HIV lipodystrophy (HIV-LD) is characterized by the loss of adipose tissue from the subcutaneous compartment with increased visceral adipose tissue (VAT) and is accompanied by insulin resistance and hypertriglyceridemia.1-4 The mechanism(s) underlying the decline in the abundance of subcutaneous adipose tissue (SAT) is not completely understood, but HIV protease inhibitors (PIs)5,6 and nucleoside reverse transcriptase inhibitors (NRTIs)7,8 have been implicated in the pathogenesis of the HIV-LD syndrome.
In addition to antiretroviral therapy, the syndrome of LD in HIV occurs within the context of active viral infection. Significantly elevated levels of the soluble tumor necrosis factor-α (TNFα) receptor type 2 (sTNFR2) have been demonstrated in individuals with HIV-LD compared with HIV-infected individuals without HIV-LD or uninfected individuals.9 This elevation in sTNFR2 is often indicative of cumulative exposure to TNFα.10 Because TNFα is capable of inducing apoptosis in adipocytes,11 increased apoptosis has been suggested as the mechanism for loss of SAT in individuals with HIV-LD.
There are reports of apoptosis in cells in subcutaneous tissue from individuals with HIV-LD.12 These studies used the terminal deoxynucleotidyl transferase biotin-dUTP nick end labeling (TUNEL) approach to identify nuclei from cells with alterations in DNA suggesting apoptosis.13 The study by Domingo et al12 clearly demonstrates the presence of positive-staining nuclei in subcutaneous tissue of individuals with HIV-LD.
Because nuclei account for a small proportion of the volume of adipose tissue, making histologic assessment of apoptosis difficult, we have investigated the role of apoptosis in the loss of SAT with an alternative method. In this method, the polymerase chain reaction (PCR) is used to enhance the sensitivity of detecting the classic marker of apoptosis, the DNA ladder. DNA ladders can be visualized by the ligase-mediated PCR (LM-PCR) amplification of DNA14 to amplify evidence of internucleosomal DNA cleavage. This amplification enhances the sensitivity of the determination of apoptosis. With the LM-PCR DNA ladder technique, there was evidence of apoptosis in subjects with HIV-LD, but the degree of DNA laddering was not increased above levels in HIV-infected subjects without LD or uninfected control subjects. Using this method, we find no evidence of increased apoptosis in the subcutaneous tissue from subjects with chronic HIV-LD.
MATERIALS AND METHODS
Patients and Samples
Twenty-seven control subjects and 47 individuals with HIV disease participated in this study, which was approved by the Committee on Research Involving Human Subjects at the University Medical Center of the State University of New York at Stony Brook. All subjects gave their informed written consent. Individuals enrolled in the study included 22 HIV-infected individuals with self-reported loss of SAT and increased abdominal fat (HIV-LD). The self-reported alterations in body fat distribution were confirmed by clinical assessment. For comparison, 25 HIV-infected individuals without self-reported loss of SAT and 27 individuals who were seronegative for HIV were also studied. All study subjects had been free of acute illness for the 3 months preceding their participation in the study.
Study subjects were admitted to the General Clinical Research Center on the evening before the procedure. Subjects fasted overnight after eating a snack at 10:00 PM. On the next morning between 7:00 AM and 8:00 AM, subcutaneous fat biopsy material was harvested from the outer aspect of the thigh under sterile conditions and local anesthesia. The biopsy material was immediately blotted free of excess blood, frozen in liquid nitrogen, and stored at −140°C.
Body fat distribution was assessed with dual-energy x-ray absorptiometry (DEXA) as described previously.9 The percentage of body fat present in the limbs was calculated as the proportion of limb fat to the total of limb and trunk fat multiplied by 100.
Genomic DNA Isolation
Frozen SAT was pulverized under liquid nitrogen. DNA was extracted from frozen tissue powder using the Puregene DNA purification kit from Gentra Systems (Minneapolis, MN) according to the manufacturer's protocol. Briefly, 10 to 20 mg of the frozen tissue powder was extracted with 600 μL of the cell lysis solution by gentle disruption of the tissue in a microcentrifuge tube fitted with a polypropylene pestle (Kontes, Vineland, NJ). The lysates were supplemented with 3 μL of proteinase K solution (20 mg/mL), mixed, and incubated at 55°C overnight. RNA was removed by digestion with RNase A for 60 minutes at 37°C. After precipitation, protein was removed by centrifugation and DNA was precipitated with isopropanol.
Ligase-Mediated Polymerase Chain Reaction
The LM-PCR ladder assay was performed using the Apo-Alert Ladder Assay Kit from Clontech (San Jose, CA). DNA was mixed with ligation mixture containing the ligation primers. The primer pair consisted of a 12-mer with a dephosphorylated 5′ terminus and sequence 5′-TGCGGTGAGAGG-3′ and the intact 24-mer 5′-AGCACTCTCGAGCCTCTCACCGCA-3′. The reaction mixture was overlaid with mineral oil and incubated at 55°C for 10 minutes. The reaction mixture was cooled to 10°C gradually over 1 hour to allow DNA annealing and was then treated with T4 DNA ligase. The reaction was transferred to 16°C and incubated overnight to yield the ligated DNA product.
The LM-PCR was carried out according to the manufacturer's instructions. Briefly, the amplification mixture, consisting of 150 ng of ligated genomic DNA, deoxynucleotide triphosphates (dNTPs), ligation primers, and Taq DNA polymerase, was incubated at 72°C for 8 minutes to fill in with complement to the 24-mer ligation adapter. Cycle parameters were 94°C for 1 minute and 72°C for 3 minutes for 25 cycles. The reaction products were separated in 1.2% agarose gels, stained with ethidium bromide, and photographed.
Confirmation of DNA Mass Equivalence
The PCR amplification of a single copy gene (En-2) was used to establish DNA mass equivalence in the LM-PCR ladder assay. The materials and protocols were included in the Apo-Alert Kit. En-2 sequence amplification was carried out as recommended by the manufacturer. En-2 primers used in the amplification were 5′-TCTACCCAGTTCCCAGGTCTCG-3′ (forward) and 5′-ACTTGCCCTCCTTGGCTGTGGT-3′ (reverse). The En-2 sequence was amplified from 150 ng of ligated genomic DNA using cycling parameters of 1 minute at 94°C, 1 minute at 70°C, and 1 minute at 72°C for 26 cycles. Reaction products were separated in 1.2% agarose gels, stained with ethidium bromide, and photographed. Evidence of the 290-base pair (bp) En-2 PCR product was used to confirm that sufficient ligated genomic DNA was used in the LM-PCR reaction. The presence of apoptosis was concluded if the ratio of the scanning intensity of DNA ladders relative to the scanning intensity of En-2 achieved a threshold value (1.5). A positive control sample was included with each gel.
Healthy uninfected control subjects (control), HIV-positive individuals without evidence of LD, and HIV-positive individuals with LD were all of similar age and body mass index (BMI; wt/h2; Table 1). Subjects in the HIV and HIV-LD groups had been on stable antiretroviral regimens for at least 3 months before the study, and the pattern of antiretroviral medications (see Table 1) was similar in both groups. Zidovudine was used by 7 subjects with HIV but no LD and by 6 subjects with HIV-LD. Lamivudine was used by 15 non-LD subjects and by 12 HIV-LD subjects, and stavudine was used by 12 non-LD subjects and by 10 subjects with HIV-LD. Two subjects in the HIV group and 1 subject in the HIV-LD group were naive to antiretroviral medication. The numbers of CD4 cells and viral burden were similar in the HIV and HIV-LD groups. In the individuals with HIV-LD, the percentage of body fat present in the limbs was significantly lower compared with the control or HIV group without LD (P < 0.05, Student-Newman-Keuls; see Table 1).
Ligase-Mediated Polymerase Chain Reaction
Evidence of DNA laddering suggesting apoptosis was observed in all groups of subjects. In the control group, 13 of 27 subjects displayed evidence of DNA ladders by LM-PCR. This was similar to the proportion of individuals displaying evidence of LM-PCR DNA ladders in the HIV group (13 of 25 subjects) and the HIV-LD group (10 of 22 subjects). Individuals in the HIV-LD group displayed no more evidence of apoptotic DNA laddering after LM-PCR amplification than individuals in the control or HIV group (Fig. 1; P = 0.9). Of the subjects with HIV disease and apoptosis, there were 12 female subjects (67% of all HIV-infected female subjects, 6 with LD and 6 with HIV but no LD) and 11 male subjects (38% of all HIV-infected male subjects, 4 with HIV-LD). There is almost a statistical significance in gender distribution of subjects with apoptosis (P = 0.055, χ2 test). The racial/ethnic distribution of subjects with apoptosis was 6 white, 15 African American, and 2 Hispanic subjects, and this was not statistically significant (P = 0.8, Fisher exact test). Adjusting for NRTIs, nonnucleoside reverse transcriptase inhibitors (NNRTIs), and PIs in a multiple logistic regression model, there was no statistical evidence of increased apoptosis in the LD group compared with the non-LD group (odds ratio [OR] = 0.88, 95% confidence interval [CI]: 0.25, 3.04; P = 0.84). A multivariate logistic regression model combining HIV and HIV-LD subjects and adjusting for PI and NRTI use showed a protective effect of NNRTI for apoptosis (OR = 0.17, 95% CI: 0.03, 0.93; P = 0.04). If the model was altered so that the NRTIs were treated as 3 separate drug categories (lamivudine, stavudine, and zidovudine), none was a significant risk factor and the protective effect of NNRTIs was weaker and not significant. If PI use was expanded to include not just current PI use but previous PI use as well, the model still did not indicate a significant relation of apoptosis to PI exposure. Analysis with a model in which there are 5 variables but only 23 cases may fail to detect associations that might be evident with a larger data set, however.
The data from the present study demonstrate that there is no increased incidence of apoptosis in the SAT from individuals with chronic HIV-LD compared with uninfected control subjects or subjects with HIV disease but without HIV-LD.
The loss of SAT in HIV-LD has been associated with HIV PI use15 as well as with NRTI use.7,24 Apoptosis has been suggested as a mechanism for the subcutaneous fat loss associated with HIV-LD.12,18,19 This suggestion is supported by the observation of elevated levels of TNFα16 and STNFR2,9 which are indicative of TNFα activity.10 Because in vitro studies of adipocytes indicate that TNFα induces apoptosis,11 this mechanism has been suggested to account for the subcutaneous loss of adipose tissue in HIV-LD. In support of this hypothesis, Domingo et al12 have provided evidence of apoptosis in subcutaneous tissue from individuals with the HIV-LD syndrome using the TUNEL procedure. This assay technique adds appropriately labeled deoxynucleotide triphosphates to available 3′-hydroxyl termini as a way of assessing which nuclei display evidence of internucleosomal DNA cleavage, a hallmark of apoptosis.17 Although there seems to be ample evidence of apoptotic nuclei in the subcutaneous tissue samples from individuals with HIV-LD in the work of Domingo et al,12 there were no data presented to address the question of whether the extent of apoptosis observed in samples from individuals with HIV-LD was increased relative to similar material obtained from uninfected controls or from individuals with HIV but no evidence of LD.
The surprising finding of the present study (see Fig. 1) is the demonstration that there is no increase in the frequency of individuals with HIV-LD displaying apoptotic DNA ladders in subcutaneous tissue compared with uninfected controls or individuals with HIV but without LD. Furthermore, the intensity of DNA ladders in the material from the HIV-LD group is similar to that for material from the other 2 groups. Thus, these data do not support the hypothesis that increased apoptosis is a mechanism for the loss of SAT in HIV-LD.
The discrepancies between the studies of Domingo and his colleagues and the present study need to be explored. Use of the TUNEL assay, which requires histologic localization of nuclei, is complicated in adipose tissue by the fact that so little of an adipocyte is nucleus. The size discrepancy between the nucleus (∼5 μm in diameter) and the adipocyte (≥100 μm in diameter) means that in a 5-μm section, the probability of obtaining a section through an adipocyte with its associated nucleus is ≤5%. This means that many sections need to be examined to determine the rate of apoptosis in adipose tissue. With the DNA ladder technique, there are also potential problems relating to specificity of the assay for adipocytes. Because a tissue sample is used for the analysis, there is a possibility of contamination with cells other than adipocytes. Inclusion of adipocyte markers colocalized histologically or assessed from mRNA in the ladder assay might help to dispel uncertainty in either technique arising from ambiguity of cell type.
Also, neither the studies by Domingo and his colleagues12,18,19 nor the present study addresses the question of constancy in the rate of loss of SAT in subjects with HIV-LD. The assumption in both studies is that measurement of a single point in time can be taken as representative of the rate of apoptosis over time. It is possible that there are periods of rapid loss of SAT that may have accelerated rates of apoptosis, although varying rates of SAT loss have not been reported.
Although there are limitations to the present study, it seems unlikely that accelerated apoptosis is responsible for the loss of SAT associated with chronic HIV-LD in this population. Multivariate logistic regression modeling of the data combining the HIV non-LD and HIV-LD subjects with adjustment for the current PI and NRTI use indicated a protective effect of NNRTI for apoptosis (OR = 0.17, 95% CI: 0.03,0.93; P = 0.04), suggesting a differential effect on apoptosis among antiretroviral therapies. In contrast, there are recent reports indicating accelerated apoptosis assessed with the TUNEL assay in individuals with HIV-associated lipoatrophy, with a decline in apoptosis markers when stavudine was discontinued.20,21 In the present study, there is not an obvious relation between apoptosis and stavudine treatment; only 3 of 9 subjects in the HIV-LD group and 6 of 13 subjects in the HIV-non-LD group with evidence of DNA ladders had a history of stavudine treatment. The switch study of Thompson et al21 is clearly a more appropriate study design for determining the effect of a particular drug, however.21 Interestingly, in that study, the decline in apoptosis when stavudine was discontinued was associated with an increase in adipose tissue of approximately 25%.
In addition to the decline in apoptosis with discontinuation of stavudine, Thompson et al21 report accelerated rates of apoptosis before the change in medication compared with subjects without HIV disease.21 Although there are obvious differences in methodology between that study and the current study, there are also differences of gender (predominantly male compared with 12 female subjects and 11 male subjects in the present study), race (predominantly white compared with predominantly African American in the present study), and the nature of fat loss in the subject population. Unlike the population studied by Thompson et al,21 subjects with HIV-LD in the present study did not have loss of body fat compared with non-LD subjects; rather, there was a redistribution of body fat with loss of peripheral fat and increased fat in the trunk. Further, although the present study and the basal conditions in the study of Thompson et al21 represent chronic conditions of antiretroviral therapy, these conditions may not be the same. The present study was carried out in 1999 through 2001. At that time, with increased awareness of the role of PIs in the development of HIV-LD, subjects were changed from PI-containing regimens to those with NRTIs or NNRTIs. All subjects had been on a stable drug regimen for 3 months before the study, but they may not have been on stavudine as along as the subjects studied more recently. Clearly, longitudinal assessment of apoptosis along with changes in body composition and changes in drug regimens in defined racial populations are needed to clarify the role of apoptosis in the loss of body fat associated with HIV disease.
Increased apoptosis may not be the only mechanism to explain loss of SAT. There is evidence that an additional mechanism for the loss of SAT may involve a reduction in the rate of adipose-specific gene expression5,22 and subsequent differentiation of preadipocytes into adipocytes.
In conclusion, the syndrome of HIV-LD is characterized by the loss of adipose tissue mass from the subcutaneous region.1 This regional fat loss is associated not only with dysregulation of carbohydrate9 and lipid23 metabolism but with distressing alterations in body fat distribution that may even lead to noncompliance with antiretroviral medication in an attempt to reverse HIV-LD. Establishing the plausible mechanism(s) for this loss of subcutaneous fat is integral to the design of therapeutic strategies to restore peripheral adipose tissue. It seems that reduced chronic loss of peripheral fat may not be caused entirely by accelerated apoptosis but may involve multiple mechanisms, including reduced adipocyte differentiation.
The authors acknowledge help from Mark Kaplan at the Department of Infectious Diseases, North Shore University Hospital, and Roy Steigbigel at the State University of New York at Stony Brook in recruiting subjects from their clinical practice. They also thank the nursing staff of the General Clinical Research Center for meticulously carrying out this protocol, Bernice Sealy for co-coordinating this study, and Joan Kavanaugh for outstanding administrative assistance.
1. Carr A. HIV lipodystrophy: risk factors, pathogenesis, diagnosis and management. AIDS
. 2003;17(Suppl 1):S141-S148.
2. Hadigan C, Miller K, Corcoran C, et al. Fasting hyperinsulinemia and changes in regional body composition in human immunodeficiency virus-infected women. J Clin Endocrinol Metab
3. Lo JC, Mulligan K, Tai VW, et al. “Buffalo hump” in men with HIV-1 infection. Lancet
4. 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
5. 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
6. Riddle TM, Kuhel DG, Woollett LA, et al. HIV protease inhibitor induces fatty acid and sterol biosynthesis in liver and adipose tissues due to the accumulation of activated sterol regulatory element-binding proteins in the nucleus. J Biol Chem
7. 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
8. Nolan D, Hammond E, Martin A, et al. Mitochondrial DNA depletion and morphologic changes in adipocytes associated with nucleoside reverse transcriptase inhibitor therapy. AIDS
9. 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
10. Dri P, Gasparini C, Menegazzi R, et al. TNF-induced shedding of TNF receptors in human polymorphonuclear leukocytes: role of the 55-kDa TNF receptor and involvement of a membrane-bound and non-matrix metalloproteinase. J Immunol
11. Prins JB, Niesler CU, Winterford CM, et al. Tumor necrosis factor-alpha induces apoptosis of human adipose cells. Diabetes
12. Domingo P, Matias-Guiu X, Pujol RM, et al. Subcutaneous adipocyte apoptosis in HIV-1 protease inhibitor-associated lipodystrophy. AIDS
13. Gavrieli Y, Sherman Y, Ben-Sasson SA. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol
14. Staley K, Blaschke AJ, Chun J. Apoptotic DNA fragmentation is detected by a semi-quantitative ligation-mediated PCR of blunt DNA ends. Cell Death Differ
15. Carr A, Samaras K, Burton S, et al. A syndrome of peripheral lipodystrophy, hyperlipidaemia and insulin resistance in patients receiving HIV protease inhibitors. AIDS
16. Johnson JA, Albu JB, Engelson ES, et al. Increased systemic and adipose tissue cytokines in patients with HIV-associated lipodystrophy. Am J Physiol Endocrinol Metab
17. Wyllie AH. Glucocorticoid-induced thymocyte apoptosis is associated with endogenous endonuclease activation. Nature
18. 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
19. Lloreta J, Domingo P, Pujol RM, et al. Ultrastructural features of highly active antiretroviral therapy-associated partial lipodystrophy. Virchows Arch
20. Cherry CL, McComsey G, Lal L, et al. Increased adipocyte apoptosis is present in lipoatrophic subjects on d4T and reverses within 48 weeks of switching NRTIs. Presented at: Australian Society for HIV Medicine Annual Meeting; 2003; Cairns, Australia.
21. Thompson KA, McComsey G, Paulsen D, et al. Improvements in body fat and mitochondrial DNA levels are accompanied by decreased adipose tissue cell apoptosis after replacement of stavudine therapy with either abacavir or zidovudine. Presented at: 10th Conference on Retroviruses and Opportunistic Infections; 2003; Boston.
22. Mynarcik D, McNurlan M, Steigbiegel R, et al. Elevated triglycerides and loss of limb fat are associated with a decrease in peripheral adipose tissue lipoprotein lipase gene expression [abstract]. Presented at: 82nd Annual Meeting of the Endocrine Society; Toronto, ON, Canada, 2000.
23. Hui DY. Effects of HIV protease inhibitor therapy on lipid metabolism. Prog Lipid Res
24. Brinkman K, Smeitink JA, Romijn JA, et al. Mitochondrial toxicity induced by nucleoside analogue reverse transcriptase inhibitors is a key factor in the pathogenesis of antiretroviral therapy-related lipodystrophy. Lancet