HIV-associated lipodystrophy syndrome has been recognized since the late 1990s.1 Features of this syndrome may include progressive subcutaneous fat wasting (lipoatrophy) and/or central fat accumulation, along with dyslipidemia and insulin resistance. Lipodystrophy was noted to occur predominantly in individuals with HIV infection after prolonged exposure to highly active antiretroviral therapy (HAART). Protease inhibitors were initially believed to be the most likely cause of this syndrome,1 but more recently the role of nucleoside analogs (NRTIs) has been recognized. Data from the Western Australian HIV Cohort Study found that NRTI exposure contributes independently to the development of fat wasting, with stavudine exposure associated with accelerated fat loss compared with zidovudine use.2 This finding has since been confirmed in a number of other cohort studies,3-5 including those using objective measure of fat volume.6 Findings from prospective, randomized clinical trials comparing antiretroviral combinations, both with and without protease inhibitors, provide even stronger evidence that stavudine is the main risk factor for lipoatrophy.7-9 Nondrug factors such as genetics,10 older age, white race,2 and the presence of HIV infection itself11 may also modulate this effect.12
Peripheral lipoatrophy is a prominent morphologic feature of the lipodystrophy syndrome.13 It has been associated with significant negative impact on the quality of life of affected patients.14 An understanding of the pathogenesis of lipoatrophy is critical both to enable more informed treatment choices and for the development of future strategies to prevent or reverse this process. The precise pathogenic mechanisms are currently imperfectly understood, although the pathology has been well described,15 with evidence that subcutaneous fat wasting is associated with increased adipocyte apoptosis.16,17 This is consistent with the association between lipoatrophy and particular NRTIs. Mitochondrial dysfunction is thought to underpin cell death through apoptosis.18,19 It is therefore plausible that, in addition to possible cytokine-mediated mechanisms, fat wasting may be driven by the adverse effects of NRTIs on mitochondria.
NRTIs inhibit DNA polymerase γ, the enzyme responsible for replication of mitochondrial DNA (mtDNA), and many of the side effects of NRTIs are believed to result from mitochondrial dysfunction.20 The association between stavudine exposure and accelerated fat loss is consistent with this theory, because this agent is known to be 1 of the more potent NRTIs in terms of inhibition of mtDNA synthesis and has been shown to be particularly associated with mitochondrial pathology, including mtDNA depletion,21-23 altered morphology,22 and a reduction in respiratory chain activity23 in subcutaneous adipocytes. These abnormalities may be most pronounced in those individuals with lipoatrophy.15,24 Improvements in subcutaneous fat have followed switching from stavudine to another NRTI (but not with switching away from protease inhibitors)25 in studies of patients with lipoatrophy.26-28 To our knowledge, the effect on levels of apoptosis in adipose tissue of switching from stavudine has not previously been studied.
The TARHEEL Study involved 118 HIV-infected patients who had been receiving stavudine-containing HAART for at least 6 months and who had lipoatrophy and/or hyperlactatemia. Participants were followed for 48 weeks after substituting either abacavir or zidovudine for stavudine, and increases in subcutaneous limb and trunk fat were observed.27 The cellular changes accompanying this change were examined in more detail in a body composition substudy that included the collection of fat biopsy specimens at baseline and at 48 weeks. Significant increases in adipocyte mtDNA were demonstrated after switching from stavudine to another NRTI, but no change occurred in levels of multiple cytokine mRNAs in adipose tissue.29
We hypothesize that 1 of the pathogenic mechanisms underlying lipoatrophy in patients treated with stavudine is increased adipocyte apoptosis driven by the mitochondrial toxicity of this drug. Further, a reduction in the degree of adipocyte apoptosis would be expected after a change from stavudine-containing HAART to therapy with an NRTI that exhibits less mitochondrial toxicity in fat. We set out to test these hypotheses by examining the degree of apoptosis present in adipose tissue samples obtained from individuals enrolled in the body composition substudy of the TARHEEL Study before and 48 weeks after substituting abacavir or zidovudine for stavudine.
Excisional biopsy specimens of subcutaneous fat were taken from the lateral thigh of individuals involved in the body composition substudy of the TARHEEL Study. All biopsy samples were collected with the written informed consent of subjects and with the approval of the local institutional review board. Control tissue specimens were obtained from HIV-uninfected individuals who were undergoing elective cosmetic surgery (Zenbios, Inc., Research Triangle Park, NC).
In Situ Terminal Transferase dUTP Nick End Labeling
Apoptotic adipocytes were identified using in situ terminal transferase dUTP nick end labeling, as previously described,30 in which the terminal transferase end labels fragmented DNA with biotin dUTP, which is then detected using the avidin-biotin complex method and a diamino benzidine hydrogen peroxide product as the colorimetric substrate. In brief, 6-μm sections of formalin-fixed, paraffin-embedded adipose tissue were deparaffinized and hydrated through decreasing concentrations of alcohols to water. Sections were then incubated for 10 minutes at room temperature in 4 μg/mL proteinase K in buffer (100 mM Tris and 50 mM EDTA, pH 8.0). Eighty microliters of 5 mM cobalt chloride, 5× terminal transferase reaction buffer, 7.5 μM biotin dUTP, 0.01 mM dATP, and 15 units terminal transferase were added to each slide. Slides were then incubated at 37°C for 60 minutes. Slides were blocked in phosphate-buffered saline with 2% normal goat serum and 0.2% Triton X-100 for 30 minutes, followed by incubation in methanol and 0.3% hydrogen peroxide for 15 minutes. Slides were then incubated in 100 μL of peroxidase-conjugated streptavidin for 60 minutes. One hundred microliters of a color solution was added to each slide and left for 2-5 minutes until a brown reaction product was visualized.
Quantitation of Apoptotic Cells
Apoptotic adipocytes were quantified per unit area of adipose tissue and as a proportion of all adipocytes present using video image analysis. A Dage DC330 camera (Scitech, Melbourne, Victoria, Australia) was used to photograph a minimum of 20 images per slide, chosen by capturing alternate fields of view at a magnification of ×20 using a BX50 Olympus microscope (Olympus, Tokyo, Japan). A positive control image was obtained for each slide before capturing these images and used for white color correction. A video image analysis software program (Image Pro Plus; Media Cybernetics, Silver Spring, MD, USA) was used to determine the precise area of adipose tissue present in each image, the number of adipocytes present, and the number of terminal transferase dUTP nick end labeling-positive (apoptotic) adipocytes present. Cell counts were confirmed manually. All quantitation was undertaken by investigators blinded to specimen date as well as clinical and treatment details for the individuals from whom the biopsy specimens were taken.
Statistical analysis was undertaken using StatView version 5.0 (SAS Institute, Cary, NC). Up to 3 slides were studied per biopsy specimen, and mean values for each biopsy sample were used in the analyses. Results were log transformed to obtain normally distributed data. Analysis of variance was then used to determine differences between the frequency of apoptotic adipocytes in biopsy specimens from the 3 groups of subjects (HIV-negative controls, HIV-infected individuals with lipoatrophy treated with stavudine, and HIV-infected individuals 48 weeks after switching from stavudine to either abacavir or zidovudine).
Samples of adipose tissue were available from 15 HIV-infected subjects with lipoatrophy treated with long-term stavudine-containing HAART and 20 HIV-uninfected controls. Tissue specimens were also available from 10 of the HIV-infected subjects who underwent a second fat biopsy 48 weeks after substituting another NRTI for stavudine. Nine of these individuals switched to abacavir and 1 switched to zidovudine without any changes to protease inhibitor or NNRTI therapy. All 15 HIV-infected subjects had been receiving stavudine for at least 2 years and had lipoatrophy at study entry. Other treatment and baseline demographic details are shown in Table 1.
Apoptotic Adipocytes Per Unit Area of Adipose Tissue
Few apoptotic cells were found per unit area of adipose tissue samples from HIV-uninfected controls (mean, 0.03; range, 0-0.18) (Fig. 1). Conversely, many apoptotic cells were found in tissue specimens from the 15 subjects with lipoatrophy treated with stavudine (mean, 0.27; range, 0.03-1.09; for difference, P < 0.0001). Forty-eight weeks after substituting abacavir or zidovudine for stavudine, the number of apoptotic adipocytes per unit area had significantly decreased (mean, 0.1; range, 0.01-0.31; for difference, P = 0.01). The number of apoptotic adipocytes seen in these 48-week biopsy samples was no longer significantly different from that seen in biopsy specimens from uninfected controls (mean, 0.1 vs. 0.03, respectively; P = 0.18).
Apoptotic Adipocytes as a Proportion of All Adipocytes Present
Results were similar when reanalyzed looking at apoptotic cells as a proportion of all adipocytes present in the tissue section to take account of any possible overall loss of cellularity in the tissue specimens from subjects with lipoatrophy (Fig. 2). Few cells were apoptotic in specimens from HIV-uninfected controls (mean, 3%; range, 0 to 14%), but a greater proportion were apoptotic in samples from subjects with lipoatrophy treated with stavudine (mean, 22%; range, 2%-85%; vs. HIV-uninfected controls, P < 0.0001). Forty-eight weeks after substituting abacavir or zidovudine for stavudine, there was a significant reduction in the level of apoptosis in tissue samples from HIV-infected subjects (mean, 9%; range, 0.2%-24%; P = 0.02). Furthermore, 48 weeks after the switch, the amount of apoptosis present was no longer significantly greater than that seen in biopsy specimens from controls (mean, 9% vs. 3%, respectively; P = 0.12).
Our study confirms that increased apoptosis is present in adipose tissue from HIV-infected patients with lipoatrophy treated with stavudine-based HAART. A novel finding is that adipocyte apoptosis decreases significantly within 48 weeks after substitution of abacavir or zidovudine for stavudine. It is likely that stavudine-associated mitochondrial dysfunction underlies this process. Further evidence supporting this theory has previously been reported from the TARHEEL Study, including a significant increase in fat mtDNA levels accompanying the reduction in apoptosis.29 Together these findings provide additional evidence for the important role of mitochondrial toxicity due to NRTIs, particularly stavudine, in the pathogenesis of HAART-associated fat wasting.
It is now widely accepted that many NRTI side effects are driven by mitochondrial dysfunction. Our finding of increased apoptosis in fat tissue specimens from patients with lipoatrophy treated with stavudine is consistent with this and with earlier descriptions.31 The novel finding that this increase reduces with switching NRTIs adds significant weight to the theory that stavudine-induced mitochondrial dysfunction may be involved in this process.
The cellular effects of NRTIs are likely to be tissue specific due to their need for intracellular metabolism via complex phosphorylation pathways to reach their active forms.32 Stavudine, once activated, is known to be a potent inhibitor of mtDNA replication. The available data suggest that adipocytes may be efficient activators of stavudine and, therefore, particularly likely to manifest mitochondrial toxicity when exposed to this compound. In addition to its association with accelerated fat loss in multiple cohort studies,3-5 stavudine exposure is strongly associated with mtDNA depletion in adipose tissue.21-23 The finding of significantly increased apoptosis in subcutaneous fat specimens from individuals with lipoatrophy treated with stavudine is in keeping with these observations and provides important pathogenic information that may form the basis of future therapeutic strategies.
Previous work has demonstrated that adipocyte apoptosis does not improve when patients with lipodystrophy switch from protease inhibitors to nevirapine without changing NRTIs, despite improvements in dyslipidemia.31 This description lacked a control group of HIV-infected subjects without HAART-associated lipoatrophy and, therefore, did not exclude the possibility of increased adipocyte apoptosis being a normal feature of HIV infection. However, there are other data supporting enhanced adipocyte apoptosis as 1 of the pathogenic mechanisms underlying HAART-associated lipoatrophy. For instance, increased local production of tumor necrosis factor-α (TNF-α), a cytokine that can induce apoptosis of adipocytes in vitro,33 has been shown in fat samples from subjects with lipodystrophy.34
Levels of circulating TNF-α as well as soluble TNF receptors increase with disease progression in individuals with uncontrolled HIV infection35 and fall toward normal after effective HAART.36 More recent studies, however, have demonstrated elevated levels of circulating soluble TNF receptors34,37 together with increased production of TNF-α34 in adipose tissue samples from subjects with lipodystrophy relative to other individuals with HIV infection. Additional work is needed to clarify whether cytokines such as TNF-α play a significant role in fat wasting, because it is possible that antiapoptotic agents, including cytokine-based therapies, may 1 day be used to prevent or reverse lipoatrophy in individuals exposed to HAART. However, the fact that levels of TNF-α did not change during the course of the TARHEEL Study38 suggests that it did not play a significant role in the reversal of apoptosis reported here.
The novel finding of histologic improvements in adipocyte apoptosis after 48 weeks of substitution of abacavir or zidovudine for stavudine is consistent with clinical studies showing that HAART-associated lipoatrophy is at least partly reversible with similar switching strategies.26,27 However, increases in limb fat volume with switching NRTIs are modest, and clinically severe lipoatrophy is not subjectively improved within 24 weeks of follow-up,26 although ongoing gradual improvements continue to at least 2 years.28 The increased levels of apoptosis found here in subjects with lipoatrophy treated with long-term stavudine provide evidence of ongoing, active destruction of subcutaneous adipose tissue in this population. Although our findings support switching stavudine for patients as an effective method of slowing this process, the regenerative capacity of adipose tissue may be limited. Therefore, future work is needed to establish whether levels of adipocyte apoptosis are predictive of lipoatrophy in patients receiving stavudine who have not yet had clinically significant fat wasting. An accurate method of predicting the development of this problem is urgently needed to guide optimal use of this effective antiretroviral agent.
The authors thank the study participants, Drs. Tyler Lonergan, Danielle Milano, and Sheldon Brown, the other study investigators, the staff at all of the ESS40010 sites participating in this substudy, and the following GlaxoSmithKline personnel: Vanessa Williams, Tracey Lancaster, Carol Humphries, Ilise Minto, Laura Lindsey, Brian Wine, Robin Fisher, and Siegrid Hessenthaler.
1. Carr A, Samaras K, Chisholm D, et al. Pathogenesis of HIV
-1-protease inhibitor-associated peripheral lipodystrophy
, hyperlipidaemia, and insulin resistance. Lancet
2. Mallal S, John M, Moore C, et al. Contribution of nucleoside analogue reverse transcriptase inhibitors to subcutaneous fat wasting in patients with HIV
3. White A. Mitochondrial toxicity
therapy. Sex Transm Infect
4. Bernasconi E, Boubaker K, Junghaus C, et al. Abnormalities of body fat distribution in HIV
-infected persons treated with antiretroviral drugs. J Acquir Immune Defic Syndr
5. Heath KV, Hogg RS, Singer J, et al. Antiretroviral treatment patterns and incident HIV
-associated morphologic and lipid abnormalities in a population-based cohort. J Acquir Immune Defic Syndr
6. Saint-Marc T, Partisani M, Poizot-Martin I, et al. Fat distribution evaluated by computed tomography and metabolic abnormalities in patients undergoing antiretroviral therapy: preliminary results of the LIPOCO study. AIDS
7. Amin J, Moore A, Carr A, et al. Combined analysis of two-year follow-up from two open-label randomized trials comparing efficacy of three nucleoside reverse transcriptase inhibitor backbones for previously untreated HIV
-1 infection: OzCombo 1 and 2. HIV Clin Trials
8. Joly V, Flandre P, Meiffredy V, et al. Increased risk of lipoatrophy
under stavudine in HIV
-1-infected patients: results of a substudy from a comparative trial. AIDS
9. Gallant JE, Staszewski S, Pozniak AL, et al. Efficacy and safety of tenofovir DF vs. stavudine in combination therapy in antiretroviral-naive patients. JAMA
10. Maher B, Alfirevic A, Vilar FJ, et al. TNF-alpha promoter region gene polymorphisms in HIV
-positive patients with lipodystrophy
11. Miller J, Carr A, Emery S, et al. HIV lipodystrophy
: prevalence, severity and correlates of risk in Australia. HIV Med
12. McComsey G, Maa J-F. Host factors may be more important than choice of antiretrovirals in the development of lipoatrophy
. AIDS Read
13. Gripshover BM, Tien P, Saag MS, et al. Lipoatrophy
is the Dominant Feature of the lipodystrophy
syndrome in HIV
-infected men. 10th Conference on Retroviruses and Opportunistic Infections; San Francisco; 2004.
14. Dukers N, Stolte I, Albrecht N, et al. The impact of experiencing lipodystrophy
on the sexual behaviour and well-being among HIV
-infected homosexual men. AIDS
15. Walker UA, Bickel M, Volksbeck S, et al. Evidence of nucleoside analogue reverse transcriptase inhibitor-associated genetic and structural defects of mitochondria in adipose tissue from HIV
-infected patients. J Acquir Immune Defic Syndr
16. Domingo P, Matias-Guiu X, Pujol RM, et al. Subcutaneous adipocyte apoptosis
-1 protease inhibitor-associated lipodystrophy
17. Lloreta J, Domingo P, Pujol RM, et al. Ultrastructural features of highly active antiretroviral therapy-associated partial lipodystrophy
. Virchows Arch
18. Green DR, Reed JC. Mitochondria and apoptosis
19. Mignotte B, Vayssiere J-L. Mitochondria and apoptosis
. Eur J Biochem
20. Brinkman K, ter Hofstede J, Burger D, et al. Adverse effects of reverse transcriptase inhibitors: mitochondrial toxicity
as common pathway. AIDS
21. Cherry C, Gahan M, Lewin S, et al. Treatment with dideoxynucleosides is reflected in lowered mitochondrial DNA in subcutaneous fat. JAIDS
22. Nolan D, Hammond EL, Martin A, et al. Mitochondrial DNA depletion and morphologic changes in adipocytes associated with nucleoside reverse transcriptase inhibitor therapy. AIDS
23. Hammond E, Nolan D, James I, et al. Reduction of mitochondrial DNA content and respiratory chain activity occurs in adipocytes within 6-12 months of commencing nucleoside reverse transcriptase inhibitor therapy (research letter). AIDS
24. 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
25. Dreschler H, Powderly WG. Switching effective antiretroviral therapy: a review. Clin Infect Dis
26. Carr A, Workman C, Smith D, et al. Abacavir substitution for nucleoside analogs in patients with HIV lipoatrophy
: a randomized trial. JAMA
27. McComsey G, Ward D, Hessenthaler S, et al. Improvement in lipoatrophy
associated with highly active antiretroviral therapy in human immunodeficiency virus-infected patients switched from stavudine to abacavir or zidovudine: the results of the TARHEEL Study. Clin Infect Dis
28. Martin A, Smith D, Carr A, et al. Reversibility of lipoatrophy
-infected patients 2 years after switching from a thymidine analogue to abacavir: the MITOX Extension Study. AIDS
29. McComsey G, Paulsen D, Lonergan J, et al. Improvements in lipoatrophy
, mitochondrial DNA levels and fat apoptosis
after replacing stavudine with abacavir or zidovudine. AIDS
30. Thompson K, McArthur JC, Wesselingh SL. Correlation between neurological progression and astrocyte apoptosis
-associated dementia. Ann Neurol
31. Domingo P, Matias-Guiu X, Pujol RM, et al. Switching to nevirapine decreases insulin levels but does not improve subcutaneous adipocyte apoptosis
in patients with highly active antiretroviral therapy-associated lipodystrophy
. J Infect Dis
32. Stein DE, Moore KH. Phosphorylation of nucleoside analog
antiretrovirals: a review for clinicians. Pharmacotherapy
33. Prins JB, Niesler CU, Winterford CM, et al. Tumor necrosis factor-α induces apoptosis
of human adipocytes. Diabetes
34. Johnson JA, Albu JR, Engelson ES, et al. Increased systemic and adipose tissue cytokines in patients with HIV
. Am J Physiol Endocrinol Metab
35. Aukrust P, Liabakk N, Muller F, et al. Serum levels of tumor necrosis factor-alpha (TNF alpha) and soluble TNF receptors in human immunodeficiency virus type 1 infection-correlations to clinical, immunologic, and virologic parameters. J Infect Dis
36. Aukrust P, Muller F, Lien E, et al. Tumor necrosis factor (TNF) system levels in human immunodeficiency virus-infected patients during highly active antiretroviral therapy: Persistent TNF activation is associated with virologic and immunologic treatment failure. J Infect Dis
37. 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
38. McComsey G, Paulsen D, Lonergan T, et al. Improvements in body fat, mitochondrial DNA levels and adipocyte apoptosis
48 weeks after replacing stavudine (d4T) with abacavir (ABC) or zidovudine (ZDV) appear independent of cellular cytokine mRNA levels. The Infectious Diseases Society of America Annual Meeting; San Diego; 2003.