HIV-associated lipodystrophy (LD) is a common adverse effect during antiretroviral therapy. It is characterized by the depletion of subcutaneous fat,1 often in association with visceral adiposity, hyperlipidemia,2 and insulin resistance.3,4 Loss of limb fat occurs most rapidly in regimens containing the nucleoside reverse transcriptase inhibitor (NRTI) stavudine (d4T), although fat loss may also occur during zidovudine (ZDV)-based therapy.5 Some of the effects on adipose tissue are thought to be caused by inhibition of mitochondrial DNA (mtDNA) polymerase-γ.6 By contrast, some protease inhibitors (PIs) have clearly been shown to impede the differentiation of murine preadipocytes7 and to induce insulin resistance in vivo.3,4 Some PIs can affect the transcription of genes involved in the differentiation of adipocytes8; specifically, sterol regulatory element-binding protein (SREBP1c), perilipin, and hormone-sensitive lipase (HSL).9 Biopsies taken from peripheral adipose tissue in patients with LD receiving PI-containing regimens had a 93% reduction in SREBP1c messenger RNA (mRNA) levels and an increase in apoptotic subcutaneous adipocytes when compared with healthy controls.10
The phenotypic and metabolic abnormalities in LD suggest that adipose tissue is a major site for drug-induced events, although the mechanisms for these effects remain unclear.8 In addition to storing triglyceride, adipose tissue is an endocrine organ capable of exerting profound effects on adipocyte biology via the secretion of adipokines.8,11 Altered production of adipokines has been implicated in the metabolic abnormalities associated with obesity11 and type 2 diabetes12 as well as with LD more recently. In particular, adipose tissue secretes tumor necrosis factor-α (TNFα), interleukin (IL)-6, leptin, resistin, and the newly identified protein adiponectin.12 Adiponectin, in contrast to the other adipokines, has insulin-sensitizing effects13 and is downregulated in patients with LD,14,15 correlating with markers of insulin resistance.14,16 Conversely, the mRNA expression and secretion of TNFα are upregulated in patients with LD.10,16 To date, adipokine expression has only been examined in patients treated with both PIs and NRTIs, leading to the conclusion that PIs are responsible for these adipokine alterations.14
To evaluate the role of antiretroviral drug choice on adipokines, we assessed plasma and/or tissue mRNA expression of the adipokines adiponectin, TNFα, IL-6, and 2 SREBP1 isoforms in HIV-positive persons on a range of treatment regimens as well as in HIV-negative controls. In addition, we have examined mtDNA levels and the expression of mtDNA-encoded protein in NRTI-treated PI-naive patients with LA in comparison to HIV-negative controls.
The design of the study is shown in Figure 1. Five groups of male subjects were examined in this study (see Fig. 1). Evidence of self-reported lipoatrophy (LA) was confirmed by investigators in both population groups before enrollment on the basis of loss of fat in the face and extremities or increased vein prominence, as per the LD case definition study-screening questionnaire.17 The first 2 groups consisted of PI-naive patients with LA receiving NRTI treatment (NRTI+LA+) recruited from the outpatient clinic at St. Stephens Centre, Chelsea and Westminster Hospital in London. These patients consisted of individuals who had only ever received d4T-based (d4T+LA+) or ZDV-based (ZDV+LA+) therapy. PI plus NRTI-treated patients with LA (HAART+LA+) and control patients treated with HAART but without clinically evident LA (HAART+LA−) were recruited from the Infectious Disease Department, University Central Hospital, Helsinki. The HAART+LA− patients were specifically selected on the basis that they had no evidence of any alterations in body composition over the duration of antiretroviral therapy. All HIV-positive groups had received therapy (listed previously) for at least 28 months. Finally, HIV-negative controls were recruited from staff members at the Chelsea and Westminster Hospital (HIV-negative). The Institutional Review Board at the Chelsea and Westminster Hospital and the Local Research Ethics Committee at University Central Hospital, Helsinki, approved the study protocols before commencement. Informed consent was obtained from all participants.
Fasting blood samples were obtained 1 hour before the biopsies, and glucose, total cholesterol, high-density lipoprotein (HDL), calculated low-density lipoprotein (LDL), and triglycerides were analyzed using standard methods. Additionally, HIV-1 RNA load and CD4 cell count were assessed in all participants, including the HIV-negative controls.
Insulin sensitivity was assessed from fasting plasma insulin and glucose using the homeostasis model. The formula for HOMA is HOMA = insulin (μU/mL) × glucose (mmol/L)/22.5.18 Peripheral blood mononuclear cells (PBMCs) were isolated by density gradient centrifugation on lymphoprep and used as a surrogate marker for mtDNA depletion and function in the NRTI and HIV-negative control groups.
Subcutaneous adipose biopsies were obtained from the superior aspect of the gluteal region under local anesthesia from the PI-naive NRTI+LA+ and HIV-negative control groups. After excision, the fat biopsies were cleaned of any visible blood before being snap-frozen in liquid nitrogen. Biopsies were stored at −80°C until extraction for mtDNA quantification or RNA extraction in Freiburg and Liverpool, respectively. Needle aspiration biopsies for the highly active antiretroviral therapy (HAART)-treated subjects were obtained from abdominal subcutaneous adipose tissue under local anesthesia and cleaned of any visible blood before being snap-frozen in liquid nitrogen.19
Total RNA and Complementary DNA Preparation
Total RNA from fat (50-150 mg) was extracted using TRIZOL reagent (Invitrogen, Paisley, UK), followed by a spin column cleanup with DNase treatment (Qiagen, Hilden, Germany) for the London-based groups. Correspondingly, frozen fat tissue from the HAART groups from Helsinki was homogenized in 2 mL of RNA STAT-60 (Tel Test, Friendswood, TX), and total RNA was isolated according to the manufacturer's instructions. All RNA was quantified by spectrophotometry and stored at −80°C until use. A total of 0.1 μg of RNA was transcribed into complementary DNA (cDNA) using a commercially available kit containing random hexamers (Applied Biosystems, Warrington, UK). In addition, a reaction containing 0.5 μg of RNA was also reverse-transcribed to create a standard. After cDNA synthesis, the standard was serially diluted at a 1:2 ratio in DNase-free water to create a standard curve, which was then used to quantitate the relative amount of each target.
Real-Time Polymerase Chain Reaction Analysis of Adipokine and Transcription Factors
TNFα, IL-6, SREBP1a/1cm, and adiponectin mRNA levels were analyzed by real-time quantitative polymerase chain reaction (PCR) using the ABI 7000 sequence detection system. The design of adiponectin primers and probes was carried out using the assay by design service from Applied Biosystems (Foster City, CA), whereas TNFα, β-actin, and IL-6 were purchased via the assays on demand service (Applied Biosystems). For quantitation of SREBP1a and SREBP1c isoforms, the same reverse primer and fluorogenic probe but different forward primers were used. The sequences were as follows: SREBP1c forward: 5′-CCATGGATTGCACTTTCGAA-3′, SREBP1a forward: 5′-TGCTGACCGACATCGAAGAC-3′, reverse: 5′-CCAGCATAGGG TGGGTCAAA-3′, and probe: 5′-FAM-TATCAACA ACCAAGACAG TGACTTCCCTGGC-3′-TAMRA.20
Each reaction was carried out in duplicate in a 20-μL reaction containing 2 μL of cDNA, 1× PCR Master Mix, and 1× probe/primer mixture for the assays on demand. For the SREBP1a and SREBP1c analysis, forward (900 and 300 nmol/L for SREBP1a and SREBP1c, respectively) and reverse (900 and 300 nmol/L for SREBP1a and SREBP1c, respectively) primers and Taqman probe (125 and 50 nmol/L for SREBP1a and SREBP1c, respectively) were added to each reaction. All reactions were carried out using standard cycling parameters: 50°C for 2 minutes and 95°C for 15 minutes, followed by 45 cycles of 95°C for 15 seconds and 60°C for 1 minute. Quantification of the target mRNA used the standard curve method on log-transformed data. Intra- and interassay coefficients of variation for each target ranged from 0.25% to 3.8% and from 0.6% to 4.2%, respectively.
Mitochondrial DNA Quantification
Total DNA was extracted with the QIAamp DNA isolation kit (Qiagen). mtDNA and nuclear DNA (nDNA) copy numbers were determined by quantitative PCR using the ABI 7700 sequence detection system (Applied Biosystems). The mtDNA ATP-6 gene was amplified between nucleotide positions 462 and 540 with the forward primer 5′-ACCAATAGCCCTGGCCGTAC-3′ and the backward primer 5′-GGTGGCGCTTC-CAATTAGGT-3′. mtDNA was quantified with the FAM-fluorophore-labeled probe, 5′-6FAM-CCTAACCGCTAACATTACTGCAGGCCACC-TAMRA-3′. For the detection of nDNA, exon number 8 of the GAPDH gene was selected between nucleotide positions 4280 and 4342, using the forward primer 5′-CGGGGCTCTCCAGAACATC-3′ and the backward primer 5′-ATGACCTTGCCCACAGCCT-3′. In this case, a VIC-fluorophore-labeled probe, 5′-VIC-CCCTGCCTCTACTGGCGCTGCC-TAMRA-3′, was used.
Each 25-μL reaction contained 25 ng of genomic DNA, 100 nM of probe, 200 nM of primers, and Taqman Universal Master Mix (Applied Biosystems). Amplifications of mitochondrial and nuclear products were separately performed in optical 96-well plates (Applied Biosystems). An initial incubation at 50°C for 2 minutes was followed by 10 minutes at 95°C and 40 denaturing steps at 95°C (15 seconds), alternating with combined annealing and extension at 60°C (1 minute). All samples were run in triplicate. Measurements of mtDNA and nDNA copy numbers were standardized using serial dilutions of plasmids with known copy numbers.21 The variation of a single (nontriplicate) repeat measurement was 20%, and the interassay variation of triplicates was 5%, based on non-logged values.
Mitochondrial DNA-Encoded Respiratory Chain Protein
The subunit II of cytochrome c-oxidase (COX) is encoded by mtDNA, whereas the subunit IV of COX is encoded by nDNA. COXII was quantified by immunoblot and normalized to the signal of a simultaneously used antibody against COXIV as described elsewhere.21,22 The intensities of the signals were densitometrically quantified using Scion-image (Scion Corporation, Frederick, MD).
Plasma adiponectin was measured by enzyme-linked immunosorbent assay (ELISA; B-Bridge International, San Jose, CA). TNFα and IL-6 concentrations were determined by ELISAs from R&D Systems (Abingdon, UK).
Independent t tests were used to compare differences between the groups within each study cohort using logarithmic-transformed data on SPSS (SPSS, Chicago, IL). Data are shown as the mean ± standard error of the mean (SEM). A P value less than 0.05 was considered statistically significant.
Demographics: London Groups
The characteristics of the subjects are presented in Tables 1 and 2. Patients from the Chelsea and Westminster Hospital were offered participation in the study if they had been treated with ZDV (ZDV+LA+ [n = 6]) or d4T (d4T+LA+ [n = 12]) for at least 28 months. Patients were on ZDV or d4T and dual therapy or triple therapy with a nonnucleoside reverse transcriptase inhibitor (NNRTI) as the third agent.
Demographics: Helsinki Groups
Most of the HAART+LA+ (n = 8) group had been treated with d4T for at least 32 months (see Table 2); however, apart from 1 subject, they had not been exposed to NNRTIs. All patients had been treated with PIs, with 4 patients on ritonavir in a boosted or unboosted regimen. HAART-treated patients who had not developed LD (HAART+LD− [n = 8]) during antiretroviral therapy were also recruited. Four patients had been treated with PIs, whereas the remainder had been treated with NNRTIs. Six patients were currently receiving d4T, whereas the remainder had been treated with ZDV. Significant differences were evident for PI duration between HAART+LA+ patients and HAART patients who did not develop LA (see Table 2). Cholesterol was significantly elevated in the HAART+LA+ group when compared with the other groups. Overall, there were no statistical differences in body mass index (BMI), age, viral load, CD4 count, duration of HIV infection, or fasting glucose measurements between the treated groups (see Table 2).
Adiponectin Plasma and Messenger RNA Expression
Plasma adiponectin levels were reduced (see Fig. 2A) in the ZDV+LA+ (2.06 ± 0.7; P < 0.01) and d4T+LA+ (1.73 ± 0.7; P < 0.01) groups when compared with the HIV-negative (4.10 ± 0.6) control group. The same trend (see Fig. 2B) was apparent when comparing the HAART+LA+ group (see Fig. 2B) with the HAART+LA− control group (3.33 ± 0.9 vs. 8.88 ± 1.2; P < 0.01).
Adiponectin mRNA expression in subcutaneous adipose tissue was 44% lower in the d4T+LA+ group (P < 0.05) when compared with the HIV-negative control group (0.62 ± 0.13 vs. 1.10 ± 0.13; P < 0.05). No difference was evident in the ZDV+LA+ patients when compared with the HIV-negative control group (Fig. 3A). A significant reduction was also evident in the HAART+LA+ patient group (see Fig. 3B) when compared with the HAART+LA− controls (0.35 ± 0.5 vs. 1.04 ± 0.5; P < 0.01).
Cytokine Messenger RNA and Protein Expression
No differences were evident between the LA subgroups and controls for TNFα at the protein (see Table 1) or mRNA levels (data not shown).
SREBP1c/1a Messenger RNA Levels
Tissue SREBP1c mRNA levels were significantly different from those of HIV-negative controls in both NRTI treatment groups (Fig. 4A); specifically, d4T (−70%; P < 0.05) induced a greater reduction than ZDV (−54%; P < 0.05). Marked reductions were seen in the HAART+LA+ group when compared with the HAART+LA− group (see Fig. 4B). SREBP1c was expressed at a greater level than the 1a isoform (data not shown). SREBP1a mRNA levels were not significantly different between the groups (data not shown).
Expression of Mitochondrial DNA-Encoded Protein
As indicated in Figure 4A, d4T+LA+ patients showed a significant reduction in the COXII/COXIV ratio (60% reduction vs. controls; P < 0.001). ZDV+LA+ patients also exhibited a similar significant reduction (53% reduction vs. controls; P < 0.01), although no differences were evident between the 2 NRTI groups. In contrast, no significant differences were evident for adipose or PBMC mtDNA levels in both treatment groups in comparison to HIV-negative controls (see Figs. 5B-D).
In vivo studies of adipose tissue from HIV-infected patients with LD have repeatedly shown reductions in adipocyte adiponectin mRNA and protein.14,15,23 Such studies, however, have included patients treated with combinations of NRTIs and PIs, making it difficult to assess which component of HAART is responsible for these alterations. In vitro evidence has supported the hypothesis that PIs are largely responsible for reduced adiponectin levels.16 Our results suggest that NRTIs and PIs both contribute to reduced adipocyte secretion and mRNA expression of adiponectin, although the PI+LA+ group had a greater reduction when compared with the HAART+LA− group. Our findings of a 44% decrease in the mRNA and a 50% decrease in adiponectin secretion in NRTI-treated PI-naive groups is comparable to reductions reported in PI-treated patients,16,24 although the absence of a PI-alone group studied prospectively makes it difficult to assess the relative contribution of PI therapy to reduced adiponectin levels, because patients in the HAART+LA+ group were treated with both PIs and NRTIs in combination.
These findings are in accordance with 2 recent studies, which have implicated d4T as working as an additive factor to PI therapy in the hypoadiponectinemia observed in patients with LA.14,24 It is possible that the low levels reported in our study as well as in other studies may simply reflect the reduction in adipose tissue rather than a direct effect of the drugs on adiponectin secretion. In support of this, analysis of patients with congenital LD found similarly low adiponectin levels.15 This suggests a response to the LA per se rather than a pathologic instigating mechanism.12 Treatment with agents such as the peroxisome proliferator-activated receptor-γ (PPARγ) agonist rosiglitazone increases adiponectin levels25 and improves insulin sensitivity without increasing fat mass.26 These data suggest that low adiponectin levels may contribute to insulin resistance in persons on antiretroviral therapy.27 Whether NRTI-induced reductions in adiponectin contribute to this reduction in insulin sensitivity remains unknown, although recent data from a randomized trial in initial therapy suggest that differences exist between the use of d4T plus didanosine (ddI) relative to abacavir plus lamivudine (3TC) in this regard.28
Having established that adiponectin is reduced in NRTI- and PI-treated patients, we explored whether both drug classes were associated with reduced expression of SREBP1c. Analysis of the 2 NRTI treatment groups revealed that d4T was associated with a greater reduction (70%) compared with a 54% reduction in the ZDV group when compared with the HIV-negative control group. Previously, several PIs have been implicated in impairing SREBP function,10 a key adipocyte regulator of various genes involved in lipid and glucose metabolism in vitro. Specifically, SREBP activates the differentiation program of the adipocyte by stimulating the expression of PPARγ. Previously, Bastard et al10 implicated PPARγ and SREBP as key targets of PI therapy after a 75% and 93% decrease in their expression, respectively, in subcutaneous fat from patients with LA. Because SREBP controls the expression of PPARγ, it is likely that any alterations have a knock-on effect on PPARγ.20 In support of this, we found reduced expression of the SREBP1c isoform in HAART+LA+ patients, who had been heavily exposed to PIs in comparison to the HAART+LA− group, although this group had a shorter treatment duration of NRTIs. It is therefore difficult without prospectively designed trials to conclude whether SREBP1c alterations eventually lead to clinical symptoms of LA. LA has previously been shown to be a progressive event related to length and combination of therapies, specifically d4T use.29 Previously, Bastard et al10 only included an HIV-negative control group, thereby not enabling a PI effect to become evident. Because of extensive use of d4T within the HAART+LA+ group, however, we were unable to separate an NRTI-associated effect from a PI-associated effect, although we present the first evidence that NRTI use in PI-naive patients is associated with reduced SREBP1c mRNA expression.
A recent study showed that 2 weeks' administration of ZDV/3TC or d4T/3TC in HIV-negative individuals was associated with a 51% and 65% reduction in the mRNA expression of PPARγ, respectively,30 which seemed to occur independent of any alterations in SREBP, despite a 45% and 10% reduction in the ZDV-treated and d4T-treated subgroups, respectively. We were unable to measure PPARγ mRNA levels; however, it is likely that the effects on SREBP after NRTI treatment may become more pronounced after longer treatment duration or as a direct consequence of the peripheral fat wasting. Alternatively, the finding by Mallon et al30 that the NRTIs exert profound effects quite acutely could indicate that the primary defect observed in LA occurs at the molecular level through PPARγ or SREBP.
The mechanisms for LA in NRTI-treated patients have centered around mitochondrial toxicity.6,22 The speed of onset and severity of LA are typically more apparent in d4T-treated patients than in ZDV-treated patients.5 Only d4T has been implicated in reduced expression of PPARγ.30 In the present study, only modest nonsignificant reductions in mtDNA in either of the NRTI subgroups were witnessed, although significant alterations were evident in the COXII/COXIV ratios. Various groups have reported profound reductions in mtDNA, with a recent study finding a 87% and 52% reduction in d4T- and ZDV-treated patients, respectively.28,31 In the study of Nolan et al,32 the adipose samples were treated with collagenase, thereby removing any confounding effects from other cell types. In the present study, whole adipose samples were used for the mitochondrial assays, potentially leading to a nonhomogeneous sample. Data derived using the same methodology for handling adipose tissue samples have, however, previously noted significant mtDNA depletion,22 whereas studies using PBMCs as a surrogate marker of mtDNA depletion seem to be far less precise in predicting mitochondrial toxicity than studies using adipose samples.33,34 In the present study, significant alterations were noted only for the COXII/COXIV ratio in adipose samples, suggesting that the degree of mitochondrial toxicity was not as profound as seen in other reports.
TNFα has a structural homology similar to adiponectin12 and is increased in subcutaneous adipose and plasma samples from patients with HAART-associated LD.10,16 We found no differences in TNFα or IL-6 levels in patients in comparison to controls, with no differences evident between LA subgroups. Indeed, although Vigouroux et al24 recently found upregulated levels of TNFα in 21 patients with LA, these changes were relatively small and could perhaps be reflective of an altered immune state. Similarly, a recent study also failed to find any association between LA and altered TNFα and IL-6 levels.35 Previous studies that have found upregulated levels have often included patients treated with PIs or patients with LA as well as hypertrophy within their study populations. A limitation of the present study, however, was our failure to measure soluble TNFα or TNFα receptor levels, which are thought to be better indicators of adipose tissue, as well as systemic production of TNF.35 Another limitation of this study relates to the cross-sectional design of the study. Two different patient populations were examined for this study: the first was undergoing NRTI treatment in London, whereas the second was treated in Helsinki. Analysis was conducted on population-matched controls, although the biopsies were taken from different anatomic locations, which may contribute to the low expression of SREBP1c in the PI treatment groups from Finland. In contrast, however, the adiponectin mRNA (but not plasma expression) was remarkably similar in the HIV-negative controls and the HIV-positive HAART+LA− patient group. The increased level of plasma adiponectin may therefore be reflective of population differences between the 2 cohorts.
In summary, our data show that patients with LA treated with NRTI-based regimens have evidence of profound changes in the secretion and mRNA expression of adiponectin and SREBP1c. Further studies are required to delineate whether these findings instigate fat redistribution and metabolic abnormalities or merely reflect loss of peripheral fat.
1. Carr A, Samaras K, Chisholm DJ, et al. Abnormal fat distribution and use of protease inhibitors. Lancet
2. Saint-Marc T, Partisani M, Poizot-Martin I, et al. A syndrome of peripheral fat wasting (lipodystrophy) in patients receiving long-term nucleoside analogue therapy. AIDS
3. Behrens G, Dejam A, Schmidt H, et al. Impaired glucose tolerance, beta cell function and lipid metabolism in HIV patients under treatment with protease inhibitors. AIDS
4. Vigouroux C, Gharakhanian S, Salhi Y, et al. Diabetes, insulin resistance and dyslipidaemia in lipodystrophic HIV-infected patients on highly active antiretroviral therapy (HAART). Diabetes Metab
5. Saves M, Chene G, Ducimetiere P, et al. Risk factors for coronary heart disease in patients treated for human immunodeficiency virus infection compared with the general population. Clin Infect Dis
6. Walker UA, Brinkman K. NRTI induced mitochondrial toxicity as a mechanism for HAART related lipodystrophy: fact or fiction? HIV Med
7. Caron M, Auclair M, Sterlingot H, et al. Some HIV protease inhibitors alter lamin A/C maturation and stability, SREBP-1 nuclear localization and adipocyte differentiation. AIDS
8. Gougeon ML, Penicaud L, Fromenty B, et al. Adipocytes targets and actors in the pathogenesis of HIV-associated lipodystrophy and metabolic alterations. Antivir Ther
9. Rudich A, Vanounou S, Riesenberg K, et al. The HIV protease inhibitor nelfinavir induces insulin resistance and increases basal lipolysis in 3T3-L1 adipocytes. Diabetes
10. Bastard JP, Caron M, Vidal H, et al. 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
11. Arner P. The adipocyte in insulin resistance: key molecules and the impact of the thiazolidinediones. Trends Endocrinol Metab
12. Haque WA, Garg A. Adipocyte biology and adipocytokines. Clin Lab Med
13. Hotta K, Funahashi T, Arita Y, et al. Plasma concentrations of a novel, adipose-specific protein, adiponectin, in type 2 diabetic patients. Arterioscler Thromb Vasc Biol
14. Addy CL, Gavrila A, Tsiodras S, et al. Hypoadiponectinemia is associated with insulin resistance, hypertriglyceridemia, and fat redistribution in human immunodeficiency virus-infected patients treated with highly active antiretroviral therapy. J Clin Endocrinol Metab
15. Kosmiski L, Kuritzkes D, Lichtenstein K, et al. Adipocyte-derived hormone levels in HIV lipodystrophy. Antivir Ther
16. Lihn AS, Richelsen B, Pedersen SB, et al. Increased expression of TNF-alpha, IL-6, and IL-8 in HALS: implications for reduced adiponectin expression and plasma levels. Am J Physiol Endocrinol Metab
17. Carr A, Law M. An objective lipodystrophy severity grading scale derived from the lipodystrophy case definition score. J Acquir Immune Defic Syndr
18. Bastard JP, Rabasa-Lhoret R, Maachi M, et al. What kind of simple fasting index should be used to estimate insulin sensitivity in humans? Diabetes Metab
19. Sutinen J, Korsheninnikova E, Funahashi T, et al. Circulating concentration of adiponectin and its expression in subcutaneous adipose tissue in patients with highly active antiretroviral therapy-associated lipodystrophy. J Clin Endocrinol Metab
20. Sewter C, Berger D, Considine RV, et al. Human obesity and type 2 diabetes are associated with alterations in SREBP1 isoform expression that are reproduced ex vivo by tumor necrosis factor-alpha. Diabetes
21. Hammond EL, Sayer D, Nolan D, et al. Assessment of precision and concordance of quantitative mitochondrial DNA assays: a collaborative international quality assurance study. J Clin Virol
22. Walker UA, Setzer B, Venhoff N. Increased long-term mitochondrial toxicity in combinations of nucleoside analogue reverse-transcriptase inhibitors. AIDS
23. Mynarcik DC, McNurlan MA, Steigbigel RT, et al. Association of severe insulin resistance with both loss of limb fat and elevated serum tumor necrosis factor receptor levels in HIV lipodystrophy. J Acquir Immune Defic Syndr
24. Vigouroux C, Maachi M, Nguyen TH, et al. Serum adipocytokines are related to lipodystrophy and metabolic disorders in HIV-infected men under antiretroviral therapy. AIDS
25. Sutinen J, Hakkinen AM, Westerbacka J, et al. Rosiglitazone in the treatment of HAART-associated lipodystrophy-a randomized double-blind placebo-controlled study. Antivir Ther
26. Carr A, Workman C, Carey D, et al. No effect of rosiglitazone for treatment of HIV-1 lipoatrophy: randomised, double-blind, placebo-controlled trial. Lancet
27. Moyle G, Sutinen J. Managing HIV lipoatrophy. Lancet
28. Schlay J. Insulin resistance in patients treated with d4T, but not abacavir and 3TC. J Acquir Immune Defic Syndr
29. Mallal SA, John M, Moore CB, et al. Contribution of nucleoside analogue reverse transcriptase inhibitors to subcutaneous fat wasting in patients with HIV infection. AIDS
30. Mallon P, Unemori P, Bowen J, et al. Nucleoside reverse transcriptase inhibitors decrease mitochondrial and PPAR-gamma gene expression in adipose tissue after only 2 weeks in HIV-uninfected healthy Adults. Presented at: 11th Conference on Retroviruses and Opportunistic Infections; 2004; San Francisco.
31. Pace CS, Martin AM, Hammond EL, et al. Mitochondrial proliferation, DNA depletion and adipocyte differentiation in subcutaneous adipose tissue of HIV-positive HAART recipients. Antivir Ther
32. Nolan D, Hammond E, James I, et al. Contribution of nucleoside-analogue reverse transcriptase inhibitor therapy to lipoatrophy from the population to the cellular level. Antivir Ther
33. Cherry C, Gahan M, McArthur J, et al. Exposure to dideoxynucleosides is reflected in lowered mitochondrial DNA in subcutaneous fat. J Acquir Immune Defic Syndr
34. Christensen E, Stegger M, Jensen-Fangel S, et al. Mitochondrial DNA levels in fat and blood cells from patients with lipodystrophy or peripheral neuropathy and the effect of 90 days of high dose co-enzyme Q treatment: a randomised double blind, placebo controlled pilot study. Clin Infect Dis
35. Lindegaard B, Keller P, Bruunsgaard H, et al. Low plasma level of adiponectin is associated with stavudine treatment and lipodystrophy in HIV-infected patients. Clin Exp Immunol
Keywords:© 2005 Lippincott Williams & Wilkins, Inc.
adipokines; NRTIs; PIs and HIV associated lipodystrophy