Medical Hypotheses (2004) 62, 970975

http://intl.elsevierhealth.com/journals/mehy

Is autism a disorder of fatty acid metabolism? Possible dysfunction of mitochondrial b-oxidation by long chain acyl-CoA dehydrogenase
Tonya Clark-Taylor*, Benjamin E. Clark-Taylor
P.O. Box 1524, Woden, 2606 Canberra, Australia
Received 19 January 2004; accepted 26 January 2004

Summary Long chain acyl-CoA dehydrogenase (LCAD) has recently been shown to be the mitochondrial enzyme responsible for the b-oxidation of branched chain and unsaturated fatty acids [Biochim. Biophys. Acta 1393 (1998) 35; Biochim. Biophys. Acta 1485 (2000) 121]. Whilst disorders of short, medium and very long chain acyl dehydrogenases are known, there is no known disorder of LCAD deciency in humans. Experimental LCAD deciency in mice shows an acyl-carnitine prole with prominent elevations of unsaturated fatty acid metabolites C14:1 and C14:2 [Hum. Mol. Genet. 10 (2001) 2069]. A child with autism whose acyl-carnitine prole also shows these abnormalities is presented, and it is hypothesized that the child may have LCAD deciency. Additional metabolic abnormalities seen in this patient include alterations of TCA energy production, ammonia detoxication, reduced synthesis of omega-3 DHA, and abnormal cholesterol metabolism. These metabolic changes are also seen as secondary abnormalities in dysfunction of fatty acid b-oxidation, and have also been reported in autism. It is hypothesized that LCAD deciency may be a cause of autism. Similarities between metabolic disturbances in autism, and those of disorders of fatty acid b-oxidation are discussed. c 2004 Elsevier Ltd. All rights reserved.

Introduction
Disorders of fatty acid b-oxidation are a group of inherited diseases that may either be caused by failure of a single mitochondrial or peroxisomal enzyme of b-oxidation, such as deciency of short chain acyl-CoA dehydrogenase (SCAD), medium chain acyl-CoA dehydrogenase (MCAD), very long chain acyl-CoA dehydrogenase (VLCAD), long
Corresponding author. Tel.: +61-2-62847096; fax: +61262945231. E-mail address: tbct@bigpond.net.au (T. Clark-Taylor).
*

chain 3-hydroxyacyl-CoA dehydrogenase (LCHAD) or acyl-CoA oxidase, or be secondary to dysfunction of dependent processes, such as deciencies of the carnitine fatty acid transporter system, mitochondrial electron transfer avoprotein system (collectively referred to as multiple acyl-CoA dehydrogenase deciency or MADD), peroxisomal fatty acid transporter, or peroxisomal biogenesis disorders. There is considerable variation in clinical presentation and severity both between and within each disorder, from life threatening to asymptomatic, presumed to be due in part to individual variation in residual enzyme activity, to the over-

0306-9877/$ - see front matter c 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.mehy.2004.01.011

Is autism a disorder of fatty acid metabolism? lap of acyl substrate specicity between the different acyl dehydrogenases and to the metabolic demand for pathways of oxidation (for review see [4,5]). Mitochondrial fatty acid oxidation (FAO) deciencies usually present in the neonate or toddler with any of hypoketotic hypoglycaemia, metabolic acidosis, mitochondrial dysfunction, hyperammonaemia, hyperuricosuria, dicarboxylic aciduria, hypotonia, muscle weakness, cardiomyopathy, seizures, failure to thrive, psychomotor delay, developmental regression, behavioural disorders and attention decit disorder [36]. Neonatal presentations are usually severe with poor prognosis and include cardiac arrythmia and sudden death [7], however in mild phenotypes there may be an initial period of normal development and function before decompensation in association with metabolic stress or immune activation, such as fasting, illness or vaccination [8]. With the exception of cardiac involvement and sudden death, all of the metabolic and developmental abnormalities listed above may occur in autism, and onset of autism may also be regressive following a period of initially normal infant development. Human mitochondrial long chain acyl-CoA dehydrogenase (LCAD) was initially thought to be the mitochondrial enzyme responsible for the rst step in the b-oxidation of long chain saturated fatty acids, particularly palmitate (C16:0), which is a major substrate for cardiac and skeletal muscle and source of energy during depletion of glucose and glycogen stores. Patients with abnormalities of palmitate b-oxidation who were initially presumed to have LCAD deciency, were in 1993 found to actually have a deciency of VLCAD [4]. Recently, human LCAD was demonstrated to be important in mitochondrial b-oxidation of branched [1] and unsaturated fatty acids [2]. Despite known disorders of the other mitochondrial acyl dehydrogenases SCAD, MCAD and VLCAD, to date there is no known disorder of LCAD dysfunction and it is hypothesized that this may be because LCAD deciency may present differently from other disorders of b-oxidation [3]. Whilst there is substrate overlap between the various mitochondrial acyl dehydrogenases, and LCAD does have some activity towards long saturated chains [2] and may provide a secondary route of energy production from palmitate, given its primary role in the metabolism of unsaturated and branched chains in theory LCAD deciency could be less likely to present with classical symptoms of muscle weakness and cardiomyopathy. Instead, degradation of mono- and polyunsaturates from membrane phospholipid trafcking and lipid signalling in tissues such as

971 neurones, glia, immune and endocrine cells and degradation of dietary unsaturates, such as by hepatic and gastrointestinal tissues, might possibly dominate symptoms of LCAD dysfunction. In autism, dysfunction of the CNS, immune [9] and gastrointestinal [10] systems is common. In this paper, we present a child with autism who has mild abnormalities of acyl-carnitine prole and mitochondrial dysfunction that are not consistent with any known disorders of b-oxidation, electron transport chain or urea cycle dysfunction, or other inborn errors of metabolism. We propose that this clinical picture might be caused by an abnormality of mitochondrial LCAD, and discuss similarities between metabolic disturbances in autism, and those of disorders of fatty acid b-oxidation.

Case report
The patient (AB) is an 8-year-old male with Autism Spectrum Disorder, eczema, and multiple food allergies demonstrated on RAST. Pregnancy was complicated by gestational diabetes that was well controlled with diet. Birth and early infantile development was unremarkable. Developmental regression was rst noted by the parents at around 18 months. The patient has hypotonia but normal muscle mass and strength and also has chronic constipation. There is a family history of type II diabetes in the mother and maternal grandmother and type I diabetes in the paternal grandmother. Investigations repeatedly show elevated serum lactate or pyruvate, low bicarbonate, and various elevations of TCA cycle intermediates citrate, isocitrate, a-ketoglutarate or succinate. There was no abnormality of the pyruvate dehydrogenase complex, no biotinidase deciency and normal mitochondrial electron transport complexes were demonstrated on muscle biopsy. Plasma fatty acids showed low levels of the essential fatty acids C22:6 n 3 (DHA) and C20:5 n 3 (EPA) with normal levels of their omega-3 precursor C18:3, and normal levels of omega-6 fatty acids. Initial plasma acyl-carnitine prole showed mildly increased esters of unsaturates C14:2 and C14:1 along with high normal C10:1. Repeat acyl-carnitine prole while on supplemental carnitine showed general reduction of these elevated levels, with a broadening of involved species, showing high normal C10:1, C12:1, C12:0, C14:2 and C18:2. Urinary adipate and subarate were intermittently high normal. Ketones were within normal range, however the b-hydroxybutyrate:acetoacetate ratio was high. Glucose

972 levels were repeatedly normal. Cholesterol metabolism showed a tendency to be high, with an elevation of urinary 3-methylglutaconate. Various abnormalities of plasma amino acids on repeat testing included low or high tryptophan, glutamate, glutamine, aspartate and asparagine with mildly elevated ammonia. Repeatedly low uric acid with a normal purine and pyrimidine prole were also present. Arginine was low, however other urea cycle intermediates citrulline, ornithine and argininosuccinate, as well as orotate were normal.

Clark-Taylor, Clark-Taylor Disorders of FAO may also produce secondary abnormalities of mitochondrial function, including altered energy production [14,15], impaired ammonia detoxication, and secondary carnitine and DHA deciency [16]. This might explain the patients abnormalities of TCA cycle intermediates, ammonia, carnitine and DHA levels. Disorders of long chain FAO also impair pyruvate metabolism [14], which is consistent with this childs abnormalities of pyruvate, lactate and bicarbonate. Disorders of long chain FAO are associated with increased triglyceride storage, and experimental inhibition of mitochondrial long chain oxidation results in increased triglyceride and cholesterol synthesis [17]. The patients tendency to mildly elevated cholesterol may be secondary to this mechanism.

Interpretation
The patient shows an abnormality of acyl-carnitine prole, abnormal ammonia detoxication and altered mitochondrial energy production together with hypotonia and possible intermittent dicarboxylic aciduria. These ndings are suggestive of a FAO decit, however the patients acyl-carnitine prole is not consistent with any known pattern of abnormality. The acyl-carnitine proles in this patient show an abnormality that is mainly of unsaturated species, particularly C14:2 and C14:1. Elevated levels of the long chain unsaturate C14:1 may be seen in VLCAD deciency, minor elevations occur in LCHAD and MAD deciency [11], and elevated C10:1 may be seen in MCAD deciency [12]. However acylcarnitine proles in each of these disorders typically do not show a prominent elevation of C14:2. Instead they usually show other associated abnormalities, particularly of saturated species, which are not seen in this patient. While the other acyl dehydrogenases predominate in the metabolism of saturates, recent work shows that LCAD has a primary role in mitochondrial metabolism of unsaturated fatty acid metabolites with C14:1 (n 9), C14:2 (n 6) and C14:3 (n 3) among the preferred substrates for LCAD [2]. In a prole very similar to that found in this patient, the unsaturates C14:2 and C14:1 were demonstrated to be the most prominently elevated acyl-carnitines in experimental LCAD decient mice, with additional minor increases of C10 and C12 species [3]. In humans, experimental LCAD decient broblast cultures have been shown to also produce elevated unsaturates C14:2 and C16:3 from the b-oxidation of arachidonic acid (C20:4) [13]. Since C14:2 and C14:1 were the prominently elevated species in this patient, with other minor abnormalities of C10 and C12 species, LCAD deciency is a theoretical possibility, although its deciency has never been documented in humans.

Discussion
Function of LCAD
Mitochondrial LCAD has a broad specicity, with most activity towards medium and long chain C10 to C18 species [4]. As mono- and polyunsaturated fatty acids such as oleate (C18:1) and linoleate (C18:2) are b-oxidized and repeatedly chain shortened, the double bond reaches the 5,6 (or 4,5) position producing C14:1 and C14:2, respectively, at which point LCAD is the acyl dehydrogenase responsible for initiating the next round of b-oxidation in the mitochondria. Consistent with this role in the metabolism of unsaturates, accumulation of acyl-carnitines of C14:1 and C14:2 has been demonstrated in LCAD decient mice [3]. Both the mitochondria and the peroxisome are active in b-oxidation of unsaturated and branched chain fatty acids, with separate enzyme systems in each organelle. Where substrate overlap occurs, control of partitioning between the mitochondrial and peroxisomal systems is not well understood, however when mitochondrial b-oxidation is inhibited in rats, polyunsaturated fatty acids have been demonstrated to be diverted to the peroxisome for metabolism [18]. Unfortunately, peroxisomes lack the mitochondrial electron transport chain and the FADH2 generated by b-oxidation in peroxisomes is not available for energy production. Instead, H2 O2 is produced and requires adequate antioxidant capacity of catalase for its safe disposal [19]. These differences make peroxisomal b-oxidation less efcient than the mitochondrial pathway and may possibly create an oxidant stress under load. There is evidence of oxidant stress in autism [20,21] and

Is autism a disorder of fatty acid metabolism? some autistic children appear to respond to antioxidant support such as glutathione, vitamins A, C and E, and alpha-lipoic acid. In contrast to energy production from saturated fats, a major role of unsaturated fatty acids is in membrane phospholipids, prostaglandins, leukotrienes and in cellular lipid signalling mechanisms. While oxidation of fatty acids for energy production is largely intermittent, and some FAO disorders may have an episodic course, degradation of unsaturated fatty acids used in membrane trafcking and lipid signalling is likely to be a constant process, with constant demand for LCAD, producing a more constant symptomatology in a hypothetical deciency state. While cardiomyopathy and muscle weakness is prominent in other disorders of FAO such as VLCAD [4], human LCAD has little activity in the heart [22] and only 1/10th the activity towards palmitoyl-CoA as VLCAD [23] which may explain why these classical symptoms are not common in autism. Instead, autistic symptoms include CNS, immune and gastrointestinal, which are tissues that may have increased physiologic turnover of unsaturated fatty acids.

973 might explain the reported increase in the rate of autism over the last 30 years [30].

Metabolic abnormalities in disorders of fatty acid oxidation that are also seen in autism
It is now known that far from being merely an energy source, long chain fatty acids have important roles in the regulation of many cellular functions, especially in the control of metabolism and cellular energy production. As well as their effects on cell membrane uidity, unsaturated fatty acids have nuclear hormone effects, controlling gene expression by peroxisome proliferator activator receptors (PPAR), liver X receptors (LXR), sterol response element binding protein (SREBP) and hepatic nuclear factor 4a (HNF4a) and control many cellular functions including glucose and fatty acid metabolism, insulin sensitivity, cholesterol metabolism and immune function [31]. Long chain fatty acids or their CoA or carnitine esters inhibit many enzymes including those involved in fatty acid synthesis (acetyl CoA carboxylase, fatty acid synthase and citrate ligase [31]), TCA cycle functioning (pyruvate dehydrogenase, citrate synthase [32], aketoglutarate dehydrogenase [33]), glutamate metabolism (glutamate dehydrogenase [31]), and oxidative phosphorylation [15,34]. Accumulation of intermediates of FAO may be directly toxic via any of these mechanisms, or may induce secondary effects by sequestration of coenzyme A, carnitine or fatty acid binding proteins, resulting in alterations of cellular metabolism and energy production. There is evidence of disturbed energy production in some autistic children with altered pyruvate, lactate and metabolic acidosis [35,36], and decreased brain high-energy phosphates and phosphocreatine that correlates with increased autistic decits [37]. Lombard [38] has proposed that autism is a disorder of mitochondrial dysfunction noting that carnitine deciency and lactic acidosis are frequent ndings. Decreased levels of the essential fatty acid DHA with normal levels of its omega-3 precursor C18:3, and normal omega-6 series have also been reported in autism [39]. Carnitine deciency in combination with decreased DHA is a marker for disorders of FAO [16] and mild to moderate increases in lactate are seen in long chain FAO disorders [14]. Inhibition of mitochondrial FAO also causes an increase in triglyceride and cholesterol synthesis [17] and elevations of the activity of the rate limiting enzyme for cholesterol synthesis, HMG-CoA reductase, are known to occur in autism [40]. In

Possible causes of LCAD deciency

Possible causes of dysfunction of LCAD might include primary enzyme deciency caused by genetic mutation of ACADL. The subtelomeric region of chromosome 2q has been implicated in autism [24] and the location of ACADL at 2q3435 [25] is close to an autism genetic susceptibility locus of 2q2133 [26]. Other possible causes of enzyme dysfunction include inhibition of LCAD by poorly oxidizable substrates, such as dietary conjugated linoleic acids [27] or trans fatty acids. Increased demand for LCAD by high substrate concentrations, as in abnormal turnover of unsaturated fatty acids or abnormal enrichment of the fatty acid pool, might stress systems that have suboptimal LCAD activity. The widespread availability of, and preference for, commercial vegetable oils in the last century has enriched the diet with polyunsaturated fatty acids. Prior to this, there may have been little genetic pressure to select for polymorphisms encoding a high enzymatic activity. Accumulated dietary polyunsaturated fatty acids in maternal tissue may cause secondary enrichment in fetal tissues, since preferential transport of maternal polyunsaturated fatty acids occurs across the placenta [28]. This may be consistent with the increased rate of autism in rstborn or lateborn children [29], as the maternal exposure is maximal at these times. Such a model

974 SmithLemiOpliz Syndrome (SLOS), an inborn error of cholesterol synthesis associated with autism, treatment of the cholesterol decit results in improvement of autistic behaviour [41].

Clark-Taylor, Clark-Taylor order of FAO, and may be the rst documented case of LCAD deciency. This childs general metabolic prole is representative of others in the autistic community and it is hypothesized that LCAD deciency may be a cause of autism. Further work is required to determine if acyl-carnitine proles showing abnormalities of unsaturated or branched chain fatty acid intermediates are found in other children with autism, and if such a prole may be caused by LCAD dysfunction.

Altered ammonia metabolism in fatty acid oxidation disorders and in autism

Ammonia detoxication is disturbed in disorders of FAO, and ammonia and fatty acids are synergistic neurotoxins [33]. Disorders of the urea cycle also result in elevations of serum ammonia and are associated with learning and behavioural abnormalities, speech delay, ADHD and seizures [42]. This patient had mildly elevated serum ammonia, as did a case reported by Cohen [43] of a child whose functioning was inversely proportional to elevations of serum ammonia and GABA levels. Ammonia alters glutamate neurotransmission, receptors and uptake [44,45]. Disturbance of the glutamate receptor system has been demonstrated in postmortem autistic brain [46] and Carlsson [47] has proposed a hypoglutamate theory of autism. Glutamate and a-ketoglutarate levels are controlled by glutamate dehydrogenase, which is pivotal in balancing the needs for anaplerosis (lling up) of the TCA cycle, glutamate and GABA neurotransmitter production, ammonia detoxication and synthesis of the antioxidant glutathione. This enzyme is part of the brain glutamate/GABA/glutamine cycle and it is subject to many allosteric controls, including control by long chain fatty acyl-CoA esters [32]. Detoxication of ammonia by non-hepatic tissues such as brain is maintained by amination of the TCA cycle intermediates a-ketoglutarate and oxaloacetate, producing glutamate, glutamine, aspartate and asparagine which are then transported to the liver for disposal of the ammonia into the urea cycle. Disturbance of these amino acids is common in autism [35,48] as are minor alteration of hepatic transaminases and this may represent altered ammonia detoxication. Treatment with vitamin B6, the cofactor for these transaminations, has been associated with improvements in function in autism [35].

References
[1] Wanders RJ, Denis S, Ruiter JP, IJlst L, Dacremont G. 2,6Dimethylheptanoyl-CoA is a specic substrate for longchain acyl-CoA dehydrogenase (LCAD): evidence for a major role of LCAD in branched-chain fatty acid oxidation. Biochim Biophys Acta 1998;1393:3540. [2] Le W, Abbas AS, Sprecher H, Vockley J, Schulz H. Longchain acyl-CoA dehydrogenase is a key enzyme in the mitochondrial beta-oxidation of unsaturated fatty acids. Biochim Biophys Acta 2000;1485:1218. [3] Cox KB, Hamm DA, Millington DS, et al. Gestational, pathologic, and biochemical differences between very long-chain acyl-CoA dehydrogenase deciency and longchain acyl-CoA dehydrogenase deciency in the mouse. Hum Mol Genet 2001;10:206977. [4] Wanders RJ, Vreken P, den Boer ME, Wijburg FA, van Gennip AH, IJlst L. Disorders of mitochondrial fatty acyl-CoA beta-oxidation. J Inherit Metab Dis 1999;22: 44287. [5] Roe CR, Ding JH. Mitochondrial fatty acid oxidation disorders. In: Scriver CR, Beaudet AL, Sly WS, Valle D, editors. The metabolic and molecular basis of inherited disease. eighth ed. New York: McGraw-Hill; 2001. p. 2297326. [6] Wang SS, Fernhoff PM, Hannon WH, Khoury MJ. Medium chain acyl-CoA dehydrogenase deciency human genome epidemiology review. Genet Med 1999;1:3329. [7] Bonnet D, Martin D, De Lonlay Pascale, et al. Arrhythmias and conduction defects as presenting symptoms of fatty acid oxidation disorders in children. Circulation 1999;100: 224853. [8] Ziadeh R, Hoffman EP, Finegold DN, et al. Medium chain acyl-CoA dehydrogenase deciency in Pennsylvania: neonatal screening shows high incidence and unexpected mutation frequencies. Pediatr Res 1995;37:6758. [9] Korvatska E, Van de Water J, Anders TF, Gershwin ME. Genetic and immunologic considerations in autism. Neurobiol Dis 2002;9:10725. [10] Horvath K, Perman JA. Autistic disorder and gastrointestinal disease. Curr Opin Pediatr 2002;14:5837. [11] Divry P, Vianey-Saban C, Mathieu M. Determination of total fatty acids in plasma: cis-5-tetradecenoic acid (C14:1 omega-9) in the diagnosis of long-chain fatty acid oxidation defects. J Inherit Metab Dis 1999;22:2868. [12] Nada MA, Chace DH, Sprecher H, Roe CR. Investigation of beta-oxidation intermediates in normal and MCAD-decient human broblasts using tandem mass spectrometry. Biochem Mol Med 1995;54:5966. [13] Spector AA, Williard DE, Kaduce TL, Gordon JA. Conversion of arachidonic acid to tetradecadienoic acid by peroxisomal oxidation. Prostaglandins Leukot Essent Fatty Acids 1997; 57:1015.

Conclusion
Autism is a developmental disorder of unknown cause, which can show many metabolic abnormalities similar to FAO disorders. This paper presents the case of an autistic child who has evidence of an abnormality of b-oxidation of unsaturated fatty acids which is not consistent with any known dis-

Is autism a disorder of fatty acid metabolism?

[14] Ventura FV, Ruiter JP, IJlst L, de Almeida IT, Wanders RJ. Lactic acidosis in long-chain fatty acid beta-oxidation disorders. J Inherit Metab Dis 1998;21:64554. [15] Ventura FV, Ruiter JP, Ijlst L, Almeida IT, Wanders RJ. Inhibition of oxidative phosphorylation by palmitoyl-CoA in digitonin permeabilized broblasts: implications for longchain fatty acid beta-oxidation disorders. Biochim Biophys Acta 1995;1272:1420. [16] Infante JP, Huszagh VA. Secondary carnitine deciency and impaired docosahexaenoic (22:6 n 3) acid synthesis: a common denominator in the pathophysiology of diseases of oxidative phosphorylation and beta-oxidation. FEBS Lett 2000;468:15. [17] Yamamoto K, Fukuda N, Fukui M, Kai Y, Ikeda H, Sakai T. Increased secretion of triglyceride and cholesterol following inhibition of long-chain fatty acid oxidation in rat liver. Ann Nutr Metab 1996;40:15764. [18] Tran TN, Christophersen BO. Partitioning of polyunsaturated fatty acid oxidation between mitochondria and peroxisomes in isolated rat hepatocytes studied by HPLC separation of oxidation products. Biochim Biophys Acta 2002;1583:195204. [19] Wanders RJA, Vreken P, Ferdinandusse S, et al. Peroxisomal fatty acid a- and b-oxidation in humans: enzymology, peroxisomal metabolite transporters and peroxisomal diseases. Biochem Soc Trans 2001;29:25067. [20] Sogut S, Zoroglu SS, Ozyurt H, et al. Changes in nitric oxide levels and antioxidant enzyme activities may have a role in the pathophysiological mechanisms involved in autism. Clin Chim Acta 2003;331:1117. [21] Golse B, Debray-Ritzen P, Durosay P, Puget K, Michelson AM. [Alterations in two enzymes: superoxide dismutase and glutathion peroxidase in developmental infantile psychosis (infantile autism)]. Rev Neurol 1978;134:699705 [Abstract only]. [22] Andresen BS, Bross P, Vianey-Saban C, et al. Cloning and characterization of human very-long-chain acyl-CoA dehydrogenase cDNA, chromosomal assignment of the gene and identication in four patients of nine different mutations within the VLCAD gene. Hum Mol Genet 1996;5:46172. [23] Aoyama T, Souri M, Ushikubo S, et al. Purication of human very-long-chain acyl-coenzyme A dehydrogenase and characterization of its deciency in seven patients. J Clin Invest 1995;95:246573. [24] Wolff DJ, Clifton K, Karr C, Charles J. Pilot assessment of the subtelomeric regions of children with autism: detection of a 2q deletion. Genet Med 2002;4:104. [25] Indo Y, Yang-Feng T, Glassberg R, Tanaka K. Molecular cloning and nucleotide sequence of cDNAs encoding human long-chain acyl-CoA dehydrogenase and assignment of the location of its gene (ACADL) to chromosome 2. Genomics 1991;11:60920. [26] Bacchelli E, Blasi F, Biondolillo M, et al. Screening of nine candidate genes for autism on chromosome 2q reveals rare nonsynonymous variants in the cAMP-GEFII gene. Mol Psychiat 2003;8:91624. [27] Demizieux L, Degrace P, Gresti J, et al. Conjugated linoleic acid isomers in mitochondria: evidence for an alteration of fatty acid oxidation. J Lipid Res 2002;43:211222. [28] Haggarty P, Page K, Abramovich DR, Ashton J, Brown D. Long-chain polyunsaturated fatty acid transport across the perfused human placenta. Placenta 1997;18:63542.

975
[29] Bolton PF, Murphy M, Macdonald H, Whitlock B, Pickles A, Rutter M. Obstetric complications in autism: consequences or causes of the condition? J Am Acad Child Adolesc Psychiat 1997;36:27281. [30] Department of Developmental Services. Autistic Spectrum Disorders Changes in the California Caseload An Update: 19992002. California: California Health and Human Services Agency; 2000. [31] Jump DB. Dietary polyunsaturated fatty acids and regulation of gene transcription. Curr Opin Lipidol 2002;13:15564. [32] Lai JC, Liang BB, Jarvi EJ, Cooper AJ, Lu DR. Differential effects of fatty acyl coenzyme A derivatives on citrate synthase and glutamate dehydrogenase. Res Commun Chem Path Pharmacol 1993;82:3318. [33] Lai JC, Cooper AJ. Neurotoxicity of ammonia and fatty acids: differential inhibition of mitochondrial dehydrogenases by ammonia and fatty acyl coenzyme A derivatives. Neurochem Res 1991;16:795803. [34] Ventura FV, Ruiter JP, Ijlst L, de Almeida IT, Wanders RJ. Inhibitory effect of 3-hydroxyacyl-CoAs and other longchain fatty acid beta-oxidation intermediates on mitochondrial oxidative phosphorylation. J Inherit Metab Dis 1996;19:1614. [35] Moreno H, Borjas L, Arrieta A, et al. Clinical heterogeneity of the autistic syndrome: a study of 60 families. Invest Clin 1992;33:1331 [Abstract only]. [36] Chugani DC, Sundram BS, Behen M, Lee ML, Moore GJ. Evidence of altered energy metabolism in autistic children. Prog Neuropsychopharmacol Biol Psychiat 1999;23:63541. [37] Minshew NJ, Goldstein G, Dombrowski SM, Panchalingam K, Pettegrew JW. preliminary 31P MRS study of autism: evidence for undersynthesis and increased degradation of brain membranes. Biol Psychiatry 1993;33:76273. [38] Lombard J. Autism: a mitochondrial disorder? Med Hypotheses 1998;50:497500. [39] Vancassel S, Durand G, Barthelemy C, et al. Plasma fatty acid levels in autistic children. Prostaglandins Leukotr Essent Fatty Acids 2001;65:17. [40] Kurup RK, Kurup PA. A hypothalamic digoxin-mediated model for autism. Int J Neurosci 2003;113:153759. [41] Kelley RI. Inborn errors of cholesterol biosynthesis. Adv Pediatr 2000;47:153. [42] Berry GT, Steiner RD. Long-term management of patients with urea cycle disorders. J Pediatr 2001;138:S5660. [43] Cohen BI. Use of a GABA-transaminase agonist for treatment of infantile autism. Med Hypotheses 2002;59:1156. [44] Albrecht J. Roles of neuroactive amino acids in ammonia neurotoxicity. J Neurosci Res 1998;51:1338. [45] Monfort P, Munoz MD, El Ayadi A, Kosenko E, Felipo V. Effects of hyperammonemia and liver failure on glutamatergic neurotransmission. Metab Brain Dis 2002;17:23750. [46] Purcell AE, Jeon OH, Zimmerman AW, Blue ME, Pevsner J. Postmortem brain abnormalities of the glutamate neurotransmitter system in autism. Neurology 2001;57:161828. [47] Carlsson ML. Hypothesis: is infantile autism a hypoglutamatergic disorder? Relevance of glutamate serotonin interactions for pharmacotherapy. J Neural Transm 1998;105:52535. [48] Aldred S, Moore KM, Fitzgerald M, Waring RH. Plasma amino acid levels in children with autism and their families. J Autism Dev Disord 2003;33:937.