-
vive: transcriptional control of antioxidant defense
programs for neuroprotection in Parkinson’s dis-
ease. Antioxid Redox Signal 11, 509–528.
Collino, M., Patel, N.S., and Thiemermann, C. (2008).
PPARs as new therapeutic targets for the treatment
181
Chapter 20: Ketogenic Diet and PPARgamma
of cerebral ischemia/reperfusion injury. Ther Adv
Cardiovasc Dis 2, 179–197.
Combs, C.K., Johnson, D.E., Karlo, J.C., Cannady, S.B.,
and Landreth, G.E. (2000). Inflammatory mecha-
nisms in Alzheimer’s disease:  inhibition of beta
amyloid-stimulated proinflammatory responses
and neurotoxicity by PPARgamma agonists. J
Neurosci 20, 558–567.
Corona, J.C., and Duchen, M.R. (2015). PPARγ and
PGC-1α as therapeutic targets in Parkinson’s.
Neurochem Res 40, 308–316.
Cullingford, T.E., Bhakoo, K., Peuchen, S., Dolphin,
C.T., Patel, R., and Clark, J.B. (1998). Distribution
of mRNAs encoding the peroxisome proliferator-
activated receptor alpha, beta, and gamma and
the retinoid X receptor alpha, beta, and gamma
in rat central nervous system. J Neurochem 70,
1366–1375.
Cullingford, T.E., Dolphin, C.T., and Sato, H. (2002a).
The peroxisome proliferator-activated receptor
alpha-selective activator ciprofibrate upregu-
lates expression of genes encoding fatty acid
oxidation and ketogenesis enzymes in rat brain.
Neuropharmacology. 42, 724–730.
Cullingford, T.E., Eagles, D.A., and Sato, H. (2002b).
The ketogenic diet upregulates expression of the
gene encoding the key ketogenic enzyme mito-
chondrial 3-hydroxy-3-methylglutaryl-CoA syn-
thase in rat brain. Epilepsy Res 49, 99–107.
Dehmer, T., Heneka, M.T., Sastre, M., Dichgans, J., and
Schulz, J.B. (2004) Protection by pioglitazone in
the MPTP model of Parkinson’s disease correlates
with I  kappa B alpha induction and block of NF
kappa B and iNOS activation. J Neurochem 88,
494–501.
Desvergne, B., and Wahli, W. (1999) Peroxisome
proliferator-activated receptors: nuclear control of
metabolism. Endocr Rev 20, 649–688.
Diab, A., Deng, C., Smith, J.D., Hussain, R.Z.,
Phanavanh, B., Lovett-Racke, A.E., Drew, P.D.,
and Racke, M.K. (2002). Peroxisome proliferator-
activated receptor-gamma agonist 15-deoxy-
Delta(12,14)-prostaglandin J(2) ameliorates
experimental autoimmune encephalomyelitis. J
Immunol 168, 2508–2515.
Diano, S., Liu, Z.W., Jeong, J.K., Dietrich, M.O., Ruan,
H.B., Kim, E., Suyama, S., Kelly, K., Gyengesi,
E., Arbiser, J.L., et  al. (2011). Peroxisome
proliferation-associated control of reactive oxygen
species sets melanocortin tone and feeding in diet-
induced obesity. Nat Med 17, 1121–1127.
DiMauro, S., Kulikova, R., Tanji, K., Bonilla, E., and
Hirano, M. (1999). Mitochondrial genes for gener-
alized epilepsies. Adv Neurol 79, 411–419.
Dineley, K.T., Jahrling, J.B., and Denner, L. (2014).
Insulin resistance in Alzheimer’s disease.
Neurobiol Dis 72, 92–103.
Doonan, F., Wallace, D.M., O’Driscoll, C., and Cotter,
T.G. (2009). Rosiglitazone acts as a neuroprotectant
181
in retinal cells via up-regulation of sestrin-1 and
SOD-2. J Neurochem 109, 631–643.
Dreyer, C., Krey, G., Keller, H., Givel, F., Helftenbein,
G., and Wahli, W. (1992). Control of the peroxi-
somal beta-oxidation pathway by a novel family of
nuclear hormone receptors. Cell 68, 879–887.
Dutton, S.B., Sawyer, N.T., Kalume, F., Jumbo-Lucioni,
P., Borges, K., Catterall, W.A., and Escayg, A.
(2011). Protective effect of the ketogenic diet in
Scn1a mutant mice. Epilepsia 52, 2050–2056.
Elbrecht, A., Chen, Y., Cullinan, C.A., Hayes, N.,
Leibowitz, Md, Moller, D.E., and Berger, J. (1996).
Molecular cloning, expression and character-
ization of human peroxisome proliferator acti-
vated receptors gamma 1 and gamma 2. Biochem
Biophys Res Commun 224, 431–437.
Engel, T., Sanz-Rodgriguez, A., Jimenez-Mateos, E.M.,
Concannon, C.G., Jimenez-Pacheco, A., Moran,
C., Mesuret, G., Petit, E., Delanty, N., Farrell, M.A.,
et al. (2013). CHOP regulates the p53-MDM2 axis
and is required for neuronal survival after seizures.
Brain 136, 577–592.
Esposito, G., Scuderi, C., Valenza, M., Togna,
G.I., Latina, V., De Filippis, D., Cipriano, M.,
Carratù, M.R., Iuvone, T., and Steardo, L. (2011).
Cannabidiol reduces Aβ-induced neuroinflam-
mation and promotes hippocampal neurogenesis
through PPARγ involvement. PLoS One 6, e28668.
Fenoglio-Simeone, K., Mazarati, A., Sefidvash-
Hockley, S., Shin, D., Wilke, J., Milligan, H.,
Sankar, R., Rho, J.M., and Maganti, R. (2009a).
Anticonvulsant effects of the selective melato-
nin receptor agonist ramelteon. Epilepsy Behav
16, 52–57.
Fenoglio-Simeone, K.A., Wilke, J.C., Milligan, H.L.,
Allen, C.N., Rho, J.M., and Maganti, R.K. (2009b).
Ketogenic diet treatment abolishes seizure period-
icity and improves diurnal rhythmicity in epileptic
Kcna1-null mice. Epilepsia 50, 2027–2034.
Fong, W.H., Tsai, H.D., Chen, Y.C., Wu, J.S., and Lin,
T.N. (2010). Anti-apoptotic actions of PPAR-
gamma against ischemic stroke. Mol Neurobiol 41,
180–186.
Fraser, D.D., Whiting, S., Andrew, R.D., Macdonald,
E.A., Musa-Veloso, K., and Cunnane, S.C. (2003).
Elevated polyunsaturated fatty acids in blood
serum obtained from children on the ketogenic
diet. Neurology 60, 1026–1029.
Fuenzalida, K., Quintanilla, R., Ramos, P., Piderit, D.,
Fuentealba, R.A., Martinez, G., Inestrosa, N.C., and
Bronfman, M. (2007). Peroxisome proliferator-
activated receptor gamma up-regulates the Bcl-2
anti-apoptotic protein in neurons and induces
mitochondrial stabilization and protection against
oxidative stress and apoptosis. J Biol Chem 282,
37006–37015.
Gahring, L.C., Persiyanov, K., Days, E.L., and Rogers,
S.W. (2005). Age-related loss of neuronal nic-
otinic receptor expression in the aging mouse
182
182 section III: Ketogenic Diet in the Laboratory
hippocampus corresponds with cyclooxygenase-
2 and PPAR gamma expression and is altered by
long-term NS398 administration. J Neurobiol 62,
453–468.
Gano, L.B., Patel, M., and Rho, J.M. (2014). Ketogenic
diets, mitochondria, and neurological diseases. J
Lipid Res 55, 2211–2228.
Girnun, G.D., Domann, F.E., Moore, S.A., and
Robbins, M.E. (2002). Identification of a func-
tional peroxisome proliferator-activated receptor
response element in the rat catalase promoter. Mol
Endocrinol 16, 2793–2801.
Guzmán, M., and Blázquez, C. (2004). Ketone body
synthesis in the brain:  possible neuroprotective
effects. Prostaglandins Leukot Essent Fatty Acids
70, 287–292.
Han, Y., Xie, N., Cao, L., Zhao, X., Liu, X., Jiang, H.,
and Chi, Z. (2011). Adenosine monophosphate-
activated protein kinase and peroxisome
proliferator-activated receptor gamma coactivator
1α signaling provides neuroprotection in status
epilepticus in rats. Neurosci Lett 500, 133–138.
Harris, J.J., Jolivet, R., and Attwell, D. (2012). Synaptic
energy use and supply. Neuron 75, 762–777.
Heneka, M.T., Gavrilyuk, V., Landreth, G.E., O’Banion,
M.K., Weinberg, G., and Feinstein, D.L. (2003).
Noradrenergic depletion increases inflammatory
responses in brain: effects on IkappaB and HSP70
expression. J Neurochem 85, 387–398.
Heneka, M.T., Klockgether, T., and Feinstein, D.L.
(2000). Peroxisome proliferator activated receptor-
gamma ligands reduce neuronal inducible nitric
oxide synthase expression and cell death in vivo, J.
Neurosci. 20, 6862–6867.
Heneka, M.T., and Landreth, G.E. (2007). PPARs in
the brain. Biochim Biophys Acta 1771, 1031–1045.
Hong, S., Huang, Y., Yu, X., Li, Y., Yang, J., Li, R., Deng,
Y., and Zhao, G. (2008). Peroxisome proliferator-
activated receptor gamma agonist, rosiglitazone,
suppresses CD40 expression and attenuates
inflammatory responses after lithium pilocarpine-
induced status epilepticus in rats. Int J Dev
Neurosci 26, 505–515.
Hong, S., Xin, Y., HaiQin, W., GuiLian, Z., Ru, Z.,
ShuQin, Z., HuQing, W., Li, Y., and Yun, D. (2012).
The PPARγ agonist rosiglitazone prevents cogni-
tive impairment by inhibiting astrocyte activation
and oxidative stress following pilocarpine-induced
status epilepticus. Neurol Sci 33, 559–566.
Hong, S., Xin, Y., HaiQin, W., GuiLian, Z., Ru, Z.,
ShuQin, Z., HuQing, W., Li, Y., Ning, B., and
YongNan, L. (2013). The PPARγ agonist rosigli-
tazone prevents neuronal loss and attenuates devel-
opment of spontaneous recurrent seizures through
BDNF/TrkB signaling following pilocarpine-
induced status epilepticus. Neurochem Int 63,
405–412.
Hughes, S.D., Kanabus, M., Anderson, G., Hargreaves,
I.P., Rutherford, T., O’Donnell, M., Cross, J.H.,
Rahman, S., Eaton, S., and Heales, S.J. (2014) The
ketogenic diet component decanoic acid increases
mitochondrial citrate synthase and complex
I  activity in neuronal cells. J Neurochem 129,
426–433.
Itoh, T., Fairall, L., Amin, K., Inaba, Y., Szanto, A.,
Balint, B.L., Nagy, L., Yamamoto, K., and Schwabe,
J.W. (2008). Structural basis for the activation of
PPARgamma by oxidized fatty acids. Nat Struct
Mol Biol 15, 924–931.
Jarrett, S.G., Milder, J.B., Liang, L.P., and Patel, M.
(2008). The ketogenic diet increases mitochondrial
glutathione levels. J Neurochem 106, 1044–1051.
Jeong, E.A., Jeon, B.T., Shin, H.J., Kim, N., Lee, D.H.,
Kim, H.J., Kang, S.S., Cho, G.J., Choi, W.S., and
Roh, G.S. (2011). Ketogenic diet-induced peroxi-
some proliferator-activated receptor-γ activation
decreases neuroinflammation in the mouse hip-
pocampus after kainic acid-induced seizures. Exp
Neurol 232, 195–202.
Johri, A., Chandra, A., and Beal, M.F. (2013). PGC-1α,
mitochondrial dysfunction, and Huntington’s dis-
ease. Free Radic Biol Med 62, 37–46.
Katsouri, L., Blondrath, K., and Sastre, M. (2012).
Peroxisome proliferator-activated receptor-γ
cofactors in neurodegeneration. IUBMB Life 64,
958–964.
Kersten, S., and Wahli, W. (2000). Peroxisome pro-
liferator activated receptor agonists. EXS 89,
141–151
Kim, D.Y., Simeone, K.A., Simeone, T.A., Pandya, J.D.,
Wilke, J.C., Ahn, Y., Geddes, J.W., Sullivan, P.G.,
and Rho, J.M. (2015). Ketone bodies mediate anti-
seizure effects through mitochondrial permeabil-
ity transition. Ann Neurol 78, 77–87.
Kitamura, Y., Shimohama, S., Koike, H., Kakimura, Ji.,
Matsuoka, Y., Nomura, Y., Gebicke-Haerter, P.J., and
Taniguchi, T. (1999). Increased expression of cyclo-
oxygenases and peroxisome proliferator-activated
receptor-gamma in Alzheimer’s disease brains.
Biochem Biophys Res Commun 254, 582–586.
Kobow, K., Auvin, S., Jensen, F., Löscher, W., Mody, I.,
Potschka, H., Prince, D., Sierra, A., Simonato, M.,
Pitkänen, A., et  al. (2012). Finding a better drug
for epilepsy: antiepileptogenesis targets. Epilepsia
53, 1868–1876.
Kroker, A.J., and Bruning, J.B. (2015). Review of the
structural and dynamic mechanisms of PPARγ par-
tial agonism. PPAR Res. doi: 10.1155/2015/816856.
Kudin, A.P., Zsurka, G., Elger, C.E., and Kunz, W.S.
(2009). Mitochondrial involvement in temporal
lobe epilepsy. Exp Neurol 218, 326–332.
Kunz, W.S., Kudin, A.P., Vielhaber, S., Blümcke, I.,
Zuschratter, W., Schramm, J., Beck, H., and Elger,
C.E. (2000). Mitochondrial complex I  deficiency
in the epileptic focus of patients with temporal
lobe epilepsy. Ann Neurol 48, 766–773.
Lee, D., Lee, K.H., Ho, W.K., and Lee, S.H. (2007).
Target cell-specific involvement of presynaptic
183
Chapter 20: Ketogenic Diet and PPARgamma
mitochondria in post-tetanic potentiation at hip-
pocampal mossy fiber synapses. J Neurosci 27,
13603–13613.
Lee, S.H., Kim, K.R., Ryu, S.Y., Son, S., Hong, H.S.,
Mook-Jung, I., Lee, S.H., and Ho, W.K. (2012).
Impaired short-term plasticity in mossy fiber syn-
apses caused by mitochondrial dysfunction of den-
tate granule cells is the earliest synaptic deficit in a
mouse model of Alzheimer’s disease. J Neurosci
32, 5953–5963.
Lim, G. P., Yang, F., Chu, T., Chen, P., Beech, W., Teter,
B., Tran, T., Ubeda, O., Ashe, K. H., Frautschy, S. A.,
and Cole, G. M. (2000). Ibuprofen suppresses plaque
pathology and inflammation in a mouse model for
Alzheimer’s disease. J Neurosci 20, 5709–5714.
Lu, J., Wu, D.M., Zheng, Y.L., Hu, B., Cheng, W.,
Zhang, Z.F., and Li, M.Q. (2013). Troxerutin coun-
teracts domoic acid-induced memory deficits in
mice by inhibiting CCAAT/enhancer binding
protein beta-mediated inflammatory response and
oxidative stress. J Immunol 190, 3466–3479.
Lu, M., Sarruf, D.A., Talukdar, S., Sharma, S., Li, P.,
Bandyopadhyay, G., Nalbandian, S., Fan, W.,
Gayen, J.R., Mahata, S.K., Webster, N.J., et  al.
(2011). Brain PPAR-γ promotes obesity and is
required for the insulin-sensitizing effect of thia-
zolidinediones. Nat Med 17, 618–622.
Luna-Medina, R., Cortes-Canteli, M., Sanchez-
Galiano, S., Morales-Garcia, J.A., Martinez,
A., Santos, A., and Perez-Castillo, A. (2007).
NP031112, a thiadiazolidinone compound, pre-
vents inflammation and neurodegeneration under
excitotoxic conditions:  potential therapeutic role
in brain disorders. J Neurosci 27, 5766–5776.
Maguire, J., and Salpekar, J.A. (2013). Stress, seizures,
and hypothalamic-pituitary-adrenal axis targets
for the treatment of epilepsy. Epilepsy Behav 26,
352–362.
Malapaka, R.R., Khoo, S., Zhang, J., Choi, J.H., Zhou,
X.E., Xu, Y., Gong, Y., Li, J., Yong, E.L., Chalmers,
M.J., et  al. (2012). Identification and mechanism
of 10-carbon fatty acid as modulating ligand of
peroxisome proliferator-activated receptors. J Biol
Chem 287, 183–195.
Malthankar-Phatak, G.H., de Lanerolle, N., Eid, T.,
Spencer, D.D., Behar, K.L., Spencer, S.S., Kim, J.H.,
and Lai, J.C. (2006). Differential glutamate dehy-
drogenase (GDH) activity profile in patients with
temporal lobe epilepsy. Epilepsia 47, 1292–1299.
Mandrekar-Colucci, S., and Landreth, G.E. (2011).
Nuclear receptors as therapeutic targets for
Alzheimer’s disease. Expert Opin Ther Targets 15,
1085–1097.
Mandrekar-Colucci, S., Sauerbeck, A., Popovich, P.G.,
and McTigue, D.M. (2013). PPAR agonists as
therapeutics for CNS trauma and neurological dis-
eases. ASN Neuro 5:e00129.
Masino, S.A., and Rho, J.M. (2012). Mechanisms of
ketogenic diet action. In Jasper’s Basic Mechanisms
183
of the Epilepsies, 4th edition, J.L. Noebels, M.
Avoli, M.A. Rogawski, R.W. Olsen, and A.V.
Delgado-Escueta, eds. (Bethesda, MD:  National
Center for Biotechnology Information).
Maurois, P., Rocchi, S., Pages, N., Bac, P., Stables,
J.P., Gressens, P., and Vamecq, J. (2008). The
PPARgamma agonist FMOC-L-leucine pro-
tects both mature and immature brain. Biomed
Pharmacother 62, 259–263.
Medina-Gomez, G., Gray, S., and Vidal-Puig, A.
(2007a). Adipogenesis and lipotoxicity:  role of
peroxisome proliferator-activated receptor gamma
(PPARgamma) and PPARgammacoactivator-1
(PGC1). Public Health Nutr 10, 1132–1137.
Medina-Gomez, G., Gray, S.L, Yetukuri, L.,
Shimomura, K., Virtue, S., Campbell, M., Curtis,
R.K., Jimenez-Linan, M., Blount, M., Yeo, G.S.,
et al. (2007b). PPAR gamma 2 prevents lipotoxic-
ity by controlling adipose tissue expandability and
peripheral lipid metabolism. PLoS Genet 3:e64.
Miglio, G., Rosa, A.C., Rattazzi, L., Collino,
M., Lombardi, G., and Fantozzi, R. (2009).
PPARgamma stimulation promotes mitochon-
drial biogenesis and prevents glucose deprivation-
induced neuronal cell loss. Neurochem Int 55,
496–504.
Milder, J.B., Liang, L.P., and Patel, M. (2010). Acute
oxidative stress and systemic Nrf2 activation by
the ketogenic diet. Neurobiol Dis 40, 238–244.
Moreno, S., Farioli-Vecchioli, S., and Cerù, M.P. (2004).
Immunolocalization of peroxisome proliferator-
activated receptors and retinoid X receptors in the
adult rat CNS. Neuroscience 123, 131–145.
Mueller, S.G., Kollias, S.S., Trabesinger, A.H., Buck, A.,
Boesiger, P., and Wieser, H.G. (2001). Proton mag-
netic resonance spectroscopy characteristics of a
focal cortical dysgenesis during status epilepticus
and in the interictal state. Seizure 10, 518–524.
Okada, K., Yamashita, U., and Tsuji, S. (2006).
Ameliorative effect of pioglitazone on seizure
responses in genetically epilepsy-susceptible EL
mice. Brain Res 1102, 175–178.
Overington, J.P., Al-Lazikani, B., and Hopkins, A.L.
(2006) How many drug targets are there? Nat Rev
Drug Discov 5, 993–996.
Porta, N., Vallee, L., Lecointe, C., Bouchaert, E., Staels,
B., Bordet, R., and Auvin, S. (2009). Fenofibrate,
a peroxisome proliferator-activated receptor-alpha
agonist, exerts anticonvulsive properties. Epilepsia
50, 943–948.
Puigserver, P., Adelmant, G., Wu, Z., Fan, M., Xu, J.,
O’Malley, B., and Spiegelman, B. (1999). Activation
of PPARgamma coactivator-1 through transcrip-
tion factor docking. Science 286, 1368–1371.
Puligheddu, M., Pillolla, G., Melis, M., Lecca, S.,
Marrosu, F., De Montis, M.G., Scheggi, S., Carta,
G., Murru, E., Aroni, S., et al. (2013). PPAR-alpha
agonists as novel antiepileptic drugs:  Preclinical
findings. PLoS One 8, e64541.
184
184 section III: Ketogenic Diet in the Laboratory
Quintanilla, R. A., Godoy, J. A., Alfaro, I., Cabezas, D.,
von Bernhardi, R., Bronfman, M., and Inestrosa,
N.C. (2013). Thiazolidinediones promote axonal
growth through the activation of the JNK pathway.
PLoS ONE 8, e65140.
Quintanilla, R.A., Utreras, E., and Cabezas-Opazo,
F.A. (2014). Role of PPAR γ in the Differentiation
and Function of Neurons. PPAR Res. doi: 10.1155/
2014/768594.
Rho, J.M., Kim, D.W., Robbins, C.A., Anderson, G.D.,
Schwartzkroin, P.A. (1999). Age-dependent differ-
ences in flurothyl seizure sensitivity in mice treated
with a ketogenic diet. Epilepsy Res 37, 233–240.
Ryan, K., Backos, D.S., Reigan, P., and Patel, M. (2012).
Post-translational oxidative modification and
inactivation of mitochondrial complex I in epilep-
togenesis. J Neurosci 32, 11250–11258.
Sarruf, D.A., Yu, F., Nguyen, H.T., Williams, D.L.,
Printz, R.L., Niswender, K.D., and Schwartz, M.W.
(2009). Expression of peroxisome proliferator-
activated receptor-gamma in key neuronal subsets
regulating glucose metabolism and energy homeo-
stasis. Endocrinology 150, 707–712.
Sauer, S. (2015). Ligands for the nuclear peroxisome
proliferator-activated receptor gamma. Trends
Pharmacol Sci 36, 688–704.
Schrader, M., Costello, J.L., Godinho, L.F., Azadi, A.S.,
and Islinger, M. (2015). Proliferation and fission of
peroxisomes—an update. Biochim Biophys Acta.
doi: 10.1016/j.bbamcr.2015.09.024
Shao, D., Rangwala, S.M., Bailey, S.T., Krakow,
S.L., Reginato, M.J., and Lazar, M.A. (1998).
Interdomain communication regulating ligand
binding by PPARgamma. Nature 396, 377–380.
Shen, Y., Lu, Y., Yu, F., Zhu, C., Wang, H., and Wang,
J. (2015). Peroxisome proliferator-activated
receptor-γ and its ligands in the treatment of
tumors in the nervous system. Curr Stem Cell Res
Ther. doi: 10.2174/1574888X10666150728122034
Simeone, T.A., Simeone, K.A., Samson, K.K., Kim, do
Y., and Rho, J.M. (2013). Loss of the Kv1.1 potas-
sium channel promotes pathologic sharp waves
and high frequency oscillations in in vitro hippo-
campal slices. Neurobiol Dis 54, 68–81.
Simeone, K.A., Matthews, S.A., Samson, K.K., and
Simeone, T.A. (2014a). Targeting deficiencies in
mitochondrial respiratory complex I  and func-
tional uncoupling exerts anti-seizure effects in a
genetic model of temporal lobe epilepsy and in a
model of acute temporal lobe seizures. Exp Neurol
251, 84–90.
Simeone, T.A., Matthews, S.A., Samson, K.K., and Simeone,
K.A. (2016). Regulation of brain PPARgamma2 con-
tributes to ketogenic diet anti-seizure efficacy. Exp
Neurol. doi: 10.1016/j.expneurol.2016.08.006.
Simeone, T.A., Samson, K.K., Matthews, S.A., and
Simeone, K.A. (2014b). In vivo ketogenic diet
treatment attenuates pathologic sharp waves
and high frequency oscillations in in vitro
hippocampal slices from epileptic Kv1.1α knock-
out mice. Epilepsia 55, e44–49.
Smart, S.L., Lopantsev, V., Zhang, C.L., Robbins,
C.A., Wang, H., Chiu, S.Y., Schwartzkroin, P.A.,
Messing, A., and Tempel, B.L. (1998). Deletion of
the K(V)1.1 potassium channel causes epilepsy in
mice. Neuron 20, 809–819.
Steger, D.J., Grant, G.R., Schupp, M., Tomaru, T.,
Lefterova, M.I., Schug, J., Manduchi, E., Stoeckert,
C.J., Jr., Lazar, M.A. (2010). Propagation of adipo-
genic signals through an epigenomic transition
state. Genes Dev 24, 1035–1044.
St-Pierre, J., Drori, S., Uldry, M., Silvaggi, J.M., Rhee,
J., Jäger, S., Handschin, C., Zheng, K., Lin, J., Yang,
W., et  al. (2006). Suppression of reactive oxygen
species and neurodegeneration by the PGC-1 tran-
scriptional coactivators. Cell 127, 397–408.
Sudha, K., Rao, A.V., and Rao, A. (2001). Oxidative
stress and antioxidants in epilepsy. Clin Chim Acta
303, 19–24.
Sugii, S., and Evans, R.M. (2011). Epigenetic codes
of PPARγ in metabolic disease. FEBS Lett 585,
2121–2128.
Sullivan, P.G., Dube, C., Dorenbos, K., Steward, O., and
Baram, T.Z. (2003). Mitochondrial uncoupling
protein-2 protects the immature brain from exci-
totoxic neuronal death. Ann Neurol 53, 711–717.
Sullivan, P.G., Rippy, N.A., Dorenbos, K., Concepcion,
R.C., Agarwal, A.K., and Rho, J.M. (2004). The
ketogenic diet increases mitochondrial uncoupling
protein levels and activity. Ann Neurol 55, 576–580.
Sundararajan, S., Gamboa, J.L., Victor, N.A., Wanderi,
E.W., Lust, W.D., and Landreth, G.E. (2005).
Peroxisome proliferator-activated receptor-gamma
ligands reduce inflammation and infarction size
in transient focal ischemia. Neuroscience 130,
685–696.
Takahashi, S., Iizumi, T., Mashima, K., Abe, T., and
Suzuki, N. (2014). Roles and regulation of keto-
genesis in cultured astroglia and neurons under
hypoxia and hypoglycemia. ASN Neuro 6.
doi: 10.1177/1759091414550997
Tan, Y., Muise, E.S., Dai, H., Raubertas, R., Wong, K.K.,
Thompson, G.M., Wood, H.B., Meinke, P.T., Lum,
P.Y., Thompson, J.R., and Berger, J.P. (2012). Novel
transcriptome profiling analyses demonstrate that
selective peroxisome proliferator-activated receptor
γ (PPARγ) modulators display attenuated and selec-
tive gene regulatory activity in comparison with
PPARγ full agonists. Mol Pharmacol 82, 68–79.
Thouennon, E., Cheng, Y., Falahatian, V., Cawley,
N.X., and Loh, Y.P. (2015). Rosiglitazone-activated
PPARγ induces neurotrophic factor-α1 transcrip-
tion contributing to neuroprotection. J Neurochem
134, 463–470.
Tontonoz, P., Hu, E., Graves, R.A., Budavari, A.I., and
Spiegelman, B.M. (1994). mPPAR gamma 2: tissue-
specific regulator of an adipocyte enhancer. Genes
Dev 8, 1224–1234.
185
Chapter 20: Ketogenic Diet and PPARgamma
Tontonoz, P., and Spiegelman, B.M. (2008). Fat and
beyond:  the diverse biology of PPARgamma.
Annu Rev Biochem 77, 289–312.
Victor, N.A., Wanderi, E.W., Gamboa, J., Zhao, X.,
Aronowski, J., Deininger, K., Lust, W.D., Landreth,
G.E., and Sundararajan, S. (2006). Altered
PPARgamma expression and activation after tran-
sient focal ischemia in rats. Eur J Neurosci 24,
1653–1663.
Vidal-Puig, A., Jimenez-Liñan, M., Lowell, B.B.,
Hamann, A., Hu, E., Spiegelman, B., Flier, J.S.,
and Moller, D.E. (1996). Regulation of PPAR
gamma gene expression by nutrition and obesity
in rodents. J Clin Invest 97, 2553–2561.
Vidal-Puig, A.J., Considine, R.V., Jimenez-Liñan, M.,
Werman, A., Pories, W.J., Caro, J.F., and Flier, J.S.
(1997). Peroxisome proliferator-activated receptor
gene expression in human tissues. Effects of obe-
sity, weight loss, and regulation by insulin and glu-
cocorticoids. J Clin Invest 99, 2416–2422.
Vielhaber, S., Niessen, H.G., Debska-Vielhaber, G.,
Kudin, A.P., Wellmer, J., Kaufmann, J., Schönfeld,
M.A., Fendrich, R., Willker, W., Leibfritz, D.,
et  al. (2008). Subfield-specific loss of hippocam-
pal N-acetyl aspartate in temporal lobe epilepsy.
Epilepsia 49, 40–50.
Wada, K., Nakajima, A., Katayama, K., Kudo, C.,
Shibuya, A., Kubota, N., Terauchi, Y., Tachibana, M.,
Miyoshi, H., Kamisaki, Y., et al. (2006). Peroxisome
proliferator-activated receptor gamma-mediated
regulation of neural stem cell proliferation and dif-
ferentiation. J Biol Chem 281, 12673–12681.
Wallace, D.C. (1999). Mitochondrial diseases in man
and mouse. Science 283, 1482–1488.
Wang, H., Jiang, R., He, Q., Zhang, Y., Zhang, Y., Li,
Y., Zhuang, R., Luo, Y., Li, Y., Wan, J., et al. (2012).
Expression pattern of peroxisome proliferator-
activated receptors in rat hippocampus following
cerebral ischemia and reperfusion injury. PPAR
Res. doi: 10.1155/2012/596394.
Wenzel, H.J., Vacher, H., Clark, E., Trimmer, J.S.,
Lee, A.L., Sapolsky, R.M., Tempel, B.L., and
Schwartzkroin, P.A. (2007). Structural conse-
quences of Kcna1 gene deletion and transfer in the
mouse hippocampus. Epilepsia 48, 2023–2046.
Werman, A., Hollenberg, A., Solanes, G., Bjorbaek,
C., Vidal-Puig, A.J., and Flier, J.S. (1997). Ligand-
independent activation domain in the N termi-
nus of peroxisome proliferator-activated receptor
gamma (PPARgamma). Differential activity of
PPARgamma1 and -2 isoforms and influence of
insulin. J Biol Chem 272, 20230–20235.
Wu, J.S., Cheung, W.M., Tsai, Y.S., Chen, Y.T., Fong,
W.H., Tsai, H.D., Chen, Y.C., Liou, J.Y., Shyue, S.K.,
Chen, J.J., et al. (2009a). Ligand-activated peroxi-
some proliferator activated receptor-gamma pro-
tects against ischemic cerebral infarction and
neuronal apoptosis by 14-3-3 epsilon upregula-
tion. Circulation 119, 1124–1134.
185
Wu, J.S., Lin, T.N., and Wu, K.K. (2009b). Rosiglitazone
and PPAR-gamma overexpression protect mito-
chondrial membrane potential and prevent apop-
tosis by upregulating anti-apoptotic Bcl-2 family
proteins. J Cell Physiol 220, 58–71.
Yakunin, E., Kisos, H., Kulik, W., Grigoletto, J.,
Wanders, R.J., and Sharon, R. (2014). The regula-
tion of catalase activity by PPAR γ is affected by
α-synuclein. Ann Clin Transl Neurol 1, 145–159.
Yamamoto, K., Itoh, T., Abe, D., Shimizu, M., Kanda,
T., Koyama, T., Nishikawa, M., Tamai, T., Ooizumi,
H., and Yamada, S. (2005). Identification of puta-
tive metabolites of docosahexaenoic acid as potent
PPARgamma agonists and antidiabetic agents.
Bioorg Med Chem Lett 15, 517–522.
Yan, Q., Zhang, J., Liu, H., Babu-Khan, S., Vassar, R.,
Biere, A. L., Citron, M., and Landreth, G. (2003).
Antiinflammatory drug therapy alters β-amyloid
processing and deposition in an animal model of
Alzheimer’s disease. J Neurosci 23, 7504–7509.
Yang, X., and Cheng, B. (2010). Neuroprotective and
anti-inflammatory activities of ketogenic diet on
MPTP-induced neurotoxicity. J Mol Neurosci 42,
145–153.
Youle, R.J., and Strasser, A. (2008). The Bcl-2 protein
family: opposing activities that mediate cell death.
Nat Rev Mol Cell Biol 9, 47–59
Yu, X., Shao, X.G., Sun, H., Li, Y.N., Yang, J., Deng,
Y.C., and Huang, Y.G. (2008). Activation of cere-
bral peroxisome proliferator-activated receptors
gamma exerts neuroprotection by inhibiting oxi-
dative stress following pilocarpine-induced status
epilepticus. Brain Res 1200, 146–158.
Zhao, X., Strong, R., Zhang, J., Sun, G., Tsien, J.Z., Cui,
Z., Grotta, J.C., and Aronowski, J. (2009). Neuronal
PPARgamma deficiency increases susceptibility to
brain damage after cerebral ischemia. J Neurosci
29, 6186–6195.
Zhao, Y., Patzer, A., Gohlke, P., Herdegen, T., and
Culman, J. (2005). The intracerebral application
of the PPARgamma-ligand pioglitazone confers
neuroprotection against focal ischaemia in the rat
brain. Eur J Neurosci 22, 278–282.
Zhao, Y., Patzer, A., Herdegen, T., Gohlke, P., and
Culman, J. (2006). Activation of cerebral peroxi-
some proliferator-activated receptors gamma pro-
motes neuroprotection by attenuation of neuronal
cyclooxygenase-2 overexpression after focal cere-
bral ischemia in rats. FASEB J 20, 1162–1175.
Zhu, Y., Alvares, K., Huang, Q., Rao, M.S., and Reddy,
J.K. (1993). Cloning of a new member of the per-
oxisome proliferator-activated receptor gene family
from mouse liver. J Biol Chem 268, 26817–26820.
Zolezzi, J.M., Bastías-Candia, S., Santos, M.J., and
Inestrosa, N.C. (2014). Alzheimer’s disease:  rel-
evant molecular and physiopathological events
affecting amyloid-β brain balance and the puta-
tive role of PPARs. Front Aging Neurosci 6, 176.
doi: 10.3389/fnagi.2014.00176
186
21
Ketogenic Diet in a Hippocampal Slice
Models and Mechanisms
M A S A H I T O K AWA M U R A J R . , M D , P H D
HIPPOCAMPUS AS A MODEL
FOR TEMPORAL LOBE
EPILEPSY
The hippocampus is well known as a brain area
involved in learning and memory and also as the
key region underlying the form of epilepsy known
as mesial temporal lobe epilepsy. The temporal
lobe refers to the ventrolateral middle part of
neocortex, and abnormal neuronal discharge or a
lesion affecting this lobe causes seizures (Gastaut,
1973). There are two main types of temporal lobe
epilepsy classified by the epileptic focus:  mesial
and lateral. The focus of mesial temporal lobe
epilepsy is the hippocampus, amygdala, or para-
hippocampal gyrus, and that of lateral temporal
lobe epilepsy is in neocortex. Over 80% of patients
with temporal lobe epilepsy have the mesial form
(Quarato et al., 2005; Schramm et al., 2001). The
typical symptom of mesial temporal lobe epilepsy
is complex partial seizures, which have a high
probability of an accompanying aura. The aura
is characteristic, usually presenting as epigastric
discomfort sometimes described as nausea or
psychiatric symptoms including fear. Complex
partial seizures often begin with arrest of motor
activities or staring. Autonomic motor behaviors
are usually oroalimentary automatisms or com-
plex automatisms. Dystonic posturing lasting 1–2
minutes often occurs involving the arm contralat-
eral to the ictal discharge (Engel, 2001). In a major-
ity of patients, mesial temporal lobe epilepsies are
associated with hippocampal sclerosis (Watson,
2003), which is atrophy with global gliosis and
loss of CA1 and/or CA3 pyramidal neurons in
the hippocampus (Thom, 2009). The structure of
the hippocampus is simple. It includes principal
cells (granular cells of dentate gyrus and CA1-4
pyramidal neurons) and surrounding interneu-
rons. The principal cells form an excitatory cir-
cuit that is modulated by inhibitory internurons.
Interestingly, CA3 pyramidal neurons are con-
nected to each other by excitatory recurrent col-
laterals. Thus, the hippocampal circuit is regulated
by a balance between excitation from recurrent
collaterals and inhibition from interneurons.
When the balance collapses, the hippocampal cir-
cuit becomes hyperexcitable and causes seizures
easily. Therefore, the atrophy in the hippocampus
is thought be one of the main focuses of mesial
temporal lobe epilepsy, and excision of hippocam-
pal sclerosis with selective amygdalohippocam-
pectomy successfully improves ~70% of surgical
patients (Paglioli et al., 2006; Wiebe et al., 2001).
From these clinical findings, the hippocampus
is thought to be a good experimental target for
investigating epileptogenesis and therapy of tem-
poral lobe epilepsy.
The ketogenic diet was designed in the 1920s to
treat epilepsy by mimicking fasting (Wilder, 1921).
Despite almost 100 years of clinical use, the mech-
anisms underlying the success of ketogenic diet
therapy are not well understood. In recent decades,
ketogenic diet has increasingly been noted as a
useful therapy for medically refractory epilepsy
(Hallbook et  al., 2007). The patients of temporal
lobe epilepsy are well known to be frequently resis-
tant to antiepileptic drugs (Wiebe and Jette, 2012).
As mentioned previously, the first choice of treat-
ment for mesial temporal lobe epilepsy associated
with hippocampal sclerosis is surgery (anterior
temporal lobectomy or selective amygdalohippo-
campectomy) because of good therapeutic out-
comes (Tanriverdi et al., 2008). For temporal lobe
epilepsy patients who are not good candidates
for surgery, however, ketogenic diet is one of the
therapeutic options (Ray and Wyllie, 2005). Thus,
a natural question is how the ketogenic diet pro-
duces its beneficial effects in temporal lobe epi-
lepsy. In this chapter, we focus on investigating
ketogenic diet antiseizure mechanisms with acute
187
Chapter 21: Ketogenic Diet in a Hippocampal Slice
slice preparations of the hippocampus, reviewing
our own and other laboratories’ work.
T H E I M P O R TA N C E
O F   H I P P O C A M PA L
S L I C E P R E PA R AT I O N S
F O R   S T U DY I N G
ANTIEPILEPTIC
MECHANISMS
O F   K E TO G E N I C   D I E T
Because epilepsy is caused by abnormal neuro-
nal discharges in the brain, electrophysiologi-
cal measurements are the most direct and useful
approach for researching epilepsy and its treat-
ments. There are two approaches for electrophysi-
ological recordings of any brain region:  in vivo
and in vitro. In vivo electrophysiological recording
of hippocampus is usually done by extracellular
recording of electrically evoked activity (Stewart
and Reid, 1993), or continuous recording of spon-
taneous field activity (Li et  al., 2008)  or multiple
unit activity (Lin et al., 2006); these preparations
can be acute or chronic. In vitro electrophysiologi-
cal recording is done using single-cell intracel-
lular sharp electrodes (Abe and Ogata, 1981)  or
patch-clamp electrodes (Kawamura et  al., 2004),
or extracellular field recording with single elec-
trodes (Masino and Dunwiddie, 1999) or electrode
arrays (Knowles et al., 1987) using acute slices of
hippocampus. As compared with in vivo hippo-
campal recordings, the advantages of hippocampal
slices are several-fold: (1) Ease of use: acute brain
slices must be maintained by perfusion with oxy-
genated artificial cerebrospinal fluid (Sakmann
et  al., 1989). Continuous perfusion allows us to
change the extracellular fluid easily. Thus, it is
easy to apply and wash out several agonists and/
or antagonists of various proteins such as chan-
nels, receptors, and transporters, and it is easy
to examine the detail of functional mechanisms
of neuronal activities; (2)  Reduction:  we usually
make 3–6 brain slices from one rodent and get
3–6 recordings from them, allowing us to reduce
the number of animals used; (3)  History:  a huge
number of electrophysiological experiments have
been done using hippocampal slice preparations
in the last half-century. Several methods for caus-
ing seizure-like bursting in vitro have been used
in the hippocampal slice preparation, including
kindling (Sayin et al., 1999), kainic acid treatment
(Congar et  al., 2000; Smith and Dudek, 2001),
inhibition of GABA receptors (Kohling et al., 2000;
Stafstrom et al., 2009), inhibition of potassium ion
channels (Stafstrom et  al., 2009), and neuronal
hyperexcitability by changing extracellular ion
187
concentrations (Congar et  al., 2000; Dulla et  al.,
2005; Kojima et  al., 1989; Stafstrom et  al., 2009).
All these approaches support the use of hippocam-
pal slice preparations to elucidate epileptic mech-
anisms. On the other hand, the pitfall of in vitro
recording is that the environment of acute brain
slice preparations is different from the in vivo
condition. We make a slice by cutting brain tissue
and that causes traumatic injury:  reactive gliosis
occurs in the acute hippocampal slice (Takano
et  al., 2014). Also, artificial cerebrospinal fluid
does not exactly reproduce actual cerebrospinal
fluid. Thus, results from in vitro hippocampal slice
preparations should be confirmed by in vivo elec-
trophysiological recordings or behavioral tests as
much as possible. For that reason, both in vivo and
in vitro electrophysiological recordings are useful,
and both are essential for epilepsy research.
T H R E E A P P ROAC H E S
FOR REPRODUCING
THE CONDITIONS
O F   K E TO G E N I C
DIET FEEDING
I N   H I P P O C A M PA L   S L I C E S
The major difficulty in using acute hippocampal
slices for ketogenic diet research is the inability to
precisely reproduce or maintain the condition of
diet therapy in vitro preparation. In vivo recording
clearly does not have this problem, because it uses
the whole body of experimental animals and diet-
altered metabolism is maintained (Koranda et al.,
2011; Masino et al., 2011). Therefore, special strat-
egies must be implemented for examining mecha-
nisms of the ketogenic diet using hippocampal
slice preparations; in this section, we review three
of these approaches (Figure 21.1).
Direct Application of Ketone Bodies
The ketogenic diet was developed to mimic fast-
ing, which alleviates epileptic seizures (Wilder,
1921). Ketogenic diet increases ketone bodies
(β-hydroxybutyrate, acetoacetate, acetone), syn-
thesized from free fatty acids in the liver (Masino
and Rho, 2012)  and then used for energy in the
brain instead of glucose (Masino et  al., 2009).
Chronic ketosis is one of hallmarks of the keto-
genic diet. Therefore, one approach to reproduc-
ing a ketogenic diet in a hippocampal slice is direct
application of ketone bodies, revealing whether
ketone bodies modulate neuronal activity directly.
In this paradigm, hippocampal slices are taken
from control diet–fed animals and dissolved ketone
bodies are applied in an extracellular solution such
as artificial cerebrospinal fluid (Figure  21.1A1,
188

188 section III: Ketogenic Diet in the Laboratory

(a) A1 Ketone body

Control diet (CD)

A2 Reduced glucose
Hippocampal slices

Increased ATP

(b)

Ketogenic diet Reduced glucose

CD (~2 weeks)

Hippocampal slices

FIGURE  21.1 Three approaches for reproducing the conditions of ketogenic diet feeding in hippocampal slices.
(A1) The first approach is the direct application of ketone bodies to hippocampal slices from control diet-fed animals.
(A2) The second approach mimics ketogenic diet conditions by using a whole-cell patch clamp recording of a single
cell in a hippocampal slice from control diet-fed animals. To mimic the ketogenic diet milieu, extracellular glucose is
reduced and intracellular ATP is increased. (b) The third approach is recording hippocampal slices made from ketogenic
diet-fed animals. Extracellular glucose is reduced during incubation and recording to maintain the in vivo effect of the
ketogenic diet.

Table 21.1). Thio et al. reported that direct appli-
cation of ketone bodies had no effect on synaptic
activity in acute hippocampal slices from Sprague-
Dawley (SD) rats. They recorded evoked field
excitatory postsynaptic potentials (fEPSP) and
population spikes (PS) in the CA1 region stimu-
lated by Schaffer collateral fibers, and applied
mixed ketone bodies (1 mM acetoacetate and
2 mM β-hydroxybutyrate) to the slices (Thio et al.,
2000). A  20-minute application of ketone bodies
did not change either fEPSP slope or PS amplitude.
They also recorded potassium channel blocker
4-aminopyridine-induced epileptiform discharge
from the dentate granule cell layer and CA3
region, and reported that application for 105 min-
utes did not change the frequency or duration of
these ictal events. Kimura et al. also reported that
application of mixed ketone bodies (1 mM each
acetoacetate and β-hydroxybutyrate) for 20 min-
utes did not change fEPSP slope and an 80-minute
application did not change the high-frequency
tetanic stimulation-induced long-term potentia-
tion (LTP) recorded from the CA1 region in acute
hippocampal slices from Wistar rats (Kimura et al.,
2012). Similar results were reported that mixed
ketone bodies (1 mM acetoacetate and 2 mM β-
hydroxybutyrate) had no effect on CA1 region
synaptic transmission or theta burst-induced LTP
in SD rat acute hippocampal slices (Youssef, 2015).
A  unique approach was used by Samoilova et  al.
(Samoilova et  al., 2010). They made organotypic
hippocampal slices, which were cultured with low
glucose and 10 mM β-hydroxybutyrate medium
for at least 3  days. This chronic in vitro ketosis,
however, did not alleviate intrinsic or induced epi-
leptiform discharges (but was neuroprotective).
All of these studies concluded that ketone bodies
do not directly affect synaptic transmission in the
rat hippocampus.
Other studies, however, have found positive
results. Juge et al. (2010) made acute hippocam-
pal slices from C57BL/6 mice and incubated the
 189

TABLE 21.1 REFERENCE LISTS OF THREE APPROACHES FOR REPRODUCING THE CONDITIONS OF KETOGENIC DIET FEEDING

Reference Animal Manipulation Recording Effect

Direct Application of Ketone Bodies

Thio et al., 2000
Kimura et al., 2012
Youssef, 2015
Samoilova et al., 2010
Juge et al., 2010
Tanner et al., 2011
Kim et al., 2015
Kawamura et al., 2010
Stafstrom et al., 1999
Simeone et al., 2015
Bough et al., 2006
Kawamura et al., 2014
SD rat 1 mM AA and 2 mM βHB for 20 min. fEPSP and PS recordings No change
Wistar rat 1 mM AA and 1 mM βHB for 20 min fEPSP recording No change
SD rat 1 mM AA and 2 mM βHB fEPSP recording No change
Wistar rat 10 mM βHB for 3 days (slice culture) PS recording No change
C57BL/6 mouse 10 mM AA for 2 hrs Whole-cell patch clamp Reducing frequency and amplitude of mEPSC
recording in CA1 pyramidal neurons
C57BL/6 mouse 2 mM βHB for 40 min Single channel recording Increasing open probability of KATP channels in
granule cells
Kcna1-null mouse 5 mM βHB and 1 mM AA for 2 weeks Multielectrode recordings Reducing spontaneous seizure-like events
(slice culture)
Controlling glucose and ATP levels with whole-cell patch clamp.
SD rat and C57BL/6 Increased [ATP]i and reduced Whole-cell patch clamp Causing hyperpolarization in CA3 pyramidal
mouse [glucose]e recording neurons
Acute hippocampal slices from ketogenic diet-fed animals.
SD rat KD (4.8:1 ratio) for 6-8 weeks fEPSP and PS recordings Reducing spontaneous seizure-like events
Kcna1-null mouse KD (6:1 ratio) for 11-15 days Multielectrode recordings Reducing spontaneous seizure-like events
SD rat KD (6:1 ratio) for 20 days fEPSP recording Inhibiting reduced glucose-induced depressions
SD rat and C57BL/6 KD (6:1 ratio) for 13-18 days PS recording Reducing evoked seizure-like bursting
mouse

AA, acetoacetate; βHB, β-hydroxybutyrate; fEPSP, field excitatory postsynaptic potentials; PS, population spikes; mEPSC, miniature excitatory postsynaptic currents; KATP channels, ATP-sensitive potassium channels;
[ATP]i, intracellular ATP concentration; [glucose]e, extracellular glucose concentration; KD, ketogenic diet.
190
190 section III: Ketogenic Diet in the Laboratory
slices with 10 mM acetoacetate for over 2 hours,
after which they recorded EPSPs from CA1 pyra-
midal neurons using whole-cell patch clamp.
Frequency and amplitude of miniature EPSPs
(mEPSP) from acetoacetate-incubated slices
were significantly reduced compared with con-
trol slices. Ketone bodies inhibited valinomycin-
evoked glutamate uptake by purified vesicular
glutamate transporter (VGLUT), suggesting that
ketone bodies inhibit synaptic transmission with
reduction of glutamate release via direct ketone
body-induced suppression of glutamate uptake
into vesicles. Importantly, they also investigated
the behavioral effects of ketone bodies. Seizures
in Wistar rats induced by intrahippocampal
4-aminopyridine were moderated by intrahip-
pocampal 10 mM acetoacetate, both infused by
microdialysis. These results clearly show that
direct application of ketone bodies modulates
synaptic transmission in hippocampal slices and
reduces seizure activity in vivo (Juge et al., 2010).
In addition, Tanner et al. (2011) recorded single
channel activity from dentate granule neurons
after incubating acute hippocampal slices from
C57BL/6 mice with 2 mM β-hydroxybutyrate for
over 40 minutes. Preincubation with this ketone
body increased steady-state and stimulus-
elevated open probability of ATP-sensitive
potassium (KATP) channels, which contribute
to the slow afterhyperpolarization after action
potential bursts to modulate spontaneous firing,
suggesting that direct ketone body-mediated
opening of KATP channels in dentate granule
neurons may act as a seizure gate in the hippo-
campus. Similar results were reported from neu-
rons of the substantia nigra in coronal midbrain
slices of rats and mice from same laboratory (Ma
et  al., 2007). Kim et  al. recorded from organo-
typic hippocampal slices that were cultured with
5 mM β-hydroxybutyrate and 1 mM acetoacetate
medium for 2 weeks (Kim et al., 2015). They used
Kcna1-null mice (C3HeB/FeJ background) lack-
ing voltage-gated potassium (Kv 1.1) channels,
which is thought to be a model for several types
of epilepsy including human temporal lobe epi-
lepsy. Extracellular multielectrode array record-
ings showed spontaneous seizure-like events in
organotypic hippocampal slice cultures from
Kcna1-null mice. The application of ketone bod-
ies for 2 weeks attenuated the seizure-like events
generated by the mutant tissue. They also applied
5 mM β-hydroxybutyrate in vivo using subcuta-
neously implanted osmotic minipumps to Kcna1-
null mice and reported that administration of
this ketone body reduced the number of seizures
(Kim et al., 2015).
In sum, several studies have used direct appli-
cation of ketone bodies in hippocampal acute
slices or organotypic cultures, and both positive
and negative results have been found. Negative and
positive studies used rats and mice, respectively, so
a simple explanation is that the discrepancy arises
from species differences. However, this is unlikely,
because ketogenic diet is known to reduce behav-
ioral seizures in both rats (Appleton and DeVivo,
1974; Bough et al., 2002; Bough et al., 2006; Hori
et al., 1997; Zhao et al., 2004) and mice (Hartman
et  al., 2008; Kwon et  al., 2008; Noh et  al., 2003;
Rho et  al., 1999; Uhlemann and Neims, 1972).
The methods for applying ketone bodies varied in
these reports, including concentration of ketone
bodies, time for application, and application path-
way (perfusion or preincubation), and these might
contribute to interstudy variation. Using similar
protocols for direct application of ketone bodies
would be useful for finding common mechanisms.
Controlling Glucose and ATP Levels
with Whole-Cell Patch Clamp
Another approach mimics the altered metabolism
during ketogenic diet treatment using single-cell
patch clamp recording. Fasting and ketogenic
diet are thought to cause anticonvulsant effect by
changing brain metabolism. It is well known that
ketogenic diet causes a stable, mild hypoglyce-
mia in humans (Huttenlocher, 1976) and rodents
(Bough et al., 2006). Interestingly, plasma glucose
level correlates with the antiepileptic effect of the
ketogenic diet (Mantis et  al., 2004), indicating
that extracellular glucose is mechanistically rel-
evant. Intracellular conditions are also thought to
be changed by ketogenic diet-altered metabolism,
and the final product of brain energy metabolism
is ATP. Several studies have reported that keto-
genic diet increases brain ATP levels in humans
(Pan et  al., 1995)  and rodents (DeVivo et  al.,
1978; Nakazawa et  al., 1983; Nylen et  al., 2009).
Therefore, intracellular ATP is a promising target
for mimicking the ketogenic diet in vitro. From
these reports, reducing and increasing extracel-
lular glucose and intracellular ATP, respectively,
might in combination reproduce mechanistically
important ketogenic diet conditions in acute hip-
pocampal slices (Figure 21.1A2). As mentioned in
section 2, it is easy to change extracellular solu-
tion with an in vitro brain slice (including moder-
ately lowering glucose), but trickier to change the
intracellular milieu. The whole-cell patch clamp
191

Chapter 21: Ketogenic Diet in a Hippocampal Slice 191

Ketogenic diet

Glucose Ketone bodies

NADH AcetylCoA
ADP Pi
ETC
ATP synthase
Panx
NADH
TCA cycle
ATP ATP FADH2

Nucleotidase

Adenosine A1R
KATP

K+
Hyperpolarization

Antiseizure
FIGURE 21.2 Schematic of pannexin-1 channel-adenosine A1 receptor-KATP channel autocrine regulation in CA3
pyramidal neurons and relationship to ketogenic diet. Ketogenic diet increases ketone body levels and reduces glucose
levels. Ketone bodies might increase intracellular ATP production, and increased ATP is released to the extracellular
space due to reduced glucose-induced opening of pannexin-1 channels (panx). After breakdown to adenosine by
nucleotidases, activated adenosine A1 receptors (A1R) open KATP channels (KATP), which hyperpolarizes CA3 pyramidal
neurons. This hyperpolarization reduces neuronal hyperexcitability and causes the ketogenic diet’s antiseizure effects.
TCA cycle, tricarboxylic acid cycle; ETC, electron transport chain.

technique is one of the methods for recording a
single cell (Kawamura et al., 2004). This technique
allows physical exchange between the intracellu-
lar fluid and the artificial intracellular solution in
the recording pipette, which allows us to modify
the intracellular fluid composition of the recorded
neuron experimentally (Figure 21.1A2, Table 21.1)
including elevating intracellular ATP. We found
that increased intracellular ATP and reduced
extracellular glucose caused hyperpolarization in
hippocampal neurons (Kawamura et  al., 2010).
We recorded from CA3 pyramidal neurons with
the whole-cell patch clamp technique in acute
hippocampal slices from control diet–fed SD rats
or C57BL/6 mice. Reduced extracellular glucose
and increased intracellular ATP caused outward
current (or hyperpolarization when recording
membrane potential) in CA3 pyramidal neu-
rons. This outward current was dose-dependent
for both extracellular glucose and intracellular
ATP, and importantly it was found in both rats
and mice. Pharmacological and genetic experi-
ments demonstrated that when intracellular ATP
was sufficient or increased, reduced extracellular
glucose opened pannexin-1 channels and released
intracellular ATP to the extracellular space.
Released ATP was rapidly hydrolyzed to adenos-
ine, which activated adenosine A1 receptors with
subsequent opening of KATP channels. Opening of
these potassium channels caused hyperpolariza-
tion and reduced excitability. These results indi-
cate that mimicking the ketogenic diet condition
with increased ATP and reduced glucose reduces
excitability in hippocampal CA3 pyramidal neu-
rons with autocrine modulation via adenosine
A1 receptors, and this might be a one of the key
mechanisms of the anticonvulsant effects of the
ketogenic diet in vivo (Figure 21.2). This approach
for reproducing ketogenic diet conditions in acute
hippocampal slice is useful to elucidate detailed
mechanisms within single neurons. However, it
mimics only two of the aspects of ketogenic diet
feeding. Further examinations using behavioral
tests and recordings from in vivo ketogenic diet-
fed animals are needed to link the results from this
approach to the effects of ketogenic diet feeding.
192
192 section III: Ketogenic Diet in the Laboratory
Acute Hippocampal Slices from Ketogenic
Diet–Fed Animals
A third approach is possibly the most direct and
useful way for investigating mechanisms of keto-
genic diet because it uses acute hippocampal slices
from ketogenic diet-fed animals (Figure 21.1b,
Table 21.1). The question about this approach is
whether or not the intra- and extracellular milieu
produced by ketogenic diet feeding can be main-
tained after making and incubating brain slices.
However, four reports show that it can work suc-
cessfully. Stafstrom et al. reported that ketogenic
diet (fat:[protein+carbohydrate] ratio of 4.8:1)
induced antiseizure effects in acute hippocampal
slices from kainic acid–treated rats (Stafstrom et
al., 1999). They recorded fEPSP and PS from area
CA1 from SD rats fed a ketogenic diet for 6–8
weeks. Synaptic transmission was not significantly
different between slices from control diet–fed and
ketogenic diet–fed rats. The frequency of kainic
acid–induced spontaneous seizures was signifi-
cantly lower in slices from ketogenic diet–fed rats
than from control diet–fed rats. Similar results
were reported by Simeone et al. using extracellular
multielectrode array recordings in acute hippo-
campal slices from ketogenic diet–fed Knca1-null
mice (Simeone et al., 2014). The pathologic sei-
zure-like events generated in Knca1-null mice
slice were diminished by ketogenic diet (6:1 ratio)
treatment for 11–15 days. However, mossy fiber-
CA3 dendritic field potential slopes and fiber vol-
ley amplitudes of mossy fiber are not significantly
different between slices from control diet–fed and
ketogenic diet–fed Knca1-null mice (Simeone
et al., 2014). Bough et al. recorded medial per-
forant pathway-evoked fEPSPs from the dentate
molecular layer in acute hippocampal slices from
SD rats fed a ketogenic diet (6:1 ratio) for over 20
days (Bough et al., 2006). Reducing extracellu-
lar glucose concentration from 10 mM to 2 mM
depressed the slope of fEPSP reversibly in slices
from control diet–fed rats, and this depression was
inhibited in slices from ketogenic diet–fed rats.
The effects were lost after slices were incubated in
10 mM glucose for over 3.5 hours. They concluded
that synaptic transmission in hippocampal slices
from ketogenic diet–fed rats were more resistant
to reduced glucose than slices from control diet–
fed rats (Bough et al., 2006). We also reported that
ketogenic diet caused antiseizure effects in acute
hippocampal slices of rats and mice (Kawamura
et al., 2014). We recorded PS and GABA receptor
blocker bicuculline-induced seizure-like bursting
from the CA3 region in acute hippocampal slices
from SD rats or C57BL/6 mice fed a ketogenic
diet (6:1 ratio) for 13–18 days. Excitability and
bicuculline-induced bursting were significantly
inhibited by reduced extracellular glucose concen-
tration in slices from ketogenic diet–fed rats and
mice but were not changed by reduced extracellu-
lar glucose in slices from control diet–fed rodents.
Ketogenic diet–induced suppression of bicucul-
line-induced bursting was inhibited by adenos-
ine A1 receptor antagonist and did not occur in
slices from adenosine A1 receptor knockout mice.
Antagonism of KATP channels or pannexin-1 chan-
nels inhibited the ketogenic diet-induced suppres-
sion of bicuculline-induced bursting. These results
suggest that ketogenic diet causes antiseizure
effects through a pannexin-1 channel-adenos-
ine A1 receptor-KATP channel autocrine pathway
(Figure 21.2), the same pathway revealed by the
whole-cell patch clamp technique for mimicking
ketogenic diet conditions as described previously
(Kawamura et al., 2010).
These studies used successfully acute hippo-
campal slices from ketogenic diet–fed rodents
to elucidate the changing of neuronal activities
underlying this treatment. Interestingly, reducing
extracellular glucose concentration is thought be
one of the most important points for reproduc-
ing the effects of ketogenic diet in this approach.
Synaptic transmission in hippocampal slices from
ketogenic diet–fed rodents was not different from
that in slices from control diet–fed rodents when
extracellular glucose concentration in artificial
cerebrospinal fluid is standard in all three reports
(however, evidence is mounting that this stan-
dard glucose concentration for acute brain slices
is higher than physiological brain glucose levels;
Kealy et al., 2013; Lowry and Fillenz, 2001; Shram
et  al., 1997). Reduced glucose reveals the differ-
ence between ketogenic diet– and control diet–
fed animals in two of these studies (Bough et al.,
2006; Kawamura et al., 2014), which parallels the
finding that the anticonvulsant effect of the keto-
genic diet is correlated with plasma glucose levels
(Mantis et  al., 2004). Therefore it would be use-
ful to make extracellular glucose concentrations
lower than standard to reproduce or maintain
effects of the ketogenic diet in acute hippocampal
slices.
CONCLUSIONS
Here we illustrate three approaches for research-
ing anticonvulsant mechanisms of ketogenic diets
by using electrophysiological recording from
hippocampal slices. The usefulness of hippocam-
pal slices is that it is easy to elucidate the details
of neuronal modulation by ketogenic diet, as
193
Chapter 21: Ketogenic Diet in a Hippocampal Slice
shown in Figure 21.2. All three approaches have
contributed to finding detailed potential mecha-
nisms underlying ketogenic diet effects including
VGLUT modulation, KATP opening, activation of
adenosine receptors, and ATP release from pan-
nexin channels. It is clear that electrophysiologi-
cal recording from hippocampal slices is a good
tool for ketogenic diet research. However, all three
approaches need some steps for reproducing keto-
genic diet effects in vitro such as ketone applica-
tion, reduced extracellular glucose, and increased
intracellular ATP. These methods are not direct
measurements of the in vivo anticonvulsant effects
of ketogenic diet. A combination of both in vivo
and in vitro recordings should provide further
explanations concerning the key anticonvulsant
mechanisms underlying ketogenic diet treatment.
AC K N OW L E D G M E N T
We thank Dr. David N. Ruskin for comments on
this manuscript. We acknowledge the support of
the Jikei University Research Fund to MK.
REFERENCES
Abe, H., and Ogata, N. (1981). Effects of penicillin on
electrical activities of neurons in guinea-pig hip-
pocampal slies. Jpn J Pharmacol 31, 661–675.
Appleton, D.B., and DeVivo, D.C. (1974). An ani-
mal model for the ketogenic diet. Epilepsia 15,
211–227.
Bough, K.J., Gudi, K., Han, F.T., Rathod, A.H., and
Eagles, D.A. (2002). An anticonvulsant profile
of the ketogenic diet in the rat. Epilepsy Res 50,
313–325.
Bough, K.J., Wetherington, J., Hassel, B., Pare, J.F.,
Gawryluk, J.W., Greene, J.G., Shaw, R., Smith,
Y., Geiger, J.D., and Dingledine, R.J. (2006).
Mitochondrial biogenesis in the anticonvulsant
mechanism of the ketogenic diet. Ann Neurol 60,
223–235.
Congar, P., Gaiarsa, J.L., Popovici, T., Ben-Ari, Y., and
Crepel, V. (2000). Permanent reduction of seizure
threshold in post-ischemic CA3 pyramidal neu-
rons. J Neurophysiol 83, 2040–2046.
DeVivo, D.C., Leckie, M.P., Ferrendelli, J.S., and
McDougal, D.B., Jr. (1978). Chronic ketosis and
cerebral metabolism. Ann Neurol 3, 331–337.
Dulla, C.G., Dobelis, P., Pearson, T., Frenguelli, B.G.,
Staley, K.J., and Masino, S.A. (2005). Adenosine
and ATP link PCO2 to cortical excitability via pH.
Neuron 48, 1011–1023.
Engel, J., Jr. (2001). Mesial temporal lobe epilepsy: what
have we learned? Neuroscientist 7, 340–352.
Gastaut, H.  (1973). Dictionary of Epilepsy:  Part  1.
Definitions. (Geneva: World Health Organization).
Hallbook, T., Kohler, S., Rosen, I., and Lundgren, J.
(2007). Effects of ketogenic diet on epileptiform
193
activity in children with therapy resistant epilepsy.
Epilepsy Res 77, 134–140.
Hartman, A.L., Lyle, M., Rogawski, M.A., and Gasior,
M. (2008). Efficacy of the ketogenic diet in the
6-Hz seizure test. Epilepsia 49, 334–339.
Hori, A., Tandon, P., Holmes, G.L., and Stafstrom,
C.E. (1997). Ketogenic diet: effects on expression
of kindled seizures and behavior in adult rats.
Epilepsia 38, 750–758.
Huttenlocher, P.R. (1976). Ketonemia and seizures:
metabolic and anticonvulsant effects of two keto-
genic diets in childhood epilepsy. Pediatr Res 10,
536–540.
Juge, N., Gray, J.A., Omote, H., Miyaji, T., Inoue, T.,
Hara, C., Uneyama, H., Edwards, R.H., Nicoll,
R.A., and Moriyama, Y. (2010). Metabolic con-
trol of vesicular glutamate transport and release.
Neuron 68, 99–112.
Kawamura, M., Gachet, C., Inoue, K., and Kato, F.
(2004). Direct excitation of inhibitory interneu-
rons by extracellular ATP mediated by P2Y1
receptors in the hippocampal slice. J Neurosci 24,
10835–10845.
Kawamura, M., Jr., Ruskin, D.N., and Masino, S.A.
(2010). Metabolic autocrine regulation of neurons
involves cooperation among pannexin hemichan-
nels, adenosine receptors, and KATP channels. J
Neurosci 30, 3886–3895.
Kawamura, M., Jr., Ruskin, D.N., Geiger, J.D., Boison,
D., and Masino, S.A. (2014). Ketogenic diet sensi-
tizes glucose control of hippocampal excitability. J
Lipid Res 55, 2254–2260.
Kealy, J., Bennett, R., and Lowry, J.P. (2013).
Simultaneous recording of hippocampal oxygen
and glucose in real time using constant potential
amperometry in the freely-moving rat. J Neurosci
Methods 215, 110–120.
Kim, D.Y., Simeone, K.A., Simeone, T.A., Pandya, J.D.,
Wilke, J.C., Ahn, Y., Geddes, J.W., Sullivan, P.G.,
and Rho, J.M. (2015). Ketone bodies mediate anti-
seizure effects through mitochondrial permeabil-
ity transition. Ann Neurol 78, 77–87.
Kimura, R., Ma, L.Y., Wu, C., Turner, D., Shen, J.X.,
Ellsworth, K., Wakui, M., Maalouf, M., and Wu,
J. (2012). Acute exposure to the mitochondrial
complex I  toxin rotenone impairs synaptic long-
term potentiation in rat hippocampal slices. CNS
Neurosci Ther 18, 641–646.
Knowles, W.D., Traub, R.D., and Strowbridge, B.W.
(1987). The initiation and spread of epilepti-
form bursts in the in vitro hippocampal slice.
Neuroscience 21, 441–455.
Kohling, R., Vreugdenhil, M., Bracci, E., and Jefferys,
J.G. (2000). Ictal epileptiform activity is facilitated
by hippocampal GABAA receptor-mediated oscil-
lations. J Neurosci 20, 6820–6829.
Kojima, H., Kowada, M., and Katsuta, Y. (1989).
Cellular substrates for epileptiform activ-
ity induced by repeated tetanic stimulation in
194
194 section III: Ketogenic Diet in the Laboratory
hippocampal slices. Neurol Med Chir (Tokyo)
29, 1–5.
Koranda, J.L., Ruskin, D.N., Masino, S.A., and Blaise,
J.H. (2011). A ketogenic diet reduces long-term
potentiation in the dentate gyrus of freely behav-
ing rats. J Neurophysiol 106, 662–666.
Kwon, Y.S., Jeong, S.W., Kim, D.W., Choi, E.S., and
Son, B.K. (2008). Effects of the ketogenic diet on
neurogenesis after kainic acid-induced seizures in
mice. Epilepsy Res 78, 186–194.
Li, T., Ren, G., Lusardi, T., Wilz, A., Lan, J.Q., Iwasato,
T., Itohara, S., Simon, R.P., and Boison, D. (2008).
Adenosine kinase is a target for the prediction and
prevention of epileptogenesis in mice. J Clin Invest
118, 571–582.
Lin, L., Chen, G., Xie, K., Zaia, K.A., Zhang, S., and
Tsien, J.Z. (2006). Large-scale neural ensemble
recording in the brains of freely behaving mice. J
Neurosci Methods 155, 28–38.
Lowry, J.P., and Fillenz, M. (2001). Real-time moni-
toring of brain energy metabolism in vivo using
microelectrochemical sensors: the effects of anes-
thesia. Bioelectrochemistry 54, 39–47.
Ma, W., Berg, J., and Yellen, G. (2007). Ketogenic
diet metabolites reduce firing in central neurons
by opening K(ATP) channels. J Neurosci 27,
3618–3625.
Mantis, J.G., Centeno, N.A., Todorova, M.T., McGowan,
R., and Seyfried, T.N. (2004). Management of multi-
factorial idiopathic epilepsy in EL mice with caloric
restriction and the ketogenic diet:  role of glucose
and ketone bodies. Nutr Metab (Lond) 1, 11.
Masino, S.A., and Dunwiddie, T.V. (1999).
Temperature-dependent modulation of excitatory
transmission in hippocampal slices is mediated by
extracellular adenosine. J Neurosci 19, 1932–1939.
Masino, S.A., Kawamura, M., Wasser, C.A., Pomeroy,
L.T., and Ruskin, D.N. (2009). Adenosine, keto-
genic diet and epilepsy: the emerging therapeutic
relationship between metabolism and brain activ-
ity. Curr Neuropharmacol 7, 257–268.
Masino, S.A., Li, T., Theofilas, P., Sandau, U.S., Ruskin,
D.N., Fredholm, B.B., Geiger, J.D., Aronica, E., and
Boison, D. (2011). A ketogenic diet suppresses sei-
zures in mice through adenosine A(1) receptors. J
Clin Invest 121, 2679–2683.
Masino, S.A., and Rho, J.M. (2012). Mechanisms of
ketogenic diet action. In Jasper’s Basic Mechanisms
of the Epilepsies, J.L. Noebels, M. Avoli, M.A.
Rogawski, R.W. Olsen, and A.V. Delgado-Escueta,
eds. (Bethesda: National Center for Biotechnology
Information), pp. 1001–1022.
Nakazawa, M., Kodama, S., and Matsuo, T. (1983).
Effects of ketogenic diet on electroconvulsive
threshold and brain contents of adenosine nucleo-
tides. Brain Dev 5, 375–380.
Noh, H.S., Kim, Y.S., Lee, H.P., Chung, K.M., Kim,
D.W., Kang, S.S., Cho, G.J., and Choi, W.S. (2003).
The protective effect of a ketogenic diet on kainic
acid-induced hippocampal cell death in the male
ICR mice. Epilepsy Res 53, 119–128.
Nylen, K., Velazquez, J.L., Sayed, V., Gibson, K.M.,
Burnham, W.M., and Snead, O.C., 3rd (2009). The
effects of a ketogenic diet on ATP concentrations
and the number of hippocampal mitochondria in
Aldh5a1(−/−) mice. Biochim Biophys Acta 1790,
208–212.
Paglioli, E., Palmini, A., Portuguez, M., Azambuja,
N., da Costa, J.C., da Silva Filho, H.F., Martinez,
J.V., and Hoeffel, J.R. (2006). Seizure and memory
outcome following temporal lobe surgery: selec-
tive compared with nonselective approaches for
hippocampal sclerosis. J Neurosurg 104, 70–78.
Pan, W.J., Osmanovic, S.S., and Shefner, S.A. (1995).
Characterization of the adenosine A1 receptor-
activated potassium current in rat locus ceruleus
neurons. J Pharmacol Exp Ther 273, 537–544.
Quarato, P.P., Di Gennaro, G., Mascia, A., Grammaldo,
L.G., Meldolesi, G.N., Picardi, A., Giampa, T.,
Falco, C., Sebastiano, F., Onorati, P., et al. (2005).
Temporal lobe epilepsy surgery: different surgical
strategies after a non-invasive diagnostic protocol.
J Neurol Neurosurg Psychiatry 76, 815–824.
Ray, A., and Wyllie, E. (2005). Treatment options and
paradigms in childhood temporal lobe epilepsy.
Expert Rev Neurother 5, 785-801.
Rho, J.M., Kim, D.W., Robbins, C.A., Anderson, G.D.,
and Schwartzkroin, P.A. (1999). Age-dependent
differences in flurothyl seizure sensitivity in mice
treated with a ketogenic diet. Epilepsy Res 37,
233–240.
Sakmann, B., Edwards, F., Konnerth, A., and Takahashi,
T. (1989). Patch clamp techniques used for study-
ing synaptic transmission in slices of mammalian
brain. Q J Exp Physiol 74, 1107–1118.
Samoilova, M., Weisspapir, M., Abdelmalik, P.,
Velumian, A.A., and Carlen, P.L. (2010). Chronic
in vitro ketosis is neuroprotective but not anti-
convulsant. J Neurochem 113, 826–835.
Sayin, Rutecki, P., and Sutula, T. (1999). NMDA-
dependent currents in granule cells of the dentate
gyrus contribute to induction but not permanence
of kindling. J Neurophysiol 81, 564–574.
Schramm, J., Kral, T., Grunwald, T., and Blumcke, I.
(2001). Surgical treatment for neocortical tempo-
ral lobe epilepsy: clinical and surgical aspects and
seizure outcome. J Neurosurg 94, 33–42.
Shram, N.F., Netchiporouk, L.I., Martelet, C., Jaffrezic-
Renault, N., and Cespuglio, R. (1997). Brain glu-
cose:  voltammetric determination in normal and
hyperglycaemic rats using a glucose microsensor.
Neuroreport 8, 1109–1112.
Simeone, T.A., Samson, K.K., Matthews, S.A., and
Simeone, K.A. (2014). In vivo ketogenic diet treat-
ment attenuates pathologic sharp waves and high
frequency oscillations in in vitro hippocampal
slices from epileptic Kv 1.1alpha knockout mice.
Epilepsia 55, e44–49.
195
Chapter 21: Ketogenic Diet in a Hippocampal Slice
Smith, B.N., and Dudek, F.E. (2001). Short- and
long-term changes in CA1 network excitability
after kainate treatment in rats. J Neurophysiol
85, 1–9.
Stafstrom, C.E., Ockuly, J.C., Murphree, L., Valley,
M.T., Roopra, A., and Sutula, T.P. (2009).
Anticonvulsant and antiepileptic actions of 2-
deoxy-D-glucose in epilepsy models. Ann Neurol
65, 435–447.
Stafstrom, C.E., Wang, C., and Jensen, F.E. (1999).
Electrophysiological observations in hippocam-
pal slices from rats treated with the ketogenic diet.
Dev Neurosci 21, 393–399.
Stewart, C., and Reid, I. (1993). Electroconvulsive
stimulation and synaptic plasticity in the rat. Brain
Res 620, 139–141.
Takano, T., He, W., Han, X., Wang, F., Xu, Q., Wang,
X., Oberheim Bush, N.A., Cruz, N., Dienel, G.A.,
and Nedergaard, M. (2014). Rapid manifestation
of reactive astrogliosis in acute hippocampal brain
slices. Glia 62, 78–95.
Tanner, G.R., Lutas, A., Martinez-Francois, J.R.,
and Yellen, G. (2011). Single K ATP channel
opening in response to action potential firing in
mouse dentate granule neurons. J Neurosci 31,
8689–8696.
Tanriverdi, T., Olivier, A., Poulin, N., Andermann,
F., and Dubeau, F. (2008). Long-term seizure
outcome after mesial temporal lobe epilepsy sur-
gery:  corticalamygdalohippocampectomy versus
195
selective amygdalohippocampectomy. J Neurosurg
108, 517–524.
Thio, L.L., Wong, M., and Yamada, K.A. (2000). Ketone
bodies do not directly alter excitatory or inhibitory
hippocampal synaptic transmission. Neurology
54, 325–331.
Thom, M. (2009). Hippocampal sclerosis:  progress
since Sommer. Brain Pathol 19, 565–572.
Uhlemann, E.R., and Neims, A.H. (1972).
Anticonvulsant properties of the ketogenic diet in
mice. J Pharmacol Exp Ther 180, 231–238.
Watson, C. (2003). Hippocampal sclerosis and the syn-
drome of medial temporal lobe epilepsy. Expert
Rev Neurother 3, 821–828.
Wiebe, S., Blume, W.T., Girvin, J.P., and Eliasziw, M.
(2001). A randomized, controlled trial of surgery for
temporal-lobe epilepsy. N Engl J Med 345, 311–318.
Wiebe, S., and Jette, N. (2012). Pharmacoresistance
and the role of surgery in difficult to treat epilepsy.
Nat Rev Neurol 8, 669–677.
Wilder, R.M. (1921). The effects of ketonemia on the
course of epilepsy. Mayo Clinic Proc 2, 307–308.
Youssef, F.F. (2015). Ketone bodies attenuate excito-
toxic cell injury in the rat hippocampal slice under
conditions of reduced glucose availability. Neurol
Res 37, 211–216.
Zhao, Q., Stafstrom, C.E., Fu, D.D., Hu, Y., and Holmes,
G.L. (2004). Detrimental effects of the ketogenic
diet on cognitive function in rats. Pediatr Res 55,
498–506.
196
22
Metabolic Therapy and Pain
D AV I D N. R U S K I N , P H D
INTRODUCTION
Across all cultures, patients with chronic pain have
among the lowest reported quality‐of‐life scores of
any medical condition and have comorbid mood
or anxiety disorders more frequently than the nor-
mal population (Becker et al., 1997; Gureje et al.,
1998). In addition, pain has a notable negative
impact on economies, with high direct costs (e.g.,
physician visits, physiotherapy, medication) and,
more strikingly, lowered productivity (e.g., lost
work days, reduced performance at work) charac-
terized in numerous countries around the world
(Phillips, 2006).
Much chronic pain falls into two distinct cat-
egories: inflammatory pain and neuropathic pain.
Chronic inflammation is typically accompanied
by pain due to the release of prostaglandins and
their consequent sensitization of sensory neurons
(Mense, 1983). Common forms of inflammatory
pain are rheumatoid arthritis, chronic inflamma-
tory bowel disease, pancreatitis, back pain, and
some cancer pain. In contrast, neuropathic pain
is caused by lesion or dysfunction of nervous tis-
sue; common forms are associated with diabetes,
herpes zoster infection (shingles), stroke, mul-
tiple sclerosis, cancer chemotherapy, and both
HIV infection and antiretroviral HIV treatment.
Neuropathic pain can also be idiopathic (Lauria
et al., 2014).
Severe pain is typically treated with medica-
tion, most commonly cyclooxygenase inhibitors
such as nonsteroidal anti‐inflammatory drugs
(NSAIDs) and opioids. Unfortunately, these drugs
can present serious dilemmas for both patients and
healthcare providers. Many patients view opioids
as undesirable because of their fear of addiction
or their experiences with cognitive or other side
effects (Welshman, 2005); prescription of opioids
is contraindicated specifically in patients with pre-
existing substance abuse issues. In addition, given
present levels of prescription drug abuse (Allread
and Paul, 2014; Manchikanti, 2006), there is an
appropriate hesistancy of healthcare providers
to prescribe or even fill prescriptions for opioids
(Glajchen, 2001; Kim et  al., 2000). The NSAIDs,
conversely, have little to no addiction liability
but are poorly effective against neuropathic pain
(Moore et al., 2015) and can have serious cardiovas-
cular and gastrointestinal side effects (Brune and
Patrignani, 2015). Thus, effective and nonaddictive
treatments with few side effects are needed. With
the understanding that chronic pain syndromes
likely involve derangement of homeostasis, there
is an increased interest in applying treatments
that aim to normalize metabolism, namely, meta-
bolic therapies. Many metabolic therapies involve
dietary changes. Whereas some dietary metabolic
therapies narrowly address specific genetic disor-
ders that include pain as a symptom (Roe et  al.,
2008), this survey of the literature focuses on
more wholesale dietary changes, some long estab-
lished, and their effects on common types of acute,
inflammatory, and neuropathic pain.
FA S T I N G
One type of dietary metabolic therapy with a his-
tory of use in pain is fasting, or a total or near
total (<500 kcal/d) reduction in calorie intake
(Michalsen, 2010). Fasting has been practiced for
thousands of years in many cultures as a treatment
for various disorders (as well as for religious rea-
sons). Biochemically, fasting results in lowered
levels of circulating glucose, as intake stops and
glycogen stores are metabolized, and in greatly
enhanced circulating levels of molecules known
as ketone bodies, which are produced by the liver
from fatty acids and can be used as fuel for the tri-
carboxylic acid cycle in the relative absence of glu-
cose. Subjects report a counterintuitive reduction
in hunger after a few days, as use of ketone bod-
ies becomes maximal (Michalsen, 2010). Fasting
can be used as a treatment for obesity and meta-
bolic syndrome, and is anticonvulsant in epileptic
patients (and so spawned the ketogenic diet, see
below). In the context of pain, some of the earliest
work concerned rheumatoid arthritis (Eisenberg,
197
Chapter 22: Metabolic Therapy and Pain
1956). Seven days of fasting significantly alleviated
joint pain and stiffness in rheumatoid arthritis suf-
ferers (Kroker et al., 1984; Michalsen et al., 2005;
Sköldstam et  al., 1979), apparently through anti‐
inflammatory actions including modified neutro-
phil function (Hafström et  al., 1988; Udén et  al.,
1983). Eight fasting days significantly improved
pain scores in osteoarthritis, an effect that lasted
weeks after the end of fasting (Schmidt et  al.,
2010). One to 2 weeks of fasting significantly
alleviated pain in the pain disorder fibromyalgia
(Michalsen et  al., 2005; Michalsen et  al., 2013).
Given the commonness of mood and anxiety dis-
orders in chronic pain (Gureje et al., 2008), fasting
has the additional benefit of enhancing mood after
several days, an effect not limited to pain sufferers
(Michalsen, 2010).
These clinical antinociceptive effects are sup-
ported by animal experiments. In a rat model of
postischemic pain, allodynia (i.e., the perception
of normally nonpainful stimuli as painful) to cold
and tactile stimulation in the ischemic paw was
significantly reduced if the animal had been fasted
for 18 hours prior to ischemia. (Given the higher
metabolic rate of rats, this time is comparable to a
longer fast in a human.) In addition, the effect of
the fast was totally reversed by glucose injection
during the ischemic event, highlighting the impor-
tance of lowered glucose (Ross‐Huot et al., 2011).
The anti‐inflammatory effects of fasting seem to
be mediated by ketosis (a state of elevated ketone
bodies) which limits oxidative stress and produc-
tion of free radicals and reactive oxygen species
(Veech, 2014).
Unfortunately, the clinical benefits of prolonged
fasting for rheumatoid arthritis are not long lasting
(unlike for osteoarthritis; Sköldstam et al., 1979),
and fasting is necessarily a time‐limited under-
taking, especially for subjects who do not have
extra adipose tissue to burn through over days to
weeks. An alternative procedure is intermittent
fasting, usually taking the form of every‐other‐
day (EOD) fasting, which can be implemented for
extended periods. Notably, in humans a 24-h fast
is sufficient to limit activation of the NLR fam-
ily, pyrin domain containing 3 (NLRP3) inflam-
masome activation (Traba et  al., 2015), which is
associated with hyperalgesia (Bullón et al., 2015);
therefore EOD fasting might be beneficial in rheu-
matoid arthritis. This effect on the NLRP3 inflam-
masome seems to be mediated by ketone bodies
(Youm et al., 2015). Studies in humans (Rothschild
et al., 2014) and rodents (Poon et al., 2006; Tajes
et  al., 2010; Tikoo et  al., 2007)  have shown posi-
tive effects of EOD fasting on various aspects of
health, but none have addressed pain. Although
197
EOD fasting generally causes weight loss clinically
(Varady et al., 2015), there is minimal body weight
change in many rodent studies (Tajes et al., 2010;
Tikoo et al., 2007). Therefore, studies of EOD fast-
ing in animals could dissociate weight loss from
antinociceptive effects, essentially impossible in
clinical studies of prolonged standard fasting.
Animal studies suggest that subjects who do not
need or wish to lose weight may limit this effect
of EOD fasting by exercising (Sakamoto and
Grunewald, 1987).
CALORIE RESTRICTION
Those who wish for pain alleviation but who can-
not tolerate long periods of time without food
might consider instead calorie restriction (CR).
Typically, this procedure limits total daily calorie
intake to 60% of normal (not less than ~900 kcal/d).
There is abundant experimental evidence for anti-
nociceptive effects of CR on acute pain (Wideman
et  al., 1996), inflammatory pain (Hargraves and
Hentall, 2005), and heat hyperalgesia in arthri-
tis models (Jurcovicova et  al., 2001). Clinically,
CR (applied for 12 weeks to 1 year, and typically
combined with exercise) in overweight patients
reduced back pain (thought to involve inflamma-
tion; Roffey et al., 2011), knee pain (White et al.,
2015), overall body pain (Landaeta‐Díaz et  al.,
2013), osteoarthritic pain (de Luis et al., 2012), and
prediabetic neuropathic pain (Smith et al., 2006).
All these benefits could be secondary to weight
loss, which would lessen physical wear on joints
and backs (as suggested in most of these papers),
though this explanation does not appear to apply
for the positive outcome in diabetic neuropathy, in
which pain reduction related to actual reinnerva-
tion of the skin (Smith et al., 2006). Testing of CR
treatment in fibromyalgia (see below), which is not
limited to overweight individuals, could dissociate
weight loss from CR‐related pain relief.
Apart from possible weight loss effects, it is
unclear whether ketosis contributes to this anti-
nociception, as none of these clinical studies
measured blood or urinary ketones. However,
improved glycemic control might be involved. As
already noted, low circulating glucose is neces-
sary for fasting amelioration of postischemic pain
(Ross‐Huot et al., 2011). A meta‐analysis of stud-
ies of diabetic neuropathy (including almost 8,000
patients) concluded that more aggressive glucose
control was associated with a lower incidence of
neuropathic pain and improved electrophysiologi-
cal measures (Callaghan et  al., 2012). In animal
studies, glycemic control partially alleviated dia-
betic neuropathy (Yorek et  al., 2014). Also, par-
tial glycolytic inhibition by 2‐deoxy‐D‐glucose
198

198 section III: Ketogenic Diet in the Laboratory

decreases sensation of acute pain (Bodnar et  al.,
1979) by central mechanisms (Bodnar et al., 1981).
This action does not involve endogenous opioids,
but seems to involve the ubiquitous neuromodu-
lator adenosine, which has broad analgesic effects
(Sawynok, 2015; Yamaoka et  al., 2013)  and is
released by 2‐deoxy‐D‐glucose or mildly lowered
glucose (Kawamura et al., 2014; Zhao et al., 1997).
Notably, fibromyalgia has been associated with
disordered glycolysis (Eisinger et al., 1994). One of
the regulators of glycolysis is the enzyme adenosine
monophosphate‐activated protein kinase (AMPK;
Carling, 2005), which has also been linked to sen-
sitization of nociceptors in pain states (Tillu et al.,
2012). Cells from fibromyalgia sufferers displayed
underactive AMPK and a related range of meta-
bolic impairments (Alcocer‐Gómez et  al., 2015;
Bullón et al., 2016). These problems were reversed
by incubation with serum from CR‐treated mice
(Alcocer‐Gómez et  al., 2015). Pharmacological
blockade of AMPK produced hyperalgesia in mice
that was reversed by CR (Bullón et  al., 2015). In
fibromyalgia sufferers, pharmacological activation
of AMPK reversed metabolic impairments and sig-
nificantly improved a range of symptoms includ-
ing pain (Bullón et  al., 2015). Thus CR seems to
restore balanced metabolism through an AMPK/
glycolysis pathway, one that can be engaged phar-
macologically with AMPK activators.
K E TO G E N I C   D I E T
Those who wish for pain alleviation but who can
tolerate neither CR nor fasting might consider the
ketogenic diet (KD). This diet was based on the
hypothesis that the beneficial effects of fasting in
epilepsy were due to the elimination of carbohy-
drates (Wilder, 1921a, 1921b); thus, the KD mini-
mizes carbohydrates and replaces the lost calories
with fats, and is in fact strongly anticonvulsant
(Dressler et al., 2015; Hallböök et al., 2015; Megaw
and Wilmshurst, 2015; Walker and Said, 2015) and
mounting evidence suggests that it is antiepilepto-
genic (Gama et al., 2015; Hu et al., 2011; Jiang et al.,
2012; Lusardi et al., 2015; Muller‐Schwarze et al.,
1999; Neal et  al., 2008; Su et  al., 2000; Todorova
et  al., 2000; Xu et  al., 2006). The KDs result in
ketosis and moderately lowered glucose like fast-
ing or sufficiently strong CR. Our animal stud-
ies showed that KDs have a clear antinociceptive
effect in acute thermal pain. This effect was pres-
ent in both male and female rats (Figure 22.1) and
was long‐lasting (Figure 22.2). Notably, the effect

Plate temperature (°C) Plate temperature (°C)

46 47 48 49 50 51 46 47 48 49 50 51

60 60
Latency to nocifensive behavior (s)

Males Females
50 50

40 40

30 30

20 20
Control diet
10 Ketogenic diet 10

0 0
14 15 16 17 18 19 14 15 16 17 18 19
Days on diet treatment Days on diet treatment

FIGURE 22.1 Sprague‐Dawley rats on a control diet or a 78% fat KD were challenged with a temperature‐response
curve for nocifensive behavior. One temperature was tested per day. Top and bottom X‐axes illustrate hotplate
temperature and days on diet, respectively. Dietary treatments began at 3–3½ weeks of age. Y‐axis indicates latency
to avoidance behavior, e.g. paw lifting, licking, or fluttering. All tests had a 60-s ceiling. At left, KD treatment in young
male rats causes heat hypoalgesia at moderate 49°C, 50°C, and 51°C temperatures. Diet F = 5.7, p = .19; Temperature
F = 158.4, p < .001; Diet‐x‐Temperature interaction F = 4.0, p < .001; n = 18–20. Neuman‐Keuls comparisons at each
temperature: *p < .05, ** p < .01, ***p < .001. Adapted from Ruskin et al. (2013). At right, KD treatment in young female
rats also causes heat hypoalgesia at the same temperatures. Diet F = 28.2, p < .001; Temperature F = 134.2, p < .001; Diet ×
Temperature interaction F = 9.6, p < .001; n = 12–16. All points are mean ± standard error. Neuman‐Keuls comparisons
at each temperature: ***p < .001. Unpublished data.
199
40
Control diet
Latency to nocifensive response (s)
Ketogenic diet
30
20
10
0 for fasting. Such findings might appear contradic-
0 5 10 15 20
Days on diet treatment
FIGURE  22.2 The development of KD effects on
pain over time. Male Sprague‐Dawley rats were placed
on a control diet or a 78% fat KD at 3–3½ weeks of age.
Temperature‐response curves as in Figure  22.1 were
used; for clarity of the progressive effect, data from
only the highest (51°C) tested temperature are shown.
Each point represents a different squad of rats. Several
days of treatment are necessary for the KD‐related heat
hypoalgesia to emerge. T‐tests at each length of treatment
were protected for multiple comparisons. (*)  p  =  .056,
**p < .01, ***p < .001. Adapted from Ruskin et al. (2009)
and Ruskin et al. (2013).
had a slower onset than ketosis and lowered blood
glucose (Figure 22.2), both of which were already
significant at 2 days of KD feeding (Ruskin et al.,
2013), suggesting that these biochemical changes
are not directly involved.
Clinically, the KD was used very early as a treat-
ment for pain (Maggioni et al., 2011). For instance,
in one study KD feeding improved or completely
controlled migraine in 39 of 50 patients (Baborka,
1930). More recent results support its antimi-
graine efficacy (Di Lorenzo et al., 2013, 2015). The
KD treatment was effective in treating inflamma-
tory bowel syndrome, including pain symptoms
(Austin et  al., 2009). Overall body pain was sig-
nificantly improved in one study of overweight
subjects (Guldbrand et  al., 2014)  and was very
near significance in another (Yancy et  al., 2009);
quality of life was significantly improved in both
studies. One of these studies specifically involved
overweight diabetics (Guldbrand et al., 2014), but
did not parse the types of pain involved, so it is
unknown whether KD feeding alleviated neuro-
pathic pain in these patients. We found no benefi-
cial effect of KD in models of nerve constriction
and chemotherapeutic neuropathic pain (Masino
and Ruskin, 2013); it remains unclear whether KD
Chapter 22: Metabolic Therapy and Pain 199
is poorly effective against neuropathic pain gen-
erally, or rather has type‐specific positive effects.
Regardless, abundant evidence shows that KDs are
clinically beneficial in diabetes (e.g. Adams, 1931;
Mobbs et al., 2013).
Antinociceptive effects of the KD might
be strongest in inflammatory pain, as this
diet is anti‐inflammatory (Pérez‐Guisado and
Muñoz‐Serrano, 2011; Ruskin et al., 2009; Sullivan
et  al., 2004; Tendler et  al., 2007), and this effect
might be due largely to ketosis, as reviewed earlier
25 30 tory to much published literature showing that
high‐fat diets promote inflammation (Johnson
and Makowski, 2015); however, this literature
refers to diets such as the so‐called Western diet,
high in fat but not low in carbohydrates. The
metabolic response to dietary fats differs greatly
depending on the presence of carbohydrates:  the
high‐fat‐plus‐carbohydrate diet promotes fat stor-
age, whereas the high fat, low‐carbohydrate diet
promotes fat metabolism.
Apart from anti‐inflammatory effects, some
antinociceptive effects of the KD might relate
directly to its high fat content. The receptor
GPR40 binds and is activated by free medium‐
and long‐chain fatty acids, and thus is sometimes
called free fatty acid receptor 1 (Covington et al.,
2006). Although this receptor may be considered
a nutrient sensor, it is expressed in the central
nervous system and turns out to affect nocicep-
tion. In animal testing, synthetic GPR40 agonists
and natural ligands injected intrathecally or intra-
cerebroventricularly reduce inflammatory and
neuropathic pain in numerous models (Harada
et  al., 2014; Karki et  al., 2015; Nakamoto et  al.,
2013; Nakamoto et  al., 2015). There are elec-
trophysiological correlates of these behavioral
effects (Karki et  al., 2015). Conversely, GPR40
knockout augments the arthritic phenotype in
an osteoarthritis model (Monfoulet et  al., 2015),
although pain behavior was not studied in this
report. In “Western” high‐fat diets, antinocicep-
tive effects of GPR40 activation might be unable
to compete against the pro-inflammatory nature
of these diets.
Regarding epilepsy, there is some controversy
over whether the KD is more effective when com-
bined with CR (Bough et  al., 1999b; Raffo et  al.,
2008). Certainly, KD plus CR will lead to stronger
ketosis and a more pronounced lowering of circu-
lating glucose. If the KD’s antinociceptive effects
are primarily mediated by fatty acids, however,
the addition of CR should add little benefit to this
outcome.
200
200 section III: Ketogenic Diet in the Laboratory
P O LY U N S AT U R AT E D
FAT T Y   A C I D S
Some have suggested that many of the beneficial
effects of the KD in epilepsy and general health are
due to enhanced dietary levels of polyunsaturated
fatty acids (PUFAs; (Fraser et al., 2003; Fuehrlein
et al., 2004; Paoli et al., 2015; Taha et al., 2005; Xu
et  al., 2008), based on anticonvulsant effects of
PUFAs in some studies (Taha et al., 2009; Voskuyl
et al., 1998; Yehuda et al., 1994). These effects are
controversial (Dahlin et  al., 2007; Dupuis et  al.,
2015; Willis et  al., 2009). Nevertheless, whereas
saturated fatty acids bind to GPR40, PUFAs addi-
tionally act on other molecular targets that could
modulate pain (see below), and PUFAs and their
epoxide metabolites indeed help in inflammatory
and neuropathic pain (Wagner et al., 2014). In ani-
mal studies, for instance, elimination of dietary
PUFAs promotes the neuropathic pain phenotype
after nerve injury (Martin and Avendaño, 2009).
Conversely, addition of fish oil (rich in PUFAs)
to the diet reverses diabetic neuropathy includ-
ing derangement of pain sensation (Coppey et al.,
2012; Coppey et  al., 2015; Redivo et  al., 2015).
Clinically, PUFA efficacy is good in inflammatory
pain (Goldberg and Katz, 2007; Lee et al., 2012) but
not osteoarthritis (Boe and Vangsness, 2015).
Some researchers believe that the various
health benefits of PUFA supplements are due to
the high ratio of ω‐3 to ω‐6 PUFAs, two types of
PUFA that differ in the location of double bonds
in their carbon backbones. Fish oil has a high ω‐3/
ω‐6 ratio. Thus, animal studies have specifically
enriched diets with ω‐3 PUFAs and found that this
reduces acute pain and neuropathic pain after spi-
nal cord contusion (Escudero et al., 2015; Figueroa
et  al., 2013). An additional benefit was that ω‐3
PUFAs reduced the development of tolerance to
morphine (Escudero et  al., 2015). On the other
hand, there is other evidence that ω‐3 PUFAs pro-
mote neuropathic pain (Pérez et al., 2005). In clini-
cal studies, ω‐3 PUFA supplementation improved
pain and quality of life in chronic headache suffer-
ers (Ramsden et al., 2013; Ramsden et al., 2015).
How might PUFAs influence pain? PUFAs are
agonists of the peroxisome proliferator‐activated
receptors (PPARs), another type of nutrient sen-
sor that also happens to be involved in nociception
(Freitag and Miller, 2014). For instance, activation
of PPARγ subtype alleviates inflammatory and
neuropathic pain in animal models (Morgenweck
et  al., 2010; Morgenweck et  al., 2013); PPARα
agonists given to patients reduce pain in a num-
ber of inflammatory and neuropathic conditions
(Freitag and Miller, 2014). Additionally, PUFAs
augment voltage‐activated A‐type (Kv1.4) potas-
sium channels (Tigerholm et  al., 2012; Xu et  al.,
2008), which could reduce hyperexcitability of
sensitized neural tissue. Certain PUFAs also block
pain due to activation of the heat‐sensing transient
receptor potential vanilloid subtype I  (the “cap-
saicin receptor”), an inflammation‐related recep-
tor expressed by peripheral nociceptive neurons
(Matta et al., 2007).
OT H E R M E C H A N I S M S
Ketone bodies, augmented in fasting, CR, and
KD feeding, might act on pain through mul-
tiple mechanisms. As already noted, a ketone
body–based metabolism limits inflammation.
A  ketone body–based metabolism also produces
ATP more efficiently than one based on glucose
(Veech, 2004). Thus, ketone body treatment leads
to elevated levels of ATP and phosphocreatine
(for instance, Deng‐Bryant et al., 2011; Kim et al.,
2010; Tieu et al., 2003), which might allow restora-
tion of energy homeostasis in distressed and pain-
ful tissue. Besides effects on energy, ketone body
treatment modulates glutamate and its recep-
tors (Chmiel‐Perzyńska et  al., 2011; Donevan
et  al., 2003; Juge et  al., 2010; Lund et  al., 2009;
Żarnowsky et al., 2012), γ‐aminobutyric acid and
its receptors (Erecińska et  al., 1996; Yang et  al.,
2007), potassium channels (Giménez‐Cassina
et  al., 2012; Ma et  al., 2007; Tanner et  al., 2011),
and the receptor PUMA‐G (Taggart et al., 2005) in
a manner often leading to reduced excitability
that could limit activity in sensitized nociceptive
systems. Multiple mechanisms acting in concert
might be especially effective in normalizing the
deranged metabolism related to inflammation and
sensitized neural tissue. Future work on ketone
bodies will surely be aided by the availability of
ketone esters (Kashiwaya et  al., 2010; Viggiano
et al., 2015), which can be taken orally and are rap-
idly metabolized to ketone bodies, thus bypassing
the need for wholesale dietary changes.
On the other hand, regarding the KD, some
have concluded that ketone bodies are not
directly involved in its anticonvulsant effect
(Bough et  al., 1999a, 2000; Harney et  al., 2002;
Likhodii et  al., 2000; Raffo et  al., 2008), similar
to our interpretation of KD reductions in ther-
mal pain (Ruskin et  al., 2013). In the context
of pain relief, what might low blood glucose
do besides promote ketosis? Such a question
is necessarily difficult to address given that the
former produces the latter. Fatty acid oxidation,
and hence ketone body synthesis, can be blocked
pharmacologically in vivo, but such work has
201
Chapter 22: Metabolic Therapy and Pain
mostly focused on studies of ingestive behavior
(Langhans et  al., 2011). Fortunately, the ketosis
complication is removed in in vitro prepara-
tions. With single‐neuron recording in hippo-
campal slices, we investigated a model of the KD
in which intracellular ATP was at medium‐to‐
high concentration and extracellular glucose was
moderately low. This treatment hyperpolarized
the recorded neuron, an effect mediated by ATP
released through pannexin channels, metabo-
lized extracellularly to adenosine, which acted
on adenosine A1 receptors to open potassium
channels (Kawamura et al., 2010). We suggested
that this autocrine inhibitory mechanism is the
major contributor to the KD’s anticonvulsant
effect (Masino et al., 2011). If this same mecha-
nism were recruited in nociceptive brainstem
and spinal cord tissue during KD (or fasting
or CR) treatment, it would have clear analgesic
effects, possibly most strikingly in hyperexcitable
sensitized tissue.
Promotion of adenosine transmission by phar-
macological means (promoting synthesis, block-
ing uptake/metabolism, allosterically modulating
receptors) ameliorates pain in a wide variety of
models (e.g., Hurt and Zylka, 2012; Imlach et al.,
2015; Little et al., 2015; Maes et al., 2012; Martins
et  al., 2013; Street et  al., 2011). Even pain treat-
ment by acupuncture and exercise seems to
involve enhanced adenosine (Goldman et  al.,
2010; Martins et al., 2013). The presence of a low‐
glucose‐activated adenosine‐based autoinhibitory
mechanism remains to be directly shown in brain-
stem and spinal cord. However, adenosine also
has antinociceptive effects in the periphery (Lima
et  al., 2010; Liu et  al., 2013; Poon and Sawynok,
1999; Valério et  al., 2009). In addition, immune
cells express several types of adenosine recep-
tor that can influence inflammation (Kumar and
Sharma, 2009).
There are other nutrient sensors besides
GPR40 and the PPARs that could be involved in
nociception. GPR41 (free fatty acid receptor 3)
is expressed in the peripheral nervous system
and is activated by short‐chain fatty acids and is
yet another molecular target for ketone bodies
(Kimura et  al., 2011; Won et  al., 2013). GPR120
(free fatty acid receptor 4), activated by PUFAs,
is known to be anti‐inflammatory (Cintra et  al.,
2012). Studies have yet to investigate any specific
involvement in pain for either of these receptors.
Many of the postulated antinociceptive
mechanisms of fasting, CR, KD feeding, and
PUFA supplementation can be recruited phar-
macologically:  for example, metformin to
201
control glycemia, selective agonists to activate
nutrient sensors, adenosine A1 receptor ago-
nists to limit neural tissue hyperexcitability.
Whereas such shortcuts may be useful in cases
where the dietary treatment cannot be followed,
they can have their own problems; for instance,
the striking bradycardia that occurs after suf-
ficiently strong A1 receptor activation (Jonzon
et al., 1986). Dietary recruitment of these mech-
anisms will be more physiological. In addition,
this review has suggested that some of these
dietary treatments are likely to involve multiple
mechanisms that might work in synergy—drug
treatments would lack this advantage. Multiple
mechanisms acting together could be ideal for
restoring disordered metabolism. Furthermore,
these dietary treatments are all possess a wide
range of health benefits, making them attrac-
tive for people to optimize their health as well as
manage their pain.
AC K N OW L E D G M E N T S
Supported by National Institutes of Health
NS065446 and AT008742.
REFERENCES
Adams, S.F. (1931). The benefit of prolonged low car-
bohydrate feeding in diabetes:  report of a case.
Mayo Clin Proc 6, 534–535.
Alcocer‐Gómez, E., Garrido‐Maraver, J., Bullón, P.,
Marín‐Aguilar, F., Cotán, D., Carrión, A.M.,
Alvarez‐Suarez, J.M., Giampieri, F., Sánchez‐
Alcazar, J.A., Battino, M., et al. (2015). Metformin
and caloric restriction induce an AMPK‐dependent
restoration of mitochondrial dysfunction in
fibroblasts from fibromyalgia patients. Biochim
Biophys Acta 1852, 1257–1267.
Allread, V., and Paul, S. (2014). The misuse and abuse
of prescription medications: part 1 current trends.
MD Advis 7, 12–20.
Austin, G.L., Dalton, C.B., Hu, Y., Morris, C.B.,
Hankins, J., Weinland, S.R., Westman, E.C., Yancy,
W.S., Jr., and Drossman, D.A. (2009). A very low‐
carbohydrate diet improves symptoms and quality
of life in diarrhea‐predominant irritable bowel syn-
drome. Clin Gastroenterol Hepatol 7, 706–708.e1.
Baborka, C.J. (1930). Migraine:  results of treatment
by ketogenic diet in fifty cases. Mayo Clin Proc 5,
190–191.
Becker, N., Bondegaard‐Thomsen, A., Olsen, A.K.,
Sjøgren, P., Bech, P., and Eriksen, J. (1997). Pain
epidemiology and health related quality of life in
chronic non‐malignant pain patients referred to
a Danish multidisciplinary pain center. Pain 73,
393–400.
Bodnar, R.J., Kelly, D.D., and Glusman, M. (1979).
2‐Deoxy‐D‐glucose analgesia: influences of opiate
202
202 section III: Ketogenic Diet in the Laboratory
and non‐opiate factors. Pharmacol Biochem Behav
11, 297–301.
Bodnar, R.J., Merrigan, K.P., and Wallace, M.M. (1981).
Analgesia following intraventricular administra-
tion of 2‐deoxy‐D‐glucose. Pharmacol Biochem
Behav 14, 579–581.
Boe, C., and Vangsness, C.T. (2015). Fish oil and osteo-
arthritis:  current evidence. Am J Orthop (Belle
Mead NJ) 44, 302–305.
Bough, K.J., Chen, R.S., and Eagles, D.A. (1999a). Path
analysis shows that increasing ketogenic ratio, but
not β‐hydroxybutyrate, elevates seizure threshold
in the rat. Dev Neurosci 21, 400–406.
Bough, K.J., Valiyil, R., Han, F.T., and Eagles, D.A.
(1999b). Seizure resistance is dependent upon age
and calorie restriction in rats fed a ketogenic diet.
Epilepsy Res 35, 21–28.
Bough, K.J., Yao, S.G., and Eagles, D.A. (2000).
Higher ketogenic diet ratios confer protection
from seizures without neurotoxicity. Epilepsy Res
38, 15–25.
Brune, K., and Patrignani, P. (2015). New insights into
the use of currently available non‐steroidal anti‐
inflammatory drugs. J Pain Res 8, 105–118
Bullón, P., Alcocer‐Gómez, E., Carrión, A.M.,
Marín‐Aguilar, F., Garrido‐Maraver, J., Román‐
Malo, L., Ruiz‐Cabello, J., Culic, O., Ryffel, B.,
Apetoh, L., et al. (2016). AMPK phosphorylation
modulates pain by activation of NLRP3 inflamma-
some. Antioxid Redox Signal 24, 157–170.
Callaghan, B.C., Little, A.A., Feldman, E.L., and
Hughes, R.A. (2012). Enhanced glucose control
for preventing and treating diabetic neuropathy.
Cochrane Database Syst Rev 6, CD007543.
Carling, D. (2005). AMP‐activated protein kinase: bal-
ancing the scales. Biochimie 87, 87–91.
Chmiel‐Perzyńska, I., Kloc, R., Perzyński, A., Rudzki,
S., and Urbańska, E.M. (2011). Novel aspect of
ketone action:  β‐hydroxybutyrate increases brain
synthesis of kynurenic acid in vitro. Neurotox Res
20, 40–50.
Cintra, D.E., Ropelle, E.R., Moraes, J.C., Pauli, J.R.,
Morari, J., Souza, C.T., Grimaldi, R., Stahl, M.,
Carvalheira, J.B., Saad, M.J., et  al. (2012).
Unsaturated fatty acids revert diet‐induced hypo-
thalamic inflammation in obesity. PLoS One 7,
e30571.
Coppey, L.J., Holmes, A., Davidson, E.P., and Yorek,
M.A. (2012). Partial replacement with menhaden
oil improves peripheral neuropathy in high‐fat‐fed
low‐dose streptozotocin type 2 diabetic rat. J Nutr
Metab 2012, 950517.
Coppey, L.J., Davidson, E.P., Obrosov, A., and Yorek,
M.A. (2015). Enriching the diet with menhaden
oil improves peripheral neuropathy in streptozoto-
cin‐induced type 1 diabetic rats. J Neurophysiol
113, 701–708.
Covington, D.K., Briscoe, C.A., Brown, A.J., and
Jayawickreme, C.K. (2006). The G‐protein‐coupled
receptor 40 family (GPR40‐GPR43) and its role in
nutrient sensing. Biochem Soc Trans 34, 770–773.
Dahlin, M., Hjelte, L., Nilsson, S., and Åmark, P.
(2007). Plasma phospholipid fatty acids are influ-
enced by a ketogenic diet enriched with n‐3 fatty
acids in children with epilepsy. Epilepsy Res 73,
199–207.
de Luis, D.A., Izaola, O., Alonso, M.G., Aller, R.,
Cabezas, G., and de la Fuente, B. (2012). Effect
of a hypocaloric diet with a commercial for-
mula in weight loss and quality of life in obese
patients with chronic osteoarthritis. Nutr Hosp
27, 1648–1654.
Deng‐Bryant, Y., Prins, M.L., Hovda, D.A., and Harris,
N.G. (2011). Ketogenic diet prevents alterations
in brain metabolism in young but not adult rats
after traumatic brain injury. J Neurotrauma 28,
1813–1825.
Di Lorenzo, C., Currà, A., Sirianni, G., Coppola, G.,
Bracaglia, M., Cardillo, A., De Nardis, L., and
Pierelli, F. (2013). Diet transiently improves
migraine in two twin sisters: possible role of keto-
genesis? Funct Neurol 28, 305–308.
Di Lorenzo, C., Coppola, G., Sirianni, G., Di Lorenzo,
G., Bracaglia, M., Di Lenola, D., Siracusano, A.,
Rossi, P., and Pierelli, F. (2015). Migraine improve-
ment during short lasting ketogenesis: a proof‐of‐
concept study. Eur J Neurol 22, 170–177.
Donevan, S.D., White, H.S., Anderson, G.D., and
Rho, J.M. (2003). Voltage‐dependent block of
N‐methyl‐D‐aspartate receptors by the novel
anticonvulsant dibenzylamine, a bioactive con-
stituent of L‐(+)‐β‐hydroxybutyrate. Epilepsia 44,
1274–1279.
Dressler, A., Trimmel‐Schwahofer, P., Reithofer, E.,
Mühlebner, A., Gröppel, G., Reiter‐Fink, E.,
Benninger, F., Grassl, R., and Feucht, M. (2015).
Efficacy and tolerability of the ketogenic diet in
Dravet syndrome: comparison with various stan-
dard antiepileptic drug regimen. Epilepsy Res
109, 81–89.
Dupuis, N., Curatolo, N., Benoist, J.‐F., and Stephane
Auvin, S. (2015). Ketogenic diet exhibits anti‐
inflammatory properties. Epilepsia 56, e95–e98.
Eisenberg, W. (1956). Zur Behandlung des primär‐
chronischen Gelenkrheumatismus durch Fasten.
Hippokrates 27, 688–692.
Eisinger, J., Plantamura, A., and Ayavou, T. (1994).
Glycolysis abnormalities in fibromyalgia. J Am
Coll Nutr 13, 144–148.
Erecińska, M., Nelson, D., Daikhin, Y., and Yudkoff,
M. (1996). Regulation of GABA level in rat brain
synaptosomes:  fluxes through enzymes of the
GABA shunt and effects of glutamate, calcium,
and ketone bodies. J Neurochem 67, 2325–2334.
Escudero, G.E., Romañuk, C.B., Toledo, M.E.,
Olivera, M.E., Manzo, R.H., and Laino, C.H.
(2015). Analgesia enhancement and preven-
tion of tolerance to morphine:  beneficial effects
203
Chapter 22: Metabolic Therapy and Pain
of combined therapy with omega‐3 fatty acids. J
Pharm Pharmacol 67, 1251–1262.
Figueroa, J.D., Cordero, K., Serrano‐Illan, M.,
Almeyda, A., Baldeosingh, K., Almaguel, F.G.,
and De Leon, M. (2013). Metabolomics uncovers
dietary omega‐3 fatty acid‐derived metabolites
implicated in anti‐nociceptive responses after
experimental spinal cord injury. Neuroscience
255, 1–18.
Fraser, D.D., Whiting, S., Andrew, R.D., Macdonald,
E.A., Musa‐Veloso, K., and Cunnane, S.C. (2003).
Elevated polyunsaturated fatty acids in blood
serum obtained from children on the ketogenic
diet. Neurology 60, 1026–1029.
Freitag, C.M., and Miller, R.J. (2014). Peroxisome
proliferator‐activated receptor agonists modulate
neuropathic pain: a link to chemokines? Front Cell
Neurosci 8, 238.
Fuehrlein, B.S., Rutenberg, M.S., Silver, J.N., Warren,
M.W., Theriaque, D.W., Duncan, G.E., Stacpoole,
P.W., and Brantly, M.L. (2004). Differential meta-
bolic effects of saturated versus polyunsaturated
fats in ketogenic diets. J Clin Endocrinol Metab
89, 1641–1645.
Gama, I.R., Trindade‐Filho, E.M., Oliveira, S.L.,
Bueno, N.B., Melo, I.T., Cabral‐Junior, C.R.,
Barros, E.M., Galvão, J.A., Pereira, W.S., Ferreira,
R.C., et al. (2015). Effects of ketogenic diets on the
occurrence of pilocarpine‐induced status epilepti-
cus of rats. Metab Brain Dis 30, 93–98.
Giménez‐Cassina, A., Martínez‐François, J.R., Fisher,
J.K., Szlyk, B., Polak, K., Wiwczar, J., Tanner, G.R.,
Lutas, A., Yellen, G., and Danial, N.N. (2012).
BAD‐dependent regulation of fuel metabolism
and KATP channel activity confers resistance to epi-
leptic seizures. Neuron 74, 719–730.
Glajchen, M. (2001). Chronic pain:  treatment barri-
ers and strategies for clinical practice. J Am Board
Fam Pract 14, 211–218.
Goldberg, R.J., and Katz, J. (2007). A meta‐analysis of
the analgesic effects of omega‐3 polyunsaturated
fatty acid supplementation for inflammatory joint
pain. Pain 129, 210–223.
Goldman, N., Chen, M., Fujita, T., Xu, Q., Peng, W.,
Liu, W., Jensen, T.K., Pei, Y., Wang, F., Han, X.,
et  al. (2010). Adenosine A1 receptors mediate
local anti‐nociceptive effects of acupuncture. Nat
Neurosci 13, 883–888.
Guldbrand, H., Lindström, T., Dizdar, B.,
Bunjaku, B., Östgren, C.J., Nystrom, F.H., and
Bachrach‐Lindström, M. (2014). Randomization
to a low‐carbohydrate diet advice improves health
related quality of life compared with a low‐fat diet
at similar weight‐loss in Type 2 diabetes mellitus.
Diabetes Res Clin Pract 106, 221–227.
Gureje, O., Von Korff, M., Simon, G.E., and Gater, R.
(1998). Persistent pain and well‐being:  a World
Health Organization study in primary care. J Am
Med Assoc 280, 147–151.
203
Gureje, O., Von Korff, M., Kola, L., Demyttenaere, K.,
He, Y., Posada‐Villa, J., Lepine, J.P., Angermeyer,
M.C., Levinson, D., de Girolamo, G., et al. (2008).
The relation between multiple pains and mental
disorders:  results from the World Mental Health
Surveys. Pain 135, 82–91.
Hafström, I., Ringertz, B., Gyllenhammar, H.,
Palmblad, J., and Harms‐Ringdahl, M. (1988).
Effects of fasting on disease activity, neutrophil
function, fatty acid composition, and leukotriene
biosynthesis in patients with rheumatoid arthritis.
Arthritis Rheum 31, 585–592.
Hallböök, T., Sjolander, A., Amark, P., Miranda, M.,
Bjurulf, B., and Dahlin, M. (2015). Effectiveness of
the ketogenic diet used to treat resistant childhood
epilepsy in Scandinavia. Eur J Paediatr Neurol
19, 29–36.
Harada, S., Haruna, Y., Aizawa, F., Matsuura, W.,
Nakamoto, K., Yamashita, T., Kasuya, F., and
Tokuyama, S. (2014). Involvement of GPR40,
a long‐chain free fatty acid receptor, in the
production of central post‐stroke pain after
global cerebral ischemia. Eur J Pharmacol 744,
115–123.
Hargraves, W.A., and Hentall, I.D. (2005). Analgesic
effects of dietary caloric restriction in adult mice.
Pain 114, 455–461.
Harney, J.P., Madara, J., Madara, J., and I’Anson, H.
(2002). Effects of acute inhibition of fatty acid oxi-
dation on latency to seizure and concentrations of
β hydroxybutyrate in plasma of rats maintained
on calorie restriction and/or the ketogenic diet.
Epilepsy Res 49, 239–246.
Hu, X.‐L., Cheng, X., Fei, J., and Xiong, Z.‐Q. (2011).
Neuron‐restrictive silencer factor is not required
for the antiepileptic effect of the ketogenic diet.
Epilepsia 52, 1609–1616.
Hurt, J.K., and Zylka, M.J. (2012). PAPupuncture has
localized and long‐lasting antinociceptive effects
in mouse models of acute and chronic pain. Mol
Pain 8, 28.
Imlach, W.L., Bhola, R.F., May, L.T., Christopoulos,
A., and Christie, M.J. (2015). A positive allosteric
modulator of the adenosine A1 receptor selectively
inhibits primary afferent synaptic transmission
in a neuropathic pain model. Mol Pharmacol 88,
460–468.
Jiang, Y., Yang, Y., Wang, S., Ding, Y., Guo, Y., Zhang,
M.‐M., Wen, S.‐Q., and Ding, M.‐P. (2012).
Ketogenic diet protects against epileptogenesis as
well as neuronal loss in amygdaloid‐ kindling sei-
zures. Neurosci Lett 508, 22–26.
Johnson, A.R., and Makowski, L. (2015). Nutrition
and metabolic correlates of obesity and inflam-
mation:  clinical considerations. J Nutr 145,
1131S‐1136S.
Jonzon, B., Bergquist, A., Li, Y.‐O., and Fredholm, B.B.
(1986). Effects of adenosine and two stable ade-
nosine analogues on blood pressure, heart rate and
204
204 section III: Ketogenic Diet in the Laboratory
colonic temperature in the rat. Acta Physiol Scand
126, 491–498.
Juge, N., Gray, J.A., Omote, H., Miyaji, T., Inoue, T.,
Hara, C., Uneyama, H., Edwards, R.H., Nicoll,
R.A., and Moriyama, Y. (2010). Metabolic con-
trol of vesicular glutamate transport and release.
Neuron 68, 99–112.
Jurcovicova, J., Stanciková, M., Svík, K., Ondrejicková,
Krsova, D., Seres, J., and Rokyta, R. (2001). Stress
of chronic food restriction attenuates the develop-
ment of adjuvant arthritis in male Long Evans rats.
Clin Exp Rheumatol 19, 371–376.
Karki, P., Kurihara, T., Nakamachi, T., Watanabe, J.,
Asada, T., Oyoshi, T., Shioda, S., Yoshimura, M.,
Arita, K., and Miyata, A. (2015). Attenuation of
inflammatory and neuropathic pain behaviors in
mice through activation of free fatty acid receptor
GPR40. Mol Pain 11, 6.
Kashiwaya, Y., Pawlosky, R., Markis, W., King, M.T.,
Bergman, C., Srivastava, S., Murray, A., Clarke,
K., and Veech, R.L. (2010). A ketone ester diet
increases brain malonyl‐CoA and uncoupling pro-
teins 4 and 5 while decreasing food intake in the
normal Wistar rat. J Biol Chem 285, 25950–25956.
Kawamura, M., Jr., Ruskin, D.N., and Masino, S.A.
(2010). Metabolic autocrine regulation of neurons
involves cooperation among pannexin hemichan-
nels, adenosine receptors and KATP channels. J
Neurosci 30, 3886–3895.
Kawamura, M., Jr., Ruskin, D.N., Geiger, J.D., Boison,
D., and Masino, S.A. (2014). Ketogenic diet sensi-
tizes glucose control of hippocampal excitability. J
Lipid Res 55, 2254–2260.
Kim, D.Y., Vallejo, J., and Rho, J.M. (2010). Ketones
prevent synaptic dysfunction induced by mito-
chondrial respiratory complex inhibitors. J
Neurochem 114, 130–141.
Kim, K.S., Glajchen, M., and Portenoy, R.K. (2000).
Unwillingness to fill opioid analgesic prescriptions
by local pharmacies. J Pain Symptom Manage 20,
160–162.
Kimura, I., Inoue, D., Maeda, T., Hara, T., Ichimura,
A., Miyauchi, S., Kobayashi, M., Hirasawa, A., and
Tsujimoto, G. (2011). Short‐chain fatty acids and
ketones directly regulate sympathetic nervous sys-
tem via G protein‐coupled receptor 41 (GPR41).
Proc Natl Acad Sci USA 108, 8030–8035.
Kroker, G.F., Stroud, R.M., Marshall, R., Bullock, T.,
Carroll, F.M., Greenberg, M., Randolph, T.G., Rea,
W.J., and Smiley, R.E. (1984). Fasting and rheu-
matoid arthritis: a multicenter study. Clin Ecol 2,
137–144.
Kumar, V., and Sharma, A. (2009). Adenosine:  An
endogenous modulator of innate immune sys-
tem with therapeutic potential. Eur J Pharmacol
616, 7–15.
Landaeta‐Díaz, L., Fernández, J.M., Da
Silva‐Grigoletto, M., Rosado‐Alvarez, D., Gómez‐
Garduño, A., Gómez‐Delgado, F., López‐Miranda,
J., Pérez‐Jimenez, F., and Fuentes‐Jiménez, F.
(2013). Mediterranean diet, moderate‐to‐high
intensity training, and health‐related quality of
life in adults with metabolic syndrome. Eur J Prev
Cardiol 20, 555–564.
Langhans, W., Leitner, C., and Arnold, M. (2011).
Dietary fat sensing via fatty acid oxidation in
enterocytes:  possible role in the control of eat-
ing. Am J Physiol Regul Integr Comp Physiol 300,
R554–565.
Lauria, G., Ziegler, D., Malik, R., Merkies, I.S.,
Waxman, S.G., Faber, C.G., and group, P.S. (2014).
The role of sodium channels in painful diabetic
and idiopathic neuropathy. Curr Diab Rep 14, 538.
Lee, Y.H., Bae, S.C., and Song, G.G. (2012). Omega‐3
polyunsaturated fatty acids and the treatment of
rheumatoid arthritis:  a meta‐analysis. Arch Med
Res 43, 356–362.
Likhodii, S.S., Musa, K., Mendonca, A., Dell, C.,
Burnham, W.M., and Cunnane, S.C. (2000).
Dietary fat, ketosis, and seizure resistance in rats
on the ketogenic diet. Epilepsia 41, 1400–1410.
Lima, F.O., Souza, G.R., Verri, W.A., Jr., Parada, C.A.,
Ferreira, S.H., Cunha, F.Q., and Cunha, T.M.
(2010). Direct blockade of inflammatory hyper-
nociception by peripheral A1 adenosine recep-
tors:  Involvement of the NO/cGMP/PKG/KATP
signaling pathway. Pain 151, 506–515.
Little, J.W., Ford, A., Symons‐Liguori, A.M., Chen, Z.,
Janes, K., Doyle, T., Xie, J., Luongo, L., Tosh, D.K.,
Maione, S., et al. (2015). Endogenous adenosine A3
receptor activation selectively alleviates persistent
pain states. Brain 138, 28–35.
Liu, J., Reid, A.R., and Sawynok, J. (2013). Spinal sero-
tonin 5‐HT7 and adenosine A1 receptors, as well
as peripheral adenosine A1 receptors, are involved
in antinociception by systemically administered
amitriptyline. Eur J Pharmacol 698, 213–219.
Lund, T.M., Risa, Ø., Sonnewald, U., Schousboe, A.,
and Waagepetersen, H.S. (2009). Availability of
neurotransmitter glutamate is diminished when
β‐hydroxybutyrate replaces glucose in cultured
neurons. J Neurochem 110, 80–91.
Lusardi, T.A., Akula, K.K., Coffman, S.Q., Ruskin, D.N.,
Masino, S.A., and Boison, D. (2015). Ketogenic
diet prevents epileptogenesis and disease progres-
sion in adult mice and rats. Neuropharmacology
99, 500–509.
Ma, W., Berg, J., and Yellen, G. (2007). Ketogenic diet
metabolites reduce firing in central neurons by
opening KATP channels. J Neurosci 27, 3618–3625.
Maes, S.S., Pype, S., Hoffmann, V.L.H., Biermans, M.,
and Meert, T.F. (2012). Antihyperalgesic activity
of nucleoside transport inhibitors in models of
inflammatory pain in guinea pigs. J Pain Res 5,
391–400.
Maggioni, F., Margoni, M., and Zanchin, G. (2011).
Ketogenic diet in migraine treatment:  a brief but
ancient history. Cephalalgia 31, 1150–1151.
205
Chapter 22: Metabolic Therapy and Pain
Manchikanti, L. (2006). Prescription drug abuse: what
is being done to address this new drug epidemic?
Pain Physician 9, 287–321.
Martin, Y.B., and Avendaño, C. (2009). Effects of
removal of dietary polyunsaturated fatty acids on
plasma extravasation and mechanical allodynia
in a trigeminal neuropathic pain model. Mol
Pain 5, 8.
Martins, D.F., Mazzardo‐Martins, L., Soldi, F.,
Stramosk, J., Piovezan, A.P., and Santos, A.R.S.
(2013). High‐intensity swimming exercise reduces
neuropathic pain in an animal model of complex
regional pain syndrome type I: evidence for a role of
the adenosinergic system. Neuroscience 234, 69–76.
Masino, S.A., Li, T., Theofilas, P., Sandau, U., Ruskin,
D.N., Fredholm, B.B., Geiger, J.D., Aronica, E.,
and Boison, D. (2011). A ketogenic diet suppresses
seizures in mice through adenosine A1 receptors. J
Clin Invest 121, 2679–2683.
Masino, S.A., and Ruskin, D.N. (2013). Ketogenic diets
and pain. J Child Neurol 28, 993–1001.
Matta, J.A., Miyares, R.L., and Ahern, G.P. (2007).
TRPV1 is a novel target for omega‐3 polyunsatu-
rated fatty acids. J Physiol 578, 397–411.
Megaw, K., and Wilmshurst, J.M. (2015). The Keilah
Foundation: making the ketogenic diet viable for
children in Africa. Epilepsia 56, 514–516.
Mense, S. (1983). Basic neurobiological mechanisms
of pain. Am J Med 75, 4–14.
Michalsen, A. (2010). Prolonged fasting as a method of
mood enhancement in chronic pain syndromes: a
review of clinical evidence and mechanisms. Curr
Pain Headache Rep 14, 80–87.
Michalsen, A., Li, C., Kaiser, K., Lüdtke, R., Meier, L.,
Stange, R., and Kessler, C. (2013). In-patient treat-
ment of fibromyalgia: a controlled nonrandomized
comparison of conventional medicine versus inte-
grative medicine including fasting therapy. Evid
Based Complement Alternat Med 2013, 908610.
Michalsen, A., Riegert, M., Lüdtke, R., Bäcker, M.,
Langhorst, J., Schwickert, M., and Dobos, G.J.
(2005). Mediterranean diet or extended fasting’s
influence on changing the intestinal microflora,
immunoglobulin A secretion and clinical outcome
in patients with rheumatoid arthritis and fibromy-
algia:  an observational study. BMC Complement
Altern Med 5, 22.
Mobbs, C.V., Mastaitis, J.W., Isoda, F., and Poplawski,
M. (2013). Treatment of diabetes and diabetic
complications with a ketogenic diet. J Child
Neurol 28, 1009–1014.
Monfoulet, L.E., Philippe, C., Mercier, S., Coxam, V.,
and Wittrant, Y. (2015). Deficiency of G‐ protein
coupled receptor 40, a lipid‐activated receptor,
heightens in vitro‐ and in vivo‐induced murine
osteoarthritis. Exp Biol Med (Maywood) 240,
854–866.
Moore, R.A., Chi, C.C., Wiffen, P.J., Derry, S., and Rice,
A.S. (2015). Oral nonsteroidal anti‐inflammatory
205
drugs for neuropathic pain. Cochrane Database
Syst Rev 10, CD010902.
Morgenweck, J., Abdel‐aleem, O.S., McNamara, K.C.,
Donahue, R.R., Badr, M.Z., and Taylor, B.K.
(2010). Activation of peroxisome proliferator‐
activated receptor γ in brain inhibits inflamma-
tory pain, dorsal horn expression of Fos, and local
edema. Neuropharmacology 58, 337–345.
Morgenweck, J., Griggs, R.B., Donahue, R.R., Zadina,
J.E., and Taylor, B.K. (2013). PPARγ activation blocks
development and reduces established neuropathic
pain in rats. Neuropharmacology 70, 236–246.
Muller‐Schwarze, A.B., Tandon, P., Liu, Z., Lang,
Y., Holmes, G.L., and Stafstrom, C.E. (1999).
Ketogenic diet reduces spontaneous seizures and
mossy fiber sprouting in the kainic acid model.
NeuroReport 10, 1517–1522.
Nakamoto, K., Nishinaka, T., Sato, N., Mankura, M.,
Koyama, Y., Kasuya, F., and Tokuyama, S. (2013).
Hypothalamic GPR40 signaling activated by free
long chain fatty acids suppresses CFA‐ induced
inflammatory chronic pain. PLoS One 8, e81563.
Nakamoto, K., Nishinaka, T., Sato, N., Aizawa, F.,
Yamashita, T., Mankura, M., Koyama, Y., Kasuya,
F., and Tokuyama, S. (2015). The activation of
supraspinal GPR40/FFA1 receptor signalling
regulates the descending pain control system. Br J
Pharmacol 172, 1250–1262.
Neal, E.G., Chaffe, H., Schwartz, R.H., Lawson, M.S.,
Edwards, N., Fitzsimmons, G., Whitney, A., and
Cross, J.H. (2008). The ketogenic diet for the treat-
ment of childhood epilepsy:  a randomised con-
trolled trial. Lancet Neurol 7, 500–506.
Paoli, A., Moro, T., Bosco, G., Bianco, A., Grimaldi,
K.A., Camporesi, E., and Mangar, D. (2015).
Effects of n‐3 polyunsaturated fatty acids (ω‐3)
supplementation on some cardiovascular risk
factors with a ketogenic Mediterranean diet. Mar
Drugs 13, 996–1009.
Pérez‐Guisado, J., and Muñoz‐Serrano, A. (2011). The
effect of the Spanish Ketogenic Mediterranean
Diet on nonalcoholic fatty liver disease:  a pilot
study. J Med Food 14, 677–680.
Pérez, J., Ware, M.A., Chevalier, S., Gougeon, R., and
Shir, Y. (2005). Dietary omega‐3 fatty acids may
be associated with increased neuropathic pain in
nerve‐injured rats. Anesth Analg 101, 444–448,
table of contents.
Phillips, C.J. (2006). Economic burden of chronic
pain. Expert Rev Pharmacoecon Outcomes Res 6,
591–601.
Poon, A., and Sawynok, J. (1999). Antinociceptive
and anti‐inflammatory properties of an adenos-
ine kinase inhibitor and an adenosine deaminase
inhibitor. Eur J Pharmacol 1999, 123–138.
Poon, H.F., Shepherd, H.M., Reed, T.T., Calabrese,
V., Stella, A.M., Pennisi, G., Cai, J., Pierce, W.M.,
Klein, J.B., and Butterfield, D.A. (2006). Proteomics
analysis provides insight into caloric restriction
206
206 section III: Ketogenic Diet in the Laboratory
mediated oxidation and expression of brain pro-
teins associated with age‐related impaired cellular
processes:  Mitochondrial dysfunction, glutamate
dysregulation and impaired protein synthesis.
Neurobiol Aging 27, 1020–1034.
Raffo, E., Francois, J., Ferrandon, A., Koning, E., and
Nehlig, A. (2008). Calorie‐restricted ketogenic
diet increases thresholds to all patterns of pentyl-
enetetrazole‐induced seizures: importance of elec-
troclinical assessment. Epilepsia 49, 320–328.
Ramsden, C.E., Faurot, K.R., Zamora, D., Suchindran,
C.M., Macintosh, B.A., Gaylord, S., Ringel, A.,
Hibbeln, J.R., Feldstein, A.E., Mori, T.A., et  al.
(2013). Targeted alteration of dietary n‐3 and
n‐6 fatty acids for the treatment of chronic head-
aches: a randomized trial. Pain 154, 2441–2451.
Ramsden, C.E., Faurot, K.R., Zamora, D., Palsson,
O.S., MacIntosh, B.A., Gaylord, S., Taha, A.Y.,
Rapoport, S.I., Hibbeln, J.R., Davis, J.M., et  al.
(2015). Targeted alterations in dietary n‐3 and n‐6
fatty acids improve life functioning and reduce
psychological distress among patients with chronic
headache:  a secondary analysis of a randomized
trial. Pain 156, 587–596.
Redivo, D.D., Schreiber, A.K., Adami, E.R., Ribeiro,
D.E., Joca, S.R., Zanoveli, J.M., and Cunha, J.M.
(2015). Effect of omega‐3 polyunsaturated fatty
acid treatment over mechanical allodynia and
depressive‐like behavior associated with experi-
mental diabetes. Behav Brain Res 298, 57–64.
Roe, C.R., Yang, B.Z., Brunengraber, H., Roe, D.S.,
Wallace, M., and Garritson, B.K. (2008). Carnitine
palmitoyltransferase II deficiency:  successful
anaplerotic diet therapy. Neurology 71, 260–264.
Roffey, D.M., Ashdown, L.C., Dornan, H.D., Creech,
M.J., Dagenais, S., Dent, R.M., and Wai, E.K.
(2011). Pilot evaluation of a multidisciplinary,
medically supervised, nonsurgical weight loss
program on the severity of low back pain in obese
adults. Spine J 11, 197–204.
Ross‐Huot, M.‐C., Laferrière, A., Gi, C.M., Khorashadi,
M., Schrincker, T., and Coderre, T.J. (2011). Effects
of glycemic regulation on chronic postischemia
pain. Anesthesiology 115, 614–625.
Rothschild, J., Hoddy, K.K., Jambazian, P., and Varady,
K.A. (2014). Time‐restricted feeding and risk of
metabolic disease: a review of human and animal
studies. Nutr Rev 72, 308–318.
Ruskin, D.N., Kawamura, M., Jr., and Masino, S.A.
(2009). Reduced pain and inflammation in juve-
nile and adult rats fed a ketogenic diet. PLoS One
4, e8349.
Ruskin, D.N., Suter, T.A.C.S., Ross, J.L., and Masino,
S.A. (2013). Ketogenic diets and thermal pain: dis-
sociation of hypoalgesia, elevated ketones, and
lowered glucose in rats. J Pain 14, 467–474.
Sakamoto, K., and Grunewald, K.K. (1987). Beneficial
effects of exercise on growth of rats during inter-
mittent fasting. J Nutr 117, 390–395.
Sawynok, J. (2015). Adenosine receptor targets for
pain. Neuroscience. in press.
Schmidt, S., Stange, R., Lischka, E., Kiehntopf, M.,
Deufel, T., Loth, D., and Uhlemann, C. (2010).
[Uncontrolled clinical study of the efficacy of
ambulant fasting in patients with osteoarthritis].
Forsch Komplementmed 17, 87–94.
Sköldstam, L., Larsson, L., and Lindström, F.D.
(1979). Effect of fasting and lactovegetarian diet
on rheumatoid arthritis. Scand J Rheumatol 8,
249–255.
Smith, A.G., Russell, J., Feldman, E.L., Goldstein, J.,
Peltier, A., Smith, S., Hamwi, J., Pollari, D., Bixby,
B., Howard, J., et al. (2006). Lifestyle intervention
for pre‐diabetic neuropathy. Diabetes Care 29,
1294–1299.
Street, S.E., Walsh, P.L., Sowa, N.A., Taylor‐Blake,
B., Guillot, T.S., Vihko, P., Wightman, R.M., and
Zylka, M.J. (2011). PAP and NT5E inhibit noci-
ceptive neurotransmission by rapidly hydrolyzing
nucleotides to adenosine. Mol Pain 7, 80.
Su, S.W., Sogawa, M.R.C.Y., Silveira, D.C., Holmes,
G.L., and Stafstrom, C.E. (2000). Timing of keto-
genic diet initiation in an experimental epilepsy
model. Dev Brain Res 125, 131–138.
Sullivan, P.G., Rippy, N.A., Dorenbos, K., Concepcion,
R.C., Agarwal, A.K., and Rho, J.M. (2004). The
ketogenic diet increases mitochondrial uncou-
pling protein levels and activity. Ann Neurol 55,
576–580.
Taggart, A.K.P., Kero, J., Gan, X., Cai, T.‐Q., Cheng,
K., Ippolito, M., Ren, N., Kaplan, R., Wu, K.,
Wu, T.‐J., et  al. (2005). (D)‐β‐hydroxybutyrate
inhibits adipocyte lipolysis via the nicotinic
acid receptor PUMA‐G. J Biol Chem 280,
26649–26652.
Taha, A.Y., Filo, E., Ma, D.W.L., and Burnham, W.M.
(2009). Dose‐dependent anticonvulsant effects
of linoleic and α‐linolenic polyunsaturated fatty
acids on pentylenetetrazol induced seizures in rats.
Epilepsia 50, 72–82.
Taha, A.Y., Ryan, M.A.A., and Cunnane, S.C. (2005).
Despite transient ketosis, the classic high‐fat
ketogenic diet induces marked changes in fatty
acid metabolism in rats. Metab Clin Exp 54,
1127–1132.
Tajes, M., Gutierrez‐Cuesta, J., Folch, J.,
Ortuño‐Sahagun, D., Verdaguer, E., Jiménez, A.,
Junyent, F., Lau, A., Camins, A., and Pallàs, M.
(2010). Neuroprotective role of intermittent fast-
ing in senescence‐accelerated mice P8 (SAMP8).
Exp Gerontol 45, 702–710.
Tanner, G.R., Lutas, A., Martínez‐François, J.R., and
Yellen, G. (2011). Single KATP channel opening in
response to action potential firing in mouse den-
tate granule neurons. J Neurosci 31, 8689–8696.
Tendler, D., Lin, S., Yancy, W.S., Jr., Mavropoulos, J.,
Sylvestre, P., Rockey, D.C., and Westman, E.C.
(2007). The effect of a low‐carbohydrate, ketogenic
207
Chapter 22: Metabolic Therapy and Pain
diet on nonalcoholic fatty liver disease:  a pilot
study. Dig Dis Sci 52, 589–593.
Tieu, K., Perier, C., Caspersen, C., Teismann, P.,
Wu, D.‐C., Yan, S.‐D., Niani, A., Vila, M.,
Jackson‐Lewis, V., Ramasamy, R., et  al. (2003).
D‐β‐hydroxybutyrate rescues mitochondrial res-
piration and mitigates features of Parkinson dis-
ease. J Clin Invest 112, 892–901.
Tigerholm, J., Börjesson, S.I., Lundberg, L., Elinder,
F., and Fransén, E. (2012). Dampening of hyper-
excitability in CA1 pyramidal neurons by polyun-
saturated fatty acids acting on voltage‐ gated ion
channels. PLoS One 7, e44388.
Tikoo, K., Tripathi, D.N., Kabra, D.G., Sharma, V., and
Gaikwad, A.B. (2007). Intermittent fasting pre-
vents the progression of type I diabetic nephropa-
thy in rats and changes the expression of Sir2 and
p53. FEBS Lett 581, 1071–1078.
Tillu, D.V., Melemedjian, O.K., Asiedu, M.N., Qu, N.,
De Felice, M., Dussor, G., and Price, T.J. (2012).
Resveratrol engages AMPK to attenuate ERK and
mTOR signaling in sensory neurons and inhib-
its incision‐induced acute and chronic pain. Mol
Pain 8, 5.
Todorova, M.T., Tandon, P., Madore, R.A., Stafstrom,
C.E., and Seyfried, T.N. (2000). The ketogenic diet
inhibits epileptogenesis in EL mice: a genetic model
for idiopathic epilepsy. Epilepsia 41, 933–940.
Traba, J., Kwarteng‐Siaw, M., Okoli, T.C., Li, J.,
Huffstutler, R.D., Bray, A., Waclawiw, M.A., Han,
K., Pelletier, M., Sauve, A.A., et al. (2015). Fasting
and refeeding differentially regulate NLRP3
inflammasome activation in human subjects. J
Clin Invest 125, 4592–4600.
Udén, A.M., Trang, L., Venizelos, N., and Palmblad, J.
(1983). Neutrophil functions and clinical perfor-
mance after total fasting in patients with rheuma-
toid arthritis. Ann Rheum Dis 42, 45–51.
Valério, D.A., Ferreira, F.I., Cunha, T.M., Alves‐Filho,
J.C., Lima, F.O., De Oliveira, J.R., Ferreira, S.H.,
Cunha, F.Q., Queiroz, R.H., and Verri, W.A., Jr.
(2009). Fructose‐1,6‐bisphosphate reduces inflam-
matory pain‐like behaviour in mice:  role of ade-
nosine acting on A1 receptors. Br J Pharmacol 158,
558–568.
Varady, K.A., Hoddy, K.K., Kroeger, C.M.,
Trepanowski, J.F., Klempel, M.C., Barnosky, A.,
and Bhutani, S. (2015). Determinants of weight
loss success with alternate day fasting. Obes Res
Clin Pract.
Veech, R.L. (2004). The therapeutic implications of
ketone bodies:  the effects of ketone bodies in
pathological conditions:  ketosis, ketogenic diet,
redox states, insulin resistance, and mitochondrial
metabolism. Prostaglandins Leukot Essent Fatty
Acids 70, 309–319.
Viggiano, A., Pilla, R., Arnold, P., Monda, M.,
D’Agostino, D., and Coppola, G. (2015).
Anticonvulsant properties of an oral ketone ester
207
in a pentylenetetrazole‐model of seizure. Brain
Res 1618, 50–54.
Voskuyl, R.A., Vreugdenhil, M., Kang, J.X., and Leaf,
A. (1998). Anticonvulsant effect of polyunsatu-
rated fatty acid in rats, using the cortical stimula-
tion model. Eur J Pharmacol 341, 145–152.
Wagner, K., Vito, S., Inceoglu, B., and Hammock,
B.D. (2014). The role of long chain fatty acids
and their epoxide metabolites in nociceptive
signaling. Prostaglandins Other Lipid Mediat
113–115, 2–12.
Walker, I., and Said, R.R. (2015). Predictors of ketogenic
diet efficacy in children based on the electroenceph-
alogram (EEG). J Child Neurol 30, 1270–1274.
Welshman, A. (2005). Palliative care. Some organi-
zational considerations. Minerva Anestesiol 71,
439–443.
White, D.K., Neogi, T., Rejeski, W.J., Walkup, M.P.,
Lewis, C.E., Nevitt, M.C., Foy, C.G., Felson, D.T.,
and Group, L.A.R. (2015). Can an intensive diet
and exercise program prevent knee pain among
overweight adults at high risk? Arthritis Care Res
(Hoboken) 67, 965–971.
Wideman, C.H., Murphy, H.M., and McCartney, S.B.
(1996). Interactions between vasopressin and food
restriction on stress‐induced analgesia. Peptides
17, 63–66.
Wilder, R.M. (1921a). The effects of ketonemia on the
course of epilepsy. Mayo Clin Bull 2, 307.
Wilder, R.M. (1921b). High fat diets in epilepsy. Mayo
Clin Bull 2, 308.
Willis, S., Samala, R., Rosenberger, T.A., and Borges,
K. (2009). Eicosapentaenoic and docosahexae-
noic acids are not anticonvulsant or neuroprotec-
tive in acute mouse seizure models. Epilepsia 50,
138–142.
Won, Y.‐J., Lu, V.B., Puhl, H.L., III, and Ikeda, S.R.
(2013). β‐hydroxybutyrate modulates N‐ type
calcium channels in rat sympathetic neurons by
acting as an agonist for the G‐protein‐ coupled
receptor FFA3. J Neurosci 33, 19314–19325.
Xu, X.P., Sun, R.P., and Jin, R.F. (2006). [Effect of keto-
genic diet on hippocampus synaptic reorganiza-
tion and GluR5 expression in kainic acid induced
rat model of epilepsy]. Zhonghua Er Ke Za Zhi 44,
100–104.
Xu, X.‐P., Erichsen, D., Börjesson, S.I., Dahlin, M.,
Åmark, P., and Elinder, F. (2008). Polyunsaturated
fatty acids and cerebrospinal fluid from children
on the ketogenic diet open a voltage‐gated K chan-
nel:  a putative mechanism of antiseizure action.
Epilepsy Res 80, 57–66.
Yamaoka, G., Horiuchi, H., Morino, T., Miura, H.,
and Ogata, T. (2013). Different analgesic effects of
adenosine between postoperative and neuropathic
pain. J Orthop Sci 18, 130–136.
Yancy, W.S., Jr., Almirall, D., Maciejewski, M.L.,
Kolotkin, R.L., McDuffie, J.R., and Westman,
E.C. (2009). Effects of two weight‐loss diets on
208
208 section III: Ketogenic Diet in the Laboratory
health‐related quality of life. Qual Life Res 18,
281–289.
Yang, L., Zhao, J., Milutinovic, P.S., Brosnan,
R.J., Eger, E.I., II, and Sonner, J.M. (2007).
Anesthetic properties of the ketogenic bod-
ies β‐hydroxybutyric acid and acetone. Anesth
Analg 105, 673–679.
Yehuda, S., Carasso, R.L., and Mostofsky, D.I. (1994).
Essential fatty acid preparation (SR‐3) raises the
seizure threshold in rats. Eur J Pharmacol 254,
193–198.
Yorek, M.S., Obrosov, A., Shevalye, H., Lupachyk, S.,
Harper, M.M., Kardon, R.H., and Yorek, M.A.
(2014). Effect of glycemic control on corneal
nerves and peripheral neuropathy in streptozoto-
cin‐induced diabetic C57Bl/6J mice. J Peripher
Nerv Syst 19, 205–217.
Youm, Y.H., Nguyen, K.Y., Grant, R.W., Goldberg, E.L.,
Bodogai, M., Kim, D., D’Agostino, D., Planavsky,
N., Lupfer, C., Kanneganti, T.D., et  al. (2015).
The ketone metabolite β‐ hydroxybutyrate blocks
NLRP3 inflammasome‐mediated inflammatory
disease. Nature Med 21, 263–269.
Żarnowsky, T., Chorągiewicz, T., Tulidowicz‐Bielak,
M., Thaler, S., Rejdak, R., Żarnowska, I., Turski,
W.A., and Gasior, M. (2012). Ketogenic diet
increases concentrations of kynurenic acid in dis-
crete brain structures of young and adult rats. J
Neural Transm 119, 679–684.
Zhao, Y.T., Tekkök, S., and Krnjević, K. (1997).
2‐Deoxy‐D‐glucose‐induced changes in mem-
brane potential, input resistance, and excitatory
postsynaptic potentials of CA1 hippocampal neu-
rons. Can J Physiol Pharmacol 75, 368–374.
209
23
Ketogenic Diet, Adenosine, Epigenetics,
and Antiepileptogenesis
T H E R E S A A . L U S A R D I , P H D A N D D E T L E V B O I S O N, P H D
INTRODUCTION
Fasting has been known as a means to control
epilepsy for thousands of years. Formal studies of
fasting in the early years of the 20th century led
to the critical observation of elevated ketone bod-
ies excreted during fasting, and diets tailored to
elevate ketones were established (Wheless, 2008;
Peterman, 1925, 1924). The difficulty of compli-
ance in combination with the availability of anti-
epileptic drugs diminished enthusiasm for the KD
throughout much of the 20th century, but more
recent studies in patients refractory to conven-
tional drugs have demonstrated the KD’s clear
efficacy for seizure suppression. While the earli-
est studies suggested efficacy of the KD only in
children, several studies show similar therapeu-
tic effects in adults (Barborka, 1930; Sirven et al.,
1999; Coppola et al., 2002).
The precise mechanisms by which the KD exerts
antiepileptic effects remain unknown. Beneficial
effects of the KD can manifest clinically as quickly
as several days or as long as 8  months after diet
initiation. Diet reversal yields even more perplex-
ing outcomes. In some individuals, seizures return
within hours of glucose reintroduction, suggesting
an antiseizure mechanism. Others are able to return
to a “normal” diet and remain seizure-free, suggest-
ing an antiepileptogenic mechanism. A  detailed
review of the proposed mechanisms of the KD has
been published recently (Masino and Rho, 2012),
illustrating that the influences of the diet on meta-
bolic processes are broad. Indeed, this broad range
of metabolic influences suggests that the therapeu-
tic benefits of the KD are not limited to epilepsy
but might extend into other neurological disorders
(Stafstrom & Rho, 2012; Kossoff & Hartman, 2012).
The dramatic reduction or elimination of
seizures in up to 40% of persons with epilepsy
on a KD is a strong argument for continued use
of the diet. This is particularly compelling when
considering that the individuals in controlled
trials of the diet are refractory to conventional
antiepileptic drugs and have been suffering from
epilepsy for many years at the time the diet is initi-
ated. A more comprehensive understanding of the
underlying mechanisms could yield evidence for
predictive biomarkers that can identify patients
who are most likely to respond well to the KD.
A D E N O S I N E H O M E O S TA S I S
AND SEIZURES
The purine ribonucleoside adenosine is a ubiqui-
tous homeostatic regulator of brain activity acting
at four G-protein-coupled receptors. Receptors
are classified by their second messenger activity
(Gi/o:  A1R and A3R; or Gs:  A2AR and A2BR) and
their affinity for adenosine (nM:  A1R, A2AR;
μM: A2BR, A3R). In the brain, basal adenosine lev-
els are typically in the nM range, regulated largely
through a balance between ATP breakdown by
ectonucleotideases (CD39, CD73) (Fredholm
et al., 2005; Cunha, 2001) and metabolic clearance
by astrocytic adenosine kinase (ADK) (Boison,
2013; Lloyd and Fredholm, 1995). In the epileptic
brain, this balance is disrupted by overexpression
of ADK in astrogliotic cells (Fedele et  al., 2005),
which results in increased clearance of adenos-
ine and loss of A1R-mediated synaptic inhibition
(Boison, 2008).
The role of adenosine in seizure termination has
been demonstrated directly (During and Spencer,
1992; Van Gompel et  al., 2014). Astrogliosis and
ADK up-regulation is well documented in the
epileptic brain (Sofroniew and Vinters, 2010;
Aronica et  al., 2011), and strategies to restore
adenosine signaling have been quite successful
for seizure suppression. Pharmacologic activation
of the A1R blocks seizures (Gouder et  al., 2003),
as does inhibition of ADK with 5-iodotubercidin
(ITU) (Gouder et al., 2004). Transgenic mice have
210
210 section III: Ketogenic Diet in the Laboratory
provided further evidence of the importance of
the A1R in seizure suppression and a causal role
for ADK overexpression in seizure susceptibility.
A1R knockout mice are susceptible to lethal status
epilepticus (Fedele et  al., 2006; Kochanek et  al.,
2006). Overexpression of ADK in the hippocam-
pus is associated with spontaneous seizures, which
can be blocked by ITU (Fedele et  al., 2005), and
ADK suppression in the hippocampus confers
protection against epileptogenesis (Li et al., 2007).
The mechanistic rationale and therapeutic efficacy
of adenosine homeostasis in seizure suppression is
well established; however, the ubiquitous nature of
adenosine signaling makes direct clinical transla-
tion of these therapies impractical. Chronic sys-
temic treatment has sedative, cardiovascular, and
renal side effects that have led to the near aban-
donment of pharmaceutical drug development
efforts. To avoid limitations of systemic adenosine
augmentation, focal delivery strategies have been
developed. It was shown that focal adenosine sup-
plementation provided similar efficacy to systemic
treatment. Cells engineered to release adenosine
inhibited kindled seizures in an A1R dependent
manner (Huber et al., 2001) and blocked seizures
in a model of drug-resistant epilepsy (Li et  al.,
2008). Similarly, a silk-based scaffold for con-
trolled focal adenosine release suppressed kindled
seizures and epileptogenesis in a kindling model
(Wilz et al., 2008; Szybala et al., 2009). Of particu-
lar interest, in the clinically relevant kainic acid
status epilepticus (KASE) model of epilepsy, silk-
based transient adenosine release halted epilepsy
progression in epileptic rats with fully generalized
spontaneous seizures, with lack of seizure progres-
sion lasting at least 3 months after the conclusion
of the adenosine supplementation, whereas seizure
rate and severity continued to progress in all con-
trol animals (Williams-Karnesky et al., 2013). This
finding for the first time suggested a potent antiep-
ileptogenic role of adenosine. Maladaptive changes
in adenosine metabolism are tightly linked to epi-
leptogenesis, and restoration of adenosine homeo-
stasis therefore presents a potentially powerful
means to prevent epilepsy or its progression.
The influence of adenosine on biologic systems
is complex. It may be pro- or anti-inflammatory
(Hasko et  al., 2005)  and can inhibit or activate
synaptic activity (Sebastiao and Ribeiro, 2009).
Adenosine concentration is under strict local con-
trol, with both intra- and extracellular enzymes
contributing to the maintenance of adenosine
“tone.” While our understanding of the local
mix of adenosine receptors and their subcellu-
lar localization explains some of the apparently
contradictory roles proposed for adenosine, a
broader view of adenosine as a “retaliatory metab-
olite” has been proposed (Newby, 1984) to explain
the emerging role of adenosine in response to
physiologic stress. This concept was applied to the
KD, postulating a role for adenosine as the molec-
ular link between cellular metabolism and seizure
suppression (Masino et  al., 2009). Indeed, subse-
quent studies have shown that administration of
the KD to epileptic rats in the pilocarpine model
restores adenosine levels in the hippocampus to
normal (nonepileptic) levels, and that returning to
a normal-fat/carbohydrate diet results in a return
to adenosine deficiency (Lusardi et  al., 2015).
A  causal role for adenosine in the antiseizure
effects of the KD has been demonstrated in A1R
knockout mice and ADK overexpressing mice,
both of which have spontaneous focal seizures.
Spontaneous seizures were eliminated in the ADK
overexpressing mice fed with the KD, an effect
that was blocked by the A1R antagonist dipropyl-
cyclopentylxanthine. No reduction in spontane-
ous seizures was observed in A1R−/− mice on the
KD, though a partial reduction was measured in
A1R+/− mice, consistent with reduced A1R expres-
sion levels in the heterozygotic mice (Masino et al.,
2011). These results support the hypothesis that
the antiseizure efficacy of the KD exerts some of its
influence through restoration of local adenosine
homeostasis in the epileptic brain.
EPIGENETICS IN EPILEPSY
Heritable factors confer a small but appre-
ciable increase in risk of epilepsy development
(Kullmann, 2002). Genetic loci have been iden-
tified in subpopulations of particular epilepsy
syndromes, but only account for a fraction of the
incidence, and are not well correlated with seizure
severity and comorbidities (Myers and Mefford,
2015). In general, large-scale genomewide stud-
ies have not identified broadly applicable genomic
variations that would indicate a common genetic
risk in epilepsy (Kasperaviciute et al., 2010; Mefford
et  al., 2010; Cavalleri et  al., 2007). However, a
recent study suggests that genetic variants of the
Adk gene are associated with an increased risk
for the development of posttraumatic epilepsy
(Diamond et  al., 2015). While the role of genet-
ics in epilepsy and epileptogenesis needs further
investigation, changes in gene expression and
regulation might play an additional role. Studies
of gene expression changes following seizures, in
the latent phase of epileptogenesis before spon-
taneous seizures develop, and in the epileptic
brain agree that many genes are (dys)regulated in
211
Chapter 23: Ketogenic Diet, Adenosine, Epigenetics, and Antiepileptogenesis
the epileptogenic brain (Lukasiuk and Pitkanen,
2004). However, the specific genes regulated are
not consistent across studies, regardless of species
or model specificity (Aronica and Gorter, 2007).
Pathway analyses across multiple epilepsy studies
have revealed that despite variations in the specific
genes regulated, certain functional pathways are
commonly identified across studies (Aronica and
Gorter, 2007). Thus, expression changes targeting
intra- and extracellular signaling, transcription
and protein biosynthesis, and immune responses
are all well represented in all time windows evalu-
ated (Aronica and Gorter, 2007). This network
view of epilepsy risk and development presents a
model that can account for the confounding fac-
tors in genomic studies of epilepsy, including phe-
notypic inconsistencies associated with heritable
risk factors, and the variable susceptibility and dis-
ease progression in acquired epilepsies.
Epigenetic changes are biochemical altera-
tions of the DNA or the chromatin structure,
which do not affect the coding sequence of the
DNA, but contribute to the regulation of gene
transcription and entire networks. Epigenetic
changes include DNA methylation and hydroxy-
methylation, as well as histone methylation and
acetylation (Jaenisch and Bird, 2003; Kiefer, 2007).
Importantly, epigenetic changes may affect several
genes thought to represent a risk factor for epi-
lepsy simultaneously. In contrast to genetic muta-
tions, epigenetic changes are potentially reversible
and may constitute a novel target for therapeutic
intervention. DNA methylation was once believed
to be stable, but a critical role for the dynamic
regulation of DNA methylation has been identi-
fied in learning and memory (Roth and Sweatt,
2009; Day and Sweatt, 2010). Environmental dis-
ruption of physiologic epigenetic regulation has
been implicated in a broad range of diseases, and
has been widely studied in the context of exposure
to environmental toxins (Pacchierotti and Spano,
2015; Bollati and Baccarelli, 2010). In rodent stud-
ies environmental factors such as maternal stimu-
lation and social interaction have been linked to
epigenetic changes, which in turn influence stress
hormones, brain development, and neuropsychi-
atric phenotypes (Curley et al., 2011). Clinical evi-
dence supports the conclusions from those rodent
studies (McGowan et al., 2009; Zhang et al., 2013).
Beyond exposure to overt toxins, diet can influ-
ence the epigenome. Naturally occurring com-
pounds in our food have a known influence on
epigenetic regulation. Their role has been exten-
sively studied in cancer risk and treatment, but
their general mechanisms are likely to influence
211
other disease processes as well (Lim and Song,
2012; Hardy and Tollefsbol, 2011). Food-derived
bioactive compounds can act as substrate donors
or reaction inhibitors (Choi and Friso, 2010), and
dietary imbalances can promote or limit disease
progression. Thus, epigenetic regulation of gene
expression not only drives the development of our
gross anatomy but also shapes our intellectual,
emotional, and physiologic phenotype.
K E T O G E N I C D I E T,
A D E N O S I N E H O M E O S TA S I S ,
A N D A N T I E P I L E P TO G E N E S I S
IN TEMPORAL LOBE
EPILEPSY
The most common form of focal epilepsy,
with about 80% of seizures originating in the
hippocampus—a critical structure for consolida-
tion, storage, and retrieval of memories—temporal
lobe epilepsy (TLE) is refractory to conventional
antiepilepsy drugs in up to 40% of patients, leav-
ing removal of the hippocampus as the antiseizure
treatment of last resort for many. The growing field
of epigenetic research has led to the discovery that
significant epigenetic changes occur in the epilep-
tic hippocampus and may present a unique thera-
peutic opportunity.
In both clinical and experimental settings the
epileptic hippocampus was shown to be character-
ized by altered DNA methylation when compared
with nonepileptic hippocampus. In human TLE
samples from resected hippocampus, gene targets
with both increased and decreased methylation
were identified, with approximately 80% of signifi-
cantly regulated sites hypermethylated, an equally
prominent signature in sclerotic and nonsclerotic
hippocampal tissue (Miller-Delaney et  al., 2015).
In rodent models of TLE, similar patterns of hyper-
methylation in the epileptic hippocampus have
been demonstrated in the KASE model (Williams-
Karnesky et al., 2013; Ryley Parrish et al., 2013) and
pilocarpine status epilepticus models (Kobow et al.,
2013; Lusardi et al., 2015). A receptor-independent
role for adenosine in regulating DNA methyla-
tion has recently been demonstrated (Williams-
Karnesky et al., 2013). Adenosine has a mass effect
on biochemical enzyme reactions and is an obliga-
tory end product of the S-adenosylmethionine
(SAM) dependent transmethylation pathway
(Figure 23.1), necessary for the transfer of methyl
groups onto DNA (Boison et al., 2002; Mato et al.,
2008). As predicted by this biochemical pathway,
exogenous application of either adenosine or its
complementary end product homocysteine inhib-
its the reaction and reduces DNA methylation,
212

212 section III: Ketogenic Diet in the Laboratory

Methionine
SAM adenosyltransferase

X-CH3
DNMT
Met

SAH
Methionine
SAH synthase
hydrolase

ADO
HCY
ADK

AMP

FIGURE 23.1 Transmethylation pathway. The amino acid methionine (Met) is converted to S-adenosylmethionine
(SAM), a methyl group donor, by methionine adenosyl transferase. DNA methyltransferase (DNMT) enzymes catalyze
the transfer of a methyl group from SAM to DNA, with S-adenosyl homocysteine (SAH) as a rate-limiting end product.
The enzyme SAH hydrolase catalyzes SAH to adenosine (ADO) and homocysteine (HCY) in a reversible reaction.
Thus, ADO and HCY homeostasis can regulate DNMT activity and DNA methylation levels. ADO is metabolized
by adenosine kinase (ADK) into adenosine monophosphate (AMP), and HCY is recycled back to Met by methionine
synthase. Methionine synthase and methionine adenosyltransferase activity depends on the availability of bioactive
compounds such as Vitamin B12.

whereas addition of the methyl group donor SAM
increases DNA methylation in the naïve rodent
brain (Williams-Karnesky et  al., 2013). Similar
effects can be obtained pharmacologically using
an ADK inhibitor (5-iodotubercidin, ITU), which
reduced hippocampal DNA methylation by 50%,
even in the presence of caffeine, a nonspecific
adenosine receptor inhibitor, further demonstrat-
ing that the effects of adenosine on methylation are
receptor independent. The ADK is up-regulated
within epileptogenic brain areas in rodent models
of TLE and in human TLE with hippocampal scle-
rosis (Boison, 2012), resulting in adenosine defi-
ciency, increasing the flux of methyl groups through
the transmethylation pathway. Concurrently,
adenosine deficiency shifts the equilibrium of the
S-adenosylhomocysteine (SAH) hydrolase reac-
tion away from the formation of SAH (Mandaviya
et  al., 2014), which is also known to block DNA
methyltransferase activity by product inhibi-
tion (James et  al., 2002), thereby reinforcing the
increase in DNA methylation in the epileptic
brain. Seizures resulting from the proconvulsant
L-methionine-dl-sulfoximine, which increases
the methylation flux by increasing the SAM/
SAH ratio, can be blocked by adenosine and
homocysteine (Gill and Schatz, 1985; Schatz et al.,
1983; Sellinger et  al., 1984). In the KASE model,
direct ventricular administration of adenosine for
10  days significantly reduced epilepsy disease
progression, including the progressive increase of
spontaneous convulsive seizures and additional
mossy fiber sprouting, and restored global DNA
methylation levels to control levels, lasting well
after the conclusion of the adenosine delivery
(Williams-Karnesky et  al., 2013). These findings
show that global DNA methylation levels are under
the direct control of adenosine, and that disrup-
tion of adenosine homeostasis (due to ADK up-
regulation at the epileptogenic focus) affects DNA
methylation levels and alters gene expression in the
epileptic brain.
As described above, through mechanisms
that are not yet well understood, strict adherence
to a KD reduces seizures through an adenosine-
receptor dependent mechanism (Masino et  al.,
2011). Using a kindling model of epileptogenesis,
the KD also delayed the acquisition of kindling in
mice, an effect that persisted even after a return
to a standard lab diet, while the conventional
antiepileptogenic drug valproic acid attenuated
only the seizures without blocking the kindling
213
Chapter 23: Ketogenic Diet, Adenosine, Epigenetics, and Antiepileptogenesis
process. These data demonstrate persistent
effects of the KD that are not merely due to sei-
zure suppression (Lusardi et al., 2015). When fed
to rats following status epilepticus, the KD not
only reduced spontaneous seizure development
but also reduced DNA methylation levels both
during diet administration and after a return to
standard diet (Kobow et al., 2013; Lusardi et al.,
2015). Though a direct link between the KD and
DNA methylation levels must still be demon-
strated, when evaluated together these results
show that the lasting effects of the KD may be
conferred via adenosine regulation of the DNA
methylome, supporting a key mechanism impli-
cated in epilepsy and epileptogenesis.
CONCLUSION
With up to 35% of persons with epilepsy consid-
ered to be refractory to treatment and no therapies
available that prevent epilepsy or its progression,
the novel epigenetic functions of KD therapy dis-
cussed here might be of therapeutic value to not
only relieve the seizure burden in patients with
epilepsy but also to modify the development of epi-
lepsy with its sequelae of pharmacoresistance and
the development of epilepsy-associated comorbid-
ities. The KD presents a powerful adjunct therapy
to existing pharmacologic and surgical approaches
to seizure relief. Renewed interest in the KD has
led to refinements in the diet formulation and
administration (Kossoff et  al., 2009; Wibisono
et  al., 2015)  and improved understanding of the
potential positive and negative interactions with
conventional antiepileptic drugs (Morrison et al.,
2009; van der Louw et al., 2015), improving com-
pliance and seizure suppression rates. However,
the diet requires close monitoring by physicians
and dietitians, and seemingly minor deviations
from the ketogenic regimen can negate its benefi-
cial effects. While a “ketogenic diet in a pill” may
be unlikely, ongoing studies to understand the bio-
chemical mechanisms of the KD are an essential
step in the continued refinement of antiepileptic
and antiepileptogenic therapies. The neuroprotec-
tive mechanisms of the KD are varied, and diet
efficacy may rely on their combined influences.
Among the metabolites regulated by the KD,
however, adenosine has both a direct relevance to
seizure suppression by A1R activation and an indi-
rect influence on epilepsy and epileptogenesis via
regulation of DNA methylation. A clearer under-
standing of how KD therapy affects adenosine
metabolism and its epigenetic sequelae may help
us understand adenosine dysregulation in epi-
lepsy, and may guide the development of therapies
213
designed to directly restore adenosine homeosta-
sis, with the goal of developing a novel class of
antiepileptogenic drugs.
REFERENCES
Aronica, E., and Gorter, J.A. (2007). Gene expression
profile in temporal lobe epilepsy. Neuroscientist
13, 100–108.
Aronica, E., Zurolo, E., Iyer, A., et  al. (2011).
Upregulation of adenosine kinase in astrocytes in
experimental and human temporal lobe epilepsy.
Epilepsia 52, 1645–1655.
Barborka, C. (1930). Epilepsy in adults. Arch Neurol
Psychiatry 23, 904–914.
Boison, D. (2008). The adenosine kinase hypothesis of
epileptogenesis. Prog Neurobiol 84, 249–262.
Boison, D. (2012). Adenosine dysfunction in epilepsy.
Glia 60, 1234–1243.
Boison, D. (2013). Adenosine kinase: exploitation for
therapeutic gain. Pharmacol Rev 65, 906–943.
Boison, D., Scheurer, L., Zumsteg, V., Rulicke, T.,
Litynski, P., Fowler, B., Brandner, S., and Mohler,
H. (2002). Neonatal hepatic steatosis by disruption
of the adenosine kinase gene. Proc Natl Acad Sci
USA 99, 6985–6990.
Bollati, V., and Baccarelli, A. (2010). Environmental
epigenetics. Heredity 105, 105–112.
Cavalleri, G.L., Weale, M.E., Shianna, K.V., et al. (2007).
Multicentre search for genetic susceptibility loci in
sporadic epilepsy syndrome and seizure types:  a
case-control study. Lancet Neurol 6, 970–980.
Choi, S.W., and Friso, S. (2010). Epigenetics:  a new
bridge between nutrition and health. Adv Nutr
1, 8–16.
Coppola, G., Veggiotti, P., Cusmai, R., et al. (2002). The
ketogenic diet in children, adolescents and young
adults with refractory epilepsy:  an Italian multi-
centric experience. Epilepsy Res 48, 221–227.
Cunha, R.A. (2001). Regulation of the ecto-
nucleotidase pathway in rat hippocampal nerve
terminals. Neurochem Res 26, 979–991.
Curley, J.P., Jensen, C.L., Mashoodh, R., and
Champagne, F.A. (2011). Social influences on
neurobiology and behavior: epigenetic effects dur-
ing development. Psychoneuroendocrinology 36,
352–371.
Day, J.J., and Sweatt, J.D. (2010). DNA methylation and
memory formation. Nat Neurosci 13, 1319–1323.
Diamond, M.L., Ritter, A.C., Jackson, E.K., Conley,
Y.P., Kochanek, P.M., Boison, D., and Wagner, A.K.
(2015). Genetic variation in the adenosine regu-
latory cycle is associated with posttraumatic epi-
lepsy development. Epilepsia 56, 1198–1206.
During, M.J., and Spencer, D.D. (1992). Adenosine: a
potential mediator of seizure arrest and postictal
refractoriness. Ann Neurol 32, 618–624.
Fedele, D.E., Gouder, N., Guttinger, M., Gabernet, L.,
Scheurer, L., Rulicke, T., Crestani, F., and Boison, D.
214
214 section III: Ketogenic Diet in the Laboratory
(2005). Astrogliosis in epilepsy leads to overex-
pression of adenosine kinase, resulting in seizure
aggravation. Brain, 128, 2383–2395.
Fedele, D.E., Li, T., Lan, J.Q., Fredholm, B.B., and
Boison, D. (2006). Adenosine A(1). receptors are
crucial in keeping an epileptic focus localized. Exp
Neurol, 200, 184–190.
Fredholm, B.B., Chen, J.F., Cunha, R.A., Svenningsson,
P., and Vaugeois, J.M. (2005). Adenosine and brain
function. Int Rev Neurobiol 63, 191–270.
Gill, M.W., and Schatz, R.A. (1985). The effect of diaz-
epam on brain levels of S-adenosyl-L-methionine
and S-adenosyl-L-homocysteine: possible correla-
tion with protection from methionine sulfoximine
seizures. Research communications in chemical
pathology and pharmacology 50, 349–363.
Gouder, N., Fritschy, J.M., and Boison, D. (2003).
Seizure suppression by adenosine A1 receptor
activation in a mouse model of pharmacoresistant
epilepsy. Epilepsia 44, 877–885.
Gouder, N., Scheurer, L., Fritschy, J.M., and Boison, D.
(2004). Overexpression of adenosine kinase in epi-
leptic hippocampus contributes to epileptogenesis.
J Neurosci 24, 692–701.
Hardy, T.M., and Tollefsbol, T.O. (2011). Epigenetic
diet:  impact on the epigenome and cancer.
Epigenomics 3, 503–518.
Hasko, G., Pacher, P., Vizi, E.S., and Illes, P. (2005).
Adenosine receptor signaling in the brain immune
system. Trends Pharmacol Sci 26, 511–516.
Huber, A., Padrun, V., Deglon, N., Aebischer, P., Mohler,
H., and Boison, D. (2001). Grafts of adenosine-
releasing cells suppress seizures in kindling epi-
lepsy. Proc Natl Acad Sci USA 98, 7611–7616.
Jaenisch, R., and Bird, A. (2003). Epigenetic regulation
of gene expression:  how the genome integrates
intrinsic and environmental signals. Nat Genet, 33
Suppl, 245–254.
James, S.J., Melnyk, S., Pogribna, M., Pogribny,
I.P., and Caudill, M.A. (2002). Elevation in S-
adenosylhomocysteine and DNA hypometh-
ylation:  potential epigenetic mechanism for
homocysteine-related pathology. J Nutr. 132,
2361S–2366S.
Kasperaviciute, D., Catarino, C.B., Heinzen, E.L., et al.
(2010). Common genetic variation and suscepti-
bility to partial epilepsies: a genome-wide associa-
tion study. Brain 133, 2136–2147.
Kiefer, J.C. (2007). Epigenetics in development. Dev
Dyn 236, 1144–1156.
Kobow, K., Kaspi, A., Harikrishnan, K.N., et al. (2013).
Deep sequencing reveals increased DNA methyla-
tion in chronic rat epilepsy. Acta Neuropathol 126,
741–76.
Kochanek, P.M., Vagni, V.A., Janesko, K.L., et  al.
(2006). Adenosine A1 receptor knockout mice
develop lethal status epilepticus after experimental
traumatic brain injury. J Cereb Blood Flow Metab
26, 565–575.
Kossoff, E.H., and Hartman, A.L. (2012). Ketogenic
diets:  new advances for metabolism-based thera-
pies. Curr Opin Neurol 25, 173–178.
Kossoff, E.H., Zupec-Kania, B.A., Amark, P.E., et  al.
(2009). Optimal clinical management of children
receiving the ketogenic diet:  recommendations
of the International Ketogenic Diet Study Group.
Epilepsia 50, 304–317.
Kullmann, D.M. (2002). Genetics of epilepsy. J Neurol
Neurosurg Psychiatry, 73 (Suppl 2), II32–35.
Li, T., Quan Lan, J., Fredholm, B.B., Simon, R.P., and
Boison, D. (2007). Adenosine dysfunction in
astrogliosis: cause for seizure generation? Neuron
Glia Biol 3, 353–366.
Li, T., Ren, G., Lusardi, T., Wilz, A., Lan, J.Q., Iwasato,
T., Itohara, S., Simon, R.P., and Boison, D. (2008).
Adenosine kinase is a target for the prediction and
prevention of epileptogenesis in mice. J Clin Invest
118, 571–582.
Lim, U., and Song, M.A. (2012). Dietary and lifestyle
factors of DNA methylation. Methods Mol Biol
863, 359–376.
Lloyd, H.G., and Fredholm, B.B. (1995). Involvement
of adenosine deaminase and adenosine kinase in
regulating extracellular adenosine concentration in
rat hippocampal slices. Neurochem Int 26, 387–395.
Lukasiuk, K., and Pitkanen, A. (2004). Large-scale
analysis of gene expression in epilepsy research: is
synthesis already possible? Neurochem Res 29,
1169–1178.
Lusardi, T.A., Akula, K.K., Coffman, S.Q., Ruskin, D.N.,
Masino, S.A., and Boison, D. (2015). Ketogenic
diet prevents epileptogenesis and disease progres-
sion in adult mice and rats. Neuropharmacology
99, 500–509.
Mandaviya, P.R., Stolk, L., and Heil, S.G. (2014).
Homocysteine and DNA methylation: a review of
animal and human literature. Mol Genet Metab
113, 243–252.
Masino, S.A., Kawamura, M., Wasser, C.D., Pomeroy,
L.T., and Ruskin, D.N. (2009). Adenosine, keto-
genic diet and epilepsy: the emerging therapeutic
relationship between metabolism and brain activ-
ity. Curr Neuropharmacol 7, 257–268.
Masino, S.A., Li, T., Theofilas, P., Sandau, U.S., Ruskin,
D.N., Fredholm, B.B., Geiger, J.D., Aronica, E., and
Boison, D. (2011). A ketogenic diet suppresses sei-
zures in mice through adenosine A(1). receptors. J
Clin Invest, 121, 2679–2683.
Masino, S.A., and Rho, J.M. (2012) Mechanisms
of ketogenic diet action. In Jasper’s Basic
Mechanisms of the Epilepsies, J.L. Noebels, M.
Avoli, M.A. Rogawski, R.W. Olsen and A.V.
Delgado-Escueta eds. (Bethesda:  National
Center for Biotechnology Information), pp.
1001–1022.
Mato, J.M., Martinez-Chantar, M.L., and Lu, S.C.
(2008). Methionine metabolism and liver disease.
Annu Rev Nutr 28, 273–293.
215
Chapter 23: Ketogenic Diet, Adenosine, Epigenetics, and Antiepileptogenesis
McGowan, P.O., Sasaki, A., D’Alessio, A.C., Dymov,
S., Labonte, B., Szyf, M., Turecki, G., and Meaney,
M.J. (2009). Epigenetic regulation of the gluco-
corticoid receptor in human brain associates with
childhood abuse. Nat Neurosci 12, 342–348.
Mefford, H.C., Muhle, H., Ostertag, P., et  al. (2010).
Genome-wide copy number variation in epi-
lepsy:  novel susceptibility loci in idiopathic
generalized and focal epilepsies. PLoS Genet 6,
e1000962.
Miller-Delaney, S.F., Bryan, K., Das, S., McKiernan,
R.C., Bray, I.M., Reynolds, J.P., Gwinn, R., Stallings,
R.L., and Henshall, D.C. (2015). Differential DNA
methylation profiles of coding and non-coding
genes define hippocampal sclerosis in human tem-
poral lobe epilepsy. Brain 138, 616–631.
Morrison, P.F., Pyzik, P.L., Hamdy, R., Hartman, A.L.,
and Kossoff, E.H. (2009). The influence of concur-
rent anticonvulsants on the efficacy of the keto-
genic diet. Epilepsia 50, 1999–2001.
Myers, C.T., and Mefford, H.C. (2015). Advancing
epilepsy genetics in the genomic era. Genome
medicine 7, 91.
Newby, A. (1984). Adenosine and the concept of “retal-
iatory metabolites.” Trends Biochem Sci 9, 42–44.
Pacchierotti, F., and Spano, M. (2015). Environmental
impact on DNA methylation in the germline: state
of the art and gaps of knowledge. BioMed Res Int,
2015, 123484.
Peterman, M.G. (1924). The ketogenic diet in the
treatment of epilepsy, a preliminary report. Am J
Dis Child 28, 28–33.
Peterman, M.G. (1925). The ketogenic diet in epilepsy.
JAMA 84, 1979–1983.
Roth, T.L., and Sweatt, J.D. (2009). Regulation of chro-
matin structure in memory formation. Curr Opin
Neurobiol 19, 336–342.
Ryley Parrish, R., Albertson, A.J., Buckingham, S.C.,
Hablitz, J.J., Mascia, K.L., Davis Haselden, W., and
Lubin, F.D. (2013). Status epilepticus triggers early
and late alterations in brain-derived neurotrophic
factor and NMDA glutamate receptor Grin2b
DNA methylation levels in the hippocampus.
Neuroscience 248C, 602–619.
Schatz, R.A., Wilens, T.E., Tatter, S.B., Gregor, P., and
Sellinger, O.Z. (1983). Possible role of increased
brain methylation in methionine sulfoximine epi-
leptogenesis:  effects of administration of adenos-
ine and homocysteine thiolactone. J Neurosci Res
10, 437–447.
Sebastiao, A.M., and Ribeiro, J.A. (2009). Tuning and
fine-tuning of synapses with adenosine. Curr
Neuropharmacol 7, 180–194.
215
Sellinger, O.Z., Schatz, R.A., Porta, R., and Wilens, T.E.
(1984). Brain methylation and epileptogenesis: the
case of methionine sulfoximine. Ann Neurol, 16
(Suppl), S115–120.
Sirven, J., Whedon, B., Caplan, D., Liporace, J.,
Glosser, D., O’Dwyer, J., and Sperling, M.R.
(1999). The ketogenic diet for intractable epi-
lepsy in adults:  preliminary results. Epilepsia 40,
1721–1726.
Sofroniew, M.V., and Vinters, H.V. (2010).
Astrocytes:  biology and pathology. Acta
Neuropathol 119, 7–35.
Stafstrom, C.E., and Rho, J.M. (2012). The ketogenic
diet as a treatment paradigm for diverse neuro-
logical disorders. Front. Pharmacol. 3, 59.
Szybala, C., Pritchard, E.M., Lusardi, T.A., Li, T.,
Wilz, A., Kaplan, D.L., and Boison, D. (2009).
Antiepileptic effects of silk-polymer based ade-
nosine release in kindled rats. Exp. Neurol., 219,
126–135.
van der Louw, E.J., Desadien, R., Vehmeijer, F.O.,
van der Sijs, H., Catsman-Berrevoets, C.E., and
Neuteboom, R.F. (2015). Concomitant lamotrig-
ine use is associated with decreased efficacy of the
ketogenic diet in childhood refractory epilepsy.
Seizure 32, 75–77.
Van Gompel, J.J., Bower, M.R., Worrell, G.A., et  al.
(2014). Increased cortical extracellular adenosine
correlates with seizure termination. Epilepsia 55,
233–244.
Wheless, J.W. (2008). History of the ketogenic diet.
Epilepsia, 49 (Suppl 8), 3–5.
Wibisono, C., Rowe, N., Beavis, E., Kepreotes, H.,
Mackie, F.E., Lawson, J.A., and Cardamone,
M. (2015). Ten-year single-center experience
of the ketogenic diet:  factors influencing effi-
cacy, tolerability, and compliance. J Pediatr, 166,
1030–1036, e1031.
Williams-Karnesky, R.L., Sandau, U.S., Lusardi, T.A.,
Lytle, N.K., Farrell, J.M., Pritchard, E.M., Kaplan,
D.L., and Boison, D. (2013). Epigenetic changes
induced by adenosine augmentation therapy pre-
vent epileptogenesis. J Clin Invest 123, 3552–3563.
Wilz, A., Pritchard, E.M., Li, T., Lan, J.Q., Kaplan, D.L.,
and Boison, D. (2008). Silk polymer-based ade-
nosine release:  therapeutic potential for epilepsy.
Biomaterials 29, 3609–3616.
Zhang, T.Y., Labonte, B., Wen, X.L., Turecki, G., and
Meaney, M.J. (2013). Epigenetic mechanisms for
the early environmental regulation of hippocam-
pal glucocorticoid receptor gene expression in
rodents and humans. Neuropsychopharmacology
38, 111–123.
216
24
Ketogenic Diet, Aging, and Neurodegeneration
K U I X U, M D , P H D , J O S E P H C . L A M A N N A , P H D ,
A N D M I C H E L L E A . P U C H OW I C Z , P H D
OV E RV I E W
It has been established for over 50 years that brain
can utilize ketone bodies under carbohydrate-
(glucose-) sparing conditions (Cahill and Owen,
1967; Owen et al., 1967; Randle et al., 1963). The
use of ketogenic diet (KD) to treat intractable
epilepsy in children continues to be an option
for clinicians (DeVivo et al., 1978; Freeman et al.,
1998; Lennox, 1928; Millichap et al., 1964; Tallian
et al., 1998; Schwartzkroin, 1999), but the mecha-
nisms of action remain diverse (Bridge, 1931;
Jarrett et  al., 2008; Lund et  al., 2009; Maalouf
et al., 2009; Milder and Patel, 2012; Millichap et al.,
1964; Nakazawa et  al., 1983; Prins and Hovda,
2009; Yudkoff et al., 2007; Masino and Rho, 2012;
Maalouf et  al., 2007; Stafstrom and Rho, 2012).
The mechanistic link of ketosis and neuroprotec-
tion also remains unclear and exploratory. The
mechanisms appear to be related to the change in
the regulation of the cell’s stress responses (Milder
et al., 2010) and the changes in oxidative (glucose)
metabolism (Prins and Hovda, 2009; Puchowicz
et  al., 2008; Bough and Rho, 2007; Masino and
Rho, 2012). Neuroprotection by ketosis is thought
to be associated with improved mitochondrial
function, decreased reactive oxygen species (ROS)
and apoptotic and inflammatory mediators, and
increased protective pathways (Veech, 2004; Julio-
Amilpas et al., 2015).
This review focuses on two potential mecha-
nisms that elucidate the neuroprotective effects
of ketosis in the brain. We also concentrate on
the applications toward the treatment of the aged
brain, and present current data from rodent stud-
ies where KD was used to induce a chronic state of
ketosis and where neuroprotection was assessed.
Our view is that the neuroprotection associated
with diet-induced ketosis is mainly through sta-
bilization of glucose metabolism and downstream
metabolic pathways such as the citric acid cycle and
cytosolic-mitochondrial redox shuttle systems. We
also maintain that ketone bodies can act as signal-
ing molecules that target HIF1α-related cellular
defenses and regulatory systems. HIF1α activa-
tion can result in stabilization of the “redox state”
of the cell (Semenza, 2011). We have previously
shown neuroprotection by ketosis (in model of
middle cerebral artery occlusion, MCAO) in rat.
The hypothesis that the mechanistic link between
ketosis and neuroprotection is through HIF1α and
its downstream effects on metabolic enzymes was
established (Puchowicz et al., 2008).
K E TO N E B O D I E S A R E
A LT E R N AT E E N E R G Y
S U B S T R AT E S T O   G L U C O S E
Ketone bodies (or the state of ketosis) have been
shown to be neuroprotective following oxidative
stress and metabolic challenges, such as those asso-
ciated with stroke, ischemia, injury, Alzheimer’s
disease, Parkinson’s disease, glucose transporter
deficiency, and seizures (Maalouf et  al., 2009;
Prins and Hovda, 2009; Kashiwaya et  al., 2000;
Cunnane et al., 2011). Ketone bodies are alternate
energy substrates to glucose and thus they have the
potential to restore energy balance via stabilization
of ATP supply. They are thought to be more effi-
cient energy substrates to glucose for utilization
by brain (Cahill and Veech, 2003; Kashiwaya et al.,
2000; Sato et  al., 1995; Veech et  al., 2001; Veech,
2004). One application for the use of ketone bodies
as an alternative substrate is for those with aging
defects in glucose metabolism.
Hypometabolism is a clinical presentation in
some Alzheimer’s disease patients and is thought
to be associated with defects in glucose metabo-
lism (Cunnane et  al., 2011). Studies measuring
the ratio of 13C-amino acid /13C- lactate in brain
showed the fraction of amino acid carbon derived
from glucose decreased with ketosis, reflecting
the utilization of ketones as carbon precursors
for amino acid synthesis (aspartate, glutamine,
217
Chapter 24: Ketogenic Diet, Aging, and Neurodegeneration
glutamate, GABA) (Yudkoff et al., 2007). The neu-
roprotective role of ketone bodies may be through
improvement in metabolic efficiency, by sparing
glucose and the degradation of muscle-derived
amino acids for substrates.
Ketone body utilization has properties that
are considered metabolically favorable over glu-
cose: (1) the direct generation of NADH, FADH2
and succinate, which makes it readily available
(compartmentalized) to the mitochondria for
energy production; (2)  the direct generation of
acetyl-CoA, which enters the citric acid cycle via
the citrate synthase reaction, thus relieving oxi-
dative stress induced by pyruvate dehydrogenase
inhibition, as well as making available ATP and
lactate generated from glycolysis for other func-
tions, such as the shuttling of substrates between
neurons and glia; (3)  having the potential to
reduce mitochondrial ROS production associated
with reperfusion injury; and (4)  stabilization of
HIF1α independent of hypoxia.
U S E O F   K E TO G E N I C
D I E T TO   I N D U C E
K E TO S I S :   A   M O D E L
O F   S TA B I L I Z E D C H R O N I C
K E TO S I S
The metabolic modulations associated with ketone
bodies require a state of stable ketosis. The use of
a KD is an approach to establish chronic stable
ketosis. Diet-induced approaches avoid the con-
straints (sodium or fluid overloads, pH imbal-
ance, or starvation with chronic fasting) of using
alternate modes of inducing ketosis, such as with
infusions of ketone bodies or ketone body precur-
sors (sodium salts or acids), or long-term fasting
(Desrochers et  al., 1995; Puchowicz et  al., 2000).
It is well described that ketosis can be induced
in rodents and humans by fasting or feeding
KDs (Gano et  al., 2014; Leino et  al., 200; Milder
et  al., 2010; Pan et  al., 2000; Puchowicz et  al.,
2007; Tallian et  al., 1998; Veech, 2004; Veldhorst
et al., 2010; Yudkoff et al., 2007). We have previ-
ously reported that 3 weeks of diet-induced stable
ketosis is necessary to induce metabolic adapta-
tions in brain of rats resulting in neuroprotection
(Puchowicz et  al., 2007; Puchowicz et  al., 2008;
Zhang et al., 2013; Xu et al., 2012), which is consis-
tent with another study (Milder et al., 2010). This
regime can be inferred to as a preconditioning
state where blood concentrations of ketone bod-
ies remain stable over time. Adaptation to chronic
ketosis includes up-regulation of monocarboxyl-
ate transporters (MCT) at the blood-brain bar-
rier (Gjedde and Crone, 1975; Leino et  al., 2001;
217
Vannucci and Simpson, 2003)  and moderate to
high blood levels of ketosis (2–6 mM) (Leino et al.,
2001; Prins and Hovda, 2009). More recently, we
have established in our rat model of ketosis that
metabolic adaptations also include a correla-
tion between cerebral metabolic rate for glucose
(CMRglc) and blood ketone levels (Zhang et  al.,
2013). Using positron emission tomography imag-
ing technology we showed a decrease of almost
10% in CMRglc for each 1-mM increase of total
plasma ketone bodies in cortical and cerebellar
brain regions. Together with our meta-analysis,
these data revealed that the degree and duration
of ketosis played a major role in determining the
corresponding change in CMRglc with ketosis and
that preconditioning with KG diet resulted in sig-
nificant glucose sparing (Zhang et al., 2013; Zhang
et al., 2015).
O X I D AT I V E S T R E S S
I N D U C E D A LT E R E D
G L U C O S E M E TA B O L I S M
Supply of energy substrate to the mitochondria
is critical, especially during conditions of high
energy demand where glucose metabolism may be
deficient, such as with oxidative injury. The abil-
ity of the central nervous system to recover from
an ischemic and/or hypoxic event is its capac-
ity to recover from metabolic stress. It is well
known that the status of cellular bioenergetics of
the neurovascular unit is a major determinate of
the pathophysiologic outcome. Stoichiometric
analysis would predict that glycolysis cannot sus-
tain NADH demand (energy demand) for more
than a few hours, depending on glycogen stores.
The dogma of tissue reperfusion for cell sur-
vival together with associated oxidative damage,
immune responses, and increased risk for cell
death remains problematic.
Under pathologic conditions the brain’s energy
demand and supply are mismatched, resulting
in energy inefficiency. For example, survival and
recovery of neurologic function after resuscita-
tion following cardiac arrest is limited by the
ability of the central nervous system to recover
from an ischemic event due to metabolic stress
to brain. Following cardiac arrest and resuscita-
tion, increased glycolytic rates are associated with
lactic acidosis and downstream metabolic blocks
in energetics. After 24 hours of reperfusion there
is an “apparent” defect in glucose metabolism,
which may continue and occur as a second-
ary response to oxidative stress-induced injury
(Selman et  al., 1990). At this stage of oxidative
damage there is a decrease in CMRglc (Bentourkia
218
218 section III: Ketogenic Diet in the Laboratory
100
90
Survival rate (%)
80
70
60
STD
KG
50
0 1
FIGURE 24.1 Overall survival rates 4 days following cardiac arrest and resuscitation. The rats fed with 3 weeks of
ketogenic diet (KD) had a significantly higher survival rates compared with the rats fed with standard diet (STD) (86%
vs. 55%, p < .05, Wilcoxon survival analysis) (Xu et al., 2012).
et al., 2000; DiVivo et al., 1973; Willis et al., 2002).
Mitochondria are among the most susceptible
organelles to oxidative stress. Brain mitochon-
drial function is known to decrease 2 days follow-
ing ischemia reperfusion (Xu et al., 2008), which
more often results in delayed neuronal death after
4 days post recovery (Xu et al., 2006). The degree
of neurological deficit is highly dependent on the
length of time to restored blood reflow and on the
severity of oxidative damage. The generation of
ROS and cytotoxic products of lipid peroxidation,
such as 4-hydroxy-2-nonenal (HNE) and glutathi-
one (GSH), also play a major role in the cell’s abil-
ity to recover. In a previous study we have shown
that KD-induced ketosis resulted in significantly
higher overall survival rates 4  days post cardiac
arrest and resuscitation compared with the stan-
dard diet group (86% vs. 55%, p < .05, Figure 24.1).
These data support the idea that diet-induced
ketosis is neuroprotective under conditions when
glucose utilization is limited (Xu et al., 2012).
Studies, past and present, continue to show
evidence that altered energy metabolism is asso-
ciated with aging and/or oxidative damage to
key enzymes that regulate glucose metabolism
(Cunnane et  al., 2011; Gibson et  al., 1998). The
imbalance between nonoxidative- and oxidative-
derived ATP is recognized but not clearly under-
stood (Hemmila and Drewes, 1993; Hempel et al.,
1977; Hoxworth et  al., 1999; Xu et  al., 2006; Xu
et al., 2008). One explanation may be the lack of
glucose carbon entry (flux) into the citric acid
cycle (i.e., via pyruvate dehydrogenase complex)
resulting in a dysregulation or “leakiness” of the
citric acid cycle (cataplerosis). These consequences
KG
STD
Number of deaths Survival rate
(4d/total)
<1d 1–2 d 2–3 d 3–4 d
2 2 3 2 55% (10/11)
1 0 0 0 86% (6/7)
2 3 4
Recovery (days)
can lead to irreversible neurological deficit.
Anaplerosis, a process that balances cataplero-
sis, ordinarily maintains/supplies intermediates
to the citric acid cycle, which becomes vital dur-
ing increased leakiness associated with ischemia
reperfusion injury (Brunengraber and Roe, 2006;
Kasumov et al., 2007) and other neuropathologies.
The balance between cataplerosis and anaplerosis
is essential to maintaining energy balance and,
hence, mitochondrial function. In a recent study,
we investigated the role of ketosis on the partition-
ing of glucose and BHB as substrates at the level of
the acetyl-CoA and citric acid cycle (Zhang et al.,
2015). This study revealed that glucose entry into
the citric acid cycle was decreased with ketosis,
whereas BHB entry was increased. The increased
contribution of BHB to central metabolism sug-
gests that carbon backbones from ketones play a
role in the balancing of the citric acid cycle. The
possibility of increased pyruvate carboxylation
via malic enzyme should also be considered as
a mechanism of ketosis in energy stabilization
(Hassel, 2000).
D I E T- I N D U C E D K E T O S I S
A S   A N E U RO P ROT E C T I V E
S T R AT E G Y I N   T H E
AGED: CLINICAL
R E L E VA N C E
The aging population is at risk for increased mor-
bidity and mortality following ischemic or hypoxic
events, such as those related to stroke or other
neurodegenerative conditions. Stroke is a leading
disease/disability in the United States, as there are
few (if any) clinical treatment strategies that can
219

Chapter 24: Ketogenic Diet, Aging, and Neurodegeneration 219

STD KG
160

120
Infarct volume (mm3)

80

40

0
STD KG

FIGURE 24.2 The total brain infarct volume artery occlusion (MCAO) in 18-month-old rats fed with ketogenic or
standard diet. Rats were randomly assigned to two diet groups, fed ad libitum, ketogenic (high fat, no carbohydrate; KG)
or standard lab-chow (STD) diet for 3 weeks prior to ischemic stroke. Rats underwent 90 minutes of MCAO, the total
infarct volume was evaluated by triphenyltetrazolium chloride (TTC) staining 24 hours after reperfusion. Results: After
fed with 3 weeks of ketogenic diet, plasma ketone bodies (mM) were increased 5-fold (3.1 ± 0.8 vs. 0.6 ± 0.1, mean ±
SEM, ketogenic diet vs. Standard diet, n = 4 each). After 24 hours of reperfusion following MCAO, the infarct volumes
were 5-fold lower in the KG diet group compared with the STD diet group (22 ± 10 vs. 124 ± 26 mm3, mean ± SEM,
n = 4 each).

reverse cellular damage. All brain cells are suscep-
tible to infarction following ischemia-reperfusion
(Pundik et al., 2012). Another example of a condi-
tion that results in moderate to severe ischemia/
reperfusion injury is cardiac arrest and resuscita-
tion, as the brain is exposed to ischemic-hypoxia
and reperfusion. In the United States about one
million cardiac arrests occur per year; about half
of the population (aged or adult) having first-time
cardiac arrests will not survive the first few days.
Of those who do survive, about 90% will have
short- or long-term neurological deficits (Chugh
et  al., 2004; Goldberger et  al., 2008; Safar,1993;
Zheng et al., 2001). These deficits are more often
related to oxidative stress induced post resuscita-
tion mortality and delayed selective neuronal cell
loss (Hoxworth et  al., 1999; Xu et  al., 2006; Xu
et al., 2008). In the aged, failure to recover is espe-
cially pronounced, as the aged brain is susceptible
to oxidative stress damage as result of declines in
repair systems and decreased antioxidant capacity.
Brain pathophysiology associated with oxidative
stress and injury, such as with ischemic reperfusion
injury, often results in energy imbalances related to
dysregulation of glucose (oxidative) metabolism. It
could be that ketones are effective against pathol-
ogy associated with altered glucose metabolism
(Cahill and Veech, 2003; Prins and Hovda, 2009;
Sato et  al., 1995; Veech et  al., 2001; Veech, 2004;
Veech et al., 2012). Ketones may also play a role as
signaling molecules that directly or indirectly act
through the regulation of cell salvation pathways,
such as HIF1α. In a recent study, we showed that
ketosis was induced in aged rats (18 months old)
following 3 weeks of feeding a KD and that diet-
induced ketosis was neuroprotective against focal
cerebral ischemia (MCAO; Figure 24.2).
N E U RO P ROT E C T I O N
T H R O U G H   S TA B I L I Z AT I O N
OF HIF1Α
The notion that ketone bodies can act as metabolic
signaling molecules in addition to being alternate
oxidative substrates to glucose has been considered
(Shimazu et al., 2013; Bough and Rho, 2007; Gano
et al., 2014; Maalouf et al., 2009; Milder et al., 2010;
220
220 section III: Ketogenic Diet in the Laboratory
Tannahill et al., 2013). The role of HIF1α as a key
regulator of oxygen homeostasis during hypoxia
is well described (Agani et al., 2002; Chavez et al.,
2000; Chavez et  al., 2006; Semenza, 2007, 2011;
Semenza et al., 2000; Sharp et al., 2001). However,
the metabolic role of HIF1α on neuroprotection
remains unclear.
HIF1α is a nuclear factor associated with neu-
roprotection via regulation of energy metabolism
and is a key regulator of oxygen homeostasis dur-
ing hypoxia, but may also play a role in neuropro-
tection following ischemia through gene regulation
(Shi, 2009; Guo et al., 2008). HIF1 activation can
also promote cell survival (Bergeron et al., 2000).
The translational impact on understanding the
role of ketone bodies as metabolic regulators can
lead to a more definitive approach on treatment
strategies that target defects in energy metabolism
associated with neuropathologies and degenera-
tive disorders. Since the aged population has an
increased risk for ischemic events, such as those
associated with transient global or focal stroke, the
development of treatment strategies that incorpo-
rate ketosis may be wise.
The mechanism of HIF1α stabilization through
ketosis has been proposed (Puchowicz et  al.,
2008) to be most likely through changes in cyto-
plasmic/mitochondrial redox state (Guo et  al.,
2008). This is consistent with others reporting
stabilization of HIF1α through impairment of pro-
teasome function, as a result of very low ratios of
alpha-ketoglutarate /fumarate (Serra-Perez et  al.,
2010). A  mechanism that explains diet-induced
HIF1α stabilization (via KD) is through the feed-
back inhibition of the prolyl-hydroxylase (PHD)
reaction by succinate, an intermediate of energy
metabolism (see scheme below). Elevated cellu-
lar levels of succinate and fumarate compete with
alpha-ketoglutarate PHD binding, resulting in
inhibition of PHD and thus HIF1α stabilization
(Hou et  al., 2014; Mills and O’Neill, 2014). The
mechanisms of ketosis that are relevant to neuro-
protection and stabilization of HIF1α (via succi-
nate), need to be further studied. We propose that
PHD inhibition occurs as a result of ketone body
utilization by the cell via the activation of aceto-
acetate to acetoacetyl-CoA (via 3-ketoacyl-CoA
transferase), which involves the generation of suc-
cinate in the mitochondria. Once activated, HIF1
modulates energy metabolism through down-
stream regulation of metabolism and related target
genes. HIF1α target genes include those related to
angiogenesis, erythropoiesis, glucose metabolism,
and cell survival (Agani et  al., 2002; Baranova
et al., 2007; Chavez et al., 2006; Chen et al., 2015;
Semenza, 2001; Sharp et al., 2001; Puchowicz et al.,
2008; Helton et al., 2005).
We have reported that HIF1α response to
hypoxia is blunted in the aged brain due to
increased PHD levels (Ndubuizu et  al., 2010).
The lack of HIF1α induction following an oxida-
tive insult, such as with reperfusion injury and/
or aging process, increases vulnerability to oxida-
tive damage that can result in poor recovery or
cell death (Xu et al., 2007; Xu et al., 2008). These
findings are consistent with other reports on mice
with neuron-specific knockdown of HIF1α and
subjected to transient focal cerebral ischemia;
the results showed increased tissue damage and
reduced cell survival (Baranova et al., 2007).
Neuroprotection by HIF target genes such as
erythropoietin (Epo) has been described (Chong
et al., 2003a; Ghezzi et al., 2010). As a neuroprotec-
tive agent Epo has many functions:  antagonizing
glutamate cytotoxic action, enhancing antioxidant
enzyme expression, reducing free radical produc-
tion rate, and affecting neurotransmitter release
(Bartesaghi et al., 2005; Leist et al., 2004). It exerts
its neuroprotective effect indirectly through res-
toration of blood flow or directly by activating
transmitter molecules in neurons that also play a
role in erythrogenesis. Although the mechanism is
unclear, it is apparent that Epo has anti-apoptotic
action after central and peripheral nerve injury
(Chong et  al., 2003b). Although apoptosis is not
reversible, early intervention with neuroprotective
therapeutic procedures such as Epo administra-
tion or preconditioning with KD may reduce the
number of neurons that undergo apoptosis and
thus improve outcome.
An additional target gene of HIF is the tran-
scription factor vascular endothelial growth factor
(VEGF). The VEGF is an endothelial cell-specific
growth factor (Storkebaum et al., 2004), but recent
evidence indicates that VEGF also has direct effects
on neuronal and glial cell types (Silverman et al.,
1999). A neuroprotective mechanism of VEGF is
thought to be through increased perfusion of the
penumbra via vasodilation and angiogenesis, such
as with ischemic brain insults (Zhang et al., 2002).
We have previously reported that preconditioning
with KD results in an angiogenic response similar
to what we have found with mild hypoxic expo-
sure (Puchowicz et al., 2007). However, we found
no changes in cerebral regional blood flow with
KD-induced ketosis in adult rats. In another study
we reported that VEGF expression was decreased
in cerebral cortex of aged mice (Benderro and
LaManna, 2011). Taken all together, KD might be
a therapeutic strategy that targets neuroprotective
221

Chapter 24: Ketogenic Diet, Aging, and Neurodegeneration 221

mechanisms through HIF-regulated VEGF, inde- 2008; Julio-Amilpas et al., 2015; Sato et al., 1995;
pendent of modifications in blood perfusion Veech, 2004). Second, ketosis induces the stabili-
(Chong et al., 2003b). zation of HIF1α and its downstream target genes.
One possible link between ketosis and the sta-
DISCUS SION bilization of HIF1α is through feedback product
The mechanism of HIF1 regulation under various inhibition of PHD via elevated tissue succinate
metabolic conditions, such as with stroke, inflam- (Semenza, 2007), as a result of a change in the
mation, and altered glucose metabolism, continues metabolic redox state (see Scheme figure 24.3)
to be explored. We describe two mechanisms that (Guo et al., 2008; Serra-Perez et al., 2010; Mills
explain neuroprotection through diet-induced and O’Neill, 2014). This is consistent with a study
ketosis. First, ketone bodies are alternate energy where mitochondrial redox in hippocampal tis-
substrates to glucose, thus reducing oxidative sue was improved by KD via the activation of NF
stress by relieving metabolic blocks downstream E2-related factor 2(Nrf2) transcription factor, a
of glycolysis at the level of the citric acid cycle and, primary responder to cellular stress (Milder et al.,
hence, enabling mitochondrial electron transfer 2010). One known mechanism of Nrf2 is to up-
and generation of ATP (Semenza, 2011; Prins, regulate GSH biosynthesis. On the other hand,

Cytosol Mitochondrion
Ubiquitin
HIF1α HIF1α 1
OH
VHL
OH Succinate

Proteasome O2 PHD1
PHD2
Fe2+
5 Vit C PHD3 Succinyl-CoA Succinate
Angiogenesis HIF1α α-Ketoglutarate
2
α-Ketoglutarate Fumarate
EPO Malate Malate
OAA OAA
Glut-1
VEGF
5 5 Isocitrate 3 Malate
Malate Aspartate
4 Shuttle
NADH NAD
+ +
NAD NADH
Citrate Oxaloacetate
3
Glutamate Glutamate
Aspartate Aspartate
+
NAD NADH

1 Acetyl-CoA
4
Glucose Pyruvate
Legend 6
NADH

1 Succinate Transporter
2 a-Ketoglutarate/Malate Transporter Lactate NAD
+

3 Aspartate/Glutamate Transporter Succinyl-CoA Succinate

4 Pyruvate Transporter
5 Malate Dehydrogenase 2 Ketone Bodies
BHB, AcAc
AcAc-CoA
6 Hexokinase

FIGURE 24.3 Scheme: Proposed mechanism of neuroprotection by diet-induced ketosis.

Our scheme emphasizes that cytosolic generated succinate (via PHD-coupled activity with alpha-ketoglutarate), must be
balanced by mitochondrial redox of the citric acid cycle (shown in gray text). An accumulation of succinate would result
in an imbalance of cytosolic redox; HIF1α would then be stabilized through feedback inhibition of PHD (text shown in
black, Puchowicz et al., 2008). Ketone body utilization also results in the generation of succinate. Steps 1–5 (circled):
HIF1α stabilization is shown to be through cellular redox, which is balanced by malate-aspartate shuttle. A link to the
stabilization HIF1α is through feedback product inhibition of prolyl-hydroxylase (PHD) via elevated tissue succinate
(Semenza, 2007) or reduced α-ketoglutarate (McFate et al., 2008) (steps 3,4). Thus, an accumulation of succinate (or
reduction in α-ketoglutarate) would result in an altered in redox state (malate aspartate shuttle, step 4) and stabilization
of HIF1α (via inhibition of PHD) (step 5). Other intermediates of energy metabolism acting on HIF1α regulation have
been recently described (Chen et al., 2015; Mills and O’Neill, 2014; Tannahill et al., 2013). The potential mechanism by
which ketosis stabilizes HIF1α is through a shift in “redox state” of the cell via changes in carbon flux into / out of the
malate-aspartate shuttle (Semenza, 2011); a target would include changes in succinate-to-α-ketoglutarate ratio.
222
222 section III: Ketogenic Diet in the Laboratory
oxidizing environments (high levels of ROS) may
suppress stabilization of HIF1α (Guo et al., 2008).
These findings are consistent with a study show-
ing that HIF1α played a role in N-acetylcysteine-
mediated neuroprotection in rats subjected to
MCAO (Zhang et al., 2014).
Our view is that there is a feedback regulation
on HIF1α activation through shifts in metabolic
redox, such as with KD or fasting which is inde-
pendent of hypoxia-induced stabilization of HIF1α
(Mills and O’Neill, 2014). Thus, HIF1α acts as a
metabolic sensor through a continued feedback
loop directly coupled to oxidative metabolism via
the citric acid cycle and malate aspartate shuttle
systems (Figure 24.3). It has been suggested that
HIF1α may also play a role promoting apoptosis
(Helton et al., 2005; Serra-Perez et al., 2010), how-
ever there is little known about this mechanism.
Neuroprotection by HIF1α may be cell specific.
It has been reported that neuron-specific inac-
tivation of the HIF1α results in increased brain
injury, as studied in a mouse model of transient
focal cerebral ischemia (Baranova et al., 2007). The
significance of these findings suggests that HIF1α
and target genes are regulated distinctly, especially
under pathological conditions. Although HIF1α is
expressed in neurons and glia, the neuroprotective
efficacy of a KG diet (via HIF1α) may be cell spe-
cific and conditional (Bergeron et al., 2000; Chavez
and LaManna, 2002). Previously, we have reported
that HIF1α response to hypoxia is suppressed in the
aged and proposed that this was an explanation to
why the aged are at greater risk for morbidity and
mortality following an oxidative insult (Ndubuizu
et al., 2010). Additionally, the potential role of KD
relative to HIF2α should also be explored (Chavez
et  al., 2006). Our approach to understanding the
central role of ketosis in the regulation of HIF1α
as a signaling molecule in brain under pathophysi-
ological conditions will help to delineate the role of
HIF1α in neuroprotection.
CONCLUSION AND
FUTURE
Over the past decade our research has consistently
shown that ketosis is neuroprotective against isch-
emic insults in rats (Puchowicz et al., 2008) and,
more recently, in mice (unpublished observa-
tion). Studies in ketotic rats showed significantly
decreased infarct volumes (via MCAO model)
and improved survival and recovery after cardiac
arrest and resuscitation (Xu et al., 2012). We pur-
port that the mechanisms of ketosis include glu-
cose sparing and ketone body utilization in brain
as well as a shift in metabolic redox, which plays a
role in energy balance and the downstream innate
signaling of HIF1α. We view that the mecha-
nism of neuroprotection by KD-induced HIF1α
is through the regulation and activation of genes
associated with salvation pathways and stabili-
zation of energy metabolism. The neuroprotec-
tion by ketosis may also be through pathways
independent of HIF1α regulation, such as AKT-
modulation of NRF2 and nuclear factor kappa
B (NF-kB), and these mechanisms remain to be
elucidated.
REFERENCES
Agani, F.H., Puchowicz, M., Chavez, J.C., Pichiule,
P., and LaManna, J. (2002). Role of nitric oxide
in the regulation of HIF-1alpha expression dur-
ing hypoxia. Am J Physiol Cell Physiol 283,
C178-C186.
Baranova, O., Miranda, L.F., Pichiule, P., Dragatsis, I.,
Johnson, R.S., and Chavez, J.C. (2007). Neuron-
specific inactivation of the hypoxia inducible fac-
tor 1 alpha increases brain injury in a mouse model
of transient focal cerebral ischemia. J Neurosci 27,
6320–6332.
Bartesaghi, S., Marinovich, M., Corsini, E., Galli, C.L.,
and Viviani, B. (2005). Erythropoietin:  a novel
neuroprotective cytokine. Neurotoxicology 26,
923–928.
Benderro, G.F., and LaManna, J.C. (2011). Hypoxia-
induced angiogenesis is delayed in aging mouse
brain. Brain Res 1389, 50–60.
Bentourkia, M., Bol, A., Ivanoiu, A., Labar, D.,
Sibomana, M., Coppens, A., Michel, C., Cosnard, G.,
and De Volder, A.G. (2000). Comparison of
regional cerebral blood flow and glucose metabo-
lism in the normal brain: effect of aging. J Neurol
Sci 181, 19–28.
Bergeron, M., Gidday, J.M., Yu, A.Y., Semenza, G.L.,
Ferriero, D.M., and Sharp, F.R. (2000). Role of
hypoxia-inducible factor-1 in hypoxia-induced
ischemic tolerance in neonatal rat brain. Ann
Neurol 48, 285–296.
Bough, K.J., and Rho, J.M. (2007). Anticonvulsant
mechanisms of the ketogenic diet. Epilepsia
48, 43–58.
Bridge, E.M. (1931). The mechanism of the keto-
genic diet in epilepsy. Bull Johns Hopkins Hos 48,
373–389.
Brunengraber, H., and Roe, C.R. (2006). Anaplerotic
molecules: current and future. J Inherit Metab Dis
29, 327–331.
Cahill, G.F., Jr., and Veech, R.L. (2003). Ketoacids?
Good medicine? Trans Am Clin Climatol Assoc
114, 149–161.
Cahill, G.F.J., and Owen, O.E. (1967). Starvation and
survival. Trans Am Clin Climat Ass 79, 13–18.
Chavez, J.C., Agani, F., Pichiule, P., and LaManna,
J.C. (2000). Expression of hypoxia-inducible
223
Chapter 24: Ketogenic Diet, Aging, and Neurodegeneration
factor-1alpha in the brain of rats during chronic
hypoxia. J Appl Physiol (1985) 89, 1937–1942.
Chavez, J.C., Baranova, O., Lin, J., and Pichiule, P.
(2006). The transcriptional activator hypoxia
inducible factor 2 (HIF-2/EPAS-1) regulates the
oxygen-dependent expression of erythropoietin in
cortical astrocytes. J Neurosci 26, 9471–9481.
Chavez, J.C., and LaManna, J.C. (2002). Activation
of hypoxia-inducible factor-1 in the rat cerebral
cortex after transient global ischemia:  potential
role of insulin-like growth factor-1. J Neurosci 22,
8922–8931.
Chen, T.T., Maevsky, E.I., and Uchitel, M.L. (2015).
Maintenance of homeostasis in the aging hypo-
thalamus: the central and peripheral roles of suc-
cinate. Front Endocrinol (Lausanne) 6, 7.
Chong, Z.Z., Kang, J.Q., and Maiese, K. (2003a). Apaf-
1, Bcl-xL, cytochrome c, and caspase-9 form the
critical elements for cerebral vascular protection
by erythropoietin. J Cereb Blood Flow Metab 23,
320–330.
Chong, Z.Z., Kang, J.Q., and Maiese, K. (2003b).
Erythropoietin fosters both intrinsic and extrin-
sic neuronal protection through modulation of
microglia, Akt1, Bad, and caspase-mediated path-
ways. Br J Pharmacol 138, 1107–1118.
Chugh, S.S., Jui, J., Gunson, K., Stecker, E.C., John,
B.T., Thompson, B., Ilias, N., Vickers, C., Dogra,
V., Daya, M., et al. (2004). Current burden of sud-
den cardiac death:  multiple source surveillance
versus retrospective death certificate-based review
in a large U.S. community. J Am Coll Cardiol 44,
1268–1275.
Cunnane, S., Nugent, S., Roy, M., Courchesne-Loyer,
A., Croteau, E., Tremblay, S., Castellano, A., Pifferi,
F., Bocti, C., Paquet, N., et  al. (2011). Brain fuel
metabolism, aging, and Alzheimer’s disease.
Nutrition 27, 3–20.
Desrochers, S., Dubreuil, P., Brunet, J., Jetté, M., David,
F., Landau, B.R., and Brunengraber, H. (1995).
Metabolism of (R, S)-1, 3-butanediol acetoacetate
esters, potential parenteral and enteral nutrients in
conscious pigs. Am J Physiol 268, Pt 1):E660–667.
DeVivo, D.C., Leckie, M.P., Ferrendelli, J.S., and
McDougal, D.B., Jr. (1978). Chronic ketosis and
cerebral metabolism. Ann Neurol 3, 331–337.
DiVivo, D.C., Pagliara, A.S., and Prensky, A.L. (1973).
Ketotic hypoglycemia and the ketogenic diet ther-
apy. Neurology 23, 640–649.
Freeman, J.M., Vining, E.P.G., Pillas, D.J., Pyzik, P.L.,
Casey, J.C., and Kelly, M.T. (1998). The efficacy of
the ketogenic diet—1998: a prospective evaluation
of intervention in 150 children. Pediatrics 102,
1358–1363.
Gano, L.B., Patel, M., and Rho, J.M. (2014). Ketogenic
diets, mitochondria, and neurological diseases. J
Lipid Res 55, 2211–2228.
Ghezzi, P., Bernaudin, M., Bianchi, R., Blomgren,
K., Brines, M., Campana, W., Cavaletti, G.,
223
Cerami, A., Chopp, M., Coleman, T., et al. (2010).
Erythropoietin:  not just about erythropoiesis.
Lancet 375, 2142.
Gibson, G.E., Sheu, K.F., and Blass, J.P. (1998).
Abnormalities of mitochondrial enzymes in
Alzheimer disease. J Neural Transm 105, 855–870.
Gjedde, A., and Crone, C. (1975). Induction processes
in blood-brain transfer of ketone bodies during
starvation. Am J Physiol 229, 1165–1169.
Goldberger, J.J., Cain, M.E., Hohnloser, S.H., Kadish,
A.H., Knight, B.P., Lauer, M.S., Maron, B.J., Page,
R.L., Passman, R.S., Siscovick, D., et  al. (2008).
American Heart Association/American College of
Cardiology Foundation/Heart Rhythm Society sci-
entific statement on noninvasive risk stratification
techniques for identifying patients at risk for sud-
den cardiac death: a scientific statement from the
American Heart Association Council on Clinical
Cardiology Committee on Electrocardiography
and Arrhythmias and Council on Epidemiology
and Prevention. Circulation 118, 1497–1518.
Guo, S., Bragina, O., Xu, Y., Cao, Z., Chen, H., Zhou,
B., Morgan, M., Lin, Y., Jiang, B.H., et al. (2008).
Glucose up-regulates HIF-1 alpha expression in
primary cortical neurons in response to hypoxia
through maintaining cellular redox status. J
Neurochem 105, 1849–1860.
Hassel, B. (2000). Carboxylation and anaplerosis in
neurons and glia. Mol Neurobiol 22, 21–40.
Helton, R., Cui, J., Scheel, J.R., Ellison, J.A., Ames,
C., Gibson, C., Blouw, B., Ouyang, L., Dragatsis,
I., Zeitlin, S., et  al. (2005). Brain-specific knock-
out of hypoxia-inducible factor-1alpha reduces
rather than increases hypoxic-ischemic damage. J
Neurosci 25, 4099–4107.
Hemmila, J.M., and Drewes, L.R. (1993). Glucose
transporter (GLUT1) expression by canine
brain microvessel endothelial cells in culture:  an
immunocytochemical study. Adv Exp Med Biol
331, 13–18.
Hempel, F.G., Jobsis, F.F., LaManna, J.L., Rosenthal,
M.R., and Saltzman, H.A. (1977). Oxidation of
cerebral cytochrome aa3 by oxygen plus carbon
dioxide at hyperbaric pressures. J Appl Physiol 43,
873–879.
Hou, P., Kuo, C.Y., Cheng, C.T., Liou, J.P., Ann, D.K.,
and Chen, Q. (2014). Intermediary metabolite
precursor dimethyl-2-ketoglutarate stabilizes
hypoxia-inducible factor-1alpha by inhibiting
prolyl-4-hydroxylase PHD2. PLoS One 9, e113865.
Hoxworth, J.M., Xu, K., Zhou, Y., Lust, W.D., and
LaManna, J.C. (1999). Cerebral metabolic pro-
file, selective neuron loss, and survival of acute
and chronic hyperglycemic rats following cardiac
arrest and resuscitation. Brain Res 821, 467–479.
Jarrett, S.G., Milder, J.B., Liang, L.P., and Patel,
M. (2008). The ketogenic diet increases mito-
chondrial glutathione levels. J Neurochem 106,
1044–1051.
224
224 section III: Ketogenic Diet in the Laboratory
Julio-Amilpas, A., Montiel, T., Soto-Tinoco, E.,
Geronimo-Olvera, C., and Massieu, L. (2015).
Protection of hypoglycemia-induced neuronal
death by beta-hydroxybutyrate involves the pres-
ervation of energy levels and decreased produc-
tion of reactive oxygen species. J Cereb Blood Flow
Metab 35, 851–860.
Kashiwaya, Y., Takeshima, T., Mori, N., Nakashima,
K., Clarke, K., and Veech, R.L. (2000). D-beta-
hydroxybutyrate protects neurons in models of
Alzheimer’s and Parkinson’s disease. Proc Natl
Acad Sci USA 97, 5440–5444.
Kasumov, T., Cendrowski, A.V., David, F., Jobbins,
K.A., Anderson, V.E., and Brunengraber, H.
(2007). Mass isotopomer study of anaplerosis
from propionate in the perfused rat heart. Arch
Biochem Biophys 463, 110–117.
Leino, R.L., Gerhart, D.Z., Duelli, R., Enerson, B.E.,
and Drewes, L.R. (2001). Diet-induced ketosis
increases monocarboxylate transporter (MCT1)
levels in rat brain. Neurochem Int 38, 519–527.
Leist, M., Ghezzi, P., Grasso, G., Bianchi, R., Villa, P.,
Fratelli, M., Savino, C., Bianchi, M., Nielsen, J.,
Gerwien, J., et  al. (2004). Derivatives of erythro-
poietin that are tissue protective but not erythro-
poietic. Science 305, 239–242.
Lennox, W.B. (1928). Ketogenic diet in the treatment
of epilepsy. N Engl J Med 199, 74–75.
Lund, T.M., Risa, O., Sonnewald, U., Schousboe, A.,
and Waagepetersen, H.S. (2009). Availability of
neurotransmitter glutamate is diminished when
beta-hydroxybutyrate replaces glucose in cultured
neurons. J Neurochem 110, 80–91.
Maalouf, M., Sullivan, P.G., Davis, L., Kim, D.Y., and
Rho, J.M. (2007). Ketones inhibit mitochondrial
production of reactive oxygen species production
following glutamate excitotoxicity by increasing
NADH oxidation. Neuroscience 145, 256–264.
Maalouf, M., Rho, J.M., and Mattson, M.P. (2009). The
neuroprotective properties of calorie restriction,
the ketogenic diet, and ketone bodies. Brain Res
Rev 59, 293–315.
Masino, S.A., and Rho, J.M. (2012). Mechanisms of
ketogenic diet action. In Jasper’s Basic Mechanisms
of the Epilepsies, J.L. Noebels, M. Avoli, M.A.
Rogawski, R.W. Olsen, and A.V. Delgado-Escueta,
eds. (Bethesda: National Center for Biotechnology
Information), pp. 1001–1022.
McFate, T., Mohyeldin, A., Lu, H., Thakar, J., Henriques,
J., Halim, N.D., Wu, H., Schell, M.J., Tsang, T.M.,
Teahan, O., et al. (2008). Pyruvate dehydrogenase
complex activity controls metabolic and malig-
nant phenotype in cancer cells. J Biol Chem 283,
22700–22708.
Milder, J., and Patel, M. (2012). Modulation of oxida-
tive stress and mitochondrial function by the keto-
genic diet. Epilepsy Res 100, 295–303.
Milder, J.B., Liang, L.P., and Patel, M. (2010). Acute
oxidative stress and systemic Nrf2 activation by
the ketogenic diet. Neurobiol Dis 40, 238–244.
Millichap, J.G., Jones, J.D., and Rudis, B.P. (1964).
Mechanism of anti-convulsant actin of ketogenic
diet. Am J Dis Child 107, 593–604.
Mills, E., and O’Neill, L.A. (2014). Succinate: a meta-
bolic signal in inflammation. Trends Cell Biol 24,
313–320.
Nakazawa, M., Kodama, S., and Matsuo, T. (1983).
Effects of ketogenic diet on electroconvulsive
threshold and brain contetents of adenosine nucle-
otides. Brain Dev 5, 375–380.
Ndubuizu, O.I., Tsipis, C.P., Li, A., and LaManna,
J.C. (2010). Hypoxia-inducible factor-1 (HIF-1)-
independent microvascular angiogenesis in the
aged rat brain. Brain Res 1366, 101–109.
Owen, O.E., Morgan, A.P., Kemp, H.G., Sullivan,
J.M., Herrera, M.G., and Cahill, G.F., Jr. (1967).
Brain metabolism during fasting. J Clin Invest 46,
1589–1595.
Pan, J.W., Rothman, T.L., Behar, K.L., Stein, D.T., and
Hetherington, H.P. (2000). Human brain beta-
hydroxybutyrate and lactate increase in fasting-
induced ketosis. J Cereb Blood Flow Metab 20,
1502–1507.
Prins, M.L. (2008). Cerebral metabolic adaptation
and ketone metabolism after brain injury. J Cereb
Blood Flow Metab 28, 1–16.
Prins, M.L., and Hovda, D.A. (2009). The effects of
age and ketogenic diet on local cerebral metabolic
rates of glucose after controlled cortical impact
injury in rats. J Neurotrauma 26, 1083–1093.
Puchowicz, M.A., Smith, C.L., Bomont, C., Koshy, J.,
David, F., and Brunengraber, H. (2000). Dog model
of therapeutic ketosis induced by oral administra-
tion of R, S-1, 3-butanediol diacetoacetate. J Nutr
Biochem 11, 281–287.
Puchowicz, M.A., Xu, K., Sun, X., Ivy, A., Emancipator,
D., and LaManna, J.C. (2007). Diet-induced keto-
sis increases capillary density without altered
blood flow in rat brain. Am J Physiol Endocrinol
Metab 292, E1607–E1615.
Puchowicz, M.A., Zechel, J.L., Valerio, J., Emancipator,
D.S., Xu, K., Pundik, S., LaManna, J.C., and Lust,
W.D. (2008). Neuroprotection in diet-induced
ketotic rat brain after focal ischemia. J Cereb Blood
Flow Metab 28, 1907–1916.
Pundik, S., Xu, K., and Sundararajan, S. (2012).
Reperfusion brain injury: focus on cellular bioen-
ergetics. Neurology 79, S44-S51.
Randle, P.J., Garland, P.B., Hales, C.N., and Newsholme,
E.A. (1963). The glucose-fatty acid cycle: its role in
insulin sensitivity and the metabolic disturbances
of diabetes mellitus. Lancet 1, 785–789.
Safar, P. (1993). Systematic development of cerebral
resuscitation after cardiac arrest. Three promising
treatments:  cardiopulmonary bypass, hyperten-
sive hemodilution, and mild hypothermia. Acta
Neurochir Suppl (Wien) 57, 110–121.
Sato, K., Kashiwaya, Y., Keon, C.A., Tsuchiya, N., King,
M.T., Radda, G.K., Chance, B., Clarke, K., and
Veech, R.L. (1995). Insulin, ketone bodies, and
225
Chapter 24: Ketogenic Diet, Aging, and Neurodegeneration
mitochondrial energy transduction. FASEB J 9,
651–658.
Schwartzkroin, P.A. (1999). Mechanisms underlying
the anti-epileptic efficacy of the ketogenic diet.
Epilepsy Res 37, 171–180.
Selman, W.R., Ricci, A.J., Crumrine, R.C., LaManna,
J.C., Ratcheson, R.A., and Lust, W.D. (1990). The
evolution of focal ischemic damage:  a metabolic
analysis. Metab Brain Dis 5, 33–44.
Semenza, G.L. (2001). Hypoxia-inducible factor
1:  oxygen homeostasis and disease pathophysiol-
ogy. Trends Mol Med 7, 345–350.
Semenza, G.L. (2007). Hypoxia-inducible factor 1
(HIF-1) pathway. Sci STKE 2007, cm8.
Semenza, G.L. (2011). Regulation of metabolism by
hypoxia-inducible factor 1. Cold Spring Harb
Symp Quant Biol 76, 347–353.
Semenza, G.L., Agani, F., Feldser, D., Iyer, N., Kotch,
L., Laughner, E., and Yu, A. (2000). Hypoxia, HIF-
1, and the pathophysiology of common human
diseases. Adv Exp Med Biol 475, 123–130.
Serra-Perez, A., Planas, A.M., Nunez-O’Mara, A.,
Berra, E., Garcia-Villoria, J., Ribes, A., and
Santalucia, T. (2010). Extended ischemia pre-
vents HIF1alpha degradation at reoxygenation by
impairing prolyl-hydroxylation: role of Krebs cycle
metabolites. J Biol Chem 285, 18217–18224.
Sharp, F.R., Bergeron, M., and Bernaudin, M. (2001).
Hypoxia-inducible factor in brain. Adv Exp Med
Biol 502, 273–291.
Shi, H. (2009). Hypoxia inducible factor 1 as a thera-
peutic target in ischemic stroke. Curr Med Chem
16, 4593–4600.
Shimazu, T., Hirschey, M.D., Newman, J., He,
W., Shirakawa, K., Le, M.N., Grueter, C.A.,
Lim, H., Saunders, L.R., Stevens, R.D., et  al.
(2013). Suppression of oxidative stress by beta-
hydroxybutyrate, an endogenous histone deacety-
lase inhibitor. Science 339, 211–214.
Silverman, W.F., Krum, J.M., Mani, N., and Rosenstein,
J.M. (1999). Vascular, glial and neuronal effects
of vascular endothelial growth factor in mes-
encephalic explant cultures. Neuroscience 90,
1529–1541.
Stafstrom, C.E., and Rho, J.M. (2012). The ketogenic
diet as a treatment paradigm for diverse neuro-
logical disorders. Front Pharmacol 3, 59.
Storkebaum, E., Lambrechts, D., and Carmeliet, P.
(2004). VEGF: once regarded as a specific angio-
genic factor, now implicated in neuroprotection.
Bioessays 26, 943–954.
Tallian, K.B., Nahata, M.C., and Tsao, C.-Y. (1998).
Role of the ketogenic diet in children with intrac-
table seizures. Ann. Pharmacother. 32, 349–385.
Tannahill, G.M., Curtis, A.M., Adamik, J., Palsson-
McDermott, E.M., McGettrick, A.F., Goel, G.,
Frezza, C., Bernard, N.J., Kelly, B., Foley, N.H.,
et al. (2013). Succinate is an inflammatory signal
that induces IL-1beta through HIF-1alpha. Nature
496, 238–242.
225
Vannucci, S.J., and Simpson, I.A.( 2003). Developmental
switch in brain nutrient transporter expression
in the rat. Am J Physiol Endocrinol Metab 285,
E1127–1134.
Veech, R.L. (2004). The therapeutic implications of
ketone bodies:  the effects of ketone bodies in
pathological conditions:  ketosis, ketogenic diet,
redox states, insulin resistance, and mitochondrial
metabolism. Prostaglandins Leukot Essent Fatty
Acids 70, 309–319.
Veech, R.L., Chance, B., Kashiwaya, Y., Lardy, H.A.,
and Cahill, G.F., Jr. (2001). Ketone bodies, poten-
tial therapeutic uses. IUBMB Life 51, 241–247.
Veech, R.L., Valeri, C.R., and Vanitallie, T.B. (2012).
The mitochondrial permeability transition
pore provides a key to the diagnosis and treat-
ment of traumatic brain injury. IUBMB Life 64,
203–207.
Veldhorst, M.A., Westerterp, K.R., van Vught, A.J.,
and Westerterp-Plantenga M.S. (2010). Presence
or absence of carbohydrates and the proportion of
fat in a high-protein diet affect appetite suppres-
sion but not energy expenditure in normal-weight
human subjects fed in energy balance. Br J Nutr
104, 1395–1405.
Willis, M.W., Ketter, T.A., Kimbrell, T.A., George,
M.S., Herscovitch, P., Danielson, A.L., Benson,
B.E., and Post, R.M. (2002). Age, sex and laterality
effects on cerebral glucose metabolism in healthy
adults. Psychiatry Res 114, 23–37.
Xu, K., Puchowicz, M.A., Lust, W.D., and LaManna,
J.C. (2006). Adenosine treatment delays post-
ischemic hippocampal CA1 loss after cardiac
arrest and resuscitation in rats. Brain Res 1071,
208–217.
Xu, K., Sun, X., Puchowicz, M.A., and LaManna, J.C.
(2007). Increased sensitivity to transient global
ischemia in aging rat brain. Adv Exp Med Biol 599,
199–206.
Xu, K., Puchowicz, M.A., Sun, X., and LaManna, J.C.
(2008). Mitochondrial dysfunction in aging rat
brain following transient global ischemia. Adv Exp
Med Biol 614, 379–386.
Xu, K., LaManna, J.C., and Puchowicz, M.A. (2012).
Neuroprotective properties of ketone bodies. Adv
Exp Med Biol 737, 97–102.
Yudkoff, M., Daikhin, Y., Melo, T.M., Nissim, I.,
Sonnewald, U., and Nissim, I. (2007). The keto-
genic diet and brain metabolism of amino
acids:  relationship to the anticonvulsant effect.
Annu Rev Nutr 27, 415–430.
Zhang, Y., Kuang, Y., Xu, K., Harris, D., Lee, Z.,
LaManna, J., and Puchowicz, M.A. (2013). Ketosis
proportionately spares glucose utilization in brain.
J Cereb Blood Flow Metab 33, 1307–1311.
Zhang, Y., Zhang, S., Marin-Valencia, I., and
Puchowicz, M.A. (2015). Decreased carbon shunt-
ing from glucose toward oxidative metabolism in
diet-induced ketotic rat brain. J Neurochem 132,
301–312.
226
226 section III: Ketogenic Diet in the Laboratory
Zhang, Z., Yan, J., Taheri, S., Liu, K.J., and Shi, H.
(2014). Hypoxia-inducible factor 1 contributes to
N-acetylcysteine’s protection in stroke. Free Radic.
Biol. Med. 68, 8–21.
Zhang, Z.G., Zhang, L., Tsang, W., Soltanian-Zadeh,
H., Morris, D., Zhang, R., Goussev, A., Powers,
C., Yeich, T., and Chopp, M. (2002). Correlation
of VEGF and angiopoietin expression with dis-
ruption of blood-brain barrier and angiogenesis
after focal cerebral ischemia. J. Cereb. Blood Flow
Metab 22, 379–392.
Zheng, Z.J., Croft, J.B., Giles, W.H., and Mensah,
G.A. (2001). Sudden cardiac death in the United
States, 1989 to 1998. Circulation. 104, 2158–2163.
227
25
Endocrine and Reproductive Effects
of Ketogenic Diets
J A C O B P. H A R N E Y, P H D , K AT H RY N G U D S N U K , M S , A M I PAT E L , M D ,
A N A N T H A R . V E L L I P U R A M , M D , S AT H YA J I T B A N D A R U, M S ,
A N D D AV I D B U T L E R , P H D
INTRODUCTION
The impacts of maternal nutrition during gesta-
tion on reproductive success have been appreci-
ated for centuries (Wu et al., 2004). The fact that
mothers provide nutrients, oxygen, and the means
to eliminate metabolic waste products from the
developing offspring via the placenta links the ulti-
mate health of both individuals to the resources
available to the mother and her ability to distribute
them effectively. It is becoming increasingly clear
that the metabolic status of the mother impacts
the phenotype and metabolic disease susceptibil-
ity of the offspring (Rando and Simmons, 2015).
The global epidemic of metabolic syndrome has
increased the concern over the effects of parental
nutrition (lifestyle and environment) on offspring
growth, development, cognitive function, and
metabolic health (Hartil et  al., 2009). With both
extremes (starvation and obesity) prevalent across
the world, a better understanding of the impacts
of diet and dietary therapies on health is critical to
having any hope of improving quality of life world-
wide (Fontana and Partridge, 2015).
H I G H - FAT /
K E TO G E N I C   D I E T S
The complex hormonal and metabolic mecha-
nisms that control and are impacted by substrate
availability drive changes in behavior that can be
uncovered through manipulation of diet. A high-fat
or ketogenic diet (KD) has been in use as a treat-
ment for intractable epilepsy in children for close
to 100 years (Vining et al., 1998; Neal et al., 2008;
Stafstrom and Rho, 2012). The neuroprotective
effects of the diet are compelling, and in recent
years the potential therapeutic uses of the diet in
humans (Kossoff et al., 2012; Kossoff and Hartman,
2013)  has been extended through animal studies
alone into thermal pain management (Ruskin et al.,
2013), neural degeneration produced by a variety of
insults/pathologies including cerebral ischemia (Yin
et  al., 2015), traumatic brain injury (Prins, 2008,
2012; Prins and Matsumoto, 2014), amyotrophic
lateral sclerosis (Zhao et al., 2006), metastatic (Poff
et  al., 2015)  and brain (Maroon et  al., 2015)  can-
cers, and in Alzheimers (Brownlow et  al., 2013),
Huntington’s (Ruskin et al., 2011), and Parkinson’s
disease (Cheng et al., 2009) mouse models. In each
case investigators assessed the effectiveness of the
diet, or ketones themselves, on limiting the states
of injury, disorder, or dysfunction. The ketone body
acetone has anticonvulsant effects, and it could play
a role in seizure control. Multiple mechanisms have
been implicated in the neuroprotective effects of
a KD (Schwartzkroin, 1999; Hartman et  al., 2007;
Ruskin and Masino, 2012; Stafstrom and Rho, 2012;
Danial et  al., 2013; Giordan et  al., 2014; Mantis
et al., 2014; Meidenbauer and Roberts, 2014; Rho,
2015; Rogawski et al., 2016) and while some effects
may be temporary and/or reversible there is mount-
ing evidence that long-term use of the diet can
impact gene expression and behavioral phenotype
on a more permanent basis.
ENDOCRINE EFFECTS
O F   H I G H - FAT /
K E TO G E N I C   D I E T S
Regulation of nutrient intake, metabolism, energy
generation, storage, distribution, and utilization is
mediated in an endocrine manner and is depen-
dent on the nature of the foodstuff (carbohydrate,
protein, or fat) and the present energy needs of the
organism. From an evolutionary perspective there
is nothing more important than the consistent
provision of food/energy to allow for growth and
successful reproduction.
228
228 section III: Ketogenic Diet in the Laboratory
Hormones involved in regulation of food
intake and energy expenditure come from the gas-
trointestinal tract (GI), adipose tissue, pancreas,
adrenal gland, and peripheral sensory receptors
and converge on the central nervous system. The
drive to eat comes, in part, from the hypothalamus,
where the “hunger/satiety” center of the brain is
influenced by nutrients, hormones, and other sig-
naling factors. Specific signals convey informa-
tion regarding overall body metabolism, dietary
composition, and the status of carbohydrate and
lipid stores, in muscle and liver (carbohydrate)
and adipose (lipid) tissue, respectively. During
emergencies the sympathetic division of the auto-
nomic nervous system is activated, resulting in
the “fight or flight” response that diminishes any
drive for food intake until the situation is resolved.
Similarly, elevated stress levels result in elevated
glucocorticoids (corticosterone in animals, corti-
sol in humans), which also decrease GI tract activ-
ity and the drive for food intake.
During nonemergencies the parasympathetic
division of the autonomic nervous system pre-
dominates and eating, digestion, absorption, and
transport occur without interruption. The hor-
mones involved include insulin, amylin, glucagon,
and pancreatic peptide (PP) from the endocrine
pancreas; neurotransmitters neuropeptide Y
(NPY), agouti gene-related protein (AGRP), and
pro-opiomelanocortin (POMC) in the CNS; and
ghrelin, peptide YY (PYY), somatostatin, secre-
tin, and cholecystokinin (CCK) from the GI
tract. Normal physiology involves integration of
these signals that converge on the brain and reg-
ulate food intake and energy metabolism (Paoli
et al., 2015).
During and immediately after a meal is the
absorptive stage, when mechanical and chemical
digestion are complete and amino acids, triglyc-
erides, and glucose elevation stimulate insulin
release from the beta cells of the endocrine pan-
creas. Insulin stimulates uptake of glucose and
glycogenesis by liver and muscle cells, uptake and
immediate use of glucose for energy by all cells
except skeletal muscle, uptake of amino acids and
protein synthesis by all tissues, and uptake of lipid
by skeletal muscle for ATP and by adipose tissue
for storage as triglycerides. The postabsorptive
state is the time between meals when glucocorti-
coids generally rise and work in concert with glu-
cagon and epinephrine to tap into energy reserves
that were stored during the absorptive state.
Glycogen and triglycerides are broken down via
glycogenolysis and lipolysis, respectively, increas-
ing circulating levels of glucose and fatty acids
and glycerol. The fatty acids can be converted to
acetyl-CoA, moved into the citric acid cycle, and
used as energy to produce ATP. The use of fatty
acids by systemic tissues reduces the need for
glucose and thus helps to maintain homeostasis.
In order for glucose levels to be sufficient for the
brain, glycogen breakdown, glycerol conversion
to glucose, and gluconeogenesis (production of
glucose from noncarbohydrate sources) by the
liver must be maintained during the postabsorp-
tive stage. In addition to gluconeogenesis, the
liver also carries out ketogenesis (production
of ketone bodies from lipid and protein break-
down), resulting in the generation of acetoacetate,
β-hydroxybutyrate, and acetone. Ketone bodies
produced can be used by the entire body, includ-
ing the brain, for energy. The efficiency of energy
conversion is actually higher for ketones than it
is for glucose (Schwartzkroin, 1999). The produc-
tion of ketone bodies serves as the signal for fast-
ing, or starvation, to the organism, as most diets
in the wild are not “ketogenic.”
Paoli et  al. (2012) published blood levels of
glucose, insulin, ketone bodies, and blood pH (see
Table 25.1) in response to normal diets, KDs, and the
development of diabetic ketoacidosis. The elevation
in blood ketones and the decrease in blood glucose,
at least in the short term, are well established. Of par-
ticular interest is the lower range in concentrations
of insulin in response to KDs versus normal diets.
Dysfunctional insulin signaling, possibly due to
hypoinsulinemia, may be central to a link reported
between type II diabetes and the development of
Alzheimer’s disease (Gotz et al., 2009; Bandaru et al.,
2010). Further investigations into insulin signaling
in the brain are imperative and ongoing.
There are two modes of energy regulation,
short-term, regulated by satiety signals, for
single-meal control, and long-term, regulated by
TABLE 25.1 BLOOD LEVELS DURING A
NORMAL DIET, KETOGENIC DIET, AND
DIABETIC KETOACIDOSIS
Blood levels Normal Ketogenic Diabetic
diet diet ketoacidosis
Glucose (mg/dL) 80–120 65–80 >300
Insulin (μU/L) 6–23 6.6–9.4 ≅0
Ketone bodies 0.1 7–8 >25
(mmol/L)
pH 7.4 7.4 <7.3
Source: Paoli et  al. (2012). Published with permission from author
(open access).
229
Chapter 25: Endocrine and Reproductive Effects of Ketogenic Diets
Ketosis
Orexigenic
AMPK phosphorylation
GABA
ROS production
Circulating Adiponectin
FIGURE 25.1 Schematic diagram demonstrating both the orexigenic and anorexigenic properties of ketosis. Figure
reprinted with permission from author (Paoli et al., 2015; open access).
adiposity signals, which are a reflection of energy
availability in storage, especially body fat. While
the body can only store enough glycogen in liver
and muscle for one day’s worth of energy demand,
it can store lipid capable of supporting weeks of
energy demand. The arcuate nucleus in the hypo-
thalamus responds with signals that either stimu-
late (orexigenic) or inhibit (anorexigenic) food
intake. It is here where the effects of ketones can
be either orexigenic or anorexigenic (Paoli et  al.,
2015; see Figure 25.1), the resultant behavioral
effects being essentially opposite.
Adiponectin (released from adipose tissue),
ghrelin (from the GI tract), and insulin and glu-
cagon (from the pancreas) are all impacted by
ketosis, resulting in mixed endocrine profiles and
resultant behaviors of increased or decreased food
intake. The biochemical make-up of that food will
dictate the metabolic response. Predicting the
endocrine and behavioral effects of KDs is going
to require step-by-step analysis of the impact of
changing both the lipid and protein profiles under
229
Anorexigenic
Circulating FFA
NPY
CCK
or = post prandial glucose
Circulating Ghrelin
AMPK phosphorylation (directly by BHB)
AgRP expression
different physiological scenarios and assessing the
effects on ketosis and other metabolic endocrine
signals (Bielohuby et al., 2011).
Ellenbroek and colleagues (2014) reported
that long-term KD treatment (22 weeks) resulted
in glucose intolerance, reduced beta cell cluster
size and mass (see Figure 25.2), and the estab-
lishment of plasma markers associated with dys-
lipidemia and inflammation in mice (see Table
25.2). Reduced beta cell mass and subsequent
inadequate insulin secretory responses result
in glucose intolerance and type II diabetes in
humans (Kahn et  al., 2006; Nolan et  al., 2011).
The regulatory mechanism was also compro-
mised with respect to glucagon as alpha cell
mass loss was even greater than beta cell mass
loss (Ellenbroek et al., 2014). Glucagon is criti-
cal for counter-regulatory control of blood glu-
cose, and if insufficient the individual is subject
to hypoglycemic episodes resulting in HAAF
(hypoglycemic associated autonomic failure).
This occurs in both type I  and II diabetes and
230

230 section III: Ketogenic Diet in the Laboratory

(a) (b) (c)

2.0

Beta cell mass (mg)

1.5

1.0

0.5

0.0
(d) (e) (f) Control KD
1.5 2000 80
Is let density (Is lets per mm2)

Control
Beta cell cluster size (µm2)

**
* KD

% Beta cell clusters

1500 60
1.0

1000 40

0.5
500 20

0.0 0 0
Control KD Control KD

00

0
0

00

50

00
50

00
–5

–7

10
<2

–1
00

00

>
00
25

50

75
Beta cell cluster size (µm2)

FIGURE 25.2 Beta cell mass in control and KD-fed mice after 22 weeks. (A) representative image of beta cells (brown) in
control mice; (B) representative image of beta cells (brown) from KD-fed mice; (C) beta cell mass (n = 8–10 mice); (D) islet
density (n = 8–10 mice); (E) median beta cell cluster size (n = 8–10 mice); (F) beta cell cluster distribution (n = 8–10 mice).
* p < .05; ** p < .01 vs. control.
Ellenbroek, J.H., van Dijck, L., Töns, H.A., Rabelink, T.J., Carlotti, F., Ballieux, B.E.P.B., de Koning, E. J. P. (2014). Long-term ketogenic diet
causes glucose intolerance and reduced ß- and α-cell mass but no weight loss in mice. Am J Physiol Endocrinol Metab 306, E552–E558.
License # 3824960598813 from AJP.

TABLE 25.2 PLASMA MARKERS OF THE contributes to long-term complications asso-

METAB OLIC SYNDROME IN CONTROL ciated with low energy, fatigue, and potential
for loss of consciousness due to insufficient
MICE AND MICE FED A KETOGENIC DIET
counter-regulatory mechanisms to raise glucose
FOR 22 WEEKS
(Dagogo-Jack, 2015).
Plasma markers Control KD Collectively, results are consistent with the
notion that long-term KD treatment, while
Total cholesterol, mg/dL 92.7 ± 3.1 141.3 ± 9.5b manageable and not life-threatening from a
Triglyceride, mg/dL 42.0 ± 1.5 64.9 ± 6.8b ketoacidotic standpoint, is not ideal for long-
Leptin, ng/mL 0.63 ± 0.37 2.35 ± 0.99a term metabolic health. Enhanced and extended
MCP-1, pg/mL 6.20 ± 0.73 12.65 ± 1.40c inflammatory responses are counter to organis-
IL-1β, pg/mL 1.19 ± 0.31 3.86 ± 1.17a mal well-being and long-term survival. For the
IL-6, pg/mL 3.20 ± 2.11 19.84 ± 8.02a KD to be an effective, long-term therapy, its del-
ALT, U/L 31.3 ± 0.9 80.6 ± 15.8b eterious side effects of dyslipidemia and inflam-
AST, U/L 89.0 ± 10.7 155.2 ± 31.4a mation will have to be addressed. Whether the
Values are means ± SE; n = 8–10 mice. MCP, monocyte chemotactic beneficial effects are mediated through ketone
protein; IL, interleukin; ALT, alanine aminotransferase; AST, aspar- bodies (Kim et  al., 2015), specific fatty acids,
tate aminotransferase. a p < .05, b p < .01, c p < .001.
Source: Ellenbroek, J.H., van Dijck, L., Töns, H.A., Rabelink, T.J., altered glucose, altered insulin and glucagon, or
Carlotti, F., Ballieux, B.E.P.B., de Koning, E. J. P. (2014). Long-term a combination of the above, an understanding of
ketogenic diet causes glucose intolerance and reduced ß- and α-cell the mechanism(s) of action of the various con-
mass but no weight loss in mice. Am J Physiol Endocrinol Metab 306,
E552–E558. License # 3824960598813 from AJP. stituents is necessary.
231
Chapter 25: Endocrine and Reproductive Effects of Ketogenic Diets
B E H AV I O R A L E F F E C T S
O F   H I G H - FAT /
K E TO G E N I C   D I E T S
Initial animal studies investigating the neuropro-
tective effects of KDs on seizure susceptibility in
rodent models were relatively short-term feeding
regimens (3 weeks) followed by quantitative (time
to onset; time of total seizure) and qualitative
(tonic, clonic, or grand mal) assessment of seizure
severity (Hori et al., 1997; Bough and Eagles, 1999;
Su et al., 2000; Harney et al., 2002; Loscher, 2011).
The success of the KD in the treatment of pediatric
epilepsy (Lefevre and Aronson, 2000; Vasconcelos
et al., 2004; Neal et al., 2008; Suo et al., 2013) and
the relative ease with which a dietary therapy can
be administered led to the natural hope that the
KD could impact positively numerous other neu-
rological or metabolic disorders (Kossoff et  al.,
2012; Stafstrom and Rho, 2012). The therapeutic
use of KDs has been primarily limited to children
(Lefevre and Aronson, 2000) but studies indicate
that they may be effective in adult patients suf-
fering from pharmacoresistant epilepsy (Coppola
et al., 2002).
Behaviors reportedly impacted by KDs
include epilepsy, sleep, cognition (Hallbook
et  al., 2012), and motor performance, but not
cognition, in two mouse models of Alzheimer’s
pathology (Brownlow et  al., 2013). In addition,
high-fat (nonketogenic) diets alone have report-
edly impacted hippocampal-dependent spatial
learning and memory in rats (Alzoubi et  al.,
2009)  that was exacerbated further in combi-
nation with chronic stress. Maternal high-fat
diets (HFD) themselves can act as a stressor,
increasing maternal glucocorticoids and dis-
rupting maternal behavior and brain activa-
tion (measured as increased c-Fos) in C57BL/6J
mice (Bellisario et al., 2015). Changes in mouse
brain gene expression in response to both keto-
genic and nonketogenic HFDs (Selfridge et  al.,
2015) suggest that the specific lipid components
of the diet likely dictate the ultimate biological
effects, including alterations in the brain’s aero-
bic infrastructure and mitochondrial DNA tran-
scriptional efficiency. Greenwood and Winocur
(1996) reported that the degree of cognitive
impairment is highly associated with the level of
saturated fatty acids fed and independent of the
mono- or polyunsaturated fatty acids in the HFD.
Davidson and coworkers (2013) reported that
cognitive impairment induced by western diets
(high in saturated fats and sugar) may be dimin-
ished by ketones. Collectively, present results
suggest that specific dietary components and not
simply the level of ketone production, ultimately
231
determine the levels and changes in gene expres-
sion that result in behavioral responses of the
organisms.
Central to the development of type II diabe-
tes is the ineffectiveness of the hormone insulin
to lower blood glucose by stimulating uptake in
peripheral tissues. The brain has been considered
to be insulin independent for glucose uptake, how-
ever, there are insulin receptors located in several
brain regions including the hippocampus, hypo-
thalamus, and cerebellum (Kahn, 1994; Ho et al.,
2004, Plum et al., 2005; Banks and Kastin, 1998).
Over the past decade a potential link between
type II diabetes and Alzheimer’s disease has been
established and a dysfunctional insulin signaling
pathway has been proposed to be involved (Gotz
et  al., 2009; Bandaru et  al., 2010). The cognitive
decline reported with long-term KD treatment in
this study and others (Su et al., 2000; Zhao et al.,
2004; Snead 2004; Behassan and Jan, 2006) might
be the result of dysfunctional insulin signaling, a
likely outcome of long-term KD treatment and its
metabolic side effects.
Despite the potential metabolic side effects,
individuals may choose any available means to
experience relief from their primary ailment, espe-
cially epileptic seizures that can occur at any time.
As such, it is imperative that the impacts of KDs
on normal physiology, including reproduction, be
investigated. The timing and duration of exposure
to dietary treatments will likely dictate whether
effects are temporal or more long-lasting (epigen-
etic modifications), as would be expected during
embryonic, fetal, or neonatal development (Rando
and Simmons, 2015).
EFFECTS OF LONG-
T E R M K E TO G E N I C D I E T
CONSUMPTION ON PUBERTY
AND REPRODUCTION
If children are going to be consuming KDs grow-
ing up, what is the impact going to be on normal
development? Our laboratory has begun to inves-
tigate the effects of a KD on growth, onset of nor-
mal female development including day of vaginal
opening, onset of nonnormal cycles, and onset of
normal 4-day cycles in rodents. Three KDs con-
taining 8%, 14%, or 18% protein, with ketogenic
ratios (fat/carb + protein) of 6.2, 4.2, and 3.2,
respectively (Bio-Serv, NJ), or rodent chow (RC)
were administered to 48 female Long-Evans rats
starting on Day 23. Dates of vaginal opening were
recorded, and only the KD 8% group was signifi-
cantly delayed (see Figure 25.3), this effect likely
the result of the more shallow growth curve (see
Figure 25.4).
232

*
39

37

Day of vaginal opening

35

33

31

29

27

25
Control KD8% KD14% KD18%
Dietary groups

FIGURE 25.3 The average date of vaginal opening in female Long-Evans rats maintained on rodent chow (control),
KD8%, KD14%, or KD18%. The KD8% were significantly delayed compared with controls (*; p < .05). Statistical analyses
included one-way ANOVA and post hoc Fisher’s least significant difference (LSD).

Average Weight of Rats

500
450
400
350
Weight (g)

300 Control
250
8% KD
200
150 14% KD
100 18% KD
50
0
22 33 43 53 63 73 83 93 120 180
Measurements by Day

FIGURE  25.4 The average weights of female Long-Evans rats fed rodent chow (control) or KD8%, KD14%, or
KD18% from weaning (day 22) to 180 days. KD8% were significantly lower (p < .05) at days 43 and 73, but not 93.
Statistical analyses included one-way ANOVA and post hoc Fisher’s least significant difference (LSD).

TABLE 25.3 AVERAGE AGES AND WEIGHTS OF CONTROL RODENT CHOW- AND

KD- FED FEMALE LONG- EVANS RATS AT THE FIRST REGULAR 4- DAY ESTROUS CYCLE
AND ON COMPLETION OF THREE CONSECUTIVE 4- DAY ESTROUS CYCLES

Groups n Age of first regu- Weight at first n Age after three Mean Weight on
lar 4-day cycle regular cycle (g) consecutive 4-day repeatable cycling (g)
(days) estrous cycles (days)

Control 12 45.1 ± 2.4 219.6 ± 15.3 12 65.2 ± 5.4 249.5 ± 18.6

KD8% 11 51.4 ± 6.8 123 ± 18.6 9 81.7 ± 5.7 272.3 ± 22.6
KD14% 12 46.8 ± 2.9 168.4 ± 18.9 11 65.2 ± 4.5 256.3 ± 17.4
KD18% 12 45.6 ± 3.1 172.3 ± 24.1 12 64.7 ± 3.7 262.7 ± 10.8
233

Chapter 25: Endocrine and Reproductive Effects of Ketogenic Diets 233

140

120

Concentration of glucose (mg/dL)

100

80

60

40

20

0
Control KD8% KD14% KD18%
Dietary groups

FIGURE 25.5 Mean concentrations of blood glucose in control rodent chow- and KD-fed Long-Evans female rats
at day 70 (postpubertal), 43 days after onset of dietary treatments. No significant differences were detected. Statistical
analyses included one-way ANOVA and post hoc Fisher’s least significant difference (LSD).

The time till completion of three consecutive
4-day (normal) estrous cycles was similarly only
significantly delayed in the KD8% group (see
Table 25.3), once again likely delayed in concert
with the delayed growth. The KD8% rats were the
lightest of the groups at first regular estrous cycle
despite the additional 16 days of eating.
Blood glucose levels were not significantly dif-
ferent between any of the groups after 47 days on
the diet (see Figure 25.5). This result is consistent
with long-term KD treatment in our laboratory.
Blood ketones were significantly different, reflect-
ing the elevated ketogenic ratio of the diets as total
protein plus carbohydrates drops and total fat
increases (see Figure 25.6).
With the establishment of multiple KD groups
(varying in circulating ketones), investigations
continued into effects of maternal KD administra-
tion prior to and throughout gestation.
Pregnancy establishment and maintenance
requires signaling between the developing con-
ceptus (embryo and extraembryonic membranes)

*
2.5
Concentration of ketones (mM)

2 * *

1.5

0.5

0
Control KD8% KD14% KD18%
Dietary groups

FIGURE 25.6 Mean concentrations of blood β-hydroxybutyrate (mM) in control rodent chow- and KD-fed Long-
Evans female rats at day 70 (postpubertal), 43 days after onset of dietary treatments. All KD groups had significantly
elevated ketones (*; p < .01) compared with rodent chow-fed controls, but were not significantly different from each
other. Statistical analyses included one-way ANOVA and post hoc Fisher’s least significant difference (LSD).
234
234 section III: Ketogenic Diet in the Laboratory
and the mother. Prior to development of a func-
tional placenta, commonly referred to as the
peri-implantation period, the conceptus relies on
uterine secretions or histotroph for support (Bazer
et al., 2013; Harney et al., 1990; Harney et al., 1993;
Harney et  al., 1994). The components of histo-
troph can be impacted by the maternal diet (Wu
et  al., 2004)  as supplementation of the diet with
the amino acid arginine enhances fetal-placental
development in rodents, swine, and humans
(Bazer at al., 2013; Bazer et al., 2015). Arginine is
a nutritionally essential amino acid for embryonic
survival and fetal and neonatal growth (Wu et al.,
2009). The hormonal milieu (levels of estrogen
and progesterone) combined with diet and over-
all metabolic and energy status of the mother will
play a key role in establishing the components of
uterine histotroph and the substrates available to
the developing conceptus. Continued success-
ful development requires a functional placenta,
delivery of necessary nutrients and oxygen, and
removal of metabolic waste and carbon dioxide
(Bazer et al., 2015).
Just as the quality of the environment impacts
health in adults, the quality of the intrauterine
(IU) and early postnatal environment impacts
the development of offspring. An adverse IU can
result in intrauterine growth restriction (IUGR).
The health of offspring later in life is impacted
by the occurrence of IUGR, or complications in
the mother, including diabetes or obesity, that
affect nutrient availability (Hales and Barker,
1992, 2001; Ravelli et al., 1976; Ravelli et al., 1998;
Valdez et al., 1994). Models used to examine IUGR
include dietary calorie or protein restriction, glu-
cocorticoid administration, and uteroplacental
insufficiency induced by ligation of the uterine
artery. All conditions induced a stressor into the
maternal system during gestation and examined
an impact on offspring glucose metabolism into
adulthood (Fowden and Forhead, 2004; McMillen
and Robinson, 2005). Any nutritional deficiency
will impact energy availability, metabolism, and
biochemical substrate availability. In studies where
nutritional overabundance is the standard (obe-
sity and type 2 diabetes), results show significant
increases in establishment of the same chronic
diseases in offspring (Rich-Edwards et  al., 1999).
Collectively, these results suggest that the IU
environment is likely impacted by any significant
changes in maternal nutrition or behavior (espe-
cially stress-induced), which, in turn, can have
any number of phenotypic effects on the offspring.
The mechanism(s) by which maternal diet influ-
ences the phenotype of offspring have yet to be
elucidated; however, epigenetic (external modi-
fications to DNA) changes appear to be involved
(Masuyama and Hiramatsu, 2012; Rando and
Simmons, 2015).
EFFECTS OF LONG-TERM
K E TO G E N I C D I E T
CONSUMPTION BEFORE AND
D U R I N G   G E S TAT I O N
There have been limited studies focused on KD
administration throughout gestation and its effects
on maternal and offspring-phenotypes; however
findings suggest, as we have always appreciated,
that maternal diet does indeed impact pregnancy
outcome. Sussman and coworkers examined the
effects of KD administration during pregnancy
on embryonic growth (Sussman et al., 2013b),
offspring physiological growth and brain struc-
ture (Sussman et al., 2013a), and brain structure
and susceptibility to depression and anxiety in
the adult mouse offspring (Sussman et al., 2015).
Dams fed a KD (67% fat; 15% protein) prior to
and during gestation exhibited a significant reduc-
tion in maternal fertility and litter size, which the
authors indicated could be due to protein defi-
ciency. However, others have reported contrary
results (Derrickson, 2007), and the 15% protein
should have been sufficient for normal growth. In
addition, since the dams were prone to fatal keto-
acidosis by midlactation (Sussman et al., 2013a),
pups were fostered by dams on standard chow,
allowing examination of the effects of KD expo-
sure during gestation alone. The results from their
studies in mice indicate that KD exposure during
all of gestation and through weaning is not condu-
cive to normal anatomical, behavioral, and physi-
ological development of the offspring. They report
delayed growth and significant relative decreases
bilaterally in the cortex, fimbria, hippocampus,
corpus callosum, and lateral ventricle. The hypo-
thalamus and medulla experienced volumetric
enlargements as well. Differences were detected
as early as postnatal day 11.5 and were greater at
postnatal day 21.5 (Sussman et al., 2013a). In an
earlier report, Sussman and colleagues (2013b)
demonstrated that anatomical comparisons of
embryos at day 13.5 detected larger volumetric
size of embryos and their hearts from dams on
a KD but smaller brains, pharynx, cervical spi-
nal cord, hypothalamus, midbrain, and pons.
By day 17.5 embryos from KD dams were volu-
metrically smaller with smaller hearts and thy-
mus glands but enlarged cervical spine, thalamus,
midbrain, and pons (Sussman et al., 2013b). The
most recent study by Sussman’s group examines
235

Chapter 25: Endocrine and Reproductive Effects of Ketogenic Diets 235

Group 1 Group 2 Group 3 Group 4

Control 8% KD 14% KD 18% KD
Mothers Mothers Mothers Mothers
(n = 12) (n = 12) (n = 12) (n = 12)

29 45 65 67
offspring offspring offspring offspring

Group 1: Group 2: Group 3: Group 4:

12 female 12 female 12 female 12 female
offspring offspring offspring offspring

FIGURE  25.7 Treatment groups of Long-Evans dams maintained on control rodent chow or KD8%, KD14%, or
KD18% from weaning (day 22) through puberty, breeding, gestation, and out to 180 days postpartum and postweaning
of their pups. Twelve female offspring per group were isolated and maintained on rodent chow till day 70, when MWM
was performed.

both brain structure and susceptibility to depres-
sion and anxiety in adult mouse offspring from
KD-fed dams throughout gestation. Eight-week-
old adult CD-1 mice exhibited reduced suscep-
tibility to anxiety and depression and increased
physical activity compared with standard chow
controls. Our studies in adult rats on KDs have
generated similar findings with decreased depres-
sion appearing as early as 14 days after onset of
treatment and continuing through 6 months (A.
Patel, unpublished results). In addition, recent
evidence suggests that maternal high-fat diets
might make offspring susceptible to behav-
ioral disorder responses to stressful challenges.
Collectively these results suggest that KD exposure
throughout gestation and weaning has significant
anatomical and developmental implications on
the offspring. Likely, changes in gene expression,
which may be mediated epigenetically (Masuyama
and Hiramatsu, 2012; Rando and Simmons, 2015),
ultimately impact anatomy and the resultant phys-
iology and behavior into adulthood.
Our lab has conducted experiments in Long
Evans rats (see Figure 25.7; Gudsnuk unpub-
lished), where females were administered rodent
chow (RC) or KDs with 8%, 14%, or 18% protein
from day of weaning (day 21)  through puberty,
breeding, gestation, and weaning of their pups.
Biweekly weights and blood samples for glucose
and β-hydroxybutyrate were collected. Dams were
tested for spatial memory by the Morris water
maze (MWM) at day 180 postpartum, and all

TABLE 25.4 EFFECTS OF DIETS ON PREGNANCY, LIT TER SIZE, PUPS DELIVERED,

AND PUPS WEANED

n % Pregnant % of Pregnant # Rats died Average Total live Total pups % of Pups
Rats rats to have during labor litter size pups weaned weaned
pups

Control 6 83.3 83.3 0 10.2 51 29 56.8

KD8% 6 100 100 1 8 48 45 93.7
KD14% 6 100 100 0 12.7 76 65 83.5
KD18% 6 100 100 0 12.5 75 67 89.3
236
236 section III: Ketogenic Diet in the Laboratory
80
70
Time in correct quadrant (s)
60
50
40
30
20
10
0
Control pups
FIGURE  25.8 Performance of female Long-Evans pups during diestrous (day 70–72), 47 to 49 days postweaning
from control rodent chow- or KD-fed dams. Offspring of the KD-fed dams spent less time (s) in the correct quadrant
of the water maze (where the hidden platform is located) than controls. KD8% again performed significantly worse
(p < .02) than the controls. The KD14% and KD18% also performed better than the KD8%, but not as well as the
controls. Statistical analyses included one-way ANOVA and post hoc Fisher’s least significant difference (LSD).
female pups were weaned at day 21, put on stan-
dard RC, and maintained till day 70, when their
spatial memory was assessed by the MWM.
KD dams showed no significant difference
in blood glucose at day 70 (see Figure 25.5) but
had significantly elevated β-hydroxybutyrate (see
Figure 25.6), prior to breeding compared with
controls.
Two breeding trials were conducted (n =
6/group/trial) and the results from the second
trial are presented above. The KD rats all suc-
cessfully produced litters and actually weaned
a greater percentage of pups than the controls
(Table 25.4). Average litter sizes were smaller for
the KD8% than the other groups. This was not
surprising, as the low protein delayed onset of
puberty and likely has other deleterious effects
on the gonads.
The results from the MWM of the pups of KD-
or RC-fed dams tend to follow the reverse rela-
tionship, where elevated ketones are coincident
with less time spent in the correct quadrant and
thus worse spatial memory. The KD8% group per-
formed significantly worse (p < .02) than the RC-
fed controls. The KD14% and KD18% tended to
perform better than the KD8%, but not as well as
the RC-fed controls (see Figure 25.8). The KD-fed
dams at 180  days postpartum tended to perform
worse than the RC-fed controls (data not shown).
Although just a few studies have been con-
ducted in mice and rats, the results to date sug-
gest that, while effective for some neurological
disorders, possibly in both children and adults,
*
KD8% pups KD14% pups KD18% pups
Dietary groups
the long-term use of KDs by women during ges-
tation may result in diminished cognitive func-
tion and other phenotypic effects in the offspring
(Hartil et  al., 2009). Bellisario and coworkers
(2015) reported changes in dams’ behavior includ-
ing increased aggression and decreased locomo-
tor activity and decreased enzymatic activity
(11β-dehydrogenase-2) in the placenta, which is
designed to protect the fetus from maternal glu-
cocorticoids. Increased exposure to glucocorti-
coids may have detrimental consequences to the
developing fetus.
S U M M A RY
The increased interest in KDs is not surprising,
as the accumulated data for the successful treat-
ment of epileptic seizures, especially, makes giv-
ing it at least a try a “no-brainer.” The known
potential side effects of KDs, including hypogly-
cemia and its associated symptoms, especially in
the short term; elevated lipids and cholesterol;
and an elevated risk of kidney stones, may be
tolerable to patients whose options are limited
(Schwartzkroin, 1999).
Dietary changes, especially radical ones,
bring about metabolic changes throughout the
body as the substrates for reactions change and
the by-products of those metabolic changes
impact homeostasis. It is certainly clear that
HFDs and KDs can vary in percentage of lipid,
source of lipid, and saturation state of lipid, thus
inducing potential orexigenic or anorexigenic
effects that will ultimately change the behavior
237
Chapter 25: Endocrine and Reproductive Effects of Ketogenic Diets
of the organism (Paoli et  al., 2015). The possi-
bility of deleterious consequences to offspring’s
metabolic regulatory system as well as cognitive
function means it is imperative that investiga-
tions continue into the specific effects of HFD,
both ketogenic and nonketogenic to uncover the
components (saturated and nonsaturated fatty
acids, protein content and source) necessary for
safe therapies in relation to the endocrinology,
development, and neuroendocrinology of ani-
mals and ultimately humans.
AC K N OW L E G M E N T S
The authors thank Dr.  William Neace for assis-
tance with statistical analyses. This research was
supported by the Neuroscience Graduate Program
and the College of Arts and Sciences Dean’s
Research Fund at the University of Hartford.
REFERENCES
Alzoubi, K.H., Abdul-Razzak, K.K., Khabour, O.F.,
Al-Tuweiq, G.M., Alzubi, M.A., and Alkadhi, K.A.
(2009). Adverse effect of combination of
chronic psychosocial stress and high fat diet on
hippocampus-dependent memory in rats. Behav
Brain Res 204, 117–123.
Bandaru, S.S., Lin, K., Roming, S., Vellipuram, R.,
and Harney, J.P. (2010). Effects of PI3K inhibition
and low docosahexaenoic acid on cognition and
behavior. Physiol Behav 100, 239–244.
Banks, W.A., and Kastin, A.J. (1998). Differential per-
meability of the blood-brain barrier to two pan-
creatic peptides: insulin and amylin. Peptides 19,
883–889.
Bazer, F.W., Johnson, G.A., and Wu, G. (2015). Amino
acids and conceptus development during the peri-
implantation period of pregnancy. Adv Exper Med
Biol 843, 23–52.
Bazer, F.W., Kim, J.Y., Song, G.W., Ka, H., Wu, G., Johnson,
G.A., and Vallet, J.L. (2013). Roles of selected nutri-
ents on development of the conceptus during preg-
nancy. Soc Reprod Fertil Suppl 68, 159–174.
Behassan, N.A., and Jan, M.M. (2006). Ketogenic Diet
Update and Application. Neurosciences (Riyadh)
11, 235–240.
Bellisario, V., Panetta, P., Balsevich, G., Baumann, V.
Noble, J., Raggi, C., Nathan, O., Berry, A., Seckl, J.,
Schmidt, M., Holmes, M., et al. (2015). Maternal
high-fat diet acts as a stressor increasing maternal
glucocorticoids’ signaling to the fetus and disrupt-
ing maternal behavior and brain activation in
C57BL/6J mice. Psychoneuroendo 60, 138–150.
Bielohuby, M., Menhofer, D., Kirchner, H., Stoehr, B.J.,
Muller, T.D., Stock, P., Hempel, M., Stemmer, K.,
Pfluger, P.T., Kienzle, E., et  al. (2011). Induction
of ketosis in rats fed low-carbohydrate, high-fat
diets depends on the relative abundance of dietary
237
fat and protein. Am J Physiol Endocrinol Metab
300, 65–76.
Bough, K.J., and Eagles, D.A. (1999). A ketogenic diet
increases the resistance to pentylenetetrazole-
induced seizures in the rat. Epilepsia 40, 138–143.
Brownlow, M.L., Benner, L., D’Agostino, D., Gordon,
M.N., and Morgan, D. (2013). Ketogenic diet
improves motor performance but not cognition in
two mouse models of Alzheimer’s pathology. PLoS
One 8(9), e75713.
Cheng, B., Yang, X., An, L., Gao, B., Liu, X., and Liu, S.
(2009). Ketogenic diet protects dopaminergic
neurons against 6-OHDA neurotoxicity via up-
regulating glutathione in a rat model of Parkinson’s
disease. Brain Res 1286, 25-31.
Coppola, G., Veggiotti, P., Cusmai, R., Bertoli, S.,
Cardinali, S., and Dionisi-Vici, C. (2002). The
ketogenic diet in children, adolescents and young
adults with refractory epilepsy:  an Italian multi-
centric experience. Epilepsy Res 48, 221–227.
Dagogo-Jack, S. (2015). Philip E. Cryer, MD: seminal
contributions to the understanding of hypoglyce-
mia and glucose counter-regulation and the dis-
covery of HAAF (Cryer syndrome). Diabetes Care
38, 2193–2199.
Danial, N.N., Hartman, A.L., Stafstrom, C.E., and
Thio, L.L. (2013). How does the ketogenic diet
work? Four potential mechanisms. J Child Neurol
28(8), 1027–1033.
Davidson, T.E., Hargrave, S.L., Swithers, S.E., Sample,
C.H., Fu, X., Kinzig, K.P., and Zheng, W. (2013).
Inter-relationships amount diet, obesity and
hippocampal-dependent cognitive function.
Neuroscience 253,110–122.
Derrickson, E.M. (2007). The effects of dietary protein
levels on milk protein levels and postnatal growth
in laboratory mice (Mus Musculus). J Mammal 88,
1475–1481.
Ellenbroek, J.H., van Dijck, L., Töns, H.A., Rabelink,
T.J., Carlotti, F., Ballieux, B.E.P.B., and de Koning,
E. J.  P. (2014). Long-term ketogenic diet causes
glucose intolerance and reduced ß- and α-cell
mass but no weight loss in mice. Am J Physiol
Endocrinol Metab 306, E552–E558.
Fontana, L., and Partridge, L. (2015). Promoting health
and longevity through diet: from model organisms
to humans. Cell Press 161(1), 106–118.
Fowden, A.L., and Forhead, A.J. (2004). Endocrine
mechanisms of intrauterine programming,
Reproduction 127, 515–526.
Giordan, C., Marchio, M., Timofeeva, E., and Biagini,
G. (2014). Neuroactive peptides as putative medi-
ators of antiepileptic ketogenic diets. Frontiers
Neurol 5, 63, 1–14.
Gotz, J., Ittner, L.M., and Lim, Y.A. (2009). Common
features between diabetes mellitus and Alzheimer’s
disease. Cell Mol Life Sci 66, 1321–1325.
Greenwood, C.E., and Winocur, G. (1996). Cognitive
impairment in rats fed high-fat diets:  A  specific
238
238 section III: Ketogenic Diet in the Laboratory
effect of saturated fatty-acid intake. Behavioral
Neurosci 110, 451–459.
Hales, C.N., and Barker, D.J. (1992). Type 2 (non-
insulin-dependent) diabetes mellitus; the thrifty
phenotype hypothesis. Diabetologia 35, 595–601.
Hales, C.N., and Barker, D.J. (2001). The thrifty pheno-
type hypothesis. Br Med Bull 60, 5–20.
Hallböök, T., Ji, S., Maudsley, S., Martin, B. (2012). The
effects of the ketogenic diet on behavior and cogni-
tion. Epilepsy Res 100:304–309.
Harney, J.P., Ali, M., Vedeckis, W.V., and Bazer, F.W.
(1994). Porcine conceptus and retinoid-binding
proteins. Reprod Fertil Dev 6, 211–219.
Harney, J.P., Madara, J., Madara, J., and I’Anson, H.
(2002). Effects of acute inhibition of fatty acid oxi-
dation on latency to seizure and concentrations of
β hydroxybutyrate in plasma of rats maintained
on calorie restriction and/or the ketogenic diet.
Epilepsy Res 49, 239–246.
Harney, J.P., Mirando, M.A., Smith, L.C., and Bazer,
F.W. (1990). Retinol-binding protein:  a major
secretory product of the pig conceptus. Biol
Reprod 42, 523–532.
Harney, J.P., Ott, T.L., Geisert, R.D., and Bazer, F.W.
(1993). Retinol-binding protein gene expression
in cyclic and pregnant endometrium of pigs, sheep
and cattle. Biol Reprod 49, 1066–1073.
Hartil, K., Vuguin, P.M., Kruse, M., Schmuel, E.,
Fiallo, A., Vargas, C., Warner, M.J., Duran, J.L.,
Jelicks, L.A., and Charron, M.J. (2009). Adverse
effect of combination of chronic psychoso-
cial stress and high fat diet on hippocampus-
dependent memory in rats. Pediatr Res 66,
368–373.
Hartman, A.L., Gasior, M., Vining, E.P., and Rogawski,
M.A. (2007). The neuropharmacology of the keto-
genic diet. Pediatr Neurol 36, 281–292.
Ho, L., Qin, W., Pompl, P.N., Xiang, Z., Wang, J., Zhao,
Z, Peng, Y., Cambareri, G., Rocher, A., Mobbs,
C.V., et  al. Diet-induced insulin resistance pro-
motes amyloidosis in a transgenic mouse model of
Alzheimer’s disease. FASEB J. 18:902–904.
Hori, A., Tandon, P., Holmes, G.L., and Stafstrom,
C.E. (1997). Ketogenic diet: effects on expression
of kindled seizures and behavior in adult rats.
Epilepsia 38, 750–758.
Kahn, C. (1994). Insulin action, diabetogenesis,
and the cause of type II diabetes. Diabetes 43,
1066–1084.
Kahn, S.E., Hull, R.L., and Utzschneider, K.M. (2006).
Mechanisms linking obesity to insulin resistance
and type 2 diabetes. Nature 444, 840–846.
Kim, D.Y., Abdelwahab, M.G., Lee, S.H., O’Neill, D.,
Thompson, R.J., Duff, H.J., Sullivan, P.G., and Rho,
J.M. (2015). Ketones prevent oxidative impairment
of Hippocampal Synaptic Integrity through KATP
channels. PLoS One 10, e0119316.
Kossoff, E.H., Caraballo, R.H., du Toit, T., Kim, H.D.,
MacKay, M.T., Nathan, J.K., and Philip, S.G.
(2012). Dietary therapies:  a worldwide phenom-
enon. Epilepsy Res 100, 205–209.
Kossoff, E.H., and Hartman, A.L. (2013). Ketogenic
diets:  new advances for metabolism-based thera-
pies. Curr Opin Neurol 25. 173–178.
Lefevre, F., and Aronson, N. (2000). Ketogenic diet
for the treatment of refractory epilepsy in chil-
dren:  a systematic review of efficacy. Pediatrics
105, E46.
Lin, C., Shao, B., Huang, H., Zhou, Y., and Lin, Y.
(2015). Maternal high fat diet programs stress-
induced behavioral disorder in adult offspring.
Physiol Behav 152, 119–127.
Loscher, W. (2011). Critical review of current animal
models of seizures and epilepsy used in the discov-
ery and development of new antiepileptic drugs.
Seizure 20, 359–368.
Mantis, J.G., Meidenbauer, J.J., Zimick, N.C.,
Centeno, N.A., and Seyfried, T.N. (2014).
Glucose reduces the anticonvulsant effects of
the ketogenic diet in EL mice. Epilepsy Res 108,
1137–1144.
Maroon, J.C., Seyfried, T.N., Donohue, J.P., and
Bost, J. (2015). The role of metabolic therapy in
treating glioblastoma multiforme. Surg Neurol
Int 6, 61.
Masuyama, H., and Hiramatsu, Y. (2012). Effects of
a high-fat diet exposure in utero on the meta-
bolic syndrome-like phenomenon in mouse
offspring through epigenetic changes in adipo-
cytokine gene expression. Endocrinology 153,
2823–2830.
McMillen, I.C., and Robinson, J.S. (2005).
Developmental origins of the metabolic syn-
drome:  prediction, plasticity, and programming.
Physiol Rev 85, 571–633.
Meidenbauer, J.J., and Roberts, M.F. (2014). Reduced
glucose utilization underlies seizure protection
with dietary therapy in epileptic EL mice. Epilepsy
Behav 39, 48–54.
Neal, E.G., Chaffe, H., Schwartz, R.H., Lawson,
M.S., Edwards, N., Fitzsimmons, G., Whitney,
A., and Cross, J.H. (2008). The ketogenic
diet for the treatment of childhood epilepsy:  a
randomized controlled trial. Lancet Neurol 7,
500–506.
Nolan, C.J., Damm, P., and Prentki, M. (2011). Type
2 diabetes across generations: from pathophysiol-
ogy to prevention and management. Lancet 387,
169–181.
Paoli, A., Grimaldi, K., Toniolo, L., Canato, M.,
Bianco, A., and Fratter, A. (2012). Nutrition and
acne: therapeutic potential of ketogenic diets. Skin
Pharmacol Physiol 25, 111–117. doi:  10.1159/
000336404
239
Chapter 25: Endocrine and Reproductive Effects of Ketogenic Diets
Paoli, A., Bosco, G., Camporesi, E.M., and Mangar,
D. (2015). Ketosis, ketogenic diet and food intake
control:  a complex relationship. Frontiers in
Psychol 6, 27. doi: 10.3389/fpsyg.2015.00027
Plum, L., Schubert, M., and Bruning, J.C. (2005). The
role of insulin receptor signaling in the brain.
Trends Endocrinol Metab 16, 59–65.
Poff, A., M., Ward, N., Seyfried, T.N., Arnold, P.,
and D’Agostino, D.P. (2015). Non-toxic meta-
bolic management of metastatic cancer in VM
mice: novel combination of ketogenic diet, ketone
supplementation, and hyperbaric oxygen therapy.
PLoS One 10, 6.
Prins, M.L. (2012). Cerebral ketone metabolism dur-
ing development and injury. Epilepsy Res 100,
218–223.
Prins, M.L. (2008). Cerebral metabolic adaptation
and ketone metabolism after brain injury. J Cereb
Blood Flow Metab 28, 1–16.
Prins, M.L., and Matsumoto, J.H. (2014). The col-
lective therapeutic potential of cerebral ketone
metabolism in traumatic brain injury. J Lipid Res
55, 2450–2457.
Rando, O.J., and Simmons, R.A. (2015). I’m eating for
two: parental dietary effects on offspring metabo-
lism. Cell 161, 93–105.
Ravelli, G.P., Stein, Z.A., and Susser, M.W. (1976).
Obesity in young men after famine exposure
in utero and early infancy. N Engl J Med 295,
349–353.
Ravelli, A.C., van der Meulen, J.H., Michels, R.P.,
Osmond, C., Barker, D.J., Hales, C.N., and
Bleker, O.P. (1998). Glucose tolerance in adults
after prenatal exposure to famine. Lancet 351,
173–177.
Rho, J.M. (2015). How does the ketogenic diet induce
anti-epileptic effects?. Neurosci Lett 7, 30.
Rich-Edwards, J.W., Colditz, G.A., Stampfer, M.J.,
Willett, W.C., Gillman, M.W., Hennekens, C.H.,
Speizer, F.E., and Manson, J.E. (1999). Birthweight
and the risk for type 2 diabetes mellitus in adult
women. Ann Intern Med 130, 278–284.
Rogawski, M.A., Loscher, W., and Rho, J.M. (2016).
Mechanisms of action of anti-seizure drugs and the
ketogenic diet. Cold Spring Harb Perspect Med,
May 2, 6(5) 397–424. pii:  a022780. doi:  10.1101/
cshperspect.a022780
Ruskin, D.N., and Masino, S.A. (2012). The nervous
system and metabolic dysregulation:  emerging
evidence converges on ketogenic diet therapy.
Frontiers in Neurosci 6(33), 1–12.
Ruskin, D.N., Ross, J.L., Kawamura Jr, M., Ruiz, T.L.,
Geiger, J.D., and Masino, S.A. (2011). A keto-
genic diet delays weight loss and does not impair
working memory or motor function in the R6/2
1J mouse model of Huntington’s disease. Physiol
Behav 103, 501–507.
239
Ruskin, D.N., Suter, T.A., Ross, J.L., and Masino, S.A.
(2013). Ketogenic diets and thermal pain: dissoci-
ation of hypoalgesia, elevated ketones, and lowered
glucose in rats. J Pain 14(5):467–474.
Schwartzkroin, P.A. (1999). Mechanisms underlying
the antiepileptic efficacy of the ketogenic diet.
Epilepsy Res 37, 171–180.
Selfridge, J.E., Wilkins, H.M., E.L., Carl, S.M., Koppel,
S., Funk, E., Fields, T., Lu, J., Tang, E.P., Slawson, C.,
et al. (2015). Effects of one month duration keto-
genic and non-ketogenic high fat diets on mouse
brain bio-energetic infrastructure. J Bioenerg
Biomembr 47, 1–11.
Snead, C.O. (2004). The ketogenic diet:  a cautionary
note. Pediatric Res 55, 368–369.
Stafstrom, C.E., and Rho, J.M. (2012). The ketogenic
diet as a treatment paradigm for diverse neurolog-
ical disorders. Frontiers in Pharmacology 3,1–8.
Su, S.W., Cilio, M.R., Sogawa, Y., Silveira, D.C.,
Holmes, G.L., and Stafstrom, C.E. (2000). Timing
of ketogenic diet initiation in an experimental epi-
lepsy model. Dev Brain Res 125, 131–138.
Suo, C., Liao, J., Lu, X., Fang, K., Hu, Y., Chen, L., Cao,
D., Huang, T., Li, B., and Li, C. (2013). Efficacy and
safety of the ketogenic diet in Chinese children.
Seizure 22, 174–178.
Sussman, D., Ellegood, J., and Henkelman, M. (2013a).
A gestational ketogenic diet  alters maternal met-
abolic status as well as offspring physiological
growth and brain structure in the neonatal mouse.
BMC Pregnancy Childbirth 13,1–10.
Sussman, D., van Eede, M., Wong, M.D., Adamson,
S.L., Henkelman, M. (2013b). Effects of a ketogenic
diet during pregnancy on embryonic growth in the
mouse. Pregnancy and Childbirth 13, 109–120.
Sussman, D., Germann, J., and Henkelman, M. (2015).
Gestational ketogenic diet programs brain struc-
ture and susceptibility to depression and anxiety
in the adult mouse offspring. Brain Behav 5(2),
e00300.
Valdez, R., Athens, M.A., Thompson, G.H., Bradshaw,
B.B., and Stern, M.P. (1994). Birthweight and adult
health outcomes in a biethnic population in the
USA. Diabetologia 37, 624–631.
Vasconcelos, M.M., Azevedo, P.M., Esteves, L., Brito,
A.R., Olivaes, M.C., and Herdy, G.V. (2004).
Ketogenic diet for intractable epilepsy in children
and adolescents:  report of six cases. Rev Assoc
Med Bras 5, 380–385.
Vining, E.P., Freeman, J.M., Ballaban-Gil, K., Camfield,
C.S., Camfield, P.R., Holmes, G.L., Shinnar, S.,
Shuman, R., Trevathan, E., and Wheless, J.W.
(1998). A multicenter study of the efficacy of the
ketogenic diet. Arch Neurol 55, 1433–1437.
Wu, G., Bazer, F.W., Cudd, T.A., Meininger, C.J., and
Spencer, T.A. (2004). Maternal nutrition and fetal
development. J Nutr 134, 2169–2172.
240
240 section III: Ketogenic Diet in the Laboratory
Wu, G., Bazer, F.W., Davis, T.A., Kim, S.W., Li, P.,
Rhoads, J.M., Satterfield, M.C., Smith, S.B.,
Spencer, T.E., and Yin, Y. (2009). Arginine metab-
olism and nutrition in growth, health and disease.
Amino Acids 37, 153–168.
Yin, J., Han, P., Tang, Z., Liu, Q., Shi, J. (2015). Sirtuin
3 mediates neuroprotection of ketones against
ischemic stroke. J Cereb Blood Flow Metab 35,
1783–1789.
Zhao, Q, Stafstrom, C.E., Fu, D.D., Hu, Y., and Holmes,
G.L. (2004). Detrimental effects of the ketogenic
diet on cognitive function in rats. Pediatr Res
55,498–506.
Zhao, Z., Lange, DJ, Voustianiouk, A., MacGrogan,
D., Ho, L., Suh, J., Humala, N., Thiyagarajan, M.,
Wang, J., and Pasinetti, G.M. (2006). A ketogenic
diet as a potential novel therapeutic intervention
in amyotropic lateral sclerosis. Neurosci 7, 29.
241
26
Alzheimer’s Disease
Causes and Treatment
RICHARD L. VEECH, MD, PHD, DPHIL AND M. TODD KING
INCIDENCE
The prevalence of Alzheimer’s disease is strongly
correlated with age. In a general community, 3%
of those between 65 and 74 years of age had prob-
able Alzheimer’s disease, compared with 18.7% of
those between 75 and 84  years. In those over 85
the prevalence rose to 47.2% (Evans et al., 1989).
In the United States it was estimated that in 2013
there were 5.2  million patients with Alzheimer’s
disease, costing $203 billion in direct medical care
and another $216 billion in care from unpaid care-
givers (Thies & Bleiler 2013).
P R E D I S P O S I N G FA C T O R S
The major predisposing factor for Alzheimer’s dis-
ease is age. Women are more likely to be affected
than men. Diabetes, obesity and lack of exercise
predispose to the disease. The genetic environ-
mental, nutritional and metabolic risk factors are
discussed below.
Genetic
Only about 10% of the cases of Alzheimer’s disease
are familial; about 90% are classed as “sporadic,”
with age being the major risk factor. A  primary
cause of the aging-associated damage is free radi-
cal damage (Harman, 1956). Factors other than
aging can predispose individuals to the onset
of Alzheimer’s disease, the major genetic factor
being the apolipoprotein E genotype (Strittmatter
and Roses, 1995). Apolipoprotein E (ApoE) is the
lipoprotein involved in transport of lipids and
cholesterol in the circulatory system. There are
three major alleles, ApoE-2, ApoE-3, and ApoE-4.
In one large kindred group of the ApoE-4 geno-
type, the onset of clinical symptoms was between
55 and 78 years of age (Martin et al., 1997), dem-
onstrating that this genetic factor predisposes
individuals to a “late-onset” form of the disease.
Early-onset Alzheimer’s disease is associated
with three genes that increase the production or
deposition of β-amyloid protein. These include the
precursor of amyloid protein, found in Alzheimer’s
disease and in Down syndrome. Missense muta-
tions in the presenilin 1 and 2 genes, which code
for several secretases, involved in the catabolism
of amyloid, lead to the early onset of an aggressive
familial form of Alzheimer’s disease. Mutations in
these genes result in the onset of Alzheimer’s dis-
ease, often in the 40s or 50s, but occasionally in the
30s. The increased incidence of early Alzheimer’s
disease with genetic abnormalities in amyloid sug-
gests that amyloid can exacerbate the inhibition
of pyruvate dehydrogenase (PDH) (Hoshi et  al.,
1996). It follows that inhibition of cerebral PDH
could be the primary cause of Alzheimer’s disease
and the primary factor to be overcome in its treat-
ment and prevention (Kashiwaya et al., 2000).
Environmental
Radiation is known to induce production of free
radicals, or reactive oxygen species (ROS), and
induce oxidant damage (Alexander and Stacey,
1959; Riley, 1994; Szilard, 1959). Low levels of
radiation, under 5 Gy, can induce both cogni-
tive dysfunction and the pathological changes of
neurodegeneration seen in both Parkinson’s and
Alzheimer’s disease (Kempf et  al., 2013). In cul-
tured neurons, radiation also induced tau phos-
phorylation (Li et al., 2014), characteristic of both
Alzheimer’s disease and frontotemporal dementia.
These effects have application to cancer therapy,
prolonged space flight, and exposure to nuclear
radiation (Li et al., 2014).
Traumatic brain injury (TBI) prematurely
causes pathology similar to both Alzheimer’s
and Parkinson’s disease (Dekosky et  al., 2013;
DeKosky et  al., 2010). In addition to the
242
242 section III: Ketogenic Diet in the Laboratory
accumulation of amyloid and phosphorylated
tau, patients with TBI can develop both cogni-
tive impairment and Parkinsonian symptoms. In
contrast to the damaging effects of brain trauma,
physical exercise can retard neurodegeneration
and decrease inflammation in brain (Cotman
et al., 2007).
Nutritional and Metabolic
Type II diabetes is strongly associated with both
vascular dementia and Alzheimer’s dementia (Ott
et  al., 1996; Xu et  al., 2009). Diabetes has been
shown to accompany an increase in dementia in
twin studies (Kuusisto et  al., 1997). Conversely,
caloric restriction or intermittent fasting ame-
liorates amyloid accumulation (Mouton et  al.,
2009)  and the cognitive deficits in the triple
transgenic mouse models of Alzheimer’s disease
(Halagappa et al., 2007).
PAT H O P H Y S I O L O G Y
The original clinical and pathological descrip-
tion of neurofibrillary tangles, amyloid plaques,
paranoid ideation, and memory loss was given
by Alois Alzheimer in 1907 (Alzheimer, 1907).
Transgenic mouse models based on mutations
in amyloid precursor protein and its processing
show accumulation of amyloid and phosphory-
lated tau protein in their brains (Johnson-Wood
et  al., 1997). The most prevalent hypothesis
to explain the pathophysiology of Alzheimer’s
disease derives from the pathological findings
described by Alzheimer of increased amyloid
plaques and phosphorylated tau tangles. The
theory that Alzheimer’s disease results from amy-
loid accumulation in the brain has been called the
amyloid hypothesis (Selkoe, 1997). Alternatively,
it is possible that an inhibition of brain PDH due
to insulin resistance in the brain results in a defi-
ciency in substrate availability in the tricarboxylic
acid (TCA) cycle. Thus brain energy deficit and
mild cognitive impairment can precede amyloid
accumulation (Biessels and Reagan, 2015; Hoyer,
1991, 1996; Kuusisto et  al., 1997; Talbot et  al.,
2012; Willette et al., 2015). The accumulation of
amyloid and phosphorylated tau protein exacer-
bates the preexisting inhibition of PDH (Hoshi
et  al., 1997a; Hoshi et  al., 1996). Mutations in
amyloid precursor protein (APP) are found in
comparatively rare early-onset forms of the dis-
ease. Amyloid accumulates in brains of subjects
with Alzheimer’s disease, but its accumulation
does not correlate, either quantitatively or tem-
porally, with the cognitive impairment or loss of
brain volume (Josephs et al., 2008).
Alterations in the alleles of the ApoE-4 lipo-
protein (Roses, 2006) involved in cholesterol and
lipid transport, are associated with predisposi-
tion to the more common late-onset form of the
disease. While the association of the ApoE-4
allele with late-onset Alzheimer’s has been con-
firmed by exhaustive genomewide association
studies (GWAS), no other major genetic links to
Alzheimer’s disease were found (Naj et al., 2014).
The majority of cases of late-onset Alzheimer’s dis-
ease are not associated with any identified genetic
factor but are simply correlated with increased age.
Alzheimer’s disease is, however, strongly correlated
with insulin resistance both epidemiologically and
pathophysiologically (Kuusisto et  al., 1997; Ott
et al., 1996; Reaven et al., 1990; Xu et al., 2009). The
incidence of the disease can be increased by free
radical damage, radiation, TBI, immune dysfunc-
tion, and indolence (Selkoe, 2001), all of which
are also associated with insulin resistance (Zhai
et al., 2011).
Evidence suggests that amyloid β-1-42 frag-
ments? can stimulate the phosphorylation and
hence the inhibition of PDH. This amyloid frag-
ment can exert toxic effects on hippocampal neu-
rons in culture (Kashiwaya et al., 2000), which is
compatible with the report that amyloid can inhibit
PDH (Hoshi et  al., 1997b; Hoshi et  al., 1996).
Triple transgenic mouse models of Alzheimer’s
disease, in addition to showing increased amy-
loid and phosphorylated tau deposits, also exhibit
decreased fluorodeoxyglucose (FDG) uptake in
brain (Nicholson et  al., 2010). These findings are
also compatible with the ability of ketone body
metabolisms to bypass the inhibition of PDH and
produce large amounts of acetyl CoA required to
supply the Krebs cycle (Sato et  al., 1995). These
and other reports have led to the proposal that the
amyloid cascade hypothesis be rejected (Herrup,
2015), although defenders of the amyloid hypoth-
esis remain and propose that amyloid accumula-
tion is a necessary accompaniment of the disease
(Musiek and Holtzman, 2015).
Central to Alzheimer’s disease is the decrease
in cerebral glucose use. Decreases in cerebral glu-
cose metabolism occurs well before the clinical
or pathological changes of Alzheimer’s disease
(Blum-Degen et  al., 1995; Cunnane et  al., 2011;
Hoyer et  al., 1991). Inhibition of brain PDH has
long been reported in the autopsy specimens
from brains of patients with Alzheimer’s disease
(Gibson et  al., 1998). Indeed, intracerebral injec-
tion of streptozotocin, a drug that inhibits insulin
secretion, leads to a decrease in insulin synthe-
sis in brain (Grunblatt et  al., 2007). Decreases in
243
brain oxidative metabolism lead to altered pro-
cessing of amyloid precursor protein, increased
amyloid accumulation (Hoyer, 1996), and to
neuronal death. These changes in brain nutri-
ent metabolism, which precede either the clini-
cal or pathological changes of the disease, suggest
that decreased brain insulin sensitivity (Talbot
et  al., 2012; Willette et  al., 2015)  and decreased
brain energy production is an important factor
in understanding the etiology of Alzheimer’s dis-
ease (Ding et al., 2013; Morgen and Frolich, 2015).
Indeed, a deficit in brain energy metabolism alters
the processing of brain APP (Gabuzda et al., 1994),
suggesting the possibility that the accumulation of
amyloid in the brains of patients with Alzheimer’s
disease results from a deficit in cerebral energy
metabolism. The accumulation of protein is not
unique to Alzheimer’s disease, but occurs in other
neurodegenerative conditions as reflected in the
accumulation of α-synuclein in Parkinson’s dis-
ease and huntintin in Huntington’s disease. These
observations suggest that the accumulation of
amyloid in the brains of patients with Alzheimer’s
disease is secondary to a primary defect in cerebral
energy metabolism resulting from PDH inhibi-
tion due to a decrease in cerebral insulin sensitiv-
ity. There is mounting evidence suggesting that
brain insulin resistance is the major factor in the
etiology of Alzheimer’s disease (Carro and Torres-
Aleman, 2004; Craft et al., 2012; de la Monte, 2012;
Bomfim et  al., 2012; Kleinridders et  al., 2014;
Talbot et  al., 2012). A  recent study of patients
with mild Alzheimer’s dementia were compared
with cognitively normal age-matched controls
Glycolysis
Glucose Pyruvate
Acetoacetate
Alternative
Energy Source
D-β-hydroxybutyrate
FIGURE  26.1 D-β-hydroxybutyrate metabolism produces acetyl CoA, which enters the Krebs, tricarboxylic acid
cycle by an alternative pathway that does not require the usual path through pyruvate dehydrogenase (PDH).
Chapter 26: Alzheimer’s Disease 243
and were found to have a decrease in cerebral 18F-
fluorodeoxyglucose uptake but no decrease in cere-
bral metabolism of 11C-acetoacetate (Castellano
et al., 2015). These results suggest that in patients
with mild cognitive impairment, perfusion with a
ketone body can overcome the metabolic defect
resulting from the brain’s inability to use glucose.
Indeed Alzheimer’s disease has been called type
3 diabetes (Steen et al., 2005) due to the decrease
in insulin and insulin-like growth factor signaling
mechanisms in autopsy specimens of patients with
Alzheimer’s disease.
The metabolism of ketone bodies can, there-
fore, correct this insulin deficiency in brain
because the metabolism of ketone bodies mimics
the metabolic effects of insulin by increasing the
production of acetyl CoA for use in the Krebs cycle
in the absence of normal PDH activity (Kashiwaya
et  al., 1997; Sato et  al., 1995). This property of
ketone body metabolism renders the administra-
tion of ketone bodies a rational method for the
therapy of this disease (Figure 26.1).
In addition to insulin resistance, patients with
Alzheimer’s disease also show multiple physiologi-
cal signs indicative of free radical damage (Wang
et al., 2013). The metabolism of ketone bodies can
reduce free radical damage by reducing the free
cytosolic [NADP+]/[NADPH] ratio making it the
lowest redox potential in the cell (Krebs, 1969).
NADPH is the terminal destroyer of oxygen free
radicals. Additionally, the β-hydroxybutyrate mol-
ecule itself inhibits histone deacetylase (HDAC).
This inhibition of HDAC allows the acetyl groups
on histones to remain in place so that the tightly
Mitochondria
Pyruvate
Pyruvate
dehydrogenase
Acetyl CoA
3 oxoacid Krebs
CoA transferase cycle
βHBDH
D-β-hydroxybutyrate
244
244 section III: Ketogenic Diet in the Laboratory
packed structure can be relaxed and available to
the FOXO3 transcription factors which leads to the
transcription of the enzymes responsible for the
destruction of ROS (Offermanns, 2006; Shimazu
et al., 2013). The actual removal of the ROS is pow-
ered by NADPH of the free cytosolic [NADP+]/
[NADPH] ratio (Krebs and Veech, 1969).
T R E AT M E N T
The hypothesis that amyloid accumulation is the
central etiological factor leading to Alzheimer’s dis-
ease has led multiple pharmaceutical companies to
develop monoclonal antibodies to removed amy-
loid. Active amyloid vaccine trials were suspended
due to the development of meningoencephali-
tis (Schenk, 2002). Passive immunotherapeutics
with two antibodies have failed to significantly
improve cognitive function or prevent progression
but are now being tested in the early stage of the
disease (Panza et al., 2014). Newer antibodies are
currently being tested in asymptomatic individu-
als at risk of developing the disease, based on the
amyloid β cascade hypothesis, in the hope that this
approach will retard the development of clinically
observable disease. It is fair to say that, to date, the
results have been very disappointing (Prins and
Scheltens, 2013).
An alternative treatment for Alzheimer’s dis-
ease has been developed based on the hypothesis
that the disease results from a primary failure of
the brain to develop both phosphorylation and
redox energy. The therapy has been based on the
observations that during starvation ketone bod-
ies can replace glucose as the major metabolic
fuel in brain (Cahill and Aoki, 1980; Owen et al.,
1967). The detailed effects of the metabolism of
ketone bodies were not well understood. A study
was undertaken to explore the metabolic effects of
ketone body metabolism in the working perfused
rat heart (Kashiwaya et al., 1994; Sato et al., 1995).
The working perfused heart provides the complete
data needed on the effects on redox and phosphor-
ylation states, O2 consumption, and metabolic effi-
ciency. Addition of ketone bodies to the glucose
perfused heart resulted in a reduction of the mito-
chondrial NAD- couple, an increase in the ΔG’ of
ATP, and a decrease in glucose use, indicating that
the metabolism of ketone bodies replaced in large
part the heart’s need for glucose to supply its meta-
bolic energy (Sato et al., 1995).
More remarkably, the addition and metabo-
lism of ketone bodies, like addition of insulin,
increased the joules of hydraulic work put out by
the heart per mole of O2 consumed (Kashiwaya
et  al., 1994), demonstrating that either ketone
bodies or insulin could improve the metabolic
efficiency of the working heart. Moreover, ketone
body metabolism can mimic the metabolic effects
of insulin (Kashiwaya et  al., 1994). This increase
in the efficiency of hydraulic work of the heart
induced by either ketone bodies or insulin indi-
cated an increase in the efficiency of mitochondrial
energy generation. That increase in mitochondrial
energy generation could later be translated into an
increased physiological performance in elite ath-
letes or in disease states where metabolic energy
generation is deficient.
A detailed examination of the metabolites
of the Krebs tricarboxcylic acid cycle showed
that addition of insulin to the working glucose-
perfused heart increased acetyl CoA content
ninefold, while addition of 4-mM ketone bod-
ies increased acetyl CoA 15-fold. This very large
increase in acetyl CoA in the case of insulin addi-
tion was caused by that hormone’s increased PDH
activity, the enzymatic gateway to the Krebs cycle
(Coore et al., 1971; Taylor and Jungas, 1974; Taylor
et  al., 1975). The even larger increase in acetyl
CoA caused by the metabolism of ketone bodies
demonstrated that metabolism of ketone bodies
could bypass the major metabolic block in insu-
lin sensitivity. The metabolism of ketone bodies
mimics the effect of insulin’s activation of PDH
by producing acetyl CoA from the ketone body
acetoacetate via the succinyl CoA transferase reac-
tion (Sato et al., 1995). The metabolism of ketone
bodies mimics a major metabolic effect of insulin
(Kashiwaya et al., 1997) and could therefore over-
come insulin resistance, which is a common factor
in many disease (Scheuermann-Freestone et  al.,
2004; Zhai et  al., 2011)  or injury states (Li and
Messina, 2009), including the brain (Talbot et al.,
2012) of Alzheimer’s disease patients.
The Krebs cycle is the major energy-producing
metabolic pathway, which results in the aerobic
production of ATP by the complete oxidation of
glucose. In brain, glucose is essentially the only
energy-producing substrate under fed conditions.
The entry of glucose into the Krebs cycle occurs
through the production of acetyl CoA from pyru-
vate in a reaction catalyzed by PDH. The speed of
that essential reaction is increased by insulin and
decreased by loss of sensitivity to insulin. The vari-
ation in the speed of this reaction strongly suggests
the loss of sensitivity of the brain’s PDH to insulin
is the root cause of Alzheimer’s disease.
Other metabolic changes of importance also
occur with addition of ketone bodies or insu-
lin. The ratio of [isocitrate]/[α-ketoglutarate] is
increased by both, indicating a reduction of the
free NADP system (Krebs and Veech, 1969). The
free mitochondrial [NAD+]/[NADH] ratio is also
245

Chapter 26: Alzheimer’s Disease 245

Lactate–
[NAD+]c/[NADH]c [AcCoA] G vs GI
NAD+c 3.39 nmol/mL ICW
1233
1 * *
LDH 18 G vs GK
2
3 NADHc+H+c 16 G vs GIK
[Pyr] Pyruvate– 14
0.055 µmol/mL ICW 12
2
1 10
2 * *
8
6 [Cit]
[CoA]
132 µmol/mL ICW
FAD NADHm 4
0.412 µmol/mL ICW
PDH 4
1 * * 3
2
2 CO2 FADH2 NAD+m 0
2
1
* * * * * *
CoA
Acetyl CoA [IsoCit]
H+m 0.017 µmol/mL ICW
5
4
3
Citrate3– 2
1 [αKG]
[ASP] * * *
0.194 µmol/mL ICW
4.54 µmol/mL ICW
* * *
Citrate systhase Aconitase 3
1 2
2 1
3 * *
4 Oxaloacetate2– Isocitrate3– [Glut]
[OAA]
2.33 µmol/mL ICW
0.63 µmol/mL ICW NADHm + H+m NAD(P)+m 6
1 * * *
MDH 5
2 ICDH 4
3
3 NAD+m CO2 NAD(P)Hm 2
4 1
NH4+ * * *
[Malate] Malate2– GLDH
3
0.112 µmol/mL ICW [Fum]/[Succ] α-Ketoglutarate2– Glutamate–
2 Fumarase 0.076
1 5
4
H+m
* *
2– 32
Fumarate 1
FAD NADHm [NH4+]
0.879 µmol/mL ICW
* * *
αKGDH 2
1
[Fumarate] CoQH2
FAD +
FADH2 NAD m
2 * * *
0.071 µmol/mL ICW
Succ DH CO2
2 [H+]
1
2 * FADH2 CoQ 8.16x10–3 mol/L MW
1 * * *
– 2
[Succinate] Succinate2– Succinyl CoA 3
1.00 µmol/mL ICW [SuccCoA] [NAD+]m/[NADH]m
1 * * 16.0
33.5 µmol/mL ICW
2 2 * * *
1 1
3 Succinyl CoA Ligase 2
2 *

2x [Pi]m 4
28.2 µmol/mL MW 6
* * *
1
Acetyl CoA Transferase 2 8
GTP GDP Pi 10
4
3-Oxoacid CoA Transferase 12
CoA 6
8 D-β-hydroxybutyrate–
10
12
[AcAcCoA] NAD+m
Acetoacetyl CoA 4.45 µmol/mL ICW Acetoacetate–
4 βHBDH
3
2
1 NADHm+H+m
*

FIGURE  26.2 The study of perfused rat hearts treated with glucose, glucose + insulin, glucose + ketone (D-β-
hydroxybutyrate), or glucose +insulin + ketone (D-β-hydroxybutyrate). The relevant pathways are shown along with
concentration changes relative to just glucose. Images are modified from Sato 1995 (Sato et al., 1995) originally drawn
by Y. Kashiwaya.

reduced 3- to 10-fold by the addition of either in the ratio of [Coenzyme Q]/[Coenzyme QH2]
ketone bodies or insulin. Also of major signifi- (Bergman et al., 2010). The increase in the redox
cance is the increase in the ratio of [fumarate]/ span between the free mitochondrial NAD and
[succinate], indicating an oxidation of Q as seen Q couples indicates an increase in the ΔG of ATP
 246
E m/c
E m/c G GI GK*GKI*
Mitochondrial area: 32.1-36.8% –100
Water: 17 – 20% of ICW –110
–120

mV
–130
–140
Eh NAD+/NADH m Mitochondrial pH –150
Mitochondrial 7.6
* *
–260 G GI GK GKI
* inner membrane 7.4 7.6 Cytosolic pH
–270 7.2 7.4
–280 Matrix 7.0 7.2
mV

–290 6.8 7.0

–300 H+ H+ 6.6 6.8
G GI* GK* GKI* 6.6
–310 G GI GK GKI
NH4+ NH3 NH3 NH4+ ATP/ADP
Eh Q/QH2 exchanger
NAD+ DβHB–
20 ADP3– ATP4– ATP
16 4H+ 4H+
12 ∆G ATP hydrolysis
8 NADH DH 2e– Glut– ATP4–
mV

4 ADP3–
0 G 2e– NADH + H+
–4 GI* GK* GKI*
Succ DH Pi
–8
ADP
QH2 ATP
Cit3– ATP synthase Cytosolic
∆G of Q/QH2-NAD+/NADH m
CoQH2 – ∆ G of ATP hydrolysis
Fum2– 3H+ 3H+
G GI* GK* GKI* Cytochrome C G GI* GK* GKI*
–50
–50 reductase 2e– TCA cycle αKG2– –52
–52 4H+ 4H+ ADP

Kj/mol
–54 –54
Kj/mol

Succ2–
–56 –56
–58 Cyt C Red Succ CoA– Pi –58
–60 +
H -Pi –60
–62 Cytochrome Ox contransporter
oxidase 4H+ H2PO4–
4H+ 2e– AcAc CoA AcAc– H2PO4–
Red Ox 2H+ Cytosolic
Cyt aa3 EhCyt zz3 Ox/Red H+ H+ [ADP] and [Pi]
VO2 620 [Pi]c [ADP]c
10 0.16
20 Measured 600
µmol/min ICW

H2O 0.14

µmol/min ICW
19 8
Calculated
mV

580 6 0.12
18 0.10
17 560 4
½ O2 0.08
16 540
15 G GI* GK* GKI* 2 0.06
14 G GI* GK* GKI* 0 G GI GK GKI
0.04

FIGURE  26.3 The study of perfused rat hearts treated with glucose, glucose + insulin, glucose + ketone (D-β-hydroxybutyrate), or glucose +insulin +
ketones (D-β-hydroxybutyrate). The relevant pathways are shown along with the relevant measures of ∆G, Gibbs free energy calculated for the nonstandard
concentrations, E, reduction potentials based on actual concentrations, VO2 is oxygen consumption in mL/kg/min. Images are modified from Sato 1995 (Sato
et al., 1995) originally drawn by Y. Kashiwaya.
247
hydrolysis from −53 to −60 kJ /mole in spite of
a decrease in O2 consumption from 18.5 to 16–
17 μmol/min. An increase in the energy of ATP
hydrolysis per O2 consumed was observed with
the addition of either ketone bodies or insulin (see
Figures 26.2 and 26.3) (Sato et al., 1995).
W H AT F O R M S O F   K E T O N E
B O D I E S A R E AVA I L A B L E ?
Following the realization of the widespread effects
of ketone body metabolism on the redox and
phosphorylation states (Sato et  al., 1995)  and
metabolic efficiency, it was recognized that thera-
peutic effects could result from the administration
of ketone bodies (Cahill and Veech, 2003; Veech
et  al., 2001). During prolonged fasting (3  days)
blood ketone body levels reach 5–7 mM (Cahill
and Aoki, 1980). This blood level of ketone bodies
results from the endogenous production of about
150 g of ketone bodies per day (Reichard et  al.,
1974). Therefore, to mimic the effects of the keto-
sis of starvation, about 150 g of ketone bodies, or
about 1.5 moles, would be required to mimic the
effects of starvation while maintaining safe levels
(5–7 mM) of ketones. It is important to note that
the minimum effective dose of ketone bodies in
various disease states has not been established and
may be much less than the 150 g per day produced
during prolonged starvation.
Administration of ketone bodies as a simple
acid or salt would present an unsafe load of either
counter ion. Accordingly, examination of a variety
of alcohols was undertaken. R-1,3 butanediol was
the alcohol chosen, because it was converted in the
liver to ketone bodies (Mehlman and Veech, 1972).
Thus, a monoester of d-β-hydroxybutyrate –R-1,3
butanediol monoester presents no other metabo-
lite load than the ketone body.
Additionally, the oxidation state of the ketone
body being administered needs to be considered.
Ketone bodies are readily interconverted between
the oxidized form, acetoacetate, and the reduced
form, d-β-hydroxybutyrate.
d-β-hydroxybutyrate + NAD+ ↔ acetoacetate
+ NADH + H+
While both forms of the ketone body can be metab-
olized, the energetic effects are quite different.
Acetoacetate can readily be used in the Langendorf
perfused heart, which is a simpler model than the
working perfused heart (Williamson and Krebs,
1961). In the working perfused rat heart, d-β-
hydroxybutyrate can be used to power the beating
heart effectively. When acetoacetate is substituted,
Chapter 26: Alzheimer’s Disease 247
the heart fails (Taegtmeyer et  al., 1980), showing
that the excessive oxidation of the mitochondria
NAD couple produced by acetoacetate impairs
cardiac energy production. Although acetoacetate
is attractive because of its low cost, clearly it should
be avoided as a supplement and the more expen-
sive chiral molecule d-β-hydroxybutyrate should
be considered instead.
R-1,3 butanediol is converted to d-β-
hydroxybutyrate, the physiological form of the
ketone body, which is first oxidized to acetoac-
etate and then activated by succinyl CoA trans-
ferase to acetoacetyl CoA and hence to 2 acetyl
CoA s by thiolase (see Figure 26.1). In contrast, S
1,3 butanediol is converted to l-β hydroxybutyr-
ate, which is activated by ATP and CoA to form
l-β-hydroxybutyryl CoA and further metabolized
by the β fatty acid oxidation pathway (Lehninger
& Greville, 1953). The metabolic fates of the d
versus the l form of β-hydroxybutyrate are quite
different, as shown in Figure 26.4. Metabolism of
one molecule of the physiological enantiomer d
β-hydroxybutyrate consumes one molecule each
of NAD+, succinyl-CoA, CoASH, and CoQ and
forms two molecules of acetyl-CoA, one molecule
of fumarate, and one molecule each of the reduced
forms of NAD+ and coenzyme Q (QH2). On the
other hand, metabolism of the nonphysiological
enantiomer, l β-hydroxybutyrate consumes two
molecules of CoA-SH, one molecule of NAD+, and
one molecule of ATP and forms two molecules
of acetyl-CoA and one molecule each of AMP
and pyrophosphate as well as one molecule of
the reduced form of NAD+. A second molecule of
ATP is also consumed to convert the AMP formed
to ADP.
In metabolizing mid-chain triglycerides (see
Figure 26.5), a reduced flavoprotein is produced in
the β oxidation pathway, which results in a reduc-
tion of mitochondrial coenzyme Q, narrowing
the redox span between mitochondrial NAD and
Q and hence decreasing the energy available for
ATP synthesis. This is in contrast to the increase
in redox span that acompanies the metabolism of
the physiological form of the ketone body, d-β-
hydroxybutyrate (Sato et al., 1995).
It follows from the above, that the oxidized
form of ketone body, acetoacetate, should not
be included in esters for administration for the
purpose of elevating ketone bodies because such
solutions would oxidize the mitochondrial NAD
couple and lower the energy available for the for-
mation of ATP. Similarly, the use of racemic R,
S 1.3 butanediol should not be used because the
unphysiological S form undergoes β oxidation,
248

248 section III: Ketogenic Diet in the Laboratory

(a) Metabolic fate of L β-Hydroxybutyrate (b) Metabolic fate of D β-Hydroxybutyrate

L β-Hydroxybutyrate D β-Hydroxybutyrate

ATP + CoASH NAD+

AMP + PPi NADH, H+

L β-Hydroxybutyryl-CoA Acetoacetate

NAD+ Succ-CoA

NADH, H+ Succinate CoQ

CoQH2
Acetoacetyl-CoA Acetoacetyl-CoA
Fumarate
CoASH CoASH

2 Acetyl-CoA 2 Acetyl-CoA

(c) Sum reaction for L b-Hydroxybutyrate

L βHB + ATP + 2CoASH + NAD+ 2Ac-CoA + AMP + PPi + NADH + H+

(d) Sum reaction for L b-Hydroxybutyrate

D βHB + NAD+ + Succ-CoA + CoASH + CoQ 2Ac-CoA + NADH + H+ Fun + CoQH2

FIGURE 26.4 The metabolism and sum reactions for both D and L β-hydroxybutyrate.
Metabolism of the L form of β-hydroxybutyrate (A) uses two molecules of CoASH and one molecule each of ATP and
NAD+ and forms 2 acetyl-CoAs and one molecule each of AMP, PPi, and reduced NAD+ (NADH, H+). Metabolism
of the D form of β-hydroxybutyrate (B) on the other hand uses one molecule each of NAD+, succ-CoA, CoASH, and
CoQ and produces two molecules of acetyl-CoA and one molecule each of fumarate and the reduced forms of NAD+
(NADN, H+), and CoQ (CoQH2). Sum reactions for each isomer are given in C and D.

producing mitochondrial flavoprotein, thus cerebral energy generation, primarily due to a

decreasing the oxidation of the Q couple, which
results from the metabolism of the physiological
R-3-hydroxybutyrate. Several studies using the
acetoacetate and racemic 1,3 butanediol ester have
appeared (D’Agostino, 2013; Desrochers et  al.,
1995). The effects of these ketone ester formula-
tions would be quite different from those produced
by the d-β-hydroxybutyrate–R-1, 3 butanediol
monoester.
Evidence presented here suggests that
Alzheimer’s disease results from a deficit in
loss of insulin sensitivity in brain, and that amy-
loid accumulation results from this decrease in
cerebral metabolic energy. The primary thera-
peutic target would therefore become not the
removal of amyloid but the overcoming of the
block of cerebral metabolic energy production
caused by cerebral insulin resistance. This could
be achieved by (1) feeding a therapeutic amount
of d-β-hydroxybutyrate–R 1,3 butanediol mono-
ester, producing acetyl CoA without requir-
ing the activity of brain PDH (Kashiwaya et  al.,
249
Fatty acyl-CoA
FAD
FADH2
2
Trans-∆ –Enoyl-CoA
H2O
L-β–Hydroxyacyl-CoA
NAD
NADH + H+
β–Ketoacyl-CoA
CoASH
Fatty acyl-CoA + Acetyl-CoA
(2 C atoms shorter)
FIGURE 26.5 The beta-oxidation pathway of metabolizing fatty acids. The β fatty acid oxidation pathway produces
1 NADH and 1 flavoprotein compared with the two NADH’s produced by the oxidation of the physiologically normal
D-β-hydroxybutyrate. Once you cycle down to the last two carbon atoms the terminal L-β-hydroxyacyl-CoA is L-β-
hydroxybutyryl-CoA, not the D form of β-hydroxybutyrate.
2013; Newport et  al., 2015). This approach has
decreased amyloid and phosphorylated tau in
the brains of triple transgenic mice and improved
cognitive performance. In a single human sub-
ject, behavioral improvement on feeding ketone
has been noted.
Alternatively, increasing the brain’s concentra-
tion of insulin by introducing insulin into brain
through intranasal administration of insulin has
been reported. This therapy also has proven suc-
cessful in treating Alzheimer’s disease in a limited
number of cases. Cerebrospinal fluid insulin levels
were shown to be significantly lower in patients
with Alzheimer’s than normal controls (Craft et al.,
1998). The olfactory lobe contains the highest con-
centration of insulin receptors of any brain area
(Kleinridders et al., 2014). Intranasal administration
of large doses of insulin improve cognitive perfor-
mance in patients with early cognitive impairment
due to Alzheimer’s disease (Reger et al., 2008).
Ketone bodies mimic the metabolic effects of
insulin. They increase the energy of ATP hydroly-
sis, decrease free radical damage, and increase the
transcription of antioxidant enzymes by inhibiting
histone deacetylase (Shimazu et  al., 2013; Veech,
2014). In the genetic mouse model of early-onset
Alzheimer’s disease, feeding mice ketone esters
Chapter 26: Alzheimer’s Disease 249
ETF:ubiquinone
ETFred Q
oxidoreductaseox
ETF:ubiquinone
ETFox QH2
oxidoreductasered
When the number of
carbons left reaches 4,
this molecule becomes
L-β-hydroxybutyryl-CoA.
Krebs cycle
resulted in a decrease in brain amyloid and phos-
phorylated tau accumulation as well as improve-
ment in cognitive function (Kashiwaya et al., 2013).
Feeding ketone ester to a patient with advanced
Alzheimer’s disease improved his behavior and
cognitive function (Newport et  al., 2015). These
results support the need for larger clinical trials of
ketone esters in the treatment of Alzheimer’s dis-
ease and other diseases where insulin resistance is
a major etiological factor.
REFERENCES
Alexander, P., and Stacey, K.A. (1959). Cross-linking
of deoxyribonucleic acid in sperm heads by ion-
izing radiations. Nature 184, 958–960.
Alzheimer, A. (1907). Uber eine eigenartige
Erkrankung der Hirnrinde. Allgemeine Zeitschrift
fur Psychiatrie und physchish-Gerichtliche
Medizin 64, 146–148.
Bergman, C., Kashiwaya, Y., and Veech, R.L. (2010).
The effect of pH and free Mg2+ on ATP linked
enzymes and the calculation of Gibbs free energy of
ATP hydrolysis. J Phys Chem B 114, 16137–16146.
Biessels, G.J., and Reagan, L.P. (2015). Hippocampal
insulin resistance and cognitive dysfunction. Nat
Rev Neurosci 16(11), 660–671.
Blum-Degen, D., Frolich, L., Hoyer, S., and Riederer, P.
(1995). Altered regulation of brain glucose
250
250 section III: Ketogenic Diet in the Laboratory
metabolism as a cause of neurodegenerative
disorders? J Neural Transm Suppl 46, 139–147,
139–147.
Bomfim, T.R., Forny-Germano, L., Sathler, L.B., Brito-
Moreira, J., Houzel, J.C., Decker, H., Silverman,
M.A., Kazi, H., Melo, H.M., McClean, P.L., et  al.
(2012). An anti-diabetes agent protects the mouse
brain from defective insulin signaling caused by
Alzheimer’s disease- associated Abeta oligomers. J
Clin Invest122, 1339–1353.
Cahill, G.F., Jr., and Aoki, T.T. (1980). Alternate fuel uti-
lization in brain. In Cerebral metabolism and neu-
ral function, J.V. Passonneau, R.A. Hawkins, W.D.
Lust, and F.A. Welsh, eds. (Baltimore: Williams &
Wilkins), pp. 234–242.
Cahill, G.F., Jr., and Veech, R.L. (2003). Ketoacids?
Good medicine? Trans Am Clin Climatol Assoc
114, 149–161; discussion 162–163, 149–161.
Carro, E., and Torres-Aleman, I. (2004). The role of
insulin and insulin-like growth factor I  in the
molecular and cellular mechanisms underlying
the pathology of Alzheimer’s disease. Eur J of
Pharmacol 490, 127–133.
Castellano, C.A., Nugent, S., Paquet, N., Tremblay, S.,
Bocti, C., Lacombe, G., Imbeault, H., Turcotte, E.,
Fulop, T., and Cunnane, S.C. (2015). Lower brain
18F-fluorodeoxyglucose uptake but normal 11C-
acetoacetate metabolism in mild Alzheimer’s dis-
ease dementia. J Alzheimers Dis 43, 1343–1353.
Coore, H.G., Denton, R.M., Martin, B.R., and Randle,
P.J. (1971). Regulation of adipose tissue pyruvate
dehydrogenase by insulin and other hormones.
Biochem J 125, 115–127.
Cotman, C.W., Berchtold, N.C., and Christie, L.A.
(2007). Exercise builds brain health:  key roles of
growth factor cascades and inflammation. Trends
Neurosci 30, 464–472.
Craft, S., Cholerton, B., and Baker, L.D. (2012). Insulin
and Alzheimer’s disease:  untangling the web.
Journal of Alzheimer’s Disease:  JAD 33 Suppl 1,
S263–275.
Craft, S., Peskind, E., Schwartz, M.W., Schellenberg,
G.D., Raskind, M., and Porte, D., Jr. (1998).
Cerebrospinal fluid and plasma insulin levels
in Alzheimer’s disease:  relationship to sever-
ity of dementia and apolipoprotein E genotype.
Neurology 50, 164–168.
Cunnane, S., Nugent, S., Roy, M., Courchesne-Loyer,
A., Croteau, E., Tremblay, S., Castellano, A., Pifferi,
F., Bocti, C., Paquet, N., et  al. (2011). Brain fuel
metabolism, aging, and Alzheimer’s disease.
Nutrition 27, 3–20.
D’Agostino, D. P., Pilla, R., Held, H. E., Landon, C. S.,
Puchowicz, M., Brunengraber, H., … Dean, J. B.
(2013). Therapeutic ketosis with ketone ester delays
central nervous system oxygen toxicity seizures in
rats. American Journal of Physiology – Regulatory,
Integrative and Comparative Physiology, 304(10),
R829–R836.
de la Monte, S.M. (2012). Brain insulin resistance and
deficiency as therapeutic targets in Alzheimer’s
disease. Curr Alzheimer Res 9, 35–66.
DeKosky, S.T., Ikonomovic, M.D., and Gandy, S.
(2010). Traumatic brain injury:  football, war-
fare, and long-term effects. N Engl J Med 363,
1293–1296.
Dekosky, S.T., Blennow, K., Ikonomovic, M.D., and
Gandy, S. (2013). Acute and chronic traumatic
encephalopathies:  pathogenesis and biomarkers.
Nat Rev Neurol 9, 192–200.
Desrochers, S., Quinze, K., Dugas, H., Dubreuil, P.,
Bomont, C., David, F., Agarwal, K.C., Kumar, A.,
Soloviev, M.V., Powers, L., et al. (1995). (R,S) 1,3
butandiol acetoacetate esters, potential alternates
to lipid emulsions for total parenteral nutrition. J
Nutr Biochem 6, 109–116.
Ding, F., Yao, J., Rettberg, J.R., Chen, S., and Brinton,
R.D. (2013). Early decline in glucose transport and
metabolism precedes shift to ketogenic system in
female aging and Alzheimer’s mouse brain: impli-
cation for bioenergetic intervention. PloS One 8,
e79977.
Evans, D.A., Funkenstein, H.H., Albert, M.S.,
Scherr, P.A., Cook, N.R., Chown, M.J., Hebert,
L.E., Hennekens, C.H., and Taylor, J.O. (1989).
Prevalence of Alzheimer’s disease in a commu-
nity population of older persons. Higher than
previously reported [see comments]. JAMA 262,
2551–2556.
Gabuzda, D., Busciglio, J., Chen, L.B., Matsudaira, P.,
and Yankner, B.A. (1994). Inhibition of energy
metabolism alters the processing of amyloid pre-
cursor protein and induces a potentially amyloido-
genic derivative. J Biol Chem 269, 13623–13628.
Gibson, G.E., Sheu, K.F., and Blass, J.P. (1998).
Abnormalities of mitochondrial enzymes in
Alzheimer disease. J Neural Transm 105, 855–870.
Grunblatt, E., Salkovic-Petrisic, M., Osmanovic, J.,
Riederer, P., and Hoyer, S. (2007). Brain insulin
system dysfunction in streptozotocin intracere-
broventricularly treated rats generates hyperphos-
phorylated tau protein. J Neurochem 101, 757–770.
Halagappa, V.K., Guo, Z., Pearson, M., Matsuoka,
Y., Cutler, R.G., Laferla, F.M., and Mattson, M.P.
(2007). Intermittent fasting and caloric restriction
ameliorate age-related behavioral deficits in the
triple-transgenic mouse model of Alzheimer’s dis-
ease. Neurobiol Dis 26, 212–220.
Harman, D. (1956). Aging:  a theory based on free
radical and radiation chemistry. J Gerontol 11,
298–300.1
Herrup, K. (2015). The case for rejecting the amyloid
cascade hypothesis. Nat Neurosci 8, 794–799.
Hoshi, M., Takashima, A., Noguchi, K., Murayama,
M., Sato, M., Kondo, S., Saitoh, Y., Ishiguro, K.,
Hoshino, T., and Imahori, K. (1996). Regulation
of mitochondrial pyruvate dehydrogenase activ-
ity by tau protein kinase I/glycogen synthase
251
kinase 3beta in brain. Proc Natl Acad Sci USA 93,
2719–2723.
Hoshi, M., Takashima, A., Murayama, M., Yasutake,
K., Yoshida, N., Ishiguro, K., Hoshino, T., and
Imahori, K. (1997a). Nontoxic amyloid á peptide
1-42 suppresses acetylcholine synthesis. J Biol
Chem 272, 2038–2041.
Hoshi, M., Takashima, A., Murayama, M., Yasutake,
K., Yoshida, N., Ishiguro, K., Hoshino, T., and
Imahori, K. (1997b). Nontoxic amyloid beta
peptide 1-42 suppresses acetylcholine synthe-
sis. Possible role in cholinergic dysfunction in
Alzheimer’s disease. J Biol Chem 272, 2038–2041.
Hoyer, S. (1991). Abnormalities of glucose metabo-
lism in Alzheimer’s disease. Ann N Y Acad Sci
640, 53–58.
Hoyer, S. (1996). Oxidative metabolism deficiencies in
brains of patients with Alzheimer’s disease. Acta
Neurol Scand Suppl 165, 18–24, 18–24.
Hoyer, S., Nitsch, R., and Oesterreich, K. (1991).
Predominant abnormality in cerebral glucose uti-
lization in late-onset dementia of the Alzheimer
type:  a cross-sectional comparison against
advanced late-onset and incipient early-onset
cases. J Neural Transm Park Dis Dement Sect
3, 1–14.
Johnson-Wood, K., Lee, M., Motter, R., Hu, K.,
Gordon, G., Barbour, R., Khan, K., Gordon, M.,
Tan, H., Games, D., et al. (1997). Amyloid precur-
sor protein processing and A beta42 deposition in
a transgenic mouse model of Alzheimer disease.
Proc Natl Acad Sci USA 94, 1550–1555.
Josephs, K.A., Whitwell, J.L., Ahmed, Z., Shiung, M.M.,
Weigand, S.D., Knopman, D.S., Boeve, B.F., Parisi,
J.E., Petersen, R.C., Dickson, D.W., et  al. (2008).
Beta-amyloid burden is not associated with rates
of brain atrophy. Ann Neurol 63, 204–212.
Kashiwaya, Y., Sato, K., Tsuchiya, N., Thomas, S., Fell,
D.A., Veech, R.L., and Passonneau, J.V. (1994).
Control of glucose utilization in working perfused
rat heart. J Biol Chem 269, 25502–25514.
Kashiwaya, Y., King, M.T., and Veech, R.L. (1997).
Substrate signaling by insulin:  a ketone bodies
ratio mimics insulin action in heart. Am J Cardiol
80, 50A–64A.
Kashiwaya, Y., Takeshima, T., Mori, N., Nakashima,
K., Clarke, K., and Veech, R.L. (2000). D-beta -
Hydroxybutyrate protects neurons in models of
Alzheimer’s and Parkinson’s disease. Proc Natl
Acad Sci USA 97, 5440–5444.
Kashiwaya, Y., Bergman, C., Lee, J.H., Wan, R., Todd
King, M., Mughal, M.R., Okun, E., Clarke, K.,
Mattson, M.P., and Veech, R.L. (2013). A ketone
ester diet exhibits anxiolytic and cognition-
sparing properties, and lessens amyloid and tau
pathologies in a mouse model of Alzheimer’s dis-
ease. Neurobiol Aging 34, 1530–1539.
Kempf, S.J., Azimzadeh, O., Atkinson, M.J., and Tapio,
S. (2013). Long-term effects of ionising radiation
Chapter 26: Alzheimer’s Disease 251
on the brain:  cause for concern? Radiat Environ
Biophys 52, 5–16.
Kleinridders, A., Ferris, H.A., Cai, W., and Kahn, C.R.
(2014). Insulin action in brain regulates systemic
metabolism and brain function. Diabetes 63,
2232–2243.
Krebs, H.A., and Veech, R.L. (1969). Pyridine nucle-
otide interrelations. In The Energy Level and
Metabolic Control in Mitochondria, S. Papa,
Tager, J.M., Quagliariello, E., & Slater, E.C., eds.
(Bari, Italy: Adriatica Editrice), pp. 329–383.
Kuusisto, J., Koivisto, K., Mykkanen, L., Helkala,
F.L., Vanhunen, M., Kervinen, K., Kesaniemi,
Y.A., Riekkinen, P.J., and Laakso, M. (1997).
Association between features of the insulin resis-
tance syndrome and Alzheimer’s disease inde-
pendently of apolipoprotein E4 phenotype: cross
sectional population based study. Brit Med J 315,
1045–1049.
Lehninger, A.L., and Greville, G.D. (1953). The enzy-
matic oxidation of d and l b hydroxybutyrate.
Biochem Biophys Acta 12, 188–202.
Li, L., and Messina, J.L. (2009). Acute insulin resis-
tance following injury. Trends Endocrinol Metab
20, 429–435.
Li, L., Wang, W., Welford, S., Zhang, T., Wang, X.,
and Zhu, X. (2014). Ionizing radiation causes
increased tau phosphorylation in primary neu-
rons. J Neurochem 131, 86–93.
Martin, E.S., Martin, S.E., Edelsohn, L., and
Borgaonkar, D.S. (1997). Studies in a large fam-
ily with late-onset Alzheimer disease (LOAD).
Alzheimer Dis Assoc Disord 11, 163–170.
Mehlman, M.A., and Veech, R.L. (1972). Redox and
phosphorylation states and metabolite concentra-
tions in frozen clamped livers of rats fed diets con-
taining 1,3- butanediol and DL-carnitine. J Nutr
102, 45–51.
Morgen, K., and Frolich, L. (2015). The metabolism
hypothesis of Alzheimer’s disease:  from the con-
cept of central insulin resistance and associated
consequences to insulin therapy. J Neural Transm
122, 499–504.
Mouton, P.R., Chachich, M.E., Quigley, C., Spangler,
E., and Ingram, D.K. (2009). Caloric restriction
attenuates amyloid deposition in middle-aged dtg
APP/PS1 mice. Neurosci Lett 464(3), 184–187.
Musiek, E.S., and Holtzman, D.M. (2015). Three
dimensions of the amyloid hypothesis: time, space
and ‘wingmen’. Nat Neurosci 18, 800–806.
Naj, A.C., Jun, G., Reitz, C., Kunkle, B.W., Perry, W.,
Park, Y.S., Beecham, G.W., Rajbhandary, R.A.,
Hamilton-Nelson, K.L., Wang, L.S., et al. (2014).
Effects of multiple genetic loci on age at onset in
late-onset Alzheimer disease: a genome-wide asso-
ciation study. JAMA Neurol 71, 1394–1404.
Newport, M.T., VanItallie, T.B., Kashiwaya, Y., King,
M.T., and Veech, R.L. (2015). A new way to pro-
duce hyperketonemia: Use of ketone ester in a case
252
252 section III: Ketogenic Diet in the Laboratory
of Alzheimer’s disease. Alzheimers Dement 11,
99–103.
Nicholson, R.M., Kusne, Y., Nowak, L.A., LaFerla,
F.M., Reiman, E.M., and Valla, J. (2010). Regional
cerebral glucose uptake in the 3xTG model of
Alzheimer’s disease highlights common regional
vulnerability across AD mouse models. Brain Res
1347, 179–185.
Offermanns, S. (2006). The nicotinic acid receptor
GPR109A (HM74A or PUMA-G) as a new thera-
peutic target. Trends Pharmacol Sci 27, 384–390.
Ott, A., Stolk, R.P., Hofman, A., van Harskamp,
F., Grobbee, D.E., and Breteler, M.M. (1996).
Association of diabetes mellitus and dementia: the
Rotterdam Study. Diabetologia 39, 1392–1397.
Owen, O.E., Morgan, A.P., Kemp, H.G., Sullivan,
J.M., Herrera, M.G., and Cahill, G.F. (1967).
Brain metabolism during fasting. J Clin Invest 46,
1589–1595.
Panza, F., Logroscino, G., Imbimbo, B.P., and Solfrizzi,
V. (2014). Is there still any hope for amyloid-based
immunotherapy for Alzheimer’s disease? Curr
Opin Psychiatry 27, 128–137.
Prins, N.D., and Scheltens, P. (2013). Treating
Alzheimer’s disease with monoclonal antibod-
ies:  current status and outlook for the future.
Alzheimers Res Ther 5, 56.
Reaven, G.M., Thompson, L.W., Nahum, D., and
Haskins, E. (1990). Relationship between hyper-
glycemia and cognitive function in older NIDDM
patients. Diabetes Care 13, 16–21.
Reger, M.A., Watson, G.S., Green, P.S., Wilkinson,
C.W., Baker, L.D., Cholerton, B., Fishel, M.A.,
Plymate, S.R., Breitner, J.C., DeGroodt, W., et  al.
(2008). Intranasal insulin improves cognition and
modulates beta-amyloid in early AD. Neurology
70, 440–448.
Reichard, G.A., Owen, O.E., Haff, A.C., Paul, P., and
Bortz, W.M. (1974). Ketone-body production and
oxidation in fasting obese humans. J Clin Invest
53, 508–515.
Riley, P.A. (1994). Free radicals in biology:  oxidative
stress and the effects of ionizing radiation. Int J
Radiat Biol 65, 27–33.
Roses, A.D. (2006). On the discovery of the genetic
association of apolipoprotein E genotypes
and common late-onset Alzheimer disease. J
Alzheimers Dis 9, 361–366.
Sato, K., Kashiwaya, Y., Keon, C.A., Tsuchiya, N., King,
M.T., Radda, G.K., Chance, B., Clarke, K., and
Veech, R.L. (1995). Insulin, ketone bodies, and
mitochondrial energy transduction. FASEB J 9,
651–658.
Schenk, D. (2002). Amyloid-beta immunotherapy for
Alzheimer’s disease: the end of the beginning. Nat
Rev Neurosci 3, 824–828.
Scheuermann-Freestone, M., Neubauer, S., and Clarke,
K. (2004). Abnormal cardiac muscle function
in heart failure is related to insulin resistance.
Cardiovasc J S Afr 15, S12.
Selkoe, D.J. (1997). Alzheimer’s disease:  genotypes,
phenotypes, and treatments. Science 275, 630–631.
Selkoe, D.J. (2001). Alzheimer’s disease:  genes, pro-
teins, and therapy. Physiol Rev 81, 741–766.
Shimazu, T., Hirschey, M.D., Newman, J., He, W.,
Shirakawa, K., Le Moan, N., Grueter, C.A.,
Lim, H., Saunders, L.R., Stevens, R.D., et  al.
(2013). Suppression of oxidative stress by beta-
hydroxybutyrate, an endogenous histone deacety-
lase inhibitor. Science 339, 211–214.
Steen, E., Terry, B.M., Rivera, E.J., Cannon, J.L., Neely,
T.R., Tavares, R., Xu, X.J., Wands, J.R., and de la
Monte, S.M. (2005). Impaired insulin and insulin-
like growth factor expression and signaling
mechanisms in Alzheimer’s disease: is this type 3
diabetes? J Alzheimers Dis 7, 63–80.
Strittmatter, W.J., and Roses, A.D. (1995).
Apolipoprotein E and Alzheimer disease. Proc
Natl Acad Sci USA 92, 4725–4727.
Szilard, L. (1959). On the nature of the aging process.
Proc Natl Acad Sci USA 45, 30–45.
Taegtmeyer, H., Hems, R., and Krebs, H.A. (1980).
Utilization of energy-providing substrates in the
isolated working rat heart. Biochem J 186, 701–711.
Talbot, K., Wang, H.Y., Kazi, H., Han, L.Y., Bakshi,
K.P., Stucky, A., Fuino, R.L., Kawaguchi, K.R.,
Samoyedny, A.J., Wilson, R.S., et  al. (2012).
Demonstrated brain insulin resistance in
Alzheimer’s disease patients is associated with
IGF-1 resistance, IRS-1 dysregulation, and cogni-
tive decline. J Clin Invest 122, 1316–1338.
Taylor, S.I., and Jungas, R.L. (1974). Regulation of lipo-
genesis in adipose tissue:  the significance of the
activation of pyruvate dehydrogenase by insulin.
Arch Biochem Biophys 164, 12–19.
Taylor, S.I., Mukherjee, C., and Jungas, R.L. (1975).
Regulation of pyruvate dehydrogenase in iso-
lated rat liver mitochondria. Effects of octanoate,
oxidation-reduction state, and adenosine triphos-
phate to adenosine diphosphate ratio. J Biol Chem
250, 2028–2035.
Thies, W., and Bleiler, L. (2013). 2013 Alzheimer’s
disease facts and figures. Alzheimers Dement 9,
208–245.
Veech, R.L. (2014). Ketone ester effects on metabolism
and transcription. J Lipid Res 55, 2004–2008.
Veech, R.L., Chance, B., Kashiwaya, Y., Lardy, H.A.,
and Cahill, G.F., Jr. (2001). Ketone bodies, poten-
tial therapeutic uses. IUBMB Life 51, 241–247.
Wang, X., Wang, W., Li, L., Perry, G., Lee, H.G., and
Zhu, X. (2013). Oxidative stress and mitochon-
drial dysfunction in Alzheimer’s disease. Biochim
Biophys Acta 1842(8), 1240–1247.
Willette, A.A., Bendlin, B.B., Starks, E.J., Birdsill, A.C.,
Johnson, S.C., Christian, B.T., Okonkwo, O.C., La
Rue, A., Hermann, B.P., Koscik, R.L., et al. (2015).
253
Association of insulin resistance with cerebral
glucose uptake in late middle-aged adults at risk
for Alzheimer disease. JAMA Neurology 72(9),
1013–1020.
Williamson, J.R., and Krebs, H.A. (1961). Acetoacetate
as fuel of respiration in the perfused rat heart.
Biochem J 80, 540–547.
Chapter 26: Alzheimer’s Disease 253
Xu, W., Qiu, C., Gatz, M., Pedersen, N.L., Johansson,
B., and Fratiglioni, L. (2009). Mid- and late-life
diabetes in relation to the risk of dementia:  a
population-based twin study. Diabetes 58, 71–77.
Zhai, L., Ballinger, S.W., and Messina, J.L. (2011). Role
of reactive oxygen species in injury-induced insu-
lin resistance. Mol Endocrinol 25, 492–502.
254
27
Mitigation of Damage from Reactive Oxygen Species
and Ionizing Radiation by Ketone Body Esters
WILLIAM CURTIS*, MARTIN KEMPER, PHD,
A L E X A N D R A M I L L E R , P H D , R O B E R T PAW L O S K Y, P H D ,
M . T O D D K I N G, A N D R I C H A R D L . V E E C H , M D, P H D, D P H I L
INTRODUCTION
Free radicals play an important role in a number
of chronic degenerative diseases including auto-
immune disorders, aging (Harman, 1956), cata-
racts, rheumatoid arthritis, Parkinson’s disease
(Kashiwaya et  al., 2000), cardiovascular disease,
and other neurodegenerative diseases (Pham-Huy
et  al., 2008). However, the pathophysiology of
these diseases can be complex. The simplest form
of ROS-induced disease is radiation sickness.
ORIGINS OF REACTIVE
OXYGEN AND NITROGEN
SPECIES
Reactive oxygen species (ROS) and reactive nitro-
gen species (RNS) arise from numerous sources
including cellular metabolism, ionizing radiation,
and various enzymes. Once created, individual ROS
and RNS participate in a cascade of interconversions
(Figure  27.1) both enzymatic and not, ultimately
being detoxified by being converted into inert mol-
ecules or by damaging various biomolecules.
Basal production of ROS occurs in the mito-
chondria as a byproduct of normal metabolism by
the spontaneous oxidation of coenzyme Q semi-
quinone by oxygen, creating superoxide (O2•−)
(Chance et al., 1979).
Q • − + O2 → O2 • − + Q
Another common origin of ROS is radiation;
ionizing radiation with energy (hν) above 33eV
can ionize water, which makes up the vast major-
ity of human cell volume. This process occurs
with many radiation types including particles
* Authors contributed equally.
(e.g., α or β), electromagnetic (e.g., γ or X), and
neutrons. Radiation of sufficient energy will split
water into a hydroxyl radical (HO•), H+, and a free
electron, e–. These three species interact with a
variety of other molecules to form the cascade of
RNS and ROS shown in Figure 27.1.
Superoxide can also be generated without radi-
ation by NADPH oxidases (NOXs) in white blood
cells in response to pathogens (Bedard and Krause,
2007) or as an inducible response to inflammation
and cellular stress. Enzymes producing ROS are
also induced after exposure to radiation, particu-
larly the NOX family members DUOX 1 and 2,
which continue producing superoxide for a num-
ber of days post radiation (Ameziane-El-Hassani
et  al., 2015). This prolonged ROS production
leaves a therapeutic window for post-radiation
treatment that targets these molecules.
Numerous other enzymes produce other types
of ROS as part of their normal function, such
as monoamine oxidase (MAO) (EC 1.4.3.4), an
enzyme integral in monoaminergic neurotransmis-
sion and xenobiotic metabolism. This enzyme pro-
duces hydrogen peroxide as part of its breakdown
of monoamines, such as dopamine in this reac-
tion: Dopamine + H2O +O2 → Dopaldehyde + NH3 +
H2O2. This is an important issue in the treatment
of Parkinsonian patients, who are often given high
doses of l-dopa, which is known to increase ROS
load (Acquier et al., 2013) and can lead to further
loss of dopaminergic neurons, often seen in patients
on high doses of l-dopa (Yamato et al., 2010).
Figure 27.1 provides a roadmap showing the
cascade of ROS and RNS. Superoxide plays a central
role in the cascade. In the case of radiation, super-
oxide results from the free electron combining with
oxygen. The enzyme superoxide dismutase (SOD)
catalyzes a reaction where two superoxide ions
255

Chapter 27: Mitigation of Damage from Reactive Oxygen Species 255

2H2O + O2
e h
las
O2 ta H+ + e–
ca
+ 1e–

2
D

X
e–

SO
O2 – H2O2 HO H2O
O2 X2
oxygen superoxide hydrogen peroxide hydroxyl radical
+ 2H+ (ROS)
(ROS) (ROS)
+ 1e– + 1e– + 1e–
Fe2+ Fe3+ + OH–
2 GSH NADP+
GSH peroxidase GSH reductase
GSSG NADPH damage DNA/RNA
2H2O
damage Protein

NO H+ damage Lipids
Nitric oxide ONOOH
peroxynitrous
(RNS) acid
(RNS)
– ONOOCO2– –
ONOO Nitrosoperoxy- CO3 + NO2
peroxynitrite carbonate carbonate nitrogen dioxide
(RNS) CO2 (RNS) radical (RNS)

Protein modification
Tyr, Cys, Met
Zn finger
Hemes

FIGURE 27.1 The origin, reactions, and target molecules of reactive oxygen species (ROS) and reactive nitrogen species
(RNS) are shown. Free radicals include superoxide, hydroxyl and nitric oxide, carbonate radical, and nitrogen dioxide. The
oxidation steps show the sequence of four single-electron reductions (addition of one electron) from oxygen to superoxide
to peroxide to hydroxyl to water. The outlined arrow +1e− indicates reduction. The reaction of ionizing radiation with
water is represented by hν, where h is Planck’s constant and the Greek letter ν is the frequency, which gives the energy
of the electromagnetic radiation by E = hν where h = 6.626 × 10–34 Js. Abbreviations: e− —electron, SOD—superoxide
dismutase, GSH—glutathione, GSSG—glutathione disulfide, TYR—tyrosine, MET—methionine, CYS—cysteine.

react to produce oxygen and hydrogen peroxide.
The hydrogen peroxide can decompose to hydroxyl
radical and hydroxide in the presence of iron and
other metals via the Fenton reaction. Superoxide
reacts with nitric oxide (NO) to produce peroxyni-
trite and begins the cascade of RNS. The reaction of
NO with superoxide is diffusion limited and non-
enzymatic. This is especially problematic as NO
is uncharged, enabling it to pass quickly through
cellular membranes, and possesses a relatively long
half-life. The kinetics of the reaction of NO with
superoxide make the production of peroxynitrite
inevitable whenever NO is produced within a few
cell diameters of superoxide (Pacher et  al., 2007).
Peroxynitrite in turn can either react directly with
CO2 to form nitrosoperoxycarbonate, which disso-
ciates to form other radical species, or be proton-
ated to form peroxynitrous acid, which dissociates
into hydroxyl radical and nitrogen dioxide radical.
The formation of hydroxyl radical from peroxyni-
trite may contribute more to the formation of free
radicals than the iron catalyzed Fenton reaction
(Pacher et al., 2007).
Oxygen-containing free radicals and other
ROS or their nitrogenous counterparts perform
useful functions in cell signaling and physiol-
ogy (Pham-Huy et  al., 2008). While the causes
of excessive ROS and RNS in diseases may have
complex or indeterminate origins, the expo-
sure to radiation provides a model with a single
known initiator. Generation of ROS/RNS is a
component of many common diseases including
cataracts, rheumatoid arthritis, neurodegenera-
tive diseases such as amyotrophic lateral scle-
rosis (ALS) and Parkinson’s (Kashiwaya et  al.,
2000)  and Alzheimer’s disease, cancer (Pham-
Huy et  al., 2008), and aging (Harman, 1956).
Table 27.1 summarizes reactions that generate or
destroy RNS and ROS.
D A M A G E C AU S E D
BY REACTIVE OXYGEN
AND NITROGEN SPECIES
Free radicals attack nucleic acids, proteins, and
lipids indiscriminately. Nucleic acids are attacked
either at the sugar (ribose or deoxyribose) or the
256

256 section III: Ketogenic Diet in the Laboratory

or within one protein to form a disulfide bridge

TABLE 27.1 THE GENERATION AND
and change the folding of the protein. This can be
DESTRUCTION OF RNS AND ROS reversed as a protein disulfide bond is reduced by
ROS/RNS formation ROS/RNS Catalyst thioredoxin reduction reaction shown in Figure 27.4.
reactions destruction However, further oxidation leads to SOH, SO2H
reactions
and SO3H, which makes the protein permanently
damaged.
H2O + hν → HO• + hν Damage to lipids is primarily to polyun-
H+ + e– saturated fatty acids (PUFAs). There are three
O2 + e– → O2•– NOX, xanthine stages, initiation, propagation, and termination.
oxidase Initiation causes the PUFA to become a free radi-
Fe2+ + H2O2 → Fe3+ + spontaneous cal. Propagation generates other free radicals, oxy-
HO• + OH– gen is often involved. Oxygen is a dual free radical,
O2•– + NO• → spontaneous and one might draw it as •O2• to show the two
ONOO– unpaired electrons. Being a dual radical makes
HO• + O2•– → spontaneous •O2• especially reactive with transition metals
OH– + O2 that are free radicals and also other free radicals.
2 O2•– + 2H+ → Superoxide Having the two unpaired electrons allows •O2• to
H2O2 + O2 dismutase propagate damage as in R• + O2 → ROO• → further
2 H2O2 → O2 + Catalase reactions. NO• being a single free radical would
2H2O result in a terminating reaction with other free
2GSH + H2O2 Glutathione
radicals as RS• + NO• → RS−NO.
→ GSSG + peroxidase
Peroxynitrite (ONOO−) is very reactive but it
2H2O
is not a free radical, therefore it tends to modify
proteins on specific residues and motifs such as
methionine, tyrosine, zinc fingers, and heme (see
base that is responsible for pairing. If hydroxyl Figure 27.1).
reacts with the sugar, it results in strand breaks,
recombination, chromosomal changes and muta- THE ROLE OF REACTIVE
tions. If hydroxyl radical reacts with the base, it OXYGEN SPECIES IN AGING
adds alcohol groups, OH, to the base where none In 1956, Denham Harmon, working in the Donner
belong and interferes with the hydrogen bonding Biophysics Lab at Berkeley, wrote a paper postulating
and fit of the matching base pair, leading to point that the process of aging was the result of free radicals
mutations. Modified guanine interferes with the produced either endogenously or by environmen-
telomerase binding proteins TRF1 and TRF2 and tal sources (Harman, 1956). In 1961 Hayflick and
causes a loss of telomere function, which requires Moorman showed that cultured cells could undergo
these binding proteins for both the protective shel- only a finite number of replications for reasons that
terin complex and for telomerase to interact with were unclear. In 1973 Olovnikov postulated that the
shelterin and extend the telomere (Sfeir, 2012). See end replication of DNA would result in the short-
for example Figure 27.2. ening of that strand at every replication (Olovnikov,
Proteins have various reactions depending on 1973). In 1978, Elizabeth Blackburn published a
the amino acid. One that is vulnerable is cysteine. paper describing a special form of DNA at the end
Oxidative stress can cause cysteine on two proteins of chromosomes, which is the telomere (Blackburn

O O O
•
HN N HN N OH Oxidation N
HN
+ HO• OH
H2N N H2N N N H N
N H2N N
R R R
Guanine (G) C8-OH-adduct radical of 8-hydroxyquanine
guanine (8-OH-G)

FIGURE 27.2 Guanine reacts with hydroxyl radical to produce 8-OH-G. There are three guanines in the telomere
repeat sequence, TTAGGG, making it vulnerable to oxidative damage. R is deoxyribose.
257
Chapter 27: Mitigation of Damage from Reactive Oxygen Species
and Gall, 1978). In 1990 it was found that the telo-
mere of human fibroblasts shortened with aging
(Harley et al., 1990). This sequence seemed to pro-
vide a mechanism to explain the development of
senescence, at least in culture.
Cell senescence can be triggered by a number
of mechanisms including telomere uncapping,
DNA damage, oxidative stress, oncogene activ-
ity, lack of nutrients, and other factors via various
signaling pathways (Ben-Porath and Weinberg,
2005). A  number of studies reported that oxida-
tive stress shortened telomeres (von Zglinicki,
2002). The formation of 8-oxo-guanine lesions in
telomeric tandem repeats by ROS-damaged telo-
meric protein binding to telomeres (Opresko et al.,
2005) impaired their functioning, causing replica-
tion to cease.
In the absence of disease, ROS damage depends
on the rate of metabolism and with it, the rate of
ROS production by mitochondria, which roughly
correlate with the life span (Speakman, 2005), a
phenomenon known as scaling. From these and
other observations it was thought that the rate of
metabolism inversely correlated with life span and
slowing it by measures such as caloric restriction
would lengthen life span (Sohal and Weindruch,
1996), in part by limiting mitochondrial ROS
generation and hence ROS damage. Studies also
appeared which reported that overexpression of
the antioxidant enzyme catalase increased life
span in transgenic mice (Schriner et  al., 2005).
Other antioxidant enzymes also increased life span
in Drosophila (Orr et al., 2013).
The most important system providing antioxi-
dants and defending against ROS is the NADPH
defense system described later in this chapter. The
preceding reports show that ROS can accelerate
telomere shortening, damage DNA directly, induce
senescence, and hasten the aging process (Correia-
Melo et al., 2014). Suppressing ROS may slow the
shortening of telomeres and decrease activation
of other mechanisms implicated in the pathology
of aging (Harman, 1956, 1981). We postulate that
using ketone body esters to suppress ROS would
be effective in retarding the aging process itself.
Telomere shortening occurs during normal
aging, leading to uncapping, activating senes-
cence and apoptotic programs in tissues with
high cell turnover. Rapidly dividing cells—gut,
skin, hematopoietic, and immune cells—replicate
constantly during life, and aging is visible to any-
body in skin as it changes from thick and smooth
in youth to thin and wrinkled in the aged. The
effects of nuclear radiation are likewise ampli-
fied in these tissues, exposure in the range of
257
2 to 5 Gy suppresses replication of blood and gut
cells through generation of ROS, causing death in
weeks. In these rapidly dividing cell types, a telo-
meric process as described by Hayflick could play
a more dominant role.
There is, however, another large group of cells
that undergo almost no replication after embry-
onic life. They include heart, skeletal muscle, and
brain cells. The death of these slowly dividing cells
does not have a unitary mechanism. The gradual
onset of senescence that characterizes normal
aging may in part be due to the accumulation of
genetic damage over a lifetime (Szilard, 1959).
Mice lacking the gene for dystrophin (MDX)
demonstrate little cardiac abnormalities. However,
mice in which the mRNA component of telomer-
ase, mTR, was deleted show progressive shortening
of telomeres with age. Crossing these two strains
to form the mdx/mTRko mouse with deleted dys-
trophin and TERC produced a mouse with heart
abnormalities, mitochondrial dysfunction, and
oxidative stress with reduced levels of the mito-
chondrial regulators peroxisome proliferator acti-
vated receptor gamma coactivator 1 α and 1 β
(PGC1-α and PGC1-β), both reduced by 80%. The
pathological changes in hearts of these animals
were ameliorated by administration of antioxidants
(Mourkioti et al., 2013). These findings show that
both mitochondrial ROS production and telomere
shortening play a role in the death and dysfunc-
tion of non-rapidly dividing cells such as heart cells
and that these functional abnormalities induced in
mdx/mTRko can be reduced by administration of
antioxidants.
The toxicity from radiation is therefore both
acute and chronic. Acute death occurs from effects
on the central nervous system (CNS) at doses
of 20 Gy, or slow, long-term accumulated DNA
damage leading to cell death from multiple hits
(Szilard, 1959). An earlier speculation by Erwin
Schrodinger attributed life to the maintenance
of negative entropy by homeostatic processes
(Schrodinger, 1944). In this view, aging and death
would be the result of an increase in entropy. This
general statement without a biochemical or physi-
ological mechanism is of limited practical use,
whereas applications of thermodynamics as exem-
plified by the energetics of the various redox states
and the process of oxidative phosphorylation have
heuristic value.
H OW A N T I OX I DA N T S   WO R K
There is a basic pattern to how antioxidant small
molecules work. They follow the pattern of reac-
tion shown in Figure 27.3.
258
258 section III: Ketogenic Diet in the Laboratory
Antioxidant + R•
HO
+ R•
O O
Tocopherol
HO
HO
O O + R•
H H
HO OH
Ascorbic Acid
FIGURE 27.3 The basic pattern of antioxidant reactions results in an antioxidant radical, which must be reduced
before it can participate in additional reactions. The antioxidant free radicals are sometimes obscured as they are the first
step in a reaction mechanism that may go to other forms.
The antioxidant is oxidized and the free radical
is reduced. Why then is it not a problem that the
antioxidant has become a free radical? The answer
is that the unpaired electron in the now antioxi-
dant radical is delocalized and thus less reactive.
This property makes ascorbate and tocopherol
distinct from other cellular reducing agents, as
they remain stable compounds after a single elec-
tron transfer. This is in contrast to glutathione, a
powerful and important reducing agent, but one
that must transfer two electrons at once to remain
stable, as the glutathione radical is highly reactive.
Tocopherol is able to transfer a single electron but
only a single electron, ascorbate is able to trans-
fer two—and, importantly, two ascorbate radicals
are able to react, forming a single molecule of
dihydroascorbate and one molecule of ascorbate.
Dihydroascorbate in turn is able to accept two
electrons from glutathione and be regenerated,
bridging single electron redox couples with two
electron couples. Glutathione itself is reduced by
the free [NADP+]/[NADPH] redox couple. It is
important to understand factors controlling the
cellular redox states to understand how the anti-
oxidants get recharged (reduced).
THE CELLULAR REDOX
S TAT E S D I F F E R F R O M   T H E
S TA N D A R D R E D O X
P OT E N T I A L S
A great deal of confusion in the literature on oxy-
gen free radicals results from the use of tables of
standard reduction potentials, ΔEo, done at pH 7.0
with all reactants at 1 M.
RH + Antioxidant •
O
RH +
Tocopheroxyl Radical
HO
HO
RH + O
O
O O
Ascorbyl Radical
The redox state of the cytoplasmic NAD-
couple in vivo was first defined in 1955 by Lynen
and his coworkers (Holzer et  al., 1956)  in yeast
by measurement of the ratio of the metabolites
[ethanol]/[acetaldehyde] in a reaction brought to
near-equilibrium by the very active cytoplasmic
enzyme alcohol dehydrogenase. Soon thereafter it
was shown by Bucher and Klingenberg (1958) that
the free [NAD+]/[NADH] in rat liver cytoplasm
could be estimated by measurement of either
the [lactate]/[pyruvate] or [dihydroxyacetone-P]/
[3-glycerophosphate] ratio. Because of the rapid-
ity with which these cellular redox state estimates
change with hypoxia, accurate determinations
were facilitated by the development of freeze
clamping (Wollenberger et  al., 1960). That ratio
of [lactate]/[pyruvate] was found to be between 5
and 10, giving a free cytosolic [NAD+]/[NADH]
calculated from the measured Keq of the reac-
tion to be about 1,000 in the fed liver and 250
in a starved liver. Following these principles,
in 1967 Krebs and his coworkers (Williamson
1967)  showed that the free [NAD+]/[NADH]
ratio in mitochondria could be estimated by
measurement of ether the [acetoacetate]/[β-d-
hydroxybutyrate] or [α-ketoglutarate]×[NH4+]/
[glutamate] ratio and the ratio of free mitochon-
drial [NAD+ ]/[NADH] was found to be between
2 to 10. Following a question to Krebs by George
Cahill, Krebs assigned Richard Veech to deter-
mine the free cytoplasmic [NADP+]/[NADPH]
ratio, which was known be used in the reduc-
tive synthesis of fats. This ratio was found to be
about 0.01 to 0.02 and could be estimated by
259

Chapter 27: Mitigation of Damage from Reactive Oxygen Species 259

TABLE 27.2 THE ACTIVIT Y OF NADP- LINKED ENZYMES IN RAT LIVER

Enzyme Activity in Keq = Metabolite couple Keq value

µmol/min/g

Isocitrate DH 22 ⎡αKG2− ⎤⎦ [CO2 ][ NADPH ] 1.17 M

EC 1.1.1.41 K eq = ⎣
⎡⎣Isocitrate3− ⎤⎦ ⎡⎣ NADP + ⎤⎦
Glutathione Reductase 7
K eq =
[GSSG][NADPH] ⎡⎣H+ ⎤⎦ 100 × 10–7 (Scott et al.,
EC 1.6.4.2 1963)
[GSH]2 ⎣⎡ NADP+ ⎤⎦
6 P Gluconate DH EC 2.8 ⎡Ribulose 5P2− ⎤⎦ [CO2 ][ NADPH ] 1.72 × 10–1 M (Villet and
1.1.1.44 K eq = ⎣ Dalziel, 1969)
⎡⎣6P Gluconate3− ⎤⎦ ⎡⎣ NADP+ ⎤⎦
Glucose 6-P DH EC 1.4 Not applicable because of lactonase
1.1.1.49
Malic Enzyme EC 1.27 ⎡ pyruvate− ⎤⎦ [CO2 ][ NADPH ] 3.44 × 10–2 M
1.1.1.39 K eq = ⎣
⎡⎣ Malate2− ⎤⎦ ⎡⎣ NADP + ⎤⎦

Source: Veech et al. (1969)

measurement of [α-ketoglutarate]×[CO2]/[iso- two major intracellular antioxidants, glutathione

citrate] or several other couples (see Table 27.2)
The standard potential of the NAD- couple is
about −0.32 V, but these couples have three
potentials in the cell depending on the intracel-
lular location of the enzymes involved. The three
redox couples have redox potentials appropriate
to the metabolic pathways they drive.
First, in cytoplasm, the ΔE of the free NAD-
couple is −0.19 V, this smaller potential pois-
ing it to receive reducing power from glycolysis.
Second, for NAD in mitochondria, it is −0.28 V,
allowing for more energy to be available for the
electron transport pathway to convert ADP to
ATP. Third, the NADP- couple in the cytosol was
estimated to be about −0.42 V, the lowest reduc-
tion potential in the cell (Krebs & Veech 1969).
This strongly reducing potential allows this couple
not only to synthesize fats but also to favor the
inactivation of oxygen free radicals through the
(GSH) and ascorbic acid (Vitamin C). As these
antioxidants are in near equilibrium with the free
cytoplasmic [NADP+]/[NADPH] couple, whose
large reducing potential favors the reduced, active
forms of GSH and Vitamin C (see Table 27.2 for
a list of these redox couples and Figure 27.4 for a
diagram of the ROS-quenching reactions driven
by the NADP couple).
These major cellular redox states are related
to one another through common intermedi-
ary metabolites (Bergman et  al., 2010; Krebs,
1969)  and also to the cellular phosphorylation
potential or ATP/ADPxPi ratio keeping the energy
of ATP hydrolysis in the range of −53 to −57 kJ/
mole (Veech et al., 1979). It can be seen that these
intracellular redox potentials cannot be inferred by
study to the standard potential of redox half reac-
tions, which have led to much confusion in writ-
ing about cellular redox potentials (Winterbourn,

S
NADP+ TrxR-S2 TrxR Trx-S2 Protein-S2
FAD Se

SH
NADPH
+
TrxR-(SH)2 TrxR Trx-(SH)2 Protein-SH2
+H
FAD Se–

FIGURE  27.4 Thioredoxin is a family of three proteins in mammals—TrxR1 (cytosolic), TrxR2 (mitochondrial),
TrxR3 (testis specific)—that can reduce protein disulfide bonds formed by ROS/RNS damage. Abbreviations: TrxR—
thioredoxin, FAD—flavin adenine dinucleotide, S2—disulfide, SH2—two sulfhydryl moieties.
260

260 section III: Ketogenic Diet in the Laboratory

2008). This history of the development of the mea- For [pyruvate]/[lactate] = 1:10

surement of in vivo redox states has been reviewed
(Veech, 2005, 2006). E NAD/NADH = −0.32V + 0.091V = −0.229V
Measurement of each of the metabolite cou-
ples listed in Table  27.2 yields the same value of The free cytosolic [NADP+]/[NADPH] redox
the free cytosolic [NADP+]/[NADPH] ratio with a potential calculated from isocitrate dehydroge-
redox potential of about −0.4 V, the most negative nase reaction and from the malic enzyme agree
of any intracellular potential (Krebs, 1969). With well and are the most negative potential in the
small refinements of the Keq, CO2 concentrations cell, although with improved metabolite mea-
and intracellular metabolite concentrations in the surements are slightly above the −0.42V esti-
intervening years, the potential of the NADP sys- mated in the original Krebs and Veech paper
tem is more likely to be in the range of −0.36 to (Krebs, 1969).
−0.38 V, but is still the most negative potential in The redox potential from isocitrate dehydroge-
the cell. nase reaction:
The formalism to calculate the redox potential
in the cell estimated from measured intracellular RT ⎛ NADP + ⎞
metabolites and the measured Keq of the reac- E NADP / NADPH = Eo NADP / NADPH + ln
tions involved is as follows. The standard redox nF ⎜⎝ NADPH ⎟⎠
potential is derived from the equilibrium con-
stant of the reaction. The standard potential of [Isocitrate−3 ] + [NADP + ] → [α-Ketoglutarate2 − ]
both the NAD and NADP couples, Eo, is, within + [NADPH] +[CO2 ]
experimental error, −0.32 V (Burton, 1974).
The redox potential within the cell is calculated
at follows (Veech, 2005): [NADP + ] [αKG2 − ] × [CO2 ] 1
= 3−
×
[NADPH] [Isocit ] K ICDH
RT ⎛ NAD+ ⎞
E NAD / NADH = Eo NAD / NADH + ln Eqn 1
nF ⎜⎝ NADH ⎟⎠
RT ⎛ [αKG2− ] × [CO2 ] 1 ⎞
E = Eo + ln ×
Where R = 8.1345 × 10−3 kJ/mole/oK
NADP
NADPH
NADP
NADPH
nF ⎜⎝ [Isocit 3− ] K ICDH ⎟⎠
T = 311.15 oK
n = number of electrons = 2 for lactate For [α-Ketoglutarate]:[Isocitrate] = 7:1 (Liver) and
dehydrogenase [CO2] = 1.985mM
F = 96.485 kJ/mole/V
E NADP = −0.32V − 0.044 V = −0.364 V
The free cytosolic [NAD+ ]/[NADH] calculated NADPH
from the lactate dehydrogenase reaction:
The redox potential from from the malic enzyme
[lactate − ] + [ NAD + ] → [pyruvate − ]
reaction:
+ [ NADH] + [H + ]
[NADP+ ] [Pyr − ] × [CO2 ] 1
= ×
At pH 7 [NADPH] [Mal 2 − ] K MalEnz

[Pyr − ] [NADH]
× = K LDH =1.11 × 10−4 where [pyruvate]/[malate] = 1:3 in liver
[Lac − ] [NAD+ ]
E NADP = −0.32V − 0.055V = −0.375V
Rearranged and substituted into equation 1 above NADPH

RT ⎛ [Pyr − ] 1 ⎞ Because the glutathione couple reacts with the

E NAD / NADH = Eo NAD / NADH + ln ⎜ × NADP- couple (see Table 27.3), the potential of the
nF ⎝ [Lac ] K LDH ⎟⎠
−
NADP- couple determines the redox potential of
the glutathione redox couple [GSH]2/[GSSG]) with
For [pyruvate]/[lactate] = 1:1 which it is in near-equilibrium at −0.37V, not the
standard potential of −0.32V at pH 7.0. The termi-
E NAD / NADH = −0.32V + 0.122V = −0.198V nal reduction of the oxygen free radicals involves
261

Chapter 27: Mitigation of Damage from Reactive Oxygen Species 261

TABLE 27.3 ENZYMATIC SYSTEMS WHICH ASSUME THE REDOX POTENTIAL

OF THE NADP- SYSTEM

Enzyme Reaction

Glutathione Reductase EC 1.6.4.2 NADPH + GSSG + H + → NADP + + 2GSH

Glutathione Dehydrogenase 2GSH + Dehydroascorbate → GSSG+ Ascorbate

EC 1.8.5.1

Glutathione Peroxidase EC 1.11.1.9 2GSH + H2O2 → GSSG + H2O

Glutathione Transferase EC 2.5.1.8 HO–Xenobiotic + GSH → GS–Xenobiotic + H2O

Thioredoxin Reductase EC 1.6.4.5 NADPH + H + + Oxidized Thioredoxin → NADP+ + Reduced Thioredoxin

glutathione, which constitutes the most abundant
intracellular antioxidant at 5 mM (Krebs, 1969).
Ascorbate, at 2–3 mM, is the second most
abundant reducing agent and antioxidant, and is
also the cofactor for a number of metal-dependent
oxygenases and dioxygenases, such as collagen
prolyl and lysyl hydroxylases (Linster and Van
Schaftingen, 2007). Ascorbate, when acting as a
reducing agent, loses an electron and a proton to
form semidehydroascorbate (SDA), which upon
losing another electron forms dehydroascorbate
(DHA). Regeneration of ascorbate by glutathione
occurs by the reaction: DHA + 2GSH ↔ ascorbate +
GSSG, which can occur spontaneously (Linster
and Van Schaftingen, 2007)  but can also be cata-
lyzed by the enzyme glutathione dehydrogenase
(EC 1.8.5.1), an enzyme with an activity in rat liver
of about 3 µmol/min/g wet weight (Carlberg and
Mannervik, 1975). These reactions (Table 27.3)
ensure that the redox potential of ascorbic acid and
the glutathione couple in vivo assume the redox
potential of the cytosolic NADP- couple of −0.37
V and not the standard potential—this is in sharp
contrast to ascorbate’s standard reduction potential
of −0.06 V (Table  27.4) or the standard potential
of the glutathione couple at −0.23 V (Clark, 1960).
The semidehydroascorbte (SDA) is also
reduced to ascorbate by thioredoxin reductase
(EC 1.6.4.5) (Linster and Van Schaftingen, 2007),
an NADP-linked enzyme that is the major cellular
protein disulfide reductase (Arner and Holmgren,
2000). This enzyme catalyzes the reaction: NADPH
+ thioredoxinox → NADP+ + thioredoxinred. This
enzyme is critical in signaling as part of thiol redox
control, and regulates the activity of a number of
transcription factors. The potential of the thiore-
doxin system is likewise set by the potential of the
free cytosolic [NADP+]/[NADPH] couple.
Thus, the potential of the ascorbate couple and
the glutathione couple are the same as that of the
NADP- couple. Likewise, disulfide (SS) groups
in protein are reduced to –SH by the thioredoxin
system.
T H E R E D O X S TAT E
OF OXYGEN FREE RADICALS
A N D O T H E R R E L E VA N T
R E A C TA N T S
A list of relevant redox potentials is listed in
Table 27.4 below. The values for the redox poten-
tials given are taken from Bergman et  al. (2010),
Clark (1960), Dawes (1971), and Krebs (1969).
This table shows the oxidizing potential of
ROS compounds. However, it should be pointed
out that these half reactions do not represent the
redox potential inside of cells, which are set by the
NAD and NADP couples which carry the domi-
nant metabolic flux during metabolism.
H O W D - Β H B A LT E R S
C E L L U L A R R E D O X   S TAT E S
The effects of insulin, glucose, and d-β-
hydroxybutyrate (d-βHB) on a perfused rat heart
gave insights into the changes in relative concen-
trations of key metabolites involved in the Krebs
cycle and oxidative phosphorylation and the free
[NADP+]/[NADPH] ratio. Prior work on the
effects of hearts perfused with d-βHB showed
that the [isocitrate]/[α-ketoglutarate] (Sato et  al.,
1995) couple was increased. This change indicated
a reduction of the NADP- couple (Kashiwaya
et al., 1997). Figure 27.5 shows the increased ratio
of [NADPH]/[NADP+] found in the perfused rat
heart studies, which supports a highly significant
increase in the capacity for cellular ROS/RNS
detoxification. (We are intentionally breaking a
262

TABLE 27.4 REDOX POTENTIAL OF SOME BIOCHEMICALLY IMPORTANT

HALF- REACTIONS IN THE BIOCHEMICAL HALF REACTIONS, ALL
REACTANTS ARE AT 1 M EXCEPT H + WHICH IS AT PH 7.

Half reaction Redox Potential of half reactions

H + + e − → 1 2 H2 -0.41

Acetyl − CoA + 2H + + 2e − → Acetaldehyde + CoA –0.42

Acetoacetate− + 2H + 2e − → β − hydroxybutyrate − –0.36

+ + − +
NAD + 2H + 2e → NADH+H –0.32

NADP + + 2H + + 2e − → NADPH+H + –0.32

2− + − 3−
α ketoglutarate + CO2 + 2H + 2e → isocitrate –0.318

Pyruvate + 2H + + 2e − → Lactate –0.19

FAD + 2H + + 2e − → FADH2 –0.06

Fumarate + 2H + + 2e − → Succinate +0.031

Dehydroascorbate + 2H + 2e → Ascorbate + −
+0.06
2− 3−
α − ketoglutarate + CO2 → isocitrate +0.002
+
Ubiquinone + 2H + 2e → Dihydroubiquinone −
+0.10
−
1
2 O 2 + H 2 O + 2e → H 2O 2 +0.30

O2 + 2H + + 2e − → H 2 O2 +0.605
3+ − 2+
Fe + e → Fe +0.77
+ −
2 O 2 + H + 2e → H 2O +0.816
1

− +
O2 + 2H → H 2 O2 +0.89
− −
•OH + e → OH +1.89

1500
Glucose
Insulin
1000 Ketones
Insulin + Ketones
Ratio

500

0
[NADH]/[NAD+] [NADPH]/[NADP+]
(x10–5)

FIGURE  27.5 Studies of a perfused rat heart demonstrated that the ketone body D-βOHB raised the ratio of
[NADPH]/[NADP+]. Modified presentation of data published in Kashiwaya et al. 1997.
263

Chapter 27: Mitigation of Damage from Reactive Oxygen Species 263

D-β-hydroxybutyrate NAD+
H2O2
βHBDH 2GSH
glutathione peroxidase
acetoacetate NADH
3 oxoacid 2H2O
CoA transferase
acetyl CoA
citrate synthase oxaloacetate Vit Cox
H2O
citrate HO•
aconitase NADP+ R•
glutathione
dehydrogenase Vit Ered
isocitrate RH
glutathione
Isocitrate dehydrogenase reductase
LOO•
α-ketoglutarate tetrahydrobiopterin
Vit Cred Vit Eox
Dihydropiopterin reductase
LOOH
dihydrobiopterin ONOO–
2 protein-SH
thioredoxin
H2O + NO2
protein-SS-protein

NADPH GSSG

FIGURE 27.6 The “NADP system” is driven by the [NADP+]/[NADPH] couples redox potential. [NADPH] was
elevated relative to [NADP+] in perfused rat heart studies with d-bOHB. To the left of NADPH shown in the diagram
is the relevant pathway to reducing NADP+ to NADPH. NADPH also provides the reducing potential to counteract
ROS and RNS through antioxidants that must be reduced each time they are oxidized as they quench a ROS or RNS.
Abbreviations: NAD—nicotinamide adenine dinucleotide, NADH—reduced NAD+, NADP—nicotinamide adenine
dinucleotide phosphate, NADPH—reduced NADP+, GSH—glutathione, GSSG glutathione disulfide, Vit C—vitamin
C or ascorbate, Vit E—vitamin E or a tocopherol, R•—free radical on a carbon atom, RH—an organic molecule with
C-H bond, LOO•—lipid peroxyl radical, LOOH—lipid peroxide.

long-standing rule by putting the reduced form
in the numerator to simplify the discussion where
an increase in [NADP+]/[NADPH] would other-
wise be reported as a decrease in the inverse of the
ratio.)
Figure 27.6 shows some of the “NADP system”
that is relevant to removal of ROS and RNS as well
as the link back to d-βHB.
K E TO S I S A S   A M E A N S
O F   P ROT E C T I O N
AGAINST IONIZING
R A D I AT I O N
At present there are very few approved treatments
for exposure to nuclear accidents or rogue nuclear
bombs (Buddemeier, 2011). Experiments have
been conducted at the Armed Forces Radiobiology
Research Institute (AFRRI). The study investigat-
ing d-βHB protection against acute ROS injury
utilized in vitro radiation lethality on HOS
(human osteoblast) cells exposed to several doses
and types of radiation. For the studies described
below, a novel ester of D-β-hydroxybutyrate and
R-1,3-butanediol referred to as ketone ester (KE),
was used. This compound has been shown to safely
increase levels of blood d-βHB to 5–7 mM (Clarke
et  al., 2012a; Clarke et  al., 2012b). The ester is
hydrolyzed in the gut and blood by esterases, and
the resulting R-1,3 butane diol is converted to
d-βHB in the liver (Figure 27.7).
T H E R E M OVA L O F   F R E E
R A D I C A L S I N   V I VO
Clinical trials of administration of antioxidants
have often been disappointing (Winterbourn,
2008). Most placebo trials of antioxidants have
failed (Steinhubl, 2008), most notably in the
long-advocated use of high-dose ascorbic acid.
The use of vitamin antioxidants has not objec-
tively been proven helpful in the absence of vita-
min deficiency and may even be harmful (Howes,
2011). More recently, high doses of Vitamin E are
reported to increase melanoma metastases in mice
(LeGal, 2015).
While it is true that, quantitatively, glutathione
at about 5 mM and ascorbic acid at about 2–3 mM
264
264 section III: Ketogenic Diet in the Laboratory
OH O OH
esterases
O
(R)-((R)-3-hydroxybutyl) 3 -hydroxybutanoate
FIGURE 27.7 The monoester of D-βOHB and R-1,3 butanediol is (R)-((R)3- hydroxybutyl) 3- hydroxybutyrate is
converted to two molecules of D-βOHB and provides an alternative fuel to produce ATP.
in most cells are the predominant intracellular
antioxidants, one cannot change the redox poten-
tial of an antioxidant in vivo by simply adding
more of the reduced form of the couple.
R E M OVA L O F   R E A C T I V E
OXYGEN SPECIES
B Y   G E N E R AT I O N
OF NADPH
We have discussed earlier the failure to remove
ROS by addition of antioxidants such as ascorbic
acid (Steinhubl, 2008). It is likewise not possible
to add the reduced component of most NADPH
producing metabolite couples (see Table 27.2
above), since these multiply-charged com-
pounds do not readily enter the interior of cells.
It is, however, possible to introduce the reduced
ketone body, d-βHB, which readily enters cells
on the monocarboxylate transporter, MCT.
While not an NADP- coupled metabolite, d-βHB
produces acetyl CoA and, via the Krebs cycle,
increased concentrations of citrate and isoci-
trate, but causes little change in α-ketoglutarate
(Sato et  al., 1995). This increase in the [isoci-
trate]/[α-ketoglutarate ] ratio causes a reduc-
tion of the NADP- system (Kashiwaya et  al.,
1997) generated by the metabolism involving the
most active NADP linked dehydrogenase, isoci-
trate dehydrogenase. The cellular metabolic pro-
cesses thus produce an increased reducing power
of the NADP system, which is regenerated by the
cellular metabolism.
THE ENZYMES
D E S T R OY I N G R E A C T I V E
OXYGEN SPECIES
Superoxide dismutase also can combine two super-
oxide radicals, along with two protons to form the
less toxic hydrogen peroxide.
O2 − + O2 − + 2H + → H 2O2 + O2
OH O
OH
Gut
D-β-hydroxybutyrate Alcohol
+ dehydrogenase
OH OH
R 1, 3 butanediol
Hydrogen peroxide can react with catalase (EC
1.11.1.6) in cytoplasm, but not mitochondria, to
form water and O2.
2 H 2O 2 → O 2 + 2H 2O
Alternatively, H2O2 can react with glutathione in a
reaction catalyzed by glutathione peroxidase (EC
1.11. 1.9) to form oxidized glutathione and water.
2GSH + H 2O2 → GSSG + 2H 2O
THE EFFECTS OF D-ΒHB
ON TRANSCRIPTION
In addition to the ability of d-βHB metabolism
to generate NADPH, the metabolite d-βHB is the
natural inhibitor of class I and II histone deacety-
lases (Shimazu et al., 2013; Veech, 2014) that were
formerly identified by their ability to bind nico-
tinic acid (Offermanns, 2006; Tunaru et al., 2003).
The inhibition of histone deacetylase increases
transcription of the enzymes of the antioxidant
pathway controlled by the FOXO3A and Mt1
transcription factors, thus increasing the activ-
ity of Mn superoxide dismutase and catalase (see
Table 27.3). d-βHB thus increases the reductive
capacity of the pathways used to destroy free radi-
cals by increasing the enzymes in the pathway used
for free radical destruction. Recent studies have
shown that HDAC inhibitors may have a signifi-
cant benefit in radiation protections (Brown et al.,
2008; Miller et al., 2011). Therefore, in vitro studies
were initiated to test whether KE could improve
radiation survival and decrease radiation-induced
damage.
IN VITRO STUDIES
Human osteoblast cells (HOS) were chosen as the
model system to study KE’s potential as a radia-
tion countermeasure. The HOS cells have been
265
Chapter 27: Mitigation of Damage from Reactive Oxygen Species
(a) Pre f-Radiation
1 Control
KE
Surviving Fraction
0.1
0.01
0.001
0.0001
0 2 4 6 8 10 12
Radiation Dose (Gy)
FIGURE 27.8 Survival curves for human osteoblasts (HOS) in culture exposed to either gamma (60Co, 66 gCy/min)
or proton radiation (4MeV). KE-treated cell cultures received 50 µM of ketone ester (KE) in the cell media 30 minutes
prior to or for 30 minutes after radiation exposure.
previously used to evaluate potential radiation
countermeasures (Miller et al., 1998; Miller et al.,
2011). The HOS cells were incubated with KE
(50 µM, 15 min to 1 h) either before (Figure 27.8A)
or immediately after (Figure  27.8B) exposure to
gamma radiation (60Co, 0.66 cGy/min). Following
radiation, HOS cells were plated in 100-mm petri
dishes to assess survival of colony-forming cells; at
10 days post radiation, colonies were fixed, stained,
and counted to determine the survival fraction.
The survival curves show that KE increased sur-
vival when delivered either immediately before or
immediately after radiation. These data strongly
suggest that KE would have benefit to human
health both as a protectant delivered to popula-
tions at risk of exposure or as a mitigant after radi-
ation. The ability of KE to improve survival even
when given 24 hours after radiation is compatible
with the persistent production of superoxide from
NADPH oxidase s, DUOX 1 and 2, for several days
after radiation exposure (Ameziane-El-Hassani
et al., 2015).
Proton radiation was used to determine the
ability of the KE to protect against particle radia-
tion, which produces less oxidative damage than
gamma radiation but is more likely to induce
double-stranded DNA breaks (Lomax et al., 2013).
Cells were exposed to KE before proton radia-
tion exposure, similar to that performed above in
Figure 27.8A with gamma radiation. The KE was
able to confer significant protection against proton
radiation in this model (Figure  28.8C). Overall
cell death was higher for this type of radiation,
which is to be expected, based on previous studies.
Protection against proton radiation is of particular
interest to NASA, as astronauts have a substantially
265
(b) Post f-Radiation
1
Surviving Fraction
0.1
0.01
0.001
0.0001
0 2 4 6 8 10 12
Radiation Dose (Gy)
greater likelihood of exposure to proton radiation
than other populations.
Genomic effects of ROS begin as discrete events
such as a break in the phosphate backbone in DNA,
nitrogenous base oxidation, or dimerization.
Breaks in the phosphate bonds occur as either sin-
gle, clustered, or double-stranded breaks, depend-
ing on the level of the ROS or radiation dose and
type. As the frequency of breaks increases, macro
effects such as abnormal chromosomes, muta-
genesis, and inhibited cell division become more
frequent (Lomax et  al., 2013). In general, single-
stranded breaks are quickly repaired, have low
cytotoxicity and occur at high levels even without
exogenous ROS or radiation (Cadet et  al., 2008).
Single-stranded breaks are induced at levels four-
to eightfold more than double-stranded breaks
with exposure to electromagnetic radiation such
as gamma or x-rays (Lomax et al., 2013). Double-
stranded or clustered single-stranded breaks occur
with particle radiation, especially α and proton, or
when ROS is focused in a confined area (Lomax
et  al., 2013). A  double-stranded break is defined
as two breaks within 10 bases (Lomax et al., 2013).
These breaks are much more difficult to repair and
can persist for more than 24 hours when large
numbers are generated (Schmid et al., 2010). These
breaks are repaired by either homologous recombi-
nation or by nonhomologous end joining (Lomax
et al., 2013), with the latter being the predominant
mechanism in most phases of the cell cycle. It is
possible to observe these breaks by examining
some of the hallmarks of their repair, by observ-
ing different types of aberrant chromosomes, or by
observing cell cycle arrest. Indeed, when cultured
cells were treated with d-βHB before an exposure
266

266 section III: Ketogenic Diet in the Laboratory

Percentage Aberrant Metaphases Fragments

100 1.5
Control
80
∆G
1.0
60

40
0.5
20

0 0.0
0 1 2 3 4 5 0 1 2 3 4 5

Breaks Rings & Dicentrics

0.020 0.05

0.04
0.015
0.03
0.010
0.02
0.005
0.01

0.000 0.00
0 1 2 3 4 5 0 1 2 3 4 5
Radiation Dose (Gy) Radiation Dose (Gy)

FIGURE 27.9 The percentage of chromosomal anomalies (aberrant metaphases, fragments, breaks, and rings and
dicentrics) occurring from γ-radiation-exposed human osteoblast cells (HOS) with or without KE. The 50-µM KE was
added to the cell culture media 30 minutes prior to radiation exposure and removed immediately prior to radiation
exposure. Cells were scored using a light microscope.

to γ radiation, production of chromosomal aber-
rations and cell cycle arrest were both significantly
suppressed (Figure 27.9).
A novel approach for stimulating repair of
radiation or ROS-induced DNA damage utilizes
HDAC inhibitors. Inhibitors of HDAC activate
important DNA repair pathways such as the Nrf2
and Rad51 and decrease markers of damage, even
if the inhibitor is given after radiation exposure
(Brown et  al., 2008; Miller et  al., 2011). Histone
deacetylase (HDAC) inhibitors have broad activity,
their primary function is to increase the acetyla-
tion of histones by preventing their deacetylation,
in turn making the genome less tightly packed
and more transcriptionally active (Verdin and
Ott, 2015). The deacetylating activity of HDACs
can also extend to various other cellular proteins,
modulating activity of enzymes and protein-
protein binding (Verdin and Ott, 2015). With
huge numbers of potential effects resulting from
inhibiting HDACs, it is difficult to determine what
the final effectors of the protection are, but stud-
ies have demonstrated that a series of chemically
unrelated compounds that all target HDACs seem
to confer similar protection and enhanced repair
after exposure (Brown et  al., 2008; Miller et  al.,
2011; Shimazu et al., 2013).
The effects of persistent DNA damage are most
quickly seen in tissues with high cell turnover that
rely on actively dividing stem cells such as skin,
the gastrointestinal system, and blood. Damage
in these systems produces some of the rapid and
most visible radiation effects, such as hair loss,
nausea/vomiting, bone marrow suppression, and
skin lesions. Damage to DNA is also responsible
for much of the increased risk of cancer that results
from radiation exposure.
E F F E C T S O F   K E TO N E
ESTERS AFTER EXPOSURE
T O   R A D I AT I O N I N   V I VO
A project is underway to evaluate the ability of KE
to increase murine survival after gamma radia-
tion exposure. Previous studies have shown that
exposure of the DBA2 inbred mouse strain to 6
Gy of γ-radiation results in reproducible bone
267

Chapter 27: Mitigation of Damage from Reactive Oxygen Species 267

(a) (b)
110 100

90 Control
KE 80

Incidence per 1000 cells

Incidence per 100 cells
70 *
50 * 60
*
30
12 40
10 *
8 *
6 * 20
4 * * *
2
0 0
MPCE MNCE
s
Fr ells

dy

S
om ice s

eB s
Po aks
nt

D ak

om tric

SD
oi
om gme

re
tC

Br

pl
an

ly
id
a
rr

at
be

as
an
A

hr

gs
%

hr
in

C
R

FIGURE 27.10 The effect of KE on the incidence rate of chromosome anomalies and micronuclei in mouse bone
marrow 48 h after 6 Gy γ-radiation. KE (750 mg/kg) or saline were delivered by gavage 24 hours after radiation exposure.
(A) Chromosomal malformation incidence per 100 cells, all cells in bone marrow were examined and malformations were
scored using a light microscope. For metaphases approximately 1,000 total were counted, 200/animal. (B) Micronuclei
were counted only in reticulocytes and erythrocytes, 10,000 total were counted per animal. Data presented as average ±
SEM, N = 5 animals, *, p < .01 by t-test with a Sidack-Bonferonni multiple comparison correction.

marrow chromosomal damage in 85-day-old
mice (Grahn, 1958). In our study we have tested a
single KE administration of 750 mg/kg by gavage
24 hours after radiation exposure. This postradia-
tion treatment led to a consistent 50% decrease
in numerous types of radiation-induced chromo-
somal malformations and abnormal cell divisions
as observed by microscopy 24 hours after KE
doses of γ-radiation (Figure 27.11) in control ani-
mals. This decrease was significantly attenuated
by 750mg/kg KE 24 hours after radiation. Further
experiments to determine KE effects on radiation
lethality are ongoing.
1.2 Control
Reticulocyte: Erythrocyte Ratio

administration (Figure 27.10). The importance of KE

decreasing these indicators of DNA damage and 1.0
genomic instability is underscored by the cytotox-
icity of this type of damage and the potential for +
0.8
neoplastic transformation, ultimately resulting in
+
cancer. This dose of KE also blunted the radiation-
induced increase in incidence of micronuclei, 0.6 *
which arise from improper chromatid separation +
during anaphase, in both reticulocytes and eryth-
0.4
rocytes, which were counted using the methods of 0 2 4 6
Schmid (1975). -Radiation Dose (Gy)
In order to explore KE effects on radiation
bone marrow suppression and impact on hema- FIGURE  27.11 The effect of KE on the relative
topoiesis, an important mechanism for radiation- abundance of polychromatic and normochromatic
induced mortality at the 6 Gy level, the ratio erythrocytes in mouse bone marrow 48 h after 0, 0.5, or 6
between total reticulocytes and erythrocytes in Gy γ-radiation. KE (750 mg/kg) or saline were delivered
bone marrow was determined. Bone marrow was by gavage 24 hours after radiation exposure. 10,000
extracted as before, 24 hours after KE or saline erythrocytes were counted per animal. Data presented
gavage, a total of 48 hours after either 0.5 or 6 Gy as average ± SEM, N  =  5 animals. *, p < .01 significant
γ-radiation. This ratio, which is approximately difference between KE and control, +, p < 0.05 significant
1 in bone marrow under normal circumstances, difference versus no radiation, by t-test with a Sidack-
was significantly decreased by both 0.5 and 6 Gy Bonferonni multiple comparison correction.
268
268 section III: Ketogenic Diet in the Laboratory
CONCLUSION
Reactive oxygen species play a central role in both
radiation damage and aging. It is proposed that
both processes can be ameliorated by the admin-
istration of KE to increase the amount of anti-
oxidant, enzymes, and the reducing power of the
NADP- system.
REFERENCES
Acquier, A.B., Mori Sequeiros Garcia, M., Gorostizaga,
A.B., Paz, C., and Mendez, C.F. (2013). Reactive
oxygen species mediate dopamine-induced signal-
ing in renal proximal tubule cells. FEBS Lett 587,
3254–3260.
Ameziane-El-Hassani, R., Talbot, M., de Souza Dos
Santos, M.C., Al Ghuzlan, A., Hartl, D., Bidart,
J.M., De Deken, X., Miot, F., Diallo, I., de Vathaire,
F., et  al. (2015). NADPH oxidase DUOX1 pro-
motes long-term persistence of oxidative stress
after an exposure to irradiation. Proc Natl Acad
Sci U S A 112, 5051–5056.
Arner, E.S., and Holmgren, A. (2000). Physiological
functions of thioredoxin and thioredoxin reduc-
tase. Eur J Biochem 267, 6102–6109.
Bedard, K., and Krause, K.H. (2007). The NOX family
of ROS-generating NADPH oxidases:  physiology
and pathophysiology. Physiol Rev 87, 245–313.
Ben-Porath, I., and Weinberg, R.A. (2005). The signals
and pathways activating cellular senescence. Int J
Biochem Cell Biol 37, 961–976.
Bergman, C., Kashiwaya, Y., and Veech, R.L. (2010).
The effect of pH and free Mg2+ on ATP linked
enzymes and the calculation of Gibbs free energy of
ATP hydrolysis. J Phys Chem B 114, 16137–16146.
Blackburn, E.H., and Gall, J.G. (1978). A tandemly
repeated sequence at the termini of the extrachro-
mosomal ribosomal RNA genes in Tetrahymena. J
Mol Biol 120, 33–53.
Brown, S.L., Kolozsvary, A., Liu, J., Ryu, S., and Kim,
J.H. (2008). Histone deacetylase inhibitors protect
against and mitigate the lethality of total-body
irradiation in mice. Radiat Res 169, 474–478.
Bucher, T., and Klingenberg, M. (1958). Wege des
Wasserstoffs in der lebendigen Organisation.
Angewandte Chemie 70, 552–570.
Buddemeier, B.R. (2011). Key Response Planning Factors
for the Aftermath of Nuclear Terrorism (FEMA,
Homeland Security, Sandia National Laboratory,
Lawrence Livermore National Laboratory).
Burton, K. (1974). The enthalpy change for the reduc-
tion of nicotinamide-adenine dinucleotide.
Biochem J 143, 365–368.
Cadet, J., Douki, T., and Ravanat, J.L. (2008).
Oxidatively generated damage to the guanine moi-
ety of DNA: mechanistic aspects and formation in
cells. Acc Chem Res 41, 1075–1083.
Carlberg, I., and Mannervik, B. (1975). Purification
and characterization of the flavoenzyme
glutathione reductase from rat liver. J Biol Chem
250, 5475–5480.
Chance, B., Sies, H., and Boveris, A. (1979).
Hydroperoxide metabolism in mammalian organs.
Physiol Rev 59, 527–605.
Clark, W.M. (1960). Oxidation Reduction Potentials
of Organic Systems (Baltimore:  Williams and
Wilkins).
Clarke, K., Tchabanenko, K., Pawlosky, R., Carter, E.,
Knight, N.S., Murray, A.J., Cochlin, L.E., King,
M.T., Wong, A.W., Roberts, A., et al. (2012a). Oral
28-day and developmental toxicity studies of (R)-
3-hydroxybutyl (R)-3-hydroxybutyrate. Regul
Toxicol Pharmacol 63, 196–208.
Clarke, K., Tchabanenko, K., Pawlosky, R., Carter,
E., Todd King, M., Musa-Veloso, K., Ho, M.,
Roberts, A., Robertson, J., Vanitallie, T.B., et  al.
(2012b). Kinetics, safety and tolerability of (R)-
3-hydroxybutyl (R)-3-hydroxybutyrate in healthy
adult subjects. Regul Toxicol Pharmacol 63,
401–408.
Correia-Melo, C., Hewitt, G., and Passos, J.F. (2014).
Telomeres, oxidative stress and inflammatory
factors:  partners in cellular senescence? Longev
Healthspan 3, 1.
Dawes, E.A. (1971). Quantitative problems in bio-
chemistry, Vol. 3, 5th edition (Edinburgh &
London: E. & S. Livingstone).
Grahn, D. (1958). Acute radiation response of mice
from a cross between radiosensitive and radiore-
sistant strains. Radiat Res 4, 189–198.
Harley, C.B., Futcher, A.B., and Greider, C.W. (1990).
Telomeres shorten during ageing of human fibro-
blasts. Nature 345, 458–460.
Harman, D. (1956). Aging:  a theory based on free
radical and radiation chemistry. J Gerontol 11,
298–300.
Harman, D. (1981). The aging process. Proc Natl Acad
Sci USA 78, 7124–7128.
Holzer, H., Schultz, G., and Lynen, F. (1956).
Bestimmung des quotienten DPNH/DPN in leb-
enden hefezellen durch analyse stationarer alkohol
und acetaldehyde konzentrationen. Biochem Z
332, 252–263.
Howes, R. (2011). Mythology of antioxidant
enzymes. J Evidence Based Compl Altern Med
16, 149–159.
Kashiwaya, Y., King, M.T., and Veech, R.L. (1997).
Substrate signaling by insulin:  a ketone bodies
ratio mimics insulin action in heart. Am J Cardiol
80, 50A–64A.
Kashiwaya, Y., Takeshima, T., Mori, N., Nakashima,
K., Clarke, K., and Veech, R.L. (2000). D-beta-
Hydroxybutyrate protects neurons in models of
Alzheimer’s and Parkinson’s disease. Proc Natl
Acad Sci USA 97, 5440–5444.
Krebs, H.A., and Veech, R.L. (1969). Pyridine nucle-
otide interrelations. In The Energy Level and
Metabolic Control in Mitochondria, S. Papa,
269
Chapter 27: Mitigation of Damage from Reactive Oxygen Species
Tager, J.M., Quagliariello, E., & Slater, E.C., eds.
(Bari, Italy: Adriatica Editrice), pp. 329–384.
LeGal, K. (2015). Antioxidants can increase melanoma
metastasis in mice. Sci Transl Med 7, 1–7.
Linster, C.L., and Van Schaftingen, E. (2007). Vitamin
C:  biosynthesis, recycling and degradation in
mammals. FEBS J 274, 1–22.
Lomax, M.E., Folkes, L.K., and O’Neill, P. (2013).
Biological consequences of radiation-induced
DNA damage:  relevance to radiotherapy. Clin.
Oncol. 25, 578–585.
Miller, A.C., Blakely, W.F., Livengood, D., Whittaker,
T., Xu, J., Ejnik, J.W., Hamilton, M.M., Parlette,
E., John, T.S., Gerstenberg, H.M., et  al. (1998).
Transformation of human osteoblast cells to the
tumorigenic phenotype by depleted uranium-
uranyl chloride. Environ Health Perspect 106,
465–471.
Miller, A.C., Cohen, S., Stewart, M., Rivas, R., and
Lison, P. (2011). Radioprotection by the histone
deacetylase inhibitor phenylbutyrate. Radiat
Environ Biophys 50, 585–596.
Mourkioti, F., Kustan, J., Kraft, P., Day, J.W., Zhao,
M.M., Kost-Alimova, M., Protopopov, A.,
DePinho, R.A., Bernstein, D., Meeker, A.K., et al.
(2013). Role of telomere dysfunction in cardiac
failure in Duchenne muscular dystrophy. Nat. Cell
Bio 15, 895–904.
Offermanns, S. (2006). The nicotinic acid receptor
GPR109A (HM74A or PUMA-G) as a new thera-
peutic target. Trends Pharmacol Sci 27, 384–390.
Olovnikov, A.M. (1973). A theory of marginotomy.
The incomplete copying of template margin in
enzymic synthesis of polynucleotides and biologi-
cal significance of the phenomenon. J Theor Biol
41, 181–190.
Opresko, P.L., Fan, J., Danzy, S., Wilson, D.M., 3rd, and
Bohr, V.A. (2005). Oxidative damage in telomeric
DNA disrupts recognition by TRF1 and TRF2.
Nucleic Acids Res 33, 1230–1239.
Orr, W.C., Radyuk, S.N., and Sohal, R.S. (2013).
Involvement of redox state in the aging of
Drosophila melanogaster. Antioxid Redox Signal
19, 788–803.
Pacher, P., Beckman, J.S., and Liaudet, L. (2007). Nitric
oxide and peroxynitrite in health and disease.
Physiol Rev 87, 315–424.
Pham-Huy, L.A., He, H., and Pham-Huy, C. (2008).
Free radicals, antioxidants in disease and health.
Int J Biomed Sci 4, 89–96.
Sato, K., Kashiwaya, Y., Keon, C.A., Tsuchiya, N., King,
M.T., Radda, G.K., Chance, B., Clarke, K., and
Veech, R.L. (1995). Insulin, ketone bodies, and
mitochondrial energy transduction. FASEB J 9,
651–658.
Schmid, T.E., Dollinger, G., Beisker, W., Hable, V.,
Greubel, C., Auer, S., Mittag, A., Tarnok, A.,
Friedl, A.A., Molls, M., et al. (2010). Differences in
the kinetics of gamma-H2AX fluorescence decay
269
after exposure to low and high LET radiation. Int J
Radiat Biol 86, 682–691.
Schmid, W. (1975). The micronucleus test. Mutation
Research 31, 9–15.
Schriner, S.E., Linford, N.J., Martin, G.M., Treuting, P.,
Ogburn, C.E., Emond, M., Coskun, P.E., Ladiges,
W., Wolf, N., Van Remmen, H., et  al. (2005).
Extension of murine life span by overexpression
of catalase targeted to mitochondria. Science 308,
1909–1911.
Schrodinger, E. (1944). What Is Life Mind and
Matter (Cambridge, UK:  Cambridge University
Press).
Scott, E.M., Duncan, I.W., and Ekstrand, V. (1963).
Purification and properties of glutathione reduc-
tase of human erythrocytes. J Biol Chem 238,
3928–3933.
Sfeir, A. (2012). Telomeres at a glance. J Cell Sci 125,
4173–4178.
Shimazu, T., Hirschey, M.D., Newman, J., He, W.,
Shirakawa, K., Le Moan, N., Grueter, C.A.,
Lim, H., Saunders, L.R., Stevens, R.D., et  al.
(2013). Suppression of oxidative stress by beta-
hydroxybutyrate, an endogenous histone deacety-
lase inhibitor. Science 339, 211–214.
Sohal, R.S., and Weindruch, R. (1996). Oxidative
stress, caloric restriction, and aging. Science
273, 59–63.
Speakman, J.R. (2005). Body size, energy metabolism
and lifespan. J Exp Biol 208, 1717–1730.
Steinhubl, S.R. (2008). Why have antioxidants failed in
clinical trials? Am J Cardiol 101, 14D-19D.
Szilard, L. (1959). On the nature of the aging process.
Proc Natl Acad Sci USA 45, 30–45.
Tunaru, S., Kero, J., Schaub, A., Wufka, C., Blaukat,
A., Pfeffer, K., and Offermanns, S. (2003). PUMA-
G and HM74 are receptors for nicotinic acid
and mediate its anti-lipolytic effect. Nat Med 9,
352–355.
Veech, R.L. (2005). Mitochondrial bioenergetics in
intact tissues. In Mitochondria in Health and
Disease, C.D. Berdanier, ed. (New  York:  CRC
Press, Taylor & Francis), pp. 65–94.
Veech, R.L. (2006). The determination of the redox
states and phosphorylation potential in living tis-
sues and their relationship to metabolic control
of disease phenotypes. Biochem Mol Biol Edu 34,
168–179.
Veech, R.L. (2014). Ketone ester effects on metabolism
and transcription. J Lipid Res 55, 2004–2008.
Veech, R.L., Eggleston, L.V., and Krebs, H.A. (1969).
The redox state of free nicotinamide-adenine
dinucleotide phosphate in the cytoplasm of rat
liver. Biochem J 115, 609–619.
Veech, R.L., Lawson, J.W.R., Cornell, N.W., and Krebs,
H.A. (1979). Cytosolic phosphorylation potential.
Journal of Biological Chemistry 254, 6538–6547.
Verdin, E., and Ott, M. (2015). 50  years of protein
acetylation:  from gene regulation to epigenetics,
270
270 section III: Ketogenic Diet in the Laboratory
metabolism and beyond. Nat Rev Mol Cell Biol 16,
258–264.
Villet, R.H., and Dalziel, K. (1969). The nature of the
carbon dioxide substrate and equilibrium constant
of the 6-phosphogluconate dehydrogenase reac-
tion. Biochem J 115, 633–638.
von Zglinicki, T. (2002). Oxidative stress shortens
telomeres. Trends in Biochemical Sciences 27,
339–344.
Williamson, D. H., Lund, P., & Krebs, H. A. (1967). The
redox state of free nicotinamide-adenine dinucleo-
tide in the cytoplasm and mitochondria of rat liver.
Biochem J, 103(2), 514–527.
Winterbourn, C.C. (2008). Reconciling the chemistry
and biology of reactive oxygen species. Nat Chem
Biol 4, 278–286.
Wollenberger, A., Ristau, O., and Schoffa, G. (1960).
Eine einfache Technik der extrem schnellen
Abkuhlung grssáerer Gewebestacke. Pflugers
Archiv 270, 399–412.
Yamato, M., Kudo, W., Shiba, T., Yamada,
K.I., Watanabe, T., and Utsumi, H. (2010).
Determination of reactive oxygen species associ-
ated with the degeneration of dopaminergic neu-
rons during dopamine metabolism. Free Radic Res
44, 249–257.
271
28
Metabolic Seizure Resistance via BAD
and KATP Channels
J UA N R A M Ó N M A R T Í N E Z - F R A N Ç O I S , P H D , N I K A N. D A N I A L , P H D ,
A N D G A RY Y E L L E N, P H D
INTRODUCTION
The mechanisms of the ketogenic diet are poorly
understood, but it has been used for almost a
century and is arguably the most effective single
treatment for epilepsy (Hartman et al., 2007; Neal
et al., 2009; Thiele, 2003). This diet, when adhered
to strictly, may reduce seizures in children by up to
50%, with 10%–15% of children becoming seizure-
free. Unfortunately, maintaining the ketogenic
diet’s strict regimen is challenging. Thus, it would
be very valuable to understand the ketogenic diet’s
mechanisms and harness them to create better epi-
lepsy therapies.
On a ketogenic diet the liver produces the
ketone bodies β-hydroxybutyrate and acetoacetate
from fatty acids, elevating ketone bodies to mil-
limolar concentrations in the blood. Under these
conditions, ketone bodies provide an alternative
fuel source to tissues, including the brain, which
otherwise would primarily utilize glucose (DeVivo
et al., 1978; Mergenthaler et al., 2013; Owen et al.,
1967; Zielke et al., 2009). The remarkable antiepi-
leptic effect of increased ketone body metabolism
points to a link between fuel utilization and neu-
ronal excitability.
However, the molecular mechanisms of this
link are poorly understood (Danial et  al., 2013;
Hartman et  al., 2007; Lutas and Yellen, 2013).
Some of the mechanisms that have been proposed
include increased adenosine signaling through A1
purinergic receptors (Masino et al., 2011), changes
in gene expression following glycolysis reduction
(Garriga-Canut et  al., 2006), suppression of glu-
tamate release (Juge et  al., 2010), and changes in
excitability mediated by lactate dehydrogenase
activity (Sada et al., 2015).
A mouse model with altered metabolism simi-
lar to the ketogenic diet produces similar anti-
seizure effects, and this nondietary manipulation
provides a better opportunity to dissect the
mechanisms involved in metabolic seizure resis-
tance (Giménez-Cassina et al., 2012). This model
involves metabolic changes in brain cells medi-
ated by the protein BAD (BCL-2-associated ago-
nist of cell death). Alteration of BAD’s function
reduces the capacity of cells to utilize glucose and
increases the capacity to use ketone bodies. The
key metabolic change involved in seizure protec-
tion appears to be the switch away from glucose
as the preferred energy source toward utilization
of other fuels, rather than elevation of circulating
ketone bodies. In addition, BAD alteration leads to
increased activity of metabolically sensitive ATP-
sensitive potassium (KATP) channels in the brain.
Modulation of KATP channel activity may also play
an important role in the seizure-protective actions
of the ketogenic diet.
B A D M O D U L AT E S G L U C O S E
A N D K E TO N E B O DY
M E TA B O L I S M
BAD is best-known as a pro-apoptotic protein
that is a member of the BCL-2 family of proteins
(Czabotar et  al., 2014; Danial and Korsmeyer,
2004; Moldoveanu et  al., 2014). Besides its role
in apoptosis, BAD modulates glucose metabo-
lism in multiple cell types, including hepatocytes,
pancreatic β-cells, and fibroblasts (Giménez-
Cassina and Danial, 2015). The switch between
BAD’s apoptotic and metabolic roles is mediated
through phosphorylation of its serine 155, located
within an alpha helical segment known as the BH-
3 domain (Danial et al., 2008; Datta et al., 2000).
In hepatocytes and pancreatic β-cells, serine 155
(equivalent to serine 118 in human BAD) phos-
phorylation promotes mitochondrial metabolism
of glucose. In the presence of apoptotic signals,
dephosphorylated BAD promotes apoptosis.
272
272 section III: Ketogenic Diet in the Laboratory
The reduction in glucose metabolism pro-
voked by BAD modification is reminiscent of met-
abolic changes associated with the ketogenic diet,
that is, reduced glycolysis and increased ketone
body metabolism. Thus, modifying BAD may be
a productive avenue to study the mechanisms that
underlie the seizure protective effects of changes
in metabolism. Given the systemic, nonspecific
effects of dietary manipulation, the study of BAD
alteration enables the study of metabolic seizure
resistance mechanisms in a more specific manner
and without the use of dietary interventions.
M E TA B O L I C C H A N G E S
I N   B A D - A LT E R E D
P R I M A RY N E U R O N S A N D
ASTROCYTES
Two key mouse models have been used to study
the effects of BAD on neurons and astrocytes:  a
Bad knockout and a Bad knockin. The knockin
mouse expresses a mutant of BAD that contains
a nonphosphorylatable residue, alanine, at posi-
tion 155 (BADS155A). In knockin animals, BAD’s
metabolic role is altered while leaving its apop-
totic role intact, as BADS155A cannot be phosphory-
lated (Danial et al., 2008; Giménez-Cassina et al.,
2014). Importantly, these mutant mice display nei-
ther gross neuroanatomical nor neurobehavioral
abnormalities. Therefore, it is unlikely that genetic
modification in these mouse models substantially
impairs normal brain function.
To elucidate whether BAD has similar effects
on metabolism in the brain as it does in other tis-
sues, Giménez-Cassina et  al. (2012) measured
the mitochondrial oxygen consumption rates
in primary cultures of neurons and astrocytes
using mitochondrial respirometry. Bad-ablated
cells exhibited reduced glucose metabolism and
elevated capacity to metabolize ketone bodies
compared with wild-type cells. Similar changes in
brain metabolism are produced by the presence
of ketone bodies. In hippocampal brain slices, β-
hydroxybutyrate competes with glucose, lactate,
and pyruvate for the generation of acetyl-CoA,
inhibiting total glycolytic flux upstream of pyruvate
kinase (Valente-Silva et al., 2015). Additionally, in
humans, glucose flux in the brain is inversely cor-
related with the degree of ketosis, allowing ketone
bodies to partially replace glucose in cerebral
metabolism (Haymond et  al., 1983; Owen et  al.,
1967). These observations cement the idea that
disrupting BAD function may mimic brain meta-
bolic changes produced by the ketogenic diet and
is a useful model to study the cellular mechanisms
linking metabolism to excitability.
To corroborate that the effects of Bad disrup-
tion in the metabolism of cultured neurons and
astrocytes were due to the metabolic, and not
the apoptotic role of BAD, mitochondrial res-
pirometry was also performed in BadS155A cells
(Giménez-Cassina et  al., 2012). Indeed, BadS155A
neurons and astrocytes also preferentially utilize
β-hydroxybutyrate over glucose, like the BAD
knockout. Conversely, BAD phosphorylation on
serine 155 inhibits ketone body utilization. This
confirms that BAD, through its metabolic role,
changes the ability of brain cells to metabolize
different fuels.
A LT E R AT I O N O F   B A D
FUNCTION PRODUCES
S E I Z U R E R E S I S TA N C E
Alteration of BAD’s metabolic role by either knock-
out of the protein or knockin of a phosphorylation-
resistant mutant variant leads to altered metabolism
without dietary manipulations. This metabolic
switch from brain glucose to ketone body utiliza-
tion also influences excitability and produces sei-
zure resistance in vivo (Giménez-Cassina et  al.,
2012). The seizure resistance has been studied in
two acute chemoconvulsant models of seizure,
involving intraperitoneal injection of kainate (cf.
Ben-Ari et al., 1980) or subcutaneous injection of
pentylenetretrazole (PTZ) (cf. Ferraro et al., 1999).
When injected with kainate, wild-type mice
display a series of seizures of increasing severity
that peak 1–2 hours after kainate injection and
then slowly decay (Giménez-Cassina et  al., 2012;
Figure  28.1A). Most wild-type mice experience
status epilepticus with very severe tonic-clonic sei-
zures, and many die. In contrast, seizure severity
in Bad−/− mice is much milder than in wild-type
mice, and mutant mice rarely go into status epilep-
ticus or die (Figures 28.1 and 28.3). Electrographic
seizures are also substantially milder (Figure 28.3).
The in vivo seizure protection is due to BAD’s met-
abolic, not apoptotic role, because BadS155A mice
were similarly resistant to kainate-induced sei-
zures. Thus, disruption of the metabolic function
of BAD alleviates both forebrain seizure activity
and generalized behavioral seizures during status
epilepticus.
BAD deletion is similarly protective for PTZ-
induced seizures (Figure  28.1C); PTZ produces
seizures by a different mechanism than kainate, by
inhibiting GABA action rather than by stimulat-
ing glutamate receptors (Bough and Eagles, 1999;
Ferraro et al., 1999). It remains to be seen whether
BAD mutations can also be protective in chronic
epilepsy models.
273

Chapter 28: Metabolic Seizure Resistance via BAD and KATP Channels 273

(a) Kainate (b) Kainate (c) PTZ

Seizure Severity (Racine Scale) 4 120 *** 120 ***
WT
Bad –/– 100 100
3

Seizure severity (%)

Seizure severity (%)

80 80
2 60 60

40 40
1
20 20
0
0 1 2 3 4 0 0
Time (hours) WT Bad –/– WT Bad –/–

FIGURE 28.1 Bad−/− mice are resistant to kainate- or pentylenetetrazole-induced acute seizures. (A) Raw seizure
scores in Bad−/− 8- to 10-week-old male mice compared with wild-type (WT) mice over a 4-hour period after a single
intraperitoneal injection of kainate (30 mg/kg). (B) Kainate-induced seizure severity calculated as ∑(all scores of a given
mouse)/(time of experiment) for wild-type (WT) or Bad−/− mice. The mean of the seizure severity values from WT mice
was assigned a value of 100%. This value was then used to normalize the severity of the other tested genotypes and/
or conditions within the same scale. WT, n = 42; Bad–/–, n = 13. (C) Integrated seizure severity (calculated as in B) in
Bad−/− and WT mice (n = 16) subjected to a single subcutaneous injection of pentylenetetrazole (PTZ) at 80 mg/kg
monitored over a 70-minute period. Data in A–C are presented as mean ± SEM. *** p < .001; two-tailed Student’s t-test.
(Adapted from Giménez-Cassina et al., 2012)

BAD EFFECTS ON SEIZURE BAD has also been implicated in regulating

ARE UNLIKE EFFECTS
O F   O T H E R B C L - 2 FA M I LY
P ROT E I N S
Due to their apoptotic roles, certain BCL-2 fam-
ily proteins such as BIM and PUMA have been
implicated in protection against neuronal loss and
seizure damage after prolonged periods of status
epilepticus (Engel et al., 2010, 2011; Murphy et al.,
2010). In a chronic seizure model (intra-amygdala
microinjection of kainate), neuronal death in the
hippocampus 24 hours after seizure initiation is
decreased in Bim−/− or Puma−/− mice, demonstrat-
ing a role for these pro-apototic proteins in cell
death after chronic seizures. In contrast, loss of
BIM or PUMA is not seizure protective to seizures
immediately after kainate injection (Engel et  al.,
2010; Murphy et  al., 2010). Similarly, knockout
of BIM or another BH3-only protein, BID, does
not produce resistance to acute seizures elicited
by intraperitoneal kainate injection (Giménez-
Cassina et al., 2012). Unlike BAD, neither of these
proteins affects glucose metabolism (Giménez-
Cassina and Danial, 2015). It appears that certain
BCL-2 family members may play a role in epilepto-
genesis by promoting hippocampal neuron death,
but that this is distinct from the role of BAD in
altering acute seizure susceptibility, which involves
a switch in fuel preference rather than apoptosis.
synaptic transmission. The BAD-BAX-caspase-
3 cascade can induce long-term depression in
CA1 hippocampal neurons (Jiao and Li, 2011).
Additionally, ABT-737, an inhibitor of BCL-2,
BCL-XL, and BCL-w survival proteins, slows the
recovery of neurotransmission after intense synap-
tic activity and decreases hypoxia-triggered dam-
age to synaptic function (Hickman et al., 2008).
This effect of ABT-737 could result from BAD
function, since this compound is likely acting on
downstream signaling partners of BAD. However,
ABT-737 also blocks binding of other BH3-only
proteins to BCL-2 and BCL-XL, which involves
the same binding pocket used by BAD. Thus,
disruption of these protein interactions by ABT-
737 is not indicative of specific actions of BAD in
synaptic transmission. In any case, these synaptic
changes are unlikely to be part of the anticonvul-
sant action of BAD deletion. For these synaptic
effects, Bad−/− and BadS155A mice would have oppo-
site phenotypes, as these alterations are dependent
on the interaction of nonphosphorylated BAD
with the downstream BCL-2 proteins; but in fact,
both genotypes produce seizure protection. These
observations highlight the varied and complex
roles of BAD in neuronal physiology, and uniquely
distinguish the switch in fuel preference as a potent
mechanism to produce seizure protection.
274
274 section III: Ketogenic Diet in the Laboratory
T H E M E TA B O L I C S E I Z U R E
R E S I S TA N C E O F   B A D − / −
M I C E L I K E LY O C C U R S
IN THE BRAIN
Upon BAD modification, the capacity to metabo-
lize glucose is decreased. This recapitulates changes
in brain fuel utilization during fasting or on the
ketogenic diet. In contrast, the switch in fuel choice
produced by BAD alteration occurs without any
dietary alterations. As opposed to dietary manip-
ulations that affect the whole organism, seizure
protection arising from BAD modification seems
to specifically originate from metabolic changes
in the brain. Knockdown of BAD by adenovirus-
produced shRNA in the liver, the major source of
ketone body production, does not protect against
kainate-induced seizures, though it does produce
the same metabolic effects observed in the liver
of Bad−/− animals (Giménez-Cassina et  al., 2012;
Giménez-Cassina et al., 2014).
Interestingly, circulating levels of ketone bod-
ies are similar in Bad−/− and wild-type mice, but
β-hydroxybutyrate levels are increased in whole-
brain tissue derived from Bad−/−. This suggests
that in the knockout animals, ketogenesis is being
increased specifically in the brain; it also argues
that changes in brain metabolism upon BAD
alteration are unlikely to be generated by systemic
modifications in metabolism. The precise changes
in metabolism responsible for BAD’s seizure-
protective effect remain to be established. Given
that the ketogenic diet’s antiseizure effect can be
reversed in patients promptly after carbohydrate
consumption (Huttenlocher, 1976; Pfeifer et  al.,
2008), it seems likely that the transition in fuel
type utilization is the crucial factor. Metabolomic
studies comparing wild-type to Bad-altered geno-
types could shed light on the metabolic pathways
implicated in BAD’s effect.
K AT P C H A N N E L S
L I N K M E TA B O L I S M
W I T H   E X C I TA B I L I T Y
One likely candidate for linking altered brain
metabolism with altered excitability, in the keto-
genic diet and in BAD-altered animals, is the
ATP-sensitive potassium (KATP) channel. KATP
channels are inhibited by intracellular ATP,
activated by intracellular ADP, and are widely
expressed throughout the body, including the
brain (Liss and Roeper, 2001; Nichols, 2006; Proks
and Ashcroft, 2009). Their role linking metabo-
lism with cell excitability has been best studied
in control of insulin release by pancreatic β-cells
(Ashcroft, 2005; Ashcroft et al., 1984). In these
cells, the intracellular ATP:ADP ratio determines
KATP channel activity. In basal conditions, KATP
channels are spontaneously active and maintain
the cell’s resting membrane potential near the
potassium equilibrium potential, that is, hyperpo-
larized. Upon an increase in extracellular glucose
concentration, and consequently an increase in
the intracellular ATP concentration, KATP chan-
nels are inhibited. This decrease in KATP channel
activity depolarizes the cell and leads to insulin
release. This exemplifies how KATP channels can
link metabolism to cell excitability, but the effect
is likely not unique to glucosensing cells like the
β-cells.
Two brain regions that are particularly
enriched in KATP channels are the hippocampus
and the substantia nigra pars reticulata (SNr)
(Dunn-Meynell et al., 1998; Karschin et al., 1997;
Ma et al., 2007; Tanner et al., 2011; Yamada et al.,
2001; Zawar et al., 1999). KATP channels present in
these two areas are octamers comprising two types
of subunits:  four Kir6.2 pore-forming subunits
and four SUR1 regulatory subunits. Both of these
brain regions are involved in the generation of
seizure activity (Depaulis et al.,1994; Heinemann
et  al., 1992; Hsu, 2007; Iadarola and Gale, 1982;
Krook-Magnuson et  al., 2015; Lothman et  al.,
1992; McNamara et  al., 1984), and excitability of
neurons in these areas can be modulated by KATP
channels (Lutas et  al., 2014; Stanford and Lacey,
1996; Tanner et  al., 2011; Yamada et  al., 2001;
Zawar et al., 1999).
Neuronal KATP channel activity appears gen-
erally to be low, implying that the intracellular
ATP concentration in neurons is often sufficiently
high to inhibit these channels (Giménez-Cassina
et al., 2012; Haller et al., 2001; Pelletier et al., 2000;
Schwanstecher and Panten, 1993; Tanner et  al.,
2011; Yamada et al., 2001). Substantial metabolic
insults to the brain are known to be capable of
activating the channels; for instance in brain isch-
emia, intracellular ATP and ATP/ADP markedly
decrease after ~2 minutes (Katsura et al., 1992). In
the CA1 area of the hippocampus, a brain region
particularly susceptible to ischemic insult (Davolio
and Greenamyre, 1995; Schmidt-Kastner and
Freund, 1991; Smith et al., 1984), hypoxia triggers
neuronal hyperpolarization by activating potas-
sium conductances, including KATP channel cur-
rents (Yamada and Inagaki, 2005; Yamada et  al.,
2001). Consistent with this, KATP channel inhibi-
tors, such as the sulfonylurea drugs glibenclamide
or tolbutamide, can suppress the hypoxia-induced
hyperpolarization. Furthermore, mice lacking
either Kir6.2 or SUR1 exhibit a markedly reduced
275
Chapter 28: Metabolic Seizure Resistance via BAD and KATP Channels
threshold for generalized seizures upon a hypoxic
insult, that is, they are hypersensitive to hypoxia-
induced seizures (Hernández-Sánchez et al., 2001;
Yamada et al., 2001). Interestingly, similar to Bad-
altered mice, mice overexpressing SUR1 display a
decrease in kainate-induced seizure susceptibility.
These observations indicate that KATP channels
play a crucial role in mediating seizure protec-
tive effects provoked by metabolic insufficiency or
hypoxia.
Besides ischemic/hypoxic insult, other meta-
bolic changes in brain regions implicated in sei-
zure activity can also modulate KATP channel
activity. Poisoning of GABAergic neurons in the
SNr or hippocampal CA1 neurons with cyanide
activates KATP channels (Matsumoto et  al., 2002;
Schwanstecher and Panten, 1993). In addition, in
dentate granule neurons (DGNs) of the hippo-
campus, a much more subtle energetic challenge
can lead to KATP channel opening: neuronal firing
elicited by antidromic stimulation increases single
KATP channel activity (Tanner et al., 2011). Firing
of action potentials produces metabolic changes
by activating the pumps, for example the Na/
K ATPase, that maintain intracellular ionic con-
centrations (Ivannikov et  al., 2010; Mercer and
Dunham, 1981). Activation of these pumps trig-
gers ATP consumption, leading to an opening of
KATP channels (Haller et  al., 2001; Tanner et  al.,
2011). It has also been shown that KATP channel
open probability in DGNs increases by extracel-
lular application of β-hydroxybutyrate (Tanner
et  al., 2011). These observations are consistent
with the idea that KATP channels could mediate
the anticonvulsant effect of the ketogenic diet.
First, KATP channels might open in conditions of
hyperexcitability, such as during a seizure. Second,
elevated levels of circulating ketone bodies could
also increase KATP channel activity. These effects
are synergistic, and the resulting increase in KATP
channel current would reduce excitability and pro-
duce seizure resistance.
BA D D I S RU P T I O N
I N C R E A S E S K AT P C H A N N E L
ACTIVITY
As previously stated, KATP channels link metabo-
lism to excitability and are activated by both
ketone bodies and electrical activity in the brain.
This raises the question of whether these channels
mediate the seizure resistance elicited by BAD dis-
ruption. In single-channel, cell-attached experi-
ments in DGNs in hippocampal brain slices, the
basal KATP channel open probability is dramatically
275
increased in Bad−/− mouse DGNs, from ~0.5% in
wild-type to ~20% in Bad−/− (Giménez-Cassina
et  al., 2012). Confirming this increase in basal
KATP channel activity in DGNs with disrupted
BAD function, whole-cell KATP current is also
inhibited by dialyzing the intracellular medium
of either Bad−/− or BadS155A DGNs with high ATP
(4 mM)—a phenomenon denominated “wash-
down” (Figure 28.2A). On the other hand, wash-
down does not occur in wild-type DGNs (as KATP
channels are mostly closed in this case) or in
DGNs lacking both BAD and Kir6.2.
These whole-cell results corroborate the
fact that basal KATP channel open probability is
increased in Bad−/− DGNs. Additionally, the num-
ber of functional KATP channels in DGNs is not
affected by altering BAD’s metabolic role. The
maximal KATP conductance, observed in whole-
cell recordings dialyzing the neuron’s intracellu-
lar medium with a low concentration of ATP (0.3
mM), is unchanged between wild-type and Bad−/−
or BadS155A DGNs.
The increased KATP channel activity observed
in Bad-altered DGNs suggests that KATP channels
might lead to decreased neuronal excitability. Even
though further studies are needed to demonstrate
such a scenario, consistent with this idea, studies
using a hippocampal neuron cell line have shown
that KATP channels can mediate changes in hip-
pocampal neuron excitability (Huang et al., 2006,
2007). Thus, a decrease in excitability mediated by
KATP channels could be a central mechanism in the
ketogenic diet’s seizure protective effects.
K AT P C H A N N E L S M E D I AT E
S E I Z U R E P ROT E C T I O N
ELICITED BY BAD
D I S RU P T I O N
Lack of BAD leads to a remarkable increase in KATP
channel activity and also produces seizure protec-
tion. It appears that these two effects are causally
related:  eliminating Kir6.2 (Kcnj11) in a Bad−/−
background nearly abolishes the seizure resistance
produced by Bad ablation (Figure  28.2B). This
substantial attenuation of seizure resistance is not
due to an increase in seizure severity produced by
lack of Kir6.2; deletion of Kir6.2 alone does not
increase mouse seizure sensitivity compared with
wild-type animals. These results constitute genetic
evidence that KATP channels are necessary in medi-
ating BAD’s effect on seizure protection.
KATP channels are well-known mediators of
changes in metabolism with neuronal excitability.
Indeed, either glucose deprivation or ketone body
276

276 section III: Ketogenic Diet in the Laboratory

(a) (b) n.s.

***
WT 120
2.0 *** **
Bad –/–
Normalized Conductance to G3min

BadS155A 100
Bad –/–; Kir6.2 –/–

Seizure Severity (%)

1.5
80

1.0 60

40
0
20

0 0
0 1 2 3 WT Bad –/– Bad –/–; Kir6.2 –/–
Time from break-in (min) Kir6.2 –/–

FIGURE 28.2 KATP channels mediate effects of BAD alteration. (A) Whole-cell conductance recorded from dentate
granule neurons (DGNs) in acute hippocampal slices showing “washdown” of an initially high KATP channel conductance
with high ATP (4 mM) in the recording pipette. Washdown is observed in Bad–/– and BadS155A DGNs but not in wild-type
(WT) or Bad–/–;Kir6.2–/– double-mutant DGNs. The time course of slope conductance during whole-cell recording was
normalized to the value 3 min after break-in. (B) Integrated seizure severity in Bad–/– or Kir6.2–/– single mutants, or Bad–/–;
Kir6.2–/– double mutant mice compared with WT mice after a single intraperitoneal injection of kainate monitored and
analyzed as in Figure 28.1. Data are presented as mean ± SEM. ** p < .01; *** p < .001; n.s., nonsignificant; two-tailed
Student’s t-test. (Adapted from Giménez-Cassina et al., 2012).

metabolism can increase KATP channel activity.
Furthermore, these effects can be potentiated by
neuronal firing, for example, during seizure activity.
The mechanism by which a metabolic fuel switch
could increase KATP channel activity is unknown.
Given that ATP and ADP directly control the open
probability of KATP channels, changes in ATP/ADP
local to KATP channels might be different in the pres-
ence or absence of BAD’s metabolic effects.
Why would ATP/ADP change in the presence
of abundant alternative fuels? One possibility is
that KATP channel activity is controlled by local
glycolysis. Glycolysis is bypassed and inhibited
by mitochondrial metabolism of ketone bod-
ies (DeVivo et  al.,1978; Haymond et  al., 1983;
Valente-Silva et al., 2015). This metabolic change
would increase ATP produced by mitochondria
and decrease ATP produced by glycolysis.
Indeed, KATP channels and glycolytic enzymes
can be found associated in large complexes in
the plasma membrane (Dhar-Chowdhury et  al.,
2007; Dubinsky et  al., 1998; Hong et  al., 2011).
Evidence suggests that these glycolytic enzyme
complexes may produce ATP compartmentation
in the cell (Chu et al., 2012; Hoffman et al., 2009;
Proverbio and Hoffman, 1977). Furthermore, gly-
colysis inhibition can trigger the opening of KATP
channels (Matsumoto et al., 2002; Schwanstecher
and Panten, 1993; Tantama et al., 2013). Another
possibility is that glucose metabolism is special
in its ability to respond rapidly to acute energy
challenges, as might occur in the elevated brain
activity leading to seizures. Thus, BAD alteration,
by switching metabolism away from glucose con-
sumption, may produce a decrease in the ATP con-
centration at the plasma membrane, which then
could trigger KATP channel opening and reduced
neuronal excitability.
Even though modulation of KATP channel activ-
ity by ATP/ADP is a straightforward explanation,
KATP channel regulation is complex, and other
processes, such as PIP2 binding or phosphoryla-
tion, can also alter KATP channel activity (Nichols,
2006; Proks and Ashcroft, 2009). In addition,
β-hydroxybutyrate levels are increased in Bad−/−
compared with wild-type brains, and, as men-
tioned above, this can induce KATP channel open-
ing. While it is unlikely that the effects of ketone
bodies on KATP channel activity explain the total-
ity of the observed KATP channel activation, these
metabolites may constitute part of multiple actions
needed to exert the full effect observed upon BAD
alteration.
How does KATP channel activation in certain
brain cells elicit seizure protection? A  simple
explanation is that elevated activity, which would
ordinarily lead to a seizure in a susceptible indi-
vidual, produces increased KATP channel current,
277

Chapter 28: Metabolic Seizure Resistance via BAD and KATP Channels 277

Glucose
BAD deletion produces BAD
altered metabolism Alternative fuels

WT Bad –/–

Increased KATP channel

activity in dentate
granule neurons

2 pA
0.5 s

Bad –/– 0.5 *

High Energy Spiking State

0.4

Fraction of Time in
KA inj KA inj

0.3

Seizure protection 0.2

0.1

500 V 0
WT Bad –/–
1 min

FIGURE 28.3 Summary of effects through BAD and KATP channels that lead to seizure resistance. Ablation of Bad
leads to decreased glucose metabolism and increased consumption of alternative fuels. This metabolic switch increases
KATP channel activity, as seen in cell-attached recordings of KATP channels in dentate granule neurons in hippocampal
brain slices. Compared with wild-type (WT) mice, Bad–/– mice are resistant to acute kainate-induced seizures exhibiting
reduced cortical seizure activity recorded by electroencephalogram (EEG) as shown by representative EEG traces of
approximately1 hour for WT or Bad–/– mice (bottom left and middle panels), and reduced fraction of time spent in high
energy spiking state (bottom right panel). WT, n = 17; Bad–/–, n = 20; mean ± SEM, * p < .05 by two-tailed Student’s t-test.
(Adapted from Giménez-Cassina et al., 2012).

which then reduces subsequent neuronal activity CONCLUSION

by hyperpolarizing neurons. It will be important
to reveal the relevant brain regions associated with
this seizure protective effect. KATP channels are
present in two brain regions associated with sei-
zure activity: the DGNs in the hippocampus and
the GABAergic neurons of the SNr. It has been
shown that reducing neuronal firing in either
of these two brain areas inhibits seizure activity
(Akman et  al., 2015; Coulter and Carlson, 2007;
Krook-Magnuson et  al., 2013; Paz et  al., 2007;
Vercueil et  al., 1998). Studies on KATP channel
activity or in vivo seizure models in conditional
Kir6.2 (Kcnj11) knockout lines could be used to
define the brain regions and brain cell types impli-
cated in seizure resistance.
In summary (Figure 28.3), certain manipula-
tions of BAD’s metabolic role produce a meta-
bolic change in brain cells, switching away from
glucose utilization toward consumption of alter-
native fuels. The metabolic changes triggered
by BAD modification are reminiscent of those
elicited by a ketogenic diet. Remarkably, BAD
disruption produces robust seizure protection
in the absence of dietary treatments. At the cel-
lular level, it appears that KATP channels mediate
BAD’s anticonvulsant effect. Disruption of BAD
function increases KATP channel activity in neu-
rons in the dentate gyrus, a brain region that is
likely a gate for seizure activity. The mechanisms
by which BAD modulates brain metabolism and
278
278 section III: Ketogenic Diet in the Laboratory
how these metabolic changes affect excitability
have just started to be revealed. These seizure-
protective mechanisms will hopefully serve as
the basis for the development of more effective
therapies for epilepsy.
AC K N OW L E D G M E N T S
We thank members of the Yellen lab for critical
review and discussion of our manuscript. This
work was supported by grants from the National
Institutes of Health (R01 NS083844 to G.Y.  and
N.N.D. and R01 NS055031 to G.Y.)
REFERENCES
Akman, O., Gulcebi, M.I., Carcak, N., Ketenci Ozatman,
S., Eryigit, T., Moshé, S.L., Galanopoulou, A.S.,
and Yilmaz Onat, F. (2015). The role of the sub-
stantia nigra pars reticulata in kindling resistance
in rats with genetic absence epilepsy. Epilepsia 56,
1793–1802.
Ashcroft, F.M. (2005). ATP-sensitive potassium chan-
nelopathies:  focus on insulin secretion. J Clin
Invest 115, 2047–2058.
Ashcroft, F.M., Harrison, D.E., and Ashcroft, S.J.
(1984). Glucose induces closure of single potas-
sium channels in isolated rat pancreatic beta-cells.
Nature 312, 446–448.
Ben-Ari, Y., Tremblay, E., Ottersen, O.P., and Meldrum,
B.S. (1980). The role of epileptic activity in hippo-
campal and “remote” cerebral lesions induced by
kainic acid. Brain Res 191, 79–97.
Bough, K.J., and Eagles, D.A. (1999). A ketogenic diet
increases the resistance to pentylenetetrazole-
induced seizures in the rat. Epilepsia 40,
138–143.
Chu, H., Puchulu-Campanella, E., Galan, J.A.,
Tao, W.A., Low, P.S., and Hoffman, J.F. (2012).
Identification of cytoskeletal elements enclos-
ing the ATP pools that fuel human red blood cell
membrane cation pumps. Proc Natl Acad Sci USA
109, 12794–12799.
Coulter, D.A., and Carlson, G.C. (2007). Functional
regulation of the dentate gyrus by GABA-mediated
inhibition. Prog Brain Res 163, 235–243.
Czabotar, P.E., Lessene, G., Strasser, A., and Adams,
J.M. (2014). Control of apoptosis by the BCL-2
protein family:  implications for physiology and
therapy. Nat Rev Mol Cell Biol 15, 49–63.
Danial, N.N., and Korsmeyer, S.J. (2004). Cell
death: critical control points. Cell 116, 205–219.
Danial, N.N., Walensky, L.D., Zhang, C.-Y., Choi,
C.S., Fisher, J.K., Molina, A.J.A., Datta, S.R.,
Pitter, K.L., Bird, G.H., Wikstrom, J.D., et  al.
(2008). Dual role of proapoptotic BAD in insu-
lin secretion and beta cell survival. Nat Med 14,
144–153.
Danial, N.N., Hartman, A.L., Stafstrom, C.E., and
Thio, L.L. (2013). How does the ketogenic diet
work? Four potential mechanisms. J Child Neurol
28, 1027–1033.
Datta, S.R., Katsov, A., Hu, L., Petros, A., Fesik, S.W.,
Yaffe, M.B., and Greenberg, M.E. (2000). 14-3-3
proteins and survival kinases cooperate to inacti-
vate BAD by BH3 domain phosphorylation. Mol
Cell 6, 41–51.
Davolio, C., and Greenamyre, J.T. (1995). Selective
vulnerability of the CA1 region of hippocampus
to the indirect excitotoxic effects of malonic acid.
Neurosci Lett 192, 29–32.
Depaulis, A., Vergnes, M., and Marescaux, C. (1994).
Endogenous control of epilepsy: the nigral inhibi-
tory system. Prog Neurobiol 42, 33–52.
DeVivo, D.C., Leckie, M.P., Ferrendelli, J.S., and
McDougal, D.B. (1978). Chronic ketosis and cere-
bral metabolism. Ann Neurol 3, 331–337.
Dhar-Chowdhury, P., Malester, B., Rajacic, P., and
Coetzee, W.A. (2007). The regulation of ion chan-
nels and transporters by glycolytically derived
ATP. Cell Mol Life Sci 64, 3069–3083.
Dubinsky, W.P., Mayorga-Wark, O., and Schultz,
S.G. (1998). Colocalization of glycolytic enzyme
activity and KATP channels in basolateral mem-
brane of Necturus enterocytes. Am J Physiol 275,
C1653–C1659.
Dunn-Meynell, A.A., Rawson, N.E., and Levin, B.E.
(1998). Distribution and phenotype of neurons
containing the ATP-sensitive K+ channel in rat
brain. Brain Res 814, 41–54.
Engel, T., Hatazaki, S., Tanaka, K., Prehn, J.H.M., and
Henshall, D.C. (2010). Deletion of Puma protects
hippocampal neurons in a model of severe status
epilepticus. Neuroscience 168, 443–450.
Engel, T., Plesnila, N., Prehn, J.H.M., and Henshall,
D.C. (2011). In vivo contributions of BH3-only
proteins to neuronal death following seizures,
ischemia, and traumatic brain injury. J Cereb
Blood Flow Metab 31, 1196–1210.
Ferraro, T.N., Golden, G.T., Smith, G.G., St Jean, P.,
Schork, N.J., Mulholland, N., Ballas, C., Schill, J.,
Buono, R.J., and Berrettini, W.H. (1999). Mapping
loci for pentylenetetrazol-induced seizure suscep-
tibility in mice. J Neurosci 19, 6733–6739.
Garriga-Canut, M., Schoenike, B., Qazi, R.,
Bergendahl, K., Daley, T.J., Pfender, R.M.,
Morrison, J.F., Ockuly, J., Stafstrom, C., Sutula, T.,
et al. (2006). 2-Deoxy-D-glucose reduces epilepsy
progression by NRSF-CtBP-dependent metabolic
regulation of chromatin structure. Nat Neurosci 9,
1382–1387.
Giménez-Cassina, A., and Danial, N.N. (2015).
Regulation of mitochondrial nutrient and energy
metabolism by BCL-2 family proteins. Trends
Endocrinol Metab 26, 165–175.
Giménez-Cassina, A., Martínez-François, J.R., Fisher,
J.K., Szlyk, B., Polak, K., Wiwczar, J., Tanner, G.R.,
Lutas, A., Yellen, G., and Danial, N.N. (2012).
BAD- dependent regulation of fuel metabolism
279
Chapter 28: Metabolic Seizure Resistance via BAD and KATP Channels
and KATP channel activity confers resistance to epi-
leptic seizures. Neuron 74, 719–730.
Giménez-Cassina, A., Garcia-Haro, L., Choi, C.S.,
Osundiji, M.A., Lane, E.A., Huang, H., Yildirim,
M.A., Szlyk, B., Fisher, J.K., Polak, K., et al. (2014).
Regulation of hepatic energy metabolism and glu-
coneogenesis by BAD. Cell Metab 19, 272–284.
Haller, M., Mironov, S.L., Karschin, A., and Richter,
D.W. (2001). Dynamic activation of KATP chan-
nels in rhythmically active neurons. J Physiol 537,
69–81.
Hartman, A.L., Gasior, M., Vining, E.P.G., and
Rogawski, M.A. (2007). The neuropharmacology
of the ketogenic diet. Pediatr Neurol 36, 281–292.
Haymond, M.W., Howard, C., Ben-Galim, E., and
DeVivo, D.C. (1983). Effects of ketosis on glucose
flux in children and adults. Am J Physiol 245,
E373–E378.
Heinemann, U., Beck, H., Dreier, J.P., Ficker, E., Stabel,
J., and Zhang, C.L. (1992). The dentate gyrus as a
regulated gate for the propagation of epileptiform
activity. Epilepsy Res (Suppl 7), 273–280.
Hernández-Sánchez, C., Basile, A.S., Fedorova, I.,
Arima, H., Stannard, B., Fernandez, A.M., Ito, Y.,
and LeRoith, D. (2001). Mice transgenically over-
expressing sulfonylurea receptor 1 in forebrain
resist seizure induction and excitotoxic neuron
death. Proc Natl Acad Sci USA 98, 3549–3554.
Hickman, J.A., Hardwick, J.M., Kaczmarek, L.K., and
Jonas, E.A. (2008). Bcl-xL inhibitor ABT-737
reveals a dual role for Bcl-xL in synaptic transmis-
sion. J Neurophysiol 99, 1515–1522.
Hoffman, J.F., Dodson, A., and Proverbio, F. (2009). On
the functional use of the membrane compartmen-
talized pool of ATP by the Na+ and Ca++ pumps in
human red blood cell ghosts. J Gen Physiol 134,
351–361.
Hong, M., Kefaloyianni, E., Bao, L., Malester, B.,
Delaroche, D., Neubert, T.A., and Coetzee, W.A.
(2011). Cardiac ATP-sensitive K+ channel associ-
ates with the glycolytic enzyme complex. FASEB J
25, 2456–2467.
Hsu, D. (2007). The dentate gyrus as a filter or gate: a
look back and a look ahead. Prog Brain Res 163,
601–613.
Huang, C.-W., Huang, C.-C., and Wu, S.-N. (2006).
The opening effect of pregabalin on ATP-sensitive
potassium channels in differentiated hippocampal
neuron–derived H19-7 Cells. Epilepsia 47, 720–72
Huang, C.-W., Huang, C.-C., Cheng, J.-T., Tsai, J.-J.,
and Wu, S.-N. (2007). Glucose and hippocampal
neuronal excitability: Role of ATP-sensitive potas-
sium channels. J Neurosci Res 85, 1468–1477.
Huttenlocher, P.R. (1976). Ketonemia and sei-
zures: metabolic and anticonvulsant effects of two
ketogenic diets in childhood epilepsy. Pediatr Res
10, 536–540.
Iadarola, M.J., and Gale, K. (1982). Substantia
nigra:  site of anticonvulsant activity mediated
279
by gamma-aminobutyric acid. Science 218,
1237–1240.
Ivannikov, M.V., Sugimori, M., and Llinás, R.R. (2010).
Calcium clearance and its energy requirements in
cerebellar neurons. Cell Calcium 47, 507–513.
Jiao, S., and Li, Z. (2011). Nonapoptotic function of
BAD and BAX in long-term depression of synaptic
transmission. Neuron 70, 758–772.
Juge, N., Gray, J.A., Omote, H., Miyaji, T., Inoue, T.,
Hara, C., Uneyama, H., Edwards, R.H., Nicoll,
R.A., and Moriyama, Y. (2010). Metabolic con-
trol of vesicular glutamate transport and release.
Neuron 68, 99–112.
Karschin, C., Ecke, C., Ashcroft, F.M., and Karschin,
A. (1997). Overlapping distribution of KATP chan-
nel-forming Kir6.2 subunit and the sulfonylurea
receptor SUR1 in rodent brain. FEBS Lett 401,
59–64.
Katsura, K., Minamisawa, H., Ekholm, A., Folbergrová,
J., and Siesjö, B.K. (1992). Changes of labile metab-
olites during anoxia in moderately hypo- and
hyperthermic rats: correlation to membrane fluxes
of K+. Brain Res 590, 6–12.
Krook-Magnuson, E., Armstrong, C., Oijala, M., and
Soltesz, I. (2013). On-demand optogenetic control
of spontaneous seizures in temporal lobe epilepsy.
Nat Commun 4, 1376.
Krook-Magnuson, E., Armstrong, C., Bui, A., Lew, S.,
Oijala, M., and Soltesz, I. (2015). In vivo evalua-
tion of the dentate gate theory in epilepsy. J Physiol
593, 2379–2388.
Liss, B., and Roeper, J. (2001). Molecular physiology
of neuronal KATP channels. Mol Membr Biol 18,
117–127.
Lothman, E.W., Stringer, J.L., and Bertram, E.H.
(1992). The dentate gyrus as a control point for
seizures in the hippocampus and beyond. Epilepsy
Res (Suppl 7), 301–313.
Lutas, A., and Yellen, G. (2013). The ketogenic
diet: metabolic influences on brain excitability and
epilepsy. Trends Neurosci 36, 32–40.
Lutas, A., Birnbaumer, L., and Yellen, G. (2014).
Metabolism regulates the spontaneous firing of
substantia nigra pars reticulata neurons via KATP
and nonselective cation channels. J Neurosci 34,
16336–16347.
Ma, W., Berg, J., and Yellen, G. (2007). Ketogenic
diet metabolites reduce firing in central neu-
rons by opening KATP channels. J Neurosci 27,
3618–3625.
Masino, S.A., Li, T., Theofilas, P., Sandau, U.S., Ruskin,
D.N., Fredholm, B.B., Geiger, J.D., Aronica, E., and
Boison, D. (2011). A ketogenic diet suppresses sei-
zures in mice through adenosine A1 receptors. J
Clin Invest 121, 2679–2683.
Matsumoto, N., Komiyama, S., and Akaike, N. (2002).
Pre- and postsynaptic ATP- sensitive potassium
channels during metabolic inhibition of rat hippo-
campal CA1 neurons. J Physiol 541, 511–520.
280
280 section III: Ketogenic Diet in the Laboratory
McNamara, J.O., Galloway, M.T., Rigsbee, L.C., and
Shin, C. (1984). Evidence implicating substantia
nigra in regulation of kindled seizure threshold. J
Neurosci 4, 2410–2417.
Mercer, R.W., and Dunham, P.B. (1981). Membrane-
bound ATP fuels the Na/K pump. Studies on
membrane-bound glycolytic enzymes on inside-
out vesicles from human red cell membranes. J
Gen Physiol. 78, 547–568.
Mergenthaler, P., Lindauer, U., Dienel, G.A., and
Meisel, A. (2013). Sugar for the brain: the role of
glucose in physiological and pathological brain
function. Trends Neurosci 36, 587–597.
Moldoveanu, T., Follis, A.V., Kriwacki, R.W., and
Green, D.R. (2014). Many players in BCL-2 family
affairs. Trends Biochem Sci 39, 101–111.
Murphy, B.M., Engel, T., Paucard, A., Hatazaki, S.,
Mouri, G., Tanaka, K., Tuffy, L.P., Jimenez-Mateos,
E.M., Woods, I., Dunleavy, M., et  al. (2010).
Contrasting patterns of Bim induction and neuro-
protection in Bim-deficient mice between hippo-
campus and neocortex after status epilepticus. Cell
Death Differ 17, 459–468.
Neal, E.G., Chaffe, H., Schwartz, R.H., Lawson, M.S.,
Edwards, N., Fitzsimmons, G., Whitney, A., and
Cross, J.H. (2009). A randomized trial of classical
and medium-chain triglyceride ketogenic diets in
the treatment of childhood epilepsy. Epilepsia 50,
1109–1117.
Nichols, C.G. (2006). KATP channels as molecular sen-
sors of cellular metabolism. Nature 440, 470–476.
Owen, O.E., Morgan, A.P., Kemp, H.G., Sullivan, J.M.,
Herrera, M.G., and Cahill, G.F. (1967). Brain metab-
olism during fasting. J Clin Invest 46, 1589–1595.
Paz, J.T., Chavez, M., Saillet, S., Deniau, J.M., and
Charpier, S. (2007). Activity of ventral medial tha-
lamic neurons during absence seizures and modu-
lation of cortical paroxysms by the nigrothalamic
pathway. J Neurosci 27, 929–941.
Pelletier, M.R., Pahapill, P.A., Pennefather, P.S., and
Carlen, P.L. (2000). Analysis of single KATP chan-
nels in mammalian dentate gyrus granule cells. J
Neurophysiol 84, 2291–2301.
Pfeifer, H.H., Lyczkowski, D.A., and Thiele, E.A.
(2008). Low glycemic index treatment: implemen-
tation and new insights into efficacy. Epilepsia 49
(Suppl 8), 42–45.
Proks, P., and Ashcroft, F.M. (2009). Modeling KATP
channel gating and its regulation. Prog Biophys
Mol Biol 99, 7–19.
Proverbio, F., and Hoffman, J.F. (1977). Membrane
compartmentalized ATP and its preferential use by
the Na,K-ATPase of human red cell ghosts. J Gen
Physiol 69, 605–632.
Sada, N., Lee, S., Katsu, T., Otsuki, T., and Inoue,
T. (2015). Epilepsy treatment. Targeting LDH
enzymes with a stiripentol analog to treat epilepsy.
Science 347, 1362–1367.
Schmidt-Kastner, R., and Freund, T.F. (1991). Selective
vulnerability of the hippocampus in brain isch-
emia. Neuroscience 40, 599–636.
Schwanstecher, C., and Panten, U. (1993). Tolbutamide-
and diazoxide-sensitive K+ channel in neurons
of substantia nigra pars reticulata. Naunyn.
Schmiedebergs Arch Pharmacol 348, 113–117.
Smith, M.L., Auer, R.N., and Siesjö, B.K. (1984). The
density and distribution of ischemic brain injury
in the rat following 2-10 min of forebrain isch-
emia. Acta Neuropathol (Berl) 64, 319–332.
Stanford, I.M., and Lacey, M.G. (1996).
Electrophysiological investigation of adenosine
trisphosphate-sensitive potassium channels in the
rat substantia nigra pars reticulata. Neuroscience
74, 499–509.
Tanner, G.R., Lutas, A., Martínez-François, J.R.,
and Yellen, G. (2011). Single KATP channel open-
ing in response to action potential firing in
mouse dentate granule neurons. J Neurosci 31,
8689–8696.
Tantama, M., Martínez-François, J.R., Mongeon, R.,
and Yellen, G. (2013). Imaging energy status in live
cells with a fluorescent biosensor of the intracellu-
lar ATP-to-ADP ratio. Nat Commun 4, 2550.
Thiele E.A. (2003). Assessing the efficacy of antiepi-
leptic treatments: the ketogenic diet. Epilepsia 44
(Suppl 7), 26–29.
Valente-Silva, P., Lemos, C., Köfalvi, A., Cunha,
R.A., and Jones, J.G. (2015). Ketone bodies effec-
tively compete with glucose for neuronal acetyl-
CoA generation in rat hippocampal slices. NMR
Biomed 28, 1111–1116.
Vercueil, L., Benazzouz, A., Deransart, C., Bressand,
K., Marescaux, C., Depaulis, A., and Benabid, A.L.
(1998). High-frequency stimulation of the subtha-
lamic nucleus suppresses absence seizures in the
rat: comparison with neurotoxic lesions. Epilepsy
Res 31, 39–46.
Yamada, K., and Inagaki, N. (2005). Neuroprotection
by KATP channels. J Mol Cell Cardiol 38, 945–949.
Yamada, K., Ji, J.J., Yuan, H., Miki, T., Sato, S.,
Horimoto, N., Shimizu, T., Seino, S., and Inagaki,
N. (2001). Protective role of ATP-sensitive potas-
sium channels in hypoxia- induced generalized
seizure. Science 292, 1543–1546.
Zawar, C., Plant, T.D., Schirra, C., Konnerth, A., and
Neumcke, B. (1999). Cell-type specific expression
of ATP-sensitive potassium channels in the rat
hippocampus. J Physiol 514 ( Pt 2), 327–341.
Zielke, H.R., Zielke, C.L., and Baab, P.J. (2009). Direct
measurement of oxidative metabolism in the living
brain by microdialysis: a review. J Neurochem 109
(Suppl 1), 24–29.
281
29
Lactate Dehydrogenase
A Novel Metabolic Target
N AG I S A S A D A , P H D A N D T S U Y O S H I I N O U E , P H D
T H E K E TO G E N I C D I E T
F O R   D R U G - R E S I S TA N T
EPILEPSY
Epilepsy is a neurological disorder that is charac-
terized by the hyperexcitation of electrical activi-
ties in the brain. Approximately 1% of the world’s
population has epilepsy, and one-third of epileptic
patients are resistant to currently available antiepi-
leptic drugs (Kwan and Brodie, 2000). Therefore,
it is important to develop new antiepileptic drugs
that can treat the drug-resistant epilepsy. As a step
toward achieving this goal, “endogenous proteins”
that have the ability to suppress epileptic seizures
need to be identified, and will become target mol-
ecules for the development of antiepileptic drugs.
“Chemical compounds” that act on these target
molecules and suppress epileptic seizures must
then be identified, and will become prototypes
of new antiepileptic drugs. Thus, the search for
new target molecules in basic neuroscience will
ultimately lead to drug development for drug-
resistant epilepsy.
Since epileptic seizures are caused by the
hyperexcitation of electrical activities in the brain,
antiepileptic drugs also need to be designed in
order to suppress the neuronal hyperexcitation.
Electrical activities in the brain are generated by
ion channels, synaptic receptors, and neurotrans-
mitter transporters. Therefore, currently avail-
able antiepileptic drugs have been designed to act
on the molecules generating electrical currents
(Meldrum and Rogawski, 2007). However, these
antiepileptic drugs are ineffective for one-third of
epileptic patients (Kwan and Brodie, 2000); that
is, other molecular targets need to be identified.
The ketogenic diet is known to be effective for
some patients with drug-resistant epilepsy (Neal
et  al., 2008), but there are no antiepileptic drugs
that mimic the ketogenic diet. Based on this back-
ground, neuroscientists have recently focused on
the molecular mechanisms underlying the antiepi-
leptic effects of the ketogenic diet.
MECHANISMS RESPONSIBLE
FOR THE ANTIEPILEPTIC
EFFECTS OF THE
K E TO G E N I C   D I E T
The ketogenic diet mainly elicits two metabolic
changes in the body; increases in ketone bodies
and decreases in glucose (Bough et al., 2006). The
neural inhibition and antiepileptic effects elicited
by these two metabolic changes have been actively
studied (Lutas and Yellen, 2013). The following
rodent studies have demonstrated the antiepileptic
effects of ketone bodies. Ketone bodies consist of
acetoacetate, β-hydroxybutyrate, and acetone. The
ketogenic diet was previously shown to increase
β-hydroxybutyrate levels up to ~8 mM in the
blood plasma of rodents (Bough et al., 1999). An
intraperitoneal injection of ketone bodies directly
suppresses seizures in vivo in audiogenic seizure-
susceptible mice (Rho et al., 2002). At the molecu-
lar and cellular levels, ketone bodies decrease the
firing rate of neurons in the substantia nigra via
adenosine 5’-triphosphate (ATP)-sensitive K+
channels (KATP channels) (Ma et al., 2007). Single
channel recordings have revealed that ketone
bodies open KATP channels in the dentate granule
cells of the hippocampus (Tanner et  al., 2011).
Acetoacetate is also known to be an inhibitor of
vesicular glutamate transporters (VGLUTs) (Juge
et al., 2010). The role of the VGLUTs is to fill gluta-
mate into synaptic vesicles and regulate excitatory
synaptic transmission in the brain, and thus, ace-
toacetate reduces miniature excitatory postsyn-
aptic currents (miniature EPSCs) and suppresses
4-aminopyridine-induced acute seizures in vivo
(Juge et al., 2010).
In terms of decreases in glucose, seizures are
suppressed by 2-deoxy-D-glucose, a glycolytic
282
282 section III: Ketogenic Diet in the Laboratory
inhibitor (Garriga-Canut et  al., 2006). Glucose
metabolism is reduced in BAD (BCL-2-associated
agonist of cell death) knockout mice, which opens
metabolically sensitive KATP channels and leads to
resistance to convulsants such as kainic acid and
pentylenetetrazol (Giménez-Cassina et al., 2012).
The ketogenic diet has also been shown to sup-
press seizures in mice by increasing the activation
of adenosine A1 receptors (Masino et  al., 2011).
In vitro recordings from hippocampal slices have
revealed that reductions in glucose hyperpolarize
pyramidal neurons, and this is mediated by ade-
nosine A1 receptors and KATP channels (Kawamura
et  al., 2010). Consistent with these findings, the
direct activation of adenosine A1 receptors has
been shown to suppress chronic seizures in a
mouse model of mesial temporal lobe epilepsy
(Gouder et al., 2003).
These extensive studies have uncovered the
mechanisms underlying the antiepileptic effects
of the ketogenic diet. However, all of the mol-
ecules revealed by these studies are ion channels
(KATP channels), synaptic receptors (adenosine
A1 receptors), and neurotransmitter transporters
(VGLUTs); namely, molecules regulating elec-
trical currents. Since the ketogenic diet changes
energy metabolites (glucose and ketone bodies),
it is hypothesized that the brain has “metabolic
pathways” that play key roles in the antiepilep-
tic effects of the ketogenic diet. In other words,
there are assumed to be “metabolic enzymes”
that regulate electrical activities in neurons and
suppress seizures in vivo in the same manner
with the ketogenic diet. In order to explore these
metabolic pathways and enzymes, we focused on
the astrocyte-neuron lactate shuttle in the brain,
because this lactate shuttle was known to be a met-
abolic pathway involved in the regulation of elec-
trical activities in the brain (Rouach et  al., 2008,
Parsons and Hirasawa, 2010).
ASTROCYTE-NEURON
L A C TAT E S H U T T L E
Glucose is the main energy source in the brain, and
produces ATP in neurons and glial cells. The con-
centration of glucose in vivo is approximately 2.5
mM in the brain (Silver and Erecińska, 1994), and
is maintained at a lower concentration than that
in the blood. Glucose is directly transported into
neurons and then converted to pyruvate by glycol-
ysis in neurons, which produces ATP in the TCA
cycle (Figure 29.1). The role of astrocytes (a type
of glial cell) as the energy supplier to neurons has
recently received increasing attention (Bélanger
et al., 2011). As shown in Figure 29.1, glucose is
initially transported into astrocytes, and converted
to lactate by lactate dehydrogenase (LDH). This
lactate is then released to extracellular spaces,
transported into neurons, and converted to pyru-
vate by LDH in glycolysis, which produces ATP in
the TCA cycle (Figure 29.1). This metabolic com-
munication from astrocytes to neurons is called
the astrocyte-neuron lactate shuttle (Bélanger et
al., 2011).
Several lines of evidence support that this
astrocyte-neuron lactate shuttle is used in the sup-
ply of energy to neurons. First, astrocytes show
lower rates of oxidative metabolism than neu-
rons, and consequently release a large amount
of lactate to the extracellular spaces (Itoh et  al.,
2003; Bouzier-Sore et  al., 2006). Second, lactate
appears to be used as an energy source preferred
over glucose in the brain (Larrabee, 1995; Smith
et al., 2003). Third, although glucose uptakes into
neurons and astrocytes are similar in the resting
state of the brain (Nehlig et  al., 2004; Chuquet
et al., 2010), increases in neuronal activities elicit
the higher glucose uptakes in astrocytes (Chuquet
et  al., 2010)  and also elicit sustained increases
in extracellular lactate (Hu and Wilson, 1997).
Thus, the energy supply to neurons is achieved by
not only the direct glucose pathway to neurons
but also the astrocyte-neuron lactate shuttle (see
Figure 29.1).
There is also growing evidence to show that the
astrocyte-neuron lactate shuttle regulates electri-
cal activities in neurons. First of all, the following
studies have demonstrated that lactate regulates
electrical activities in neurons. The inhibition of
glycolysis reduces synaptic transmission in the
hippocampus, and this is rescued by the presence
of lactate (Schurr et al., 1988), which demonstrates
that the metabolism of lactate to pyruvate in neu-
rons contributes to the maintenance of synaptic
transmission. Consistent with this finding, the
inhibition of monocarboxylate transporters that
transport lactate into neurons also reduces synap-
tic transmission in the hippocampus (Izumi et al.,
1997). Furthermore, orexin-containing neurons in
the hypothalamus are hyperpolarized by the inhi-
bition of monocarboxylate transporters (Parsons
and Hirasawa, 2010). Importantly, they are also
hyperpolarized by fluoroacetate, a glial toxin, and
this hyperpolarization is recovered by the applica-
tion of lactate (Parsons and Hirasawa, 2010), These
findings suggest that lactate released from astro-
cytes regulates membrane potentials in neurons.
Rouach and colleagues provided more direct
evidence to show that electrical activities in neu-
rons are regulated by the metabolic pathway from
283
Astrocyte
Lactate
LDH
Glucose
Ketone
body
FIGURE 29.1 The astrocyte-neuron lactate shuttle. Glucose is directly transported into neurons, and then converted
to pyruvate by glycolysis. As an alternative pathway, glucose is transported into astrocytes and converted to lactate by
LDH, which is released into extracellular spaces: lactate is then transported into neurons, and converted to pyruvate
by glycolysis. This pathway is called the astrocyte-neuron lactate shuttle. Decreases in glucose and increases in ketone
bodies are elicited by the ketogenic diet. Ketone bodies are transported into neurons, which bypass glycolysis and
directly enter the TCA cycle in neurons.
astrocytes (Rouach et al., 2008). Astrocytes in the
hippocampus are connected to one another by gap
junctions, and thus, small molecules injected into
single astrocytes using patch pipettes diffuse into
many neighboring astrocytes (D’Ambrosio et  al.,
1998). By using this property, a group of nearby
astrocytes can be controlled by selective filling of
active small molecules into single astrocytes using
patch pipettes (Rouach et  al., 2008; Henneberger
et  al., 2010). Rouach and colleagues made filling
of glucose into single astrocytes in hippocampal
slices, and then examined the effects of the depri-
vation of extracellular glucose on hippocampal
synaptic transmission. They demonstrated that,
although glucose deprivation reduced synaptic
transmission, these reductions were rescued by the
selective filling of glucose into astrocytes (Rouach
et  al., 2008), showing that hippocampal synaptic
transmission is regulated by metabolic communi-
cation from astrocytes. Rouach and colleagues also
showed that epileptiform activities in hippocam-
pal slices were regulated by the selective filling of
glucose into astrocytes (Rouach et al., 2008).
L A C TAT E D E H Y D R O G E N A S E :
A   M E TA B O L I C TA R G E T
FOR EPILEPSY
The findings of these studies prompted us to inves-
tigate electrical regulation by the astrocyte-neuron
lactate shuttle, and we recently reported that this
lactate shuttle is involved in the antiepileptic
effects of the ketogenic diet (Sada et al., 2015). The
ketogenic diet is known to increase ketone bodies
and decrease glucose in the body (Bough et  al.,
2006). Therefore, we made patch clamp recordings
from neurons in the subthalamic nucleus in mouse
brain slices, and switched glucose in the extra-
cellular perfusate (artificial cerebrospinal fluid;
Chapter 29: Lactate Dehydrogenase 283
Neuron
Glycolysis
LDH
Pyruvate
TCA cycle
ACSF) to ketone bodies. The switch from glucose
to ketone bodies hyperpolarized subthalamic neu-
rons; and this hyperpolarization was recovered by
the addition of lactate in ACSF (Sada et al., 2015).
Based on the metabolic pathways shown in Figure
29.1, it is likely that the hyperpolarization is elic-
ited by the decrease in extracellular lactate derived
from the switch of glucose to ketone bodies, and
recovered by the addition of lactate to ACSF. Thus,
the astrocyte-neuron lactate shuttle is likely to reg-
ulate membrane potentials in neurons.
In order to confirm this idea, we focused on
LDH, a metabolic enzyme located in the astrocyte-
neuron lactate shuttle (Figure 29.1). Thus, this lac-
tate shuttle can be suppressed by inhibiting LDH.
We used oxamate, an LDH inhibitor that has been
used to inhibit LDH in the brain (Lam et al., 2005),
and examined the effects of the LDH inhibitor on
membrane potentials in neurons (Sada et al., 2015;
see Figure 29.2). The inhibition of LDH hyperpo-
larized neurons in the subthalamic nucleus, and
also hyperpolarized pyramidal cells in the hippo-
campus (Figure  29.2A). Furthermore, the hyper-
polarization elicited by the LDH inhibition was not
recovered by the addition of lactate, but was recov-
ered by the addition of pyruvate (Figure  29.2B).
Lactate is an upstream metabolite of LDH in neu-
rons, whereas pyruvate is a downstream metabo-
lite in neurons (see Figure 29.1). Thus, neuronal
LDH in the lactate shuttle regulates membrane
potentials in neurons (Figure 29.2C).
We also obtained direct evidence to show that
membrane potentials in neurons are regulated by
lactate released from astrocytes (Sada et al., 2015;
see Figure 29.3). Patch-clamp recordings were
obtained from neighboring pairs of pyramidal cells
and astrocytes in hippocampal slices, and oxam-
ate was selectively applied to recorded astrocytes
284

284 section III: Ketogenic Diet in the Laboratory

(a)

–60 mV
3 mV
5 min

Oxamate
(b)

Oxamate Oxamate
28 min 2 mV 28 min 2 mV
2 min 2 min

Oxamate Oxamate

Lactate Pyruvate

(c)
Astrocyte Neuron
Lactate
Glycolysis
LDH
LDH

Pyruvate

Glucose

TCA cycle
Neuronal
Inhibition

FIGURE 29.2 LDH is an electrical regulator in neurons. (A) Pyramidal cells in the hippocampus are hyperpolarized
by oxamate, an LDH inhibitor. (B) The hyperpolarization elicited by the LDH inhibitor is recovered by pyruvate, but not
by lactate. Lactate and pyruvate are located upstream and downstream of neuronal LDH on the astrocyte-neuron lactate
shuttle (see Figure 29.1), respectively. Thus, the lactate-to-pyruvate conversion via LDH in neurons is critical for this
electrical regulation. (C) The summary scheme showing the neuronal inhibition elicited by inhibiting LDH in neurons.
Reproduced from Sada et al. (2015).

through the patch pipette (Figure  29.3A). The
selective inhibition of LDH in astrocytes elicited
hyperpolarization in neighboring pyramidal cells
(Figure 29.3B), which demonstrated that LDH in
astrocytes regulates membrane potentials in neu-
rons. In order to further confirm that this hyper-
polarization in pyramidal cells was actually due to
reductions in the release of lactate from astrocytes
(see Figure 29.3A), we performed the same experi-
ments under the extracellular perfusion of lactate.
The hyperpolarization induced by filling oxamate
into astrocytes (see Figure 29.3B) was not observed
when lactate was present in the extracellular per-
fusate (Figure 29.3C). Thus, astrocytic LDH in this
lactate shuttle also regulates membrane potentials
in neurons (Figure 29.3D).
These in vitro studies revealed that the release
of lactate from astrocytes to neurons regulates
membrane potentials in neurons. We also obtained
evidence to show that this lactate release is actually
weakened by the ketogenic diet (Sada et al., 2015;
see Figure 29.4). Mice were fed a ketogenic diet
for about 3 weeks, and the lactate concentration
in the hippocampus was measured and compared
with that in the hippocampus of age-matched
mice fed a standard diet. The measurement of lac-
tate concentration revealed that the hippocampal
lactate was lower in mice fed the ketogenic diet
than in those fed the standard diet (Figure 29.4A).
Thus, the release of lactate in this lactate shuttle
is lowered by the ingestion of the ketogenic diet
(Figure 29.4B).
Taken together, these findings show that the
astrocyte-neuron lactate shuttle is regulated by the
ketogenic diet, and LDH in this lactate shuttle is a
key enzyme that regulates membrane potentials in
285

(A) (C) Paired recording in lactate

O Glucose Glucose Oxamate Lactate Oxamate Lactate
xa
m
at Rupture
e
Patch Patch
pipette pipette “Rupture”
LDH Lactate LDH

Pyruvate

Astrocyte Pyramidal cell

Astrocyte
(B) Paired recording –77 mV
10 mV
Oxamate Oxamate
2 min
Rupture
“Rupture”

–59 mV
Pyramidal 1 mV
Astrocyte cell 2 min

–76 mV (D)
10 mV
2 min
Astrocyte Neuron
Lactate Glycolysis
LDH
–61 mV LDH

Pyruvate
Pyramidal
cell
Glucose

Neuronal TCA cycle

1 mV Inhibition
2 min

FIGURE  29.3 Electrical regulation by the astrocyte-neuron lactate shuttle. (A) The experimental design used for
simultaneous recordings from pyramidal cells and astrocytes in the hippocampus. LDH in astrocytes are selectively
inhibited by the intracellular application of oxamate via patch pipettes. (B)  Pyramidal cells are hyperpolarized by
the LDH inhibition in astrocytes. Neurons are recorded in the whole-cell mode using normal intracellular solution,
whereas astrocytes are recorded in the cell-attached mode using intracellular solution including oxamate. Oxamate
is selectively applied by rupturing the patch membranes in astrocytes. (C) Pyramidal cells are not hyperpolarized by
the LDH inhibition in astrocytes when lactate is present extracellularly. Thus, lactate is a mediator for this astrocyte-
induced electrical regulation in neurons. (D) The summary scheme showing the neuronal inhibition by inhibiting LDH
in astrocytes. Reproduced from Sada et al. (2015).

(A) (B)
0.05
Astrocyte Neuron
Lactate in the Hippocampus

0.04 **
Lactate Glycolysis
(nmol/μg protein)

LDH
0.03 LDH

Pyruvate
0.02

0.01 Glucose

TCA cycle
0 Ketogenic
Standard Ketogenic diet
diet diet

FIGURE 29.4 Decrease in hippocampal lactate by the ketogenic diet. (A) Lactate concentrations in the hippocampus
are lower in mice fed a ketogenic diet than in those fed a standard diet. **p < .01 (unpaired t-test; n = 11 in each group).
(B) The summary scheme showing the reduction in the release of lactate elicited by the ketogenic diet. Reproduced from
Sada et al. (2015).
286
286 section III: Ketogenic Diet in the Laboratory
neurons. We further reported that seizures in vivo
were suppressed by the inhibition of LDH (Sada
et  al., 2015). Microinjection of kainic acid into
the hippocampus is known to produce a mouse
model of mesial temporal lobe epilepsy, which
exhibits spontaneous paroxysmal discharges with
abnormal morphology in the hippocampus (Riban
et  al., 2002; Gouder et  al., 2003; Heinrich et  al.,
2006). The LDH inhibitor oxamate suppressed
the paroxysmal discharges in the hippocampus
of this chronic seizure model (Sada et  al., 2015).
The in vivo knockdown of lactate dehydrogenase
A  (LDHA), a subunit of LDH, by an antisense
oligodeoxynucleotide also exerted antiepileptic
actions in this seizure model. Thus, the inhibition
of LDH suppresses seizures in vivo.
FUTURE DIRECTIONS
T O WA R D   D R U G
DEVELOPMENT
The findings of our recent study (Sada et  al.,
2015) indicate that LDH is a metabolic target for
epilepsy, and also that the inhibition of LDH mim-
ics the antiepileptic effects of the ketogenic diet.
This metabolic target is clearly different from mol-
ecules that have already been identified in previ-
ous studies about the ketogenic diet; KATP channels
(Ma et al., 2007; Tanner et al., 2011), adenosine A1
receptors (Masino et al., 2011), and VGLUTs (Juge
et  al., 2010)  are molecules that directly regulate
electrical currents in neurons. Lactate dehydro-
genase as a metabolic target against epilepsy will
be useful in the development of new antiepileptic
drugs. The next step toward drug development is
to identify chemical compounds that act on LDH.
Although many antiepileptic drugs are now clini-
cally used, none of antiepileptic drugs have been
designed to act on LDH. Antiepileptic drugs to
act on LDH are also not known. Therefore, we
explored LDH inhibitors from clinically used anti-
epileptic drugs, and found that stiripentol, a clini-
cal drug for Dravet syndrome (Chiron et al., 2000),
is an LDH inhibitor (Sada et al., 2015). By modify-
ing its chemical structure, we further identified a
stiripentol analog that inhibits LDH and strongly
suppresses seizures in vivo (Sada et  al., 2015).
Thus, new antiepileptic drugs to mimic the keto-
genic diet can be developed in the future by target-
ing LDH enzymes with stiripentol derivatives.
In order to develop new antiepileptic drugs,
it will be important to create more powerful and
selective LDH inhibitors. The design and synthe-
sis of selective inhibitors for each LDH subunit
(LDHA or LDHB) and/or for the unidirectional
conversion of LDH activities (pyruvate-to-lactate
or lactate-to-pyruvate conversion) will be useful
because these are expected to act on more spe-
cific components and functions in the brain. To
date, LDH inhibitors have been synthesized to
create antimalarial compounds (Deck et al., 1998;
Cameron et  al., 2004). More importantly, LDHA
is known to be involved in antitumor actions
based on the Warburg effect (Fantin et al., 2006).
Also, LDHA inhibitors have already been syn-
thesized and studied as antitumor reagents (Le
et al., 2010; Granchi et al., 2011; Farabegoli et al.,
2012). Drug development for epilepsy by targeting
LDH enzymes, which is based on the antiepileptic
mechanism of the ketogenic diet, may also lead to
new antitumor drugs based on the Warburg effect.
REFERENCES
Bélanger, M., Allaman, I., and Magistretti, P.J. (2011).
Brain energy metabolism:  focus on astrocyte-
neuron metabolic cooperation. Cell Metab 14,
724–728.
Bough, K.J., Valiyil, R., Han, F.T., and Eagles, D.A.
(1999). Seizure resistance is dependent upon age
and calorie restriction in rats fed a ketogenic diet.
Epilepsy Res. 35, 21–28.
Bough, K.J., Wetherington, J., Hassel, B., Pare, J.F.,
Gawryluk, J.W., Greene, J.G., Shaw, R., Smith,
Y., Geiger, J.D., and Dingledine, R.J. (2006).
Mitochondrial biogenesis in the anticonvulsant
mechanism of the ketogenic diet. Ann Neurol 60,
223–235.
Bouzier-Sore, A.K., Voisin, P., Bouchaud, V.,
Bezancon, E., Franconi, J.M., and Pellerin, L.
(2006). Competition between glucose and lactate
as oxidative energy substrates in both neurons
and astrocytes:  a comparative NMR study. Eur J
Neurosci 24, 1687–1694.
Cameron, A., Read, J., Tranter, R., Winter, V.J., Sessions,
R.B., Brady, R.L., Vivas, L., Easton, A., Kendrick,
H., Croft, S.L., et  al. (2004). Identification and
activity of a series of azole-based compounds
with lactate dehydrogenase-directed anti-malarial
activity. J Biol Chem 279, 31429–31439.
Chiron, C., Marchand, M.C., Tran, A., Rey, E.,
d’Athis, P., Vincent, J., Dulac, O., and Pons, G.
(2000). Stiripentol in severe myoclonic epilepsy
in infancy:  a randomised placebo-controlled
syndrome-dedicated trial. STICLO study group.
Lancet 356, 1638–1642.
Chuquet, J., Quilichini, P., Nimchinsky, E.A., and
Buzsáki G. (2010). Predominant enhancement
of glucose uptake in astrocytes versus neurons
during activation of the somatosensory cortex. J
Neurosci 30, 15298–15303.
D’Ambrosio, R., Wenzel, J., Schwartzkroin, P.A.,
McKhann, G.M., 2nd, and Janigro, D. (1998).
Functional specialization and topographic
287
segregation of hippocampal astrocytes. J Neurosci
18, 4425–4438.
Deck, L.M., Royer, R.E., Chamblee, B.B., Hernandez,
V.M., Malone, R.R., Torres, J.E., Hunsaker, L.A.
Piper, R.C., Makler, M.T., and Vander Jagt, D.L.
(1998). Selective inhibitors of human lactate dehy-
drogenases and lactate dehydrogenase from the
malarial parasite Plasmodium falciparum. J Med
Chem 41, 3879–3887.
Fantin, R.F., St-Pierre, J., and Leder, P. (2006).
Attenuation of LDH-A expression uncovers a link
between glycolysis, mitochondrial physiology, and
tumor maintenance. Cancer Cell 9, 425–434.
Farabegoli, F., Vettraino, M., Manerba, M., Fiume, L.,
Roberti, M., and Di Stefano, G. (2012). Galloflavin,
a new lactate dehydrogenase inhibitor, induces the
death of human breast cancer cells with different
glycolytic attitude by affecting distinct signaling
pathways. Eur J Pharm Sci 47, 729–738.
Garriga-Canut, M., Schoenike, B., Qazi, R., Bergendahl,
K., Daley, T.J., Pfender, R.M., Morrison, J.F.,
Ockuly, J., Stafstrom, C., Sutula, T., and Roopra,
A. (2006). 2-Deoxy-D-glucose reduces epilepsy
progression by NRSF-CtBP-dependent metabolic
regulation of chromatin structure. Nat Neurosci 9,
1382–1387.
Giménez-Cassina, A., Martínez-François J.R., Fisher,
J.K., Szlyk, B., Polak, K., Wiwczar, J., Tanner, G.R.,
Lutas, A., Yellen, G., and Danial, N.N. (2012).
BAD-dependent regulation of fuel metabolism
and KATP channel activity confers resistance to epi-
leptic seizures. Neuron 74, 719–730.
Gouder, N., Fritschy, J.M., and Boison, D. (2003).
Seizure suppression by adenosine A1 receptor acti-
vation in a mouse model of pharmacoresistant epi-
lepsy. Epilepsia 44, 877–885.
Granchi, C., Roy, S., Giacomelli, C., Macchia, M.,
Tuccinardi, T., Martinelli, A., Lanza, M., Betti, L.,
Giannaccini, G., Lucacchini, A., Funel, N., et  al.
(2011). Discovery of N-hydroxyindole-based
inhibitors of human lactate dehydrogenase iso-
form A (LDH-A) as starvation agents against can-
cer cells. J Med Chem 54, 1599–1612.
Heinrich, C., Nitta, N., Flubacher, A., Müller, M.,
Fahrner, A., Kirsch, M., Freiman, T., Suzuki, F.,
Depaulis, A., Frotscher, M., et  al. (2006). Reelin
deficiency and displacement of mature neurons,
but not neurogenesis, underlie the formation of
granule cell dispersion in the epileptic hippocam-
pus. J Neurosci 26, 4701–4713.
Henneberger, C., Papouin, T., Oliet, S.H., and Rusakov,
D.A. (2010). Long-term potentiation depends on
release of D-serine from astrocytes. Nature 463,
232–236.
Hu, Y., and Wilson, G.S. (1997). A temporary local
energy pool coupled to neuronal activity:  fluc-
tuations of extracellular lactate levels in rat brain
monitored with rapid-response enzyme-based
sensor. J Neurochem 69, 1484–1490.
Chapter 29: Lactate Dehydrogenase 287
Itoh, Y., Esaki, T., Shimoji, K., Cook, M., Law,
M.J., Kaufman, E., and Sokoloff, L. (2003).
Dichloroacetate effects on glucose and lactate oxi-
dation by neurons and astroglia in vitro and on
glucose utilization by brain in vivo. Proc Natl Acad
Sci USA 100, 4879–4884.
Izumi, Y., Benz, A.M., Katsuki, H., and Zorumski, C.F.
(1997). Endogenous monocarboxylates sustain
hippocampal synaptic function and morphologi-
cal integrity during energy deprivation. J Neurosci
17, 9448–9457.
Juge, N., Gray, J.A., Omote, H., Miyaji, T., Inoue, T.,
Hara, C., Uneyama, H., Edwards, R.H., Nicoll,
R.A., and Moriyama, Y. (2010). Metabolic con-
trol of vesicular glutamate transport and release.
Neuron 68, 99–112.
Kawamura, M., Jr., Ruskin, D.N., and Masino, S.A.
(2010). Metabolic autocrine regulation of neurons
involves cooperation among pannexin hemichan-
nels, adenosine receptors, and KATP channels. J
Neurosci 30, 3886–3895.
Kwan, P., and Brodie, M.J. (2000). Early identification
of refractory epilepsy. N Engl J Med 342, 314–319.
Lam, T.K., Gutierrez-Juarez, R., Pocai, A., and Rossetti,
L. (2005). Regulation of blood glucose by hypotha-
lamic pyruvate metabolism. Science 309, 943–947.
Larrabee, M.G. (1995). Lactate metabolism and its
effects on glucose metabolism in an excised neural
tissue. J Neurochem 64, 1734–1741.
Le, A., Cooper, C.R., Gouw, A.M., Dinavahi, R.,
Maitra, A., Deck, L.M., Royer, R.E., Vander Jagt,
D.L., Semenza, G.L., et al. (2010). Inhibition of lac-
tate dehydrogenase A induces oxidative stress and
inhibits tumor progression. Proc Natl Acad Sci
USA 107, 2037–2042.
Lutas, A., and Yellen, G. (2013). The ketogenic
diet: metabolic influences on brain excitability and
epilepsy. Trends Neurosci 36, 32–40.
Ma, W., Berg, J., and Yellen, G. (2007). Ketogenic
diet metabolites reduce firing in central neu-
rons by opening KATP channels. J Neurosci 27,
3618–3625.
Masino, S.A., Li, T., Theofilas, P., Sandau, U.S., Ruskin,
D.N., Fredholm, B.B., Geiger, J.D., Aronica, E., and
Boison, D. (2011). A ketogenic diet suppresses sei-
zures in mice through adenosine A1 receptors. J
Clin Invest 121, 2679–2683.
Meldrum, B.S., and Rogawski, M.A. (2007). Molecular
targets for antiepileptic drug development.
Neurotherapeutics 4, 18–61.
Neal, E.G., Chaffe, H., Schwartz, R.H., Lawson, M.S.,
Edwards, N., Fitzsimmons, G., Whitney, A., and
Cross, J.H. (2008). The ketogenic diet for the treat-
ment of childhood epilepsy:  a randomised con-
trolled trial. Lancet Neurol 7, 500–506.
Nehlig, A., Wittendorp-Rechenmann, E., and
Lam, C.D. (2004). Selective uptake of [14C]2-
deoxyglucose by neurons and astrocytes: high-
resolution microautoradiographic imaging by
288
288 section III: Ketogenic Diet in the Laboratory
cellular 14C-trajectography combined with immu-
nohistochemistry. J Cereb Blood Flow Metab 24,
1004–1014.
Parsons, M.P., and Hirasawa, M. (2010). ATP-sensitive
potassium channel-mediated lactate effect on
orexin neurons: implications for brain energetics
during arousal. J Neurosci 30, 8061–8070.
Rho, J.M., Anderson, G.D., Donevan, S.D., and
White, H.S. (2002). Acetoacetate, acetone, and
dibenzylamine (a contaminant in L-(+)-β-
hydroxybutyrate) exhibit direct anticonvulsant
actions in vivo. Epilepsia 43, 358–361.
Riban, V., Bouilleret, V., Pham-Lê, B.T., Fritschy, J.M.
Marescaux, C., and Depaulis, A. (2002). Evolution
of hippocampal epileptic activity during the devel-
opment of hippocampal sclerosis in a mouse
model of temporal lobe epilepsy. Neuroscience
112, 101–111.
Rouach, N., Koulakoff, A., Abudara, V., Willecke, K.,
and Giaume, C. (2008). Astroglial metabolic net-
works sustain hippocampal synaptic transmission.
Science 322, 1551–1555.
Sada, N., Lee, S., Katsu, T., Otsuki, T., and Inoue,
T. (2015). Targeting LDH enzymes with a sti-
ripentol analog to treat epilepsy. Science 347,
1362–1367.
Schurr, A., West, C.A., and Rigor, B.M. (1988).
Lactate-supported synaptic function in the rat
hippocampal slice preparation. Science 240,
1326–1328.
Silver, I.A., and Erecińska, M. (1994). Extracellular
glucose concentration in mammalian brain:  con-
tinuous monitoring of changes during increased
neuronal activity and upon limitation in oxygen
supply in normo-, hypo-, and hyperglycemic ani-
mals. J Neurosci 14, 5068–5076.
Smith, D., Pernet, A., Hallett, W.A., Bingham, E.,
Marsden, P.K., and Amiel, S.A. (2003). Lactate: a
preferred fuel for human brain metabolism in vivo.
J Cereb Blood Flow Metab 23, 658–664.
Tanner, G.R., Lutas, A., Martínez-François, J.R., and
Yellen, G. (2011). Single KATP channel opening in
response to action potential firing in mouse den-
tate granule neurons. J Neurosci 31, 8689–8696.
289
30
Effects of the Ketogenic Diet on 
the Blood-Brain Barrier
MANOJ BANJARA, PHD AND DAMIR JANIGRO, PHD
THE BLOOD-BRAIN
BARRIER
The blood-brain barrier (BBB) is a selectively per-
meable cellular boundary that protects the mam-
malian brain from systemic factors (Daneman and
Prat, 2015). Even though the microvessel endothe-
lial cells (ECs) have a primary role in the formation
of the BBB, several other cells are equally important
to maintain BBB integrity. Endothelial cells, astro-
cytes, pericytes, neurons, and other glial cells form
a “neurovascular unit” (Figure 30.1; Abbott et  al.,
2006; Grant and Janigro, 2010; Neuwelt et al., 2011).
In addition to the layer of capillary ECs, other lay-
ers exist between blood and brain. These include a
basement membrane (BM), consisting of type IV
collagen, fibronectin, and laminin that cover capil-
laries completely; pericytes embedded in the BM;
and astrocytes whose processes surround the BM
(Figure 30.1). Each of these layers has potential to
restrict the movement of solutes (Hawkins et  al.,
2006). Thus, the microenvironment of the BBB acts
in concert with tight junctions to maintain the selec-
tive permeability of the neurovascular unit barrier.
Simply stated, BBB integrity is an essential player
involved in the maintenance of brain homeostasis.
Disruption of the BBB occurs in various conditions
such as neoplasia, trauma, epilepsy, Alzheimer’s
disease, infections (e.g., meningitis), and sterile
inflammation (e.g., multiple sclerosis) (Huber et al.,
2001; Zlokovic, 2008). Thus, in these conditions it is
highly desirable to reverse the disruption of the BBB.
However, a highly restricted BBB is a huge challenge
for the delivery of drugs into the diseased brain
(Gloor et al., 2001). This dual nature of the BBB has
been recently reviewed by Saunders et al. (2014).
C E L L U L A R A S S O C I AT I O N S
AT   T H E   B B B
The BBB maintains the proper ionic composition
of the brain interstitial fluid, which is essential for
optimum neuronal function. It achieves a function
of “transport barrier” by facilitating the uptake
of necessary nutrients, while at the same time
preventing the uptake or actively effluxing other
molecules. Additionally, the BBB functions as a
“metabolic barrier,” possessing intracellular (e.g.,
cytochrome P450) and extracellular (e.g., pepti-
dases and nucleotidases) enzymes (Abbott et  al.,
2010; Ghosh et  al., 2013; Ghosh et  al., 2011a;
Ghosh et al., 2011b). Functions of individual cells
in brain microvessel are discussed in detail in this
section
Endothelial Cells
Endothelial cells line the interior of blood vessels.
The membranes of the capillary ECs are divided
into two distinct sides:  luminal (blood side) and
abluminal (brain side) (Daneman and Prat, 2015).
The ECs form blood vessels, and mural cells sit
on the abluminal side of the EC. These brain ECs
are less “leaky” than those of the peripheral ves-
sels. However, it has been shown that brain ECs
become “leaky” when they are allowed to vascu-
larize peripheral tissue, while peripheral ECs form
tight junctions resembling the BBB when allowed
to vascularize the brain parenchyma (Rubin et al.,
1991). Central nervous system (CNS) micro-
vascular ECs are 39% thinner than muscle ECs
(Coomber and Stewart, 1985).
The type and amount of membrane lipids and
proteins vary between luminal and abluminal sides
of the blood vessel (Tewes and Galla, 2001). Thus,
nutrients must transverse two sheaths of mem-
brane; their combined characteristics determine
which particles cross the barrier, and how fast. The
EC transporters are divided into two major catego-
ries: efflux transporters and nutrient transporters.
Efflux transporters are polarized to the luminal
surface and transport a broad variety of lipophilic
molecules (Cordon-Cardo et  al., 1989; Thiebaut
290

290 section III: Ketogenic Diet in the Laboratory

(a)
P

E
A E Lumen of blood A
vessels

S P
TJ
BM A: Astrocytes
N
A BM: Basement membrane
E: Endothelial cells
N: Neurons
P: Pericytes
S: Smooth muscle cells
(b) TJ: Tight junction

Blood
LAM
JAM
OP
GLUT1 MCT LAT1 Tfr JAC Z03 Claudin
Z02
Z01

ins
Caten CD
31 Z0-
Pgp BCRP MRP
E Tfr E
Z0-

N-cadherin
3
Ocln Z0-

OP
2

VE - JAC
1

Cad
BM herin
P
Connexin

CNS
Parenchyma

FIGURE 30.1 Schematic diagram of a brain capillary. A. Cell-to-cell interaction at the BBB. The endothelial cells
(E) are joined together by tight junctions (TJ) and surrounded by a basement membrane (BM). Pericytes (P) reside
within the BM and are distributed discontinuously along the length on cerebral capillaries. The cerebral capillaries are
surrounded by astrocytic foot processes or end-feet from astrocytes (A). Axonal projections from neurons (N) containing
vasoactive neurotransmitters and peptide onto arteriolar smooth muscle cells (S). BBB permeability may be regulated by
the molecules released from cells associated with endothelium (e.g., microglia, astrocytes). B. Schematic representation
of structural and transport components at the BBB. GLUT1, glucose transporter 1; MCT, monocarboxylate transporter;
LAT1, L-type amino acid transporter 1; Tfr, transferrin receptor; JAM, junctional adhesion molecules; LAM, leukocyte
adhesion molecules; ZO-1/2/3, zonula occludin-1/2/3; JACOP, junction-associated coiled-coil protein; Ocln, Occludin;
CD31, cluster of differentiation 31; Pgp, p-glycoprotein; BCRP, breast cancer resistant protein; MRP, multidrug
resistance-associated protein.

et al., 1987). Nutrient transporters are specific for
the transport of nutrients into, and the removal of
waste products from, the CNS (Mittapalli et  al.,
2010). The ECs of the BBB contain high numbers
of mitochondria, which are important to generate
a large amount of adenosine triphosphate (ATP)
to maintain ion gradients critical for the function
of transporters. In addition, CNS ECs express low
levels of leukocyte adhesion molecules, which ulti-
mately limits the number of immune cells that can
enter the CNS (Daneman et al., 2010). Moreover,
these ECs generate a barrier by altering the physical
properties of molecules, which then can modulate
their solubility, reactivity, and transport proper-
ties. These metabolic properties are further regu-
lated by the flow (Cucullo et al., 2011; Desai et al.,
2002). The combination of a physical barrier (e.g.,
tight junctions and low transcytosis), a molecular
barrier (e.g., low leukocyte adhesion molecules,
efflux transporters, and specific metabolism), and
291

Chapter 30: Effects of the Ketogenic Diet on the Blood-Brain Barrier 291

presence of specific nutrient transporters (e.g., Astrocytes

Glut1; McAllister et  al., 2001)  allows CNS ECs
to firmly regulate CNS homeostasis (Daneman
and Prat, 2015; Ghosh et  al., 2010). The specific
nutrient transporters are probably topographi-
cally organized for the development or function
of brain-specific region or particular subclass of
neurons. A major unanswered question regarding
the BBB and ECs is whether they possess unique
properties in different brain regions.
The following features distinguish BBB ECs
from peripheral vasculature (Arcangeli et al., 1999;
Dalvi et al., 2014; Gloor et al., 2001; Hawkins and
Davis, 2005; Rubin et al., 1991):
1. Presence of a large number of tight
junctions to limit the paracellular
movement of macromolecules (Figures
30.1 and 30.2).
2. Controlled fluid-phase endocytosis
rate to limit transcellular passage of
macromolecules.
3. Presence of highly specific transporters and
carrier molecules.
4. Absence of fenestrations.
5. Few cytoplasmic vesicles.
6. Increased mitochondrial content.
Astrocytes are important glial cells that help in
conditioning and developing the brain microves-
sel ECs. The interactions of ECs and astrocytes
are known to regulate the phenotype of the BBB
under (patho)physiological conditions (Prat
et  al., 2001). Astrocytes are known to alter the
properties of ECs in multiple ways (Grant et al.,
1998; Haseloff et al., 2005; Janzer and Raff, 1987;
Lee et  al., 2007; Stanness et  al., 1996; Stanness
et al., 1997):
1. Tightening the BBB, as evidenced by the
decreased paracellular permeability of
sucrose.
2. Elevating transendothelial electrical
resistance.
3. Increasing the activity of barrier-
related marker enzymes such as
alkaline phosphatase and γ-glutamyl
transpeptidase.
4. Enhancing the number, length, and
complexity of tight junctions.
5. Increasing the expression of glucose
transporters.
6. Secreting angiogenic factors like vascular
endothelial growth factor (VEGF),

1 2 3 4 6
5
Blood

Endothelial
Cells

Brain Pericyte

Neuron Astrocyte
Microglia

Lipid-soluble molecules 1: Transcellular pathway

Water-soluble molecules
Tight junctions
Transport proteins
Receptors
Positively charged molecules
2: Paracellular pathway
3: Transport protein
4: Receptor-mediated transcytosis
5: Adsorptive transcytosis
6: Efflux transport

FIGURE 30.2 Schematic diagram of transport pathways in cerebrovascular endothelial cells. The transport of diverse
molecules from the circulation across the BBB may include (1) transcellular lipophilic pathway, (2) paracellular aqueous
pathway across tight junctions, (3) transport proteins, (4) receptor-mediated transcytosis, (5) adsorptive transcytosis,
and (6) efflux transporters.
292
292 section III: Ketogenic Diet in the Laboratory
which is essential for the formation and
remodeling of embryonic blood vessels,
and decreasing vascular stability in adult
blood vessels.
Angiotensin-converting enzyme-I produced
by astrocytes converts angiotensin I  into angio-
tensin II, which in turn acts on type I angiotensin
receptors expressed by brain ECs. Angiotensin II
promotes the recruitment of junctional proteins
into the lipid raft, restricting BBB permeability
(Wosik et  al., 2007). Furthermore, retinoic acid
(RA) secreted by radial glial cells acts on RA-
receptor β expressed in the developing vascula-
ture, which increases transendothelial electrical
resistance by enhancing the expression of vascular
endothelial-cadherin, p-glycoprotein, and zonula
occludin-1 (Figure 30.1; Mizee et al., 2013). Other
factors such as transforming growth factor-β,
glial cell-derived neurotrophic factor, and src-
suppressed C-kinase substrate are postulated to
play a role in maintaining tightness of the BBB
(Haseloff et al., 2005).
Pericytes
Pericytes, the mural cells of blood microvessels,
are specialized cells of mesenchymal lineage that
surround the surface of the vascular tube. They
are on the abluminal side of the microvascular
endothelial tube and are embedded on the vascu-
lar basement membrane (Abbott et al., 2010). The
CNS microvasculatures have a higher pericyte-to-
endothelial ratio (1:3 to 1:1) than muscle (1:100)
(Shepro and Morel, 1993). Pericytes in the BBB
have several major functions (Armulik et  al.,
2011; Hall et al., 2014; Lai and Kuo, 2005):
1. Formation and maintenance of tight
junctions.
2. Autoregulation of cerebrovascular
blood flow.
3. Secretion of angiopoietin-1 and brain
angiogenesis.
4. Initiation of extrinsic blood coagulation
pathway after cerebrovascular injury.
5. Regulation of inflammation by the
secretion of cytokines (e.g., IL-1β and
IL-6), leukocyte transmigration, antigen
presentation, and T-cell activation.
6. Contraction of capillary diameter by
contractile proteins in pericytes.
The lack of a pericyte-specific marker, how-
ever, often leads to misidentification of pericytes
as other cells occupying the perivascular space.
All blood vessels, including those of the brain,
are ensheathed by extracellular matrix layers called
basement membranes (BMs). The inner vascular
BM and outer parenchymal BM are two BMs that
surround the vascular tube (Daneman and Prat,
2015; Sorokin, 2010). The BM is an extracellular
matrix secreted specially by ECs and pericytes in
vascular BM, and by astrocytic processes in paren-
chymal BM. Type IV collagens, nidogen, heparin
sulfate proteoglycans, and laminin are some mol-
ecules present in BMs. These membranes anchor
many signaling processes and provide an addi-
tional barrier that molecules and cells must cross
prior to accessing neural tissue. The degradation of
BMs by matrix metalloproteinases is observed in
different neurological diseases, allowing the infil-
tration of leukocytes (Daneman and Prat, 2015;
Sorokin, 2010).
M O L E C U L E S AT   T H E
BLOOD-BRAIN BARRIER
The structure and function of the BBB are
highly specialized by the presence of typical com-
ponents that are discussed in this section.
Tight Junctions
The tight junction (TJ) consists of membrane and
cytoplasmic proteins (Figure 30.1). Identification
of molecules expressed by CNS ECs led us to under-
stand the structural and transport components of
the BBB (Ohtsuki et  al., 2014). Claudins, occlu-
din, and junctional adhesion molecules (JAM) are
the integral membrane proteins. Cingulin, zonula
occludin proteins (ZO-1, -2, -3), AF6, 7H6 anti-
gen, and symplekin are the cytoplasmic accessory
proteins that form plaque and function as adap-
tor proteins. The tight junctional complexes are
dynamic entities that can “bend without break-
ing,” thereby sustaining structural integrity (Huber
et al., 2001).
Claudins are the largest family of transmem-
brane phosphoproteins; so far 24 members (clau-
dins 1–24) have been characterized (Ballabh
et al., 2004). Among these, claudins 1, 3, 5, and
12 have a role in tight-junction formation at
the BBB. Claudin-1 is an integral component of
tight junctions, whose absence is associated with
several pathological conditions, such as stroke,
inflammatory diseases, and tumors (Liebner
et al., 2000).
Occludin is another transmembrane phospho-
protein, and its subcellular localization parallels
that of claudin. In adult brain microvessel ECs,
occludin expression is higher than the peripheral
ECs that interact with claudins. Jointly, they form
293
Chapter 30: Effects of the Ketogenic Diet on the Blood-Brain Barrier
channels to regulate the paracellular flow of ions
and other hydrophilic molecules (Hirase et  al.,
1997; Huber et al., 2001; Morita et al., 1999).
The JAMs are members of the immunoglobu-
lin superfamily that form homotypic interactions
at the endothelial and epithelial cells TJ, which
are known to regulate leukocyte extravasation
and paracellular permeability (Aurrand-Lions
et al., 2001).
Several cytoplasmic proteins are also essen-
tial components of the TJs. Actin is the cyto-
skeleton protein that plays a major role in
maintenance of the TJ. Actin-degrading mol-
ecules (e.g., cytochalasin-D and phalloidin) can
disrupt the actin cytoskeleton, compromising the
TJ (Huber et al., 2001). Zonula occludin proteins
(ZO-1, -2, -3) are also important. These proteins
make direct contact with claudins, occludins,
and JAMs on one side and the actin cytoskele-
ton on the other (Ballabh et al., 2004). Cingulin
is another protein that links the TJ accessory
proteins with the cytoskeleton (Huber et  al.,
2001). Several intracellular processes involving
calcium-signaling, phosphorylation, G-proteins,
proteases, and inflammatory cytokine secretion
can modulate TJ proteins (Huber et  al., 2001;
Kniesel and Wolburg, 2000).
Leukocyte Adhesion Molecules
The CNS of healthy individuals has an extremely
low immune surveillance, lacking neutrophils
and lymphocytes in the parenchyma (Galea et al.,
2007). Entry of leukocytes into the brain paren-
chyma requires multiple steps, including rolling
adhesion, firm adhesion, and extravasation. This
requires several leukocyte adhesion molecules
(LAMs) that include selectins (E-, P- selectin)
for rolling adhesion and immunoglobulin family
members for firm adhesion (Huang et al., 2006). In
a normal brain, the expression of these adhesion
molecules is extremely lower in CNS ECs than
in peripheral ECs. However, LAMs are elevated
in neuroinflammatory conditions like stroke and
multiple sclerosis (Daneman et  al., 2010; Huang
et al., 2006).
TRANSPORTERS
The paracellular junctions, “having low perme-
ability,” control the transport of molecules and
ions between the blood and brain. The ECs of the
CNS are highly polarized and have distinct lumi-
nal and abluminal compartments. As mentioned
earlier, there are two major transporter types that
are expressed in CNS ECs: efflux transporters and
nutrient transporters (Figures 30.1–30.4).
293
Efflux transporters (e.g., multidrug resistance
protein 1, MDR1; breast cancer resistance pro-
tein, BCRP; multidrug resistance proteins, MRPs)
hydrolyze ATP to transport substrates against their
concentration gradient (Ha et  al., 2007). Many
efflux transporters are localized on the luminal
surface in order to transport substrates into the
blood compartment. These transporters facilitate
the movement of a diverse range of substrates, and
provide a barrier to many small lipophilic mol-
ecules, which would otherwise passively diffuse
through the membrane of ECs. MDR1, also called
p-glycoprotein (Pgp), is a well-studied transporter,
which is associated with drug-resistant epilepsy
(Dombrowski et al., 2001; Marchi et al., 2004) and
tumors (Abbott et  al., 2001). Endogenous sub-
strates of efflux transporters are still not well
studied.
Nutrient transporters facilitate the movement
of nutrients according to their concentration
gradients. A wide variety of such transporters are
expressed in CNS ECs to deliver specific nutri-
ents into the brain parenchyma. Most of these
nutrient transporters are in the solute carrier
class of facilitated transporters (slc2a1 transports
glucose, slc16a1 transports lactate and pyruvate,
SCL7A1 transports cationic amino acids, slc7a5
transports neural amino acids and L-Dopa)
(Zlokovic, 2008). Slc2a1, also called glucose
transporter 1 (Glut1), is a well-studied nutrient
transporter that provides glucose to the CNS
(Figure 30.4). It is more frequently expressed by
brain ECs than nonbrain ECs (Cornford et  al.,
1994). In humans, Glut1 deficiency leads to an
epileptic syndrome, which can be treated with a
high ketone diet (De Vivo et  al., 2002; De Vivo
et al., 1991). Most of the transporters in the BBB-
ECs provide nutrients to the brain, however, in
mild cognitive impairment and Alzheimer’s dis-
ease a receptor (receptor for advanced glycation
end products) is thought to assist in transporting
waste (amyloid β) from blood to brain (Daneman
and Prat, 2015).
OT H E R C O M P O N E N T S
The rate of transcytosis is lower in brain ECs; how-
ever, it is up-regulated upon BBB dysfunction dur-
ing brain injury and disease. In ECs, transcytosis
(Figure 30.2) is mediated via caveoline-based ves-
icle trafficking (Gu et al., 2012). Additionally, the
plasmalemma vesicle-associated protein number
is lower in the ECs of a healthy BBB than the ECs
of a BBB following traumatic brain injury (Shue
et al., 2008). The significance of these changes in
the context of disease etiology is unknown.
294
294 section III: Ketogenic Diet in the Laboratory
FUNCTIONS OF THE
BLOOD-BRAIN BARRIER
Regulation of Transit across
the Blood-Brain Barrier
The BBB provides a suitable environment for neu-
ral functions because it regulates ionic composi-
tion in the brain. The combination of ion-specific
channels and transporters at the BBB regulates the
composition and quantity of ions in the CNS. The
concentration of potassium in mammalian brain
cerebrospinal fluid (CSF) and interstitial fluid
(ISF) is maintained at around 2.5–2.9 mM, though
it is approximately 4.5 mM in plasma. Even though
plasma K+ concentration changes following exer-
cise or meals, it remains within 2.5–2.9 mM range
in the brain (see Janigro, 1999, 2012). In addition,
Ca+, Mg2+, and pH are actively regulated in the BBB
(Abbott et al., 2010; Jeong et al., 2006; Nischwitz et
al., 2008; Somjen, 2004).
The BBB acts as a shield to prevent the entry
of most macromolecules into the brain. The total
protein content in plasma is much higher than in
CSF, and the individual protein composition is
distinctly different (Abbott et  al., 2010). Proteins
that are markedly high in plasma (e.g., albumin,
prothrombin, and plasminogen) are detrimental
to nervous tissue, as they can trigger cell apoptosis
(Gingrich and Traynelis, 2000). Factor Xa and tis-
sue plasminogen activator are widely expressed in
the CNS, and can convert prothrombin and plas-
minogen to thrombin and plasmin, respectively.
If present, thrombin and plasmin can initiate cas-
cades resulting in seizures, glial activation, glial
cell division, and cell death (Abbott et  al., 2010).
Hence, the ingress of these macromolecules upon
BBB disruption can have serious pathological
consequences.
After taking meals, blood levels of a neuro-
excitatory amino acid, glutamate, are elevated.
If glutamate is freely transposed into the brain
ISF, it can be detrimental. For example, in an
ischemic stroke patient, uncontrolled glutamate
release into the brain can cause permanent neu-
roexcitatory damage of neural tissue. The periph-
eral and central nervous systems share many of
the same neurotransmitters; the BBB, however,
keeps them separate and minimizes the cross-talk
(Abbott et al., 2010; Abbott et al., 2006; Bernacki
et al., 2008).
The BBB is a protective barricade that shields
the brain from neurotoxic substances circulat-
ing in the blood (e.g., endogenous metabolites or
proteins, and xenobiotics ingested in the diet or
acquired from the environment). The ECs in the
CNS express a number of energy-dependent efflux
transporters (Figures 30.1 and 30.2) that actively
extrude neurotoxins out of the brain (Abbott et al.,
2010). These transporters also prevent neuronal
cell death where neurons are exposed to toxins
(Marchi et al., 2004).
Transport of Nutrients to the Brain
Most of the essential water-soluble nutrients and
metabolites required for neuronal survival and
differentiation are transported into the brain by
specific transport systems (Figures  30.2–30.4) at
the BBB. The endothelium begins to differentiate
into a barrier layer from the embryonic angiogen-
esis stage, and is maintained in adults by its asso-
ciation with other cell types, especially the endfeet
of astroglial cells. Astrocytic glial cells promote
the up-regulation of tight junction proteins and
the differential expression of luminal and ablu-
minal membrane-specific transporters (Abbott
et al., 2006; Wolburg et al., 2009). Other cell types,
namely pericytes, microglia, and nerve terminals
play supporting roles in barrier induction, mainte-
nance, and function (Abbott et al., 2006; Nakagawa
et al., 2009; Shimizu et al., 2008).
Together with oxygen, glucose is the obligatory
nutrient for the human brain. The BBB regulates
the transport of glucose in the CNS and prevents
equalization of plasma levels (~ 5 mM) to the
CSF and ISF levels (< 2.5 mM) (McAllister et al.,
2001). It is used for ATP production, which is
primarily used for ion transport and the mainte-
nance of the ion gradient (Magistretti and Pellerin,
1995). However, a small fraction is also used for
biosynthetic processes. The diffusion of this vital
nutrient across the BBB is facilitated by the glu-
cose transporter type 1 (Glut1). In the brain, Glut1
interacts with other Glut1 isoforms that mediate
glucose transport into neurons and astrocytes
(Klepper, 2008).
Brain glucose levels in a healthy and normally
active person are kept adequate by a complex regu-
latory mechanism to ensure an appropriate supply
to neurons, in spite of massive drops in plasma lev-
els. Glucose is controlled by intracellular enzymes
(e.g., hexokinase) and Glut1 proteins that are
expressed asymmetrically in the ECs (lower con-
centration in abluminal membrane than in lumi-
nal membrane) (McAllister et  al., 2001). Thus,
glucose levels in the brain parenchyma and CSF
ultimately depend on the “barrier” nature of the
brain endothelium.
Under normal conditions, asymmetrically
distributed Glut1 in the BBB transports blood
glucose into the brain, in which barrier integrity
295
Chapter 30: Effects of the Ketogenic Diet on the Blood-Brain Barrier
is maintained by tight junctions, preventing potas-
sium leakage in the brain (Figure 30.5). However,
under pathological conditions GLUT1 expression
is altered (Cornford et al., 1998a; Cornford et al.,
1998c). Cerebral blood flow also fails to meet the
brain metabolic demand when the BBB is damaged
(Bruehl et al., 1998), and in such cases, potassium
leaks across the open tight junctions in the brain
(Cornford et  al., 1998a; Cornford et  al., 1998b;
Cornford et al., 1998c). The elevated extracellular
[K+] further increases the metabolic demand due
to exaggerated neuronal firing and lost homeosta-
sis by voltage-dependent channels.
However, increased plasma ketone levels pro-
duced due to starvation, exercise, or ketogenic
diet (KD) ingestion provide an alternative energy
source to fulfill the energy demand in BBB-
damaged conditions (Figures 30.3 and 30.5). In
addition, ketone bodies (KBs) directly enhance
Glut1 activity (Janigro, 1999).
K E TO G E N I C   D I E T
The KD is a strict high-fat, low-carbohydrate,
and low-protein diet. As a result of this diet, the
body receives a minimal dietary source of glu-
cose, which is required for all metabolic needs
(Freeman et  al., 2006). Fatty acids become an
obligatory source of cellular energy production,
however, they do not readily cross the BBB due
to the presence of tight junctions (Mitchell et al.,
2009). Their transport into the brain by diffu-
sion or via specific protein-mediated transport is
still debated and controversial (Abumrad et  al.,
1998; Hamilton, 1998). The KD is characterized
by elevated serum levels of circulating KBs:  ace-
toacetate (AcAc), β-hydroxybutyrate (BHB), and
acetone (Ac), which are mainly produced by the
liver. The plasma concentration of KBs increases to
three-to-four times its basal levels (AcAc: 100 µM,
BHB: 200 µM) (Musa-Veloso et al., 2002). The KBs
are utilized by extrahepatic tissues as an energy
source. However, Ac is not an energy source and
is exhaled or excreted as a waste. In the absence
of glucose, KBs are the preferred source of brain
energy. They are transported into the brain by
simple (Ac) or facilitated diffusion (AcAc and
BHB) mediated by the monocarboxylate trans-
porter (MCT) (Figures 30.3–30.5) (Cremer, 1982;
Klepper, 2008). These MCTs are expressed in the
plasma membrane of choroid plexus and BBB cells
(endothelial and epithelial), glia, and neurons,
which ultimately oxidize ketones in the mito-
chondrial matrix, releasing acetyl-CoA that enters
into the tricarboxylic acid (TCA) or Krebs cycle
(Morris, 2005) (Figures 30.3 and 30.4).
295
Once KBs cross the BBB, brain cells take up
KBs via diffusion or a carrier-mediated process to
support cellular energy requirements. Even though
neurons and glial cells express MCTs, since KBs
must transverse the BBB first, their role appears
more significant at the BBB.
Regulation of Cerebral Ketone Uptake
In the absence of glucose, monocarboxylic acids
(e.g., ketones, lactate, and pyruvate) generate a
substantial amount of energy for the brain (Pierre
and Pellerin, 2005). Prolonged fasting elevates
normal circulating KB concentrations from ~5.8
mM to 9 mM (White and Venkatesh, 2011). The
plasma concentration of KBs is a major factor
regulating the rate of cerebral uptake. Since KBs
are hydrophilic compounds in the blood, specific
transporters are required to facilitate KB diffu-
sion across the BBB and maintain proper levels of
these metabolic products in the brain (Pierre and
Pellerin, 2005). Monocarboxylic acid transporters
(e.g., MCT1 and MCT2) involved in transport-
ing monocarboxylic acids (e.g., lactic acid and
pyruvate) are thought to facilitate the transport
of KBs. The first monocarboxylic acid transporter
identified in brain ECs was MCT1. Additionally,
small amounts of MCT1 were found in astrocytic
endfeet surrounding the capillaries. Interestingly,
MCT2 was found only in BBB ECs and neurons,
but not in astrocytes. On the other hand, MCT4
was present exclusively in the astrocytes and glial
cells. Glucose and monocarboxylate transporters
expressed in the major brain cells are summarized
in a schematic diagram (Figure 30.4). This distri-
bution of MCT is associated with the distinctive
functional characteristics. Brain MCT1 and MCT4
facilitate the release of monocarboxylates into the
extracellular space and allow cells to maintain a
high glycolytic rate (Pierre and Pellerin, 2005).
These monocarboxylates produced and released by
astrocytes in the extracellular space are an energy
substrate for active neurons; monocarboxylates
are taken up by MCT2 expressed in the dendrites
and axons of neurons (Pierre and Pellerin, 2005;
Takimoto and Hamada, 2014).
At a given arterial concentration, AcAc uptake
is double that of BHB; however, it is not yet under-
stood how the expression of MCT is regulated
(White and Venkatesh, 2011). Prolonged fasting
increases BBB uptake of KBs by increasing the
expression of MCT1 as high as eightfold in mice
(Leino et  al., 2001). Similarly, another report in
human subjects showed a 13-fold increase in
cerebral BHB uptake following several days of
fasting (Hasselbalch et  al., 1994). Interestingly, a
296

Ketogenic Diet

Glucose GLUT1 Glucose

Acetone Acetone
Fatty acid
FFA Oxidation Acetyl-
CoA
Acetoacetate MCT Acetoacetate
FFA
-hydroxybutyrate MCT
-hydroxybutyrate
Oxaloacetate

TCA cycle

ix
atr
lM
ria
Stomach
ito
ch
ond
Blood BBB
M

Cytosol
Liver
Brain
BBB: Blood-brain barrier
FFA: Free fatty acid
GLUT1: Glucose transporter 1
MCT: Monocarboxylate transporter
TCA: Tricarboxylic acid

FIGURE  30.3 Schematic diagram of ketogenic diet metabolism and brain uptake. The ketogenic diet is deficient
in carbohydrates, but provides large amounts of long chain free fatty acids (FFA). The acetyl-CoA produced by fatty
acid oxidation goes into the Krebs/tricarboxylic acid (TCA) cycle or is converted to ketone body (KB) acetoacetate
(AcAc), which spontaneously degrades to acetone (Ac). AcAc is further converted to β-hydroxybutyrate (BHB) in a
reversible reaction. These KBs represent alternative energy substrates for the brain. GLUT1, glucose transporter 1; MCT,
monocarboxylate transporter.

Lumen of blood
vessels A

GLUT1 MCT1 A: Astrocyte

GLUT3 MCT2 C: Choroid Plexus
GLUT4 MCT3 E: Endothelial Cell
GLUT5 MCT4 N: Neuron
GLUT6 M: Microglia
GLUT8

FIGURE  30.4 Schematic diagram of glucose and monocarboxylate transporters in the brain. Multiple types of
glucose transporters (GLUTs) and monocarboxylate transporters (MCTs) are responsible to transport glucose and
monocarboxylic acids (ketone bodies, lactate, and pyruvate), respectively.
297
Chapter 30: Effects of the Ketogenic Diet on the Blood-Brain Barrier
Normal BBB Damaged BBB
A A
Glucose
Glucose
KB
P
Glucose
E E
Blood Brain Blood
FIGURE 30.5 Schematic diagram of cellular and molecular events in normal and BBB damaged tissue. The ketogenic
diet potentially has a role to reestablish physiologic metabolic supply and repair damaged BBB by inducing the expression
levels of monocarboxylate transporter (MCT) and connexin 43 (Cx43), and enhancing glucose transporter 1 (GLUT1)
activity.
rapid increase in plasma ketone levels, following
an intravenous infusion of BHB, does not induce
cerebral uptake as significantly as prolonged fast-
ing; this suggests that MCT up-regulation is par-
tially dependent on the length of exposure to
increased plasma KBs (Pan et al., 2000; Pan et al.,
2001). Thus, a prolonged high-fat diet increases
blood ketone concentrations, leading to an ele-
vated expression of MCT and cerebral uptake of
KBs (Figure 30.5).
Anti-Inflammatory Effects
of Ketone Bodies
Neuroinflammation is defined as inflammation
of CNS that occurs upon invasion of pathogens,
trauma, and/or neurodegeneration. Cells such as
macrophages, astrocytes, and oligodendrocytes
and molecules such as cytokines and comple-
ment and pattern-recognition components are
the known contributors to neuroinflammation
(Banjara, 2014; de Vries et  al., 2012; Glass et  al.,
2010; Marchi et al., 2014). Proinflammatory mol-
ecules are either generated locally within the CNS
or transported from the peripheral system upon
BBB disruption. Moreover, inflammatory media-
tors have been shown to influence tight junctions
and to activate astrocytes and microglia (David
297
Damaged BBB + Ketogenic Diet
A
Glucose
KB
P KB P
E
Brain Blood Brain
A: Astrocyte
E: Endothelial Cell
P: Pericytes
GLUT1
MCT
Cx43
et al., 2009; Ivens et al., 2007; Tomkins et al., 2008),
ultimately compromising the BBB integrity. An
optimum amount of neuroinflammation is con-
sidered neuroprotective, and this generally occurs
for a short period of time. Chronic neuroinflam-
mation is often detrimental, inducing further cell
and tissue damage. However, the opposing mecha-
nisms of this ostensibly paradoxical exacerbation
or amelioration of parenchymal brain injury are
largely unknown (Cederberg and Siesjo, 2010;
Finnie, 2013).
The therapeutic efficacy of large spectrum anti-
neuroinflammatory drugs is limited by side effects
after even transient immune suppression. Recent
studies suggest that the KD could be useful, par-
ticularly in patients with recent worsening of epi-
lepsy and inflammation (Janigro, 1999; Marchi
et al., 2012; Nabbout et al., 2011). The KD has been
used as an effective treatment in patients with
inflammation-induced epileptic encephalopathies
(e.g., fever-induced refractory epileptic encepha-
lopathy in school-aged children) (Nabbout et  al.,
2011). Additionally, polysaturated fatty acids,
mostly eicosapentaenoic acid and docosahexae-
noic acid, are thought to have anti-inflammatory
actions, mediated typically by their hydroxylated
metabolites (Grimble, 1998; Porta et  al., 2009).
298

298 section III: Ketogenic Diet in the Laboratory

Furthermore, KD attenuated thermal nociception immune system, which is primarily responsible for
and decreased peripheral edema, indicating an IL-1β derived autoinflammatory disorders such as
anti-inflammatory effect (Ruskin et al., 2009). gout, obesity, atherosclerosis, and neurodegenera-
A systemic review published by Gibson et al., tive diseases (Levy et al., 2015). And BHB, but not
which analyzed 16 independent studies contain- structurally similar AcAc or butyrate, was shown
ing a total of 733 rodent studies, demonstrated to limit the cytokine production by inhibiting
a significant protective (both pathological and the activation of these NLRP3 inflammasome in
functional) effect of KD or exogenous adminis- human monocytes in vitro (Yasui et  al., 2011).
tration of KBs (p < .001) on outcomes following Since NLRP3 is a common player of sterile neu-
cerebral ischemia. Moreover, regardless of dietary roinflammation, the identification of BHB as an
intervention or administration-induced ketogenic NLRP3 inhibitor provides a rationale to further
state, ketones were found to have beneficial/anti- investigate the effectiveness of KD for the treat-
inflammatory effects on functional/behavioral ment of neuroinflammatory diseases.
tests, lesion volume, and brain water content
(Gibson et al., 2012).
Neuroprotective Effects of Ketone Bodies
In traumatic brain injury (TBI) models, con-
Ketones, an alternative energy source for normal
cussion or mild TBI models, and spinal cord
or injured brain, may also have neuroprotective
injury, KD intervention (pretreatment or admin-
effects (Prins, 2008). Ketones are considered neu-
istration to postinjury animals) revealed improved
roprotective due to the following:
structural and functional outcomes (Prins, 2008;
Prins and Matsumoto, 2014). Within minutes after
1. BHB is a more efficient energy source than
injury, the ionic equilibrium across the neuronal
glucose. It can also stimulate mitochondrial
membrane is disrupted, which requires cellular
biogenesis by up-regulating mitochondrial
energy to reestablish homeostasis. Thus, the cere-
enzymes and genes that stimulate energy
bral glucose uptake is increased in both rodent and
metabolism (Veech et al., 2001).
human within 30 min after fluid percussion and
2. KBs protect cells against glutamate-
within 8 days after TBI, respectively (Bergsneider
mediated apoptosis and necrosis by
et al., 1997). However, this transient “hyperglycol-
attenuating the formation of reactive
ysis” is followed by a prolonged period of reduc-
oxygen species (Ziegler et al., 2003). They
tion in glucose metabolism. Hence, utilizing brain
also attenuate apoptosis by reducing the
ketone metabolism after TBI as a therapeutic
activation of caspase-3 (Gasior et al., 2006).
approach is appealing as it can bypass the early glu-
3. KBs enhance the conversion of glutamate
cose metabolic derangements and also offers mul-
to gamma-aminobutyric acid (GABA)
tiple consequences that are beneficial to the brain
and boost the GABA-mediated inhibition
(Prins and Matsumoto, 2014). Ketone metabo-
(Gasior et al., 2006).
lism is particularly advantageous to TBI patients
4. Cerebral blood flow elevated by ketone
because it decreases the production of free radi-
metabolism is often considered as an effect
cals in mitochondria and cytosol (Sullivan et  al.,
of neuroprotective therapy (Hasselbalch
2004; Ziegler et  al., 2003). In addition, animals
et al., 1996).
on the KD produced less Bcl-2 associated protein,
ultimately decreasing cellular apoptosis and brain
Even though studies so far demonstrated
swelling. Furthermore, KD-induced decrease in
beneficial neuroprotective effects of KD, fur-
mitochondrial release of the cytochrome c into the
ther research is necessary to clarify issues such
cytosol inhibits the stimulation of apoptotic sig-
as dosing, timing, and the route and duration of
naling cascade (Hu et al., 2009a; Hu et al., 2009b).
administration.
More recently, it has been shown that KD can
modulate lipopolysaccharide-induced fever by
decreasing peripheral inflammation and CNS IL- Effects of Ketone Bodies on 
1β expression (Dupuis et al., 2015). These proper- the Blood-Brain Barrier
ties potentially contribute to the beneficial effect Data on the effects of the KD and/or KBs on the
of KD in neuroinflammatory conditions (e.g., BBB are based largely on animal studies. There
epilepsy, Alzheimer’s disease, Parkinson’s disease, is evidence of rapid changes in cerebral blood
multiple sclerosis) and/or brain trauma. flow and cellular transporters (decreased Glut1
The NOD-like receptor P3 (NLRP3) inflam- expression) that favor KBs’ uptake and metabo-
masome is a multiprotein complex of the innate lism in several neurological conditions such as
299
Chapter 30: Effects of the Ketogenic Diet on the Blood-Brain Barrier
TBI, hemorrhagic shock, ischemia, and hypoxia
(Gasior et al., 2006; Gibson et al., 2012; Prins and
Matsumoto, 2014; White and Venkatesh, 2011).
Prins et al. demonstrated an increase in expres-
sion of MCT (MCT1 and MCT2) levels and ketone
transport following TBI in rats (Prins and Giza,
2006). Additionally, ketone metabolizing enzyme,
β-hydroxybutyrate dehydrogenase, is elevated fol-
lowing cerebral injury that converts BHB to AcAc,
which is scarce in the adult brain (Tieu et al., 2003).
These evidences suggest brain’s improved ability to
utilize exogenous KBs. Thus, taking advantage of
improved transport and cellular metabolism of
ketones, brain’s reliance for energy may shift from
glucose to KBs in injured brain. It is beneficial for
the brain, particularly as hyperglycemia is deleteri-
ous to the injured brain (Salim et al., 2009).
Astrocyte endfeet lacking the gap junction pro-
tein connexin 43 (Cx43) are inefficient in trans-
membrane receptor anchoring. This weakens the
BBB, which further fails upon increased hydro-
static vascular pressure and shear stress (Ezan
et al., 2012). Interestingly, higher concentrations of
KBs, either alone or in combination, significantly
up-regulated the expression of the Cx43 at both
mRNA and protein levels (Ho et al., 2013) (Figure
30.5). Further, the up-regulation of Cx43 pro-
tein in ECs and glial cells by KBs accelerated cell
migration (Bates et al., 2007).
In a separate report, BHB (0.5 mM) has been
shown to promote human endothelial cell prolif-
eration (Cheng et al., 2006). These studies suggest
that the KD plays a role in repairing damaged BBB
and restoring normal brain function. In contrast,
Freeman et al. demonstrated AcAc-induced inhi-
bition of EC proliferation by elevation of oxida-
tive stress (Freeman et  al., 2006; Freeman et  al.,
1998). Hence, the response of cell types to dif-
ferent KBs might vary depending on cell-specific
characteristics.
In severely uncontrolled diabetes, KBs are pro-
duced in massive quantities, which is described
as ketoacidosis; this causes high concentrations
of protons and overwhelms the buffering system
of the body by activating a pH sensitive sodium-
proton pump (Bohn and Daneman, 2002). In
contrast, during high-fat/low-carbohydrate
intake, ketone bodies are produced in a regulated
manner causing a harmless physiological state
called dietary ketosis. There is minimal data on
the adverse effects of ketogenic diet on the BBB.
Depending on concentrations, acute administra-
tion of KBs alters the pH, sodium level, lipids,
and glucose in the blood. The acute intravenous
infusion of BHB also significantly increases the
299
pH and sodium concentrations in the brain
(Hiraide et  al., 1991). Thus, hypoglycemia and
dehydration are two predicted side effects of
continuous ketone consumption. The long-term
consequences of reduction in glucose cerebral
metabolism are thought to increase cerebral
blood flow, increase expression levels of MCT,
and enhance KB’s uptake and metabolism on the
BBB, however details are not yet known. In con-
trast, a report showed that the prolonged intake of
KD can significantly raise mean blood cholesterol
levels, leading to lipid deposition in blood vessels
(Freeman et al., 1998).
CONCLUDING REMARKS
It has long been recognized that increased con-
centration of KBs associated with the KD is effi-
cient fuel for the brain. Dramatic effects of such
high-fat dietary treatment were shown in patients
with multiple neurological disorders and Glut1
deficiency syndrome. Studies suggest that the
ketogenic diet potentially has the ability to rees-
tablish physiologic metabolic supply by activating
glucose transporters, enhancing the expression of
ketone transporters, and repairing damaged BBB
by inducing the expression levels of a gap junction
proteins. However, additional studies are essential
to validate these initial observations and explore
further.
REFERENCES
Abbott, N.J., Khan, E.U., Rollinson, C., Reichel, A.,
Janigro, D., Dombrowski, S., Dobbie, M., and
Begley, D.J. (2001). Drug resistance in epilepsy: the
role of the blood-brain barrier. In Mechanisms
of Drug Resistance in Epilepsy:  Lessons
from Oncology, Bock, G. and Goode, J.A. ed.
(Chichester, UK: John Wiley), pp. 38–47.
Abbott, N.J., Patabendige, A.A., Dolman, D.E., Yusof,
S.R., and Begley, D.J. (2010). Structure and func-
tion of the blood-brain barrier. Neurobiol Dis
37, 13–25.
Abbott, N.J., Ronnback, L., and Hansson, E. (2006).
Astrocyte-endothelial interactions at the blood-
brain barrier. Nat Rev Neurosci 7, 41–53.
Abumrad, N., Harmon, C., and Ibrahimi, A. (1998).
Membrane transport of long-chain fatty acids: evi-
dence for a facilitated process. J Lipid Res 39,
2309–2318.
Arcangeli, A., Rosati, B., Crociani, O., Cherubini, A.,
Fontana, L., Passani, B., Wanke, E., and Olivotto,
M. (1999). Modulation of HERG current and herg
gene expression during retinoic acid treatment of
human neuroblastoma cells: potentiating effects of
BDNF. J Neurobiol 40, 214–225.
Armulik, A., Genov, G., and Betsholtz, C. (2011).
Pericytes:  developmental, physiological, and
300
300 section III: Ketogenic Diet in the Laboratory
pathological perspectives, problems, and prom-
ises. Dev Cell 21, 193–215.
Aurrand-Lions, M., Johnson-Leger, C., Wong, C., Du
Pasquier, L., and Imhof, B.A. (2001). Heterogeneity
of endothelial junctions is reflected by differen-
tial expression and specific subcellular localiza-
tion of the three JAM family members. Blood 98,
3699–3707.
Ballabh, P., Braun, A., and Nedergaard, M. (2004). The
blood-brain barrier: an overview: structure, regu-
lation, and clinical implications. Neurobiol Dis
16, 1–13.
Banjara, M. (2014). Lipocalin-2:  a new regulator of
non-pathogen-associated neuroinflammation. Int
J Clin Exp Neurol 2, 8–15.
Bates, D.C., Sin, W.C., Aftab, Q., and Naus, C.C.
(2007). Connexin43 enhances glioma invasion by
a mechanism involving the carboxy terminus. Glia
55, 1554–1564.
Bergsneider, M., Hovda, D.A., Shalmon, E., Kelly, D.F.,
Vespa, P.M., Martin, N.A., Phelps, M.E., McArthur,
D.L., Caron, M.J., and Kraus, J.F. (1997). Cerebral
hyperglycolysis following severe traumatic brain
injury in humans:  a positron emission tomogra-
phy study. J Neurosurg 86, 241–251.
Bernacki, J., Dobrowolska, A., Nierwiska, K., and
Malecki, A. (2008). Physiology and pharmacologi-
cal role of the blood-brain barrier. Pharmacol Rep
60, 600–622.
Bohn, D., and Daneman, D. (2002). Diabetic ketoaci-
dosis and cerebral edema. Curr Opin Pediatr 14,
287–291.
Bruehl, C., Hagemann, G., and Witte, O.W. (1998).
Uncoupling of blood flow and metabolism in epi-
lepsy. Epilepsia 39, 1235–1242.
Cederberg, D., and Siesjo, P. (2010). What has inflam-
mation to do with traumatic brain injury? Childs
Nerv Syst 26, 221–226.
Cheng, S., Chen, G.Q., Leski, M., Zou, B., Wang, Y., and
Wu, Q. (2006). The effect of D, L-hydroxybutyric
acid on cell death and proliferation in L929 cells.
Biomaterials 27, 3758–3765.
Coomber, B.L., and Stewart, P.A. (1985). Morphometric
analysis of CNS microvascular endothelium.
Microvasc Res 30, 99–115.
Cordon-Cardo, C., O’Brien, J.P., Casals, D., Rittman-
Grauer, L., Biedler, J.L., Melamed, M.R., and
Bertino, J.R. (1989). Multidrug-resistance gene (P-
glycoprotein) is expressed by endothelial cells at
blood-brain barrier sites. PNAS 86, 695–698.
Cornford, E.M., Hyman, S., and Swartz, B.E.
(1994). The human brain GLUT1 glucose trans-
porter:  ultrastructural localization to the blood-
brain barrier endothelia. J Cereb Blood Flow
Metab 14, 106–112.
Cornford, E.M., Hyman, S., Cornford, M.E., and
Damian, R.T. (1998a). Glut1 glucose transporter
in the primate choroid plexus endothelium. J
Neuropathol Exp Neurol 57, 404–414.
Cornford, E.M., Hyman, S., Cornford, M.E., Damian,
R.T., and Raleigh, M.J. (1998b). A single glucose
transporter configuration in normal primate brain
endothelium:  comparison with resected human
brain. J Neuropathol Exp Neurol 57, 699–713.
Cornford, E.M., Hyman, S., Cornford, M.E., Landaw,
E.M., and Delgado-Escueta, A.V. (1998c). Interictal
seizure resections show two configurations of endo-
thelial Glut1 glucose transporter in the human
blood-brain barrier. J Cereb Blood Flow Metab
18, 26–42.
Cremer, J.E. (1982). Substrate utilization and brain
development. J Cereb Blood Flow Metab 2, 394–407.
Cucullo, L., Hossain, M., Puvenna, V., Marchi, N.,
and Janigro, D. (2011). The role of shear stress in
Blood-Brain Barrier endothelial physiology. BMC
Neurosci 12, 40.
Dalvi, S., On, N., Nguyen, H., Pogorzelec, M., Miller,
D.W., and Hatch, G.M. (2014). e blood brain bar-
rier-regulation of fatty acid and drug transport. In
Neurochemistry, T. Heinbockel, ed. (InTech). DOI:
10.5772/57604. Available from: http://www.intecho-
pen.com/books/neurochemistry/the-blood-brain-
barrier-regulation-of-fatty-acid-and-drug-transport
Daneman, R., and Prat, A. (2015). The blood-brain
barrier. Cold Spring Harb Perspect Biol 7, a020412.
Daneman, R., Zhou, L., Agalliu, D., Cahoy, J.D.,
Kaushal, A., and Barres, B.A. (2010). The mouse
blood-brain barrier transcriptome: a new resource
for understanding the development and func-
tion of brain endothelial cells. PLoS One 5,
e13741–e13741.
David, Y., Cacheaux, L.P., Ivens, S., Lapilover, E.,
Heinemann, U., Kaufer, D., and Friedman, A.
(2009). Astrocytic dysfunction in epileptogen-
esis:  Consequence of altered potassium and glu-
tamate homeostasis? J Neurosci 29, 10588–10599.
De Vivo, D.C., Trifiletti, R.R., Jacobson, R.I., Ronen,
G.M., Behmand, R.A., and Harik, S.I. (1991).
Defective glucose transport across the blood-brain
barrier as a cause of persistent hypoglycorrhachia,
seizures, and developmental delay. N Engl J Med
325, 703–709.
De Vivo, D.C., Leary, L., and Wang, D.O.N.G. (2002).
Glucose transporter 1 deficiency syndrome and other
glycolytic defects. J Child Neurol 17, 3S15–13S23.
de Vries, H.E., Kooij, G., Frenkel, D., Georgopoulos, S.,
Monsonego, A., and Janigro, D. (2012). Inflammatory
events at blood-brain barrier in neuroinflammatory
and neurodegenerative disorders:  Implications for
clinical disease. Epilepsia 53, 45–52.
Desai, S.Y., Marroni, M., Cucullo, L., Krizanac-Bengez,
L., Mayberg, M.R., Hossain, M.T., Grant, G.G., and
Janigro, D. (2002). Mechanisms of endothelial sur-
vival under shear stress. Endothelium 9, 89–102.
Dombrowski, S.M., Desai, S.Y., Marroni, M., Cucullo,
L., Goodrich, K., Bingaman, W., Mayberg, M.R.,
Bengez, L., and Janigro, D. (2001). Overexpression
of multiple drug resistance genes in endothelial
301
Chapter 30: Effects of the Ketogenic Diet on the Blood-Brain Barrier
cells from patients with refractory epilepsy.
Epilepsia 42, 1501–1506.
Dupuis, N., Curatolo, N., Benoist, J.F., and Auvin, S.
(2015). Ketogenic diet exhibits anti-inflammatory
properties. Epilepsia 56, e95–e98.
Ezan, P., Andr, P., Cisternino, S., Saubama, B., Boulay,
A.C., Doutremer, S., Thomas, M.A., Quenech’du,
N., Giaume, C., and Cohen-Salmon, M. (2012).
Deletion of astroglial connexins weakens the
blood-brain barrier. J Cereb Blood Flow Metab 32,
1457–1467.
Finnie, J.W. (2013). Neuroinflammation:  benefi-
cial and detrimental effects after traumatic brain
injury. Inflammopharmacology 21, 309–320.
Freeman, J.M., Vining, E.P., Pillas, D.J., Pyzik, P.L., and
Casey, J.C. (1998). The efficacy of the ketogenic
diet-1998: a prospective evaluation of intervention
in 150 children. Pediatrics 102, 1358–1363.
Freeman, J., Veggiotti, P., Lanzi, G., Tagliabue, A.,
and Perucca, E. (2006). The ketogenic diet:  from
molecular mechanisms to clinical effects. Epilepsy
Res 68, 145–180.
Galea, I., Bechmann, I., and Perry, V.H. (2007). What
is immune privilege (not)? Trends Immunol
28, 12–18.
Gasior, M., Rogawski, M.A., and Hartman, A.L. (2006).
Neuroprotective and disease-modifying effects of
the ketogenic diet. Behav Pharmacol 17, 431.
Ghosh, C., Gonzalez-Martinez, J., Hossain, M.,
Cucullo, L., Fazio, V., Janigro, D., and Marchi, N.
(2010). Pattern of P450 expression at the human
blood-brain barrier:  roles of epileptic condition
and laminar flow. Epilepsia 51, 1408–1417.
Ghosh, C., Marchi, N., Desai, N.K., Puvenna, V.,
Hossain, M., Gonzalez-Martinez, J., Alexopoulos,
A.V., and Janigro, D. (2011a). Cellular localiza-
tion and functional significance of CYP3A4 in the
human epileptic brain. Epilepsia 52, 562–571.
Ghosh, C., Puvenna, V., Gonzalez-Martinez, J.,
Janigro, D., and Marchi, N. (2011b). Blood-brain
barrier P450 enzymes and multidrug transporters
in drug resistance: a synergistic role in neurologi-
cal diseases. Curr Drug Metab 12, 742–749.
Ghosh, C., Hossain, M., Puvenna, V., Martinez-
Gonzalez, J., Alexopolous, A., Janigro, D., and
Marchi, N. (2013). Expression and functional
relevance of UGT1A4 in a cohort of human drug-
resistant epileptic brains. Epilepsia 54, 1562–1570.
Gibson, C.L., Murphy, A.N., and Murphy, S.P. (2012).
Stroke outcome in the ketogenic state-systematic
review of the animal data. J Neurochem 123, 52–57.
Gingrich, M.B., and Traynelis, S.F. (2000). Serine pro-
teases and brain damage is there a link? Trends
Neurosci 23, 399–407.
Glass, C.K., Saijo, K., Winner, B., Marchetto, M.C., and
Gage, F.H. (2010). Mechanisms underlying inflam-
mation in neurodegeneration. Cell 140, 918–934.
Gloor, S.M., Wachtel, M., Bolliger, M.F., Ishihara, H.,
Landmann, R., and Frei, K. (2001). Molecular and
301
cellular permeability control at the blood-brain
barrier. Brain Res Rev 36, 258–264.
Grant, G.A., Abbott, N.J., and Janigro, D. (1998).
Understanding the physiology of the blood-brain
barrier:  in vitro models. News Physiol Sci 13,
287–293.
Grant, G.A., and Janigro, D. (2004). The blood-brain
barrier. In Youmans Neurological Surgery, H.R.
Winn, ed. (Philadelphia, PA:  Saunders), pp.
153–174.
Grimble, R.F. (1998). Nutritional modulation of cyto-
kine biology. Nutrition 14, 634–640.
Gu, Y., Zheng, G., Xu, M., Li, Y., Chen, X., Zhu, W.,
Tong, Y., Chung, S.K., Liu, K.J., and Shen, J. (2012).
Caveolin-1 regulates nitric oxide-mediated matrix
metalloproteinases activity and blood-brain bar-
rier permeability in focal cerebral ischemia and
reperfusion injury. J Neurochem 120, 147–156.
Ha, S.N., Hochman, J., and Sheridan, R.P. (2007). Mini
review on molecular modeling of P-glycoprotein
(Pgp). Curr Top Med Chem 7, 1525–1529.
Hall, C.N., Reynell, C., Gesslein, B., Hamilton, N.B.,
Mishra, A., Sutherland, B.A., Farrell, F.M., Buchan,
A.M., Lauritzen, M., and Attwell, D. (2014).
Capillary pericytes regulate cerebral blood flow in
health and disease. Nature 508, 55–60.
Hamilton, J.A. (1998). Fatty acid transport: difficult or
easy? J Lipid Res 39, 467–481.
Haseloff, R.F., Blasig, I.E., Bauer, H.C., and Bauer, H.
(2005). In search of the astrocytic factor (s) modu-
lating blood-brain barrier functions in brain capil-
lary endothelial cells in vitro. Cell Mol Neurobiol
25, 25–39.
Hasselbalch, S.G., Knudsen, G.M., Jakobsen, J.,
Hageman, L.P., Holm, S., and Paulson, O.B.
(1994). Brain metabolism during short-term star-
vation in humans. J Cereb Blood Flow Metab 14,
125–131.
Hasselbalch, S.G., Madsen, P.L., Hageman, L.P., Olsen,
K.S., Justesen, N., Holm, S., and Paulson, O.B.
(1996). Changes in cerebral blood flow and carbo-
hydrate metabolism during acute hyperketonemia.
Am J Physiol Endocrinol Metab 270, E746–E751.
Hawkins, B.T., and Davis, T.P. (2005). The blood-brain
barrier/neurovascular unit in health and disease.
Pharmacol Rev 57, 173–185.
Hawkins, R.A., O’Kane, R.L., Simpson, I.A., and Via,
J.R. (2006). Structure of the blood-brain barrier
and its role in the transport of amino acids. J Nutr
136, 218S–226S.
Hiraide, A., Katayama, M., Sugimoto, H., Yoshioka,
T., and Sugimoto, T. (1991). Effect of 3-
hydroxybutyrate on posttraumatic metabolism in
man. Surgery 109, 176–181.
Hirase, T., Staddon, J.M., Saitou, M., Ando-Akatsuka,
Y., Itoh, M., Furuse, M., Fujimoto, K., Tsukita, S.,
and Rubin, L. (1997). Occludin as a possible deter-
minant of tight junction permeability in endothe-
lial cells. J Cell Sci 110, 1603–1613.
302
302 section III: Ketogenic Diet in the Laboratory
Ho, C.F., Chan, K.W., Yeh, H.I., Kuo, J., Liu, H.J., and
Wang, C.Y. (2013). Ketone bodies upregulate
endothelial connexin 43 (Cx43) gap junctions. Vet
J 198, 696–701.
Hu, Z.G., Wang, H.D., Jin, W., and Yin, H.X. (2009a).
Ketogenic diet reduces cytochrome c release and
cellular apoptosis following traumatic brain injury
in juvenile rats. Ann Clin Lab Sci 39, 76–83.
Hu, Z.G., Wang, H.D., Qiao, L., Yan, W., Tan, Q.F.,
and Yin, H.X. (2009b). The protective effect of the
ketogenic diet on traumatic brain injury-induced
cell death in juvenile rats. Brain Inj 23, 459–465.
Huang, J., Upadhyay, U.M., and Tamargo, R.J. (2006).
Inflammation in stroke and focal cerebral isch-
emia. Surg Neurol 66, 232–245.
Huber, J.D., Egleton, R.D., and Davis, T.P. (2001).
Molecular physiology and pathophysiology of
tight junctions in the blood-brain barrier. Trends
Neurosci 24, 719–725.
Ivens, S., Kaufer, D., Flores, L.P., Bechmann, I.,
Zumsteg, D., Tomkins, O., Seiffert, E., Heinemann,
U., and Friedman, A. (2007). TGF-beta receptor-
mediated albumin uptake into astrocytes is
involved in neocortical epileptogenesis. Brain 130,
535–547.
Janigro, D. (1999). Blood-brain barrier, ion homeo-
stasis and epilepsy: possible implications towards
the understanding of ketogenic diet mechanisms.
Epilepsy Res 37, 223–232.
Janigro, D. (2012). Are you in or out? Leukocyte, ion,
and neurotransmitter permeability across the
epileptic blood-brain barrier. Epilepsia 53 (Suppl
1), 26–34.
Janzer, R.C., and Raff, M.C. (1987). Astrocytes induce
blood-brain barrier properties in endothelial cells.
Nature 325, 253–257.
Jeong, S.M., Hahm, K.D., Shin, J.W., Leem, J.G., Lee,
C., and Han, S.M. (2006). Changes in magnesium
concentration in the serum and cerebrospinal fluid
of neuropathic rats. Acta Anaesthesiol Scand 50,
211–216.
Klepper, J. (2008). Glucose transporter deficiency
syndrome (GLUT1DS) and the ketogenic diet.
Epilepsia 49, 46–49.
Kniesel, U., and Wolburg, H. (2000). Tight junctions
of the blood-brain barrier. Cell Mol Neurobiol
20, 57–76.
Lai, C.H., and Kuo, K.H. (2005). The critical compo-
nent to establish in vitro BBB model:  pericyte.
Brain Res Rev 50, 258–265.
Lee, S., Chen, T.T., Barber, C.L., Jordan, M.C.,
Murdock, J., Desai, S., Ferrara, N., Nagy, A., Roos,
K.P., and Iruela-Arispe, M.L. (2007). Autocrine
VEGF signaling is required for vascular homeo-
stasis. Cell 130, 691–703.
Leino, R.L., Gerhart, D.Z., Duelli, R., Enerson, B.E.,
and Drewes, L.R. (2001). Diet-induced ketosis
increases monocarboxylate transporter (MCT1)
levels in rat brain. Neurochem Int 38, 519–527.
Levy, M., Thaiss, C.A., and Elinav, E. (2015). Taming
the inflammasome. Nat Med 21, 213–215.
Liebner, S., Fischmann, A., Rascher, G., Duffner,
F., Grote, E.H., Kalbacher, H., and Wolburg, H.
(2000). Claudin-1 and claudin-5 expression and
tight junction morphology are altered in blood
vessels of human glioblastoma multiforme. Acta
Neuropathol 100, 323–331.
Magistretti, P.J., and Pellerin, L. (1995). Cellular basis
of brain energy metabolism and their relevance to
brain imaging-evidence for a prominent role for
astrocytes. Cereb Cortex 5, 301–306.
Marchi, N., Granata, T., Alexopoulos, A., and Janigro,
D. (2012). The blood-brain barrier hypothesis in
drug resistant epilepsy. Brain 135, e211.
Marchi, N., Granata, T., and Janigro, D. (2014).
Inflammatory pathways of seizure disorders.
Trends Neurosci 37, 55–65.
Marchi, N., Hallene, K.L., Kight, K.M., Cucullo, L.,
Moddel, G., Bingaman, W., Dini, G., Vezzani, A.,
and Janigro, D. (2004). Significance of MDR1 and
multiple drug resistance in refractory human epi-
leptic brain. BMC Med 2, 37.
McAllister, M.S., Krizanac-Bengez, L., Macchia,
F., Naftalin, R.J., Pedley, K.C., Mayberg, M.R.,
Marroni, M., Leaman, S., Stanness, K.A., and
Janigro, D. (2001). Mechanisms of glucose trans-
port at the blood-brain barrier: an in vitro study.
Brain Res 904, 20–30.
Mitchell, R.W., Edmundson, C.L., Miller, D.W., and
Hatch, G.M. (2009). On the mechanism of oleate
transport across human brain microvessel endo-
thelial cells. J Neurochem 110, 1049–1057.
Mittapalli, R.K., Manda, V.K., Adkins, C.E.,
Geldenhuys, W.J., and Lockman, P.R. (2010).
Exploiting nutrient transporters at the blood-brain
barrier to improve brain distribution of small mol-
ecules. Ther Deliv 1, 775–784.
Mizee, M.R., Wooldrik, D., Lakeman, K.A., van het
Hof, B., Drexhage, J.A., Geerts, D., Bugiani, M.,
Aronica, E., Mebius, R.E., and Prat, A. (2013).
Retinoic acid induces blood-brain barrier devel-
opment. J Neurosci 33, 1660–1671.
Morita, K., Furuse, M., Fujimoto, K., and Tsukita, S.
(1999). Claudine multigene family encoding four-
transmembrane domain protein components of
tight junction strands. PNAS 96, 511–516.
Morris, A.A.M. (2005). Cerebral ketone body metabo-
lism. J Inherit Metab Dis 28, 109–121.
Musa-Veloso, K., Likhodii, S.S., and Cunnane, S.C.
(2002). Breath acetone is a reliable indicator of
ketosis in adults consuming ketogenic meals. Am J
Clin Nutr 76, 65–70.
Nabbout, R., Vezzani, A., Dulac, O., and Chiron, C.
(2011). Acute encephalopathy with inflammation-
mediated status epilepticus. Lancet Neurol 10,
99–108.
Nakagawa, S., Deli, M.A., Kawaguchi, H., Shimizudani,
T., Shimono, T., Kittel, A., Tanaka, K., and Niwa,
303
Chapter 30: Effects of the Ketogenic Diet on the Blood-Brain Barrier
M. (2009). A new blood-brain barrier model using
primary rat brain endothelial cells, pericytes and
astrocytes. Neurochem Int 54, 253–263.
Neuwelt, E.A., Bauer, B., Fahlke, C., Fricker, G., Iadecola,
C., Janigro, D., Leybaert, L., Molnar, Z., O’Donnell,
M.E., Povlishock, J.T., et al. (2011). Engaging neu-
roscience to advance translational research in brain
barrier biology. Nat Rev Neurosci 12, 169–182.
Nischwitz, V., Berthele, A., and Michalke, B. (2008).
Speciation analysis of selected metals and deter-
mination of their total contents in paired serum
and cerebrospinal fluid samples:  an approach to
investigate the permeability of the human blood-
cerebrospinal fluid-barrier. Anal Chim Acta 627,
258–269.
Ohtsuki, S., Hirayama, M., Ito, S., Uchida, Y., Tachikawa,
M., and Terasaki, T. (2014). Quantitative targeted
proteomics for understanding the blood-brain
barrier: towards pharmacoproteomics. Expert Rev
Proteomics 11, 303–313.
Pan, J.W., Rothman, D.L., Behar, K.L., Stein, D.T., and
Hetherington, H.P. (2000). Human brain [bgr]-
hydroxybutyrate and lactate increase in fasting-
induced ketosis. J Cereb Blood Flow Metab 20,
1502–1507.
Pan, J.W., Telang, F.W., Lee, J.H., de Graaf, R.A.,
Rothman, D.L., Stein, D.T., and Hetherington, H.P.
(2001). Measurement of &beta;-hydroxybutyrate
in acute hyperketonemia in human brain. J
Neurochem 79, 539–544.
Pierre, K., and Pellerin, L. (2005). Monocarboxylate
transporters in the central nervous system:  dis-
tribution, regulation and function. J Neurochem
94, 1–14.
Porta, N., Valle, L., Boutry, E., Fontaine, M., Dessein,
A.F., Joriot, S., Cuisset, J.M., Cuvellier, J.C., and
Auvin, S. (2009). Comparison of seizure reduc-
tion and serum fatty acid levels after receiving the
ketogenic and modified Atkins diet. Seizure 18,
359–364.
Prat, A., Biernacki, K., Wosik, K., and Antel, J.P. (2001).
Glial cell influence on the human blood-brain bar-
rier. Glia 36, 145–155.
Prins, M.L. (2008). Cerebral metabolic adaptation
and ketone metabolism after brain injury. J Cereb
Blood Flow Metab. 28, 1–16.
Prins, M.L., and Giza, C.C. (2006). Induction of mono-
carboxylate transporter 2 expression and ketone
transport following traumatic brain injury in juve-
nile and adult rats. Dev Neurosci 28, 447–456.
Prins, M.L., and Matsumoto, J.H. (2014). The col-
lective therapeutic potential of cerebral ketone
metabolism in traumatic brain injury. J Lipid Res
55, 2450–2457.
Rubin, L.L., Hall, D.E., Porter, S., Barbu, K., Cannon,
C., Horner, H.C., Janatpour, M., Liaw, C.W.,
Manning, K., Morales, J., et al. (1991). A cell cul-
ture model of the blood-brain barrier. J Cell Biol
115, 1725–1735.
303
Ruskin, D.N., Kawamura, M., and Masino, S.A.
(2009). Reduced pain and inflammation in juve-
nile and adult rats fed a ketogenic diet. PLoS One
4, e8349.
Salim, A., Hadjizacharia, P., Dubose, J., Brown, C.,
Inaba, K., Chan, L.S., and Margulies, D. (2009).
Persistent hyperglycemia in severe traumatic brain
injury: an independent predictor of outcome. Am
Surg 75, 25–29.
Saunders, N.R., Dreifuss, J.J., Dziegielewska, K.M.,
Johansson, P.A., Habgood, M.D., Mllgrd, K., and
Bauer, H.C. (2014). The rights and wrongs of
blood-brain barrier permeability studies:  a walk
through 100 years of history. Front Neurosci 8.
Shepro, D., and Morel, N.M. (1993). Pericyte physiol-
ogy. FASEB Journal 7, 1031–1038.
Shimizu, F., Sano, Y., Maeda, T., Abe, M., Nakayama,
H., Takahashi, R., Ueda, M., Ohtsuki, S., Terasaki,
T., and Obinata, M. (2008). Peripheral Nerve peri-
cytes originating from the blood-nerve barrier
expresses tight junctional molecules and trans-
porters as barrier-forming cells. J Cell Physiol 217,
388–399.
Shue, E.H., Carson-Walter, E.B., Liu, Y., Winans,
B.N., Ali, Z.S., Chen, J., and Walter, K.A. (2008).
Plasmalemmal vesicle associated protein-1 (PV-1)
is a marker of blood-brain barrier disruption in
rodent models. BMC Neurosci 9, 29.
Somjen, G.G. (2004). Ions in the brain: normal func-
tion, seizures, and stroke (New York, USA: Oxford
University Press).
Sorokin, L. (2010). The impact of the extracellular
matrix on inflammation. Nat Rev Immunol 10,
712–723.
Stanness, K.A., Guatteo, E., and Janigro, D. (1996).
A dynamic model of the blood-brain barrier “in
vitro.” Neurotoxicology 17, 481–496.
Stanness, K.A., Westrum, L.E., Fornaciari, E.,
Mascagni, P., Nelson, J.A., Stenglein, S.G., Myers,
T., and Janigro, D. (1997). Morphological and
functional characterization of an in vitro blood-
brain barrier model. Brain Res 771, 329–342.
Sullivan, P.G., Rippy, N.A., Dorenbos, K., Concepcion,
R.C., Agarwal, A.K., and Rho, J.M. (2004). The
ketogenic diet increases mitochondrial uncou-
pling protein levels and activity. Ann Neurol 55,
576–580.
Takimoto, M., and Hamada, T. (2014). Acute exer-
cise increases brain region-specific expression of
MCT1, MCT2, MCT4, GLUT1, and COX IV pro-
teins. J Appl Physiol 116, 1238–1250.
Tewes, B.J., and Galla, H.J. (2001). Lipid polarity in
brain capillary endothelial cells. Endothelium 8,
207–220.
Thiebaut, F., Tsuruo, T., Hamada, H., Gottesman,
M.M., Pastan, I., and Willingham, M.C. (1987).
Cellular localization of the multidrug resistance
gene product in normal human tissues. PNAS 84,
7735–7738.
304
304 section III: Ketogenic Diet in the Laboratory
Tieu, K., Perier, C., Caspersen, C., Teismann, P., Wu,
D.C., Yan, S.D., Naini, A., Vila, M., Jackson-
Lewis, V., and Ramasamy, R. (2003). D-beta-
hydroxybutyrate rescues mitochondrial respiration
and mitigates features of Parkinson disease. J Clin
Invest 112, 892.
Tomkins, O., Shelef, I., Kaizerman, I., Eliushin,
A., Afawi, Z., Misk, A., Gidon, M., Cohen, A.,
Zumsteg, D., and Friedman, A. (2008). Blood-
brain barrier disruption in post-traumatic epi-
lepsy. J Neurol Neurosurg Psychiatry 79, 774–777.
Veech, R.L., Chance, B., Kashiwaya, Y., Lardy, H.A.,
and Cahill, G.F. (2001). Ketone bodies, potential
therapeutic uses. IUBMB Life 51, 241–247.
White, H., and Venkatesh, B. (2011). Clinical
review: ketones and brain injury. Crit Care 15, 219.
Wolburg, H., Noell, S., Mack, A., Wolburg-Buchholz,
K., and Fallier-Becker, P. (2009). Brain endothelial
cells and the glio-vascular complex. Cell Tissue
Res 335, 75–96.
Wosik, K., Cayrol, R., Dodelet-Devillers, A.,
Berthelet, F., Berna 32rd, M., Moumdjian, R.,
Bouthillier, A., Reudelhuber, T.L., and Prat, A.
(2007). Angiotensin II controls occludin function
and is required for blood-brain barrier mainte-
nance: relevance to multiple sclerosis. J Neurosci
27, 9032–9042.
Yasui, T., Hirose, J., Tsutsumi, S., Nakamura, K.,
Aburatani, H., and Tanaka, S. (2011). Epigenetic
regulation of osteoclast differentiation:  possible
involvement of Jmjd3 in the histone demethylation
of Nfatc1. J Bone Miner Res 26, 2665–2671.
Ziegler, D.R., Ribeiro, L.C., Hagenn, M., Arajo,
E., Torres, I.L., Gottfried, C., Netto, C.A., and
Gonalves, C.A. (2003). Ketogenic diet increases
glutathione peroxidase activity in rat hippocam-
pus. Neurochem Res 28, 1793–1797.
Zlokovic, B.V. (2008). The blood-brain barrier in
health and chronic neurodegenerative disorders.
Neuron 57, 178–201.
305

SECTION IV
Ketone-Based Metabolism

General Health and Metabolic Alternatives

D O M I N I C P. D ’ AG O S T I N O , P H D , S E C T I O N E D I T O R
306
307
31
Overview of Ketone-Based Metabolism
General Health and Metabolic Alternatives
D O M I N I C P. D ’ AG O S T I N O , P H D
T he ketone bodies β-hydroxybutyrate (BHB)
and acetoacetate (AcAc) are produced
from fatty acids in the liver and serve as alterna-
tive energy sources for the brain, heart, skeletal
muscle, and peripheral tissues during prolonged
fasting, calorie restriction, strenuous exercise,
or adherence to a high-fat, low-carbohydrate
ketogenic diet (KD) (Cahill and Veech, 2003).
Ketones have historically been labeled as abnor-
mal metabolic byproducts of a pathological state
(VanItallie and Nufert, 2003), particularly due to
their association with diabetic ketoacidosis (DKA)
in uncontrolled type 1 diabetics, but emerging
interest has focused on nutritional ketosis as a
powerful metabolic therapy for general health and
a growing number of medical conditions in addi-
tion to drug-resistant epilepsy, where its use is well
established (Stafstrom and Rho, 2012). Nutritional
ketosis produces a nonpathological hyperketone-
mia resulting from decreased glucose availability,
lower insulin, and increased fat oxidation; how-
ever, long-term maintenance of the KD can be
difficult for some and requires strict adherence.
The restrictive nature of the KD has limited the
clinical applicability of therapeutic ketosis due to
practical considerations. Emerging data suggests
that many of the benefits of the KD are mecha-
nistically attributable to the ketone bodies or
specific medium chain triglycerides (MCTs), and
this has motivated investigators to develop strate-
gies to further augment the efficacy of the KD or
use metabolic-based supplements to circumvent
the need for dietary restriction to improve com-
pliance and the maintenance of this therapeutic
state (Newman and Verdin, 2014; Veech, 2004).
This section, “Ketone-Based Metabolism: General
Health and Metabolic Alternatives” includes chap-
ters that discuss the expanding medical and per-
formance applications of nutritional ketosis and
the emerging science of ketones and other related
metabolites as alternative fuels and potent signal-
ing molecules.
Induction of hyperketonemia through fasting
or the KD is known to produce acute and chronic
changes in metabolic physiology and molecu-
lar signaling pathways that provide therapeutic
effects in varied disease states. Metabolic-based
mechanisms of ketone therapies include an eleva-
tion of blood ketones and associated anaplerosis
with simultaneous suppression of blood glucose,
increased insulin sensitivity, improved mitochon-
drial efficiency, and suppressed oxidative stress
and inflammation. The reported benefits of hyper-
ketonemia and similar metabolic alternatives
have generated significant interest in the science
and application of implementing strategies for
inducing and sustaining blood levels of specific
metabolites (Kesl et  al., 2016). Most exogenous
ketone supplements and engineered anapleurotic
agents are currently under investigation for under-
standing their potential benefits for both healthy
and diseased individuals alike. It is likely that
most, if not all, of the conditions that are known
to benefit from the KD would receive some ben-
efit from exogenous ketone supplementation by
elevating blood ketones and lowering blood glu-
cose (Cahill and Veech, 2003). Currently the most
studied type of exogenous ketone would be ketone
esters, which induce a dose-dependent hyperke-
tonemia (1–7mM) in rats, mice, dogs, pigs, and
humans (Birkhahn and Border, 1978; Clarke et al.,
2012; Desrochers et al., 1995; Pascual et al., 2014).
A  recent case study of Alzheimer’s disease high-
lights its application in humans (Newport et  al.,
2015). There are a growing number of promising
metabolic alternatives to ketone esters, and many
of these agents and formulas are being evaluated
for their therapeutic efficacy, practical application
and potential synergy with the KD (Borges and
Sonnewald, 2012; Kesl et  al., 2016). Importantly,
308
308 section IV: Ketone-Based Metabolism
ketone supplementation and metabolic alterna-
tives provide a tool for achieving ketosis in patients
who are unable, unwilling, or uninterested in con-
suming a low-carbohydrate or ketogenic diet.
Included in this section are chapters that dis-
cuss conditions for which ketones and metabolic
alternatives have the most potential, including
seizure disorders, glucose transporter type 1 defi-
ciency syndrome (GLUT1DS), Alzheimer’s disease
(AD), brain and metastatic cancer, insulin resis-
tance and type 2 diabetes mellitus (T2DM), weight
loss, and performance. GLUT1DS is a rare genetic
disorder caused by a mutation in the SLC2A1
gene, which encodes the glucose transporter pro-
tein type 1. Patients with GLUT1DS have impaired
glucose transport into the brain and other tissues,
and thus alternative energy substrates in the form
of ketones or other metabolic alternatives offer a
means for the metabolic management of this dis-
order (Pascual et al., 2014). A known hallmark of
AD is impaired brain glucose metabolism, which
is associated with neurodegeneration and rapid
progression (Cunnane et  al., 2016). Preclinical
data demonstrate that ketones protect against AD
development and slow its progression (Kashiwaya
et al., 2013). Cancer is another disorder potentially
treated with nutritional ketosis due a metabolic
phenotype that is highly glycolytic, has abnormal
mitochondrial structure and function, and has a
deficiency in many ketolytic enzymes (Seyfried
and Shelton, 2010). Indeed, numerous studies have
shown that cancer cells lack the capacity to metab-
olize BHB for energy, and thus glucose restriction
can be lethal (Maurer et  al., 2011). This section
discusses the research showing that ketone ester
administration simultaneously decreases blood
glucose and insulin concentrations, suggesting
that exogenous ketone supplementation could be
used for insulin resistance and T2DM (Kashiwaya
et  al., 1997). The beneficial effects of ketones in
T2DM may be via direct regulation of HDAC-
dependent glucose metabolism and by inducing
resistance to oxidative stress, which helps restore
insulin sensitivity.
Included in this section is a chapter by Poff
et. al (Chapter  32) that describes the develop-
ment and testing of exogenous ketone supple-
ments, the diseases they are being investigated
for use in, and the potential mechanisms of
action of their therapeutic efficacy. The chapter
by Walker and Williams (Chapter  33) gives an
overview of the acute in vitro and in vivo stud-
ies of the MCT, decanoic acid, which imparts
specific signaling effects by directly inhibiting
AMPA receptors. Thus, this data demonstrates
a novel metabolic-independent mechanism of
MCTs, which were previously thought only to be
nutritional support enhancing ketogenesis. Also
included in this section is a chapter by Borges
(Chapter  34) on triheptanoin, the triglyceride of
heptanoate (C7 fatty acid), a metabolic alternative
to the KD that is being used to treat patients with
rare genetic metabolic disorders (Pascual et  al.,
2014). The compelling animal work associated
with triheptanoin shows that metabolic altera-
tions found in brains of rodent seizure models
can be restored by this compound, which appears
to cross the blood-brain barrier (Borges and
Sonnewald, 2012). Preclinical and clinical studies
indicate that triheptanoin is beneficial in numer-
ous neurological and neuro-muscular disorders,
making it a promising treatment for a variety of
clinical conditions (Borges and Sonnewald, 2012).
Very little attention has been given to the potential
function of protein and specific amino acids in the
ketogenic diet. Hartman (Chapter 35) focuses on
this topic by discussing the antiseizure potential
of select amino acids and the cellular and molecu-
lar mechanisms associated with their effects. The
chapter by Stafstrom et al. (Chapter 36) discusses
inhibition of glycolysis as one way that the keto-
genic diet suppresses seizures and how 2-deoxy-
D-glucose (2DG) is an anti-glycolytic tool that
has anti-seizure effects in numerous seizure mod-
els. Westman et al. (Chapter 37) review the ratio-
nale and recent clinical research supporting the
use of a low-carbohydrate, ketogenic diet in indi-
viduals with obesity and T2DM. The last chapter
in this section, Hyde et  al. (Chapter  38) discuss
the many health benefits of nutritional ketosis
that are relevant to athletes, including enhanced
fat oxidation, reduced inflammation, reduction in
blood lipids, and suppression of oxidative stress.
Zupec-Kania et al. (Chapter 39), reviews the gen-
esis and ongoing efforts of the Charlie Foundation
and Matthew’s Friends in promoting ketogenic
therapies for epilepsy and more recently as an
adjuvant for the metabolic management of can-
cer and other neurological and neurodegenerative
disorders. We hope the chapters in this section
will help the reader gain an appreciation for the
science and emerging applications of nutritional
ketosis, ketone supplementation, and other meta-
bolic alternatives that will inevitably be utilized
for the treatment and prevention for a broad
range of disease states.
REFERENCES
Birkhahn, R.H., and Border, J.R. (1978). Intravenous
feeding of the rat with short chain fatty acid
309
Chapter 31: Overview of Ketone-Based Metabolism
esters:  II. Monoacetoacetin. Am J Clin Nutr 31,
436–441.
Borges, K., and Sonnewald, U. (2012). Triheptanoin: a
medium chain triglyceride with odd chain fatty
acids: a new anaplerotic anticonvulsant treatment?
Epilepsy Res 100, 239–244,
Cahill, G.F., Jr., and Veech, R.L. (2003). Ketoacids?
Good medicine? Trans Am Clin Climatol Assoc
114, 149–161; discussion 162–143.
Clarke, K., Tchabanenko, K., Pawlosky, R., Carter,
E., Todd King, M., Musa-Veloso, K., Ho, M.,
Roberts, A., Robertson, J., Vanitallie, T.B., et  al.
(2012). Kinetics, safety and tolerability of (R)-3-
hydroxybutyl (R)-3-hydroxybutyrate in healthy
adult subjects. Regul Toxicol Pharmacol 63,
401–408,
Cunnane, S.C., Courchesne-Loyer, A., St-Pierre,
V., Vandenberghe, C., Pierotti, T., Fortier, M.,
Croteau, E., and Castellano, C.A. (2016). Can
ketones compensate for deteriorating brain glu-
cose uptake during aging? Implications for the
risk and treatment of Alzheimer’s disease. Ann NY
Acad Sci,Mar;1367(1):12–20.
Desrochers, S., Dubreuil, P., Brunet, J., Jette, M.,
David, F., Landau, B.R., and Brunengraber, H.
(1995). Metabolism of (R,S)-1,3-butanediol ace-
toacetate esters, potential parenteral and enteral
nutrients in conscious pigs. Am J Physiol 268,
E660–667.
Kashiwaya, Y., King, M.T., and Veech, R.L. (1997).
Substrate signaling by insulin:  a ketone bodies
ratio mimics insulin action in heart. Am J Cardiol
80, 50A–64A.
Kashiwaya, Y., Bergman, C., Lee, J.H., Wan, R., King,
M.T., Mughal, M.R., Okun, E., Clarke, K., Mattson,
M.P., and Veech, R.L. (2013). A ketone ester diet
exhibits anxiolytic and cognition-sparing proper-
ties, and lessens amyloid and tau pathologies in a
mouse model of Alzheimer’s disease. Neurobiol
Aging 34, 1530–1539.
309
Kesl, S.L., Poff, A.M., Ward, N.P., Fiorelli, T.N., Ari,
C., Van Putten, A.J., Sherwood, J.W., Arnold, P.,
and D’Agostino, D.P. (2016). Effects of exogenous
ketone supplementation on blood ketone, glucose,
triglyceride, and lipoprotein levels in Sprague-
Dawley rats. Nutr Metab (Lond) 13, 9.
Maurer, G., Brucker, D., Bähr, O., Harter, P., Hattingen,
E., Walenta, S., Mueller-Klieser, W., Steinbach, J.,
and Rieger, J. (2011). Differential utilization of
ketone bodies by neurons and glioma cell lines: a
rationale for ketogenic diet as experimental glioma
therapy. BMC Cancer 11, 315.
Newman, J.C., and Verdin, E. (2014). Ketone bodies as
signaling metabolites. Trends Endocrinol Metabol
25, 42–52.
Newport, M.T., VanItallie, T.B., Kashiwaya, Y., King,
M.T., and Veech, R.L. (2015). A new way to pro-
duce hyperketonemia: use of ketone ester in a case
of Alzheimer’s disease. Alzheimers Dement11,
99–103.
Pascual, J.M., Liu, P., Mao, D., Kelly, D.I., Hernandez,
A., Sheng, M., Good, L.B., Ma, Q., Marin-Valencia,
I., Zhang, X., et  al. (2014). Triheptanoin for glu-
cose transporter type I deficiency (G1D): modula-
tion of human ictogenesis, cerebral metabolic rate,
and cognitive indices by a food supplement. JAMA
Neurol 71, 1255–1265.
Seyfried, T.N., and Shelton, L.M. (2010). Cancer as a
metabolic disease. Nutr Metab (Lond) 7, 7.
Stafstrom, C.E., and Rho, J.M. (2012). The ketogenic
diet as a treatment paradigm for diverse neuro-
logical disorders. Front Pharmacol 3, 59.
VanItallie, T.B., and Nufert, T.H. (2003). Ketones: metab-
olism’s ugly duckling. Nutr Rev 61, 327–341.
Veech, R.L. (2004). The therapeutic implications of
ketone bodies:  the effects of ketone bodies in
pathological conditions:  ketosis, ketogenic diet,
redox states, insulin resistance, and mitochondrial
metabolism. Prostaglandins Leukot Essent Fatty
Acids 70, 309–319.
310
32
Ketone Supplementation for Health and Disease
A N G E L A M . P O F F, P H D , S H A N N O N L . K E S L , P H D ,
A N D D O M I N I C P. D ’ AG O S T I N O , P H D
INDUCING THERAPEUTIC
K E TO S I S W I T H   K E TO N E
S U P P L E M E N TAT I O N
Ketones are typically produced in the liver only
under certain physiological conditions associated
with the suppression of the hormone insulin: star-
vation, fasting, calorie restriction, prolonged
exercise, or during the consumption of a high-
fat, low-carbohydrate ketogenic diet (KD). The
restrictive nature of these states has limited the
clinical applicability of therapeutic ketosis due to
practical considerations. In an effort to circumvent
this dilemma, researchers have recently developed
a number of exogenous ketogenic supplements—
ketogenic precursors that are metabolized
to produce a dose-dependent elevation of β-
hydroxybutyrate (βHB) and acetoacetate (AcAc)
in the blood (Clarke et al., 2012b; D’Agostino et al.,
2013; Kesl et  al., 2016; Puchowicz et  al., 2000).
These ketone supplements allow for a rapid and
controlled induction of physiologic ketosis with-
out the need for severe dietary restrictions. Since
many of the benefits of ketosis are mechanistically
attributable to the ketone bodies, it is possible that
exogenous ketone supplementation could mimic
the therapeutic efficacy of the KD for certain dis-
orders. In addition, the adjuvant use of exogenous
ketone supplements for patients consuming a KD
could provide a method for faster adaptation, eas-
ier sustainment of ketosis, and perhaps help return
a sense of normalcy to patients and their caregiv-
ers. Furthermore, in some scenarios, an exogenous
ketone-supplemented KD may be more effective
than KD or ketone supplementation alone.
DEVELOPMENT AND
T E S T I N G O F   K E TO N E
SUPPLEMENTS
There are numerous sources of ketones and keto-
genic precursors being developed and tested,
including medium chain triglycerides, diols, salts,
and esters that have been shown to elevate blood
ketone levels independent of caloric or carbo-
hydrate restriction. The investigation of natural
and synthetically derived ketogenic precursors to
establish nutritional ketosis without the need for
dietary restriction has revealed that each formu-
lation has distinct properties in terms of extent
and duration of ketosis as well as metabolic sig-
naling properties, including anticonvulsant effects
(D’Agostino et  al., 2013; Kesl et  al., 2015a; Kesl
et al., 2016). Most of the developed ketone supple-
ments are currently under investigation for safety
and efficacy in a number of disease states.
Medium Chain Triglycerides
Medium chain triglycerides (MCTs) contain a
glycerol backbone esterified to medium-chain
fatty acids (MCFAs), which are fatty acids with
hydrocarbon side chains 6 to 12 carbons in length.
The MCFAs include caproic acid (C6:0, hexanoic
acid), caprylic acid (C8:0, octanoic acid), cap-
ric acid (C10:0, decanoic acid), and lauric acid
(C12:0, dodecanoic acid). The MCTs are natu-
rally found in coconut oil (~60%), palm kernel oil
(~30%), cheese (7.3%), whole milk (6.9%), but-
ter (6.8%), and full-fat yogurt (6.6%) (Karen and
Welma, 2015; Koji and Truyoshi, 2010). Compared
to long-chain fatty acids (LCFAs), MCFAs have a
lower melting point, smaller molecule size, and are
less calorically dense (8.3 calories per gram ver-
sus 9.2). These distinct physiochemical properties
allow MCTs to be absorbed directly into the blood-
stream through the hepatic portal vein without the
need for bile or pancreatic enzymes for degrada-
tion. Additionally, MCTs do not require carnitine
to enter the mitochondria, but rather quickly cross
the mitochondrial matrix, where they are metabo-
lized to acetyl co-A and subsequently to ketone
bodies. Thus, they are easily and rapidly digested,
transferred to the liver, and used for energy rather
than stored as fat. In comparison, LCFAs require
311
Chapter 32: Ketone Supplementation for Health and Disease
re-esterification in the small intestine, transport
by chylomicrons via the lymphatic and vascular
systems, and oxidation in the liver for energy or
storage. The MCTs are metabolized as rapidly as
glucose but have roughly twice the energy density
(Bell et  al., 1997). In the early 1980s, Dr.  Vigen
K. Babayan of the Nutrition Laboratory at Harvard
University developed a method to produce MCTs
in large quantities (Bach and Babayan, 1982).
These commercialized MCTs are acquired through
lipid fractionation from natural fats such as coco-
nut oil and milk and are predominantly composed
of C:8 and C:10 MCTs (Babayan, 1987; Hashim
and Tantibhedyangkul, 1987).
Since the 1970s, MCTs have been used in a
modification of the classical KD as an alternative
fat source (Huttenlocher et  al., 1971). The keto-
genic properties of MCTs allow patients to eat less
total fat in their diet and include more carbohy-
drate and protein without sacrificing their nutri-
tional ketosis. Although MCTs have the potential
to be an efficient and beneficial ketogenic fat,
they are currently limited in clinical usage due to
gastrointestinal (GI) side effects stimulated by
the large dose needed to induce ketonemia (typi-
cally greater than 40 g/day) (Bell et al., 1997). The
original modified KD with MCT allowed 60% of
its energy to be derived from MCTs; however, this
caused GI distress in some children (Huttenlocher,
1976; Mak et  al., 1999; Sills et  al., 1986; Trauner,
1985). For this reason, an additional modified
KD with MCT was developed using only 30% of
its energy from MCTs; however, it induced much
lower levels of ketosis (Elizabeth et al., 2009; Ruby
et  al., 1989). In a recent study, juvenile Sprague-
Dawley rats were given a daily intragastric bolus
of MCT oil (10 g/kg) which rapidly elevated blood
βHB levels (4mM βHB, <30 min) and remained
significantly elevated for up to 12 hours (Kesl
et al., 2016). In addition to their metabolic effects,
emerging data suggests that octanoic acid (C8) has
specific anticonvulsant and neuroprotective prop-
erties independent of ketones (Chang et al., 2015).
Research and development is currently being con-
ducted to formulate MCT oil into a powder with
soluble fiber compounds, which may delay gastric
emptying and enhance absorption to improve tol-
erability and ketogenic potential.
1,3-Butanediol
1,3-Butanediol (BD; also known as 1,3-butylene
glycol) is an FDA-approved organic di-alcohol
(diol), naturally present in some species of pepper
(Capsicum annuum), and is used as a food flavor-
ing solvent, an intermediate in the manufacture
311
of certain polyester plasticizers, and a humec-
tant for cosmetics, and has been considered as a
synthetic food for long-duration space missions
(Budavari, 1989; Dymsza, 1975; Falbe et al., 1985).
When ingested orally, BD is metabolized by the
liver via alcohol dehydrogenase (ADH) to β-
hydroxybutyraldehyde, which is rapidly oxidized
to βHB by aldehyde dehydrogenase (Tate et  al.,
1971). Further, BD contributes approximately 6
kcal/g of energy and can produce dose-dependent
millimolar concentrations of ketones in the blood
in a ratio of 6:1 of βHB to AcAc (D’Agostino et al.,
2013; Desrochers et  al., 1992; Drackley et  al.,
1990; Tobin et  al., 1972). Extensive toxicology
studies have concluded that BD is safe with very
few adverse health effects in humans or animals
(Dymsza, 1975; Hess et al, 1981; Opitz, 1958; Scala
and Paynter, 1967).
In a 28-day study, a daily 5-g/kg dose of BD
administered via intragastric gavage in Sprague-
Dawley rats showed a significant elevation of
blood ketones (1mM βHB, <30 min) without an
effect on blood glucose, triglyceride, or lipoprotein
levels (Kesl et  al., 2016). Recently, BD has been
investigated as a backbone of ketone mono- and
diesters, which are discussed later in this chapter.
Ketone Salts
Originally, researchers attempted to administer
βHB or AcAc in their free acid forms; however,
this was shown to be too expensive and ineffec-
tive at producing sustained ketosis. Subsequently,
it was suggested to buffer the free acid of βHB
with sodium, but this causes potentially harmful
sodium overload and mineral imbalance at thera-
peutic levels of ketosis, and is it has been shown
that βHB alone is largely ineffective at preventing
seizures in animal models (Bough and Rho, 2007).
A  study showed that oral administration of Na+/
βHB in doses from 80 to 900 mg/kg/day elevated
blood ketone levels to 0.19–0.36 mM, which was
therapeutic in children with acyl CoA dehydro-
genase deficiency (Hove et  al, 2003). However,
for a 70-kg man to achieve these levels of ketosis
would require ingesting between 5.6 and 6.3 g/day.
Considering the potential effects of such a large
sodium load, the costs of the administration of Na+/
βHB salts to achieve ketosis made this approach
unrealistic (Veech, 2004).
More recently, ketone salts with a balanced
electrolyte formulation to prevent sodium over-
load have been developed and tested and can
include potassium, calcium, magnesium, lithium,
arginine, lysine, histidine, ornithine, creatine,
agmatine, or citrulline. Maintaining a balanced
312
312 section IV: Ketone-Based Metabolism
electrolyte ratio should help offset any potential
adverse effects of sodium on blood pressure. It is
speculated that this formulation will be especially
beneficial for elderly patients most susceptible
to sodium-induced hypertension (Veech, 2004).
Furthermore, when first attempting to follow a
KD, many people experience headaches and leth-
argy during the initial stages of “keto-adaptation,”
a term that describes the adaptive changes in
metabolic physiology associated with transition-
ing more toward fat and ketone metabolism. The
symptoms are largely a result of reduced glucose
availability to the brain and a transient depletion
of minerals, especially sodium, potassium, and
magnesium in the plasma (Zhang et  al., 2013b).
These symptoms can be attenuated or reversed
with sufficient supplementation of sodium, potas-
sium, calcium, and magnesium, and thus a ketone
supplement that delivers ketones with these elec-
trolytes would be favorable.
Several mineral ketone salt combinations are
currently being tested for safety and efficacy. In a
recent study, neither a 5-g/kg nor a 10-g/kg intra-
gastric gavage of Na+/K+ βHB salts significantly
elevated blood ketone levels or reduced blood glu-
cose levels in juvenile Sprague-Dawley rats (Kesl
et  al., 2016). However, there appears to be inter-
species variability in absorption, as in a recent case
study a 100-kg male subject was administered a
4% Na+/K+ βHB salt solution (containing 11 grams
of sodium and 7.1 grams βHB) and his plasma
levels of βHB were significantly elevated at 30–60
minutes after administration. After receiving the
same dose for 3  days, the patient had sustained
elevated blood ketone levels from 15 to 120 min-
utes after administration. Similar results were seen
in a 70-kg male who fasted 3 days prior to admin-
istration of the ketone salt supplement (D’Agostino
et al., 2014). In a 15-week study, Sprague-Dawley
rats were administered Na+/Ca+2 ketone salt 20%
by weight (~25 g/kg/day) in their food fed ad libi-
tum. The Na+/Ca+2 βHB–supplemented rats exhib-
ited sustained elevated blood ketone levels (1 mM
βHB) at 1, 4, 8, 10, and 13 weeks of chronic feed-
ing, which did not affect blood glucose levels com-
pared with controls (Poff et al., 2016).
Combination βHB Mineral Salt
and Medium Chain Triglyceride
Considering the variety of ketogenic precursors
available, researchers are investigating unique
combinations of the individual supplements in
hopes of optimizing their benefits. A  combina-
tion of βHB mineral salts (BMS) and MCT oil has
been administered in ratios of 1:1 to 1:2 mixtures.
Formulating in this way allows for rapid and sus-
tained elevation of ketosis by delivering exogenous
ketones while simultaneously stimulating endog-
enous ketogenesis with MCTs. In addition, the
combination formulation allows for a lower dosing
of the components as compared to administering
the individual compounds, thus reducing poten-
tial for side effects (gastric hyperosmolality) and
resulting in a distinct blood ketone profile that is
sustained over a longer period of time (D’Agostino
et al., 2014).
In a 28-day study, the combination of a 50%
Na+/K+ βHB salt mixed in a 1:1 solution with
MCT oil (BMS+MCT) significantly elevated and
sustained blood ketone levels and reduced blood
glucose levels in a dose-dependent manner (Kesl
et al., 2016). Additionally, the study demonstrated
a significant correlation between elevated blood
ketone levels and reduced blood glucose levels post
intragastric gavage administration with no effect
on blood triglyceride or lipoprotein levels. A note-
worthy observation from the study revealed that
in rats, MCT caused a rise in ketones (>3mM) that
exceeds what is possible in humans due to gastro-
intestinal intolerance. This is likely due to the fact
that rats have a higher rates of absorption for MCT
and higher rates of hepatic fat metabolism, which
stimulates greater ketone production. Considering
these results it is important to take into account
interspecies variability in the metabolic response to
ketone supplements, which will need to be further
characterized to fully understand how we could
extrapolate findings and translate into human dos-
ing equivalents. In the rats, the BMS+MCT sup-
plement elevated blood ketones similar to that of
MCT alone; however, the gastric side effects were
not observed, suggesting a potential method for
avoiding this unwanted adverse effect (Kesl et al.,
2016). In a case study, a 100-kg male was admin-
istered a combination of a 4% Na+/K+ βHB salt
solution (containing 11 grams of sodium and 7.1
grams βHB) + 20 mL MCT oil. This combination
demonstrated higher elevation of blood ketone
levels than either βHB salts or MCT oil alone,
starting at 15 minutes post consumption and last-
ing for 4 hours. Similar results were observed in
a 70-kg male who fasted 3  days prior to admin-
istration; however elevated blood ketone levels
were observed sooner after supplementation and
were sustained for a considerably longer time after
administration (8 hours) (D’Agostino et al., 2015).
This effect was accompanied by a reduction in
blood glucose, and a lower starting blood glucose
concentration on each subsequent day of supple-
mentation (D’Agostino et al., 2014). In a 15-week
313
Chapter 32: Ketone Supplementation for Health and Disease
study, Sprague-Dawley rats were administered
a 1:1 mixture of Na+/Ca+2 ketone salt + MCT oil
(20% by weight, ~25 g/kg/day) in their food fed ad
libitum. The combination-supplemented rats had
significantly sustained and elevated blood ketone
levels at weeks 3, 4, 8, 10, and 13 without signifi-
cantly affecting blood glucose levels during the
study (Kesl et al., 2014).
Ketone Esters
Researchers have developed and investigated sev-
eral synthetic ketone mono- and di-esters to induce
nutritional ketosis independent of dietary calorie
or carbohydrate restriction. When ketone esters
are administered, gastric esterases liberate ketones
(βHB and AcAc) as a free acid from a backbone
molecule, which varies depending on the specific
formulation, but is favorably a ketogenic precur-
sor such as BD. As previously discussed, BD is
subsequently metabolized by the liver to produce
βHB (D’Agostino et  al., 2013). Thus, the ketone
esters currently available are unique among the
aforementioned ketone supplements in that they
can directly elevate ketones and supply ketogenic
precursors that can further sustain ketogenesis.
Additionally, synthetically derived ketone esters
are currently the most potent form of exogenous
ketones available, but their potency also necessi-
tates a thorough investigation of their long-term
safety and toxicity.
In the late 1970s, Birkhahn et  al. were the
first to synthesize a monoester of glycerol and
AcAc (monoacetoacetin) for parenteral nutri-
tion. These studies demonstrated that monoace-
toacetin induced hyperketonemia comparable to
fasted rats at a dose of 50 g/kg per day (Birkhahn
and Border, 1978; Birkhahn et al., 1977; Birkhahn
et  al., 1979). In attempts to increase the caloric
density of monoacetoacetin, they synthesized
both a monoester and triester of glycerol and βHB.
These esters are hydrolyzed to release free βHB in a
way that elevates and sustains blood ketones. Later,
Desrochers and colleagues synthesized mono-
and diesters of AcAc with BD that elevated both
AcAc and βHB (Desrochers et  al., 1995b). These
and other ketone esters developed by or in col-
laboration with Henri Brunengraber and Richard
Veech have demonstrated an ability to induce a
dose-dependent hyperketonemia (1–7 mM) in
rats, mice, dogs, pigs, and humans (Brunengraber,
1997; Ciraolo et al., 1995; Desrochers et al., 1995a;
Puchowicz et  al., 2000; Srivastava et  al., 2012;
Sylvain et  al., 1995). Clarke and colleagues dem-
onstrated the safety of a ketone ester in rats and
humans that has also been documented in a recent
313
case study of Alzheimer’s disease (AD; Clarke et al.,
2012a; Clarke et al., 2012b; Newport et al., 2015).
Recently, the ketone ester R,S-1,3-butanediol ace-
toacetate diester (BD-AcAc2), given as an intragas-
tric gavage, elevated both AcAc and βHB blood
levels to >3mM in rats (D’Agostino et  al., 2013).
The induction of therapeutic ketosis was rapid
(within 30 minutes) and sustained at high levels
for over 4 hours. In a 28-day study, a daily intra-
gastric gavage (5 g/kg body weight) of BD-AcAc2
induced significantly elevated blood ketone lev-
els and significantly reduced blood glucose levels
without significantly altering blood triglyceride or
lipoprotein levels (Kesl et al., 2016). In a 15-week
chronic feeding study, the BD-AcAc2 was admin-
istered to Sprague-Dawley rats in a low dose (5%
food weight, 10 g/kg/day) (LKE) and a high dose
(20% food weight, 25 g/kg/day) (HKE) ad libitum.
Both doses significantly elevated blood ketone lev-
els without reducing blood glucose levels. Serum
clinical chemistry of both LKE and HKE did not
reveal any alterations in markers of kidney and
liver function compared with rats fed standard
chow (Poff et al., 2016).
P OT E N T I A L T H E R A P E U T I C
M E C H A N I S M S O F   K E TO N E
S U P P L E M E N TAT I O N
Induction of hyperketonemia produces acute and
chronic changes in metabolic physiology and
molecular signaling pathways that provide thera-
peutic effects in varied disease states. Metabolic-
based mechanisms of ketone therapies include
an elevation of blood ketones and associated
anaplerosis with simultaneous suppression of
blood glucose, enhancement of insulin sensitivity,
mitochondrial efficiency, suppression of specific
inflammatory mediators and inhibition of oxi-
dative stress and preservation of mitochondrial
health and function.
Suppression of Blood Glucose and
Enhancement of Insulin Sensitivity
Hyperglycemia and hyperinsulinemia are patho-
logically linked to numerous disorders, includ-
ing cancer, cardiovascular disease, obesity, type
2 diabetes, impaired wound healing, and neu-
rodegenerative diseases, among others (Laakso
and Kuusisto, 2014; Ryu et al., 2014). These states
are associated with chronic systemic inflamma-
tion, oxidative stress, impairment of the immune
system, and vascular and metabolic dysfunction
(Bornfeldt and Tabas, 2011; de Carvalho Vidigal
et al., 2012; Turina et al., 2005). Exogenous ketone
supplements may provide therapeutic benefits
314
314 section IV: Ketone-Based Metabolism
in diseases characterized by hyperglycemia or
hyperinsulinemia as reports have demonstrated
that ketone administration lowers blood glucose
by increasing insulin sensitivity. To demonstrate
this, male rats were fed a standard diet with 30% of
calories replaced with the R-3-hydroxybutyrate-R-
1,3-butanediol monoester for 14 days. The ketone
ester-supplemented diet induced nutritional keto-
sis (3.5 mM βHB), and both plasma glucose and
insulin were decreased by approximately 50%
(Srivastava et  al., 2012). Glucose was decreased
from 5 mM to 2.8 mM, and insulin was decreased
from 0.54 ng/mL to 0.26 ng/mL. In a similar study
by the same group, mice receiving a KE diet exhib-
ited a 73% increase in the Quantitative Insulin-
Sensitivity Check Index (QUICKI), a surrogate
marker of insulin sensitivity, compared with con-
trol, calorie-matched mice (Srivastava et al., 2012).
Fasting plasma glucose levels were not altered in
these mice, but fasting plasma insulin levels were
reduced by approximately 85% in the KE-fed mice
compared with controls, demonstrating that exog-
enous ketones enhance insulin sensitivity.
In a study examining the potential use of
ketogenic supplements as a cancer therapy, mice
consuming a standard high carbohydrate diet
mixed with the R,S-1,3-butanediol diacetoacetate
ester (BD-AcAc2) at 20% by weight had approxi-
mately 30% lower blood glucose than control
animals (Poff et  al., 2014). The administration of
BD did not significantly elevate blood ketones in
this study, nor did it decrease blood glucose. This
result is likely due in part to the comparatively
small dose of BD being consumed at any point in
time according to the route of administration, as
BD in higher doses is known to elevate ketones.
Insulin levels were not investigated in this study.
In a study designed to assess the dose-dependent
effects of exogenous ketone supplements on blood
glucose, ketones, and lipids, healthy male rats
were administered one of five ketogenic agents
daily via intragastric gavage (Kesl et al., 2016). The
ketogenic agents investigated included:  BD, BD-
AcAc2, a sodium/potassium β-hydroxybutyrate
mineral salt (BMS), medium chain triglyceride oil
(MCT oil), BMS+MCT 1:1 mixture (BMS+MCT),
and the R,S-1,3 butanediol diacetoacetate ester
(BD-AcAc2). BD-AcAc2, BMS, MCT oil, and
BMS+MCT-treated rats demonstrated a decrease
in blood glucose following ketone supplement
administration given as an acute bolus. Similarly to
the previously mentioned mouse study, BD did not
lower blood glucose in these animals, although it is
known to be a hypoglycemic agent. The duration
of the reduction in blood glucose varied between
supplements and lasted anywhere from 30 minutes
to 12 hours, suggesting that ketone supplementa-
tion could potentially provide a novel method of
glycemic control.
Enhanced Metabolic Efficiency
The superior metabolic efficiency of ketone bod-
ies has been known since the 1940s, when Henry
Lardy compared the energetic efficiency of 16
major carbohydrate, lipid, and intermediary
metabolites (Lardy and Phillips, 1945). He dem-
onstrated that βHB and AcAc were unique among
the panel of metabolites tested in their ability to
increase bull sperm mobility while simultaneously
decreasing oxygen consumption. Nearly 50  years
later, Richard Veech and colleagues confirmed
Lardy’s observation and elucidated the molecular
mechanisms underlying the phenomenon in the
working perfused rat heart (Kashiwaya et al., 1994).
They demonstrated that supplementation of 5 mM
ketones (4 mM βHB, 1 mM AcAc) to glucose-
containing perfusate (10 mM glucose) increased
cardiac hydraulic work by approximately 25%
while simultaneously reducing oxygen consump-
tion (Kashiwaya et al., 1994). Their study revealed
that this effect was mediated by a reduction of the
mitochondrial NAD couple and an oxidation of the
coenzyme Q couple, which increases the energy of
the redox span between these sites. This results in
an increase in energy released by electrons in the
ETC, causing more protons to be pumped into
the inner mitochondrial space, thus enhancing the
electrochemical gradient established there and
increasing the energy of ATP hydrolysis. Indeed,
thermodynamic tables for heat of combustion, cal-
culated with bomb calorimeter experiments, show
that βHB produces more energy than glucose per
2-carbon moiety (Cahill and Veech, 2003). These
results are supported by human studies that dem-
onstrated a reduction in blood flow and oxygen
consumption in the brains of fasted obese sub-
jects in ketosis (McHenry, 1966). Thus, ketones
appear to be a superior fuel for ATP production
per unit oxygen compared to glucose, and can be
considered among the most metabolically efficient
energy metabolites (Figure 32.1).
Anti-Inflammatory Effects
The physiological state of ketosis has known
anti-inflammatory properties, as several stud-
ies have demonstrated that the KD reduces cir-
culating inflammatory markers in animals and
in humans (Cassandra et  al., 2007; Sharman and
Volek, 2004; Torres-Gonzalez et al., 2008). Recent
evidence suggests that this effect is, at least in
315

Chapter 32: Ketone Supplementation for Health and Disease 315

Interstitial 3Na+
space

Cytosol ADP+Pi ATP

2K+
Glucose ADP+Pi ATP
Major Energy Glucose ADP+Pi
∆G ATP
Source NAD+ ATP synthase
ATP Mitochondrion
NADH
Pyruvate– Pyruvate–
PDH
3H+ Cytochrome
Glycolysis oxidase 1/2O2
4H+
Cyt aa3
Acetyl CoA Electron 2e– H2O
transport Cyt C
system
CoQH2-CytC
Reductase
D-B-hydroxybutyrate
– TCA cycle 4H+
CoQ
CoQH2
2e–
Alternative Energy NAD+ NADH DH
Source
NAD+
Acetoacetyl CoA NADH 4H+
NADH
Acetoacetate–
– –
BHB BHB

FIGURE 32.1 Ketone bodies added in the presence of glucose fundamentally enhance mitochondrial metabolism.
When added to glucose, a physiological level of ketone bodies reduces the mitochondrial NAD couple, oxidizes the
coenzyme Q couple and increases the ΔG0 of ATP hydrolysis. These changes are shown on the right side of the figure.
The arrows illustrate the effects of ketone bodies compared with glucose alone. Ketone bodies provide an alternative
metabolic fuel, which can preserve ATP production during the impairment of glycolysis, as occurs in diabetes or
insulin resistance, or during inhibition of pyruvate dehydrogenase complex (PDH) as occurs in the presence of amyloid.
Oxidation of the coenzyme Q couple decreases superoxide anion production. Increases in the ΔG0 of ATP hydrolysis
widen the extra/intracellular ionic gradients leading to hyperpolarization of cells, which may play a role in treating
certain forms of epilepsy. The actions of ketone bodies mimic the acute effects of insulin in insulin-sensitive tissue and
tissues with high metabolic demands, including heart and brain. Figure taken with permission from Veech, et al.: Ketone
bodies—potential therapeutic uses.

part, directly mediated by the ketone bodies,
indicating that exogenous ketone supplementa-
tion could be used to suppress inflammation.
Dixit and colleagues reported that βHB inhibits
the NLRP3 inflammasome, an important compo-
nent of the innate immune system that controls
activation of caspase-1 and production of the
pro-inflammatory cytokines IL-1β and IL-18 by
macrophages (Youm et  al., 2015). This effect was
mediated specifically by βHB, as neither AcAc nor
the structurally related fatty acids butyrate and
acetate elicited this response. The βHB-induced
inhibition of the NLRP3 inflammasome occurs
without its being oxidized in the TCA cycle, and
is thus independent of its function as an energy
metabolite. Furthermore, the anti-inflammatory
changes associated with βHB were not dependent
on alterations in AMPK, reactive oxygen species
(ROS), glycolytic inhibition, UCP, or SIRT2 signal-
ing, further validating its function as a signaling
metabolite. To investigate the in vivo translatabil-
ity of this finding, researchers administered the
BD-AcAc2 to a familial cold anti-inflammatory
syndrome (FCAS) mouse model with an induced
missense mutation in NLRP3. The ketone ester
protected the mice from neutrophilia and hyper-
glycemia, and did not affect infiltration of perito-
neal macrophages or overall frequency of splenic
T cells, macrophages, or neutrophils, leading the
authors to conclude that elevating blood ketones
could be a therapeutic option for patients with
NLRP3-mediated chronic inflammatory diseases.
Indeed, other research has shown that inhibition
of the NLRP3 inflammasome mitigates the sever-
ity of numerous inflammatory diseases, includ-
ing atherosclerosis, type 2 diabetes, AD, and gout,
316
316 section IV: Ketone-Based Metabolism
among others (Duewell et al., 2010; Heneka et al.,
2013; Martinon et al., 2006; Vandanmagsar et al.,
2011; Youm et al., 2013; Youm et al., 2015).
In recent studies rats were fed one of three
exogenous ketone supplements (BD-AcAc2, BMS,
or a 1:1 mixture of BMS:MCT mixed into stan-
dard rodent chow at 18% by weight for 15 weeks.
Inflammatory profiling was performed on serum
collected from the animals at the end of the
chronic feeding study and revealed decreases in
several pro-inflammatory cytokines including
IL-1β, IL-6, IFN-γ, MCP-1, and RANTES (Poff
et al., 2016).
Inhibition of Oxidative Stress
The ketogenic diet has been reported to reduce
oxidative stress in vivo in a number of preclini-
cal and clinical reports (Jarrett et  al., 2008).
Studies suggest that this effect is mediated by
the ketone bodies themselves, and therefore the
effect would likely be recapitulated with exog-
enous ketone supplementation. Some important
molecular mechanisms underlying the effects
of ketone metabolism on oxidative stress were
delineated by studies investigating the bioener-
getic efficiency and mitochondrial respiration in
the working perfused rat heart following admin-
istering of a glucose-containing perfusate supple-
mented with exogenous 5 mM βHB (Kashiwaya
et  al., 1994). As described previously, ketone
metabolism increases the oxidation of ubiqui-
nol (Q) in the electron transport chain, reducing
semiquinone radical (Q∙), an intermediate in the
reduction of ubiquinone that is sensitive to oxida-
tion by molecular oxygen to produce superoxide
anion (O2−∙). O2−∙ is an important precursor for
the generation of many ROS; therefore, ketone
metabolism suppresses mitochondrial ROS pro-
duction (Kashiwaya et al., 1994). Simultaneously,
ketone metabolism suppresses oxidative stress
by enhancing endogenous antioxidant capac-
ity. Ketone metabolism induces reduction of the
mitochondrial NAD and cytoplasmic NADP cou-
ples. And NADH and NADPH are necessary for
the regeneration of reduced glutathione (GSH),
which is required for the neutralization of ROS
by glutathione peroxidase, an important endog-
enous antioxidant enzyme. These molecular
effects are observed in vivo, as the KD has been
shown to increase the ratio of reduced to oxidized
mitochondrial glutathione (GSH:GSSG) in rat
brains (Jarrett et  al., 2008). These effects appear
to be ubiquitous in various tissues; however, the
brain has been the most well characterized in this
regard. For example, in vitro treatment with βHB
and AcAc decreases neuronal ROS production
following glutamate exposure (Maalouf et  al.,
2007)  and inhibits apoptosis in cortical slices
exposed to hydrogen peroxide (H2O2) (Kim do
et al., 2007).
Eric Verdin and colleagues recently demon-
strated that βHB functions as an endogenous his-
tone deacetylase inhibitor (HDACI) in vitro and
in vivo at physiologic concentrations easily achiev-
able with exogenous ketone supplementation
(Shimazu et al., 2013). Fasting, calorie restriction,
and exogenous βHB administration all increased
global histone acetylation in mice and induced the
transcriptional activation of the oxidative stress
resistance factors FOXO3A and MT2. This effect
was found to be mediated directly by inhibition of
class 1 and 2 HDACs. Furthermore, the authors
demonstrated that exogenous ketone supplemen-
tation could prevent oxidative stress. Mice were
administered βHB via a subcutaneous pump for
24 hours prior to receiving an injection of para-
quat, which induces the production and accumula-
tion of ROS. Protein carbonylation was suppressed
by 54%, and lipid peroxidation was completely
suppressed, in the kidneys of βHB pretreated mice.
Immunoblotting of kidney tissue from these ani-
mals revealed a significant increase in the mito-
chondrial superoxide dismutase (MnSOD) and
catalase (CAT) endogenous antioxidant systems.
This study strongly supports the feasibility and
applicability of exogenous ketone supplements for
the prevention of oxidative stress.
Preservation of Mitochondrial
Health and Function
Ketone metabolism is generally recognized to
support or enhance mitochondrial health (Veech,
2004). As previously described, ketones suppress
mitochondrial ROS production and enhance
endogenous antioxidant systems. The resultant
reduction in oxidative stress protects the mito-
chondrial DNA and membranes from damage
that would impair respiratory and mitochondrial
function. Interestingly, exogenous ketone supple-
mentation with a ketone ester has been shown to
induce mitochondrial biogenesis (Srivastava et al.,
2012). Mice in this study were fed a diet from
which approximately 30% of calories were derived
from the ketone ester D-β-hydroxybutyrate-R-
1,3-butanediol monoester for one month. The
mitochondrial content and expression of elec-
tron transport chain proteins were significantly
increased in the intrascapular brown adipose tis-
sue as compared with control mice, although calo-
rie intake was matched between the two groups.
317
Chapter 32: Ketone Supplementation for Health and Disease
CONDITIONS WHERE
K E T O N E S U P P L E M E N TAT I O N
I S L I K E LY B E N E F I C I A L
Because of the previously described therapeutic
mechanisms of ketone metabolism, exogenous
ketone supplementation is being investigated as
a potential treatment for a number of disorders.
Here we discuss some conditions for which ketone
supplements have been most well demonstrated to
be useful, including epilepsy and seizure disorders,
glucose transporter type 1 deficiency syndrome,
AD, cancer, insulin resistance and type 2 diabetes
mellitus, and weight loss.
Epilepsy/Seizure Disorders
The KD is a proven, effective therapy for epilepsy
in children and adults (Klein et al., 2014; Li et al.,
2013). In many cases, it is more effective than anti-
epileptic drugs (AEDs) and is therefore routinely
used as a front-line treatment for children with
retractable (drug-resistant) epilepsy (Levy et  al.,
2011). Although the mechanisms of KD therapy
are largely unknown, achieving and sustaining
therapeutic ketonemia (>1 mM blood ketones) or
ketonuria (>40 mg/dL) is generally necessary for
antiseizure efficacy. Despite its success, the dietary
restrictions of the KD can be unpalatable for some
patients and difficult for caregivers, a contributing
factor for cessation of treatment (Klein et al., 2014;
Levy et  al., 2011). Exogenous ketogenic supple-
mentation mimics the metabolic and physiologic
effects of the KD, including enhancing mito-
chondrial biogenesis, anaplerosis, suppression
of glycolysis, and increasing ATP and adenosine
production, all thought to mediate the therapeu-
tic effects of KD in epilepsy (Kesl et al., 2014; Kesl
et al., 2016; Kovac et al., 2013; Masino and Geiger,
2009; Srivastava et al., 2012; Stafstrom et al., 2009;
Stafstrom et al., 2008).
One of the earliest reports of the antiseizure
efficacy of exogenous ketogenic supplementation
was performed in a unique seizure model that uses
hyperbaric hyperoxia (HBO) to reliably induce
epileptic-like (tonic-clonic) seizures in wild-type
rats, a condition known as central nervous system
oxygen toxicity (CNS-OT). A  single oral dose of
the ketone ester BD-AcAc2 induced rapid (within
30 minutes) and sustained (>4 hours) ketosis
(>3mM βHB and >3mM AcAc) and prolonged
the latency to seizure by 574% (D’Agostino et al.,
2013). An elevation in AcAc and acetone appear to
be required for the anticonvulsant effects of keto-
sis. BD elevated blood βHB (>5mM) but did not
elevate AcAc or acetone nor did it affect latency
to seizure. This encouraging response prompted
317
preliminary investigation into preventing or
delaying seizures with ketogenic supplements
in a variety of transgenic rodent and chemical-
induced seizure models. Pentylenetetrazole (PTZ)
is a GABA antagonist and epileptogenic agent
that is used to induce seizures in rodents for pre-
clinical analysis of anticonvulsant therapies. In
a recent study by Coppola and colleagues, the
dosage threshold for seizure induction of PTZ
was assessed in control (water) and ketone ester-
treated rats (Viggiano et  al., 2015). A  single oral
dose (4 g/kg body weight) of BD-AcAc2 elevated
blood βHB (2.7mM) and increased the thresh-
old of PTZ seizure from 122±6 mg/kg to 140±11
mg/kg. Unpublished data from preliminary stud-
ies also suggest an anticonvulsant effect of ketone
ester treatment in the WAG/Rij rat model of
absent epilepsy, the Ube3a m-/p+ mouse model of
Angelman syndrome, and the kainic acid–induced
mouse seizure model.
Glucose Transporter Type 1
Deficiency Syndrome
Glucose transporter type 1 deficiency syndrome
(GLUT1 DS) is a rare genetic disorder caused by
a mutation in the SLC2A1 gene, which encodes
the glucose transporter protein type 1 (GLUT1).
This mutation results in a glucose deficiency in the
brain, which causes seizures as well as cognitive
and physical developmental delay. In a seminal
study in 1967, Cahill and colleagues discovered
that ketones replace glucose as the predominant
energy substrate for the brain during prolonged
fasting and starvation (Cahill, 2006). GLUT1 DS
is treated with the KD, which circumvents the
metabolic blockade by inducing ketosis and is
effective at suppressing seizures and enhancing
cognitive and motor development in most patients.
Maintaining therapeutic levels of ketosis is criti-
cal to support the development of children with
this disorder. Thus, it is clear how an exogenous
ketone supplement could be useful in this patient
population. Triheptanoin is a triglyceride contain-
ing three heptanoates, a 7-carbon fatty acid whose
metabolism produces the 5-carbon ketone bod-
ies β-ketopentanoate and β-hydroxypentanoate.
Because it possesses odd carbon FAs, heptanoate
is metabolized through β-oxidation to produce
propionyl-CoA, which can be subsequently car-
boxylated to succinyl-CoA, replenishing the TCA
cycle via anaplerosis (Borges and Sonnewald,
2012). Triheptanoin has shown therapeutic efficacy
in children and adults with GLUT1 DS, reducing
seizure activity and improving neuropsychological
performance and cerebral metabolic rate (Pascual
318
318 section IV: Ketone-Based Metabolism
et al., 2014). There are ongoing studies investigat-
ing the therapeutic effects of ketone salts and esters
in a GLUT1 DS mouse model and may provide an
alternative or adjunctive treatment to the KD.
Alzheimer’s Disease
In the early stages of AD, the brain exhibits a defi-
ciency in glucose metabolism, which contributes
to the neurodegeneration and progression of the
disorder (Chu and Jiao, 2015; de la Monte, 2012).
Patients with preclinical and clinical AD have
decreased cerebral glucose metabolism as visu-
alized by fluorodeoxyglucose positron emission
tomography (FDG-PET) (Mosconi et al., 2011). It
is thought that this decrease in glucose metabolism
is associated with brain insulin resistance (Talbot
et  al., 2012). As ketones are the principal alter-
native fuel for the brain during fasting or starva-
tion, elevating blood ketone levels in AD patients
would theoretically bypass the deficiencies in glu-
cose metabolism and provide energy to the starv-
ing neurons. Data to support this hypothesis was
recently reported by Cunnane and colleagues, who
demonstrated that while brain glucose uptake is
impaired in AD, ketone uptake remains unaf-
fected. Indeed, the authors conclude that supply-
ing ketones to the brains of AD patients could
restore the brain fuel supply to serve as a potential
therapeutic (Cunnane et  al., 2016). Thus, exog-
enous ketone supplementation could be a useful
tool for supporting cerebral energy metabolism in
this regard.
There is preclinical data to suggest that
ketones could protect against AD development
or slow its progression. The ability of a βHB- and
BD-containing ketone ester to suppress AD pro-
gression was evaluated in a triple transgenic AD
mouse model (3xTgAD) (Kashiwaya et al., 2013).
Presymptomatic mice were fed a diet with approxi-
mately 20% kcal from the ketone ester and com-
pared with isocaloric standard diet–fed control
animals. Ketone ester-treated animals exhibited
less anxiety and improved performance on learn-
ing and memory tests at 4 and 7 months after ini-
tiation of the diet. Immunohistochemical analysis
of the brain revealed that the ketone ester-fed mice
had less Aβ and hyperphosphorylated tau deposi-
tion in the cortex, hippocampus, and amygdala.
In a similar study, 3-hydroxybutyrate methyl ester
(HBME), a derivative of βHB, was assessed for its
therapeutic efficacy in a double transgenic mouse
model of AD (Zhang et  al., 2013a). The HBME-
treated mice (40 mg/kg/d via intragastric gavage)
exhibited improved spatial learning and working
memory and decreased anxiety compared with
control mice. Also, MRI analysis revealed that
HBME prevented the development of asymmetri-
cal ventricle morphology, which was observed
in the untreated AD mice. Following 2.5 months
of treatment, the brains of the animals were ana-
lyzed via immunohistochemistry. The authors also
performed in vitro studies to further investigate
mechanism of protection. The HBME-treated
mice exhibited reduced Aβ plaque deposition in
the cortex and hippocampus. The authors reported
that HBME supported neuronal survival following
glucose deprivation and prevented NaN3-induced
mitochondrial dysfunction by rescuing expres-
sion of the respiratory complexes, reducing ROS
production, and maintaining mitochondrial mem-
brane potential.
In an encouraging case report by Newport
and Veech, supplementation with medium chain
fatty acids (MCFA) and (R)-3-hydroxybutyl (R)-
3-hydroxybutyrate ketone monoester (KME)
was investigated as a potential therapy for AD
(Newport et al., 2015). The patient described was
an APOE4-positive, 63-year-old Caucasian male
with early-onset, sporadic AD diagnosed 12 years
prior to the publication of the report. His disease
rapidly progressed prior to initiation of ketosis
therapy in 2008, characterized by increasingly
severe memory loss and an inability to carry out
normal activities of daily living. His Mini-Mental
State Examination (MMSE) score declined from
23 to 12 between 2004 and 2008, and his MRI
scans revealed diffuse involutional changes of
his frontal and parietal lobes with atrophy of the
amygdala and hippocampus. The patient initially
began ketosis treatment in May 2008 by consum-
ing sources of MCFAs—MCT oil and coconut oil
(CO). His dose and regimen of administration was
optimized, and eventually reached 165 mL of a
4:3 mixture of MCT:CO divided into 3–4 servings
over the course of the day. Within 75 days of treat-
ment, the patient’s MMSE score improved from
12 to 20. The patient continued MCFA treatment
for 20  months, and during that time he exhib-
ited remarkable cognitive and physical improve-
ments. His Alzheimer’s Disease Assessment
Scale-Cognitive (ADAS-Cog) rose 6 points and
his activities of daily living (ADLs) score rose 14
points over that time period. Further, MRI scans
revealed no changes in brain atrophy from June
2008 to April 2010, suggesting stabilization of the
disease process. In 2010, the patient also began
taking the KME (28.7 g KME, three times daily).
The patient exhibited many improvements follow-
ing KME treatment. Within a few days of initiating
KME therapy, the patient regained the ability to
319

Chapter 32: Ketone Supplementation for Health and Disease 319

recite and write out the alphabet, and dress himself.
He began improving in a number of ADLs such as
showering, shaving, and putting away dishes, and
demonstrated improvements in abstract thinking,
insight, and sense of humor. The patient himself
reported feeling happier and more energetic after
beginning treatment with KME. Over time, he
exhibited significant improvements in memory
retrieval and regained the ability to perform com-
plex tasks such as vacuuming and yard work. Both
MCFA and KME treatment significantly elevated
blood ketones, even while the patient continued to
consume a normal diet.
The KME was particularly effective at induc-
ing ketosis, elevating blood ketones up to 7 mM
within 1 hour of administration (Figure 32.2),
a level approximately 5–10 times greater than
is possible with the classical ketogenic diet or
MCFA consumption. The patient’s caregiver, a
physician, noted that his improved cognitive and
physical performance seemed to track his plasma
ketone concentrations, with the greatest improve-
ments seen during peak elevation in blood
ketones. Importantly, there were no adverse effects
observed in the patient over this 2-year study, sug-
gesting that prolonged hyperketonemia is likely
safe. The authors note that not all patients may
respond to such therapy in a similar manner, but
that appropriately designed trials should be con-
ducted to evaluate the percentage of Alzheimer’s
patients responsive to ketone ester therapy.
An oral ketogenic compound and prescrip-
tion medical food called AC-1202 (trade name
Axona) was developed by Accera, Inc., as an AD
therapy. AC-1202 elevates blood ketone levels, as it
contains MCFAs, which are natural ketogenic pre-
cursors. AC-1202 was evaluated in a randomized,
double-blind, placebo-controlled, multicenter trial
in patients with mild to moderate AD (Henderson
et  al., 2009). AC-1202 induced a mild level of
ketosis in patients (up to 0.3–0.4 mM), which
was significantly higher than placebo controls.
AC-1202 treatment induced small improvements
in ADAS-Cog, MMSE, and ADCS-CGIC (AD
Cooperative Study—Clinical Global Impression
of Change) scores compared with placebo in
some subgroups of AD patients tested. In many
of the patient subgroups, AC-1202 did not affect

Responses of whole blood ß-hydroxybutyrate levels to

different doses of ketone monoester (KME)
7
25 grams
35 grams
6 50 grams

5
ß-hydroxybutyrate mM

0
1 2 3 4 5 6
Hours after KME ingestion

FIGURE 32.2 β-Hydroxybutyrate (βHB) concentrations rose to 3 to 7 mM 1 hour after ingestion of ketone monoester
(KME) in three different doses, 25, 35, and 50 g, taken on separate days. The peak levels measured are in the range of
those obtained during adherence to the classical ketogenic diet and are about 10-fold the concentrations achievable by
MCFA administration. The findings suggest that therapeutic ketosis can be maintained throughout the day if KME is
taken every 3 to 4 hours. Precision Xtra Glucose and Ketone Monitoring System (Abbott Diabetes Care, Inc., Alameda,
CA, USA) was used to measure βHB levels in capillary blood samples. Acetoacetate (AcAc) was not measured. This
figure was reprinted with permission from Newport et al. (2015).
320
320 section IV: Ketone-Based Metabolism
performance of APOE4 negative patients on these
tests. The modest improvements observed in this
study may potentially be due to the comparatively
low level of ketosis induced by AC-1202, consider-
ing what is attainable with KME or similar keto-
genic supplements. Furthermore, the lack of effect
observed in APOE4 positive patients could also
be a result of the mild level of ketosis induced, or
alternatively because the study may have lacked
the statistical power necessary to reveal a signifi-
cant effect in this subpopulation (Newport et  al.,
2015). Indeed, the case report by Newport and
colleagues suggests that ketosis can be an effective
therapy for AD in some APOE4 positive patients.
Taken together, these reports clearly demonstrate
the potential utility of exogenous ketone supple-
mentations to confer the therapeutic benefits of
ketosis in a population of patients for which severe
dietary restrictions would be extremely difficult, if
not impossible.
Cancer
Unlike healthy tissues, many cancers do not appear
capable of efficiently metabolizing ketone bod-
ies for energy. Cancer cells often lack expression
of the ketone utilization enzymes, like succinyl-
coenzyme A:3-oxoacid coenzyme A  transferase
(SCOT) (Chang et  al., 2013; Sawai et  al., 2004;
Skinner et  al., 2009; Tisdale and Brennan, 1983).
Abnormal mitochondrial function and impaired
OXPHOS capacity is another ubiquitous feature
of cancers which likely limits the use of ketones,
which are metabolized exclusively within the
mitochondria, as a fuel for cancer. Indeed, a study
on five different brain cancer cell lines concluded
that glioma cells lack the capacity to metabolize
βHB during glucose restriction, unlike healthy
brain cells (Fredericks, and Ramsey, 1978; Maurer
et al., 2011).
The ketogenic diet, fasting, and calorie restric-
tion are dietary regimens that have been shown to
inhibit cancer progression in both preclinical and
clinical studies (Fine et al., 2012; Freedland et al.,
2008; Mavropoulos et  al., 2009; Nebeling et  al.,
1995; Poff et  al., 2013; Rossifanelli et  al., 1991;
Shelton et al., 2010; Wheatley et al., 2008; Zuccoli
et al., 2010). Until recently, the generally accepted
mechanism of action of these therapies was
decreasing glucose availability to the tumor and
suppressing the insulin and IGF signaling path-
ways. However, all three of these therapies also
elevate blood ketones, and recent evidence sug-
gests that ketones themselves may possess inher-
ent anticancer properties. One small clinical trial
studied the effects of a 28-day KD in patients with
late-stage, metastatic cancer (Fine et  al., 2012).
Prior to beginning the dietary intervention, all
patients exhibited progressive disease. Following
1 month of treatment, over 50% of patients showed
stable disease or partial remission. Interestingly,
there was no significant drop in blood glucose in
the patients over the course of the diet, but rather,
patient response was most strongly correlated with
degree of ketosis relative to baseline. In vitro and
preclinical studies have confirmed the hypoth-
esis that ketones are damaging to cancer. In 1979,
Magee et  al. demonstrated that βHB inhibited
proliferation in a dose-dependent manner up to
20 mM in multiple cancer cell lines of varied ori-
gin (Magee et al., 1979). Similarly, both AcAc and
βHB inhibited viability and induced apoptosis in
neuroblastoma cells, but had no effect on control
fibroblasts (Skinner et  al., 2009). Another study
demonstrated that 5 mM βHB slows proliferation
and decreases viability in VM-M3 glioma cells,
even in the presence of excess (25 mM) glucose.
Exogenous ketone supplementation with BD or
the BD-AcAc2 elicited potent anticancer effects in
the VM-M3 model of metastatic cancer, slowing
tumor growth and prolonging survival by 51% and
69%, respectively (Poff et al., 2014). These obser-
vations strongly suggest that exogenous ketone
supplements could be used as an effective adjuvant
therapy for cancer.
There are multiple mechanisms by
which ketones may be damaging to cancer
cells: (1) Cancer is particularly reliant on the gly-
colytic pathway for energy production and bio-
synthesis (Gillies et al., 2008), and βHB inhibits
the first and third enzymatic reactions of glycoly-
sis (Randle et al., 1964). (2) Excess fermentation
in cancer cells causes lactate production and a
subsequent acidification of the tumor microenvi-
ronment, which promotes malignancy. Both lac-
tate and the ketone bodies are transported across
the plasma membrane by the monocarboxylic
transporters family of transporters (Halestrap
and Price, 1999). And βHB has been shown to
inhibit lactate export from isolated rat hepato-
cytes in vitro (Metcalfe et al., 1986). It is possible
that ketones may damage cancer cells by inhibit-
ing lactate export through competitive inhibition
of monocarboxylic transporters, subsequently
inducing intracellular acidification and prevent-
ing the tumor-promoting effects of lactate in
the tumor microenvironment. (3)  Inflammation
and oxidative stress are both known to promote
cancer development and progression (Coussens
and Werb, 2002; Wang and Yi, 2008); thus, the
inhibitory effects of ketone metabolism on both
321
Chapter 32: Ketone Supplementation for Health and Disease
of these pathways could contribute to its antican-
cer efficacy. (4)  Cancers exhibit widespread dif-
ferences in such epigenetic patterns compared
to their normal tissue counterparts, allowing
them to increase expression of oncogenes and
inhibit the expression of tumor suppressor genes
(Hassler and Egger, 2012). Histone deacetylase
inhibitors (HDACI) are being investigated for
their use as antineoplastic agents, and have been
shown to elicit a plethora of anticancer effects in
vitro including activation of apoptosis, induction
of ROS generation and DNA damage, and inhibi-
tion of DNA repair (Bose et al., 2014). As men-
tioned, βHB functions as an endogenous HDACI,
a mechanism that may underlie its potential use
as a cancer treatment. (5) Finally, mitochondrial
transfer studies have demonstrated that healthy
mitochondria act as a tumor suppressor (Seyfried,
2012). The potential for ketone metabolism to
support or enhance mitochondrial health could
also account for its therapeutic effects and its
potential for preventing carcinogenesis.
Insulin Resistance/ Type 2
Diabetes Mellitus
As previously described, ketone ester adminis-
tration simultaneously decreases blood glucose
and blood insulin concentrations (Srivastava
et al., 2012). Thus, less insulin is required to pro-
mote peripheral glucose uptake, exemplifying an
enhancement in insulin sensitivity. These results
suggest a potential therapeutic use of exogenous
ketone supplementation for insulin resistance and
type 2 diabetes mellitus (T2DM). Indeed, exog-
enous ketones qualitatively mimic the acute meta-
bolic effects of insulin (Kashiwaya et  al., 1997).
It is known that insulin activates pyruvate dehy-
drogenase (PDH) to increase the production of
acetyl CoA. The administration of 5 mM ketones
mimicked this effect, increasing acetyl CoA pro-
duction 15-fold in the glucose-perfused isolated
rat heart (Kashiwaya et al., 1994). Furthermore, in
this model, ketones and insulin increased cardiac
hydraulic efficiency to a similar degree, approxi-
mately 25%–35% (Kashiwaya et  al., 1994). In
another study by Srivastava et al., mice that were
fed a diet formulated with ketone ester (30% kcal)
had a 73% increase in the Quantitative Insulin-
Sensitivity Check Index (QUICKI), a surrogate
marker of insulin sensitivity, compared with con-
trol, calorie-matched mice (Srivastava et al., 2012).
Fasting plasma glucose levels were not altered in
these mice, but fasting plasma insulin levels were
reduced by approximately 85% in the KE-fed mice
compared with controls. The authors therefore
321
hypothesized that ketones could be therapeutic by
correcting metabolic defects of acute insulin defi-
ciency or in the insulin-resistant state (Kashiwaya
et al., 1997).
The HDACI activity of βHB could also be
beneficial in T2DM by altering the direct regula-
tion of HDAC-dependent glucose metabolism
and by inducing resistance to oxidative stress. The
HDACs regulate the expression of genes encoding
many metabolic enzymes, and HDAC3 knock-
out animals exhibit reduced glucose and insu-
lin. Suberoylanilide hydroxamic acid (SAHA), a
class I HDAC inhibitor, has been shown to improve
insulin sensitivity and increase oxidative metabo-
lism and metabolic rate in a mouse model of diabe-
tes (Galmozzi et al., 2013). Butyrate, a short-chain
fatty acid that is structurally similar to βHB and
also acts as a HDACI, lowers blood glucose and
insulin levels and improves glucose tolerance and
respiratory efficiency (Gao et al., 2009). The vascu-
lar dysfunction in T2DM is thought to be caused
by oxidative stress (Giacco and Brownlee, 2010).
And HDAC inhibition prevents renal damage in
mouse models of diabetic nephropathy through
modulation of redox mechanisms (Advani et  al.,
2011). Therefore, βHB suppression of oxidative
stress through HDAC inhibition may help restore
insulin sensitivity and manage complications of
diabetes.
Weight Loss
The KD is hypothesized to induce weight loss by
reducing appetite through the satiety effect of
protein and ketone bodies, reducing de novo lipo-
genesis and increasing lipolysis, and enhancing
metabolic efficiency with fat and ketone metabo-
lism (Paoli, 2014). However, the KD can be difficult
to maintain long-term for many individuals, and
many people regain weight rapidly upon returning
to a standard diet (Paoli, 2014). Recently, prelimi-
nary studies have shown that exogenous ketone
supplementation can also induce weight loss.
It should be noted that ketone supplements are
sources of calories, with each ketone supplement
providing on average 5–8 kcal/gram ingested;
therefore, patients would need to decrease dietary
caloric intake in order to prevent weight gain. The
reported satiating effect of ketosis may aid in this
adjustment.
The administration of both βHB and BD has
been shown to decrease food intake in rats and
pigmy goats (Arase et  al., 1988; Carpenter and
Grossman, 1983; Davis et  al., 1981; Langhans
et al., 1983; Rossi et al., 2000). Similarly, it is sug-
gested that MCTs increase satiety, resulting in
322
322 section IV: Ketone-Based Metabolism
reduced food intake and weight loss as a con-
sequence of their rapid oxidation into ketone
bodies (Krotkiewski, 2001; Poppitt et  al., 2010;
Wymelbeke and Himaya, 1998). The MCTs may
further counteract fat deposition in adipocytes by
increasing thermogenesis (Dulloo, 2011). Several
studies in animals and humans have revealed
increased energy expenditure and lipid oxida-
tion with MCTs compared with LCTs (Baba et al.,
1982; Bach and Babayan, 1982; Crozier et  al.,
1987; Dulloo et  al., 1996; Ferreira et  al., 2014;
Karen and Welma, 2015; Scalfi et al., 1991; Seaton
et al., 1986; St-Onge et al., 2003a, 2003b). Ketone
esters have also been shown to affect weight in
mice, rats, and humans. Mice administered a
diet supplemented with ketone ester resulted in
reduced voluntary food intake, increased insulin
sensitivity, increased resting energy expenditure,
and increased brown fat activity, demonstrat-
ing the potential utility of the ketone ester as an
antiobesity compound (Srivastava et  al., 2012).
In another study, mice fed a ketone ester diet
maintained a body weight of approximately 10%–
12% less than control, standard diet–fed animals
(Kashiwaya et  al., 2013). Kashiwaya et  al. sug-
gested that the ketone ester–supplemented diet
may benefit obesity as it decreases brain malonyl
CoA, an important metabolic mediator of appetite
(Kashiwaya et al., 2010). In a 28-day oral gavage
study, all five ketogenic supplements tested (BD,
BD-AcAc2, MCT, BMS, BMS+MCT) caused a sig-
nificant reduction in weight gain compared with
control animals (Kesl et al., 2016). Similarly, a 15-
week chronic feeding study in which BD-AcAc2,
BMS, or BMS+MCT replaced approximately 20%
of the diet by weight, fed ad libitum, led to sig-
nificant reduction in weight gain compared with
control animals (unpublished data). In a 7-day
study, eight humans with T2DM consumed three
KE-supplemented drinks a day in conjunction
with their normal diet. The group lost an aver-
age of 2% of their body weight (Hashim and
VanItallie, 2014).
CONCLUSION
Exogenous ketone supplements are being devel-
oped as an alternative or adjuvant method of
inducing therapeutic ketosis aside from the clas-
sic ketogenic diet. Emerging evidence has dem-
onstrated that these novel compounds have the
potential to offer benefits for both healthy and dis-
eased individuals alike. It is likely that most, if not
all, of the conditions that are known to benefit from
the KD would receive some benefit from exog-
enous ketone supplementation by elevating blood
ketones and lowering blood glucose. Importantly,
ketone supplementation provides a tool for achiev-
ing ketosis in patients who are unable, unwilling,
or uninterested in consuming a low carbohydrate
or ketogenic diet. It may also help circumvent
some of the difficulties associated with KD therapy,
as it allows for a rapid dose-dependent induction
of ketosis, which can be sustained with prolonged
consumption and monitored precisely with com-
mercially available technologies (e.g., blood ketone
meters). Simultaneously, it could provide patients
with the opportunity to reap the benefits of keto-
sis without the practical and social difficulties of a
highly restrictive diet. Further research is needed
to fully investigate the clinical utility and feasibility
of exogenous ketone supplements as a method of
inducing therapeutic ketosis.
REFERENCES
Advani, A., Huang, Q., Thai, K., Advani, S.L., White,
K.E., Kelly, D.J., Yuen, D.A., Connelly, K.A.,
Marsden, P.A., and Gilbert, R.E. (2011). Long-
term administration of the histone deacetylase
inhibitor vorinostat attenuates renal injury in
experimental diabetes through an endothelial
nitric oxide synthase-dependent mechanism. Am
J Pathol 178, 2205–2214.
Arase, K., Fisler, J.S., Shargill, N.S., York, D.A., and
Bray, G.A. (1988). Intracerebroventricular infu-
sions of 3-OHB and insulin in a rat model of
dietary obesity. Am J Physiol 255, 81.
Baba, N., Bracco, E.F., and Hashim, S.A. (1982).
Enhanced thermogenesis and diminished depo-
sition of fat in response to overfeeding with diet
containing medium chain triglyceride. Am J Clin
Apr;35(4):678–82. .
Babayan, V.K. (1987). Medium chain triglycerides and
structured lipids. Lipids Jun;22(6):417–20.
Bach, A.C., and Babayan, V.K. (1982). Medium-chain
triglycerides:  an update. Am J Clin Nutr 36,
950–962.
Bell, S.J., Bradley, D., Forse, R.A., and Bistrian, B.R.
(1997). The new dietary fats in health and disease.
J Am Diet Assoc. 97, 280–286.
Birkhahn, R.H., and Border, J.R. (1978). Intravenous
feeding of the rat with short chain fatty acid esters.
II. Monoacetoacetin. Am J Clin Nutr 31, 436–441.
Birkhahn, R.H., McMenamy, R.H., and Border, J.R.
(1977). Intravenous feeding of the rat with short
chain fatty acid esters:  I. Glycerol monobutyrate.
Am J Clin Nutr 30, 2078–2082.
Birkhahn, R.H., McMenamy, R.H., and Border, J.R.
(1979). Monoglyceryl acetoacetate: a ketone body-
carbohydrate substrate for parenteral feeding of
the rat. J Nutr 109, 1168–1174.
Borges, K., and Sonnewald, U. (2012). Triheptanoin: a
medium chain triglyceride with odd chain fatty
323
Chapter 32: Ketone Supplementation for Health and Disease
acids: a new anaplerotic anticonvulsant treatment?
Epilepsy Res 100, 239–244.
Bornfeldt, K.E., and Tabas, I. (2011). Insulin resis-
tance, hyperglycemia, and atherosclerosis. Cell
Metabol 14, 575–585.
Bose, P., Dai, Y., and Grant, S. (2014). Histone deacetylase
inhibitor (HDACI) mechanisms of action:  emerg-
ing insights. Pharmacol Ther Sep;143(3):323–36.
Bough, K.J., and Rho, J.M. (2007). Anticonvulsant
mechanisms of the ketogenic diet. Epilepsia
48, 43–58.
Brunengraber, H. (1997). Potential of ketone body
esters for parenteral and oral nutrition. Nutrition
(Burbank, Los Angeles County, Calif) 13, 233–235.
Budavari, S.E. (1989). The Merck index: An encyclope-
dia of chemicals, drugs, and biologicals. Rahway,
N.J: Merck.
Cahill, G. (2006). Fuel metabolism in starvation.
Annual review of nutrition 26, 1–22.
Cahill, G.F., Jr., and Veech, R.L. (2003). Ketoacids?
Good medicine? Trans Am Clin Climatol Assoc
114, 149–161; discussion 162–143.
Carpenter, R.G., and Grossman, S.P. (1983). Plasma fat
metabolites and hunger. Physiol Behav 30, 57–63.
Cassandra, E.F., Stephen, D.P., Maria Luz, F., Erin,
E.Q., Richard, J.W., Doug, M.B., William, J.K.,
Richard, D.F., and Jeff, S.V. (2007). Comparison of
low fat and low carbohydrate diets on circulating
fatty acid composition and markers of inflamma-
tion. Lipids 43(1):65–77
Chang, H.T., Olson, L.K., and Schwartz, K.A. (2013).
Ketolytic and glycolytic enzymatic expression pro-
files in malignant gliomas:  implication for keto-
genic diet therapy. Nutr Metab (Lond) 10, 47.
Chang, P., Zuckermann, A.M., Williams, S., Close, A.J.,
Cano-Jaimez, M., McEvoy, J.P., Spencer, J., Walker,
M.C., and Williams, R.S. (2015). Seizure control by
derivatives of medium chain fatty acids associated
with the ketogenic diet show novel branching-
point structure for enhanced potency. J Pharmacol
Exp Ther 352, 43–52.
Chu, C.B., and Jiao, L.D. (2015). (R)-3-oxobutyl 3-
hydroxybutanoate (OBHB) induces hyperketone-
mia in Alzheimer’s disease. Int J Clin Exp Med 8,
7684–7688.
Ciraolo, S.T., Previs, S.F., Fernandez, C.A., Agarwal,
K.C., David, F., Koshy, J., Lucas, D., Tammaro, A.,
Stevens, M.P., and Tserng, K.Y. (1995). Model of
extreme hypoglycemia in dogs made ketotic with
(R,S)-1,3-butanediol acetoacetate esters. Am J
Physiol 269, 75.
Clarke, K., Tchabanenko, K., Pawlosky, R., Carter,
E., Knight, N., Murray, A., Cochlin, L., King, M.,
Wong, A., Roberts, A., et  al. (2012a). Oral 28-
day and developmental toxicity studies of (R)-
3-hydroxybutyl (R)-3-hydroxybutyrate. Regul
Toxicol Pharmacol: RTP 63, 196–208.
Clarke, K., Tchabanenko, K., Pawlosky, R., Carter, E.,
Todd King, M., Musa-Veloso, K., Ho., M., Roberts,
323
A., Robertson, J., Vanitallie, T., and Veech, R.
(2012b). Kinetics, safety and tolerability of (R)-
3-hydroxybutyl (R)-3-hydroxybutyrate in healthy
adult subjects. Regul Toxicol Pharmacol: RTP 63,
401–408.
Coussens, L.M., and Werb, Z. (2002). Inflammation
and cancer. Nature 420, 860–867.
Crozier, G., Bois-Joyeux, B., Chanez, M., Girard, J.,
and Peret, J. (1987). Metabolic effects induced by
long-term feeding of medium-chain triglycerides
in the rat. Metabolism.1987 Aug;36(8):807–14
Cunnane, S.C., Courchesne-Loyer, A., St-Pierre, V.,
Vandenberghe, C., Pierotti, T., Fortier, M., Croteau,
E., and Castellano, C.A. (2016). Can ketones com-
pensate for deteriorating brain glucose uptake
during aging? Implications for the risk and treat-
ment of Alzheimer’s disease. Ann NY Acad Sci.
D’Agostino, D., Arnold, P., and Kesl, S.  (2014).
Compositions and methods for producing ele-
vated and sustained ketosis. In: Compositions and
Methods for Producing Elevated and Sustained
Ketosis: US Patent: WO2014153416 A1; University
of South Florida):  http://www.google.com/pat-
ents/WO2014153416A1?cl=en.
D’Agostino, D.P., Pilla, R., Held, H.E., Landon, C.S.,
Puchowicz, M., Brunengraber, H., Ari, C., Arnold,
P., and Dean, J.B. (2013). Therapeutic ketosis with
ketone ester delays central nervous system oxygen
toxicity seizures in rats. Am J Physiol Regul Integr
Comp Physiol 304, R829–836.
Davis, R.J., Brand, M.D., and Martin, B.R. (1981).
The effect of insulin on plasma-membrane and
mitochondrial-membrane potentials in isolated
fat-cells. Biochem J 196, 133–147.
de Carvalho Vidigal, F., Guedes Cocate, P., Goncalves
Pereira, L., and de Cassia Goncalves Alfenas, R.
(2012). The role of hyperglycemia in the induction
of oxidative stress and inflammatory process. Nutr
Hosp 27, 1391–1398.
de la Monte, S.M. (2012). Brain insulin resistance and
deficiency as therapeutic targets in Alzheimer’s
disease. Curr Alzheimer Res 9, 35–66.
Desrochers, S., David, F., Garneau, M., Jetté, M., and
Brunengraber, H. (1992). Metabolism of R- and
S-1,3-butanediol in perfused livers from meal-fed
and starved rats. Biochem J 285 (Pt 2), 647–653.
Desrochers, S., Dubreuil, P., Brunet, J., Jetté, M., David,
F., Landau, B.R., and Brunengraber, H. (1995a).
Metabolism of (R,S)-1,3-butanediol acetoacetate
esters, potential parenteral and enteral nutrients in
conscious pigs. Am J Physiol 268, 7.
Desrochers, S., Quinze, K., Dugas, H., Dubreuil, P.,
Bomont, C., David, F., Agarwal, K.C., Kumar, A.,
Soloviev, M.V., Powers, L., et al. (1995b). R,S-1,3-
butanediol acetoacetate esters, potential alternates
to lipid emulsions for total parenteral-nutrition. J
Nutr Biochem 6, 111–118.
Drackley, J.K., Richard, M.J., and Young, J.W. (1990).
In vitro production of beta-hydroxybutyrate from
324
324 section IV: Ketone-Based Metabolism
1,3-butanediol by bovine liver, rumen mucosa,
and kidney. J Dairy Sci 73, 679–682.
Duewell, P., Kono, H., Rayner, K.J., Sirois, C.M.,
Vladimer, G., Bauernfeind, F.G., Abela, G.S.,
Franchi, L., Nunez, G., Schnurr, M., et al. (2010).
NLRP3 inflammasomes are required for ath-
erogenesis and activated by cholesterol crystals.
Nature 464, 1357–1361.
Dulloo, A.G. (2011). The search for compounds that
stimulate thermogenesis in obesity manage-
ment:  from pharmaceuticals to functional food
ingredients. Obes Rev 12, 866–883.
Dulloo, A.G., Fathi, M., and Mensi, N. (1996). Twenty-
four-hour energy expenditure and urinary
catecholamines of humans consuming low-to-
moderate amounts of medium-chain triglycer-
ides: a dose-response. Eur J Clin 50(3):152–158.
Dymsza, H.A. (1975). Nutritional application and impli-
cation of 1,3-butanediol. Fed Proc 34, 2167–2170.
Elizabeth, G.N., Hannah, C., Ruby, H.S., Margaret,
S.L., Nicole, E., Georgiana, F., Andrea, W., and
Cross, J.H. (2009). A randomized trial of classical
and medium-chain triglyceride ketogenic diets in
the treatment of childhood epilepsy. Epilepsia 50,
1109–1117.
Falbe, J., Bahrmann, H., and Lipps, W. (1985). Alcohols,
aliphatic. Ullmann’s Encyclopedia.http://onlineli-
brary.wiley.com/doi/10.1002/14356007.a01_279/
abstract
Ferreira, L., Lisenko, K., and Barros, B. (2014).
Influence of medium‐chain triglycerides on con-
sumption and weight gain in rats:  a systematic
review. J Anim Physiol Anim Nutr (Berl)
Fine, E., Segal-Isaacson, C., Feinman, R., Herszkopf, S.,
Romano, M., Tomuta, N., Bontempo, A., Negassa,
A., and Sparano, J. (2012). Targeting insulin inhi-
bition as a metabolic therapy in advanced can-
cer: a pilot safety and feasibility dietary trial in 10
patients. Nutrition 28, 1028–1035.
Fredericks, M., and Ramsey, R.B. (1978). 3-Oxo acid
coenzyme A  transferase activity in brain and
tumors of the nervous system. J Neurochem 31,
1529–1531.
Freedland, S., Mavropoulos, J., Wang, A., Darshan,
M., Demark-Wahnefried, W., Aronson, W.,
Cohen, P., Hwang, D., Peterson, B., et  al. (2008).
Carbohydrate restriction, prostate cancer growth,
and the insulin-like growth factor axis. The
Prostate 68, 11–19.
Galmozzi, A., Mitro, N., Ferrari, A., Gers, E., Gilardi, F.,
Godio, C., Cermenati, G., Gualerzi, A., Donetti, E.,
Rotili, D., et al. (2013). Inhibition of class I histone
deacetylases unveils a mitochondrial signature and
enhances oxidative metabolism in skeletal muscle
and adipose tissue. Diabetes 62, 732–742.
Gao, Z., Yin, J., Zhang, J., Ward, R.E., Martin, R.J.,
Lefevre, M., Cefalu, W.T., and Ye, J. (2009). Butyrate
improves insulin sensitivity and increases energy
expenditure in mice. Diabetes 58, 1509–1517.
Giacco, F., and Brownlee, M. (2010). Oxidative
stress and diabetic complications. Circ Res 107,
1058–1070.
Gillies, R., Robey, I., and Gatenby, R. (2008). Causes
and consequences of increased glucose metabo-
lism of cancers. J Nucl Med 49 (Suppl 2): 24S–42S.
Halestrap, A., and Price, N. (1999). The proton-
linked monocarboxylate transporter (MCT) fam-
ily: structure, function and regulation. Biochem J
343 Pt 2, 281–380.
Hashim, S.A., and Tantibhedyangkul, P. (1987).
Medium chain triglyceride in early life: effects on
growth of adipose tissue. Lipids 22(6):429–434
Hashim SA and VanItallie, T.B. (2014). Ketone body
therapy: from the ketogenic diet to the oral admin-
istration of ketone ester. J Lipid Res 55, 1818–1826.
Hassler, M.R., and Egger, G. (2012). Epigenomics of
cancer:  emerging new concepts. Biochimie 94,
2219–2230.
Henderson, S.T., Vogel, J.L., Barr, L.J., Garvin, F.,
Jones, J.J., and Costantini, L.C. (2009). Study of
the ketogenic agent AC-1202 in mild to moderate
Alzheimer’s disease:  a randomized, double-blind,
placebo-controlled, multicenter trial. Nutr Metab
(Lond) 6, 31.
Heneka, M.T., Kummer, M.P., Stutz, A., Delekate, A.,
Schwartz, S., Vieira-Saecker, A., Griep, A., Axt,
D., Remus, A., Tzeng, T.C., et al. (2013). NLRP3 is
activated in Alzheimer’s disease and contributes to
pathology in APP/PS1 mice. Nature 493, 674–678.
Hess, F.G., Cox, G.E., Bailey, D.E., Parent, R.A., and
Becci, P.J. (1981). Reproduction and teratology
study of 1,3-butanediol in rats. J Appl Toxicol 1,
202–209.
Hove JLK, Grünewald, S., Jaeken, J., and Demaerel,
P. (2003). D, L-3-hydroxybutyrate treatment of
multiple acyl-CoA dehydrogenase deficiency
(MADD). Lancet 361(9367):1433–5
Huttenlocher, P.R. (1976). Ketonemia and sei-
zures: metabolic and anticonvulsant effects of two
ketogenic diets in childhood epilepsy. Pediat Res
10, 536–540.
Huttenlocher, P.R., Wilbourn, A.J., and Signore, J.M.
(1971). Medium-chain triglycerides as a therapy
for intractable childhood epilepsy. Neurology 21,
1097–1103.
Jarrett, S.G., Milder, J.B., Liang, L.P., and Patel, M.
(2008). The ketogenic diet increases mitochondrial
glutathione levels. J Neurochem 106, 1044–1051.
Karen, M., and Welma, S. (2015). Effects of medium-
chain triglycerides on weight loss and body com-
position: a meta-analysis of randomized controlled
trials. J Acad Nutr Diet 115.
Kashiwaya, Y., Sato, K., Tsuchiya, N., Thomas, S., Fell,
D.A., Veech, R.L., and Passonneau, J.V. (1994).
Control of glucose utilization in working perfused
rat heart. J Biol Chem 269, 25502–25514.
Kashiwaya, Y., King, M.T., and Veech, R.L. (1997).
Substrate signaling by insulin:  a ketone bodies
325
Chapter 32: Ketone Supplementation for Health and Disease
ratio mimics insulin action in heart. Am J Cardiol
80, 50A–64A.
Kashiwaya, Y., Pawlosky, R., Markis, W., King, M.T.,
Bergman, C., Srivastava, S., Murray, A., Clarke,
K., and Veech, R.L. (2010). A ketone ester diet
increases brain malonyl-CoA and Uncoupling pro-
teins 4 and 5 while decreasing food intake in the
normal Wistar Rat. J Biol Chem 285, 25950–25956.
Kashiwaya, Y., Bergman, C., Lee, J.H., Wan, R., King,
M.T., Mughal, M.R., Okun, E., Clarke, K., Mattson,
M.P., and Veech, R.L. (2013). A ketone ester diet
exhibits anxiolytic and cognition-sparing proper-
ties, and lessens amyloid and tau pathologies in a
mouse model of Alzheimer’s disease. Neurobiol
Aging 34, 1530–1539.
Kesl, S., Poff, A., Ward, N., Fiorelli, T., Ari, C.,
D’Agostino, D. (2014). Methods of Sustaining
Dietary Ketosis in Sprague-Dawley Rats. (San
Diego, CA:  Federation of the American Societies
for Experimental Biology).
Kesl, S., Poff, A., Ward, N., Fiorelli, T., Ari, C., Van
Putten, A., Sherwood, J., Arnold, P., and D’Agostino,
D. (2015a). Effect of Sustaining Dietary Ketosis
on the Hippocampal and Serum Metabolome of
Sprague-Dawley Rats. (Boston, MA: Federation of
American Societies of Experimental Biology).
Kesl, S.L., Poff, A.M., Ward, N.P., Fiorelli, T.N., Ari,
C., Van Putten, A.J., Sherwood, J.W., Arnold, P.,
and D’Agostino, D.P. (2016). Effects of exogenous
ketone supplementation on blood ketone, glucose,
triglyceride, and lipoprotein levels in Sprague-
Dawley rats. Nutr Metab (Lond) 13, 9.
Kim do, Y., Davis, L.M., Sullivan, P.G., Maalouf, M.,
Simeone, T.A., van Brederode, J., and Rho, J.M.
(2007). Ketone bodies are protective against oxi-
dative stress in neocortical neurons. J Neurochem
101, 1316–1326.
Klein, P., Tyrlikova, I., and Mathews, G.C. (2014).
Dietary treatment in adults with refractory epi-
lepsy: a review. Neurology 83, 1978–1985.
Koji, N., and Teruyoshi, Y. (2010). Medium-chain fatty
acids:  Functional lipids for the prevention and
treatment of the metabolic syndrome. Pharmacol
Res. 61(3), 208–212
Kovac, S., Abramov, A.Y., and Walker, M.C. (2013).
Energy depletion in seizures: anaplerosis as a strat-
egy for future therapies. Neuropharmacology 69,
96–104.
Krotkiewski, M. (2001). Value of VLCD supplementa-
tion with medium chain triglycerides. Int J Obes
Relat Metab Disord. 25(9), 1393–1400.
Mosconi, L., McHugh, P., de Leon, M.J., Berti, V., and
Pupi, A. (2011). A Tale of Two Tracers:  Glucose
Metabolism and Amyloid Positron Emission
Tomography Imaging in Alzheimer’s Disease.
(Amsterdam: IOS Press), 219–234.
Laakso, M., and Kuusisto, J. (2014). Insulin resistance
and hyperglycaemia in cardiovascular disease
development. Nat Rev Endocrinol 10, 293–302.
325
Langhans, W., Wiesenreiter, F., and Scharrer, E.
(1983). Different effects of subcutaneous D,L-3-
hydroxybutyrate and acetoacetate injections on
food intake in rats. Physiol Behav 31, 483–486.
Lardy, H.A., and Phillips, P.H. (1945). Studies of fat
and carbohydrate oxidation in mammalian sper-
matozoa. Arch Biochem 6, 53–61.
Levy, R.G., Cooper, P.N., and Giri, P. (2011). Ketogenic
diet and other dietary treatments for epilepsy.
Cochrane Database Syst Rev 3.
Li, H.-F.F., Zou, Y., and Ding, G. (2013). Therapeutic
success of the ketogenic diet as a treatment option
for epilepsy: a meta-analysis. Iranian J Pediatr 23,
613–620.
Maalouf, M., Sullivan, P.G., Davis, L., Kim, D.Y., and
Rho, J.M. (2007). Ketones inhibit mitochondrial
production of reactive oxygen species production
following glutamate excitotoxicity by increasing
NADH oxidation. Neuroscience 145, 256–264.
Magee, B.A., Potezny, N., Rofe, A.M., and Conyers,
R.A. (1979). The inhibition of malignant cell
growth by ketone bodies. Aust J Exp Biol Med Sci
57, 529–539.
Mak, S.C., Chi, C.S., and Wan, C.J. (1999). Clinical expe-
rience of ketogenic diet on children with refrac-
tory epilepsy. Acta paediatrica Taiwanica = Taiwan
er ke yi xue hui za zhi 40, 97–100.
Martinon, F., Petrilli, V., Mayor, A., Tardivel, A., and
Tschopp, J. (2006). Gout-associated uric acid crys-
tals activate the NALP3 inflammasome. Nature
440, 237–241.
Masino, S.A., and Geiger, J.D. (2009). The ketogenic
diet and epilepsy:  is adenosine the missing link?
Epilepsia 50, 332–333.
Maurer, G., Brucker, D., Bähr, O., Harter, P., Hattingen,
E., Walenta, S., Mueller-Klieser, W., Steinbach, J.,
and Rieger, J. (2011). Differential utilization of
ketone bodies by neurons and glioma cell lines: a
rationale for ketogenic diet as experimental glioma
therapy. BMC Cancer 11, 315.
Mavropoulos, J., Buschemeyer, W., Tewari, A.,
Rokhfeld, D., Pollak, M., Zhao, Y., Febbo, P.,
Cohen, P., Hwang, D., Devi, G., et al. (2009). The
effects of varying dietary carbohydrate and fat con-
tent on survival in a murine LNCaP prostate can-
cer xenograft model. Cancer Prev Res 2, 557–565.
McHenry LC, Jr. (1966). Cerebral blood flow. N Engl J
Med 274, 82–91.
Metcalfe, H.K., Monson, J.P., Welch, S.G., and Cohen,
R.D. (1986). Inhibition of lactate removal by
ketone bodies in rat liver:  evidence for a quanti-
tatively important role of the plasma membrane
lactate transporter in lactate metabolism. J Clin
Invest 78, 743–747.
Nebeling, L.C., Miraldi, F., Shurin, S.B., and Lerner, E.
(1995). Effects of a ketogenic diet on tumor metab-
olism and nutritional status in pediatric oncology
patients:  two case reports. J Am Coll Nutr 14,
202–208.
326
326 section IV: Ketone-Based Metabolism
Newport, M.T., VanItallie, T.B., Kashiwaya, Y., King,
M.T., and Veech, R.L. (2015). A new way to pro-
duce hyperketonemia: use of ketone ester in a case
of Alzheimer’s disease. Alzheimers Dement 11,
99–103.
Opitz, K. (1958). Über die glykogenbildende Wirkung
von Glykolen. Naunyn Schmiedebergs Arch
Pharmacol 234(5), 448–454
Paoli, A. (2014). Ketogenic diet for obesity: friend or
foe? Int J Environ Res Publ Health 11, 2092–2107.
Pascual, J.M., Liu, P., Mao, D., Kelly, D.I., Hernandez,
A., Sheng, M., Good, L.B., Ma, Q., Marin-Valencia,
I., Zhang, X., et  al. (2014). Triheptanoin for glu-
cose transporter type I deficiency (G1D): modula-
tion of human ictogenesis, cerebral metabolic rate,
and cognitive indices by a food supplement. JAMA
Neurol 71, 1255–1265.
Poff, A.M., Ari, C., Seyfried, T.N., and D’Agostino, D.P.
(2013). The ketogenic diet and hyperbaric oxygen
therapy prolong survival in mice with systemic
metastatic cancer. PLoS One 8: e65522.
Poff, A.M., Ari, C., Arnold, P., Seyfried, T.N., and
D’Agostino, D.P. (2014). Ketone supplementation
decreases tumor cell viability and prolongs sur-
vival of mice with metastatic cancer. Int J Cancer
1711–1720.
Poff, A., Kesl, S.,Ward, N., D’Agostino, D.  (2016).
Metabolic effects of exogenous ketone
supplementation—an alternative or adjuvant to
the ketogenic diet as a cancer therapy? Keystone
Symposia—New Frontiers in Tumor Metabolism,
Banff, Alberta, Canada.
Poppitt, S.D., Strik, C.M., MacGibbon AKH, McArdle,
B.H., Budgett, S.C., and McGill, A.T. (2010). Fatty
acid chain length, postprandial satiety and food
intake in lean men. Physiol Behav 101, 161–167.
Puchowicz, M.A., Smith, C.L., Bomont, C., Koshy, J.,
David, F., and Brunengraber, H. (2000). Dog model
of therapeutic ketosis induced by oral administra-
tion of R,S-1,3-butanediol diacetoacetate. J Nutr
Biochem 11, 281–287.
Randle, P.J., Newsholme, E.A., and Garland, P.B.
(1964). Regulation of glucose uptake by muscle.
8.  Effects of fatty acids, ketone bodies and pyru-
vate, and of alloxan-diabetes and starvation, on the
uptake and metabolic fate of glucose in rat heart
and diaphragm muscles. Biochem J 93, 652–665.
Rossi, R., Dörig, S., Del Prete, E., and Scharrer, E.
(2000). Suppression of feed intake after paren-
teral administration of D-beta-hydroxybutyrate in
pygmy goats. J Vet Med A, Physiol, Pathol, Clinic
Med 47, 9–16.
Rossifanelli, F., Franchi, F., Mulieri, M., Cangiano, C.,
Cascino, A., Ceci, F., Muscaritoli, M., Seminara, P.,
and Bonomo, L. (1991). Effect of energy substrate
manipulation on tumor-cell proliferation in paren-
terally fed cancer-patients. Clin Nutr 10, 228–232.
Ruby, H.S., Jane, E., Bower, B.D., and Aynsley-Green,
A. (1989). Ketogenic diets in the treatment of
epilepsy:  short-term clinical effects. Dev Med
Child Neurol 31, 145–151.
Ryu, T.Y., Park, J., and Scherer, P.E. (2014).
Hyperglycemia as a risk factor for cancer progres-
sion. Diabetes Metab J 38, 330–336.
Sawai, M., Yashiro, M., Nishiguchi, Y., Ohira, M., and
Hirakawa, K. (2004). Growth-inhibitory effects
of the ketone body, monoacetoacetin, on human
gastric cancer cells with succinyl-CoA, 3-oxoacid
CoA-transferase (SCOT) deficiency. Anticancer
Res 24, 2213–2217.
Scala RA and Paynter, O.E. (1967). Chronic oral tox-
icity of 1, 3-butanediol. Toxicol Appl Pharmacol
10(1),160–164
Scalfi, L., Coltorti, A., and Contaldo, F. (1991).
Postprandial thermogenesis in lean and obese
subjects after meals supplemented with medium-
chain and long-chain triglycerides. Am J Clin Nutr
53(5),1130–1133.
Seaton, T.B., Welle, S.L., and Warenko, M.K. (1986).
Thermic effect of medium-chain and long-chain
triglycerides in man. Am J Clin Nutr. 44(5),
630–634
Seyfried, T.N. (2012). Cancer as a Metabolic
Disease:  On the Origin, Management, and
Prevention of Cancer. (Hoboken, NJ: Wiley).
Sharman, M., and Volek, J. (2004). Weight loss leads
to reductions in inflammatory biomarkers after a
very-low-carbohydrate diet and a low-fat diet in
overweight men. Clinical Science 107, 365–369.
Shelton, L.M., Huysentruyt, L.C., Mukherjee, P., and
Seyfried, T.N. (2010). Calorie restriction as an
anti-invasive therapy for malignant brain cancer
in the VM mouse. ASN Neuro 2: e00038.
Shimazu, T., Hirschey, M.D., Newman, J., He, W.,
Shirakawa, K., Le Moan, N., Grueter, C.A.,
Lim, H., Saunders, L.R., Stevens, R.D., et  al.
(2013). Suppression of oxidative stress by beta-
hydroxybutyrate, an endogenous histone deacety-
lase inhibitor. Science 339, 211–214.
Sills, M.A., Forsythe, W.I., Haidukewych, D.,
MacDonald, A., and Robinson, M. (1986). The
medium chain triglyceride diet and intractable
epilepsy. Archives of Disease in Childhood 61,
1168–1172.
Skinner, R., Trujillo, A., Ma, X., and Beierle, E. (2009).
Ketone bodies inhibit the viability of human neu-
roblastoma cells. Journal of Pediatric Surgery
44, 212.
Srivastava, S., Kashiwaya, Y., King, M., Baxa, U., Tam, J.,
Niu, G., Chen, X., Clarke, K., and Veech, R. (2012).
Mitochondrial biogenesis and increased uncou-
pling protein 1 in brown adipose tissue of mice fed
a ketone ester diet. FASEB Journal: Official publi-
cation of the Federation of American Societies for
Experimental Biology 26, 2351–2362.
St-Onge, M.P., Bourque, C., Jones PJH, and Ross, R.
(2003a). Medium-versus long-chain triglycerides
for 27  days increases fat oxidation and energy
327
Chapter 32: Ketone Supplementation for Health and Disease
expenditure without resulting in changes in body
composition in overweight … . Int J Obes Relat
Metab Disord 27(1), 95–102.
St‐Onge, M.P., Ross, R., and Parsons, W.D. (2003b).
Medium‐chain triglycerides increase energy
expenditure and decrease adiposity in overweight
men. Obes Res 11(3), 395–402.
Stafstrom, C.E., Roopra, A., and Sutula, T.P. (2008).
Seizure suppression via glycolysis inhibition with
2-deoxy-D-glucose (2DG). Epilepsia 49(Suppl 8),
97–100.
Stafstrom, C.E., Ockuly, J.C., Murphree, L., Valley, M.T.,
Roopra, A., and Sutula, T.P. (2009). Anticonvulsant
and antiepileptic actions of 2-deoxy-D-glucose in
epilepsy models. Ann Neurol 65, 435–447.
Sylvain, D., Khadijah, Q., Hermann, D., Pascal, D.,
Catherine, B., France, D., Kamlesh, C.A., Alok,
K., Maxim, V.S., Lisa, P., Bernard, R.L., and Henri,
B. (1995). R,S-1,3-butanediol acetoacetate esters,
potential alternates to lipid emulsions for total
parenteral nutrition. J Nutr Biochem 6.
Talbot, K., Wang, H.Y., Kazi, H., Han, L.Y., Bakshi,
K.P., Stucky, A., Fuino, R.L., Kawaguchi, K.R.,
Samoyedny, A.J., Wilson, R.S., et  al. (2012).
Demonstrated brain insulin resistance in
Alzheimer’s disease patients is associated with
IGF-1 resistance, IRS-1 dysregulation, and cogni-
tive decline. J Clin Invest 122, 1316–1338.
Tate, R.L., Mehlman, M.A., and Tobin, R.B. (1971).
Metabolic fate of 1,3-butanediol in the rat:  con-
version to -hydroxybutyrate. J Nutr 101,
1719–1726.
Tisdale, M., and Brennan, R. (1983). Loss of acetoac-
etate coenzyme A transferase activity in tumours
of peripheral tissues. Br J Cancer 47, 293–297.
Tobin, R.B., Mehlman, M.A., and Parker, M. (1972).
Effect of 1,3-butanediol and propionic acid on
blood ketones, lipids and metal ions in rats. J Nutr
102, 1001–1008.
Torres-Gonzalez, M., Volek, J.S., Leite, J.O., Fraser,
H., and Luz Fernandez, M. (2008). Carbohydrate
restriction reduces lipids and inflammation
and prevents atherosclerosis in Guinea pigs. J
Atheroscler Thromb 15, 235–243.
Trauner, D.A. (1985). Medium-chain triglycer-
ide (MCT) diet in intractable seizure disorders.
Neurology 35, 237–238.
Turina, M., Fry, D.E., and Polk, H.C., Jr. (2005).
Acute hyperglycemia and the innate immune sys-
tem: clinical, cellular, and molecular aspects. Critic
Care Med 33, 1624–1633.
Vandanmagsar, B., Youm, Y.H., Ravussin, A., Galgani,
J.E., Stadler, K., Mynatt, R.L., Ravussin, E., Stephens,
J.M., and Dixit, V.D. (2011). The NLRP3 inflamma-
some instigates obesity-induced inflammation and
insulin resistance. Nat Med 17, 179–188.
327
Veech, R. (2004). The therapeutic implications of ketone
bodies: the effects of ketone bodies in pathological
conditions:  ketosis, ketogenic diet, redox states,
insulin resistance, and mitochondrial metabo-
lism. Prostaglandins Leukot Essent Fatty Acids 70,
309–319.
Veech RL, Chance B, Kashiwaya Y, et al: Ketone bod-
ies, potential therapeutic uses. IUBMB life 51:
241–7, 2001
Veech, R.L. (2014). Ketone ester effects on metabolism
and transcription. J Lipid Res 55, 2004–2006.
Viggiano, A., Pilla, R., Arnold, P., Monda, M.,
D’Agostino, D., and Coppola, G. (2015).
Anticonvulsant properties of an oral ketone ester
in a pentylenetetrazole-model of seizure. Brain
Res 1618, 50–54.
Wang, J., and, Yi, J. (2008). Cancer cell killing via
ROS: to increase or decrease, that is the question.
Cancer Biol Ther 7, 1875–1884.
Wheatley, K.E., Williams, E.A., Smith, N.C., Dillard, A.,
Park, E.Y., Nunez, N.P., Hursting, S.D., and Lane,
M.A. (2008). Low-carbohydrate diet versus caloric
restriction:  effects on weight loss, hormones, and
colon tumor growth in obese mice. Nutr Cancer
60, 61–68.
Wymelbeke, V.V., and Himaya, A. (1998). Influence of
medium-chain and long-chain triacylglycerols on
the control of food intake in men. Am J Clin Nutr
68(2), 226–234.
Youm, Y.H., Grant, R.W., McCabe, L.R., Albarado,
D.C., Nguyen, K.Y., Ravussin, A., Pistell, P.,
Newman, S., Carter, R., Laque, A., et  al. (2013).
Canonical Nlrp3 inflammasome links systemic
low-grade inflammation to functional decline in
aging. Cell Metabol 18, 519–532.
Youm, Y.H., Nguyen, K.Y., Grant, R.W., Goldberg, E.L.,
Bodogai, M., Kim, D., D’Agostino, D., Planavsky,
N., Lupfer, C., Kanneganti, T.D., et al. (2015). The
ketone metabolite beta-hydroxybutyrate blocks
NLRP3 inflammasome-mediated inflammatory
disease. Nat Med 21(3), 263–269
Zhang, J., Cao, Q., Li, S., Lu, X., Zhao, Y., Guan, J.S.,
Chen, J.C., Wu, Q., and Chen, G.Q. (2013a). 3-
Hydroxybutyrate methyl ester as a potential
drug against Alzheimer’s disease via mitochon-
dria protection mechanism. Biomaterials 34,
7552–7562.
Zhang, Y., Kuang, Y., LaManna, J., and Puchowicz, M.
(2013b). Contribution of brain glucose and ketone
bodies to oxidative metabolism. Adv Exp Med Biol
765, 365–370.
Zuccoli, G., Marcello, N., Pisanello, A., Servadei, F.,
Vaccaro, S., Mukherjee, P., and Seyfried, T.N. (2010).
Metabolic management of glioblastoma multiforme
using standard therapy together with a restricted
ketogenic diet: Case Report Nutr Metab (Lond) 7, 33.
328
33
Identifying the Molecular Mechanism of the Medium
Chain Triglyceride (Ketogenic) Diet
M AT T H E W C . WA L K E R , F R C P, P H D A N D R O B I N S . B . W I L L I A M S , P H D
INTRODUCTION
The medium chain triglyceride (MCT) ketogenic
diet was first introduced as a more palatable alter-
native to the classical ketogenic diet for the treat-
ment of refractory epilepsy by Huttenlocher et al.
(1971). It now provides a key therapeutic approach
for the treatment of children with drug-resistant
epilepsy (Levy et  al., 2012; Liu, 2008; Neal and
Cross, 2010; Neal et  al., 2009). The diet involves
a stringent reduction in carbohydrate intake, with
an elevated consumption of medium chain fatty
acids within a triglyceride backbone. Typically,
the fats in this diet provide 65% to 75% of total
daily energy requirement. These fats comprise
two medium, straight chain fats:  octanoic acid
(comprising eight carbons) and decanoic acid
(comprising ten carbons); 60%–80% of the MCT
intake is octanoic acid and the remainder is mostly
decanoic acid (Sills et  al., 1986b). Both fats are
rapidly hydrolyzed off the triglyceride backbone
in the gut, and absorbed as free fatty acids (Bach
and Babayan, 1982) and metabolized into ketones
(β-hydroxybutyrate, acetone, and acetoacetate)
and carbon dioxide, or catabolized into long chain
fatty acids. This diet results in an elevation of the
concentration of both fatty acids in peripheral
blood:  octanoic acid to 104–859 µM and averag-
ing around 306 µM; and decanoic acid to 87–552
µM with an average of 157 µM (Dean et al., 1989;
Haidukewych et  al., 1982; Sills et  al., 1986a). In
animal models, decanoic acid has been found to
penetrate the blood-brain barrier and to be pres-
ent in brain at 60%–80% of serum levels (Wlaz
et al., 2012).
The dietary restrictions and high fat intake
of the MCT ketogenic diet can cause a variety
of gastrointestinal-related side effects, such as
cramps, bloating, diarrhea, and vomiting (Liu,
2008). Use of the diet has also been limited by
poor tolerability, especially in adults, resulting
in a high attrition rate (Levy et al., 2012). Due to
these adverse effects, and a desire to increase the
efficacy of the diet, many studies have sought to
identify the therapeutic mechanism of the diet. As
the diet was based on the classical ketogenic diet,
it was considered to act through the generation of
ketones (Bough and Rho, 2007; Rho and Stafstrom,
2011). However, the presence of these ketones
poorly correlates with anticonvulsant efficacy, and
this ketone-based mechanism has not been widely
supported in animal model studies (Likhodii et al.,
2000; Thavendiranathan et al., 2000). Furthermore,
ketones do not directly alter hippocampal synaptic
transmission (Thio et al., 2000) nor do they affect
epileptiform activity induced by 4-aminopyridine
in ex vivo seizure models (Thio et al., 2000). Thus
the therapeutic mechanism of the MCT ketogenic
diet has, until recently, been unclear.
M E D I U M C H A I N FAT T Y
ACIDS IN SEIZURE CONTROL
A clinical role for medium chain fatty acids in
direct seizure control was proposed over 30 years
ago (Dean et al., 1989; Haidukewych et al., 1982;
Sills et al., 1986a), but due to limited study sizes, a
direct correlation between plasma concentrations
and seizure control was not established. Recently,
however, an unbiased screen for medium and
short chain fatty acids related to valproate (Chang
et al., 2012) in a simple cellular model identified
a number of fatty acids as potential seizure-con-
trol treatments. In this study, a simple non-animal
model, in which valproate has been shown to reg-
ulate phosphoinositide turnover (Xu et al., 2007),
was used to identify potential anti-seizure com-
pounds (Cunliffe et al., 2015). In this study, over 60
compounds were screened for an inhibitory effect
on rapid phosphoinositide turnover. Interestingly,
this inhibitory effect has since been confirmed as
a therapeutic mechanism for valproate in in vitro
329

Chapter 33: Molecular Mechanism of the Medium Chain Triglyceride (Ketogenic) Diet 329

and in vivo animal seizure models (Chang et al.,
2014). Thus, using this phosphoinositide screen,
Chang et al. (2012) were able to investigate a wide
range of fatty acids and related compounds, with
some structures producing little effect, and some
structures showing enhanced activity over valpro-
ate. These potent compounds included decanoic
acid, nonanoic acid, and the branched 8-carbon
backbone 4-methyloctanoic acid. However, these
compounds did not act through regulating inosi-
tol levels in this simple model that was previously
suggested as a mechanism of valproate (Shaltiel et
al., 2007; Vaden et al., 2001; Williams et al., 2002;
Williams, 2005). These compounds were further
tested in a well-established ex vivo mammalian
model for drug-resistant epilepsy, the low mag-
nesium hippocampal-entorhinal cortex model
(Chang et al., 2012). Here, brain slices were kept
“alive” in a bath perfused with oxygenated arti-
ficial cerebrospinal fluid, and seizure-like activ-
ity was induced by reducing magnesium levels,
so enhancing NMDA receptor currents. In these
experiments, equimolar concentrations of a range
of fatty acids were more potent than valproate,
within 10 minutes of compound addition (the time
for the compounds to perfuse the bath and pen-
etrate the slice), thus significant levels of ketosis in
these experiments is unlikely. These experiments
therefore suggested a direct role for medium chain
fatty acids in seizure control.
Further studies pursued the analysis of the
activity of these medium chain fatty acids in
seizure control and related effects (Chang et al.,
2013). Here, decanoic acid and nonanoic acid
were shown to provide protection against an ex
vivo model of epileptiform activity generated by
decreasing GABAergic inhibition (pentylenetet-
razol, PTZ; Figure 33.1). In these studies, deca-
noic acid completely blocked seizure activity 35
minutes post addition. Interestingly, octanoic
acid, also prescribed in the MCT ketogenic diet,
did not reduce seizure activity in this model.
However, branched derivatives of octanoic acid
showed variable efficacy, with some compounds
providing strong seizure control (e.g., 4-methy-
loctanoic acid) and some showing no activ-
ity (3,7-dimethyloctanoic acid). These results
confirmed earlier studies (Chang et al., 2012)
but extended these findings to establish a wide
chemical space showing potential efficacy in
seizure control. This study also examined these
chemicals for inhibition of histone deacetylase
activity (HDAC), an effect associated with tera-
togenicity (Gottlicher et al., 2001; Gurvich et al.,
2004; Phiel et al., 2001), and thus limiting their
use during pregnancy (Jentink et al., 2010; Koren
et al., 2006). Of the compounds analyzed in this
study, only valproate and 2-proplyoctanoic acid
caused significant inhibition of HDAC activity
at 1 mM, a concentration considerably higher
than that of medium chain fatty acids found
during MCT ketogenic diet treatment (Sills et
al., 1986a). This suggests that medium chain
fatty acid are unlikely to exhibit the teratogenic

Treatment (1 mM)
120 Control
discharge (100 % of baseline)

O
Frequency of epileptiform

100 Octanoic acid HO

O

80 Valproate HO
O

4-methyloctanoic acid HO
60 O
Nonanoic acid
40 HO
O
4-ethyloctanoic acid
20 HO
O
Decanoic acid
0 O
HO

0 10 20 30 40 50 60 4-BCCA HO
Time (minutes)

FIGURE 33.1 Structurally specific medium chain fatty acids strongly reduce frequency of in vitro epileptiform activity.
In these experiments, an ex vivo hippocampal slice model was used, with seizure-like activity induced by application of
PTZ. Compounds were added (gray box) at 1 mM, and removed after 40 minutes. The frequency of epileptiform activity
is plotted against time following control (DMSO). Straight medium chain fatty acids decanoic (10 carbon) and nonanoic
acid (9 carbon), two branched chain derivatives (4-methyloctanoic acid and 4-ethyloctanoic acid), and a cyclic congener
(trans-4-butylcyclocarboxylic acid; 4-BCCA) were able to strongly reduce seizure-like activity. In contrast, octanoic acid
showed no inhibitory activity. The widely used, established epilepsy treatment, valproate, showed weak activity in this
seizure model. Data derived from Chang et al. (2014), Chang et al. (2015), and Chang et al. (2016).
330

330 section IV: Ketone-Based Metabolism

effects found following valproate treatment. This
study also showed that medium chain fatty acids
(4-methyloctanoic acid and nonanoic acid) pro-
vide control against self-sustaining status epi-
lepticus (SSSE), induced by perforant pathway
stimulation, in an in vivo rat model (Walker and
Williams, 2015). Analysis of neuronal cell death
in these animals as a result of SSSE showed that
2 months after seizure induction, nonanoic acid
significantly reduced cell death in the hilus of
the hippocampus, suggesting a neuroprotective
effect. This therapeutic effect is also unlikely to
be related to ketosis, since these compounds rap-
idly reduced seizure activity within 10 minutes
of treatment. For nonanoic acid, these effects
were also unlikely to occur due to sedation, since
sedative effects were not shown even at high con-
centrations (600 mg/kg). Together these studies
provide strong evidence for the direct activity of
a range of medium chain fatty acids in seizure
control and neuroprotection.
One further study has been reported, investi-
gating the breadth of chemical space for medium
chain fatty acids in seizure control (Chang et al.,
2015). This study specifically investigated octanoic
acid–related compounds, including a systematic
analysis of methyloctanoic acid derivatives. Using
an ex vivo model of seizure activity, induced in
rat hippocampal slices by the application of PTZ,
a clear structure activity relationship was seen
in seizure control, provided by the position of
branching on the octanoic acid backbone. The
most potent compounds branched around the
fifth carbon. This study also examined these com-
pounds for protection against excitotoxic cell
death, an effect that is similar to that resulting
from status epilepticus (DeLorenzo et al., 2005).
In these experiments, exposure of hippocampal
neurons in culture to low magnesium for 4 hours
triggered cell death (Deshpande et al., 2008),
and consistent with the seizure control experi-
ments, neuroprotection was seen with the addi-
tion of compounds with methyl branching from
the fourth to the seventh carbon (Chang et al.,
2015). Based on these results, three related com-
pounds were then examined: 4-ethyloctanoic
acid, containing a longer side chain; 4-methyl-
nonanoic acid, containing a longer backbone,
and trans-4-butylcyclohexane carboxylic acid
(4-BCCA), a related cyclic compound. All three
compounds also showed strong seizure control
in the PTZ model (Figure 33.1), and were not
active against HDAC activity. These compounds
were further screened in a range of in vivo seizure
models (Table 33.1), providing promising activity
for trans-4-butylcyclohexane carboxylic acid in
multiple models. In a related study, mice treated
with decanoic acid by gastric gavage (30 mmol/
kg) also showed seizure control in the 6-Hz sei-
zure and maximal electroshock seizure threshold
(MEST) models (Wlaz et al., 2012). These data
provide evidence that medium chain fatty acids,
of defined structure, show activity in a range of in
vivo seizure models.

TABLE 33.1 IN VIVO SEIZURE CONTROL DATA FOR ACTIVE COMPOUNDS

Compound Species Seizure Dose Animals Animals ED50

model (mg/kg) (protected/tested) (Toxic/tested) (mg/kg)

4-EOAa Mice 6Hz 150 7/8 — 110

Rat MES 125 12/16 4/15† 100
Mice scMET 200 8/8 4/8¥ 142
Mice CKM 110 6/8 0/8 71
4-BCCAa Mice 6Hz 100 3/4 0/4 81
Rat MES 100 4/8 1/8† ~100*
Mice scMET 150 4/8 0/8 ~150*
Mice CKM 80 8/8 0/8 44
VPA Mice 6Hz 263b
Rat MES 485c
Mice scMET 191d
Mice CKM 174e
Summary of in vivo seizure control in multiple models, with data provided by collaborative research with (a) NINDS, or previously deter-
mined by (b) Barton et al. (2001), (c) Loscher (1999), (d) Rowley and White (2010), or (e) NINDS.
* based on single dose data; — not determined; † unable to grasp rotorod; ¥ ataxia/loss of righting reflex
Source: Partially reproduced from Chang et al. (2015), with permission.
331
Chapter 33: Molecular Mechanism of the Medium Chain Triglyceride (Ketogenic) Diet
M E D I U M C H A I N FAT T Y
A C I D S A R E A M PA
R E C E P TO R AG O N I S T S
The molecular mechanism of decanoic acid in
direct seizure control has recently been estab-
lished (Chang et  al., 2016). In this study, deca-
noic acid (1 mM) was shown to abolish seizure
activity in vitro in both the PTZ- and low-
magnesium-induced drug-resistant seizure
models. As previously described, decanoic acid
blocks seizure activity within 15 minutes of addi-
tion to each model, strongly suggesting that the
effect is directly related to the fatty acid, not to
ketone generation. Moreover acetone and β-
hydroxybutyrate (two ketones) have no effect in
either model at high concentrations (10 mM).
The mechanism of decanoic acid for this effect
was then shown using whole-cell patch clamp
recordings of evoked AMPA receptor–mediated
excitatory postsynaptic currents (EPSCs) from
CA1 pyramidal neurons, where decanoic acid
reduced EPSC amplitude by 17.0 ± 5.6% at a con-
centration comparable to its steady state level in
children on the MCT diet (100 µM) (Sills et al.,
1986a). These data, for this first time, provided
evidence of a direct mechanism of the MCT keto-
genic diet in seizure control through inhibition of
AMPA receptor currents.
To investigate a direct effect of decanoic acid
on AMPA receptor activity, a range of experi-
ments were carried out using a Xenopus heter-
ologous expression system (Chang et  al., 2016).
Here, AMPA receptor subunits (GluA1-3) were
expressed in oocytes individually or in pairs to
produce homotetramers (GluA1) or heterotetra-
mers (GluA1/2 and GluA2/3), and glutamate was
applied, resulting in inward currents, on which
direct application of decanoic acid and related
compounds was tested. This approach allowed the
quantification of the direct inhibition of AMPA
receptor–mediated currents by medium chain
fatty acids (Table 33.2). From these experiments,
decanoic acid was shown to be an AMPA receptor
antagonist, with efficacy across all subunit com-
position, and greatest potency against GluA2/3
heterotetramers (IC50 = 0.52 mM). Decreasing the
length of the backbone reduced potency, such that
nonanoic acid and octanoic acid were increasingly
less potent (IC50 = 1.48 and 3.82 mM, respectively),
although addition of a side chain to octanoic acid
to produce 4-methyloctanoic acid enhanced
potency (IC50  =  0.84 mM). These data indicate a
direct action of decanoic acid and specific related
structures on inhibition of AMPA receptors to
regulate neuronal function.
331
This study also investigated some kinetic
aspects of medium chain fatty acid-dependent
AMPA receptor inhibition (Chang et al., 2016).
First, since during seizure activity, membrane
depolarization occurs, increasing excitability,
decanoic acid was analyzed at varying membrane
potentials. These studies identified that under
depolarized membrane potentials, decanoic acid
showed enhanced inhibitory activity, thus poten-
tially providing more potent inhibition during
seizure activity. Second, since AMPA receptors
are activated by glutamate, and glutamate levels
increase during seizure activity (Van Den Pol et
al., 1996), competition assays were used to explore
whether altered glutamate levels modulate deca-
noic acid-dependent AMPA receptor inhibition.
These experiments indicate that decanoic acid is a
noncompetitive AMPA receptor antagonist (Table
33.2), thus its inhibitory activity is independent of
glutamate concentrations.
These studies have suggested that medium
chain fatty acids function in seizure control
through inhibition of AMPA receptors. However,
it remained possible that these compounds could
give rise to seizure control through alternative
mechanisms. To address this, Chang et al. (2016)
employed the hippocampal/PTZ seizure model,
treated with the well-characterized AMPA recep-
tor antagonist GYKI 52466 (Arai, 2001), to pro-
vide a similar level of AMPA receptor inhibition
as decanoic acid (1 mM). Under these conditions,
TABLE 33.2 SUMMARY OF MEDIUM
CHAIN FAT T Y ACIDS INHIBITORY
EFFECTS ON AMPA RECEPTORS
GluA2/3 GluA1/2 GluA1
Subunit specificity: IC50 (mM)
Decanoic acid 0.52 1.16 2.09
Nonanoic acid 1.48 1.79
4-methyloctanoic 0.84
acid
Octanoic acid 3.82
Voltage dependence (IC50 mM, Decanoic acid, GluA2/3)
–80 mV 1.11
–40 mV 0.43
Kinetics
Decanoic acid is noncompetitive toward glutamate
Data was derived from heterologous expression of rat glutamate
receptors in a Xenopus oocyte model, activated with 100 μM
glutamate.
Source: Data derived from Chang et al. (2016).
332
332 section IV: Ketone-Based Metabolism
GYKI 52466 blocked seizure activity. These
results indicate that direct AMPA inhibition
is sufficient to explain decanoic acid’s antisei-
zure effect in in vitro seizure models. Moreover,
AMPA receptors are widely recognized as tar-
gets for seizure control (Meldrum and Rogawski,
2007; Russo et al., 2012; Szenasi et al., 2008).
Perampanel, a selective AMPA receptor antago-
nist, has also recently been marketed for the
treatment of refractory partial epilepsy (Rektor,
2013), confirming the relevance of this mecha-
nism to patient treatment. Chang et al. (2016)
also showed that inhibition of AMPA receptors
by decanoic acid at therapeutically relevant con-
centrations provided similar inhibitory effects
as Perampanel at levels found during clinical
use (Rogawski and Hanada, 2013; Ceolin et al.,
2012). It is interesting to note however, that the
use of Perampanel is limited by neuropsychiatric
side effects, in particular aggression (Rugg-Gunn,
2014; Steinhoff et al., 2014). This effect is not seen
during the MCT ketogenic diet, suggesting that
these side effects may be peculiar to Perampanel
rather than a class effect of drugs that act at
AMPA receptors; thus the development of phar-
maceutical reagents based on the mechanisms of
decanoic acid will hopefully not cause this side
effect.
Finally, the binding site for decanoic acid on
AMPA receptors was investigated by in silico
modeling (Chang et  al., 2016). Using the trans-
membrane domains of GluA2, a putative deca-
noic acid–binding region was identified in the
M3 helix, thought to be involved in regulating
and gating inward currents. This site is distinct
from that reported for Perampanel (Szenasi et al.,
2008), suggesting potential differences in the
effects of decanoic acid and Perampanel on AMPA
receptor function. This binding site is consistent
with results from electrophysiology experiments
described here, although further studies will need
to confirm this site.
IS THERE STILL A ROLE
F O R   K E TO N E S
IN SEIZURE CONTROL?
Over the last 30  years, many studies have sug-
gested a role for ketones in seizure control and
neuroprotection (Bough and Rho, 2007; Rho
and Stafstrom, 2011). In some types of epilepsy,
ketones may provide an alternative energy source
for the brain (Kim et al., 2015; Masino et al., 2011),
for example in Glut1 deficiency when glucose is
poorly transported across the blood-brain bar-
rier. Ketones have also been demonstrated to alter
transcriptional regulation relevant to a putative
disease-modifying role (Kim et  al., 2015; Masino
et  al., 2011), and to have indirect effects on ion
channels (Sada et al., 2015). Thus ketones may still
provide some beneficial effects related to specific
causes of epilepsy and in protection against long-
term epileptogenic changes.
I M P L I C AT I O N S
Understanding the mechanism of action of the
ketogenic diet is essential for the development of
diets or treatments that are less restrictive, and
that do not produce the same diet-associated side
effects. The recent data outlined here provide a
strong case for the antiseizure effects of the MCT
ketogenic diet resulting from increases in plasma
fatty acid concentrations rather than through the
production of ketones. Thus, due to the burgeon-
ing evidence that fatty acids play a primary role
in the diet’s therapeutic effect and despite the
potential beneficial roles of ketones, it would per-
haps be more appropriate to rename the diet, the
MCT diet.
Decanoic acid may have beneficial effects
beyond its action at AMPA receptors. A  recent
study has identified a role for decanoic acid in
regulating mitochondrial proliferation (Hughes
et al., 2014). Here, decanoic acid acts through reg-
ulation of a fatty acid receptor, PPARγ (Malapaka
et al., 2012; Zuckermann et al., 2015), at therapeu-
tically relevant concentrations (250 μM) over a 6-
day period in a neuronal cell line and in human
fibroblasts. Interestingly, this effect was not shown
by octanoic acid. The consequent increase in
mitochondrial load has been suggested to protect
against seizure induction (Bough et  al., 2006),
and to protect against mitochondrial dysfunction
in epilepsy that has been clearly demonstrated in
animal models of status epilepticus (Cock et  al.,
2002) and in patients (Kunz et al., 2000).
Medium chain fatty acids are rapidly metabo-
lized in vivo (Bach and Babayan, 1982). It remains,
therefore, a key consideration in the design and
management of seizure-control treatments based
on decanoic acid, to ensure maintenance of thera-
peutic fatty acid levels. Early studies of medium
chain fatty acids show considerable variation in
plasma fatty acid levels over a 24-hour period (Sills
et al., 1986b). It remains unclear if a reduced car-
bohydrate load with the diet is necessary to main-
tain decanoic acid levels at an effective level, and
further studies will need to address this in greater
detail. However, a corollary of this is that clinicians
using the MCT diet should focus on monitoring
blood fatty acids rather than just ketone levels.
333
Chapter 33: Molecular Mechanism of the Medium Chain Triglyceride (Ketogenic) Diet
The discovery of a molecular target for the MCT
diet  also opens the possibility of a pharmaceutic
approach to replace the diet. Several considerations
are necessary here. Due to the rapid metabolism
of decanoic acid, a chemical that provides similar
inhibitory activity against AMPA receptors but is
resistant to metabolic degradation may overcome
the stringent dietary regime. Developing novel
chemical structures that show enhanced binding to
the target site on the receptor may also reduce the
necessary dose of the compound. This would addi-
tionally help to reduce the side effects associated
with high fatty acid intake, in particular gastric irri-
tation. Studies outlined here have provided some
evidence for the efficacy of decanoic acid congeners
in seizure control and in direct inhibition of AMPA
receptors. Further studies will be necessary to iden-
tify suitable candidates for clinical testing, but may
result in the “diet in a pill.”
REFERENCES
Arai, A.C. (2001). GYKI 52466 has positive modula-
tory effects on AMPA receptors. Brain Res 892,
396–400.
Bach, A.C., and Babayan, V.K. (1982). Medium-
chain triglycerides: an update. Am J Clin Nutr 36,
950–962.
Barton, M.E., Klein, B.D., Wolf, H.H., and White, H.S.
(2001). Pharmacological characterization of the 6
Hz psychomotor seizure model of partial epilepsy.
Epilepsy Res 47, 217–227.
Bough, K.J., and Rho, J.M. (2007). Anticonvulsant
mechanisms of the ketogenic diet. Epilepsia
48, 43–58.
Bough, K.J., Wetherington, J., Hassel, B., Pare, J.F.,
Gawryluk, J.W., Greene, J.G., Shaw, R., Smith,
Y., Geiger, J.D., and Dingledine, R.J. (2006).
Mitochondrial biogenesis in the anticonvulsant
mechanism of the ketogenic diet. Ann Neurol 60,
223–235.
Ceolin, L., Bortolotto, Z.A., Bannister, N., Collingridge,
G.L., Lodge, D., and Volianskis, A. (2012). A novel
anti-epileptic agent, perampanel, selectively inhib-
its AMPA receptor-mediated synaptic transmis-
sion in the hippocampus. Neurochem Int 61,
517–522.
Chang, P., Orabi, B., Deranieh, R.M., Dham, M.,
Hoeller, O., Shimshoni, J.A., Yagen, B., Bialer,
M., Greenberg, M.L., Walker, M.C., et  al. (2012).
The antiepileptic drug valproic acid and other
medium-chain fatty acids acutely reduce phos-
phoinositide levels independently of inositol in
Dictyostelium. Dis Model Mech 5, 115–124.
Chang, P., Terbach, N., Plant, N., Chen, P.E., Walker,
M.C., and Williams, R.S. (2013). Seizure control
by ketogenic diet-associated medium chain fatty
acids. Neuropharmacology 69, 105–114.
333
Chang, P., Walker, M.C., and Williams, R.S. (2014).
Seizure-induced reduction in PIP3 levels contrib-
utes to seizure-activity and is rescued by valproic
acid. Neurobiol Dis 62, 296–306.
Chang, P., Zuckermann, A.M., Williams, S., Close, A.J.,
Cano-Jaimez, M., McEvoy, J.P., Spencer, J., Walker,
M.C., and Williams, R.S. (2015). Seizure control by
derivatives of medium chain fatty acids associated
with the ketogenic diet show novel branching-
point structure for enhanced potency. J Pharmacol
Exp Ther 352, 43–52.
Chang, P., Augustin, A.M., Boddum, K., Williams, S.,
Sun, M., Terschak, J.A., Hardege, J.D., Chen, P.E.,
Walker, M.C., and Williams, R.S.B. (2016). Seizure
control by decanoic acid through direct AMPA
receptor inhibition. Brain 139(Pt 2):431–443. In
press, online release 25 Nov 2015.
Cock, H.R., Tong, X., Hargreaves, I.P., Heales, S.J.,
Clark, J.B., Patsalos, P.N., Thom, M., Groves,
M., Schapira, A.H., Shorvon, S.D., et  al. (2002).
Mitochondrial dysfunction associated with neu-
ronal death following status epilepticus in rat.
Epilepsy Res 48, 157–168.
Cunliffe, V.T., Baines, R.A., Giachello, C.N., Lin, W.H.,
Morgan, A., Reuber, M., Russell, C., Walker, M.C.,
and Williams, R.S. (2015). Epilepsy research meth-
ods update:  understanding the causes of epileptic
seizures and identifying new treatments using non-
mammalian model organisms. Seizure 24, 44–51.
Dean, H.G., Bonser, J.C., and Gent, J.P. (1989). HPLC
analysis of brain and plasma for octanoic and dec-
anoic acids. Clin Chem 35, 1945–1948.
DeLorenzo, R.J., Sun, D.A., and Deshpande, L.S.
(2005). Cellular mechanisms underlying acquired
epilepsy: the calcium hypothesis of the induction
and maintainance of epilepsy. Pharmacol Ther
105, 229–266.
Deshpande, L.S., Lou, J.K., Mian, A., Blair, R.E.,
Sombati, S., Attkisson, E., and DeLorenzo, R.J.
(2008). Time course and mechanism of hippocam-
pal neuronal death in an in vitro model of status
epilepticus: role of NMDA receptor activation and
NMDA dependent calcium entry. Eur J Pharmacol
583, 73–83.
Gottlicher, M., Minucci, S., Zhu, P., Kramer, O.H.,
Schimpf, A., Giavara, S., Sleeman, J.P., Lo, C.F.,
Nervi, C., Pelicci, P.G., et al. (2001). Valproic acid
defines a novel class of HDAC inhibitors inducing
differentiation of transformed cells. EMBO J 20,
6969–6978.
Gurvich, N., Tsygankova, O.M., Meinkoth, J.L., and
Klein, P.S. (2004). Histone deacetylase is a target
of valproic acid-mediated cellular differentiation.
Cancer Res 64, 1079–1086.
Haidukewych, D., Forsythe, W.I., and Sills, M. (1982).
Monitoring octanoic and decanoic acids in plasma
from children with intractable epilepsy treated
with medium-chain triglyceride diet. Clin Chem
28, 642–645.
334
334 section IV: Ketone-Based Metabolism
Hughes, S.D., Kanabus, M., Anderson, G., Hargreaves,
I.P., Rutherford, T., O’Donnell, M., Cross, J.H.,
Rahman, S., Eaton, S., and Heales, S.J. (2014).
The ketogenic diet component decanoic acid
increases mitochondrial citrate synthase and com-
plex I activity in neuronal cells. J Neurochem 129,
426–433.
Huttenlocher, P.R., Wilbourn, A.J., and Signore, J.M.
(1971). Medium-chain triglycerides as a therapy
for intractable childhood epilepsy. Neurology 21,
1097–1103.
Jentink, J., Loane, M.A., Dolk, H., Barisic, I., Garne, E.,
Morris, J.K., and de Jong-van den Berg LT (2010).
Valproic acid monotherapy in pregnancy and
major congenital malformations. N Engl J Med
362, 2185–2193.
Kim, D.Y., Simeone, K.A., Simeone, T.A., Pandya, J.D.,
Wilke, J.C., Ahn, Y., Geddes, J.W., Sullivan, P.G.,
and Rho, J.M. (2015). Ketone bodies mediate anti-
seizure effects through mitochondrial permeabil-
ity transition. Ann Neurol 78(1), 77–87.
Koren, G., Nava-Ocampo, A.A., Moretti, M.E.,
Sussman, R., and Nulman, I. (2006). Major mal-
formations with valproic acid. Can Fam Physician
52, 441–442, 444, 447.
Kunz, W.S., Kudin, A.P., Vielhaber, S., Blumcke, I.,
Zuschratter, W., Schramm, J., Beck, H., and Elger,
C.E. (2000). Mitochondrial complex I  deficiency
in the epileptic focus of patients with temporal
lobe epilepsy. Ann Neurol 48, 766–773.
Levy, R.G., Cooper, P.N., and Giri, P. (2012). Ketogenic
diet and other dietary treatments for epilepsy.
Cochrane Database Syst Rev 3, CD001903.
Likhodii, S.S., Musa, K., Mendonca, A., Dell, C.,
Burnham, W.M., and Cunnane, S.C. (2000).
Dietary fat, ketosis, and seizure resistance in rats
on the ketogenic diet. Epilepsia 41, 1400–1410.
Liu, Y.M. (2008). Medium-chain triglyceride (MCT)
ketogenic therapy. Epilepsia 49 (Suppl 8), 33–36.
Loscher, W. (1999). Animal models of epilepsy and epi-
leptic seizures. In Antiepileptic Drugs, Pharmacology
and Therapeutics, M.J. Eadie and F.J.E. Vajda, eds.
(Berlin: Springer-Verlag), pp. 19–62.
Malapaka, R.R., Khoo, S., Zhang, J., Choi, J.H., Zhou,
X.E., Xu, Y., Gong, Y., Li, J., Yong, E.L., Chalmers,
M.J., et  al. (2012). Identification and mechanism
of 10-carbon fatty acid as modulating ligand of
peroxisome proliferator-activated receptors. J Biol
Chem 287, 183–195.
Masino, S.A., Li, T., Theofilas, P., Sandau, U.S., Ruskin,
D.N., Fredholm, B.B., Geiger, J.D., Aronica, E., and
Boison, D. (2011). A ketogenic diet suppresses sei-
zures in mice through adenosine A(1) receptors. J
Clin Invest 121, 2679–2683.
Meldrum, B.S., and Rogawski, M.A. (2007). Molecular
targets for antiepileptic drug development.
Neurotherapeutics 4, 18–61.
Neal, E.G., Chaffe, H., Schwartz, R.H., Lawson, M.S.,
Edwards, N., Fitzsimmons, G., Whitney, A., and
Cross, J.H. (2009). A randomized trial of classical
and medium-chain triglyceride ketogenic diets in
the treatment of childhood epilepsy. Epilepsia 50,
1109–1117.
Neal, E.G., and Cross, J.H. (2010). Efficacy of dietary
treatments for epilepsy. J Hum Nutr Diet 23,
113–119.
Phiel, C.J., Zhang, F., Huang, E.Y., Guenther, M.G.,
Lazar, M.A., and Klein, P.S. (2001). Histone
deacetylase is a direct target of valproic acid, a
potent anticonvulsant, mood stabilizer, and terato-
gen. J Biol Chem 276, 36734–36741.
Rektor, I. (2013). Perampanel, a novel, non-
competitive, selective AMPA receptor antago-
nist as adjunctive therapy for treatment-resistant
partial-onset seizures. Expert Opin Pharmacother
14, 225–235.
Rho, J.M., and Stafstrom, C.E. (2011). The ketogenic
diet: what has science taught us? Epilepsy Res
100(3), 210–217.
Rogawski, M.A., and Hanada, T. (2013). Preclinical
pharmacology of perampanel, a selective non-
competitive AMPA receptor antagonist. Acta
Neurol Scand Suppl (197), 19–24.
Rowley, N.M., and White, H.S. (2010). Comparative
anticonvulsant efficacy in the corneal kindled
mouse model of partial epilepsy: correlation with
other seizure and epilepsy models. Epilepsy Res
92, 163–169.
Rugg-Gunn, F. (2014). Adverse effects and safety
profile of perampanel:  a review of pooled data.
Epilepsia 55 (Suppl 1), 13–15.
Russo, E., Gitto, R., Citraro, R., Chimirri, A., and De,
S.G. (2012). New AMPA antagonists in epilepsy.
Expert Opin Investig Drugs 21, 1371–1389.
Sada, N., Lee, S., Katsu, T., Otsuki, T., and Inoue,
T. (2015). Epilepsy treatment:  targeting LDH
enzymes with a stiripentol analog to treat epilepsy.
Science 347, 1362–1367.
Shaltiel, G., Mark, S., Kofman, O., Belmaker, R.H.,
and Agam, G. (2007). Effect of valproate deriva-
tives on human brain myo-inositol-1-phosphate
(MIP) synthase activity and amphetamine-
induced rearing. Pharmacol Rep 59, 402–407.
Sills, M.A., Forsythe, W.I., and Haidukewych, D.
(1986a). Role of octanoic and decanoic acids in the
control of seizures. Arch Dis Child 61, 1173–1177.
Sills, M.A., Forsythe, W.I., Haidukewych, D.,
MacDonald, A., and Robinson, M. (1986b). The
medium chain triglyceride diet and intractable
epilepsy. Arch. Dis. Child 61, 1168–1172.
Steinhoff, B.J., Hamer, H., Trinka, E., Schulze-
Bonhage, A., Bien, C., Mayer, T., Baumgartner, C.,
Lerche, H., and Noachtar, S. (2014). A multicenter
survey of clinical experiences with perampanel in
real life in Germany and Austria. Epilepsy Res 108,
986–988.
Szenasi, G., Vegh, M., Szabo, G., Kertesz, S., Kapus, G.,
Albert, M., Greff, Z., Ling, I., Barkoczy, J., Simig,
335
Chapter 33: Molecular Mechanism of the Medium Chain Triglyceride (Ketogenic) Diet
G., et  al. (2008). 2,3-benzodiazepine-type AMPA
receptor antagonists and their neuroprotective
effects. Neurochem Int 52, 166–183.
Thavendiranathan, P., Mendonca, A., Dell, C.,
Likhodii, S.S., Musa, K., Iracleous, C., Cunnane,
S.C., and Burnham, W.M. (2000). The MCT keto-
genic diet:  effects on animal seizure models. Exp
Neurol 161, 696–703.
Thio, L.L., Wong, M., and Yamada, K.A. (2000). Ketone
bodies do not directly alter excitatory or inhibitory
hippocampal synaptic transmission. Neurology
54, 325–331.
Vaden, D.L., Ding, D., Peterson, B., and Greenberg,
M.L. (2001). Lithium and valproate decrease inosi-
tol mass and increase expression of the yeast INO1
and INO2 genes for inositol biosynthesis. J Biol
Chem 276, 15466–15471.
Van Den Pol, A.N., Obrietan, K., and Belousov, A.
(1996). Glutamate hyperexcitability and seizure-
like activity throughout the brain and spinal cord
upon relief from chronic glutamate receptor block-
ade in culture. Neuroscience 74, 653–674.
Walker, M.C., and Williams, R.S. (2015). New
experimental therapies for status epilepticus
in preclinical development. Epilepsy Behav 49,
290–293.
335
Williams, R.S.B. (2005). Pharmacogenetics in
model systems:  Defining a common mecha-
nism of action for mood stabilisers. Prog
Neuropsychopharmacol Biol Psychiatry 29,
1029–1037.
Williams, R.S.B., Cheng, L., Mudge, A.W., and
Harwood, A.J. (2002). A common mechanism of
action for three mood-stabilizing drugs. Nature
417, 292–295.
Wlaz, P., Socala, K., Nieoczym, D., Luszczki, J.J.,
Zarnowska, I., Zarnowski, T., Czuczwar, S.J.,
and Gasior, M. (2012). Anticonvulsant profile of
caprylic acid, a main constituent of the medium-
chain triglyceride (MCT) ketogenic diet, in mice.
Neuropharmacology 62, 1882–1889.
Xu, X., Muller-Taubenberger, A., Adley, K.E.,
Pawolleck, N., Lee, V.W., Wiedemann, C., Sihra,
T.S., Maniak, M., Jin, T., and Williams, R.S. (2007).
Attenuation of phospholipid signaling provides a
novel mechanism for the action of valproic acid.
Eukaryot Cell 6, 899–906.
Zuckermann, A.M., La Ragione, R.M., Baines, D.L.,
and Williams, R.S. (2015). Valproic acid protects
against haemorrhagic shock-induced signalling
changes via PPARgamma activation in an in vitro
model. Br J Pharmacol 172(22), 5306–5317.
336
34
Triheptanoin in Epilepsy and Beyond
KARIN BORGES, PHD
INTRODUCTION
TO   A NA P L E RO S I S A N D
T R I H E P TA N O I N
Glucose, under nonfasting conditions, is usu-
ally the main fuel for the CNS. Glycolysis, which
occurs in cytosol, is the main pathway of metabolic
breakdown of glucose into pyruvate. When there
is limited oxygen, glycolysis produces lactate, oth-
erwise acetyl-CoA is produced via the pyruvate
dehydrogenase pathway for further oxidation in
the tricarboxylic acid (TCA) cycle in mitochon-
dria (Figure 34.2). In aerobic metabolism, after
citrate synthase transfers the acetyl-group onto
oxaloacetate forming citrate, the TCA cycle oxi-
dizes acetyl-CoA to two carbon dioxide molecules
in a series of chemical reactions. At the same
time, the TCA cycle provides the electron trans-
port chain with reducing equivalents, which then
produces the majority of ATP during aerobic con-
ditions (Figure 34.1). The TCA cycle is also part
of many other key metabolic pathways. The TCA
cycle intermediates containing four or five carbons
not only are part of the cycle involved in energy
production but also are used for other metabolic
pathways, for example, synthesis of various amino
acids and neurotransmitters, such as glutamate
and γ-aminobutyric acid (GABA). It is therefore
important that C4 intermediates of the TCA cycle
get “refilled” (anaplerosis) (Brunengraber and Roe,
2006; Hassel, 2000; Kornberg, 1966). Anaplerosis
is required for TCA cycling to occur, as it enhances
acetyl-CoA entry into the TCA cycle by providing
oxaloacetate, thus increasing TCA cycling and ATP
production. Moreover with increased amounts of
TCA cycle metabolites, more substrate for com-
plex II can be produced and reduction of NADH+
will be enhanced, which again will contribute to
ATP production. Please note that the TCA cycle
can also be short-circuited and still produce ATP,
for example with substrate entry as 2-oxoglutarate
or succinate. In many different models of disease
there is evidence of reduced amounts of TCA cycle
metabolites, for example, epilepsy (Alvestad et
al., 2008; Melo et al., 2005; Smeland et al., 2013;
Willis et al., 2010), stroke (Haberg et al., 2009), and
amyotrophic lateral sclerosis (ALS) (Niessen et al.,
2007). This is expected to lead to low oxaloacetate
levels, which then can reduce the binding of ace-
tyl-CoA by citrate synthase, its oxidation, and ulti-
mately decrease ATP production.
Various enzymes partake in the refilling of the
TCA cycle. In the CNS, carboxylation of pyru-
vate to oxaloacetate (number 1 in Figure 34.2)
is the main refilling pathway, thought to be tak-
ing place in astrocytes (Patel, 1974). Also, several
enzymes facilitate the formation of 2-oxoglutarate
(also called α-ketoglutarate) from glutamate.
This includes glutamate dehydrogenase, which
forms 2-oxoglutarate and ammonia from gluta-
mate and vice versa (2 in Figure 34.2). The reac-
tion catalyzed by glutamic pyruvic transaminases
(alanine aminotransferases), pyruvate + glutamate
↔ 2-oxoglutarate + alanine refills the cycle with
2-oxoglutarate (3 in Figure 34.2). In addition,
aspartate transaminase (also called glutamic oxa-
loacetic transaminase) can bypass part of the TCA
cycle to produce oxaloacetate from 2-oxoglutarate
in the reaction: Asp + 2-OG ↔ OAA + Glu (4 in
Figure 34.2).
Another anaplerotic pathway, the propionyl-
CoA carboxylation pathway (5 in Figure 34.2), has
largely been studied in peripheral tissues (Deng
et  al., 2009; Martini et  al., 2003; Nuutinen et  al.,
1981; Owen et  al., 2002; Reszko et  al., 2003). It
involves the carboxylation of propionyl-CoA to
methylmalonyl-CoA by propionyl-CoA carboxyl-
ase (EC 6.4.1.3). Methylmalonyl-CoA epimerase
(EC 5.1.99.1) and methylmalonyl-CoA mutase (EC
5.4.99.2) then produce succinyl-CoA. The branched
chain amino acids, isoleucine and valine, as well
as uneven fatty acids are anaplerotic via this path-
way, all providing propionyl-CoA (Figure 34.1).
Treatment with triheptanoin, the triglyceride of
heptanoate (C7 fatty acid), appears to currently be
337

Chapter 34: Triheptanoin in Epilepsy and Beyond 337

Glucose Triheptanoin

Acetyl~CoA Heptanoate

Citrate

OAA Propionyl~CoA
2-OG
NADH
Succinyl~CoA

Fumarate Succinate
ATP
O2 H2O ADP
II
I III IV V

FIGURE 34.1 Boosting the TCA cycling capacity when in need via the propionyl-CoA carboxylation pathway can
increase ATP production. This schematic includes the reduction of NAD+ to NADH by the TCA cycle, which then feeds
electrons into complex I  of the mitochondrial electron transport chain, while succinate directly feeds electrons into
complex II. Ultimately ATP is produced by ATP synthase.

the least toxic treatment when fueling this pathway
to a large extent. Unlike the branched chain amino
acids, triheptanoin does not overload the body
with nitrogen. In addition, providing the uneven
medium chain fat as a triglyceride avoids excessive
levels of sodium or acid, which otherwise could
challenge physiological homeostasis.
Triheptanoin is a tasteless oil and can be mixed
with various foods or made into an emulsion. As a
medium chain triglyceride, it is hydrolyzed in the
gastrointestinal tract, and due to its lipophilicity
the free medium chain heptanoate is thought to
diffuse directly into blood and mitochondria of all
tissues. This is unlike long chain fatty acids, which
are much more slowly metabolized, because they
first enter the lymph and require various trans-
port proteins in the blood and for final transport
into mitochondria for β-oxidation. Similar to even
medium chain fats, heptanoate is converted by the
liver to “ketones.” While even chain fats octanoate
and decanoate (C8 and C10) are turned into “C4
ketones,” β-hydroxybutyrate and acetoacetate, hep-
tanoate is metabolized to “C5 ketones,” β-hydroxy-
pentanoate and β-ketopentanoate (Brunengraber
and Roe, 2006; Gu et al., 2010; Kinman et al.,
2006; Roe and Mochel, 2006; Roe et al., 2002).
After release into the blood, C4 and C5 ketones are
taken up into cells by monocarboxylate transport-
ers. Regarding the crossing of the blood-brain bar-
rier, monocarboxylate transporters of the MCT1
type (Broer et al., 1998; Meredith and Christian,
2008) are thought to transport C5 ketones into
the brain, while heptanoate is more likely to enter
the brain via diffusion (Oldendorf, 1973). Once in
cells, both heptanoate and C5 ketone molecules
are metabolized to their CoA adducts via medium
chain acyl-CoA synthetases or 3-oxoacid CoA
transferase, respectively, and then converted into
the main TCA cycle fuel acetyl-CoA and propio-
nyl-CoA. After carboxylation the latter can refill
the TCA cycle with succinyl-CoA (see above) and
thus promote acetyl-CoA oxidation and ATP pro-
duction (Figure 34.2).
E N E R G Y M E TA B O L I S M
IN EPILEPSY
An epileptic seizure or hypersynchronous acti-
vation of neuronal networks can be initiated by
imbalances between excitation and inhibition. In
addition, impairments in energy metabolism can
also cause seizures and/or contribute to epilepsy.
Numerous studies have attempted to shed light
on energy metabolism in brains of patients with
epilepsy and rodent epilepsy models. Assessments
of metabolic functions or metabolite levels have
revealed dysfunction in energy metabolism in
patients with temporal lobe and extra-temporal
lobe epilepsy (reviewed in Li et al., 2000; Pan et al.,
2005; Pan et al., 2008). Similar changes have been
found in rodent epilepsy models (Alvestad et  al.,
2008; Melo et al., 2010; Melo et al., 2005; Smeland
et al., 2013; Willis et al., 2010) (Figure 34.3).
Furthermore, mitochondrial dysfunction and
mutations within mitochondrial constituents have
been described (Kann and Kovacs, 2007; Kudin
et al., 2009; Waldbaum and Patel, 2010), which can
also contribute to energetic imbalances contribut-
ing to seizures.
Many studies in the literature discuss interictal
glucose hypometabolism in patients with temporal
338

Glucose

Pyruvate

CO2 1 Acetyl~CoA
Glutamine
Ala
3
Citrate
Glutamate
OAA 2
Aspartate GABA
2-OG OAA
4

Malate Aspartate

Fumarate Succinyl~CoA Heptanoate,

C5 ketones,
Succinate Ile, Val
Methylmalonyl~CoA
5 CO2
Propionyl~CoA

FIGURE 34.2 The TCA cycle and anaplerotic pathways. The numbers indicate five different anaplerotic pathways.
1 Pyruvate carboxylase forms oxaloactetate by carboxylastion of pyruvate. 2 denotes the activity of glutamate
dehydrogenase, which metabolizes glutamate into 2-oxoglutarate and ammonia and vice versa. 3 shows the reversible
reaction catalyzed by glutamic pyruvic transaminases (also called alanine aminotransferases):  pyruvate + glutamate
<=>2-oxoglutarate + alanine. 4 aspartate transaminase (also called glutamic oxaloacetic transaminase) produces
oxaloacetate from aspartate, while transferring the amino group onto 2-oxoglutarate forming glutamate and vice versa.
5 denotes the activity of the propionyl-CoA carboxylation pathway forming succinyl-CoA. 2-OG—2-oxoglutarate.

(a) MEST test (b) 2nd hit PTZ

18 90
*
**
P TZ threshold (mg/kg i.v.)

80
16
70
CC50 (mA)

14
60

12 50

0 0
CON 35% Trih CON 35% Trih
6 weeks 3 weeks
(c)
Absence seizure model: GABAAR φ 2 (R43Q)

CON

0.5 mV

Trih 2s

FIGURE  34.3 Anticonvulsant effects of oral triheptanoin in three models. (a) Triheptanoin increased the critical
current at which 50% of mice seize (CC50) in the maximal electroshock threshold test (Thomas et al., 2012). (b) It also
increased the threshold to PTZ-induced tonic seizures in chronically epileptic mice (Willis et al., 2010) and (c) decreased
spike wave discharges in a mouse model for absence seizures (Kim et al., 2013).
339
Chapter 34: Triheptanoin in Epilepsy and Beyond
lobe epilepsy (TLE), which is a common epilepsy
type in adults that is largely treatment resistant.
In TLE, results from positron emission tomogra-
phy with 18F-labeled fluorodeoxyglucose (18FDG)
have been interpreted to show interictal glucose
hypometabolism (Arnold et al., 1996; Henry et al.,
1993; Henry et al., 1990). However, care needs to
be taken when evaluating these studies, as results
only revealed decreased 18FDG amounts in cer-
tain brain areas, which reflects enhanced uptake of
18
FDG. Levels of 18FDG cannot be used to assess
glucose metabolism, because after it is metabo-
lized by hexokinase 18FDG cannot be further
metabolized via glycolysis. Similar to these stud-
ies in epilepsy patients, decreased uptake of 14C-
deoxyglucose was found in the chronic phase of
the lithium-pilocarpine adult rat model of TLE
(Dube et al., 2001). Higher total glucose amounts
were found in extracts from brain areas involved
in seizure activity compared to tissue from con-
trol rats (Melo et al., 2005). However, total glucose
amounts by themselves are difficult to interpret, as
they depend on uptake and rates of metabolism
and no alterations in total glucose amounts were
found in the chronic stage of the mouse pilocar-
pine model (Smeland et  al., 2013). In summary,
decreased glucose uptake appears to be common
in some “epileptic” tissues, but there is little knowl-
edge about the metabolism of this major fuel.
Given the anticonvulsant efficacy of dietary treat-
ments, such as the ketogenic and modified Atkins
diets, there is a critical need for more knowledge
about glycolysis in epilepsy.
Intermediates of the TCA cycle, such as
2-oxogluarate and oxaloacetate, are precursors of
the amino acids and neurotransmitters glutamate,
GABA, and aspartate. Decreased levels of TCA
intermediates and amino acids have been found in
the chronic seizure stage of rat and mouse epilepsy
models, which display recurrent spontaneous
seizures (Alvestad et  al., 2008; Melo et  al., 2005;
Smeland et al., 2013; Willis et al., 2010). For exam-
ple, in the mouse pilocarpine epilepsy model, we
found lower forebrain levels of malate, aspartate,
and acetyl- and propionyl-CoA during the chronic
epileptic stage compared with mice without sei-
zures (Willis et  al., 2010). While several other
TCA cycle intermediates were not quantified,
TCA intermediate levels seemed to be decreased
overall, which can result in decreased oxidative
phosphorylation.
Based on this knowledge, providing additional
alternative anaplerotic fuel, which can increase
the amounts of C4 intermediates of the TCA cycle
339
in the brain, is a biochemically valid approach to
improve energy metabolism in “epileptic” brains
(Brunengraber and Roe, 2006). Thus, we tested
the metabolic effects and anticonvulsant profile of
triheptanoin treatment as an anaplerotic approach,
which at that time had already been explored in
a few animal models and in patients with various
genetic metabolic disorders. Please also note two
reviews on the effects of other approaches to sup-
plement TCA cycle substrates (Kovac et al., 2013;
Tan et al., 2015).
T R I H E P TA N O I N
A LT E R S B R A I N E N E R G Y
M E TA B O L I S M I N   VA R I O U S
SETTINGS, INCLUDING
EPILEPSY MODELS
Few publications shed light on the metabolic
effects of triheptanoin or its metabolite heptanoate
on the brain.
An elegant study by the group of Professor
Pascual (Marin-Valencia et  al., 2013), infused
13
C-labeled 5,6,7-heptanoate into the jugular vein
of mice. The 13C-carbons were largely found in
brain glutamine, but not glutamate, indicating that
astrocytes primarily metabolize heptanoate and its
C5 ketone metabolites.
McDonald et  al. (2014) compared the
amounts of various hippocampal metabolites
within energy metabolism pathways (adeno-
sine nucleotides, NAD+, NADH, and NADPH),
glycolysis, the pentose phosphate pathway, and
the TCA cycle in healthy mice after feeding tri-
heptanoin and trioctanoin (the triglyceride of
octanoate) for 3 weeks. While the even medium
chain triglyceride altered the levels of various
metabolites, no significant changes were found
with triheptanoin, indicating that it would have
few metabolic side effects.
A recent study in patients with Huntington’s
disease patients found that 1-month triheptanoin
treatment corrected the energetic response to
brain activation (Adanyeguh et  al., 2015). In
untreated Huntington’s disease patients, a visual
stimulus fails to alter the ratio of inorganic phos-
phate to phosphocreatine, while healthy control
and treated patients showed an increase in this
ratio, indicating that triheptanoin could restore
metabolism of high-energy phosphates.
Two studies (Hadera et al., 2014; Willis et
al., 2010) aimed to increase the understanding
of the metabolic effects of triheptanoin treat-
ment in mouse “epileptic” brain tissue. In the
pilocarpine model, mice that experience status
340
340 section IV: Ketone-Based Metabolism
epilepticus (SE) show one motor seizure per day
on average in the chronic stage of the model,
while mice that did not get SE (no SE) do not
show any seizure activity or neuropathological
changes (Benson et al., 2015; Borges et al., 2003).
Willis et al. (2010) compared the levels of various
forebrain metabolites related to the TCA cycle in
the chronic stage of this model. Forebrain lev-
els of malate and propionyl-CoA were reduced
in SE versus no SE mice on control diet, but this
was prevented by triheptanoin feeding, indicat-
ing that triheptanoin can improve a low capacity
of the TCA cycle. Hadera et al. (2014) followed
the metabolism of [1,2-13C]glucose in the same
mouse model to be able to evaluate the activities
of pyruvate dehydrogenase and pyruvate carbox-
ylase, the latter of which provides C4 carbons to
the TCA cycle. The analyses of 13C label incor-
poration into TCA cycle intermediates showed
that the percentage of enrichment for two 13C
atoms in malate, citrate, succinate, and GABA
was reduced in control treated SE mice versus no
SE mice. Except for succinate, triheptanoin feed-
ing alleviated these reductions in SE mice, which
provides additional evidence that triheptanoin
can increase the metabolism of glucose via the
TCA cycle. The author believes that an adequate
ATP supply is very important in “epileptic” tis-
sue to keep neuronal membrane potentials stable
and prevent seizure generation. Taken together,
all studies are consistent with the brain being
able to metabolize heptanoate or “C5 ketones,”
and improvements in cases of impaired energy
metabolism could be detected.
ANTICONVULSANT EFFECTS
O F   T R I H E P TA N O I N A N D
CLINICAL TRIALS
Triheptanoin shows a unique anticonvulsant pro-
file. It was found to be anticonvulsant in various
mouse seizure models (McDonald et al., 2014;
Thomas et al., 2012; Willis et al., 2010) (Figure
34.3), with efficacy in three chronic mouse sei-
zure models. In the chronic corneal kindling
model we found a reproducible delay in the kin-
dling process in CF1 mice (Willis et al., 2010),
which is similar to effects found with established
anticonvulsant drugs in the rat kindling model,
namely phenobarbital and low concentrations
of valproate (Brandt et al., 2006; Matagne et al.,
2008; Silver et al., 1991). Protective effects of var-
ious compounds in kindling models correlate
well with efficacy in humans against absence sei-
zures. Also, in a mouse model for genetic absence
seizures (Tan et al., 2007), namely mice with a
missense (R43Q) mutation in the GABAA recep-
tor γ2 subunit, triheptanoin decreased the num-
ber and duration of spike wave discharges (SWD),
resulting in a halved time with seizures (Kim et
al., 2013) (Figure 34.3C). In chronically epileptic
mice after pilocarpine-induced SE, triheptanoin
reproducibly increased the pentylenetetrazole
(PTZ) seizure threshold (Figure 34.3B). Efficacy
in this model suggests efficacy against drug-
resistant seizures, based on the finding that a sim-
ilar second hit rat model is resistant to valproate,
phenytoin, and phenobarbital (Blanco et al., 2009;
Borges and Sonnewald, 2011). In the acute max-
imal electroshock threshold test (MEST), a test
for the efficacy against generalized seizures, we
observed a reproducible increase of the critical
current at which 50% of mice seize (Thomas et al.,
2012) (CC50, Fig 34.3A). Effects in other acute
seizure models have been variable, for example,
in the 6-Hz model the threshold to motor sei-
zures was only elevated in some experiments, but
not others (McDonald et al., 2014; Thomas et al.,
2012). Similarly, anticonvulsant effects in models
using the GABAA receptor channel blockers, fluo-
rothyl or PTZ, have been inconsistent (Thomas et
al., 2012). This lack of consistent effects in acute
mouse models is not surprising, because energy
metabolism in healthy mice is likely to be opti-
mal and unlikely to require additional fuel and/
or anaplerosis. In summary, triheptanoin’s anti-
convulsant profile suggests efficacy against a vari-
ety of seizure types, including focal and general
motor seizures, absence seizures, and possibly
pharmacoresistant motor seizures (Smith et al.,
2007; White, 2003).
Currently, there are three clinical trials of tri-
heptanoin in adult and children with medically
refractory epilepsy in Australia (http://www.
anzctr.org.au/). The first phase IIa randomized
double-blind placebo controlled study to evaluate
the safety and tolerability of oral triheptanoin as an
add-on treatment to adolescent and adult patients
with medically refractory epilepsy took place with
Professor O’Brien as the lead doctor at the Royal
Melbourne Hospital. Results are currently being
analyzed. Patients who took part in this trial have
been invited to continue in an open-label exten-
sion study of oral triheptanoin as an add-on treat-
ment. This study recruited patients in 2015/2016
and is going well. In addition, a trial in Brisbane is
evaluating the effects of triheptanoin in children
ages 3–18 with medically refractory epilepsy and
has stopped recruiting.
341
Chapter 34: Triheptanoin in Epilepsy and Beyond
T R I H E P TA N O I N A N D
OT H E R D I S O R D E R S
Metabolic Disorders: Glucose Transporter
Type 1 Deficiency, Long Chain Fatty
Acid Oxidation Disorders, Pyruvate
Decarboxylase Deficiency
While this review focused on epilepsies of vari-
ous or unknown etiologies, triheptanoin is also
being developed as a new treatment for D-glu-
cose transporter type 1 deficiency (Pascual et al.,
2014). In this genetic disorder glucose cannot be
taken up into the brain, which results in epilep-
tic seizures around age 3–4 and later in paroxys-
mal exercise-induced dyskinesias, which can be
very disabling (Klepper and Leiendecker, 2007;
Suls et al., 2009). The treatments for this disor-
der are discussed in an other chapter in this book
(Klepper, chapter 5).
Originally, triheptanoin was given to patients
with rare inborn metabolic disorders, including
long chain fatty acid oxidation disorders due to
deficiencies in different enzymes, namely car-
nitine palmitoyltransferase I  or II, very long-
chain acyl-CoA dehydrogenase, L-3-hydroxy
acyl-CoA dehydrogenase, and mitochondrial
trifunctional protein (Roe et al., 2002; Roe et al.,
2008). The use of dietary or body fat as energy
sources is preempted in these disorders, which
can result in severe clinical problems, such as
cardiomyopathy, intermittent rhabdomyolysis,
hypoglycemia, and sudden death. Treatment with
traditional medium chain triglycerides has been
unsatisfactory in the past, while triheptanoin as
an alternatively medium chain and anaplerotic
fuel, which is also gluconeogenic, can decrease
hospitalizations due to major clinical manifesta-
tions and incidences in hypoglycemia (Roe and
Brunengraber, 2015; Roe and Mochel, 2006; Roe
et al., 2002; Vockley et al., 2015). Another meta-
bolic disorder that may benefit from triheptanoin
is pyruvate carboxylase deficiency (Breen et  al.,
2014; Mochel et al., 2005).
Other Disorders: Neurological
Conditions, Muscle Disorders,
Cardiac Hypertrophy
Triheptanoin appears to be a promising treat-
ment for other neurological disorders, including
Canavan disease, stroke, ALS, Alzheimer’s and
Huntington’s disease, autism, and glycogenoses. In
three neurological disorders models, oral trihep-
tanoin feeding was found to prevent neuronal cell
death, and it is conceivable that increased energy
levels can protect cells from degeneration.
341
The pediatric leukodystrophy Canavan dis-
ease is caused by mutations in aspartoacylase, an
enzyme that is largely found in oligodendrocytes
and catalyzes the hydrolysis of neuronally derived
N-acetylaspartate to provide acetyl groups for lipid
synthesis. In nur7 mutant mice containing a non-
sense mutation in the aspartoacylase gene, early
triheptanoin treatment prevented the loss of oli-
godendrocytes, dysmyelination, and impairments
in motor function (Francis et al., 2014). The early
loss of various brain metabolites in this model,
such as ATP and acetyl-, malonyl-, and propionyl-
CoA, was largely prevented, indicating that tri-
heptanoin can rescue metabolic deficits also in
oligodendrocytes.
The brain infarct area after middle cere-
bral artery occlusion (MCAO), a mouse model
of stroke, was smaller when mice were fed tri-
heptanoin before the insult (Schwarzkopf et  al.,
2015). In addition, mitochondrial functions were
preserved in mitochondria isolated from brains
1 hour after onset of MCAO (Schwarzkopf et al.,
2015), indicating that improved mitochondrial
energetic functions contribute to triheptanoin’s
neuroprotective effects.
Similarly, in the ALS mouse model overex-
pressing the human SOD1 G93A mutation, tri-
heptanoin treatment resulted in 33% less motor
neuron loss in the L4-L5 spinal cord, and loss of
motor functions was significantly delayed (Tefera
et al. PLOS One, in press).
A study in an Alzheimer’s disease model indi-
cated that triheptanoin in the context of a keto-
genic diet increased the expression of the mRNA
levels of Sirt1, Pparg, Sod1, and Sod2 (Aso et  al.,
2013). As sirtuin 1 and PPARγ are involved in the
regulation of lipid and glucose metabolism as well
as mitochondrial respiration and oxidative stress,
this suggests that the neuroprotective effects of
triheptanoin may also include other mechanisms
that can improve energy metabolism, such as
reduced oxidative stress and preserved mitochon-
drial functions.
Adult polyglucosan body disease (APBD) is a
rare autosomal recessive disorder due to partial
deficiency of the glycogen brancher enzyme (GBE)
and leads to deposition of polyglucosan bodies in
neurons and glia. Similar to other glycogenoses,
for example, Pompe’s disease, in which glycogen
cannot be metabolized, it is thought that the pro-
gressive deposition of polysaccharides causes cells
to be disrupted. The APBD patients show gradual
progression with difficulty walking, impaired bal-
ance, neurogenic bladder, weakness, and, in about
half of the cases, dementia. Five APBD patients
342
342 section IV: Ketone-Based Metabolism
treated with triheptanoin experienced stabiliza-
tion of disease progression or some functional
improvement (Roe et al., 2010).
There is increasing hope that Huntington’s
disease patients might benefit from long-term
triheptanoin therapy, as there is evidence that tri-
heptanoin improves energy metabolism in patients
in the CNS and also skeletal muscle (Adanyeguh
et  al., 2015; Mochel et  al., 2010). Moreover,
improvements in muscle function were found in
skeletal muscle in a Rett syndrome model (Park
et  al., 2014)  and rat heart (Nguyen et  al., 2015).
In Sydney, a pilot clinical study is currently taking
place with patients with inclusion body myositis
and Pompe’s disease (http://www.anzctr.org.au/;
Corbett et al., 2015).
Among the autism spectrum disorders, Rett
syndrome is a genetic disorder largely caused
by mutations in the X-linked gene for the tran-
scription factor methyl-CpG binding protein 2
(MeCP2), resulting in severe disability regarding
cognitive and motor function. In male MeCP2-
deficient mice triheptanoin increased life span and
improved social interaction and motor function
as well as mitochondrial morphology in skeletal
muscle (Park et  al., 2014). The mutant mice also
showed increased adiposity and lower glucose tol-
erance and insulin sensitivity, which all improved
with triheptanoin treatment, indicating that this
medium chain triglyceride can counteract obesity
and normalize glucose metabolism.
It had been known for some time that propio-
nate is an anaplerotic substrate for the heart, for
example (Martini et al., 2003). Recently, the effects
of triheptanoin in a rat model of cardiac hypertro-
phy were assessed. A 30% triheptanoin diet reduced
left ventricular hypertrophy and improved diastolic
function and myocardial glucose oxidation (Nguyen
et  al., 2015), indicating that anaplerotic actions of
heptanoate or C5 ketones can benefit the ailing heart.
Studies on the effect of triheptanoin on altera-
tions in fatty acid composition have just begun.
Traditionally, it was thought that medium chain
fats do not get elongated and subsequently stored.
However, in mice, feeding of even and uneven
MCT for 1 year altered the fatty acid profile of liver
and heart significantly, with unusual fatty acids
becoming detectable, for example, C15, C17:1, and
C22:1 with triheptanoin and C22:5n3 with trioc-
tanoin (Tucci et  al., 2015). The potential impacts
of these changes remain to be explored.
CONCLUSIONS
Triheptanoin was first used as a novel alternative
approach to satisfy energy needs in patients with
rare metabolic enzyme deficiencies, and many
patients with fatty acid oxidation disorders depend
on its availability. Within the last decade new
research has shown promising effects of trihep-
tanoin in a variety of human disorders and their
animal models, including epilepsy. Larger-scale
controlled clinical trials are now needed to prove
triheptanoin’s safety and tolerability, and there is
hope that a variety of beneficial effects reported so
far can be confirmed.
DISCLOSURE/CONFLICT
OF INTEREST
KB applied for two full US patents on the use of
triheptanoin for epilepsy and ALS.
AC K N OW L E D G M E N T S
I am grateful for funding by the American
Epilepsy Foundation, Parents against Childhood
Epilepsy, UniQuest, the Thrasher Research Fund,
and Ultragenyx Pharmaceuticals, Inc., for funding
ongoing clinical trials of triheptanoin in adults and
children with treatment resistant epilepsy. I  also
thank the Australian National Health and Medical
Research Council (grant 1044407) for funding my
laboratory research.
REFERENCES
Adanyeguh, I.M., Rinaldi, D., Henry, P.G., Caillet, S.,
Valabregue, R., Durr, A., and Mochel, F. (2015).
Triheptanoin improves brain energy metabolism
in patients with Huntington disease. Neurology
84, 490–495.
Alvestad, S., Hammer, J., Eyjolfsson, E., Qu, H.,
Ottersen, O.P., and Sonnewald, U. (2008). Limbic
structures show altered glial-neuronal metabolism
in the chronic phase of kainate induced epilepsy.
Neurochem Res 33, 257–266.
Arnold, S., Schlaug, G., Niemann, H., Ebner, A.,
Luders, H., Witte, O.W., and Seitz, R.J. (1996).
Topography of interictal glucose hypometabolism
in unilateral mesiotemporal epilepsy. Neurology
46, 1422–1430.
Aso, E., Semakova, J., Joda, L., Semak, V., Halbaut, L.,
Calpena, A., Escolano, C., Perales, J.C., and Ferrer,
I. (2013). Triheptanoin supplementation to keto-
genic diet curbs cognitive impairment in APP/PS1
mice used as a model of familial Alzheimer’s dis-
ease. Curr Alzheimer Res 10, 290–297.
Benson, M.J., Thomas, N.K., Talwar, S., Hodson, M.P.,
Lynch, J.W., Woodruff, T.M., and Borges, K. (2015).
A novel anticonvulsant mechanism via inhibition
of complement receptor C5ar1 in murine epilepsy
models. Neurobiol Dis 76, 87–97.
Blanco, M.M., dos Santos, J.G., Jr., Perez-Mendes,
P., Kohek, S.R., Cavarsan, C.F., Hummel, M.,
Albuquerque, C., and Mello, L.E. (2009).
343
Chapter 34: Triheptanoin in Epilepsy and Beyond
Assessment of seizure susceptibility in pilocarpine
epileptic and nonepileptic Wistar rats and of sei-
zure reinduction with pentylenetetrazole and elec-
troshock models. Epilepsia 50, 824–831.
Borges, K., Gearing, M., McDermott, D.L., Smith, A.B.,
Almonte, A.G., Wainer, B.H., and Dingledine, R.
(2003). Neuronal and glial pathological changes
during epileptogenesis in the mouse pilocarpine
model. Exp Neurol 182, 21–34.
Borges, K., and Sonnewald, U. (2011). Triheptanoin-
A medium chain triglyceride with odd chain fatty
acids:  A  new anaplerotic anticonvulsant treat-
ment? Epilepsy Res. 100, 239–44.
Brandt, C., Heile, A., Potschka, H., Stoehr, T., and
Loscher, W. (2006). Effects of the novel antiepilep-
tic drug lacosamide on the development of amyg-
dala kindling in rats. Epilepsia 47, 1803–1809.
Breen, C., White, F.J., Scott, C.A., Heptinstall, L.,
Walter, J.H., Jones, S.A., and Morris, A.A. (2014).
Unsuccessful treatment of severe pyruvate carbox-
ylase deficiency with triheptanoin. Eur J Pediatr
173, 361–366.
Broer, S., Schneider, H.P., Broer, A., Rahman,
B., Hamprecht, B., and Deitmer, J.W. (1998).
Characterization of the monocarboxylate trans-
porter 1 expressed in Xenopus laevis oocytes by
changes in cytosolic pH. Biochem J 333 (Pt 1),
167–174.
Brunengraber, H., and Roe, C.R. (2006). Anaplerotic
molecules: current and future. J Inherit Metab Dis
29, 327–331.
Corbett, A., Garg, N., Crosbie, A., and Borges, K.
(2015). A pilot study of triheptanoin treatment
in sporadic inclusion body myositis (sIBM). Ann
Neurol 78, S102.
Deng, S., Zhang, G.F., Kasumov, T., Roe, C.R., and
Brunengraber, H. (2009). Interrelations between
C4 ketogenesis, C5 ketogenesis, and anaplero-
sis in the perfused rat liver. J Biol Chem 284,
27799–27807.
Dube, C., Boyet, S., Marescaux, C., and Nehlig, A. (2001).
Relationship between neuronal loss and interictal
glucose metabolism during the chronic phase of the
lithium-pilocarpine model of epilepsy in the imma-
ture and adult rat. Exp Neurol 167, 227–241.
Francis, J.S., Markov, V., and Leone, P. (2014). Dietary
triheptanoin rescues oligodendrocyte loss, dysmy-
elination and motor function in the nur7 mouse
model of Canavan disease. J Inherit Metab Dis 37,
369–381.
Gu, L., Zhang, G.F., Kombu, R.S., Allen, F., Kutz, G.,
Brewer, W.U., Roe, C.R., and Brunengraber, H.
(2010). Parenteral and enteral metabolism of
anaplerotic triheptanoin in normal rats: II. Effects
on lipolysis, glucose production, and liver acyl-
CoA profile. Am J Physiol Lung Cell Mol Physiol
298, E362–371.
Haberg, A.K., Qu, H., and Sonnewald, U. (2009).
Acute changes in intermediary metabolism in
343
cerebellum and contralateral hemisphere fol-
lowing middle cerebral artery occlusion in rat. J
Neurochem 109 (Suppl 1), 174–181.
Hadera, M.G., Smeland, O.B., McDonald, T.S., Tan,
K.N., Sonnewald, U., and Borges, K. (2014).
Triheptanoin partially restores levels of tricar-
boxylic acid cycle intermediates in the mouse
pilocarpine model of epilepsy. J Neurochem 129,
107–119.
Hassel, B. (2000). Carboxylation and anaplerosis in
neurons and glia. MolNeurobiol 22, 21–40.
Henry, T.R., Mazziotta, J.C., and Engel, J., Jr. (1993).
Interictal metabolic anatomy of mesial temporal
lobe epilepsy. Arch Neurol 50, 582–589.
Henry, T.R., Mazziotta, J.C., Engel, J., Jr., Christenson,
P.D., Zhang, J.X., Phelps, M.E., and Kuhl, D.E.
(1990). Quantifying interictal metabolic activity in
human temporal lobe epilepsy. J Cereb Blood Flow
Metab 10, 748–757.
Kann, O., and Kovacs, R. (2007). Mitochondria and
neuronal activity. Am J PhysiolCell Physiol 292,
C641–657.
Kim, T.H., Borges, K., Petrou, S., and Reid, C.A.
(2013). Triheptanoin reduces seizure susceptibility
in a syndrome-specific mouse model of general-
ized epilepsy. Epilepsy Res 103, 101–105.
Kinman, R.P., Kasumov, T., Jobbins, K.A., Thomas,
K.R., Adams, J.E., Brunengraber, L.N., Kutz, G.,
Brewer, W.U., Roe, C.R., and Brunengraber, H.
(2006). Parenteral and enteral metabolism of
anaplerotic triheptanoin in normal rats. Am J
Physiol Endocrinol Metab 291, E860–866.
Klepper, J., and Leiendecker, B. (2007). GLUT1 defi-
ciency syndrome--2007 update. Dev Med Child
Neurol 49, 707–716.
Kornberg, H.L. (1966). Anaplerotic sequences and
their role in metabolism. Essays Biochem 2, 1–31.
Kovac, S., Abramov, A.Y., and Walker, M.C. (2013).
Energy depletion in seizures: anaplerosis as a strat-
egy for future therapies. Neuropharmacology 69,
96–104.
Kudin, A.P., Zsurka, G., Elger, C.E., and Kunz, W.S.
(2009). Mitochondrial involvement in temporal
lobe epilepsy. Exp Neurol 218, 326–332.
Li, L.M., Cendes, F., Andermann, F., Dubeau, F.,
and Arnold, D.L. (2000). Spatial extent of neu-
ronal metabolic dysfunction measured by pro-
ton MR spectroscopic imaging in patients
with localization-related epilepsy. Epilepsia 41,
666–674.
Marin-Valencia, I., Good, L.B., Ma, Q., Malloy, C.R.,
and Pascual, J.M. (2013). Heptanoate as a neural
fuel:  energetic and neurotransmitter precursors
in normal and glucose transporter I-deficient
(G1D) brain. J Cereb Blood Flow Metab 33,
175–182.
Martini, W.Z., Stanley, W.C., Huang, H., Rosiers,
C.D., Hoppel, C.L., and Brunengraber, H. (2003).
Quantitative assessment of anaplerosis from
344
344 section IV: Ketone-Based Metabolism
propionate in pig heart in vivo. Am J Physiol
Endocrinol Metab 284, E351–356.
Matagne, A., Margineanu, D.G., Kenda, B., Michel,
P., and Klitgaard, H. (2008). Anti-convulsive and
anti-epileptic properties of brivaracetam (ucb
34714), a high-affinity ligand for the synaptic vesi-
cle protein, SV2A. Br J Pharmacol 154, 1662–1671.
McDonald, T.S., Tan, K.N., Hodson, M.P., and Borges,
K. (2014). Alterations of hippocampal glucose
metabolism by even versus uneven medium
chain triglycerides. J Cereb Blood Flow Metab 34,
153–160.
Melo, T., Bigini, P., Sonnewald, U., Balosso, S.,
Cagnotto, A., Barbera, S., Uboldi, S., Vezzani, A.,
and Mennini, T. (2010). Neuronal hyperexcit-
ability and seizures are associated with changes in
glial-neuronal interactions in the hippocampus of
a mouse model of epilepsy with mental retarda-
tion. J Neurochem 115, 1445–1454.
Melo, T.M., Nehlig, A., and Sonnewald, U. (2005).
Metabolism is normal in astrocytes in chronically
epileptic rats: a (13)C NMR study of neuronal-glial
interactions in a model of temporal lobe epilepsy. J
Cereb Blood Flow Metab 25, 1254–1264.
Meredith, D., and Christian, H.C. (2008). The SLC16
monocarboxylate transporter family. Xenobiotica
38, 1072–1106.
Mochel, F., DeLonlay, P., Touati, G., Brunengraber, H.,
Kinman, R.P., Rabier, D., Roe, C.R., and Saudubray,
J.M. (2005). Pyruvate carboxylase deficiency: clin-
ical and biochemical response to anaplerotic diet
therapy. Mol Gen Metab 84, 305–312.
Mochel, F., Duteil, S., Marelli, C., Jauffret, C., Barles,
A., Holm, J., Sweetman, L., Benoist, J.F., Rabier,
D., Carlier, P.G., et al. (2010). Dietary anaplerotic
therapy improves peripheral tissue energy metab-
olism in patients with Huntington’s disease. Eur J
Hum Genet 18, 1057–1060.
Nguyen, T.D., Shingu, Y., Amorim, P.A., Schwarzer,
M., and Doenst, T. (2015). Triheptanoin alleviates
ventricular hypertrophy and improves myocardial
glucose oxidation in rats with pressure overload. J
Card Fail. 21, 906–15.
Niessen, H.G., Debska-Vielhaber, G., Sander, K.,
Angenstein, F., Ludolph, A.C., Hilfert, L., Willker,
W., Leibfritz, D., Heinze, H.J., Kunz, W.S., et  al.
(2007). Metabolic progression markers of neu-
rodegeneration in the transgenic G93A-SOD1
mouse model of amyotrophic lateral sclerosis. Eur
J Neurosci 25, 1669–1677.
Nuutinen, E.M., Peuhkurinen, K.J., Pietilainen,
E.P., Hiltunen, J.K., and Hassinen, I.E. (1981).
Elimination and replenishment of tricarboxylic
acid-cycle intermediates in myocardium. Biochem
J 194, 867–875.
Oldendorf, W.H. (1973). Carrier-mediated blood-
brain barrier transport of short-chain mono-
carboxylic organic acids. Am J Physiol 224,
1450–1453.
Owen, O.E., Kalhan, S.C., and Hanson, R.W. (2002).
The key role of anaplerosis and cataplerosis for
citric acid cycle function. J Biol Chem 277,
30409–30412.
Pan, J.W., Kim, J.H., Cohen-Gadol, A., Pan, C.,
Spencer, D.D., and Hetherington, H.P. (2005).
Regional energetic dysfunction in hippocampal
epilepsy. Acta Neurol Scand 111, 218–224.
Pan, J.W., Williamson, A., Cavus, I., Hetherington,
H.P., Zaveri, H., Petroff, O.A., and Spencer, D.D.
(2008). Neurometabolism in human epilepsy.
Epilepsia 49 (Suppl 3), 31–41.
Park, M.J., Aja, S., Li, Q., Degano, A.L., Penati, J.,
Zhuo, J., Roe, C.R., and Ronnett, G.V. (2014).
Anaplerotic triheptanoin diet enhances mitochon-
drial substrate use to remodel the metabolome and
improve lifespan, motor function, and sociability
in MeCP2-null mice. PloS One 9, e109527.
Pascual, J.M., Liu, P., Mao, D., Kelly, D.I., Hernandez,
A., Sheng, M., Good, L.B., Ma, Q., Marin-Valencia,
I., Zhang, X., et  al. (2014). Triheptanoin for glu-
cose transporter type I deficiency (G1D): modula-
tion of human ictogenesis, cerebral metabolic rate,
and cognitive indices by a food supplement. JAMA
Neurology 71, 1255–1265.
Patel, M.S. (1974). The relative significance of CO2-
fixing enzymes in the metabolism of rat brain. J
Neurochem 22, 717–724.
Reszko, A.E., Kasumov, T., Pierce, B.A., David, F.,
Hoppel, C.L., Stanley, W.C., Des Rosiers, C., and
Brunengraber, H. (2003). Assessing the reversibil-
ity of the anaplerotic reactions of the propionyl-
CoA pathway in heart and liver. J Biol Chem 278,
34959–34965.
Roe, C.R., Bottiglieri, T., Wallace, M., Arning, E.,
and Martin, A. (2010). Adult polyglucosan
body disease (APBD):  anaplerotic diet therapy
(triheptanoin) and demonstration of defective
methylation pathways. Mol Gen Metabol 101,
246–252.
Roe, C.R., and Brunengraber, H. (2015). Anaplerotic
treatment of long-chain fat oxidation disorders
with triheptanoin: Review of 15 years experience.
Mol Gen Metabol. 116, 260–8.
Roe, C.R., and Mochel, F. (2006). Anaplerotic diet
therapy in inherited metabolic disease: therapeu-
tic potential. J Inherit Metab Dis 29, 332–340.
Roe, C.R., Sweetman, L., Roe, D.S., David, F., and
Brunengraber, H. (2002). Treatment of cardiomy-
opathy and rhabdomyolysis in long-chain fat oxi-
dation disorders using an anaplerotic odd-chain
triglyceride. J Clin Invest 110, 259–269.
Roe, C.R., Yang, B.Z., Brunengraber, H., Roe, D.S.,
Wallace, M., and Garritson, B.K. (2008). Carnitine
palmitoyltransferase II deficiency:  successful
anaplerotic diet therapy. Neurology 71, 260–264.
Schwarzkopf, T.M., Koch, K., and Klein, J. (2015).
Reduced severity of ischemic stroke and improve-
ment of mitochondrial function after dietary
345
Chapter 34: Triheptanoin in Epilepsy and Beyond
treatment with the anaplerotic substance trihep-
tanoin. Neuroscience 300, 201–209.
Silver, J.M., Shin, C., and McNamara, J.O. (1991).
Antiepileptogenic effects of conventional anticon-
vulsants in the kindling model of epilepsy. Ann
Neurol 29, 356–363.
Smeland, O.B., Hadera, M.G., McDonald, T.S.,
Sonnewald, U., and Borges, K. (2013). Brain mito-
chondrial metabolic dysfunction and glutamate
level reduction in the pilocarpine model of tem-
poral lobe epilepsy in mice. J Cereb Blood Flow
Metab 33, 1090–1097.
Smith, M., Wilcox, K.S., and White, H.S.
(2007). Discovery of antiepileptic drugs.
Neurotherapeutics 4, 12–17.
Suls, A., Mullen, S.A., Weber, Y.G., Verhaert, K.,
Ceulemans, B., Guerrini, R., Wuttke, T.V., Salvo-
Vargas, A., Deprez, L., Claes, L.R., et  al. (2009).
Early-onset absence epilepsy caused by mutations
in the glucose transporter GLUT1. Ann Neurol 66,
415–419.
Tan, H.O., Reid, C.A., Single, F.N., Davies, P.J., Chiu,
C., Murphy, S., Clarke, A.L., Dibbens, L., Krestel,
H., Mulley, J.C., et  al. (2007). Reduced cortical
inhibition in a mouse model of familial childhood
absence epilepsy. Proc Natl Acad Sci USA 104,
17536–17541.
Tan, K.N., McDonald, T.S., and Borges, K. (2015).
Metabolic dysfunctions in epilepsy and novel
345
metabolic treatment approaches. In Bioactive
Nutraceuticals and Dietary Supplements in
Neurological and Brain Disease:  Prevention
and Therapy, R.R. Watson and V.R. Preedy, eds.
(Amsterdam: Elsevier), pp. 461–470.
Thomas, N.K., Willis, S., Sweetman, L., and Borges, K.
(2012). Triheptanoin in acute mouse seizure mod-
els. Epilepsy Res 99, 312–317.
Tucci, S., Behringer, S., and Spiekerkoetter, U. (2015).
De novo fatty acid biosynthesis and elongation
in very long-chain acyl-CoA dehydrogenase-
deficient mice supplemented with odd or even
medium-chain fatty acids. FEBS J 282, 4242–4253.
Vockley, J., Marsden, D., McCracken, E., DeWard, S.,
Barone, A., Hsu, K., and Kakkis, E. (2015). Long-
term major clinical outcomes in patients with long
chain fatty acid oxidation disorders before and
after transition to triheptanoin treatment-A retro-
spective chart review. Mol Gen Metab. 116, 53–60
Waldbaum, S., and Patel, M. (2010). Mitochondria,
oxidative stress, and temporal lobe epilepsy.
Epilepsy Res 88, 23–45.
White, H.S. (2003). Preclinical development of antiep-
ileptic drugs: past, present, and future directions.
Epilepsia 44 (Suppl 7), 2–8.
Willis, S., Stoll, J., Sweetman, L., and Borges, K. (2010).
Anticonvulsant effects of a triheptanoin diet in two
mouse chronic seizure models. Neurobiol Dis 40,
565–572.
346
35
Amino Acids in the Treatment
of Neurological Disorders
A D A M L . H A R T M A N, M D
INTRODUCTION
Studies of the therapeutic mechanisms of nutrient-
based therapies mainly focus on fats and carbo-
hydrates (Masino and Rho, 2012). The role of
protein, the third major diet component, has not
been studied as thoroughly as fats and carbohy-
drate, but its relevance was demonstrated by the
finding that protein must be restricted for systemic
ketosis to occur in the ketogenic diet (Laeger et al.,
2014; Yudkoff et  al., 2007). The building blocks
of proteins, amino acids, have been studied in a
variety of neurological disorders. This chapter
focuses on the therapeutic use of naturally occur-
ring proteinogenic amino acids (and some of their
D-enantiomers) for two model neurological disor-
ders, epilepsy and traumatic brain injury. Organic
acids (e.g., taurine) and modified/synthetic amino
acids are not discussed.
SEIZURES AND EPILEPSY
Seizures and epilepsy (the propensity for recurrent
unprovoked seizures) represent one of the most
common neurological disorders in all age groups.
Current treatments still leave 20-30% of patients
without adequate seizure control. This has fueled
interest in drug discovery for epilepsy, including
amino acids.
L-Amino Acids
L-leucine and L-isoleucine (300 mg/kg, a high
dose) protect against seizure onset by prolong-
ing the latency to onset of spike-wave discharges
and clinical tonic-clonic seizures induced by the
GABA-A receptor antagonist pentylenetetrazol
(PTZ); L-leucine also decreases the duration of
spike-wave discharges, but neither amino acid
affects seizure duration (Dufour et  al., 1999).
Interestingly, α-ketoisocaproic acid, a metabo-
lite of L-leucine that results from a transamina-
tion reaction that also produces L-glutamate, is
inactive against PTZ. Also, each of the branched-
chain amino acids (L-leucine, L-isoleucine, and
L-valine) increase the latency to onset of seizures
induced by another GABA-A receptor antagonist,
picrotoxin (Skeie et al., 1994).
Using a different seizure test and dosing para-
digm, we showed that pretreatment with L-leucine
terminates seizures induced by the excitotoxin
kainic acid; L-valine was ineffective, suggest-
ing that in this test, there is not a class effect of
branched-chain amino acids (Hartman et  al.,
2015). Interestingly, L-leucine was ineffective in
terminating kainic acid-induced seizures when it
was injected after seizure onset.
Glycine is somewhat unique among amino
acids in that it can have either excitatory or
inhibitory effects, depending on which receptor
is involved:  generally speaking, glycine binding
to NMDA receptors in the brain is excitatory but
binding to glycine receptors in the spinal cord is
inhibitory (Johnson and Ascher, 1987; Werman
et al., 1968), although there are exceptions to this
general rule. Inhibition of the glycine transporter
(GlyT1) protects against acute and chronic sei-
zures in rodents (Kalinichev et  al., 2010b; Shen
et al., 2015). However, binding of glycine to the N-
methyl-D-aspartate (NMDA) receptor likely does
not mediate this effect (Zellinger et al., 2014), and
in fact, inhibitors of this site have acute antiseizure
effects (Bristow et al., 1996; Nichols and Yielding,
1998). The exact role of glycine in seizure initiation
versus the progression or propagation of seizure
activity remains unclear.
These data suggest a number of potential mech-
anisms for an antiseizure effect. First, the amino
acids may be working through either cell surface
transporters or receptors, as in the case of glycine
and the GlyT1 transporter. Alternatively, they
may activate intracellular amino acid “sensors,”
with downstream effects on signaling pathways.
347
Chapter 35: Amino Acids in the Treatment of Neurological Disorders
L-leucine activates the master integrator of cellu-
lar metabolism, mammalian target of rapamycin
complex 1 (mTORC1) via Sestrin2 (Bar-Peled and
Sabatini, 2014; Cota et al., 2006; Sancak et al., 2008;
Wolfson et  al., 2016). However, data suggest that
mTORC1 activation is not the mechanism for sei-
zure control of L-leucine, because pathologically
increased mTORC1 activity has been measured
in both humans and rodent models of tuberous
sclerosis complex (Orlova and Crino, 2010; Zeng
et  al., 2008)  as well as some chemoconvulsant
rodent models of temporal lobe epilepsy; similarly,
pharmacological inhibition of mTORC1 decreases
seizures in some models (Buckmaster and Lew,
2011; Huang et al., 2010; Sosanya et al., 2015; Zeng
et  al., 2009). Thus, it is unlikely that L-leucine,
an mTORC1 activator, protects against seizures
by increasing activity in this particular pathway.
Nonetheless, the role of L-leucine in mTORC1-
related disorders is unknown. This mechanism
also does not explain the antiseizure effect of L-
isoleucine and L-valine.
Another potential mechanism for the action
of L-leucine (and other amino acids) on seizure
activity is activation via amino acid transport-
ers. Examples include LAT1/SLC7A5, which has
implications for transport of other small molecules
into the brain (Geier et  al., 2013), and B0AT2/
SLC6A15 (Broer et  al., 2006; Hagglund et  al.,
2013). Interestingly, LAT1/SLC7A5 is expressed in
pathological balloon cells in tissue from patients
with tuberous sclerosis (which supports the role of
this transporter in cell growth, one of the cardi-
nal abnormal features in this disease) (Lim et al.,
2011). An antiseizure mechanism may involve
countertransport of other amino acids, which
would in turn lead to decreased synaptic concen-
tration of excitatory amino acids and a resulting
decrease in seizure activity (Yudkoff et  al., 2007;
Yudkoff et  al., 2005). A  number of other amino
acid transporters also may transport leucine
(Box 35.1). Further testing of these transporters
(using either pharmacological inhibitors or genet-
ically modified organisms) is needed to determine
their importance in chronic epilepsy.
L-serine is less well known for a role in neuro-
transmission than its D-enantiomer. Mice lacking
the ASC1 transporter (alanine-serine-cysteine)
have seizures (Xie et al., 2005). Indicating stereo-
specific requirements, L-serine does not potentiate
NMDA-induced seizures, although D-serine, an
endogenous ligand of this receptor, does (Singh
et al., 1990a). D-serine is discussed below.
L-arginine gained attention because it is the
metabolic precursor of nitric oxide, which itself
347
BOX 35.1
NEUROLOGICALLY IMPORTANT
L- LEUCINE TRANSPORTERS
(NUMERICAL ORDER, MODIFIED
FROM GALLUS.REACTOME.ORG)
SLC3A2 (ATA2)
SLC6A14
SLC6A15
SLC7A5 (LAT1)
SLC7A6 (y+LAT2)
SLC7A7 (y+LAT1)
SLC7A9
has shown mixed results in seizure and epilepsy
studies (Banach et  al., 2011). Not surprisingly,
similarly mixed results have been shown with L-
arginine. In an electrical kindling paradigm of
epilepsy, L-arginine did not affect acquisition of
kindling or seizure severity in rats (Herberg et al.,
1995). In a developmental model of seizures, L-
arginine prolonged PTZ-induced EEG-recorded
seizure duration in postnatal day #10 (P10) rat
pups, although somewhat paradoxically, P21 rats
had increased survival (with no change in seizure
duration) (de Vasconcelos et al., 2000). The type of
seizure (i.e., tonic vs. tonic-clonic) also was differ-
ent in L-arginine-treated P21 pups compared with
controls, suggesting a form of neuromodulation
that needs further clarification (and as the authors
note, this may have an impact on survival).
D-Amino Acids
Our studies of L-leucine led us to consider
transporter-mediated effects or contaminants in
commercial preparations of L-leucine. The most
abundant other amino acid “species” in this prep-
aration was the enantiomer, D-leucine, which
represented up to 0.5% of the L-leucine in our com-
mercial source (Sigma-Aldrich technical informa-
tion). Surprisingly, D-leucine terminates the
behavioral manifestations of kainic acid–induced
seizures, even when administered after seizure
onset (Hartman et  al., 2015). As mentioned pre-
viously, L-leucine was ineffective when given after
seizure onset in this test. D-leucine (administered
in drinking water for 14 days) also protects against
seizures in the 6-Hz electroshock test, which mod-
els focal-onset seizures, demonstrating seizure
protection in a different assay. Although D-leucine
348
348 section IV: Ketone-Based Metabolism
is more ketogenic than L-leucine (i.e., its degra-
dation results in the production of ketone bodies
but not glucose) (Embden, 1908), D-leucine treat-
ment does not induce systemic ketosis (Hartman
et al., 2015). D-leucine does not bind to the kainic
acid receptor (thus eliminating competition on the
receptor as a mechanism of seizure protection),
nor does it bind to a panel of other CNS receptors
or transporters. D-leucine is a known ligand of the
taste receptors Tas1R2/R3 (which are expressed
in the hippocampus, a major seizure-generating
region of the brain), but it is unclear whether this
represents the mechanism of D-leucine antiseizure
action (Bassoli et al., 2014; Shin et al., 2010). Other
ligands of this receptor have been shown to have
limited antiseizure activity in the maximal electro-
shock test (Talevi et al., 2012), supporting a poten-
tial role for this receptor in epilepsy.
Produced by bacteria, D-leucine is found
in food products, particularly those that are
plant-based (Ekborg-Ott and Armstrong, 1996;
Mutaguchi et  al., 2013). D-leucine has been iso-
lated from rat and mouse hippocampus, mouse
neocortex, and other areas of the brain at lower
concentrations (Hamase et al., 1997; Hamase et al.,
2001). Newly-synthesized mammalian proteins do
not include D-leucine (Fox et al., 1998). The only
other data on therapeutic use of D-leucine were in
analgesia studies, some of which also included use
of D-phenylalanine in combination with D-leucine
(Cheng and Pomeranz, 1980; McKibbin and
Cheng, 1982; Ninomiya et al., 1990). Importantly,
the doses required of each amino acid in the anal-
gesia studies were ~80x greater than the lowest
effective dose in our seizure studies, suggesting a
different mechanism for D-leucine in terminating
seizures.
Other D-amino acids may play a role in seizure
activity, although the literature is mixed in terms
of efficacy. D-serine binds to the glycine site on the
NMDA receptor, and decreased endogenous con-
centrations have been shown to be partly responsi-
ble for cognitive dysfunction in rats with epilepsy
induced by pilocarpine status epilepticus (Klatte
et  al., 2013; Schell et  al., 1995). Results of stud-
ies on the anticonvulsant effects of D-serine have
been mixed. D-serine protects weakly in the max-
imal electroshock test, where it also potentiates
the effects of some antiseizure drugs (Kalinichev
et  al., 2010a; Peterson, 1991). D-serine increases
afterdischarge thresholds in amygdala-kindled rats
(Loscher et al., 1994). Conversely, serine racemace
knockout mice, which have decreased extracellu-
lar levels of D-serine in the dentate gyrus, are rel-
atively protected against seizures induced by PTZ
(Harai et al., 2012; Singh et al., 1990a). Similarly,
exogenously administered D-serine potentiates
seizures induced by PTZ (Singh et al., 1990a). D-
serine does not have an effect on spike wave dis-
charges in GAERS (Genetic Absence Epilepsy Rats
from Strasbourg) Wistar rats (Koerner et al., 1996).
D-serine potentiates seizures induced by NMDA,
as might be expected (Singh et al., 1990a). In stud-
ies designed to test the efficacy of antiseizure med-
icines at the D-serine/glycine binding site on the
NMDA receptor, D-serine decreases the antisei-
zure effect of the clinical medicine felbamate, the
experimental compound L-687,414, and the opioid
kappa-receptor agonist CI-977, among others (De
Sarro et al., 1994; Singh et al., 1990b; Tricklebank
et al., 1994; White et al., 1995). Notably, the latter
experiments were not designed to directly test the
antiseizure efficacy of D-serine but rather used it
as a coactivator of NMDA receptors.
D-alanine binds to the D-serine/glycine site on
the NMDA receptor and may play a role in circa-
dian endocrine function (Kleckner and Dingledine,
1988; Morikawa et  al., 2008). Treatment with D-
alanine does not change seizure-related param-
eters in fully amygdala-kindled rats (Croucher
and Bradford, 1991). A  high dose (1,400 mg/kg)
of exogenously administered D-arginine has CNS
depressant effects and increases the latency to
onset of PTZ-induced convulsions (but 700 mg/kg
was ineffective) (Navarro et al., 2005).
D-amino acids are considered by some to be
“unnatural,” but the extant literature shows that
other endogenous D-amino acids have biological
functions. These are discussed briefly here for the
sake of interest and completeness, although they
do not have a reported direct role in epilepsy. D-
aspartate has neurotransmitter properties and may
play a role in neurodevelopment, learning, and
neuroendocrine function (D’Aniello et  al., 2000;
D’Aniello et al., 2011); it also decreases chemically
induced schizophrenia-like symptoms in rodents
(Errico et al., 2011; Errico et al., 2008). The brain
concentration of D-proline is highest in pineal
and pituitary tissue, but its endogenous func-
tion has not been reported (Hamase et al., 2005).
D-glutamate has been detected in subsets of
neuronal cell bodies in the mesencephalon and
thalamus by immunohistochemical techniques,
but its role is unknown (Mangas et  al., 2007).
D-methionine (1 mM) protects cultured auditory
neurons against cisplatin-induced neurotoxicity
(Gopal et al., 2012). A role for these D-amino acids
in seizures has not been reported. Interestingly,
plasma concentrations of D-serine, D-aspartate,
D-alanine, D-leucine, and D-proline are reduced
349
Chapter 35: Amino Acids in the Treatment of Neurological Disorders
BOX 35.2
D- AMINO ACIDS
WITH ANTISEIZURE EFFECTS
D-leucine
D-serine
D-arginine
in the rat beta-amyloid hippocampus injection
model of Alzheimer’s disease (Xing et  al., 2016).
Further work should identify whether D-amino
acid detection will be useful as a biomarker for
either identifying the disease or monitoring dis-
ease progression (Box 35.2).
T R AU M AT I C B R A I N
I N J U RY   ( T B I )
Traumatic brain injury represents one of the major
causes of injury and disability in all age groups
(although the causes vary between age groups).
Branched chain amino acids (particularly leucine)
represent a major source of glutamate in the cen-
tral nervous system (Sakai et al., 2004). Glutamate
serves as an excitatory neurotransmitter, an exci-
totoxin, and an energy store (used indirectly for
generation of ATP). In the mouse lateral fluid
percussion model of traumatic brain injury, hip-
pocampal concentrations of each of the branched
chain amino acids (total levels, i.e., L- and D-
amino acids) are decreased; supplementation of
these amino acids for 5 or 10 days (but not shorter
durations) reverses injury-induced deficits in
both anterograde and retrograde memory (Cole
et  al., 2010; Elkind et  al., 2015). In human tri-
als, supplementation of all three branched chain
amino acids led to a substantial improvement
in Disability Rating Scale scores (which evalu-
ate factors such as self-care and employability)
in patients with severe TBI or in a posttraumatic
vegetative state (in both studies, mean Glasgow
Coma scores were in the 5.5–5.9 range, represent-
ing significant impairment) (Aquilani et al., 2008;
Aquilani et al., 2005).
L-serine is neuroprotective in a mouse weight-
drop model in the acute phase of TBI (measured
morphologically), and the mechanism may involve
activation of glycine receptors (Zhai et al., 2015).
L-serine also decreases astrocytosis, microglial
activation, and concentrations of inflammatory
cytokines in the later phases of this model of TBI
(Zhai et al., 2015).
349
CONCLUSIONS
Select amino acids have been shown to protect
against seizure activity and the sequelae of TBI.
Interestingly, these compounds have a variety of
different mechanisms in these disorders, including
receptor binding and transporter effects, and as
metabolic intermediates for other signaling mol-
ecules. They also may affect intracellular signaling
pathways, although evidence for the latter is scant
at this point. Further research will elucidate addi-
tional roles for these amino acids in the treatment
of neurological disorders.
REFERENCES
Aquilani, R., Boselli, M., Boschi, F., Viglio, S., Iadarola,
P., Dossena, M., Pastoris, O., and Verri, M. (2008).
Branched-chain amino acids may improve recov-
ery from a vegetative or minimally conscious state
in patients with traumatic brain injury:  a pilot
study. Arch Phys Med Rehab 89, 1642–1647.
Aquilani, R., Iadarola, P., Contardi, A., Boselli, M.,
Verri, M., Pastoris, O., Boschi, F., Arcidiaco, P.,
and Viglio, S. (2005). Branched-chain amino acids
enhance the cognitive recovery of patients with
severe traumatic brain injury. Arch Phys Med
Rehab 86, 1729–1735.
Banach, M., Piskorska, B., Czuczwar, S.J., and
Borowicz, K.K. (2011). Nitric oxide, epileptic
seizures, and action of antiepileptic drugs. CNS
Neurol Disord-Dr 10, 808–819.
Bar-Peled, L., and Sabatini, D.M. (2014). Regulation
of mTORC1 by amino acids. Trends Cell Biol 24,
400–406.
Bassoli, A., Borgonovo, G., Caremoli, F., and Mancuso,
G. (2014). The taste of D- and L-amino acids: IN
vitro binding assays with cloned human bitter
(TAS2Rs) and sweet (TAS1R2/TAS1R3) receptors.
Food Chem 150, 27–33.
Bristow, L.J., Hutson, P.H., Kulagowski, J.J., Leeson,
P.D., Matheson, S., Murray, F., Rathbone, D.,
Saywell, K.L., Thorn, L., Watt, A.P., et  al. (1996).
Anticonvulsant and behavioral profile of L-
701,324, a potent, orally active antagonist at the
glycine modulatory site on the N-methyl-D-
aspartate receptor complex. J Pharmacol Exp Ther
279, 492–501.
Broer, A., Tietze, N., Kowalczuk, S., Chubb, S.,
Munzinger, M., Bak, L.K., and Broer, S. (2006).
The orphan transporter v7-3 (slc6a15) is a Na+-
dependent neutral amino acid transporter (B(0)
AT2). Biochem J 393, 421–430.
Buckmaster, P.S., and Lew, F.H. (2011). Rapamycin
suppresses mossy fiber sprouting but not seizure
frequency in a mouse model of temporal lobe epi-
lepsy. J Neurosci 31, 2337–2347.
Cheng, R.S., and Pomeranz, B. (1980). A combined
treatment with D-amino acids and electroacu-
puncture produces a greater analgesia than either
350
350 section IV: Ketone-Based Metabolism
treatment alone; naloxone reverses these effects.
Pain 8, 231–236.
Cole, J.T., Mitala, C.M., Kundu, S., Verma, A., Elkind,
J.A., Nissim, I., and Cohen, A.S. (2010). Dietary
branched chain amino acids ameliorate injury-
induced cognitive impairment. Proc Natl Acad Sci
USA 107, 366–371.
Cota, D., Proulx, K., Smith, K.A., Kozma, S.C.,
Thomas, G., Woods, S.C., and Seeley, R.J. (2006).
Hypothalamic mTOR signaling regulates food
intake. Science 312, 927–930.
Croucher, M.J., and Bradford, H.F. (1991). The influ-
ence of strychnine-insensitive glycine receptor
agonists and antagonists on generalized seizure
thresholds. Brain Res 543, 91–96.
D’Aniello, A., Di Fiore, M.M., Fisher, G.H., Milone, A.,
Seleni, A., D’Aniello, S., Perna, A.F., and Ingrosso,
D. (2000). Occurrence of D-aspartic acid and N-
methyl-D-aspartic acid in rat neuroendocrine tis-
sues and their role in the modulation of luteinizing
hormone and growth hormone release. FASEB J
14, 699–714.
D’Aniello, S., Somorjai, I., Garcia-Fernandez, J., Topo,
E., and D’Aniello, A. (2011). D-Aspartic acid is a
novel endogenous neurotransmitter. FASEB J 25,
1014–1027.
De Sarro, G., Ongini, E., Bertorelli, R., Aguglia, U.,
and De Sarro, A. (1994). Excitatory amino acid
neurotransmission through both NMDA and
non-NMDA receptors is involved in the anticon-
vulsant activity of felbamate in DBA/2 mice. Eur J
Pharmacol 262, 11–19.
de Vasconcelos, A.P., Gizard, F., Marescaux, C.,
and Nehlig, A. (2000). Role of nitric oxide
in pentylenetetrazol-induced seizures:  age-
dependent effects in the immature rat. Epilepsia
41, 363–371.
Dufour, F., Nalecz, K.A., Nalecz, M.J., and Nehlig, A.
(1999). Modulation of pentylenetetrazol-induced
seizure activity by branched-chain amino acids and
alpha-ketoisocaproate. Brain Res 815, 400–404.
Ekborg-Ott, K.H., and Armstrong, D.W. (1996).
Evaluation of the concentration and enantiomeric
purity of selected free amino acids in fermented
malt beverages (beers). Chirality 8, 49–57.
Elkind, J.A., Lim, M.M., Johnson, B.N., Palmer, C.P.,
Putnam, B.J., Kirschen, M.P., and Cohen, A.S.
(2015). Efficacy, dosage, and duration of action of
branched chain amino acid therapy for traumatic
brain injury. Front Neurol 6, 73.
Embden, G. (1908). Uber das Verhalten der optish-
isomeren Leucine in der Leber. Beitr z Chem
Physiol u Path 11, 348–355.
Errico, F., Nistico, R., Napolitano, F., Mazzola, C.,
Astone, D., Pisapia, T., Giustizieri, M., D’Aniello,
A., Mercuri, N.B., and Usiello, A. (2011). Increased
D-aspartate brain content rescues hippocampal
age-related synaptic plasticity deterioration of
mice. Neurobiol Aging 32, 2229–2243.
Errico, F., Rossi, S., Napolitano, F., Catuogno, V.,
Topo, E., Fisone, G., D’Aniello, A., Centonze, D.,
and Usiello, A. (2008). D-aspartate prevents cor-
ticostriatal long-term depression and attenuates
schizophrenia-like symptoms induced by amphet-
amine and MK-801. J Neurosci 28, 10404–10414.
Fox, H.L., Pham, P.T., Kimball, S.R., Jefferson, L.S., and
Lynch, C.J. (1998). Amino acid effects on transla-
tional repressor 4E-BP1 are mediated primarily by
L-leucine in isolated adipocytes. Am J Physiol-Cell
Ph 275, C1232–C1238.
Geier, E.G., Schlessinger, A., Fan, H., Gable, J.E.,
Irwin, J.J., Sali, A., and Giacomini, K.M. (2013).
Structure-based ligand discovery for the Large-
neutral Amino Acid Transporter 1, LAT-1. Proc
Natl Acad Sci USA 110, 5480–5485.
Gopal, K.V., Wu, C., Shrestha, B., Campbell, K.C.,
Moore, E.J., and Gross, G.W. (2012). d-Methionine
protects against cisplatin-induced neurotoxic-
ity in cortical networks. Neurotoxicol Teratol 34,
495–504.
Hagglund, M.G.A., Roshanbin, S., Lofqvist, E.,
Hellsten, S.V., Nilsson, V.C.O., Todkar, A., Zhu,
Y.N., Stephansson, O., Drgonova, J., Uhl, G.R.,
et  al. (2013). B(0)AT2 (SLC6A15) is local-
ized to neurons and astrocytes, and is involved
in mediating the effect of leucine in the brain.
PLoS One 8.
Hamase, K., Homma, H., Takigawa, Y., Fukushima, T.,
Santa, T., and Imai, K. (1997). Regional distribu-
tion and postnatal changes of D-amino acids in rat
brain. Biochim Biophys Acta 1334, 214–222.
Hamase, K., Inoue, T., Morikawa, A., Konno, R., and
Zaitsu, K. (2001). Determination of free D-proline
and D-leucine in the brains of mutant mice lack-
ing D-amino acid oxidase activity. Anal Biochem
298, 253–258.
Hamase, K., Konno, R., Morikawa, A., and Zaitsu, K.
(2005). Sensitive determination of D-amino acids
in mammals and the effect of D-amino-acid oxi-
dase activity on their amounts. Biol Pharm Bull 28,
1578–1584.
Harai, T., Inoue, R., Fujita, Y., Tanaka, A., Horio, M.,
Hashimoto, K., Hongou, K., Miyawaki, T., and
Mori, H. (2012). Decreased susceptibility to sei-
zures induced by pentylenetetrazole in serine race-
mase knockout mice. Epilepsy Res 102, 180–187.
Hartman, A.L., Santos, P., O’Riordan, K.J., Stafstrom,
C.E., and Marie Hardwick, J. (2015). Potent
anti-seizure effects of D-leucine. Neurobiol Dis
82, 46–53.
Herberg, L.J., Grottick, A., and Rose, I.C. (1995).
Nitric-oxide synthesis, epileptic seizures and kin-
dling. Psychopharmacology 119, 115–123.
Huang, X., Zhang, H., Yang, J., Wu, J., McMahon, J.,
Lin, Y., Cao, Z., Gruenthal, M., and Huang, Y.
(2010). Pharmacological inhibition of the mam-
malian target of rapamycin pathway suppresses
acquired epilepsy. Neurobiol Dis 40, 193–199.
351
Chapter 35: Amino Acids in the Treatment of Neurological Disorders
Johnson, J.W., and Ascher, P. (1987). Glycine potenti-
ates the NMDA response in cultured mouse-brain
neurons. Nature 325, 529–531.
Kalinichev, M., Starr, K.R., Teague, S., Bradford, A.M.,
Porter, R.A., and Herdon, H.J. (2010a). Glycine
transporter 1 (GlyT1) inhibitors exhibit anticon-
vulsant properties in the rat maximal electroshock
threshold (MEST) test. Brain Res 1331, 105–113.
Kalinichev, M., Starr, K.R., Teague, S., Bradford, A.M.,
Porter, R.A., and Herdon, H.J. (2010b). Glycine
transporter 1 (GlyT1) inhibitors exhibit anticon-
vulsant properties in the rat maximal electroshock
threshold (MEST) test. Brain Research 1331,
105–113.
Klatte, K., Kirschstein, T., Otte, D., Pothmann, L.,
Muller, L., Tokay, T., Kober, M., Uebachs, M.,
Zimmer, A., and Beck, H. (2013). Impaired D-
serine-mediated cotransmission mediates cog-
nitive dysfunction in epilepsy. J Neurosci 33,
13066–13080.
Kleckner, N.W., and Dingledine, R. (1988).
Requirement for glycine in activation of NMDA-
receptors expressed in Xenopus oocytes. Science
241, 835–837.
Koerner, C., Danober, L., Boehrer, A., Marescaux, C.,
and Vergnes, M. (1996). Thalamic NMDA trans-
mission in a genetic model of absence epilepsy in
rats. Epilepsy Res 25, 11–19.
Laeger, T., Henagan, T.M., Albarado, D.C., Redman,
L.M., Bray, G.A., Noland, R.C., Munzberg, H.,
Hutson, S.M., Gettys, T.W., Schwartz, M.W., et al.
(2014). FGF21 is an endocrine signal of protein
restriction. J Clin Invest 124, 3913–3922.
Lim, B.C., Cho, K.Y., Lim, J.S., Lee, R.S., Kim, H.S.,
Kim, M.K., Kim, J.H., Woo, Y.J., Kim, J.K., Kim,
D.K., et al. (2011). Increased expression of l-amino
acid transporters in balloon cells of tuberous scle-
rosis. Child Nerv Syst 27, 63–70.
Loscher, W., Wlaz, P., Rundfeldt, C., Baran, H., and
Honack, D. (1994). Anticonvulsant effects of the
glycine/NMDA receptor ligands D-cycloserine
and D-serine but not R-(+)-HA-966 in amygdala-
kindled rats. Br J Pharmacol 112, 97–106.
Mangas, A., Covenas, R., Bodet, D., Geffard, M., Aguilar,
L.A., and Yajeya, J. (2007). Immunocytochemical
visualization of D-glutamate in the rat brain.
Neuroscience 144, 654–664.
Masino, S.A., and Rho, J.M. (2012). Mechanisms of
ketogenic diet action. In Jasper’s Basic Mechanisms
of the Epilepsies, J.L. Noebels, M. Avoli, M.A.
Rogawski, O.R. W., and A.V. Delgado-Escueta,
eds. (Bethesda: National Center for Biotechnology
Information), pp. 1001–1022.
McKibbin, L.S., and Cheng, R.S. (1982). Systemic d-
phenylalanine and d-leucine for effective treat-
ment of pain in the horse. Canadian VetJ/La Revue
Veterinaire Canadienne 23, 39–40.
Morikawa, A., Hamase, K., Miyoshi, Y., Koyanagi, S.,
Ohdo, S., and Zaitsu, K. (2008). Circadian changes
351
of D-alanine and related compounds in rats and
the effect of restricted feeding on their amounts.
J Chromatogr B Analyt Technol Biomed Life Sci
875, 168–173.
Mutaguchi, Y., Ohmori, T., Wakamatsu, T., Doi, K., and
Ohshima, T. (2013). Identification, purification,
and characterization of a novel amino acid race-
mase, isoleucine 2-epimerase, from Lactobacillus
species. J Bacteriol 195, 5207–5215.
Navarro, E., Alonso, S.J., Martin, F.A., and Castellano,
M.A. (2005). Toxicological and pharmacologi-
cal effects of D-arginine. Basic Clin Pharmacol
Toxicol 97, 149–154.
Nichols, A.C., and Yielding, K.L. (1998). Anticonvulsant
activity of 4-urea-5,7-dichlorokynurenic acid
derivatives that are antagonists at the NMDA-
associated glycine binding site. Mol Chem
Neuropathol 35, 1–12.
Ninomiya, Y., Kawamura, H., Nomura, T., Uebayashi,
H., Sabashi, K., and Funakoshi, M. (1990).
Analgesic effects of D-amino acids in four inbred
strains of mice. Comp Biochem Physiol C 97,
341–343.
Orlova, K.A., and Crino, P.B. (2010). The tuberous
sclerosis complex. Ann NY Acad Sci 1184, 87–105.
Peterson, S.L. (1991). Anticonvulsant drug potentia-
tion by glycine in maximal electroshock seizures
is mimicked by D-serine and antagonized by
7-chlorokynurenic acid. Eur J Pharmacol 199,
341–348.
Sakai, R., Cohen, D.M., Henry, J.F., Burrin, D.G., and
Reeds, P.J. (2004). Leucine-nitrogen metabolism
in the brain of conscious rats:  its role as a nitro-
gen carrier in glutamate synthesis in glial and
neuronal metabolic compartments. Journal of
Neurochemistry 88, 612–622.
Sancak, Y., Peterson, T.R., Shaul, Y.D., Lindquist, R.A.,
Thoreen, C.C., Bar-Peled, L., and Sabatini, D.M.
(2008). The Rag GTPases bind raptor and medi-
ate amino acid signaling to mTORC1. Science 320,
1496–1501.
Schell, M.J., Molliver, M.E., and Snyder, S.H. (1995).
D-serine, an endogenous synaptic modula-
tor:  localization to astrocytes and glutamate-
stimulated release. Proc Natl Acad Sci USA 92,
3948–3952.
Shen, H.Y., van Vliet, E.A., Bright, K.A., Hanthorn,
M., Lytle, N.K., Gorter, J., Aronica, E., and Boison,
D. (2015). Glycine transporter 1 is a target for the
treatment of epilepsy. Neuropharmacology 99,
554–565.
Shin, Y.J., Park, J.H., Choi, J.S., Chun, M.H., Moon,
Y.W., and Lee, M.Y. (2010). Enhanced expression
of the sweet taste receptors and alpha-gustducin
in reactive astrocytes of the rat hippocampus
following ischemic injury. Neurochem Res 35,
1628–1634.
Singh, L., Oles, R.J., and Tricklebank, M.D. (1990a).
Modulation of seizure susceptibility in the mouse
352
352 section IV: Ketone-Based Metabolism
by the strychnine-insensitive glycine recognition
site of the NMDA receptor/ion channel complex.
Br J Pharmacol 99, 285–288.
Singh, L., Vass, C.A., Hunter, J.C., Woodruff, G.N.,
and Hughes, J. (1990b). The anticonvulsant
action of CI-977, a selective kappa-opioid recep-
tor agonist:  a possible involvement of the gly-
cine/NMDA receptor complex. Eur J Pharmacol
191, 477–480.
Skeie, B., Petersen, A.J., Manner, T., Askanazi, J., and
Steen, P.A. (1994). Effects of valine, leucine, isoleu-
cine, and a balanced amino acid solution on the
seizure threshold to picrotoxin in rats. Pharmacol
Biochem Behav 48, 101–103.
Sosanya, N.M., Brager, D.H., Wolfe, S., Niere, F., and
Raab-Graham, K.F. (2015). Rapamycin reveals an
mTOR-independent repression of Kv1.1 expres-
sion during epileptogenesis. Neurobiol Dis 73,
96–105.
Talevi, A., Enrique, A.V., and Bruno-Blanch, L.E.
(2012). Anticonvulsant activity of artificial sweet-
eners: a structural link between sweet-taste recep-
tor T1R3 and brain glutamate receptors. Bioorg
Med Chem Lett 22, 4072–4074.
Tricklebank, M.D., Bristow, L.J., Hutson, P.H., Leeson,
P.D., Rowley, M., Saywell, K., Singh, L., Tattersall,
F.D., Thorn, L., and Williams, B.J. (1994). The anti-
convulsant and behavioural profile of L-687,414,
a partial agonist acting at the glycine modulatory
site on the N-methyl-D-aspartate (NMDA) recep-
tor complex. Br J Pharmacol 113, 729–736.
Werman, R., Davidoff, R.A., and Aprison, M.H.
(1968). Inhibitory action of glycine on spinal neu-
rons in cat. J Neurophysiol 31, 81–95.
White, H.S., Harmsworth, W.L., Sofia, R.D., and
Wolf, H.H. (1995). Felbamate modulates the
strychnine-insensitive glycine receptor. Epilepsy
Res 20, 41–48.
Wolfson, R.L., Chantranupong, L., Saxton, R.A.,
Shen, K., Scaria, S.M., Cantor, J.R., and Sabatini,
D.M. (2016). Sestrin2 is a leucine sensor for the
mTORC1 pathway. Science 351, 43–48.
Xie, X.M., Dumas, T., Tang, L., Brennan, T., Reeder,
T., Thomas, W., Klein, R.D., Flores, J., O’Hara, B.F.,
Heller, H.C., et  al. (2005). Lack of the alanine-
serine-cysteine transporter 1 causes tremors, sei-
zures, and early postnatal death in mice. Brain Res
1052, 212–221.
Xing, Y., Li, X., Guo, X., and Cui, Y. (2016).
Simultaneous determination of 18 D-amino
acids in rat plasma by an ultrahigh-performance
liquid chromatography-tandem mass spectrom-
etry method:  application to explore the potential
relationship between Alzheimer’s disease and D-
amino acid level alterations. Anal Bioanal Chem
408, 141–150.
Yudkoff, M., Daikhin, Y., Melo, T.M., Nissim, I., and
Sonnewald, U. (2007). The ketogenic diet and
brain metabolism of amino acids:  relationship
to the anticonvulsant effect. Annu Rev Nutr 27,
415–430.
Yudkoff, M., Daikhin, Y., Nissim, I., Horyn, O.,
Lazarow, A., Luhovyy, B., and Wehrli, S. (2005).
Response of brain amino acid metabolism to keto-
sis. Neurochem Int 47, 119–128.
Zellinger, C., Salvarnoser, J.D., Soerensen, J., van
Vliet, E.A., Aronica, E., Gorter, J., and Potschka,
H. (2014). Pre-treatment with the NMDA recep-
tor glycine-binding site antagonist L-701,324
improves pharmacosensitivity in a mouse kindling
model. Epilepsy Res 108, 634–643.
Zeng, L.H., Xu, L., Gutmann, D.H., and Wong, M.
(2008). Rapamycin prevents epilepsy in a mouse
model of tuberous sclerosis complex. Ann Neurol
63, 444–453.
Zeng, L.H., Rensing, N.R., and Wong, M. (2009). The
mammalian target of rapamycin signaling path-
way mediates epileptogenesis in a model of tem-
poral lobe epilepsy. J Neurosci 29, 6964–6972.
Zhai, P.P., Xu, L.H., Yang, J.J., Jiang, Z.L., Zhao, G.W.,
Sun, L., Wang, G.H., and Li, X. (2015). Reduction
of inflammatory responses by L-serine treatment
leads to neuroprotection in mice after traumatic
brain injury. Neuropharmacology 95, 1–11.
353
36
2-Deoxyglucose
Metabolic Control of Seizures through Inhibition of Glycolysis
C A R L E . S TA F S T R O M , M D , P H D A N D T H O M A S P. S U T U L A , M D , P H D
INTRODUCTION
Conventional anticonvulsant medications reduce
neuronal excitability through effects on ion chan-
nels or synaptic function. In recent years, it has
become clear that metabolic factors also play a role
in the modulation of neuronal excitability (Reid
et al., 2014). This volume contains many examples
of potentially beneficial metabolic treatments for
epilepsy and other neurological disorders. In par-
ticular, the high-fat, low-carbohydrate ketogenic
diet (KD) is effective in controlling seizures in many
children whose seizures are refractory to anticon-
vulsant medications (Neal et al., 2008). However,
the mechanisms of action of the KD and its vari-
ants (e.g., medium chain triglyceride diet, modi-
fied Atkins diet, low glycemic index treatment)
are very complex and are not fully characterized
(Rho and Stafstrom, 2012; Lutas and Yellen, 2013;
Gano et  al., 2014). Possible mechanisms include
reduction of excitability by ketone bodies or fatty
acids, altered neurotransmitter synthesis or action,
improved mitochondrial function, or a combina-
tion of these or other factors. The key observation
that ingestion of small amounts of carbohydrate by
children on the KD results in loss of seizure con-
trol (Huttenlocher, 1976) led to the idea that car-
bohydrate restriction could exert a protective effect
against seizures (Greene et al., 2003). In addition
to limiting carbohydrate intake, restricting calorie
intake also suppresses seizures and affords neuro-
protection (Greene et al., 2003; Ingram and Roth,
2011; Pani, 2015; Yuen and Sander, 2014), and in
fact, the KD was initially formulated to mimic the
physiological effects of fasting. Although some
data supports intermittent fasting for seizure con-
trol (Hartman et al., 2013), fasting is not a feasible
long-term treatment option.
As an alternative to fasting, the KD restricts
dietary carbohydrates and generates ketone bodies
as the proximate energy source, thereby reducing
glycolysis (Figure 36.1). This effect and the obser-
vation that minimal carbohydrate intake can abol-
ish seizure control achieved by the KD suggests
that inhibitors of glycolysis may mimic some of the
favorable therapeutic effects of the diet. Among
compounds that inhibit glycolysis, the glucose
analog 2-deoxy-D-glucose (2DG) is a promising
novel agent for seizure protection. 2DG differs
from glucose by removal of a single oxygen atom
from the 2-position (Figure 36.2A). 2DG is taken
up by cells and undergoes phosphorylation at the
6-position to 2DG-6P, but glycolytic flux is reduced
because 2DG-6P cannot undergo isomerization by
phosphoglucose isomerase (Figure 36.2B). Uptake
of glucose and 2DG is enhanced in energetically
active cells. 2DG has been used for decades as a
fluorinated positron emitted tracer (F18-2DG) for
measurement and imaging of regional glucose uti-
lization by positron emission tomography (PET)
(Wree, 1990). 2DG has also been investigated as
an adjuvant chemotherapeutic agent for several
types of cancer, since rapidly dividing, metaboli-
cally active neoplastic cells with enhanced glucose
uptake are vulnerable to glycolytic inhibition by
2DG (Pelicano et al., 2006; Cheong et al., 2011).
ANTICONVULSANT
ACTIONS OF 2DG
We tested the effects of 2DG in several models
of acute seizures in vivo and in vitro (Stafstrom
et  al., 2009). When exposed to elevated extracel-
lular potassium (Ko+, 7.5 mM), hippocampal CA3
neurons in slices from adult rats developed high-
frequency interictal (epileptiform) bursts at a fre-
quency of about 30 per minute. Addition of 2DG
(10 mM) to the bathing medium reversibly reduced
the burst frequency by about 50% (Stafstrom
et al., 2009). Similarly, 2DG reduced epileptiform
bursts induced by bath application of the chemi-
cal convulsants bicuculline, a γ-aminobutyric acid
354

354 section IV: Ketone-Based Metabolism

KD

Ketone bodies Anaplerosis

(β-OHB, AcAc)
2DG

Glycolysis
PDH
Glucose Pyruvate ACoA TCA

Gluconeogenesis

FDP

PPP Lactate

FIGURE  36.1 Glucose metabolism and points at which interventions could affect neuronal excitability and
seizure control. Glucose can be diverted to the pentose phosphate shunt (PPP) via fructose-1,6-diphosphate (FDP).
2-Deoxyglucose (2DG) inhibits glycolysis by blocking the phosphoglucose isomerase step. The ketogenic diet (KD), via
ketone bodies, bypasses glycolysis by providing acetyl-CoA (ACoA) to the TCA (tricarboxylic acid cycle) after glycolysis.
Anaplerosis refers to the “refilling” of intermediate compounds depleted from the TCA cycle. Other abbreviations: β-
OHB, beta-hydroxybutyrate; AcAc, acetoacetate; PDH, pyruvate dehydrogenase. Reprinted with permission from Rho
and Stafstrom, 2012.

(GABA) receptor antagonist, or 4-aminopyridine, seconds; 2DG decreased the occurrence of these
a potassium channel blocker. In slices from juve- ictal bursts. Therefore, 2DG has an anticonvul-
nile animals (P10-13), 7.5 mM Ko+ induced pro- sant effect on both interictal and ictal epileptiform
longed ictal discharges in area CA3 for 10–30 activity in the CA3 region of hippocampus.

(a) (b)
CH2OH CH2OPO32– Glucose
3–O CH2OH
H OH H OH 3POH2C
H H O
OH H OH H HHO Glucose-6-P
HO OH HO OH H OH
GPI 2DG
H OH H OH HO H
Glucose Glucose-6P Fructose-6P Fructose-6-P

CH2OH CH2OPO32– PFK Pyruvate

3–O CH2OH
H OH H OH 3PCH2C
H H O Fructose-1,6-P
OH H OH H HHO NAD
HO OH HO OH H OH
H H H H HO H NADH

2DG 2DG-6P TCA cycle

FIGURE 36.2 Glucose, 2DG, and the glycolytic pathway. (a) Chemical structures of glucose, 2-deoxyglucose (2DG),
and intermediates of the initial steps of glycolysis. Phosphorylation of 2DG yields 2DG-6P, which cannot undergo
isomerization by glucose-6-phosphate (glucose-6-P) isomerase (GPI) to fructose-6-phosphate (fructose-6-P), thereby
preventing subsequent steps of glycolysis. (b) Schematic diagram of key steps of glycolysis illustrating the rate-limiting
step involving phosphofructokinase, which is inhibited by pyruvate, the end product of the pathway. Oxidation of
phosphoenol-pyruvate (structure not shown) to pyruvate generates nicotinamide adenine dinucleotide (NADH) before
entry into the tricarboxylic acid (TCA) cycle. Reprinted with permission from Stafstrom et al., 2009.
355
An anticonvulsant effect was also seen in in
vivo seizure models. Kindling refers to the pro-
gressive decrease in seizure threshold in response
to repeated subconvulsive electrical stimuli.
Kindled seizures can be elicited by stimulation of
many brain pathways. We assessed the effect of
2DG on kindling development by stimulation of
the perforant path or the olfactory bulb (Garriga-
Canut et al., 2006; Stafstrom et al., 2009). 2DG was
administered at a dose of 250 mg/kg intraperitone-
ally (i.p.) 30 minutes prior to daily kindling stimu-
lation of either pathway. In response to kindling
of the perforant path, but not the olfactory bulb,
2DG-pretreated rats displayed an increase in mean
afterdischarge (AD) threshold over time, defining
a region-specific anticonvulsant effect.
In addition, an anticonvulsant effect of 2DG
was seen in two other acute seizure models in ani-
mals. First, in the 6 Hz stimulation model (Barton
et al., 2001), psychomotor seizures were induced by
corneal stimulation using electrical pulses at a fre-
quency of 6 Hz. Pretreatment with 2DG resulted in
seizure protection in 75% of rats tested (Stafstrom
et  al., 2009). Second, in Fring’s mice with audio-
genic seizures, 50% of mice were protected from
sound-induced seizures after pretreatment with
2DG (Stafstrom et  al., 2009). However, 2DG did
not protect against pentylenetetrazole- or maximal
electroshock-induced seizures in rats. Therefore,
anticonvulsant effects of 2DG were demonstrated
in several but not all in vivo seizure models, with
a pattern of effectiveness unlike that afforded by
any currently available antiseizure medicine. For
example, no conventional antiseizure medication
exhibits region-specific activity, as indicated by
2DG’s differential action on perforant path versus
olfactory bulb. Thus, 2DG’s mechanisms are likely
to be broad, affecting seizure-induced plasticity
across many seizure types and syndromes, but an
exact correlation with human epilepsies is not yet
possible (Holmes and Zhao, 2008).
Not all studies have confirmed that 2DG exerts
an anticonvulsant effect. In the presence of 2DG,
latency to seizure onset was slightly decreased in
mice when chemoconvulsants (pentylenetetrazole,
kainic acid) were coadministered intravenously
with 2DG (Gasior et al., 2010). While this outcome
has been interpreted as a possible proconvulsant
effect, an alternative explanation is that the well-
documented acute increase in cerebral blood flow
induced by 2DG enhances brain delivery of par-
enteral convulsants such as pentylenetetrazole and
kainic acid, thereby shortening the latency to ini-
tial seizure manifestations as a threshold outcome
measure.
Chapter 36: 2-Deoxyglucose 355
It is worth mentioning that the KD and 2DG
both suppress seizure activity, but our studies with
2DG were not intended to investigate KD mecha-
nisms. For instance, 2DG does not cause ketosis.
Both 2DG and the KD suppress seizures in Fring’s
mice and in the 6-Hz model (Hartman et  al.,
2008), but the KD is effective in the MES model
while 2DG is not. The former two models are used
to model focal onset seizures, whereas the MES is
considered to mimic generalized tonic-clonic sei-
zures. Kindling is abrogated by both the KD and
2DG, though exact kindling protocols differ widely
among studies (Hori et  al., 1997; Garriga-Canut
et al., 2006; Stafstrom et al., 2009). Therefore, 2DG
and the KD are not comparable directly, though
both modify seizure susceptibility by altering met-
abolic pathways involved in energy regulation.
Therefore, potentially useful clinical information
can be gained from studies of both compounds.
ANTIEPILEPTIC
ACTIONS OF 2DG
An antiepileptic (or antiepileptogenic) effect is
defined as the slowing or prevention of epilepsy
development. Antiepileptogenesis is “disease-
modifying,” meaning that the process of epilepsy
development is hindered. Using the kindling
model, with stimulation of either the olfactory bulb
or perforant path, we investigated whether 2DG
mediates an antiepileptogenic effect. Kindling
progression was quantified by the number of ADs
required to elicit specific stages of seizure sever-
ity, using the Racine scoring scale (Racine, 1972).
Pretreatment with 2DG at doses ranging from 37.5
mg/kg intraperitoneally (i.p.) (we found this to be
the minimally effective dose) up to 250 mg/kg i.p.,
given 30 minutes before each kindling stimulation,
resulted in an ~2-fold increase in the number of
ADs required to achieve class III, IV, or V kindled
seizures in both regions (Garriga-Canut et  al.,
2006; Sutula and Franzoso, 2008; Stafstrom et al.,
2009). These results indicate a disease-modifying
antiepileptic action. Note that the olfactory bulb
kindling experiments described in the preced-
ing paragraph, testing the anticonvulsant effects
of 2DG, showed no significant effect of this com-
pound on kindled seizure threshold. Therefore,
2DG has different anticonvulsant effects in differ-
ent brain regions, whereas its antiepileptic effects
were similar in hippocampus and olfactory bulb.
Furthermore, 2DG retarded kindled seizure pro-
gression when administered immediately after
or 10 minutes after a kindled seizure, raising the
possibility that 2DG can be used as a “postseizure”
356
356 section IV: Ketone-Based Metabolism
treatment (e.g., for clusters of seizures or even
status epilepticus) (Sutula and Franzoso, 2008;
Stafstrom et al., in preparation).
2DG exerts neuroprotective actions in several
other models as well. In hippocampal cell cultures,
2DG increased neuronal resistance to oxidative
and metabolic insults by inducing stress proteins,
and in rats, 2DG allayed kainic acid seizure-
induced memory deficits and hippocampal neu-
ron loss (Lee et al., 1999). In another model, 2DG
pretreatment of mice undergoing bilateral carotid
artery occlusion protected against seizures that
typically occur in this protocol (Redjak et  al.,
2001). Numerous mechanisms of 2DG-mediated
neuroprotection have been hypothesized, includ-
ing activation of adenosine monophosphate
(AMP)-activated protein kinase (Park et al., 2009),
reduction in oxidative stress (Yao et  al., 2011),
and disruption of glycosylation causing protein
unfolding (Zhang et al., 2014), to name a few.
POS SIBLE MECHANISMS
OF 2DG EFFECTS
The acute and chronic actions of 2DG probably
involve different cellular and molecular mecha-
nisms. The chronic antiepileptic effects of 2DG
have been associated with decreased expression
of brain derived neurotrophic factor (BDNF) and
its receptor, tyrosine kinase B (trkB), which are
required for kindling progression (He et al., 2004).
2DG suppression of seizure-induced increases
in BDNF and trkB is mediated by the transcrip-
tional repressor neuron restrictive silencing factor
(NRSF) and its nicotinamide adenine dinucle-
otide hydride (NADH)–sensitive corepressor
carboxy-terminal binding protein (CtBP) acting
at the promoter regions of BDNF and trkB genes.
In pathological conditions such as seizures, glycol-
ysis is enhanced to meet the energy demands of
activated neurons. The increase in glycolysis ele-
vates NADH, which in turn causes dissociation of
CtBP from NRSF, thus decreasing transcriptional
repression and resulting in increased expression of
BDNF and trkB. In the presence of 2DG, which
reduces NADH levels as a consequence of glyco-
lytic inhibition, the NRSF-CtBP complex main-
tains repression of BDNF and trkB, and kindling
progression is slowed (Garriga-Canut et al., 2006;
Huang and McNamara, 2006).
Compared to its pivotal role in mediating
chronic 2DG effects, NRSF is not required for the
antiepileptic effect of the KD (Hu et al., 2011). The
antiepileptic effect of 2DG was abolished in mice
with conditional knockout of NRSF, but the KD
continued to afford protection against kindling
progression in these transgenic animals.
Others have reported evidence that the effect
of 2DG is mediated through up-regulation of KATP
channel subunits Kir6.1 and Kir6.2 (Yang et  al.,
2013). KATP channels are closed in the presence
of intracellular ATP and open when intracellular
ATP is depleted. Open KATP channels efflux K+,
hyperpolarizing the cell and decreasing its excit-
ability. It has been hypothesized that ketones,
by decreasing glycolysis and thus ATP produc-
tion, may lower cellular excitability (Ma et  al.,
2007). Decreased ATP levels, perhaps restricted
to submembrane compartments adjacent to KATP
channels, might lead to enhanced K+ efflux and
hyperpolarization. Such effects could be limited to
certain neuron types (e.g., substantia nigra, den-
tate gyrus) that function as a gate for pathologi-
cal discharges (Ma et al., 2007; Tanner et al., 2011;
Lutas and Yellen, 2013).
The rapid onset of anticonvulsant effects
of 2DG in vitro and in vivo suggests that 2DG
may be exerting direct actions at the synaptic or
membrane levels. The effects of 2DG on excit-
atory synaptic transmission were investigated
in pilot experiments using hippocampal slices
(Pan et  al., 2008; Pan et  al., 2014; Rutecki et  al.,
in preparation). In hippocampal area CA3, there
was no effect of 10 mM 2DG on the frequency or
amplitude of spontaneous excitatory postsynaptic
currents (sEPSCs). However, after the induction
of epileptiform bursting in this region by appli-
cation of elevated (7.5 mM) Ko+, 2DG reduced
sEPSC frequency and amplitude, suggesting that
the effects of 2DG are activity-dependent—2DG
is taken up by actively firing cells and works pref-
erentially in conditions of intense neuronal activ-
ity such as seizures. The same dose of 2DG had no
effect on miniature EPSCs isolated by exposure to
tetrodotoxin, which blocks sodium channels and
thus reduces neuronal activity-mediated synap-
tic release of neurotransmitters, but miniature
EPSCs were significantly reduced by 10 mM 2DG
in the elevated Ko+ condition. These effects are not
a general consequence of glycolysis inhibition,
since other glycolysis blockers depress sEPSCs in
both normal and epileptic slices (Pan et al., 2009;
Devinney et  al., 2009). 2DG also reduces spon-
taneous inhibitory postsynaptic current (sIPSC)
frequency in conditions of elevated Ko+ (7.5
mM), but to a lesser extent than sEPSCs, result-
ing in a net reduction in excitatory transmission
(Pan et al., 2015). 2DG has no significant effects
on intrinsic membrane properties, and its acute
effects on synaptic transmission appear to be
357
presynaptic based on analysis of miniature EPSCs
(Pan et al, 2015).
Therefore, one of the unique features about
2DG is its use-dependence. That is, 2DG is taken
up only by neurons that are metabolically active, as
occurs in areas of circuitry involved in seizure activ-
ity. This represents a distinct advantage when con-
sidering the goal of a medication in targeting only
brain areas displaying pathological seizure activity.
The acute anticonvulsant effects of 2DG may
be influenced by other, as yet undetermined met-
abolic or electrophysiological consequences of
glycolytic inhibition. For example, 2DG’s effects
might be spatially limited to certain submem-
brane compartments, it might alter systemic lipid
metabolism, or it might modify mitochondrial
metabolism in a manner that subsequently influ-
ences neuronal excitability.
Another compound in the glycolytic pathway,
fructose-1,6-diphosphate (FDP), has been shown
to exert acute anticonvulsant activity in several
seizure models in adult rats including kainic acid,
pilocarpine, and pentylenetetrazole (Lian et  al.,
2007). In that study, the effectiveness of FDP as
an anticonvulsant surpassed that of 2DG, KD,
and valproate. The mechanism of FDP’s anticon-
vulsant effect is unclear. Fructose-1,6-diphosphate
increases glucose flux from glycolysis into the pen-
tose phosphate pathway (PPP) (Figure 36.1), and
NADPH generated in the PPP reduces glutathione,
resulting in an anticonvulsant action. Therefore,
FDP may exert an endogenous anticonvulsant
action (Stringer and Xu, 2008). Subsequent stud-
ies have established that FDP retards kindling pro-
gression by attenuating BDNF and trkB expression
(Ding et al., 2010), like 2DG. Furthermore, during
the kindling process, FDP inhibits the kindling-
induced down-regulation of the expression of
potassium-chloride cotransporter 2 (KCC2) and
decreases the expression of sodium-potassium-
chloride cotransporter 1 (NKCC1), suggesting that
FDP might alter the switch between GABAergic
excitation and inhibition (Ding et al., 2013). These
findings remain to be verified, but together, results
from several laboratories have identified modifica-
tion of glycolysis as a possible novel mechanism
for treatment of seizures.
2 D G A S   A P OT E N T I A L
CLINICAL AGENT
Preclinical studies to evaluate safety and toxicity of
2DG have demonstrated that 2DG is well tolerated
in rats and dogs at doses associated with anticon-
vulsant and antiepileptic effects. We used both acute
and chronic protocols to test the effects of 2DG on
Chapter 36: 2-Deoxyglucose 357
spatial learning and memory in the Morris water
maze (Ockuly et al., 2012). For acute testing, 2DG
was injected 15 minutes prior to the water maze
trial each testing day. For chronic testing, 2DG (250
mg/kg or 500 mg/kg i.p., twice daily) was injected
daily for 14 days before water maze testing began.
Neither protocol caused a difference in the latency
to platform acquisition (spatial learning) or reten-
tion of platform location (probe test) by either dose
of 2DG, suggesting that 2DG has no obvious delete-
rious effect on spatial memory or learning.
Rats were also tested on the open field test,
which assesses exploratory activity and has also
been regarded as a measure of anxiety (Ockuly
et  al., 2012). Rats were pretreated (30 min prior
to open field testing) i.p. with saline, 50 mg/kg
2DG, or 250 mg/kg 2DG in a crossover design.
The exploratory activity (number of lines crossed
in the open field arena) did not differ between the
saline- and 50 mg/kg 2DG-treated groups, but
the rats receiving 250 mg/kg 2DG had decreased
motor activity (fewer lines crossed). When the
saline group was crossed over to receive 2DG 250
mg/kg and the prior 250 mg/kg group was crossed
over to saline, the results reversed—rats receiving
250 mg/kg 2DG had diminished open field activity
and the group now receiving saline had increased
motor activity. Therefore, the 250 mg/kg dose of
2DG caused a transient, reversible decrease in
exploratory activity. Taken together, these find-
ings suggest that 2DG has no permanent adverse
behavioral effects on spatial learning and memory,
exploratory activity, or anxiety at doses that sup-
press seizures and retard epilepsy progression,
supporting its potential for clinical use.
Some animal studies have shown that 2DG at
high doses is associated with adverse cardiac effects.
Oral or intravenous doses of 2DG 250–2,000
mg/kg given to rats or mice over 7 days caused a
dose-dependent fall in mean arterial pressure,
decreased respiratory rate, and increased mortality
(Vijayaraghavan et al., 2006). Detailed pathological
evaluation of cardiac tissue after chronic oral 2DG
ingestion in two rat strains revealed cardiotoxic
effects with vacuolization of cardiac myocytes,
increased incidence of pheochromocytomas, and
reduced life spans (Minor et  al., 2010). Cardiac
effects of 2DG in rats had features consistent with
autophagy, a process by which a cell degrades
unnecessary cellular components in response to
nutrient stress. Comprehensive preclinical safety
and toxicology studies have subsequently dem-
onstrated reversible species-specific cardiac tox-
icity at high doses in F344 rats but not in Beagle
dogs; this toxicity was detectable by monitoring
358

358 section IV: Ketone-Based Metabolism

TABLE 36.1 EFFECTS OF 2DG IN MODELS OF SEIZURES AND EPILEPSY

Anticonvulsant Effects

In vitro
High Ko+ Decreased interictal and ictal bursting (CA3) Stafstrom et al., 2009
Bicuculline Decreased interictal bursting (CA3) Stafstrom et al., 2009
4-Aminopyridine Decreased interictal bursting (CA3) Stafstrom et al., 2009
In vivo
6-Hz corneal stimulation Seizure protection Stafstrom et al., 2009;
in mice Gasior et al., 2010
Audiogenic seizures in Seizure protection Stafstrom et al., 2009
Fring’s mice
Kindling of perforant path Increased afterdischarge threshold Garriga-Canut et al., 2006
Pilocarpine Decreased seizure severity and duration; Lian et al., 2007
increased seizure latency
Kainic acid i.p. No effect Lian et al., 2007
Kainic acid i.v. Decreased seizure threshold Gasior et al., 2010
Pentylenetetrazole i.v. or s.c. Decreased seizure threshold Gasior et al., 2010
Shortened seizures Lian et al., 2007
No effect Stafstrom et al., 2009
Maximum electroshock No effect Stafstrom et al., 2009
Decreased seizure threshold Gasior et al., 2010

Antiepileptic Effects

Kindling of perforant path Greater number of afterdischarges required Garriga-Canut et al., 2006

to reach class III, IV and V seizures
Kindling of olfactory bulb Greater number of afterdischarges required Stafstrom et al., 2009
to reach class III, IV and V seizures
Kindling of amygdala No effect Gasior et al., 2010

Behavioral Effects

Morris water maze No adverse effect on spatial learning or  Ockuly et al., 2012

memory
Open field test No lasting effect on exploratory activity Ockuly et al., 2012

Legend: Ko+, extracellular potassium; i.p., intraperitoneal; i.v., intravenous; s.c., subcutaneous

plasma N-terminal probrain natriuretic peptide
(Nt-proBNP) (Terse et al., 2016).
Experience regarding human safety of 2DG
has been obtained in cancer trials (Pelicano et al.,
2006). Tumor cells are dependent on glycolysis to
support their metabolic requirements. By reducing
glycolysis, 2DG deprives rapidly growing cells of
their required cellular fuel (Cheong et  al., 2011).
For example, 2DG inhibits breast cancer cell
growth in a dose-dependent manner and causes
cell death through apoptosis (Aft et  al., 2002).
In combination with adriamycin, 2DG slows the
growth of solid tumors such as osteosarcoma
(Maschek et al., 2004). In patients with advanced
prostate cancer, a 2-week dose-escalating study
revealed dose-limiting toxicity of grade 3 asymp-
tomatic QTc prolongation at a dose of 60 mg/kg,
but doses of 45 mg/kg were well tolerated (Stein
et al., 2010).
CONCLUSIONS
The inhibition of glycolysis with compounds such
as 2DG represents a novel therapeutic approach to
epilepsy. The effects of 2DG in animal models are
summarized in Table 36.1. 2DG exerts acute anti-
convulsant effects in vitro that are independent of
359
the method of seizure induction. 2DG has acute
anticonvulsant effects in vivo in several models of
seizure induction and also possesses novel chronic
antiepileptic effects against progression of seizures
and the adverse consequences of seizure-induced
plasticity that are associated with alterations in
neuronal gene expression. Finally, 2DG has a
favorable preliminary toxicity profile. Together,
these observations support 2DG or similar agents
as feasible for treating epilepsy in patients. The
detailed anticonvulsant and antiepileptic mecha-
nisms of 2DG actions remain to be clarified.
AC K N OW L E D G M E N T S
This work was supported by The Charlie
Foundation (CES), NIH RO1 25020 (TPS), the
Epilepsy Research Foundation New Therapy
Development Project, and the Wisconsin Alumni
Research Foundation.
REFERENCES
Aft, R.L., Zhang, F.W., and Gius, D. (2002). Evaluation
of 2-deoxy-D-glucose as a chemotherapeutic
agent: mechanism of cell death. Brit J Cancer 87,
805–812.
Barton, M.E., Klein, B.D., Wolf, H.H., and White, H.S.
(2001). Pharmacological characterization of the 6
Hz psychomotor seizure model of partial epilepsy.
Epilepsy Res 47, 217–227.
Cheong, J.H., Park, E.S., Liang, J., Dennison, J.B.,
Tsavachidou, D., Nguyen-Charles, C., WaCheng,
K., Hall, H., Zhang, D., Lu, Y., et  al. (2011).
Dual inhibition of tumor energy pathway by 2-
deoxyglucose and metformin is effective against a
broad spectrum of preclinical cancer models. Mol
Cancer Ther 10, 2350–2362.
Devinney, M.J., Pan, Y.-Z., Rutecki, P.A., and Sutula,
T.P. (2009). Differential effects of glycolytic and
mitochondrial metabolism on epileptic network
synchronization in the CA3 subfield of the hippo-
campus. Soc Neurosci Abstr 331.9/I5.
Ding, Y., Wang, S., Jiang, Y., Yang, Y., Zhang, M., Guo,
Y., Wang, S., and Ding, M.P. (2013). Fructose-1,6-
diphosphate protects against epileptogenesis by
modifying cation-chloride co-transporters in a
model of amygdaloid-kindling temporal epilepti-
cus. Brain Res 1539, 87–94.
Ding, Y., Wang, S., Zhang, M.M., Guo, Y., Yang, Y.,
Weng, S.Q., Wu, J.M., Qiu, X., and Ding, M.P.
(2010). Fructose-1,6-diphosphate inhibits seizure
acquisition in fast hippocampal kindling. Neurosci
Lett 477, 33–36.
Gano, L.B., Patel, M., and Rho, J.M. (2014). Ketogenic
diets, mitochondria, and neurological diseases. J
Lipid Res 55, 2211–2228.
Garriga-Canut, M., Schoenike, B., Qazi, R., Bergendahl,
K., Daley, T.J., Pfender, R.M., Morrison, J.F.,
Chapter 36: 2-Deoxyglucose 359
Ockuly, J., Stafstrom, C., Sutula, T., et  al. (2006).
2-Deoxy-D-glucose reduces epilepsy progres-
sion by NRSF-CtBP-dependent metabolic regu-
lation of chromatin structure. Nature Neurosci 9,
1382–1387.
Gasior, M., Yankura, J., Hartman, A.L., French, A.,
and Rogawski, M.A. (2010). Anticonvulsant and
proconvulsant actions of 2-deoxy-D-glucose.
Epilepsia 5, 1385–1394.
Greene, A.E., Todorova, M.T., and Seyfried, T.N.
(2003). Perspectives on the metabolic manage-
ment of epilepsy through dietary reduction of glu-
cose and elevation of ketone bodies. J Neurochem
86, 529–537.
Hartman, A. L., Lyle, M., Rogawski, M. A., and Gasior,
M. (2008). Efficacy of the ketogenic diet in the 6-
Hz seizure test. Epilepsia 49, 334–339.
Hartman, A.L., Rubenstein, J.E., and Kossoff, E.H.
(2013). Intermittent fasting:  a “new” historical
strategy for controlling seizures? Epilepsy Res 104,
275–279.
He, X.P., Kotloski, R., Nef, S., Luikart, B.W., Parada,
L.F., and McNamara, J.O. (2004). Conditional
deletion of TrkB but not BDNF prevents epi-
leptogenesis in the kindling model. Neuron
43, 31–42.
Holmes, G. L., and Zhao, Q. (2008). Choosing the cor-
rect antiepileptic drugs: from animal studies to the
clinic. Pediatr Neurol 38, 151–162.
Hori, A., Tandon, P., Holmes, G. L., and Stafstrom,
C. E. (1997). Ketogenic diet:  effects on expres-
sion of kindled seizures and behavior in adult rats.
Epilepsia 38, 750–758.
Hu, X.L., Cheng, X., Fei, J., and Xiong, Z.Q. (2011).
Neuron-restrictive silencer factor is not required
for the antiepileptic effect of the ketogenic diet.
Epilepsia 52, 1609–1616.
Huang, Y.Z. and McNamara, J.O. (2006). Inhibiting
glycolysis to reduce seizures:  how it might work.
Nature Neurosci 9, 1351–1352.
Huttenlocher, P.R. (1976). Ketonemia and sei-
zures: metabolic and anticonvulsant effects of two
ketogenic diets in childhood epilepsy. Pediatr Res
10, 536–540.
Ingram, D.K., and Roth, G.S. (2011). Glycolytic inhibi-
tion as a strategy for developing calorie restriction
mimetics. Exp Gerontol 46, 148–154.
Lee, J., Bruce-Keller, A.J., Kruman, Y., Chan, S.L., and
Mattson, M.P. (1999). 2-Deoxy-D-glucose pro-
tects hippocampal neurons against excitotoxic and
oxidative injury:  evidence for the involvement of
stress proteins. J Neurosci Res 57, 48–61.
Lian, X.-Y., Khan, F.A., and Stringer, J.L. (2007).
Fructose-1,6-biphosphate has anticonvulsant
activity in models of acute seizures in adult rats. J
Neurosci 27, 12007–12011.
Lutas, A., and Yellen, G. (2013). The ketogenic
diet: metabolic influences on brain excitability and
epilepsy. Trends Neurosci 36, 32–40.
360
360 section IV: Ketone-Based Metabolism
Ma, W., Berg, J., and Yellen, G. (2007). Ketogenic
diet metabolites reduce firing in central neurons
by opening K(ATP) channels. J Neurosci 27,
3618–3625.
Maschek, G., Savaraj, N., Priebe, W., Braunschweiger,
P., Hamilton, K., Tidmarsh, G.F., DeYoung, L.R.,
and Lampidis, T.J. (2004). 2-Deoxy-D-glucose
increases the efficacy of adriamycin and paclitaxel
in human osteosarcoma and non-small cell lung
cancers in vivo. Cancer Res 64, 31–34.
Minor, R.K., Smith Jr, D.L., Sossong, A.M., Kaushik,
S., Poosala, S., Spangler, E.L., Roth, G.S., Lane, M.,
Allison, D.B., deCabo, R., et  al. (2010). Chronic
ingestion of 2-deoxy-D-glucose induces cardiac
vacuolization and increases mortality in rats.
Toxicol Appl Pharmacol 243, 332–339.
Neal, E.G., Chaffe, H., Schwartz, R.H., Lawson, M.S.,
Edwards, N., Fitzsimmons, G., Whitney, A., and
Cross, J.H. (2008). The ketogenic diet for the treat-
ment of childhood epilepsy:  a randomised con-
trolled trial. Lancet Neurol 7, 500–506.
Ockuly, J.C., Gielissen, J.M., Levenick, C.V., Zeal, C.,
Groble, K., Munsey, K., Sutula, T.P., and Stafstrom,
C.E. (2012). Behavioral, cognitive, and safety
profile of 2-deoxy-2-glucose (2DG) in adult rats.
Epilepsy Res 101, 246–252.
Pan, Y.Z., Ockuly, J.C., Rutecki, P.A., and Sutula, T.P.
(2008). Activity-dependent effects of 2DG influ-
encing synaptic transmission, LTP, LTD, and epi-
leptogenesis. Soc Neurosci Abstr, 147.146/W129.
Pan, Y.Z., Devinney, M.J., Rutecki, P.A., and Sutula,
T.P. (2009). Effect of glycolytic metabolism on
synaptic currents in CA3 pyramidal neurons. Soc
Neurosci Abstr 331.10/16.
Pan, Y.Z., Sutula, T., and Rutecki, P.A. (2014). Evidence
for use-dependent and presynaptic actions of 2DG
on abnormal synaptic network activity in the CA3
region of the hippocampus. Soc Neurosci Abstr
485.01.
Pan Y.Z., Sutula, T.P., Rutecki, P.A. (2015). 2DG
reduces abnormal hippocampal network excit-
ability without changing membrane properties.
Abstr #3.247, American Epilepsy Society Annual
Meeting (www.aesnet.org).
Pani, G. (2015). Neuroprotective effects of dietary
restriction: evidence and mechanisms. Semin Cell
Dev Biol 40, 106–114.
Park, M., Song, K.S., Kim, H.K., Park, Y.J., Kim, H.S.,
Bae, M.I., and Lee, J. (2009). 2-Deoxy-d-glucose
protects neural progenitor cells against oxidative
stress through the activation of AMP-activated
protein kinase. Neurosci Lett 449, 201–206.
Pelicano, H., Martin, D.S., Xu, R.-H., and Huang, P.
(2006). Glycolysis inhibition for anticancer treat-
ment. Oncogene 25, 4633–4646.
Racine, R.J. (1972). Modification of seizure activ-
ity by electrical stimulation:  II. Motor seizure.
Electroencephalogr Clin Neurophysiol 32,
281–294.
Redjak, K., Redjak, R., Sieklucka-Dziuba, M.,
Stelmasiak, Z., and Grieb, P. (2001). 2-
Deoxyglucose enhances epileptic tolerance evoked
by transient incomplete brain ischemia in mice.
Epilepsy Res 43, 271–278.
Reid, C.A., Mullen, S., Kim, T.H., and Petrou, S.
(2014). Epilepsy, energy deficiency and new thera-
peutic approaches including diet. Pharmacol Ther
144, 192–201.
Rho, J.M., and Stafstrom, C.E. (2012). The ketogenic
diet: what has science taught us? Epilepsy Res 100,
210–217.
Stafstrom, C.E., Ockuly, J.C., Murphree, L., Valley,
M.T., Roopra, A., and Sutula, T.P. (2009).
Anticonvulsant and antiepileptic actions of 2-
deoxy-D-glucose in epilepsy models. Ann Neurol
65, 435–447.
Stein, M., Lin, H., Jeyamohan, C., Dvorzhinski, D.,
Gounder, M., Bray, K., Eddy, S., Goodin, S.,
White, E., DiPaola. R.S. (2010). Targeting tumor
metabolism with 2-Deoxyglucose in patients with
castrate-resistant prostate cancer and advanced
malignancies. Prostate 70, 1388–1394.
Stringer, J.L., and Xu, K. (2008). Possible mechanisms
for the anticonvulsant activity of fructose-1,6-
diphosphate. Epilepsia 49 (Suppl. 8), 101–103.
Sutula, T., and Franzoso, S. (2008). Dose-response and
time course of action of disease-modifying effects
of 2DG on kindling progression:  effectiveness of
administration after seizures. Epilepsia 49 (Suppl
7), 382.
Tanner, G.R., Lutas, A., Martínez-François, J.R.,
and Yellen, G. (2011). Single K ATP channel
opening in response to action potential firing in
mouse dentate granule neurons. J Neurosci 31,
8689–8696.
Terse, P.S., Joshi, P.S., Bordelon, N.R., Brys, A.M.,
Patton, K.M., Arndt, T.P., and Sutula, T.P. (2016).
2-Deoxy–D-Glucose (2-DG) induced cardiac
toxicity in rat: NT-proBNP and BNP as potential
early cardiac safety biomarkers. Int J Toxicol 35,
284–293.
Vijayaraghavan, R., Kumar, D., Dube, S.N., Singh, R.,
Pandey, K.S., Bag, B.C., Kaushik, M.P., Sekhar, K.,
Dwarakanath, B.S., and Ravindranath, T. (2006).
Acute toxicity and cardio-respiratory effects of
2-deoxy-2-glucose:  a promising radio sensitiser.
Biomed Environ Sci 19, 96–103.
Wree, A. (1990). Principles of the 2-deoxyglucose
method for the determination of the local cerebral
glucose utilization. Eur J Morphol 28, 132–138.
Yang, H., Guo, R., Wu, J., Peng, Y., Xie, D., Zheng, W.,
Huang, X., Liu, D., Liu, W., Huang, L., et al. (2013).
The antiepileptic effect of the glycolytic inhibitor
2-deoxy-D-glucose is mediated by upregulation
of K(ATP) channel subunits Kir6.1 and Kir6.2.
Neurochem Res 38, 677–685.
Yao, J., Chen, S., Mao, Z., Cadenas, E., and Brinton,
R.D. (2011). 2-Deoxy-D-glucose treatment
361
induces ketogenesis, sustains mitochondrial
function, and reduces pathology in female
mouse model of Alzheimer’s disease. PLoS One
6, e21788.
Yuen, A.W.C., and Sander, J.W. (2014). Rationale for
using intermittent calorie restriction as a dietary
Chapter 36: 2-Deoxyglucose 361
treatment for drug resistant epilepsy. Epilepsy
Behav 33, 110–114.
Zhang, D., Li, J., Wang, F., Hu, J., Wang, S., and Sun,
Y. (2014). 2-Deoxy-D-glucose targeting of glucose
metabolism in cancer cells as a potential therapy.
Cancer Lett 355, 176–183.
362
37
Ketogenic Diets as Highly Effective Treatments
for Diabetes Mellitus and Obesity
E R I C C . W E S T M A N , M D , M H S , E M I LY M AG U I R E , M S C ,
A N D W I L L I A M S . YA N C Y J R . , M D , M H S
INTRODUCTION
The prevalence of overweight and obesity has dra-
matically increased in the past 40 years, with levels
now reaching epidemic proportions in both devel-
oped and developing countries (Lobstein and Baur,
2005; Mastorakos et al., 2010). Obesity brings with
it an increased risk for premature mortality and
development of comorbidities including hyperten-
sion, dyslipidemia, and type 2 diabetes mellitus
(T2DM) (Stein and Colditz, 2004). Obesity has
been shown to be one of the main etiological fac-
tors in the development of T2DM, and the current
epidemic of the disease is believed to be largely
attributable to the increased incidence of obe-
sity (The Look AHEAD Research Group, 2003).
Epidemiological studies have shown that the risk
of developing T2DM increases exponentially with
a body mass index (BMI) >28kg/m² (Clark, 2004).
And T2DM is fast becoming a serious public
health problem all over the world—the disease is
expected to affect 380  million people globally by
2025 (Tahrani et al., 2010).
Once developed, T2DM brings with it many
health, social, and economic burdens (Huang
et  al., 2010). In particular, individuals with the
disease can experience a life expectancy shortened
by up to 10 years when compared with individu-
als without the disease, due to an increased risk
of both macrovascular and microvascular com-
plications (Bottomley and Raymond, 2007). As
the worldwide prevalence of the disease increases,
so too does the demand for and cost of medi-
cal care in order to treat this progressive disease
(Brandle et  al., 2003; Bottomley and Raymond,
2007). Reports show that effective glycemic con-
trol is one of the strongest predictors in decreasing
annual healthcare costs among individuals with
T2DM (Hansen et  al., 2010). As well as having
direct costs to society and to an individual’s health,
T2DM brings with it many indirect costs. These
include loss of productivity for employers and loss
of income for both employer and employee due
to chronic sickness and absence from work, early
retirement, or premature mortality (Bottomley
and Raymond 2007).
Type 2 diabetes mellitus has long been
regarded as a chronic progressive condition that
is heterogeneous in nature (Lim et  al., 2011;
Nyenwe, 2011). It is a complex disorder in which
the interaction of both the environment and an
individual’s genetics results in the development
of insulin resistance (IR) and, eventually, β-cell
dysfunction (Leahy, 2005). The β-cell function
declines over a period of time, and the develop-
ment of IR to T2DM is progressive, with some
evidence showing it can take up to 12  years for
the disease to develop (Nyenwe, 2011). Obesity
worsens IR, as adipose tissue releases increased
amounts of hormones (e.g., adiponectin), pro-
inflammatory cytokines (e.g., tumor-necrosis-α),
and other factors that are involved in the develop-
ment of IR (Kahn et al. 2000). As obesity becomes
a chronic state, there is an increase in the release
of nonesterified fatty acids (NEFAs), particularly
from intra-abdominal fat (Leahy, 2005). This
is believed to be one of the crucial factors link-
ing obesity, specifically central obesity, to IR and
eventually T2DM (Stumvoll et al., 2005). Insulin
resistance is thought to impair liver and pancre-
atic β-cell function, consequently reducing the
target tissues’ sensitivity to insulin (Tahrani et al.,
2010). In addition to defective insulin action and
secretion, nonsuppressible glucagon secretion
after a meal is also evident in individuals with
T2DM (Nyenwe, 2011).
With the significant health, social, and eco-
nomic burdens associated with T2DM, it is clear
that devising effective interventions for treating
363

Chapter 37: Ketogenic Diets as Highly Effective Treatments 363

and, most importantly, preventing the disease
are imperative. A  fundamental approach for the
effective prevention and treatment of T2DM is
weight management (Miles et  al., 2002; Nyenwe,
2011). Currently, it is understood that a moderate
weight loss of between 5% and 10% of body weight
has been shown to offer improvements in glyce-
mic control, blood pressure, and dyslipidemia in
patients with T2DM.
ORIGINS OF THE USE
O F   K E TO G E N I C D I E T S
T O   T R E AT O B E S I T Y
One of the first descriptions of a low-carbohydrate
weight-loss program was written in 1863 by
William Banting, called A Letter on Corpulence
(Banting, 1863). As a result of Banting’s letter, a
low-carbohydrate, high-fat diet was a widely rec-
ommended diet for obesity until the mid-20th
century. John Yudkin, a professor of nutrition in
the United Kingdom, brought attention to carbo-
hydrate restriction as a method of losing weight,
and suspected that sugar was a culprit in modern
illnesses (Yudkin, 1959).
an increase in blood or urine ketone bodies (typ-
ically < 20 grams of dietary carbohydrate/day).
These recent studies for the treatment of obesity
have been summarized in several meta-analy-
ses (Nordmann et al., 2006, Hession et al., 2008,
Santos 2012, Krieger 2006, Hu 2012, Johnston
2014). These studies are notable in that, despite
the higher fat content of the low-carbohydrate
diet, the adverse changes in serum cholesterol
that were predicted from other diet studies did
not occur. The low-carbohydrate diet improves
the cardiometabolic risk factors by lowering
serum triglyceride and raising HDL-cholesterol,
both components of the metabolic syndrome
(Table 37.1; Volek and Feinman, 2005). There are
very few dietary disease endpoint outcome stud-
ies, but a low-carbohydrate diet was an inter-
vention arm in a study assessing carotid intimal
thickness after 2 years (Shai et al, 2010). In this
study, all diet interventions, including the low-
carbohydrate diet led to significant regression of
measurable carotid intimal thickness. The effect
appeared to be mediated by weight loss–induced
decline in blood pressure.

RANDOMIZED, ORIGINS OF THE USE

CONTROLLED TRIALS
O F   C A R B O H Y D R AT E
RESTRICTION FOR OBESITY
Marginalized as a “fad diet,” the low-carbohy-
drate diet was forgotten as a mainstream treat-
ment for obesity in the last half of the 20th
century. Over the last 15 years, several indepen-
dent groups have reexamined low-carbohydrate
diets and given them the new name of “carbohy-
drate-restricted diets,” or “ketogenic diets” if the
dietary carbohydrate is sufficiently low to cause
O F   K E TO G E N I C D I E T S
T O   T R E AT D I A B E T E S
MELLITUS
The discovery of insulin and dissemination of
its use in the early 1920s was truly miraculous
for individuals with type 1 diabetes mellitus
(T1DM), who had no endogenous insulin pro-
duction. Insulin treatment was quickly incor-
porated into clinical practice (Bliss, 1982). The
widespread use of insulin and subsequently
developed antiglycemic medications for T2DM,

TABLE 37.1 OUTPATIENT RANDOMIZED, CONTROLLED TRIALS OF CARB OHYDRATE

VERSUS FAT RESTRICTION FOR OBESIT Y: RESULTS

Fat Restricted Diet Carbohydrate Restricted Diet

Reference n Weight LDL Trig HDL Weight LDL Trig HDL Duration

Sondike 30 –4.1 kg –17%* –6% +2% –9.9 kg* +4% –48%* +4% 3 months
Brehm 42 –3.9 kg –5% +2% +8% –8.5 kg* 0% –23%* +13% 6 months
Samaha 132 –3.1 kg –3% +2% –12% –5.1 kg* +6% –29%* –2%* 12 months
Foster 63 –4.5 kg –6% +1% +3% –7.3 kg +1% –28%* +18%* 12 months
Yancy 119 –6.5 kg –3% –15% –1% –12.0 kg* +2% –42%* +13%* 6 months
Brinkworth 40 –11.5 kg +3%* –12%* 0%* –14.5 kg +3%* –35%* +21%* 12 months
Shai** 211 –2.9 kg* –0.1% –2%* +17%* –4.7 kg* –3% –13%* +22%* 24 months
* p < .05 for between group comparison
** Study also included a third Mediterranean diet intervention (not shown here)
364
364 section IV: Ketone-Based Metabolism
however, have not allowed individuals with dia-
betes to live entirely normal lives. Like T1DM,
T2DM is known as a chronic disease with high
rates of retinopathy, nephropathy, and vas-
culopathy (Feudtner, 2003). Importantly, the
pathophysiology of the majority (95%) of indi-
viduals with diabetes today is related to insulin
resistance (T2DM), not to insulin deficiency
(T1DM). Today, individuals with either T1DM
or T2DM are instructed to self-monitor their
blood glucose and self-administer subcutane-
ous insulin to achieve normoglycemia. Despite
intensive management with insulin and oral
medications, normoglycemia remains an elusive
goal (DCCT 1993, UKPDS 1998).
In hindsight, the rapid adoption of insulin ther-
apy and the hope that insulin would “cure” T2DM
may have led to a premature departure from suc-
cessful treatments for T2DM prior to the discovery
of insulin. In the pre-insulin era, the pharmaco-
logical treatment of diabetes mellitus included
a combination of modalities including alcohol,
opioids, arsenic, and potassium bromide. More
potent than medication, though, was the high-fat,
low-carbohydrate diet. From a textbook written in
1877: “There are few diseases which present to the
practitioner so clear an indication of what is to be
done … a Diabetic should exclude all saccharine
[sugary] and farinaceous [starchy] materials from
his diet” (Morgan, 1877).
The leading authorities in the early 1900s,
Frederick Madison Allen and Elliott P. Joslin, pub-
lished several studies supporting the use of a low-
carbohydrate, high-fat diet for diabetes. Allen used
a pancreatectomized dog model of diabetes and
observed that feeding a high-carbohydrate diet
led to glycosuria. If, however, the dogs were fed a
high-fat, low-carbohydrate diet, the glycosuria was
no longer detectable. In the treatment of diabetes
in humans, Allen employed fasting, then a step-
wise reintroduction of macronutrients to find the
threshold at which glycosuria developed (Allen,
1914, 1915a, 1915b, 1920). Using this method, the
average diet recommendation for diabetes was a
diet containing 70% fat, 18% protein, 4% alcohol,
and only 8% of calories from carbohydrate (Allen
et al., 1919; Westman et al., 2006).
Like Allen, Joslin recommended a 70% fat,
10% carbohydrate diet for the treatment of diabe-
tes (Joslin, 1928). Joslin categorized carbohydrate-
containing foods by their carbohydrate content,
and advised his patients to eat vegetables with
less than 5% carbohydrate content (Joslin, 1919).
According to his classification, vegetables with car-
bohydrate content from 1% to 3% included lettuce,
cucumbers, spinach, asparagus, rhubarb, endive,
dandelion greens, swiss chard, celery, and mush-
rooms. Vegetables with carbohydrate content from
3% to 5% were tomatoes, brussels sprouts, water-
cress, sea kale, okra, cauliflower, eggplant, cabbage,
canned string beans, broccoli, and canned arti-
chokes. Non-carbohydrate-containing foods were
not limited (meat, eggs, fish).
So, before the modern medical therapies of
medication and insulin, the latter being discov-
ered in 1921, the leaders in diabetes used a low-
carbohydrate, high-fat diet for the treatment of
diabetes.
D I E TA RY
R E C O M M E N D AT I O N S
FOR DIABETES IN THE
P O S T- I N S U L I N   E R A
The discovery and therapeutic use of insulin
changed the management of T1DM and T2DM,
including its dietary management (Franz et  al.,
2002). Sansum achieved the absence of glycosu-
ria with increased levels of dietary carbohydrate
by increasing the dose of insulin (Sansum, 1928).
From the calculated means of the cases reported,
an increase in dietary carbohydrate from 41.9 g
to 196 g could be made with an increase of insu-
lin from 20.4 units to 85.6 units/day on average.
It was also observed that in patients already on
insulin, glycosuria remained absent and carbohy-
drate could be increased without increasing the
insulin dosage if the caloric intake was reduced
(Richardson, 1929; Rabinowitch, 1930). From
the calculated means of the cases reported, an
increase from 73g to 144g of carbohydrate could
be made by reducing the caloric intake from 1,788
kcal to 1,427 kcal on average. However, it is inter-
esting to note that Richardson studied individu-
als with T1DM. In regard to T2DM, Richardson
wrote, “it has been evident that with patients who
are fat or markedly overweight it is not possible
to make this change in the fat and carbohydrate
of the diet. The tolerance has been exceeded and
sugar in the urine has resulted in all of those cases
which we have tried” (Richardson, 1929). Higher
carbohydrate intakes could be consumed, but
glycosuria would appear unless the insulin dos-
age was increased, or the total caloric intake was
decreased.
Note that these observations were made
using glycosuria as the measure of glycemic con-
trol. While individual variability exists, glucose
does not typically appear in the urine until the
serum level is greater than 180 mg/dL (Buse
et al., 2003).
365
Chapter 37: Ketogenic Diets as Highly Effective Treatments
E VO L U T I O N
O F   T H E D I E TA RY
R E C O M M E N D AT I O N S
FOR TYPE 2 DIABETES
After the use of insulin became routine, the
amount of carbohydrate in the “diabetic diet”
was gradually increased. Although carbohydrate
counting was still an important part of the diabe-
tic diet recommended by the American Dietetic
Association as late as 1950, by 1971 the guidelines
read “Important dietary concepts have developed
in the last decade which require some alteration in
long-held precepts. There no longer appears to be
any need to restrict disproportionately the intake
of carbohydrate” (Bierman et al., 1971). (It is inter-
esting in retrospect that there are no references
included in this 1971 position paper.)
It appears that the change in guidelines to
soften the limitation of dietary carbohydrate was
based on a few small studies. In 1963, a study of
insulin-dependent diabetic patients compared two
2,200-kcal eucaloric diets containing two relatively
high levels of carbohydrate (41% vs. 64%), and
found that both diets led to similar glucose control
(Stone and Connor, 1963). Blood glucose meas-
urement was not used, however, and good glucose
control was defined as no glycosuria and few hypo-
glycemic episodes. According to the authors, “In
practical terms, this meant the avoidance of more
than minimal glycosuria, the avoidance of more
than 10 gm. of glucose per twenty-four-hour spec-
imen of urine and of more than one mild hypogly-
cemic episode every two weeks.”
In 1971, a paper was published involving 13
patients with mild diabetes who were given liq-
uid formula diets containing 85%CHO/15%PRO/
0%FAT or 45%CHO/15%PRO/40%FAT diet for
a period of 8–10 days (Brunzell et al., 1971). The
carbohydrate was either dextrose or a mixture of
dextrins and maltose; calories were adjusted to
maintain a constant body weight. On the 85% car-
bohydrate diet, the fasting blood glucose was 91
mg/dL and the fasting insulin was 16.2, compared
with 100.2 mg/dL and 20.8, respectively, on the
lower carbohydrate diet. The conclusion was that
the glucose control was similar for a moderate or
high carbohydrate diet.
It appears that the recommendation of a 55%
carbohydrate diet was based on extremely small
studies that used either glycosuria or hypoglyce-
mia as the measure of diabetes control (not precise
enough by today’s standards) and compared rela-
tively high levels of carbohydrate intake, which does
not take into account the clinical experience of low-
carbohydrate diets prior to the discovery of insulin.
365
Clearly, the discovery and clinical application
of insulin therapy revolutionized the treatment of
diabetes mellitus due to insulin deficiency, and no
one would question the use of insulin for T1DM.
For diabetes mellitus related to insulin resistance,
however, it is not clear that medication therapy
(including insulin) is superior to a high-fat, low-
carbohydrate diet for glycemic control and avoid-
ance of long-term complications. This study was
never done. It can be argued that because the path-
ophysiology of T2DM involves insulin resistance
and hyperinsulinemia, using insulin therapy is
counterintuitive.
Several limitations temper our ability to
directly apply this historical information today.
Due to the possibility of spectrum bias, it is dif-
ficult to ascertain the severity of diabetes that was
treated by Allen and Joslin. Their patients may
have had less severe or progressed T2DM and
therefore might not be representative of the dia-
betic population of that time, since individuals
with more severe disease are likely to have died.
There was also no formal distinction between type
1 and type 2 diabetes mellitus at the time, though
it was noted that children and young adults pre-
sented with weight loss (probably type 1), while
older patients were often obese (probably type 2).
In the sample of Allen’s patients, with a mean age
of 40, most of these patients probably had type 2
diabetes.
In summary, one of the widely recommended
treatments of diabetes mellitus in the early 1900s
before the introduction of medication therapy was
a high-fat, low-carbohydrate diet.
NUTRITION THERAPY
F O R   D I A B E T E S :   L OW
G LY C E M I C D I E TA RY
PAT T E R N S
The amount of glucose in the human bloodstream
is regulated very closely, and chronic, small eleva-
tions of serum glucose lead to glucose intolerance,
T2DM, and the associated complications. The
amount of serum glucose in an adult with a serum
glucose of 100 mg/dL and 5–7 liters of blood is
about 5–7 grams (about the amount contained in a
heaping teaspoon of table sugar or a few medium-
sized strawberries). Normal serum glucose ranges
from 80 to 99 mg/dL, and when a fasting serum
glucose is elevated to 100–124 mg/dL, the diagno-
sis of impaired fasting glucose is made. When the
fasting glucose is above 124 mg/dL or a random
glucose is elevated above 200 mg/dL with com-
mon symptoms such as polydipsia or polyuria, the
diagnosis of T2DM is made (American Diabetes
366

366 section IV: Ketone-Based Metabolism

Association, 2010). These small changes in concen-
tration in serum glucose represent changes in the
amount of glucose in the blood of only a few grams.
The “glycemic index” is a concept that cat-
egorizes single foods containing carbohydrate
on the basis of the rise in blood glucose after the
ingestion of  a standard amount of  carbohydrate
from the particular food (Jenkins et al., 1981; Holt
et al., 1997; Ludwig, 2002). This glucose response
is then compared with the glucose response to a
standard weight of  white bread or glucose. The
“glycemic load” is a variation of  the “glycemic
index” in that the usual serving size of the food
is the unit of comparison, in the attempt to make
the construct more clinically useful. One problem
with  these constructs is that the  glycemic effect
of a meal is a function of not only the glycemic
index of  the carbohydrate, but also of  the other
macronutrients and foods ingested at  the same
time. Despite these criticisms, low glycemic index
diets can lead to improvement in diabetic control
(Brand-Miller et al., 2003). However, the magni-
tude of this effect on glycemic control in patients
with  T2DM is modest, with  a typical response
of  0.43% reduction compared with  higher gly-
cemic diets. Because the  “low glycemic” diets
in these studies have contained from 40% to 60%
of  calories from  carbohydrate, it is possible that
a more potent effect on  lowering blood glucose
may be observed with a reduction in the percent-
age of  carbohydrate and not just the  glycemic
index.

Glucose response
900
Net Areas 50 Total Areas 500
a 8000
600 400
25

mmol • h/L
mmol • h/L
mg • h/dL

mg • h/dL

300 6000 a 300

0 0
4000 b 200
–300 c
–25
b 2000 100
–600 b
–900 –50 0 0 20
350 Standard CHO Fasting Standard CHO Fasting
Diet Free Diet Free 18
300
16

250 14
mmol/L
12
mg/dL

200
10

150 8

6
100
Fasting 4
50 CHO Free
Standard Diet 2
B L D
0 0
0 2 4 6 8 10 12 14 16 18 20 22 24
Time (Hours)

FIGURE  37.1 Fasting and Postprandial Glucose Response of a Ketogenic Diet 24-h glucose response. The open
circle–solid line represents the mean glucose concentration at several time points during the first 24 h of both days
during which the standard diet was ingested (i.e. day 1 of each arm of the study). The triangle–dotted line represents
the mean glucose concentration during the last 24 h on a carbohydrate-free diet. The closed circle–solid line represents
the mean glucose concentration during the last 24 h of the fast (energy-free) diet. B, L, D, indicate the times at which
breakfast, lunch, and dinner were ingested.The net area response (Left Insert) indicates the area under the curve using
the fasting concentration as baseline. Different letters on bars indicate statistically significant differences (Friedman
P = <0.0012).The total area response (Right Insert) indicates the area under the curve, using zero as baseline. Different
letters on bars indicate statistically significant differences (Friedman: P = <0.0001) (Nuttall, Almokayyad, Gannon, 2015).
367
Chapter 37: Ketogenic Diets as Highly Effective Treatments
G LY C E M I C E F F E C T O F
D I E TA RY C A R B O H Y D R AT E
In controlled studies, consuming foods of lower gly-
cemic index versus foods of higher glycemic index
leads to a reduction in postprandial glucose and insu-
lin. When the glycemic concept is taken further and
applied to the reduction of the amount of carbohy-
drate to less than 20–40 g/d—all of the carbohydrate-
containing foods are “very low” to “no” glycemic
index foods. (Fat has a glycemic index of 0 because
the consumption of fat does not raise the blood glu-
cose.) When the carbohydrate intake is this low, the
glucose and insulin responses to the meals are even
lower. In fact, the fasting and postprandial glucose
and insulin levels after carbohydrate-deficient meals
are almost as low as total fasting (not eating any-
thing at all) (Figure 37.1; Nuttall et  al., 2015). The
lack of postprandial rise in glucose and insulin for an
extremely low carbohydrate diet (<20 g/day) has been
replicated in two other studies in subjects without
diabetes (Bisschop et al., 2000; Noakes et al., 2006).
In summary, lowering the glycemic index and
the absolute amount of carbohydrate in the diet
can have a profound effect on lowering the blood
glucose; dietary protein has a small effect on blood
glucose; dietary fat has none. This is the rationale
for use of a low-carbohydrate diet for the treat-
ment and prevention of diabetes.
C O N T R O L L E D O U T PAT I E N T
S T U D I E S O F   L OW E R
C A R B O H Y D R AT E D I E T S
FOR TYPE 2 DIABETES
MELLITUS
From 1998 to 2004, several controlled studies
showed that lowering the dietary carbohydrate can
lead to a reduction in fasting blood glucose and
HbA1c, and that this effect can be independent of
weight loss (Table 37.2). However, none of these
studies lowered the carbohydrate intake to the lev-
els of the ketogenic diet, which had been employed
clinically in the early 1900s, as discussed earlier.
A review of existing studies in 2008 concluded that
a low-carbohydrate diet would be useful for the
treatment of T2DM (Dyson, 2008).
M E TA B O L I C WA R D S T U D Y
O F   T H E K E TO G E N I C D I E T
FOR TYPE 2 DIABETES
MELLITUS
A major limitation of out-patient clinical trials is
that the dietary intake is estimated by self-reported
intake questionnaires and may not reflect actual
dietary intake. This limitation was overcome in a
carefully controlled metabolic ward study among
367
individuals with T2DM. In this study, an ad libi-
tum low-carbohydrate, ketogenic diet (<20 g/day)
led to a spontaneous reduction in caloric intake
and improvement in insulin sensitivity measured
by insulin clamp technique (Boden et  al., 2005).
Additionally, serum glucose levels improved substan-
tially and consistently enough in the 10 subjects that
hemoglobin A1c improved significantly from base-
line after only 14 days on the low-carbohydrate diet.
RANDOMIZED,
CONTROLLED TRIALS
O F   C A R B O H Y D R AT E
RESTRICTION FOR TYPE 2
DIABETES MELLITUS
There has been a recent interest in studying the
use of low-carbohydrate diets for the treatment of
T2DM. Several of the recent studies have limited
carbohydrate intake to the degree that was used by
Allen and Joslin (Table 37.3, Daly 2006, Westman
2008, Gulbrand 2012, Saslow 2014, Tay 2014,
Yamada 2014, Mayer 2014). The consistent find-
ing is that the lower carbohydrate, even ketogenic,
diets lead to a greater improvement in T2DM than
the diets containing higher levels of carbohydrate.
Lowering dietary carbohydrate intake demon-
strated benefits on glycemic control beyond its
weight loss effects in some of the studies, while at
the same time lowering antiglycemic medication
requirements.Two meta-analyses have also sum-
marized these studies (Kirk 2008, Dyson 2008).
P I L OT S T U DY O F   A
K E TO G E N I C D I E T
FOR WEIGHT LOS S
IN OBESITY VERSUS TYPE 2
DIABETES MELLITUS
While clinical experience suggests that weight loss
is slower for individuals with T2DM than obesity
alone, there are few studies directly comparing this
clinical observation for ketogenic diets. One of this
chapter’s authors (EM) conducted a retrospective
cohort study to assess the rate of weight change that
occurred between a group of overweight or obese
type 2 diabetics compared with age-, weight-, and
gender-matched control groups of overweight or
obese individuals (unpublished data).
The intervention was the Go Lower Low
Carbohydrate Diet, which provided around 10%
of total energy from carbohydrates, 30% of total
energy from protein, and 60% of total energy from
fat. All meals were delivered to individuals, pro-
viding them with breakfast, lunch, dinner, and up
to 2 snacks per day. Weekly support was provided
to participants with additional nutrition support
 368
TABLE 37.2 CONTROLLED STUDIES OF LOWER CARB OHYDRATE DIETS FOR T YPE 2 DIABETES MELLITUS

Reference N BMI KCAL CHO PRO FAT Duration Post Pre Post Weight
Glucose HgbA1c HgbAIc Change
kg/m2 % % % mg/dL % % kg

Gutierrez 19 27.9 Varied 55 20 25 12 weeks 183 8.1 8.9 –0.8

1998 25 45 30 8 weeks 192 9.9 8.1 –1.4
Heilbronn 35 33 1541 73 17 10 12 weeks 131 8.5 7.0 –6.6
1999 1596 50 18 32 MUF 119 7.8 6.8 –6.6
1613 50 17 33 SF 122 7.5 6.6 –6.6
Heilbronn 45 22 1440 60 HGI 22 18 12 weeks 110 6.7 6.1 –4.8
2002 1440 60 LGI 22 18 116 6.7 6.0 –4.4
Gannon 12 31 2250 55 15 34 5 weeks 114 8.0 7.7 0
2003 2235 40 30 30 114 8.1 7.3 0
Rizkalla 12 31 2291 38 HGI 20 37 4 weeks 176 7.5 7.6 –0.6
2004 2222 36 LGI 21 37 165 7.6 7.2 –0.6
Gannon 8 31 2825 55 15 30 5 weeks 198 9.8 9.8 –1.8
2004 2835 20 30 50 126 9.8 7.6 –1.8

BMI = body mass index, HGI = high glycemic index, LGI = low glycemic index, MUF = high monounsaturated fat, SF = high saturated fat
 369
TABLE 37.3 OUTPATIENT RANDOMIZED, CONTROLLED TRIALS OF CARB OHYDRATE
RESTRICTION FOR T YPE 2 DIABETES MELLITUS

Reference N BMI Carbohydrates Duration Start End Difference Difference

HgbA1c HgbA1c HgbA1c Weight
kg/m2 grams/day months percent percent Percent kilograms

Daly 102 36 170 3 9.0 8.8 –0.2 –0.9

2006 110 9.0 8.5 –0.5 –3.6
Westman 49 38 100 6 8.3 7.8 –0.5* –6.9
2008 20 8.8 7.3 –1.5* –11.0
Guldbrand 61 32 225 24 7.2 7.4 +0.2 –4.0
2012 90 7.5 7.5 –0.0 –4.0
Saslow 34 36 138 3 6.9 6.9 0* –2.6
2014 58 6.6 6.0 –0.6* –5.5
Tay 115 34 205 6 7.3 6.6 –0.7* –12
2014 57 7.4 6.2 –0.6* –12
Yamada 24 25 203 6 7.7 7.5 –0.2* –1.4
2014 126 7.6 7.0 –0.6* –2.6*
Mayer 46 40 156 + O 12 7.6 7.7 +0.1* –8.1
2014 76 7.6 6.9 –0.7* –7.5
*p < .05 for between group comparison.
n = sample size, O = Orlistat
370
370 section IV: Ketone-Based Metabolism
provided by qualified nutritionists. Between weeks
9 and 12, only breakfast and snacks were provided,
which required individuals to prepare two meals
per day.
A sample of 20 participants (10 obese with dia-
betes and 10 obese without diabetes) was identified
for inclusion in this study. At baseline, the mean
age was 55.5 years, mean BMI was 36.0 kg/m2,
and 70% were female, and the mean HgbA1c was
12.3%. After 3 months of following the dietary
protocol, obese nondiabetic participants had lost
significantly more weight than obese type 2 dia-
betic participants (-20.56 ± 8.8 kg versus -12.84 ±
8.1 kg; p = .05). However, there was no significant
difference between these groups in mean weight
change at 12 months. These results are in line with
other previous studies that have demonstrated that
patients with type 2 diabetes do not lose as much
weight as their nondiabetic counterparts (Khan et
al., 2000; Miles et al., 2002).
NONRANDOMIZED STUDIES
U S I N G T H E   K E TO G E N I C
DIET IN TYPE 2 DIABETES
MELLITUS
In order to assess the effect of the ketogenic diet
on T2DM in clinical practice, one of the authors
(ECW) performed chart reviews of physicians
who had used carbohydrate restriction for many
years in their clinical practices (Hickey et al., 2003;
O’Neill et  al., 2003; Vernon et  al., 2003; Vernon
et al., 2004) (Table 37.4). Nielsen also reported the
results of long-term follow-up in a clinical setting
(Nielsen and Joensson, 2008). These cases show
that reversal of T2DM is possible in the context of
selection of highly motivated individuals, or as a
result of programs that have excellent adherence.
One of these research projects was a prospec-
tive study which enrolled 28 overweight sub-
jects with T2DM into a 16-week single-arm diet
TABLE 37.4 CLINICAL PRACTICE AND NONRANDOMIZED STUDIES
OF CARBOHYDRATE RESTRICTION FOR T YPE 2 DIABETES MELLITUS
Reference N Baseline Carbohydrates
Weight (kg) Pct of Calories/day
Vernon 2003 14 123.2 10%
O’Neill 2003 20 82 10%
Nielsen 2005 16 100 20%
Yancy 2005 28 131.4 7%
intervention trial at the Durham Veterans Affairs
Medical Center (Yancy et al., 2003). At diet ini-
tiation, diabetic medication was reduced to avoid
hypoglycemia. At baseline, the mean age was 56.3
years and the mean BMI was 42.2 kg/m2. After
16 weeks, the mean hemoglobin A1c decreased
by 15% from baseline (7.4% to 6.3%, p < .001).
Medications for diabetes were discontinued in 7
subjects, reduced in 10 subjects, and unchanged
in 5 subjects. The mean body weight decreased by
6.5% from 130.9 kg to 122.4 kg (p < .001). Fasting
serum triglyceride decreased 40% (2.61 mmol/L to
1.54 mmol/L, p < .001), while other serum lipid
measurements did not change significantly.
CASE REPORT
OF REMIS SION OF SEVERE
TYPE 2 DIABETES MELLITUS
W I T H   K E TO G E N I C   D I E T
One of the authors (ECW) recently treated an oth-
erwise healthy 60-year-old Caucasian male with a
BMI of 26.9 kg/m2 who had a new onset of T2DM.
The initial hemoglobin A1C was 10.5% on routine
annual testing, and serum glucose levels were
also reflective of hyperglycemia. He was given
instruction on how to follow a low-carbohydrate,
ketogenic diet (<20 grams of carbohydrate/day),
and the repeat hemoglobin A1C at 1, 5, 12, and
24 months was 6.4%, 5.5%, 5.6%, and 5.5% respec-
tively (Figure 37.2). The current recommended
approach to manage severe diabetes mellitus is
to start chronic insulin therapy. This case report
suggests that potent lifestyle approaches like the
ketogenic diet may be useful instead of medication
therapy.
CONCLUSION
In summary, lifestyle modification that reduces
carbohydrate intake is effective in the treatment
of obesity, T2DM, and metabolic syndrome.
Duration Pre Post Weight
Months HbA1c HBA1c Difference
% % %
8 10.0 5.9 –9.7
10 8.4 5.8 –6.7
12 8.0 6.6 –11.9
4 7.5 6.3 –6.6
371

Chapter 37: Ketogenic Diets as Highly Effective Treatments 371

64 y/o Male, BMI 29.1, New Onset Diabetes

205 12
200 11
10.5
195 198 10
190 9
8
185 189
7
180 6.3
5.9 6
175 5.5
5
170
Weight (pounds) 4
165 170 169 169 3
HgbA1c (%)
160 2
155 1
150 0
Oct-10
Jan-11
Apr-11
Jul-11
Oct-11
Jan-12
Apr-12
Jul-12
Oct-12
Jan-13
Apr-13
Jul-13
Oct-13
Jan-14
Apr-14
Jul-14
Oct-14
Jan-15
Low Carbohydrate Ketogenic Diet
(No medications)

FIGURE 37.2 Treatment of New Onset Type 2 Diabetes Mellitus with a Ketogenic Diet

A low-carbohydrate, ketogenic diet combines two D.B., Larson, G., Lustig, R.H., et al. (2008). Dietary
approaches that, on their own, improve blood glu- carbohydrate restriction in type 2 diabetes mel-
cose control: weight loss and reduction of glycemic litus and metabolic syndrome:  time for a critical
index/load diet response to dietary intake. appraisal. Nutr Metab (Lond) 5, 9.
As for most ongoing lifestyle change programs, Allen, F.M. (1914). Studies concerning diabetes. JAMA
63, 939–943.
research efforts to improve dietary adherence are
Allen, F.M. (1915a). Prolonged fasting in diabetes.
needed. We have found that introducing the con-
Trans Assoc Am Physicians 30, 323–329.
cept that T2DM is possibly reversible, and that Allen, F.M. (1915b). The treatment of diabetes. Boston
individuals may be able to have normal blood Med Surg J 172, 241–247.
glucose off medications is highly motivating, and Allen, F.M., Stillman, E., and Fitz, R. (1919). Total
the reduction or elimination of medications may dietary regulation in the treatment of diabetes.
improve long-term adherence. Most individuals Monograph No. 11. (New  York:  The Rockefeller
with T2DM have been told repeatedly that they Institute for Medical Research).
will always have diabetes and always require Allen, F.M. (1920). Protein diets and undernutrition in
medication—but this is clearly not the case in treatment of diabetes. JAMA 74, 571–577.
some individuals with amelioration of the under- American Diabetes Association. (2010). Diagnosis
lying insulin resistance. and classification of diabetes mellitus. Diabetes
The underlying principle of carbohydrate Care 33 (Suppl 1), S62.
restriction and the history of using the low- Banting, W. (2005) [1863]. Letter on Corpulence.
(New York: Cosimo Classics).
carbohydrate diet for T2DM suggest that the
Bierman, E.L., Albrink, M.J., and Arky, R.A. (1971).
low-carbohydrate approach may play an impor- Special report: principles of nutrition and dietary
tant nonpharmacologic role in reversing the recommendations for patients with diabetes mel-
current epidemic of “diabesity” (Accurso et  al., litus: 1971. Diabetes 20, 633–634.
2008; Feinman et  al., 2015; Khazrai et  al., 2014). Bisschop, P.H., Pereira Arias, A.M., Ackermans,
Furthermore, it suggests that the common M.T., Endert, E., Pijl, H., Kuipers, F., Meijer, A.J.,
approach of antiglycemic medication + higher car- Sauerwein, H.P., and Romjin, J.A. (2000). The
bohydrate diets should be more rigorously tested effects of carbohydrate variation in isocaloric diets
against the low-carbohydrate, ketogenic approach on glycogenolysis and gluconeogenesis in healthy
for the treatment of T2DM. men. J Clin Endocrinol Metab 85, 1963–1967.
Bliss, M. (1982). The Discovery of Insulin.
REFERENCES (Chicago: University of Chicago Press).
Accurso, A., Bernstein, R.K., Dahlqvist, A., Draznin, Boden, G., Sargrad, K., Homko, C., Mozzoli, M., and
B., Feinman, R.D., Fine, E.J., Gleed, A., Jacobs, Stein, T.P. (2005). Effect of a low-carbohydrate diet
372
372 section IV: Ketone-Based Metabolism
on appetite, blood glucose levels, and insulin resis-
tance in obese patients with type 2 diabetes. Ann
Intern Med 142, 403–411.
Bottomley, J.M., and Raymond, F.D. (2007). Pharmaco-
economic issues for diabetes therapy. Best Pract
Res Clin Endocrinol Metab 21, 657–685.
Brandle, M., Zhou, H., Smith, B.R., Marriott, D.,
Burke, R., Tabaei, B.P., Brown, M.B., and Herman,
W.H. (2003). The direct medical cost of type 2 dia-
betes. Diabetes Care 26, 2300–2304.
Brand-Miller, J., Hayne, S., Petocz, P., and Colagiuri, S.
(2003). Low-glycemic index diets in the manage-
ment of diabetes:  a meta-analysis of randomized
controlled trials. Diabetes Care 26, 2261–2267.
Brehm, B.J., Seeley, R.J., Daniels, S.R., and D’Alessio,
D.A. (2003). A randomized trial comparing a very
low carbohydrate diet and a calorie-restricted
low fat diet on body weight and cardiovascular
risk factors in healthy women. J Clin Endocrinol
Metab 88, 1617–1623.
Brinkworth, G.D., Noakes, M., Buckley, J.D., Keogh,
J.B., and Clifton, P.M. (2009). Long-term effects
of a very-low-carbohydrate weight loss diet com-
pared with an isocaloric low-fat diet after 12 mo.
Am J Clin Nutr 90, 23–32.
Brunzell, J.D., Lerner, R.L., Hazzard, W.R., Porte, D.,
Jr., and Bierman, E.L. (1971). Improved glucose
tolerance with high carbohydrate feeding in mild
diabetes. N Engl J Med 284, 521–524.
Buse, J.B., Polonsky, K.S., and Burant, C.F. (2003).
Type 2 Diabetes Mellitus. In Williams Textbook
of Endocrinology, 10th ed., P.R. Larsen, H.M.
Kronenberg, S.  Melmed, and K.S. Polonsky, eds.
(Philadelphia: Saunders), p. 1457.
Clark, M. (2004). Is weight loss a realistic goal of
treatment in type 2 diabetes? The implications of
restraint theory. Patient Educ Couns, 53, 277–283.
DCCT. (1993). The effect of intensive treatment of dia-
betes on the development and progression of long-
term complications in insulin-dependent diabetes
mellitus. The Diabetes Control and Complications
Trial Research Group. N Engl J Med 329, 977–986.
Daly, M.E., Paisey, R., Paisey, R., Milward, B.A., Eccles,
C. Williams, K., Hammersley, S., MacLeod, K.M.,
and Gale, T.J. (2006). Short-term effects of severe
dietary carbohydrate-restriction advice in Type 2
diabetes—a randomized controlled trial. Diabet
Med 23, 15–20.
Dashti, H.M., Al-Zaid, N.S., Mathew, T.W., Al-
Mousawi, M., Talib, H., Asfar, S.K., and Behbahani,
A.I. (2006). Long-term effects of ketogenic diet in
obese subjects with high cholesterol level. Mol Cell
Biochem 286, 1–9.
Dyson, P.A. (2008). A review of low and reduced car-
bohydrate diets and weight loss in type 2 diabetes.
J Hum Nutr Diet 21(6), 530–538.
Feinman, R.D., Pogozelski, W.D., Astrup, A.,
Bernstein, R.K., Fine, E.J., Westman, E.C. Accurso,
A., Frassetto, L., Gower, B.A., McFarlane, S.I., et al.
(2015). Dietary carbohydrate restriction as the first
approach in diabetes management: critical review
and evidence base. Nutrition 31, 1–13.
Feudtner, C. (2003). Bittersweet:  Diabetes, Insulin,
and the Transformation of Illness. (Chapel
Hill: University of North Carolina Press).
Foster, G.D., Wyatt, H.R., Hill, J.O., McGuckin, B.G.,
Brill, C., Mohammed, B.S., Szapary, P.O., Rader,
D.J., Edman, J.S., and Klein, S. (2003). A random-
ized trial of a low-carbohydrate diet for obesity. N
Engl J Med 348, 2082–2090.
Foster, G.D., Wyatt H.R., Hill, J.O., Makris, A.P.,
Rosenbaum, D.L., Brill, C., Stein, R.I., Mohammed,
B.S., Miller, B., Rader, D.J., et  al. (2010). Weight
and metabolic outcomes after 2  years on a low-
carbohydrate versus low fat diet. Ann Intern Med
153, 147–157.
Franz, M.J., Bantle, J.P., Beebe, C.A., Brunzell, J.D.,
Chiasson, J.L., Garg, A., Holzmeister, L.A.,
Hoogwerf, B., Mayer-Davis, E., Mooradian, A.D.,
et  al. (2002). Evidence-based nutrition principles
and recommendations for the treatment and pre-
vention of diabetes and related complications.
Diabetes Care 25, 148–198.
Gannon, M.C., and Nuttall, F.Q. (2004). Effect of a
high-protein, low-carbohydrate diet on blood
glucose control in people with type 2 diabetes.
Diabetes 53, 2375–2382.
Gannon, M.C., Nuttall, F.Q., Saeed, A., Jordan, K.,
and Hoover, H. (2003). An increase in dietary
protein improves the blood glucose response in
persons with type 2 diabetes. Am J Clin Nutr 78,
734–741.
Garg, A., Bonanome, A., Grundy, S.M., Zhang, Z.J.,
and Unger, R.H. (1988). Comparison of a high-
carbohydrate diet with a high-monounsaturated-
fat diet in patients with non-insulin-dependent
diabetes mellitus. N Engl J Med 319, 829–834.
Garg, A., Bantle, J.P., Henry, R.R., Coulston, A.M.,
Griver, K.A., Raatz, S.K., Brinkley, L., Chen, Y.D.,
Grundy, S.M., and Huet, B.A. (1994). Effects of
varying carbohydrate content of diet in patients
with non-insulin-dependent diabetes mellitus.
JAMA 271, 1421–1428.
Gutierrez, M., Akhavan, M., Jovanovic, L., and
Peterson, C.M. (1998). Utility of a short-term 25%
carbohydrate diet on improving glycemic control
in type 2 diabetes mellitus. J Am Coll Nutr 17,
595–600.
Hansen, R.A., Farley, J.F., Droege, M., and Maciejewski,
M.L. (2010). A Retrospective cohort study of eco-
nomic outcomes and adherence to monotherapy
with metformin, pioglitazone, or a sulfonylurea
among patients with Type 2 Diabetes Mellitus in
the United States From 2003 to 2005. Clin Ther 32,
1308–1319.
Heilbronn, L.K., Noakes, M., and Clifton, P.M. (1999).
Effect of energy restriction, weight loss, and diet
composition on plasma lipids and glucose in
373
Chapter 37: Ketogenic Diets as Highly Effective Treatments
patients with type 2 diabetes. Diabetes Care 22,
889–895.
Heilbronn, L.K., Noakes, M., and Clifton, P.M. (2002).
The effect of a high- and low-glycemic index
energy restricted diets on plasma lipid and glucose
profiles in type 2 diabetic subjects with varying
glycemic control. J Am Coll Nutr 21, 120–127.
Hession, M., Rolland, C., Kulkarni, U., Wise, A., and
Broom, J. (2009). A systematic review of random-
ized controlled trials of low-carbohydrate vs. low-
fat/low-calorie diets in the management of obesity
and its comorbidities. Obes Rev, 10, 36–50.
Hickey, J.T., Hickey, L., Yancy Jr., W.S., Hepburn, J.,
and Westman, E.C. (2003). Clinical use of a car-
bohydrate-restricted diet to treat the dyslipidemia
of the metabolic syndrome. Metab Syndr Relat
Disord 1, 227–232.
Holt, S.H., Brand-Miller, J.C., and Petocz, P. (1997). An
insulin index of foods: the insulin demand gener-
ated by 1000-kJ portions of common foods. Am J
Clin Nutr 66, 1264–1276.
Hu, T., Mills, K.T., Demanelis, K., Eloustaz, M., Yancy
Jr., W.S., Kelly, T.N., He, J., and Bazzano, L.A.
(2012). Effects of low-carbohdyrate diets versus
low-fat diets on metabolic risk factors: a meta-
analysis of randomized controlled clinical trials.
Am J Epidemiol 176, S44–S54.
Huang, M.F., Courtney, M., Edwards, H., and
McDowell, J. (2010). Factors that affect health
outcomes in adults with type 2 diabetes:  a cross-
sectional study. Int J Nurs Stud 47, 542–549.
Jenkins, D.J., Wolever, T.M., Taylor, R.H., Barker,
H., Fielden, H., Baldwin, J.M., Bowling, A.C.,
Newman, H.C., Jenkins, A.L., and Goff, D.V.
(1981). Glycemic index of foods:  a physiological
basis for carbohydrate exchange. Am J Clin Nutr
34, 362–366.
Johnston, B.C., Kanters, S., Bandayrel, K., Naji,
F., Siemieniuk, R.A., Ball, G.D.C., Busse, J.W.,
Thorlund, K., Guyatt, G., Jansen, J.P., and Mills, E.J.
(2014). Comparison of weight loss among named
diet programs in overweight and obese adults. A
meta-analysis. JAMA 312:923-933.
Joslin, E.P. (1928). Ideals in the treatment of diabetes
and methods for their realization. N Engl J Med
198, 379–382.
Joslin, E.P. (1919). A Diabetic Manual for the Mutual
Use of Doctor and Patient. (Philadelphia:  Lea &
Febiger).
Khazrai, Y.M., Defeudis, G., and Pozzilli, P. (2014).
Effect of diet on type 2 diabetes mellitus: a review.
Diabetes Metab Res Rev 30 (Suppl 1), 24–33.
Khan, M.A., St Peter, J.V., Breen, G.A., Hartley, G.G.,
and Vessey, J.T. (2000). Diabetes disease stage pre-
dicts weight loss outcomes with long-term appetite
suppressants. Obesity Res 8, 43–48.
Kirk, J.K., Graves, D.E., Craven, T.E., Lipkin, E.W.,
Austin, M., and Margolis, K.L. (2008). Restricted-
carbohydrate diets in patients with type 2
373
diabetes:  a meta-analysis. J Am Diet Assoc 108,
91–100.
Krieger, J.W., Sitren, H.S., Daniels, M.J., and
Langkamp-Henken, B. (2006). Effects of varia-
tion in protein and carbohydrate intake on body
mass and composition during energy restriction: a
meta-regression. Am J Clin Nutr 83, 260–274.
Leahy, J.L. (2005). Pathogenesis of type 2 diabetes mel-
litus. Arch Med Res 36, 197–209.
Lim, E.L., Hollingsworth, K.G., Aribisala, B.S., Chen,
M.J., Mathers, J.C. and Taylor, R. (2011). Reversal
of type 2 diabetes: normalisation of beta cell func-
tion in association with decreased pancreas and
liver triacylglycerol. Diabetologia, 54, 2506–2514.
Ludwig, D.S. (2002). The glycemic index: physiologi-
cal mechanisms relating to obesity, diabetes, and
cardiovascular disease. JAMA 287, 2414–2423.
Mayer, S.B., Jeffreys, A.S., Olsen, M.K., McDuffie, J.R.,
Feinglos, M.N., and Yancy Jr., W.S., (2014). Two
diets with different haemoglobin A1c and antigly-
caemic medication effects despite similar weight
loss in type 2 diabetes. Diabetes Obes Metab 16,
90–93.
Miles, J.M., Leiter, J., Hollander, P., Wadden, T.,
Anderson, J.W., Doyle, M., Foreyt, J., Aronne, L.,
and Klein S. (2002). Effect of orlistat in overweight
and obese patients with type 2 diabetes treated
with metformin. Diabetes Care 25, 1123–1128.
Morgan, W. (1877). Diabetes Mellitus:  Its History,
Chemistry, Anatomy, Pathology, Physiology, and
Treatment. (London:  Homeopathic Publishing
Company).
Nielsen, J.V., and Joensson, E.A. (2008). Low-
carbohydrate diet in type 2 diabetes:  stable
improvement of bodyweight and glycaemic con-
trol during 44  months follow-up. Nutr Metab
(Lond) 5, 14.
Noakes, M., Foster, P.R., Keogh, J.B., James, A.P.,
Mamo, J.C., and Clifton, P.M. (2006). Comparison
of isocaloric very low carbohydrate/high saturated
fat and high carbohydrate/low saturated fat diets
on body composition and cardiovascular risk.
Nutr Metab (Lond). 3, 7.
Nordmann, A.J., Nordmann, A., Briel, M., Keller,
U., Yancy Jr., W.S., Brehm, B.J., and Bucher, H.C.
(2006). Effects of low-carbohydrate vs low-fat diets
on weight loss and cardiovascular risk factors.
Arch Intern Med 166, 285–293.
Nuttall, F.Q., Almokayyad, R.M., and Gannon, M.C.
(2015). Comparison of a carbohydrate-free diet vs.
fasting on plasma glucose, insulin and glucagon in
type 2 diabetes. Metabolism 64, 253–262.
Nyenwe, E.A. (2011). Management of type 2 diabe-
tes: evolving strategies for the treatment of patients
with type 2 diabetes. Metabolism 60, 1–23.
O’Neill, D.F., Westman, E.C., and Bernstein, R.K.
(2003). The effects of a low-carbohydrate regimen
on glycemic control and serum lipids in diabetes
mellitus. Metab Syndr Relat Disord 1, 291–298.
374
374 section IV: Ketone-Based Metabolism
Rabinowitch, I.M. (1930). Experiences with a high
carbohydrate-low calorie diet for the treatment of
diabetes mellitus. CMAJ 23, 489–498.
Richardson, R. (1929). High-carbohydrate diets in dia-
betes mellitus. Am J Med Sci 177, 426–430.
Rizkalla, S.W., Taghrid, L., Laromiguiere, M., Huet, D.,
Boillot, J., Rogoir, A., Elgrably, F., and Slama, G.
(2004). Improved plasma glucose control, whole-
body glucose utilization, and lipid profile on a
low-glycemic index diet in type 2 diabetic men: a
randomized controlled trial. Diabetes Care 27,
1866–1872.
Samaha, F.F., Iqbal, N., Seshadri, P., Chicano, K.L.,
Daily, D.A., McGrory, J., Williams, T., Williams,
M., Gracely, E.J., and Stern, L. (2003). A low-
carbohydrate as compared with a low-fat diet in
severe obesity. N Engl J Med 348, 2074–2081.
Sansum, W.D., Blatherwick, N.R., and Bowden, R.
(1928). The use of high carbohydrate diets in
the treatment of diabetes mellitus. JAMA 86,
178–181.
Santos, F.L., Esteves, S.S., da Costa Pereira, A., Yancy,
W.S.Jr., and Nunes, J.P.L. (2012). Systematic review
and meta-analysis of clinical trials of the effects of
low carbohydrate diets on cardiovascular risk fac-
tors. Obes Rev 13(11), 1048-1066.
Saslow, L.R., Kim, S., Daubenmier, J.J., Moskowitz,
J.T., Phinney, S.D., Goldman, V., Murphy, E.J.,
Cox, R.M., Moran, P., and Hecht, F.M. (2014).
A randomized pilot trial of a moderate carbohy-
drate diet compared to a very low carbohydrate
diet in overweight or obese individuals with type
2 diabetes mellitus or prediabetes. PLoS One 9,
e91027.
Shai, I., Schwarzfuchs, D., Henkin, Y., Shahar, D.R.,
Witkow, S., Greenberg, I., Golan, R., Fraser, D.,
Bolotin, A., Vardi, H., et  al. (2008). Weight loss
with a low-carbohydrate, Mediterranean, or low-
fat diet. N Engl J Med 359, 229–241.
Shai, I., Spence, D., Schwarzfuchs, D., Henkin, Y.,
Parraga, G., Rudich, A., Fenster, A., Mallett, C.,
Liel-Cohen, N., Tirosh, A., et  al. (2010). Dietary
intervention to reverse carotid atherosclerosis.
Circulation, 121, 1200–1208.
Sondike, S.B., Copperman, N., and Jacobson, M.S.
(2003). Effects of a low-carbohdyrate diet on
weight loss and cardiovascular risk factor in over-
weight adolescents. J Pediatr 142, 253–258.
Stein, C.J., and Colditz, G.A. (2004). The epidemic of
obesity. J Clin Endocrinol Metab 89, 2522–2525.
Stone, D.B., and Connor, W.E. (1963). The prolonged
effects of a low cholesterol, high carbohydrate
diet upon the serum lipids in diabetic patients.
Diabetes 12, 127–132.
Stumvoll, M., Goldstein, B.J., and Van Heaften T.W.
(2005). Type 2 diabetes: principles of pathogenesis
and therapy. Lancet 365, 1333–1346.
Tahrani, A.A., Piya, M.K., Kennedy, A., and Barnett,
A.H. (2010). Glycaemic control in type 2 diabe-
tes:  targets and new therapies. Pharmacol Ther
125, 328–361.
Tay, J., Luscombe-Marsh, N.D., Thompson, C.H.,
Noakes, M., Buckley, J.D., Wittert, G., A., Yancy,
W.S., and Brinkworth, G.D. (2014). A very low-
carbohydrate, low-saturated fat diet for type 2 dia-
betes management:  a randomized trial. Diabetes
Care 37, 2909–2918.
The Look Ahead Research Group. (2003). Look
AHEAD (Action for Health in Diabetes):  design
and methods for a clinical trial of weight loss for
the prevention of cardiovascular disease in type 2
diabetes. Control Clin Trials 24, 610–628.
UKPDS. United Kingdom Prospective Diabetes Study
33. (1998). Intensive blood-glucose control with
sulphonylurea or insulin compared with con-
ventional treatment and risk of complications in
patients with type 2 diabetes. Lancet 352, 837–853.
UKPDS. United Kingdom Prospective Diabetes Study
34. (1998). Effect of intensive blood-glucose control
with metformin on complications in overweight
patients with type II diabetes. Lancet 352, 854–865.
Vernon, M.C., Mavropoulos, J., Transue, M., Yancy Jr.,
W.S., and Westman, E.C. (2003). Clinical experi-
ence of a carbohydrate-restricted diet: effect on
diabetes mellitus. Metab Syndr Relat Disord 1,
233–237.
Vernon, M.C., Kueser, B., Transue, M., Yates, H.E.,
Yancy Jr., W.S., and Westman, E.C. (2004). Clinical
experience of a carbohydrate-restricted diet for the
metabolic syndrome. Metab Syndr Relat Disord 2,
180–186.
Volek, J.S., and Feinman, R.D. (2005). Carbohydrate
restriction improves the features of metabolic syn-
drome: metabolic syndrome may be defined by the
response to carbohydrate restriction. Nutr Metab
(Lond) 2, 31.
Westman, E.C., Yancy, W.S., Jr., and Humphreys, M.
(2006). Dietary treatment of diabetes mellitus in
the pre-insulin era (1914–1922). Perspect Biol
Med 49, 77–83.
Yamada, Y., Uchida, J., Izumi, H., Tsukamoto, Y., Inoue,
G., Watanabe, Y., Irie, J., and Yamada, S. (2014).
A non-calorie-restricted low-carbohydrate diet is
effective as an alternative therapy for patients with
type 2 diabetes. Intern Med 53, 13–19.
Yancy Jr., W.S., Vernon, M.C., and Westman, E.C.
(2003). A pilot trial of a low-carbohydrate keto-
genic diet in patients with Type 2 diabetes. Metab
Syndr Rel Dis 1, 239–244.
Yancy Jr., W.S., Olsen, M.K., Guyton, J.R., Bakst, R.P.,
and Westman, E.C. (2004). A low-carbohydrate,
ketogenic diet versus a low-fat diet to treat obe-
sity and hyperlipidemia: a randomized, controlled
trial. Ann Intern Med 140, 769–777.
375

Chapter 37: Ketogenic Diets as Highly Effective Treatments 375

Yancy Jr., W.S., Foy, M., Vernon, M.C., and Westman,
E.C. (2005). A low-carbohydrate ketogenic diet
to treat Type 2 Diabetes. Nutr Metab (London)
2, 34.
Yancy Jr., W.S., Westman, E.C., McDuffie, J.R.,
Grambow, S.C., Jeffreys, A.S., Bolton, J., Chalecki, A.,
and Oddone, E.Z. (2010). A randomized trial of
a low-carbohydrate diet vs orlistat plus a low-
fat diet for weight loss. Arch Intern Med 170,
136–145.
Yudkin, J. (1959). The causes and cure of obesity.
Lancet 274, 1135–1138.
376
38
Keto-Adaptation in Health and Fitness
PA R K E R H Y D E , C S C S , C I S S N , V I N C E N T J. M I L L E R , M S ,
AND JEFF S. VOLEK, PHD, RD
INTRODUCTION
Not completely dissimilar from the gravitational
potential energy of a ball sitting atop a tower, our
bodies have an abundance of metabolic potential
energy (MPE). This MPE can be found in two
primary storage forms, carbohydrate and fats.
Carbohydrates stored predominantly as glycogen
in skeletal muscle and liver represent approxi-
mately 2,000 kcals of energy, whereas fat stores in
adipose tissue are at least an order of magnitude
greater. Thus, from an evolutionary standpoint,
dietary carbohydrate more aptly represents a
single matchbook of MPE, better suited for tran-
sient fight-or-flight responses and not the pri-
mary fuel source. Since storage is limited, when
carbohydrate is consumed it must be oxidized or
converted to something else. The glucose-insulin
axis supports this hierarchy of fuel use, as acute
elevations in insulin following consumption of
carbohydrate promptly inhibit both adipose tissue
lipolysis and systemic fat oxidation (Bonadonna
et al., 1990; Jensen et al., 1989). Thus, when energy
expenditure is high there is a constant need to sup-
ply a steady source of carbohydrate-based fuel. As
exemplified by the metaphor “hitting the wall,”
when carbohydrate stores are near depletion dur-
ing exercise, the body cannot simply transition to
using fat efficiently, unless, the person has under-
gone a very low-carbohydrate diet for several
weeks, which results in a complex set of coordi-
nated adaptations that dramatically accelerate the
body’s ability to access and use lipid-based fuels.
The rest of this chapter discusses the concept that
humans function more efficiently when they are
utilizing predominantly fat for fuel.
Modern diets emphasize carbohydrate-dense
foods, and therefore bias metabolism toward
carbohydrate oxidation while impairing fat oxi-
dation. Whereas some individuals can maintain
weight and metabolic health consuming low-fat/
high-carbohydrate diets, the majority cannot, as
evidenced by the fact that half the adults in the
United States are prediabetic or diabetic (Menke
et al., 2015).
The parallel increase in dietary carbohydrate
and the sharp rise in obesity and diabetes in the
United States is prima facie evidence that carbo-
hydrate intake has an important effect on health
status. The mechanistic links between dietary
carbohydrate, oversecretion of insulin, and exac-
erbation of insulin resistance make the over-
consumption of sugars and starches the most
obvious driver of the obesity and diabetes epidem-
ics. Intuitively, decreasing carbohydrate is an obvi-
ous solution. To what extent should carbohydrate
be decreased? The answer is not simple and varies
from person to person and even within an indi-
vidual over time. For many with insulin resistance
(i.e., carbohydrate intolerance), a modest reduc-
tion in dietary carbohydrate may be all that is
necessary to maintain metabolic health. For others
with a higher degree of insulin resistance, effective
metabolic correction requires keeping carbohy-
drates under 40–50 g/day for an extended period
of time, resulting in keto-adaptation.
Keto-adaptation is a process that sustains
optimum fuel flow to all organs through use of
metabolic pathways we have acquired over 2 mil-
lion years as hunters/gatherers/herders. The most
important aspect of keto-adaptation is the ability
of the brain to displace glucose for ketones as the
primary fuel source. Pioneering work in the 1960s
and 1970s examining starvation ketosis showed
that the brain will switch to preferential use of
ketone oxidation (~60%) for derivation of energy
(Cahill et al., 1970; Cahill and Owen, 1968; Owen
et al., 1967). This preferential use of ketone bodies
as a primary fuel source provides both an energy-
and neuroprotective effect (Cahill et  al., 1970).
Similar effects are observed in non-starvation-
induced ketosis, when carbohydrate is restricted
and dietary protein and fat are present in the diet.
377

Chapter 38: Keto-Adaptation in Health and Fitness 377

Glucose flux

Fat Oxidation Capacity

Insulin Chronic Disease

Body Composition
Triglycerides Longevity

Oxidative Stress

Fat Utilization Quality of Life

Inflammation

Ketones Physical Resilience

Membrane Integrity

Satiety

Days Weeks/Months Long-term

Time

FIGURE 38.1 Theoretical accumulation of benefit during sustained nutritional ketosis leading to keto-adaptation.

The metabolic pathways responsible for synthesis,
transport, uptake, and oxidation of ketones are all
up-regulated. Within days, blood levels of ketones
increase from less than 0.1  mmol/L to 0.5 up to
3–5  mmol/L. This range is referred to as nutri-
tional ketosis, and is associated with a positive
effect on a broad range of cardiometabolic risk
factors. It is noteworthy that ketoacidosis does not
occur until ketones are above 10 mmol/L, a level
only reached in individuals with insulin insuf-
ficiency (i.e., poorly controlled type-1 diabetics).
Even in a starvation-induced ketogenic state with
extreme exercise, an individual with a functioning
pancreas will not experience the increased acidity
associated with ketoacidosis.
The near-exclusive reliance on lipid-based fuel
is a fundamental characteristic of keto-adaptation,
but a new perspective of nutritional ketosis
has emerged following the identification of β-
hydroxybutyrate (βOHB) as a potent epigenetic and
signaling molecule (Newman and Verdin, 2014a).
We now appreciate that keto-adaptation involves
cellular adaptations that can phenotypically mani-
fest in different ways between people, probably
because of genetic and epigenetic effects. The dura-
tion of this chapter focuses on mechanisms associ-
ated with keto-adaptation as it pertains to health
outcomes, and explores the emerging potential
of well-formulated ketogenic diets to extend the
physical and cognitive performance of athletes
and military personnel. Although many studies
have documented acute and short-term responses
of ketogenic diets, the precise temporal pattern of
responses to nutritional ketosis ultimately leading
to the keto-adapted phenotype requires additional
work (Figure 38.1).
K E T O - A D A P TAT I O N A N D
H E A LT H I M P L I C AT I O N S
Ketogenic diets have been shown to improve
numerous health outcomes, especially those asso-
ciated with insulin resistance, the underlying com-
mon feature in metabolic syndrome (MetS), type
2 diabetes, cardiovascular disease (CVD), obesity,
polycystic ovarian syndrome (PCOS), and many
other chronic health conditions.
Ketogenic Diets Improve Insulin
Resistance, Metabolic Syndrome,
and Type 2 Diabetes
A primary application of ketogenic diets in adults
is the management of insulin-resistant conditions.
Insulin resistance is the primary feature underlying
type 2 diabetes, but it also exists across a contin-
uum in the general population. It is defined by an
impaired functioning of the insulin response at the
cellular level, which given the pleiotropic effects of
insulin can manifest in many different signs and
symptoms. Most commonly, insulin resistance
is viewed from the perspective of impaired skel-
etal muscle glucose uptake and increased hepatic
glucose output, manifesting in hyperglycemia.
378
378 section IV: Ketone-Based Metabolism
Thus, individuals with insulin resistance have a
fundamental problem metabolizing dietary car-
bohydrate and maintaining a normal glucose level.
Fasting blood glucose values greater than 100 mg/
dL are indicative of impaired glucose tolerance,
and values greater than 7.0 mmol/L (126 mg/dL)
on two occasions indicate type 2 diabetes. Since
type 2 diabetes is discussed in a separate chapter,
we focus our discussion on the impact of ketogenic
diets on milder forms of insulin resistance that
manifest in the cluster of disturbances that encom-
pass metabolic syndrome.
Metabolic syndrome is best described as pre-
diabetes and provides an early sign that the body
is mismanaging dietary carbohydrate. One aspect
of this mismanagement is that a greater proportion
of incoming carbohydrate is converted to fat in the
liver (Petersen et  al., 2007). This greater conver-
sion of dietary carbohydrate into fat (i.e., de novo
lipogenesis), much of it entering the circulation
as saturated fat, is a metabolic abnormality that
causes a significant amount of collateral damage in
the body. For all these reasons, insulin resistance is
a problem that effectively manifests itself as carbo-
hydrate intolerance. Like other food intolerances,
the most logical and effective approach to man-
aging carbohydrate intolerance is to restrict the
offending nutrient. When dietary carbohydrate is
restricted to a level below which it is not signifi-
cantly converted to fat (a threshold that varies from
person to person), signs and symptoms of insulin
resistance improve or often disappear completely.
Based on our research, a ketogenic diet can resolve
all the signs and symptoms of metabolic syndrome
(Volek and Feinman, 2005; Volek et al., 2008).
The main features of metabolic syndrome are
abdominal obesity; high blood levels of glucose,
insulin, and triglycerides; low HDL-cholesterol;
and high blood pressure. Other features not typi-
cally included in the diagnosis of metabolic syn-
drome include elevated constitutive inflammation,
vascular dysfunction, and disturbances in fatty acid
composition. It is commonly stated that obesity is
the cause of metabolic syndrome, but this perspec-
tive fails to acknowledge the cause of obesity, which
for many involves the overconsumption of dietary
carbohydrate relative to a person’s tolerance. Thus,
the primary driver of metabolic syndrome is over-
consumption of sugars and starches relative to the
level a person can metabolize without resorting to
increased de novo lipogenesis and other manifes-
tations of carbohydrate intolerance (e.g., increased
oxidative stress). Although the molecular details of
insulin resistance have not been fully elucidated,
a well-formulated ketogenic diet results in broad
spectrum improvements across a range of risk fac-
tors relative to a traditional low-fat diet in patients
with metabolic syndrome, in effect putting the dis-
order into remission (Volek et al., 2009b). A large
number of studies have shown that the features of
metabolic syndrome improve, often in dramatic
fashion, with adequate carbohydrate restriction
(Volek and Feinman, 2005; Volek et al., 2008).
Insulin resistance is one of the primary predic-
tors of nonalcoholic fatty liver disease (NAFLD),
which is characterized by increased liver fat,
inflammation, and resultant hypertriglyceride-
mia, all of which increase CVD risk. Although
the pathophysiology and mechanisms of NAFLD
are yet to be fully elucidated, there is a clear link
to increased carbohydrate consumption, de novo
lipogenesis, and increased adiposity. Insulin resis-
tance in the peripheral tissues inhibits the cellu-
lar uptake of glucose, thus a shunting of dietary
carbohydrate toward the liver occurs. Glucose
flux into hepatic tissue is not insulin dependent,
thus an increased rate of glycogen synthesis or de
novo lipogenesis, during the postprandial state, is
experienced. Hepatic insulin resistance is associ-
ated with increased concentration of malonyl-
CoA; inhibition of carnitine palmitoyl transferase
1 (CPT-1), a transporter for fatty acids into the
mitochondria; inhibition of beta-oxidation; and
increased accumulation of triglycerides, cho-
lesterol esters, and very low density lipoprotein
(Browning et  al., 2011; Zivkovic et  al., 2007).
Since keto-adaptation results in significantly
decreased de novo lipogenesis and accelerated
beta-oxidation, as well as decreased inflammation
and serum triglycerides, it would seem obvious to
also have beneficial effects on NAFLD. However,
only a few studies have examined effects on liver
fat. Early during keto-adaptation, approximately
2 weeks, there is significantly decreased hepatic
triglyceride content, as compared to a higher car-
bohydrate diet (Browning et al., 2011). Six months
of a ketogenic diet has been demonstrated to
improve steatosis, necroinflammatory grade, and
centrilobular fibrosis, all of which are histological
markers of NAFLD (Tendler et al., 2007).
Weight Loss and Body
Composition Responses
Dozens of studies have examined weight-loss
responses to low-carbohydrate and low-fat diets
over the last 15  years. Although there is great
variability in the comparison of diets of any dura-
tion, it is noteworthy that low-carbohydrate diets
do at least as well and usually better than low-fat
diets according to recent meta-analyses (Ajala
379
Chapter 38: Keto-Adaptation in Health and Fitness
et  al., 2013; Bueno et  al., 2013). In fact a recent
meta-analysis, which used Bayesian hierarchi-
cal modeling to provide an estimate of the likeli-
hood of achieving a desired degree of weight loss,
reported that the probability of greater weight loss
associated with a low-carbohydrate diet was >99%
(Sackner-Bernstein et  al., 2015). It is noteworthy
that the degree of insulin resistance has an effect
on expected weight loss. In a post hoc analysis of
the A to Z study, it was reported that individuals
who were insulin sensitive had similar degrees of
weight loss on a low- and high-carbohydrate diet
after 1  year, but the those with higher levels of
insulin resistance lost significantly more weight on
a low-carbohydrate diet (McClain et al., 2013).
Several weight-loss studies have also mea-
sured changes in fat and lean body mass. Similar
to changes in total body mass, low-carbohydrate
diets have been shown to result in a greater loss of
fat mass (Krieger et al., 2006). Outcomes for lean
mass, however, are less clear due to variations in
diet and limitations in diet formulation and assess-
ment methods. Over 30  years ago Phinney et  al.
(1983) reported that nitrogen balance was positive
after the first week of a eucaloric ketogenic diet.
An important factor not always appreciated in
studies of ketogenic diets is the need to adjust elec-
trolyte intake to offset the natriuretic effect of car-
bohydrate restriction. Unfortunately, the majority
of studies conducted since then have not consid-
ered this natriuretic effect. However, our research
has shown it is possible not only to preserve lean
mass during weight loss with a ketogenic diet but
also to increase it (Volek et al., 2002). An impor-
tant consideration is the water retention associ-
ated with high-carbohydrate diets and glycogen
storage (Olsson and Saltin, 1970), and hence the
water loss that accompanies glycogen loss during
carbohydrate restriction. Readings from dual-
energy X-ray absorptiometry DXA (Going et  al.,
1993) and biological impedance (Koulmann et al.,
2000), which are two commonly used methods of
assessing body composition, fluctuate in accor-
dance with body water content. Furthermore,
these methods generally include total body water
as a component of lean mass. Therefore, the water
loss that typically occurs during the initiation of
carbohydrate restriction can give a false indication
of functional muscle loss.
As previously described, nutritional ketosis
is characterized by an elevation in blood ketones
and, depending on the extent of carbohydrate
restriction, is not always synonymous with a low-
carbohydrate diet. Infusion of βOHB, the primary
ketone elevated during nutrition ketosis, reduces
379
nitrogen excretion (Sherwin et al., 1975) and leu-
cine oxidation (Nair et  al., 1988)  and increases
muscle protein synthesis (Nair et  al., 1988). As
such, nutritional ketosis may be an important
determinant of lean mass preservation during pro-
longed periods of energy deficit induced by caloric
restriction and/or high energy expenditure.
Preservation of lean mass can further be sup-
ported through resistance training. In fact, our
research has shown 12 weeks of resistance training
to result in increases in lean body mass, along with
greater fat loss, during a ketogenic diet versus a
low-fat diet (Quann et al., 2007; Volek et al., 2010).
Similar results have been observed in response to a
ketogenic diet versus a typical Western diet (Rauch
et al., 2014).
While the metabolic adaptations that occur
during carbohydrate restriction clearly have an
important role in weight loss, satiety is an impor-
tant factor as well. In weight-loss trials comparing
low-carbohydrate to low-fat diets, energy deficit
for the low-carbohydrate diet is often achieved
without intentional restriction, indicating greater
satiety. Furthermore, very low calorie diets, which
are often ketogenic due to the inherently low
level of carbohydrates and protein, are known
to reduce appetite. This is supported by a recent
meta-analysis in which both ketogenic and very-
low calorie diets were considered (Gibson et  al.,
2015). The similarities in satiety between these two
diets, despite the considerable differences in pro-
tein, fat, and energy content, suggests that ketosis
may suppress appetite. This is supported by stud-
ies showing a eucaloric or ad libitum ketogenic
diet to be more satiating than nonketogenic diets
with higher protein content (Johnstone et al., 2008;
Veldhorst et al., 2010).
Improved Lipoprotein Profile
Although higher probability of weight loss is a
major benefit of a ketogenic diet, there are also pro-
found changes in lipoprotein metabolism (Volek
et  al., 2008). The reduced conversion of carbohy-
drate to fat in the liver (lipogenesis) plus the low
insulin state enabling accelerated fat oxidation are
major factors contributing to the improvements
in processing of lipoproteins commonly observed.
The most consistent response is a sharp decrease
in plasma triglycerides, most dramatically in those
with preexisting hypertriglyceridemia. A ketogenic
diet also demonstrates a striking reduction in the
postprandial lipemic response to a high-fat meal
(Volek et  al., 2000; Volek et  al., 2009b) and has
positive effects on postabsorptive and postpran-
dial vascular function (Volek et al., 2009a). Almost
380
380 section IV: Ketone-Based Metabolism
as consistent as their triglyceride-lowering effects,
ketogenic diets increase HDL-C. The increase in
HDL-C may not occur as quickly as the decrease
in triglycerides, but there is some evidence that it
develops more slowly over time (Shai et al., 2008).
The response in LDL-C concentration to a keto-
genic diet is highly variable, with many people
showing a decrease and many an increase, with a
subset (~15%) showing a marked rise. While there
is variability in the LDL-C concentration response,
there are more uniform patterns in qualitative fea-
tures of the LDL particle. Dietary carbohydrate
restriction consistently decreases small, dense LDL
particles, which are more atherogenic due to their
longer residence time in the circulation and greater
propensity for oxidation, whereas low-fat/high-
carbohydrate diets have the opposite effect. This
inverse relationship between dietary carbohydrate
content and LDL particle size is evident over a wide
range of carbohydrate intakes (Krauss, 2005), and
it can have quite dramatic (i.e., positive) effects in
response to a ketogenic diet (Volek et al., 2009b).
K E T O - A D A P TAT I O N
IN SPORT AND FITNES S
Beyond the clinical applications of keto-adaptation
for promoting weight loss and managing insulin
resistant conditions, there is a growing interest in
ketogenic diets by athletes and active individu-
als interested in better health, performance, and
recovery.
Brief History
More than 70 years ago (Christensen and Hansen,
1939) described that consumption of a high-car-
bohydrate diet, compared to a high-fat diet, for 1
week prior to a submaximal event was associated
with enhanced performance. Nearly 3 decades
later, Bergstrom and Hultman (1966, 1967) dis-
covered that muscle glycogen depletion was asso-
ciated with fatigue, and that a high-carbohydrate
diet maintained muscle glycogen and perfor-
mance. This set the stage for the next 40+ years,
during which time the supremacy of carbohydrate
has become deeply embedded in the minds of most
scientists, coaches, and athletes (Hargreaves et al.
2004). There has been a strong confirmation bias
to support high-carbohydrate diets, reinforced by
a burgeoning sports beverage industry profiting
from this paradigm.
It needs to be acknowledged that a lot of great
work has been done by many researchers detail-
ing how carbohydrate affects exercise metab-
olism and performance, but there has been a
surprising lack of recognition, or willful neglect,
of the alternative paradigm that deemphasizes
carbohydrate. For example, there are few placebo-
controlled randomized control trials, few studies
of very low carbohydrate intake, and even fewer
studies that allowed adequate time to adapt to a
low-carbohydrate diet. Carbohydrate loading has
not shown improved performance in all studies,
despite significantly increased muscle glycogen
levels (Burke et  al., 2000). The most obvious fact
that has been downplayed in prior research has
been the profound effect of carbohydrate intake
on inhibition of fat utilization, and the perspective
that fat is a preferred fuel.
Fat as a Premium Fuel
Conventional wisdom has been that carbohydrate
is the preferred source of fuel for exercise, but this
perspective is rapidly changing considering the
many benefits associated with emphasizing fatty
acids and ketones as primary fuels, and relegat-
ing blood sugar and glycogen to secondary status.
Ten hypothetical reasons a high capacity to utilize
lipid-based fuels might be beneficial for athletes
are listed here:
1. Fat is stored in adipose tissue in greater
quantities than carbohydrates by at least an
order of magnitude (>20,000 kcal in even the
thinnest athlete vs. 2,000 kcal as glycogen).
A key element contributing to deteriorating
cognitive and physical performance during
physically demanding exercise is reduced
carbohydrate availability coupled with an
inability to effectively utilize the thousands
of calories stored as fat. Paradoxically,
deteriorating performance associated with
glycogen depletion occurs in the presence of
an abundance of fuel stored as body fat that
the athlete cannot access efficiently, unless
they have previously keto-adapted for
several weeks.
2. Fat is a more efficient source of fuel than
carbohydrates. A molecule of palmitic
acid has over twice as many carbons as a
molecule of glucose, and over twice the
energy per unit weight (i.e., 9 vs. 4 kcal/g).
Fat is also more efficient, since it is not
stored along with a lot of extra water
weight like glycogen. Finally, ketones have
been shown to have greater efficiency
in providing cellular energy and work
output (i.e., more work per unit oxygen
consumed) (Sato et al., 1995).
3. Relying on body fat for energy significantly
decreases the need to fuel during exercise.
381
Chapter 38: Keto-Adaptation in Health and Fitness
This has physiologic benefits as it avoids
the need to digest and absorb nutrients at
a time when the majority of blood flow is
diverted to active skeletal muscles. It also
alleviates the gastrointestinal problems
associated with fueling during exercise that
are common among endurance athletes.
4. When oxidized inside the mitochondria,
ketones generate less reactive oxygen
species relative to glucose or other fuels,
in essence making them a cleaner burning
fuel (Maalouf et al., 2007; Veech, 2004).
5. Deriving greater amounts of energy from
fat at higher exercise intensities translates
into less reliance on limited glycogen
stores, glycolysis, and generation of
hydrogen ions.
6. Making fat the primary fuel often results
in greater ease of fat loss and improved
body composition, which translates
into improved efficiency and power-to-
weight ratio. Hypothetically this would
be advantageous for endurance as well as
many strength/power athletes.
7. When ketone concentrations increase
to the 1–5 mmol/L range, the brain will
begin to preferentially oxidize ketones
(Cahill et al., 1970; Owen et al., 1967). In
a non-keto-adapted athlete, continuous
exercise often results in profound mental
confusion (i.e., “hitting the wall” or
“bonking”). A brain adapted to extracting
the majority of its energy requirements
from ketones is resistant to this energy
crisis, since ketones are derived from
fatty acids and actually increase in
concentration as exercise duration
increases. A keto-adapted brain is able to
tolerate extraordinarily low levels of blood
sugar without any signs or symptoms
hypoglycemia (Cahill and Aoki, 1980;
Drenick et al., 1972).
8. Independent of its role as a vital fuel
source for the brain and other organs/
tissues when carbohydrates are limited,
βOHB has recently been shown to
directly signal cellular processes, acting
more like a hormone (Newman and
Verdin, 2014a, 2014b). Specifically, it
was demonstrated that at blood levels
characteristic of nutritional ketosis, βOHB
acts as a class I histone deactylase inhibitor,
switching on specific genes that protect
cells from free radical damage (Shimazu
et al., 2013). Oxidative stress is a root
381
cause of aging and inflammation, and the
reduction by βOHB of these processes was
remarkable.
9. One of the most common perceived
benefits following keto-adaptation
is enhanced recovery from exercise.
This is frequently demonstrated by
an expedited return to training and
competition following physically
demanding events and long training
sessions. This area of research has
not been investigated thoroughly, but
we speculate that less oxidative stress
and the anti-inflammatory effects of
a ketogenic diet may contribute to the
perceived improvement in recovery by
many athletes. Carbohydrate ingestion
and acute hyperglycemia activate a
host of inflammatory and free radical-
generating pathways (Aljada et al.,
2006; Dhindsa et al., 2004). Some of
these include: stimulation of NAPDH
oxidase, superoxide generation by
leukocytes, TNF-α production, and
activation of NF-κB which regulates the
transcriptional activity of over 100 pro-
inflammatory genes. Consistent with the
notion that carbohydrate can aggravate
inflammatory balance, many studies
have reported that low carbohydrate
diets decrease markers of constitutive
inflammation in athletes (Phinney
et al., 1983) and non-athletes (Youm
et al., 2015).
10. A low carbohydrate intake enhances the
effects of exercise on several positive cel-
lular adaptations associated with health
(Hawley and Burke, 2010). For example,
in the post-exercise period consumption
of even small amounts of carbohydrate
rapidly decreases the release of fatty acids
from fat stores and oxidation of fat in the
muscle (Long et al., 2008). In some ath-
letes, a surge in insulin may be followed
by a low blood sugar eliciting a stress
response characterized by a counter-reg-
ulatory hormonal response that can man-
ifest as carbohydrate cravings, lethargy,
poor physical/mental performance and
suboptimal recovery. Carbohydrate con-
sumption during recovery has also been
shown to diminish the beneficial effects of
exercise on insulin sensitivity and other
cardio-metabolic risk markers (Holtz
et al., 2008; Stephens and Braun, 2008).
382
382 section IV: Ketone-Based Metabolism
T H E K E TO - A DA P T E D
AT H L E T E
Although a critical mass of experimental data on
ketogenic diets and ketone physiology has been
generated in the last decade, demonstrating safety
and therapeutic efficacy in managing a range of
clinical conditions, there has been little effort
focused on studying how ketogenic diets affect
performance and recovery. The two notable stud-
ies that shed light on the metabolic phenotype
of the keto-adapted athlete come from Phinney
in 1983 (Phinney et al., 1983) and Volek in 2016
(Volek et al., 2016).
Phinney first demonstrated in a small (n=5) but
highly controlled metabolic ward study that 4 weeks
of adaptation to a ketogenic diet in high-level cyclists
resulted in profound metabolic adaptations and res-
toration of endurance performance following an
initial decline. These cyclists showed a remarkable
near exclusive reliance on fat as the sole fuel at 64%
maximal oxygen consumption, while significantly
decreasing glycogen breakdown and glucose oxida-
tion. For more than 3 decades these findings have
been largely ignored but never refuted.
We designed the FASTER (Fat Adapted
Substrate Use in Trained Elite Runners) study
to validate and extend Phinney’s work by docu-
menting the keto-adapted phenotype in elite ath-
letes after long-term keto-adaptation (Volek et al.
2016). The primary goal was to characterize the
metabolic differences between high-caliber ultra-
endurance runners habitually consuming a very
low-carbohydrate ketogenic diet versus a more
traditional high-carbohydrate diet. Ten low-carb
and 10 high- carb high-caliber ultra-runners were
recruited and studied. They were carefully matched
on age, body mass, maximum aerobic capacity,
and competition times; the main difference was
their habitual diet. The high-carbohydrate group
consumed over half their intake as carbohydrate,
whereas the low-carbohydrate athletes consumed
~10% of their energy intake from carbohydrate
for an average of 20  months. Athletes performed
a maximal oxygen consumption test during which
peak fat oxidation was determined, and on a sepa-
rate day they ran on a treadmill at 64% of maximal
oxygen consumption for 3 hours.
The most remarkable finding was an extraor-
dinary ability to oxidize fat in low-carbohydrate
athletes, which was on average two-fold higher
(1.5 vs. 0.7 g/min, respectively). Even the low-
est “fat burner” in the low-carbohydrate group
exceeded the highest “fat burner” in the high-
carbohydrate group. These rates of fat oxidation
are even more remarkable considering that the
highest values reported in the literature (with the
notable exception of the cyclists in Phinney et al.,
1983) are 50% lower (i.e., ~1.0 g/min). Fuel use
during the 3-hour run was also dramatically dif-
ferent. The high-carbohydrate group used an even
mix of fuel, whereas nearly 90% of energy expend-
iture was derived from fat in the low-carb athletes.
This translated into a two-fold greater use of fat
(~1.2 vs 0.6 g/min) during exercise. Figure 38.2
shows how this enhanced reliance on fat oxida-
tion could enable endurance athletes to compete
without need for exogenous fuel. Serum glycerol
concentrations, a marker of adipose tissue lipoly-
sis (fat breakdown), were two-fold higher during
exercise and recovery in low-carbohydrate run-
ners, further supporting the dramatically acceler-
ated rates of fat metabolism in the keto-adapted
runners (Figure 38.2).
Despite the extraordinary transformation
to extract the majority of energy from fat dur-
ing submaximal exercise in the low-carbohydrate
athletes, they showed a surprisingly high level
of muscle glycogen at rest, glycogen breakdown
during exercise, and synthesis of glycogen dur-
ing recovery. This result was different from that
observed in keto-adapted cyclist after 4 weeks,
suggesting that complete adaptations in glycogen
metabolism may occur over a longer period than
1  month. Although speculative, we hypothesize
there is a need to provide a source of oxaloacetate
for optimal Krebs cycle functioning in the keto-
adapted state, and breakdown of glycogen, without
terminal oxidation, provides a source of carbons
for that purpose.
S U M M A RY
The long-standing dogma in sports nutrition
has been that athletes need to consume a high-
carbohydrate diet in order to excel physically and
cognitively. This approach may work for some
athletes, but for many it is associated with less
than desirable health and performance outcomes.
Ketogenic diets represent an alternative approach
with a rapidly growing level of scientific and obser-
vational support. Although it is clear that keto-
genic diets are an effective weight-loss intervention
often leading to greater fat loss than low-fat diets,
it remains to be determined whether the benefit is
exclusively a result of carbohydrate restriction or
whether nutritional ketosis provides added benefit.
Given the new perspective of βOHB as more than
a metabolite, there are now more reasons to spec-
ulate why a well-formulated ketogenic diet would
be associated with more robust health and perfor-
mance effects. In regard to metabolic syndrome,
383

Chapter 38: Keto-Adaptation in Health and Fitness 383

120
Phinney et al., 1983
110

100

Fat oxidation @ 64% VO2 max (g/h)

90 Volek et al., 2016

80

70

60

50

40
Phinney et al., 1983
30
Volek et al., 2016
20
High-Carb (non- Keto-Adapted
keto-adapted) Athletes
Athletes

FIGURE 38.2 Rate of fat oxidation during submaximal exercise in non-keto-adapted (left) vs. keto-adapted (right)
endurance athletes. The dotted line represents the estimated rate of fat oxidation required to run the marathon leg of an
Ironman triathlon in 2 hr 40 min with minimal glycogen contribution and no exogenous fuel (Volek et al., 2015). Circles
= individual data points; Xs = mean responses.

ketogenic diets are more likely than low-fat diets
to effect global improvement in the diverse signs
and symptoms of this complex disease highly asso-
ciated with type 2 diabetes. Moreover, treating any
of the individual metabolic syndrome markers
with carbohydrate restriction holds the promise to
benefit the others. Therefore, a ketogenic diet is the
preferred primary intervention.
In regard to physical activity, ketogenic diets
dramatically transform energy metabolism during
exercise and may also facilitate recovery through
modulation of inflammation, free radical gen-
eration, and antioxidant defense mechanisms.
However, as with weight loss, it is not clear whether
nutritional ketosis provides added benefit beyond
carbohydrate restriction without ketosis. There are
many health benefits associated with nutritional
ketosis that are relevant for athletes. In regard to
exercise performance, field testing indicates a high
level of success among many endurance athletes,
but there has been very little laboratory testing
focused on performance outcomes. Hopefully
that will change as we dawn on a new era of sports
nutrition that looks beyond the high-carbohydrate
paradigm to alternative approaches that empha-
size a fat-based metabolism.
REFERENCES
Ajala, O., English, P., and Pinkney, J. (2013). Systematic
review and meta-analysis of different dietary
approaches to the management of type 2 diabetes.
Am J Clin Nutr 97, 505–516.
Aljada, A., Friedman, J., Ghanim, H., Mohanty, P.,
Hofmeyer, D., Chaudhuri, A., and Dandona, P.
(2006). Glucose ingestion induces an increase in
intranuclear nuclear factor kappaB, a fall in cel-
lular inhibitor kappaB, and an increase in tumor
necrosis factor alpha messenger RNA by mononu-
clear cells in healthy human subjects. Metabolism
55, 1177–1185.
Bergstrom, J., and Hultman, E. (1966). Muscle glyco-
gen synthesis after exercise:  an enhancing factor
localized to the muscle cells in man. Nature 210,
309–310.
Bergstrom, J., and Hultman, E. (1967). A study of the
glycogen metabolism during exercise in man.
Scand J Clin Lab Invest 19, 218–228.
Bonadonna, R.C., Groop, L.C., Zych, K., Shank, M.,
and DeFronzo, R.A. (1990). Dose-dependent effect
of insulin on plasma free fatty acid turnover and
oxidation in humans. Am J Physiol 259, E736–750.
Browning, J.D., Baker, J.A., Rogers, T., Davis, J.,
Satapati, S., and Burgess, S.C. (2011). Short-
term weight loss and hepatic triglyceride reduc-
tion:  evidence of a metabolic advantage with
dietary carbohydrate restriction. Am J Clin Nutr
93, 1048–1052.
Bueno, N.B., de Melo, I.S., de Oliveira, S.L., and da
Rocha Ataide, T. (2013). Very-low-carbohydrate
ketogenic diet v. low-fat diet for long-term weight
loss: a meta-analysis of randomised controlled tri-
als. Br J Nutr 110, 1178–1187.
384
384 section IV: Ketone-Based Metabolism
Burke, L.M., Hawley, J.A., Schabort, E.J., St Clair
Gibson, A., Mujika, I., and Noakes, T.D. (2000).
Carbohydrate loading failed to improve 100-km
cycling performance in a placebo- controlled trial.
J Appl Physiol (1985) 88, 1284–1290.
Cahill, G., Jr., Felig, P., Owen, O., and Wahren, J.
(1970). Metabolic adaptation to prolonged starva-
tion in man. Nord Med 83, 89.
Cahill, G.F., Jr., and Owen, O.E. (1968). Starvation and
survival. Trans Am Clin Climatol Assoc 79, 13–20.
Christensen, E.H., and Hansen, O. (1939).
Untersuchungen iber die verbrennungsvorgange
bei langdavernder, schwerer Muskelarbeit. Skand
Arch Physiol 81.
Dhindsa, S., Tripathy, D., Mohanty, P., Ghanim, H.,
Syed, T., Aljada, A., and Dandona, P. (2004).
Differential effects of glucose and alcohol on reac-
tive oxygen species generation and intranuclear
nuclear factor-kappaB in mononuclear cells.
Metabolism 53, 330–334.
Drenick, E.J., Alvarez, L.C., Tamasi, G.C., and
Brickman, A.S. (1972). Resistance to symptom-
atic insulin reactions after fasting. J Clin Invest 51,
2757–2762.
Forsythe, C.E., Phinney, S.D., Fernandez, M.L., Quann,
E.E., Wood, R.J., Bibus, D.M., Kraemer, W.J.,
Feinman, R.D., and Volek, J.S. (2008). Comparison
of low fat and low carbohydrate diets on circulat-
ing fatty acid composition and markers of inflam-
mation. Lipids 43, 65–77.
Gibson, A.A., Seimon, R.V., Lee, C.M., Ayre, J.,
Franklin, J., Markovic, T.P., Caterson, I.D., and
Sainsbury, A. (2015). Do ketogenic diets really
suppress appetite? A systematic review and meta-
analysis. Obes Rev 16, 64–76.
Going, S.B., Massett, M.P., Hall, M.C., Bare, L.A., Root,
P.A., Williams, D.P., and Lohman, T.G. (1993).
Detection of small changes in body composition
by dual-energy x-ray absorptiometry. Am J Clin
Nutr 57, 845–850.
Hargreaves M, Hawley JA, Jeukendrup A. Pre-exercise
carbohydrate and fat ingestion: effects on metabo-
lism and performance. J Sports Sci. 2004
Hawley, J.A., and Burke, L.M. (2010). Carbohydrate
availability and training adaptation: effects on cell
metabolism. Exerc Sport Sci Rev 38, 152–160.
Holtz, K.A., Stephens, B.R., Sharoff, C.G., Chipkin,
S.R., and Braun, B. (2008). The effect of carbo-
hydrate availability following exercise on whole-
body insulin action. Appl Physiol Nutr Metab 33,
946–956.
Jensen, M.D., Caruso, M., Heiling, V., and Miles, J.M.
(1989). Insulin regulation of lipolysis in nondia-
betic and IDDM subjects. Diabetes 38, 1595–1601.
Johnstone, A.M., Horgan, G.W., Murison, S.D.,
Bremner, D.M., and Lobley, G.E. (2008). Effects of
a high-protein ketogenic diet on hunger, appetite,
and weight loss in obese men feeding ad libitum.
Am J Clin Nutr 87, 44–55.
Koulmann, N., Jimenez, C., Regal, D., Bolliet, P.,
Launay, J.C., Savourey, G., and Melin, B. (2000).
Use of bioelectrical impedance analysis to estimate
body fluid compartments after acute variations of
the body hydration level. Med Sci Sports Exerc 32,
857–864.
Krauss, R.M. (2005). Dietary and genetic probes of
atherogenic dyslipidemia. Arterioscler Thromb
Vasc Biol 25, 2265–2272.
Krieger, J.W., Sitren, H.S., Daniels, M.J., and
Langkamp-Henken, B. (2006). Effects of varia-
tion in protein and carbohydrate intake on body
mass and composition during energy restriction: a
meta-regression 1. Am J Clin Nutr 83, 260–274.
Long, W., 3rd, Wells, K., Englert, V., Schmidt, S.,
Hickey, M.S., and Melby, C.L. (2008). Does prior
acute exercise affect postexercise substrate oxida-
tion in response to a high carbohydrate meal? Nutr
Metab (Lond) 5, 2.
Maalouf, M., Sullivan, P.G., Davis, L., Kim, D.Y., and
Rho, J.M. (2007). Ketones inhibit mitochondrial
production of reactive oxygen species production
following glutamate excitotoxicity by increasing
NADH oxidation. Neuroscience 145, 256–264.
McClain, A.D., Otten, J.J., Hekler, E.B., and Gardner,
C.D. (2013). Adherence to a low-fat vs. low-
carbohydrate diet differs by insulin resistance sta-
tus. Diabetes Obes Metab 15, 87–90.
Menke, A., Casagrande, S., Geiss, L., and Cowie, C.C.
(2015). Prevalence of and trends in diabetes among
adults in the United States, 1988–2012. JAMA 314,
1021–1029.
Nair, K.S., Welle, S.L., Halliday, D., and Campbell, R.G.
(1988). Effect of beta-hydroxybutyrate on whole-
body leucine kinetics and fractional mixed skeletal
muscle protein synthesis in humans. J Clin Invest
82, 198–205.
Newman, J.C., and Verdin, E. (2014a). Beta-
hydroxybutyrate:  much more than a metabolite.
Diabetes Res Clin Pract 106, 173–181.
Newman, J.C., and Verdin, E. (2014b). Ketone bodies
as signaling metabolites. Trends Endocrinol Metab
25, 42–52.
Olsson, K.E., and Saltin, B. (1970). Variation in total
body water with muscle glycogen changes in man.
Acta Physiol Scand 80, 11–18.
Owen, O.E., Morgan, A.P., Kemp, H.G., Sullivan,
J.M., Herrera, M.G., and Cahill, G.F., Jr. (1967).
Brain metabolism during fasting. J Clin Invest 46,
1589–1595.
Petersen, K.F., Dufour, S., Savage, D.B., Bilz, S.,
Solomon, G., Yonemitsu, S., Cline, G.W., Befroy,
D., Zemany, L., Kahn, B.B., et al. (2007). The role
of skeletal muscle insulin resistance in the patho-
genesis of the metabolic syndrome. Proc Natl Acad
Sci USA 104, 12587–12594.
Phinney, S.D., Bistrian, B.R., Wolfe, R.R., and
Blackburn, G.L. (1983). The human metabolic
response to chronic ketosis without caloric
385
Chapter 38: Keto-Adaptation in Health and Fitness
restriction:  physical and biochemical adaptation.
Metabolism 32, 757–768.
Quann, E.E., Ballard, K.D., Puglisi, M.J., Forsythe, C.E.,
Volk, B.M., Silvestre, R., and Volek, J.S. (2007).
Carbohydrate restriction and resistance training
have additive effects on body composition during
weight loss in men. J Am Diet Assoc 107, A14.
Rauch, J.T., Silva, J.E., Lowery, R.P., McCleary, S.A.,
Shields, K.A., Ormes, J.A., and Wilson, J.M.
(2014). The effects of ketogenic dieting on skeletal
muscle and fat mass. J Int Soc Sports Nutr 11, 1.
Sackner-Bernstein, J., Kanter, D., and Kaul, S. (2015).
Dietary intervention for overweight and obese
adults: comparison of low-carbohydrate and low-
fat diets: a meta-analysis. PLoS One 10, e0139817.
Sato, K., Kashiwaya, Y., Keon, C.A., Tsuchiya, N., King,
M.T., Radda, G.K., Chance, B., Clarke, K., and
Veech, R.L. (1995). Insulin, ketone bodies, and
mitochondrial energy transduction. FASEB J 9,
651–658.
Seshadri, P., Iqbal, N., Stern, L., Williams, M., Chicano,
K.L., Daily, D.A., McGrory, J., Gracely, E.J., Rader,
D.J., and Samaha, F.F. (2004). A randomized study
comparing the effects of a low-carbohydrate diet
and a conventional diet on lipoprotein subfrac-
tions and C-reactive protein levels in patients with
severe obesity. Am J Med 117, 398–405.
Shai, I., Schwarzfuchs, D., Henkin, Y., Shahar, D.R.,
Witkow, S., Greenberg, I., Golan, R., Fraser, D.,
Bolotin, A., Vardi, H., et  al. (2008). Weight loss
with a low-carbohydrate, Mediterranean, or low-
fat diet. N Engl J Med 359, 229–241.
Sherwin, R.S., Hendler, R.G., and Felig, P. (1975). Effect
of ketone infusions on amino acid and nitrogen
metabolism in man. J Clin Invest 55, 1382–1390.
Shimazu, T., Hirschey, M.D., Newman, J., He, W.,
Shirakawa, K., Le Moan, N., Grueter, C.A., Lim,
H., Saunders, L.R., Stevens, R.D., et  al. (2013).
Suppression of oxidative stress by beta- hydroxy-
butyrate, an endogenous histone deacetylase
inhibitor. Science 339, 211–214.
Stephens, B.R., and Braun, B. (2008). Impact of nutri-
ent intake timing on the metabolic response to
exercise. Nutr Rev 66, 473–476.
Tendler, D., Lin, S., Yancy, W.S., Jr., Mavropoulos, J.,
Sylvestre, P., Rockey, D.C., and Westman, E.C.
(2007). The effect of a low-carbohydrate, ketogenic
diet on nonalcoholic fatty liver disease:  a pilot
study. Dig Dis Sci 52, 589–593.
Veech, R.L. (2004). The therapeutic implications of
ketone bodies:  the effects of ketone bodies in
pathological conditions:  ketosis, ketogenic diet,
redox states, insulin resistance, and mitochondrial
metabolism. Prostaglandins Leukot Essent Fatty
Acids 70, 309–319.
Veldhorst, M.A., Westerterp, K.R., van Vught, A.J., and
Westerterp-Plantenga, M.S. (2010). Presence or
absence of carbohydrates and the proportion of fat
in a high-protein diet affect appetite suppression
385
but not energy expenditure in normal-weight
human subjects fed in energy balance. Br J Nutr
104, 1395–1405.
Volek, J.S., Gomez, A.L., and Kraemer, W.J. (2000).
Fasting lipoprotein and postprandial triacylglyc-
erol responses to a low-carbohydrate diet supple-
mented with n-3 fatty acids. J Am Coll Nutr 19,
383–391.
Volek, J.S., Sharman, M.J., Love, D.M., Avery, N.G.,
Gomez, A.L., Scheett, T.P., and Kraemer, W.J.
(2002). Body composition and hormonal responses
to a carbohydrate-restricted diet. Metabolism 51,
864–870.
Volek, J.S., and Feinman, R.D. (2005). Carbohydrate
restriction improves the features of metabolic syn-
drome: metabolic syndrome may be defined by the
response to carbohydrate restriction. Nutr Metab
(Lond) 2, 31.
Volek, J.S., Fernandez, M.L., Feinman, R.D., and
Phinney, S.D. (2008). Dietary carbohydrate
restriction induces a unique metabolic state posi-
tively affecting atherogenic dyslipidemia, fatty acid
partitioning, and metabolic syndrome. Prog Lipid
Res 47, 307–318.
Volek, J.S., Ballard, K.D., Silvestre, R., Judelson, D.A.,
Quann, E.E., Forsythe, C.E., Fernandez, M.L.,
and Kraemer, W.J. (2009a). Effects of dietary car-
bohydrate restriction versus low-fat diet on flow-
mediated dilation. Metabolism 58, 1769–1777.
Volek, J.S., Phinney, S.D., Forsythe, C.E., Quann,
E.E., Wood, R.J., Puglisi, M.J., Kraemer, W.J.,
Bibus, D.M., Fernandez, M.L., and Feinman, R.D.
(2009b). Carbohydrate restriction has a more
favorable impact on the metabolic syndrome than
a low fat diet. Lipids 44, 297–309.
Volek, J.S., Quann, E.E., and Forsythe, C.E. (2010).
Low-carbohydrate diets promote a more favor-
able body composition than low-fat diets. Strength
Cond Journal 32, 6.
Volek, J.S., Freidenreich, D.J., Saenz, C., Kunces, L.J.,
Creighton, B.C., Bartley, J.M., Davitt, P.M., Munoz,
C.X., Anderson, J.M., Maresh, C.M., Lee, E.C.,
Schuenke, M.D., Aerni, G., Kraemer, W.J., Phinney,
S.D. (2016). Metabolic characteristics of keto-
adapted ultra-endurance runners. Metabolism.
Mar;65(3):100–10.
Volek, J.S., Noakes, T., and Phinney, S.D. (2015).
Rethinking fat as a fuel for endurance exercise. Eur
J Sport Sci 15, 13–20.
Youm, Y.H., Nguyen, K.Y., Grant, R.W., Goldberg, E.L.,
Bodogai, M., Kim, D., D’Agostino, D., Planavsky,
N., Lupfer, C., Kanneganti, T.D., et al. (2015). The
ketone metabolite beta-hydroxybutyrate blocks
NLRP3 inflammasome-mediated inflammatory
disease. Nat Med 21, 263–269.
Zivkovic, A.M., German, J.B., and Sanyal, A.J. (2007).
Comparative review of diets for the metabolic syn-
drome:  implications for nonalcoholic fatty liver
disease. Am J Clin Nutr 86, 285–300.
386
39
Advancing the Awareness and Application
of Ketogenic Therapies Globally
The Charlie Foundation and Matthew’s Friends
BETH ZUPEC-KANIA, RD, CD, JIM ABRAHAMS,
EMMA WILLIAMS, MBE, AND SUSAN A. MASINO, PHD
OV E RV I E W
The Charlie Foundation was born out of the
desire to spare children from the unnecessary
suffering that Charlie Abrahams endured before
he achieved seizure freedom with the ketogenic
diet. Jim and Nancy Abrahams shared their story
in 1994 on Dateline and through a 1999 movie
called First Do No Harm. Across the Atlantic in
the United Kingdom, Matthew Williams suf-
fered for 6  years with a devastating seizure dis-
order before becoming seizure-free within 2
weeks of starting the diet. Based on this experi-
ence, Matthew’s Friends was founded in 2004 by
Emma Williams, Matthew’s mother, with a simi-
lar mission as the Charlie Foundation. Despite
these dramatic testimonials and dedicated foun-
dations, the diet remained underutilized. Several
key breakthroughs came in 2008:  The Charlie
Foundation had commissioned medical profes-
sionals with ketogenic experience to collaborate
on guidelines for prescribing the diet. This cul-
minated in a publication in Epilepsia, a presti-
gious international medical journal. During the
same year, a Class I study was published in Lancet
Neurology confirming the diet’s efficacy as a treat-
ment for epilepsy; more positive Class  I  studies
followed. Use of ketogenic diet therapy spread
rapidly worldwide, and with increased use came
a broader understanding of its benefits for other
disorders. Less restrictive versions of the diet were
developed to meet the needs of older children and
adults. In 2012 the Charlie Foundation also began
educating all people with epilepsy to eliminate
sugar, reduce refined carbohydrates, and choose a
predominantly whole foods diet. In addition, both
foundations have expanded efforts to address
other conditions that can benefit from ketogenic
therapies including neurological and neurodegen-
erative disorders and certain cancers.
GENESIS OF THE CHARLIE
F O U N D AT I O N
Charlie Abrahams, the youngest in a family of
three children, had no apparent health problems
until his first birthday in March 1993, when he
experienced his first seizure. This was followed by
a series of tumultuous events that changed his life
as well as the lives of his siblings, Joseph and Jamie;
his parents, Jim and Nancy; and ultimately thou-
sands (possibly millions) of others. His father Jim
described those months as follows:
After thousands of epileptic seizures, an
incredible array of drugs, dozens of blood
draws, eight hospitalizations, a mountain of
EEGs, MRIs, CST scans and PET scans, one
fruitless brain surgery, five pediatric neurolo-
gists in three cities, two homeopaths, one faith
healer, and countless prayers, Charlie’s seizures
were unchecked, his development “delayed”
and he had a prognosis of continued seizures
and “progressive retardation.”
Charlie lived that period in his bed, on a blan-
ket, or in an infant seat, where he was safely
padded during the violent seizures that caused
him to thrash uncontrollably. The adverse effects
of the antiseizure medications were nearly as
destructive as the seizures themselves. Between
seizures, he was groggy, chronically constipated,
and in the words of his father, seemingly “drunk.”
Charlie also underwent a high-risk brain sur-
gery to attempt to remove the area of the brain
suspected to be the source of his seizures. He
387
Chapter 39: Awareness and Application of Ketogenic Therapies Globally
experienced zero improvement, and the seizures
continued.
Exhausted by the failure of the medical com-
munity to provide an effective therapy and reel-
ing from the severe disruption inflicted on the
rest of the family, Jim began looking for another
option. He spent hours in the medical library of
UCLA reading through literature on epilepsy, and
it was there that he came across an epilepsy book
containing a chapter on ketogenic diet as a treat-
ment for epilepsy. He subsequently discovered a
1992 study from Johns Hopkins University that
had been published in Epilepsia documenting 29%
seizure freedom with an additional 30% significant
improvement in 58 consecutive cases of children
who were placed on a ketogenic diet (Kinsman
et al., 1992). Notably, these children were as com-
promised by their disease as Charlie. Jim brought
this book and the study to Charlie’s pediatric neu-
rologist. Although he was aware of the diet, he
immediately dismissed it as a treatment that “he
had never seen work.” Despite this discourage-
ment, Jim and Nancy made the decision to take
Charlie to Johns Hopkins Hospital in Baltimore,
Maryland. (At the time, Johns Hopkins was one
of the few hospitals in the United States that had
provided the ketogenic diet continuously since its
discovery in 1921.) Dr. John Freeman and the die-
titian Millicent Kelly formed the hospital team that
implemented Charlie’s ketogenic diet.
Within 3 days of starting this treatment, Charlie
was seizure-free. Over the course of the following
month he was weaned off of the four failed antisei-
zure medications. Free from the sedating effects of
antiseizure medication, Charlie could finally hold
his head up and once again interact with his family.
Although the diet required careful measurement
and specific foods, Jim described this as a “walk
on the beach” when compared with their previous
circumstances. This abrupt and almost incompre-
hensible reversal left the Abrahams with a diverse
range of emotions; exhilaration and gratitude over
Charlie’s sudden recovery was mixed with anger
over the 9 months of unnecessary suffering that he
had endured.
Jim and Nancy appeared on NBC-TVs
Dateline in 1994 to recap those months. They soon
found their mailbox overflowing with letters from
other families, thanking them for bringing the
diet out into the light of day. Each letter described
a similar journey to the one the Abrahams expe-
rienced with Charlie. The names and ages of the
kids were different, but the stories were the same.
Uncontrolled seizures, multiple medications,
trips to the emergency room, mounting medical
387
bills, and a drugged, poorly functioning child in
a rapid decline. Adoption of the ketogenic diet was
the common denominator, and, for some, their
lifesaver.
One letter came from a family who had to
mortgage their farm to pay their son’s medical
bills. The child was admitted to a hospital for treat-
ment of uncontrollable seizures and placed under
heavy sedation. The family requested a trial of the
ketogenic diet, but the treating physicians refused
to consider this. Instead, they wanted to add more
medications and had recommended surgery. With
the help of a compassionate nurse, the family made
the difficult decision to leave without medical
authorization. They flew directly to Johns Hopkins
Hospital. Several days later, this child, like Charlie,
became seizure-free.
Jim was compelled to use his professional skills
as a movie producer and his connections in the
industry to find ways to help more families learn
about the diet. He wrote, directed, and produced
a dramatization of the devastating impact epilepsy
can have, and the potential for the diet to stop sei-
zures when nothing else has helped. In 1997 this
story aired in a made-for-TV film called First Do
No Harm. The film starred Meryl Streep (who
received a Humanitas Award for her performance),
and the movie became a conduit for spreading the
word about the ketogenic diet worldwide. Twenty-
two years later, the Charlie Foundation continues
to hear from people who learned about the diet
through this movie. Despite the diet’s over-90-year
history, many physicians first learn about the diet
from the families of their patients.
The Abrahams founded the Charlie Foundation
to Cure Pediatric Epilepsy in 1994. Funding was
provided for public awareness and research in ani-
mal models. Videos were developed for families
and medical professionals, and the Johns Hopkins
team received financial support for a book, The
Ketogenic Diet: A Treatment for Epilepsy (now in its
6th edition). Over the last two decades, the Charlie
Foundation has organized educational confer-
ences, maintained a website, and trained medical
teams at hospitals worldwide.
G E N E S I S O F   M AT T H E W ’ S
FRIENDS
Matthew Williams was born in 1995. He suffers
with a catastrophic form of epilepsy called Dravet
syndrome, and his seizures started when he was
9  months old. Like Charlie, his seizures were
severe, frequent, and uncontrolled by medication.
Emma, his mother, asked if Matthew could try the
ketogenic diet when he was 2  years old; she was
388
388 section IV: Ketone-Based Metabolism
told the diet “didn’t work.” As none of the medi-
cations helped, she continued to ask periodically
about the ketogenic diet and was given a variety of
responses; she was told the diet was “too hard” and
caused “terrible side effects.” Like the Abrahams,
Emma battled on, trying innumerable medica-
tions that did not help Matthew’s seizures and
themselves caused devastating side effects. Neither
family was told that it has long been known that
once the first medication fails the chance of sei-
zure control with another medication or an added
medication is lower, and even lower with succes-
sively added drugs (down to 2.7% seizure control
at the third round; Mohanraj and Brodie, 2006).
There is even evidence that polypharmacy can
aggravate seizures (Perucca et  al., 1998; Shorvon
and Reynolds, 1979). Matthew continued to suffer
from thousands of seizures, and his situation did
not improve on what Emma termed the “merry-
go-round” of medications. Emma was told that if
Matthew lived to age 12 he would likely be con-
fined to a residential home.
When Matthew was 7, at a routine appoint-
ment Emma insisted on trying the diet—the only
other option was new combinations of drugs that
had already been proven ineffective and shown to
cause terrible side effects in Matthew:
“The side effects of the medications were awful,
he was on a considerable amount of medica-
tion and his quality of life was so poor already
that it really could not have got any worse for
any of us.”
Fortunately there was a new option. Six years
after Emma first asked about the ketogenic diet,
Matthew and 144 other children with severe epi-
lepsy were enrolled in a clinical research trial of
the ketogenic diet spearheaded by Dr.  Helen
Cross at Great Ormond Street Hospital (GOSH),
London, England (Neal et  al., 2008). Within 2
weeks of starting the diet Matthew’s seizures had
reduced by 90% and within 8 months he was off all
medication. His behavior improved dramatically.
Sadly for Matthew, however, the damage had been
done. Years of thousands of seizures had caused
terrible brain damage, his family had broken apart,
and Emma was now a single mum to Matthew and
his younger sister, Alice.
Inspired by Matthew, and hoping she could
spare other families from what they had endured,
and particularly prevent others from lifelong dis-
abilities, Emma started Matthew’s Friends in
2004—initially running the charity out of her
kitchen. She, like Jim and Nancy Abrahams, was
shocked and angry at the lack of information
and the underutilization of the diet—even when
Emma had asked about it years earlier! She was
determined to raise awareness and access to the
ketogenic diet.
TEN YEARS
O F   A C C E L E R AT I N G A N D
C O M P L E M E N TA RY E F F O R T S
The past 10 years have been a period of enormous
growth and revitalization for the field of metabolic
therapy and for both foundations. Beth Zupec-
Kania, a registered dietitian and nutritionist,
joined the Charlie Foundation in 2006 to develop
resources and trainings for ketogenic therapies.
She designed an online ketogenic diet calculator
program (KetoDietCalculator.org), which allows
nutritionists to efficiently create diets including
meal plans, special recipes, and infant and feeding
tube formulas. A  Help Line within the program
supports clinicians by answering daily questions.
KetoDietCalculator is provided without cost to
licensed nutritionists, who can choose to extend
access to their patients or clients. It receives regular
updates and additions and has been used by over
50,000 people worldwide. Figure 39.1 displays a
screenshot of a typical meal (4:1 ratio of fat to non-
fat grams) designed using the KetoDietCalculator.
In 2007, the Charlie Foundation estimated
that fewer than 15 of the 200 children’s hospitals
in the United States had a ketogenic diet therapy
program. Frustrated by this lack of progress, the
Charlie Foundation commissioned a group of phy-
sicians and dietitians to collaborate on methods of
providing ketogenic diets. Their report, published
in 2008 in Epilepsia was titled “Optimal Clinical
Management of Children Receiving the Ketogenic
Diet:  Recommendations of the International
Ketogenic Diet Study Group” (Kossoff et al., 2009).
This consensus statement became the cornerstone
for developing new programs. The report identi-
fied which patients were most likely to benefit,
outlined methods of starting the diet, and pro-
vided monitoring guidelines. The group summa-
rized its conclusions as follows:
The ketogenic diet (KD) should be strongly
considered in a child who has failed two to
three anticonvulsant therapies, regardless of
age or gender, and particularly in those with
symptomatic generalized epilepsies. It can
be considered the treatment of choice for
two distinct disorders of brain metabolism,
Glut-1 deficiency syndrome and Pyruvate
Dehydrogenase Deficiency Disorder. In the
389
Chapter 39: Awareness and Application of Ketogenic Therapies Globally
FIGURE 39.1 Screen shot of a 4:1 ketogenic meal designed in Ketodietcalculator.org.
particular epilepsy syndromes of Dravet syn-
drome, infantile spasms, myoclonic-astatic
epilepsy, tuberous sclerosis complex, the KD
could be offered earlier.
In the same year as this collaborative report, a
randomized-controlled study of children receiv-
ing the classic and medium chain triglyceride-
supplemented ketogenic diets was published in
Lancet Neurology (Neal et  al., 2008). This is the
clinical trial in which Matthew Williams partici-
pated. Children between the ages of 2 and 16 with
refractory seizures continued their current treat-
ment and were randomized to a ketogenic diet
or continued standard of care for 3  months. The
results were clear:  children treated with the diet
were significantly improved compared with base-
line and with the control group; some had over a
90% reduction in their seizures or were seizure-
free. Seizure frequency in children who continued
to receive the standard of care worsened compared
with baseline, and none had over a 90% reduction
or were seizure-free. This Class I study is consid-
ered the highest level of conclusive evidence and
was the affirmation that doctors and advocates
had been waiting for. It had been nearly 90 years
since the ketogenic diet was developed and finally
a comparison of the diet as treatment for difficult-
to-control epilepsy against multiple antiseizure
medications proved that the diet was superior in
effectiveness. Recently, another Class I study found
similar results (Lambrechts et al., 2016).
The combination of the Epilepsia and Lancet
Neurology publications gave further credence to
the diet, and in 2010 led to the requirement for
inclusion of ketogenic diet therapies in Level 3
and 4 Epilepsy Centers in the United States. This
distinction is granted by the NAEC (National
389
Association of Epilepsy Centers) to centers that
provide medically approved epilepsy treatments.
A S P E C T RU M
O F   K E TO G E N I C D I E T
THERAPIES
Expanding use of the ketogenic diet for epilepsy
paved the way to variations in the diet, particularly
for adults and for older children who find it diffi-
cult to adhere to a restrictive diet. Dr. Eric Kossoff
at Johns Hopkins developed a modified version of
the ketogenic diet known as the modified Atkins
diet (Kossoff et al., 2003). At about the same time,
Heidi Pfeifer, a Massachusetts General Hospital
dietitian, developed the low glycemic index treat-
ment (LGIT; Pfeiffer and Thiele, 2005). These two
newer diets are similar to the classic ketogenic diet
in their high-fat, low-carbohydrate content, but
are different in that they could be initiated outside
of the hospital and do not require careful weigh-
ing of each food. The modified Atkins diet was
shown to be effective in a Class I study (Sharma
et al., 2013). Clinical reviews began to report the
efficacy of the modified Atkins and LGIT diets as
being “nearly as effective” as the classic ketogenic
diet (Coppola et al., 2011; Karimzadeh et al., 2014;
Thibert et al., 2012). These two diets are often eas-
ier for older children and adults to manage and
are sometimes used as predicates to the classic
ketogenic diet.
For now, the classic ketogenic diet remains the
most effective diet therapy based on current sci-
entific evidence. However, less strict versions may
be more realistic for certain individuals, especially
older children and adults. These metabolic diet
treatments are referred to collectively as “ketogenic
therapies.” The Charlie Foundation published a
chart (Table 39.1) that outlines the differences
 390
TABLE 39.1 COMPARISON OF DIET THERAPIES FOR EPILEPSY

QUESTIONS Ketogenic Therapies MCT Oil Low Glycemic Index Modified Atkins
Treatment

Is medical supervision required Yes Yes Yes Yes

Is diet high in fat? Yes Yes Yes Yes
Is diet low in carbohydrate? Yes Yes Yes Yes
What is the ratio of fat to carbo- 4:1, 3:1, 2:1, 1:1 Approximately 1:1 Approximately 1:1 Approximately 2:1
hydrate & protein?
How much carbohydrate is 8gm carb on a 4:1 40-50gm 40-60gm 10gm adolescents or 15gm adults for
allowed on a 1000 Calorie 16gm carb on a 3:1 1 month
diet? 30gm carb on a 2:1 20gm afterwards
40-60gm carb on a 1:1
How are foods measured? Weighed Weighed or measured Measured or estimated Estimated
Are meal plans used? Yes Yes Yes Optional
Where is the diet started? Hospital Hospital Home Home
Are calories controlled? Yes Yes Yes No
Are vitamin and mineral supple- Yes Yes Yes Yes
ments required?
Are liquids (fluids) restricted? No No No No
Is a pre-diet laboratory evalua- Yes Yes Yes Yes
tion required?
Can there be side-effects? Yes Yes Yes Yes
What is the overall difference in This is an individualized and An individualized and This is individualized but This diet focuses on limiting the amount
design of these diets? structured diet that pro- structured diet contain- less structured diet than of carbohydrate while encouraging
vides specific meal plans. ing Medium Chain the ketogenic diet. It uses fat. Carbohydrate may be consumed
Foods are weighed and Triglycerides (MCT) which exchange lists for planning at any time during the day as long as
meals should be consumed are highly ketogenic. This meal and emphasizes com- it is within limits and should be con-
in their entirety for best allows more carbohydrate plex carbohydrates. The sumed with fat. Suggested meal plans
results. The ratio of this diet and protein than the clas- balance of low glycemic are used as a guide. Protein is not
can be adjusted to effect sic ketogenic diet. A 2008 carbohydrates in combi- limited but too much is discouraged
better seizure-control and study showed that both nation with fat result in
also liberalized for better diets are equal in eliminat- steady glucose levels. It is
tolerance. This diet is also ing seizures. A source of not intended to promote
considered a low glycemic essential fatty acids must be ketosis.
therapy and results in included with this diet.
steady glucose levels.
The Charlie Foundation for Ketogenic Therapies © 2015 www.charliefoundation.org
391

FIGURE 39.2 Charlie Foundation logo.

FIGURE 39.3 Charlie thanking Meryl Streep for her role in First Do No Harm.
392
392 section IV: Ketone-Based Metabolism
FIGURE 39.4 Charlie reading to his preschool students in Los Angeles.
between the therapies to assist families and health
professionals in selecting the best option.
Both the Charlie Foundation and Matthew’s
Friends offer a wealth of recipes and expertise,
including demonstration videos. Their tireless
dedication to developing and sharing recipes has
helped make the diet enormously more palatable
and accessible to patients and their families.
O T H E R C O N D I T I O N S T H AT
M AY B E N E F I T
A study published in 2008 in Lipids laid the ground-
work for an entirely new application of ketogenic
diet therapy. Metabolic syndrome is a diagnosis
that includes three or more of the following abnor-
malities; abdominal obesity, elevated blood pres-
sure, elevated fasting blood glucose, and elevated
lipids or triglycerides. This syndrome is currently
an epidemic in many areas of the world, includ-
ing the United States and the United Kingdom.
Improvements in metabolic syndrome were signif-
icantly better in a group that followed a modified
ketogenic diet versus those who followed a low-fat,
FIGURE 39.5 Matthew’s Friends logo
high-carbohydrate diet (Volek et al., 2008). Due in
part to the appetite-suppressing effect of ketosis,
fewer calories are consumed—resulting in weight
and adipose tissue loss along with lowered blood
glucose, triglycerides, and saturated fatty acids.
People inquiring about ketogenic therapy
for brain cancer have become the second-largest
group (next to epilepsy sufferers and their fami-
lies) requesting information from both organiza-
tions. These individuals often experience seizures;
therefore the role that the ketogenic diet offers as
an antiseizure, anti-inflammatory, and antitumor
therapy offers multifaceted benefits (Seyfried and
Mukherjee, 2005; Veech, 2004). Several studies
are currently in progress through the National
Institutes of Health (NIH) investigating ketogenic
diet effects on cancer. Recently Matthews Friends
has partnered with the Astro Brain Tumour Fund
to help adults and children with brain cancer.
Additional applications of ketogenic diet ther-
apies for other conditions have grown exponen-
tially in recent years. Benefits have been reported
in a multitude of neurological disorders and
393
Chapter 39: Awareness and Application of Ketogenic Therapies Globally
FIGURE 39.6 Emma and Matthew Williams
animal models of disorders including autism, cer-
tain mitochondrial diseases, diabetes, migraine,
Prader-Willi syndrome, neurodegenerative dis-
eases including Alzheimer’s and Parkinson’s, trau-
matic brain injury, and stroke. New research may
expand the mission of the Charlie Foundation
and Matthew’s Friends—or may bring new part-
ners supporting metabolic therapies for diverse
applications.
In response to the increasing demands for
adult resources, the Charlie Foundation pro-
duced a guide in 2014 titled “Modified Ketogenic
Therapy for Neurologic and Other Conditions.”
Intended for use under medical supervision, it
describes portion sizes for protein, fat, and car-
bohydrate and advice for optimal nutrition. The
Charlie Foundation has shared resources with
oncology medical professionals and expanded its
website information to include this new popula-
tion of users. A new link on the charliefoundation.
org landing page has been added to ketogenic diet
studies conducted through the NIH for cancer,
epilepsy, and other disorders.
Medical supervision for ketogenic therapies
and choosing the appropriate diet therapy for each
individual is an evolving focus. The spectrum of
393
therapies facilitates selecting the best diet for indi-
viduals taking into account their condition, diag-
nosis, age, and ability to comply. Matthew’s Friends
developed a support system to help families prior
to and during the diet to enhance the likelihood of
success. Adjusting the therapy during the course of
treatment to optimize effectiveness is often neces-
sary. Discontinuing the diet is usually the goal in
childhood epilepsy: two to three years is the typi-
cal course of treatment. However, in some cases of
epilepsy and other neurological disorders it may
be necessary to continue with a modified version
of the diet indefinitely. These discussions are fre-
quently addressed at the Charlie Foundation and
have become the topics of collaborative journal
articles and professional guidelines spearheaded
by nutritionists.
News of ketogenic diet therapies spreads
quickly through social media and the popu-
lar press. Thousands of copies of the Charlie
Foundation’s Parent’s Guide to the Ketogenic Diet
have been distributed in English and Spanish.
Consultant chef Dawn Martenz and nutrition-
ist Beth Zupec-Kania have collaborated on a new
cookbook, Keto Cookbook II (Demos Publications).
Cooking videos of the most popular keto recipes
have been added to the charliefoundation.org
website, and posts of creative snacks and meals
are added to the Facebook page regularly. In addi-
tion, an App is under development for the modi-
fied ketogenic diet. Similarly, Matthew’s Friends is
dedicated to developing and publishing delicious
keto-friendly meals.
Both the Charlie Foundation and Matthew’s
Friends have been continuously supportive of
research and have partnered in funding global
symposia on ketogenic diets. They stay abreast of
new research related to ketogenic diet therapies.
A recent study in the journal Obesity (2015) showed
dramatic improvements in the health of obese chil-
dren in just 10 days after eliminating sugar in their
diets (Lustig et al., 2016). The metabolic impact of
a sugar-free diet is one element of ketogenic diets
that can be undertaken by anyone without the
need for medical supervision. Eliminating sugar
can be a difficult task for most people and requires
strong motivation. A  2007 study published in
Neuroscience and Behavioral Reviews showed that
rats that were fed sugar intermittently had behav-
ior and neurochemical changes that resemble the
effects of substance abuse (Avena et  al., 2008).
Eliminating refined (processed) foods is a second
step that can be taken to improve one’s diet.
The Charlie Foundation has translated these
and other similar research findings into pragmatic
394
394 section IV: Ketone-Based Metabolism
guidelines. A  new publication, Does What I  Eat
Affect My Epilepsy? outlines steps that can be taken
to eliminate sugar and refined foods and to con-
sume a mostly whole foods diet. This has been
distributed widely (in English and Spanish) for all
people with epilepsy regardless of their interest in
ketogenic diet therapy.
For conditions that may benefit from keto-
genic therapies, the Foundation designed a Pre-
ketogenic Diet. This document is distributed (free)
to health professionals to provide to their patients
or clients who are potential candidates for keto-
genic diet therapy. Similar to the classic ketogenic
diet, it eliminates gluten and sugar; however, it is
not intended to induce ketosis. Instead, it can be
described as a whole foods Mediterranean-style
diet that prepares the user for the transition to
ketogenic therapy and also aids in determining
which diet in the spectrum is best suited for their
needs and preferences. In addition, healthcare
professionals can use the Pre-ketogenic Diet as a
screening tool to identify which individuals are
able to adhere to the lifestyle changes required for
ketogenic diets.
T H E C H A R L I E F O U N D AT I O N
A N D M AT T H E W ’ S F R I E N D S —
L O O K I N G TO   T H E   F U T U R E
In 2013 the Charlie Foundation renamed itself the
Charlie Foundation for Ketogenic Therapies to
better define its mission for the future: advocacy,
awareness, and education. An expansion of edu-
cation efforts includes adult-focused media and
training. Development of new resources in both
English and Spanish remains a goal. A list of cur-
rent publications is shown in Box 39.1.
Throughout the 22  years of its existence, the
Charlie Foundation has continued to receive daily
e-mails and letters from families and health pro-
fessionals requesting assistance. Jim Abrahams
has responded to the majority of these requests,
triaging some of them to support staff. Although
Charlie has been seizure- and drug-free and off of
the diet since 1999, Charlie’s family continues to
represent ketogenic therapies in their Los Angeles
community and on a national level through part-
nerships with other nonprofits and with supportive
commercial organizations. Over 160 medical cen-
ters have been trained in the spectrum of ketogenic
therapies in the United States and beyond, includ-
ing Canada, Portugal, Austria, Jamaica, Slovenia,
Kuwait, Saudi Arabia, and the Republic of Georgia.
Beginning in 2008, the Charlie Foundation has
also sponsored and organized global symposia
bringing together leading scientists and medical
BOX 39.1
CHARLIE FOUNDATION
EDUCATION RESOURCES
Parent’s Guide to the Ketogenic Diet
Modified Ketogenic Diet Therapy:  1:1 and
2:1 Prescriptions
Does What I Eat Affect My Epilepsy?
Preketogenic Diet; Low-Carb, Gluten-Free,
High-Fat
KetoDietCalculator Guide for the Nutritionist
and the User
Professional’s Guide to the Ketogenic Diet
Frequently Asked Questions about
Ketogenic Diets
Comparison of Ketogenic Diet Therapies
Ketogenic Diet Primer for Health Professionals
professionals with the intent of advancing research
and clinical use of ketogenic therapies. Plans for
future symposia are in-the-making through the
year 2022. As new clinical and scientific research
continues to emerge, the Foundation will respond
with further resources to promote safe and effec-
tive use of ketogenic therapies and make it less
daunting for the user to manage.
Matthew’s Friends grew rapidly from Emma’s
kitchen table, and in 2011 opened its own clinic.
The charity has continued to grow rapidly, and
recently expanded into New Zealand and Canada.
In 2016 it embarked on a new program called
KetoCollege, to offer training for teams around the
globe, adding to and complementing the efforts
of the Charlie Foundation to train ketogenic diet
teams at their home institutions around the world.
Now seizure-free young adults—no longer on
a special diet—Charlie Abrahams and Matthew
Williams are two clear examples that even cata-
strophic epilepsy can be cured by a ketogenic
diet. A  question frequently asked about Charlie,
Matthew, and others who have become seizure-
free on a ketogenic therapy is “How is it possible
that they can come off of the diet and remain
seizure-free?” At the present time, there is no clear
answer to this question. Emerging genetic research
may soon clarify whether early application of keto-
genic therapy may either ameliorate the disease
or even cure it. The field of nutritional genomics
395
Chapter 39: Awareness and Application of Ketogenic Therapies Globally
looks at the effect of nutritional changes on genes
and suggests that maladaptive epigenetic changes
can be altered by diet. Targeted diet therapies are
already being prescribed along with genetic testing
for certain metabolic disorders.
Implementation of a restrictive diet is easiest
in infants and children (given their reliance on the
parents) and becomes more challenging as children
gain independence. Early initiation of ketogenic
therapies is better tolerated in young children—
which may result in improved compliance and
outcomes. Nevertheless the diet does work in
adults, and has ever-increasing applications.
REFERENCES
Avena, N.M., Rada, P., and Hoebel, B.G. (2008).
Evidence for sugar addiction: behavioral and neu-
rochemical effects of intermittent, excessive sugar
intake. Neurosci Biobehav Rev 32, 20–39.
Coppola, G., D’Aniello, A., Messana, T., Di Pasquale,
F., della Corte, R., Pascotto, A., and Verrotti, A.
(2011). Low glycemic index diet in children and
young adults with refractory epilepsy: first Italian
experience. Seizure 20, 526–528.
Karimzadeh, P., Sedighi, M., Beheshti, M., Azargashb,
E., Ghofrani, M., and Abdollahe-Gorgi, F. (2014).
Low glycemic index treatment in pediatric refrac-
tory epilepsy: the first Middle East report. Seizure
23, 570–572.
Kinsman, S.L., Vining, E.P., Quaskey, S.A., Mellits, D.,
and Freeman, J.M. (1992). Efficacy of the keto-
genic diet for intractable seizure disorders: review
of 58 cases. Epilepsia 33, 1132–1136.
Kossoff, E.H., Krauss, G.L., McGrogan, J.R., and
Freeman, J.M. (2003). Efficacy of the Atkins diet
as therapy for intractable epilepsy. Neurology 61,
1789–1791.
Kossoff, E.H., Zupec-Kania, B.A., Åmark, P.E.,
Ballaban-Gil, K.R., Bergqvist, A.G.C., Blackford,
R., Buchhalter, J.R., Caraballo, R.H., Cross,
J.H., Dahlin, M.G., et  al. (2009). Optimal clini-
cal management of children receiving the keto-
genic diet: recommendations of the International
Ketogenic Diet Study Group. Epilepsia 50,
304–317.
Lambrechts, D.A.J.E., de Kinderen, R.J.A., Vles, J.S.H.,
de Louw, A.J.A., Aldenkamp, A.P., and Majoie,
H.J.M. (2016). A randomized controlled trial of
395
the ketogenic diet in refractory childhood epi-
lepsy. Acta Neuro Scand. doi: 10.1111/ane.12592
Lustig, R.H., Mulligan, K., Noworolski, S.M., Tai, V.W.,
Wen, M.J., Erkin-Cakmak, A., Gugliucci, A., and
Schwarz, J.M. (2016). Isocaloric fructose restric-
tion and metabolic improvement in children with
obesity and metabolic syndrome. Obesity (Silver
Spring) 24, 453–460.
Mohanraj, R., and Brodie, M.J. (2006). Diagnosing
refractory epilepsy:  response to sequential treat-
ment schedules. Eur J Neurol 13, 277–282.
Neal, E.G., Chaffe, H., Schwartz, R.H., Lawson, M.S.,
Edwards, N., Fitzsimmons, G., Whitney, A., and
Cross, J.H. (2008). The ketogenic diet for the treat-
ment of childhood epilepsy:  a randomised con-
trolled trial. Lancet Neurol 7, 500–506.
Perucca, E., Gram, L., Avanzini, G., Dulac, O. (1998)
Antiepileptic drugs as a cause of worsening sei-
zures. Epilepsia 39, 5–17
Pfeiffer, H.H., and Thiele, E.A. (2005). Low-glycemic-
index treatment:  a liberalized ketogenic diet for
treatment of intractable epilepsy. Neurology 65,
1810–1812.
Seyfried, T.N., and Mukherjee, P. (2005). Targeting
energy metabolism in brain cancer:  review and
hypothesis. Nutr Metab 2, 30.
Sharma, S., Sankhyan, N., Gulati, S., and Agarwala, A.
(2013). Use of the modified Atkins diet for treat-
ment of refractory childhood epilepsy: a random-
ized controlled trial. Epilepsia 54, 481–486.
Shorvon, S.D., and Reynolds, E.H. (1979) Reduction
in polypharmacy for epilepsy. Br Med J 2:1023–5.
Thibert, R.L., Pfeifer, H.H., Larson, A.M., Raby,
A.R., Reynolds, A.A., Morgan, A.K., and Thiele,
E.A. (2012). Low glycemic index treatment for
seizures in Angelman syndrome. Epilepsia 53,
1498–1502.
Veech, R.L. (2004). The therapeutic implications of
ketone bodies:  the effects of ketone bodies in
pathological conditions:  ketosis, ketogenic diet,
redox states, insulin resistance, and mitochondrial
metabolism. Prostaglandins Leukot Essent Fatty
Acids 70, 309–319.
Volek, J.S., Fernandez, M.L., Feinman, R.D., and
Phinney, S.D. (2008). Dietary carbohydrate
restriction induces a unique metabolic state posi-
tively affecting atherogenic dyslipidemia, fatty acid
partitioning, and metabolic syndrome. Prog Lipid
Res 47, 307–318.
396
397

INDEX

Page references followed by an italics f indicate figures, t indicate tables, and b indicate boxes.

Abbott, N. J., 289, 292–294 adverse effects. see side effects

Abdallah, D. M., 174–175 aerobic fermentation, 80
Abrahams, Charlie, 386 aerobic glycolysis, 80
Abrahams, Jim, 3 aging
AC-1202, 319–320 cell senescence, 257
acetoacetate. see also ketone bodies neuroprotection, ketosis, 221–222, 221f
acetone from, 113 reactive oxygen species, 256–257
administration, 124 telomere shortening, 256–257
Alzheimer’s, 122f aging and neurodegeneration, 216–222
brain energy source, 113 cataplerosis–anaplerosis balance, 218
brain uptake, 296f glucose and ketone metabolism, 113–127 (see also
hippocampal slices, 187–190, 189t Alzheimer’s disease)
inflammation, 150–151 glucose metabolism, oxidative stress, 217–218, 218f
ketogenic diet, 90, 136 HIF1α stabilization, 219–221
Krebs cycle, 243f ketone bodies, 216–217
mitochondrial metabolism, 315f ketosis, 217
NADP system, 263f ketosis neuroprotection, clinical relevance,
neuroprotection, 135 218–219, 219f
prolyl-hydroxylase inhibition, 220 Ahn, Y., 107
redox potential, 262t D-alanine, seizure onset protection, 348
synthesis and metabolism, 177, 187, 228, 244, Allen, B. G., 92, 367
247–248, 248f, 296f Allen, Frederick Madison, 364, 365
VGLUTs, 281 Alpers disease, status epilepticus, 61
acetone “alternative” ketogenic diets, 5–13. see also specific types
from acetoacetate, 113 choice, 11–12
ketogenesis, 119f effectiveness, vs. classical ketogenic diet, 10–12, 10t
acidosis, 67 initiation and follow-up, 12–13
actin, 293 low glycemic index treatment, 6f, 9–10, 9b
Adabi Mohazab, R., 174 medium chain triglyceride, 5–7, 6b, 6f
adenosine modified Atkins diet, 6f, 7–9, 8b, 8t
on inflammation, 151f, 152–153 Alzheimer, Alois, 242
pain, 201 Alzheimer’s disease, 113–127, 241–249
retaliatory metabolite, 210 anaplerosis, 124–125
adenosine homeostasis arteriovenous difference, brain, 114–115, 115t
seizures, 209–210 brain energy status, neuroprotection, and
temporal lobe epilepsy, 211–213, 212f mitochondrial function, 125–126
adenosine kinase, epilepsy overexpression, 209–210 early brain development and evolution, 126
adherence, 19 environmental factors, 241–241–242
Adk gene, epilepsy, 210–211 genetics, 241
adult polyglucosan body disease, triheptanoin, 341–342 glucose, cerebral use, 243–244
398

398 Index

Alzheimer’s disease, (cont.) astrocytes

glucose, presymptomatic hypometabolism, 116–118, BAD-altered, metabolic changes, 272
116t, 117f blood-brain barrier, 290f, 292–293
glucose, uptake and metabolism, 113–114 athlete, keto-adapted, 382, 383f
hyperketonemia with cognitive deficit, 123–124, 125t autism spectrum disorder, 101–107
hypometabolism, 216–217 animal models, 105–107
incidence, 241 definition, 101
ketone bodies, available forms, 247–249, 248f, 249f effects, 105
ketone kinetics and transport, 118–120, 120f, 120t epidemiology, 101
ketone regional uptake, early disease, 121–122, etiology, 101
121f, 122f future research, 107
ketones, brain fuel, 113, 118, 119f ketogenic diet, 103–105
ketone supplementation, 318–320, 319f mitochondrial and metabolic dysfunction, 102–103,
medium chain triglycerides, dose-response 104t–105t
ketogenesis, 6f, 122–123, 123f pathophysiology, 101–102
medium chain triglycerides, safety, 126
nutritional and metabolic factors, 241–242 Babayan, Vigen K., 311
pathophysiology, 242–244, 243f BAD, 271
perspectives, 126–127, 127t on KATP channels, 271, 275
PET-FDG protocol, 114 BAD channels, metabolic seizure resistance,
predisposing factors, 241–242 271–274, 277f
prevention, 113 glucose and ketone body metabolism, 271–272, 282
risk factors, 113 KATP channels, 275–277, 276f
treatment, 244–247, 245f–246f metabolic function alteration, 272, 273f, 277f
amyloid, 244 neurons and astrocyte metabolic changes, 272
ketone body metabolism, 244–247, 245f–246f vs. other BCL-2 family proteins, 273
triheptanoin, 341 Bansal, S., 29
amino acids, for neurological disorders, 346–349 Banting, William, 363
seizures and epilepsy Barborka, C. J., 16–17, 23
D-amino acids, 347–349, 349b basement membranes, blood-brain barrier, 290f, 292
L-amino acids, 346–347, 347b Bastible, C., 16–17
traumatic brain injury, 349 BCL-2-associated agonist of cell death (BAD), 271. see
AMPA receptor agonists, medium chain fatty acids, also BAD channels, metabolic seizure resistance
331–332, 331t behavioral effects, 231
amyloid precursor protein (APP), 242, 243 Bellisario, V., 236
amyotrophic lateral sclerosis (ALS), 157–158 Bergqvist, A. G. C., 69
triheptanoin, 341 beta-oxidation metabolic pathway, 248, 249f
anaplerosis, 124–125, 218 β-hydroxybutyrate (BHB), 307. see also ketone bodies
propionyl-CoA carboxylation pathway, aging brain and Alzheimer’s disease, 113, 115t,
336–337, 338f 121–122, 121f
TCA cycling, 336, 337f, 338f aging brain and Alzheimer’s disease, kinetics,
Angelman syndrome, 52–53 120, 120t
low glycemic index treatment, 9–10, 9b antioxidant function, 138
anorexigenic food intake, 229, 229f brain energy source, 115t, 121–122, 121f
antiepileptogenesis. see epileptogenesis actions brain uptake, 296f
anti-excitotoxic effects, 137 cancer, 81–82, 83
antioxidants. see also specific types on cellular processes, 381
mechanisms of action, 257–258, 258f hippocampal slices, 187–190, 189t
neuroprotective mechanisms, 137–138 inflammation, 150–151, 151f
apolipoprotein E (ApoE), 241 KATP channels, 157
apolipoprotein E-4 (ApoE-4), 241, 242 ketogenesis, 119f
L-arginine ketogenic diet, 90, 281
seizure onset protection, 347 liver synthesis, 156
traumatic brain injury, 349 medium chain triglycerides, 123f
ascorbate, 258, 261, 262t metabolism and sum reactions, 243, 243f, 246f–248f,
ascorbic acid, 258f, 259 247–248
astrocyte-neuron lactate shuttle, 282–285, 283f–285f mineral salts supplementation, with medium chain
antiepileptic effects, 283–286, 284f, 285f triglycerides, 311–312
electrical regulation, 284, 285f neuroprotection, 124, 124f, 125t, 134
399

Index 399

nutritional ketosis, 377, 379 brain development, ketones, 126

Parkinson disease, 158 brain energy
on reactive oxygen species, 83 acetoacetate, 113
rodent model, 233f, 235–236 Alzheimer’s disease, 126
synthesis and metabolism, 187, 228, 296f β-hydroxybutyrate, 115t, 121–122, 121f
D-β-hydroxybutyrate bioenergetics, 77
administration, 124 ketones, 113, 118, 119f, 380–381
ionizing radiation protection, 263, 264f mitochondria, 125–126
ketogenesis, 119f requirements, 113
ketogenic diets, 136 brain energy metabolism
mitochondrial metabolism, 315f PET-FDG protocol, 114
neuroprotection, 135 triheptanoin, 339–340
on redox states, cellular, 261–263, 262f, 263f brain evolution, ketones, 126
on transcription, 261t, 264 brain injury, traumatic. see traumatic brain injury
Blackburn, Elizabeth, 256–257 brain ketones, 113, 118, 119f
Blomqvist, G., 121, 121f, 122 Bucher, T., 258
blood-brain barrier, 289–299 Buckley, J. A., 105
caveoline-based vesicle trafficking, 291f, 293 1,3-butanediol supplementation, 311
cellular associations, 289–292
astrocytes, 290f, 291–292 Cahill, G. F., 81, 258, 307, 317, 376
basement membranes, 290f, 292 calorie restriction
endothelial cells, 289–292, 290f, 291f classic ketogenic diet, 29
pericytes, 290f, 292 on inflammation, 151–152, 151f
endothelial cells, 289, 290f pain, 197–198
functions on tumor growth, 81–82
nutrient transport to brain, 291f, 294–295, 296f Canavan disease, triheptanoin, 341
transit regulation, 290f, 291f, 294 cancer, 79–83. see also brain cancer
glucose transporters, 294–295, 296f, 297f aspartate, 80–81
ketogenic diet, 295–299 calorie restriction, 81–82
cerebral ketone uptake regulation, 295–297, 297f cell metabolism, 80–81
ketone bodies on, 298–299 genetics and somatic mutation theory, 79
neuroinflammation, 297–298 as genetic vs. mitochondrial
neuroprotection, 298 metabolic disease, 79–80
molecules glucose, 80–81
leukocyte adhesion molecules, 293 Glucose/Ketone Index calculator, 82–83
tight junctions, 290f, 292–293 glutamine, 81
plasmalemma vesicle-associated hypoxia-inducible factor 1, 89–90
protein number, 293 ketogenic diet, 82–83
structure and functions, 289, 290f ketone supplementation, 320–321
transporters, 290f, 291f, 293, 296f mitochondrial dysfunction, 80
body composition, keto-adaptation, 378–379 nuclear factor-kappa B, 90
Bough, K. J., 192 PI3K/AKT signaling pathway, 89
brain radiation therapy, 92
ketones, 113 reactive oxygen species, 90
metabolic rate, glucose, 115 Caraballo, R. H., 44–47
PPARγ, 170–172, 172f carbohydrate
brain, aging, glucose and ketone metabolism, 113–127. depletion, 376
see also Alzheimer’s disease stored, 376
brain cancer, 392 carbohydrate, dietary
brain cancer, malignant, 88–94 decreasing, 376
glioblastoma multiforme, 88 glycemic effect, 366f, 367
ketogenic diet, humans, 93–94 carbohydrate loading, 380
ketogenic diet, with standard carbohydrate restriction, dietary
therapies, 90–91 diabetes mellitus, controlled studies, 367, 368t
metabolic remodeling, 88 diabetes mellitus, randomized controlled trials,
preclinical evidence, 91 367, 369t
standard therapies, 91–93, 92f obesity, randomized controlled
tumor metabolism and Warburg trials, 363, 363t
effect, 88–90, 89f carboxy-terminal binding protein (CtBP), 356
400

400 Index

cardiac arrest and resuscitation diabetes mellitus type 2, 362–371

ketosis neuroprotection, 218–219, 219f carbohydrate, glycemic effect, 366f, 367
survival after, 217–218, 218f carbohydrate restriction, randomized controlled trials,
cardiac hypertrophy, triheptanoin, 342 367, 369t
cardiac side effects, 69–70 dietary recommendations, evolution, 365
cardiomyopathy, 69–70 dietary recommendations, post-insulin era, 364
Castellano, C. A., 121, 121f, 122f keto-adaptation, 378
cataplerosis, 124–125, 218 ketogenic diet
caveoline-based vesicle trafficking, 291f, 293 metabolic ward study, 367
cell senescence, 257 nonrandomized studies, 370, 370t
cerebral metabolic rate, glucose, 115 origins and use, 363–364
Champ, C. E., 94 pilot study, 367–370
Chang, P., 7, 328–332, 329f, 330t, 331t remission case report, 370, 371f
Charlie Foundation, 3, 27, 386–392, 391f, 392f ketone supplementation, 321
achievements, 388–389, 389f low carbohydrate diets, controlled studies, 367, 368t
education resources, 393–394, 394b nutrition therapy, low glycemic dietary patterns,
future, 394–395 365–367
genesis, 386–387 pathophysiology, 362
history, 386 prevalence, 362
ketogenic diet therapy spectrum, 389–392, 390t progression and morbidity, 362
research support, 393–394 dietary therapy, adults, 16–24. see also specific types
childhood absence epilepsy (CAE), 56–57 demand, 17–19
Chuang, Y. C., 173–174 children transitioning to adult epilepsy
claudins, 292 providers, 17–18
clinical trials, 77 refractory epilepsy, 18
cognition, 22 super-refractory status epilepticus, 18–19
cognitive impairment, mild effects, adverse
ketogenic treatments, 127, 127t gastrointestinal, 22–23
medium chain triglycerides, 123f, 124, 125t lipids, 23
cognitive impairment, severe. see menstrual cycle, 23
Alzheimer’s disease other, 23
constipation, 67–68 effects, beneficial, 22t
convulsive status epilepticus, 61 cognition and mood, 22
Coppola, G., 317 weight loss, 22
cortical spreading depression, 160 history, 16–17
Couch, S. C., 70 results
counseling, prediet, 27 efficacy, 19–22, 20f, 21t
Crabtree effect, 80 feasibility, tolerability, and adherence, 19
Cross, Helen, 388 discontinuation, ketogenic diet, 70–71
prevalence, 66, 67t
D’Agostino, D. P., 91, 310–313, 317 DNA methylation levels, epilepsy, 211–213, 212f
Darlington, C. D., 79 Does What I Eat Affect Epilepsy?, 394
Davidson, T. E., 231 Doose syndrome, 35, 46–47
Deanna Protocol, 158 Dravet syndrome, 45–46
decanoic acid, 331. see also Drenick, E. J., 114, 118, 121f
medium chain fatty acids Dressler, A., 43, 45–46
AMPA receptors and seizure control, 331 drug resistant epilepsy, 40
blood-brain barrier penetration, 328 duration, treatment, 31
mitochondrial proliferation, 332 dyslipidemia, 229, 230t
seizure control, 330, 332
2-deoxyglucose, 353–359 education, prediet, 27
anticonvulsant actions, 353–355 effectiveness, timing, 29–30
antiepileptic actions, 355–356 efficacy, 19–22, 20f, 21t
chemical structure, 354f electrolyte imbalances, 69
glucose metabolism and glycolytic Ellenbroek, J. H., 229, 229f
pathway, 353, 354f El-Rashidy, O. F., 30
depression, 159 endocrine effects, 227–230, 228t, 229f, 230f, 230t
diabetes mellitus, Alzheimer’s disease and vascular endothelial cells, blood-brain barrier, 289–292, 290f, 291f
dementia, 242 energy, brain. see brain energy
401

Index 401

energy regulation fitness, keto-adaptation, 380–381

dysregulation, 77 food refusal, 70
long-term, 228–229 Freeman, J. M., 3, 40, 387
short-term, 228 Freemantle, E., 123, 124f
epigenetics, epilepsy, 210–211 free radicals. see also reactive nitrogen species (RNS);
epilepsy. see also specific topics reactive oxygen species (ROS)
adenosine kinase overexpression, 209–210 Alzheimer’s disease, 243–244
Adk gene, 210–211 redox state, 261, 262t
2-deoxyglucose for, 355–356 removal, in vivo, 263–264
DNA methylation levels, 211–213, 212f fructose-1,6-diphosphate (FDP), 357
energy metabolism and triheptanoin, 337–339, 338f
epigenetics, 210–211 Gahring, L. C., 170–171
fasting, 209 gene theory of cancer, 79
kainic acid status epilepticus (KASE) model, 210, 211 Geyelin, H. R., 16
ketogenic diet mechanisms, 281–282 Gibson, C. L., 298
ketone supplementation, 317 Giménez-Cassina, A., 272–274
PPARγ, 173–174 glioblastoma multiforme, 88. see also brain cancer,
epilepsy indications, established, 40–47 malignant
Doose syndrome, 35, 46–47 glioma, malignant, 88. see also brain cancer, malignant
Dravet syndrome, 45–46 glucagon, 229–230
infantile spasms, 42–44 glucose, 336
Lennox Gastaut syndrome, 44–45 blood levels, ketogenic diet, 228, 228t
refractory nonsurgical epilepsy, 40–42 chemical structure, 354f
epileptic encephalopathies, 44. see also specific types decreases, 281–282
epileptogenesis actions, 209–213 pain, 200–201
adenosine homeostasis and seizures, 209–210 glucose, brain
definition, 355 blood-brain barrier, 294–295, 296f, 297f
DNA methylation levels, 211–213, 212f cerebral metabolic rate, 115
epigenetics, 210–211 requirements, 113
ketogenic diet, 137, 165, 198 (see also ketogenic diet; glucose, brain, Alzheimer’s
seizure control) hypometabolism, presymptomatic, 116–118, 116t, 117f
ketogenic diet and adenosine homeostasis, uptake vs. metabolism, 118
211–213, 212f glucose-insulin axis, 376
erythropoietin, neuroprotection, 220 Glucose/Ketone Index Calculator (GKIC), 82–83
Eun, S. H., 43–44 glucose metabolism
evaluation, pre-treatment, 66–67, 67t, 71t BAD on, 271–272
Evangeliou, A., 105 2-deoxyglucose, 353, 354f
excitotoxic effects, 137 oxidative stress, 217–218, 218f
glucose suppression, 313–314
family-centered, team-based approach, 26 glucose transporters (GLUTs), 294–295, 296f, 297f
fasting GLUT 1 (glucose transporter type 1), 293, 297f
classic ketogenic diet, 28–29 GLUT 1 (glucose transporter type 1) deficiency, 18,
epilepsy, 209 35–37, 308
hyperketonemia, 307 clinical presentation and phenotype, 35–36, 36f
neuroprotection, 134 diagnosis, 35
pain, 196–197 ketogenic diet efficacy, 20
spinal cord injury, 134 ketogenic diet therapies, 36–37
fats ketone supplementation, 317–318
dietary, as fuel, 380–381 (see also brain energy) metabolic concept, 35, 36f
stored, 376 open questions, 37
fatty acid metabolism, beta-oxidation metabolic pathway, triheptanoin, 341
248, 249f glutamate, 349
Feinman, R. D., 378 D-glutamate, seizure onset protection, 348
fever, on inflammation, 147–148 glutathione
fever-induced refractory epileptic encephalopathy in mechanisms of action, 259, 261, 262t
school-age children (FIRES), 147 on reactive oxygen species, 264
status epilepticus, 60, 61 glycemic dietary patterns, diabetes mellitus, 365–367
fibroblast growth factor 21, 147–148 glycemic index (GI), 9, 365
First Do No Harm, 3, 41, 386, 391f glycemic load, 366
402

402 Index

glycine, seizure onset protection, 346 implementation, 26–31

glycolysis, 336, 354f classic ketogenic diet
glycolytic flux, neuroprotection, 157 caloric restriction, 29
glycolytic pathway, 2-deoxyglucose, 353, 354f fasting, 28–29
Go Lower Low Carbohydrate Diet, 367–370 liquid/formula vs. food, 29
Gottstein, U., 114 education and counseling, prediet, 27
GPR40, 199, 200 effectiveness, timing of determination, 29–30
GPR41, 201 family-centered, team-based approach, 26
GPR120, 201 initiation, in-patient vs. out-patient, 27–28
Greenwood, C. E., 231 ketogenic diet team, composition, 26–27
growth, 70 ratios, importance, 30–31
Gudsnuk, K., 233f, 235–236, 235f, 235t response predictor, 31
Guelpa, G., 16 seizure freedom, 30
treatment duration, 31
Hadera, M. G., 340 infantile spasms, 42–44
Harmon, Denon, 256 infections, 68
Hayflick, L., 256, 257 inflammasome, 138–139
HDAC inhibitor, 321 NOD-like receptor P3, 298
ionizing radiation protection, 264 inflammation, 147–153
hepatitis, 69 animal models, neurological disease, 148–150
Herbert, M. R., 105 multiple sclerosis, 148–150
HIF1α pain, 148
activation, redox state, 216 Parkinson’s disease, 150
ketones, 219 studies, 148, 149t
neuroprotection clinical evidence, 153
ketosis, 221–222, 221f fever, 147–148
stabilization, 219–221 fibroblast growth factor 21, 147–148
oxygen homeostasis, hypoxia, 220 FIRES, 147
regulation, 221 future indications, 153
hippocampal slice, 186–193 ketone supplementation, 314–316
for ketogenic diet and epilepsy studies, 187 mechanisms, 150–153, 151f
ketogenic diet–fed animal slices, 188f, 189t, 191 adenosine modulation, 151f, 152–153
ketone body direct application, caloric restriction, 151–152, 151f
187–190, 188f, 189t ketone bodies, 150–151
temporal lobe epilepsy model, 186–187 mitochondrial membrane potential, 153
whole-cell patch clamp for glucose and ATP control, polyunsaturated fatty acids, 151f, 152
188f, 189t, 190–191, 191f reactive oxygen species, 151f, 153
Hippocrates, 16, 28 uncoupling proteins, 153
Hong, A. M., 43 neuroinflammation, 297–298
Hong, S., 175 neuroprotection, 138–140
hormonal mechanisms, 227 pain, 147, 148, 196 (see also pain)
Hoyer, S., 114–115, 122 initiation, in-patient vs. out-patient, 27–28
Huntington’s disease, triheptanoin, 342 Inoue, T., 283–286, 284f, 285f
Huttenlocher, P. R., 5, 328 insulin
hydrogen peroxide, 264 blood levels, ketogenic diet, 228, 228t
synthesis, monoamine oxidase, 254–255, 255f glucose-insulin axis, 376
3-hydroxybutyrate ketone monoester (KME), hyperinsulinemia, 313–314
318–320, 319f insulin resistance
3-hydroxybutyrate methyl ester (HBME), 318 brain, Alzheimer’s disease, 243
hyperglycemia, 313–314 keto-adaptation, 377–378
hyperinsulinemia, 313–314 ketone supplementation, 321
hyperketonemia nonalcohoic fatty liver disease, 378
benefits, 307 weight loss, keto-adaptation, 379
with cognitive deficit, clinical insulin sensitivity
studies, 123–124, 125t brain ketone use, 118
hyperlipidemia, 69 ketone supplementation, 313–314
hypoglycemia, 68 International Consensus Statement for Ketogenic Diet
hypometabolism, Alzheimer’s disease, 216–217 (ICSKD), 26
hypoxia-inducible factor 1 (HIF-1), 89–90 intrauterine growth restriction (IUGR), 234
403

Index 403

ionizing radiation other applications, 392–393

HDAC inhibitor protection, 264 research support, 393–394
ketone ester effects, post-exposure, 266–267, 267f ketone-based metabolism, 282
ketosis protection, 263 ketone bodies and ketones,271. see also acetoacetate; β-
in vitro studies, 264–266, 265f, 266f hydroxybutyrate (BHB); specific types
L-isoleucine, seizure onset protection, 346 alternative energy substrates, 216–217
Alzheimer’s disease, 244–247, 245f–246f
JAMs, 292–293 available forms, 247–249, 248f, 249f
Jaworski, D. M., 80–81 on blood-brain barrier, 298–299
Jeong, E. A., 176, 177 blood levels, ketogenic diet, 228, 228t
Joensson, E. A., 370 brain, 113, 118, 119f
Joslin, Elliott P., 364, 365, 367 development and evolution, 126
Juge, N., 188–190 fuel, 113, 118, 119f, 380–381
junctional adhesion molecules (JAMs), 292–293 uptake, 295–297, 297f
juvenile myoclonic epilepsy (JME), 18, 57–58 uptake, regional, early Alzheimer’s, 121–122,
121f, 122f
kainic acid status epilepticus (KASE) model, 210, 211 breakdown, 28
Kang, H., 69 exogenous, 307
Kashiwaya, Y., 314, 316, 318, 321, 322 vs. glucose, 217
KATP channels, BAD on, 271 as “good medicine,” 81
KATP channels, metabolic seizure resistance, increased, 281
274–277, 277f on inflammation, 138–140, 150–151
BAD disruption on, 275 kinetics and transport, 118–120, 120f, 120t
metabolism and excitability link, 274–275 metabolism, BAD on, 271–272
seizure protection, BAD-elicited, 275–277, 276f neuroinflammation, 297–298
Kelly, Millicent, 387 neuroprotection, 126, 156–157, 298
Kessler, S. K., 30–31 on pain, 200
keto-adaptation, 312, 376–383 on radiation effects, post-exposure, 266–267, 267f
athlete, 382, 383f on tumors, 91
effects, 376–377 ketone ester supplementation, 312
fundamentals, 376–377 ketone salts supplementation, 311–312
health implications, 377–380 ketone supplementation, 310–322
insulin resistance, metabolic syndrome, diabetes applications, 317–322
type 2, 377–378 Alzheimer’s disease, 318–320, 319f
lipoprotein profile, 379–380 cancer, 320–321
weight loss and body composition responses, epilepsy/seizure disorders, 317
378–379 GLUT1 deficiency syndrome, 317–318
history, 376 insulin resistance/type 2 diabetes mellitus, 321
nutritional ketosis, 377, 377f weight loss, 321–322
sport and fitness, 380–381 development and testing, 310–312
starvation ketosis, 377 1,3-butanediol, 311
Keto Cookbook II, 393 formulations, 310
KetoDietCalculator, 388 ketone esters, 312
ketogenesis, 118, 119f ketone salts, 311–312
ketogenic diet. see also specific types and topics medium chain triglycerides, 310–311
history and early research, 3–4, 26 βHB mineral salts and medium chain triglycerides,
ketosis, 217 311–312
mechanisms of action, 90–91 (see also specific types) therapeutic ketosis from, 310
pediatrics indications, 4 therapeutic mechanisms, 313–316
spectrum, 389–392, 390t glucose suppression and insulin sensitivity
types, 3–4 enhancement, 313–314
use, 26 on inflammation, 314–316
ketogenic diet, advancing awareness, 386–395 metabolic efficiency, enhanced, 314, 315f
brain cancer, 392 mitochondrial health and function, 316
Charlie Foundation, 3, 27, 386–392, 391f, 392f (see oxidative stress, inhibition, 316
also Charlie Foundation) ketosis, nutritional (therapeutic), 217, 307, 310, 379.
Matthew’s Friends, 386, 388–389, 392f (see also see also ketone supplementation
Matthew’s Friends) benefits, 377, 377f
metabolic syndrome, 392 β-hydroxybutyrate, 377, 379
404

404 Index

ketosis, starvation, 377 Maurois, P., 174

kidney stones, 68 McDonald, T. S., 339
Kilaru, S., 47 medium chain fatty acids, 329f
Kim, D. Y., 190 AMPA receptor agonists, 331–332, 331t
Kimura, R., 188 seizure control, 328–330, 329f, 329t
kindling, 355 medium chain triglyceride (MCT) ketogenic diet,
Klein, P., 19 5–7, 6b, 6f
Klement, R. J., 94 history, 328
Klingenberg, M., 258 initiation and follow-up, 12–13
Kossoff, E. H., 7, 8b, 20, 30–31, 57, 389 mechanisms, 328–333
Krebs, H. A, 258–261 AMPA receptor agonism, 331–332, 331t
Kreb’s cycle, 244 implications, 332–333
Kverneland, M., 58 seizure control, 328–330, 329f, 329t
seizure control, ketones, 332
laboratory, 165–166 molecular mechanism, 328–333
lactate dehydrogenase (LDH), 281–286 medium chain triglycerides (MCTs), 5
astrocyte-neuron lactate shuttle, 282–283, 283f coconut oil and palm kernel oil, 123
epilepsy ketogenesis, dose-response, 6f, 122–123, 123f
mechanisms, 281–282 products, 123–124, 125t
metabolic target, 283–286, 284f, 285f safety, 126
future directions, 286 supplementation, 310–311
lactate fermentation, 80 supplementation, with βHB
Lambrechts, D. A., 20 mineral salts, 311–312
Lardy, Henry, 314 MELAS, status epilepticus, 61
Laux, L., 45 menstrual cycle, 23
Lemmon, M. E., 44 metabolic acidosis, 67
Lennox Gastaut syndrome, 44–45 metabolic diseases
D-leucine, 347–348 autism spectrum disorders, 102–103, 104t–105t
L-leucine, 346–347 cancer, 79–80
leukocyte adhesion molecules, metabolic potential energy, 376
blood-brain barrier, 293 metabolic remodeling, 88
lipids, 23 metabolic syndrome
lipogenesis, 379 keto-adaptation, 378
lipoprotein profile, keto-adaptation, 379–380 ketogenic therapy, 392
liquid/formula vs. food, classic ketogenic diet, 29 plasma markers, ketogenic diet, 229, 230t
Liu, Christiana, 5 metabolism. see also specific types
Lomax, M. E., 265–266 ketogenic diet mechanisms, 227
long chain fatty acid oxidation disorders, ketone supplementation, 314, 315f
triheptanoin, 341 metabolism, mitochondrial
long-chain triglycerides, 5 acetoacetate, 315f
low glycemic index treatment (LGIT), 6f, 9–10, 9b D-β-hydroxybutyrate, 315f
development, 389 D-methionine, seizure onset protection, 348
initiation and follow-up, 12–13 middle cerebral artery occlusion, triheptanoin, 341
Luna-Medina, R., 175 migraine, 159–160
Lying-Tunell, U., 114, 115, 115t, 121f, 122 mild cognitive impairment
ketogenic treatments, 127, 127t
Magee, B. A., 320 medium chain triglycerides, 123f, 124, 125t
Magrath, G., 9 mitochondrial dysfunction
Mantis, J. G., 106 autism spectrum disorders, 102–103
Marie, A., 16 brain energy status and neuroprotection, 125–126
Marin-Valencia, I., 339 cancer, 79–80
Martinez, C. C., 17 cytopathies, 27
Matthew’s Friends, 27, 392f, 393f function, 102
achievements, 388–389 ketone supplementation, 316
future, 394–395 membrane potential, inflammation, 153
genesis, 387–388 nervous system, 102
ketogenic diet therapy spectrum, mitochondrial metabolism
389–392, 390t acetoacetate, 315f
research support, 393–394 D-β-hydroxybutyrate, 315f
405

Index 405

modified Atkins diet (MAD), 6f, 7–9, 8b, 8t neuroprotection, mechanisms, 216, 221–222, 221f
development, 389 anaplerotic, 136
effectiveness, vs. classical ketogenic diet, 10–11, 10t anti-excitotoxic, 137
efficacy, 19–22, 20f, 21t antioxidant, 137–138
initiation and follow-up, 12–13 inflammation, 138–140
mood, 22 non-anaplerotic, 137
mood disorders, 159 neurovascular unit, 289, 290f
Moorman, M. A., 256, 257 Newport, M. T., 318, 320
Moreno, S., 170 NF E2-related factor 2(Nrf2) transcription factor, 138,
multiple sclerosis, 148–150 172, 221, 266
myoclonic astatic epilepsy (MAE), 35, 46–47 Nielsen, J. V., 370
Nizamuddin, J., 69
Nabbout, R., 45, 60, 61, 63 NLRP3 inflammasome, 138–139, 148, 149t, 150, 151f,
NADPH, 264 153, 197, 298, 315–316
NADP-linked enzymes, liver activity, 259, 259t NOD-like receptor P3 (NLRP3) inflammasome, 138–
NADP system, 261–263, 262f, 263f 139, 148, 149t, 150, 151f, 153, 197, 298, 315–316
D-β-hydroxybutyrate on, 264 nonalcohoic fatty liver disease, insulin resistance, 378
Neal, E. G., 3, 4, 10, 10t nonconvulsive status epilepticus, 61–62
Nebeling, L. C., 83, 93 nonsurgical epilepsy, refractory, 40–42
Nei, M., 20 Nordli, D. R., 43
neurodegeneration and aging, 216–222. see also aging NORSE, status epilepticus, 60, 61
and neurodegeneration Nrf2, 138, 172, 221, 266
neuroinflammation, 297–298 nuclear factor-kappa B, 90
neurological disorders, 156–160. see also Numis, A. L., 43–44
specific types nutritional ketosis. see ketosis, nutritional
amino acid treatment, 346–349 (see also amino acids, Nylen, K., 106
for neurological disorders)
amyotrophic lateral sclerosis, 157–158 obesity, 362–371
glycolytic flux, 157 carbohydrate restrictions, randomized controlled
ketone bodies, 156–157 trials, 363, 363t
migraine, 159–160 diabetes mellitus type 2, 362 (see also diabetes mellitus
mood disorders, 159 type 2)
neuroprotection, 156–157 ketogenic diet
Parkinson disease, 158–159 origins and use, 363
neuron restrictive silencing factor (NRSF), 356 pilot study, 367–370
neurons, BAD-altered, metabolic changes, 272 prevalence, 362
neuropathic pain, 196. see also pain occludin, 292–293
neuroprotection, 156–157, 298, 376 Ogawa, M., 115, 121f, 122
acetoacetate, 136 Oguni, H., 46
aging, 218–219, 219f Olovnik, A. M., 256
aging, ketosis, 221–222, 221f oncogenic paradox, 80
antioxidants, 137–138 orexigenic food intake, 229, 229f
β-hydroxybutyrate, 124, 124f, 125t, 134 osteopenia, 70
D-β-hydroxybutyrate, 134 overweight, prevalence, 362
cardiac arrest and resuscitation, 218–219, 219f Owen, O. E., 114, 118, 121f, 376
erythropoietin, 220 oxidative phosphorylation, 80
fasting, 134 oxidative stress
glycolytic flux, 157 on glucose metabolism, 217–218, 218f
HIF1α, 219–222, 221f ketone supplementation, 316
inflammation, 138–140 oxygen free radicals. see also reactive oxygen
ketone bodies, 156–157, 298 species (ROS)
ketones, 126 redox state, 261, 262t
neurological disorders, 156–157
Parkinson’s disease, 158–159 pain, 196–201
polyunsaturated fatty acids, 156–157 adenosine transmission, 201
PPARγ, post-SE and during SRS, 171, 172f calorie restriction, 197–198
spinal cord injury, drug candidates, 133–134 economic cost, 196
stroke, 218–219, 219f fasting, 196–197
vascular endothelial growth factor, 220–221 glucose, 200–201
406
406 Index
pain, (cont.)
GPR40, 199, 200
GPR41, 201
GPR120, 201
inflammation, 147, 148, 196
inflammatory, 196
ketogenic diet, 198–199, 198f, 199f
ketone bodies, 200
neuropathic, 196
polyunsaturated fatty acids, 200
quality of life, 196
severe, treatment, 196
pancreatitis, 69
Paoli, A., 228–229, 228t, 229f, 236–237
Parent’s Guide to the Ketogenic Diet, 393
Parkinson’s disease
on inflammation, animal models, 150
neuroprotection, 158–159
Pasteur effect, 80
Patel, S. P., 136
pentose phosphate shunt (PPP), 354f, 357
pentylenetetrazole (PTZ), 317, 346
pericytes, blood-brain barrier, 290f, 292
peroxisome proliferator-activated receptors (PPARs)
historical overview, 167
isoforms and domains, 168, 169f
polyunsaturated fatty acids, 200
structure and functional diversity, 167–168
peroxisome proliferator-activated receptor γ (PPARγ), 7,
137, 151, 151f, 152, 157, 167–179
activation, 168
agonists, 174
neuroprotection post-SE and during SRS, 175
pretreatment, acute seizure models, 174
on seizure thresholds and kindling models, 174–175
on SRS and spatial learning, 175
brain, 170–172, 172f
epilepsy, 173–174
functional consequences, 177–178
gene regulation and activity, 169–170, 169f
ketogenic diet
activation, evidence, 175–176
activation, potential mechanisms, 176–177
regulation, 173–177
mechanisms of action, 178–179, 179f
neuroprotection, homeostatic, 171, 172f
splice variants and activity, 168–169, 169f
structure and domains, 168, 169f
Pfeifer, Heidi, 389
Phinney, S. D., 379, 382
PI3K/AKT signaling pathway, 89
Pires, M. E., 43
plasmalemma vesicle-associated protein number, 293
polycystic ovary syndrome, 117
polyunsaturated fatty acids (PUFAs)
on inflammation, 151f, 152
neuroprotection, 156–157
pain, 200
peroxisome proliferator-activated receptors, 200
pregnancy
establishment and maintenance, 233–234
before and during gestation, 233f, 234–236, 235f,
235t, 236f
Pre-ketogenic diet, 394
presenilin 1/2 genes, 241
Prins, M. L., 298–299
D-proline, seizure onset protection, 348
prolyl-hydroxylase (PHD), 220, 221, 221f
propionyl-CoA carboxylation pathway, anaplerosis,
336–337, 338f
puberty, ketogenic diet long-term effects, 231–234, 232f,
232t, 233f
pyruvate decarboxylase deficiency, triheptanoin, 341
pyruvate dehydrogenase
deficiency, 50
inhibition, Alzheimer’s disease, 242, 243
QT interval prolongation, 69
radiation, ionizing
HDAC inhibitor protection, 264
ketone ester effects, post-exposure, 266–267, 267f
ketosis protection, 263, 264f
in vitro studies, 264–266, 265f, 266f
radiation therapy, cancer, 92
Raju, K. N., 30
Rasmussen syndrome, status epilepticus, 61
ratios, importance, 30–31
reactive nitrogen species (RNS), 92, 254–268
damage caused by, 255–256, 255f, 256f
origins, reactions, and target molecules, 254–255,
255f, 256t
reactive oxygen species (ROS), 254–268
in aging, 256–257
Alzheimer’s disease, 243–244
D-β-hydroxybutyrate
on cellular redox states, 261–263, 262f, 263f
on transcription, 261t, 264
cancer, 90
cellular redox states vs. standard redox potentials,
258–261, 259f, 259t, 261t, 262t
damage, 255–256, 255f, 256f
enzymes destroying, 264
on inflammation, 151f, 153
mechanisms of action, 257–258, 258f
origins, reactions, and target molecules, 254–255,
255f, 256t
redox state of oxygen free radicals and other reactants,
261, 262t
removal, in vivo, 263–264
removal, NADPH generation, 264
redox potentials, standard, vs. cellular redox states,
258–261, 259f, 259t, 261t, 262t
redox states, cellular
D-β-hydroxybutyrate on, 261–263, 262f, 263f
oxygen free radicals and other reactants, 261, 262t
vs. standard redox potentials, 258–261, 259f, 259t,
261t, 262t
407

Index 407

refractory epilepsy, 18 growth, 70

refractory nonsurgical epilepsy, 40–42 hepatitis, 69
renal calculi, 68 hyperlipidemia, 69
reproduction, 227–237 hypoglycemia, 68
behavioral, 231 infections, 68
endocrine, 227–230, 228t, 229f, 230f, 230t kidney stones, 68
high-fat/ketogenic diets, 227 osteopenia, 70
long-term effects, 233–234 pancreatitis, 69
pregnancy vitamin deficiencies, 68–69
establishment and maintenance, 233–234 vomiting, 67
before and during gestation, 233f, 234–236, 235f, Simard-Tremblay, S., 47
235t, 236f Slc2a1, 293
puberty, 231–233, 232f, 232t, 233f social screening, 66–67
reproduction, 233–234 somatic mutation theory, 79
response predictor, 31 Spilioti, M., 105
resting energy expenditure (REE), 29 spinal cord injury, 133–141
Rett syndrome, 50–51, 106 epidemiology, 133
triheptanoin, 342 inflammation, 138–140
Richardson, R., 364 intermittent fasting, 134
Rouach, N., 282–283 ketogenic diet, mechanisms
Ruskin, D. N., 106 anaplerotic, 136
anti-excitotoxic, 137
Sada, N., 283–286, 284f, 285f antioxidant, 137–138
S-adenosylhomocysteine (SAH) hydrolase, 212, 212f non-anaplerotic, 137
S-adenosylmethionine (SAM) dependent ketogenic diet, outcome, 134–135
transmethylation pathway, 211–212, 212f ketogenic diet, translation to
Samoilova, M., 188 clinical setting, 140–141
Sansum, W. D., 364 management, acute, 133
Sarruf, D. A., 170 neuroprotective drug candidates, 133–134
Scheck, A. C., 83, 91–92, 92f rehabilitation, 133
Schmid, W., 267 spontaneous recurrent seizures, PPARγ
Schoeler, N. E., 41 agonists on, 175
Schrödinger, Erwin, 257 sport, keto-adaptation, 380–381
SCN1A gene mutations, 45 Stafstrom, C. E., 192
screening, pre-treatment, 66–67, 67t, 71t status epilepticus (SE), 60–63
medical, 66 cause, 60
social, 66–67 challenges, 61t, 62–63
seizure. see also epilepsy convulsive, 61
fasting, 209 duration, 60
freedom from, 30 efficacy, ketogenic diet, 60, 61t
ketone supplementation, 317 implementation, 62, 62f
seizure control. see also epileptogenesis actions incidence, 60
ketones, 332 nonconvulsive, 61–62
medium chain fatty acids, 328–330, 329f, 329t super-refractory, 60
senescence, cell, 257 treatment, first-line, 60
D-serine, seizure onset protection, 348 types, 60
L-serine stroke
seizure onset protection, 347 neuroprotection, 218–219, 219f
traumatic brain injury, 349 triheptanoin, 341
severe myoclonic epilepsy of infancy (SMEI), 45–46 Sturge-Weber syndrome, 53
Seyfried, T. N., 80 succinic semialdehyde dehydrogenase (SSADH)
Sharma, S., 3, 10, 10t deficiency, 105–106
Shimazu, T., 81 sugar, dietary, children, 393
side effects, 66–71 superoxide dismutase (MnSOD), 138, 264
acidosis, 67 oxidative stress, 172f
cardiac, 69–70 PPARγ deficiency on, 172
constipation, 67–68 reactive oxygen species, 254–255, 255f, 256t, 264
electrolyte imbalances, 69 super-refractory status epilepticus (SRSE), 18–19, 60
food refusal, 70 Sussman, D., 234–235
408

408 Index

Tanner, G. R., 190 neurological conditions, 341–342

Taub, K. S., 30 pyruvate decarboxylase deficiency, 341
TCA cycle, 336, 337f, 338f tuberous sclerosis complex, 51–52
team-based approach tumor metabolism, 88–90, 89f
family-centered, 26
team composition, 26–27 uncoupling proteins, on inflammation, 153
telomere shortening, 256–257 unlimited protein diet. see modified Atkins diet (MAD)
temporal lobe epilepsy
hippocampus model, 186–187 vascular endothelial growth factor (VEGF),
ketogenic diet and adenosine homeostasis, neuroprotection, 220–221
211–213, 212f Veech, R. L., 81, 258–260, 259t, 307, 314, 318
Thakur, K. T., 19 Verdin, ERic, 316
therapeutic ketosis, 310. see also vesicular glutamate transporters (VGLUTs), 190,
ketone supplementation 193, 286
Thio, L. L., 188 acetoacetate on, 281
thioredoxin, 259, 259f, 261 functions, 281
tight junctions, blood-brain ketogenic diet, 282
barrier, 290f, 292–293 Viggiano, A., 317
tocopherol, 258, 258f Villeneuve, N., 61
transcription, D-β-hydroxybutyrate on, 261t, 264 Vining, E., 70
transmethylation pathway, 211, 212f vitamin deficiencies, 68–69
transporters. see also specific types Volek, J. S., 378, 379–380, 382, 383f
blood-brain barrier, 290f, 291f, 293, 296f vomiting, 67
traumatic brain injury, 133–141, 298
amino acid treatment, 349 Warburg, Otto, 88
on inflammation, 138–140 Warburg effect, 80, 82, 88–90, 89f
ketogenic diet, translation to Warburg’s theory, 79–80
clinical setting, 140–141 weight loss, 22
ketogenic diet effects, 135–136 keto-adaptation, 378–379
ketogenic diet mechanisms ketone supplementation, 321–322
anaplerotic, 136 Westman, E. C., 367–370, 370t
anti-excitotoxic, 137 Wilder, R. M., 16, 26, 165
antioxidant, 137–138 Williams, Matthew, 386, 387–388, 393f
non-anaplerotic, 137 Williams, S., 70
triglycerides, 5 Williams-Karnesky, R. L., 210, 211–212
triheptanoin, 317–318, 336–342 Willis, S., 340
anaplerosis, 336–337, 337f, 338f Winocur, G., 231
anticonvulsant effects, 338f, 340 Wirrell, E. C., 28
brain energy metabolism, 339–340 Worden, L. T., 71
cardiac hypertrophy, 341
epilepsy, energy metabolism, Yudkin, John, 363
337–339, 338f
GLUT1 deficiency, 341 zona occludin proteins, 293
long chain fatty acid oxidation disorders, 341 Zuccoli, G., 93
muscle disorders, 342 Zupec-Kania, Beth, 388, 393