Toxicidad Mitocondrial

ANTIVIRAL THERAPY
Proceedings of the 1st Meeting on Mitochondrial Toxicity and HIV Infection: Understanding the Pathogenesis for a Therapeutic Approach
Lectures presented in this meeting are published as part of Antiviral Therapy Volume 10. The page number given for each paper is correct for Volume 10, Supplement 2. The correct citation for a paper is: Authors. Title. Antiviral Therapy 2005 10(Suppl 2):M Page number.
1st Meeting on Mitochondrial Toxicity & HIV Infection: Understanding the Pathogenesis for a Therapeutic Approach
ANTIVIRAL THERAPY
EDITORS-IN-CHIEF
Joep MA Lange
National AIDS Therapy Evaluation Centre Academic Medical Centre University of Amsterdam, Pietersbergweg 9 1105 BM Amsterdam, the Netherlands Tel: +31 20 314 9300 Fax: +31 20 314 9399 E-mail: j.lange@amc.uva.nl EMERGING VIRUSES AND BIODEFENCE PRE-CLINICAL HIV
Clarence J Peters
University of Texas Medical Branch 3.146 Keiller Building 301 University Boulevard Galveston, TX 77555-0609, USA Fax: +1 409 747 2429 HEPATITIS VIRUSES
Amalio Telenti
Institute of Microbiology, CHUV 1011 Lausanne, Switzerland Fax: +41 21 314 4095 PAPILLOMAVIRUSES
William Bonnez Stephen Locarnini

Victorian Infectious Diseases Reference Laboratory WHO Collaborating Centres for Virus Reference and Research, and Biosafety 10 Wrekyn Street, North Melbourne 3051, Victoria, Australia Fax: +61 3 9342 2666 CLINICAL HIV Infectious Diseases Unit - Box 689 University of Rochester Medical Center 601 Elmwood Avenue Rochester, NY 14642, USA Fax: +1 585 442 9328 RESPIRATORY VIRUSES
Douglas D Richman
University of California, San Diego Departments of Pathology and Medicine 0679 9500 Gilman Drive, La Jolla CA 92093-0679, USA Tel: +1 858 552 7439 Fax: +1 858 552 7445 E-mail: drichman@ucsd.edu
Frederick G Hayden
Department of Internal Medicine Box 473, University of Virginia Health Sciences Center, Charlottesville VA 22908, USA Fax: +1 434 924 9065
Peter Reiss
National AIDS Therapy Evaluation Centre Academic Medical Centre University of Amsterdam, Pietersbergweg 9 1105 BM Amsterdam, the Netherlands Fax: +31 20 314 9399
SECTION EDITORS
HERPESVIRUSES
Richard J Whitley
University of Alabama at Birmingham School of Medicine, Division of Clinical Virology 616 Childrens Hospital 1600 7th Avenue South, Birmingham AL 35233-0001, USA Fax: +1 205 934 8559
Joe Eron
University of North Carolina at Chapel Hill 211a W Cameron St, CB 7030, APCF Chapel Hill, NC 27599-7030, USA Fax: +1 919 966 1576
EDITORIAL BOARD
Fred Y Aoki (Winnipeg, Canada) Lawrence Banks (Trieste, Italy) Karen K Biron (Research Triangle Park, NC, USA) Michael Bray (Bethesda, MD, USA) David M Burger (Nijmegen, the Netherlands) Jeffrey I Cohen (Bethesda, MD, USA) Ann Collier (Seattle, WA, USA) Robert B Couch (Houston, TX, USA) Janet Englund (Houston, TX, USA) Delia Enria (Pergamino, Argentina) Michael Gale (Dallas, TX, USA) Jos M Gatell (Barcelona, Spain) Lutz Gissmann (Heidelberg, Germany) Scott M Hammer (Boston, MA, USA) Martin Hirsch (Boston, MA, USA) Stanley M Lemon (Galveston, TX, USA) Yun-Fan Liaw (Taipei, Taiwan) Susan J Little (San Diego, CA, USA) Anna Lok (Ann Arbour, MI, USA) Michael Peter Manns (Hannover, Germany) Patrick Marcellin (Clichy, France) Julio SG Montaner (Vancouver, Canada) Luc Perrin (Geneva, Switzerland) Thierry Poynard (Paris, France) Jonathan M Schapiro (Tel Hashomer, Israel) Raymond F Schinazi (Atlanta, GA, USA) Robert T Schooley (Denver, CO, USA) John J Treanor (Rochester, NY, USA) Stefano Vella (Rome, Italy) Patrick Yeni (Paris, France)
EDITORIAL OFFICE
MANAGING EDITOR Stephen Cameron SENIOR PRODUCTION EDITOR Gavin Jamieson PRODUCTION EDITOR Angela Willcox PRODUCTION ASSISTANT Sarah Leboff International Medical Press 36 St Mary at Hill London EC3R 8DU UK Tel: +44 20 7398 0700 Fax: +44 20 7398 0701 E-mail: info@intmedpress.com http://www.intmedpress.com
ii
Antiviral Therapy 10, Supplement 2
CONTENTS
Organizing and Scientific Committee Workshop supporters Programme Review contents Introduction Reviews Speakers affiliation iv iv v vii M1 M3 M131
iii
ORGANIZING & SCIENTIFIC COMMITTEE

Professor Andrea Cossarizza Chair of Immunology, Department of Biomedical Sciences, University of Modena and Reggio Emilia, Modena, Italy Professor Roberto Esposito Infectious Diseases Clinics, University of Modena and Reggio Emilia, Modena, Italy Dr Cristina Mussini Infectious Diseases Clinics, University of Modena and Reggio Emilia, Modena, Italy
www.unimore.it
Support for the 1st Meeting on Mitochondrial Toxicity and HIV Infection: Understanding the Pathogenesis for a Therapeutic Approach:
The meeting was supported by an unrestricted educational grant from:
This supplement was supported by:
iv
Proceedings of the 1st Meeting on Mitochondrial Toxicity and HIV Infection: Understanding the Pathogenesis for a Therapeutic Approach
May 19-21, 2005, Modena, Italy
PROGRAMME
TIME TITLE PRESENTING AUTHOR
Thursday, May 19, 2005

18:00 Opening ceremony Opening lecture: Mitochondria in phenoptosis, programmed death of the organism Chairmen: Roberto Esposito, Fernando Aiuti Vladimir Skulachev
Friday, May 20, 2005

SESSION 1 9.00-13.00 From the organelle to the clinic Introductory remarks Mitochondrial nucleoside pools in AIDS: impact of NRTIs on a delicate cellular balance Methodological considerations in human studies of gene expression in HIV-associated lipodystrophy Mitochondrial RNA as a new marker for drug toxicity Chairmen: Gianni Di Perri, Adriano Lazzarin Kees Brinkman William Lewis Patrick W. Mallon
Andrea Cossarizza
Diagnosis of mitochondrial dysfunction in Jordi Casademont HIV-infected patients under HAART: possibilities beyond the standard procedures HIV protease inhibitors as regulators of mitochondrial-induced T cell apoptosis SESSION 2 14.00-18.30 Clinical evidence of mitochondrial damage Lipodystrophy and mitochondria: from experimental hypothesis to clinical findings Predictors of lipodystrophy and metabolic alterations in the lipoicona cohort Mitochondrial function in fat from HIV patients Role of NRTI-induced adipocyte mtDNA depletion in lipoatrophy Mitochondrial toxicity in children perinatally exposed to nucleoside analogues. An update Mitochondrial toxicity in HIV-infected children Chairmen: Roberto Cauda, Mauro Moroni Oscar Mir Massimo Galli Mariana Gerschenson Simon Mallal Stephane Blanche Alessandra Vigan Walter Malorni
Viral hepatitis and its therapy influence Vincent Soriano mitochondrial damage in HIV-infected patients
Saturday, May 21, 2005

SESSION 3 9.00-13.00 The challenge for mitochondrial recovery Kidney toxicity in HIV antiretroviral therapy Chairmen: Andrea Cossarizza, Roberto Esposito Hlne C.F. Ct
Efficacy and safety of medical and surgical Giovanni Guaraldi interventions for treating morphological and metabolic alterations of HIV-related lipodystrophy in women: a 48-week observational study Peripheral neuropathy: recent advances in pathophysiology and treatment Effects of reducing stavudine dose or switching from stavudine to tenofovir in HIV+ patients treated with standard dose stavudine-based triple therapy Mitochondrial DNA during treatment interruptions Uridine in the treatment of mitochondrial toxicity Concluding remarks Mike Youle Ana Milinkovic
Cristina Mussini Ulrich Walker Peter Reiss, Andrea Cossarizza
vi
REVIEW CONTENTS
M1 Introduction Mitochondria, HIV infection and its treatment: where do we go from here? A Cossarizza & P Reiss On the possible ways nucleoside analogues can affect mitochondrial DNA content and gene expression during HIV therapy H Ct Nucleoside reverse transcriptase inhibitors, mitochondrial DNA and AIDS therapy W Lewis HIV-protease inhibitors prevent mitochondrial hyperpolarization & redox imbalance and decrease endogenous UCP-2 expression in GP-120-activated human T lymphocytes P Matarrese et al. Mechanisms of HIV and NRTI injury to the mitochondria G Moyle Adverse effects of nucleoside analogues treatment: focus on HIV-infected children A Vigano & V Giacomet Diagnosis of mitochondrial dysfunction in HIV-infected patients under HAART: possibilities beyond the standard procedures J Casademont et al. Mitochondrial studies in HAART-related lipodystrophy: from experimental hypothesis to clinical findings O Mir et al. Mitochondrial DNA levels of PBMCs and subcutaneous adipose tissue from thigh, fat and abdomen of HIV-1 seropositive and negative individuals M Gerschenson et al. Altered mitochondrial RNA production in adopocytes from HIV infected individuals with lipodystrophy L Galluzzi et al. Methodological considerations in human studies of gene expression in HIV-associated lipodystrophy P Mallon et al. The role of hepatitis C virus (HCV) on mitochondrial DNA damage in HIV/HCV co-infected individuals C de Mendoza & V Soriano Uridine in the prevention and treatment of NRTI-related mitochondrial toxicity U Walker & N Venhoff HIV-associated antiretroviral toxic neuropathty (ATN): recent advances in pathophysiology and treatment M Youle
vii
M3
Reviews
M13
M29
M47
M53
M65
M73
M83
M91
M101
M109
M117
M125
Modena, Italy 1921 May 2005
Introduction Mitochondria, HIV infection and its treatment: where do we go from here?
Andrea Cossarizza1* and Peter Reiss2
1 2
University of Modena and Reggio Emilia, Modena, Italy Academic Medical Center, University of Amsterdam, The Netherlands & Clinical HIV Section Editor, Antiviral Therapy
*Corresponding author: Tel: +39 059 205 5415; Fax: +39 059 205 5426; E-mail: cossarizza.andrea@unimore.it
The idea of organizing a meeting on different aspects of the interactions between mitochondria, HIV infection and its treatment emerged as a result of the growing interest the scientific community has shown in this topic during recent years. Many of the scientists who have been conducting clinical as well as more basic research in this field responded enthusiastically to the invitation to present and discuss their ongoing research at the first international conference on Mitochondrial toxicity and HIV infection: understanding the pathogenesis for a therapeutic approach, held in Modena (Italy) from May 19-21, 2005. The initiative for this conference was taken by Professor Andrea Cossarizza, Chair of Immunology at the University of Modena and Reggio Emilia, and his colleagues from the Department of Infectious Diseases at the same University, Professor Roberto Esposito and Dr Cristina Mussini. The meeting was made possible by un unrestricted educational grant from Gilead Sciences, Italy, and publication of part of the proceedings of the meeting in this supplement of Antiviral Therapy was made possible by an unrestricted grant from Sigma Tau, Italy. We wish to underline that the conference organizers had full responsibility for the content of the programme as well as the choice of speakers and session chairs, without any interference from the pharmaceutical sponsors. All papers appearing in this supplement underwent peer-review prior to being considered for publication. The meeting started with a lectura magistralis given by Professor Vladimir Skulachev (State University of Moscow, Russia), one of the founding fathers of the field of bioenergetics, who provided an overview on programmed death in relation to ageing, at the level of individual cells (apoptosis), whole organisms (phenoptosis) as well as subcellular organelles including
mitochondria (mitoptosis). This stimulated discussion on whether several of the adverse effects of treatment for HIV infection might be a reflection of accelerated ageing by mechanisms including mitochondrial toxicity. The talks and related discussions that took place in the following days revealed that several issues concerning the detrimental effects which HIV, other concomitant infections such as those with hepatitis C, and HIV therapy may have on mitochondria warrant further investigation and clarification in the years to come. What is becoming increasingly clear is that antiretrovirals, and nucleoside analogue reverse transcriptase inhibitors (nRTI) in particular, may affect mitochondria in more ways than one. Inhibition by nRTI of DNA -polymerase resulting in mitochondrial DNA (mtDNA) depletion remains one of the cornerstones by which nRTI may induce mitochondrial toxicity. It is evident however that nRTI and even other classes of antiretrovirals such as HIV protease inhibitors, may have additional effects on mitochondria, for instance by way of exerting direct effects on mtRNA transcription and mitochondrial enzymes. Differences between individual agents as well as differences in the extent to which these effects may play a role in different cell types remain to be further delineated. An important problem still remaining is the choice of the type of cell in which to best measure mitochondrial markers as a possible reflection of treatment toxicity. Blood obviously remains the easiest tissue to obtain from patients, but contamination with platelets which contain numerous mitochondria may yield poorly interpretable results when using whole blood or peripheral blood mononuclear cells (PBMCs). Better purified cell populations such as isolated CD4+ or CD8+ T lymphocytes, or platelet-depleted lymphocytes
M1
or monocytes may provide more reliable results and need to be investigated further. Nevertheless, other tissues and cells coming directly from the organ systems affected by the treatment toxicity being studied (for instance adipose, hepatic or renal) may be more informative, and more appropriate when trying to assess the possible mitochondrial pathogenesis underlying a specific drug toxicity. In order for any marker of mitochondrial toxicity to become clinically useful for the early detection of drug toxicity, its sensitivity, specificity and predictive value would have to be adequately demonstrated in the context of appropriate clinical studies. Several technologies, assays and methodologies are nowadays used by several groups to analyse different aspects of mitochondrial biology. Flow cytometry and confocal microscopy, utilizing specific fluorescent probes, can detect at the single cell level changes in several mitochondrial parameters (including changes in internal membrane potential or organelle mass), as well as modification in the morphology of the cell or tissue under investigation, and are widely used in different laboratories. Classic biochemical assays, which can measure different aspects of mitochondrial respiratory chain function, are also quite popular, even if they often require large amounts of biological material, classically skeletal muscle. Thanks to the improvement in molecular biological techniques, including assays based upon the real time polymerase chain reaction, the measurement of the amount of mitochondrial DNA per cell is becoming easier, and a variety of scientists even if different genes are used as target are obtaining similar results, also as far as the absolute number of mtDNA copies in a given cell is concerned. Nevertheless, adequate quality control and assurance programmes are key in order to allow comparison of results obtained by different groups of investigators. Of note, a task force focussing on the further standardization of mtDNA quantification, involving investigators in the US, Canada, Australia, and Europe, has just been created. Similar approaches should be advocated for the evaluation of additional mitochondrial markers such as the quantification of mitochondrial RNA. Conferences such as the one held in Modena will hopefully contribute to the further unravelling of the role mitochondria play in the setting of HIV infection and its treatment, ultimately leading to safer treatments for our patients.
M2
Possible ways nucleoside analogues can affect mitochondrial DNA content and gene expression during HIV therapy
Hlne CF Ct
British Columbia Centre for Excellence in HIV/AIDS, Department of Pathology & Laboratory Medicine/University of British Columbia, Vancouver, BC, Canada Tel: +1 604 822 9777; Fax: +1 604 822 7635; E-mail: helene.cote@ubc.ca
In recent years, research into nucleoside reverse transcriptase inhibitor (NRTI)-related mitochondrial (mt) toxicity in HIV therapy has led to conflicting results and many unanswered questions regarding the molecular mechanisms that lead to such toxicity. From the early hypothesis that inhibition of the human mt polymerase by NRTIs was responsible for the drugs mt toxicity, an increasingly complex picture is emerging that probably involves
multiple mt pathways. Results have been presented suggesting that NRTIs affect not only mtDNA but also mtRNA, nucleotide phosphorylation and the mt respiratory chain. Based on the current level of knowledge, this overview addresses some of the potential mechanisms through which NRTIs could affect mitochondria and ultimately cause the toxicity symptoms observed in HIV patients receiving NRTI-containing antiretroviral therapy.
Introduction
Since mitochondrial (mt) toxicity was suggested as a common pathway for the adverse effects of nucleoside reverse transcriptase inhibitors (NRTIs) [13], a body of evidence has accumulated that, for the most part, supports the model. The intention of this review is to point out some of the limitations of HIV drug toxicity research, review the evidence relating NRTI toxicity to mtDNA damage (or not), and present the various pathways of mt toxicity that could potentially contribute to such damage. It has been known for some time that mt toxicity tends to be drug and tissue specific [49,39]. More recently, evidence has been mounting that NRTIrelated mt toxicity in HIV is probably a complex, multifactorial phenomenon. Drug- and tissue-specific differences in inhibitory activity toward polymerases could, of course, play a part, but a number of other variables are emerging. Different rates of NRTI uptake and/or intracellular phosphorylation between tissues could modulate specificity [11]. Patient characteristics such as gender, body weight or coinfection status can influence the incidence and severity of mt toxicity symptoms. A plausible association between proinflammatory and immune activation cytokine levels such as interferon and tissue necrosis factor, and the heightened risk of NRTI toxicity in specific populations has been suggested [11,12], but the mechanism behind this effect remains unclear. Co-morbidities, concomitant medications and genetic factors, as well as HIV therapy-independent mt damage can all influence and modulate mt toxicity and the development of symptoms.
Limitations of HIV mt toxicity research

Mitochondrial toxicity is a complex phenomenon, with multiple factors involved. Several factors may have contributed to the inconsistencies currently found in the literature on HIV-related mt toxicity (Table 1). Firstly, the relatively low incidence of severe mt toxicity-related adverse events has presented a challenge to studies in the HIV patient population. Secondly, many clinical symptoms suspected to have their aetiology in mt toxicity such as peripheral neuropathy, myopathy or even lipoatrophy can be challenging to quantify objectively in a standardized fashion. The majority of publications to date have studied relatively small observational cohorts, often in cross-sectional designs. In small studies, a lack of statistically significant association between mtDNA levels and a given clinical condition, although important to report, may be imputable to insufficient power, thereby limiting the conclusions that can be reached. Another issue is the potential time lag between the NRTI-induced mt damage taking place and the actual manifestation of the toxicity through clinical symptoms. As is the case of hyperlactataemia [16] and
M3
HCF Ct
Table 1. Limitations and factors that may contribute to the inconsistency of findings in the HIV antiretroviral therapy mt toxicity literature
Low incidence of severe adverse event Challenge to objectively grade or quantify certain adverse effects Time lag between mt damage and clinical symptoms Limited sample size Cross-sectional as opposed to longitudinal design Potential biases inherent to observational studies Rapid cell turnover Differences in sample type and sample processing Differences in techniques and methodologies
mt, mitochondrial.
Potential pathways of mt toxicity (Figure 1)

mtDNA depletion
Cellular mtDNA content is significantly lower in HIVinfected patients than in uninfected controls [1621]. This important observation strongly suggests a virusinduced effect on the mitochondria and its genome, by an as yet unknown mechanism. That NRTI-containing therapy exacerbates this effect is highly plausible. It has long been known that NRTIs used in HIV therapy can inhibit the mt polymerase enzyme in vitro [22,23]. Cell culture [2429] and animal studies [3034] provide strong evidence in support of the NRTI inhibiting mtDNA synthesis model. However, whether this mtDNA depletion is the pathogenesis for all clinical symptoms of toxicity observed in HIV patients on highly active antiretroviral therapy (HAART) is less clear. The DNA polymerase hypothesis by itself fails to explain the entire array of metabolic deficiencies associated with NRTI-induced disorders [35]. Table 2 presents a summary of some of the published studies on NRTI and related mt toxicity, divided in two groups: consistent and inconsistent with mtDNA depletion as the pathogenesis of mt toxicity. mtDNA depletion was first reported in muscle [54], fat [55] and nerve tissue [4] of HIV patients experiencing drug-related symptoms of myopathy, lipoatrophy and neuropathy, respectively. We reported mtDNA depletion in peripheral blood buffycoat from HIV-infected patients receiving NRTIs and having symptoms of mt toxicity including hyperlactataemia
probably lipoatrophy, this time lag can also present a challenge to the demonstration of any relationship, let alone causation. In certain tissues with rapid turnover such as blood cells, depleted mtDNA levels usually rebound upon withdrawal of drug pressure [16]. This readily available sample is thereby susceptible to treatment interruptions, planned or not. Finally, differences in the type of sample studied as well as the techniques and methodologies used could also explain some of the discrepancies in the literature [1315].
Figure 1. Schematic representation of several potential mechanisms by which NRTIs could affect mtDNA and mt gene expression (mtRNA), leading to mt toxicity
Direct mtDNA depletion through inhibition of replication by polymerase and early chain termination Direct mtRNAs (mRNA, tRNA or rRNA) depletion through inhibition of RNA polymerase or down-regulation of transcription
mtRNAs NRTI
1 2
1
mt proteins
mtDNA
mtDNA
Indirect depletion of mtDNA and mtRNAs through nucleoside/nucleotide pool imbalance caused by:
NRTI
5
dNTP
4
MRC
3 4 5 6
MRC dysfunction inhibiting de novo pyrimidine synthesis Reduced ATP production inhibiting endogenous nucleoside phosphorylation NRTI phosphorylation by mt kinases inhibiting endogenous nucleoside phosphorylation Direct inhibition of MRC by non-phosphorylated nucleoside analogues
(de novo)
ATP endogenous nucleosides
kinases nucleoside analogues

6
Pathways recently suggested or that are more speculative in nature are indicated by numbered squares. mtRNA: mRNA, rRNA or tRNA. dNTP, deoxynucleotides; MRC, mt respiratory chain; mt, mitochondrial; NRTI, nucleoside reverse transcriptase inhibitor.
M4
Mitochondrial nucleic acids in HIV therapy
[16]. The mtDNA depletion was reversible upon removal of the NRTIs suspected of causing mt toxicity, as are most toxicity symptoms if the antiretroviral therapy is changed or interrupted. We also found that the NRTI combination of stavudine (d4T) with didanosine (ddI), two drugs associated with a greater risk of hyperlactataemia [5,56,57], was also associated with greater blood mtDNA depletion than the other NRTI combinations studied [7]. These findings all appear consistent with the involvement of mtDNA depletion in NRTI-induced mt toxicity. However, not all studies of HIV therapy-related symptomatic mt toxicity show mtDNA depletion (Table 2). Discrepancy and controversy also remains with respect to the effect of NRTI-mediated mtDNA depletion on mt function. Several studies, using various tissues from HIV patients, have concluded that mtDNA levels and mt activities are positively correlated [38,44,58]. However, others would suggest otherwise and the relationship between mtDNA and mt function is somewhat unclear based on the current HIV literature. For example, one study found that a decrease in blood cell mt mass and mtDNA content was not accompanied by a decrease in cytochrome c oxidase gene expression or activity, leading the authors to suggest a compensatory mechanism up-regulating mt transcription or translation [59]. Another study from the same group found that decreased peripheral blood mononuclear cell (PBMC) mtDNA was accompanied by a decrease in mt respiratory chain (MRC) complex IV activity, but without evidence of mt dysfunction such as altered oxygen consumption or lipid peroxidation [51]. Indeed, results have also been inconsistent with respect to the association of mtDNA depletion with specific clinical adverse effects widely attributed to mt toxicity such as lipoatrophy and hyperlactataemia (Table 2). For example, in cases of hyperlactataemia and lipoatrophy, several studies have shown an association between the symptoms and mtDNA depletion in blood cells [16,10,18] while others have not [47,50,20]. In fat tissue, a few studies have suggested an association between fat mtDNA copies/cell and the presence of lipoatrophy [36,40,45]. Confocal microscopy analyses even offered direct evidence of a relationship between severity of adipose tissue toxicity and mtDNA depletion [40]. Recently, a significant decrease in fat cell mtDNA but not PBMC mtDNA levels was found in patients with peripheral neuropathy or lipodystrophy [43]. In contrast, another group found PBMC mtDNA decreased significantly in patients with lipoatrophy compared with those without lipoatrophy while mtDNA levels in adipose tissue did not differ significantly between the two groups [10]. Thus, contradictory results prevail in adipose tissue as well. These and other studies [60] clearly emphasize the
importance of the tissue assayed. The mtDNA measurements are also likely to be influenced by the severity of the toxicity symptoms, as well as the timing of the tissue sample collection, issues that are rarely addressed in the current research. Drug-related adverse events remain a common reason for intentional lack of adherence or for therapy change or interruption in adherent patients [61,62]. We carried out an observational study on a cohort of HIVinfected individuals starting their first regimen, to examine the relationship between blood buffycoat mtDNA and therapy change or discontinuation for any reason. Within the first 6 months on therapy, we measured a greater blood cell mtDNA decrease from baseline in those receiving the NRTI combination d4T/ddI. A greater proportion of these patients also eventually changed or discontinued therapy for any reason during the study period. However, in the final analysis, despite the fact that all NRTI combinations studied showed a decline in mtDNA levels at the last follow-up, an inverse relationship was observed a shorter time to therapy change/stop was associated with an increase in mtDNA from baseline, rather than a decline [63]. A number of confounding factors that could contribute to therapy changes or cessation could not be controlled for and may have contributed to this counterintuitive result. Alternatively, it could reflect some compensatory mechanism as suggested by others [59], which would be consistent with the in vitro observation that mt mass and mtDNA content initially increase in response to oxidative stress [64]. Overall, poor correlation between mtDNA levels and various mt toxicity symptoms strongly suggest additional mechanism(s), some of which are presented below.
mtRNA depletion
The mt genome encodes 13 polypeptides, all subunits of the mt respiratory chain (MRC) and essential for oxidative phosphorylation. It is tempting to assume that, in a manner similar to the effect of aging on muscle [65], depletion of mtDNA by NRTIs may lead to decreases in mt gene expression and ultimately a decrease in mt function. Recent work, however, has thrown doubt on this simplistic model. Non-HIV studies have shown that despite severe mtDNA depletion, mt genes can be transcribed at normal levels [66,67]. In HIV patients, results once again have been somewhat inconsistent. One study found that mtDNA depletion was not accompanied by a decrease in PBMC mt gene expression [59]. A second showed no change in PBMC mtDNA or mtRNA levels between two groups of patients initiating an antiretroviral regimen with or without d4T [52]. In contrast, Mallon et al. reported significantly decreased mt gene expression in adipocytes
M5
HCF Ct
Table 2. Published studies with findings appearing consistent versus inconsistent with a relationship between mtDNA damage and NRTI-related mt toxicity
Consistent with relationship between mtDNA damage and mt toxicity Specimen studied (tissue) Subcutaneous fat Blood buffycoat Subcutaneous fat PBMCs Blood buffycoat Subcutaneous fat Subcutaneous fat Subcutaneous fat Cord blood, placenta Peripheral blood Subcutaneous fat PBMCs Subcutaneous fat Subcutaneous fat PBMCs Liver biopsy Subcutaneous fat Subcutaneous fat PBMCs, muscle, fat Study/control, n 11/13 8/47 33/16 12/24 214 69/7 20/11 21/11 8/5 16/10 28 28 25/10 7 69 34/35 6/7 96/34 7 mtDNA Decrease Decrease Decrease Decrease Decrease Decrease Decrease Decrease Decrease Mutations Decrease Decrease Decrease Decrease Decrease Decrease Decrease Decrease Increase Associated with Lipoatrophy at biopsy site versus no lipoatrophy Symptomatic hyperlactataemia D drugs HAART +/ lipodystrophy, decline in MRC activity Didanosine/stavudine, didanosine use Stavudine, fat wasting Stavudine or zidovudine versus no HAART NRTI therapy versus no HAART NRTI therapy versus no HIV Lipoatrophy Stavudine versus zidovudine HIV lipoatrophy Peripheral neuropathy, lipodystrophy, HAART Reduced COX activity Stavudine versus other drugs, higher lactate D drugs HIV lipoatrophy Thymidine analogue versus non-thymidine Substituting stavudine, improved lipoatrophy Reference Walker et al., 2002 [36] Cote et al., 2002 [9] Cherry et al., 2002 [37] Miro et al., 2003 [38] Cote et al., 2003 [7] Nolan et al., 2003 [6] Pace et al., 2003 [39] Nolan et al., 2003 [40] Shiramizu et al., 2003 [41] Martin et al., 2003 [42] Van der Valk et al., 2004 [10] Van der Valk et al., 2004 [10] Rabing Christensen et al., 2004 [43] Hammond et al., 2004 [44] de Mendoza et al., 2004 [18] Walker et al., 2004 [20] Gerschenson et al., 2004 [14] Hammond et al., 2004 [45] McComsey et al., 2005 [46]
Inconsistent with relationship between mtDNA damage and mt toxicity Specimen studied (tissue) Blood leukocytes Blood lymphocytes PBMCs Blood lymphocytes PBMCs PBMCs PBMCs Blood Subcutaneous fat PBMCs PBMCs Longitudinal PBMCs Subcutaneous fat Study/control, n 10/10 6/12 33/16 23/11 33/13 94 25/10 157 28 42/25 7/7 15/63 20/0 mtDNA No difference No difference No difference Increase Decrease No difference No difference No difference No difference Decrease No difference No difference No difference Associated with Patients with or without lipoatrophy Children with or without lipodystrophy HAART with or without D drugs Lipodystrophy, cholesterolaemia Low CD4, high viral load Lipoatrophy patients on thymidine analogue versus switch to abacavir Peripheral neuropathy, lipodystrophy, HAART HIV treatment, lipodystrophy, lactate, clinical symptoms Patients with or without lipoatrophy No change in oxygen consumption and lipid peroxidation HIV lipoatrophy Clinical adverse events Decreased mtRNA Reference McComsey et al., 2002 [47] Cossarizza et al., 2002 [48] Cherry et al., 2002 [37] Cossarizza et al., 2003 [49] Miura et al., 2003 [17] Hoy et al., 2004 [50] Rabing Christensen et al., 2004 [43] Chiappini et al., 2004 [20] Van der Valk et al., 2004 [10] Lopez et al., 2004 [51] Gerschenson et al., 2004 [14] Casula et al., 2004 [52] Mallon et al., 2004 [53]
D drugs: didanosine, stavudine and zalcitabine (dideoxynucleotides); COX, cytochrome c oxidase; HAART, highly active antiretroviral therapy; mt, mitochondrial; MRC, mt respiratory chain; NRTI, nucleoside reverse transcriptase inhibitor; PBMCs, peripheral blood mononuclear cells.
M6
and blood monocytes after short periods on zidovudine (AZT)- or d4T-containing therapy [53,68]. Importantly, this was not accompanied by any significant change in mtDNA content or metabolic parameters, and was independent of HIV [53,68]. The latter results would imply that NRTIs can exert an effect, directly or indirectly, not only on mtDNA polymerization but could also regulate mt gene expression and/or inhibit mtRNA polymerization. Interestingly, the DNA polymerase is capable of catalysing reverse transcription with a high efficiency, an activity that may be physiologically significant, especially considering that RNA-primed DNA synthesis activity is required for initiation of mtDNA replication [69]. Such activity could also be inhibited by NRTIs. Another potential target of NRTIs could be the mtRNA polymerase itself and the co-factors required for mt transcription, some of which are regulated through phosphorylation [70] and are, therefore, susceptible to MRC dysfunction. In addition, NRTIs could impair ribonucleotide synthesis and/or utilization in a manner similar to that hypothesized for nucleotides. All of the above could potentially translate into altered mt messenger RNA (mRNA) levels. The mt genome also encodes 22 mt transfer RNAs (tRNAs) and 2 ribosomal RNAs (rRNAs) that are necessary for the translation of the 13 mtDNA-encoded polypeptides. Mutations in mt tRNA genes have been associated with various neuromuscular and neurodegenerative disorders, many sharing phenotypic similarities with HIV therapy mt toxicity symptoms [71,72]. Any NRTI-induced mtDNA damage or any transcription inhibition could also affect these other mtRNAs and participate in mt toxicity.
nucleotides [dTTP, deoxycytosine triphosphate (dCTP) and deoxyuridine triphosphate (dUTP)] is coupled to the MRC [78]. Therefore, MRC dysfunction, whatever the cause, leads to decreased mt ATP regeneration that, in turn, impairs pyrimidine synthesis as well as its nucleotide phosphorylation [79], further enhancing the cycle of mt toxicity. Indeed, mtDNA-lacking cells do not synthesize pyrimidine nucleosides and require exogenous uridine for growth [80]. Interestingly, this is consistent with the observations that uridine can rescue mtDNA depletion induced by the pyrimidine analogues AZT, d4T and zalcitabine (ddC) in vitro [81,82].
mtDNA damage (mutation, deletion)

Much attention has been paid to mtDNA quantity in recent research. Also important, but technically challenging to evaluate, is mtDNA quality. It is well-documented that mtDNA damage in the form of mutations and deletions accumulates over time, and plays a role in the conditions and diseases associated with aging [8385]. NRTI therapy also provides conditions permissive for the development of peripheral blood mtDNA mutations in vivo [42]. Furthermore, NRTI or oxidative agents have been shown to directly impair the energy-producing system of mitochondria, causing dysfunction of cellular redox control, which eventually leads to loss of the mtDNA integrity [86]. In utero d4T exposure in mice leads to long-term mt damage and dysfunction in heart mitochondria, lasting to adulthood [87]. This effect would not be mediated by inhibition of the DNA polymerase but more probably result from persistent mtDNA mutations in the cardiac tissue [87], and it provides indirect evidence of NRTIs causing long-term decrease in mtDNA quality. In the HIV population, large mtDNA deletions have been reported in a few cases of NRTI lactic acidosis [88,89]. Although it is unknown at this point whether these deletions predated HIV infection and HIV therapy, one can hypothesize that the presence of such deletions would, in all likelihood, diminish the mitochondrias ability to compensate under NRTI drug pressure, thereby accelerating or exacerbating the development of mt toxicity symptoms.
Nucleotide pool imbalance

mtDNA depletion can be caused by NRTI incorporation into elongating mtDNA or inhibition of the polymerase , but even if these events occur, they are probably not the only mechanisms at play. Imbalances in the mt nucleotide pool have been suggested as the cause of mtDNA abnormalities in patients with inherited thymidine phosphorylase deficiencies [73,74]. In a similar manner, nucleoside analogues used to treat HIV infection can alter the endogenous cellular and mt nucleoside/nucleotide pools, potentially leading to the disturbance of a wide range of nucleic acid pathways that depend on these building blocks (Figure 1). In its non-phosphorylated pro-drug form, AZT is a potent inhibitor of thymidine phosphorylation in heart and liver mitochondria [7577]. Thus, the toxicity of AZT in some tissues may be mediated by disruption of the substrate supply of deoxythymidine triphosphate (dTTP) for mtDNA replication. De novo synthesis of pyrimidine
Other potential mt toxicity mechanisms

Apart from the various ways by which NRTIs can affect mt nucleic acids and their synthesis, there are a number of mtDNA-independent, poorly understood potential mechanisms that have been suggested for NRTI-mediated toxicity and a few are presented below. Before any measurable effect on mtDNA, short-term AZT treatment of cultured rat myotube can induce a marked, yet reversible reduction in mt membrane
M7
HCF Ct
Figure 2. Potentially cumulative effect of HIV infection, increased ROS production and NRTIs on mtDNA damage (depletion/mutation/deletion)
HIV mtDNA damage
ROS
NRTI
NRTI, nucleoside reverse transcriptase inhibitor; mt, mitochondrial; ROS, reactive oxygen species.
infection can also damage mtDNA. This cumulative mtDNA damage eventually impairs MRC function, which itself can lead to further oxidative stress and mtDNA damage [85]. Long-term HIV therapy with NRTI-containing regimen in aging patients thus provides the context for a triple insult to the mitochondria (Figure 2). One can hypothesize a synergistic, multifactorial model for mt toxicity, which would operate at different rates depending on the tissue, the drugs and the individuals health and genetic background. These are but a few of the various ways NRTIs and their intermediates are known or suspected to lead to mt toxicity in HIV patients receiving HAART. Further research with large controlled longitudinal studies, preferably with standardized assays [13,97], will be needed to elucidate the mechanisms of NRTI-induced mt toxicity and their relative contribution to clinical toxicity symptoms. This will be crucial in determining the marker(s) and samples most clinically relevant to mt toxicity testing, with the goal of preventing, predicting and monitoring mt toxicity in HIV patients on antiretroviral therapy.
Acknowledgements
potential [90]. This suggests that AZT can physically interfere with the membrane structure and could lead to modifications of its physical characteristics. This is consistent with the earlier observations that AZT exerts a short-term effect directly on the MRC [30], causing a decrease in ATP production [86], in addition to long-term alteration of the mtDNA. Similarly, in a rat heart model, ddC can decrease ATP, increase reactive oxygen species (ROS) formation and mediate cell damage before changes to mtDNA occur [91]. Recently, thymidine analogues and a thymidine catabolite were shown to enhance hepatic fat oxidation without affecting fat mtDNA, introducing NRTI catabolic products as possible players in HIV lipoatrophy [92]. Thanks to Christopher Alexander and Brian Scarth for their assistance with the manuscript.
Support
HCFC is supported by a Scholar Investigator Award from the Michael Smith Foundation for Health Research.
References
1. 2. Lewis W & Dalakas MC. Mitochondrial toxicity of antiviral drugs. Nature Medicine 1995; 1:417422. Brinkman K, ter Hofstede HJ, Burger DM, Smeitink JA & Koopmans PP. Adverse effects of reverse transcriptase inhibitors: mitochondrial toxicity as common pathway. AIDS 1998; 12:17351744. Kakuda TN. Pharmacology of nucleoside and nucleotide reverse transcriptase inhibitor-induced mitochondrial toxicity. Clinical Therapy 2000; 22:685708. Dalakas MC, Semino-Mora C & Leon-Monzon M. Mitochondrial alterations with mitochondrial DNA depletion in the nerves of AIDS patients with peripheral neuropathy induced by 23-dideoxycytidine (ddC). Laboratory Investigation 2001; 81:15371544. John M, Moore CB, James IR, Nolan D, Upton RP, McKinnon EJ & Mallal SA. Chronic hyperlactatemia in HIV-infected patients taking antiretroviral therapy. AIDS 2001; 15:717723. Nolan D, Hammond E, James I, McKinnon E & Mallal S. Contribution of nucleoside-analogue reverse transcriptase inhibitor therapy to lipoatrophy from the population to the cellular level. Antiviral Therapy 2003; 8:617626. Cote HC, Yip B, Asselin JJ, Chan JW, Hogg RS, Harrigan PR, OShaughnessy MV & Montaner JS. Mitochondrial:nuclear DNA ratios in peripheral blood cells from human immunodeficiency virus (HIV)-infected patients who received
3. 4.
The vicious circle of mt damage

Over time, ROS and free radicals damage the mtDNA, a phenomenon that forms the basis for one of the theories on aging. This damage in turn increases ROS production and leads to further oxidative damage to the mtDNA in human tissues, such as large-scale deletions and point mutations that accumulate with age [93,94,85]. There is also growing evidence that HIV itself can affect mitochondria and its nucleic acid [9,21] by an as yet unknown mechanism, possibly predisposing patients to mt toxicity [95,96]. Finally, it is now well accepted that NRTIs used to treat HIV
M8
5.
6.
7.
8.
9.
10.
11.
12.
13. 14.
15.
16.
17.
18.
19.
20.
21.
selected HIV antiretroviral drug regimens. Journal of Infectious Diseases 2003; 187:19721976. Reiss P, Casula M, de Ronde A, Weverling GJ, Goudsmit J & Lange JM. Greater and more rapid depletion of mitochondrial DNA in blood of patients treated with dual (zidovudine+didanosine or zidovudine+zalcitabine) vs single (zidovudine) nucleoside reverse transcriptase inhibitors. HIV Medicine 2004; 5:1114. Cote HCF, Brumme ZL, Chan JW, Guillemi S, de Ronde A, de Baar MP, OShaughnessy MV, Montaner JSG & Harrigan PR. Severe mitochondrial DNA depletion in post-mortem fat and liver tissue from a fatal case of lactic acidosis during HIV therapy. 15th International AIDS Conference. 1116 July 2004, Bangkok, Thailand. Abstrast WePeB5915. van der Valk M, Casula M, Weverlingz GJ, van Kuijk K, van Eck-Smit B, Hulsebosch HJ, Nieuwkerk P, van Eeden A, Brinkman K, Lange J, de Ronde A & Reiss P. Prevalence of lipoatrophy and mitochondrial DNA content of blood and subcutaneous fat in HIV-1-infected patients randomly allocated to zidovudine- or stavudine-based therapy. Antiviral Therapy 2004; 9:385393. Anderson PL, Kakuda TN & Lichtenstein KA. The cellular pharmacology of nucleoside- and nucleotide-analogue reverse-transcriptase inhibitors and its relationship to clinical toxicities. Clinical Infectious Diseases 2004; 38:743753. Cossarizza A, Mussini C & Vigano A. Mitochondria in the pathogenesis of lipodystrophy induced by anti-HIV antiretroviral drugs: actors or bystanders? Bioessays 2001; 23:10701080. Cossarizza A. Tests for mitochondrial function and DNA: potentials and pitfalls. Current Opinion in Infectious Diseases 2003 16:510. Gerschenson M, Shiramizu B, Tran D, OCallaghan CO, LiButti DE & Shikuma CM. Mitochondrial DNA depletion in adipose tissues but not PBMC correlate with lipoatrophy in HIV-infected patients. 11th Conference on Retroviruses & Opportunistic Infections. 811 February 2004, San Francisco, CA, USA. Abstrast 710. Banas B, Kost BP & Goebel FD. Platelets, a typical source of error in real-time PCR quantification of mitochondrial DNA content in human peripheral blood cells. European Journal of Medical Research 2004; 9:371377. Cote HC, Brumme ZL, Craib KJ, Alexander CS, Wynhoven B, Ting L, Wong H, Harris M, Harrigan PR, OShaughnessy MV & Montaner JS. Changes in mitochondrial DNA as a marker of nucleoside toxicity in HIV-infected patients. New England Journal of Medicine 2002; 346:811820. Miura T, Goto M, Hosoya N, Odawara T, Kitamura Y, Nakamura T & Iwamoto A. Depletion of mitochondrial DNA in HIV-1-infected patients and its amelioration by antiretroviral therapy. Journal of Medical Virology 2003; 70:497505. de Mendoza C, de Ronde A, Smolders K, Blanco F, GarciaBenayas T, de Baar M, Fernandez-Casas P, Gonzalez-Lahoz J & Soriano V. Changes in mitochondrial DNA copy number in blood cells from HIV-infected patients undergoing antiretroviral therapy. AIDS Research & Human Retroviruses 2004; 20:271273. Miro O, Lopez S, Martinez E, Pedrol E, Milinkovic A, Deig E, Garrabou G, Casademont J, Gatell JM & Cardellach F. Mitochondrial effects of HIV infection on the peripheral blood mononuclear cells of HIV-infected patients who were never treated with antiretrovirals. Clinical Infectious Diseases 2004; 39:710716. Chiappini F, Teicher E, Saffroy R, Pham P, Falissard B, Barrier A, Chevalier S, Debuire B, Vittecoq D & Lemoine A. Prospective evaluation of blood concentration of mitochondrial DNA as a marker of toxicity in 157 consecutively recruited untreated or HAART-treated HIV-positive patients. Laboratory Investigation 2004; 84:908914. Walker UA, Bauerle J, Laguno M, Murillas J, Mauss S, Schmutz G, Setzer B, Miquel R, Gatell JM & Mallolas J. Depletion of mitochondrial DNA in liver under antiretroviral therapy with didanosine, stavudine, or zalcitabine. Hepatology 2004; 39:311317.
22. Martin JL, Brown CE, Matthews-Davis N & Reardon JE. Effects of antiviral nucleoside analogs on human DNA polymerases and mitochondrial DNA synthesis. Antimicrobial Agents & Chemotherapy 1994; 38:27432749. 23. Lim SE & Copeland WC. Differential incorporation and removal of antiviral deoxynucleotides by human DNA polymerase gamma. Journal of Biological Chemistry 2001; 276:2361623623. 24. Pan-Zhou XR, Cui L, Zhou XJ, Sommadossi JP & DarleyUsmar VM. Differential effects of antiretroviral nucleoside analogs on mitochondrial function in HepG2 cells. Antimicrobial Agents & Chemotherapy 2000; 44:496503. 25. Birkus G, Hitchcock MJ & Cihlar T. Assessment of mitochondrial toxicity in human cells treated with tenofovir: comparison with other nucleoside reverse transcriptase inhibitors. Antimicrobial Agents & Chemotherapy 2002; 46:716723. 26. Walker UA, Setzer B & Venhoff N. Increased long-term mitochondrial toxicity in combinations of nucleoside analogue reverse-transcriptase inhibitors. AIDS 2002; 16:21652173. 27. Lee H, Hanes J & Johnson KA. Toxicity of nucleoside analogues used to treat AIDS and the selectivity of the mitochondrial DNA polymerase. Biochemistry 2003; 42:1471114719. 28. Walker UA, Venhoff N, Koch EC, Olschewski M, Schneider J & Setzer B. Uridine abrogates mitochondrial toxicity related to nucleoside analogue reverse transcriptase inhibitors in HepG2 cells. Antiviral Therapy 2003; 8:463470. 29. Velsor LW, Kovacevic M, Goldstein M, Leitner HM, Lewis W & Day BJ. Mitochondrial oxidative stress in human hepatoma cells exposed to stavudine. Toxicology & Applied Pharmacology 2004; 199:1019. 30. Masini A, Scotti C, Calligaro A, Cazzalini O, Stivala LA, Bianchi L, Giovannini F, Ceccarelli D, Muscatello U, Tomasi A & Vannini V. Zidovudine-induced experimental myopathy: dual mechanism of mitochondrial damage. Journal of Neurological Sciences 1999; 166:131140. 31. Gaou I, Malliti M, Guimont MC, Letteron P, Demeilliers C, Peytavin G, Degott C, Pessayre D & Fromenty B. Effect of stavudine on mitochondrial genome and fatty acid oxidation in lean and obese mice. Journal of Pharmacology & Experimental Therapeutics 2001; 297:516523. 32. Gerschenson M, Nguyen VT, St Claire MC, Harbaugh SW, Harbaugh JW, Proia LA & Poirier MC. Chronic stavudine exposure induces hepatic mitochondrial toxicity in adult Erythrocebus patas monkeys. Journal of Human Virology 2001; 4:335342. 33. Gerschenson M, Nguyen V, Ewings EL, Ceresa A, Shaw JA, St Claire MC, Nagashima K, Harbaugh SW, Harbaugh JW, Olivero OA, Divi RL, Albert PS & Poirier MC. Mitochondrial toxicity in fetal Erythrocebus patas monkeys exposed transplacentally to zidovudine plus lamivudine. AIDS Research & Human Retroviruses 2004; 20:91100. 34. Collins ML, Sondel N, Cesar D & Hellerstein MK. Effect of nucleoside reverse transcriptase inhibitors on mitochondrial DNA synthesis in rats and humans. Journal of Acquired Immune Deficiency Syndromes 2004; 37:11321139. 35. Lund KC & Wallace KB. Direct, DNA pol-gamma-independent effects of nucleoside reverse transcriptase inhibitors on mitochondrial bioenergetics. Cardiovascular Toxicology 2004; 4:217228. 36. Walker UA, Bickel M, Lutke Volksbeck SI, Ketelsen UP, Schofer H, Setzer B, Venhoff N, Rickerts V & Staszewski S. Evidence of nucleoside analogue reverse transcriptase inhibitor-associated genetic and structural defects of mitochondria in adipose tissue of HIV-infected patients. Journal of Acquired Immune Deficiency Syndromes 2002; 29:117121. 37. Cherry CL, Gahan ME, McArthur JC, Lewin SR, Hoy JF & Wesselingh SL. Exposure to dideoxynucleosides is reflected in lowered mitochondrial DNA in subcutaneous
M9
HCF Ct
fat. Journal of Acquired Immune Deficiency Syndromes 2002; 30:271277. 38. Miro O, Lopez S, Pedrol E, Rodriguez-Santiago B, Martinez E, Soler A, Milinkovic A, Casademont J, Nunes V, Gatell JM & Cardellach F. Mitochondrial DNA depletion and respiratory chain enzyme deficiencies are present in peripheral blood mononuclear cells of HIV-infected patients with HAART-related lipodystrophy. Antiviral Therapy 2003; 8:333338. 39. Pace CS, Martin AM, Hammond EL, Mamotte CD, Nolan DA & Mallal SA. Mitochondrial proliferation, DNA depletion and adipocyte differentiation in subcutaneous adipose tissue of HIV-positive HAART recipients. Antiviral Therapy 2003; 8:323331. 40. Nolan D, Hammond E, Martin A, Taylor L, Herrmann S, McKinnon E, Metcalf C, Latham B & Mallal S. Mitochondrial DNA depletion and morphologic changes in adipocytes associated with nucleoside reverse transcriptase inhibitor therapy. AIDS 2003; 17:13291338. 41. Shiramizu B, Shikuma KM, Kamemoto L, Gerschenson M, Erdem G, Pinti M, Cossarizza A & Shikuma C. Placenta and cord blood mitochondrial DNA toxicity in HIVinfected women receiving nucleoside reverse transcriptase inhibitors during pregnancy. Journal of Acquired Immune Deficiency Syndromes 2003; 32:370374. 42. Martin AM, Hammond E, Nolan D, Pace C, Den Boer M, Taylor L, Moore H, Martinez OP, Christiansen FT & Mallal S. Accumulation of mitochondrial DNA mutations in human immunodeficiency virus-infected patients treated with nucleoside-analogue reverse-transcriptase inhibitors. American Journal of Human Genetics 2003; 72:549560. 43. Rabing Christensen E, Stegger M, Jensen-Fangel S, Laursen AL & Ostergaard L. Mitochondrial DNA levels in fat and blood cells from patients with lipodystrophy or peripheral neuropathy and the effect of 90 days of high-dose coenzyme Q treatment: a randomized, double-blind, placebocontrolled pilot study. Clinical Infectious Diseases 2004; 39:13711379. 44. Hammond E, Nolan D, James I, Metcalf C & Mallal S. Reduction of mitochondrial DNA content and respiratory chain activity occurs in adipocytes within 612 months of commencing nucleoside reverse transcriptase inhibitor therapy. AIDS 2004; 18:815817. 45. Hammond E, Nolan D, McKinnon E, James I & Mallal S. Differential effect of nucleoside reverse transcriptase inhibitor (NRTI) regimens on adipocyte mitochondrial DNA depletion in HIV-infected patients. Antiviral Therapy 2004; 9:L11. 46. McComsey GA, Paulsen DM, Lonergan JT, Hessenthaler SM, Hoppel CL, Williams VC, Fisher RL, Cherry CL, White-Owen C, Thompson KA, Ross ST, Hernandez JE & Ross LL. Improvements in lipoatrophy, mitochondrial DNA levels and fat apoptosis after replacing stavudine with abacavir or zidovudine. AIDS 2005; 19:1523. 47. McComsey G, Tan DJ, Lederman M, Wilson E & Wong LJ. Analysis of the mitochondrial DNA genome in the peripheral blood leukocytes of HIV-infected patients with or without lipoatrophy. AIDS 2002; 16:513518. 48. Cossarizza A, Pinti M, Moretti L, Bricalli D, Bianchi R, Troiano L, Fernandez MG, Balli F, Brambilla P, Mussini C & Vigano A. Mitochondrial functionality and mitochondrial DNA content in lymphocytes of vertically infected human immunodeficiency virus-positive children with highly active antiretroviral therapy-related lipodystrophy. Journal of Infectious Diseases 2002; 185:299305. 49. Cossarizza A, Riva A, Pinti M, Ammannato S, Fedeli P, Mussini C, Esposito R & Galli M. Increased mitochondrial DNA content in peripheral blood lymphocytes from HIVinfected patients with lipodystrophy. Antiviral Therapy 2003; 8:315321. 50. Hoy JF, Gahan ME, Carr A, Smith D, Lewin SR, Wesselingh S & Cooper DA. Changes in mitochondrial DNA in peripheral blood mononuclear cells from HIVinfected patients with lipoatrophy randomized to receive abacavir. Journal of Infectious Diseases 2004; 190:688692.
M10
51. Lopez S, Miro O, Martinez E, Pedrol E, RodriguezSantiago B, Milinkovic A, Soler A, Garcia-Viejo MA, Nunes V, Casademont J, Gatell JM & Cardellach F. Mitochondrial effects of antiretroviral therapies in asymptomatic patients. Antiviral Therapy 2004; 9:4755. 52. Casula M, Weverling G, de Baar M, Wit F, Stek M, Lange J & Reiss P. Longitudinal assessment of mitochondrial DNA and RNA in PBMC in a randomized comparative trial of NRTI-sparing and NRTI-containing antiretroviral combination therapy. 11th Conference on Retroviruses & Opportunistic Infections. 811 February 2004, San Francisco, CA, USA. Abstract 709. 53. Mallon PWG, Sedwell R, Unemori P, Merlin K, McGinley C, Ammaranond P, Peperias M, Rafferty M, Williams K, Samaras K, Morey AL, Chisholm D, Kelleher A, Cooper DA & Carr A. Nucleoside reverse transcriptase inhibitors (NRTI) decrease adipocyte and monocyte mitochondrial (mt) messenger RNA transcription in the absence of changes in mtDNA or cell morphology. Antiviral Therapy 2004; 9:L56. 54. Arnaudo E, Dalakas M, Shanske S, Moraes CT, DiMauro S & Schon EA. Depletion of muscle mitochondrial DNA in AIDS patients with zidovudine-induced myopathy. Lancet 1991; 337:508510. 55. Shikuma CM, Hu N, Milne C, Yost F, Waslien C, Shimizu S & Shiramizu B. Mitochondrial DNA decrease in subcutaneous adipose tissue of HIV-infected individuals with peripheral lipoatrophy. AIDS 2001; 15:18011809. 56. Boubaker K, Flepp M, Sudre P, Furrer H, Haensel A, Hirschel B, Boggian K, Chave JP, Bernasconi E, Egger M, Opravil M, Rickenbach M, Francioli P & Telenti A. Hyperlactatemia and antiretroviral therapy: the Swiss HIV Cohort Study. Clinical Infectious Diseases 2001; 33:19311937. 57. Moyle GJ, Datta D, Mandalia S, Morlese J, Asboe D & Gazzard BG. Hyperlactataemia and lactic acidosis during antiretroviral therapy: relevance, reproducibility and possible risk factors. AIDS 2002; 16:13411349. 58. Lopez S, Garrabou G, Martinez E, Domingo P, Fontdevila J, Gatell JM, Infante AB, Gallart X, Milinkovic A, Cardellach F, Casademont J & Miro O. Mitochondrial studies in adipose tissue of HIV-infected patients without fat redistribution. Antiviral Therapy 2004; 9:L20. 59. Miro O, Lopez S, Rodriguez de la Concepcion M, Martinez E, Pedrol E, Garrabou G, Giralt M, Cardellach F, Gatell JM, Vilarroya F & Casademont J. Upregulatory mechanisms compensate for mitochondrial DNA depletion in asymptomatic individuals receiving stavudine plus didanosine. Journal of Acquired Immune Deficiency Syndromes 2004; 37:15501555. 60. Vittecoq D, Jardel C, Barthelemy C, Escaut L, Cheminot N, Chapin S, Sternberg D, Maisonobe T & Lombes A. Mitochondrial damage associated with long-term antiretroviral treatment: associated alteration or causal disorder? Journal of Acquired Immune Deficiency Syndromes 2002; 31:299308. 61. Heath KV, Singer J, OShaughnessy MV, Montaner JS & Hogg RS. Intentional nonadherence due to adverse symptoms associated with antiretroviral therapy. Journal of Acquired Immune Deficiency Syndromes 2002; 31:211217. 62. Heath KV, Montaner JS, Bondy G, Singer J, OShaughnessy MV & Hogg RS. Emerging drug toxicities of highly active antiretroviral therapy for human immunodeficiency virus (HIV) infection. Current Drug Targets 2003; 4:1322. 63. Cote H, Yip B, Bruce S, Chan J, Wong H, Hogg R, Harrigan PR & Montaner J. Blood mitochondrial DNA levels in a populational cohort of HIV+ patients receiving antiretroviral therapy (ART). 44th Interscience Conference on Antimicrobial Agents & Chemotherapy. 30 October 2 November 2004, Washington, DC, USA. Abstract H-165. 64. Lee HC, Yin PH, Lu CY, Chi CW & Wei YH. Increase of mitochondria and mitochondrial DNA in response to oxidative stress in human cells. Biochemical Journal 2000; 348:425432.
65. Welle S, Bhatt K, Shah B, Needler N, Delehanty JM & Thornton CA. Reduced amount of mitochondrial DNA in aged human muscle. Journal of Applied Physiology 2003; 94:14791484. 66. Barthelemy C, Ogier de Baulny H, Diaz J, Cheval MA, Frachon P, Romero N, Goutieres F, Fardeau M & Lombes A. Late-onset mitochondrial DNA depletion: DNA copy number, multiple deletions, and compensation. Annals of Neurology 2001; 49:607617. 67. Seidel-Rogol BL & Shadel GS. Modulation of mitochondrial transcription in response to mtDNA depletion and repletion in HeLa cells. Nucleic Acids Research 2002; 30:19291934. 68. Mallon P, Unemori P, Bowen M, Miller J, Winterbotham M, Kelleher A, Williams K, Cooper D & Carr A. Nucleoside reverse transcriptase inhibitors decrease mitochondrial and PPARgamma gene expression in adipose tissue after only 2 weeks in HIV-uninfected healthy adults. 11th Conference on Retroviruses & Opportunistic Infections. 811 February 2004, San Francisco, CA, USA. Abstract 76. 69. Murakami E, Feng JY, Lee H, Hanes J, Johnson KA & Anderson KS. Characterization of novel reverse transcriptase and other RNA-associated catalytic activities by human DNA polymerase gamma: importance in mitochondrial DNA replication. Journal of Biological Chemistry 2003; 278:3640336409. 70. Prieto-Martin A, Montoya J & Martinez-Azorin F. Phosphorylation of rat mitochondrial transcription termination factor (mTERF) is required for transcription termination but not for binding to DNA. Nucleic Acids Research 2004; 32:20592068. 71. Florentz C, Sohm B, Tryoen-Toth P, Putz J & Sissler M. Human mitochondrial tRNAs in health and disease. Cellular & Molecular Life Sciences 2003; 60:13561375. 72. Finsterer J. Mitochondriopathies. European Journal of Neurology 2004; 11:163186. 73. Nishigaki Y, Marti R, Copeland WC & Hirano M. Sitespecific somatic mitochondrial DNA point mutations in patients with thymidine phosphorylase deficiency. Journal of Clinical Investigation 2003; 111:19131921. 74. Song S, Wheeler LJ & Mathews CK. Deoxyribonucleotide pool imbalance stimulates deletions in HeLa cell mitochondrial DNA. Journal of Biological Chemistry 2003; 278:4389343896. 75. McKee EE, Bentley AT, Hatch M, Gingerich J & SusanResiga D. Phosphorylation of thymidine and AZT in heart mitochondria: elucidation of a novel mechanism of AZT cardiotoxicity. Cardiovascular Toxicology 2004; 4:155167. 76. McKee EE, Lynx MD, Susan-Resiga D, Bentley AT, DHaenens JP, Cullen D & Ferguson M. Zidovudine inhibits thymidine phosphorylation: a novel site of potential toxicity in non-mitotic cells. Antiviral Therapy 2004; 9:L9. 77. Lynx MD, DHaenens JP, Bentley AT, Susan-Resiga D & McKee EE. Zidovudine inhibits thymidine phosphorylation: a novel site of potential toxicity in non-mytotic cells. Antiviral Therapy 2004; 9:L22. 78. Ruckemann K, Fairbanks LD, Carrey EA, Hawrylowicz CM, Richards DF, Kirschbaum B & Simmonds HA. Leflunomide inhibits pyrimidine de novo synthesis in mitogen-stimulated T-lymphocytes from healthy humans. Journal of Biological Chemistry 1998; 273:2168221691. 79. Gattermann N, Dadak M, Hofhaus G, Wulfert M, Berneburg M, Loeffler ML & Simmonds HA. Severe impairment of nucleotide synthesis through inhibition of mitochondrial respiration. Nucleosides, Nucleotides & Nucleic Acids 2004; 23:12751279. 80. King MP & Attardi G. Human cells lacking mtDNA: repopulation with exogenous mitochondria by complementation. Science 1989; 246:500503. 81. Walker UA, Venhoff N, Koch EC, Olschewski M, Schneider J & Setzer B. Uridine abrogates mitochondrial
toxicity related to nucleoside analogue reverse transcriptase inhibitors in HepG2 cells. Antiviral Therapy 2003; 8:463470. 82. Walker UA, Auclair M, Lebrecht D, Kornprobst M, Capeau J & Caron M. Uridine abrogates the adverse effect of stavudine and zalcitabine on adipose cell function. Antiviral Therapy 2004; 9:L10. 83. Miquel J. An integrated theory of aging as the result of mitochondrial-DNA mutation in differentiated cells. Archives of Gerontology & Geriatry 1991; 12:99117. 84. Gianni P, Jan KJ, Douglas MJ, Stuart PM & Tarnopolsky MA. Oxidative stress and the mitochondrial theory of aging in human skeletal muscle. Experimental Gerontology 2004; 39:13911400. 85. Alexeyev MF, Ledoux SP & Wilson GL. Mitochondrial DNA and aging. Clinical Science 2004; 107:355364. 86. Yamaguchi T, Katoh I & Kurata S. Azidothymidine causes functional and structural destruction of mitochondria, glutathione deficiency and HIV-1 promoter sensitization. European Journal of Biochemistry 2002; 269:27822788. 87. Walker DM, Poirier MC, Campen MJ, Cook DL Jr, Divi RL, Nagashima K, Lund AK, Cossey PY, Hahn FF & Walker VE. Persistence of mitochondrial toxicity in hearts of female B6C3F1 mice exposed in utero to 3-azido-3deoxythymidine. Cardiovascular Toxicology 2004; 4:133153. 88. Bartley PB, Westacott L, Boots RJ, Lawson M, Potter JM, Hyland VJ & Woods ML 2nd. Large hepatic mitochondrial DNA deletions associated with L-lactic acidosis and highly active antiretroviral therapy. AIDS 2001; 15:419420. 89. Walker UA & Venhoff N. Multiple mitochondrial DNA deletions and lactic acidosis in an HIV-infected patient under antiretroviral therapy. AIDS 2001; 15:14491450. 90. Cazzalini O, Lazze MC, Iamele L, Stivala LA, Bianchi L, Vaghi P, Cornaglia A, Calligaro A, Curti D, Alessandrini A, Prosperi E & Vannini V. Early effects of AZT on mitochondrial functions in the absence of mitochondrial DNA depletion in rat myotubes. Biochemical Pharmacology 2001; 62:893902. 91. Skuta G, Fischer GM, Janaky T, Kele Z, Szabo P, Tozser J & Sumegi B. Molecular mechanism of the short-term cardiotoxicity caused by 2,3-dideoxycytidine (ddC): modulation of reactive oxygen species levels and ADPribosylation reactions. Biochemical Pharmacology 1999; 58:19151925. 92. Maisonneuve C, Igoudjil A, Begriche K, Letteron P, Guimont MC, Bastin J, Laigneau JP, Pessayre D & Fromenty B. Effects of zidovudine, stavudine and betaaminoisobutyric acid on lipid homeostasis in mice: possible role in human fat wasting. Antiviral Therapy 2004; 9:801810. 93. Lee HC & Wei YH. Mitochondrial alterations, cellular response to oxidative stress and defective degradation of proteins in aging. Biogerontology 2001; 2:231244. 94. Barrientos A, Casademont J, Cardellach F, Ardite E, Estivill X, Urbano-Marquez A, Fernandez-Checa JC & Nunes V. Qualitative and quantitative changes in skeletal muscle mtDNA and expression of mitochondrial-encoded genes in the human aging process. Biochemical & Molecular Medicine 1997; 62:165171. 95. Lewis W. Defective mitochondrial DNA replication and NRTIs: pathophysiological implications in AIDS cardiomyopathy. American Journal of Physiology, Heart & Circulatory Physiology 2003; 284:H1H9. 96. Cossarizza A & Moyle G. Antiretroviral nucleoside and nucleotide analogues and mitochondria. AIDS 2004; 18:137151. 97. Hammond EL, Sayer D, Nolan D, Walker UA, Ronde A, Montaner JS, Cote HC, Gahan ME, Cherry CL, Wesselingh SL, Reiss P & Mallal S. Assessment of precision and concordance of quantitative mitochondrial DNA assays: a collaborative international quality assurance study. Journal of Clinical Virology 2003; 27:97110.
Received 10 January 2005, accepted 19 April 2005

1st Meeting on Mitochondrial Toxicity & HIV Infection: Understanding the Pathogenesis for a Therapeutic Approach M11
Nucleoside reverse transcriptase inhibitors, mitochondrial DNA and AIDS therapy

William Lewis
Department of Pathology, Emory University, Atlanta, GA, USA Tel: +1 404 712 9005; Fax: +1 404 712 9007; E-mail: wlewis@emory.edu
Introduction and overview

New, more effective antiretroviral therapeutic agents [1] and promise from HIV vaccine studies [2] have not prevented the AIDS epidemics global spread. Nucleoside reverse transcriptase inhibitors (NRTIs) in combinations called highly active antiretroviral therapy (HAART) are cornerstones of AIDS therapy in the developed world. Their extensive use has brought serious side effects to light that appear to relate to mitochondrial dysfunction. This review addresses some mitochondrial changes attributed to NRTI therapy as a brief overview of some of the relevant literature, with emphasis on the prevailing theories. As alluded to above, the number of publications that relate to AIDS and mitochondrial dysfunction have grown rapidly since the original clinical and basic observations in the latter part of the 20th century [39]. Increased clinical interest in the entity coincided with increased frequency of side effects and increased patient survival. A number of reviews addressed different aspects of the problem from basic [1019], clinical laboratory [2022] and patient care [2326] perspectives. Table 1 is an annotated list of some of the reports in the literature that have addressed issues of mitochondrial toxicity (MT) of individual NRTIs, target organs, basic and clinical findings, and diagnostic approaches to the clinical problem. The subcellular pharmacological mechanism or mechanisms of MT are incompletely understood, but are generally believed to relate (at least in part) to inhibition of DNA polymerase- (DNA pol-; the enzyme responsible for mtDNA replication in eukaryotes [27]) by the phosphorylated NRTIs. Enzyme kinetic changes in mtDNA with NRTI toxicity have been addressed by us [16,17] and other investigators [18,19]. Clinical and experimental findings related to MT from NRTI therapy with zidovudine (AZT) and related compounds were first presented in the latter part of the last century [35,8,9,28]. Hallmark features of mtDNA depletion and energy depletion related clinical [3,4,16] and in vivo [9,28,29] experimental findings to inhibition kinetics with DNA pol- and phosphorylated NRTIs in vitro [5,10,3034]. Other observations suggested that mitochondrial energy deprivation is concomitant with, or the result of, mitochondrial oxidative stress in AIDS (for example, from HIV itself) or from NRTI therapy with its mitochondrial effects. In vivo studies with NRTI treatment of inbred mice [29,35] support this latter part of the hypothesis and data from our group and others employing transgenic mice revealed that oxidative stress results from transgenic expression of HIV Tat in the heart or liver [3638]. The above-mentioned processes may result from oxidative mtDNA damage, aberrant mtDNA replication and/or altered mtRNA transcription. These events are cornerstones of the mitochondrial dysfunction hypothesis [16,17] and remain foci in my laboratorys studies of toxicity of NRTIs and events in cardiomyopathy (CM) in AIDS [8,3941]. These principles apply to MT in other tissue targets demonstrated experimentally [32,4244].
M13
Predisposition to NRTI toxicity

Presently, there is limited understanding of genetic predispositions to NRTI toxicity or associated somatic mutations that may have pharmacogenetic implications. Such information may be crucial to complete understanding of toxicological mechanisms of NRTIs in humans and, conversely, to rationally guide effective therapeutic strategies in patients receiving HAART.
Mitochondrial DNA replication defects

Mitochondrial DNA (mtDNA) replication defects, including mtDNA depletion in target (or possibly surrogate) tissues, are frequently but not universally observed in both experimental and clinical reports.
Table 1. Published reports of the mitochondrial toxicity of individual NRTIs

Clinical evidence for MT Mitochondrial myopathy; ragged red fibres; decreased mtDNA, paracrystals; phosphocreatine depletion [3,4,24,52,6471] Polymyositis versus AZT myopathy: HIV vs NRTIs [205,206] AIDS myopathy not AZT-related [207,208] Cardiomyopathy with cardiac dilatation and failure; mitochondrial cristae dissolution; elevated serum lactate; Reyes syndrome [11,40,148151] Hepatomegaly, steatosis, mitochondrial ultrastructural change [66,144147] Paediatric AIDS patients are susceptible to NRTI toxicity [156] Paediatric AIDS patients are resistant to NRTI toxicity [13,67,152155] MT to cardiovascular tissues in utero [145,157161,209211] Bone marrow toxicity of NRTIs [212] Depletion of mtDNA in fat [20,162] Salutary effect of AZT on mtDNA [213] mtDNA mutations result from NRTIs [143] Worsening of mitochondrial genetic illness by NRTIs [214] Lactic acidosis from NRTIs [130142] Combinations cause greater mtDNA depletion than single NRTIs [215] Hypertension and MT [216] Mitochondrial phenotype from NRTIs without mtDNA depletion [217] Reviews Mechanisms of NRTI toxicity that focus on mtDNA replication [1316,18,33,47,62,69,103111] Absence of mitochondrial toxicity [112] Questioning the relationship to mtDNA [218] DNA pol- hypothesis Decreased mtDNA, mtRNA, mitochondrial polypeptides and mitochondrial ultrastructural damage in vivo and in vitro. Low Ki for AZT-TP with DNA pol-. Failure of exonucleolytic excision of terminally incorporated AZT. Mixed (competitive and non-competitive) Ki with cardiac DNA pol- against various templates [5,79,28,3033,40,42,81,86,113123] Biochemical/cellular mechanisms proposed for DNA pol- inhibition [14,16,18,19,106,219222] Mechanism of NRTI toxicity unrelated to NRTI triphosphate [223] Biochemical and cell biological findings AZT is toxic to skeletal and cardiac muscle in vivo [224] AZT depletes mtDNA and releases lactate in vitro [225] AZT causes oxidative stress [29,209] AZT depletes glutathione [129] AZT inhibits adenylate kinase [125] AZT inhibits adenine nucleotide translocator [23,126] AZT inhibits NADH-cytochrome c reductase [226] AZT inhibits mitochondrial permeability transition and apoptosis [223,227] AZT inhibits NADH oxidase [127] AZT effects on human deoxynucleotide carrier in vitro [100] AZT changes cellular nucleotide pools [84,228] AZT inhibits protein glycosylation [82] AZT inhibits OXPHOS early [119] AZT has no effect on mitochondria [229] AZT causes hepatic mitochondrial defects in vivo [117] Uridine abrogates NRTI toxicity in vitro [230] NRTIs decrease mtDNA replication in rats and humans [180,231] HepG2 cells are excellent models for NRTI dysfunction [232] Other findings NRTIs have no effects on mtDNA depletion [109] Cytoplasmic bystander gene effect [128] Experimental evidence for MT
W Lewis
M14
NRTI
Anatomic or tissue target
AZT
Liver, heart, skeletal muscle and bone marrow
Table 1. Continued
Painful peripheral neuropathy [184189] ddC inhibits mtDNA replication in cells; causes mitochondrial structural changes and lipid accumulation in nerves of ddC-treated rabbits [57,33,81,115,168,179,181,190194]
ddC
Peripheral nerve
3TC Favourable Km with gapped duplex DNA [233,234] Limited MT of carbovir in vitro [235] Cytofluorometry in vitro to determine pancreatitis [201] ddI has related hepatotoxicity [236] ddI caused fatal hepatic failure in chronic hepatitis [237] ddI+tenofovir caused fatal lactic acidosis [178]
3TC muscle toxicity [195]
Carbovir
ddI
Pancreas
d4T
Peripheral nerve
Distorted cristae and decreased mtDNA in CEM cells [33,168,169] d4T mitochondrial neuropathy in vivo [170] d4T biochemical changes with DNA pol- [171] d4T hepatoxicity abrogated by antioxidant porphyrin [124]
Liver Hypersensitivity reactions prominent [241243] Mitochondrial side effects not clear; pancreatitis with co-administration of ddI may relate to MT [244,245] ()FTC has favourable safety profile [196] Lactic acidosis; hepatic failure and steatosis; renal failure; skeletal and cardiac myopathy; peripheral neuropathy; death [25,93,141,246251] Lactic acidosis related to nitric oxide [252] Absence of elevated lactate with NRTI therapy [253] Newly developed antiretroviral for salvage [202]; plasma levels established [262]; safety concerns stop clinical trial [263,264]
Painful peripheral neuropathy [163165] Lack of d4T neuropathy [238] Pathological neuropathic features with AZT administration [238] Treatment of neuropathy with acetyl-L-carnitine suggests MT [166,167] Liver toxicity from NRTIs [239]; evaluation by transmission EM [240]
Abacavir
Tenofovir
()FTC
()FTC kinetics with DNA pol- favour efficacy [197]; no effect on HepG2 mtDNA replication [198] FIAU experimental studies demonstrate toxicity [25,4244,204,254261]: FIAU incorporates into mtDNA in vivo and in vitro; mitochondrial structural defects and intracellular fat accumulates in vitro and in vivo; competitive Ki of FIAU-TP, FMAU-TP with DNA pol- (0.02 M) Potent antiretroviral activity [265] Pharmacokinetics of FddA [266] Acute cardiotoxicity in vivo [203] Pharmacological manufacture of FddA [267]; primate pharmacokinetics [268] Plasma lactate correlates with NRTI toxicity and cardiomyopathy with AZT-based HAART [39]
FIAU
Liver, skeletal and cardiac muscle, peripheral nerve
FddA
All
Clinical testing with surrogate marker
Blood lactate/pyruvate ratios are useful [5456] Peripheral blood depletion mtDNA correlates NRTI effects [22,4653] Peripheral blood mtDNA does not correlate with NRTI effects [72,73] Surrogate testing for NRTI toxicity unclear [21,7477] Lipodystrophy correlates with mtDNA depletion in affected tissues [5763] PBL mtDNA does not correlate with lipodystrophy in children with NRTIs [182]
NRTIs, mtDNA and AIDS therapy
M15
3TC, lamivudine; AZT, zidovudine; d4T, stavudine; ddC, zalcitabine; DNA pol-, DNA polymerase-; ddI, didanosine; EM, electron microscopy; FddA, iodenosine; FIAU, fialuridine; ()FTC, emitracitabine; HAART, highly active antiretroviral therapy; MT, mitochondrial toxicity; mtDNA, mitochondrial DNA; NADH, nicotinamide adenine dinucleotide; NRTI, nucleoside reverse transcriptase inhibitor; OXPHOS, oxidative phosphorylation; PBL, peripheral blood lymphocyte; -TP, trisphosphate.
W Lewis
Assessing NRTI toxicity

It is possible to make analogies between studies that address mechanisms of NRTI-induced MT and those that examine defects in genetic mitochondrial illnesses in which defective mitochondrial gene product, oxidative stress and environmental factors contribute to disease [45]. This has been the approach utilized by many clinical studies in which mtDNA depletion in surrogate tissues (for example, blood cells) [22,4653] and lactic acidaemia [5456] support NRTI MT. Some studies indicated depleted mtDNA in target tissue occurred, particularly in fat tissue and skeletal muscle of AIDS patients treated with NRTIs [3,4,24,52,5771]. Others suggested that mtDNA depletion in surrogate tissues was not a useful marker [72,73] or emphasized the intrinsic importance of sample handling and data interpretation as potential pitfalls that could confound proper interpretation of results [21,7477]. The DNA pol- hypothesis follows principles of mitochondrial medicine [78]. If the intramitochondrial pool of native nucleotides is disrupted by NRTI treatment, altered energetics may result from alterations in expression of electron transport proteins encoded by mtDNA. In another system, the importance of mitochondrial alterations in the development of low output congestive heart failure [79] related mitochondrial and cardiac dysfunction. Genetic mitochondrial illnesses manifest with clinical findings that occur at a threshold that is based in part on the heteroplasmic effects of the associated mtDNA mutation and the impact of oxidative phosphorylation (OXPHOS) on tissue function according to the OXPHOS paradigm [80]. Energy deprivation, possibly the initial phenotypic step of NRTI toxicity (based on mtDNA depletion), relates decreased energy abundance in tissues (for example, heart, as in our studies [8,39,40]) to decreased mitochondrial function. Such a threshold underscores phenotypic change, particularly in tissues like myocardium. It should be understood that this might not be the only mechanism utilized or available to explain toxicity of NRTI in mitochondria. Other mechanisms have been suggested that relate to mtDNA replication [81] or other aspects of mitochondrial and cellular homeostasis [82,83] and are addressed below.
Nucleotide pools, NRTIs and mtDNA replication

An overview of some aspects of NRTI cellular pharmacology is offered here, but it is not exhaustive. On a mass action basis, sufficient intramitochondrial NRTI concentration is required to affect mtDNA replication. Specifically, NRTI triphosphate must compete with the native moiety at the nucleotide-binding site of DNA
M16
pol- to yield inhibition of mtDNA replication. On a biochemical basis, the important active pharmacological (and toxic) element of AZT is the triphosphate (AZTTP). AZT-TP inhibits both HIV reverse transcriptase [84,85] and mammalian DNA pol- in vitro [32]. It follows that stoichiometrically sufficient phosphorylated NRTI must be made available intramitochondrially for inhibition of mtDNA synthesis, subsequent depletion of mtDNA [86] and development of toxic manifestations. The various mechanisms of internalization of nucleosides, their transport and the homeostasis of mitochondrial nucleotide pools are key elements, as is the phosphorylation of NRTIs. To date, the intramitochondrial concentration of NRTIs has been difficult to determine. AZT-like NRTIs are phosphorylated in three intracellular steps. Thymidine kinase (TK) phosphorylates AZT to AZT-MP. Thymidylate kinase phosphorylates AZT-MP to AZT-DP. Nucleoside diphosphate kinase yields the active AZT-TP [87] from AZT-DP. At each step, phosphatases exist to maintain homeostasis. Accordingly, a key step in the pathophysiology of NRTI pharmacology and toxicity is regulation of natural deoxyribonucleotide triphosphate (dNTP) and NRTI triphosphate pool sizes in mitochondria that affect mtDNA replication dysregulation of phosphorylation and dephosphorylation could impact mtDNA replication. TK exists in isoforms: TK1 (cytosolic isoform) has relatively low activity in extracts of striated skeletal muscle [88,89], while TK2 activity is higher. TK2 is reported to have a broader substrate range and phosphorylates deoxycytidine (dCyd) and 5-substituted deoxythymidine (dThd) and dCyd analogues [90]. Dideoxynucleotides (ddNTPs) function as either competitive inhibitors of the natural substrates of polymerases (reviewed in [91]) or lead to chain termination [85,92]. Since mtDNA depletion is a hallmark of treatment with AZT in skeletal muscle of humans and rodents [3,9], and fialuridine (FIAU; 1-[2-deoxy-2-fluoro--D-arabinofuranosyl]-5-iodouracil) causes mtDNA depletion in woodchucks and humans [12,93] (see below), it is possible that genetic models exist to support the working hypothesis. TK2 mutations represent an aetiology for mtDNA depletion and have been associated syndromically with that genetic finding. Two substitution mutations in TK2 (His90Asn and Ile181Asn), resulted in a phenotype of infantile myopathy and mtDNA depletion in muscle [94]. In contrast to its cytoplasmic counterpart, mitochondrial ribonucleotide reductase is not well documented, so import of deoxyribonucleosides and their phosphorylated products (or the analogous NRTIs) into mitochondria must occur in mammalian tissues. dNTPs synthesized by the cytosolic ribonucleotide reductase can be imported directly through the mitochondrial membrane [95]. Import of deoxynucleosides into
mitochondria allows for subsequent phosphorylation by mitochondrially localized kinases [96] to nucleotides. Steady-state abundance is balanced by mitochondrially localized dephosphorylations by phosphatases [9799]. Import of NRTIs or NRTI phosphates could alter the stoichiometry of the intramitochondrial pool of native nucleotides [100].
suggested (including, for example, glutathione (GSH) depletion [129]).
Zidovudine (AZT)
AZT was the first NRTI antiretroviral used in the treatment of AIDS and affords the greatest toxicological experience. Both in clinical [3,4,24,52,6471] and experimental studies [1316,18,33,47,62,69,103111], AZT has been implicated in the development of mitochondrial diseases with features of myopathy, ragged red fibres, decreased mtDNA and defective mtDNA replication. AZT has worsened mitochondrial genetic illnesses, been implicated in the genesis of lactic acidosis [130142] and has caused mtDNA mutations [143]. With respect to toxicity in various target tissues, observationally based clinical correlates were made in some of the early studies in which AZT liver toxicity was associated with obesity and female gender [116,133]. Refined genetic correlates were lacking. NRTI toxicity presents a variable and complex diagnostic phenotype in the treated population and mimics key features of mitochondrial diseases. NRTI toxicity may serve as an important model system for relevant pharmacogenetic studies. Such a pharmacologically based review is beyond the scope of this work, but has been addressed elsewhere [11,16,17,106]. Various organs and tissues have been described as susceptible (and resistant) to AZT toxicity. Table 1 enumerates some of these, including the liver. In ways that resemble FIAU (described below) hepatomegaly, steatosis and mitochondrial ultrastructural change [66,144147] have been documented with AZT. It should also be noted that hepatic toxicity from didanosine (ddI) and zalcitabine (ddC) has been reported [64,66,67]. The toxic mechanism is presumed to relate to liver mitochondria. Fatal hepatomegaly with severe steatosis [66], severe lactic acidosis [67] and adult Reyes syndrome [64] in AZT-treated HIV-seropositive patients were all pathogenetically linked to AZT-induced hepatotoxicity. Clinical features resembled some of those seen in FIAU toxicity (below). The cardiovascular system effects of AZT include CM with cardiac dilatation and failure, mitochondrial cristae dissolution and elevated serum lactate [11,40,148151]. Conflicting reports have suggested resistance [13,67, 152155] or susceptibility to toxicity from AZT [156], but MT to cardiovascular tissues has been documented in primates and in utero [145,157161] with AZT and other NRTIs. AZT therapy more recently has been implicated in lipodystrophy in AIDS patients [20,162].
Overview of current NRTI therapy

NRTIs used to treat HIV infection [101] include zidovudine (AZT; 3-azido-2,3-deoxythymidine), zalcitibine (ddC; 2,3-dideoxycytidine), didanosine (ddI; 2,3dideoxyinosine), stavudine (d4T; 2,3-didehydro-3dideoxythymidine), lamivudine (3TC; 3-thiacytidine; cis1-[2-hydroxymethyl-5-(1,3-oxathiolanyl)]cytosine), emtricitabine [()FTC], tenofovir and abacavir (these latter compounds are not NRTIs, but related moieties). Because a large number of NRTIs are currently available and most NRTIs are administered in HAART, resultant combinations vary widely in their constituents [102]. This amplifies the complexity of the problem and makes interpretation difficult. Controls are not always available for some observations in clinical populations. From a commercial standpoint, NRTIs serve an important market with features that include significant long-term patient use of therapy, a growing patient population and absence of competition from preventative or curative vaccines in the near future. Because of this, the search for new NRTIs has led to development of many important compounds with therapeutic effectiveness but also some that exhibited profound clinical toxicity (addressed below). Additionally, the diagnosis, identification, prediction and amelioration of mitochondrial toxic effects have generated increasing attention.
Mechanisms of toxicity
As mentioned previously, the prevailing theory suggests AZT-induced MT involves defective mtDNA replication (reviewed in [1316,18,33,47,62, 69,103111]). It should be noted that the hypothesis is not universally accepted [112] and alternatives may prove to be promising as data are published. On a biochemical basis, decreased mtDNA, mtRNA, mitochondrial polypeptides and defective mitochondrial ultrastructure, correlate with micromolar, mixed Kis for dideoxy-NRTI triphosphates in various experimental systems [5,79,28,3033,40,42,81,86,113124]. Some other explanations for MT from AZT and NRTIs include inhibition of adenylate kinase [125], adenine nucleotide translocator [23,126], NADH oxidase [127], protein glycosylation [82] and a bystander effect [128]. It should be noted that other mechanisms unrelated to mtDNA replication have also been
Stavudine (d4T)
d4T emerged as a first-line HAART component. A number of documented MTs [145,157161] were
M17
W Lewis
attributed to d4T (see Table 1). Like ddC (below), painful peripheral neuropathy [163165] has been described with d4T. One study suggested a lack of such a relationship. A clinical proof of principle used treatment of d4T mitochondrial neuropathy with acetyl-L-carnitine. The trial was considered successful and suggested that MT was mechanistically related to d4T effects on mitochondria [166,167]. Preclinical and basic studies further support MT of d4T. Distorted cristae and decreased mtDNA in CEM cells occurred with d4T exposure [33,168,169] and d4T mitochondrial neuropathy was generated in vivo [170]. Kinetics of inhibition with d4T and DNA pol- [171] resulted in a nanomolar Ki [124]. Clinical treatment with certain NRTIs (d4T/3TC) results in anion gap acidosis. Moreover, the lactic acidosis/hepatic steatosis syndrome may be more common than previously appreciated in adults and children treated with NRTIs. d4T treatment has also been associated with lipodystrophy [172]. Mechanisms were posited to involve altered mitochondrial biogenesis and/or oxidative changes, and adipocyte apoptosis [11,173]. d4T and other NRTIs have been suggested to relate to development of lactic acidaemia. Some patients treated with NRTIs experienced lactic acidaemia [67,93,133, 135,145,154,174178] and a phenotype of mtDNA depletion [46,123]. Depending on the biological system employed in the study [21,63,162,179, 180], deleterious effects on mitochondrial structure and function in selected targets have been documented [8,11,44,181]. Although the specificity of blood cell mtDNA depletion as a surrogate marker for NRTI toxicity has been documented in some studies [58,156,162,182,183], the impact of the studies was confounded by the control groups used [50]. In principle, mtDNA depletion is mechanistically consistent with NRTI toxicity. However, the impact of mitochondrial dysfunction in surrogate tissues remains unclear [52], even in the face of a logical working hypothesis. Methods for diagnosis usually include examination of plasma lactate or lactate/pyruvate ratios [50,51], but require careful sample preparation and handling to assure meaningful results and interpretations. Overall, depletion of mtDNA appears to be accepted as an important marker of the toxic process, and may even serve as a diagnostic hallmark [3,9] to monitor successful HAART therapy [46]. It should be emphasized that the ideal surrogate tissue to monitor mtDNA depletion from NRTIs remains to be determined and is an active focus of clinical research and commercial enterprise.
has been associated with clinical ddC toxicity [184189], and inhibition of mtDNA replication has been observed in vitro and in vivo [57,33,81,115, 168,179,181,190194].
Lamivudine (3TC) and emitracibine [()FTC]

3TC is widely used in HAART and ()FTC is becoming an important therapeutic tool since its recent addition to the armamentarium. Of the commonly used NRTIs, 3TC appears to have a favourable safety profile, but like the others, requires co-administration in a HAART regimen, particularly because of induction of HIV resistance mutations. Toxicity to muscle is reported clinically with 3TC [195], but basic evidence for toxicity of 3TC monotherapy is lacking in our in vivo systems (Lewis et al., unpublished). Newly approved ()FTC exhibits a relatively favourable safety profile [196] and kinetics with DNA pol- favour efficacy [197]. Studies with HepG2 cells also are supportive [198] of the safety of ()FTC. It remains to be seen if long-term toxicity is to occur.
Didanosine (ddI)
ddI remains an important element in HAART. Two principal clinical toxicities have been recognized. As with ddC, a painful peripheral neuropathy has been documented with ddI therapy in humans [199,200]. Early in the development of ddI, severe pancreatitis was identified as an important side effect with mortality [199]. Experimental work documented pancreatic changes by flow cytometry [201]. Fatal hepatotoxicity was described and lactic acidosis has occurred with coadministration of tenofovir.
Other NRTIs
One more recent and serious adverse event occurred with a halogenated purine. Fluoro-dideoxyadenosine (FddA; 2-fluoro-2,3-dideoxyadenosine) went into clinical trial but was discontinued about 1 year later because it exhibited severe adverse events including profound lactic acidosis [202,203]. A second serious adverse event occurred with a halogenated pyrimidine in a hepatitis B trial. This tragic event occurred at the Clinical Center at the National Institutes of Health. After in vitro studies documented significant efficacy, the pyrimidine nucleoside analogue FIAU went to clinical trial and experienced promising results early on. FIAU was later found in the trial to be extremely toxic to liver, skeletal and cardiac muscle, pancreas and peripheral nerves in treated patients. MT from FIAU was profound. Lactic acidosis and hepatic failure required heroic clinical interventions and necessitated
Zalcitibine (ddC)
ddC is less popular today; nonetheless, painful peripheral neuropathy attributed to mitochondrial dysfunction
M18
early termination of the protocol, but some deaths occurred. Abandonment of these compounds as pharmacological agents was subsequently confirmed by documenting MT in animal models [43,204] and in humans [93] and defining inhibition kinetics in vitro with DNA pol- that favoured toxicity to mitochondria [42,44]. These tragic trials underscore the necessity for examining MT of NRTIs extensively using in vitro and in vivo testing.
8.
9.
10. 11. 12. 13.
Summary
NRTIs are cornerstones of antiretroviral therapy and perhaps the most important drugs developed for AIDS treatment. Judicious use of NRTIs in the fight against HIV infection has afforded significant clinical advances in AIDS treatment. Nonetheless, it is axiomatic to expect side effects from NRTIs as well. A principal toxicity of NRTIs relates to chronic and cumulative MT in various tissues. The long-term impact of this toxicity on affected patients is not clear except in extreme cases where morbidity was severe and mortality occurred. Because of chronicity and potential for severity, NRTI MT remains an important clinical problem. Further studies will help us unravel mechanisms of NRTI MT and the natural history of mitochondrial biogenesis in humans and other mammalian systems.
14. 15.
16.
17. 18.
Acknowledgments
19.
Supported by DHHS, NIH HL072707 and HL063666.
References
1. Condra JH, Miller MD, Hazuda DJ & Emini EA. Potential new therapies for the treatment of HIV-1 infection. Annual Review of Medicine 2002; 53:541555. Graham BS. Clinical trials of HIV vaccines. Annual Review of Medicine 2002; 53:207221. Arnaudo E, Dalakas M, Shanske S, Moraes CT, DiMauro S & Schon EA. Depletion of muscle mitochondrial DNA in AIDS patients with zidovudine-induced myopathy. Lancet 1991; 337:508510. Dalakas MC, Illa I, Pezeshkpour GH, Laukaitis JP, Cohen B & Griffin JL. Mitochondrial myopathy caused by longterm zidovudine therapy. New England Journal of Medicine 1990; 322:10981105. Chen CH & Cheng YC. Delayed cytotoxicity and selective loss of mitochondrial DNA in cells treated with the antihuman immunodeficiency virus compound 2,3-dideoxycytidine. Journal of Biological Chemistry 1989; 264:1193411937. Chen CH & Cheng YC. The role of cytoplasmic deoxycytidine kinase in the mitochondrial effects of the anti-human immunodeficiency virus compound, 2,3-dideoxycytidine. Journal of Biological Chemistry 1992; 267:28562859. Chen CH, Vazquez-Padua M & Cheng YC. Effect of antihuman immunodeficiency virus nucleoside analogs on mitochondrial DNA and its implication for delayed toxicity. Molecular Pharmacology 1991; 39:625628.
20.
21. 22.
2. 3.
4.
23. 24. 25. 26. 27.
5.
6.
7.
Lewis W, Papoian T, Gonzalez B, Louie H, Kelly DP, Payne RM & Grody WW. Mitochondrial ultrastructural and molecular changes induced by zidovudine in rat hearts. Laboratory Investigation 1991; 65:228236. Lewis W, Gonzalez B, Chomyn A & Papoian T. Zidovudine induces molecular, biochemical, and ultrastructural changes in rat skeletal muscle mitochondria. Journal of Clinical Investigation 1992; 89:13541360. Wright GE & Brown NC. Deoxyribonucleotide analogs as inhibitors and substrates of DNA polymerases. Pharmacology & Therapeutics 1990; 47:447497. Lewis W & Dalakas MC. Mitochondrial toxicity of antiviral drugs. Nature Medicine 1995; 1:417422. Swartz MN. Mitochondrial toxicity new adverse drug effects. New England Journal of Medicine 1995; 333:10991105. Brinkman K, Smeitink JA, Romijn JA & Reiss P. Mitochondrial toxicity induced by nucleoside-analogue reverse-transcriptase inhibitors is a key factor in the pathogenesis of antiretroviral-therapy-related lipodystrophy. Lancet 1999; 354:11121115. Kakuda TN. Pharmacology of nucleoside and nucleotide reverse transcriptase inhibitor-induced mitochondrial toxicity. Clinical Therapeutics 2000; 22:685708. Kakuda TN, Brundage RC, Anderson PL & Fletcher CV. Nucleoside reverse transcriptase inhibitor-induced mitochondrial toxicity as an etiology for lipodystrophy. AIDS 1999; 13:23112312. Lewis W, Copeland WC & Day B. Mitochondrial DNA depletion, oxidative stress and mutation: mechanisms of nucleoside reverse transcriptase inhibitor toxicity. Laboratory Investigation 2001; 81:777790. Lewis W, Day BJ & Copeland WC. Mitochondrial toxicity of NRTI antiviral drugs: an integrated cellular perspective. Nature Reviews Drug Discovery 2003; 2:812822. Johnson AA, Ray AS, Hanes J, Suo Z, Colacino JM, Anderson KS & Johnson KA. Toxicity of antiviral nucleoside analogs and the human mitochondrial DNA polymerase. Journal of Biological Chemistry 2001; 276:4084740857. Anderson KS. Perspectives on the molecular mechanism of inhibition and toxicity of nucleoside analogs that target HIV-1 reverse transcriptase. Biochimica & Biophysica Acta 2002; 1587:296299. Cossarizza A, Mussini C & Vigano A. Mitochondria in the pathogenesis of lipodystrophy induced by anti-HIV antiretroviral drugs: actors or bystanders? Bioessays 2001; 23:10701080. Cossarizza A, Troiano L & Mussini C. Mitochondria and HIV infection: the first decade. Journal of Biological Regulators & Homeostatic Agents 2002; 16:1824. Montaner JS, Cote HC, Harris M, Hogg RS, Yip B, Chan JW, Harrigan PR & OShaughnessy MV. Mitochondrial toxicity in the era of HAART: evaluating venous lactate and peripheral blood mitochondrial DNA in HIV-infected patients taking antiretroviral therapy. Journal of Acquired Immune Deficiency Syndromes 2003; 34(Suppl 1):S85S90. Barile M, Valenti D, Quagliariello E & Passarella S. Mitochondria as cell targets of AZT (zidovudine). General Pharmacology 1998; 31:531538. Cherry CL, McArthur JC, Hoy JF & Wesselingh SL. Nucleoside analogues and neuropathy in the era of HAART. Journal of Clinical Virology 2003; 26:195207. Colacino JM. Mechanisms for the anti-hepatitis B virus activity and mitochondrial toxicity of fialuridine (FIAU). Antiviral Research 1996; 29:125139. Herman JS & Easterbrook PJ. The metabolic toxicities of antiretroviral therapy. International Journal of STD & AIDS 2001; 12:555562; quiz 563564. Kaguni LS. DNA polymerase gamma, the mitochondrial replicase. Annual Review of Biochemistry 2004; 73:293320.
M19
W Lewis
28. Lamperth L, Dalakas MC, Dagani F, Anderson J & Ferrari R. Abnormal skeletal and cardiac muscle mitochondria induced by zidovudine (AZT) in human muscle in vitro and in an animal model. Laboratory Investigation 1991; 65:742751. 29. de la Asuncion JG, del Olmo ML, Sastre J, Millan A, Pellin A, Pallardo FV & Vina J. AZT treatment induces molecular and ultrastructural oxidative damage to muscle mitochondria. Prevention by antioxidant vitamins. Journal of Clinical Investigation 1998; 102:49. 30. Izuta S, Saneyoshi M, Sakurai T, Suzuki M, Kojima K & Yoshida S. The 5-triphosphates of 3-azido-3-deoxythymidine and 2, 3-dideoxynucleosides inhibit DNA polymerase gamma by different mechanisms. Biochemical & Biophysical Research Communications 1991; 179:776783. 31. Konig H, Behr E, Lower J & Kurth R. Azidothymidine triphosphate is an inhibitor of both human immunodeficiency virus type 1 reverse transcriptase and DNA polymerase gamma. Antimicrobial Agents & Chemotherapy 1989; 33:21092114. 32. Lewis W, Simpson JF & Meyer RR. Cardiac mitochondrial DNA polymerase-gamma is inhibited competitively and noncompetitively by phosphorylated zidovudine. Circulation Research 1994; 74:344348. 33. Martin JL, Brown CE, Matthews-Davis N & Reardon JE. Effects of antiviral nucleoside analogs on human DNA polymerases and mitochondrial DNA synthesis. Antimicrobial Agents & Chemotherapy 1994; 38:27432749. 34. Parker WB & Cheng YC. Mitochondrial toxicity of NRTI analogs. Journal of NIH Research 1994; 6:5761. 35. Bialkowska A, Bialkowski K, Gerschenson M, Diwan BA, Jones AB, Olivero OA, Poirier MC, Anderson LM, Kasprzak KS & Sipowicz MA. Oxidative DNA damage in fetal tissues after transplacental exposure to 3-azido-3deoxythymidine (AZT). Carcinogenesis 2000; 21:10591062. 36. Choi J, Opalenik SR, Wu W, Thompson JA & Forman HJ. Modulation of glutathione synthetic enzymes by acidic fibroblast growth factor. Archives of Biochemistry & Biophysics 2000; 375:201209. 37. Choi J, Liu RM, Kundu RK, Sangiorgi F, Wu W, Maxson R & Forman HJ. Molecular mechanism of decreased glutathione content in human immunodeficiency virus type 1 Tat-transgenic mice. Journal of Biological Chemistry 2000; 275:36933698. 38. Raidel SM, Haase C, Jansen NR, Russ RB, Sutliff RL, Velsor LW, Day BJ, Hoit BD, Samarel AM & Lewis W. Targeted myocardial transgenic expression of HIV Tat causes cardiomyopathy and mitochondrial damage. American Journal of Physiology. Heart & Circulatory Physiology 2002; 282:H1672H1678. 39. Lewis W, Haase CP, Raidel SM, Russ RB, Sutliff RL, Hoit BD & Samarel AM. Combined antiretroviral therapy causes cardiomyopathy and elevates plasma lactate in transgenic AIDS mice. Laboratory Investigation 2001; 81:15271536. 40. Lewis W, Grupp IL, Grupp G, Hoit B, Morris R, Samarel AM, Bruggeman L & Klotman P. Cardiac dysfunction occurs in the HIV-1 transgenic mouse treated with zidovudine. Laboratory Investigation 2000; 80:187197. 41. Lewis W & Grody WW. AIDS and the heart: review and consideration of pathogenetic mechanisms. Cardiovascular Pathology 1992; 1:5364. 42. Lewis W, Meyer RR, Simpson JF, Colacino JM & Perrino FW. Mammalian DNA polymerases alpha, beta, gamma, delta, and epsilon incorporate fialuridine (FIAU) monophosphate into DNA and are inhibited competitively by FIAU triphosphate. Biochemistry 1994; 33:1462014624. 43. Lewis W, Griniuviene B, Tankersley KO, Levine ES, Montione R, Engelman L, de Courten-Myers G, Ascenzi MA, Hornbuckle WE, Gerin JL & Tennant BC. Depletion of mitochondrial DNA, destruction of mitochondria, and
M20
44.
45. 46.
47.
48.
49.
50. 51. 52. 53.
54.
55. 56.
57.
58.
accumulation of lipid droplets result from fialuridine treatment in woodchucks (Marmota monax). Laboratory Investigation 1997; 76:7787. Lewis W, Levine ES, Griniuviene B, Tankersley KO, Colacino JM, Sommadossi JP, Watanabe KA & Perrino FW. Fialuridine and its metabolites inhibit DNA polymerase gamma at sites of multiple adjacent analog incorporation, decrease mtDNA abundance, and cause mitochondrial structural defects in cultured hepatoblasts. Proceedings of the National Academy of Sciences, USA 1996; 93:35923597. Schapira AH & Cooper JM. Mitochondrial function in neurodegeneration and ageing. Mutation Research 1992; 275:133143. Cote HC, Brumme ZL, Craib KJ, Alexander CS, Wynhoven B, Ting L, Wong H, Harris M, Harrigan PR, OShaughnessy MV & Montaner JS. Changes in mitochondrial DNA as a marker of nucleoside toxicity in HIV- infected patients. New England Journal of Medicine 2002; 346:811820. Walker UA, Setzer B & Venhoff N. Increased long-term mitochondrial toxicity in combinations of nucleoside analogue reverse-transcriptase inhibitors. AIDS 2002; 16:21652173. Cote HC, Yip B, Asselin JJ, Chan JW, Hogg RS, Harrigan PR, OShaughnessy MV & Montaner JS. Mitochondrial: nuclear DNA ratios in peripheral blood cells from human immunodeficiency virus (HIV)-infected patients who received selected HIV antiretroviral drug regimens. Journal of Infectious Diseases 2003; 187:19721976. Montaner JS, Cote HC, Harris M, Hogg RS, Yip B, Harrigan PR & OShaughnessy MV. Nucleoside-related mitochondrial toxicity among HIV-infected patients receiving antiretroviral therapy: insights from the evaluation of venous lactic acid and peripheral blood mitochondrial DNA. Clinical Infectious Diseases 2004; 38(Suppl 2):S73S79. Rastegar DA. Mitochondrial DNA and nucleoside toxicity. New England Journal of Medicine 2002; 347:216218. Author reply. Chariot P, Bourokba N & Brivet F. Mitochondrial DNA and nucleoside toxicity. New England Journal of Medicine 2002; 347:216218. Author reply. Casademont J, Miro O & Cardellach F. Mitochondrial DNA and nucleoside toxicity. New England Journal of Medicine 2002; 347:216218. Author reply. Polo R, Martinez S, Madrigal P & Gonzalez-Munoz M. Factors associated with mitochondrial dysfunction in circulating peripheral blood lymphocytes from HIV-infected people. Journal of Acquired Immune Deficiency Syndromes 2003; 34:3236. Chariot P, Monnet I, Mouchet M, Rohr M, Lefaucheur JP, Dubreuil-Lemaire ML, Chousterman M & Gherardi R. Determination of the blood lactate: pyruvate ratio as a noninvasive test for the diagnosis of zidovudine myopathy. Arthritis & Rheumatism 1994; 37:583586. McComsey G & Lonergan JT. Mitochondrial dysfunction: patient monitoring and toxicity management. Journal of Acquired Immune Deficiency Sydromes 2004; 37:S30S35. ter Hofstede HJ, Willems HL & Koopmans PP. Serum Llactate and pyruvate in HIV-infected patients with and without presumed NRTI-related adverse events compared to healthy volunteers. Journal of Clinical Virology 2004; 29:4450. Pace CS, Martin AM, Hammond EL, Mamotte CD, Nolan DA & Mallal SA. Mitochondrial proliferation, DNA depletion and adipocyte differentiation in subcutaneous adipose tissue of HIV-positive HAART recipients. Antiviral Therapy 2003; 8:323331. Miro O, Lopez S, Pedrol E, Rodriguez-Santiago B, Martinez E, Soler A, Milinkovic A, Casademont J, Nunes V, Gatell JM & Cardellach F. Mitochondrial DNA depletion and respiratory chain enzyme deficiencies are present in peripheral blood mononuclear cells of HIV-infected patients with HAART-related lipodystrophy. Antiviral Therapy 2003; 8:333338.
59. Walker UA & Brinkman K. NRTI induced mitochondrial toxicity as a mechanism for HAART related lipodystrophy: fact or fiction? HIV Medicine 2001; 2:163165. 60. Nolan D, Moore C, Castley A, Sayer D, Mamotte C, John M, James I & Mallal S. Tumour necrosis factor-alpha gene 238G/A promoter polymorphism associated with a more rapid onset of lipodystrophy. AIDS 2003; 17:121123. 61. Nolan D, Hammond E, Martin A, Taylor L, Herrmann S, McKinnon E, Metcalf C, Latham B & Mallal S. Mitochondrial DNA depletion and morphologic changes in adipocytes associated with nucleoside reverse transcriptase inhibitor therapy. AIDS 2003; 17:13291338. 62. Walker UA, Bickel M, Lutke Volksbeck SI, Ketelsen UP, Schofer H, Setzer B, Venhoff N, Rickerts V & Staszewski S. Evidence of nucleoside analogue reverse transcriptase inhibitor-associated genetic and structural defects of mitochondria in adipose tissue of HIV-infected patients. Journal of Acquired Immune Deficiency Syndromes 2002; 29:117121. 63. Shikuma CM, Hu N, Milne C, Yost F, Waslien C, Shimizu S & Shiramizu B. Mitochondrial DNA decrease in subcutaneous adipose tissue of HIV-infected individuals with peripheral lipoatrophy. AIDS 2001; 15:18011809. 64. Jolliet P & Widmann JJ. Reyes syndrome in adult with AIDS. Lancet 1990; 335:1457. 65. Chariot P & Gherardi R. Partial cytochrome C oxidase deficiency and cytoplasmic bodies in patients with zidovudine myopathy. Neuromuscular Disorders 1991; 1:357363. 66. Freiman JP, Helfert KE, Hamrell MR & Stein DS. Hepatomegaly with severe steatosis in HIV-seropositive patients. AIDS 1993; 7:379385. 67. Chattha G, Arieff AI, Cummings C & Tierney LM Jr. Lactic acidosis complicating the acquired immunodeficiency syndrome. Annals of Internal Medicine 1993; 118:3739. 68. Casademont J, Barrientos A, Grau JM, Pedrol E, Estivill X, Urbano-Marquez A & Nunes V. The effect of zidovudine on skeletal muscle mtDNA in HIV-1 infected patients with mild or no muscle dysfunction. Brain 1996; 119:13571364. 69. Cherry CL & Wesselingh SL. Nucleoside analogues and HIV: the combined cost to mitochondria. Journal of Antimicrobial Chemotherapy 2003; 51:10911093. 70. Grau JM, Masanes F, Pedrol E, Casademont J, FernandezSola J & Urbano-Marquez A. Human immunodeficiency virus type 1 infection and myopathy: clinical relevance of zidovudine therapy. Annals of Neurology 1993; 34:206211. 71. Manji H, Harrison MJ, Round JM, Jones DA, Connolly S, Fowler CJ, Williams I & Weller IV. Muscle disease, HIV and zidovudine: the spectrum of muscle disease in HIVinfected individuals treated with zidovudine. Journal of Neurology 1993; 240:479488. 72. Chiappini F, Teicher E, Saffroy R, Pham P, Falissard B, Barrier A, Chevalier S, Debuire B, Vittecoq D & Lemoine A. Prospective evaluation of blood concentration of mitochondrial DNA as a marker of toxicity in 157 consecutively recruited untreated or HAART-treated HIV-positive patients. Laboratory Investigation 2004; 84:908914. 73. Hoy JF, Gahan ME, Carr A, Smith D, Lewin SR, Wesselingh S & Cooper DA. Changes in mitochondrial DNA in peripheral blood mononuclear cells from HIVinfected patients with lipoatrophy randomized to receive abacavir. Journal of Infectious Diseases 2004; 190:688692. 74. Cossarizza A. Tests for mitochondrial function and DNA: potentials and pitfalls. Current Opinion in Infectious Diseases 2003; 16:510. 75. Janes MS, Hanson BJ, Hill DM, Buller GM, Agnew JY, Sherwood SW, Cox WG, Yamagata K & Capaldi RA. Rapid analysis of mitochondrial DNA depletion by fluorescence in situ hybridization and immunocytochemistry: potential strategies for HIV therapeutic monitoring. Journal of Histochemistry & Cytochemistry 2004; 52:10111018.
76. Petit C, Mathez D, Barthelemy C, Leste-Lasserre T, Naviaux RK, Sonigo P & Leibowitch J. Quantitation of blood lymphocyte mitochondrial DNA for the monitoring of antiretroviral drug-induced mitochondrial DNA depletion. Journal of Acquired Immune Deficiency Syndromes 2003; 33:461469. 77. Henry K, Erice A, Balfour HH Jr, Schmeling M, Berthiaume J & Wallace K. Lymphocyte mitochondrial biomarkers in asymptomatic HIV-1-infected individuals treated with nucleoside reverse transcriptase inhibitors. AIDS 2002; 16:24852487. 78. Luft R. The development of mitochondrial medicine. Proceedings of the National Academy of Sciences, USA 1994; 91:87318738. 79. Katz AM. Is the failing heart energy depleted? Cardiology Clinics 1998; 16:633644. 80. Wallace DC. Mitochondrial diseases in man and mouse. Science 1999; 283:14821488. 81. Eriksson S, Xu B & Clayton DA. Efficient incorporation of anti-HIV deoxynucleotides by recombinant yeast mitochondrial DNA polymerase. Journal of Biological Chemistry 1995; 270:1892918934. 82. Hall ET, Yan JP, Melancon P & Kuchta RD. 3-Azido-3deoxythymidine potently inhibits protein glycosylation. A novel mechanism for AZT cytotoxicity. Journal of Biological Chemistry 1994; 269:1435514358. 83. Prakash O, Teng S, Ali M, Zhu X, Coleman R, Dabdoub RA, Chambers R, Aw TY, Flores SC & Joshi BH. The human immunodeficiency virus type 1 Tat protein potentiates zidovudine-induced cellular toxicity in transgenic mice. Archives of Biochemistry & Biophysics 1997; 343:173180. 84. Furman PA, Fyfe JA, St Clair MH, Weinhold K, Rideout JL, Freeman GA, Lehrman SN, Bolognesi DP, Broder S, Mitsuya H & Barry DW. Phosphorylation of 3-azido-3deoxythymidine and selective interaction of the 5-triphosphate with human immunodeficiency virus reverse transcriptase. Proceedings of the National Academy of Sciences, USA 1986; 83:83338337. 85. Mitsuya H, Weinhold KJ, Furman PA, St Clair MH, Lehrman SN, Gallo RC, Bolognesi D, Barry DW & Broder S. 3-Azido-3-deoxythymidine (BW A509U): an antiviral agent that inhibits the infectivity and cytopathic effect of human T-lymphotropic virus type III/lymphadenopathyassociated virus in vitro. Proceedings of the National Academy of Sciences, USA 1985; 82:70967100. 86. Simpson MV, Chin CD, Keilbaugh SA, Lin TS & Prusoff WH. Studies on the inhibition of mitochondrial DNA replication by 3-azido-3-deoxythymidine and other dideoxynucleoside analogs which inhibit HIV-1 replication. Biochemical Pharmacology 1989; 38:10331036. 87. Connolly KJ & Hammer SM. Antiretroviral therapy: strategies beyond single-agent reverse transcriptase inhibition. Antimicrobial Agents & Chemotherapy 1992; 36:509520. 88. Eriksson S, Munch-Petersen B, Kierdaszuk B & Arner E. Expression and substrate specificities of human thymidine kinase 1, thymidine kinase 2 and deoxycytidine kinase. Advances in Experimental Medicine & Biology 1991; 309:239243. 89. Eriksson S, Kierdaszuk B, Munch-Petersen B, Oberg B & Johansson NG. Comparison of the substrate specificities of human thymidine kinase 1 and 2 and deoxycytidine kinase toward antiviral and cytostatic nucleoside analogs. Biochemical & Biophysical Research Communications 1991; 176:586592. 90. Munch-Petersen B, Cloos L, Tyrsted G & Eriksson S. Diverging substrate specificity of pure human thymidine kinases 1 and 2 against antiviral dideoxynucleosides. Journal of Biological Chemistry 1991; 266:90329038. 91. Mitsuya H, Yarchoan R & Broder S. Molecular targets for AIDS therapy. Science 1990; 249:15331544. 92. Toji L & Cohen SS. The enzymatic termination of polydeoxynucleotides by 2,3-dideoxyadenosine triphosphate.
M21
W Lewis
Proceedings of the National Academy of Sciences, USA 1969; 63:871877. 93. McKenzie R, Fried MW, Sallie R, Conjeevaram H, Di Bisceglie AM, Park Y, Savarese B, Kleiner D, Tsokos M, Luciano C, Pruett T, Stotka J, Straus SE & Hoofnagle JH. Hepatic failure and lactic acidosis due to fialuridine (FIAU), an investigational nucleoside analogue for chronic hepatitis B. New England Journal of Medicine 1995; 333:10991105. 94. Saada A, Shaag A, Mandel H, Nevo Y, Eriksson S & Elpeleg O. Mutant mitochondrial thymidine kinase in mitochondrial DNA depletion myopathy. Nature Genetics 2001; 29:342344. 95. Bridges EG, Jiang Z & Cheng YC. Characterization of a dCTP transport activity reconstituted from human mitochondria. Journal of Biological Chemistry 1999; 274:46204625. 96. Johansson M & Karlsson A. Cloning of the cDNA and chromosome localization of the gene for human thymidine kinase 2. Journal of Biological Chemistry 1997; 272:84548458. 97. Rampazzo C, Johansson M, Gallinaro L, Ferraro P, Hellman U, Karlsson A, Reichard P & Bianchi V. Mammalian 5(3)-deoxyribonucleotidase, cDNA cloning, and overexpression of the enzyme in Escherichia coli and mammalian cells. Journal of Biological Chemistry 2000; 275:54095415. 98. Rampazzo C, Gallinaro L, Milanesi E, Frigimelica E, Reichard P & Bianchi V. A deoxyribonucleotidase in mitochondria: involvement in regulation of dNTP pools and possible link to genetic disease. Proceedings of the National Academy of Sciences, USA 2000; 97:82398244. 99. Mazzon C, Rampazzo C, Scaini MC, Gallinaro L, Karlsson A, Meier C, Balzarini J, Reichard P & Bianchi V. Cytosolic and mitochondrial deoxyribonucleotidases: activity with substrate analogs, inhibitors and implications for therapy. Biochemical Pharmacology 2003; 66:471479. 100. Dolce V, Fiermonte G, Runswick MJ, Palmieri F & Walker JE. The human mitochondrial deoxynucleotide carrier and its role in the toxicity of nucleoside antivirals. Proceedings of the National Academy of Sciences, USA 2001; 98:22842288. 101. Cohen J. Therapies. Confronting the limits of success. Science 2002; 296:23202324. 102. Yeni PG, Hammer SM, Carpenter CC, Cooper DA, Fischl MA, Gatell JM, Gazzard BG, Hirsch MS, Jacobsen DM, Katzenstein DA, Montaner JS, Richman DD, Saag MS, Schechter M, Schooley RT, Thompson MA, Vella S & Volberding PA. Antiretroviral treatment for adult HIV infection in 2002: updated recommendations of the International AIDS Society USA Panel. Journal of the American Medical Association 2002; 288:222235. 103. Pan-Zhou XR, Cui L, Zhou XJ, Sommadossi JP & DarleyUsmar VM. Differential effects of antiretroviral nucleoside analogs on mitochondrial function in HepG2 cells. Antimicrobial Agents & Chemotherapy 2000; 44:496503. 104. Moyle G. Clinical manifestations and management of antiretroviral nucleoside analog-related mitochondrial toxicity. Clinical Therapeutics 2000; 22:911936; discussion 898. 105. Brinkman K. Editorial response: hyperlactatemia and hepatic steatosis as features of mitochondrial toxicity of nucleoside analogue reverse transcriptase inhibitors. Clinical Infectious Diseases 2000; 31:167169. 106. Lewis W. Mitochondrial DNA replication, nucleoside reverse-transcriptase inhibitors, and AIDS cardiomyopathy. Progress in Cardiovascular Diseases 2003; 45:305318. 107. Lewis W. Mitochondrial dysfunction and nucleoside reverse transcriptase inhibitor therapy: experimental clarifications and persistent clinical questions. Antiviral Research 2003; 58:189197. 108. Lee H, Hanes J & Johnson KA. Toxicity of nucleoside analogues used to treat AIDS and the selectivity of the mitochondrial DNA polymerase. Biochemistry 2003; 42:1471114719.
109. Note R, Maisonneuve C, Letteron P, Peytavin G, Djouadi F, Igoudjil A, Guimont MC, Biour M, Pessayre D & Fromenty B. Mitochondrial and metabolic effects of nucleoside reverse transcriptase inhibitors (NRTIs) in mice receiving one of five single- and three dual-NRTI treatments. Antimicrobial Agents & Chemotherapy 2003; 47:33843392. 110. Macchi B & Mastino A. Pharmacological and biological aspects of basic research on nucleoside-based reverse transcriptase inhibitors. Pharmacological Research 2002; 46:473482. 111. Dieterich DT. Long-term complications of nucleoside reverse transcriptase inhibitor therapy. AIDS Reader 2003; 13:176184. 112. Moyle G. Toxicity of antiretroviral nucleoside and nucleotide analogues: is mitochondrial toxicity the only mechanism? Drug Safety 2000; 23:467481. 113. Ahluwalia GS, Gao WY, Mitsuya H & Johns DG. 2,3Didehydro-3-deoxythymidine: regulation of its metabolic activation by modulators of thymidine-5-triphosphate biosynthesis. Molecular Pharmacology 1996; 50:160165. 114. Benbrik E, Chariot P, Bonavaud S, Ammi-Said M, Frisdal E, Rey C, Gherardi R & Barlovatz-Meimon G. Cellular and mitochondrial toxicity of zidovudine (AZT), didanosine (ddI) and zalcitabine (ddC) on cultured human muscle cells. Journal of the Neurological Sciences 1997; 149:1925. 115. Cherrington JM, Allen SJW, Bischofberger N & Chen MS. Kinetic interaction of the diphosphates of 9-(2-phosphonylmethoxyethyl) adenine and other anti-HIV active purine congeners with HIV reverse transcriptase and human DNA polymerases alpha, beta and gamma. Antiviral Chemistry & Chemotherapy 1995; 6:217221. 116. Corcuera Pindado MT, Lopez Bravo A, MartinezRodriguez R, Picazo Talavera A, Gomez Aguado F, Roldan Contreras M, Perez Alvarez MJ, Fernandez Garcia A & Alonso Martin MJ. Histochemical and ultrastructural changes induced by zidovudine in mitochondria of rat cardiac muscle. European Journal of Histochemistry 1994; 38:311318. 117. Corcuera T, Alonso MJ, Picazo A, Gomez F, Roldan M, Abad M, Munoz E & Lopez-Bravo A. Hepatic morphological alterations induced by zidovudine (ZDV) in an experimental model. Pathology, Research & Practice 1996; 192:182187. 118. Hobbs GA, Keilbaugh SA & Simpson MV. The Friend murine erythroleukemia cell, a model system for studying the association between bone marrow toxicity induced by 3-azido-3-dideoxythymidine and dideoxynucleoside inhibition of mtDNA replication. Biochemical Pharmacology 1992; 43:13971400. 119. Hobbs GA, Keilbaugh SA, Rief PM & Simpson MV. Cellular targets of 3-azido-3-deoxythymidine: an early (non-delayed) effect on oxidative phosphorylation. Biochemical Pharmacology 1995; 50:381390. 120. Nusbaum NJ & Joseph PE. AZT incorporation into mitochondria: study in a human myeloid cell line. DNA & Cell Biology 1996; 15:363366. 121. Schroder JM, Kaldenbach T & Piroth W. Nuclear and mitochondrial changes of co-cultivated spinal cord, spinal ganglia and muscle fibers following treatment with various doses of zidovudine. Acta Neuropathologica 1996; 92:138149. 122. Semino-Mora MC, Leon-Monzon ME & Dalakas MC. The effect of L-carnitine on the AZT-induced destruction of human myotubes. Part II: treatment with L-carnitine improves the AZT-induced changes and prevents further destruction. Laboratory Investigation 1994; 71:773781. 123. Wang H, Lemire BD, Cass CE, Weiner JH, Michalak M, Penn AM & Fliegel L. Zidovudine and dideoxynucleosides deplete wild-type mitochondrial DNA levels and increase deleted mitochondrial DNA levels in cultured KearnsSayre syndrome fibroblasts. Biochimica & Biophysica Acta 1996; 1316:5159.
M22
124. Velsor LW, Kovacevic M, Goldstein M, Leitner HM, Lewis W & Day BJ. Mitochondrial oxidative stress in human hepatoma cells exposed to stavudine. Toxicology & Applied Pharmacology 2004; 199:1019. 125. Barile M, Valenti D, Hobbs GA, Abruzzese MF, Keilbaugh SA, Passarella S, Quagliariello E & Simpson MV. Mechanisms of toxicity of 3-azido-3-deoxythymidine. Its interaction with adenylate kinase. Biochemical Pharmacology 1994; 48:14051412. 126. Barile M, Valenti D, Passarella S & Quagliariello E. 3Azido-3-deoxythmidine uptake into isolated rat liver mitochondria and impairment of ADP/ATP translocator. Biochemical Pharmacology 1997; 53:913920. 127. Pereira LF, Oliveira MB & Carnieri EG. Mitochondrial sensitivity to AZT. Cell Biochemistry & Function 1998; 16:173181. 128. Sanda A, Zhu C, Johansson M & Karlsson A. Bystander effects of nucleoside analogs phosphorylated in the cytosol or mitochondria. Biochemical & Biophysical Research Communications 2001; 287:11631166. 129. Yamaguchi T, Katoh I & Kurata S. Azidothymidine causes functional and structural destruction of mitochondria, glutathione deficiency and HIV-1 promoter sensitization. European Journal of Biochemistry 2002; 269:27822788. 130. Gopinath R, Hutcheon M, Cheema-Dhadli S & Halperin M. Chronic lactic acidosis in a patient with acquired immunodeficiency syndrome and mitochondrial myopathy: biochemical studies. Journal of the American Society of Nephrology 1992; 3:12121219. 131. Gerard Y, Maulin L, Yazdanpanah Y, De La Tribonniere X, Amiel C, Maurage CA, Robin S, Sablonniere B, Dhennain C & Mouton Y. Symptomatic hyperlactataemia: an emerging complication of antiretroviral therapy. AIDS 2000; 14:27232730. 132. Chariot P, Dubreuil-Lemaire AL & Gherardi R. Lactic acidosis and AIDS. Annals of Internal Medicine 1993; 119:344345. 133. Olano JP, Borucki MJ, Wen JW & Haque AK. Massive hepatic steatosis and lactic acidosis in a patient with AIDS who was receiving zidovudine. Clinical Infectious Diseases 1995; 21:973976. 134. Sundar K, Suarez M, Banogon PE & Shapiro JM. Zidovudine-induced fatal lactic acidosis and hepatic failure in patients with acquired immunodeficiency syndrome: report of two patients and review of the literature. Critical Care Medicine 1997; 25:14251430. 135. Shaer AJ & Rastegar A. Lactic acidosis in the setting of antiretroviral therapy for the acquired immunodeficiency syndrome. A case report and review of the literature. American Journal of Nephrology 2000; 20:332338. 136. Bartley PB, Westacott L, Boots RJ, Lawson M, Potter JM, Hyland VJ & Woods ML 2nd. Large hepatic mitochondrial DNA deletions associated with L-lactic acidosis and highly active antiretroviral therapy. AIDS 2001; 15:419420. 137. Powderly WG. Long-term exposure to lifelong therapies. Journal of Acquired Immune Deficiency Syndromes 2002; 29(Suppl 1):S28S40. 138. Bartlett JG. Toxicity of antiretroviral agents. Hopkins HIV Report 1999; 11:2,12. 139. Antoniades C, Macdonald C, Knisely A, Taylor C & Norris S. Mitochondrial toxicity associated with HAART following liver transplantation in an HIV-infected recipient. Liver Transplantation 2004; 10:699702. 140. Walker UA, Bauerle J, Laguno M, Murillas J, Mauss S, Schmutz G, Setzer B, Miquel R, Gatell JM & Mallolas J. Depletion of mitochondrial DNA in liver under antiretroviral therapy with didanosine, stavudine, or zalcitabine. Hepatology 2004; 39:311317. 141. Claessens YE, Chiche JD, Mira JP & Cariou A. Bench-tobedside review: severe lactic acidosis in HIV patients treated with nucleoside analogue reverse transcriptase inhibitors. Critical Care 2003; 7:226232.
142. Montessori V, Harris M & Montaner JS. Hepatotoxicity of nucleoside reverse transcriptase inhibitors. Seminars in Liver Disease 2003; 23:167172. 143. Martin AM, Hammond E, Nolan D, Pace C, Den Boer M, Taylor L, Moore H, Martinez OP, Christiansen FT & Mallal S. Accumulation of mitochondrial DNA mutations in human immunodeficiency virus-infected patients treated with nucleoside-analogue reverse-transcriptase inhibitors. American Journal of Human Genetics 2003; 72:549560. 144. Radovanovic J, Todorovic V, Boricic I, Jankovic-Hladni M & Korac A. Comparative ultrastructural studies on mitochondrial pathology in the liver of AIDS patients: clusters of mitochondria, protuberances, minimitochondria, vacuoles, and virus-like particles. Ultrastructural Pathology 1999; 23:1924. 145. Chariot P, Drogou I, de Lacroix-Szmania I, Eliezer-Vanerot MC, Chazaud B, Lombes A, Schaeffer A & Zafrani ES. Zidovudine-induced mitochondrial disorder with massive liver steatosis, myopathy, lactic acidosis, and mitochondrial DNA depletion. Journal of Hepatology 1999; 30:156160. 146. Shapiro SH & Klavins JV. Concentric membranous bodies and giant mitochondria in hepatocytes from a patient with AIDS. Ultrastructural Pathology 1993; 17:557563. 147. Bissuel F, Bruneel F, Habersetzer F, Chassard D, Cotte L, Chevallier M, Bernuau J, Lucet JC & Trepo C. Fulminant hepatitis with severe lactate acidosis in HIV-infected patients on didanosine therapy. Journal of Internal Medicine 1994; 235:367371. 148. Sinnwell TM, Sivakumar K, Soueidan S, Jay C, Frank JA, McLaughlin AC & Dalakas MC. Metabolic abnormalities in skeletal muscle of patients receiving zidovudine therapy observed by 31P in vivo magnetic resonance spectroscopy. Journal of Clinical Investigation 1995; 96:126131. 149. Herskowitz A, Willoughby SB, Baughman KL, Schulman SP & Bartlett JD. Cardiomyopathy associated with antiretroviral therapy in patients with HIV infection: a report of six cases. Annals of Internal Medicine 1992; 116:311313. 150. DAmati G, Kahn HJ, Butany J & Silver MD. Altered distribution of desmin filaments in hypertrophic cardiomyopathy: an immunohistochemical study. Modern Pathology 1992; 5:165168. 151. Lewis W. Pathologic changes in the hearts of patients with AIDS. In Cardiology in AIDS 1998; pp. 233254. Edited by SE Lipshultz. New York: Chapman and Hall. 152. Lipshultz SE, Easley KA, Orav EJ, Kaplan S, Starc TJ, Bricker JT, Lai WW, Moodie DS, Sopko G, McIntosh K & Colan SD. Absence of cardiac toxicity of zidovudine in infants. Pediatric Pulmonary and Cardiac Complications of Vertically Transmitted HIV Infection Study Group. New England Journal of Medicine 2000; 343:759766. 153. Al-Attar I, Orav EJ, Exil V, Vlach SA & Lipshultz SE. Predictors of cardiac morbidity and related mortality in children with acquired immunodeficiency syndrome. Journal of the American College of Cardiology 2003; 41:15981605. 154. Brinkman K. Evidence for mitochondrial toxicity: lactic acidosis as proof of concept. Journal of HIV Therapy 2001; 6:1316. 155. Brinkman K, ter Hofstede HJ, Burger DM, Smeitink JA & Koopmans PP. Adverse effects of reverse transcriptase inhibitors: mitochondrial toxicity as common pathway. AIDS 1998; 12:17351744. 156. Shiramizu B, Shikuma KM, Kamemoto L, Gerschenson M, Erdem G, Pinti M, Cossarizza A & Shikuma C. Brief report: placenta and cord blood mitochondrial DNA toxicity in HIV-infected women receiving nucleoside reverse transcriptase inhibitors during pregnancy. Journal of Acquired Immune Deficiency Syndromes 2003; 32:370374. 157. Agbaria R, Manor E, Barak J & Balzarini J. Phosphorylation of 3-azidothymidine in maternal and fetal peripheral blood mononuclear cells during gestation and at term. Journal of Acquired Immune Deficiency Syndromes 2003; 32:477481.
M23
W Lewis
158. Ha JC, Nosbisch C, Abkowitz JL, Conrad SH, Mottet NK, Ruppenthal GC, Robinette R, Sackett GP & Unadkat JD. Fetal, infant, and maternal toxicity of zidovudine (azidothymidine) administered throughout pregnancy in Macaca nemestrina. Journal of Acquired Immune Deficiency Syndromes & Human Retrovirology 1998; 18:2738. 159. Ha JC, Nosbisch C, Conrad SH, Ruppenthal GC, Sackett GP, Abkowitz J & Unadkat JD. Fetal toxicity of zidovudine (azidothymidine) in Macaca nemestrina: preliminary observations. Journal of Acquired Immune Deficiency Syndromes 1994; 7:154157. 160. Divi RL, Walker VE, Wade NA, Nagashima K, Seilkop SK, Adams ME, Nesel CJ, ONeill JP, Abrams EJ & Poirier MC. Mitochondrial damage and DNA depletion in cord blood and umbilical cord from infants exposed in utero to Combivir. AIDS 2004; 18:10131021. 161. Gerschenson M, Nguyen V, Ewings EL, Ceresa A, Shaw JA, St Claire MC, Nagashima K, Harbaugh SW, Harbaugh JW, Olivero OA, Divi RL, Albert PS & Poirier MC. Mitochondrial toxicity in fetal Erythrocebus patas monkeys exposed transplacentally to zidovudine plus lamivudine. AIDS Research & Human Retroviruses 2004; 20:91100. 162. Cherry CL, Gahan ME, McArthur JC, Lewin SR, Hoy JF & Wesselingh SL. Exposure to dideoxynucleosides is reflected in lowered mitochondrial DNA in subcutaneous fat. Journal of Acquired Immune Deficiency Syndromes 2002; 30:271277. 163. Browne MJ, Mayer KH, Chafee SB, Dudley MN, Posner MR, Steinberg SM, Graham KK, Geletko SM, Zinner SH, Denman SL, Dunkle LM , Kaul S , Mclaren C , Skowron G, Kouttab NM, Kennedy TA, Weitberg AB & Curt GA. 2,3-didehydro-3-deoxythymidine (d4T) in patients with AIDS or AIDS-related complex: a phase I trial. Journal of Infectious Diseases 1993; 167:2129. 164. Cohen P, Sande M & Volberding P. The AIDS Knowledge Base: A Textbook on HIV Disease from the University of California, San Francisco. 2nd edn 1994. Boston: Little Brown and Company. 165. Cupler EJ & Dalakas MC. Exacerbation of peripheral neuropathy by lamivudine. Lancet 1995; 345:460461. 166. Hart AM, Wilson AD, Montovani C, Smith C, Johnson M, Terenghi G & Youle M. Acetyl-L-carnitine: a pathogenesis based treatment for HIV-associated antiretroviral toxic neuropathy. AIDS 2004; 18:15491560. 167. Venhoff N, Setzer B, Lebrecht D & Walker UA. Dietary supplements in the treatment of nucleoside reverse transcriptase inhibitor-related mitochondrial toxicity. AIDS 2002; 16:800802. 168. Medina DJ, Tsai CH, Hsiung GD & Cheng YC. Comparison of mitochondrial morphology, mitochondrial DNA content, and cell viability in cultured cells treated with three anti-human immunodeficiency virus dideoxynucleosides. Antimicrobial Agents & Chemotherapy 1994; 38:18241828. 169. Cui L, Locatelli L, Xie MY & Sommadossi JP. Effect of nucleoside analogs on neurite regeneration and mitochondrial DNA synthesis in PC-12 cells. Journal of Pharmacology & Experimental Therapeutics 1997; 280:12281234. 170. Dalakas MC, Semino-Mora C & Leon-Monzon M. Mitochondrial alterations with mitochondrial DNA depletion in the nerves of AIDS patients with peripheral neuropathy induced by 23-dideoxycytidine (ddC). Laboratory Investigation 2001; 81:15371544. 171. Lim SE, Ponamarev MV, Longley MJ & Copeland WC. Structural determinants in human DNA polymerase gamma account for mitochondrial toxicity from nucleoside analogs. Journal of Molecular Biology 2003; 329:4557. 172. Saint-Marc T, Partisani M, Poizot-Martin I, Bruno F, Rouviere O, Lang JM, Gastaut JA & Touraine JL. A syndrome of peripheral fat wasting (lipodystrophy) in patients receiving long-term nucleoside analogue therapy. AIDS 1999; 13:16591667.
173. Harrison DG. Cellular and molecular mechanisms of endothelial cell dysfunction. Journal of Clinical Investigation 1997; 100:21532157. 174. Brinkman K, Vrouenraets S, Kauffmann R, Weigel H & Frissen J. Treatment of nucleoside reverse transcriptase inhibitor-induced lactic acidosis. AIDS 2000; 14:28012802. 175. Carr A, Morey A, Mallon P, Williams D & Thorburn DR. Fatal portal hypertension, liver failure, and mitochondrial dysfunction after HIV-1 nucleoside analogue-induced hepatitis and lactic acidaemia. Lancet 2001; 357:14121414. 176. Miller KD, Cameron M, Wood LV, Dalakas MC & Kovacs JA. Lactic acidosis and hepatic steatosis associated with use of stavudine: report of four cases. Annals of Internal Medicine 2000; 133:192196. 177. Mokrzycki MH, Harris C, May H, Laut J & Palmisano J. Lactic acidosis associated with stavudine administration: a report of five cases. Clinical Infectious Diseases 2000; 30:198200. 178. Murphy MD, OHearn M & Chou S. Fatal lactic acidosis and acute renal failure after addition of tenofovir to an antiretroviral regimen containing didanosine. Clinical Infectious Diseases 2003; 36:10821085. 179. Tsai CH, Doong SL, Johns DG, Driscoll JS & Cheng YC. Effect of anti-HIV 2-beta-fluoro-2,3-dideoxynucleoside analogs on the cellular content of mitochondrial DNA and on lactate production. Biochemical Pharmacology 1994; 48:14771481. 180. Gaou I, Malliti M, Guimont MC, Letteron P, Demeilliers C, Peytavin G, Degott C, Pessayre D & Fromenty B. Effect of stavudine on mitochondrial genome and fatty acid oxidation in lean and obese mice. Journal of Pharmacology & Experimental Therapeutics 2001; 297:516523. 181. Anderson TD, Davidovich A, Feldman D, Sprinkle TJ, Arezzo J, Brosnan C, Calderon RO, Fossom LH, DeVries JT & DeVries GH. Mitochondrial schwannopathy and peripheral myelinopathy in a rabbit model of dideoxycytidine neurotoxicity. Laboratory Investigation 1994; 70:724739. 182. Cossarizza A, Pinti M, Moretti L, Bricalli D, Bianchi R, Troiano L, Fernandez MG, Balli F, Brambilla P, Mussini C & Vigano A. Mitochondrial functionality and mitochondrial DNA content in lymphocytes of vertically infected human immunodeficiency virus-positive children with highly active antiretroviral therapy-related lipodystrophy. Journal of Infectious Diseases 2002; 185:299305. 183. Shiramizu B, Shikuma KM, Kamemoto L, Gerschenson M, Erdem G, Pinti M, Cossarizza A & Shikuma C. Placenta and cord blood mitochondrial DNA toxicity in HIVinfected women receiving nucleoside reverse transcriptase inhibitors during pregnancy. Journal of Acquired Immune Deficiency Syndromes 2003; 32:370374. 184. Yarchoan R, Pluda JM, Thomas RV, Mitsuya H, Brouwers P, Wyvill KM, Hartman N, Johns DG & Broder S. Longterm toxicity/activity profile of 2,3-dideoxyinosine in AIDS or AIDS-related complex. Lancet 1990; 336:526529. 185. Dubinsky RM, Yarchoan R, Dalakas M & Broder S. Reversible axonal neuropathy from the treatment of AIDS and related disorders with 2,3-dideoxycytidine (ddC). Muscle & Nerve 1989; 12:856860. 186. Merigan TC, Skowron G, Bozzette SA, Richman D, Uttamchandani R, Fischl M, Schooley R, Hirsch M, Soo W, Pettinelli C, Schaumburg H & the ddC Study Group of the AIDS Clinical Trials Group. Circulating p24 antigen levels and responses to dideoxycytidine in human immunodeficiency virus (HIV) infections. A Phase I and II study. Annals of Internal Medicine 1989; 110:189194. 187. Berger AR, Arezzo JC, Schaumburg HH, Skowron G, Merigan T, Bozzette S, Richman D & Soo W. 2,3Dideoxycytidine (ddC) toxic neuropathy: a study of 52 patients. Neurology 1993; 43:358362.
M24
188. Dalakas MC. Peripheral neuropathy and antiretroviral drugs. Journal of the Peripheral Nervous System 2001; 6:1420. 189. Cicalini S, Forcina G & De Rosa FG. Infective endocarditis in patients with human immunodeficiency virus infection. Journal of Infection 2001; 42:267271. 190. Feldman D, Brosnan C & Anderson TD. Ultrastructure of peripheral neuropathy induced in rabbits by 2,3-dideoxycytidine. Laboratory Investigation 1992; 66:7585. 191. Anderson TD, Davidovich A, Arceo R, Brosnan C, Arezzo J & Schaumburg H. Peripheral neuropathy induced by 2,3-dideoxycytidine. A rabbit model of 2,3-dideoxycytidine neurotoxicity. Laboratory Investigation 1992; 66:6374. 192. Starnes MC & Cheng YC. Inhibition of human immunodeficiency virus reverse transcriptase by 2,3dideoxynucleoside triphosphates: template dependence, and combination with phosphonoformate. Virus Genes 1989; 2:241251. 193. Keilbaugh SA, Hobbs GA & Simpson MV. Anti-human immunodeficiency virus type 1 therapy and peripheral neuropathy: prevention of 2,3-dideoxycytidine toxicity in PC12 cells, a neuronal model, by uridine and pyruvate. Molecular Pharmacology 1993; 44:702706. 194. Feldman D & Anderson TD. Schwann cell mitochondrial alterations in peripheral nerves of rabbits treated with 2,3dideoxycytidine. Acta Neuropathologica 1994; 87:7180. 195. Ojetti V, Gasbarrini A, Migneco A, Flore R, Santoliquido A, De Martini D, Agnes S, Gentiloni Silveri N & Pola P. Lamivudine-induced muscle mitochondrial toxicity. Digestive & Liver Disease 2002; 34:384385. 196. Feng JY, Murakami E, Zorca SM, Johnson AA, Johnson KA, Schinazi RF, Furman PA & Anderson KS. Relationship between antiviral activity and host toxicity: comparison of the incorporation efficiencies of 2,3-dideoxy-5-fluoro-3thiacytidine-triphosphate analogs by human immunodeficiency virus type 1 reverse transcriptase and human mitochondrial DNA polymerase. Antimicrobial Agents & Chemotherapy 2004; 48:13001306. 197. Faraj A, Agrofoglio LA, Wakefield JK, McPherson S, Morrow CD, Gosselin G, Mathe C, Imbach JL, Schinazi RF & Sommadossi JP. Inhibition of human immunodeficiency virus type 1 reverse transcriptase by the 5-triphosphate beta enantiomers of cytidine analogs. Antimicrobial Agents & Chemotherapy 1994; 38:23002305. 198. Cui L, Schinazi RF, Gosselin G, Imbach JL, Chu CK, Rando RF, Revankar GR & Sommadossi JP. Effect of betaenantiomeric and racemic nucleoside analogues on mitochondrial functions in HepG2 cells. Implications for predicting drug hepatotoxicity. Biochemical Pharmacology 1996; 52:15771584. 199. Lambert JS, Seidlin M, Reichman RC, Plank CS, Laverty M, Morse GD, Knupp C, McLaren C, Pettinelli C, Valentine FT & Dolin R. 2,3-Dideoxyinosine (ddI) in patients with the acquired immunodeficiency syndrome or AIDS-related complex. A Phase I trial. New England Journal of Medicine 1990; 322:13331340. 200. Cooley TP, Kunches LM, Saunders CA, Ritter JK, Perkins CJ, McLaren C, McCaffrey RP & Liebman HA. Once-daily administration of 2,3-dideoxyinosine (ddI) in patients with the acquired immunodeficiency syndrome or AIDSrelated complex. Results of a Phase I trial. New England Journal of Medicine 1990; 322:13401345. 201. Foli A, Benvenuto F, Piccinini G, Bareggi A, Cossarizza A, Lisziewicz J & Lori F. Direct analysis of mitochondrial toxicity of antiretroviral drugs. AIDS 2001; 15:16871694. 202. Hanna L. FddA: antiretroviral in development. BETA Bulletin of Experimental Treatments for AIDS 1998:78. 203. Comereski CR, Kelly WA, Davidson TJ, Warner WA, Hopper LD & Oleson FB. Acute cardiotoxicity of nucleoside analogs FddA and FddI in rats. Fundamental & Applied Toxicology 1993; 20:360364.
204. Tennant BC, Baldwin BH, Graham LA, Ascenzi MA, Hornbuckle WE, Rowland PH, Tochkov IA, Yeager AE, Erb HN, Colacino JM, Lopez C, Engelhardt JA, Bowsher RR, Richardson FC, Lewis W, Cote PJ, Korba BE & Gerin JL. Antiviral activity and toxicity of fialuridine in the woodchuck model of hepatitis B virus infection. Hepatology 1998; 28:179191. 205. Walsh K, Kaye K, Demaerschalk B, Stewart S, Crukley J & Hammond R. AZT myopathy and HIV-1 polymyositis: one disease or two? Canadian Journal of Neurological Sciences 2002; 29:390393. 206. Prime KP, Edwards SG, Pakianathan MR, Holton JL, Scaravilli F & Miller RF. Polymyositis masquerading as mitochondrial toxicity. Sexually Transmitted Infections 2003; 79:417418. 207. Simpson DM, Citak KA, Godfrey E, Godbold J & Wolfe DE. Myopathies associated with human immunodeficiency virus and zidovudine: can their effects be distinguished? Neurology 1993; 43:971976. 208. Simpson DM, Slasor P, Dafni U, Berger J, Fischl MA & Hall C. Analysis of myopathy in a placebo-controlled zidovudine trial. Muscle & Nerve 1997; 20:382385. 209. Gerschenson M, Erhart SW, Paik CY, St Claire MC, Nagashima K, Skopets B, Harbaugh SW, Harbaugh JW, Quan W & Poirier MC. Fetal mitochondrial heart and skeletal muscle damage in Erythrocebus patas monkeys exposed in utero to 3-azido-3-deoxythymidine. AIDS Research & Human Retroviruses 2000; 16:635644. 210. Gerschenson M, Nguyen VT, St Claire MC, Harbaugh SW, Harbaugh JW, Proia LA & Poirier MC. Chronic stavudine exposure induces hepatic mitochondrial toxicity in adult Erythrocebus patas monkeys. Journal of Human Virology 2001; 4:335342. 211. Gerschenson M & Poirier MC. Fetal patas monkeys sustain mitochondrial toxicity as a result of in utero zidovudine exposure. Annals of the New York Academy of Sciences 2000; 918:269281. 212. Agarwal RP & Olivero OA. Genotoxicity and mitochondrial damage in human lymphocytic cells chronically exposed to 3-azido-2,3-dideoxythymidine. Mutation Research 1997; 390:223231. 213. Miura T, Goto M, Hosoya N, Odawara T, Kitamura Y, Nakamura T & Iwamoto A. Depletion of mitochondrial DNA in HIV-1-infected patients and its amelioration by antiretroviral therapy. Journal of Medical Virology 2003; 70:497505. 214. Mackey DA, Fingert JH, Luzhansky JZ, McCluskey PJ, Howell N, Hall AJ, Pierce AB & Hoy JF. Lebers hereditary optic neuropathy triggered by antiretroviral therapy for human immunodeficiency virus. Eye 2003; 17:312317. 215. Reiss P, Casula M, de Ronde A, Weverling GJ, Goudsmit J & Lange JM. Greater and more rapid depletion of mitochondrial DNA in blood of patients treated with dual (zidovudine+didanosine or zidovudine+zalcitabine) vs. single (zidovudine) nucleoside reverse transcriptase inhibitors. HIV Medicine 2004; 5:1114. 216. Newman MD. Bone disorders, hypertension, and mitochondrial toxicity in HIV disease. Topics in HIV Medicine 2003; 11:1015. 217. Miller RF, Shahmonesh M, Hanna MG, Unwin RJ, Schapira AH & Weller IV. Polyphenotypic expression of mitochondrial toxicity caused by nucleoside reverse transcriptase inhibitors. Antiviral Therapy 2003; 8:253257. 218. Cossarizza A & Moyle G. Antiretroviral nucleoside and nucleotide analogues and mitochondria. AIDS 2004; 18:137151. 219. Copeland WC, Ponamarev MV, Nguyen D, Kunkel TA & Longley MJ. Mutations in DNA polymerase gamma cause error prone DNA synthesis in human mitochondrial disorders. Acta Biochimica Polonica 2003; 50:155167. 220. Feng JY, Johnson AA, Johnson KA & Anderson KS. Insights into the molecular mechanism of mitochondrial toxicity by AIDS drugs. Journal of Biological Chemistry 2001; 276:2383223837.
M25
W Lewis
221. Graziewicz MA, Day BJ & Copeland WC. The mitochondrial DNA polymerase as a target of oxidative damage. Nucleic Acids Research 2002; 30:28172824. 222. Lewis W. Defective mitochondrial DNA replication and NRTIs: pathophysiological implications in AIDS cardiomyopathy. American Journal of Physiology. Heart & Circulatory Physiology 2003; 284:H1H9. 223. Elimadi A, Morin D, Albengres E, Chauvet-Monges AM, Allain V, Crevat A & Tillement JP. Differential effects of zidovudine and zidovudine triphosphate on mitochondrial permeability transition and oxidative phosphorylation. British Journal of Pharmacology 1997; 121:12951300. 224. Freyssenet D, DiCarlo M, Escobar P, Grey J, Schneider J & Hood DA. Zidovudine (AZT) induced alterations in mitochondrial biogenesis in rat striated muscles. Canadian Journal of Physiology & Pharmacology 1999; 77:2935. 225. Birkus G, Hitchcock MJ & Cihlar T. Assessment of mitochondrial toxicity in human cells treated with tenofovir: comparison with other nucleoside reverse transcriptase inhibitors. Antimicrobial Agents & Chemotherapy 2002; 46:716723. 226. Modica-Napolitano JS. AZT causes tissue-specific inhibition of mitochondrial bioenergetic function. Biochemical & Biophysical Research Communications 1993; 194:170177. 227. Petit F, Arnoult D, Lelievre JD, Moutouh-de Parseval L, Hance AJ, Schneider P, Corbeil J, Ameisen JC & Estaquier J. Productive HIV-1 infection of primary CD4+ T cells induces mitochondrial membrane permeabilization leading to a caspase-independent cell death. Journal of Biological Chemistry 2002; 277:14771487. 228. Frick LW, Nelson DJ, St Clair MH, Furman PA & Krenitsky TA. Effects of 3-azido-3-deoxythymidine on the deoxynucleotide triphosphate pools of cultured human cells. Biochemical & Biophysical Research Communications 1988; 154:124129. 229. Herzberg NH, Zorn I, Zwart R, Portegies P & Bolhuis PA. Major growth reduction and minor decrease in mitochondrial enzyme activity in cultured human muscle cells after exposure to zidovudine. Muscle & Nerve 1992; 15:706710. 230. Walker UA, Venhoff N, Koch EC, Olschewski M, Schneider J & Setzer B. Uridine abrogates mitochondrial toxicity related to nucleoside analogue reverse transcriptase inhibitors in HepG2 cells. Antiviral Therapy 2003; 8:463470. 231. Collins ML, Sondel N, Cesar D & Hellerstein MK. Effect of nucleoside reverse transcriptase inhibitors on mitochondrial DNA synthesis in rats and humans. Journal of Acquired Immune Deficiency Syndromes 2004; 37:11321139. 232. Pinti M, Troiano L, Nasi M, Ferraresi R, Dobrucki J & Cossarizza A. Hepatoma HepG2 cells as a model for in vitro studies on mitochondrial toxicity of antiviral drugs: which correlation with the patient? Journal of Biological Regulators & Homeostatic Agents 2003; 17:166171. 233. White EL, Parker WB, Macy LJ, Shaddix SC, McCaleb G, Secrist JAd, Vince R & Shannon WM. Comparison of the effect of Carbovir, AZT, and dideoxynucleoside triphosphates on the activity of human immunodeficiency virus reverse transcriptase and selected human polymerases. Biochemical & Biophysical Research Communications 1989; 161:393398. 234. Parker WB, White EL, Shaddix SC, Ross LJ, Buckheit RW Jr, Germany JM, Secrist JAd, Vince R & Shannon WM. Mechanism of inhibition of human immunodeficiency virus type 1 reverse transcriptase and human DNA polymerases alpha, beta, and gamma by the 5-triphosphates of carbovir, 3-azido-3-deoxythymidine, 2,3-dideoxyguanosine and 3deoxythymidine. A novel RNA template for the evaluation of antiretroviral drugs. Journal of Biological Chemistry 1991; 266:17541762. 235. Parker WB, Shaddix SC, Vince R & Bennett LL Jr. Lack of mitochondrial toxicity in CEM cells treated with carbovir. Antiviral Research 1997; 34:131136.
M26
236. Moreno A, Quereda C, Moreno L, Perez-Elias MJ, Muriel A, Casado JL, Antela A, Dronda F, Navas E, Barcena R & Moreno S. High rate of didanosine-related mitochondrial toxicity in HIV/HCV-coinfected patients receiving ribavirin. Antiviral Therapy 2004; 9:133138. 237. Fleischer R, Boxwell D & Sherman KE. Nucleoside analogues and mitochondrial toxicity. Clinical Infectious Diseases 2004; 38:E79E80. 238. von Giesen HJ, Hefter H, Jablonowski H & Arendt G. Stavudine and the peripheral nerve in HIV-1 infected patients. Journal of Neurology 1999; 246:211217. 239. Verucchi G, Calza L, Manfredi R & Chiodo F. Incidence of liver toxicity in HIV-infected patients receiving isolated dual nucleoside analogue antitretroviral therapy. Journal of Acquired Immune Deficiency Syndromes 2003; 33:546548. 240. Van Huyen JP, Landau A, Piketty C, Belair MF, Batisse D, Gonzalez-Canali G, Weiss L, Jian R, Kazatchkine MD & Bruneval P. Toxic effects of nucleoside reverse transcriptase inhibitors on the liver. Value of electron microscopy analysis for the diagnosis of mitochondrial cytopathy. American Journal of Clinical Pathology 2003; 119:546555. 241.Peyriere H, Guillemin V, Lotthe A, Baillat V, Fabre J, Favier C, Atoui N, Hansel S, Hillaire-Buys D & Reynes J. Reasons for early abacavir discontinuation in HIV-infected patients [see comment]. Annals of Pharmacotherapy 2003; 37:13921397. 242. Hughes DA, Vilar FJ, Ward CC, Alfirevic A, Park BK, Pirmohamed M. Cost-effectiveness analysis of HLS B*5701 genotyping in preventing abacavir hypersensitivity. Pharmacogenetics 2004; 14:335342. 243. Carr A, Workman C, Smith DE, Hoy J, Hudson J, Doong N, Martin A, Amin J, Freund J, Law M & Cooper DA; Mitochondrial Toxicity (MITOX) Study Group. Abacavir substitution for nucleoside analogs in patients with HIV lipoatrophy: a randomized trial. Journal of the American Medical Association 2002; 288:207215. 244. Martinez E, Milinkovic A, de Lazzari E, Ravasi G, Blanco JL, Larrousse M, Mallolas J, Garcia F, Miro JM & Gatell JM. Pancreatic toxic effects associated with co-administration of didanosine and tenofovir in HIV-infected adults. Lancet 2004; 364:6567. 245. De Clercq E. Clinical potential of the acyclic nucleoside phosphonates cidofovir, adefovir, and tenofovir in treatment of DNA virus and retrovirus infections. Clinical Microbiology Reviews 2003; 16:569596. 246. Institute of Medicine (US). Committee to Review the Fialuridine (FIAU/FIAC) Clinical Trials. Review of the Fialuridine (FIAU) Clinical Trials, 1995. Washington, DC: National Academy Press. 247. Stevenson W, Gaffey M, Ishitani M, McCullough C, Dickson R, Caldwell S, Lobo P & Pruett T. Clinical course of four patients receiving the experimental antiviral agent fialuridine for the treatment of chronic hepatitis B infection. Transplantation Proceedings 1995; 27:12191221. 248. Honkoop P, Scholte HR, de Man RA & Schalm SW. Mitochondrial injury. Lessons from the fialuridine trial. Drug Safety 1997; 17:17. 249. Brahams D. Deaths in US fialuridine trial. Lancet 1994; 343:14941495. 250. Odeh M. Lactic acidosis and acquired immunodeficiency syndrome. Journal of Internal Medicine 1994; 236:478479. 251. Crowley VE & Olukogu AO. The chemical pathology of AIDS. Annals of Clinical Biochemistry 1995; 32(Pt 5):511512. 252. Torre D & Speranza F. Is there a role for nitric oxide in hyperlactataemia syndromes and mitochondrial dysfunction associated with HIV therapy? Lancet Infectious Diseases 2003; 3:609610. 253. Wohl DA, Pilcher CD, Evans S, Revuelta M, McComsey G, Yang Y, Zackin R, Alston B, Welch S, Basar M, Kashuba A, Kondo P, Martinez A, Giardini J, Quinn J, Littles M,
Wingfield H & Koletar SL; Adult AIDS Clinical Trials Group A5129 Team. Absence of sustained hyperlactatemia in HIV-infected patients with risk factors for mitochondrial toxicity. Journal of Acquired Immune Deficiency Syndromes 2004; 35:274278. 254. de la Asuncion JG, del Olmo ML, Sastre J, Pallardo FV & Vina J. Zidovudine (AZT) causes an oxidation of mitochondrial DNA in mouse liver. Hepatology 1999; 29:985987. 255. Klecker RW, Katki AG & Collins JM. Toxicity, metabolism, DNA incorporation with lack of repair, and lactate production for 1-(2-fluoro-2-deoxy-beta-D-arabinofuranosyl)-5iodouracil in U-937 and MOLT-4 cells. Molecular Pharmacology 1994; 46:12041209. 256. Richardson FC, Engelhardt JA & Bowsher RR. Fialuridine accumulates in DNA of dogs, monkeys, and rats following long-term oral administration. Proceedings of the National Academy of Sciences, USA 1994; 91:1200312007. 257. Cui L, Yoon S, Schinazi RF & Sommadossi JP. Cellular and molecular events leading to mitochondrial toxicity of 1-(2deoxy-2-fluoro-1-beta-D-arabinofuranosyl)-5-iodouracil in human liver cells. Journal of Clinical Investigation 1995; 95:555563. 258. Lewis W & Tankersley KO. Triphosphorylated fialuridine and metabolites inhibit cardiac mitochondrial DNA polymerase gamma (DNA polymerase-gamma): alternative substrates for thymidine triphosphate (dTTP). Circulation 1995; 92:I186. 259. Horn DM, Neeb LA, Colacino JM & Richardson FC. Fialuridine is phosphorylated and inhibits DNA synthesis in isolated rat hepatic mitochondria. Antiviral Research 1997; 34:7174. 260. Wang J & Eriksson S. Phosphorylation of the anti-hepatitis B nucleoside analog 1-(2-deoxy- 2-fluoro-1-beta-D-arabinofuranosyl)-5-iodouracil (FIAU) by human cytosolic and mitochondrial thymidine kinase and implications for cyto-
toxicity. Antimicrobial Agents & Chemotherapy 1996; 40:15551557. 261. Colacino JM, Malcolm SK & Jaskunas SR. Effect of fialuridine on replication of mitochondrial DNA in CEM cells and in human hepatoblastoma cells in culture. Antimicrobial Agents & Chemotherapy 1994; 38:19972002. 262. Roth JS, Ford H Jr, Tanaka M, Mitsuya H & Kelley JA. Determination of 2-beta-fluoro-2,3-dideoxyadenosine, an experimental anti-AIDS drug, in human plasma by highperformance liquid chromatography. Journal of Chromatography. B, Biomedical Sciences & Applications 1998; 712:199210. 263. Highleyman L. Lodenosine trials stopped due to safety concerns. BETA Bulletin of Experimental Treatments for Aids 1999; 12:4. 264. Anonymous. Lodenosine trial halted. AIDS Patient Care & STDs 2000; 14:61. 265. Ruxrungtham K, Boone E, Ford H Jr, Driscoll JS, Davey RT Jr & Lane HC. Potent activity of 2-beta-fluoro-2,3dideoxyadenosine against human immunodeficiency virus type 1 infection in hu-PBL-SCID mice. Antimicrobial Agents & Chemotherapy 1996; 40:23692374. 266. Srinivas NR & Shyu WC. Single dose and multiple dose pharmacokinetics of 2-fluoro-2,3-dideoxyadenosine and 2-fluoro-2,3-dideoxyinosine, anti-HIV agents, in rats. European Journal of Drug Metabolism & Pharmacokinetics 1999; 24:329334. 267. Izawa K, Takamatsu S, Katayama S, Hirose N, Kozai S & Maruyama T. An industrial process for synthesizing lodenosine (FddA). Nucleosides, Nucleotides & Nucleic Acids 2003; 22:507517. 268. Roth JS, McCully CM, Balis FM, Poplack DG & Kelley JA. 2-beta-fluoro-2,3-dideoxyadenosine, lodenosine, in rhesus monkeys: plasma and cerebrospinal fluid pharmacokinetics and urinary disposition. Drug Metabolism & Disposition 1999; 27:11281132.
Received 13 January 2005, accepted 22 March 2005
M27
HIV protease inhibitors prevent mitochondrial hyperpolarization and redox imbalance and decrease endogenous uncoupler protein-2 expression in gp120-activated human T lymphocytes
Paola Matarrese1, Antonella Tinari2, Lucrezia Gambardella1, Elisabetta Mormone1, Piero Narilli4, Marina Pierdominici3, Roberto Cauda5 and Walter Malorni*1
Departments of Drug Research and Evaluation, 2Technology and Health, and 3Cell Biology and Neurosciences, Istituto Superiore di Sanit, viale Regina Elena 299-00161 Rome, Italy 4 Department of General Surgery and Organ Transplantation, University of Rome La Sapienza, Rome, Italy 5 Department of Infectious Diseases, Catholic University, Rome, Italy *Corresponding author: Tel: +39 06 4990 2905; Fax: +39 06 4990 3691; E-mail: malorni@iss.it
1
It has been demonstrated that HIV protease inhibitors (PIs) are able to inhibit apoptosis of both infected and uninfected T cells. It was hypothesized that the mechanisms underlying this effect are associated with a specific activity of these drugs against mitochondrial modifications occurring in the execution phase of apoptosis. In this work, we investigated the activity of PIs towards the early changes occurring in mitochondrial membrane potential in freshly isolated uninfected human T lymphocytes sensitized to CD95/Fas-induced physiological apoptosis via pre-exposure to HIV envelope protein gp120. The results obtained clearly indicate that PIs are capable of hindering
early morphogenetic changes bolstering T cell apoptosis, that is, cell polarization and mitochondrial hyperpolarization. The target effect on mitochondria appeared to be characterized by a specific activity of PIs in the maintenance of their homeostasis either in intact cells or in cell-free systems, that is, isolated mitochondria. PIs seem to act as boosters of mitochondrial defense mechanisms, including modulation of endogenous uncouplers. These results add new insights in the field of PI mitochondrial toxicity mechanisms and pharmacological perspectives for the use of these drugs in the control of immune system homeostasis.
Introduction
Given the important role of apoptosis in the pathogenesis and progression of HIV infection, several studies have been focused specifically on the mechanisms of apoptosis in both HIV-infected and uninfected CD4+ cells [1,2]. As a general rule, two different apoptotic pathways leading to activation of cell-specific programs have been proposed [3]. These two pathways refer to different initiation patterns, that is, receptordependent or independent [4]. In particular, changes of mitochondrial membrane potential (MMP) have been hypothesized to play a key role in apoptotic cascade. In fact, alterations of MMP have been associated with the release of apoptogenic factors that directly or indirectly, that is, via apoptosome formation, lead to the execution of apoptosis. A major role of mitochondria has also previously been suggested in the process of CD4+ T cell death [5,6]. Several drugs employed in clinical practice powerfully counteract the reduction of CD4+ cells by apoptosis in HIV infection. In fact, an important aspect of highly active antiretroviral therapy (HAART) is represented by immune reconstitution [7]. This is a therapeutic approach that involves, among others, drugs of different natures, such as HIV reverse transcriptase inhibitors, for example, zidovudine (AZT), and HIV protease inhibitors (PIs). Some of these, for example, indinavir (IDV), saquinavir (SQV) and lopinavir (LPV), are known to induce viral load lowering as well as the reduction of cell loss. It has also been suggested that AZT might be considered as an apoptotic inducer [8] whilst various PIs can be considered as apoptosishindering drugs [7,9]. Some in vitro and ex vivo studies have in fact suggested that PIs are able to inhibit peripheral blood mononuclear cell loss and restore impaired T-cell proliferative response [10]. Strikingly, this occurred independently from any viral infection [10]. Although a target activity of PIs towards mitochondria was hypothesized [11,12], the mechanism
M29
P Matarrese et al.
underlying this activity still remains poorly understood [9,13]. In the present work, we analysed the earlier mechanisms involved in the subcellular effects of PIs considered as mitochondriotropic drugs. The results reported herein document a series of changes, that is, cell polarization, induced by gp120 HIV protein in freshly isolated human T lymphocytes and the consequent proneness to CD95/Fas-induced apoptosis. These events were accompanied by mitochondrial hyperpolarization. PIs infer, with these subcellular events, hijacking T cells towards apoptotic resistance via a target effect on mitochondrial homeostasis. Furthermore, they also suggest for the first time that the mechanisms involved in PI activity include the modulation of the expression of endogenous mitochondrial uncoupler proteins (UCPs).
Evaluation of cell surface receptors

The surface expression of molecules associated with T cell activation (CD69, CD38 and HLA-DR) and cell death (CD95/Fas) was verified by flow cytometry on resting and activated lymphocytes. For this purpose, mAbs directly conjugated to fluorochromes PE, FITC or PerCP to human CD95, CD38, HLA-DR and CD69 (Becton Dickinson, Mountain View, CA, USA) were used. Appropriate fluorochrome-conjugated immunoglobulins were used as negative controls.
Apoptosis evaluation
Quantitative evaluation of apoptosis was performed by using the following flow and static cytometry methods: i) double staining using the annexin V-FITC apoptosis detection kit (Eppendorf, Milan, Italy). This technique allows cells that have lost membrane integrity (and are therefore considered necrotic) to show red staining with propidium iodide (40 g/ml) throughout the nucleus and to be easily distinguished from the living cells and ii) staining with the chromatin dye Hoechst (Molecular Probes, Eugene, OR, USA) as previously described [14].
Materials and methods

Isolation and activation of peripheral blood lymphocytes
Human peripheral blood lymphocytes (PBLs) from healthy donors (HDs) were isolated from freshly heparinized blood through a Ficoll-Hypaque density gradient centrifugation and washed three times in phosphate-buffer saline (PBS), pH 7.4. (Lympholyte-H; Cedarlane Laboratories, Hornby, ON, Canada). PBLs were subcultured in 25 cm2 or 75 cm2 Falcon plastic flasks at a density of approximately 1106 cells/ml in RPMI 1640 (Gibco-BRL Life Technologies, Milan, Italy) containing 10% fetal calf serum, (Flow Laboratories, Irvine, Scotland), 1% non-essential amino acids, 5 mM L-glutamine, penicillin (100 IU/ml) and streptomycin (100 mg/ml) at 37C in a humidified 5% CO2 atmosphere.
Mitochondrial membrane potential (MMP) in living cells

The MMP of control and treated lymphocytes was studied by using the JC-1 probe. Following this method, cells were stained with 10 M of 5-5,6-6-tetrachloro1,1,3,3-tetraethylbenzimidazol-carbocyanine iodide (JC-1; Molecular Probes) as previously described [15]. Tetramethylrhodamine ester (1 M TMRM (red fluorescence); Molecular Probes) was also used to confirm data obtained by JC-1.
Redox balance
Resting and differently activated lymphocytes (5105), either treated or untreated with IDV, were incubated in 495 l of Hanks balanced salt solution (HBSS) (Gibco BR2, Burlington, ON, Canada) pH 7.4 with 10 M dihydrorhodamine 123 (DHR 123; Molecular Probes) or 1 M dihydroethidium (DHE; Molecular Probes) in polypropylene test tubes for 15 min at 37C. The median values of fluorescence intensity histograms were used to provide semi-quantitative analysis of reactive oxygen intermediate (ROI) production.
In vitro treatments
Purified T cells were activated for 72 h, in the presence or absence of 100 nM IDV (Merck Sharp & Dohme Ltd, Hotteson, UK), SQV (Roche Registration, Welwyn Garden City, UK) or LPV (Abbott Laboratories, Queenborough, UK), with i) interleukin 2 (IL2, 60 IU/ml; Gibco-BRL); ii) IL2 and phytohaemoagglutinin (PHA 2 g/ml; Sigma Chemical Co, St Louis, MO, USA); iii) HIV-1 gp120 (3 g/ml; Intracell Corp, Cambridge, MA, USA); and iv) IL2 and HIV-1 gp120. Isolated human lymphocytes (resting or differently activated) were treated with 500 ng/ml of an anti-human Fas IgM monoclonal antibodies (mAbs) (-Fas, clone CH11; Upstate Biotechnology, Lake Placid, NY, USA). For dose-dependent studies, T lymphocytes were treated with 100 nM, 1 M and 10 M IDV for 72 h. Because of overlapping results obtained by using the three different PIs, only results obtained with IDV, thus considered as representative, are shown in this paper.
M30
Immunocytochemistry
Immunofluorescence analyses were performed by double staining as follows. Control and treated lymphocyte samples were incubated with mAb to CD95/Fas (Chemicon International, Temecula, CA, USA) directly conjugated with FITC for 1 h at 4C. Cells were then washed, fixed with 4% paraformaldehyde (w/v in PBS) for 1 h at room temperature and then made permeable
HIV protease inhibitors supply mitochondria homeostasis
by 0.5% (v/v) Triton X-100 for 5 min. After washing, samples were incubated for 1 h at 4C with a mAb to ezrin (Becton Dickinson, Mountain View, CA, USA). Negative controls were incubated with isotypic immunoglobulins. After washing, samples and isotypic controls were incubated for 45 min at room temperature with anti-mouse IgG TRITC-conjugate (Sigma Chemical Co). All samples were mounted with specific medium and analysed by intensified video microscopy (IVM) as previously reported [16].
Evaluation of uncoupler protein 2 (UCP-2) expression

Western blot. Aliquots of total protein extracts (20 g) from T cells after different treatments were size fractioned by 12% SDS-PAGE and, after transfer to nitrocellulose membrane, filters were incubated with primary polyclonal antibody (Pab) against UCP-2 (Calbiochem Co, Darmstadt, Germany). Detection was achieved using HRP-conjugated secondary Pab and by the ECL detection system (Amersham-Pharmacia, Arlington Heights, IL, USA). Densitometric analyses were carried out by using a densitometer (Biomed Instruments, Inc, Fullerton, CA, USA) and results were expressed as arbitrary units (AU). Flow cytometry. Control and IDV-treated lymphocytes were washed, fixed with 4% paraformaldehyde (w/v in PBS) for 1 h at room temperature and then permeabilized by 0.5% (v/v) Triton X-100 for 5 min. For quantification of uncoupler protein 2 (UCP-2), samples were incubated for 1 h at room temperature with specific Pab (Calbiochem). Negative controls were incubated with normal rabbit serum. After several washings, samples and isotypic controls were incubated in the dark for 45 min at room temperature with Alexa488conjugated anti-rabbit IgG (Molecular Probes), washed and analysed on a cytometer.
PBS, cells were resuspended in Homo buffer (10 mM Hepes, pH 7.4; 1 mM ethylene glycol-bis(-aminoethyl ether) N,N,N-tetraacetic acid (EGTA), 0.1 M sucrose, 5% bovine serum albumin (BSA), 1 mM phenylmethylsulphonyl fluoride and complete protease inhibitor cocktail (Roche, Indianapolis, IN, USA) and maintained for 10 min on ice. After this time, cells were homogenized with about 100 strokes of a Teflon homogenizer with Btype pestle as previously reported [17] for 10 min at 4C to remove intact cells and nuclei, and the supernatants were further centrifuged at 10 000g at 4C for 10 min to precipitate the heavy membrane fractions (enriched in mitochondria). These fractions were then purified by standard differential centrifugation. The mitochondrial pellet obtained was resuspended in swelling buffer (SB) containing 0.1 M sucrose, 0.5 M sodium succinate, 50 mM EGTA at pH 7.4, 1 M phosphoric acid (H3PO4), 0.5 M 3-(N-morpholino) propane sulphonic acid (MOPS) and 2 mM rotenone, kept on ice and used within 2 h of the preparation. Protein content in the mitochondrial preparation was determined by a spectrophotometric method using BSA as standard. The purity of the mitochondria preparation was assessed by Western blot, checking subunit I of cytochrome c oxidase (mAb; Chemicon, Temecula, CA, USA).
Swelling experiments
Mitochondria (0.5 mg protein/ml) from each sample (gp120, IDV+gp120 and FCCP+gp120) were resuspended in SB at a final volume of 3 ml and separated into two samples. In the first sample, to induce mitochondrial swelling mimicking physiological conditions, we used 1 g/ml of recombinant Bid protein cleaved by caspase 3 (t-Bid; PeproTech, Inc, Rocky Hill, NJ, USA). Mitochondria from gp120, IDV+gp120-treated and FCCP+gp12-treated lymphocytes were incubated with 1 g/ml t-Bid 2 h at 37C. In the second, control sample, mitochondria from the same samples were incubated in SB without t-Bid. Flow cytometry analysis. We analysed the of isolated mitochondria by using a cytofluorimetric method after staining with 1 M TMRM. Using this method, the incorporation of the dye TMRM was monitored in the FL3 channel: low levels of TMRM incorporation (revealed by a decrease of red fluorescence) indicated a low . On this basis, in our experiments we tested the effect produced by t-Bid on the of gp120-, IDV+gp120- or FCCP+gp120-treated samples. As a general rule, we recorded 5 min control samples (without t-Bid) and, after this time, we started to record t-Bid-treated samples for a total recording time of 25 min. All samples were analysed with a FACScan cytometer (Becton Dickinson) equipped with a 488 argon laser. To exclude debris, samples were gated
M31
Transmission electron microscopy (TEM)

For TEM examination, cells were fixed in 2.5% cacodylate-buffered (0.2 M, pH 7.2) glutaraldehyde for 20 min at room temperature and post-fixed in 1% OsO4 in cacodylate buffer for 1 h at room temperature. Fixed specimens were dehydrated through a graded series of ethanol solutions and embedded in Agar 100 (Agar Aids, Cambridge, UK). Serial ultrathin sections were collected on 200-mesh grids and then counterstained with uranyl acetate and lead citrate. Sections were observed with a Philips 208 electron microscope at 80 kV.
Preparation of isolated mitochondria

Gp120-, IDV+gp120- and carbonyl cyanide p-trifluoromethoxyphenyl hydrazone (FCCP)+gp120-treated PBL were collected by centrifugation. After three washings in
P Matarrese et al.
based on light-scattering properties in the side scattering and forward scattering modes during analyses. The red fluorescence emission (due to TMRM dye) of untreated mitochondria was set up in correspondence of the 102 channel and considered as basal emission. Dot plots of red fluorescence emission as a function of the time, obtained in each condition we used, were statistically analysed by using CellQuestTM software (Version 3.3) on a Macintosh (Becton Dickinson) in order to determine the percentage of mitochondria with depolarized membrane.
Analysis of cytochrome c release

Cytochrome c (CytC) was evaluated in supernatants from isolated mitochondria by a sensitive and specific immunoassay, using a commercial ELISA kit (R&D Systems, Minneapolis, MN, USA) according to the manufacturers instructions. The light emitted was quantified by using a microtitre plate reader at 405 nm. CytC concentration was expressed in ng/ml.
Data analysis and statistics

All samples were analysed with a FACScan cytometer (Becton Dickinson) equipped with a 488 argon laser. At least 20 000 events were acquired. Data were recorded and statistically analysed using CellQuestTM. The expression level of considered proteins was expressed as the median value of fluorescence emission curve and the statistical significance was calculated using the parametric KolmogorovSmirnov (K/S) test. Statistical analysis of apoptosis data was performed by using Students t-test or one-way variance analysis by using the StatView program (Version 5.0) for Macintosh (SAS Institute, Inc, Cary, NC, USA). All data reported in this paper were verified in at least four different HDs and expressed as mean standard deviation (SD). Only P<0.01 was considered as significant.
Results
IDV hinders gp120-induced T-cell polarization
Viral envelope protein gp120 was able to induce a wide variety of changes in human T lymphocytes that are typical of activation, for example, i) the increased expression of T-cell activation markers and ii) the redistribution of molecules of importance in determining lymphocyte polarization and fate, that is, survival or apoptosis. In particular, regarding the first point, we evaluated the surface expression of CD69 (Figure 1A), HLA-DR (Figure 1B) and CD38 (Figure 1C). A significant increase of these molecules on the surface of T cells was detected after gp120 administration (Figure 1). Interestingly, these results were comparable with those obtained by using typical activating factors such as IL2 and PHA (Figures 1AC). Importantly, pre-exposure of
M32
human lymphocytes to PIs was ineffective in this respect and the up-regulation of these activation markers remained still detectable (Figure 3AC). Regarding the molecules of importance in determining T cell fate, the following results were obtained. Firstly, CD95/Fas expression was significantly increased by administration of gp120 either given alone or in association with IL2 (Figure 1D). Furthermore, in consideration of the wellknown ability of various PIs to impair T cell apoptosis [11,18], specific measurements of CD95/Fas expression were carried out in the presence of IDV. Notably, as described above for activation markers (Figures 1AC), the presence of IDV did not influence this parameter and the gp120-induced over-expression of CD95/Fas was well detectable either in the presence or absence of IDV (Figure 1D). Secondly, a peculiar rearrangement of this molecule was observed. In particular, cellular localization of CD95/Fas was found to be modified by gp120. In fact, whilst control resting T cells appear nonpolarized (Figure 2A, first row), gp120-treated cells underwent profound shape changes mainly represented by cell polarization. They appeared characterized by an accumulation of CD95/Fas molecules at one pole of the cell positive patches at the cell surface, confined in the subcellular erniations called uropods were detectable by IVM analyses (Figure 2A, second row, left panel). This polarization of the CD95 molecule was paralleled by a rearrangement of a cytoskeletal molecule previously demonstrated to play a key role in CD95/Fas-induced cell death, representing a pre-requisite for Fas signalling to occur: the ezrin molecule [16,19]. In fact, after gp120 exposure, in most of the lymphocytes the ezrin molecule appeared localized at one pole of the cell. Importantly, this cytoskeleton protein also appeared to co-localize with CD95/Fas at the uropod level (Figure 2A, second row, middle and right panels; compare with resting cells shown in the first row). Notably, IDV treatment was capable of hindering these processes and both CD95/Fas and ezrin remained randomly distributed at the cell surface or in the cell cytoplasm, respectively, and no co-localization of these molecules was detectable (Figure 2A, third row). Altogether these results seem to indicate that: i) gp120 activation can predispose T lymphocytes to Fasmediated apoptosis via uropod formation, CD95 polarization and ezrin cytoskeleton reorganization and ii) the presence of IDV was capable of impairing these processes. Ultrastructural analysis by TEM also point out some morphogenetic changes caused by gp120, such as cell shape remodelling, as well as mitochondria ultrastructural changes. In particular, gp120 administration led to: i) a redistribution of organelles, mainly of mitochondria, at one pole of the cell and ii) an alteration of their ultrastructural features (Figure 2B, middle
Figure 1. IDV does not influence gp120-induced increase of activation markers
A
40 35 30 25 20 15 10 5 0
B
60 50 HLA-DR expression 40 30 20 10
CD69 expression
in est
IL2 /PHA p120 0/IL2 +IDV +IDV +IDV +IDV +IDV g 12 ing IL2 HA 20 IL2 IL2 /P gp1 20/ gp rest 1 IL2 gp
res
tin
C
250
D
45 40 CD95/Fas expression 200 35 30 25 20 15 10 5 0 0
CD38 expression
150 100 50
in est
res
tin
Quantitative cytofluorimetric analysis of surface expression of activation markers CD69, HLA-DR and CD38, and of CD95 molecule in resting and differently activated lymphocytes. Note that the significant (P<0.01) increase of (A) CD69, (B) HLA-DR, (C) CD38 and (D) CD95 due to lymphocyte activation was not modified by IDV. Data represent the mean of four independent experiments and are expressed as median values of fluorescence intensity histograms obtained by semiquantitative flow cytometry analyses. IDV, indinavir; IL2, interleukin 2; PHA, phytohaemoagglutinin.
panel; compare with resting cells shown in Figure 2B, left panel). The first was characterized by the appearance of clusters of coalescent mitochondria migrated to one pole of the cell (Figure 2B, middle panel). The second was represented by ultrastructural modifications of mitochondria in terms of increased electrondensity and changes in the morphological features of cristae (Figure 2B, middle panel, inset; compare with control cell mitochondria shown in the inset in Figure 2B, left panel). Importantly, IDV administration (Figure 2B, right panel) prevented the mitochondrial morphology and distribution changes being similar to those detected in resting lymphocytes (compare left and right panels in Figure 2B).
Mitochondrial membrane potential

It has previously been reported that during T lymphocyte activation, an early increase of MMP clearly occurs [20,21]. This alteration was associated with T cell sensitization to Fas-induced apoptosis [18,22,23]. On the basis of these published data, the aim was to assess the early effects of HIV-1 gp120 protein on mitochondrial function, that is, on mitochondrial . Thus, MMP was evaluated in T cells exposed to gp120 by using two different probes specifically employed in flow cytometry studies, that is, TMRM and JC-1 [15]. Figure 3A shows the results obtained in lymphocytes from a representative HD, whereas in Figure 3B the mean of values obtained by analysing four different
M33
P Matarrese et al.
Figure 2. IDV hinders gp120-induced T cell polarization
A
CD95 ezrin merge
resting
gp120
IDV+gp120
B
resting gp120 IDV+gp120
(A) Immunofluorescence analysis of CD95 and ezrin molecules after double staining of T lymphocytes with specific antibodies. Micrographs show that, after gp120 administration, the CD95 molecule was polarized at the uropod (second row, green fluorescence) where it co-localized with ezrin (second row, red fluorescence) as demonstrated by merge pictures in the right column (second row, yellow fluorescence). In IDV/gp120-treated lymphocytes, the uropod formation was inhibited and no CD95/ezrin co-localization was detectable (third row) as in resting cells (first row). (B) Ultrastructural analysis by TEM of resting and gp120-activated T lymphocytes in the presence or absence of IDV. Control (untreated cells) display mitochondria randomly distributed throughout the cell cytoplasm (left panel). After gp120 treatment, mitochondria appeared to have migrated to one pole of the cell, characterized by electron dense matrix and rearrangement of cristae (middle panel). These alterations were prevented by IDV (right panel). Results obtained from a representative HD out of four are shown. IDV, indinavir.
M34
Figure 3. IDV hinders mitochondria hyperpolarization and redox changes associated with T lymphocyte activation
104 103 J-aggregates 102 101 100
resting
104 103 102 101 100 103 104 100 101
gp120
104 103 102 101 100
IDV+gp120
100 101 102 J-monomers
102
103
104
100
101
102
103
104
B
% of cells with hyperpolarized mitochondria
70 60 50 40 30 20 10 0 resting IL2 IL2 gp120 IL2 resting IL2 PHA gp120 IL2 gp120 IL2 PHA gp120 IDV
Superoxide anion production (arbitrary units) Hydhrogen peroxide production (arbitrary units)
100 80 70 40 cell count 20 0 100 101 DHE (FL2) 102 103 104 resting M=44.1 gp120 M=68.5 IDV+gp120 M=46.9
70 60 50 40 30 20
res
100 80 70 40 cell count 20 0 100 101 DHR123 (FL2) 102 103 104 gp120 M=60.4 resting M=30.7 IDV+gp120 M=23.5
g 2 A 0 0 g 2 A 0 0 tin IL PH p12 p12 estin IL PH p12 p12 IL2 g IL2 g IL2 g IL2 g r IDV
70 60 50 40 30
res
g 2 A 0 0 g 2 A 0 0 tin IL PH p12 p12 estin IL PH p12 p12 IL2 g IL2 g r IL2 g IL2 g IDV
(A) Quantitative cytofluorimetric analyses of MMP in resting (left panel) and in gp120-activated lymphocytes in the presence or absence of IDV as detected using the JC-1 probe. J-aggregates (red fluorescence, ordinate) increased after gp120 administration, when mitochondrial membrane became hyperpolarized (middle panel). This mitochondria hyperpolarization was significantly (P<0.01) prevented by IDV (right panel). Numbers in the boxed areas represent the percentage of cells with hyperpolarized mitochondria, while in the area under the dashed line, the percentage of cells with depolarized mitochondria is reported. Results obtained from a representative HD out of four are shown. (B) Flow cytometry MMP analysis performed in different activating conditions. Results obtained by pooling together data from four different HDs are reported. (CF) Semiquantitative cytofluorimetric analyses of ROI production performed by using DHE (C and D) and DHR123 (E and F) in resting and gp-120-activated lymphocytes in the presence (or absence) of IDV. Values reported represent the median values of the fluorescence intensity histograms. In (E) and (F), the means SD of the results obtained form four different HDs are reported. Note that, independently from the activating condition used, IDV was capable of significantly (P<0.01) preventing either anion superoxide (C,D) or hydrogen peroxide (E and F) production. HD, healthy donors; IDV, indinavir; IL2, interleukin 2; MMP, mitochondrial membrane potential; PHA, phytohaemoagglutinin; ROI, reactive oxygen intermediate.
M35
P Matarrese et al.
HDs are reported. We found that, with respect to control samples (Figure 3A, left panel, boxed area and Figure 3B), gp120 induced a marked increase of MMP (hyperpolarization) in a significant percentage of cells (Figure 3A, middle panel, boxed area and Figure 3B). Of note is that the mitochondrial hyperpolarization phenomenon, as detected by flow cytometry, seems to correspond to changes in the mitochondrial structural features, as revealed by TEM analysis and shown in Figure 2B (middle panel). Importantly, pre-exposure to IDV completely impaired this mitochondrial hyperpolarization state and the percentage of cells with increased MMP was similar to that found in control samples (Figures 3A, compare right and left panels, boxed areas, and histograms in Figure 3B).
IDV hinders redox changes associated with gp120 administration

Production of reactive oxygen species was associated with mitochondrial changes occurring during T cell activation [23,24]. We thus evaluated two of the major parameters of importance in defining the redox state of a cell: superoxide anion and hydrogen peroxide production by a semiquantitative flow cytometry analysis. It has been found that gp120 was able to act as a mild pro-oxidant compound leading to an increased production of both superoxide anion (Figure 3C, a representative HD; Figure 3D, mean results obtained from lymphocytes from four different HDs) and hydrogen peroxide (Figure 3E, a representative HD; Figure 3F, mean results obtained from lymphocytes from four different HDs). Interestingly, IDV treatment was able to significantly protect lymphocytes from gp120-induced oxidative imbalance, as demonstrated by fluorescence median values very similar to those found in control samples.
IDV hinders T cell sensitization to CD95/Fas-mediated apoptosis

The increase of CD95/Fas cell surface expression, the ezrin/CD95/Fas polarization and co-localization, as well as mitochondria hyperpolarization and prooxidant conditions are known to be associated with an increased susceptibility to Fas-induced apoptosis [19,22,25]. In the following experiments we evaluated whether gp120 administration was able to sensitize T cells to various apoptotic stimuli. The analysis of apoptosis, after double staining of cells with annexin V-FITC and propidium iodide, clearly indicated that i) gp120 was not able to induce apoptosis per se (Figure 4A, first row, middle panel) but ii) it was able to sensitize T cells to CD95/Fas-mediated apoptotic triggering (Figure 4A, second row, middle panel) and that iii) IDV was capable of significantly preventing lymphocyte apoptosis (Figure 4A, second row, right panel).
M36
Furthermore, the anti-apoptotic activity exerted by IDV was also assessed in T lymphocytes activated by standard protocols and exposed to triggering -Fas (Figure 4B). Data reported in Figure 4B, clearly show that IDV also significantly (P<0.01) prevented Fasinduced apoptosis in lymphocytes activated by IL2, IL2/PHA or IL2/gp120. On the basis of these results, we then considered whether IDV treatment was capable of hindering apoptosis associated with agents specifically influencing mitochondrial homeostasis. To this purpose, the cytokine interferon- (IFN-) and the drug staurosporin (STS), known to be able to modify mitochondrial homeostasis [26,27] were used. However, these agents were used at low concentrations and were, therefore, unable to induce any pro-apoptotic effect per se. The results obtained are shown in Figures 5A and 5B and can be summarized as follows. Firstly, in the presence of low concentrations of IFN- (Figure 4A, boxed area) and STS (Figure 4B, boxed area) the percentage of cells with hyperpolarized mitochondria was significantly increased with respect to resting control lymphocytes without any sign of cell death (see Figure 3A, left panel). Secondly, according to the above results (see Figure 3), lymphocytes were hypersensitive to -Fas-triggering that also induced, as a late event, depolarization of mitochondrial membrane, typical of apoptotic execution. This was demonstrated by the high percentage of cells characterized by low red fluorescence emission when stained by JC-1 (see area under dotted line). Thirdly, and most importantly, in contemporaneous treatments with IDV and IFN- (IDV/IFN) or IDV and STS (IDV/STS), we observed a significant impairment of IFN- and STS-mediated increase of , a reduction of T lymphocyte -Fasinduced apoptosis and, accordingly, an inhibition of loss, that is, mitochondria depolarization typical of apoptotic execution phase (Figures 5C and 5D, respectively; in Figure 5E mean values from four different HDs are shown).
IDV inhibits t-Bid-induced swelling in isolated mitochondria

The above results seem to suport a direct effect of IDV on mitochondria polarization state. In order to further analyse this point, we first investigated whether IDV was able to prevent mitochondrial swelling in isolated mitochondria. To this end, we treated mitochondria isolated from human T lymphocytes with recombinant Bid protein cleaved by caspase 3 (t-Bid). It is in fact well known that Bid, a pro-apoptotic protein belonging to the Bcl-2 family, represents a key effector of the CD95-induced apoptotic pathway acting on mitochondria [28]. Figure 6A shows a representative profile of the swelling induced by 1 g/ml t-Bid on mitochondria
Figure 4. IDV hinders CD95/Fas-mediated apoptosis in activated T cells
104 103 102 101 100 100 104 103 102 101 PI 100 100
resting
104 103 102 101 100 100 104 103 102 101 100 100
gp120
104 103 102 101 100 100 104 103 102 101 100 100
IDV+gp120
IgM
101
102
103
104
101
102
103
104
101
102
103
104
-Fas
101
102
103
104
101
102
103
104
101
102
103
104
Annexin V-FITC
60 50 40 30 20 10 0
% apoptotic cells
IgM control -Fas (CH11)
res
tin
IL2
IL2
/PH
1 gp
20 IL
p 2/g
12
0 re
+I ng sti
DV
IL2
+I
DV IL2
A /PH
+I
DV g 2 p1
0+
ID
V /g 2 p1
ID 0+
IL2
(A) Percentage of apoptotic cells (as revealed by annexin V/propidium iodide double staining) in gp120-activated PBL from a representative HD. Note that gp120 i) did not induce apoptosis per se, ii) sensitized T cells to CD95-induced apoptosis and iii) that IDV was able to significantly reduce apoptosis (P<0.01) independently from the activating stimulus taken into consideration. (B) Mean SD of the results obtained in four different HDs. PI, propidium iodide.
isolated from gp120-activated lymphocytes monitored by means of variations in TMRM fluorescence. It is very evident that t-Bid induced a rapid decrease of MMP (swelling phenomenon) in a significant percentage of cells (19.6%, grey area). In particular, the MMP, expressed as a median value of the TMRM fluorescence curve significantly decreased from 249.4 to
89.4 (Figure 6B; control: plain black histogram; t-Bid: empty black histogram). Importantly, in accordance with the above results obtained in intact cells, mitochondria derived from IDV-treated gp120-activated lymphocytes were more resistant to t-Bid-induced swelling, being the percentage of depolarized mitochondria significantly lower than that found in
M37
P Matarrese et al.
Figure 5. IDV is capable of hindering apoptosis associated with agents specifically influencing mitochondrial homeostasis
JC-1
Annexin V/PI
JC-1
104 103 102
control IgM
IFN- 1000 IU
104
101 100 100 101 102 103 104
102 104 100 100 102 104 103 102 101 100 100 101 102 103
+ -Fas
104
102
104
100 100
102
104
104 103 102 104
control IgM
STS 1 nM
101
102
100 100
104
101
102
103
104
+ -Fas
104
100 100 102 104
103 102 101 100 100 101 102 103 104 100 100 102 104 102
Cytofluorimetric analysis of MMP (by JC-1) and of CD95-induced apoptosis (by annexin V/propidium iodide double staining) in PBL of a representative HD treated with 1000 IU IFN- (A,C) or 1 nM STS (B,D) in the absence (A,B) or presence (C,D) of IDV. In the JC-1 dot plots, numbers in the boxed areas represent the percentages of cells with hyperpolarized mitochondria, while in the area under the dashed line, the percentage of cells with depolarized mitochondria is reported. In annexin V/propidium iodide dot plots, numbers reported in the lower and upper right quadrants represent the percentages of annexin V single positive cells and annexin V/propidium iodide double positive cells, respectively. In (E) the analysis of CD95-induced apoptosis is reported as mean SD of the results obtained from four different HDs. Note that IDV was capable of significantly (P<0.01) preventing apoptosis either in INF-- or in STS-treated lymphocytes. HD, healthy donors; IDV, indinavir; IFN, interferon; IL2, interleukin 2; MMP, mitochondrial membrane potential; PBL, peripheral blood lymphocytes; STS, staurosporin.
M38
Figure 5 continued
104 103 102 104
control IgM
IDV + IFN-
101 100 100 10

4
102
101
102
103
104
+ -Fas
104 103 102 101 100 100 101 102 103 104 100 100 102
100 100 102 104
102
104
D
104 103 102 104
control IgM
IDV + STS
101 100 100 101 102 103 104
102
+ -Fas
104 104
100 100 102 104
103 102 101 100 100 101 102 103 104 100 100 102
102
104
80 70 % apoptotic cells 60 50 40 30 20 10 0 IFN- STS IFN-/IDV STS/IDV control IgM -Fas
M39
P Matarrese et al.
Figure 6. IDV inhibits t-Bid-induced mitochondrial swelling
104
B
count
200 160 120 t-Bid M=89.4 80 40 0 control M=249.4
100
TMRM (FL3) 101 102 103
time
5 min
25 min
100
101
102
103
104
104
200 160 t-Bid M=165.4 IDV M=196.3
TMRM (FL3) 101 102 103
count 5 min 25 min
120 80 40 0 100 101 102
100 0
time
103
104
104
F
200 160 count 120 80 40 0 time 5 min 25 min 100 101 102 103 104 t-Bid M=154.3 FCCP M=171.4
100 0
TMRM (FL3) 101 102 103
G
MMP
H
300 250 200 150 100 50 0
M rol nt 0 Co 2 30 L C Ca 0 0 d d 20 Bid 12 Bi 12 Bi gp 0+t- +gp 0+t- +gp1 0+t12 IDV p12 CCP p12 gp F +g g V+ CP ID FC
45 40 35 30 25 20 15 10 5 0
0 10 M n X 300 o t tri aCL2 C gp 12 I 0 + DV gp id id -B -B +t 0+t 2 1 gp +gp V ID 12 0 12 0
Cytofluorimetric dot plots represent the profile of the mitochondrial swelling monitored by means of variations in TMRM fluorescence as a function of time in mitochondria isolated from untreated lymphocytes. (A) Swelling was evaluated after treatment with t-Bid [histograms in (B) show the corresponding MMP values], (C) after treatment with IDV [histograms in (D) show the corresponding MMP values), or (E) with the uncoupler FCCP [histograms in (F) show the corresponding MMP values]. Results obtained in freshly isolated T lymphocytes from a representative HD are shown. (A,C,E) Numbers in grey areas in represent the percentage of mitochondria that underwent MMP decrease. Note that i) t-Bid induced swelling in a high percentage (19.6%) of mitochondria of untreated lymphocytes [(A), grey area], while ii) mitochondria derived from IDV- and FCCP-treated lymphocytes were significantly more resistant to t-Bid-induced swelling as shown by the percentages reported in grey areas of (B) and (C), and iii) parallel evaluations indicated the MMP loss induced by t-Bid (B) was counteracted by both IDV (D) and FCCP (F). (G) MMP mean values SD obtained by evaluating lymphocytes from three different healthy donors. Note the significant protection (P<0.001) exerted by both IDV and FCCP against t-Bid induced depolarization. Calcium chloride, able to induce MMP loss, represented a positive control. Gp120 was not able per se to induce mitochondrial membrane depolarization. (H) Results obtained by evaluating CytC release from isolated mitochondria treated with gp120 are reported. Calcium chloride, used as positive control, as well as recombinant t-Bid were able to induce the release of a significant amount of CytC. Conversely, treatment with IDV significantly (P<0.001) decreased the amount of CytC released from isolated mitochondria. Gp120 was not able to induce CytC release per se. CytC, cytochrome c; IDV, indinavir; FCCP, carbonyl cyanide p-trifluoromethoxyphenyl hydrazone; MMP, mitochondrial membrane potential, TMRM, tetramethylrhodamine ester.
M40
ng/ml
gp120-treated samples (Figure 6C; 5.4%, grey area). In fact, the median value of the TMRM fluorescence curve obtained from mitochondria isolated from IDVtreated lymphocytes was only lightly modified (median values reduced from 196.3 to 165.4, Figure 6D; control: plain black histogram; t-Bid: empty histogram). As a control for this series of experiments, we used the protonophore uncoupler carbonyl cyanide fluorophenyl-hydrazone (FCCP), known to be capable of hindering MMP loss at low concentrations [18,29]. Interestingly, FCCP, similarly to IDV, significantly prevented t-Bid-induced mitochondrial swelling (Figure 6E, grey area) as well as the decrease of MMP as measured by TMRM fluorescence emission (Figure 6F; control: plain black histogram; t-Bid: empty histogram). Mean values SD obtained from three independent experiments are shown in Figure 6G. A significant protection exerted by both IDV and FCCP towards MMP changes induced by t-Bid was detected. We have also evaluated the ability of t-Bid to induce the release of CytC from mitochondria purified from human lymphocytes. The supernatants obtained from gp120-treated mitochondria (swelling experiments), before any TMRM staining, were analysed by means of ELISA methodology. Figure 6H shows i) as expected, calcium chloride 300 M, used as positive control, induced the release of a significant amount of CytC; ii) recombinant t-Bid also induced CytC release, although in a minor extent; and iii) either IDV or a baby dose of the FCCP compound (not shown), significantly (P<0.01) decreased the amount of CytC released from isolated mitochondria. These data suggest that IDV could act similarly to FCCP given at low concentration, that is, via a mitochondrial uncoupler-like activity. A family of endogenous uncoupling proteins are normally detectable in the inner mitochondrial membrane of different tissues. These are called uncoupler proteins (UCPs) and are able to regulate cell energy and mitochondrial membrane potential protecting cells from ROI production and apoptosis [30,31]. On the basis of these considerations, an analysis of the expression of UCP-2 (an ubiquitously expressed UCP isoform) was thus performed either by flow cytometry or by Western blotting (WB) analyses. The results obtained clearly indicated an IDVdependent down-regulation of UCP-2 protein in gp120-activated T cells (Figure 7A, empty grey histogram) with respect to gp-120-activated control cells (Figure 7A, empty black histogram; UCP-2 expression in resting T cells is shown as a plain histogram). Statistical analyses of these data by K/S tests also indicated that this IDV-induced UCP-2 decrease was highly significant (Figure 7A, right panels). Mean values SD obtained on T cells from four different HDs are
reported in Figure 7B. These results show a clear-cut doseeffect response of IDV in gp-120-activated lymphocytes. In fact, a increased down-regulation of UCP-2 expression was detectable with increasing IDV concentrations. Notably, as expected [18], no effect of IDV was found in resting T cells. These results were confirmed by WB analyses, as demonstrated by the use of increasing doses of IDV, that is, 0.1, 1 and 10 M (see densitometric measurements in Figure 7C).
Discussion
Freshly isolated lymphocytes represent a peculiar model for studying apoptotic cell death. In fact, resting T cells have been considered as apoptotic-resistant cells [32]. By contrast, once activated, for example, by IL-2 and PHA, they become susceptible to different proapoptotic stimuli, including the CD95/Fas physiological receptor triggering. Interestingly, among various changes occurring in activated T cells, two of them are gathered the attention of immunologists involved in the study of immune system homeostasis: cell polarization (the migration of organelles and surface molecules at one pole of the cell) and an early mitochondrial change detectable by different flow cytometry methods an increase of MMP (a mitochondrial hyperpolarization) [21,33]. In this context, several studies have been carried out as regards the activity of HIV PIs. These compounds have been demonstrated to impair apoptosis, improving T cell survival and to act independently from the presence of HIV particles inside T cells. The ability of various PIs to lead to the so-called immune reconstitution has in turn stimulated a plethora of experimental studies aimed at the comprehension of their mechanisms of action. In these studies, PIs have been demonstrated to be able to hinder apoptosis via a target effect on mitochondria [7,18]. In particular, the mitochondrial hyperpolarization state as detectable in T lymphocytes following activation with IL-2/PHA administration, was suggested to represent a prerequisite for apoptotic susceptibility of T cells and was found to be inhibited by PI treatments [18]. Accordingly, in the present work, we shepherd through the mechanisms underlying this peculiar effect exerted by PIs by using a model system that, in our opinion, could be of great relevance in AIDS pathogenesis studies, that is, gp120-treated, freshly isolated human T lymphocytes. Engagingly, the administration of HIV envelope protein gp120, which is responsible for HIVhost cell interaction and detectable in the blood of HIV-infected patients, was capable of inducing a series of changes. Namely, gp120 induces a significant increase of some activation markers and overexpression of CD95/Fas; organelle, cytoskeletal components and surface molecule polarization; mitochondrial
M41
P Matarrese et al.
Figure 7. IDV decreases UCP-2 expression in gp120-treated T lymphocytes
A
200 160 negative control M=3.6 IDV 100 nM+GP120 M=51.4
100 80 60
% events
40 20 0
K/S analysis resting vs gp120 D=0.09 p>0.05 100 101 103 102 Channels 104
120 80 cell count resting M=68.2 gp120 M=64.1
100
% events
80 60 40 20 0
40 0 100 101 UCP-2 expression 102 103 104
K/S analysis resting vs IDV+ gp120 D=00.81 p>0.001 100 101 103 102 Channels 104
B
UCP-2 expression (arbitary units)
90 75 resting 60 45 30 control IDV 100 nM IDV 1 M IDV 10 M gp120-activated
1 gp
20
1 gp
20
/I
DV
0 10
nM 1 gp 20 /I DV
M 1 gp 20 /I DV
10
UCP-2
AU 0.82 0.64 0.51
0.16
(A) Quantitative flow cytometry analysis of UCP-2 protein expression in gp120-activated lymphocytes treated with IDV in comparison with resting untreated lymphocytes. Note UCP-2 down-regulation detectable after IDV treatment (grey curve). Statistical analyses (K/S test) indicated that IDV significantly (P<0.01) decreased UCP-2 expression in gp120-activated cells only, whilst resting lymphocytes remained unaffected (right hand panels). Numbers reported indicate the median values of fluorescence intensity histograms obtained in T lymphocytes isolated from a representative HD. (B) Semiquantitative flow cytometry analysis of UCP-2 expression in both resting and gp120-activated lymphocytes treated with different doses of IDV. Results are shown as means SD of the data obtained from six different HDs. Note the dose-dependent decrease of the UCP-2 protein. (C) Western blotting analysis of UCP-2 protein shows an IDV-induced dose-dependent decrease of UCP-2 expression in gp120-activated lymphocytes. Densitometric analyses confirmed the dose dependent activity of IDV. These results are expressed in arbitrary units (AU). IDV, indinavir; UCP-2, uncoupler protein 2.
M42
ultrastructural changes and hyperpolarization; and an increased apoptotic susceptibility. Conversely, the presence of PIs, particularly IDV, was irrelevant with regards to cell surface molecule overexpression, including that of CD95/Fas, while it significantly reduced cell polarization state and also blocked both MMP changes and apoptosis. Hence, PIs do not act by down-regulating cell surface molecules of importance in apoptotic triggering, that is, CD95/Fas, but, conversely, they specifically act by impairing cell morphogenetic changes influencing apoptotic proneness and blocking MMP increase. The importance of cell re-modeling, that is, uropod formation and cell polarization, as well as of mitochondrial stabilization, that is, the blockade of hyperpolarization state, was previously described in other in vitro and ex vivo systems [18,34,35]. Hence, our results clearly indicate that HIV-gp120 may act as a sort of booster of T cell alterations associated with the activation of apoptotic machinery and that PIs can lead to apoptosis hindering via a specific inhibitory effect on cell polarization and maintenance of mitochondrial homeostasis. In consideration of the role of apoptosis occurring in bystander cells of HIV-infected patients and of its role in the onset of immunodeficiency, our data seem to partially explain the possible mechanisms underlying the beneficial effects of HAART. They support, in fact, the concept that the immune reconstitution detected in vivo after PI exposure could also be due to a protective activity of these drugs against gp120-induced apoptosis sensitization of bystander cells. More generally, these results seem also to provide further evidence for a reappraisal of PIs in the clinical management of human immune deficiencies other than AIDS. Data herein reported also indicate some important new findings regarding the target effects of these drugs. One is represented by the ability of PIs to act by impairing reactive oxygen species formation detectable after gp120 administration. Hence, in consideration of the key role played by the redox imbalance in apoptotic cell death program [22], these results allow us to hypothesize that the IDV-induced cell survival can be the outcome of a sort of antioxidant activity exerted by PIs. It is in fact well known that mitochondria are the major source of free radical species originated during the execution phases of apoptotic cell death. On these basis, we cannot rule out the possibility that PIs, by acting as quenching agents, could contribute to maintaining the intracellular redox balance via their target activity on mitochondria, that is, via unexpected antioxidant properties. These hypotheses seem to be confirmed by the protective role of PIs towards apoptosis in model systems alternative to gp120-induced apoptosis sensitization. In fact, T cell treatment with non-cytotoxic doses of IFN- or STS, both considered
as mitochondriotropic agents [23,36], lead to an increased apoptotic proneness, which was significantly counteracted by IDV. Hence, an immunoregulatory cytokine such as IFN- as well as a drug, such as STS, directly acting on mitochondrial homeostasis, were both capable of sensitizing T cells to Fas-mediated apoptosis. Thus, the target activity of PIs to mitochondria could be responsible for their protective effects in the maintenance of mitochondrial homeostasis and, as a consequence, in the inhibition of T cell apoptosis. In consideration of the importance of IFNs in infectious diseases, these results also point to PIs as useful pharmacological modulators in a number of human innate and acquired immune diseases. Some interesting considerations derive from the analyses of the results obtained by low concentrations of the protonophore uncoupler FCCP. High concentrations of this compound were demonstrated to induce apoptosis in tumour cell lines [37]. In contrast, very low doses of FCCP could protect human T cells from CD95-induced apoptosis [38]. Our results emphasized that PIs and FCCP show impressive similarities in their mitochondrial effects. In fact, the results obtained in isolated mitochondria have shown the ability of both drugs to stabilize MMP after t-Bid administration. The truncated form of the Bid molecule was in fact considered as a pro-apoptotic signalling molecule reaching mitochondria after CD95/Fas stimulation. A partial confirmation of the above hypothesis was further offered by the PI-induced down-modulation of UCP-2 mitochondrial protein. This can reinforce the idea that mitochondria represent specific targets of PIs and also favour the intriguing hypothesis that PIs, for example, IDV, could exert a uncoupler-like activity, minimizing mitochondrial hyperpolarization occurring in the early phases of apoptosis. In fact, the mechanisms implicated in the maintenance of mitochondrial homeostasis and redox balance involve the coupling activity of the respiratory chain and are physiologically regulated by UCPs. The decreased UCP expression in the presence of PIs could also represent indirect evidence for a peculiar role of PIs that, vicariating UCP activity in activated lymphocytes, might sustain mitochondrial homeostasis impairing mitochondrial hyperpolarization, reduce redox imbalance and, later, inhibit loss. In addition, UCPs seem to play an important role also in some pathophysiological states [39] including obesity, insulin resistance [40] and atherosclerosis [41]. Finally, it is well known that UCPs also play a pivotal role in regulating the energy metabolism also in the adipose tissue, where UCPs have been isolated for the first time [42,43]. Thus, on the basis of the well-known adverse effects associated with HAART, that is, lipodystrophy, atherosclerosis and diabetes, the present results on the modulatory activity
M43
P Matarrese et al.
exerted by PIs on UCP molecule expression would also stimulate further studies on the pathogenetic mechanism underlying HAART-associated disease contributing, in the long run, to a refining of the therapeutic strategies.
protein acts as a negative regulator of apoptosis in a human lymphoblastoid cell line: possible implications for the pathogenesis of AIDS. Journal of Experimental Medicine 1998; 187:403413. 15. Cossarizza A, Franceschi C, Monti D, Salvioli S, Bellesia E, Rivabene R, Biondo L, Rainaldi G, Tinari A & Malorni W. Protective effect of N-acetylcysteine in tumor necrosis factor-alpha-induced apoptosis in U937 cells: the role of mitochondria. Experimental Cell Research 1995; 220:232240. 16. Parlato S, Giammarioli AM, Logozzi M, Lozupone F, Matarrese P, Luciani F, Falchi M, Malorni W & Fais S. CD95 (APO-1/Fas) linkage to the actin cytoskeleton through ezrin in human T lymphocytes: a novel regulatory mechanism of the CD95 apoptotic pathway. EMBO Journal 2000; 19:51235134. 17. Zamzami N, Maisse C, Metivier D & Kroemer G. Measurement of membrane permeability and permeability transition of mitochondria. Methods in Cell Biology 2001; 65:147158. 18. Matarrese P, Gambardella L, Cassone A, Vella S, Cauda R & Malorni W. Mitochondrial membrane hyperpolarization hijacks activated T lymphocytes toward the apoptoticprone phenotype: homeostatic mechanisms of HIV protease inhibitors. Journal of Immunology 2003; 170:60066015. 19. Luciani F, Matarrese P, Giammarioli AM, Lugini L, Lozupone F, Federici C, Iessi E, Malorni W & Fais S. CD95/phosphorylated ezrin association underlies HIV-1 GP120/IL-2-induced susceptibility to CD95(APO-1/Fas)mediated apoptosis of human resting CD4(+)T lymphocytes. Cell & Death Differentiation 2004; 11:574582. 20. Matarrese P, Cauda R & Malorni W. Activation-associated mitochondrial hyperpolarization hijacks T cells toward an apoptosis-sensitized phenotype. Cell & Death Differentiation 2003; 10:609611. 21. Banki K, Hutter E, Gonchoroff NJ & Perl A. Elevation of mitochondrial transmembrane potential and reactive oxygen intermediate levels are early events and occur independently from activation of caspases in Fas signalling. Journal of Immunology 1999; 162:14661479. 22. Perl A, Gergely P Jr, Puskas F & Banki K. Metabolic switches of T-cell activation and apoptosis. Antioxidants & Redox Signaling 2002; 4:427443. 23. Perl A, Gergely P Jr, Nagy G, Koncz A & Banki K. Mitochondrial hyperpolarization: a checkpoint of T-cell life, death and autoimmunity. Trends in Immunology 2004; 25:360367. 24. Puskas F, Gergely P, Niland B, Banki K & Perl A. Differential regulation of hydrogen peroxide and Fasdependent apoptosis pathways by dehydroascorbate, the oxidized form of vitamin C. Antioxidants & Redox Signaling 2002; 4:357369. 25. Sato T, Machida T, Takahashi S, Iyama S, Sato Y, Kuribayashi K, Takada K, Oku T, Kawano Y, Okamoto T, Takimoto R, Matsunaga T, Takayama T, Takahashi M, Kato J & Niitsu Y. Fas-mediated apoptosome formation is dependent on reactive oxygen species derived from mitochondrial permeability transition in Jurkat cells. Journal of Immunology 2004; 173:285296. 26. Matarrese P, Di Biase L, Santodonato L, Straface E, Mecchia M, Ascione B, Parmiani G, Belardelli F, Ferrantini M & Malorni W. Type I interferon gene transfer sensitizes melanoma cells to apoptosis via a target activity on mitochondrial function. American Journal of Pathology 2002; 160:15071520. 27. Wang GQ, Gastman BR, Wieckowski E, Goldstein LA, Rabinovitz A, Yin XM & Rabinowich H. Apoptosisresistant mitochondria in T cells selected for resistance to Fas signaling. Journal of Biological Chemistry 2001; 276:36103619. 28. Wei MC, Zong WX, Cheng EH, Lindsten T, Panoutsakopoulou V, Ross AJ, Roth KA, MacGregor GR, Thompson CB & Korsmeyer SJ. Proapoptotic BAX and BAK: a requi-
Acknowledgements
The authors thank Professor Antonio Cassone for critical reading of the manuscript. Supported by AIDS project 30F1 to WM.
References
1. Gougeon ML & Montagnier L. Programmed cell death as a mechanism of CD4 and CD8 T cell deletion in AIDS. Molecular control and effect of highly active anti-retroviral therapy. Annals of the New York Academy of Sciences 1999; 887:199212. Roshal M, Zhu Y & Plannells V. Apoptosis in AIDS. Apoptosis 2001; 6:103116. Hengartner MO. The biochemistry of apoptosis. Nature 2000; 407:770776. Schmitz I, Walczak H, Krammer PH & Peter ME. Differences between CD95 type I and type II cells detected with the CD95 ligand. Cell Death & Differentiation 1999; 6:821822. Banki K, Hutters E, Gonchoroff NJ & Perl A. Molecular ordering in HIV-induced apoptosis. Journal of Biological Chemistry 1998; 273:1194411953. Petit F, Arnoult D, Lelievre JD, Moutouh-de Parseval L, Hance AJ, Schneider P, Corbeil J, Ameisen JC & Estaquier J. Productive HIV-1 infection of primary CD4+ T cells induces mitochondrial membrane permeabilization leading to a caspase-independent cell death. Journal of Biological Chemistry 2002; 277:14771487. Phenix BN, Angel JB, Mandy F, Kravcik S, Parato K, Chambers KA, Gallicano K, Hawley-Foss N, Cassol S, Cameron DW & Badley AD. Decreased HIV-associated T cell apoptosis by HIV protease inhibitors. AIDS Research & Human Retroviruses 2000; 16:559567. Viora M, Di Genova G, Rivabene R, Malorni W & Fattorossi A. Interference with cell cycle progression and induction of apoptosis by dideoxynucleoside analogs. International Journal of Immunopharmacology 1997; 19:311321. Chavan SJ, Tamma SL, Kaplan M, Gersten M & Pahwa SG. Reduction in T cell apoptosis in patients with HIV disease following antiretroviral therapy. Clinical Immunology 1999; 93:2433.
2. 3. 4.
5.
6.
7.
8.
9.
10. Lu W & Andrieu JM. HIV protease inhibitors restore impaired T-cell proliferative response in vivo and in vitro: a viral-suppression-independent mechanism. Blood 2000; 96:250258. 11. Phenix BN, Lum JJ, Nie Z, Sanchez-Dardon J & Badley AD. Antiapoptotic mechanism of HIV protease inhibitors: preventing mitochondrial transmembrane potential loss. Blood 2001; 98:10781085. 12. Estaquier J, Lelievre JD, Petit F, Brunner T, Moutouh-De Parseval L, Richman DD, Ameisen JC & Corbeil J. Effects of antiretroviral drugs on human immunodeficiency virus type 1-induced CD4(+) T-cell death. Journal of Virology 2002; 76:59665973. 13. Air P, Torti C, Uccelli MC, Malacarne F, Palvarini L, Carosi G & Castelli F. Naive CD4+ T lymphocytes express high levels of Bcl-2 after highly active antiretroviral therapy for HIV infection. AIDS Research & Human Retroviruses 2000; 16:18051807. 14. Conti L, Rainaldi G, Matarrese P, Varano B, Rivabene R, Columba S, Sato A, Belardelli F & Gessani S. The HIV-1
M44
site gateway to mitochondrial dysfunction and death. Science 2001; 292:727730. 29. Giovannini C, Matarrese P, Scazzocchio B, Sanchez M, Masella R & Malorni W. Mitochondria hyperpolarization is an early event in oxidized low-density lipoproteininduced apoptosis in Caco-2 intestinal cells. FEBS Letters 2002; 523:200206. 30. Skulachev VP. Uncoupling: new approaches to an old problem of bioenergetics. Biochimica & Biophysica Acta 1998; 1363:100124. 31. Kagawa Y, Cha SH, Hasegawa K, Hamamoto T & Endo H. Regulation of energy metabolism in human cells in aging and diabetes: FoF(1), mtDNA, UCP, and ROS. Biochemical & Biophysical Research Communications 1999; 266:662676. 32. Roberts AI, Devadas S, Zhang X, Zhang L, Keegan A, Greeneltch K, Solomon J, Wei L, Das J, Sun E, Liu C, Yuan Z, Zhou JN & Shi Y. The role of activation-induced cell death in the differentiation of T-helper-cell subsets. Immunologic Research 2003; 28:285293. 33. Matarrese P, Testa U, Cauda R, Vella S, Gambardella L & Malorni W. Expression of P-170 glycoprotein sensitizes lymphoblastoid CEM cells to mitochondria-mediated apoptosis. Biochemical Journal 2001; 355:587595. 34. Perl A, Nagy G, Gergely P, Puskas F, Qian Y & Banki K. Apoptosis and mitochondrial dysfunction in lymphocytes of patients with systemic lupus erythematosus. Methods in Molecular Medicine 2004; 102:87114. 35. Matarrese P, Tinari A, Mormone E, Bianco GA, Toscano MA, Ascione B, Rabinovich GA & Malorni W. Galectin-1 sensitizes resting human T lymphocytes to Fas (CD95)mediated cell death via mitochondrial hyperpolarization, budding and fission. Journal of Biological Chemistry 2004; 280:69696985.
36. Carrozzo R, Rizza T, Stringaro A, Pierini R, Mormone E, Santorelli FM, Malorni W & Matarrese P. Maternallyinherited Leigh syndrome-related mutations bolster mitochondrial-mediated apoptosis. Journal of Neurochemistry 2004; 90:490501. 37. Aronis A, Melendez JA, Golan O, Shilo S, Dicter N & Tirosh O. Potentiation of Fas-mediated apoptosis by attenuated production of mitochondria-derived reactive oxygen species. Cell Death & Differentiation 2003; 10:335344. 38. Stoetzer OJ, Pogrebniak A, Pelka-Fleischer R, Hasmann M, Hiddemann W & Nuessler V. Modulation of apoptosis by mitochondrial uncouplers: apoptosis-delaying features despite intrinsic cytotoxicity. Biochemical Pharmacology 2002; 63:471483. 39. Argiles JM, Busquets S & Lopez-Soriano FJ. The role of uncoupling proteins in pathophysiological states. Biochemical & Biophysical Research Communications 2002; 293:11451152. 40. Lameloise N, Muzzin P, Prentki M & AssimacopoulosJeannet F. Uncoupling protein 2: a possible link between fatty acid excess and impaired glucose-induced insulin secretion? Diabetes 2001; 50:803809. 41. Blanc J, Alves-Guerra MC, Esposito B, Rousset S, Gourdy P, Ricquier D, Tedgui A, Miroux B & Mallat Z. Protective role of uncoupling protein 2 in atherosclerosis. Circulation 2003; 107:388390. 42. Garlid KD, Jaburek M, Jezek P & Varecha M. How do uncoupling proteins uncouple? Biochimica & Biophysica Acta 2000; 459:383389. 43. Garlid KD, Jaburek M & Jezek P. Mechanism of uncoupling protein action. Biochemical Society Transactions 2001; 29:803806.
Received 10 January 2005, accepted 23 March 2005
M45
Mechanisms of HIV and nucleoside reverse transcriptase inhibitor injury to mitochondria

Graeme Moyle
St Stephens HIV Research, Chelsea and Westminster Hospital, London, UK Tel: 020 8746 8000; Fax: 020 8746 6188 E-mail: gm@moyleg.demon.co.uk
Introduction
Mitochondria perform a range of biological functions and carry a number of factors involved in cell apoptosis. In particular, mitochondria are the key organelles in energy production in all human cells except erythrocytes. Energy, in the form of ATP, is produced through the oxidative phosphorylation pathway. HIV and other infectious agents as well as some nucleoside/nucleotide reverse transcriptase inhibitors (NRTIs) are known to affect mitochondrial (mt) function. A number of important clinical events occurring in individuals with HIV infection and on antiretroviral therapy (ART) have been linked to mt injury and dysfunction. In vitro studies have demonstrated that NRTIs may differ in their effects on mitochondria and may affect mitochondria in different cell lines in different ways. This is likely to influence the clinical syndromes associated with toxicity to these agents. Dideoxy-NRTIs (d-NRTIs) have the greatest affinity for mtDNA polymerase- [1], the enzyme responsible for mtDNA replication, whereas other nucleoside analogues may influence mt function through other mechanisms. These differences may be important when choosing techniques to evaluate the impact of antiretroviral agents on mitochondria. Lactic acidosis is the event that establishes proof that clinically important mt toxicity occurs in individuals receiving NRTIs, although causation is not necessarily established by this observation. For lactic acidosis to occur, the mt oxidative phosphorylation system, which normally removes H+ generated by the hydrolysis of ATP and the conversion of glucose-6phosphate to pyruvate, must be dysfunctional (Figure 1). Lactic acidosis in the critical care setting has a range of causes and these are not always reliably excluded in reports of individuals with HIV on ART who develop lactic acidosis. Not surprising, lactic acidosis has also been reported in individuals with HIV not currently receiving ART [2]. Other NRTI-associated adverse events where the weight of evidence favours mt injury include peripheral neuropathy, myopathy and hepatic steatosis. However, multiple aetiologies for these adverse events exist and should be routinely sought in individuals presenting with these effects. Other adverse effects observed during ART that may represent manifestations of, or be contributed to by mt toxicity include hypogonadism, pancreatitis, diabetes mellitus, proximal renal tubular dysfunction, anaemia and neutropaenia, and peripheral fat atrophy [3]. Manifestations of NRTI-related mt toxicity in adults do not commonly involve the brain, the organ most commonly affected in inherited mt disorders. The possibility of CNS mt toxicity arising in infants after in utero exposure to NRTIs has been raised in reports from France [4,5] but has not been found in a large retrospective analysis in the USA [6]. Certainly, an extremely high incidence of hyperlactataemia in infants after in utero NRTI (mainly AZT plus lamivudine) exposure has been reported [7,8], suggesting mt dysfunction may be common in infants exposed to nucleoside analogues in utero, possibly due to differential effects of NRTIs during ontogeny. Additionally,
M47
Clinical toxicities with a possible mt aetiology

A wide range of adverse events occurring in individuals with HIV infection, particularly those receiving NRTIbased therapy, have been suggested as relating to diminished mt function. The development programs of zidovudine (AZT), didanosine, zalcitabine and stavudine have all required dose de-escalations due to toxicities thought to be related to drug impact on mitochondria. Due to limitations in measurement technology, clinical studies of NRTIs, while collecting adverse event data, have not specifically evaluated mt toxicity in a systematic way.
G Moyle
Figure 1. (A) L-Lactic acid synthesis: lactic acid accumulates if pyruvate or NADH accumulates. (B) DCA effects: DCA activates PDH, favouring lactate oxidation
A
Glycogen Glucose G6P Protein Amino acids
the proportion of mutant versus functional mtDNA present. In the HIV setting, as with many other drugs, this assessment may not be appropriate as, while some drugs (mainly the d-NRTIs) may predominately affect mtDNA content (by inhibition of DNA polymerase-), others (most notably AZT) may effect mt enzymes or other nuclear gene expression [1519].
The impact of HIV infection on mitochondria

Urea Pyruvate + H+ + NADH + H+ LDH Lactate + H+ + NAD+
B
Glucose PDH Lactate Pyruvate Acetyl-CoA O2, ADP Activated by DCA ATP Fatty acids
G6P, glucose-6-phosphate; DCA, dichloroacetate; LDH, lactate dehydrogenase; PDH, pyruvate dehydrogenase. From Luft FC. Lactic acidosis update for critical care clinicians. Journal of the American Society of Nephrology 2001; 12:S15S19.
reports of a high rate of nuclear DNA (nDNA) mutation (genotoxicity) have been reported in children exposed to AZT in utero [9], underlining that DNA damage may not be localized to the mitochondria.
Familial and drug-related mt disorders

Familial or inherited mt diseases are well-recognized in medicine, particularly in paediatrics, neurology and hepatology [1014]. Mutation in either mtDNA or nDNA may lead to disorders of mt function. Drugs that may impact the mitochondria also tend to lead to adverse effects in the most metabolically active tissues, with the brain (for example, valproate, lead, alcohol), other nerves (aminoglycosides, alcohol), heart (anthracyclines, amiodarone, alcohol) and liver (valproate, aspirin, alcohol) all involved, with lactic acidosis occasionally seen (phenformin, metformin, alcohol). For clinical disease to be present in familial cases, depletion of functional mtDNA to <80% of normal is required [1014]. This is usually assessed by evaluating
M48
In patients with untreated HIV, samples from a range of tissues have been observed to have diminished mtDNA relative to age-matched uninfected individuals [2024]. Muscle and nerve biopsies from untreated individuals presenting with HIV-associated myopathy [22] or distal symmetrical peripheral neuropathy may have low mtDNA, abnormalities in mt ultrastructure or respiratory chain abnormalities [22]. These changes are similar to those reported with NRTI therapy in vitro, animal and human nerves and muscle cells. Declines in mtDNA in adipose tissue of untreated individuals have also been described [20,2225]. Changes in the mtDNA:nDNA ratio in peripheral blood mononuclear cells (PBMCs) (derived from buffy coat) were proportionally more pronounced when comparing HIV-negative controls to HIV-infected individuals (a 44% reduction) relative to the addition of antiretrovirals (a further 24% decline) [20]. Thus, reductions in mtDNA can occur with HIV infection alone and these changes may precede the use of NRTI therapy or symptoms. Indeed, mtDNA in PBMCs may rise following the initiation of effective therapy [24], even when that therapy includes agents known to inhibit DNA polymerase-. Changes in mtDNA content or mitochondria ultrastructure in NRTI-treated individuals, with or without clinical disease, are generally greater than in untreated individuals. These data raise the possibility that HIV may directly (or via cytokines released in response to HIV infection or immune reconstitution) injure mitochondria, potentially making them more vulnerable to the effects of NRTIs. Several HIV gene products, most notably tat and viral protein R (vpr), have been demonstrated to decrease or damage mitochondria and cause clinical disease in vitro or in animal models. For example, expression of TAT may lead to cardiomyopathy with mt destruction in mice [26]. The vpr directly affects the mt permeability transition pore and may trigger cell apoptosis through a mt pathway independent of caspase [27]. The HIV-1 protease may also process procaspase to cause mt release of cytochrome c triggering apoptosis [28]. As muscle, nerve or fat cells may not be infectable by HIV, cytokines may represent the key mediator of
Mitochondria and NRTIs
injury. Tumour necrosis factor (TNF)- triggers apoptosis through targeting mitochondria [29] and may reduce mtDNA number and quality [30]. It has been implicated in liver injury from alcohol [31], particularly contributing to steatohepatitis, a condition reflective of diminished mt beta-oxidation of fatty acids [32,33]. TNF- and interferon (IFN)- also inhibit mt respiration in smooth muscle and other cells [34]. IFN- is a pro-apoptotic cytokine whose effects are mediated through mt cytochrome c release and impaired mtDNA transcription. TNF- and IFN- are elevated in untreated HIV infection [35,36] but may decline with therapy [37] although subsets of CD8 lymphocytes highly productive of TNF- may persist [38]. Several studies have suggested associations between elevations of TNF- or soluble TNF- receptor type II and dyslipidaemia, insulin resistance and/or fat atrophy [3942], a condition associated with some degree of mtDNA reduction in fat cells [23,26]. IFN- has also been linked with lipoatrophy [43]. Thus, mt changes observed in a range of tissues from individuals with HIV infection may represent the consequences of cytokines and HIV gene products produced during HIV infection, and indicate that changes in mt appearance and mtDNA content may to some extent predate in the introduction of NRTI therapy. Therefore, individuals with HIV infection may be at increased risk of mt toxicity once NRTIs are introduced than may be expected from studies in vitro, in animals or in human volunteers.
Effects of NRTIs on mitochondria

The most prevalent theory regarding the effects of NRTIs on mitochondria is that these agents act as competitive inhibitors of DNA polymerase-, the enzyme responsible for copying mtDNA. The d-NRTIs, zalcitabine (ddC), didanosine (ddI) and stavudine (d4T), have the greatest affinity for this enzyme [1]. Inhibition of polymerase- is likely to lead to decline in quantity and, most likely, quality of mtDNA and eventually trigger a cellular energy crisis. With regards to peripheral neuropathy and the doselimiting toxicity of the d-NRTIs, the weight of evidence indicates that this is indeed a mt toxicity with mtDNA decline. Using an in vitro model of nerve cells, ddC and ddI have been shown to reduce mtDNA, leading to destruction of mitochondria and an increase in intracellular lactate levels [44]. Nerve biopsies from patients with NRTI-associated neuropathy have show abnormal mitochondria with enlarged size, excessive vacuolization, electron-dense concentric inclusions and degenerative myelin structures. Abnormal mitochondria represented the majority (55% vs 9% in controls) of nerve cell mitochondria in biopsies from individuals
with ddC-related neuropathy and mtDNA was reduced by as much as 80% compared with the controls [45]. This degree of mtDNA depletion is consistent with loss of functional mtDNA observed in familial mt disorders and hence is consistent with causation. Mitochondrial changes are also found in nerves of untreated individuals presenting with HIV-related peripheral neuropathy [46], hence neuropathy in individuals on therapy may in some cases represent an unmasking or exacerbation of a pre-existent mt neuropathy. However, with myopathy, a probable mt toxicity associated with AZT therapy, the effects on mtDNA mass are less evident. As with neuropathy, specific changes in muscle mitochondria have been observed in clinical biopsy samples and in animal or in vitro models. Effects have been demonstrated in both skeletal and cardiac muscle [47]. Histological features suggested to be distinguishing include ragged-red fibres and abnormal mitochondria with paracrystalline inclusions [48,49], although similar mt abnormalities have also been reported in therapy-naive individuals with HIV-related myopathy [22,50]. Tissue-specificity studies with AZT have demonstrated that this drug has greater impact on muscle cells, compared with kidney or liver cells, and may involve inhibition of succinate transport or cytochrome oxidase activity, reduction in carnitine, oxidative damage to mitochondria and destruction of myotubes rather than reduction in mtDNA [47,5155]. Thus, myopathy (and presumably other toxicity events) with AZT may relate to this agents impact on mt enzyme systems rather than on DNA polymerase-. Other studies have found that AZT may affect a range of mt functions including strongly inhibiting the ADP/ATP antiport in a competitive manner [17], and adenylate kinase [18]. These differences in mechanism of toxicity with AZT relative to d-NRTIs probably relates to the observation that some of its cytological and mt toxicities are mediated through its intermediate metabolite AZTmonophosphate rather than the triphosphate metabolites responsible for the toxicity (and activity) of other NRTIs [56]. This has important implications when assessing ways in which drug impact on mitochondria is measured, as standard measures of mtDNA content are likely to underestimate the effects of AZT on mt function. More recently, a further mechanism by which AZT may cause mt toxicity has been evidenced. In nonmitotic cells, AZT inhibits thymidine phosphorylation by thymidine kinase (TK) type 2, the TK expressed in mitochondria. TK2 is the only constitutively expressed TK in non-mitotic cells. The reduced supply of the natural thymide triphosphate (TTP) leads to limited mtDNA replication and subsequent mtDNA depletion
M49
G Moyle
[57]. As d4T is a much poorer substrate of TK2 than AZT, but a more potent inhibitor of polymerase-, it is unlikely that this mechanism is critical to d4T toxicity [58]. In vitro studies indicate that lactic acid production begins in AZT-treated cells before the decline in mtDNA, whereas declines in mtDNA coincide with lactic acid production increases in d-NRTI-treated cells [59]. In vivo, the rate of mtDNA reductions over time appear somewhat slower and may be less severe in TKdependent cells (such as adipocytes) with AZT than with d4T [25,60,61], an issue that may be linked to the different modes of toxicity.
4.
5.
6.
Summary
Available evidence suggests that a number of important clinical events in individuals with HIV infection are related to mt dysfunction. Several factors may contribute to the development of these events and the tissue(s) in which the event occurs. Some individuals are likely to have important genetic predispositions for mt disease, which may be unmasked by the presence of HIV infection or the introduction of NRTI antiretrovirals. HIV infection per se is associated with reduction in mtDNA content and changes in mt morphology and function, which in some cases leads to clinical events such as myopathy or peripheral neuropathy. NRTI antiretrovirals may impact mtDNA content and function through a number of different mechanisms and have been demonstrated to be causative of a number of clinical toxicities. In in vitro and in clinical studies, newer nucleoside and nucleotides agents such as lamivudine, emtricitabine, abacavir and tenofovir appear to be much weaker inhibitors of mtDNA polymerase- or other mt functions, and appear to be associated with a lower risk of events thought to be related to mt toxicity. Simple, non-invasive tests for mt function are not available at present in the clinical routine, and assays of mtDNA content in blood cells may miss key aspects of mt function, require careful sample handling and may not reflect events occurring in other tissues. There remains a need for the development of rapid, cheap and clinically applicable assays that would enable the prediction of increased likelihood of mt events.
7.
8.
9.
10. 11. 12. 13. 14. 15.
16. 17.
18.
References
1. Martin JL, Brown CE, Matthews-Davis N & Reardon JE. Effects of antiviral nucleoside analogs on human DNA polymerases and mitochondrial DNA synthesis. Antimicrobial Agents & Chemotherapy 1994; 38:27432749. John M, Moore CB, James IR, Nolan D, Upton RP, McKinnon EJ & Mallal SA. Chronic hyperlactatemia in HIV-infected patients taking antiretroviral therapy. AIDS 2001; 15:717723. Brinkman K, Hofsted HJ, Burger DM, Smeitink JAM & Koopmans PP. Adverse effects of reverse transcriptase
19.
20.
2.
21.
3.
inhibitors: mitochondrial toxicity as common pathway. AIDS 1998; 12:17351744. Blanche S, Tardieu M, Rustin P, Slama A, Barret B, Firtion G, Ciraru-Vigneron N, Lacroix C, Rouzioux C, Mandelbrot L, Desguerre I, Rotig A, Mayaux MJ & Delfraissy JF. Persistent mitochondrial dysfunction and perinatal exposure to antiretroviral nucleoside analogues. Lancet 1999; 354:10841089. Delfraissy JF, Blanche S, Tardieu M, Barret B, Rustin P & Mayaux MJ. Mitochondrial and mitochondrial-like symptoms in children born to HIV-infected mothers. Exhaustive evaluation in a large prospective cohort. 8th Conference on Retroviruses & Opportunistic Infections. 14 February 2001, Chicago, IL, USA. Abstract 625B. McIntosh K. Toxicity of perinatally administered zidovudine. 7th Conference on Retroviruses & Opportunistic Infections. 30 January2 February 2000, San Francisco, CA, USA. Abstract S14. Noguera A, Fortuny C, Munoz-Almagro C, Sanchez E, Vilaseca MA, Artuch R, Pou J & Jimenez R. Hyperlactatemia in human immunodeficiency virus-uninfected infants who are exposed to antiretrovirals. Pediatrics 2004; 114:e598e603. Giaquinto C, De Romeo A, Giacomet V, Rampon O, Ruga E, Burlina A, De Rossi A, Sturkenboom M & DElia R. Lactic acid levels in children perinatally treated with antiretroviral agents to prevent HIV transmission. AIDS 2001; 15:10741075. Poirier MC, Olivero OA, Walker DM & Walker VE. Perinatal genotoxicity and carcinogenicity of anti-retroviral nucleoside analog drugs. Toxicology & Applied Pharmacology 2004; 199:151161. Lombes A, Bonilla E & Dimauro S. Mitochondrial encephalomyopathies. Revista de Neurologia 1989; 145:671689. Fromenty B & Pessayre D. Impaired mitochondrial function in microvesicular steatosis. Journal of Hepatology 1997; 26(Suppl 2):4353. Treem WR & Sokol RJ. Disorders of the mitochondria. Seminars in Liver Disease 1998; 18:237253. Taylor RW, Chinnery PF, Clark KM, Lightowlers RN & Turnbull DM. Treatment of mitochondrial disease. Journal of Bioenergetics & Biomembranes 1997; 29:195205. Chinnery PF, Howell N, Andrews RM & Turnbull DM. Clinical mitochondrial genetics. Journal of Medical Genetics 1999; 36:425436. Mhiri C, Baudrimont M, Bonne G, Geny C, Degoul F, Marsac C, Roullet E & Gherardi R. Zidovudine myopathy: a distinctive disorder associated with mitochondrial dysfunction. Annals of Neurology 1991; 29:606614. Modica Napolitano JS. AZT causes tissue-specific inhibition of mitochondrial bioenergetic function. Biochemical & Biophysical Research Communications 1993; 194:170177. Barile M, Valenti D, Hobbs GA, Abruzzese MF, Keilbaugh SA, Passarella S, Quagliariello E & Simpson MV. Mechanisms of toxicity of 3-azido-3-deoxythymidine: its interaction with adenylate kinase. Biochemical Pharmacology 1994; 48:14051412. Valenti D, Barile M & Passarella S. AZT inhibition of the ADP/ATP antiport in isolated rat heart mitochondria. International Journal of Molecular Medicine 2000; 6:9396. Pan Zhou XR, Cui L, Zhou XJ, Sommadossi JP & Darley Usmar VM. Differential effects of antiretroviral nucleoside analogs on mitochondrial function in HepG2 cells. Antimicrobial Agents & Chemotherapy 2000; 44:496503. Ct HC, Brumme ZL, Craib KJ, Alexander CS, Wynhoven B, Ting L, Wong H, Harris M, Harrigan PR, OShaughnessy MV & Montaner JS. Changes in mitochondrial DNA as a marker of nucleoside toxicity in HIV-infected patients. New England Journal of Medicine 2002; 346:811820. McComsey G, Tan DJ, Lederman M, Wilson E & Wong LJ. Analysis of mitochondrial DNA genome in the peripheral blood leukocytes of HIV-infected patients with of without lipoatrophy. AIDS 2002; 16:513518.
M50
Mitochondria and NRTIs
22. Morgello S, Wolfe D, Gadfrey E, Feinstein R, Tagliati M & Simpson DM. Mitochondrial abnormalities in human immunodeficiency virus-associated myopathy. Acta Neuropathologica 1995; 90:366374. 23. Shikuma CM, Hu N, Yost F, Waslien C, Shimizu S & Shiramizu B. Mitochondrial DNA decrease in subcutaneous adipose tissue of HIV-infected individuals with peripheral lipoatrophy. AIDS 2001; 15:18011809. 24. Gallant J, Staszewski S, Pozniak A, Lu B, Miller MD, Coakley DF & Cheng A. Favorable lipid and mitochondrial (mt) DNA profile for tenofovir disoproxil fumarate (TDF) compared to stavudine (d4T) in combination with lamivudine (3TC) and efavirenz (EFV) in antiretroviral therapy (ART) nave patients: a 48 week interim analysis. 42nd Interscience Conference on Antimicrobial Agents & Chemotherapy. 2730 September 2002, San Diego, CA, USA. Abstract LB-2. 25. Nolan D, Hammond E, James I, McKinnon E & Mallal S. Contribution of nucleoside-analogue reverse transcriptase inhibitor therapy to lipoatrophy from the population to the cellular level. Antiviral Therapy 2003; 8:617626. 26. Raidel SM, Haase C, Jansen NR, Russ RB, Sutliff RL, Velsor LW, Day BJ, Hoit BD, Samarel AM & Lewis W. Targeted myocardial transgenic expression of HIV TAT causes cardiomyopathy and mitochondrial damage. American Journal of Physiology. Heart & Circulatory Physiology 2002; 282:H1672H1678. 27. Jacotot E, Ravagnan L, Loeffler M, Ferri KF, Vieira HL, Zamzami N, Costantini P, Druillennec S, Hoebeke J, Briand JP, Irinopoulou T, Daugas E, Susin SA, Cointe D, Xie ZH, Reed JC, Roques BP & Kroemer G. The HIV-1 viral protein R induces apoptosis via a direct effect on the mitochondrial permeability transition pore. Journal of Experimental Medicine 2000; 19:3346. 28. Roumier T, Vieira HL, Castedo M, Ferri KF, Boya P, Andreau K, Druillennec S, Joza N, Penninger JM, Roques B & Kroemer G. The C-terminal moiety of HIV-1 Vpr induces cell death via a caspase-independent mitochondrial pathway. Cell Death & Differentiation 2002; 9:12121219. 28. Nie Z, Phenix BN, Lum JJ, Alam A, Lynch DH, Beckett B, Krammer PH, Sekaly RP & Badley AD. HIV-1 protease processes procaspase 8 to cause mitochondrial release of cytochrome C, caspase cleavage and nuclear fragmentation. Cell Death & Differentiation 2002; 9:11721184. 29. Polla BS, Kantengwa S, Francois D, Salvioli S, Franceschi C, Marsac C & Cossarizza A. Mitochondria are selective targets for the protective effects of heat shock against oxidative injury. Proceedings of the National Academy of Sciences, USA 1996; 93:64586463. 30. Li YY, Chen D, Watkins SC & Feldman AM. Mitochondrial abnormalities in tumor necrosis factor-alpha-induced heart failure are associated with impaired DNA repair activity. Circulation 2001; 104:24922497. 31. Neuman MG, Shear NH, Bellentani S & Tiribelli C. Role of cytokines in ethanol-induced cytotoxicity in vitro in HepG2 cells. Gastroenterology 1998; 115:157166. 32. Pessayre D, Mansouri A, Haouzi D & Fromenty B. Hepatotoxicity due to mitochondrial dysfunction. Cell Biology & Toxicology 1999; 15:367373. 33. Chitturi S & Farrell GC. Etiopathogenesis of nonalcoholic steatohepatitis. Seminars in Liver Disease 2001; 21:2741. 34. Geng Y, Hansson GK & Holme E. Interferon-gamma and tumor necrosis factor synergize to induce nitric oxide production and inhibit mitochondrial respiration in vascular smooth muscle cells. Circulation Research 1992; 71:12681276. 35. Cossarizza A, Mussini C, Mongiardo N, Borghi V, Sabbatini A, De Rienzo B & Franceschi C. Mitochondria alterations and dramatic tendency to apoptosis in peripheral blood lymphocytes during acute HIV syndrome. AIDS 1997; 11:1926. 36. Cossarizza A, Ortolani C, Mussini C, Borghi V, Guaraldi G, Mongiardo N, Bellesia E, Franceschini MG, De Rienzo B & Franceschi C. Massive activation of immune cells with an
intact T cell repertoire in acute HIV syndrome. Journal of Infectious Diseases 1995; 172:105112. 37. Franco JM, Rubio A, Rey C, Leal M, Macias J, Pineda JA, Sanchez B, Sanchez-Quijano A, Nunez-Roldan A & Lissen E. Reduction of immune system activation in HIV-1-infected patients undergoing highly active antiretroviral therapy. European Journal of Clinical Microbiology & Infectious Diseases 1999; 18:733736. 38. Ledru E, Christeff N, Patey O, de Truchis P, Melchior JC & Gougeon ML. Alteration of tumor necrosis factor-alpha Tcell homeostasis following potent antiretroviral therapy: contribution to the development of human immunodeficiency virus-associated lipodystrophy syndrome. Blood 2000; 95:31913198. 39. Mynarcik DC, McNurlan MA, Steigbigel RT, Fuhrer J & Gelato MC. Association of severe insulin resistance with both loss of limb fat and elevated serum tumor necrosis factor receptor levels in HIV lipodystrophy. Journal of AIDS 2000; 25:312321. 40. Domingo P, Veloso S, Montero F, Sirvent J, Matias-Guiu X, Vialdes J, Gutierrez C, Sambeat MA, Aguilar C, Peraire J, Richart C & Richart C. Serum levels of soluble tumor necrosis factor (TNF) receptors correlate with the degree of subcutaneous adipocyte apoptosis in HIV-infected patients with lipodystrophy. 14th AIDS International Conference. 712 July 2002, Barcelona, Spain. Abstract WePeA5792. 41. Johnson JA, Albu JB, Engelson ES, Fried SK, Inada Y, Ionescu G & Kotler DP. Increased systemic and adipose tissue cytokines in patients with HIV-associated lipodystrophy. American Journal of Physiology, Endocrinology & Metabolism 2004; 286:E261E271. 42. Lagathu C, Bastard JP, Auclair M, Maachi M, Kornprobst M, Capeau J & Caron M. Antiretroviral drugs with adverse effects on adipocyte lipid metabolism and survival alter the expression and secretion of proinflammatory cytokines and adiponectin in vitro. Antiviral Therapy 2004; 9:911920. 43. Christeff N, Melchior JC, de Truchis P, Perronne C & Gougeon ML. Increased serum interferon alpha in HIV-1 associated lipodystrophy syndrome. European Journal of Clinical Investigation 2002; 32:4350. 44. Keilbaugh SA, Prusoff WH & Simpson VMV. The PC12 cell as a model for studies of mechanism of induction of peripheral neuropathy by anti-HIV dideoxynucleoside analogs. Biochemical Pharmacology 1991; 42:R5R8. 45. Dalakas MC, Semino-Mora C & Leon-Monzon M. Mitochondrial alterations with mitochondrial DNA depletion in the nerves of AIDS patients with peripheral neuropathy induced by 23-dideoxycytidine (ddC). Laboratory Investigation 2001; 81:15371544. 46. Simpson DM & Tagliati M. Nucleoside analogue-associated peripheral neuropathy in human immunodeficiency virus infection. Journal of Acquired Immune Deficiency Syndromes & Human Retrovirology 1995; 9:153161. 47. McCurdy DTr & Kennedy JM. AZT decreases rat myocardial cytochrome oxidase activity and increase beta-myosin heavy chain content. Journal of Molecular & Cellular Cardiology 1998; 30:19791989. 48. Dalakas MC, Illa I, Pezeshkpour GH, Laukaitis JP, Cohen B & Griffin JL. Mitochondrial myopathy caused by long-term zidovudine therapy. New England Journal of Medicine 1990; 322:10981105. 49. Pezeshkpour G, Illa I & Dalakas MC. Ultrastructural characteristics and DNA immunocytochemistry in human immunodeficiency virus and zidovudine-associated myopathies. Human Pathology 1991; 22:12811288. 50. Simpson DM, Citak KA, Godfrey E, Godbold J & Wolfe DE. Myopathies associated with human immunodeficiency virus and zidovudine: can their effects be distinguished? Neurology 1993; 43:971976. 51. Dalakas MC, Leon-Mozon ME, Bernardini I, Gahl WA & Jay CA. Zidovudine-induced mitochondrial myopathy is associated with muscle carnitine deficiency and lipid storage. Annals of Neurology 1994; 35:482487.
M51
G Moyle
52. Semino-Mora MC, Leon-Monzon ME & Dalakas MC. Effect of L-carnitine on the zidovudine-induced destruction of human myotubes. Part I: L-carnitine prevents the myotoxicity of zidovudine in vitro. Laboratory Investigation 1994; 71:102112. 53. Semino-Mora MC, Leon-Monzon ME & Dalakas MC. Effect of L-carnitine on the zidovudine-induced destruction of human myotubes. Part II: treatment with L-carnitine improves AZT-induced changes and prevents further destruction. Laboratory Investigation 1994; 71:773781. 54. Pereira LF, Oliveira MB & Carnieri EG. Mitochondrial sensitivity to zidovudine. Cell Biochemistry & Function 1998; 16:173181. 55. de la Asuncion JG, del Olmo ML, Sastre J, Millan A, Pellin A, Pallardo FV & Vina J. AZT treatment induces molecular and ultrastructural oxidative damage to muscle mitochondria. Prevention by antioxidant vitamins. Journal of Clinical Investigation 1998; 102:49. 56. Sales SD, Hoggard PG, Sunderland D, Khoo S, Hart CA & Back DJ. Zidovudine phosphorylation and mitochondrial
toxicity in vitro. Toxicology & Applied Pharmacology 2001; 177:5458. 57. McKee EE, Lynx MD, Susan-Resiga D, Bentley AT, DHaenens JP, Cullen D & Ferguson M. Zidovudine inhibits thymidine phosphorylation: a novel site of potential toxicity in non-mitotic cells. Antiviral Therapy 2004; 9:L9. 58. Rylova SN, Albertioni F, Flygh G & Eriksson S. Activity profiles of deoxynucleoside kinases and 5-nucleotidases in cultured adipocytes and myoblastic cells: insights into mitochondrial toxicity of nucleoside analogs. Biochemical Pharmacology 2005; 69:951960. 59. Setzer B, Schlesier M, Thomas AK & Walker UA. Mitochondrial toxicity of nucleoside analogues in primary human lymphocytes. Antiviral Therapy 2005; 10:327334. 60. van der Valk M, Casula M, Weverlingz GJ, van Kuijk K, van Eck-Smit B, Hulsebosch HJ, Nieuwkerk P, van Eeden A, Brinkman K, Lange J, de Ronde A & Reiss P. Prevalence of lipoatrophy and mitochondrial DNA content of blood and subcutaneous fat in HIV-1-infected patients randomly allocated to zidovudine- or stavudine-based therapy. Antiviral Therapy 2004; 9:385393.
Received 7 January 2005, accepted 23 May 2005
M52
Nucleoside analogues toxicities related to mitochondrial dysfunction: focus on HIV-infected children

Alessandra Vigan* and Vania Giacomet
Paediatrics, L Sacco Hospital, University of Milan, Italy *Corresponding author: Tel: +39 02 390 42265; Fax: +39 02 839 2229; E-mail: alessandra.vigano@unimi.it
Introduction
Survival in HIV-infected children has greatly improved with the introduction of highly active antiretroviral therapy (HAART) [1]. Concomitantly, morbidity from the long-term effects of antiretroviral therapy (ART) has grown in importance. Children are more vulnerable than adults to the metabolic side effects of therapy because of the potential impact on growth and the likelihood of greater cumulative exposure. A number of important clinical events have been described in HIVinfected adults receiving antiretroviral nucleoside analogues that have been linked to mitochondrial (mt) injury and dysfunction. This review discusses these complications individually in an attempt to summarize the current understanding of their pathogenesis, with a focus on HIV-infected children. from 1.33.9 cases per 1000 person-years, a mortality rate greater than 50% and is highly correlated with NRTI use [2,14]. In some cases, hyperlactataemia (<5.0 mmol/l) is associated with clinical symptoms (nausea, emesis and vague abdominal pain), mild liver abnormalities without hepatic failure and is not complicated by metabolic acidosis [26]. Symptomatic hyperlactataemia is rare in HIV-infected adults (814.5 cases per 1000 person-years) and is highly correlated with NRTI use, showing a mild clinical course usually with a good prognosis, provided that NRTIs are promptly discontinued [2,12,15]. Most (>85%) of the 818.3% of HIV-infected adults who test positive for hyperlactataemia are symptom-free [26]. Subclinical hyperlactataemia is characterized by lactate levels ranging from 2.15.0 mmol/l, is usually transient and has a poor positive predictive value for identifying cases of future symptomatic lactic acidosis/hepatic steatosis [35]. Moreover, in the largest published prospective cohort study, the detection of subclinical hyperlactataemia in 2% of ART-naive patients indicated that this syndrome might be relatively non-specific for current NRTI use [5]. Finally, a recent prospective study showed that chronic subclinical mild hyperlactataemia with a lactate concentration of 1.53.5 mmol/l occurs in a significant proportion of HIV-infected adults started on NRTI-containing regimens. Initiation of HAART containing either zidovudine (AZT) or stavudine (d4T) was associated with a mild increase in group mean lactate levels during the first 68 months of therapy and with a new serum lactate steady state after 12 months of therapy [3]. The cause(s) of subclinical hyperlactataemia remain(s) unclear. Given its transitory pattern, however, sampling variability might play a major role. In addition, the superimposition of sampling variability on an already increased serum lactate setpoint may also account for the higher
M53
Hyperlactataemia syndromes
Use of nucleoside analogue reverse transcriptase inhibitors (NRTIs) by HIV-infected adult patients may be complicated by hyperlactataemia with a frequency ranging from 818.3% in cohort studies [26]. The precise biochemical pathways involved in NRTIrelated hyperlactataemia (NRH) are not yet completely understood or defined. Nevertheless, a leading theory based on several in vitro data is that NRH derives from direct inhibition of mtDNA polymerase- by NRTIs and, consequently, mt dysfunction [710]. Three different hyperlactataemia syndromes have been described in HIV-infected adults: lactic acidosis syndrome (LAS), symptomatic hyperlactataemia and subclinical hyperlactataemia. LAS is a rare, life-threatening disease characterized by severe lactic acidosis (lactate levels usually >5.0 mmol/l), hepatic steatosis, occasional liver failure and sometimes by other NRTI-related toxicities believed to have a mt genesis, such as pancreatitis, sensory neuropathy and skeletal myopathy [1014]. In HIV adult cohorts, LAS shows an incidence ranging
A Vigan & V Giacomet
frequency of this syndrome in patients on HAART that contains NRTIs as compared with untreated controls. Lactic acidosis is an important cause of metabolic acidosis in children and may be divided into two categories: type A, which is associated with tissue hypoperfusion and hypoxaemia secondary to cardiopulmonary arrest, septic, cardiac or hypovolaemic shock and type B, which occurs without any clinical evidence of tissue hypoxia but is due to impairment of mt function caused by drugs, toxins and inherited metabolic disease. The first group is relatively common in the paediatric intensive care unit; the second is uncommon [16]. During 19972003, three cases of type B lactic acidosis were identified in HIV-infected children treated with antiretroviral drugs. The first paediatric case of lactic acidosis and hepatic steatosis was described by Miller et al. [17]. The author reported on four patients, one of whom was an HIV-infected adolescent who developed severe hepatic steatosis and lactic acidosis while receiving an antiretroviral regimen containing d4T. The paediatric case was a 16-year-old HIV-infected girl with a 3-day history of nausea, vomiting and abdominal pain. She had severe metabolic acidosis (arterial blood pH 7.33), an elevated serum lactate level (9.9 mmol/l), hepatic steatosis [liver density, at computed tomography (CT), consistent with diffuse fatty infiltration], elevated liver enzymes [alanine aminotransferase (ALT), 120 U/l; aspartate aminotransferase (AST), 166 U/l], pancreatitis (pancreatic imaging, at CT, showing inflammation and necrosis) with increased levels of amylase (561 U/l) and lipase (6150 U/l), and myopathy (increased fat droplet in myocytes, cytochrome oxidase-negative fibres, degenerating fibres at light and electron microscopy assessment of quadriceps biopsy specimen). Therapy with d4T, didanosine (ddI) and nelfinavir (NFV) (which the patient had been taking for 3 months) was discontinued. After a complicated hospital course that included a prolonged stay in the intensive care unit, the girl started a new antiretroviral regimen with AZT, nevirapine and NFV and her illness did not recur. Subsequently, Church et al. described an HIVinfected child treated with AZT, ddI and NFV who developed near-fatal lactic acidosis and liver failure with mtDNA depletion [18]. At 23 months old the child showed generalized developmental regression (she became unable to stand and physical examination revealed generalized areflexia and marked doughy texture of the musculature), signs of liver dysfunction, lactic acidosis (plasma lactic acid from 3.667.11 mmol/l and serum CO2 of 10 mEq/l), and elevation in Krebs cycle intermediate products on urine acid measurements. Examination of muscle biopsy revealed severe generalized atrophy and variation in fibre size, loss of myofibril and poorly defined mitochondria with
M54
disorganized cristae and dense irregular granules. In addition, spectrophotometric analysis of muscles mt enzymes showed 6289% reductions relative to laboratory normal values in the activities of respiratory chain enzymes, and quantitative Southern blot analysis of muscle mtDNA demonstrated a depletion of 79% as compared with the average mtDNA content of agematched controls. Coenzyme Q, riboflavin and levocarnitine were given as empirical therapy for mt failure and ART was transiently discontinued and subsequently adjusted to a regimen free of nucleoside analogues. Neurological, hepatic and metabolic abnormalities improved after nucleoside analogues were discontinued and progressively resolved during the follow-up. In addition, a repeat muscle biopsy 6 months after discontinuation of nucleoside analogues demonstrated structural, enzymatic and mtDNA normalization. Recently, Rey et al. described the first paediatric case of fatal lactic acidosis [19]. The child was a 5-year-old HIV-infected girl receiving ritonavir, d4T and ddI with a 10-day history of nausea and vomiting and a 12-hour history of abdominal pain with severe tachypnea and hypocapnia. Initial laboratory studies revealed a pH of 7.27, bicarbonate of 2.8 mmol/l and a lactate level of 13.6 mmol/l. Signs of liver dysfunction included elevated bilirubin levels, very decreased fibrinogen levels (37 mg/dl) with prolonged prothrombin time (21.5 sec) and partial thromboplastin time (40 sec), mild elevation of AST and ALT (57 and 67 U/l, respectively), and elevated lipase (3222 U/l) and amylase levels (805 U/l). She was admitted to the paediatric intensive care unit, where ART was discontinued and high-dose intravenous bicarbonate, haemodiafiltration and empirical therapy for mt failure with essential cofactors (L-carnitine, riboflavin and thiamine) was started. Serum lactate level initially decreased from 21.2 to 12.4 mmol/l and serum bicarbonate rose from 9.5 to 19.5 mmol/l; however, soon after this, lactate increased and bicarbonate decreased again and this tendency persisted until the death of the child 36 h after hospital admission. The comparison between paediatric and adult cases of LAS seems to identify several overlapping features concerning incidence, risk factors and clinical presentation and, unique to paediatric cases, the more frequent concomitant presence of other signs of NRTI-related toxicities. Thus, the above-mentioned paediatric cases of LAS share several features with those described in adult patients: the duration of exposure to NRTIs was in the expected range of 320 months, the most common complaints were gastrointestinal (except in the infant who could not be expected to be able to refer these symptoms), signs of liver dysfunction were present in
NRTI toxicity and mitochondrial dysfunction in children
all cases and lactate levels were always >10 mmol/l in the fatal cases [20,26,1114]. Unlike adult cases of LAS, the concomitant presence of other NRTI-related toxicities (pancreatitis and skeletal myopathy) seems to be more frequent in the paediatric spectrum of the disease. Use of d4T and ddI has been identified as a risk factor in adult NRH, and the combination of d4T and ddI may confer the highest risk [3,4,1113,2123]. It is noteworthy that all paediatric LAS cases included one of these two drugs in the antiretroviral regimen and that in the fatal case both drugs were present. The anecdotal nature of the paediatric observations does not allow correct estimation of the incidence of LAS in children taking NRTIs, however the report of only three cases in a time period of 5 years suggests quite a low incidence. The features of subclinical hyperlactataemia have been assessed in only two paediatric studies. Noguera et al. conducted a prospective observational study of venous lactate concentrations during a 28-month period in 80 HIV-infected children, mostly receiving antiretroviral drugs [24]. Fasting venous blood samples were obtained after the patient had rested for 15 min, avoiding tourniquets when possible. Pathological hyperlactataemia was considered only if lactate and alanine levels were concomitantly higher than the reference range (1.77 mmol/l and 390 mol/l, respectively). During follow-up, none of the children developed clinical signs or symptoms consistent with lactic acidosis and no revisions in ART were made for this reason. Overall, 23/80 children (29%) showed elevated lactate values in one or more of the determinations throughout the study. Of them, only 14 patients were identified with concomitant elevated alanine concentration, giving a prevalence of hyperlactataemia of 17%. Peak lactate plasma concentrations ranged from 2.054.9 mmol/l and in 5/14 (45%) cases, lactate values spontaneously returned to normal values. The size of this cohort did not allow the establishment of a possible relationship between hyperlactataemia and type of NRTI used, however it is remarkable that all children with hyperlactataemia were receiving at least one NRTI and that d4T was used in 11/14 of these patients. In addition, the study identified younger age at the beginning of ART as a significant risk factor for hyperlactataemia. Desay et al. reported 251 lactate levels obtained from 127 children with HIV/AIDS; 104 on HAART and 23 not on HAART [25]. Of the 104 HAARTtreated children, 20 (19%) were on non-NRTIs (NNRTIs), 102 were on one or more NRTI (98%) and 95 (91%) were on one or more protease inhibitor (PI). Lactate levels showed a mean value of 1.70 mmol/l (range 0.64.4 mmol/l). Of the 127 children, 41 (32.3%) had at least one value higher than 2 mmol/l. Of the total 251 lactate readings, 56 (25.6%) were
above 2 mmol/l. Only one patient was symptomatic, with abdominal pain, nausea and vomiting, but these symptoms disappeared after a temporary discontinuation of ART. Elevated lactate levels were associated with therapy with NRTIs (P=0.01) or PIs (P=0.04) when considered as classes. Further studies are obviously needed to fully evaluate the incidence of subclinical hyperlactataemia and to assess the exact contributory role of factors other than NRTIs in this syndrome in HIV-infected children. However, both the published studies deserve some comment. Noguera et al. [24] showed a 17% prevalence of subclinical hyperlactataemia; however, if the nine cases with transient hyperlactataemia are excluded, the frequency would further decrease to 11%. In addition, younger age was a significant risk factor for hyperlactataemia. This observation needs further confirmation, namely, it may arise either from a greater susceptibility of younger children to NRTI toxicity or from more frequent difficulties in blood sampling in infants and toddlers compared with older children. Desay et al. [25] claim a 32% prevalence of subclinical hyperlactataemia; however, no detailed information on the blood sampling method is given or whether hyperlactataemia was also detected in the 23 untreated children included in the study. The study also raises the novel possibility that therapy with PIs may contribute to elevated lactate levels. However, since almost all patients (102/104) included in the analysis were receiving at least one NRTI, it is unclear if the observation came from a true evidence or from a bias in the statistical analysis.
Cardiomyopathy
Several patterns of cardiovascular involvement have been reported in HIV-infected children [2632]. A continuum from asymptomatic left ventricular (LV) dysfunction to dilated cardiomyopathy to congestive heart failure (CHF) to hypotensive pump failure with cardiac associated mortality has been suggested [33]. Abnormalities of LV hypertrophy have also been suggested in which LV mass is excessive for body surface area but insufficient for LV dimension, resulting in a sustained elevation of LV peak wall stress, a mediator of mechanically induced hypertrophy [28,34]. Other reported cardiac problems include haemodynamic abnormalities, conduction abnormalities, dysrhythmias and sudden death, as well as pericardial and vascular involvement [2632]. Dilated cardiomyopathy appears to be more common in HIV-infected children than in seroreverted children and increases in frequency as HIV-infected children progress to AIDS [35]. CHF appears to occur chronically in some 10% of HIV-infected children and transiently in another 10%
M55
[28]. Cardiomyopathy appears to reduce survival in HIV-infected children [27,29,30,36]. One study showed a relative risk of death of 2.76 in children with cardiomyopathy compared with children without cardiomyopathy [27]. Children were more likely to be short-term survivors (<5 years) if cardiomyopathy was present [31]. Risk factors for more advanced cardiac involvement include HIV encephalopathy and a low CD4 cell count [28,37,38]. The causes of these abnormalities are likely to be multifactorial. Possible causes include HIV itself, other viruses, immune-mediated cardiac muscle dysfunction, malnutrition and cardiotoxic drugs. A possible role of AZT use has been questioned on the basis of some studies done on animal models. Distinctive histopathological changes in mitochondria of cardiac muscle have been noted in rats receiving high doses of AZT and they were speculated to be due to drug-induced inhibition of mtDNA replication [39]. In the Erytrocebus patas monkey model of in utero exposure, investigators demonstrated in full-term foetuses that transplacental exposure to AZT, at around 86% of the human daily dose for the second half of gestation, resulted in structural alterations in mitochondria of skeletal muscle (including disrupted cristae) and mtDNA depletion in cardiac and skeletal muscle [40]. Studies in transgenic adult mice suggest that AZT is associated with diffuse destruction of cardiac mt ultrastructure and inhibition of mtDNA replication [41]. An association between AZT and dilated cardiomyopathy has been noted in some HIV-infected adult patients; however, other studies showed opposite results [42]. Herskowitz et al. retrospectively studied six patients in whom congestive heart failure developed, which was temporally associated with ART; three patients received AZT, two received ddI and one received 3 weeks of ddI therapy after 7 months of AZT therapy. However, cardiac changes were not confirmed by heart biopsy [43]. Cardoso et al. did not show effects of AZT on LV end-diastolic dimension, mass or fractional shortening [44]. The first paediatric case of dilated cardiomyopathy was reported in 1986 in one vertically HIV-infected infant who had never received ART [45]. Symptomatic dilated cardiomyopathy developed at age 14 months, with cardiomegaly on a chest roentgenogram, an LV fractional shortening of 17% and clinical congestive heart failure. The infant showed a progressive course of cardiac disease, which necessitated increasingly intensive medical support until his death almost 18 months later. Similar cases were reported by others a year later, also without the involvement of antiretroviral drugs, in which the progressive cardiomyopathy was attributed to the longer survival providing more time for cardiac lesion of multifactorial etiopathogenesis to manifest [46].
M56
The relationship between AZT use and the development of cardiomyopathy has been carefully examined by Lipshultz et al. [47]. Serial echocardiograms were performed in 24 children with symptomatic HIV infection before they started AZT and a mean of 1.32 years since therapy began, 27 age-matched children with symptomatic HIV infection not treated with antiretrovirals and 191 healthy controls (HCs). As compared with HCs, children treated with AZT had progressive LV dilatation and peak wall stress; dilatation and stress were significantly elevated both before and during AZT treatment. The ratio of ventricular thickness to internal dimension was below normal before ART began. After AZT treatment, overall LV mass as well as peak wall stress increased while LV fractional shortening decreased. No significant differences were detected at follow-up in any of these measurements between HIVinfected children treated with AZT and those not treated. These data show that a progressive LV dilatation with compensatory hypertrophy inadequate to maintain normal peak systolic wall stress occurred in children with symptomatic HIV infection. AZT treatment did not appear to worsen or ameliorate these cardiac changes. In a P2C2 HIV Study Group trial, 196 children with severe and mild symptomatic HIV infection and with a median age of 2.1 years were followed with a longitudinal 2-year echocardiographic assessment. The results confirmed that subclinical cardiovascular abnormalities were common in HIV-infected children and most remained throughout follow-up, with some progression. Unfortunately, the study did not perform a separate analysis for the 124 AZT-treated and the 72 untreated children, and so the possible role of this drug cannot be inferred [48]. In another P2C2 HIV Study Group trial, infants born to HIV-infected mothers were followed from birth to the first 14 months of age [49]. Data on 382 infants without HIV infection (all exposed to AZT perinatally and 36 postnatally, for a median of 42 days) and 58 HIV-infected children (all exposed to AZT perinatally and 12 postnatally during the first 12 months of age) were analysed. No association with acute or chronic abnormalities in LV structure or function was found with perinatal and postnatal exposure to this antiretroviral drug. Domanski et al. retrospectively reviewed echocardiograms and clinical records (from January 1987December 1992) on 137 HIV-infected children (13 of whom had never received ART) with symptomatic disease in the large majority. At the date of the first echocardiogram, the antiretrovirals administered included AZT (52 patients), ddI (13 patients), both drugs (one patient) and no treatment (71 patients). Many of the 71 untreated patients were receiving their echocardiograms as a part of pre-treatment evaluation.
Children treated with AZT had a lower average in LV fractional shortening than those untreated with AZT. The odds that a cardiomyopathy would develop were 8.4 times greater in children who had previously used AZT than in those who had never taken this drug. ddI was not associated with the development of a cardiomyopathy [50]. Overall, the great majority of the studies performed in the pre-HAART era suggest that cardiomyopathy is quite common in HIV-infected children with symptomatic disease and that it is not associated with AZT use. The conflicting data obtained by Domanski et al. need to be interpreted with caution. The authors have mainly compared serial echocardiographic measurements from children with symptoms of HIV disease receiving AZT with those from children receiving ddI. The echocardiograms on 58/71 children, who were supposed to represent a group who had received neither therapy, were actually baseline measurements before the start of ART with no follow-up studies without therapy. Only 13 never-treated patients were included to represent the effect of HIV alone and these were specified as having no symptoms. Finally a recent case report showed benefits of a HAART regimen including AZT in a child with severe cardiomyopathy [51]. A dilated cardiomyopathy was diagnosed between 912 months and it severely deteriorated in the ensuing 2 years. When the child was 3.5 years old, AZT, lamivudine (3TC) and ritonavir therapy was started. Six months after therapy, CD4 count increased, viral load became undetectable and LV shortening fraction increased from 12 to 25%. One year later, the heart size was still slightly large on chest X-ray, but echocardiography revealed normal cardiac function with an LV shortening fraction of 33%.
Distal symmetric polyneuropathy

Peripheral nerve disorders are frequent complications of HIV disease and distal symmetric polyneuropathy (DSP) is the most common form of neuropathy with reported estimates of 1550% in HIV-infected adult populations [52]. While DSP is relatively uncommon early in the course of HIV disease, its incidence increases as the degree of immunosuppression progresses [53]. The major presenting complaints of DSP are paraesthesia, numbness and burning sensations in the feet, usually in a symmetrical pattern. Unexpected peripheral neuropathies have been described in Phase I/II dose-finding studies with zalcitabine (ddC), ddI and d4T. The percentage of patients who developed DSP during ddC therapy ranged from 2566% [5456]. The occurrence of ddCassociated neuropathy as well as the severity and progression of symptoms were clearly dose-dependent.
The incidence of DSP in patients treated with ddI varied from 1234% of the subjects [57]. The onset and progression of neurological toxicity were both related to the daily and cumulative doses of this drug [58]. The occurrence of DSP has been described in 6, 15 and 31% of patients receiving d4T at a daily dose of 0.5, 1.0 and 2.0 mg/kg, respectively (Bristol-Myers Squibb, data on file). The temporal relationship of onset and resolution of DSP with drug intake and discontinuation clearly indicates a toxic effect of NRTIs. There are several theories for the mechanism of this neurotoxicity but most evidence focuses on mt toxicity. Using an in vitro model of nerve cells, ddC and ddI have been shown to reduce mtDNA and cause mt damage [59]. Nerve biopsies from patients with ddC-related DSP, as compared with nerve biopsies from patients with AIDSrelated neuropathy never treated with ddC, have shown a majority of abnormal mitochondria (55 vs 9%) and mtDNA depletion as high as 80% [60]. However, mt changes were also observed in nerves of untreated individuals presenting with DSP [61]. In addition, depleted levels of acetyl-L-carnitine have been noted in patients with DSP receiving ddC, ddI or d4T therapy [62]. This depletion may impair peripheral nerve regeneration and disrupt mt metabolism providing a further mechanism for NRTI neurotoxicity [63]. Children with HIV infection develop a variety of neurological complications, but peripheral nervous system involvement is relatively infrequent compared with the frequent occurrence of peripheral neuropathy described in HIV-infected adults [64]. Neuropathy has never been observed or has been rarely reported in paediatric populations treated with ddC [65]. In 1991, a case of a 5-year-old boy with AIDS who developed an inflammatory demyelinating polyneuropathy was reported by Raphael et al. [66]. In 1997, Floeter et al. described the electrophysiological data of 50 HIV-infected children under 18 years of age referred for evaluation of suspected neuropathy to the EMG laboratory at the National Institutes of Health during 19891995 [67]. Most children had moderate to severe manifestations of HIV infection, with 31/50 classified as C3 the clinical and immunologically most severe category of the disease. All but three of the children had been treated with one or more NRTIs (AZT, ddI, ddC, 3TC) before the development of symptoms of neuropathy and before referral for nerve conduction studies (NCSs). Overall, 12 children had abnormal NCSs and a few common patterns of abnormalities were detected: DSP in seven cases, median mononeuropathy in three cases and unusual patterns in two cases (central and peripheral nervous system dysfunction in one and subacute lumbosacral polyradiculopathy following a varicella zoster infection
M57
in the other one). All children with DSP had fairly advanced HIV infection (five in CDC class C3 and two in CDC class B2) and had received ART prior to experiencing symptoms suggestive of neuropathy. Interestingly, a stepwise logistical regression of risk factors (age, % CD4 count and number of NRTIs received) for the development of neuropathy found that age alone was significantly associated with DSP. Children with DSP had a mean age of 14.4 years compared with a mean age of 8.2 years for children without DSP. Thus DSP in children, as in adults, seems to be the most common pattern of peripheral neuropathy and it occurs in subjects with advanced HIV infection. The role of NRTIs, particularly ddI and ddC, in causing distal neuropathy could not be clarified by this study. In this series, children with more severe HIV infection had received more antiretroviral drugs overall, and all children with DSP received more than one antiretroviral drug prior to developing symptoms. Although these data cannot truly untangle the role of NRTIs in the development of neuropathy in these children, most children who received NRTIs did not develop peripheral neuropathy. In 2000, a cross-sectional study on 39 children (age range 514 years) was conducted by means of a structured questionnaire (regarding symptoms of pain, paraesthesia or weakness) and a physical examination [68]. Thirteen (34%) cases had symptoms and signs of DSP. Distal paraesthesia and/or pain plus diminished ankle jerks and/or diminished vibration sense were the most common clinical findings. Symptoms were chronic and fluctuating, but pain was, in general, not severe. In eight of the 13 patients, the use of potential neurotoxic drugs was absent before the beginning of DSP, while in the remaining five patients, clinical features developed after having been on ddI therapy. The paucity of reports on DSP in HIV-infected children does not allow a definitive assessment of its incidence and correlation with the use of neurotoxic drugs in this population. However, the prevalence reported in the more recent study [68] is similar to that reported in HIV-infected adults. The lack of peripheral neuropathy in studies conducted at the beginning of the epidemic could be explained by the early mortality of children at that time. Dying before acquiring adequate language skills would preclude symptoms such as paraesthesia or pain from being confirmed. In addition, symptoms in children seem less severe and they could not be referred spontaneously to the clinicians. An interesting question, which deserves further research, is the role of age in the development of DSP. In animals, age alone is a risk factor diminishing regenerative capacity of the growing peripheral nervous system [69]. In humans, peripheral nerves normally exhibit very rapid growth shortly after birth followed by slower growth over the
M58
first few years, leading to adult nerve conduction velocities by about the fourth year [70]. If regenerative capacity in animals parallels that in humans, the peripheral nervous system may no longer be able to compensate for chronic low-level nerve injuries from HIV infection by the time the child reaches adolescence.
Peripheral lipoatrophy
Body fat abnormalities, summarized under the term lipodystrophy, are common in HIV-infected adults receiving potent ART, occurring in 3050% or more of individuals in several large, prospective studies [7173]. These abnormalities include, singularly or in combination, central fat accumulation evidenced by increased abdominal girth, development of a dorsocervical fat pad and breast enlargement, as well as loss of peripheral subcutaneous fat (lipoatrophy). The latter designation includes subcutaneous fat loss in the limbs, buttocks and face. There is no universally accepted definition of lipodystrophy, but most descriptions define this syndrome on the basis of clinical features of fat redistribution, as assessed with physical examination, CT or magnetic resonance imaging (MRI) scans, dual-energy X-ray absorptiometry (DXA) scans and anthropometry. The clinical diagnosis of lipodystrophy in children is hampered by the lack of questionnaires appropriate to different paediatric ages and by the fact that alterations in body fat frequently escape clinical detection. The latter observation stems from studies which clearly showed that children without clinical signs of lipodystrophy demonstrate a decrease in limbs fat and an increase in trunk fat with DXA assessment [74,75]. Lumbar single-slide CT or MRI scans have shown great accuracy in quantifying visceral and subcutaneous adipose tissue in patients affected by body composition abnormalities [76]. However, the use of these methods without sedation is only suitable in children older than 45 years and there are no reference values in the paediatric population. DXA quantifies the lean and fat compartments precisely and it is presently considered the gold standard for in vivo body composition studies in children; moreover, it allows a regional analysis of the fat distribution between limbs and trunk [77]. Once again, this technique is only suitable in children older than 45 years and the few available reference values are mainly derived from a population including overweight children [78]. Skinfold thickness can measure the subcutaneous fat layer, allowing a comparison with reference values at well-defined points at all paediatric ages [79]. However, this method suffers from wide inter-/intraperson variability, requires considerable training for the results to be reproducible and may also be inappropriate in subjects with severe lipoatrophy.
Reference values are available for circumferences in children; waist circumference may be a useful parameter to identify trunk adiposity although limb circumferences are not a useful marker of peripheral atrophy, being a measure of the sum of lean and fat mass. In addition, the lack of comparative studies between simple anthropometry (peripheral or central skinfolds and circumferences) and gold standard techniques (DXA and MRI) in HIV-infected children currently limits its diagnostic power, especially in subtle, early forms of lipoatrophy. Overall, the measurement of change in adiposity in children is challenging because of the effects of maturation and growth on lean muscle mass, fat mass and hydration status. This is particularly true in age periods when subcutaneous adipose tissue is minimal (pre-school and school age) or when important inter-individual gender-related variabilities in fat compartment are common (infancy and puberty) [79]. Longitudinal body composition analysis, possibly starting before beginning ART, may be required even more in children than in adults to properly assess lipodystrophy and, consequently, to adopt therapeutic strategies in order to avoid this relevant clinical and metabolic complication. Studies evaluating body composition by DXA, MRI and skinfold thickness have indicated that changes in body fat content and distribution do occur in HIVinfected children. The prevalence of these changes varies from 1833% and increases with longer duration of exposure to ART [74,75,8082]. Recently, a clinical assessment of body fat abnormalities, according to an agreed protocol, was performed on 477 HIV-infected children aged 3 years in 30 European paediatric HIV clinics [83]. Prevalence was 26.0% for any fat redistribution, 8.81% for central lipohypertrophy, 7.55% for peripheral lipoatrophy and 9.64% for the combined subtype. Independent predictors of fat redistribution included advanced stage of HIV disease, female gender, ever used versus never used PIs and d4T. Increasing time since initiation of ART was associated with increased severity of fat redistribution. Considerable controversy exists concerning the pathophysiological mechanisms underlying the development of lipodystrophy. Although many studies support the view that this syndrome is mainly a drugrelated side effect mediated from both the NRTI and PI classes, some studies have demonstrated no evidence of antiretroviral class-specific effects [84]. PIs were initially believed to be the most likely cause of this syndrome, but more recently the use of NRTIs, d4T in particular, has been linked specifically to the development of the lipoatrophic component of lipodystrophy [85,86]. Non-drug factors such as HIV itself, older age, Caucasian race, sex and genetics may also modulate this syndrome.
It has been proposed that mt toxicity induced by NRTI-mediated inhibition of DNA polymerase plays a role in the development of the lipoatrophy component of the lipodystrophy syndrome. This theory was suggested by the phenotypic similarity in fat maldistribution and metabolic abnormalities to what is seen in some patients with inherited mt enzyme deficiency. However, the precise mechanism by which NRTIs contribute to the syndrome is not yet completely understood [87]. Several data support mt toxicity as the mechanism of lipoatrophy whilst other data refute this hypothesis [88]. Mitochondrial structural abnormalities and reduced mtDNA in subcutaneous fat biopsies taken from adult patients with lipoatrophy have been observed [89,90]. Reduction of mtDNA content and respiratory chain activity in adipocytes of patients on NRTI therapy has been demonstrated and these effects were noted 612 months after the beginning of NRTI therapy [91]. Moreover, the ability to reverse peripheral fat loss to some extent following the substitution of d4T with abacavir or AZT, and the fact that such improvement is associated with decreased adipose cell apoptosis may further support the role of NRTIinduced mt toxicity in the development of lipoatrophy [9294]. On the other hand, samples from subjects with lipoatrophy have normal levels of mtDNA in some cases, while some of the samples from HCs showed diminished mtDNA [89,90]. In addition, another study demonstrated a higher mtDNA content in CD4 T cells from patient with lipodystrophy as compared with patients without lipodystrophy and HCs [95]. The existence of possible association(s) between mt toxicity and the lipodystrophy syndrome has been assessed in only one paediatric study so far [96]. Mitochondrial functionality and apoptosis were studied in peripheral blood lymphocytes (PBLs) in 18 HAARTtreated (d4T, 3TC and one protease inhibitor), HIVinfected children, 12 with (LD+) and six without (LD) lipodystrophy, and in 10 HCs. Using flow cytometry, mt membrane potential, mt mass, intra-mt cardiolipin distribution, and early and late apoptosis in fresh PBLs or in PBLs cultured with different stimuli were assessed. MtDNA content in fresh PBLs was also evaluated by competitive quantitative PCR. PBLs from LD+ and LD children and HCs were similar in mt functionality and their tendency to apoptosis. MtDNA content showed a great variability among subjects but overall, there were no relevant differences among the groups. Thus, these results seem to exclude HAART-induced mt damage in PBLs from children with LD. However, some hypotheses may partly explain our results. In HAARTtreated children, PBLs are mainly generated in the thymus and it may be reasonable to assume that the period during which these cells were exposed to drugs
M59
was too short to induce relevant damage. PBLs have a low metabolic rate with a consequent low replication rate of mtDNA, which in turn provokes a low incorporation rate of damaging drugs into mtDNA and a low requirement for polymerase. Lastly, adipocytes may be more representative, as compared with PBLs, to assess the role of NRTI-related mt damage in lipoatrophy.
Treatment of possible mitochondria-related events

The manifestations of NRTI-related mitochondrial toxicity are generally, at least partially, reversible upon withdrawal of the causative agent(s). However, choosing agents that appear to have a low risk of these toxicities and avoiding the use of combinations with overlapping mt toxicities may be the best approach for limiting the incidence of these problems. The severity of the toxicity and the degree of tissue damage may, at least in part, influence the extent of recovery from mt toxicities. Moreover, mt toxicity may recur following the reintroduction of the same or similar agents. As concerns hyperlactataemia, given the controversial role of serum lactate levels and mtDNA/nuclear DNA assessments as screening tools, maintaining a high level of vigilance remains key to the prevention and diagnosis of this condition. Children (when appropriate) on NRTIs (mainly d4T, ddI and AZT) and their families should be instructed on signs and symptoms to look out for and strongly recommended to seek prompt medical evaluation. Physicians should regularly screen children taking these agents for symptoms (particularly during the first 12 months of therapy) such as unexplained weight loss, gastrointestinal complaints, myalgia and paraesthesia, and they should pay particular attention to unexplained tachypnea and neurological deterioration in infants and toddlers. A high index of suspicion is specifically warranted during episodes of infection, as antecedent respiratory infection may precede symptomatic lactic acidosis. In all cases with symptomatic hyperlactataemia and lactic acidosis, ART should be discontinued if no other causes are evident [97]; children who are more seriously ill may require supportive care in an intensive care unit, haemodialysis and mechanical ventilation. Administration of essential cofactors such as thiamine, riboflavin, L-carnitine, coenzyme Q10 and antioxidants have been used as a therapy for congenital mt diseases [98]. The beneficial effect of these agents in subjects with NRTI-related symptomatic hyperlactataemia is largely anecdotal and dosing schedules have not yet been established. However, the low toxicity potential of these agents make them a possible adjunct to standard measures. Even after the interruption of ART, normalization of lactate levels may take 36
M60
months. Rechallenge after symptomatic hyperlactataemia should be performed with caution but carries a low risk of recurrence as long as the culprit NRTI is replaced by either a non-NRTI antiretroviral agent or one less frequently associated with hyperlactataemia, such as abacavir, 3TC or tenofovir [97]. The management of NRTI-related peripheral neuropathy may involve discontinuation or dose reduction of the causative agent, however the latter approach not only lacks a proven beneficial effect but also cannot be recommended in HAART regimens. Even after drug withdrawal, symptoms may persist and require treatment with ibuprofen, acetaminophen, tricyclic antidepressants, topical lidocain, narcotics for refractory pain or acupuncture. The choice of which intervention needs to be carefully tailored to the age of the paediatric patient. Treatment of lipodystrophy is evolving, with no clear standard of care. No proven treatment exists for HIV lipoatrophy and no studies have found that limb fat increases spontaneously over time in patients with lipoatrophy. In vivo studies have suggested that thymidine analogues, especially d4T, are important in the lipoatrophy component of the lipodystrophy syndrome. The effect on fat loss of replacing d4T with abacavir or AZT has been recently assessed in HIV-infected adults with lipoatrophy [92,93]. These studies showed that an improvement of lipoatrophy, albeit modest, occurs in the short-term (2448 weeks) and lipoatrophy is partially reversed in the long-term follow-up (104 weeks). NRTI switches could be an attractive option in HIV-infected children with lipoatrophy, however they cannot be formally suggested at the moment due to the lack of studies in this population.
Conclusions
HIV-infected children treated with potent ART can exhibit a large decrease in HIV-RNA levels and a subsequent increase in CD4 cells. These changes have resulted in significant improvements in clinical outcome and quality of life, dramatically decreasing disease progression and mortality. For many, HIV disease has been transformed from an acute life-threatening illness into a manageable chronic disease, allowing children, who were once not expected to survive, to live well into adulthood. Some clinical events related to NRTI-related mt toxicity have been reported in HIV-infected children but the available studies are scanty as compared with those performed in HIV-infected adults and further research is required to clarify the entity and the consequences of NRTI-related mt toxicity in paediatric patients. Therefore, close monitoring of these syndromes is highly advisable due to the possible relevant damage to liver, muscle and nervous system, and the negative impact on growth related to a persistent subclinical mt toxicity.
Acknowledgements
This work was supported by Grant n. 30F.54, Istituto Superiore di Sanit V Programma Nazionale di Ricerca sullAIDS 2004.
References
1. De Martino M, Tovo PA, Balducci M, Galli L, Gabiano C, Rexxa G & Pezzotti P for the Italian Register for HIV Infection in Children and the Italian National AIDS Registry. Reduction in mortality with availability of antiretroviral therapy for children with perinatal HIV-1 infection. Journal of the American Medical Association 2000; 184:190197. Gerard Y, Maulin L, Yazdanpanah Y, De La Tribonnaire X, Amiel C, Maurage CA, Robin S, Sablonniere B, Dhennain C & Mouton Y. Symptomatic hyperlactataemia: an emerging complication of antiretroviral therapy. AIDS 2000; 14:27232730. John M, Moore CB, James IR, Nolan D, Upton RP, McKinnon EJ & Mallal SM. Chronic hyperlactataemia in HIV-infected patients taking antiretroviral therapy. AIDS 2001; 15:717723. Boubaker K, Flepp M, Sudre P, Furrer H, Haensel A, Hirschel B, Boggian K, Chave JP, Bernasconi E, Egger M, Opravil M, Rickenbach M, Franciolo P & Telenti A. Hyperlactatemia and antiretroviral therapy: the Swiss HIV Cohort Study. Clinical Infectious Diseases 2001; 33:19311937. Moyle GJ, Datta D, Mandalia S, Morlese J, Asboe D & Gazzard BG. Hyperlactatemia and lactic acidosis during antiretroviral therapy: relevance, reproducibility and possible risk factors. AIDS 2002; 14:13411349. John M & Mallal S. Hyperlactactaemia syndromes in people with HIV infection. Current Opinion in Infectious Diseases 2002; 15:2329. Konig H, Behr E, Lower J & Kurth R. Azidothymidine triphosphate is an inhibitor of both human immunodeficiency virus type 1 reverse transcriptase and DNA polymerase gamma. Antimicrobial Agents & Chemotherapy 1989; 33:21092114. Chen C, Vasquez-Padua M & Cheng Y. Effect of anti human immunodefiency virus nucleoside analogues on mitochondrial DNA and its implication for delayed toxicity. Molecular Pharmacology 1991; 39:625628. Brinkman K, ter Hofstede JM, Burger DM, Smeitink JAM & Koopmans PP. Adverse effects of reverse transcriptase inhibitors: mitochondrial toxicity as common pathway. AIDS 1998; 12:17351744. White AJ. Mitochondrial toxicity and HIV therapy. Sexually Transmitted Infections 2001; 77:158173. Mokrzycki M, Harris C, May H, Laut J & Palmisano J. Lactic acidosis associated with stavudine administration: a report of five cases. Clinical Infectious Diseases 2000; 30:198200. Lonergan JL, Behling C, Pfander H, Hassanein TT & Mathews WC. Hyperlactataemia and hepatic abnormalities in ten human immunodeficiency virus-infected patients receiving nucleoside analogue combination regimens. Clinical Infectious Diseases 2000; 31:162166. Coghlan ME, Sommadossi JP, Jhala NC, Many WJ, Saag MS & Johnson VA. Symptomatic lactic acidosis in hospitalized antiretroviral-treated patients with human immunodeficiency virus infection: a report of 12 cases. Clinical Infectious Diseases 2001; 33:19111921. Falco V, Rodriguez D, Ribera E, Martinez E, Mir JM, Domingo P, Diazaraque R, Arribas JR, Gonzales-Garcia JJ, Montero F, Sanchez L & Pahissa A. Severe nucleosideassociated lactic acidosis in human immunodeficiency virus-infected patients: report of 12 cases and review of the literature. Clinical Infectious Diseases 2002; 34:838846.
2.
3.
4.
5.
6. 7.
8.
9.
10. 11.
12.
13.
14.
15. Brinkman K. Management of hyperlactataemia: no need for routine lactate measurements. AIDS 2001; 15:795797. 16. Mizock BA & Falk JL. Lactic acidosis in critical illness. Critical Care Medicine 1992; 20:8093. 17. Miller KD, Cameron M, Wood LV, Dalakas MC & Kovacs JA. Lactic acidosis and hepatic steatosis associated with use of stavudine: report of four cases. Annals of Internal Medicine 2000; 133:192196. 18. Church JA, Mitchell WG, Gonzalez-Gomez I, Christensen J, Vu TH, Dimauro S & Boles RG. Mitochondrial DNA depletion, near-fatal metabolic acidosis and liver failure in an HIV-infected child treated with combination antiretroviral therapy. Journal of Pediatrics 2001; 138:748751. 19. Rey C, Prieto S, Medina A, Perez C, Concha A & Menendez S. Fatal lactic acidosis during antiretroviral therapy. Pediatric Critical Care Medicine 2003; 4:485487. 20. Bissuel F, Bruneel F, Habersetzer F, Chassard D, Cotte L, Chevallier M, Bernuau J, Lucet JC & Trepo C. Fulminant hepatitis with severe lactate acidosis in HIV-infected patients on didanosine therapy. Journal of Internal Medicine 1994; 235:367371. 21. Bleeker-Rovers CP, Kadir SW, van Leusen R & Richter C. Hepatic steatosis and lactic acidosis caused by stavudine in an HIV-infected patient. Netherlands Journal of Medicine 2000; 57:190193. 22. Fortgang IS, Belitsos PC, Chaisson RE & Moore RD. Hepatomegaly and steatosis in HIV-infected patients receiving nucleoside analogue antiretroviral therapy. American Journal of Gastroenterology 1995; 90:14331436. 23. Hofstede HJM, de Marie S, Foudraine NA, Danner SA & Brinkman K. Clinical features and risk factors of lactic acidosis following long-term antiretroviral therapy: 4 fatal cases. International Journal of STD & AIDS 2000; 11:611616. 24. Noguera A, Fortuny C, Sanchez E, Arthuch R, Vilaseca MA, Munoz-Almagro C, Pou J & Jimenez R. Hyperlactatemia in human immunodeficiency virus-infected children receiving antiretroviral treatment. Pediatric Infectious Disease Journal 2003; 22:778782. 25. Desai N, Mathur M & Weedon J. Lactate levels in children with HIV/AIDS on highly active antiretroviral therapy. AIDS 2003; 17:15651567. 26. Lipshultz SE, Chanock S, Sanders SP, Colan SD & McIntosh K. Cardiac manifestations of human immunodeficiency infection in infants and children. American Journal of Cardiology 1989; 63:14891497. 27. Tovo PA, de Martino M, Gabiano C, Cappello N, DElia R & Loy A. Prognostic factors and survival in children with perinatal HIV-1 infection. Lancet 1992; 339:12491253. 28. Luginbuhul LM, Orav EJ, McIntosh K & Lipshultz SE. Cardiac morbidity and related mortality in children with HIV infection. Journal of the American Medical Association 1993; 269:28692875. 29. Turner BJ, Denison M, Eppes SC, Houchens R, Fanning T & Markson LE. Survival experience of 789 children with the acquired immunodeficiency syndrome. Pediatric Infectious Disease Journal 1993; 12:310320. 30. Grenier MA, Karr SS, Rakusan TA & Martin GR. Cardiac disease in children with HIV: relationship of cardiac disease to HIV symptomatology. Pediatric AIDS & HIV Infection 1994; 5:174179. 31. Italian Register for HIV Infection in Children. Features of children perinatally infected with HIV-1 surviving longer than 5 years. Lancet 1994; 343:191195. 32. Lipshultz SE, Orav EJ, Sanders SP & Colan SD. Immunoglobulin and left ventricular struture and function in paediatric HIV infection. Circulation 1995; 92:22202225. 33. Lipshultz SE, Bancroft EA & Boller A-M. Cardiovascular manifestations of HIV infection in children. In The Science and Practice of Pediatric Cardiology. 1997 2nd edn. Edited by JT Bricker, A Garson Jr, DJ Fisher & SR Neish. Baltimore, MD: Williams-Wilkins.
M61
34. Kearney DL for the P2C2 HIV Study Group. Postmortem cardiomegaly correlates with premortem measurements of left ventricular size in malnourished HIV-infected children. 3rd Conference on Retroviruses & Opportunistic Infections. 28 January1 February 1996, Washington, DC, USA. Abstract 431. 35. Lindegren ML, Caldwell B & Oxtoby M. Paediatric HIV infection in the USA. In A Guide to Pratical Management, 1995; pp. 2151. Edited by JYQ Mok & M-L Newell. New York: Cambridge University Press. 36. Lipshultz SE for the P2C2 HIV Study Group. Cardiac dysfunction predicts mortality in HIV-infected children: the prospective NHLBI P2C2 Study. 3rd Conference on Retroviruses & Opportunistic Infections. 28 January1 February 1996, Washington DC, USA. Abstract 415. 37. Lobato MN, Caldwell B, Ng P & Oxtoby MJ. Encephalopathy in children with perinatally acquired human immunodeficiency virus infection. Journal of Pediatrics 1995; 126:710715. 38. Lipshultz SE, Easley K, Kaplan S, Bricker JT, Starc T, Lai W, Moodie D & Colan SD for the P2C2 Study Group. The relation between progressive immune dysfunction and ventricular dysfunction: the prospective NHLBI P2C2 study. 2nd National Conference on Human Retroviruses & Related Infections. 29 January2 February 1995, Washington DC, USA. Abstract 193. 39. Lamperth T, Dalakas MC, Dagani F, Anderson J & Ferrari R. Abnormal skeletal and cardiac muscle mitochondrial in vitro and in an animal model. Laboratory Investigations 1991; 6:742751. 40. Gerschenson M, Erhart SW, Paik CY, Nagashima K, Skopels B, Harbaugh SW, Harbaugh JW, Quan W & Poirier MC. Fetal mitochondrial heart and skeletal muscle damage in Erytrocebus patas monkeys exposed in utero 3-azido-3deoxythymidine. AIDS Research & Human Retroviruses 2000; 16:635644. 41. Lewis W, Grupp I, Grupp G, Hoit B, Morris R, Samarel AM, Bruggeman L & Klotman P. Cardiac dysfuction in the HIV-1 transgenic mouse treated with zidovudine. Laboratory Investigations 2000; 80:187197. 42. Rerkpattanapipat P, Wongpraparut N, Jacobs LE & Kotler MN. Cardiac manifestations of acquired immunodeficiency syndrome. Archives of Internal Medicine 2000; 160:602608. 43. Herskowitz A, Willoughby SB, Baughman KI, Schulman SP & Bartleit JD. Cardiomyopathy associated with antiretroviral therapy in patients with HIV infection: a report of six cases. Annals of Internal Medicine 1992, 18:12601265. 44. Cardoso JS, Moura B, Mota-Miranda A, Goncalves FR & Lecour H. Zidovudine therapy and left ventricular function and mass in human immunodeficiency virus-infected patients. Cardiology 1997; 88:2628. 45. Steinherz L, Broschstein JA & Robins J. Cardiac involvement in congenital acquired immunodeficiency syndrome. American Journal of Diseases of Children 1986; 140:12411244. 46. Joshi VV, Gadol C, Connor E, Oleske JM, Mendelson J & Marin-Garcia J. Dilated cardiomyopathy in children with acquired immunodeficiency syndrome: a pathologic study of five cases. Human Pathology 1988; 19:6973. 47. Lipshultz SE, Orav EJ, Sanders SP, Hale AR, McIntosh K & Colan SD. Cardiac structure and function in children with human immunodeficiency infection treated with zidovudine. New England Journal of Medicine 1992; 327:12601265. 48. Lipshultz SE, Kirk A, Easley MS, Orav EJ, Kaplan S, Starc TJ, Bricker T, Lai WW, Moodie DS, McIntosh K, Schluchter MD & Colan SD. Left ventricular structure and function in children infected with human immunodeficiency virus: the prospective P2C2 HIV Multicenter Study. Pediatric Pulmonary and Cardiac Complications of Vertically Transmitted HIV Infection (P2C2 HIV) Study Group. Circulation 1998; 97:12461256. 49. Lipshultz SE, Kirk A, Easley MS, Orav EJ, Kaplan S, Starc TJ, Bricker T, Lai WW, Moodie DS, McIntosh K, Schluchter MD & Colan SD. Absence of cardiac toxicity of zidovudine
M62
in infants. Pediatric Pulmonary and Cardiac Complications of Vertically Transmitted HIV Infection Study Group. New England Journal of Medicine 2000; 343:759766. 50. Domanski MJ, Sloas MM, Follmann DA, Scalise PP, Tucker EE, Egan D & Pizzo PA. Effect of zidovudine and didanosine on heart function in children infected with human immunodeficiency virus. Journal of Pediatrics 1995; 127:137146. 51. Saulsbury FT. Resolution of organ-specific complications of immunodeficiency virus infection in children with use of highly active antiretroviral therapy. Clinical Infectious Diseases 2001; 32:464468. 52. Cornblath DR & McArthur JC. Predominantly sensory neuropathy in patients with AIDS and AIDS-related complex. Neurology 1998; 38:794796. 53. Fuller GN, Jacobs JM & Guiloff RJ. Nature and incidence of peripheral nerve syndromes in HIV infection. Journal of Neurology, Neurosurgery & Psychiatry 1993; 56:372381. 54. Yarchoan R, Perno CF, Thomas RV, Klecker RW, Allain JP, Wills RJ, McAtee N, Fischl MA, Dubinsky R, McNeely MC, Mitauya H, Pluda JM, Lawley TJ, Luther M, Safai B, Collins JM, Myers CE & Broder S. Phase I studies of 2,3dideoxycytidine in severe human immunodeficiency virus infection as a single agent and alternating with zidovudine. Lancet 1988; 1:7682. 55. Merigan TC, Skowron G, Bozzette SA, Richman D, Uttamchandani R, Fischi M, Schooley R, Hirsch M, Soo W & Pettinelli C. Circulating p24 antigen levels and responses to dideoxycytidine in human immunodeficiency virus (HIV) infections. A phase I and II study. Annals of Internal Medicine 1989; 110:189194. 56. Fischi MA, Olson RM, Follansbee SE, Lalezari JP, Henry DH, Frame PT, Remick SC, Salgo MP, Lin AH & NaussKarol C. Zalcitabine compared with zidovudine in patients with advanced HIV-1 infection who received previous zidovudine therapy. Annals of Internal Medicine 1993; 118:762769. 57. Package Insert Videx (didanosine). October 1991. New York: Bristol-Meyer-Squibb. 58. Rozeneweig M, McLaren C, Beltangady M, Ritter J, Canetta R, Schacter L, Kelley S, Nicaise C, Smaldone L & Dunkle L. Overview of phase I trials of 2,3-dideoxyinosine (ddI) conducted on adult patients. Reviews of Infectious Diseases 1990; 12:570575. 59. Keilbaugh SA, Prusoff WH & Simpson VWV. The PC12 cell as a model for studies of mechanism of induction of peripheral neuropathy by anti-HIV dideoxynucleoside analogs. Biochemical Pharmacology 1991; 42:R5R8. 60. Dalakas MC, Semino-Mora C & Leon-Monzon M. Mitochondrial alterations with mitochondrial DNA depletion in the nerves of AIDS patients with peripheral neuropathy induced by 23-dideoxycitidine (ddC). Laboratory Investigations 2001; 81:15371544. 61. Simpson DM & Tagliati M. Nucleoside analogue-associated peripheral neuropathy in human immunodeficiency virus infection. Journal of Acquired Immune Deficiency Syndromes & Human Retrovirology 1995; 9:153161. 62. Famularo G, Moretti S, Marcellini S, Trinchieri V, Tzantzoglou S, Santini G, Longo A & De Simone C. Acetylcarnitine deficiency in AIDS patients with neurotoxicity on treatment with antiretroviral NRTI. AIDS 1997; 11:185190. 63. Forloni G, Angeretti N & Smiroldo S. Neuroprotective activity of acetyl-L-carnitine: studies in vitro. Journal of Neuroscience Research 1994; 37:9296. 64. Civitello LA. Neurologic complications of HIV infection in children. Pediatric Neurosurgery 1991; 17:104112. 65. Pizzo PA, Butler K, Balis F, Brouwers E, Hawkins M, Eddy J, Einloth M, Falloon J, Husson R & Jarosinski P. Dideoxycitidine alone and in an alternating schedule with zidovudine in children with symptomatic human immunodeficiency virus infection. Journal of Pediatrics 1990; 117:799808.
66. Raphael SA, Price ML, Lischner HW, Griffin JW, Grover WD & Bagarsa O. Inflammatory demyelinating polyneuropathy in a child with symptomatic human immunodeficiency virus infection. Journal of Pediatrics 1991; 118:242245. 67. Floeter MK, Civitell LA, Everett CR, Dambrosia J & Luciano CA. Peripheral neuropathy in children with HIV infection. Neurology 1997; 49:207212. 68. Prufer de QC Araujo A, Nascimento OJM & Garcia OS. Distal sensory polyneuropathy in a cohort of HIV-infected children over five years of age. Pediatrics 2000; 106:E35. 69. Komiya Y. Axonal regeneration in bifurcating axons of dorsal root ganglion cells. Experimental Neurology 1981; 73:824826. 70. Gamstorp I. Normal nerve conduction velocity of ulnar, median and peroneal nerves in infancy, childhood and adolescence. Acta Paediatrica 1963; 14(suppl 146):6876. 71. Schambelan M, Benson CA, Carr A, Currier JS, Dube MP, Gerber JG, Grinspoon SK, Grunfeld C, Kotler DP, Mulligan K, Powderly WG & Saag MS. Management of metabolic complication associated with antiretroviral therapy for HIV1 infection: recommendation of an International AIDS Society-USA panel. Journal of Acquired Immune Deficiency Syndromes 2002; 31:257275. 72. Bernasconi E, Boubaker K, Junghans C, Flepp M, Furrer HJ, Haensel A, Hirschel B, Boggian K, Chave JP, Opravil M, Weber R, Rickenbach M & Telenti A. Abnormalities of body fat distribution in HIV-infected persons treated with antiretroviral drugs: the Swiss HIV Cohort Study. Journal of Acquired Immune Deficiency Syndromes 2002; 31:5055. 73. Health KV, Hogg RS, Singer J, Chan KJ, OShaughnessy MV & Montaner JS. Antiretroviral treatment patterns and incident HIV-associated morphologic and lipid abnormalities in a population-based cohort. Journal of Acquired Immune Deficiency Syndromes 2002; 30:440447. 74. Brambilla P, Bricalli D, Sala N, Renzetti F, Manzoni P, Vanzulli A, Chiumello G, di Natale B & Vigan A. Highly active antiretroviral-treated HIV-infected children show fat distribution changes even in absence of lipodystrophy. AIDS 2001; 15:24152422. 75. Arpadi SM, Cuff PA, Horlick M, Wang J & Kotler DP. Lipodystrophy in HIV-infected children is associated with high viral load and low CD4+-lymphocyte count and CD4+lymphocyte percentage at baseline and use of protease inhibitors and stavudine. Journal of Acquired Immune Deficiency Syndromes 2001; 27:3034. 76. Seidell JC, Bakker CJG & Van der Kooy K. Imaging techniques for measuring adipose tissue distribution. A comparison between computed tomography and 1.5 T magnetic resonance. American Journal of Clinical Nutrition 1990; 51:953957. 77. Ellis KJ, Shypailo RJ, Pratt JA & Pond WG. Accuracy of dual-energy X-ray absorptiometry for body-composition measurements in children. American Journal of Clinical Nutrition 1994; 60:660665. 78. He Q, Horlick M, Fedun B, Wang J, Pierson RN, Heshka S & Gallagher D. Trunk fat and blood pressure in children through puberty. Circulation 2002; 105:10931096. 79. Lohman TG, Roche F & Martorell R (Editors). Anthropometric standardization reference manual, 1988. Champaign, IL, USA: Human Kinetics Books. 80. Jaquet D, Levine M, Ortega-Rodriguez E, Faye A, Polak M, Vilmer E & Levy-Marchal C. Clinical and metabolic presentation of the lipodystrophic syndrome in HIV-infected children. AIDS 2000; 14:21232128. 81. Amaya RA, Kozinetz CA, McMeans A, Schwarzwald H & Kline MW. Lipodystrophy syndrome in human immunodeficiency virus-infected children. Pediatric Infectious Disease Journal 2002; 21:405410. 82. Vigan A, Mora S, Testolin C, Beccio S, Schneider L, Bricalli D, Vanzulli A, Manzoni P & Brambilla P. Increased lipodystrophy with increased exposure to highly active antiretroviral therapy in HIV-infected children. Journal of
83. 84.
85.
86.
87.
88. 89.
90.
91.
92.
93.
94.
95.
96.
Acquired Immune Deficiency Syndromes 2003; 32:482489. European Paediatric Lipodystrophy Group. Antiretroviral therapy, fat redistribution and hyperlipidaemia in HIVinfected children in Europe. AIDS 2004; 18:14431451. Lichtenstein KA, Delaney KM, Armon C, Ward DJ, Moorman AC, Wood KC & Holmberg SD. Incidence of and risk factor for lipoatrophy (abnormal fat loss) in ambulatory HIV-1-infected patients. Journal of Acquired Immune Deficiency Syndromes 2003; 32:4856. Carr A, Samaras K, Burton S, Law M, Freund J, Chisholm DJ & Cooper DA. A syndrome of peripheral lipodystrophy, hyperlipidaemia and insulin resistance in patients receiving HIV protease inhibitors. AIDS 1998; 12:F51F58. Mallal SA, John M, Moore CB, James IR & McKinnon EJ. Contribution of nucleoside analogue reverse transcriptase inhibitors to subcutaneous fat wasting in patients with HIV infection. AIDS 2000; 14:13091316. Cossarizza A, Mussini C & Vigan A. Mitochondria in the pathogenesis of lipodystrophy induced by anti-HIV antiretroviral drugs: actors or bystanders? BioEssays 2001; 23:10701080. Moyle G. Mitochondrial toxicity hypothesis for lipoatrophy: a refutation. AIDS 2001; 15:413414. Walker UA, Bickel M, Lutke Volksbeck SI, Ketelsen UP, Schofer H, Setzer B, Venhoff N, Rickerts V & Staszewski S. Evidence of nucleoside analogue reverse transcriptase inhibitor-associated genetic and structural defects of mitochondria in adipose tissue of HIV-infected patients. Journal of Acquired Immune Deficiency Syndromes 2002; 29:117121. Shikuma CM, Hu N, Milne C, Yost F, Waslien C & Shiramizu B. Mitochondrial DNA decrease in subcutaneous adipose tissue of HIV-infected individuals with peripheral lipoatrophy. AIDS 2001; 15:18011809. Hammond E, Nolan D, James I, Metcalf C & Mallal S. Reduction of mitochondrial DNA content and respiratory chain activity occurs in adipocytes within 612 months of commencing nucleoside reverse transcriptase inhibitor therapy. AIDS 2004; 18:815827. Martin A, Smith DE, Carr A, Ringland C, Amin J, Emery S, Hoy J, Workman C, Doong N, Freund J & Cooper DA; Mitochondrial Toxicity Study Group. Reversibility of lipoatrophy in HIV-infected patients 2 years after switching from a thymidine analogue to abacavir: the MITOX Extension Study. AIDS 2004; 18:10291036. McComsey GA, Ward DJ, Hessenthaler SM, Sension MG, Shalit P, Lonergan JT, Fisher RL, Williams VC & Hernandez JE; Trial to Assess the Regression of Hyperlactatemia and to Evaluate the Regression of Established Lipodystrophy in HIV1-Positive Subjects (TARHEEL; ESS40010) Study Team. Improvement in lipoatrophy associated with highly active antiretroviral therapy in human immunodeficiency virusinfected patients switched from stavudine to abacavir or zidovudine: the results of the TARHEEL study. Clinical Infectious Diseases 2004; 38:263270. McComsey GA, Paulsen DM, Lonergan JT, Hessenthaler SM, Hoppel CL, Williams VC, Fisher RL, Cherry CL, White-Owen C, Thompson KA, Ross ST, Hernandez JE & Ross LL. Improvements in lipoatrophy, mitochondrial DNA levels and fat apoptosis after replacing stavudine with abacavir or zidovudine. AIDS 2005; 19:1523. Cossarizza A, Riva A, Pinti M, Ammannato S, Fedeli P, Mussini C, Esposito R & Galli M. Increased mitochondrial DNA content in peripheral blood lymphocytes from HIVinfected patients with lipodystrophy. Antiviral Therapy 2003; 8:315321. Cossarizza A, Pinti M, Moretti L, Bricalli D, Bianchi R, Troiano L, Fernandez MG, Balli F, Brambilla P, Mussini C & Vigan A. Mitochondrial functionality and mitochondrial DNA content in lymphocytes of vertically infected human immunodeficiency virus-positive children with highly active antiretroviral therapy-related lipodystrophy. Journal of Infectious Diseases 2002; 185:299305.
M63
97. Ogedegbe AE, Thomas DL & Diehi AM. Hyperlactatemia syndromes associated with HIV therapy. Lancet Infectious Diseases 2003; 3:329337. 98. Chinnery PF & Bindoff LA; European neuromuscular center. 116th ENMC international workshop: the treatment
of mitochondrial disorders, 14th16th March 2003, Naarden, The Netherlands. Neuromuscular Disorders 2003; 13:757764.
Received 29 December 2004, accepted 14 April 2005
M64
Diagnosis of mitochondrial dysfunction in HIVinfected patients under highly active antiretroviral therapy: possibilities beyond the standard procedures
Jordi Casademont*, Eduard Sanjurjo, Gloria Garrabou and scar Mir
Laboratory of Mitochondrial Function, Department of Internal Medicine, Hospital Clnic, IDIBAPS, Medical School, University of Barcelona, Barcelona, Catalonia, Spain *Corresponding author: Tel/Fax: +34 9322 79365; E-mail: jcasa@clinic.ub.es
Introduction
Mitochondrial (mt) disorders are a complex group of diseases due to malfunctions of the mitochondrial respiratory chain (MRC). As mitochondria are ubiquous within body cells, clinical manifestations of mt diseases are very polymorphous; classically, they have been considered primary or secondary. Primary mt defects are considered to be those caused by mutations in genes encoding subunits of the MRC [1]. Because mitochondria have a dual genetic control, these defects include mutations of mtDNA as well as nuclear DNA (nDNA) genes. Mitochondrial dysfunctions not related to mutations in genes encoding subunits of the MRC are considered secondary mt disorders [2,3]. They include defects of nuclearencoded mt proteins responsible for the maintenance of MRC subunits and mtDNA biogenesis, or those involved in many other mt biochemical pathways. They also include non-genetic defects that produce derangements of mt homeostasis that have an impact on MRC function. This last situation is generally due to exposure to an environmental toxin or drug, which may interfere with MRC either directly or through a series of mechanisms including secondary genetic defects, which may appear as a consequence of mt derangement. Since the introduction of highly active antiretroviral therapy (HAART) the prognosis of HIV infection has changed dramatically, shifting the concept of HIV disease from that of a highly mortal disease to that of a chronic illness. However, the chronic use of HAART has also been associated with an increase in adverse drug effects, such as hyperlactataemia, polyneuritis and lipodystrophy. It has been proposed that some of these effects could be mediated by mt toxicity and, actually, a decrease in mtDNA copy number and a MRC dysfunction has been repeatedly described [412]. A frequent problem in clinical practice is that the diagnosis of mt dysfunction is not easy to establish. A classic and relatively simple way to approach this issue is the measurement of ketone body ratios. An enzymatic defect of MRC produces a modification of redox potential status due to the accumulation of reduced equivalents (NADH and FADH2). This accumulation partially inhibits the Krebs cycle that, in turn, favours the production of -hydroxybutyrate with respect to acetoacetate inside mitochondria, and lactate with respect to pyruvate in cytoplasm. These ratios were measured some years ago in children with perinatal HIV infection treated with zidovudine without any significant result, possibly because its sensitivity is very low [13]. To our knowledge, its measurement has not been reported in HIV infection since. A series of more sophisticated techniques have been used since then, including morphological, molecular genetic and enzymatic studies. The most important findings using these approaches are the presence of ragged-red fibres on muscle histochemical studies [4], a depletion of mtDNA and a dysfunction of MRC in diverse tissues [512]. A problem with these techniques is that they are usually not routine. Moreover, to establish a diagnosis of mt dysfunction, a combination of these studies is frequently required, and only a few medical centres in each country are equipped for thorough mt evaluation. The suspicion of a possible mt dysfunction in the context of HAART treatment is, consequently, difficult to prove and frequently only suspected. This may have serious implications because the suspicion of mt toxicity often induces a change in HAART regime with the consequent discontinuation of useful antiretrovirals. In this review, we analyse a series of techniques that may suggest the existence of a mt dysfunction not based on pathology, biochemical tissue or molecular genetic analyses. Several of these techniques have been used for the evaluation of classic mt diseases and, with some modifications, could be applied to the diagnosis of HAART-associated mt dysfunction.
M65
J Casademont et al.
Most have not been used for this purpose, their utility being speculative.
Figure 1.13CO2 breath test exhalation curves of four patients with drug-related hyperlactataemia during therapy and after therapy modification
Eight healthy controls Same four patients after HAART modification Four patients on HAART 12 10 8 6 4 2 0 0 15 30
13
Breath tests
Breath tests are simple, cost-effective and safe. For these reasons, they have been proposed for many dynamic evaluations, especially related to liver disease either of genetic or acquired origin. The rationale for their use to assess mt function is that oxidative metabolism of some substrates, such as methionine and ketoisocaproic acid, need the integrity of the hepatocyte electron transport chain and/or ATP synthesis for decarboxylation [14,15]. Methionine is the beststudied carbon-labelled substrate; the isotope usually used is 13C, which is non-radioactive, although 14C can also be used. The procedure consists of measuring the exhalation of 13CO2 after administration of 2 mg/kg body weight of [methyl-13C]-labelled methionine. Breath 13CO2 is measured with a mass spectrometer at baseline and every 15 min thereafter for 120 or 180 min [15]. What the study reflects is hepatic mt function, although to date there is still no general consensus as to its usefulness in the clinical setting. In the absence of liver disease, the 13C-methionine breath test has been used to assess mt function in patients under HAART. Decreased intramitochondrial decarboxylation capacity has been reported in HIVinfected patients with antiretroviral drug-related hyperlactataemia compared with healthy controls [16,17]. Drug discontinuation or regimen modification led to a recovery of this capacity (Figure 1). The most important limitation in the evaluation of HIV-infected patients is the high prevalence of coinfection with hepatitis C and B viruses, as well as possible alcohol abuse or hepatotoxicity from drugs, making it difficult to completely exclude an undiagnosed hepatic disease. Another minor problem in the use of breath tests is the need for mass spectrophotometry. However, this problem can be easily overcome with the development of a kit similar to those used to diagnose Helicobacter pylori infection. This kit uses 13C-urea as a substrate and breath-exhaled samples are collected into bags and mailed to a reference laboratory [18]. The general impression is that this is a young but promising field and that further studies are needed to assess the suitable substrate or substrates required for the evaluation of the complex mt metabolism.
Increase over baseline CO2 enrichment
45
60
75
90 105 120
Time from C-methionine adminstration, min

The second breath test was performed between 3 weeks and 2 months after HAART modification. Results are compared with a series of eight healthy controls (continuous line). Modified from Milazzo et al. [16]. HAART, highly active antiretroviral therapy.
Cardiopulmonary exercise tests

Cardiopulmonary exercise testing is an objective method for evaluating exercise limitations either of cardiac, pulmonary or metabolic origin, in which subjects exercise, preferably on a bicycle ergometer.
M66
During exercise they breathe through a mouthpiece that is a miniaturized pressure differential pneumotachygraph. The inspired and expired gas is continuously sampled and O2 uptake and CO2 elimination are computed [19]. Typical recordings also include cardiac output, dynamic flow volume loops and ventilationperfusion measurements performed at rest and during an incremental workload exercise test. Serum lactate can also be measured. A series of derived parameters can be obtained and correlated with workload, measured in watts. The respiratory exchange ratio (RER) is the quotient of the amount of CO2 produced to the amount of O2 consumed. Peak oxygen consumption (VO2max) in ml oxygen/kg body weight/min denotes cardiovascular or aerobic fitness. The final workload (peak watts/kg) achieved by patients is considered their peak work capacity. The point during exercise at which anaerobic metabolism is used to supplement aerobic metabolism as a source of energy is termed the anaerobic threshold. It normally occurs at >40% of VO2max [19,20]. In mt myopathies, exaggerated circulatory and ventilatory responses to exercise are governed by skeletal muscle oxidative capacity, in which more severely impaired oxidative phosphorylation elicits more active systemic responses. Exercise capacity varies widely, which is attributable to different levels of oxidative impairment, but in general, the average peak work capacity and VO2max are lower than in control
Diagnosis of mitochondrial dysfunction
Venous oxygen saturation (% respect to baseline)
subjects [21,22]. Low anaerobic threshold in a subject with unimpaired cardiovascular fitness provides additional evidence of mt dysfunction [22]. In a group of HIV-positive patients on HAART, classified according to their level of venous lactate levels, there were no significant differences with respect to peak work capacity, expired volume/min or VO2max when compared with a group of controls. However, patients with abnormal venous lactate levels had a higher RER at the peak of exercise and tended to have a lower anaerobic threshold [23]. In another group of HIV-positive patients on HAART, lactate production was higher in patients with lipodystrophy than in patients without lipodystrophy at the same level of VO2max. In the same way, peak work capacity on the cycloergometer was reduced in the lipodystrophic group. The anaerobic threshold occurred earlier in the lipodystrophic group [24]. Finally, a group of HIVinfected patients with lipodystrophy and elevated P-lactate levels had a significantly lower peak work capacity and a trend towards reduced VO2 max compared with controls [25]. One problem in interpreting these results is that a series of HIV-positive patients analysed in the late 1980s before the introduction of HAART, also exercised to a significantly lower peak work capacity compared with a group of non-infected patients. The ventilatory anaerobic threshold values and VO2max were also significantly lower [26]. There are actually several possible mechanisms to explain exercise limitations in HIV-positive patients aside from mt dysfunction: cardiac, ventilatory, peripheral nerve or muscle abnormalities, anaemia, smoking, deconditioning, decreased motivation resulting from chronic disease or a combination of these factors [27]. The extreme difficulty in excluding all these circumstances in HIVpositive patients, together with the need for well-trained personnel to interpret the results and the relatively sophisticated material used, makes cardiopulmonary exercise testing far from being an easy method to implement in usual clinical practice.
Furthermore, in a study not designed to study the rate of lactate clearance, the authors did not find any difference between HIV-infected patients with lipodystrophy and elevated P-lactate levels and a group of controls [25]. This possibly indicates that the test is not useful in the analysis of mitochondria in patients under HAART. An aerobic forearm exercise test specifically developed for the study of mt performance [32] seems more interesting for the study of HAART-related mt dysfunction. The protocol consists of intermittent handgrip exercise (squeeze 1 sec, rest 1 sec) for 3 min at 33% of intended maximal voluntary contraction (MVC) force, which is determined 30 min before initiation of data collection. Oxygen saturation is analysed in blood samples collected from the median cubital vein of the exercised arm. Patients with a mt myopathy do not develop a significant desaturation during exercise compared with healthy subjects or patients with other muscular diseases [32]. We recently had the opportunity to study a patient with symptomatic hyperlactataemia due to HAART [33]. During the acute phase of hyperlactataemia, O2 saturation did not decrease as expected in response to exercise. Antiretroviral therapy was initially stopped and afterwards modified. The patient clinically recovered and after 6 months, the aerobic forearm test was completely normal (Figure 2). Although promising and relatively easy to implement in clinical practice, more studies are required to determine the sensitivity and specificity of this method.
Figure 2. Venous oxygen saturation before a forearm aerobic exercise test (minute 0), during aerobic exercise (minutes 1, 2 and 3) and after resting (minute 4)
Forearm exercise tests

Functional blood tests
The best known is the ischaemic forearm exercise test introduced with the specific aim of screening for abnormalities of muscle glycogen metabolism such as myophosphorylase deficiency or McArdles disease [28,29]. Although not designed for the study of mt defects, it has been suggested that the rate of lactate clearance in the post-exercise period may be significantly slower in patients with mt myopathies than in healthy subjects [30,31]. To our knowledge, there are no systematic studies that sustain this hypothesis.
110 100 90 80 70 60 50 40 30 0 1 2 3 4
Healthy control HIV+ on HAART control without SHL HIV+ on HAART patient with SHL HIV+ on HAART patient after resolving SHL
Time, min
A patient on HAART who developed symptomatic hyperlactataemia did not present a normal venous desaturation during exercise as compared with healthy subjects and patients on HAART without hyperlactataemia. This is an indication of poor O2 utilization during exercise. After 6 months and once the patient was clinically recovered, the aerobic forearm test was normal. SHL, symptomatic hyperlactataemia; HAART, highly active antiretroviral therapy.
M67
J Casademont et al.
Near-infrared spectroscopy (NIRS)

NIRS is a non-invasive optical method for continuous monitoring of oxygenation and haemodynamics in tissue. It is based on the capacity of light in the nearinfrared range to penetrate tissues to a depth of several centimetres and on absorption characteristics of oxyhaemoglobin plus oxymyoglobin, compared with deoxyhaemoglobin plus deoxymyoglobin, with differential near-infrared light spectrometry. A series of optical fibres are placed on top of the muscles of the exercising limb and the difference in the absorption at 760 and 850 nm estimates the relative change in oxyhaemoblobin versus deoxyhaemoglobin. The sum of these spectra provides an estimate of the blood volume. Thus, the net extraction of oxygen from oxyhaemoglobin relative to oxygen delivery by blood circulation can be determined over the area of the tissue sampled by the device, and the effect of exercise on muscle oxygenation can be assessed [34]. Since data can be collected continuously, this device provides unique information regarding the kinetics of oxygen utilisation relative to delivery in the transition from rest to exercise, during sustained exercise and during recovery. This technique has been used to assess patients with classic mt diseases. It seems that these patients present a decrease in O2 consumption compared with controls in both low-intensity exercise and at rest [34,35]. To our knowledge, NIRS has not been used to assess the possible mt toxicity in HAARTtreated patients. The advantages of the non-invasiveness of the procedure are possibly counterbalanced by the difficulties in interpreting the results. It seems, nonetheless, to be a technique with potential utility in this setting.
Imaging techniques
Imaging studies have changed our approach to medical diagnosis in the last 20 years. Many modalities may contribute toward the diagnosis of mt diseases. Table 1 summarizes the most important methods [36]. Among them, magnetic resonance spectroscopy (MRS) seems the most promising for the diagnosis of mild mt dysfunction due to its ability to detect metabolic changes. MRS can be used to monitor tissue bioenergetic changes in both the brain and the skeletal muscle. Proton-MRS is useful in studying bioenergetic changes in the brain. The main resonances observed in the spectrum arise from N-acetylaspartate (a neuronal marker), lactate, creatine and choline. Proton-MRS is less useful in the investigation of skeletal muscle because of the large signal from the protons in the subcutaneous fat, which can obscure other signals of interest [36]. For this reason, proton-MRS seems to be of low utility in the investigation of the effects of HAART regimes. The spectra of phosphorus-MRS (P-MRS) contain several resonances: three arise from the , and phosphate groups of ATP, one from phosphocreatine (PCr) and one from inorganic phosphate (Pi) (Figure 3); two additional smaller resonances can sometimes be observed from phosphomonoesters and phosphodiesters. The area under each resonance is proportional to the amount of the corresponding metabolite. The spectral distance between Pi and PCr provides information about intracellular pH. In normal exercising muscle, there is a PCr decrease linked to Pi accumulation, while the ATP signal remains
Table 1. Clinical imaging techniques useful in the diagnosis of mitochondrial diseases

Imaging modality Magnetic resonance imaging (MRI) Computed tomography (CT) Diffusion weighted imaging Contribution to diagnosis Structural changes Structural changes Differentiation between ischemic and MELAS stroke Changes in central blood flow Detection of glucose/oxygen and oxygen/blood flow ratios Metabolic changes Comments No insight into tissue function No insight into tissue function Limited relevant studies to date
Single-photon-emission computed tomography (SPECT) Positron emission tomography (PET)
Low spatial resolution Limited availability
Phosphorous and proton magnetic resonance spectroscopy (MRS)
Measurement of metabolite concentrations at rest or under exercise
MELAS, mitochondrial encephalopathy with lactic acidosis and stroke-like episodes. Modified from Parry & Matthews [35].
M68
Figure 3. Representation of a muscle phosphorous magnetic resonance spectra from two hypothetical subjects: (A) healthy individual and (B) patient with a mitochondrial myopathy
Figure 4. Muscle dynamic P-MRS study. Scheme of the variations of PCr during a rest-exercise-recovery protocol
PCr
B
40
Rest Exercise Recovery Control
PCr ATP Pi 10 0 10 ppm 10 0 Pi ATP 10 ppm

[PCr] mM
30 Mitochondrial patient
20
10
Spectra arise from the phosphate groups , and of ATP, PCr and Pi. The spectrum from the patient with mitochondrial myopathy has a reduced PCr and an increased Pi concentration in comparison to the spectrum from the healthy individual. ppm, parts per million; Pi, inorganic phosphate; PCr, phosphocreatine.
0 0 5 10 15 Time, min
Patients with a mitochondrial myopathy have a rapid rate of PCr depletion and a prolonged rate of PCr recovery, which is a direct consequence of slower oxidative rephosphorylation of ADP to ATP. Modified from Mattei et al. [39]. PCr, phosphocreatine; P-MRS, phosphorous magnetic resonance spectroscopy.
20
25
30
unchanged due to the continuous resynthesis of ATP through different metabolic pathways. After exercise, mt oxidative phosphorylation remains the main source of ATP, which continues for a while at an accelerated rate to replenish the high-energy phosphates utilised during physical activity. Thus, during this period of recovery, PCr gradually increases, Pi and ADP decrease, and pH returns to its resting level. The initial rate of PCr recovery provides a measure of maximal oxidative rate in the tissue and the recovery of ADP is now considered as one of the most sensitive and reliable indexes of mt dysfunction in vivo [37]. P-MRS of muscle is particularly interesting in the evaluation of patients with mt diseases, which present, at rest, an increase in Pi and, less often, a decrease in PCr resulting in a low PCr/Pi (Figure 3). Abnormalities in resting skeletal muscle and in brain can be detected in patients even with relatively mild disease and normal serum lactate, or no clinical evidence of CNS involvement [38]. Muscle dynamic P-MRS studies during exercise can increase the specificity of the examination (Figure 4). Patients with mt myopathies display a rapid rate of PCr depletion. After stopping exercise, there is a prolonged rate of PCr recovery, which is a direct consequence of slower oxidative rephosphorylation of ADP to ATP, and faster than normal pH recovery [39]. These changes do not seem to be specific for primary mt diseases. They have also been described to occur secondary to traumatic muscle injury or in inflammatory myopathies [40,41]. It is conceivable that P-MRS may be a useful tool for the evaluation of HIV-infected patients on HAART, although no studies have been reported to date.
Conclusions
There is currently a general belief that mt dysfunction plays a pivotal role in some adverse effects observed in patients receiving HAART. The demonstration of this hypothetical dysfunction is complex and relies on a combination of pathological, biochemical and molecular genetic studies that are far from being considered routine in most general clinical settings. In the present review we have analysed some further diagnostic procedures that may help in deciding whether mitochondria are really affected in a given patient. Although some of these procedures are expensive and also limited to a few high technology centres (for example, MRS), others can be easily performed in more conventional outpatient clinics (for example, functional forearm exercise tests). The major inconvenience of all these methods is that they have either been introduced relatively recently or their use in diagnosing mt dysfunction is only a proposition. For these reasons their sensitivity, specificity and positive and negative predictive values remain to be elucidated in the diagnosis of HAART-related mt toxicity.
Acknowledgements
To Fundaci La Marat de TV3 (01/1710 and 02/0210), SGR 2001/00379 and V2003-REDG011-0.
M69
J Casademont et al.
References
1. 2. 3. 4. Wallace CD. Diseases of mitochondrial DNA. Annual Review of Biochemistry 1992; 61:11751212. Cardellach F, Casademont J & Urbano-Marquez A. Mitocondriopatas secundarias. Revista de Neurologia 1998; 26(Suppl 1):S81S86. Schapira AH. Primary and secondary defects of the mitochondrial respiratory chain. Journal of Inherited Metabolic Disease 2002; 25:207214. Casademont J, Barrientos A, Grau JM, Pedrol E, Estivill X, Urbano-Marquez A & Nunes V. The effect of zidovudine on skeletal muscle mtDNA in HIV-1 infected patients with mild or no muscle dysfunction. Brain 1996; 119:13571364. Zaera MG, Miro O, Pedrol E, Soler A, Picon M, Cardellach F, Casademont J & Nunes V. Mitochondrial involvement in antiretroviral therapy-related lipodystrophy. AIDS 2001; 15:16431651. Cote HC, Brumme ZL, Craib KJ, Alexander CS, Wynhoven B, Ting L, Wong H, Harris M, Harrigan PR, OShaughnessy MV & Montaner JS. Changes in mitochondrial DNA as a marker of nucleoside toxicity in HIV-infected patients. New England Journal of Medicine 2002; 346:811820. Casademont J, Mir O & Cardellach F. Mitochondrial DNA and nucleoside toxicity. New England Journal of Medicine 2002; 347:217. Walker UA, Bickel M, Lutke Volksbeck SI, Ketelsen UP, Schofer H, Setzer B, Venhoff N, Rickerts V & Staszewski S. Evidence of nucleoside analogue reverse transcriptase inhibitor-associated genetic and structural defects of mitochondria in adipose tissue of HIV-infected patients. Journal of Acquired Immune Deficiency Syndromes 2002; 29:117121. Miro O, Lopez S, Pedrol E, Rodriguez-Santiago B, Martinez E, Soler A, Milinkovic A, Casademont J, Nunes V, Gatell JM & Cardellach F. Mitochondrial DNA depletion and respiratory chain enzyme deficiencies are present in peripheral blood mononuclear cells of HIV-infected patients with HAART-related lipodystrophy. Antiviral Therapy 2003; 8:333338. Miro O, Lopez S, Martinez E, Rodriguez-Santiago B, Blanco JL, Milinkovic A, Miro JM, Nunes V, Casademont J, Gatell JM & Cardellach F. Short communication: reversible mitochondrial respiratory chain impairment during symptomatic hyperlactatemia associated with antiretroviral therapy. AIDS Research & Human Retroviruses 2003; 19:10271032. Cossarizza A & Moyle G. Antiretroviral nucleoside and nucleotide analogues and mitochondria. AIDS 2004; 18:137151. Lopez S, Miro O, Martinez E, Pedrol E, RodriguezSantiago B, Milinkovic A, Soler A, Garcia-Viejo MA, Nunes V, Casademont J, Gatell JM & Cardellach F. Mitochondrial effects of antiretroviral therapies in asymptomatic patients. Antiviral Therapy 2004; 9:4755. de Martino M, Zammarchi E, Filippi L, Donati MA, Mannelli F, Galli L & Vierucci A. Redox potential status in children with perinatal HIV-1 infection treated with zidovudine. AIDS 1995; 9:13811383. Armuzzi A, Marcoccia S, Zocco MA, De Lorenzo A, Grieco A, Tondi P, Pola P, Gasbarrini G & Gasbarrini A. Non-invasive assessment of human hepatic mitochondrial function through the 13C-methionine breath test. Scandinavian Journal of Gastroenterology 2000; 35:650653. Candelli M, Cazzato IA, Zocco MA, Nista EC, Fini L, Armuzzi A, Camise V, Santoro M, Miele L, Grieco A, Gasbarrini G & Gasbarrini A. 13C-breath tests in the study of mitochondrial liver function. European Review for Medical & Pharmacological Sciences 2004; 8:2331.
chondrial impairment in HIV-infected patients with antiretroviral drug-related hyperlactatemia. Journal of Acquired Immune Deficiency Syndromes 2004; 35:429432. 17. Banasch M, Goetze O, Hollborn I, Hochdorfer B, Brockmeyer N, Schmidt WE & Schmitz F. Non-invasive assessment of hepatic mitochondrial toxicity in HIV-infected patients with normal serum lactate by 13C-methionine breath test. Antiviral Therapy 2004; 9:L19. 18. Versalovic J. Helicobacter pylori. Pathology and diagnostic strategies. American Journal of Clinical Pathology 2003; 119:403412. 19. ATS/ACCP. Statement on cardiopulmonary exercise testing. American Journal of Respiratory & Critical Care Medicine 2003; 167:211277. 20. Mayers JN. The physiology behind exercise testing. Primary Care 2001; 28:528. 21. Dandurand RJ, Matthews PM, Arnold DL & Eidelman DH. Mitochondrial disease. Pulmonary function, exercise performance, and blood lactate levels. Chest 1995; 108:182189. 22. Taivassalo T, Jensen TD, Kennaway N, DiMauro S, Vissing J & Haller RG. The spectrum of exercise tolerance in mitochondrial myopathies: a study of 40 patients. Brain 2003; 126:413423. 23. Tesiorowski AM, Harris M, Chan KJ, Thompson CR & Montaner JS. Anaerobic threshold and random venous lactate levels among HIV-positive patients on antiretroviral therapy. Journal of Acquired Immune Deficiency Syndromes 2002; 31:250251. 24. Chapplain JM, Beillot J, Begue JM, Souala F, Bouvier C, Arvieux C, Tattevin P, Dupont M, Chapon F, Duvauferrier R, Hespel JP, Rochcongar P & Michelet C. Mitochondrial abnormalities in HIV-infected lipoatrophic patients treated with antiretroviral agents. Journal of Acquired Immune Deficiency Syndromes 2004; 37:14771488. 25. Roge BT, Calbet JA, Moller K, Ullum H, Hendel HW, Gerstoft J & Pedersen BK. Skeletal muscle mitochondrial function and exercise capacity in HIV-infected patients with lipodystrophy and elevated p-lactate levels. AIDS 2002; 16:973982. 26. Johnson JE, Anders GT, Blanton HM, Hawkes CE, Bush BA, McAllister CK & Matthews JI. Exercise dysfunction in patients seropositive for the human immunodeficiency virus. American Review of Respiratory Disease 1990; 141:618622. 27. Stringer WW. Mechanisms of exercise limitation in HIV+ individuals. Medicine & Science in Sports & Exercise 2000; 32(7 Suppl):S412S421. 28. Munsat TL. A standardized forearm ischemic exercise test. Neurology 1970; 20:11711178. 29. Sanjurjo E, Laguno M, Bedini JL, Miro O & Grau JM. Forearm ischemic exercise test. Standardization and diagnostic value in the identification of McArdle disease. Medicina Clinica 2004; 122:761766. 30. Clark JB, Bates TE, Boakye P, Kuimov A & Land JM. Investigation mitochondrial defects in brain and skeletal muscle. In A Practical Approach to the Investigation of Metabolic Disease, 1996, pp.151174. Edited by AJ Turner & HS Batchelard. Oxford: IRL Press, Oxford University Press. 31. Dysgaard Jeppesen T, Olsen D & Vissing J. Cycle ergometry is not a sensitive diagnostic test for mitochondrial myopathy. Journal of Neurology 2003; 250:293299. 32. Jensen TD, Kazemi-Esfarjani P, Skomorowska E & Vissing J. A forearm exercise screening test for mitochondrial myopathy. Neurology 2002; 58:15331538. 33. Garrabou G, Lpez S, Sanjurjo E, Infante A, Riba J, Casademont J, Cardellach F & Mir . Mitochondrial dysfunction of HAART-related hyperlactataemia is demonstrable by non-invasive studies. Antiviral Therapy 2004; 9:L24.
5.
6.
7. 8.
9.
10.
11. 12.
13.
14.
15.
16. Milazzo L, Riva A, Sangaletti O, Piazza M, Antinori S & Moroni M. 13C-methionine breath test detects liver mito-
M70
34. Abe K, Matsuo Y, Kadekawa J, Inoue S & Yanagihara T. Measurement of tissue oxygen consumption in patients with mitochondrial myopathy by noninvasive tissue oximetry. Neurology 1997; 49:837841. 35. Beekvelt M, Engelen B, Wevers R & Colier W. Quantitative near-infrared spectroscopy discriminates between mitochondrial myopathies and normal muscle. Annals of Neurology 1999; 46:667670. 36. Parry A & Matthews PM. Roles for imaging in understanding the patophysiology, clinical evaluation, and management of patients with mitochondrial disease. Journal of Neuroimaging 2003; 13:293302. 37. Argov Z, De Stefano N & Arnold DL. ADP recovery after a brief ischemic exercise in normal and diseased human muscle a 31P MRS study. NMR in Biomedicine 1996; 9:165172.
38. Matthews PM, Allaire C, Shoubridge EA, Karpati G, Carpenter S & Arnold DL. In vivo muscle magnetic resonance spectroscopy in the clinical investigation of mitochondrial disease. Neurology 1991; 41:114120. 39. Mattei JP, Bendahan D & Cozzone P. P-31 magnetic resonance spectroscopy. A tool for diagnostic purposes and pathophysiological insights in muscle diseases. Reumatismo 2004; 56:914. 40. Argov Z. Functional evaluation techniques in mitochondrial disorders. European Neurology 1998; 39:6571. 41. Cea G, Bendahan D, Manners D, Hilton-Jones D, Lodi R, Styles P & Taylor DJ. Reduced oxidative phosphorylation and proton efflux suggest reduced capillary blood supply in skeletal muscle of patients with dermatomyositis and polymyositis: a quantitative 31P-magnetic resonance spectroscopy and MRI study. Brain 2002; 125:16351645.
M71
Mitochondrial studies in HAART-related lipodystrophy: from experimental hypothesis to clinical findings

scar Mir*, Snia Lpez, Francesc Cardellach and Jordi Casademont
Mitochondrial Research Laboratory, Department of Internal Medicine, Hospital Clnic, IDIBAPS, Barcelona, Catalonia, Spain *Corresponding author: Tel: +34 93 227 9833; Fax: +34 93 227 5693; E-mail: omiro@clinic.ub.es
Chronic use of antiretrovirals (ARVs) to treat HIV infection, along with more prolonged patient survival, has been associated with an increase in adverse drug effects in HIV-infected patients on treatment. It has been proposed that some of these adverse effects (including myopathy, cardiomyopathy, anaemia, hyperlactataemia/ lactic acidosis, pancreatitis, polyneuritis and lipodystrophy) could be mediated by mitochondrial (mt) toxicity. From the experimental data, it has been proposed that nucleoside analogue reverse transcriptase inhibitors (NRTIs) also inhibit -polymerase, the enzyme devoted to replicate (and, to a lesser extent, repair) mtDNA. It is now widely accepted that the use of most NRTIs in HIVinfected patients is associated with mtDNA depletion.
Although cross-sectional studies suggest that certain ARVs, especially stavudine, are more toxic to mitochondria, the differences among different highly active ARV therapy (HAART) schedules detected in the analysis of longitudinal studies are not so clear. These types of study in previously untreated individuals suggest that the greatest mtDNA loss appears at the beginning of the treatment. Conversely, in ARV-experienced patients, the potential beneficial effects of HAART changes in terms of mtDNA content remain controversial and must be further investigated. Functional studies accompanying genetic investigations are needed for the correct pathogenic interpretation of the mtDNA abnormalities.
Introduction
The clinical use of antiretrovirals (ARVs) to treat HIV infection began in 1986 with the introduction of zidovudine (AZT), an analogue of thymidine nucleoside with proven capacity to inhibit HIV reverse transcriptase (RT). Since then, more than 20 ARVs that act at different steps of the HIV life cycle have been approved. With this wide therapeutic arsenal, the current standard treatment for HIV infection consists of a combination of several ARVs in so-called highly active ARV therapy (HAART). As a consequence, HAART use has achieved a marked decrease in patient mortality and has shifted the concept of HIV infection from a highly mortal disease to a chronic illness. However, chronic use of ARVs together with prolonged patient survival has also been associated with an increase in adverse drug effects. It has been proposed that some of these adverse effects (including myopathy, cardiomyopathy, anaemia, hyperlactataemia/lactic acidosis, pancreatitis, polyneuritis and lipodystrophy) could be mediated by mitochondrial (mt) toxicity. Nonetheless, despite the large number of basic and clinical studies published on this hypothesis during the last 10 years, a unifying theory that explains the clinical manifestations of ARV mt damage is still lacking. One of the main reasons for this is that most of these studies have been performed in vitro and were either non-reproducible or have not been examined in the setting of human studies. Consequently, it is currently unknown why certain patients develop adverse effects in a particular tissue or why patients with similar accumulated doses of ARVs express different patterns of adverse effects. It is possible that ARV-related factors (different rates of tissue incorporation, transportation into mt compartment or intra-mt phosphorylation for each ARV), patient-related factors (mtDNA polymorphisms) or tissue-related factors [oxidative phosphorylation (OXPHOS) dependence] play a significant role in this heterogeneity. Therefore, while for some of these adverse effects (such as toxic myopathy or lactic acidosis) a strong relationship with mt dysfunction has been proven [16], in others (especially lipodystrophy) controversies remain. In this review, we cover the field from experimental hypothesis to clinical studies and analyse the level of evidence with respect to the relationship between ARV adverse effects and mt toxicity, with particular attention to mt participation in lipodystrophy.
M73
Mir et al.
Molecular basis of ARV toxicity on mitochondria

In addition to inhibiting HIV-RT, in vitro studies suggest that nucleoside analogue RT inhibitors (NRTIs) are also able to inhibit and human DNA polymerases, although a current demonstration of such direct inhibition by NRTIs in humans is lacking. From a pathogenic point of view, whereas the inhibition of polymerase (devoted to the repair nuclear of DNA) seems not to be of clinical relevance, the inhibition of -polymerase (-pol, devoted to the replication and, to a lesser extent, the repair of mtDNA) has been proposed to participate in most of the secondary effects associated with the clinical use of NRTIs [7]. Although -pol inhibition is considered a specific class effect of these drugs, the magnitude is not the same for all NRTIs, at least under experimental conditions. Based on experimental laboratory data, different NRTIs have been classified from a higher potency to impair -pol activity to a lower potency as follows: zalcitabine (ddC) > didanosine (ddI) > stavudine (d4T) >>> AZT > lamivudine (3TC) > abacavir (ABV) = tenofovir (TDF) [810]. NRTIs inhibit -pol through four different mechanisms encompassing their effects as: i) mtDNA chain terminators (once incorporated into a growing strand, DNA replication is abruptly halted), ii) competitive inhibitors (competing with natural nucleotides to be incorporated into growing DNA chains by -pol), iii) inductors of errors in the fidelity of mtDNA replication (inhibiting the exonucleolytic proofreading function of -pol) and iv) contributors to the decrease of mtDNA reparatory exonuclease activity (resisting exonucleolytic removal by exonuclease activity of -pol because of the lack of the group 3OH in NRTIs) [11]. As a final consequence, the efficiency of -pol is decreased and the whole cellular content of mtDNA is reduced, while the percentage of point mutations [12] and deletions [13] in such genomes increases. Since mtDNA only codifies for functional proteins corresponding to mt respiratory chain complexes I, III, IV [cytochrome C oxidase (COX)] and V, the final and only possible consequence of mtDNA damage (if the magnitude is great enough to surpass compensatory mechanisms) is to cause respiratory chain dysfunction and low ATP synthesis. Other studies carried out on different cell lines have also demonstrated decreased synthesis of some of these mt-encoded proteins [14,15]. Furthermore, respiratory chain dysfunction can cause a loss of mt membrane potential and an increase in mtdriven apoptosis, a cascade of events that has been demonstrated for AZT and d4T [16]. Experimental studies have also invoked other additional mechanisms aside from -pol inhibition to
M74
completely explain NRTI-associated mt toxicity. For example, AZT is able to i) inhibit the ATP/ADP translocator from rat liver and heart in vitro, thereby limiting the OXPHOS mt capacity [17,18], ii) reduce protein glycosylation [19] and iii) to decrease protein synthesis [20]. The effects of ARV drug classes other than NRTIs against mitochondria have been less frequently studied and evaluated. Although it seems that non-NRTIs (NNRTIs) have no toxic effects on mtDNA, they could interfere with apoptotic pathways and eventually lead to some secondary harmful effects against mitochondria. Recent data indicate that in laboratory assays, efavirenz (EFV) acts as an inductor of the caspasadependent apoptotic mt pathway [21]. Also, few and controversial data have been reported regarding mt effects of protease inhibitors (PIs). While some authors have found that PIs can induce loss of mt membrane potential [22], others have suggested that PIs could exert a beneficial role due to their anti-apoptotic properties [23]. In a very recent review, Badley identified at least five distinct mechanisms for the anti-apoptotic effects of PIs: i) decreasing expression of apoptosis regulatory molecules, ii) caspase inhibition, iii) altering proliferation, iv) inhibiting calpain and v) avoiding loss of mt transmembrane potential [24]. However, under certain circumstances (particularly high doses of PIs) a paradoxical pro-apoptotic effect can be observed in transformed cell lines in vitro and implanted mouse models [24]. Clearly, further studies are required with this drug class to define the exact interactions with mitochondria, especially with apoptotic pathways. Finally, there are no current reports evaluating the effects of fusion inhibitors on mitochondria. The bulk of data generated under experimental laboratory conditions, briefly mentioned above, have been of crucial relevance in order to propose and, in some cases define, the exact mechanisms by which ARV cause mt side effects. However, results from in vitro experimental studies on cultured cells or animals cannot always be directly extrapolated to what occurs in vivo in HIV-infected individuals. Laboratory models usually consist of non-HIV infected cells or animals and, nowadays, there are data consistently indicating that HIV itself causes some mt disarrangements, including mtDNA depletion [4,2528] and respiratory chain dysfunction [2529]. Thus, experimental studies do not take into consideration the role of HIV in facilitating or magnifying ARV-related mt toxic effects and do not include the beneficial effects of ARV in limiting HIV-related mt damage by controlling HIV infection. Additionally, ARV pharmacokinetics are not accounted for in models based on cell cultures, and the effects of the long-term use of ARVs are probably not accurately evaluated either. Moreover, experimental studies are
Mitochondrial studies in HAART-related lipodystrophy
mainly limited to ascertaining the effects of an isolated ARV and, in clinical practice, ARVs are managed as combinations of three or four drugs that are concomitantly administered to patients. Finally, the exact pathogenic significance of experimental findings to explain certain adverse effects (such as lipodystrophy) which appear during HAART, is not established by cellular models. Therefore, this review encompasses mt data from studies performed in HIV-infected patients receiving ARV who have developed lipodystrophy.
Cross-sectional studies in HIV-infected individuals developing lipodystrophy

The first direct data of mt involvement in lipodystrophy was described in a case report of an HIVinfected woman suffering from lipodystrophy. The patient exhibited multiple mtDNA deletions in skeletal muscle and subcutaneous fat along with a marked respiratory chain dysfunction in skeletal muscle and peripheral blood mononuclear cells (PBMCs) [30]. Soon after this, mtDNA depletion was found in subcutaneous fat of eight patients developing lipodystrophy [31]. Since then, case series and cross-sectional studies have both progressively appeared. Currently, mt abnormalities (and especially mtDNA depletion) have been identified in several tissues of HIV-infected patients with lipodystrophy on HAART, including skeletal muscle [13,32,33], liver [32], adipose tissue [13,31,34,35], sperm [36] and PBMCs [12,37]. However, some other authors have not found mt abnormalities [3841]. Table 1 summarizes the published data from transversal studies of HIV patients developing lipodystrophy. As one can see, the evaluated tissues are diverse and there is great heterogeneity of the methodologies for studying mitochondria. For example, Southern blot methodology to measure mtDNA content has been replaced by realtime PCR technology, which renders greater sensitivity, accuracy and reliability than the former. Moreover, it is highly unusual for a single work to provide morphological, genetic and functional data on mitochondria. These facts can explain, at least in part, the lack of current consensus regarding the exact role of mt dysfunction in lipodystrophy. Additionally, since the majority of data comes from HIV lipodystrophic patients who have been receiving ARV for many years, there are important limitations in interpreting the results. For example, the prior drug exposure is different in practically every patient and other uncontrolled patient-related factors [such as age, smoking habit, intake of drugs (other than ARV) which can interfere with mt function] can act as confounding factors, which may severely limit the power of the cross-sectional studies.
In addition, mtDNA depletion has also been described in patients not showing clinically relevant ARV adverse effects [4244]. Some cross-sectional studies have reported that, in asymptomatic individuals, mtDNA depletion is greater in patients receiving d4T as the backbone of HAART in comparison with other HAART regimens [4345]. This is consistent with the greater in vitro affinity of this drug for pol than other currently used NRTIs. However, the finding of mtDNA depletion in asymptomatic individuals limits the pathogenic conclusions from the aforementioned transversal studies of patients with lipodystrophy, since mtDNA depletion could be interpreted as a class drug effect of NRTI that, if investigated, may be demonstrable in most patients receiving such drugs, irrespective of the presence or absence of adverse effects. In fact, it has been reported very recently that mtDNA depletion is even present in asymptomatic HIV-uninfected infants born to HIV-infected women receiving ARV during pregnancy [46]. There are, therefore, reasonable doubts regarding the real value of abnormal findings in HIV-infected individuals with lipodystrophy. In this setting, data from longitudinal studies evaluating the same mt parameters using the same tissues from the same individuals should help to better understand the findings reported by crosssectional studies.
Longitudinal studies in previously untreated HIV-infected individuals

Sequential studies performed in naive patients starting ARV are of special interest because they eliminate the hypothetical effects of ARV taken by patients prior to the current ARV or HAART regimen that is being evaluated. The main endpoint evaluated in such studies is usually the mtDNA content, since this is the direct effect of NRTIs. In a pioneering study, Reiss group [47] reported the effects on the mtDNA content of PBMCs of starting ARV therapy with different ARV strategies, prior to the HAART era. They observed that, after 1 year of treatment, patients receiving AZT in monotherapy exhibited a 27% decrease of mtDNA with respect to baseline and that, when AZT was associated with ddC or ddI, the decrease was 44% and 49%, respectively. It is remarkable that the greatest part of this molecular side effect was already present as early as 4 weeks after starting the NRTIs. More recently, these investigators have also studied HIVnaive patients starting HAART and have found that, after 4 years on treatment, both d4T- and AZT-based HAART are associated with a profound mtDNA depletion in PBMCs, with this decrease being greater in the former HAART schedules (75% of decrease) than in the latter (63% of decrease) [48]. Our laboratory has
M75
Mir et al.
Table 1. Results of cross-sectional studies published until December 2004 in HIV-patients receiving ARV with lipodystrophy
Reference Miro et al. [30] Year 2000 Tissue Skeletal muscle Fat Blood cells Shikuma et al. [31] Zaera et al. [13] 2001 2001 Fat Skeletal muscle Fat White et al. [36] Walker et al. [34] 2001 2002 Sperm Fat n 1 1 1 8 7 2 3 11 Morphology Lipid+ COX/SDH+ ? ? Lipid+ ? ? Lipid+ mt. inclusions Normal ? ? COX-/SDH+/ COX-II def. ? ? ? ? ? COX ? ? mtDNA M. deletions M. deletions Normal Depletion M. deletions M. deletions M. deletions Depletion MRC function Abnormal (C3 def., C4 def., O2 consumption def.) Abnormal (C3 def., C4 def., O2 consumption def.) ? Abnormal (C3 def., C4 def., O2 consumption def.) ? ? ?
Cossarizza et al. [40] McComsey et al. [39] Roge et al. [41] Vittecoq et al. [32]
2002 2002 2002 2002
Blood cells Blood cells Skeletal muscle Skeletal muscle Liver Blood cells
6 10 7 19 5 12 12 23 5 10 13 13
Normal Normal ? M. deletions, depletion Depletion (one out of three cases) Normal Depletion Increase Point mutations ? Depletion Normal
Normal ? Normal Abnormal (C1 def., C3 def., C4 def.) Abnormal (C4 def.) Normal Abnormal (C3 def., C4 def.) ? ? Abnormal (C3 def., C4 def.) ? ?
Miro et al. [37] Cossarizza et al. [38] Martin et al. [12] Chapplain et al. [35] Christensen et al. [33]
2003 2003 2003 2004 2004
Blood cells Blood cells Blood cells Skeletal muscle Fat Blood cells
M, multiple; C1, C3 and C4, complexes I, III and IV of the mt respiratory chain, respectively; n, number of patients studied; ?, not performed; lipid+, lipid deposition; COX, cytochrome c oxidase negative fibres; mt, mitochondrial; SDH+, succinate dehydrogenase hyper-reactive fibres; def., deficiency.
studied 11 patients naive for ARV, four of them starting with d4T+ddI-based HAART and seven of them starting with AZT+3TC-based HAART. As with previously discussed studies, we have also found greater harmful effects of d4T on PBMCs mtDNA abundance, with a 28% decline in mtDNA content 6 months after starting HAART containing d4T+ddI, a situation that it is not observed in those who started HAART containing AZT+3TC. However, not all longitudinal studies have shown greater mtDNA depletion in HAART schemes containing d4T. Petit et al. [45] also evaluated the effect
M76
of starting four different HAART regimens on mtDNA content of PBMCs after a variable time on treatment (from 5080 weeks). They found more modest negative effects on mtDNA abundance (ranging from 1538% of reduction) than those found by Reiss et al. and they did not find clear differences between schemes containing d4T and those containing AZT. Interestingly, the sequential analysis of mtDNA content evolution demonstrates again that the greatest decay is achieved during the first weeks of treatment. Nolans group has been the only team to study the effects of the introduction of HAART on mtDNA content in the subcutaneous
fat of naive patients [49] the target tissue of the lipodystrophy syndrome. They found that when HAART included AZT+3TC, patients developed a 78% decrease in mtDNA after 24 weeks of treatment, whereas when HAART included d4T+3TC, the magnitude of the decrease was 58%. They have further reported that, in such patients, the decrease in mtDNA fat content is accompanied by COX deficiencies demonstrated through immunohistochemical reactions [50]. Taken as a whole (Figure 1), the aforementioned results indicate that commencing HAART is uniformly associated with mtDNA depletion. As previously discussed, the vast majority of cross-sectional studies have found greater mt toxicity for d4T-based HAART than other HAART schedules. However, no clear conclusions regarding a greater capacity to induce mtDNA depletion by a particular ARV or HAART scheme can be depicted from the data reported in longitudinal analysis of previously untreated patients. Additionally, such a decline seems to mainly occur early after starting ARV. This observation is consistent with the cross-sectional study performed in liver by Walker et al. [51]. They found that mtDNA content declines during the initial 6 months of treatment with dideoxynucleotides, with no further decline beyond this time. One possible explanation is that while HAART adverse effects on mtDNA kinetics due to -pol inhibition are clearly manifest at initial stages of treatment, they are further counteracted
by the reduction of HIV viraemia achieved by HAART. Since it has been demonstrated that HIV infection per se is associated with a decrease in mtDNA content [4,25,28], the control of HIV infection may limit the mt side effects of HIV itself. As a result, the initial exponential decline in mtDNA content may be followed by a steady-step phase (Figure 2).
Longitudinal studies of the effects of HAART switching in HIV-infected individuals

From a clinical point of view, it seems that once lipodystrophy appears, changes in HAART schedules do not lead to a rapid improvement of the syndrome. One possible explanation could be that adipocytes involved
Figure 2. Proposed model of action of HIV and ARVs on mtDNA content
Relative effects of HIV on mtDNA content
100%
0%
Elapsed time
Figure 1. Changes in mtDNA content in HIV-naive patients starting antiretrovirals
Relative effects of HAART on mtDNA content
100%
% of mtDNA (compared with baseline)
100 80 60 40 20 0 0 25 50 75 100 125 150 175 200 225 Weeks after initiating ARV treatment
HAART with d4T (n=4) [49] d4T+ddI+RTV (n=12) [45] HAART with d4T (n=15) [48] d4T+3TC+IDV (n=6) [45] Studies on PBMCs Studies on adipose tissue
0%
Elapsed time
Overall mtDNA content
100%
0%
Normality Pure HIV effects Asymptomatic phase Mixed HIV/HAART effects Exponential Steady-state decrease HIV HAART infection initiation
Elapsed time
Symptomatic phase (that is, hyperlactataem polyneuropathy, lipodystrophy)
AZT monotherapy (n=19) [47] AZT+ddC (n=14) [47] AZT+ddI (n=13) [47] AZT+ddC+RTV (n=10) [45] HAART with AZT+3TC (n=3) [49] HAART with AZT (n=8) [48] AZT+3TC+ddI (n=5) [45]
Longitudinal studies published up to December 2004. n, number of patients studied; ARV, antiretroviral; HAART, highly active antiretroviral therapy; mt, mitochondrial; PBMCs, peripheral bloodmononuclear cells; 3TC, lamivudine; ABV, abacavir; AZT, zidovudine; ddC, zalcitabine; ddI, didanosine; d4t, stavudine; IDV, indinavir; RTV, ritonavir.
Precipitating event (rebound of HIV viraemia, switch to a more mitotoxic HAART regimen, introduction of other drugs that interact with mitochondria (ribavirine, valproic acid), failure of up-regulatory mt mechanisms, other mitotoxic factors, individual unknown factors, etc)
ARV, antiretroviral; mt, mitochondrial; HAART, highly active antiretroviral therapy.
M77
Mir et al.
in lipodystrophic changes are severely and, for the most part, inevitably damaged. Since patients with lipodystrophy usually exhibit a mixed pattern of peripheral lipoatrophy and central adiposity, it has been postulated that apoptotic activation could be responsible for lipoatrophy while decreased mt metabolism could result in lipid accumulation in visceral and central adipocytes [49]. Based on experimental data regarding the affinity of NRTIs for -pol, it has been proposed that switching to a HAART scheme that does not contain dideoxynucleotides could eventually improve lipodystrophy. Very recently, data from the MITOX study [52] have shown that, in patients receiving HAART who have developed lipodystrophy, the substitution of d4T or AZT by ABV (one of the NRTIs with the weakest affinity for -pol), is associated with a significant increase in body fat content 2 years after ABV introduction. Unfortunately, no molecular studies of mtDNA content at 2 years accompanied this report. The demonstration of the reversibility of mtDNA depletion associated with HAART changes occurring concurrently with the regression of lipodystrophy would strengthen the existence of an mt role in the development of this syndrome. This close relationship between clinical manifestations and laboratory findings has been better approached for HAART-related hyperlactataemia. For example, Montaner et al. [53] studied eight patients receiving d4T-based HAART who developed hyperlactataemia, and demonstrated that, after discontinuation or substitution of d4T, the disappearance of the clinical signs and symptoms correlated with an increase of over 200% in the mtDNA content in PBMCs. Similarly, we have reported that reversibility of mt abnormalities with HAART interruption is not limited to mtDNA depletion in patients developing hyperlactataemia, but also implicates the restoration of previously altered respiratory chain function [54]. Unfortunately, the data regarding the benefits in terms of mtDNA content of switching a HAART schedule are scarce and have often been obtained in a very limited number of patients with contradictory results [49,5558] (Figure 3). On one hand, Carrs group [57] reported the largest cohort in which sequential mtDNA content has been determined and did not find evidence of any significant mtDNA change after switching from HAART containing a thymidine analogue to HAART containing ABV, at least after 24 weeks of follow-up. Additionally, the behaviour of these patients was the same as that of a control group in which no HAART changes were introduced. Consistent with Carr et al.s results, we have found that the introduction of a nucleoside-sparing HAART (nevirapine plus lopinavir/ritonavir) to ARV-experienced patients is associated only with a trend toward increasing mtDNA
M78
Figure 3. Effects of HAART changes on mtDNA content
750 700 500 450 400 350 300 250 200 150 100 50 0 0 5 10 15 20 25 30 35 40 45 50
% of mtDNA (compared with baseline)
Weeks after switching

Switched from d4T-based to AZT- or ABV-based HAART, fat samples (n=11) [55] Switched from d4T-based HAART to AZT+3TC+ABV, fat samples (n=3) [49] Switched from d4T-based HAART to AZT or ABV-based HAART, skeletal muscle samples (n=12) [56] Switched from d4T-based HAART to AZT or ABV-based HAART, fat samples (n=13) [56] Switched from thymidine-based to ABV-based HAART, PBMCs (n=39) [57] Switched from NRTI-based HAART to NVP+LPV/rit, PBMCs (n=5) [58] Switched from d4T-based HAART to AZT- or ABV-based HAART, PBMCs (n=13) [56]
Longitudinal studies published until January 2005. n, number of patients studied; 3TC, lamivudine; ABV, abacavir; AZT, zidovudine; LPV/rit, lopinavir/ritonavir; NVP, nevirapine; HAART, highly active antiretroviral therapy; mt, mitochondrial; PBMCs, peripheral bloodmononuclear cells.
without changes in respiratory chain function evaluated through COX activity [58]. Indeed, this group of patients showed very similar changes as those in patients who continued with the same NRTI-containing HAART scheme (Figure 4) [57]. It is noteworthy that these non-relevant effects of switching the HAART regimen obtained by our group and Carrs were achieved using PBMCs of asymptomatic individuals. Although some doubts have been raised regarding the sensitivity of this biological sample to detect mt abnormalities, we have demonstrated that changes in mtDNA are present in asymptomatic patients receiving ARV, thereby demonstrating that PBMCs are reliable in the investigation of mt disorders [44]. In a clash with these results, Nolans group [49] found a mean increase of 650% of mtDNA in subcutaneous fat from three patients with lipodystrophy who were switched from d4T-based HAART to a HAART combination consisting in AZT+3TC+ABV. Similarly, McComseys group [55] has reported that mtDNA increased around 100% in the subcutaneous fat of patients with lipodystrophy in whom a d4T-based HAART was replaced by an AZT- or ABV-based HAART scheme. In a more recent and extensive paper, they studied mtDNA content in patients treated with
Figure 4. Longitudinal study comparing the effects of being maintained on a NRTI-containing HAART (NRTI group) or switched to a NRTI-sparing HAART (NVP group) on (A) mtDNA content and (B) complex IV activity
A
% mt DNA activity (compared with baseline)
180 160 140 120 100 80 60 40 20 0 0
NVP group (n=5) NRTI group (n=5) 24 Time, weeks 48
B
140 % Complex IV activity (compared with baseline) 120 100 80 60 40 20 0 0
NVP group (n=5) NRTI group (n=5) 24 Time, weeks 48
lacks a control cohort group to assess the exact effects of pharmacological intervention. Overall, it is important to note that investigations of mtDNA content are not sufficient to correctly ascertain the effects of ARV on mitochondria. The study of functional repercussions of the eventual genetic deficiencies appearing during HAART is probably a more physiopathological approach to knowing the real involvement of mitochondria in the development of lipodystrophy [59]. For example, we have recently reported the presence of up-regulatory mechanisms that compensate for mtDNA depletion developed by patients on treatment with ddI+d4T. Such mechanisms allow maintenance of correct COXII expression (a protein codified by mtDNA, which was depleted in the patients studied), as well as a normal COX activity [60]. Therefore, mtDNA studies only cover one of the multiple aspects of mt function and, consequently, can only offer partial explanations. On the other hand, lipodystrophy probably has a multifactorial mt aetiology where events other than mtDNA depletion play a pivotal role. Among them, ARV modifications of mt-dependent apoptotic pathways, as well as of overall biology of adipose tissue, are being intensively investigated [61]. Regardless of the mechanism, the important question that remains to be answered is how altered mt function leads to the massive modifications observed in adipose tissue of patients developing HAART-related lipodystrophy.
n, number of patients studied; NVP, nevirapine; HAART, highly active antiretroviral therapy; mt, mitochondrial; NRTI, nucleoside reverse transcriptase inhibitor.
Conclusions
Nowadays, it has been demonstrated that the use of most of NRTIs in HIV-infected patients is associated with mtDNA depletion. Although cross-sectional studies suggest that certain ARVs, especially d4T, are more toxic to mitochondria, the differences among different HAART schedules detected in the analysis of longitudinal studies are not so clear. This kind of study in previously untreated individuals suggests that the greatest mtDNA loss appears at the beginning of the treatment. Conversely, in ARV-experienced patients, the potential beneficial effects of HAART switches in terms of affecting mtDNA content remain controversial and must be further investigated. Functional studies accompanying genetic investigations are needed for the correct pathogenic interpretation of mtDNA abnormalities.
d4T-based HAART and lipodystrophy, and found that the replacement of d4T by AZT or ABV was accompanied by a rebound in mtDNA of 141% in skeletal muscle, 146% in adipose tissue and 369% in PBMCs at week 48 [56]. Additionally, they also observed a restoration of mt respiratory chain complex I activity in skeletal muscle [56], as well as a decrease in apoptosis scores in adipose tissue after the switching. It is noteworthy that this study simultaneously investigates, for the first time, sequential genetic and functional mt parameters along with clinical data from patients with lipodystrophy in whom HAART was switched, and it constitutes the first evidence that the substitution of ABV or AZT for d4T improves the subcutaneous fat content as well as reverting abnormal mt indices and fat apoptosis. The demonstration of deficiencies in the diverse tissues evaluated, as well as a close correlation between mtDNA content in PBMCs and in fat, reinforces the usefulness of PBMCs as subrogates of other more appropriate, but more difficult to obtain, tissues. As the only caveat, the McComsey et al. study [55]
Grants
Fundacin para la Investigacin y la Prevencin del Sida en Espaa (FIPSE 3102-00 and 3161/00A), Fundaci La Marat de TV3 (020210), Redes de Investigacin en Mitocondrias (V2003-REDC06E-0) y Sida (Rg-173) and Suport a Grups de Recerca 2001/SGR/00379.
M79
Mir et al.
References
1. Dalakas MC, Illa I, Pezeshkpour GH, Laukaitis JP, Cohen B & Griffin JL. Mitochondrial myopathy caused by longterm zidovudine therapy. New England Journal of Medicine 1990; 332:10981105. Grau JM, Masanes F, Pedrol E, Casademont J, FernandezSola J & Urbano-Marquez A. Human immunodeficiency virus type 1 infection and myopathy: clinical relevance of zidovudine therapy. Annals of Neurology 1993; 34:206211. Masanes F, Barrientos A, Cebrian M, Pedrol E, Mir O, Casademont J & Grau JM. Clinical, histological and molecular reversibility of zidovudine myopathy. Journal of the Neurological Sciences 1998; 159:226228. Cote HC, Brumme ZL, Craib KJ, Alexander CS, Wynhoven B, Ting L, Wong H, Harris M, Harrigan PR, OShaughnessy MV & Montaner JS. Changes in mitochondrial DNA as a marker of nucleoside toxicity in HIV-infected patients. New England Journal of Medicine 2002; 346:811820. Church JA, Mitchell WG, Gonzalez-Gomez I, Christensen J, Vu TH, Dimauro S & Boles RG. Mitochondrial DNA depletion, near-fatal metabolic acidosis, and liver failure in an HIV-infected child treated with combination antiretroviral therapy. Journal of Pediatrics 2001; 138:748751. Mir O, Lopez S, Martinez E, Rodriguez-Santiago B, Blanco JL, Milinkovic A, Miro JM, Nunes V, Casademont J, Gatell JM & Cardellach F. Short communication: reversible mitochondrial respiratory chain impairment during symptomatic hyperlactatemia associated with antiretroviral therapy. AIDS Research & Human Retroviruses 2003; 19:10271032. Cossarizza A & Moyle G. Antiretroviral nucleoside and nucleotide analogues and mitochondria. AIDS 2004; 18:137151. Lim SE & Copeland WC. Differential incorporation and removal of antiviral deoxynucleotides by human DNA polymerase gamma. Journal of Biological Chemistry 2001; 276:2361623623. Kakuda TN. Pharmacology of nucleoside and nucleotide reverse transcriptase inhibitor-induced mitochondrial toxicity. Clinical Therapeutics 2000; 22:685708. Nolan D & Mallal S. Complications associated with NRTI therapy: update on clinical features and possible pathogenic mechanisms. Antiviral Therapy 2004; 9:849863. Lewis W, Day BJ & Copeland WC. Mitochondrial toxicity of NRTI antiviral drugs: an integrated cellular perspective. Nature Reviews 2003; 2:812822. Martin AM, Hammond E, Nolan D, Pace C, Den Boer M, Taylor L, Moore H, Martinez OP, Christiansen FT & Mallal S. Accumulation of mitochondrial DNA mutations in human immunodeficiency virus-infected patients treated with nucleoside-analogue reverse transcriptase inhibitors. American Journal of Human Genetics 2003; 72:549560. Zaera MG, Mir O, Pedrol E, Soler A, Picon M, Cardellach F, Casademont J & Nunes V. Mitochondrial involvement in antiretroviral therapy-related lipodystrophy. AIDS 2001; 15:16431651. Pan-Zhou XR, Zhou XJ, Sommadossi JP & Darley-Usmar VM. Differential effects of antiretroviral nucleoside analogs on mitochondrial function in HepG2 cells. Antimicrobial Agents & Chemotherapy 2000; 44:496503. Walker UA, Setzer B & Venhoff N. Increased long term mitochondrial toxicity in combinations of nucleoside analogue reverse-transcriptase inhibitors. AIDS 2002; 16:21652173. Caron M, Auclair M, Lagathu C, Lombes A, Walker UA, Kornprobst M & Capeau J. The HIV-1 nucleoside reverse transcriptase inhibitors stavudine and zidovudine alter adipocyte functions in vitro. AIDS 2004; 18:21272136. Barile M, Valenti D, Passarella S & Quagliarello E. 3azido3deoxythimidine uptake into isolated rat liver mitochondria and impairment of ADP/ATP translocator. Biochemical Pharmacology 1997; 53:913920. Valenti D, Barile M & Passarella S. AZT inhibition of the ADP/ATP antiport in isolated rat heart mitochondria. International Journal of Molecular Medicine 2000; 6:9396.
2.
3.
4.
5.
6.
7. 8.
9. 10. 11. 12.
13.
14.
15.
16.
17.
18.
19. Hall ET, Yan JP, Melancon P & Kuchta RD. 3azido3deoxythimidine potently inhibits protein glycosylation. Journal of Biological Chemistry 1994; 269:1435514358. 20. Hoobs GA, Keilbaugh SA, Rief PM & Simpsom MV. Cellular targets of 3azido-3deoxythimidine: an early (nondelayed) effect on oxidative phosphorylation. Biochemical Pharmacology 1995; 50:381390. 21. Pilon AA, Lum JJ, Sanchez-Dardon J, Phenix BN, Douglas R & Badley AD. Induction of apoptosis by a nonnucleoside human immunodeficiency virus type 1 reverse transcriptase inhibitor. Antimicrobial Agents & Chemotherapy 2002; 46:26872691. 22. Nerurkar PV, Pearson L, Frank JE, Yanagihara R & Nerurkar VR. Highly active antiretroviral therapy (HAART)-associated lactic acidosis: in vitro effects of combination of nucleoside analogues and protease inhibitors on mitochondrial function and lactic acid production. Cellular & Molecular Biology 2003; 49:12051211. 23. Phenix BN, Lum JJ, Nie Z, Sanchez-Dardon J & Badley AD. Antiapoptotic mechanism of HIV protease inhibitors: preventing mitochondrial transmembrane potential loss. Blood 2001; 98:10781085. 24. Badley AD. In vitro and in vivo effects of HIV protease inhibitors on apoptosis. Cell Death & Differentiation 2005. In press. 25. Mir O, Lopez S, Martinez E, Pedrol E, Milinkovic A, Deig E, Garrabou G, Casademont J, Gatell JM & Cardellach F. Mitochondrial effects of HIV infection on the peripheral blood mononuclear cells of HIV-infected patients who were never treated with antiretrovirals. Clinical Infectious Diseases 2004; 39:710716. 26. Miura T, Goto M, Hosoya N, Odawara T, Kitamura Y, Nakamura T & Iwamoto A. Depletion of mitochondrial DNA in HIV-1-infected patients and its amelioration by antiretroviral therapy. Journal of Medical Virology 2003; 70:497505. 27. Cherry CL, Gahan ME, McArthur JC, Lewin SR, Hoy JF & Wesselingh SL. Exposure to dideoxynucleosides is reflected in lowered mitochondrial DNA in subcutaneous fat. Journal of Acquired Immune Deficiency Syndromes 2002; 30:271277. 28. Chiappini F, Teicher E, Saffroy R, Pham P, Falissard B, Barrier A, Chevalier S, Debuire B, Vittecoq D & Lemoine A. Prospective evaluation of blood concentration of mitochondrial DNA as a marker of toxicity in 157 consecutively recruited untreated or HAART-treated HIV-positive patients. Laboratory Investigations 2004; 84:908914. 29. Polo R, Martinez S, Madrigal P & Gonzalez-Munoz M. Factors associated with mitochondrial dysfunction in circulating peripheral blood lymphocytes from HIV-infected people. Journal of Acquired Immune Deficiency Syndromes 2003; 34:3236. 30. Mir O, Gmez M, Pedrol E, Cardellach F, Nunes V & Casademont J. Respiratory chain dysfunction associated with multiple mitochondrial DNA deletions in antiretroviral therapy-related lipodystrophy. AIDS 2000; 14:18551857. 31. Shikuma CM, Hu N, Milne C, Yost F, Waslien C, Shimizu S & Shiramizu B. Mitochondrial DNA decrease in subcutaneous adipose tissue of HIV-infected individuals with peripheral lipoatrophy. AIDS 2001; 15:18011809. 32. Vittecoq D, Jardel C, Barthelemy C, Escaut L, Cheminot N, Chapin S, Sternberg D, Maisonobe T & Lombes A. Mitochondrial damage associated with long-term antiretroviral treatment: associated alteration or causal disorder? Journal of Acquired Immune Deficiency Syndromes 2002; 31:299308. 33. Christensen ER, Stegger M, Jensen-Fangel S, Laursen AL & Ostgergaard L. Mitochondrial DNA levels in fat and blood cells from patients with lipodystrophy or peripheral neuropathy and the effect of 90 days of high-dose coenzyme Q treatment: a randomized, double blind, placebo-controlled pilot study. Clinical Infectious Diseases 2004; 39:13711379. 34. Walker UA, Bickel M, Lutke Volksbeck SI, Ketelsen UP, Schofer H, Setzer B, Venhoff N, Rickerts V & Staszewski S.
M80
35.
36. 37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
Evidence of nucleoside analogue reverse transcriptase inhibitor-associated genetic and structural defects of mitochondria in adipose tissue of HIV-infected patients. Journal of Acquired Immune Deficiency Syndromes 2002; 29:117121. Chapplain JM, Beillot J, Begue JM, Souala F, Bouvier C, Arvieux C, Tattevin P, Dupont M, Chapon F, Duvauferrier R, Hespel JP, Rochcongar P & Michelet C. Mitochondrial abnormalities in HIV-infected lipoatrophic patients treated with antiretroviral agents. Journal of Acquired Immune Deficiency Syndromes 2004; 37:14771488. White DJ, Mital D, Taylor S & St John JC. Sperm mitochondrial DNA deletions as a consequence of long term highly active antiretroviral therapy. AIDS 2001; 15:10611062. Mir O, Lopez S, Pedrol E, Rodriguez-Santiago B, Martinez E, Soler A, Milinkovic A, Casademont J, Nunes V, Gatell JM & Cardellach F. Mitochondrial DNA depletion and respiratory chain enzyme deficiencies are present in peripheral blood mononuclear cells of HIV-infected patients with HAART-related lipodystrophy. Antiviral Therapy 2003; 8:333338. Cossarizza A, Riva A, Pinti M, Ammannato S, Fedeli P, Mussini C, Esposito R & Galli M. Increased mitochondrial DNA content in peripheral blood lymphocytes from HIVinfected patients with lipodystrophy. Antiviral Therapy 2003; 8:315321. McComsey G, Tan DJ, Lederman M, Wilson E & Wong LJ. Analysis of the mitochondrial DNA genome in the peripheral blood leukocytes of HIV-infected patients with or without lipoatrophy. AIDS 2002; 16:513518. Cossarizza A, Pinti M, Moretti L, Bricalli D, Bianchi R, Troiano L, Fernandez MG, Balli F, Brambilla P, Mussini C & Vigano A. Mitochondrial functionality and mitochondrial DNA content in lymphocytes of vertically infected human immunodeficiency virus-positive children with highly active antiretroviral therapy-related lipodystrophy. Journal of Infectious Diseases 2002; 185:299305. Roge BT, Calbet JA, Moller K, Ullum H, Hendel HW, Gerstoft J & Pedersen BK. Skeletal muscle mitochondrial function and exercise capacity in HIV-infected patients with lipodystrophy and elevated p-lactate levels. AIDS 2002; 16:973982. Pace CS, Martin AM, Hammond EL, Mamotte CD, Nolan DA & Mallal SA. Mitochondrial proliferation, DNA depletion and adipocyte differentation in subcutaneous adipose tissue of HIV-positive HAART recipients. Antiviral Therapy 2003; 8:323331. de Mendoza C, de Ronde A, Smolders K, Blanco F, GarciaBenayas T, de Baar M, Fernandez-Casas P, Gonzalez-Lahoz J & Soriano V. Changes in mitochondrial DNA copy number in blood cells from HIV-infected patients undergoing antiretroviral therapy. AIDS Research & Human Retroviruses 2004; 20:271273. Lopez S, Mir O, Martinez E, Pedrol E, Rodriguez-Santiago B, Milinkovic A, Soler A, Garcia-Viejo MA, Nunes V, Casademont J, Gatell JM & Cardellach F. Mitochondrial effects of antiretroviral therapies in asymptomatic patients. Antiviral Therapy 2004; 9:4755. Petit C, Mathez D, Barthelemy C, Leste-Lasserre T, Naviaux RK, Sonigo P & Leibowitch J. Quantitation of blood lymphocyte mitochondrial DNA for the monitoring of antiretroviral drug-induced mitochondrial DNA depletion. Journal of Acquired Immune Deficiency Syndromes 2003; 33:461469. Divi RL, Walker VE, Wade NA, Nagashima K, Seilkop SK, Adams ME, Nesel CJ, ONeill JP, Abrams EJ & Poirier MC. Mitochondrial damage and DNA depletion in cord blood and umbilical cord from infants exposed in utero to combivir. AIDS 2004; 18:10131021. Reiss P, Casula M, de Ronde A, Weverling GJ, Goudsmit J & Lange JM. Greater and more rapid depletion of mitochondrial DNA in blood of patients treated with dual (zidovudine+didanosine or zidovudine+zalcitabine) vs single (zidovudine) nucleoside reverse transcriptase inhibitors. HIV Medicine 2004; 5:1114.
48. van der Valk M, Casula M, Weverlingz GJ, van Kuijk K, van Eck-Smit B, Hulsebosch HJ, Nieuwkerk P, van Eeden A, Brinkman K, Lange J, de Ronde A & Reiss P. Prevalence of lipoatrophy and mitochondrial DNA content of blood and subcutaneous fat in HIV-1-infected patients randomly allocated to zidovudine- or stavudine-based therapy. Antiviral Therapy 2004; 9:385393. 49. Nolan D, Hammond E, James I, McKinnon E & Mallal S. Contribution of nucleoside-analogue reverse transcriptase inhibitor therapy to lipoatrophy from the population to the cellular level. Antiviral Therapy 2003; 8:617626. 50. Hammond E, Nolan D, James I, Metcalf C & Mallal S. Reduction of mitochondrial DNA and respiratory chain activity occurs in adipocytes within 612 months of commencing nucleoside reverse transcriptase inhibitor therapy. AIDS 2004; 18:815817. 51. van der Valk M, Casula M, Weverlingz GJ, van Kuijk K, van Eck-Smit B, Hulsebosch HJ, Nieuwkerk P, van Eeden A, Brinkman K, Lange J, de Ronde A & Reiss P. Depletion of mitochondrial DNA in liver under antiretroviral therapy with didanosine, stavudine or zalcitabine. Hepatology 2004; 39:311317. 52. Martin A, Smith DE, Carr A, Ringland C, Amin J, Emery S, Hoy J, Workman C, Doong N, Freund J & Cooper DA; Mitochondrial Toxicity Study Group. Reversibility of lipoatrophy in HIV-infected patients 2 years after switching from a thymidine analogue to abacavir: the MITOX Extension Study. AIDS 2004; 18:10291036. 53. Montaner JS, Cote HC, Harris M, Hogg RS, Yip B, Harrigan PR & OShaughnessy MV. Nucleoside-related mitochondrial toxicity among HIV-infected patients receiving antiretroviral therapy: insights from the evaluation of venous lactic acid and peripheral blood mitochondrial DNA. Clinical Infectious Diseases 2004; 38:S73S79. 54. Mir O, Lopez S, Martinez E, Rodriguez-Santiago B, Blanco JL, Milinkovic A, Miro JM, Nunes V, Casademont J, Gatell JM & Cardellach F. Short communication: reversible mitochondrial respiratory chain impairment during symptomatic hyperlactatemia associated with antiretroviral therapy. AIDS Research & Human Retroviruses 2003; 19:10271032. 55. McComsey G & Lonergan JT. Mitochondrial dysfunction: patient monitoring and toxicity management. Journal of Acquired Immune Deficiency Syndromes 2004; 37:S30S35. 56. McComsey GA, Paulsen DM, Lonergan JT, Hessenthaler SM, Hoppel CL, Williams VC, Fisher RL, Cherry CL, White-Owen C, Thompson KA, Ross ST, Hernandez JE & Ross LL. Improvements in lipoatrophy, mitochondrial DNA levels and fat apoptosis after replacing stavudine with abacavir or zidovudine. AIDS 2005; 19:1523. 57. Hoy JF, Gahan ME, Carr A, Smith D, Lewin SR, Wesselingh S & Cooper DA. Changes in mitochondrial DNA in peripheral blood mononuclear cells from HIV-infected patients with lipoatrophy randomized to receive abacavir. Journal of Infectious Diseases 2004; 190:688692. 58. Negredo E, Molto J, Burger D, Cote H, Mir O, Ribalta J, Martinez E, Puig J, Ruiz L, Salazar J, Lopez S, Montaner J, Rey-Joly C & Clotet B. Lopinavir/ritonavir plus nevirapine as a nucleoside-sparing approach in antiretroviral-experienced patients (NEKA Study). Journal of Acquired Immune Deficiency Syndromes 2005; 38:4752. 59. Casademont J, Mir O & Cardellach F. Mitochondrial DNA and nucleoside toxicity. New England Journal of Medicine 2002; 347:216218. 60. Mir O, Lopez S, Rodriguez de la Concepcion M, Martinez E, Pedrol E, Garrabou G, Giralt M, Cardellach F, Gatell JM, Vilarroya F & Casademont J. Upregulatory mechanisms compensate for mitochondrial DNA depletion in asymptomatic individuals receiving stavudine plus didanosine. Journal of Acquired Immune Deficiency Syndromes 2004; 37:15501555. 61. Villaroya F, Domingo P & Giralt M. Lipodystrophy associated with highly active anti-retroviral therapy for HIV infection: adipocyte as a target of anti-retroviral-induced mitochondrial toxicity. Trends in Pharmacological Sciences 2005; 26:8893.

1st Meeting on Mitochondrial Toxicity & HIV Infection: Understanding the Pathogenesis for a Therapeutic Approach M81
Mitochondrial DNA levels of peripheral blood mononuclear cells and subcutaneous adipose tissue from thigh, fat and abdomen of HIV-1 seropositive and negative individuals
Mariana Gerschenson*, Bruce Shiramizu, Daniel E LiButti & Cecilia M Shikuma
Hawaii AIDS Clinical Research Program, Department of Medicine, John A Burns School of Medicine, University of Hawaii, Honolulu, HI, USA *Corresponding author: Tel: +1 808 737 2751; Fax: +1 808 735 7047; Email: gerschen@hawaii.edu
Mitochondrial dysfunction has been demonstrated in subcutaneous adipose tissue from lipoatrophic HIV-1infected patients treated with nucleoside reverse transcriptase inhibitors (NRTIs). To further assess mitochondrial toxicity, mitochondrial DNA (mtDNA) copies/cell were measured in subcutaneous fat from various sites. Peripheral adipose tissues were obtained from the abdominal wall near the umbilicus, anterior lateral thigh, and dorsal cervical region of the neck of individuals from four cohorts: 1) seven lipoatrophic HIV1-infected patients receiving a regimen with nucleoside reverse transcriptase inhibitors (NRTIs) as part of highly active antiretroviral therapy (HAART) for >6 months; 2) seven non-lipoatrophic HIV-1-infected patients receiving NRTIs-containing HAART; 3) five HIV-1-infected patients on antiretroviral therapy <2 weeks (naive); 4) and five HIV-1-negative participants. Along with the adipose tissue samples, peripheral blood mononuclear cells
(PBMC) were also obtained from each patient for mtDNA depletion examination. Total DNA was isolated and mtDNA copies/cell quantitated by competitive and realtime PCR. MtDNA copies/cell in abdomen, thigh, and neck fat were depleted in lipoatrophic HIV-1 seropositive compared to the seropositive naive and seronegative cohorts. MtDNA copies/cell in thigh and neck fat were also decreased in non-lipoatrophic subjects exposed to NRTIs compared with NRTI-naive and HIV seronegative controls. PBMC values did not differ among the cohorts and there was no correlation with lipoatrophy state or HIV-1 serostatus. Additionally, differences in mtDNA copies/cell were observed in the fat depots from seronegative subjects. Thigh fat mtDNA levels were 4555% lower than abdomen and neck. These studies help demonstrate that mtDNA levels can vary in different subcutaneous adipose depots suggesting possible metabolic differences.
Introduction
HIV-1 lipodystrophy is a syndrome characterized by alterations in the normal distribution of fat tissue resulting in subcutaneous lipoatrophy of the face, arms and legs as well as increases in visceral fat of the abdomen and back of the neck. Metabolic findings of hyperlipidaemia and insulin resistance often accompany these distressing body morphology changes [1,2]. Epidemiological data suggests a strong association between the use of NRTIs and the development of lipoatrophy [37]. Mitochondrial involvement in the pathogenesis of HIV-1 lipoatrophy has been hypothesized via their role in regulating energy metabolism in adipose tissue [8] and to their known sensitivity to NRTIs [9]. Mitochondria contain their own DNA consisting of a circular, double stranded DNA molecule of 16 kb. Although NRTIs preferentially inhibit HIV-1 reverse transcriptase, NRTIs can also inhibit cellular DNA polymerases, most notably mitochondrial DNA (mtDNA) polymerase . DNA polymerase is a key regulatory enzyme of mtDNA replication. It has been hypothesized that inhibition of this enzyme by NRTIs results in mtDNA depletion and/or impairment in the production of mitochondrial enzymes within the adipocyte [10,11]. It is hypothesized that tissue toxicities due to NRTIs will occur if mtDNA falls below a critical level, so that the production of mtDNAencoded protein subunits of the mitochondrial respiratory chain and RNAs are insufficient to meet cellular energy requirements. Recently several studies have shown decreases in mtDNA levels in subcutaneous adipose tissue in lipoatrophic patients taking NRTIs [1215]. This has led
M83
M Gerschenson et al.
investigators to query whether mitochondrial parameters could be monitored in a non-invasive manner, for example in peripheral blood mononuclear cells (PBMCs), as in some genetically inherited mitochondrial diseases [16]. In this study, we measure mtDNA copies/cell by quantitative PCR in subcutaneous adipose tissues from the abdominal wall near the umbilicus, anterior lateral thigh and dorsal cervical region of the neck, and PBMCs from HIV-1 seropositive lipoatrophic patients, HIV-1 seropositive non-lipoatrophic, HIV-1 seropositive nave, and seronegative patients.

Patients and specimens
As per guidelines established by the University of Hawaii Institutional Review Board, specimens for this study were obtained without personal identifiers from a previously described cross-sectional 4-cohort study [12] and 5 additional HIV seronegative age and gender matched controls for the PBMC measurements. The cohorts consisted of: 1) HIV-1 (+) individuals on NRTI (zidovudine, stavudine and/or didanosine)-containing HAART for more than six months with self-reported changes in lipodystrophy which included peripheral lipoatrophy following initiation of such therapy (n=3 were taking stavudine); 2) HIV-1 (+) individuals on similar NRTI-containing HAART without selfreported changes of lipodystrophy (n=4 were taking stavudine) (Table 1); 3) HIV-1 (+) individuals with less than two weeks total exposure to antiretroviral therapy (nave); and 4) individuals seronegative for antibodies to HIV-1 by ELISA. Subcutaneous adipose tissue specimens were previously obtained by core needle technique from the abdominal wall near the umbilicus, anterior lateral thigh, and dorsal cervical region of the neck. The tissues were flash frozen and were stored at 70C. Available sample numbers from the four cohorts of the cross-sectional study varied from n=310. PBMCs were isolated by Ficoll gradient and stored viably at 70C. Total DNA was isolated using QIAamp DNA Blood Mini kit and DNeasy Tissue kit; both are from Qiagen (Qiagen Inc, Valencia, CA, USA). DNA integrity was examined by agarose gel electrophoresis. Intact DNA was assayed for mtDNA copies/cell by real-time PCR.
exon, was used as a standard. The competitor plasmid with the embedded mtDNA and gDNA is smaller than the target DNA from the specimens, which allows for differentiation when size-fractionated on an agarose gel. In order to determine the number of mitochondrial DNA copies per cell, it was necessary to perform two QC-PCR, one to detect mitochondrial primers, and one with genomic primers. Serial dilutions of the competitor (108100) were set up with the same amounts of DNA from the patient sample. The mitochondrial primers, mt192R (GCTCTAGAAAGAGATCAGGTTCGTCCTTTAGTG) and mt27D (AAA ATGAACGAAAATCTGTTCGCT) amplified a 180 bp fragment of the gene encoding for the subunit VI of FOF1 ATPase. 276R (GGGGATCCGGT GGCAGCGGTAGTGGAG) and 60D (GGGAATTCTGAGAAGAAGTAAAACC GTTTGCTG), the genomic primers, amplified a 259 bp fragment of the Fas Ligand. PCR preparation and cycling was identical for both sets of competitive PCRs. Each reaction tube was prepared in a final volume of 25 l and contained 1 PCR Buffer, 0.2 mmol/l of each dNTP, 0.2 mol/l of each primer, 1 unit of Amplitaq (Applied Biosystems, Foster City, CA, USA), 1 l of a known pMITO dilution (ranging from 108 copies per l to 100 copies per L) and approximately 100 ng of patient DNA. Cycling conditions were as follows: 94C for 5 min, 50 cycles of 94C for 30 sec, 55C for 30 sec and 72C for 30 sec, with a final hold at 72C for 7 min. PCR products were visualized on an ethidium bromide stained 1% agarose/1.5% low molecular weight agarose gel and quantitated using a Biorad imaging densitometer (Bio-Rad Laboratories, CA, USA). In order to determine the copies of mtDNA per cell, the sample value determined with mitochondrial primers was divided by the value established by the genomic primers and this ratio was multiplied by two, due to there being two copies of the nuclear-encoded genes in each cell.
Analysis of mtDNA copies/cell by real-time PCR

All fat tissues and PBMCs were assayed using real-time PCR as previously described [18]. Standardization of real-time PCR was performed using iQ SYBR green supermix with the BioRad iCycler real-time instrument (Bio-Rad Laboratories, CA, USA). The mitochondrial primers, mtDIR (CAC AGA AGC TGC CAT CAA GTA) and mtREV (CCG GAG AGT ATA TTG TTG AAG AG), amplify a region of mtDNA (90 bp) which encodes for NADH dehydrogenase subunit 2 (90 bp). The genomic primers, GenDIR (GGC TCT GTG AGG GAT ATA AAG ACA) and GenREV (CAA ACC ACC CGA GCA ACT AAT CT) were specific for the region of the genome encoding the Fas Ligand (98 bp) [18]. The PCR products were cloned into plasmid, which were used as standard in serial dilutions. Each sample
Quantitative competitive PCR (QC-PCR) mtDNA analysis

A competitive PCR assay to semi-quantitate mtDNA copy per cell was set up as previously described [17] and used to assay the abdominal tissue. A plasmid containing the human mitochondrial gene encoding the subunit VI of the F0F1 ATPase region and a gDNA insert from the human FasL gene, spanning the first
M84
MtDNA levels in PBMCs and fat in HIV lipoatrophic patients
Table 1. Cohort characteristics of lipoatrophic and non-lipoatrophic cohorts

Prior additive exposure time on ART
Case, n Non-lipoatrophic (n=7)
ART at time of biopsy
Age
Gender
1006
11 months d4T/3TC/EFV 60 months ZDV/3TC 9 months d4T/ddI/IDV
2 months ZDV 3 months ABC None
51
2014
50
3015
12 months ZDV/3TC 24 months ABC 9 months IDV/EFV 12 months ZDV
33
4016
60 months d4T/NFV 48 months ddI 10 months d4T/ABC/EFV
42
5022
108 months ZDV 42 months 3TC 9 months EFV 24 months ZDV 12 months 3TC/NFV None
42
6027
1 month d4T/ZDV/NFV 24 month ZDV/3TC/IDV
41
7031
35
Lipoatrophic (n=7)
1002
102 months ZDV
24 months d4T 18 months NFV None
40
2003
144 months ZDV 60 months 3TC 48 months IDV 96 months ZDV 18 months 3TC 1 month ABC 36 months ZDV/ddC/NFV 42 months d4T 48 months 3TC/IDV 72 months ZDV/IDV 24 months d4T/ddI/NFV
47
3004
25 months NFV
45
4009
24 months d4T/3TC
36
5017
84 months ZDV 30 months ddI None 36 months ZDV/3TC
42 M 55 59 M F
6018 7020
ABC, abacavir; ART, antiretroviral therapy; ddI, didanosine; d4T, stavudine; EFV, efavirenz; IDV, indinavir; NFV, nelfinavir; RTV, ritonavir; 3TC, lamivudine; ZDV, zidovudine
M85
and standard was run in triplicate (50 l reaction volume) containing: 2X SYBR Green SuperMix, 0.5 M of each primer, and 10 ng sample of DNA. PCR cycling conditions were: denaturation for 1 cycle 95C for 14:30 min, amplification for 40 cycles 95C for 30 sec, 58C for 1 min, and 72C for 30 sec, final extension for 1 cycle at 72C for 6:30 min, melt for 80 cycle at 55C for 30 sec, cool 1 cycle at 4C for infinity. The results were then analysed with version 3.0 iCycler software (Bio-Rad Laboratories).
non-lipoatrophic (51 50) was not statistically different [12] (Table 1). Another interesting finding was variation in mtDNA copies/cell in the thigh, abdomen, and neck fat in the HIV-1 seronegative
Figure 1. Ethidium-stained gel of PCR products from pMITO plasmid using mtDNA and gDNA primers.
A
Statistical analysis
Statistical analysis was done using the MannWhitney rank sum test for non-parametric data with the threshold for significance set at P=0.05. The assays were compared by linear regression analyses using Sigma Stat 3.0 (SPSS Inc, IL, Chicago, USA).
mtDNA 180 bp 108 107 106 105 104 103 102 101 100 N
gDNA 259 bp
Results
Competitive PCR of abdominal subcutaneous fat was quantitated measuring mtDNA and gDNA titration curves of pMito (Figure 1A). The mean mtDNA content and standard deviation were 1057 737 copies/cell in the HIV seronegative group (n=7), 280 110 copies/cell in the antiretroviral naive cohort (n=5), 219 355 copies/cell in the non-lipoatrophic cohort (n=7) and 59 64 copies/cell in the lipoatrophic cohort (n=10). Examples of low and high mtDNA copies/cell are demonstrated in Figure 1B and C, respectively. A statistically significant 94% decrease in mtDNA was found in the HIV-1 seropositive lipoatrophic cohort compared to the seronegative cohort (P=0.001). Similarly, real-time PCR comparisons were also comparable (Table 2) in the abdomen (although, the two assays were not comparable by linear regression analyses). MtDNA copies/cell in abdomen, thigh, and neck fat were decreased in the HIV-1 seropositive lipoatrophic cohort compared to the HIV-1 seropositive nave and seronegative cohorts (P0.05) by realtime PCR. Also, thigh and neck HIV-1 seropositive non-lipoatrophic mtDNA copies/cell were decreased compared to HIV-1 seropositive naive and the seronegative controls (P0.05). These mtDNA decreases varied in the HIV-1 seropositive lipoatrophics from 40% in the thigh, 70% abdomen and 80% neck compared to the seronegative controls. There were no statistically significant differences in mtDNA copies/cell in the fat depots between the lipoatrophic and non-lipoatrophic seropositive patients or the seronegative and naive cohorts. NRTI exposure time between the lipoatrophic cohort (90 39) and non-lipoatrophic (60 50) was not statistically different [12]. Additive NRTI exposure time between the lipoatrophic cohort (101 40) and
M86
B
PT 108 107 106 105 104 103 102 101 100 + N mtDNA
gDNA
C
PT 108 107 106 105 104 103 102 101 100 N
mtDNA
gDNA
(A) Gel of standard PCR products from pMITO plasmid using mtDNA and gDNA primers. Dilution of the pMITO ranged from 108 to 100 copies and included negative (N) control with no plasmid. (B) Example of low copy (1 mtDNA copy/cell). Gel of competitive PCR products from pMITO plasmid diluted 108 to 100 copies competing with an equal amount of unknown DNA (Pt) using mtDNA and gDNA primers. Positive control (+, pMITO) and negative (N) control with no plasmid are included. : Denotes the equivalence point: (5107/108)2=1 mtDNA copy/cell. (C) Example of high copy (400 mtDNA copies/cell). Gel of competitive PCR products from pMITO plasmid diluted 108 to 100 copies competing with an equal amount of unknown DNA (Pt) using mtDNA and gDNA primers. Positive control (+, pMITO) and negative (N) control with no plasmid are included. : Denotes the equivalence point: (108/5105)2=400 mtDNA copies/cell. mtDNA, mitochondrial DNA; gDNA, genomic DNA
Table 2. MtDNA copies/cell in subcutaneous fat and PBMCs

HIV(-) Thigh 435 63 n=4 Abdomen 790 292 n=5 Neck 976 292 n=5 PBMC 201 62 n=10 105 48 n=3 157 49 n=7 148 53 n=7 676 271 n=5 396 249* n=6* 205 78* n=7* 545 190 n=5 335 158* n=6* 244 148* n=7* 489 100 n=5 267 136* n=6* 255 124* n=6* HIV (+) naive HIV (+) no lipoatrophy HIV (+) lipoatrophy
All values are representative as mtDNA copies/cell ( SD) and statistical significance are described. *Statistical significance between HIV (-) compared to either HIV(+) no lipoatrophy or lipoatrophy. Thigh HIV (+) no lipoatrophy and lipoatrophy were compared to HIV (-), P=0.05 and P=0.04 respectively. Thigh HIV (+) no lipoatrophy and lipoatrophy were compared to HIV (+) naive, P=0.01 and P=0.008. Abdomen HIV (+) no lipoatrophy and lipoatrophy were compared to HIV (-), P=0.003 and P=0.001 respectively. Abdomen HIV (+) no lipoatrophy was significant compared to HIV (-), P=0.02. Neck HIV (+) no lipoatrophy and lipoatrophy are compared to HIV (-), P=0.001 and P=0.002 respectively. Thigh HIV (+) no lipoatrophy and lipoatrophy are compared to HIV (+) naive, P=0.001 and P=0.002. Statistical significance between the HIV (-) thigh in comparison to abdomen (P=0.05) or neck (P=0.05).
cohort. The thigh fat had 4555% less mtDNA copies/cell than the abdomen and neck. Additionally, mtDNA copies/cell were evaluated in PBMCs from these cohorts and ranged from 105263 copies/cell. No statistically significant differences in PBMC values were observed among the cohorts and there was no correlation with lipoatrophy state or HIV-1 serostatus (data not shown).
Conclusion
This study was designed to measure quantitatively mtDNA copies/cell from PBMCs and different subcutaneous fat depots in a previously described cohort [12]. Our previous study measured mtDNA copies/cell in the fat tissue using semi-quantitative PCR by amplifying three different mtDNA fragments and analysing them by visual interpretation of ethidium-stained gels. The conclusion of that study showed a decrease in mtDNA content in HAART treated HIV-1 infected patients with peripheral fat wasting in comparison with subjects in the control cohorts. The findings in this current protocol confirm our previous study and show that significant mtDNA depletion is evident regardless of which PCR method was used. We observed significant differences by real-time PCR in mtDNA copies/cell in the subcutaneous adipose tissue from thigh, abdomen, and neck of seronegative subjects. Additionally,
mtDNA depletion was observed in thigh and neck fat from both lipoatrophic and non-lipoatrophic cohorts on NRTIs compared with NRTI naive controls. In contrast, we found that Ficoll-isolated PBMC mtDNA copies/cell did not vary significantly among the cohorts. This lack of differences may be due to the small number of samples available particularly in the HIV (+) naive group, n=3. Thus in our study, PBMC mtDNA copies/cell did not correspond with HIV status, current NRTI exposure, or the presence of selfreported lipoatrophy. Consistent with our findings, other investigators have found a substantial decrease in mtDNA content in fat tissue in association with HIV-1 lipoatrophy [13,15,19]. This mtDNA depletion has been observed in vitro with different NRTIs affecting DNA polymerase [9] and is more profound in subjects on stavudine compared to zidovudine [15,19,20]. More recently, the partial reversal of lipoatrophy following switch substitution of thymidine analogue NRTIs to a more mitochondrial friendly non-thymidine analogue NRTI regimen or to a completely NRTI-sparing regimen also supports the hypothesis that NRTIs and thymidine analogues in particular have a role in the development of lipoatrophy [15,21]. In our study, both the non-lipoatrophic (five of the seven patients) and lipoatrophic (four of the seven patients) cohorts had been treated with stavudine and mtDNA depletion was
M87
observed in both cohorts, adipose tissues, but we did not observe mtDNA depletion in PBMCs. Other studies with HIV-1 lipoatrophic cohorts have confirmed these findings [22]. One of the most interesting findings from this study is the difference in mtDNA copies/cell in the HIV (-) thigh, neck, and abdomen, with mtDNA copies/cell being lowest in the thigh. These results suggest there may be mitochondrial metabolic differences in different fat depots. Studies comparing visceral and subcutaneous fat in humans have shown gene expression and biochemical differences. Gene expression profiles of subcutaneous and visceral adipose tissue from 10 nondiabetic, normolipidemic obese men showed increased expression in subcutaneous fat of genes in the Wnt signalling pathway, CEPBA, HOX, lipolytic stimuli, and cytokine secretion [23]. Furthermore, uncoupling protein 2 mRNA expression has been shown to be increased in the omentum vs subcutaneous adipocytes from patients undergoing elective intra-abdominal surgery [24]. Tumour necrosis alpha levels have been shown to be lower in visceral fat in women compared to subcutaneous adipose tissue in patients undergoing elective intra-abdominal surgery [25]. All these pathways, genes, and proteins affect mitochondrial function. For instance, uncoupling proteins are mitochondrial carrier proteins that catalyze a regulated proton leak across the inner mitochondrial membrane, diverting free energy from ATP synthesis by the mitochondrial F0F1-ATP synthase to the production of heat [26]. Therefore, further investigations are warranted in understanding the differences in subcutaneous fat depots and their roles in HIV lipodystrophy.
dations of an International AIDS Society-USA panel. Journal of Acquired Immune Deficiency Syndromes 2002; 31:257275. 3. Saint-Marc T, Partisani M, Poizot-Martin I, Bruno F, Rouviere O, Lang JM, Gastaut JA & Touraine JL. A syndrome of peripheral fat wasting (lipodystrophy) in patients receiving long-term nucleoside analogue therapy. AIDS 1999; 13:16591667. Mallal SA, John M, Moore CB, James IR & McKinnon EJ. Contribution of nucleoside analogue reverse transcriptase inhibitors to subcutaneous fat wasting in patients with HIV infection. AIDS 2000; 14:13091316. Heath KV, Hogg RS, Chan KJ, Harris M, Montessori V, O'Shaughnessy MV & Montanera JS. Lipodystrophy-associated morphological, cholesterol and triglyceride abnormalities in a population-based HIV/AIDS treatment database. AIDS 2001; 15:231239. Lichtenstein KA, Ward DJ, Moorman AC, Delaney KM, Young B, Palella FJ Jr, Rhodes PH, Wood KC & Holmberg SD; HIV Outpatient Study Investigators. Clinical assessment of HIV-associated lipodystrophy in an ambulatory population. AIDS 2001; 15:13891398. Galli M, Ridolfo AL, Adorni F, Gervasoni C, Ravasio L, Corsico L, Gianelli E, Piazza M, Vaccarezza M, dArminio Monforte A & Moroni M. Body habitus changes and metabolic alterations in protease inhibitor-naive HIV-1-infected patients treated with two nucleoside reverse transcriptase inhibitors. Journal of Acquired Immune Deficiency Syndromes 2002; 29:2131. Kopecky J, Rossmeisl M, Flachs P, Bardova K & Brauner P. Mitochondrial uncoupling and lipid metabolism in adipocytes. Biochemical Society Transactions 2001; 29:791797. Dagan T, Sable C, Bray J & Gerschenson M. Mitochondrial dysfunction and antiretroviral nucleoside analog toxicities: what is the evidence. Mitochondrion 2002; 1:397412.
4.
5.
6.
7.
8.
9.
10. Lewis W, Day BJ & Copeland WC. Mitochondrial toxicity of NRTI antiviral drugs: an integrated cellular perspective. Nature Reviews. Drug Discovery 2003; 2:812822. 11. Gerschenson M & Brinkman K. Mitochondrial Dysfunction in AIDS and its treatment. Mitochondrion 2004; 4:763767. 12. Shikuma CM, Hu N, Milne C, Yost F, Waslien C, Shimizu S & Shiramizu B. Mitochondrial DNA decrease in subcutaneous adipose tissue of HIV-infected individuals with peripheral lipoatrophy. AIDS 2001; 15:18011809. 13. Walker UA, Bickel M, Lutke Volksbeck SI, Ketelsen UP, Schofer H, Setzer B, Venhoff N, Rickerts V & Staszewski S. Evidence of nucleoside analogue reverse transcriptase inhibitor associated genetic and structural defects of mitochondria in adipose tissue of HIV-infected patients. Journal of Acquired Immune Deficiency Syndromes 2002; 29:117121. 14. Cherry CL, McArthur JC, Hoy JF & Wesselingh SL. Nucleoside analogues and neuropathy in the era of HAART. Journal of Clinical Virology 2003; 26:195207. 15. McComsey GA, Paulsen DM, Lonergan JT, Hessenthaler SM, Hoppel CL, Williams VC, Fisher RL, Cherry CL, White-Owen C, Thompson KA, Ross ST, Hernandez JE & Ross LL. Improvements in lipoatrophy, mitochondrial DNA levels and fat apoptosis after replacing stavudine with abacavir or zidovudine. AIDS 2005; 19:1523. 16. Shoffner JM. Oxidative phosphorylation disease diagnosis. Seminars in Neurology 1999; 19:341351. 17. Pinti M, Pedrazzi J, Benatti F, Sorrentino V, Nuzzo C, Cavazzuti V, Biswas P, Petrusca DN, Mussini C, De Rienzo B & Cossarizza A. Differential down-regulation of CD95 or CD95L in chronically HIV-infected cells of monocytic or lymphocytic origin: cellular studies and molecular analysis by quantitative competitive RT-PCR. FEBS Letters 1999; 458:209214. 18. Shiramizu B, Shikuma KM, Kamemoto L, Gerschenson M, Erdem G, Pinti M, Cossarizza A & Shikuma C. Brief report:
Acknowledgements
We would like to thank Duy Tran and Elizabeth Westgard for their technical assistance, Dr Christopher J OCallaghan, Queens University, Kingston, Ontario, Canada for his statistical analyses and Drs Andrea Cossarizza and Marcello Pinti, University of Modena and Reggio Emilia School of Medicine, Modena, Italy for the plasmids and methodologies for quantitating mtDNA. Funding for this research was provided by the National Institutes of Health, USA, grant numbers: AI34853 and MD000173.
References
1. Shevitz A, Wanke CA, Falutz J & Kotler DP. Clinical perspectives on HIV-associated lipodystrophy syndrome: an update. AIDS 2001; 15:19171930. Schambelan M, Benson CA, Carr A, Currier JS, Dube MP, Gerber JG, Grinspoon SK, Grunfeld C, Kotler DP, Mulligan K, Powderly WG & Saag MS; International AIDS SocietyUSA. Management of metabolic complications associated with antiretroviral therapy for HIV-1 infection: recommen-
2.
M88
placenta and cord blood mitochondrial DNA toxicity in HIV-infected women receiving nucleoside reverse transcriptase inhibitors during pregnancy. Journal of Acquired Immune Deficiency Syndromes 2003; 32:370374. 19. Nolan D, Hammond E, Martin A, Taylor L, Herrmann S, McKinnon E, Metcalf C, Latham B & Mallal S. Mitochondrial DNA depletion and morphologic changes in adipocytes associated with nucleoside reverse transcriptase inhibitor therapy. AIDS 2003; 17:13291338. 20. Pace CS, Martin AM, Hammond EL, Mamotte CD, Nolan DA & Mallal SA. Mitochondrial proliferation, DNA depletion and adipocyte differentiation in subcutaneous adipose tissue of HIV+, HAART recipients. Antiviral Therapy 2003; 8:323331. 21. Carr A, Workman C, Smith DE, Hoy J, Hudson J, Doong N, Martin A, Amin J, Freund J, Law M & Cooper DA; Mitochondrial Toxicity (MITOX) Study Group. Abacavir substitution for nucleoside analogs in patients with HIV lipoatrophy: a randomized trial. Journal of the American Medical Association 2002; 288:207215. 22. Miura T, Goto M, Hosoya N, Odawara T, Kitamura Y, Nakamura T & Iwamoto A. Depletion of mitochondrial
DNA in HIV-1-infected patients and its amelioration by antiretroviral therapy. Journal of Medical Virology 2003; 70:497505. 23. Vohl MC, Sladek R, Robitaille J, Gurd S, Marceau P, Richard D, Hudson TJ & Tchernof A. A survey of genes differentially expressed in subcutaneous and visceral adipose tissue in men. Obesity Research 2004; 12:12171222. 24. Digby JE, Crowley VE, Sewter CP, Whitehead JP, Prins JB & ORahilly S. Depot-related and thiazolidinedione-responsive expression of uncoupling protein 2 (UCP2) in human adipocytes. International Journal of Obesity & Related Metabolic Disorders 2000; 24:585592. 25. Orel M, Lichnovska R, Gwozdziewiczova S, Zlamalova N, Klementa I, Merkunova A & Hrebicek J. Gender differences in tumor necrosis factor alpha and leptin secretion from subcutaneous and visceral fat tissue. Physiological Research 2004; 53:501505. 26. Lowell BB & Spiegelman BM. Towards a molecular understanding of adaptive thermogenesis. Nature 2000; 404:652660.
Received 21 January 2005, accepted 27 June 2005
M89
Altered mitochondrial RNA production in adipocytes from HIV-infected individuals with lipodystrophy
Lorenzo Galluzzi 1, Marcello Pinti 1, Giovanni Guaraldi 2, Cristina Mussini 2, Leonarda Troiano1, Erika Roat 1, Chiara Giovenzana1, Elisa Nemes1, Milena Nasi1, Gabriella Orlando 2 , Paolo Salomoni 3 and Andrea Cossarizza1*
1 2
Chair of Immunology, Department of Biomedical Sciences, University of Modena and Reggio Emilia, Modena, Italy Clinic of Infectious Diseases, University of Modena and Reggio Emilia and Azienda Policlinico, Modena, Italy 3 Laboratory of Genetic Instability II, MRC Toxicology Unit, University of Leicester, Leicester, UK *Corresponding author: Tel: +39 059 2055 415; Fax: +39 059 2055 426; E-mail: cossarizza.andrea@unimore.it L Galluzzi and M Pinti contributed equally to this work.
Background: Damage to mitochondria (mt) is a major side effect of highly active antiretroviral therapy (HAART) that includes a nucleoside reverse transcriptase inhibitor (NRTI). Such damage is associated with the onset of lipodystrophy in HAART-treated HIV+ patients. To further investigate mt changes during this syndrome, we analysed the expression of mtRNA in adipocytes from lipodystrophic HIV+ patients taking NRTI-containing HAART and compared it with similar cells from healthy individuals. Materials and methods: Total RNA was extracted from adipocytes collected from different anatomical locations of 11 HIV+ lipodystrophic patients and seven healthy
control individuals. RNA was reverse transcribed and Taqman-based real-time PCR was used to quantify three different mt transcripts (ND1, CYTB and ND6 gene products). mtRNA content was normalized versus the housekeeping transcript L13. Results: ND1, CYTB and ND6 expression was significantly reduced in HIV+ lipodystrophic patients. HIV+ men and women did not differ in a statistically significant way regarding the levels of ND1 and ND6, whereas the opposite occurred for CYTB. Conclusions: Lipodystrophy following treatment with NRTI-containing HAART is associated with a decrease in adipose tissue mtRNAs.
Introduction
It has been proved that lipodystrophy (central obesity and peripheral lipoatrophy), together with hyperlipidaemia and insulin resistance, is associated with highly active antiretroviral therapy (HAART) [1,2]. Although the first studies focused on the role of protease inhibitors (PIs), subsequent analyses revealed that nucleoside analogue reverse transcriptase inhibitors (NRTIs) are equally, if not more, involved [35]. Indeed, it has been suggested that NRTI-induced mitochondrial (mt) toxicity may be one cause of HAARTassociated lipodystrophy. In particular, as evidenced by several clinical studies, these drugs are the main cause of the onset of lipoatrophic forms: a strong association exists between NRTI-containing therapeutic regimens and either lipoatrophy or markers of mt toxicity such as lactic acidaemia, hepatic impairment and changes in mtDNA content ([612]; reviewed in [13]). Mitochondrial activity could theoretically be impaired even in the absence of systemic abnormalities that give clinical signs, or when the mtDNA content of cells is still unaffected, at least in the case of cell lines in vitro exposed to NRTIs [14]. Thus, it could be of interest to look for other markers of mt damage, such as the expression of mtRNAs. mtRNAs are synthesized, completely processed and translated within the organelle by a dedicated machinery made both of nuclear-encoded components (for example, mtRNA polymerase) and of mt-encoded ones (tRNAs) [15,16]. Human mtDNA is a doublestranded, circular molecule of 16 571 base pairs encoding 13 proteins, 22 tRNAs and two rRNAs. It is transcribed as two large polycistronic units from both template DNA strands (H or heavy and L or light); an endonuclease processes these large molecules and gives origin to all functional mtRNA species [1719]. To conveniently analyse the transcription of the whole mt genome, we chose to quantify three different mt transcripts: ND1, encoding for subunit 1 of the NADH-dehydrogenase complex and lying at the 5 end of the H strand; CYTB, encoding for cytochrome B and lying at 3 end of the H strand; and ND6, encoding for subunit 6 of the NADH-dehydrogenase complex,
M91
L Galluzzi et al.
and being the only mRNA encoded by the L strand (5 end). This study was designed to assess mtRNA content in the adipocytes from seropositive individuals who have developed lipodystrophy following NRTIcontaining HAART and to compare the results with those from seronegative subjects.

Studied subjects
The present study was performed on samples among HIV+ patients with lipodystrophy selected from the patients of the Metabolic Clinic of the Azienda Policlinico di Modena Hospital and from healthy controls undergoing minor plastic surgery. The HIV+ subjects, who gave informed consent for the analysis reported in the present study, underwent surgical procedures for the amelioration of the physical changes caused by lipodystrophy. Fat samples were obtained from different subcutaneous body areas (buffalo hump, back, breast and abdomen) by mechanical liposuction, performed in some cases to obtain a fat graft for autologous fat transplant (in the case of HIV+ patients, to correct facial lipoatrophy) or for aesthetic purposes (healthy controls). The lipodystrophic HIV+ group was formed by six men (mean age SD: 46.3 8.9 years) and five women (mean age 41.4 10.0). All these patients had a long history of exposure to NRTIs, including D-drugs [stavudine (d4T) and didanosine (ddI)], and were currently taking HAART including one or more NRTI. The control HIV group was composed of one man (aged 63) and six women (mean age 52.3 9.8).
RNA extraction and reverse transcription

Total RNA was extracted from adipocytes of study subjects by means of RNeasy Lipid Tissue Mini Kit from Qiagen (Hilden, Germany). Frozen bioptic samples of subcutaneous adipose tissue from different anatomical locations were the source of adipocytes from both HIV-infected patients and healthy control individuals. Represented locations included breast, abdomen, buffalo hump (from HIV-infected patients only) and lower limbs. Subsequently, 1 g of the extracted RNA was subjected to a standard randomprimed reverse transcription reaction [20,21].
Construction of the standard plasmid for mtRNA quantification

The amplicons of interest (L13, ND1, ND6 and CYTB, whose sequences are reported in Table 1) were cloned together in a plasmid in order to prepare a single, reference DNA molecule that could be used as the standard in real-time PCR quantification of all four transcripts (patent PCT/NL03/00545). The use of
M92
a single, cleaved plasmid was preferred to guarantee the ratio among amplicons to be fixed at the value of 1:1:1:1. As described below, the plasmid was constructed using classical molecular biology techniques [20], and its sequence checked by an automatic DNA sequencer (Abi Prism 310 Genetic Analyzer; Applied Biosystems, Foster City, CA, USA). A human cDNA sample, obtained from the random-primed reverse transcription of total RNA from peripheral blood mononuclear cells, was used as a template to amplify the four amplicons of interest by PCR (primers are reported in Table 1). Each primer was designed to contain convenient linker regions useful for the subsequent cloning steps. Tails to be added to primers in order to produce linkers were selected from the genome of an evolutionarily distant organism (Cryptococcus neoformans) and blasted against the human genome to check for absence of homology. This was necessary to avoid unspecific annealing between the tailed primers and human cDNAs, which would have lead to unspecific PCR products. All primers, as well as probes used in the quantification reactions, were purchased from Qiagen Operon (Hilden, Germany). Primers SP1/SP2 created L13 amplicon (72 bp) with a PstI restriction site immediately upstream and a linker (L1, 20 bp) immediately downstream; primers SP7/SP8 lead to ND6 amplicon (109 bp) with a different linker (L2, 21 bp) immediately upstream and a HindIII restriction site immediately downstream. Primers SP3/SP4 were used to amplify CYTB amplicon (107 bp) with L1 immediately upstream and a false linker introducing a SacI restriction site immediately downstream. This false linker contains the SacI restriction site and part of the sequence of ND1 amplicon (78 bp), which in turn was obtained with primers SP5/SP6, and serves only construction purposes. ND1 amplicon had L2 immediately downstream and a SacI restriction site immediately upstream, in the context of the false linker (in this case part of CYTB amplicon). These four PCR products were used together as templates in a single PCR reaction to obtain the full insert of interest. A small amount of each amplicon was used in a multitemplate reaction (based upon an original strategy) as follows. The starting reaction mixture included <0.5 l of the products of each described single reaction (templates); Mg2+-free buffer 1 (Promega, Madison, WI, USA); MgCl2 1.5 mM, Taq polymerase 2 U (Promega), dNTPs mix 0.2 mM (Fermentas, Vilnius, Lithuania) and water up to 49.4 l (total starting volume). No primers were present at this stage. The mixture then underwent the following thermal protocol, to fill in the gaps between templates partially annealed because of the complementary tails: starting denaturation (95C, 5 min), two main cycles
Table 1. Primers, probes and amplicons

Length, bp Source
Name
53 Sequence/labels
Amplicons 72 78 107 109 [21] Present work Present work Present work
A1L13 A2ND1 A3CYTB A4ND6
5-GCTGGAAGTACCAGGCAGTGACAGCCACCCTGGAGGAGAAGAGGAAAGAGAAAGCCAAGATCCACTACCGGT-3 5-CCTTCGCTGACGCCATAAAACTCTTCACCAAAGAGCCCCTAAAACCCGCCACATCTACCATCACCCTCTACATCACCG-3 5-AGTCCCACCCTCACACGATTCTTTACCTTTCACTTCATCTTACCCTTCATTATTGCAGCCCTAGCAGCACTCCACCTCCTATTCTTGCACGAAACGGGATCAAACAA-3 5-AACCCTGACCCCTCTCCTTCATAAATTATTCAGCTTCCTACACTATTAAAGTTTACCACAACCACCACCCCATCATACTCTTTCACCCACAGCACCAATCCTACCTCCA-3
Standard plasmid construction primers tailed primers* (Qiagen Operon) 30 41 40 40 39 41 41 28 Present work Present work Present work Present work Present work Present work Present work Present work
SP1L13 dir/PstI SP2L13 rev/L1 SP3CYTB dir/L1c SP4CYTB rev/SacI SP5ND1 dir/SacI SP6D1 rev/L2 SP7ND6 dir/L2c SP8ND6 rev/HindIII
5-AAACTGCAGGCTGGAAGTACCAGGCAGTGA-3 5-TCCAGACACCCAACAACGTCACCGGTAGTGGATCTTGGCTT-3 5-GACGTTGTTGGGTGTCTGGAAGTCCCACCCTCACACGATT-3 5-GGCGTCAGCGAAGGGAGCTCTTGTTTGATCCCGTTTCGTG-3 5-ACGGGATCAAACAAGAGCTCCCTTCGCTGACGCCATAAA-3 5-GACAACACTCCTCGGCACCTTCGGTGATGTAGAGGGTGATG-3 5-AAGGTGCCGAGGAGTGTTGTCAACCCTGACCCCTCTCCTTC-3 5-CCAAGCTTTGGAGGTAGGATTGGTGCTG-3
Real-time PCR primers (Qiagen Operon) 21 21 19 20 20 20 20 20 [21] [21] Present work Present work Present work Present work Present work Present work
P1L13 dir P2L13 rev P3ND1 dir P4ND1 rev P5CYTB dir P6CYTB rev P7ND6 dir P8ND6 rev
5-GCTGGAAGTACCAGGCAGTGA-3 5-ACCGGTAGTGGATCTTGGCTT-3 5-CCTTCGCTGACGCCATAAA-3 5-CGGTGATGTAGAGGGTGATG-3 5-AGTCCCACCCTCACACGATT-3 5-TTGTTTGATCCCGTTTCGTG-3 5-AACCCTGACCCCTCTCCTTC-3 5-TGGAGGTAGGATTGGTGCTG-3
Probes (Qiagen Operon) 27 23 27 23 [21] Present work Present work Present work
S1L13 S2ND1 S3CYTB S4ND6
5-TCTTTCCTCTTCTCCTCCAGGGTGGCT-3 Texas Red BHQ1 5-CCAAAGAGCCCCTAAAACCCGCC-3 FAM490 BHQ1 5-CAGCCCTAGCAGCACTCCACCTC-3 Texas Red BHQ1 5-CCACCACCCCATCATACTCTTTCACCC-3 FAM490 BHQ1
mtRNA in adipose tissue
*Tail nucleotides are reported in italics and inserted restriction sites are underlined. Gene bank accession numbers: NM_012423 for L13, AAO88540 for ND1, AY339474 (region: 1474715887) for CYTB; AAO88837 for ND6. dir, direct; rev, reverse.
M93
L Galluzzi et al.
(94C, 15 sec; 63C, 30 sec; 72C, 30 sec) and two additional main cycles (94C, 15 sec; 72C, 1 min). After this, one additional unit of Taq polymerase and primers SP1 and SP8 were added up to a final concentration of 0.4 M (final volume 50 l). The thermal protocol was then continued as follows: two main cycles (94C, 15 sec; 64C, 30 sec; 72C, 30 sec), 35 additional main cycles (94C, 15 sec; 70C, 30 sec; 72C, 30 sec) and final elongation (72C, 7 min). Electrophoresis on 3% agarose gel of the products confirmed the amplification of a single fragment of 430 bp (see Figure 1, which also shows the localization of the genes we studied on the mtDNA molecule). The 430 bp insert was then cloned as a PstI-HindIII fragment into the vector pQE-81L (Qiagen). The deriving plasmid (pQE-81L-mtRNA), extracted from recipient Escherichia coli JM109 cells with QIAprep Spin Miniprep Kit (Qiagen), was digested with SacI to
obtain the final standard mixture, composed of two equimolar linear fragments of 4937 and 227 bp, respectively. The digestion pattern is due to the presence of two SacI restriction sites: the first one within the insert and the second one as part of the polylinker of the parental vector pQE-81L. The products were then quantified by comparing their intensity after 1% agarose gel electrophoresis with a molecular weight marker of known concentration. The software Quantity One (Bio-Rad, Hercules, CA, USA) was used for this analysis. The digested plasmid was used in 10fold dilutions for the construction of the standard curve for the real time PCR quantification.
Real-time PCR
Each different transcript was quantified in a separate reaction to avoid non-specific amplifications and to ensure the maximal reliability of the study. Each reaction entailed the following reagents: 1 l of cDNA solution (sample); PCR Mg2+-free buffer 1 (Promega); MgCl2 1.5 mM; Taq polymerase 2 U (Promega); dNTPs mix 0.2 mM (Fermentas), direct and reverse primers 0.4 M for mt transcripts, 0.2 M for L13 (primers vary according to the measured RNA, see Table 1); Taqman probe 0.2 M (probes vary according to the measured RNA, see Table 1); and water up to 50 l. Reactions were performed in PCR multiwell plates (Bioplastics, North Ridgeville, OH, USA). iCycler (BioRad) was the real-time thermal cycler of choice and was set as follows: starting denaturation (95C, 5 min) followed by 45 cycles (94C, 15 sec; 60C, 30 sec; fluorescence reading). The reading wavelength depended on the probe in use: FAM490-labelled probes are detected at 520 nm, while those labelled with Texas Red are detected at 615 nm. Baseline cycles, threshold and threshold cycles (Ct) are calculated automatically by selecting the maximum correlation coefficient approach in the cycler software.
Figure 1. mtDNA map and insert for mtRNA standard plasmid
A
CYTB ND6 ND5 ND1 ND4 ND4L ND3 COX3 ATP6 ATP8 COX2 D-loop 12S rRNA 16S rRNA
ND2
COX1
B
L13 CYTB SP1 SP3 SP5 SP7 ND1 ND6 HindIII
Data handling and statistical analysis

Each real-time PCR quantification reaction, including those run on the standard mixture at different dilutions, was performed in triplicate. The iCycler software, at the end of the run, automatically compares the mean Ct values with the created standard curve and gives results in terms of absolute copies and standard deviation for each group of triplicate samples. Nevertheless, absolute cDNA copies are not a suitable parameter for analysis. Indeed, to have interpretable data, it is recommended that the mtRNA content of each sample is normalized to its content of a housekeeping transcript. We chose L13, which encodes a protein belonging to the large ribosomal subunit and is widely considered as a housekeeping gene [22], whose
PstI
SP2 Sacl
SP4
SP6
SP8
(A) Map of mtDNA that evidences the position of the genes chosen for the analysis of mtRNA production (in grey) and the sense of their transcription. Black lines indicate genes for tRNAs. Note that ND1 and CYTB genes are encoded by the H strand (clockwise arrows), while ND6 gene by the L strand (anticlockwise arrow). (B) Grey boxes represent the four amplicons involved in the quantification reactions. Arrows depict the annealing positions of the primers (indicated as SP) used for the construction of the molecule. Black lines show relevant restriction sites (enzymes used are in italics).
M94
transcription is, in practice, constant in all functional moments of the cell. The normalization to a housekeeping RNA avoids further mathematical analyses that would otherwise be necessary to consider the different efficiency of RNA isolation and reverse transcription in different samples. Thus, in this study the results are shown as copies of each mt transcript per copy of L13. Studied subjects were grouped into two main categories according to their HIV seropositivity. Infected individuals were subsequently divided according to sex. This was not done for healthy controls because of the large preponderance of women. Groups were compared for the expression of mtRNAs in the following ways: HIV+ patients versus HIV controls, HIV+ men versus HIV+ women and HIV+ women versus HIV women (for CYTB only). Statistical significance was determined by using two-tailed nonparametric test (MannWhitney test), with confidence intervals of 95%.
Figure 2. ND1 expression in studied subjects
200
p < 0.05
p = ns 100
0 Controls HIV+ HIV+ Men HIV+ Women
Results are reported as ND1 copies/L13 copies in terms of mean value and standard error. Statistical significance is indicated by P value.
Results
Mitochondrial transcription in healthy adipocytes
Firstly we analysed the adipocyte mtRNAs from healthy subjects and found that the transcription levels of the three genes under investigation are different. Indeed, ND1 is expressed at much lower levels (130.5 copies/ L13 copies, mean value) than the other two transcripts (573.1 and 1043.0 copies/L13 copies, mean values for ND6 and CYTB, respectively). No significant differences were present among fat samples collected in different sites.
mtRNA depletion in HIV+ patients

The quantification of mtRNAs in samples obtained from HIV+ individuals revealed a dramatic decrease of their expression. This decrease was similar in samples collected from different sites. In cells from HIV+ men, ND1 had a concentration of 65.6 copies/L13 copies, corresponding to a 50% reduction with respect to the controls. We also demonstrated a similar fall for HIV+ women (52.5 copies/L13 copies corresponding to a 60% reduction). The comparison between the control group and all HIV+ patients was statistically significant (P=0.0185), but HIV+ men and women did not significantly differ (P=0.4762). The data for ND1 are summarized in Figure 2. In the adipose tissue of HIV+ men, ND6 was present at the concentration of 183.5 copies/L13 copies (68% reduction from the controls). In samples from HIV+ women the expression of ND6 was even lower (107.8 copies/L13 copies an 81% reduction). Again, the MannWhitney test witnessed a statistically significant difference between the control group and HIV+
patients (P=0.0021), whereas HIV+ men and women did not differ (P=0.3290) (Figure 3). The concentration of CYTB transcripts in adipocytes from HIV+ men was 359.0 copies/L13 copies (66% reduction vs controls). mtRNA levels in adipose tissue from HIV+ women were much lower (63.9 copies/L13 copies, which corresponds to a 94% reduction vs controls). As a consequence, statistical analysis revealed a significant difference not only between the control group and all HIV+ patients (P=0.0104, similar to ND1 and ND6), but also between HIV+ men and women (P=0.0303). Finally, we compared CYTB expression in HIV+ and HIV women (who were represented by six individuals, aged 52.3 9.8), and found a statistically significant difference (P=0.0079) (Figure 4).
Figure 3. ND6 expression in studied subjects

p < 0.01 750 ND6 copies/ L13 copies
500 p = ns 250
Results are reported as ND6 copies/L13 copies in terms of mean value and standard error. Statistical significance is indicated by P value.
M95
L Galluzzi et al.
Figure 4. CYTB expression in studied subjects

CYTB copies/ L13 copies
1500
p < 0.01
p < 0.05 1500 CYTB copies/ L13 copies
500
Controls Women
HIV+ Women
1000
p = 0.05
500
Results are reported as CYTB copies/L13 copies in terms of mean value and standard error. Statistical significance is indicated by P value. The insert represents the comparison between control women and HIV+ women.
Discussion
Considerable attention has been paid to the role of mitochondria in the pathogenesis of lipodystrophy, and several authors have demonstrated a significant, dramatic depletion of mtDNA content in adipose tissue of lipodystrophic HIV+ patients compared with controls [2326]. The determinant role of D-drugs, especially d4T, in causing lipoatrophy has been well defined by several sets of clinical data. The functional consequences of mtDNA depletion are complex. It is known from classic studies on genetic disorders affecting the mt genome that an mtDNA decrease of the order of 80% is necessary to evidence functional alterations. However, in treated HIV+ patients, even smaller decreases in mtDNA content are significantly associated with relevant alterations in the distribution of fat tissue [3,11,27,28]. Depletion of adipocyte mtDNA positively correlates with evidence of tissue, cell and mt toxicity, including decreased expression of mt proteins, increased adipocyte pleiomorphism, infiltration of macrophages and non-adipocyte inflammatory cells, and increased apoptosis [29]. Mitochondrial damages are not irreversible, since improvements in lipoatrophy, mtDNA content and fat tissue apoptosis can be obtained after replacing a constituent drug with a less toxic one [3032]; damage can also be reduced by regimen alternations [33]. At the functional level, studies on frozen adipocyte biopsies from HIV+ individuals demonstrated that NRTI-induced mtDNA depletion can result in a decreased enzymatic activity of the respiratory chain complex IV [34]. It is not known, at least in fat cells,
M96
what relationship exists between changes in mtDNA content and changes in the number of mitochondria. Nevertheless, it has been shown that mtDNA depletion is associated with an increased organelle mass, suggesting that this may represent a compensatory response to decreased mtDNA [25]. The mechanisms that regulate the replication of mtDNA and the relationship among mtDNA levels, mtRNA levels and the levels of the encoded proteins are not completely clear. On one hand, it could be hypothesized that the reduction in mtDNA leads to a parallel decrease in its transcription and translation; on the other hand, in cell cultures, mitochondria could compensate mtDNA loss with a concomitant increase in its transcriptional rate [35]. Clinical data showed that the treatment with d4T plus ddI was associated with decreased mt mass and mtDNA content, but that the expression and activity of mtDNA-codified enzymes such as COXII remained unaltered [36]. This suggests that up-regulatory transcriptional or posttranscriptional mechanisms could compensate for mtDNA depletion before profound mtDNA depletion occurs. In cultured mouse muscle cells treated with zidovudine (AZT), the disruption of mt cristae occurred after 4 weeks of treatment and was not accompanied by changes in mtDNA content [37]. Cytochrome b and cytochrome c oxidase I mRNA significantly increased (64% and 31%, respectively), while no changes were present as far as mtDNA was concerned. This suggests that AZT can provoke complex alterations in the production of nucleic acids, that can affect also mtRNA transcription. In patients, it was reported that dual NRTI therapy decreases mtRNA production after 2 weeks of therapy, before changes in fat mass or changes in metabolic parameters [38]. There are very few data, in addition to ours, about the quantification of mtRNAs in human cells [35,3840]. The analysis of mtRNA levels, besides those on mtDNA, probably provides additional information on the consequences of mtDNA depletion. This approach could also help to clarify if and how this depletion effectively leads to a decline in terms of functionality of the mt machinery. The real-time PCR technique that we describe in this paper measures the concentration of three mt transcripts encoded at different positions in the mt genome. We selected these mtRNAs in order to conveniently represent the whole transcriptional process from the mtDNA molecule. It is indeed known that the mt genome is transcribed from both mtDNA strands into two large polycistronic units, which are cleaved by specific endonucleases to produce the mature mtRNAs [18,19]. According to this model, an equal amount of ND1 and CYTB mtRNAs should be found, since they are encoded on the same template strand. This is not
the case, probably because the amount of each mtRNA is determined not only by its transcription (presumably occurring at the same rate for ND1 and CYTB genes) but also by its degradation, which probably occurs at different rates, according to the intrinsic stability of each transcript and the cellular functional microenvironment. Our analysis allows the quantification of mtRNAs using nuclear RNA as a reference. Even if quite unlikely, because the nuclear gene we chose as reporter is quite stable [22], it cannot be excluded that modifications in nuclear transcription could influence the data we obtained. We have found different expression of CYTB in HIV+ male and females. Since CYTB is the last gene transcribed in the heavy chain and since no main differences were observed among fat samples collected in different sites, one could speculate that the efficiency of the transcription is lower in women, or that they are more prone to mt toxicity. Since the number of patients we have studied is relatively low, further data are required to investigate this aspect. We have previously shown that the relative ratios among these three mtRNAs is different in various experimental models [14]. This suggests that it cannot be exclusively due to different transcription rates, but is probably the result of a dynamic process involving rate of mtRNA synthesis, editing, stability and degradation, which all vary between different cells [14]. What we observed in ex vivo (lymphocytes) and in vitro models (U937, CEM and HUT-78 cell lines) clearly indicates that the relative ratio among ND1, ND6 and CYTB is highly variable, but is usually kept constant in a particular type of cell. The data we report in this paper reveal the presence of an association between NRTI-induced lipodystrophy in HIV+ patients and significantly reduced levels of ND1, ND6 and CYTB mtRNAs. However, we are well aware that the analysis of HIV+ patients without lipodystrophy is urgently needed to ascertain which is the role of the viral infection per se in provoking mtRNA alterations, as observed for mtDNA [41]. For this reason, we are in the process of selecting naive HIV+ patients whose adipocytes will be studied. When chronically exposed to NRTI drugs, adipose tissue from HIV+ patients with lipodystrophy shows a general, dramatic decrease in the mtRNA levels for all of the three messengers analysed. However, even in these cells, the ratio between the mtRNAs was roughly maintained, suggesting that NRTIs act by provoking a general drop in the efficiency of the transcriptional apparatus of mitochondria. Since the decrease in mtRNA levels does not affect one single transcript, we are tempted to assume that NRTIs interfere with the overall transcriptional process rather than affecting the stability of mature molecules. It will be interesting to
know whether this decrease is an indirect consequence of mtDNA depletion in these cells (a well-known phenomenon described by several authors in adipose tissue), or if it is due to a direct action of NRTIs on the transcriptional machinery. In a previous work, we demonstrated that mtRNA levels can decrease in cell lines treated with NRTIs even when mtDNA content is still preserved [14]. This additionally stresses the importance of mtRNA quantification, which is able to reveal mt impairment before a considerable decrease in mtDNA content can be measured. Direct studies on mtDNA transcription in the presence or absense of different drugs/drug combinations, as well as sequential studies in patients who start NRTI-based therapy, could clarify the causes of this phenomenon and help in a better identification of the molecular target(s) of their toxic action.
Acknowledgements
This paper was supported by ISS, Rome (Programma Nazionale di Ricerca sullAIDS, grant No. 30F.15 to AC). GeneMoRe Italy srl and GeneMoRe Holding SA are gratefully acknowledged.
References
1. Carr A, Samaras K, Burton S, Law M, Freund J, Chisholm DJ & Cooper DA. A syndrome of peripheral lipodystrophy, hyperlipidaemia and insulin resistance in patients receiving HIV protease inhibitors. AIDS 1998; 12:F51F58. Carr A. HIV lipodystrophy: risk factors, pathogenesis, diagnosis and management. AIDS 2003; 17(Suppl 1):S141S148. Brinkman K, Smeitink JA, Romijn JA & Reiss P. Mitochondrial toxicity induced by nucleoside-analogue reverse-transcriptase inhibitors is a key factor in the pathogenesis of antiretroviral-therapy-related lipodystrophy. Lancet 1999; 354:11121115. Carr A, Samaras K, Thorisdottir A, Kaufmann GR, Chisholm DJ & Cooper DA. Diagnosis, prediction, and natural course of HIV-1 protease-inhibitor-associated lipodystrophy, hyperlipidaemia, and diabetes mellitus: a cohort study. Lancet 1999; 353:20932099. Carr A, Miller J, Law M & Cooper DA. A syndrome of lipoatrophy, lactic acidaemia and liver dysfunction associated with HIV nucleoside analogue therapy: contribution to protease inhibitor-related lipodystrophy syndrome. AIDS 2000; 14:F25F32. Ct HC, Brumme ZL, Craib KJ, Alexander CS, Wynhoven B, Ting L, Wong H, Harris M, Harrigan PR, OShaughnessy MV & Montaner JS. Changes in mitochondrial DNA as a marker of nucleoside toxicity in HIV-infected patients. New England Journal of Medicine 2002; 346:811820. Cossarizza A, Troiano L & Mussini C. Mitochondria and HIV infection: the first decade. Journal of Biological Regulators & Homeostatic Agents 2002; 16:1824. Cossarizza A, Pinti M, Moretti L, Bricalli D, Bianchi R, Troiano L, Fernandez MG, Balli F, Brambilla P, Mussini C & Vigan A. Mitochondrial functionality and mitochondrial DNA content in lymphocytes of vertically infected human immunodeficiency virus-positive children with highly active antiretroviral therapy-related lipodystrophy. Journal of Infectious Diseases 2002; 185:299305.
2.
3.
4.
5.
6.
7.
8.
M97
L Galluzzi et al.
9.
Cossarizza A, Riva A, Pinti M, Ammannato S, Fedeli P, Mussini C, Esposito R & Galli M. Increased mitochondrial DNA content in peripheral blood lymphocytes from HIVinfected patients with lipodystrophy. Antiviral Therapy 2003; 8:315321.
26.
10. Hoy JF, Gahan ME, Carr A, Smith D, Lewin SR, Wesselingh S & Cooper DA. Changes in mitochondrial DNA in peripheral blood mononuclear cells from HIVinfected patients with lipoatrophy randomized to receive abacavir. Journal of Infectious Diseases 2004; 190:688692. 11. Shikuma CM, Hu N, Milne C, Yost F, Waslien C, Shimizu S & Shiramizu B. Mitochondrial DNA decrease in subcutaneous adipose tissue of HIV-infected individuals with peripheral lipoatrophy. AIDS 2001; 15:18011809. 12. Shiramizu B, Shikuma KM, Kamemoto L, Gerschenson M, Erdem G, Pinti M, Cossarizza A & Shikuma C. Placenta and cord blood mitochondrial DNA toxicity in HIVinfected women receiving nucleoside reverse transcriptase inhibitors during pregnancy. Journal of Acquired Immune Deficiency Syndromes 2003; 32:370374. 13. Cossarizza A & Moyle G. Antiretroviral nucleoside and nucleotide analogues and mitochondria. AIDS 2004; 18:137151. 14. Galluzzi L, Pinti M, Troiano L, Prada N, Nasi M, Ferraresi R, Salomoni P, Mussini C & Cossarizza A. Changes in mitochondrial RNA production in cells treated with nucleoside analogues. Antiviral Therapy 2005; 10:191195. 15. Anderson S, Bankier AT, Barrell BG, de Bruijn MH, Coulson AR, Drouin J, Eperon IC, Nierlich DP, Roe BA, Sanger F, Schreier PH, Smith AJ, Staden R & Young IG. Sequence and organization of the human mitochondrial genome. Nature 1981; 290:457465. 16. Bonen L. The mitochondrial genome: so simple yet so complex. Current Opinion in Genetics & Development 1991; 1:515522. 17. Shuey DJ & Attardi G. Characterization of an RNA polymerase activity from HeLa cell mitochondria, which initiates transcription at the heavy strand rRNA promoter and the light strand promoter in human mitochondrial DNA. Journal of Biological Chemistry 1985; 260:19521958. 18. Clayton DA. Transcription of the mammalian mitochondrial genome. Annual Reviews of Biochemistry 1984; 53:573594. 19. Clayton DA. Transcription and replication of mitochondrial DNA. Human Reproduction 2000; 15(Suppl 2):1117. 20. Sambrook J & Russell DW. Molecular Cloning: a Laboratory Manual. 3rd edn 2001. New York: Cold Spring Harbor Laboratory Press. 21. Pinti M, Troiano L, Nasi M, Monterastelli E, Moretti L, Bellodi C, Mazzacani A, Mussi C, Salvioli G & Cossarizza A. Development of real time PCR assays for the quantification of Fas and FasL mRNA levels in lymphocytes: studies on centenarians. Mechanisms of Ageing & Development 2003; 124:511516. 22. Fu X, Menke JG, Chen Y, Zhou G, MacNaul KL, Wright SD, Sparrow CP & Lund EG. 27-hydroxycholesterol is an endogenous ligand for liver X receptor in cholesterolloaded cells. Journal of Biological Chemistry 2001; 276:3837838387. 23. Cherry CL, Gahan ME, McArthur JC, Lewin SR, Hoy JF & Wesselingh SL. Exposure to dideoxynucleosides is reflected in lowered mitochondrial DNA in subcutaneous fat. Journal of Acquired Immune Deficiency Syndromes 2002; 30:271277. 24. Walker UA, Bickel M, Lutke Volksbeck SI, Ketelsen UP, Schofer H, Setzer B, Venhoff N, Rickerts V & Staszewski S. Evidence of nucleoside analogue reverse transcriptase inhibitor associated genetic and structural defects of mitochondria in adipose tissue of HIV-infected patients. Journal of Acquired Immune Deficiency Syndromes 2002; 29:117121. 25. Nolan D, Hammond E, Martin A, Taylor L, Herrmann S, McKinnon E, Metcalf C, Latham B & Mallal S.
M98
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
Mitochondrial DNA depletion and morphologic changes in adipocytes associated with nucleoside reverse transcriptase inhibitor therapy. AIDS 2003; 17:13291338. van der Valk M, Casula M, Weverlingz GJ, van Kuijk K, van Eck-Smit B, Hulsebosch HJ, Nieuwkerk P, van Eeden A, Brinkman K, Lange J, de Ronde A & Reiss P. Prevalence of lipoatrophy and mitochondrial DNA content of blood and subcutaneous fat in HIV-1-infected patients randomly allocated to zidovudine- or stavudine-based therapy. Antiviral Therapy 2004; 9:385393. Mallal SA, John M, Moore CB, James IR & McKinnon EJ. Contribution of nucleoside analogue reverse transcriptase inhibitors to subcutaneous fat wasting in patients with HIV infection. AIDS 2000; 14:3091316. Zaera MG, Mir O, Pedrol E, Soler A, Picon M, Cardellach F, Casademont J & Nunes V. Mitochondrial involvement in antiretroviral therapy-related lipodystrophy. AIDS 2001; 15:16431651. Pace CS, Martin AM, Hammond EL, Mamotte CD, Nolan DA & Mallal SA. Mitochondrial proliferation, DNA depletion and adipocyte differentiation in subcutaneous adipose tissue of HIV-positive HAART recipients. Antiviral Therapy 2003; 8:323331. Carr A, Workman C, Smith DE, Hoy J, Hudson J, Doong N, Martin A, Amin J, Freund J, Law M & Cooper DA; for the Mitochondrial Toxicity (MITOX) Study Group. Abacavir substitution for nucleoside analogs in patients with HIV lipoatrophy: a randomized trial. Journal of the American Medical Association 2002; 288:207215. Martin A, Smith DE, Carr A, Ringland C, Amin J, Emery S, Hoy J, Workman C, Doong N, Freund J & Cooper DA; for the Mitochondrial Toxicity (MITOX) Study Group. Reversibility of lipoatrophy in HIV-infected patients 2 years after switching from a thymidine analogue to abacavir: the MITOX Extension Study. AIDS 2004; 18:10291036. McComsey GA, Paulsen DM, Lonergan JT, Hessenthaler SM, Hoppel CL, Williams VC, Fisher RL, Cherry CL, White-Owen C, Thompson KA, Ross ST, Hernandez JE & Ross LL. Improvements in lipoatrophy, mitochondrial DNA levels and fat apoptosis after replacing stavudine with abacavir or zidovudine. AIDS 2005; 19:1524. Negredo E, Paredes R, Peraire J, Pedrol E, Ct H, Gel S, Fumoz CR, Ruiz L, Abril V, Rodriguez de Castro E, Ochoa C, Martinez-Picado J, Montaner J, Rey-Joly C & Clotet B (Swatch Study Team). Alternation of antiretroviral drug regimens for HIV infection. Efficacy, safety and tolerability at week 96 of the Swatch Study. Antiviral Therapy 2004; 9:889893. Hammond E, Nolan D, James I, Metcalf C & Mallal S. Reduction of mitochondrial DNA content and respiratory chain activity occurs in adipocytes within 612 months of commencing nucleoside reverse transcriptase inhibitor therapy. AIDS 2004; 18:815817. Seidel-Rogol BL & Shadel GS. Modulation of mitochondrial transcription in response to mtDNA depletion and repletion in HeLa cells. Nucleic Acids Research 2002; 30:19291934. Mir O, Lopez S, Rodriguez de la Concepcion M, Martinez E, Pedrol E, Garrabou G, Giralt M, Cardellach F, Gatell JM, Vilarroya F & Casademont J. Upregulatory mechanisms compensate for mitochondrial DNA depletion in asymptomatic individuals receiving stavudine plus didanosine. Journal of Acquired Immune Deficiency Syndromes 2004; 37:15501555. dAmati G & Lewis W. Zidovudine causes early increases in mitochondrial ribonucleic acid abundance and induces ultrastructural changes in cultured mouse muscle cells. Laboratory Investigations 1994; 71:879884. Mallon P, Unemori P, Bowen M, Miller J, Winterbotham M, Kelleher A, Williams K, Cooper D & Carr A. Nucleoside reverse transcriptase inhibitors decrease mitochondrial and PPARgamma gene expression in adipose tissue after only 2 weeks in HIV-uninfected healthy adults. 11th Conference on Retroviruses & Opportunistic Infections. 1114 February 2004, San Francisco, CA, USA. Abstract 76.
39. Suomalainen A, Majander A, Pihko H, Peltonen L & Syvanen AC. Quantification of tRNA3243(Leu) point mutation of mitochondrial DNA in MELAS patients and its effects on mitochondrial transcription. Human Molecular Genetics 1993; 2:525534. 40. Bonod-Bidaud C, Giraud S, Mandon G, Mousson B & Stepien G. Quantification of OXPHOS gene transcripts during muscle cell differentiation in patients with
mitochondrial myopathies. Experimental Cell Research 1999; 246:9197. 41. Lopez S, Mir O, Martinez E, Pedrol E, RodriguezSantiago B, Milinkovic A, Soler A, Garcia-Viejo MA, Nunes V, Casademont J, Gatell JM & Cardellach F. Mitochondrial effects of antiretroviral therapies in asymptomatic patients. Antiviral Therapy 2004; 9:4755.
Received 20 December 2004, accepted 29 March 2005
M99
Methodological considerations in human studies of gene expression in HIV-associated lipodystrophy

Patrick WG Mallon1,2,3*, Rebecca Sedwell 1,3, Patrick Unemori1,3, Anthony Kelleher 1,2,3, David A Cooper1,2,3 and Andrew Carr 2,3
1 2
National Centre in HIV Epidemiology and Clinical Research, University of New South Wales, Sydney, Australia HIV, Immunology and Infectious Diseases Clinical Services Unit, St Vincents Hospital, Sydney, Australia 3 HIV Immunovirology Research Laboratory, Centre for Immunology, St Vincents Research Campus, Sydney, Australia *Corresponding author: Tel: +61 2 8382 3107; Fax: +61 2 8382 2391; E-mail: p.mallon@cfi.unsw.edu.au
In the majority of cases, HIV-associated lipodystrophy, lipoatrophy in particular, becomes clinically apparent only after months or years of continuous exposure to antiretroviral medications and, once developed, is difficult to reverse. Many lipid-related side effects of antiretroviral medications result from drug-induced changes in gene expression. As our understanding of the pathogenic mechanisms underlying HIV-associated lipodystrophy improves, it is important to be able to explore changes at a molecular level in order to fully elucidate the mechanisms whereby antiretroviral drugs exert their toxicities.
Monitoring changes in gene expression in vivo may enable physicians to identify, predict or prevent drug toxicities early, before irreversible changes in body composition occur. However, monitoring changes in gene expression at a population level presents many methodological challenges that need to be addressed, over and above the considerable intra- and inter-individual variability inherent in the cellular expression of any gene. Careful collection and processing of adequate biological samples, robust laboratory processes and assays, and appropriate study design can help overcome many of these difficulties.
Introduction
HIV-associated lipodystrophy (HIVLD) is a complex syndrome of multifactorial aetiology and, although prevention strategies have yielded some encouraging results [1], no effective treatment exists. Lipoatrophy takes months to develop [2,3] and, even when the causal agents are substituted, undergoes only a slow and incomplete reversal [4]. In vitro, protease inhibitors (PIs) disrupt the action of sterol regulatory element binding protein 1 (SREBP1), a transcriptional activator that mediates expression of peroxisome proliferatoractivated receptor gamma (PPAR) [5,6]. The thymidine analogue nucleoside reverse transcriptase inhibitors (NRTIs) zidovudine and stavudine decrease expression of mitochondrial genes and PPAR in human adipose tissue [7] and have been linked with mitochondrial dysfunction [8,9] and depletion [10,11]. As there is no effective way to predict or monitor the onset of HIVLD before it becomes clinically apparent, a fuller understanding of drug-induced molecular changes in vivo is important if the joint goals of effective prevention and treatment of HIVLD are to be realized. In contrast to laboratory experiments using cell lines, where conditions and interventions can be tightly regulated, studies of gene expression in vivo are complicated by numerous potential confounding factors that can complicate the detection or interpretation of changes in gene expression. In the case of lipodystrophy, the incidence and severity of the syndrome can be significantly influenced by factors such as age, gender, presence of coinfection, and pretreatment weight and immune status [12]. Many of these factors may influence the molecular effects of antiretrovirals and complicate gene expression studies in this clinical setting. With any target tissue under investigation, such as subcutaneous adipose tissue, appropriate sample collection and processing is of the utmost importance. This is because the amount of tissue required is dictated by the number of planned investigations and the RNA yield from the target tissue, which varies considerably between tissues and, in the case of human subcutaneous adipose tissue, is relatively low [7,13,14]. In addition, certain genes of interest, such as transcription factors, are expressed at very low levels within tissues [7] making it more difficult to detect subtle but potentially clinically important changes in their expression. Biological and pharmacological confounders, such as ongoing opportunistic infections or use of polypharmacy, which can affect overall cellular gene expression in a target tissue, may make
M101
PWG Mallon et al.
identifying an appropriate housekeeping gene difficult. Despite these obstacles, with careful attention to study design and sample processing, it is possible to limit the impact of these factors.
Sample collection and processing

RNA yields from adipose tissue are lower per weight of tissue than other tissues such as liver and muscle. In contrast to muscle and liver, which are expected to yield 100600 g of RNA per 100 mg tissue [15], adipose tissue yields are only in the region of 23 g per 100 mg tissue [7,11,13,14]. Given the rapid degradation of RNA, in particular messenger RNA (mRNA), careful sample collection and rapid processing, such as immediate fixing and snap-freezing of biopsied samples in liquid nitrogen, can limit degradation and maintain optimal RNA yields (Figure 1). Careful consideration also needs to be applied to the quality of tissue biopsied. Collection techniques such as needle biopsies [11,13] or liposuction [14], are often
well tolerated and easy to perform, but the overall yields of tissue can be low. In addition, the blinded nature of the sampling increases the risk of contamination of biopsied fat with unwanted tissues such as connective tissue and muscle, especially in subjects with low subcutaneous fat volumes, such as those seen in patients with lipoatrophy. Open biopsy techniques [7,10] (Figure 2) have the potential advantage of yielding larger amounts of tissue, with operators being able to carefully biopsy adipose tissue, thus limiting contamination. However, these procedures are arguably more invasive and complicated than needlebased procedures.
RNA extraction
Although isolation reagents based on phenol and guanidine isothiocyanate, such as TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA, USA), are commonly used to fix tissues, avoiding RNA degradation and maintaining RNA expression as close as
Figure 1. Processing of adipose tissue for use in gene expression experiments
Extract RNA, DNase treatment, purify and store in 200 ng aliquots
4 X RNA (200 ng)
Adipose tissue
Fix in TRlzol, snap freeze and store at 70C
Homogenize (100 mg tissue per ml TRlzol)
4 X RT (oligo dT)
4 X cDNA (20 l) Low expression = fail (discard aliquots) Check PCR (-actin)
-actin
Good expression = pool aliquots
Results = gene of interest -actin Gene of interest

RT, reverse transcriptase.
PCR (duplicates)
cDNA pool (80 l)
2 l
M102
Molecular studies in HIV-associated lipodystrophy
Figure 2. Open subcutaneous adipose tissue biopsy
Figure 3. RNA yields from adipose tissue using TRIzol extraction protocol
30 25 RNA yield, g 20 15 10 5 0 0 200 400 600 800
The biopsy is taken from the lateral aspect of the buttock. The procedure is performed in an ambulatory setting under local anaesthetic.
Mass of adipose tissue, mg/ml TRlzol

RNA yields begin to plateau if RNA is extracted from more than 400 mg adipose tissue per 1 ml TRIzol. This is probably due to inhibition of the reagent due to high lipid levels within the sample. RNA was extracted from both visceral and subcutaneous adipose tissue samples.
possible to that seen in vivo at the time of sample collection, high lipid levels within the target tissue (such as seen in adipose tissue) can inhibit the chloroform/ethanol extraction process and reduce the yield of RNA obtained using this method. In the case of human adipose tissue using the TRIzol extraction protocol, RNA yields tend not to increase if more than 400 mg adipose tissue is fixed per ml of TRIzol (Figure 3), presumably because of the inhibitory effects of higher lipid concentrations on the reagent and subsequent precipitation process. Problems with RNA yield are not limited to adipose tissue. We have evaluated monocytes as a potential peripheral blood surrogate for mitochondrial toxicity [7]. As monocytes make up only a fraction of the total cell population in whole blood, we found the RNA yield from extracted monocytes also to be relatively low in the region of 2 g per monocyte sample. In monocytes, the RNA yield from extracted samples significantly correlated with the number of monocytes isolated from the original whole blood sample (Figure 4). It is therefore important to ensure that the volume of blood collected is enough to ensure an adequate number of monocytes to give an appropriate RNA yield. In the majority of cases, more than 3 million monocytes [extracted from 18 ml blood in acid citrate dextrose (ACD) buffer] provide in excess of 2 g total RNA, enough to examine the expression of more than 10 genes of interest.
Figure 4. Relationship between RNA yield and monocyte cell count

7 6 Cell count (x106 cells) 5 4 3 2 1 0 0 2 4 6 RNA yield, g
RNA yields rise proportionally to the number of monocytes in the starting sample. Starting with at least 3 million cells (extracted from 18 ml ACD blood) will give a yield >2 g RNA in more than 90% cases. ACD, acid citrate dextrose (buffer).
Rho=0.61 P=0.001
10
12
RNA and cDNA preparation

For real-time quantitative PCR, it is important that the starting RNA is of the highest purity. As a minimum, RNA samples should undergo DNase treatment and
column purification. Contamination with genomic DNA can give false positive results, especially for genes that are expressed at very low levels, whilst contamination with ethanol-based buffers used in RNA extraction can inhibit subsequent PCR reactions [16]. After RNA extraction, it is usual to store RNA at 70C for use in future PCR reactions or gene arrays. As RNA is
M103
PWG Mallon et al.
unstable and tends to degrade easily with every freezethaw episode, adequate storage of RNA is important. Storing RNA at pre-determined concentrations (either ng or g aliquots, depending on whether the planned investigations include PCR or gene array) avoids multiple freezethaws. Ideally, laboratory protocols should be developed to limit the number of freezethaw events between RNA extraction and cDNA preparation to only one. Various methods are used to quantify extracted RNA, the commonest of which are absorbance assays [17] and fluorescent nucleic acid stains [7,13]. Spectrophotometers measure the ability of RNA to absorb ultraviolet light. Maximum absorption occurs at 260 nm and the absorption can be used to calculate the concentration of RNA within a sample. Due to the potential for contaminants such as protein and DNA to affect absorbance, this method is relatively insensitive, especially when measuring low concentrations of RNA, such as those extracted from adipose tissue (for example, an A260 of only 0.1 represents 4 g/ml RNA). Fluorescent nucleic acid stains, such as SYBR Green (Applied Biosystems, Foster City, CA, USA) [7] and RiboGreen (Molecular Probes, Eugene, OR, USA) [13], rely on the use of standard curves constructed from serial dilutions of RNA of a known quantity. This method is more specific for RNA and is accurate at measuring even small quantities. First strand complementary DNA (cDNA), used in real-time PCR experiments, can be prepared from RNA using oligo dT [7,13] or random hexamer primers [17] and the reverse transcriptase (RT) enzyme. The reaction involves the RT enzyme elongating single strands of DNA from a site where the primer anneals to the RNA, using the RNA as a template. As their name suggests, random hexamer primers comprise short lengths of random nucleic acids capable of annealing to all types of RNA, while oligo dT primers are designed to specifically anneal to the poly-A tails of mRNA. For this reason, there are differences in the cDNA resulting from the use of these two primers. RT using oligo dT should result in transcription to cDNA of only transcribed genes (mRNA) and some ribosomal RNA (rRNA), whilst use of random hexamers results in the additional transcription to cDNA of non-coding RNA, transfer RNA (tRNA) and most rRNA. The presence of cDNA transcribed from non-messenger RNA introduces the risk of interference by these molecules in any subsequent PCR reaction, increasing non-specific annealing of PCR primers and the potential for false positive results. Regardless of the primers used, the efficiency of the RT reaction can vary considerably, both between samples and between reactions. If equal amounts of the same RNA sample undergo RT, the yield of cDNA, as
M104
measured by -actin expression, can vary by more than 60% (Figure 5A). The variability introduced by the RT step can significantly affect results when RT is combined with PCR in a single-step RT-PCR reaction [16]. If efficiency of a particular RT reaction is low, it may be difficult to measure the expression of a gene of interest, such as a transcription factor, that is normally expressed at much lower levels than -actin; sometimes greater than 100-fold lower for genes such as those encoding PPAR co-activator 1 or the uncoupling proteins UCP1 and 2 [7,13]. Separating the RT step from the PCR reaction, performing multiple RT reactions on an individual sample, testing the efficiency of the RT by examining the expression of a gene known to be ubiquitously expressed in the target tissue (such as -actin) and pooling the multiple cDNAs to establish a pool of adequately transcribed cDNA can help overcome much of this type of variability (Figure 5B). In this way, a large amount of cDNA of adequate quality can be made from which expression of multiple genes of interest can be determined (Figure 1).
Analysis
Given the variability introduced by, among other things, RNA extraction and RT techniques, careful interpretation of results is required. Although various corrections can be made along the way to attempt to adjust for within-sample variation, such as quantification of and correction for starting RNA or cDNA concentrations, these analyses are relatively insensitive when using nanogram quantities for real-time PCR assays. As such, it is important to perform a final correction prior to presentation of data the correction for the housekeeping gene. The principal is simple. Find a gene that is ubiquitously expressed in the tissue of interest and which isnt affected by the intervention under investigation, and present any changes in a gene of interest relative to changes in the housekeeping gene. By doing so, an investigator can say with some confidence that the intervention under investigation is the cause of the change in expression of the gene of interest, rather than other potential confounders such as insensitive quantification of RNA, failed reverse transcription or interference in cellular gene expression introduced by environmental or laboratory factors. This argument for use of a housekeeping gene carries special significance in cross-sectional human studies that measure changes in gene expression in disease states. For example, in studies in HIV-positive populations, confounders such as age, weight, cortisol and insulin responsiveness and the effects of HIV, antiretroviral therapy and changes in body composition all have the potential to individually alter cellular gene expression
Figure 5. (A) Intra-sample variation in reverse transcriptase and (B) post-pooling gene expression
A
100 000
Inter-RT, intra-sample cv=0.61
10 000 actin expression (x10-8 ng/l)
1000
100
10
1
R R R R R R R R R R R R R R R R R R R R R RR R R R R R R R R R R R R R R R R S S S S S S S S S S S S S S S 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 5 5 6 7 8 9 1 1 1 1 1 1 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 4 9 0 2 6 7 0 3 0 1 2 3 4 6
Sample ID
B
100 000
actin expression (x10 -8 ng/l)
10 000
1000
100
10
1
R R R R R R R R R R R R R RR R R R R R R R R R R R R R R R R R R R R R R R R S S S S S S S S S S S S S S 1 1 1 1 1 1 1 1 1 1 1 2 2 22 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 5 5 5 5 5 6 7 8 9 1 1 1 1 1 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 12 4 6 7 8 9 0 1 2 3 8 6 7 8 0 1 3 4 0 1 2 3 4
Sample ID
(A) shows the considerable inter-RT variability (measured by expression of -actin in the cDNA) when equal amounts of starting RNA from the same sample undergo RT. The lower limit of this assay is 1x108 ng/l. Therefore, if the check RT result falls below 1x106 ng/l (dotted line) then the sample would be unsuitable for measuring expression of genes such as transcription factors, that are expressed at much lower quantities than -actin and should be considered to have failed the RT. (B) shows duplicate results for -actin expression after discarding the failed samples and pooling the remaining cDNA. Much of the intra-sample variability is removed and a large amount of good quality cDNA remains from which expression of multiple genes of interest can be measured. cv, coefficient of variation; ID, identification; RT, reverse transcriptase. 1st Meeting on Mitochondrial Toxicity & HIV Infection: Understanding the Pathogenesis for a Therapeutic Approach M105
PWG Mallon et al.
Figure 6. SREBP1 expression in monocytes before and after correction for -actin expression
A
Gene expression (x10-8 ng/l)
100
P=0.005
HIV-negative control (n=14) HIV-positive patient with HIVLD (n=39)
75
50
25
0 SREBP1
B
20 000 Gene expression (x10-8 ng/l)
P=0.007
15 000
10 000
5000
0 -actin
0.005
P=0.8
0.004 SREBP1/-actin
0.003
0.002
0.001
0 SREBP1/ actin
HIVLD, HIV-associated lipodystrophy; SREBP1, sterol-regulatory element binding protein 1.
[1821]. Even exercise can significantly alter the expression of some commonly used housekeeping genes [22]. As a result, some authors recommend that, due to inherent changes in gene expression between systems, use of two [23] or even multiple [24] housekeeping genes should be employed in any analysis of such studies. In such diverse populations as those seen in studies of lipodystrophy, identification of a true housekeeping gene can be difficult and any results presented without correcting for expression of a housekeeping gene need to be interpreted with caution. Various housekeeping genes have been used in human studies of gene expression. These include genes encoding -actin, glyceraldehyde-3-phosphate dehydrogenase (GAPDH), 2-microglobulin and 18S ribosomal RNA [7,13,14,17]. An in-depth knowledge of the molecular characteristics of the tissue under investigation, including the likely effect of any intervention on the expression of a housekeeping gene and the laboratory processes used, helps in the choice of an appropriate housekeeping gene. For example, RT using oligo dT primers, although more specific in transcribing mRNA, may not transcribe rRNA as efficiently. As a result, most studies using 18S rRNA as a control use random hexamers to prepare cDNA. In addition, as HIVLD arises as a result of disturbances in adipose tissue lipid metabolism, use of genes encoding components of the glycolytic pathway, such as GAPDH, may be inappropriate as it is likely that their expression will be affected by disturbances in lipid metabolism. To illustrate these difficulties, we performed a crosssectional study examining the expression of the transcription factor SREBP1 in monocytes extracted from whole blood, as previously described [7]. We examined monocyte gene expression using real-time PCR (Lightcycler, Roche Applied Science, Mannheim, Germany) from samples of pooled cDNA synthesized from individual starting quantities of 20 ng RNA using oligo dT primers. Samples from a total of 53 subjects were analysed: 39 HIV-infected subjects with lipodystrophy and 14 HIV-negative healthy controls. Initial, uncontrolled analysis showed a significant decrease in SREBP1 expression in those subjects with HIVLD (Figure 6A), which would support a drug-induced down-regulation of SREBP1 in monocytes similar to that observed in fat. However, -actin expression was also significantly down-regulated in those with HIVLD compared with healthy controls (Figure 6B). When the SREBP1 results were reanalysed correcting for -actin expression, no significant difference in SREBP1 expression was detected between the two groups (Figure 6C). Similar results were found when GAPDH was used as an alternative housekeeping gene (data not shown). As such, these results tell us very little about the relationship between HIVLD and gene expression in monocytes
M106
as it is not possible to control for the inherent differences in gene expression between the two diverse populations studied using these housekeeping genes. This is a limitation seen in many cross-sectional studies and underlies the need for appropriate study design when examining tissue gene expression in human disease.
Population selection and study design

Cross-sectional human studies of gene expression in lipodystrophy, especially those involving small numbers of subjects, are prone to significant bias introduced by the many potential confounding factors described above. To overcome these difficulties in a cross-sectional setting, studies need to be carefully and adequately powered for expression of each gene of interest before comments relating to the clinical or biological relevance of negative results can be made. The target tissue under examination can also differ significantly in structure and cellular content between cross-sectional groups. In the case of subcutaneous adipose tissue, tissue biopsies from patients affected by HIVLD have been shown to differ at a microscopic level from HIV-negative control subjects, with smaller adipocytes and infiltrates of inflammatory cells observed in some diseased adipose tissue [10,14]. Although, as previously described, adequate sampling can help reduce contamination by unwanted tissue, the potential influence of tissue infiltration by cells other than adipocytes on gene expression within adipose tissue needs to be carefully considered when interpreting the results of cross-sectional studies. Prospective studies using well-defined populations can overcome many of these difficulties. As many antiretroviral drug (ARV) toxicities originate at the molecular level, changes in expression of biologically relevant genes would be expected to occur early after exposure to ARVs. By examining changes in gene expression induced by ARVs in a prospective randomized setting, environmental and disease-related factors may have less of an impact on the overall results, as any potential error introduced by these confounders would be present in both baseline and post-intervention samples, provided that the interval between sampling is relatively close. At present, the number of such studies is relatively small, but is increasing.
HIV introduces, it can be difficult to determine definite cause and effect relationships between exposure to ARVs and subsequent effects on gene expression in vivo. In addition, when expression of multiple genes changes in response to ARV exposure [7,13,14,25], it is almost impossible to elicit an exact sequence of events, that is, which changes arise directly from antiretroviral drug effects and which arise as a result of dysregulation induced by changes in other genes, without very detailed and sample-intensive time-course studies. Human studies involving multiple biopsies, although ideal in principle, would be both difficult to recruit to and ethically challenging. For these reasons, in vitro experiments remain a vital resource for robustly testing the many hypotheses that the limited in vivo gene expression studies generate.
Summary
Molecular studies of lipodystrophy in populations of HIV-infected subjects have the potential to reveal aspects of the syndrome that may help prevent the syndrome occurring and aid in the development of new, safer therapies. Effective research in this field requires an effective communication between clinical and laboratory researchers to enable them to design and implement clinical trials and laboratory assays capable of overcoming the many pitfalls inherent in this type of research. Further research is vital, particularly using newer technologies such as microarray that are able to characterize genomewide responses to exposure to ARVs, if the goal of safe, lifelong ARV therapy for HIV is to be realized.
References
1. Gallant JE, Staszewski S, Pozniak AL, DeJesus E, Suleiman JM, Miller MD, Coakley DF, Lu B, Toole JJ & Cheng AK; 903 Study Group. Efficacy and safety of tenofovir DF vs stavudine in combination therapy in antiretroviral-naive patients: a 3-year randomized trial. Journal of the American Medical Association 2004; 292:191201. Carr A, Samaras K, Burton S, Law M, Freund J, Chisholm DJ & Cooper DA. A syndrome of peripheral lipodystrophy, hyperlipidaemia and insulin resistance in patients receiving HIV protease inhibitors. AIDS 1998; 12:F51F58. Mallon PWG, Miller J, Cooper D & Carr A. Prospective evaluation of the effects of antiretroviral therapy on body composition in HIV-1 infected men starting therapy. AIDS 2003; 17:971979. Martin A, Smith DE, Carr A, Ringland C, Amin J, Emery S, Hoy J, Workman C, Doong N, Freund J & Cooper DA; Mitochondrial Toxicity Study Group. Reversibility of lipoatrophy in HIV-infected patients 2 years after switching from a thymidine analogue to abacavir: the MITOX Extension Study. AIDS 2004; 18:10291036. Liang J-S, Distler O, Cooper DA, Jamil H, Deckelbaum RJ, Ginsberg HN & Sturley SL. HIV protease inhibitors protect apolipoprotein B from degradation by the proteasome: a potential mechanism for protease inhibitor-induced hyperlipidaemia. Nature Medicine 2001; 7:15. Caron M, Auclair M, Vigouroux C, Glorian M, Forest C & Capeau J. The HIV protease inhibitor indinavir impairs
2.
3.
4.
Limitations
Molecular studies on human tissue offer the ability to demonstrate clinically relevant changes in gene expression due to ARVs. However, this approach has several important limitations. As previously mentioned, many molecular pathways involved in HIVLD are under complex regulatory control and, given the many variables
5.
6.
M107
PWG Mallon et al.
sterol regulatory element-binding protein-1 intranuclear localisation, inhibits preadipocyte differentiation, and induces insulin resistance. Diabetes 2001; 50:13781388. 7. Mallon PW, Unemori P, Sedwell R, Morey A, Rafferty M, Williams K, Chisholm D, Samaras K, Emery S, Kelleher A, Cooper DA & Carr A; SAMA Investigators. In vivo, nucleoside reverse-transcriptase inhibitors alter expression of both mitochondrial and lipid metabolism genes in the absence of depletion of mitochondrial DNA. Journal of Infectious Diseases 2005; 191:16861696. Brinkman K, Smeitink JA, Romijn HA & Reiss P. Mitochondrial toxicity induced by nucleoside-analogue reverse-transcriptase inhibitors is a key factor in the pathogenesis of antiretroviral-related lipodystrophy. Lancet 1999; 354:11121115. Lewis W & Dalakas MC. Mitochondrial toxicity of antiviral drugs. Nature Medicine 1995; 1:417422.
17. Liu YM, Lacorte JM, Viguerie N, Poitou C, Pelloux V, Guy-Grand B, Coussieu C, Langin D, Basdevant A & Clement K. Adiponectin gene expression in subcutaneous adipose tissue of obese women in response to short-term very low calorie diet and refeeding. Journal of Clinical Endocrinology & Metabolism 2003; 88:58815886. 18. Tomlinson JW, Moore JS, Clark PM, Holder G, Shakespeare L & Stewart PM. Weight loss increases 11beta-hydroxysteroid dehydrogenase type 1 expression in human adipose tissue. Journal of Clinical Endocrinology & Metabolism 2004; 89:27112716. 19. Tiikkainen M, Hakkinen AM, Korsheninnikova E, Nyman T, Makimattila S & Yki-Jarvinen H. Effects of rosiglitazone and metformin on liver fat content, hepatic insulin resistance, insulin clearance, and gene expression in adipose tissue in patients with type 2 diabetes. Diabetes 2004; 53:21692176. 20. Marchlewska A, Stenvinkel P, Lindholm B, Danielsson A, Pecoits-Filho R, Lonnqvist F, Schalling M, Heimburger O & Nordfors L. Reduced gene expression of adiponectin in fat tissue from patients with end-stage renal disease. Kidney International 2004; 66:4650. 21. Otake K, Omoto S, Yamamoto T, Okuyama H, Okada H, Okada N, Kawai M, Saksena NK & Fujii YR. HIV-1 Nef protein in the nucleus influences adipogenesis as well as viral transcription through the peroxisome proliferatoractivated receptors. AIDS 2004; 18:189198. 22. Mahoney DJ, Carey K, Fu MH, Snow R, Cameron-Smith D, Parise G & Tarnopolsky MA. Real-time RT-PCR analysis of housekeeping genes in human skeletal muscle following acute exercise. Physiological Genomics 2004; 18:226231. 23. Tricarico C, Pinzani P, Bianchi S, Paglierani M, Distante V, Pazzagli M, Bustin SA & Orlando C. Quantitative realtime reverse transcription polymerase chain reaction: normalization to rRNA or single housekeeping genes is inappropriate for human tissue biopsies. Analytical Biochemistry 2002; 309:293300. 24. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A & Speleman F. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biology 2002; 3:research0034.10034.11. 25. Sutinen J, Kannisto K, Korsheninnikova E, Fisher RM, Ehrenborg E, Nyman T, Virkamaki A, Funahashi T, Matsuzawa Y, Vidal H, Hamsten A & Yki-Jarvinen H. Effects of rosiglitazone on gene expression in subcutaneous adipose tissue in highly active antiretroviral therapyassociated lipodystrophy. American Journal of Physiology, Endocrinology & Metabolism 2004; 286:E941E949.
8.
9.
10. Nolan D, Hammond E, Martin A, Taylor L, Herrmann S, McKinnon E, Metcalf C, Latham B & Mallal S. Mitochondrial DNA depletion and morphologic changes in adipocytes associated with nucleoside reverse transcriptase inhibitor therapy. AIDS 2003; 17:13291338. 11. Shikuma CM, Hu N, Milne C, Yost F, Waslien C, Shimizu S & Shiramizu B. Mitochondrial DNA decrease in subcutaneous adipose tissue of HIV infected individuals with peripheral lipoatrophy. AIDS 2001; 15:18011809. 12. Mallon PWG, Cooper DA & Carr A. HIV-associated lipodystrophy. HIV Medicine 2001; 2:166173. 13. Kannisto K, Sutinen J, Korsheninnikova E, Fisher RM, Ehrenborg E, Gertow K, Virkamaki A, Nyman T, Vidal H, Hamsten A & Yki-Jarvinen H. Expression of adipogenic transcription factors, peroxisome proliferator-activated receptor gamma co-activator 1, IL-6 and CD45 in subcutaneous adipose tissue in lipodystrophy associated with highly active antiretroviral therapy. AIDS 2003; 17:17531762. 14. Bastard JP, Caron M, Vidal H, Jan V, Auclair M, Vigouroux C, Luboinski J, Laville M, Maachi M, Girard PM, Rozenbaum W, Levan P & Capeau J. Association between altered expression of adipogenic factor SREBP1 in lipoatrophic adipose tissue from HIV-1-infected patients and abnormal adipocyte differentiation and insulin resistance. Lancet 2002; 359:10261031. 15. TRIzol reagent Product Information. Invitrogen Life Technologies, Carlsbad, CA, USA. 16. Bustin SA & Nolan T. Pitfalls of quantitative real-time reverse-transcription polymerase chain reaction. Journal of Biomolecular Techniques 2004; 15:155166.
Received 23 January 2005, accepted 3 June 2005
M108
The role of hepatitis C virus (HCV) in mitochondrial DNA damage in HIV/HCV-coinfected individuals
Carmen de Mendoza and Vincent Soriano*
Department of Infectious Diseases, Hospital Carlos III, Madrid, Spain *Corresponding author: Tel: +34 91 453 2500; Fax: +34 91 733 6614; E-mail: vsoriano@dragonet.es
Oxidative stress accompanying hepatitis C virus (HCV) infection seems to result in mitochondrial (mt) dysfunction. In HIV/HCV-coinfected individuals, HCV-related mt damage could be further enhanced and clinical manifestations of mt damage may appear, particularly following exposure to some antiretroviral drugs. Furthermore, when HCV medications are used together with certain antiretrovirals, the risk
of developing mt adverse events may be particularly frequent, such as development of pancreatitis when ribavirin and didanosine are coadministered. The management of HIV/HCV-coinfected individuals needs to consider the high risk of mitochondria-associated toxicities in this population, which may significantly influence treatment decisions and therapeutic modalities.
Introduction
Around 200 million people are infected with hepatitis C virus (HCV) worldwide. The prevalence of HCV infection among HIV-positive individuals varies widely and mainly depends on the risk category. In regions such as southern Europe, the east coast of North America and South East Asia, where intravenous drug users represent a large proportion of HIV-infected individuals, the rate of HIV/HCV coinfection is extremely high [13]. The involvement of antiretroviral therapy (ART) drugs in mitochondrial (mt) damage, especially nucleoside analogues, has been highlighted in recent years [46]. This is clear even when taking into consideration that HIV infection itself may cause mtDNA depletion in at least some cell compartments such as peripheral blood mononuclear cells (PBMCs) [79]. On the other hand, some co-morbidities may be particularly prone to mt damage in this population. There is no doubt that ARTrelated complications such as lipoatrophy and hyperlactataemia are more frequent among individuals coinfected with HCV [10,11]. In this review, we have tried to summarize the information available regarding the role of HCV in mt damage, especially in the setting of HIV infection. The involvement of HCV drugs and their interactions with antiretrovirals leading to further mt dysfunction in HIV/HCV-coinfected patients will be discussed further. There are many challenges to reproducing in vitro HCV results due to difficulties in producing HCV in viral cultures. For this reason, is still unclear how HCV persists and replicates in hepatocytes or how it produces cellular injury. Ultrastructural studies have shown that mt abnormalities are common in liver biopsies from HCV-infected patients [12]. Mitochondrial alterations due to enhanced oxidative stress caused by HCV infection are associated with a depletion of the tissue glutathione store, which is a key cellular antioxidant [13], and with a reduction in the mtDNA/nuclear DNA ratio. In the light of various findings, Rust & Gores [14] have suggested that HCV could induce mt dysfunction by the following mechanisms: 1) direct effect of viral proteins on mitochondria; 2) inducing a persistent intracellular oxidative stress, which may lead to secondary mt dysfunction; 3) immune-mediated activation of the cell death pathways, such as the Fas system; and 4) stimulation of inflammatory cells that could target mitochondria through further oxidative stress (see Figure 1). Oxidative injury may occur as a direct result of the HCV core protein expression both in vitro and in vivo [15] and may involve a direct effect of the HCV core protein on mitochondria. HCV replication might also lead to an increase in intracellular oxidant production, damaging the mitochondria of infected cells. Accordingly, reactive oxygen species have been associated with disease activity in chronic hepatitis C [16]. In relation to the apoptotic pathways, similar assertions have been made to explain why mtDNA depletion
M109
HCV and mt dysfunction aetiological and pathophysiological aspects

Despite recent advances in the pathogenesis of HCVrelated liver disease, many questions remain unanswered.
C de Mendoza & V Soriano
Figure 1. Four possible mechanisms of HCV-mediated mt dysfunction. Modified from Rust & Gores [14]
3
T-cell receptor NS3
CTL
4
O2
Fas/Fas ligand
Monocyte
Cas-8
Bid
tBid
Apoptosis HCV
1
Viral particles and NS3 proteins
Oxidative stress
CTL, cytotoxic T lymphocyte; mt, mitochondrial.
is seen in drug-naive, HIV-infected individuals and subjects with chronic HCV infection alone [79]. Briefly, mtDNA depletion in drug-naive, HIV-infected individuals seems to be related to cell apoptosis [17,18]. HIV-1 infection may cause apoptosis through different mechanisms, some of which rely on the intricate virushost cell interaction while others involve activation of the hosts inflammatory and immune responses. Soluble HIV-1 products, such as the accessory proteins Tat, Vpr and gp120, directly promote cell death through interactions with CD4, CXCR4 and other uncharacterized receptors [19] or by induction of the caspase cascade [20]. By similar mechanisms, immunemediated apoptotic pathways are activated in chronic HCV infection. Expression of the death receptor Fas is significantly higher in HCV core antigen-positive
M110
hepatocytes compared with uninfected cells [21]. Fas expression promotes the formation of a death complex, which activates the caspase-8 cascade, which eventually leads to apoptotic cell death [22]. Likewise, a recent study has shown that expression of the HCV NS3 protein, or the NS2/NS3 precursor protein, results in induction of apoptosis and activation of the caspase8 cascade [23]. Chronic HCV infection is characterized by inflammatory liver damage and is associated with a significant risk of liver cirrhosis and hepatocellular carcinoma in the long term. There is increasing evidence suggesting that liver cell damage in chronic hepatitis C is mediated by the induction of apoptosis [24]. Indeed, a marked increase in caspase activity is noticed in the sera of HCV-infected patients [25].
Mitochondrial damage due to HCV in HIV infection
Although one quarter of individuals with chronic HCV infection may show persistently normal serum alanine aminotransferase (ALT) levels, inflammatory and fibrotic lesions in the liver may appear in more than half of these patients, which in turn exhibit elevated serum caspase activity. Moreover, 30% of patients with normal ALT levels but elevated caspase activity show a significant fibrosis (F2F4) in the liver biopsy. This is why some authors have claimed that measurement of caspase activity in serum might be a useful indirect tool for assessing liver damage in chronic hepatitis C, particularly in patients with normal ALT levels [25,26]. Although the association between HCV and mt damage seems to be clear, it remains unclear how HCV could cause mt dysfunction. An increased production of free radicals by virus-related inflammation is most probably the main cause of HCV-associated mt damage [27].
Figure 2. Comparison of mtDNA copy numbers in PBMCs from different groups. Effect of HCV infection alone and HIV/HCV coinfection on the depletion of mtDNA [34]
1400 1200 mtDNA copy number 1000 800 600 400 200 0 n= 11 HIV 56 HIV+ 18 HCV+ 107 HIV/HCV coinfection
Clinical manifestations of mt damage in HIV/HCV coinfection

The first mention of mt dysfunction in HIV infection appeared in the early 1990s, when several reports linked zidovudine (AZT)-associated myopathy with mt damage [2830]. Since Brinkman proposed his theory in 1998 [31], mt damage in HIV infection has been directly related to nucleoside reverse transcriptase inhibitor (NRTI) use. However, there is evidence suggesting that HIV infection itself may deplete mtDNA in some cell compartments [7,8], causing clinical symptoms in the absence of any ART [32]. Mitochondrial dysfunction in HIV-infected individuals may be enhanced by the presence of HCV coinfection. Lipoatrophy, pancreatitis, peripheral neuropathy, myopathy, hyperlactataemia and lactic acidosis are the most important adverse events of ART due to mt damage [33]. HCV infection seems to increase the incidence of some of these events. In a transverse study conducted at our institution, coinfection with HCV and/or hepatitis B virus (HBV) was associated with a higher decrease in the mtDNA content of PBMCs than in patients infected with HIV alone [8]. A more detailed investigation showed that HCV infection itself led to a significant depletion of mtDNA in PBMCs, which was more pronounced in patients who underwent treatment with pegylated interferon plus ribavirin (RBV) for chronic hepatitis C [34] (Figure 2).
Groups of infection
mt, mitochondrial; PBMCs, peripheral blood mononuclear cells.
than in patients without lipodystrophy (20%) [11]. Thus, HCV infection may be associated with the atrophic form of lipodystrophy in HIV-infected patients, especially after extended periods of exposure to nucleoside analogues.
Hyperlactataemia and lactic acidosis

Mitochondrial toxicity causing lactic acidosis is a rare but fatal complication, which has been described in some HIV-infected patients exposed to nucleoside analogues. It has been reported in individuals receiving both single and dual nucleoside combinations, including AZT or stavudine (d4T) with didanosine (ddI), zalcitabine (ddC) or lamivudine. The overall prevalence is in the range of five per 1000 patients on therapy per year [35]. The risk of lactic acidosis is somewhat more common in women, the obese and those with viral hepatitis coinfections [30]. Analysing the drugs in use, the combination of d4T plus ddI is associated with the highest risk [8,3537]. In a retrospective study performed in France, nine cases of severe lactic acidosis were identified [38]. The incidence was 0.9/1000 patient-years of exposure to nucleoside analogues. Six patients were coinfected by HCV and/or HBV. More recently, a case of lactic acidosis in an HIVinfected patient receiving ART after liver transplantation for HCV-induced liver disease has been reported [39]. All these observations support HCV increasing the incidence of mitochondria-related complications in the setting of HIV infection.
M111
Lipoatrophy
In a cohort of 226 HIV-infected patients, HCV infection was significantly more frequent in lipoatrophic patients (46%) than in subjects with adiposity or the mixed syndrome (15%). It was also more prevalent
Liver-related mt damage
Grade 3 or 4 liver enzyme elevations occur on average in 510% of patients who initiate triple ART [40]. The rate is significantly higher in HIV-infected individuals with underlying chronic hepatitis C [4145]. Four main mechanisms of drug-related liver toxicity have been recognized in HIV-infected individuals exposed to ART: i) direct drug toxicity, ii) immune reconstitution following initiation of highly active antiretroviral therapy (HAART) in the presence of HCV and/or HBV coinfections, iii) mt toxicity and iv) hypersensitivity reactions with liver involvement. Chronic hepatitis C is known to increase the toxic effects of drugs in the liver, most probably by impairing the mechanisms of cytoprotection in liver cells [46]. This is why exposure to drugs like ritonavir at full doses (600 mg twice daily) is associated with a high risk of hepatotoxicity [43]. Immune reconstitution may explain some episodes of liver enzyme elevations following the initiation of HAART, particularly in severely immunosuppressed patients with chronic hepatitis B or C [4749]. The rationale behind this phenomenon is that the inhibition of HIV replication with HAART leads to rapid immune recovery and, consequently, to an immune response against HBV and/or HCV antigens exposed in the surface of hepatocytes, which are destroyed [50]. Mitochondrial toxicity of some nucleoside analogues is a well-known cause of liver damage [51]. As we have discussed previously, exposure to HIV nucleoside inhibitors (particularly ddI and/or d4T), female sex and obesity are some of the predisposing factors, and underlying chronic HCV infection may increase the chance of this complication [46,5255]. Finally, hypersensitivity reactions to drugs such as nevirapine, abacavir or amprenavir may involve multiple organs including the liver. In this situation, liver enzyme elevations occur within the first days or weeks following the initiation of HAART and tend to resolve upon drug discontinuation. As expected, allergic reactions may be more frequent in patients with high CD4 counts and do not seem to be favoured by the presence of HCV coinfection [56]. HCV may induce mt alterations in the liver of patients with hepatocellular carcinoma [57,58]. Similar ultrastructural liver mt abnormalities were observed in a group of 30 HIV/HCV-coinfected patients under HAART [59]. The main ultrastructural abnormalities of mitochondria were reduction or loss of cristae, decrease in matrix density and glycogen accumulation. There is evidence of mtDNA depletion in the liver of coinfected individuals in association with the use of nucleoside analogues. For instance, the liver mtDNA content was decreased by 47% in HIV/HCV-coinfected individuals taking d4T, ddI or ddC compared with
M112
subjects never exposed to these NRTIs [60]. However, the authors of this study did not observe a clear association between increases in serum lactate levels and the extent of liver mtDNA depletion. Lactate elevations may occur only in the presence of significant reductions in the hepatic mtDNA content. Finally, a higher risk of metabolic abnormalities, including insulin resistance and changes in body composition, have been reported in HIV/HCVcoinfected individuals receiving ART, compared with HIV-monoinfected individuals exposed to the same medications [10,6164].
Interactions between HCV medications and antiretroviral drugs

The combination of pegylated interferon plus RBV, a guanosine nucleoside analogue, is the current recommended therapy for chronic hepatitis C in HIV/HCVcoinfected individuals [65]. Sustained virological response to this combination is around 50% in HCVmonoinfected individuals [60] but much lower in HIV/HCV-coinfected subjects [67,68]. Besides an intrinsically lower efficacy of HCV medication in the HIV setting, the co-administration of HCV medication and antiretroviral drugs has resulted in a substantial number of side effects that often lead to premature drug discontinuation, precluding the achievement of response to HCV therapy. Among the most important drug interactions are those of RBV with ddI and d4T, whose pathogenic mechanism relies on an enhancement of mt toxicity (Figure 3). HIV nucleoside analogues require intracellular phosphorylation to become active. The deoxyribonucleoside kinases are located in the cell cytosol as well as within the mitochondrion. The mechanism of inhibition of phosphorylated nucleoside analogues is by inhibitory competition with the natural nucleoside for
Figure 3. Interactions between RBV and nucleoside analogues AZT ddI d4T
anaemia hepatic pancreatitis decompensation & lactic acidosis
weight loss
mt DNA synthesis
lactate
AZT, zidovudine; ddI, didanosine; d4T, stavudine; RBV, ribavirin.
binding to the HIV reverse transcriptase, ultimately leading to premature DNA chain termination. Nucleoside analogues are also substrates of DNA polymerase , the enzyme responsible for mtDNA replication. This affinity explains the depletion and/or mutation of genes encoded by mtDNA in subjects exposed to these drugs [31,69]. The final consequence of mt dysfunction is a reduction in ATP production and the release of reactive oxygen radicals, which ultimately affect the mitochondrion structure. Of note, RBV is also a nucleoside analogue and might cause mt damage by the same mechanism as the HIV inhibitors, although their affinity for polymerase has not yet been determined. The concomitant use of ddI and RBV is particularly risky as it favours the development of pancreatitis [61,62,70]. RBV inhibits inosine monophosphate (IMP) dehydrogenase, promoting the phosphorylation of ddI [71]. This effect leads to an increase in intracellular and also mt concentrations of ddATP, which inhibits DNA polymerase (Figure 4) [72]. Regarding a different mechanism, a recent report has highlighted a potential synergistic interaction between d4T and RBV, leading to a rapid and severe weight loss in HIVinfected patients receiving both d4T and/or ddI along with HCV medications [10]. Interestingly, more pronounced weight loss, and lactate and amylase elevations were found in patients taking d4T or ddI compared with those taking other nucleoside analogues along with RBV.
Summary
HCV infection seems to be associated with mt alterations either causing it directly or in the context of indirect induction of cell death mediated by oxidative stress. In HIV/HCV coinfection, mt damage could be further enhanced and clinical manifestations may become more apparent when using certain antiretroviral drugs and when HCV medication is used with these antiretrovirals. The management of HIV/HCVcoinfected individuals needs to consider the high risk of mitochondria-related toxicities in this population, which may influence treatment decisions and therapeutic modalities significantly. Given the deleterious influence of HCV on HIV, efforts to eradicate hepatitis C with specific therapy will be of great value in the coinfected population.
Acknowledgements
This work was supported in part by grants from Fundacin Investigacin y Educacin en SIDA (IES), Red de Investigacin en SIDA (RIS project 173), Red de Investigacin en Hepatitis (RTIC G03/015) and the VIRGIL Network.
References
1. Saillour F, Dabis F, Dupon M, Lacoste D, Trimoulet P, Rispal P, Monlun E, Ragnaud JM, Morlat P, Pellegrin JL, Fleury H & Couzigou P. Prevalence and determinants of antibodies to hepatitis C virus and markers for hepatitis B virus infection in patients with HIV infection in Aquitaine. Groupe dEpidemiologie Clinique du SIDA in Aquitaine. British Medical Journal 1996; 313:461464. Martin J, Castilla J, Lpez M, Arranz R, Gonzalez-Lahoz J & Soriano V. Impact of chronic hepatitis C in HIV disease progression. HIV Clinical Trials 2004; 5:125131. Sherman K, Rouster D, Chung R & Rajicic N. Hepatitis C virus prevalence among patients infected with HIV: a crosssectional analysis of the US adult AIDS Clinical Trials Group. Clinical Infectious Diseases 2002; 34:381387. Gerard Y, Maulin L, Yazdanpanah Y, De La Tribonniere X, Amiel C, Maurage CA, Robin S, Sablonniere B, Dhennain C & Mouton Y. Symptomatic hyperlactatemia: an emerging complication of antiretroviral therapy. AIDS 2000; 14:27232730. Walker U, Setzer B & Venhoff N. Increased long-term mitochondrial toxicity in combinations of nucleoside analogues reverse transcriptase inhibitors. AIDS 2002; 16:21652173. Dagan T, Sable C, Bray J & Gerschenson M. Mitochondrial dysfunction and antiretroviral nucleoside analogues toxicities: what is the evidence? Mitochondrion 2002; 1:397412. Cote HC, Brumme ZL, Craib KJ, Alexander CS, Wynhoven B, Ting L, Wong H, Harris M, Harrigan PR, OShaughnessy M & Montaner JS. Changes in mitochondrial DNA as a marker of nucleoside toxicity in HIV infected patients. New England Journal of Medicine 2002; 346:811820. De Mendoza C, de Ronde A, Smolders K & Soriano V. Changes in mitochondrial DNA copy number in blood cells from HIV-infected patients undergoing antiretroviral therapy. AIDS Research & Human Retroviruses 2004; 20:271273.
Figure 4. Metabolic pathways involved in the potentiation of ddI by RBV. The increased concentrations of ddATP due to the inhibition of the IMP dehydrogenase by RBV inhibits the DNA polymerase
2.
3.
4.
RBV RBV-MP
Physiological nucleosides ddI IMP ddIMP ddAMP Inositol ddADP

7. 5.
IMP dehydrogenase
6.
HIV-RT
ddATP
8.
ddI, didanosine; -MP, -monophosphate; HIV-RT, HIV reverse transcriptase; IMP, inosine monophosphate; RBV, ribavirin
M113
9.
10. 11. 12.
13. 14.
15.
16.
17. 18. 19.
20.
21. 22. 23.
24. 25.
26. 27. 28.
Miura T, Goto M, Hosoya N, Odawara T, Kitamura Y, Nakamura T & Iwamoto A. Depletion of mitochondrial DNA in HIV-1 infected patients and its amelioration by antiretroviral therapy. Journal of Medical Virology 2003; 70:497505. Garcia-Benayas T, Blanco F & Soriano V. Weight loss in HIV-infected patients. New England Journal of Medicine 2002; 347:12871288. Zylberberg H, Nalpas B, Pol S, Brechot C & Viard JP. Is there a relationship between hepatitis C virus infection and antiretroviral-associated lipoatrophy? AIDS 2000; 14:2055. Barbaro G, Di Lorenzo G, Asti A, Ribersani M, Belloni G, Grisorio B, Filice G & Barbarini G. Hepatocellular mitochondrial alterations in patients with chronic hepatitis C: ultrastructural and biochemical findings. American Journal of Gastroenterology 1999; 94:21982205. De Leve L & Kaplowitz N. Importance and regulation of hepatic glutathione. Seminars in Liver Disease 1990; 10:251266. Rust C & Gores G. Does hepatitis C cause liver injury by pathways associated with mitochondrial dysfunction? American Journal of Gastroenterology 1999; 94:20032005. Okuda M, Li K, Beard MR, Showalter LA, Scholle F, Lemon SM & Weinman SA. Mitochondrial injury, oxidative stress, and antioxidant gene expression are induced by hepatitis C virus core protein. Gastroenterology 2002; 122:366375. De Maria N, Colantoni A, Fagiuoli S, Liu GJ, Rogers BK, Farinati F, Van Thiel DH & Floyd RA. Association between reactive oxygen species and disease activity in chronic hepatitis C. Free Radical Biology & Medicine 1996; 21:291295. Arnoult D, Petit F, Lelievre J & Estaquier J. Mitochondria in HIV-1 induced apoptosis. Biochemical & Biophysical Research Communications 2003; 304:561574. Badley A, Roumier T, Lum J & Kroemer G. Mitochondrion mediated apoptosis in HIV-1 infection. Trends in Pharmacological Sciences 2003; 24:298305. Twu C. Cardiomyocytes undergo apoptosis in HIV cardiomyopathy through mitochondrion and death receptor controlled pathways. Proceedings of the National Academy of Sciences, USA 2002; 14:386391. Nie Z. HIV protease processes pro-caspase 8 to cause mitochondrial relase of cytochrome C, caspase cleavage and nuclear fragmentation. Cell Death & Differentiation 2002; 9:11721184. Hayashi N & Mita E. Fas system and apoptosis in viral hepatits. Journal of Gastroenterology & Hepatology 1997; 12:223226. Cohen G. Caspases: the executioners of apoptosis. Biochemical Journal 1997; 326:116. Prikhodko EA, Prikhodko GG, Siegel RM, Thompson P, Major ME & Cohen JI. The NS3 protein of hepatitis C virus induces caspase-8-mediated apoptosis independent of its protease or helicase activities. Virology 2004; 329:5367. Bantel H & Schulze-Osthoff K. Apoptosis in hepatitis C virus infection. Cell Death & Differentiation 2003; 10:4858. Bantel H, Lugering A, Heidemann J, Volkmann X, Poremba C, Strassburg CP, Manns MP & Schulze-Osthoff K. Detection of apoptotic caspase activation in sera from patients with chronic HCV infection is associated with fibrotic liver injury. Hepatology 2004; 40:10781087. Feldstein A & Gores G. An apoptotic biomarker goes to the HCV clinic. Hepatology 2004; 40:10441046. Korenaga M, Okuda M, Otani K, Wang T, Li Y & Weinman S. Mitochondrial dysfunction in hepatitis C. Journal of Clinical Gastroenterology 2005; 39:S162S166. Dalakas MC, Illa I, Pezeshkpour GH, Laukaitis JP, Cohen B & Griffin JL. Mitochondrial myopathy caused by long term zidovudine toxicity. New England Journal of Medicine 1990; 322:10981105.
29. Arnaudo E, Dalakas M, Shanske S, Moraes CT, DiMauro S & Schon EA. Depletion of muscle mitochondrial DNA in AIDS patient with zidovudine-induced myopathy. Lancet 1991; 337:508510. 30. Peters BS, Winer J, Landon DN, Stotter A & Pinching AJ. Mitochondrial myopathy associated with chronic zidovudine therapy in AIDS. Quarterly Journal of Medicine 1993; 86:515. 31. Brinkman K, ter Hofstede HJ, Burger DM, Smeitink JA & Koopmans PP. Adverse effects of reverse transcriptase inhibitors: mitochondrial toxicity as a common pathway. AIDS 1998; 12:17351744. 32. Miro O, Lpez S, Martinez E, Pedrol E, Milinkovic A, Deig E, Garrabou G, Casademont J, Gatell J & Cardellach F. Mitochondrial effects of HIV infection on the peripheral blood mononuclear cells of HIV-infected patients who were never treated with antiretrovirals. Clinical Infectious Diseases 2005; 39:710716. 33. Cossarizza A & Moyle G. Antiretroviral nucleoside and nucleotide analogues and mitochondria. AIDS 2004; 18:137151. 34. De Mendoza C, Sanchez-Conde M, Timmermans E & Soriano V. Mitochondrial DNA depletion in HIV-infected patients is enhanced with chronic hepatitis C and treatment with pegylated interferon plus ribavirin. 15th International AIDS Conference. 1116 July 2004, Bangkok, Thailand. Abstract MoPeB3313. 35. Moyle G, Datta D, Mandalia S, Morlese J, Asboe D & Gazzard B. Hyperlactatemia and lactic acidosis during antiretroviral therapy: relevance, reproducibility and possible risk factors. AIDS 2002; 16:13411349. 36. Brinkman K & ter Hofstede H. Mitochondrial toxicity of nucleoside analogue reverse transcriptase inhibitors: lactic acidosis, risk factors and therapeutic options. AIDS Reviews 1999; 1:141148. 37. Boubaker K, Flepp M, Sudre P, Furrer H, Hansel A & Hirschel B. Hyperlactatemia and antiretroviral therapy: the Swiss HIV Cohort study. Clinical Infectious Diseases 2001; 33:19311937. 38. Bonnet F, Bonarek M, Abridj A, Mercie P, Dupon M, Gemain MC, Malvy D, Bernard N, Pellegrin JL, Morlat P & Beylot J. [Severe lactic acidosis in HIV-infected patients treated with nucleoside reverse transcriptase analogues: a report of 9 cases]. Revue de Medecine Interne 2003; 24:1116. French. 39. Antoniades C, Macdonald C, Knisely A, Taylor C & Norris S. Mitochondrial toxicity associated with HAART following liver transplantation in an HIV-infected recipient. Liver Transplantation 2004; 10:699702. 40. Bonnet F, Lawson-Ayayi S, Thiebaut R, Ramanampamonjy R, Lacoste D, Bernard N, Malvy D, Bonarek M, Djossou F, Beylot J, Dabis F & Morlat P; French Aquitaine Cohort. Groupe dEpidemiologie Clinique du SIDA en Aquitaine (GECSA). A cohort study of nevirapine tolerance in clinical practice: French Aquitaine Cohort, 19971999. Clinical Infectious Diseases 2002; 35:12311237. 41. Rodriguez-Rosado R, Garcia-Samaniego J & Soriano V. Hepatotoxicity after introduction of highly active antiretroviral therapy. AIDS 1998; 12:1256. 42. Saves M, Vandentorren S, Daucourt V, Marimoutou C, Dupon M, Couzigou P, Bernard N, Mercie P & Dabis F. Severe hepatic cytolysis: incidence and risk factors in patients treated by antiretroviral combinations. Aquitaine Cohort, France, 19961998. Groupe dEpidemiologie Clinique de Sida en Aquitaine (GECSA). AIDS 1999; 13:F115F121. 43. Sulkowski M, Thomas D, Chaisson R & Moore R. Hepatotoxicity associated with antiretroviral therapy in adults infected with HIV and the role of hepatitis C or B virus infection. Journal of the American Medical Association 2000; 283:7480. 44. Nuez M, Lana R, Mendoza JL & Soriano V. Risk factors for severe liver toxicity following the introduction of HAART. Journal of Acquired Immune Deficiency Syndromes 2001; 27:426431.
M114
45. Aceti A, Pasquazzi C, Zechini B & De Bac C; LIVERHAART Group. Hepatotoxicity development during antiretroviral therapy containing protease inhibitors in patients with HIV. The role of hepatitis B and C virus infection. Journal of Acquired Immune Deficiency Syndromes 2002; 29:4148. 46. Bissell D, Gores G, Laskin D & Hoofnagle J. Drug-induced liver injury: mechanisms and test systems. Hepatology 2001; 33:10091013. 47. John M, Flexman J & French A. Hepatitis C virus-associated hepatitis following treatment of HIV-infected patients with HIV protease inhibitors: an immune restoration disease? AIDS 1998; 12:22892293. 48. Vento S, Garofano T, Renzini C, Casali F, Ferraro T & Concia E. Enhancement of hepatitis C virus replication and liver damage in HIV-coinfected patients on antiretroviral combination therapy. AIDS 1998; 12:116117. 49. Gavazzi G, Bouchard O, Leclercq P, Morel-Baccard C, Bosseray A, Dutertre N, Micoud M & Morand P. Change in transaminases in hepatitis C virus- and HIV-coinfected patients after highly active antiretroviral therapy: differences between complete and partial virologic responders? AIDS Research & Human Retroviruses 2000; 16:10211023. 50. Stone S, Lee S, Keane N, Price P & French M. Association of increased hepatitis C virus (HCV)-specific IgG and soluble CD26 dipeptidyl peptidase IV enzyme activity with hepatotoxicity after highly active antiretroviral therapy in HIV-HCV-coinfected patients. Journal of Infectious Diseases 2002; 186:14981502. 51. McKenzie R, Fried M, Sallie R, Conjeevaram H, Di Bisceglie A & Park Y. Hepatic failure and lactic acidosis due to fialuridine (FIAU), an investigational nucleoside analogue for chronic hepatitis B. New England Journal of Medicine 1995; 333:10991105. 52. Lonergan J, Behling C, Pfander H, Hassanein T & Mathews W. Hyperlactatemia and hepatic abnormalities in 10 HIVinfected patients receiving nucleoside analogue combination regimens. Clinical Infectious Diseases 2000; 31:162166. 53. Coghlan M, Sommadossi J, Jhala N, Many W, Saag M & Johnson V. Symptomatic lactic acidosis in hospitalized antiretroviral-treated patients with HIV infection: a report of 12 cases. Clinical Infectious Diseases 2001; 33:19141921. 54. Barbaro G, Di Lorenzo G, Asti A, Ribersani M, Belloni G & Grisorio B. Hepatocellular mitochondrial alterations in patients with chronic hepatitis C: ultrastructural and biochemical findings. American Journal of Gastroenterology 1999; 94:21982205. 55. De Mendoza C, Snchez-Conde M, Rivera E, Domingo P & Soriano V. Could mitochondrial DNA quantitation be a surrogate marker for drug mitochondrial toxicity? AIDS Reviews 2004; 6:169180. 56. Gonzalez de Requena D, Nuez M, Jimenez-Nacher I & Soriano V. Liver toxicity caused by nevirapine. AIDS 2002; 16:290291. 57. Moriya K, Nakagawa K, Santa T, Shintani Y, Fujie H, Miyoshi H, Tsutsumi T, Miyazawa T, Ishibashi K, Horie T, Imai K, Todoroki T, Kimura S & Koike K. Oxidative stress in the absence of inflammation in a mouse model for hepatitis C virus associated hepatocarcinogenesis. Cancer Research 2001; 61:43654370. 58. Shao JY, Gao HY, Li YH, Zhang Y, Lu YY & Zeng YX. Quantitative detection of common deletion of mitochondrial DNA in hepatocellular carcinoma and hepatocellular nodular hyperplasia.World Journal of Gastroenterology 2004; 10:15601564. 59. Verucchi G, Calza L, Biagetti C, Attard L, Costigliola P, Manfredi R, Pasquinelli G & Chiodo F. Ultrastructural liver
60.
61. 62. 63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
mitochondrial abnormalities in HIV/HCV coinfected patients receiving antiretroviral therapy. Journal of Acquired Immune Deficiency Syndromes 2004; 35:326328. Walker UA, Bauerle J, Laguno M, Murillas J, Mauss S, Schmutz G, Setzer B, Miquel R, Gatell JM & Mallolas J. Depletion of mitochondrial DNA in liver under antiretroviral therapy with didanosine, stavudine, or zalcitabine. Hepatology 2004; 39:311317. Lafeuillade A, Hittinger G & Chadapaud S. Increased mitochondrial toxicity with ribavirin in HIV/HCV coinfection. Lancet 2001; 357:280281. Salmon-Ceron D, Chauvelot-Moachon L, Abad S, Silbermann B & Sogni P. Mitochondrial toxic effects and ribavirin. Lancet 2001; 357:1803. Duong M, Petit JM, Piroth L, Grappin M, Buisson M, Chavanet P, Hillon P & Portier H. Association between insulin resistance and hepatitis C virus chronic infection in HIV-hepatitis C virus coinfected patients undergoing antiretroviral therapy. Journal of Acquired Immune Deficiency Syndromes 2001; 27:245250. Lonardo A, Adinolfi LE, Loria P, Carulli N, Ruggiero G & Day CP. Steatosis and hepatitis C virus: mechanisms and significance for hepatic and extrahepatic disease. Gastroenterology 2004; 126:586597. Soriano V, Puoti M, Sulkowski M, Cargnel A, Dietrich D & Rockstroh J. Care of patients with hepatitis C and HIV co-infection. Updated recommendations from the HIVHCV International Panel. AIDS 2004; 18:112. Fried MW, Shiffman ML, Reddy KR, Smith C, Marinos G, Goncales FL Jr, Haussinger D, Diago M, Carosi G, Dhumeaux D, Craxi A, Lin A, Hoffman J & Yu J. Peginterferon alfa-2a plus ribavirin for chronic hepatitis C virus infection. New England Journal of Medicine 2002; 347:975982. Torriani FJ, Rodriguez-Torres M, Rockstroh JK, Lissen E, Gonzalez-Garcia J, Lazzarin A, Carosi G, Sasadeusz J, Katlama C, Montaner J, Sette H Jr, Passe S, De Pamphilis J, Duff F, Schrenk UM & Dieterich DT; APRICOT Study Group. Peginterferon alfa 2a plus ribavirin for chronic hepatitis C virus infection in HIV-infected patients. New England Journal of Medicine 2004; 351:438450. Carrat F, Bani-Sadr F, Pol S, Rosenthal E, Lunel-Fabiani F, Benzekri A, Morand P, Goujard C, Pialoux G, Piroth L, Salmon-Ceron D, Degott C, Cacoub P & Perronne C; ANRS HCO2 RIBAVIC Study Team. Pegylated interferon alfa-2b vs standard interferon alfa-2b, plus ribavirin, for chronic hepatitis C in HIV-infected patients: a randomized controlled trial. Journal of the American Medical Association 2004; 292:28392848. Lim S, Ponamerev M, Longley M & Copeland W. Structural determinants in human DNA polymerase account for mitochondrial toxicity from nucleoside analogs. Journal of Molecular Biology 2003: 329:4557. Moreno A, Quereda C, Moreno L, Perez-Elias MJ, Muriel A, Casado JL, Antela A, Dronda F, Navas E, Barcena R & Moreno S. High rate of didanosine-related mitochondrial toxicity in HIV/HCV coinfected patients receiving ribavirin. Antiviral Therapy 2004; 9:133138. Balzarini J, Lee C, Herdewijn P & De Clercq E. Mechanism of the potentiating effect of ribavirin on the activity of 2,3-dideoxyinosine against HIV. Journal of Biological Chemistry 1991; 266:2150921514. Soriano V, Sulkowski M, Bergin C, Katlama C, Dietrich D, Poynard T & Rockstroh J. Care of patients with chronic hepatitis C and HIV co-infection: recommendations from the HIV-HCV international panel. AIDS 2002; 16:813828.
M115
Uridine in the prevention and treatment of NRTIrelated mitochondrial toxicity

Ulrich A Walker* and Nils Venhoff
Department of Rheumatology and Clinical Immunology, Medizinische Universittsklinik, Freiburg, Germany *Corresponding author: Tel: +49 761 2703 401; Fax: +49 761 2703 446; E-mail: ulrich.walker@klinikum.uni-freiburg.de
Long-term side effects of antiretroviral therapy are attributed to the mitochondrial (mt) toxicity of nucleoside analogue reverse transcriptase inhibitors (NRTIs) and their ability to deplete mtDNA. Studies in hepatocytes suggest that uridine is able to prevent and treat mtDNA depletion by pyrimidine NRTIs [zalcitabine (ddC) and stavudine (d4T)] and to fully abrogate hepatocyte death, elevated lactate production and intracellular steatosis. Uridine was also found to improve the liver and haematopoietic toxicities of zidovudine (AZT), which are unrelated to mtDNA depletion, and to prevent neuronal cell death induced by ddC. Most recently, uridine was found to prevent the onset of a lipoatrophic phenotype (reduced intracellular lipids, increased apoptosis, mtDNA depletion and mt depolarization) in adipocytes incubated long-term with d4T and AZT. Various steps of mt nucleoside utilization may be involved
in the protective effect, but competition of uridine metabolites with NRTIs at polymerase or other enzymes is a plausible explanation. Pharmacokinetic studies suggest that uridine serum levels can be safely increased in humans to achieve concentrations which are protective in vitro (50200 M). Uridine was not found to interfere with the antiretroviral activity of NRTIs. Mitocnol, a sugar cane extract which effectively increases uridine in human serum, was beneficial in individual HIV patients with mt toxicity and is now being tested in placebo-controlled randomized trials. Until these data become available, the riskbenefit calculation of using uridine should be individualized. The current safety data justify the closely monitored use of uridine in individuals who suffer from mt toxicity but who cannot be switched to less toxic NRTIs.
Introduction
More than 8 years after the widespread introduction of highly active antiretroviral therapy (HAART), it has become clear that antiretroviral drugs have long-term effects on organs and body metabolism. Nucleoside reverse transcriptase inhibitors (NRTIs) within the antiretroviral cocktail are associated with hyperlactataemia and organ toxicities such as damage to the liver, peripheral nerves and skeletal muscle. The choice of NRTI also determines an individuals risk of developing lipoatrophy, a clinically irreversible loss of subcutaneous tissue. The main mechanism of these NRTI-related side effects has been identified as mitochondrial (mt) toxicity [17]. strand. This second step causes chain termination. As a result of polymerase impairment, mtDNA depletion (a quantitative reduction of the mtDNA copy number) ensues. The relative potency of activated nucleoside triphosphates to inhibit polymerase is not the same among all NRTIs. In vitro data indicate a relatively strong inhibitory effect of the d-drugs, that is, zalcitabine (ddC), didanosine (ddI) and stavudine (d4T), whereas abacavir, emtricitabine, lamivudine and tenofovir do not impair mtDNA replication in clinically relevant concentrations [3,8,9]. Zidovudine (AZT) is a special case because this NRTI is a mt toxin despite the fact that AZT triphosphate only has a low potency to affect polymerase and mtDNA content in clinically relevant and cytotoxic concentrations, at least in proliferating cells [811]. On one hand, the mt toxicity of AZT may, in part, involve binding to adenylate kinase (an enzyme involved in ATP formation) and inhibition of the mt ADP/ATP translocator [1214]. These mechanisms may explain why some toxicities have been observed relatively early after AZT exposure [9,13]. On the other hand, mtDNA depletion has indeed been
M117
Pathogenesis of NRTI-related mt toxicity

NRTIs are activated by triphosphorylation and then they inhibit polymerase , the enzyme which replicates mtDNA [3,8]. Polymerase inhibition is a result of several distinct steps [3]. The first step involves competition of NRTI triphosphates with the natural nucleoside triphosphates. If this competition is successful, the NRTIs are incorporated into the nascent mtDNA
UA Walker & N Venhoff
observed with AZT in vivo [6,1517]. Two observations may explain why mtDNA depletion may also occur in the absence of direct polymerase inhibition. Firstly, it has been shown in vivo that some of the administered AZT can be non-enzymatically converted into d4T, and thus a stronger polymerase inhibitor [18]. Secondly, mtDNA depletion may result from a another mechanism, namely from AZT-mediated inhibition of thymidine kinase (TK) type 2 [19]. This TK is expressed in mitochondria and responsible for the intramitochondrial phosphorylation of pyrimidine nucleosides (deoxythymidine, deoxycytidine and deoxyuridine). In non-replicating cells, the cytosolic TK type 1 (TK1) is down-regulated, making the pyrimidine supply for mtDNA synthesis dependent on the activity of TK2. Such reduced supply of the normal deoxypyrimidine phosphates limits mtDNA replication, especially in skeletal muscle, as evidenced by a mt myopathy in subjects carrying TK2 mutations [20]. As mtDNA encodes for subunits of the mt respiratory chain, mtDNA depletion therefore results in respiratory chain dysfunction. Any respiratory chain dysfunction may promote electron leakage in the mt matrix and thus the generation of reactive oxygen species (ROS). Such increased ROS formation may then in turn damage the lipid architecture of the mt membrane, attack respiratory chain proteins or damage polymerase and mtDNA itself, thereby closing several vicious circles that promote even more ROS formation [21,22]. There is also evidence for additional mechanisms of ROS formation [1]. Markers of oxidative damage and heteroplasmic mtDNA point mutations have indeed been shown to increase in patients treated with NRTIs [23,24]. Respiratory chain dysfunction also leads to the secondary impairment of several metabolic pathways. Firstly, ATP can no longer be synthesized efficiently through oxidative phosphorylation and glycolysis has to be relied upon. Secondly, the block of NADH utilization in the respiratory chain increases the intracellular NADH/NAD+ ratio. This alteration of the redox status promotes the conversion of pyruvate to lactate and inhibits key enzymes of beta oxidation, resulting in the intracellular accumulation of triglycerides [25]. The mt respiration also has a third important task: an efficient electron-flux through the respiratory chain is essential for the activity of dihydroorotatedehydrogenase (DHODH; E.C. 1.3.99.11), an enzyme located in the inner mt membrane and necessary for the de novo synthesis of all (intramitochondrial and intracytoplasmic) pyrimidines [26]. This is because DHODH catalyses the oxidation of dihydroorotate to orotate from which uridine monophosphate (UMP) and intracellular pyrimidines are synthesized (Figure 1).
M118
A defect in the respiratory chain therefore results in pyrimidine depletion. The indirect inhibition of DHODH by NRTI-related mt toxicity is likely to be similar to those caused by direct DHODH inhibitors [27]. Research into leflunomide [27,28], a direct DHODH inhibitor and a licensed immunosuppressive drug has taught us about the in vitro and in vivo consequences of DHODH inhibition (Figure 2). The depletion of UMP and derived intracellular pyrimidines activates p53 and its immediate transcriptional target p21 [27,29]. p53 also regulates the activation of Rb protein and thus of cyclins via phosphorylation [30]. Through this mechanism, the pyrimidine depletion inhibits the transition to the S-phase of the cell cycle and leads to a mitotic arrest in the G1 phase. p53 can also activate the transcription of Bax [31] and promote apoptosis. These molecular mechanisms may explain why cells with mtDNA depletion stop dividing and then die. The importance of the intracellular pyrimidine pools for the survival of cells without a functional respiratory chain is supported by the fact that cells without a single molecule of mtDNA (rho0-cells) are rescued from cell
Figure 1. Simplified scheme of pyrimidine metabolism
Carbamoyl phosphate
Dihydroorotic acid mt toxicity leflunomide DHODH Orotic acid Betaalanine
UMP
Uridine
Uracil
UTP
dTTP
dCTP
CTP
RNA UDP sugars
DNA
DNA
RNA CDP lipids
The biosynthetic pathway starts with the formation of carbamoyl phosphate. DHODH (an enzyme which is inhibited by respiratory chain dysfunction in mt toxicity of NRTIs and by leflunomide) then catalyses the synthesis of orotate. Orotate is then anabolized to UMP, which can be used to produce RNA, DNA, glycosylation products or membrane constituents. Uridine can be salvaged into UMP by uridine kinase or degraded into beta-alanine, which enters the tricarboxylic acid cycle. Dashed arrows signify pathways involving intermediate metabolites. mt, mitochondrial; CDP, cytidine diphosphate; DHODH, dihydroorotate dehydrogenase; dCTP, dideoxycytidine triphosphate; dTTP, dideoxythymidine triphosphate; UDP, uridine diphosphate; UMP, uridine monophosphate; UTP, uridine triphosphate.
Uridine in mitochondrial toxicity
Figure 2. Molecular and functional consequences of diminished intracellular pyrimidine supply
Figure 3. Vicious cycle which is hypothesized to contribute to the mt toxicity of antiretroviral pyrimidine d-drugs and which may be abrogated by uridine supplementation
UMP depletion
p63 + p21 activation
pRb not activated
cyclins not activated
mitotic arrest
Inhibition of -polymerase by NRTIs Relative excess of NRTIs over natural pyrimidine Uridine Pyrimidine pool Respiratory chain dysfunction mtDNA depletion
Bax transcription
apoptosis
UMP, uridine monophosphate.
death and grow virtually normally if the intracellular pyrimidine pools are replenished by substances that can be salvaged into pyrimidines by being converted into UMP distal to DHODH. One such substance that can bypass the block in the de novo synthesis of pyrimidines is uridine [32].
Inhibition of uridine synthesis
Inhibition of DHODH
mt, mitochondrial; DHODH, dihydroorotate dehydrogenase; NRTI, nucleoside reverse transcriptase inhibitor.
Uridine abrogates mt toxicity in vitro

The relationship between respiratory chain dysfunction and pyrimidine metabolism makes uridine an attractive candidate to alleviate symptoms of NRTI-related mt toxicity. Early work has demonstrated that neuronal cells exposed to ddC are rescued from death and improve in proliferation and neurite outgrowth if the medium was supplemented with uridine (50 M) [33]. Uridine in concentrations of 50 M also completely reversed the haematopoietic toxicity of AZT (5 M) on normal human granulocytemacrophage progenitor cells [34]. Similar strategies in mouse models of AZT-induced bone marrow suppression reversed anaemia, leucopoenia, increased peripheral reticulocytes and increased bone marrow cellularity [35]. The mechanism for the beneficial action of uridine on AZT is still unclear. As discussed above, AZT has many effects on cell metabolism [8,19]. It is conceivable that uridine or its derived pyrimidines may compete with AZT for one or several of these metabolic steps or, alternatively, for kinases or transporters responsible for the intramitochondrial presence of triphosphorylated AZT [8,34]. Investigations into a model of d-drug-related hepatotoxicity made the surprising discovery that uridine was not only able to prevent cell death (an expected finding), but also to prevent the onset of a severe mtDNA depletion and thereby normalize the synthesis of mtDNA-encoded respiratory chain subunits. This also normalized the rate of lactate production and the intracellular triglyeride content [36]. Importantly, uridine was only able to improve the mtDNA depletion caused by pyrimidine NRTIs, not that caused by purine analogues such as ddI. The ability of uridine to antagonize the polymerase inhibition by pyrimidine d-drugs may be explained by its ability to disrupt the following vicious circle (Figure 3): as discussed above, polymerase inhibition involves competition of NRTIs with the natural nucleotides as a first step. mtDNA depletion, respiratory dysfunction, DHODH inhibition and pyrimidine depletion ensue. The decrease in intracellular pyrimidines most probably allows for a more efficient competition of the exogenous nucleoside analogue at polymerase . Thus, a vicious circle is closed and drives the cell into further mtDNA depletion. We hypothesize that this circle is disrupted by supplying uridine as an exogenous source of intracellular pyrimidines. The data also suggest that the ability of uridine to abrogate mt toxicities was proportional to the concentration of uridine [33,34,36], underlining the hypothesis of a competitive process. Alternatively, uridine may compete with antiretrovirals at steps of intracellular NRTI transport and phosphorylation. Most recently, long-term exposure of adipocytes to d4T (10 M), ddC (0.2 M) or AZT (1 M) was shown to induce a lipoatrophic phenotype consisting of apoptosis, loss of lipids, mtDNA depletion, loss of mtDNA-encoded respiratory chain subunits and disruption of the mt membrane potential [37]. The addition of uridine (200 M) completely abrogated all these effects on adipocytes.
M119
Notably, uridine was not only able to prevent the onset of mt toxicity but also to treat toxicities that were already established [35,36]. Interestingly, in the absence of uridine it took considerably longer for mtDNA depletion to develop (weeks), than it took for uridine to revert such mt toxicity (days) [36]. This relatively quick therapeutic effect of uridine relative to the more prolonged development of mt toxicities may allow for intermittent uridine dosing in order to reset the mitochondrial clock.
Metabolism, pharmacokinetics and safety of uridine in humans

Normal uridine concentrations range from 38 M in human blood plasma, bone marrow and cerebrospinal fluid [38]. Although uridine is part of our everyday food, diet is not an important source of uridine [39,40]. Clinical studies and animal models suggest that uridine is mostly produced in the liver and that erythrocytes serve as carriers for distributing the uridine throughout the body [38]. Exogenous uridine rapidly disappears from plasma (t1/2=2 min), reflecting a concentrative and, under physiological conditions, unsaturated entry into tissue cells, as well as catabolism by the liver [41]. Subsequently, the tissue uridine pools turn over with half-lives of 13 to 18 h [41]. The physiological range of uridine in the human plasma was shown not to completely satisfy the pyrimidine requirements of dividing cells, making some de novo synthesis necessary for optimal proliferation [42]. Circulating uridine may nevertheless be of physiological importance by allowing dividing cells to utilize their salvage pathway [42]. Uridine has several metabolic fates in the cell (Figure 1). Exogenous uridine is rapidly incorporated into nucleotides in nucleated cells [43]. Uridine can be converted to dTTP and dCTP, which are used to produce DNA. UTP is used for the synthesis of RNA. UTP can also be converted into CTP, which upon conjugation of lipids forms cellular membrane constituents such as CDP ethanolamine. In the form of UDP sugars, uridine may help in the production of glycogen and in protein glycosylation. Uridine is degraded into beta-alanine, which can enter the tricarboxilic acid cycle (See Figure 1 for abbreviations) [38]. Pharmacokinetic and safety data for uridine were collected in several human Phase 1 and 2 trials. Parental administration of uridine as a 1 h infusion (8 g/m2) resulted in plasma levels in the millimolar range, far above those required to abrogate NRTI-related mt toxicity [44]. Half-life, volume of distribution (634 ml/kg) and total clearance (4.98 ml/kg/min) of uridine appear to be independent of dose, whereas Cmax and AUC increase with dose in a linear fashion [44]. In subjects given uridine at doses of 212 g/m2 as a single 1 h
M120
infusion, about a fourth of the administered dose was excreted in the urine [44]. Uracil, a uridine catabolite, accounted for 3% of the uridine dose [44]. These intravenous doses were tolerated without side effects. However, transient fever, lasting for 15 min, occurred with higher doses [44,45]. Intermittent infusion schedules with 3 g/m2/h for 3 h, alternating with a 3-h treatment-free interval, resulted in plasma levels of 138335 M during the treatment-free period and were tolerable over the 72-h study period [45]. Parenteral uridine necessitates central venous administration due to the onset of phlebitis if given through a peripheral vein [45]. Uridine can also be administered orally and is generally tolerated without any side effects. However, excessive oral dosing (12 g/m2) is limited by mild and reversible osmotic diarrhoea due to the relatively poor bioavailability of uridine (7%) [46]. Using other pyrimidine precursors, for example, triacetyluridine [34,47,48] or inhibitors of uridine catabolism or excretion may also be envisaged [35,49,50]. Human uridine serum levels can now be effectively increased with mitocnol, a sugar cane extract with a high content (17%) of nucleosides [51]. 24-hour pharmacokinetic data indicate that consuming 36 g of a powder that contains mitocnol increases human uridine serum levels from baseline values (5.6 M) to mean uridine serum concentrations (Cmax) of 152.0 M [51]. Adverse events were not observed. It is recommended that three sachets of mitocnol are taken on three consecutive days per month, taking into account the relatively quick improvement of mt toxicity from in vitro studies. In summary, the current data indicate that uridine concentrations that are protective in vitro can be safely achieved with oral and parenteral dosing. Oral uridine supplementation (150 mg/kg/d) is also recommended and has been safely used long-term in patients with hereditary orotic aciduria, an inborn error of pyrimidine de novo synthesis, in which uridine reverses megaloblastic anaemia and other symptoms [52].
Interaction of uridine with antiretroviral nucleotides

If uridine or its metabolites are able to compete with NRTIs at the level of polymerase , they may also do so at the level of HIV reverse transcriptase (RT). This poses a theoretical risk for the antiretroviral efficacy of nucleoside analogues. The efficiency of RT inhibition is dependent on the ratio between the normal deoxynucleoside triphosphates and the NRTI triphosphates at the enzyme. For example, mycophenolate mofetil, an inhibitor of purine synthesis, depletes intracellular deoxyguanosine triphosphate and decreases plasma HIV-1 RNA in patients treated with the guanosine
analogue abacavir [53,54]. Uridine may thus theoretically have an opposite effect on RT by increasing the normal deoxypyrimidine triphosphates. Such an effect of uridine on the antiretroviral activity of pyrimidine analogues was first analysed with regard to AZT [34]. Phenotypic HIV resistance assays demonstrated that uridine did not interfere with viral suppression [34]. Importantly, uridine did not impair the antiretroviral activity even in a 10 000-fold molar excess, whereas the maximal therapeutic effects of uridine were already achieved with a 10-fold molar surplus. Investigations in mice also came to the same conclusion [35]. The potential interference of uridine with the antiretroviral activity of NRTIs was also extensively examined in phenotypic HIV resistance assays using nucleoside analogues alone and in combinations [55]. Both X-4 tropic and R-5 tropic HIV isolates were tested and three different detection systems including primary human peripheral blood mononuclear cells were used. Uridine was added in concentrations up to 615 M. Additionally, in these investigations no effect of uridine on NRTI-mediated viral suppression was detected. Enhancement of the normal intracellular pyrimidine stores therefore does not seem to have a crucial effect on HIV replication. Taken together, the data suggest that the interaction between uridine and NRTIs in the prevention of mt damage does not necessarily imply a reduced antiretroviral efficacy. Explanations for this selectivity include a separate regulation of mt and cytoplasmic dNTP pools, either at the level of mt transport [56] or by the presence of disparate kinases in both compartments [57]. The differential action of uridine on the mt and antiretroviral replication enzymes may also be caused by differences between the polymerases in selecting the natural nucleotide over the activated NRTI.
d4T was then switched to tenofovir with no subsequent clinical or laboratory abnormalities. Mitocnol is now widely used in Germany. Several and in-part randomized and placebo-controlled clinical trials are currently being conducted to formally analyse whether mitocnol is able to prevent and treat mt toxicities such as lipoatrophy, polyneuropathy, hepatic steatosis and myopathy. Virological failure has not been reported (UA Walker, personal communication).
Perspective
The issues discussed above have several further implications. Uridine supplementation may be used to enhance the therapeutic index of pyrimidine NRTIs and thus allow higher dosing to overcome multidrug resistance in salvage therapy. The available data also suggest the possibility of mtDNA depletion in blood [17,59,60]. If the detected mtDNA depletion in blood secondary to pyrimidine NRTIs indeed also reflects reduced mtDNA copy numbers in lymphocytes and if it exceeded a certain threshold, it would have effects similar to those of the direct DHODH inhibitor leflunomide and of inherited defects in pyrimidine synthesis. From the clinical experience with leflunomide as a licensed immunosuppressive antirheumatic drug, it could then be predicted that the mt toxicity in lymphocytes impairs the proliferation of lymphocytes in response to mitotic stimuli, interferes with CD4 recovery and thus is immunosuppressive. Impaired cellmediated immune responses and reduced CD4 and CD8 lymphocyte subsets were also observed in several patients with an inherited defect in the de novo synthesis of pyrimidines; their immunodeficiency improved upon uridine therapy [52,61]. It was also shown that uridine antagonized the inhibition of lymphocytes by leflunomide [62,63]. Most recent in vitro and in vivo observations in HIV patients also support the view of mt toxicity as being immunosuppressive [11,64,65]. This also offers the potential for uridine to enhance the CD4 cell recovery of patients under antiretroviral treatment. Strategies aimed at increasing uridine also improved symptoms in patients harbouring a qualitative defect in mtDNA by carrying inherited mtDNA mutations [66]. In patients with such mtDNA mutations, however, uridine would be predicted to ameliorate only one aspect of respiratory chain dysfunction, namely the consequences of DHODH inhibition and, in contrast with antiretroviral-treated HIV patients, not to improve the underlying mtDNA pathology. Therefore, patients with mtDNA mutations are likely to have a continued defect in ATP synthesis and hyperlactataemia under uridine. Further clinical data on this group of patients are eagerly awaited.
M121
Uridine in HIV-infected patients

The selective effect of exogenous uridine on NRTIinhibited mtDNA replication, but not on NRTI antiretroviral action, implies that HIV-infected patients under treatment with pyrimidine NRTIs and suffering from mt toxicity may benefit from strategies aimed at increasing uridine. Mitocnol was used in an HIV patient with progressive hyperlactataemia, mt steatohepatitis and symptomatic elevation of creatine kinase (CK) under long-term antiretroviral treatment with d4T [58]. The patient was started on mitocnol (three sachets/day for four consecutive days). After 2 weeks, at his next visit, liver and muscle enzymes, as well as the myalgias had improved rapidly, despite unchanged medication. Lactate had normalized after 7 weeks and HIV replication remained below the limit of detection.
Many aspects of uridine are still poorly understood. For example, an oral dose of 300 mg three times daily for 6 months also improved diabetic neuropathy in a well-conducted trial [67]. Therapeutic uses of uridine were also proposed in cardiovascular disease, hypertension, liver disease and infertility, among others [38]. Until the data from formal clinical studies are available, the riskbenefit calculation of using uridine in HIV-infected patients should be individualized. The current safety data justify the current use of uridine in individuals suffering from mt toxicity who are closely monitored and who cannot be switched to an antiretroviral regimen with a lower potential of mt toxicity.
15.
16.
17.
18.
References
1. 2. Lewis W & Dalakas MC. Mitochondrial toxicity of antiviral drugs. Nature Medicine 1995; 1:417422. Brinkman K, ter Hofstede HJ, Burger DM, Smeitink JA & Koopmans PP. Adverse effects of reverse transcriptase inhibitors: mitochondrial toxicity as common pathway. AIDS 1998; 12:17351744. Lewis W, Day BJ & Copeland WC. Mitochondrial toxicity of NRTI antiviral drugs: an integrated cellular perspective. Nature Reviews Drug Discovery 2003; 2:812822. Walker UA, Buerle J, Laguno M, Murillas J, Mauss S, Schmutz G, Setzer B, Miquel R, Gatell JM & Mallolas J. Depletion of mitochondrial DNA in liver under antiretroviral therapy with didanosine, stavudine, or zalcitabine. Hepatology 2004; 39:311317. Dalakas MC, Semino-Mora C & Leon-Monzon M. Mitochondrial alterations with mitochondrial DNA depletion in the nerves of AIDS patients with peripheral neuropathy induced by 23-dideoxycytidine (ddC). Laboratory Investigation 2001; 81:15371544. Arnaudo E, Dalakas M, Shanske S, Moraes CT, DiMauro S & Schon EA. Depletion of muscle mitochondrial DNA in AIDS patients with zidovudine-induced myopathy. Lancet 1991; 337:508510. Walker UA, Bickel M, Ltke Volksbeck SI, Ketelsen UP, Schfer H, Setzer B, Venhoff N, Rickerts V & Staszewski S. Evidence of nucleoside analogue reverse transcriptase inhibitor-associated genetic and structural defects of mitochondria in adipose tissue of HIV-infected patients. Journal of Acquired Immune Deficiency Syndromes 2002; 29:117121. Kakuda TN. Pharmacology of nucleoside and nucleotide reverse transcriptase inhibitor-induced mitochondrial toxicity. Clinical Therapeutics 2000; 22:685708. Walker UA, Setzer B & Venhoff N. Increased long-term mitochondrial toxicity in combinations of nucleoside analogue reverse-transcriptase inhibitors. AIDS 2002; 16:21652173. Chen CH, Vazquez-Padua M & Cheng YC. Effect of antihuman immunodeficiency virus nucleoside analogs on mitochondrial DNA and its implication for delayed toxicity. Molecular Pharmacology 1991; 39:625628. Setzer B, Schlesier M, Thomas AK & Walker UA. Mitochondrial toxicity of nucleoside analogues in primary human lymphocytes. Antiviral Therapy 2005; 10:327334. Barile M, Valenti D, Passarella S & Quagliariello E. 3Azido-3-deoxythmidine uptake into isolated rat liver mitochondria and impairment of ADP/ATP translocator. Biochemical Pharmacology 1997; 53:913920. Hobbs GA, Keilbaugh SA, Rief PM & Simpson MV. Cellular targets of 3-azido-3-deoxythymidine: an early (non-delayed) effect on oxidative phosphorylation. Biochemical Pharmacology 1995; 50:381390. Barile M, Valenti D, Hobbs GA, Abruzzese MF, Keilbaugh SA, Passarella S, Quagliariello E & Simpson MV. 19.
20. 21. 22.
3. 4.
5.
23.
6.
24.
7.
25. 26.
8. 9.
27. 28.
10.
11. 12.
29. 30. 31. 32.
13.
14.
Mechanisms of toxicity of 3-azido-3-deoxythymidine. Its interaction with adenylate kinase. Biochemical Pharmacology 1994; 48:14051412. Chariot P, Drogou I, de Lacroix-Szmania I, Eliezer-Vanerot MC, Chazaud B, Lombes A, Schaeffer A & Zafrani ES. Zidovudine-induced mitochondrial disorder with massive liver steatosis, myopathy, lactic acidosis, and mitochondrial DNA depletion. Journal of Hepatology 1999; 30:156160. Gerschenson M & Poirier MC. Fetal patas monkeys sustain mitochondrial toxicity as a result of in utero zidovudine exposure. Annals of the New York Academy of Sciences 2000; 918:269281. Shiramizu B, Shikuma KM, Kamemoto L, Gerschenson M, Erdem G, Pinti M, Cossarizza A & Shikuma C. Placenta and cord blood mitochondrial DNA toxicity in HIV-infected women receiving nucleoside reverse transcriptase inhibitors during pregnancy. Journal of Acquired Immune Deficiency Syndromes 2003; 32:370374. Becher F, Pruvost AG, Schlemmer DD, Creminon CA, Goujard CM, Delfraissy JF, Benech HC & Grassi JJ. Significant levels of intracellular stavudine triphosphate are found in HIV-infected zidovudine-treated patients. AIDS 2003; 17:555561. McKee EE, Lynx MD, Susan-Resiga D, Bentley AT, DHaenens JP, Cullen D & Ferguson M. Zidovudine inhibits thymidine phosphorylation: a novel site of potential toxicity in non-mitotic cells. Antiviral Therapy 2005; 9:L9. Saada A, Shaag A, Mandel H, Nevo Y, Eriksson S & Elpeleg O. Mutant mitochondrial thymidine kinase in mitochondrial DNA depletion myopathy. Nature Genetics 2001; 29:342344. Graziewicz MA, Day BJ & Copeland WC. The mitochondrial DNA polymerase as a target of oxidative damage. Nucleic Acids Research 2002; 30:28172824. Corral-Debrinski M, Stepien G, Shoffner JM, Lott MT, Kanter K & Wallace DC. Hypoxemia is associated with mitochondrial DNA damage and gene induction. Implications for cardiac disease. Journal of the American Medical Association 1991; 266:18121816. de la Asuncion JG, del Olmo ML, Sastre J, Millan A, Pellin A, Pallardo FV & Vina J. AZT treatment induces molecular and ultrastructural oxidative damage to muscle mitochondria. Prevention by antioxidant vitamins. Journal of Clinical Investigation 1998; 102:49. Martin AM, Hammond E, Nolan D, Pace C, Den Boer M, Taylor L, Moore H, Martinez OP, Christiansen FT & Mallal S. Accumulation of mitochondrial DNA mutations in human immunodeficiency virus-infected patients treated with nucleoside-analogue reverse-transcriptase inhibitors. American Journal of Human Genetics 2003; 72:549560. Fromenty B & Pessayre D. Inhibition of mitochondrial betaoxidation as a mechanism of hepatotoxicity. Pharmacology & Therapeutics 1995; 67:101154. Lffler M, Jckel J, Schuster G & Becker C. Dihydroorotatubiquinone oxidoreductase links mitochondria in the biosynthesis of pyrimidine nucleotides. Molecular & Cellular Biochemistry 1997; 174:125129. Fox RI. Mechanism of action of leflunomide in rheumatoid arthritis. Journal of Rheumatology 1998; 53(Suppl 1):2026. Cherwinski HM, Cohn RG, Cheung P, Webster DJ, Xu YZ, Caulfield JP, Young JM, Nakano G & Ransom JT. The immunosuppressant leflunomide inhibits lymphocyte proliferation by inhibiting pyrimidine biosynthesis. Journal of Pharmacology & Experimental Therapeutics 1995; 275:10431049. Oren M. Regulation of the p53 tumor suppressor protein. Journal of Biological Chemistry 1999; 274:3603136034. Laliberte J, Yee A, Xiong Y & Mitchell BS. Effects of guanine nucleotide depletion on cell cycle progression in human T lymphocytes. Blood 1998; 91:28962904. Schuler M & Green DR. Mechanisms of p53-dependent apoptosis. Biochemical Society Transactions 2001; 29:684688. Bodnar AG, Cooper JM, Leonard JV & Schapira AH. Respiratory-deficient human fibroblasts exhibiting defective mitochondrial DNA replication. Biochemical Journal 1995; 305:817822.
M122
33. Keilbaugh SA, Hobbs GA & Simpson MV. Anti-human immunodeficiency virus type 1 therapy and peripheral neuropathy: prevention of 2,3-dideoxycytidine toxicity in PC12 cells, a neuronal model, by uridine and pyruvate. Molecular Pharmacology 1993; 44:702706. 34. Sommadossi JP, Carlisle R, Schinazi RF & Zhou Z. Uridine reverses the toxicity of 3-azido-3-deoxythymidine in normal human granulocyte-macrophage progenitor cells in vitro without impairment of antiretroviral activity. Antimicrobial Agents & Chemotherapy 1988; 32:9971001. 35. Calabresi P, Falcone A, St Clair MH, Wiemann MC, Chu SH & Darnowski JW. Benzylacyclouridine reverses azidothymidine-induced marrow suppression without impairment of anti-human immunodeficiency virus activity. Blood 1990; 76:22102215. 36. Walker UA, Venhoff N, Koch E, Olschweski M, Schneider J & Setzer B. Uridine abrogates mitochondrial toxicity related to nucleoside analogue reverse transcriptase inhibitors in HepG2 cells. Antiviral Therapy 2003; 8:463470. 37. Walker UA, Auclair M, Lebrecht D, Kornprobst M, Capeau J & Caron M. Uridine abrogates the adverse effects of stavudine and zalcitabine on adipose cell functions. Antiviral Therapy 2004; 9:L21L22. 38. Connolly GP & Duley JA. Uridine and its nucleotides: biological actions, therapeutic potentials. Trends in Pharmacological Science 1999; 20:218225. 39. Becroft DM, Phillips LI & Simmonds A. Hereditary orotic aciduria: long-term therapy with uridine and a trial of uracil. Journal of Pediatrics 1969; 75:885891. 40. Yamamoto T, Moriwaki Y, Takahashi S, Tsutsumi Z, Ka T, Fukuchi M & Hada T. Effect of beer on the plasma concentrations of uridine and purine bases. Metabolism 2004; 51:13171323. 41. Darnowski JW & Handschumacher RE. Tissue uridine pools: evidence in vivo of a concentrative mechanism for uridine uptake. Cancer Research 1986; 46:34903494. 42. Karle JM, Anderson LW & Cysyk RL. Effect of plasma concentrations of uridine on pyrimidine biosynthesis in cultured L1210 cells. Journal of Biological Chemistry 1984; 259:6772. 43. Hoogenraad NJ & Lee DC. Effect of uridine on de novo pyrimidine biosynthesis in rat hepatoma cells in culture. Journal of Biological Chemistry 1974; 249:27632768. 44. Leyva A, van Groeningen CJ, Kraal I, Gall H, Peters GJ, Lankelma J & Pinedo HM. Phase I and pharmacokinetic studies of high-dose uridine intended for rescue from 5-fluorouracil toxicity. Cancer Research 1984; 44:59285933. 45. van Groeningen CJ, Leyva A, Kraal I, Peters GJ & Pinedo HM. Clinical and pharmacokinetic studies of prolonged administration of high-dose uridine intended for rescue from 5-FU toxicity. Cancer Treatment Reports 1986; 70:745750. 46. van Groeningen C, Peters G, Nadal J, Leyva A, Gall H & Pinedo H. Phase I clinical and pharmacokinetics study of orally administered uridine. Proceedings of the American Association of Cancer Research 1987; 28:195. 47. Bhalla KN, Li GR, Grant S, Cole JT, MacLaughlin WW & Volsky DJ. The effect in vitro of 2-deoxycytidine on the metabolism and cytotoxicity of 2,3-dideoxycytidine. AIDS 1990; 4:427431. 48. Bhalla K, Birkhofer M, Li GR, Grant S, MacLaughlin W, Cole J, Graham G & Volsky DJ. 2-deoxycytidine protects normal human bone marrow progenitor cells in vitro against the cytotoxicity of 3-azido-3-deoxythymidine with preservation of antiretroviral activity. Blood 1989; 74:19231928. 49. Falcone A, Darnowski JW, Ruprecht RM, Chu SH, Brunetti I & Calabresi P. Different effect of benzylacyclouridine on the toxic and therapeutic effects of azidothymidine in mice. Blood 1990; 76:22162221. 50. de Miranda P, Good SS, Yarchoan R, Thomas RV, Blum MR, Myers CE & Broder S. Alteration of zidovudine pharmacokinetics by probenecid in patients with AIDS or AIDS-related complex. Clinical Pharmacology & Therapeutics 1989; 46:494500.
51. Venhoff N, Zilly M, Lebrecht D, Schirmer D, Klinker H, Thoden J, Langmann P & Walker UA. Uridine pharmacokinetics of mitocnol, a sugar cane extract. AIDS 2005; 19:739745. 52. Girot R, Hamet M, Perignon JL, Guesnu M, Fox RM, Cartier P, Durandy A & Griscelli C. Cellular immune deficiency in two siblings with hereditary orotic aciduria. New England Journal of Medicine 1983; 308:700704. 53. Margolis DM, Kewn S, Coull JJ, Ylisastigui L, Turner D, Wise H, Hossain MM, Lanier ER, Shaw LM & Back D. The addition of mycophenolate mofetil to antiretroviral therapy including abacavir is associated with depletion of intracellular deoxyguanosine triphosphate and a decrease in plasma HIV-1 RNA. Journal of Acquired Immune Deficiency Syndromes 2002; 31:4549. 54. Hossain MM, Coull JJ, Drusano GL & Margolis DM. Dose proportional inhibition of HIV-1 replication by mycophenolic acid and synergistic inhibition in combination with abacavir, didanosine, and tenofovir. Antiviral Research 2002; 55:4152. 55. Koch EC, Schneider J, Weiss R, Penning B & Walker UA. Uridine excess does not interfere with the antiretroviral efficacy of nucleoside analogue reverse transcriptase inhibitors. Antiviral Therapy 2003; 8:485487. 56. Rossi L, Serafini S, Schiavano GF, Casabianca A, Vallanti G, Chiarantini L & Magnani M. Metabolism, mitochondrial uptake and toxicity of 2,3-dideoxycytidine. Biochemical Journal 1999; 344:915920. 57. Gallinaro L, Crovatto K, Rampazzo C, Pontarin G, Ferraro P, Milanesi E, Reichard P & Bianchi V. Human mitochondrial 5-deoxyribonucleotidase. Overproduction in cultured cells and functional aspects. Journal of Biological Chemistry 2002; 277:3508035087. 58. Walker UA, Langmann P, Miehle N, Zilly M, Klinker H & Petschner F. Beneficial effects of oral uridine in mitochondrial toxicity. AIDS 2004; 18:10851086. 59. Cot HC, Brumme ZL, Craib KJ, Alexander CS, Wynhoven B, Ting L, Wong H, Harris M, Harrigan PR, OShaughnessy MV & Montaner JS. Changes in mitochondrial DNA as a marker of nucleoside toxicity in HIV-infected patients. New England Journal of Medicine 2002; 346:811820. 60. Cossarizza A. Tests for mitochondrial function and DNA: potentials and pitfalls. Current Opinion in Infectious Diseases 2003; 16:510. 61. Yazaki M, Okajima K, Suchi M, Morishita H & Wada Y. Increase of protein synthesis by uridine supplement in lectinstimulated peripheral blood lymphocytes and EB virus-transformed B cell line of hereditary orotic aciduria type I. Tohoku Journal of Experimental Medicine 1987; 153:189195. 62. Greene S, Watanabe K, Braatz-Trulson J & Lou L. Inhibition of dihydroorotate dehydrogenase by the immunosuppressive agent leflunomide. Biochemical Pharmacology 1995; 50:861867. 63. Silva HT, Cao W, Shorthouse R & Morris RE. Mechanism of action of leflunomide: in vivo uridine administration reverses its inhibition of lymphocyte proliferation. Transplantation Proceedings 1996; 28:30823084. 64. Setzer B, Schlesier M & Walker UA. Functional impairment of NRTI-related mitochondrial DNA-depletion in primary human T-lymphocytes. Journal of Infectious Diseases 2005; 191:848855. 65. Polo R, Martinez S, Madrigal P & Gonzalez-Munoz M. Factors associated with mitochondrial dysfunction in circulating peripheral blood lymphocytes from HIV-infected people. Journal of Acquired Immune Deficiency Syndromes 2003; 34:3236. 66. Naviaux RK, McGowan K, Barshop BA, Nyhan WL & Haas RH. Correction of renal tubular acidosis (RTA) in mitochondrial disease patients treated with triacetyluridine (PN401). Mitochondrion 2001; 1:107. 67. Gallai V, Mazzotta G, Montesi S, Sarchielli P & Del Gatto F. Effects of uridine in the treatment of diabetic neuropathy: an electrophysiological study. Acta Neurologica Scandinavica 1992; 86:37.
Received 14 December 2004, accepted 25 April 2005
M123
HIV-associated antiretroviral toxic neuropathy (ATN): a review of recent advances in pathophysiology and treatment
Mike Youle
Royal Free Centre for HIV Medicine, Royal Free Hospital, London *Corresponding author: Tel: +44 207 941 1835; Fax: +44 207 941 1830; E-mail: mike@mikeyoule.com
Introduction
Nucleoside analogue reverse transcriptase inhibitors (NRTIs) disrupt neuronal mitochondrial DNA synthesis, impairing energy metabolism and resulting in a distal symmetrical polyneuropathy (DSP), categorised as an antiretroviral toxic neuropathy (ATN) that causes significant morbidity in HIV disease [1]. Drugs used to treat many HIV-related conditions, such as isoniazid, vincristine, lithium carbonate, dapsone, pyridoxine (vitamin B6) and thalidomide, and opportunistic illnesses, for example, cytomegalovirus infection and syphilis, may also cause neuropathy. Other possible causes of these symptoms include alcohol and some recreational drugs such as heroin, cocaine, or amphetamines. It seems that these different causes of neuropathy can interact. People with mild neuropathy symptoms often find that they intensify after they start taking treatments such as didanosine (ddI) or stavudine (d4T). Neuropathy has been associated with a deficiency of the B vitamins, especially vitamin B12. Factors associated with the development of peripheral neuropathy in a large cohort of HIV-infected patients include older age, diabetes and white race. whose severity is associated with elevated plasma HIV1 RNA [7] levels, and which may be unresponsive to analgesia, even in combination with anticonvulsants [1], tricyclic antidepressants [3], mexiletine [3], or gabapentin [8]. Withdrawal of certain NRTIs often becomes necessary [1,4] since there are no licensed effective therapies, although lamotrigine [9] and recombinant nerve growth factor (rhNGF) [10] have shown some benefit. Ideally a pathogenesis-based treatment for ATN would allow patients to continue NRTI therapy, still the keystone of current highly active antiretroviral therapy regimes [11]. ATN is thought to result from disrupted mitochondrial oxidative metabolism [6,12] secondary to reduction in neuronal mitochondrial DNA content [6,13,14]. Consequently, neurons are unable to meet the metabolic requirements of their long peripheral axons which undergo die-back, resulting in the glove and stocking distribution of ATN [1]. In keeping with this die-back hypothesis, epidermal innervation is reduced in ATN [15,16].
Symptomatic treatment
If the neuropathy is caused by a drug, symptoms usually occur after a few weeks. Depending on severity of the neuropathy and other treatments available, the patient may be advised to stop or to reduce the dosage of the offending drug. The main purpose of treating neuropathy is to relieve symptoms. In mild cases, where the symptoms are not affecting daily life, standard painkillers such as ibuprofen may be all that is necessary. If the symptoms start to be disruptive, tricyclic antidepressants such as amitriptyline may help. These drugs can take a couple of weeks to show any effects, and may cause side-effects of a dry mouth, difficulty
M125
Pathophysiology
Antiretroviral toxic neuropathy (ATN) is the commonest HIV-associated DSP [1] causing significant morbidity in 1035% of HIV positive patients [24], and occurring in 1166% of patients [1,5] on NRTI drug therapy. Dideoxynucleotide analogue agents such as zalcitabine (ddC), d4T and ddI are implicated in its pathogenesis [1,4,6] and changes in the prescribing pattern of these drugs, including dose reduction, have led to incidence reduction in antiretroviral toxic neuropathy (ATN) [4]. A feature of HIV-associated neuropathy is development of dysaesthetic pain [4]
M Youle
urinating, high blood pressure and drowsiness. People with severe symptoms may require stronger pain killers such as fentanyl or other opiates. The anti-convulsants, phenytoin and carbamazepine, can be useful. La Spina and co-workers assessed the efficacy and safety of gabapentin as a sole analgesic in patients with HIV-related painful neuropathy [8]. Nineteen patients with HIV-related painful neuropathy were administered gabapentin. Efficacy was evaluated with two 100 mm visual analogue scales (VAS) (0: no symptom; 100: worst symptom), rating pain and interference of pain with sleep, performed at baseline and monthly intervals. Mean pain VAS score decreased from a baseline of 55.7 19.1 mm to a final 14.7 18.6 mm (P<0.0001) and mean sleep interference VAS score decreased from a baseline of 60.4 31.9 mm to a final 15.5 27.7 mm (P<0.0001). Hahn [17] carried out a multicentre, prospective, randomised, double-blind placebo-controlled study of gabapentin in 26 patients with HIV sensory neuropathy. Fifteen patients received gabapentin at 400 mg per day before being increased to 1200 mg per day over 2 weeks. This dose was maintained or increased to 2400 mg per day if not beneficial. There was a significant decrease in pain score in the gabapentin group (-44%) but not the placebo group (n=11; -30%). Sleep interference score decreased in the gabapentin group (-49%) but not the placebo group (-12%). Somnolence was reported in 80% of the gabapentin group. Capsaicin has been found to be effective in relieving pain associated with other neuropathic pain syndromes, and is mentioned as a possible topical adjuvant analgesic for the relief of DSP [18]. This multicentre, controlled, randomized, double-masked clinical trial studied patients with HIV-associated DSP and compared measures of pain intensity, pain relief, sensory perception, quality of life, mood and function for patients who received topical capsaicin to the corresponding measures for patients who received the vehicle only. Twenty-six subjects were enrolled in the study. At the end of 1 week, subjects receiving capsaicin tended to report higher current pain scores than did subjects receiving the vehicle (P=0.042). The dropout rate was higher for the capsaicin group (67%) than for the vehicle group (18%) (P=0.014). A further study showed a 40% reduction in pain in capsaicin treated individuals [19].
Pathophysiologic treatments
Vitamin B12 deficiency is a cause of neuropathy and some individuals with HIV have such a deficiency, which may cause symptoms such as fatigue, poor memory and low levels of red blood cells. A vitamin B complex supplement may be taken to relieve vitamin
M126
B12 deficiency although overdosing of vitamin B6 may exacerbate nerve damage. Recombinant human (rh) nerve growth factor (NGF) has been used to treat HIVrelated and diabetic peripheral neuropathy. One study in HIV-associated neuropathy found that NGF reduced pain and improved sensitivity [20]. A total of 270 patients with HIV-associated sensory neuropathy (SN) were randomized to receive placebo, 0.1 g/kg rhNGF, or 0.3 g/kg rhNGF by doubleblinded subcutaneous injection twice weekly for 18 weeks. The primary outcome was a change in selfreported neuropathic pain intensity (Gracely pain scale). In a subset, epidermal nerve fibre densities were determined in punch skin biopsies. Both doses of NGF produced significant improvements in average and maximum daily pain compared with placebo. Positive treatment effects were also observed for global pain assessments (P=0.001) and for pin sensitivity (P=0.019). No treatment differences were found with respect to mood, analgesic use or epidermal nerve fiber densities. Injection site pain was the most frequent adverse event, and resulted in un-blinding in 39% of subjects. Schifitto [21] reported that symptoms of pain improved in 200 people with HIV-associated distal symmetrical polyneuropathy (DSP) who were treated with neurotrophin (NGF) for 48 weeks in an open-label study. However, there was no improvement as measured by neurologic examination, quantitative sensory testing, and epidermal nerve fibre density. Following disappointing results in treating diabetic neuropathy, Genentech have ceased development of NGF [22]. Research into diabetic neuropathy has suggested that supplements of -linolenic acid (GLA), -lipoic acid, magnesium and chromium may help relieve pain due to neuropathy, although none of these supplements have been tested among people with HIV-related neuropathy. A proposed therapeutic agent for DSP is acetyl-Lcarnitine (ALCAR), the acetyl ester of L-carnitine. ALCAR is vital for normal mitochondrial function, being a transport molecule for free fatty acids, and an important acetyl-group donor in high energy metabolism and free fatty acid -oxidation [23]. Also, ALCAR potentiates NGF actions [24], promotes peripheral nerve regeneration [25], is neuroprotective in vitro [26], in vivo [27] and in animal models of diabetic neuropathy [28]. ALCAR has analgesic properties, possibly mediated by increasing ACTH and -endorphin levels [29], whilst its non-acetylated form, Lcarnitine, also has favourable immunological benefits [30] in HIV infection. ALCAR has shown improvement in pain scores and electrophysiologic parameters in a placebo controlled study in diabetics with DSP [31]. Short term ALCAR treatment [32] has shown symptomatic benefits in ATN, and further studies are
Pathophysiology and treatment of ATN
underway to define whether this effect is long-lasting, due to neuronal regeneration, or merely to a result of analgesia. Hart and co-workers reported on 21 patients with NRTI-associated DSP treated with 1500 mg ALCAR twice daily for up to 18 months [33]. Median baseline CD4 cell count was 286 cells/mm3 and 40% had HIV RNA <400 copies/ml. Eight started the study with grade 1 neuropathy, 10 with grade 2 and three with grade 3. Skin biopsies were taken from the leg at baseline and at months 6 and 12 and stained with antibodies for nerve fibres. The innervation of the epidermis increased by 34% and by 101% after 6 and 12 months. In the dermis, it increased by 65% and around the sweat glands by 75% after 6 months (Figure 1). Nerve fibre density approached normal levels after 6 months. The greatest increase was in small sensory nerve fibres, with 100% increase in the epidermis (P=0.006) and 133% in the dermis (P<0.001). Neuropathic pain improved in 76% of patients. No side-effects of ALCAR were experienced.
Figure 1. Plots of the mean of all treated individuals mean fractional immunostaining for protein gene product 9.5 in each area of cutaneous innervation at each time point up to 18 months
Patients on ALCAR Control median values Control lower range Control upper range
Fractional immunostaining (x10-2)
0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0
PGP epidermis
3 6 9 12 15 Time since commencement of ALCAR (months)
18
Discussion
Quantification of cutaneous innervation has previously been found to correlate with clinical tests of sensory function in leprosy [34] and diabetic neuropathy [32], where during disease progression, immunohistochemical changes precede those of sensory testing [31]. The morphology of cutaneous innervation in DSP has previously been described qualitatively, and the density of epidermal fibres shown to be reduced along a proximal-distal gradient in the limb in DSP [14] and ATN [15]. Epidermal nerve fibre density has recently been described as a therapeutic outcome measure in HIV [16], however fibre counts do not assess the health of surviving fibres, whose atrophy or regeneration can be determined by immunostaining area quantification, as used in this study. In established ATN, the symptoms reflect a cutaneous innervation reduction of the lower leg, with epidermal innervation being most affected, consistent with the dieback hypothesis of sensory neuropathy [4]. There is also marked atrophy of dermal and sweat gland plexi. This cutaneous denervation may explain why neuropathic symptoms frequently do not begin to resolve for some weeks after starting ALCAR treatment, and may then continue to improve for many months, since this timeframe matches the slow rate at which peripheral nerves regenerate. As expected from the ATN dysaesthetic algesic symptomatology, small sensory (C, A) fibres are most affected, and it has been demonstrated that these fibres show the greatest reduction in immunostaining area when compared to controls [33]. Furthermore, the
1.2 1 0.8 0.6 0.4 0.2 0 0
PGP dermis
18
8 7 6 5 4 3 2 1 0 0
PGP sweat gland
18
All observations defined as occurring at 6 months occurred in the window from 39 months after starting ALCAR. Similar windows were created for 12 (915) and 18 (1524) months. All averages are non-parametric (medians and ranges).
proportion of small sensory and structural fibres was markedly reduced in neuropathic patients epidermis (ATN 4%; control 36%), implying loss of epidermal fibres and reduced function of the surviving sensory fibres. Six months of oral ALCAR treatment resulted in significant increases in the innervation in epidermis,
M127
M Youle
dermis and sweat glands, an improvement maintained throughout the treatment for all patients. The results demonstrate that dermal PGP-immunostaining nerve fibres regenerated sufficiently to reach the range found in normal skin. Intra-epidermal fibres also regenerated, although more gradually, suggesting a temporal relationship with improvement. ALCAR treatment may counteract dideoxyribonucleotide NRTI toxicity (ATN) by several mechanisms. Firstly, it may reduce mitochondrial DNA damage by direct antioxidant effect [34,35]. Secondly, by promoting glucose utilization and high-energy substrate oxidative metabolism, ALCAR improves neuronal metabolic capacity [36]. In addition ALCAR may facilitate distal neurotrophic support, particularly of myelinated (A) and unmyelinated (C) fibres. Furthermore, ALCAR promotes peripheral nerve regeneration [31,37] independently of NRTI toxicity and function [3841]. Patients with NRTI-associated peripheral neuropathy also have reduced serum ALCAR levels, contrary to asymptomatic HIV positive controls [42]. Although nucleoside analogue antiretroviral agents have been partly superceded by other agents such as protease inhibitors, they remain fundamental for combination therapy [4,11] and are likely to remain important components in HIV management for the foreseeable future [4]. Peripheral neuropathy has been the principal complication limiting the use of these agents [1,4] and pathogenesis-based therapies such as ALCAR may be promising effective management approaches, allowing patients to remain on NRTI therapy.
inhibitors: mitochondrial toxicity as common pathway. AIDS 1998; 12:17351744. 7. Simpson DM, Haidich AB, Schifitto G, Yiannoutsos CT, Geraci AP, McArthur JC & Katzenstein DA. Severity of HIV-associated neuropathy is associated with plasma HIV-1 RNA levels. AIDS 2002; 16:407412. La Spina I, Porazzi D, Maggiolo F, Bottura P & Suter F. Gabapentin in painful HIV-related neuropathy: a report of 19 patients, preliminary observations. European Journal of Neurology 2001; 8:7175. Simpson DM, McArthur JC, Olney R, Clifford D, So Y, Baird BJ, Barrett P & Hammer AE. Lamotrigine for HIVassociated painful peripheral sensory neuropathies: a placebo controlled trial. Neurology 2003; 60:15081514.
8.
9.
10. McArthur JC, Yiannoutsos C, Simpson DM, Adornato BT, Singer EJ, Hollander H, Marra C, Rubin M, Cohen BA, Tucker T, Navia BA, Schifitto G, Katzenstein D, Rask C, Zaborski L, Smith ME, Shriver S, Millar L, Clifford DB & Karalnik IJ. A phase II trial of nerve growth factor for sensory neuropathy associated with HIV infection. AIDS Clinical Trials Group Team 291. Neurology 2000; 54:10801088. 11. Pozniak A, Gazzard B, Anderson J, Babiker A, Churchill D, Collins S, Fisher M, Johnson M, Khoo S, Leen C, Loveday C, Moyle G, Nelson M, Peters B, Phillips A, Pillay D, Wilkins E, Williams I & Youle M. British HIV Association (BHIVA) guidelines for the treatment of HIV-infected adults with antiretroviral therapy: BHIVA Writing Committee on behalf of the BHIVA Executive Committee. July 2003 [cited June 2005]. Available from: http://www.bhiva.org/guidelines/2003/hiv/index.html 12. Chen CH, Vazquez-Padua M & Cheng YC. Effect of antihuman immunodeficiency virus nucleoside analogs on mitochondrial DNA and its implication for delayed toxicity. Molecular Pharmacology 1991; 39:625628. 13. Keilbaugh SA, Prusoff WH & Simpson MV. The PC12 cell as a model for studies of the mechanism of induction of peripheral neuropathy by anti-HIV-1 dideoxynucleoside analogs. Biochemical Pharmacology 1991; 42:R5R8. 14. Lewis W & Dalakas MC. Mitochondrial toxicity of antiviral drugs. Nature Medicine 1995; 1:417422. 15. McCarthy BG, Hsieh ST, Stocks A, Hauer P, Macko C, Cornblath DR, Griffin JW & McArthur JC. Cutaneous innervation in sensory neuropathies: evaluation by skin biopsy. Neurology 1995; 45:18481855. 16. Polydefkis M, Yiannoutsos CT, Cohen BA, Hollander H, Schifitto G, Clifford DB, Simpson DM, Katzenstein D, Shriver S, Hauer P, Brown A, Haidich AB, Moo L & McArthur JC. Reduced intraepidermal nerve fiber density in HIV-associated sensory neuropathy. Neurology 2002; 58:115119. 17. Hahn K, Arendt G, Braun JS, von Giesen HJ, Husstedt IW, Maschke M, Straube ME & Schielke E; German NeuroAIDS Working Group. A placebo-controlled trial of gabapentin for painful HIV-associated sensory neuropathies. Journal of Neurology 2004; 251:12601266. 18. Paice JA, Ferrans CE, Lashley FR, Shott S, Vizgirda V & Pitrak D. Topical capsaicin in the management of HIVassociated peripheral neuropathy. Journal of Pain & Symptom Management 2000; 19:4552. 19. Simpson D, Brown S, Sampson J, Estanislao L, Vilahu C, Ramanathan S, Jermano J & von Stein T. Novel highconcentration capsaicin patch for the treatment of painful HIV an associated distal symmetrical polyneuropathy: results of an open label trial. 11th Conference on Retroviruses & Opportunistic Infections. 811 February 2004, San Francisco, CA, USA. Abstract 490. 20. McArthur JC, Yiannoutsos C, Simpson DM, Adornato BT, Singer EJ, Hollander H, Marra C, Rubin M, Cohen BA, Tucker T, Navia BA, Schifitto G, Katzenstein D, Rask C, Zaborski L, Smith ME, Shriver S, Millar L, Clifford DB & Karalnik IJ. A phase II trial of nerve growth factor for sensory neuropathy associated with HIV infection. AIDS Clinical Trials Group Team 291. Neurology 2000; 54:10801088.
References
1. Simpson DM & Tagliati M. Nucleoside analogue-associated peripheral neuropathy in human immunodeficiency virus infection. Journal of Acquired Immune Deficiency Syndromes & Human Retrovirology 1995; 9:153161. Hall CD, Snyder CR, Messenheimer JA, Wilkins JW, Robertson WT, Whaley RA & Robertson KR. Peripheral neuropathy in a cohort of human immunodeficiency virusinfected patients. Incidence and relationship to other nervous system dysfunction. Archives of Neurology 1991; 48:12731274. Kieburtz K, Simpson D, Yiannoutsos C, Max MB, Hall CD, Ellis RJ, Marra CM, McKendall R, Singer E, Dal Pan GJ, Clifford DB, Tucker T & Cohen B. A randomized trial of amitriptyline and mexiletine for painful neuropathy in HIV infection. AIDS Clinical Trial Group 242 Protocol Team. Neurology 1998; 51:16821688. Moyle GJ & Sadler M. Peripheral neuropathy with nucleoside antiretrovirals: risk factors, incidence and management. Drug Safety 1998; 19:48194. Moore RD, Wong WM, Keruly JC & McArthur JC. Incidence of neuropathy in HIV-infected patients on monotherapy versus those on combination therapy with didanosine, stavudine and hydroxyurea. AIDS 2000; 14:273278. Brinkman K, ter Hofstede HJ, Burger DM, Smeitink JA & Koopmans PP. Adverse effects of reverse transcriptase
2.
3.
4.
5.
6.
M128
Pathophysiology and treatment of ATN
21. Schifitto G, Yiannoutsos C, Simpson DM, Adornato BT, Singer EJ, Hollander H, Marra CM, Rubin M, Cohen BA, Tucker T, Koralnik IJ, Katzenstein D, Haidich B, Smith ME, Shriver S, Millar L, Clifford DB & McArthur JC; AIDS Clinical Trials Group Team 291. Long-term treatment with recombinant nerve growth factor for HIV-associated sensory neuropathy. Neurology 2001; 57:13131316. 22. Apfel SC. Nerve growth factor for the treatment of diabetic neuropathy: what went wrong, what went right, and what does the future hold? International Review of Neurobiology 2002; 50:393413. 23. Bremer J. The role of carnitine in intracellular metabolism. Journal of Clinical Chemistry & Clinical Biochemistry 1990; 28:297301. 24. Manfridi A, Forloni GL, Arrigoni-Martelli E & Mancia M. Culture of dorsal root ganglion neurons from aged rats: effects of acetyl-L-carnitine and NGF. International Journal of Developmental Neuroscience 1992; 10:321329. 25. Hart AM, Wiberg M & Terenghi G. Pharmacological enhancement of peripheral nerve regeneration in the rat by systemic acetyl-L-carnitine treatment. Neuroscience Letters 2002; 334:181185. 26. Virmani MA, Biselli R, Spadoni A, Rossi S, Corsico N, Calvani M, Fattorossi A, De Simone C & Arrigoni-Martelli E. Protective actions of L-carnitine and acetyl-L-carnitine on the neurotoxicity evoked by mitochondrial uncoupling or inhibitors. Pharmacological Research 1995; 32:383389. 27. Hart AM, Wiberg M, Youle M & Terenghi G. Systemic acetyl-L-carnitine eliminates sensory neuronal loss after peripheral axotomy: a new clinical approach in the management of peripheral nerve trauma. Experimental Brain Research 2002; 145:182189. 28. Sima AA, Ristic H, Merry A, Kamijo M, Lattimer SA, Stevens MJ & Greene DA. Primary preventive and secondary interventionary effects of acetyl-L-carnitine on diabetic neuropathy in the bio-breeding worcester rat. Journal of Clinical Investigation 1996; 97:19001907. 29. Onofrj M, Fulgente T, Melchionda D, Marchionni A, Tomasello F, Salpietro FM, Alafaci C, De Sanctis E, Pennisi G, Bella R, Ventura F & Morganti A. L-acetylcarnitine as a new therapeutic approach for peripheral neuropathies with pain. International Journal of Clinical Pharmacology Research 1995; 15:915. 30. Famularo G, De Simone C & Cifone G. Carnitine stands on its own in HIV infection treatment. Archives of Internal Medicine 1999; 159:11431144. 31. De Grandis D & Minardi C. Acetyl-L-carnitine (levacarnine) in the treatment of diabetic neuropathy. A long-term
randomized, double-blind, placebo controlled study. Drugs in R&D 2002; 3:223231. 32. Scarpini E, Sacilotto G, Baron P, Cusini M & Scarlato G. Effect of acetyl-L-carnitine in the treatment of painful peripheral neuropathies in HIV+ patients. Journal of the Peripheral Nervous System 1997; 2:250252. 33. Hart AM, Wilson AD, Montovani C, Smith C, Johnson M, Terenghi G & Youle M. Acetyl-L-carnitine: a pathogenesis based treatment for HIV-associated antiretroviral toxic neuropathy. AIDS 2004; 18:15491560. 34. Calvani M & Arrigoni-Martelli E. Attenuation by acetyl-Lcarnitine of neurological damage and biochemical derangement following brain ischemia and reperfusion. International Journal of Tissue Reactions 1999; 21:16. 35. Tesco G, Latorraca S, Piersanti P, Piacentini S, Amaducci L & Sorbi S. Protection from oxygen radical damage in human diploid fibroblasts by acetyl-L-carnitine. Dementia 1992; 3:5860. 36. Peluso G, Benatti P, Nicolai R, Reda E & Calvani M (1998) Carnitine system and insulin resistance. In Insulin Resistance, Metabolic Disease & Diabetic Complications; pp. 259274. Edited by Crepaldi G, Tiengo A & Del Prato S. Elsevier. 37. Fernandez E, Pallini R, Gangitano C, Del Fa A, Sangiacomo CO, Sbriccoli A, Ricoy JR & Rossi GF. Effects of L-carnitine, L-acetylcarnitine and gangliosides on the regeneration of the transected sciatic nerve in rats. Neurological Research 1989; 11:5762. 38. Kano M, Kawakami T, Hori H, Hashimoto Y, Tao Y, Ishikawa Y & Takenaka T. Effects of ALCAR on the fast axoplasmic transport in cultured sensory neurons of streptozotocin-induced diabetic rats. Neuroscience Research 1999; 33:207213. 39. Stevens MJ, Lattimer SA, Feldman EL, Helton ED, Millington DS, Sima AA & Greene DA. Acetyl-L-carnitine deficiency as a cause of altered nerve myo-inositol content, Na,K-ATPase activity and motor conduction velocity in the streptozotocin-diabetic rat. Metabolism 1996; 45:865872. 40. De Grandis D, Santoro L & Di Benedetto P. L-acetylcarnitine in the treatment of patients with peripheral neuropathies. Clinical Drug Investigation 1995; 10:317322. 41. Giammusso B, Morgia G, Spampinto A & Motta M. Improved pallesthetic sensitivity of pudendal nerve in impotent diabetic patients treated with acetyl-L-carnitine. Archivio Italiano di Urologia 1996; 10:185187. 42. Famularo G, Moretti S, Marcellini S, Trinchieri V, Tzantzoglou S, Santini G, Longo A & De Simone C. Acetyl-carnitine deficiency in AIDS patients with neurotoxicity on treatment with antiretroviral nucleoside analogues. AIDS 1997; 11:185190.
Received 29 March 2005, accepted 14 June 2005
M129
SPEAKER
Aiuti, Fernando Blanche, Stephane Brinkman, Kees Casademont, Jordi Cauda, Roberto Cossarizza, Andrea
ORGANISATION
Division of Allergy and Clinical Immunology, University of Rome "La Sapienza", Rome, Italy Unite d'Immunologie-Hematologie Pediatrique, Hopital Necker Enfants Malades, Paris, France Department of Internal Medicine, Onze Lieve Vrouwe Gasthuis, Amsterdam, The Netherlands Hospital Clnic, Department of Internal Medicine, University of Barcelona Medical School, Barcelona, Spain Infectious Diseases Clinics, Catholic University, Rome, Italy Department of Biomedical Sciences, Section of General Pathology, University of Modena and Reggio Emilia, Modena, Italy Department of Pathology & Laboratory Medicine, University of British Columbia, Vancouver, B.C., Canada Infectious Diseases Clinics, University of Modena and Reggio Emilia, Modena, Italy Infectious Diseases Clinics, L. Sacco Hospital University of Milan, Milan, Italy Director of Molecular Medicine and Infectious Diseases, Hawaii AIDS Clinical Research Program, John A. Burns School of Medicine, University of Hawaii at Manoa, Honolulu, Hawaii Infectious Diseases Clinics, University of Modena and Reggio Emilia, Modena, Italy Vita Salute San Raffaele University, Clinic of Infectious Diseases, Milano, Italy Director of Cardiovascular Pathology, Emory University School of Medicine, Robert W. Woodruff Health Sci. Ctr., Atlanta, Georgia Center for Clinical Immunology and Biomedical Statistics, Royal Perth Hospital, Wellington St, Perth, Western Australia National Centre in HIV Epidemiology and Clinical Research, University of New South Wales, Sydney, Australia Dipartimento del Farmaco, Istituto Superiore di Sanit viale Regina Elena, Roma, Italy Clinical Institute of Infections and Immunology, Institu d'Investigacions Biomediques August Pi i Sunyer, Hospital Clinic, Barcelona, Spain Mitochondrial Research Laboratory, Department of Internal Medicine, Hospital Clinic, Barcelona, Spain Infectious Diseases Clinics, L. Sacco Hospital, University of Milan, Milan, Italy Infectious Diseases Clinics, University of Modena and Reggio Emilia, Modena, Italy Department of Clinical Infectious Diseases, University of Torino, Torino, Italy Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands Belozersky Institute of Physico-Chemical Biology, Moscow State University, Moscow, Russia Department of Infectious Diseases, Hospital Carlos III, Madrid, Spain
M131
Ct, Hlne Esposito, Roberto Galli, Massimo Gerschenson, Mariana
Guaraldi, Giovanni Lazzarin, Adriano Lewis, William
Mallal, Simon
Mallon, Patrick W.
Malorni, Walter Milinkovic, Ana
Mir, Oscar Moroni, Mauro Mussini, Cristina Di Perri, Gianni Reiss, Peter Skulachev, Vladimir Soriano, Vincent
SPEAKER (continued)
Vigan, Alessandra Walker, Ulrich A.
ORGANISATION
Pediatrics Clinics, L. Sacco Hospital, Milan, Italy Medizinische Universittsklinik, Department of Rheumatology and Clinical Immunology, Freiburg, Germany Director HIV Clinical Research, Royal Free Hospital, London, UK
Youle, Mike
M132

Toxicidad Mitocondrial

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Toxicidad Mitocondrial

Uploaded by

Copyright:

Available Formats

ANTIVIRAL THERAPY

William Bonnez Stephen Locarnini

Antiviral Therapy 10, Supplement 2

ORGANIZING & SCIENTIFIC COMMITTEE

The meeting was supported by an unrestricted educational grant from:

This supplement was supported by:

Antiviral Therapy 10, Supplement 2

Thursday, May 19, 2005

Friday, May 20, 2005

Saturday, May 21, 2005

Cristina Mussini Ulrich Walker Peter Reiss, Andrea Cossarizza

Antiviral Therapy 10, Supplement 2

Modena, Italy 1921 May 2005

Modena, Italy 1921 May 2005

Antiviral Therapy 10, Supplement 2

Modena, Italy 1921 May 2005

Limitations of HIV mt toxicity research

Potential pathways of mt toxicity (Figure 1)

ATP endogenous nucleosides

kinases nucleoside analogues

Antiviral Therapy 10, Supplement 2

Mitochondrial nucleic acids in HIV therapy

Antiviral Therapy 10, Supplement 2

Mitochondrial nucleic acids in HIV therapy

mtDNA damage (mutation, deletion)

Nucleotide pool imbalance

Other potential mt toxicity mechanisms

HIV mtDNA damage

The vicious circle of mt damage

Mitochondrial nucleic acids in HIV therapy

Mitochondrial nucleic acids in HIV therapy

Received 10 January 2005, accepted 19 April 2005

Modena, Italy 1921 May 2005

Nucleoside reverse transcriptase inhibitors, mitochondrial DNA and AIDS therapy

Introduction and overview

Predisposition to NRTI toxicity

Mitochondrial DNA replication defects

Table 1. Published reports of the mitochondrial toxicity of individual NRTIs

Anatomic or tissue target

Liver, heart, skeletal muscle and bone marrow

Antiviral Therapy 10, Supplement 2

3TC muscle toxicity [195]

Liver, skeletal and cardiac muscle, peripheral nerve

Clinical testing with surrogate marker

NRTIs, mtDNA and AIDS therapy

Assessing NRTI toxicity

Nucleotide pools, NRTIs and mtDNA replication

NRTIs, mtDNA and AIDS therapy

suggested (including, for example, glutathione (GSH) depletion [129]).

Overview of current NRTI therapy

Lamivudine (3TC) and emitracibine [()FTC]

NRTIs, mtDNA and AIDS therapy

10. 11. 12. 13.

Supported by DHHS, NIH HL072707 and HL063666.

23. 24. 25. 26. 27.

50. 51. 52. 53.

NRTIs, mtDNA and AIDS therapy

Antiviral Therapy 10, Supplement 2

NRTIs, mtDNA and AIDS therapy

Antiviral Therapy 10, Supplement 2

NRTIs, mtDNA and AIDS therapy

NRTIs, mtDNA and AIDS therapy