Meroni, Luca MD; Riva, Agostino MD; Morelli, Paola MD; Galazzi, Morena BD; Mologni, Daniela PhD; Adorni, Fulvio PhD; Galli, Massimo MD
Alterations of lipid metabolism were described in the natural history of HIV infection even before the advent of highly active antiretroviral therapy (HAART). Unexpectedly, the introduction of potent drug combinations, including or not including protease inhibitors (PIs), resulted in a severe increase in the incidence of metabolic abnormalities.1 Such alterations include lipodystrophy, hypercholesterolemia, hypertriglyceridemia, and insulin resistance. Different pathogenic mechanisms associated with various antiviral drug classes were suggested, including reduced lipogenesis, increased lipolysis, impaired storage of triglycerides consequent to insulin resistance, and mitochondrial toxicity.2
Recently, a role of CD36 antigen in the genesis of HIV-related dyslipidemia has been hypothesized. CD36 is a multifunctional transmembrane glycoprotein with a wide tissue distribution and with a broad relevance in lipid metabolism. Physiologic and pathologic effects after the engagement of CD36 depend on the tissue involved in the process.3,4
In adipose tissue as well as heart and skeletal muscle, CD36 has a role in energy metabolism by transporting long-chain fatty acids (LCFAs) across the cell membrane. The role of CD36 in the regulation of glucose and lipid metabolism is essential; its genetic deficiency on plasma membrane has been demonstrated to account for insulin resistance and dyslipidemia in rodent models.5 In human beings, genetic CD36 deficiency is common in Asian populations, in which it has also been associated with insulin resistance6 and cardiomyopathy.7 On macrophages, the major role of CD36 is the specific uptake of oxidized low-density lipoproteins (ox-LDLs).8 CD36 expression on macrophages is essential to cholesterol accumulation, the formation of foam cells, and the genesis of atherosclerotic lesions.9
Two recent studies investigated the role of CD36 in the pathogenesis of HAART-related dyslipidemia. The first reported a PI-induced decrease in monocyte CD36 levels as a possible causative agent for insulin resistance and dyslipidemia.10 In direct contrast, in a more recent in vitro study, a PI-driven upregulation of CD36 leading to intracellular accumulation of cholesteryl esters has been described, suggesting a potential proatherogenic effect of HAART.11 Although the opposite findings of these studies may be partly a result of differences in experimental design, the issue regarding the role of CD36 antigen in the pathogenesis of HIV-related fat abnormalities still remains unclear.
We measured the expression of CD36 on circulating monocytes in a population of HIV-infected patients larger than those considered in the previous studies and with a longer exposure to antiretroviral therapy. We observed increased monocyte CD36 levels in HIV-positive patients compared with HIV− subjects, no significant correlation between CD36 levels and antiretroviral regimens, and a negative correlation between CD4 lymphocyte counts and monocyte CD36 levels.
CD36 antigen expression on peripheral blood monocytes was measured by means of flow cytometry in 35 healthy controls (HCs) and in 165 consecutively enrolled HIV-infected patients. Individuals taking statin or fibrate were excluded. Fifty microliters of fresh whole blood was stained with directly labeled (fluorescein [FITC], phycoerythrin, [PE]) monoclonal antibodies (anti-CD36 FITC, anti-CD14 PE; Beckman Coulter Immunotech, Marseille, France). Appropriate isotypic controls were used for CD36 and CD14 antibodies. Stained specimens were fixed with Immunoprep kit reagents (Beckman Coulter Immunotech) by means of a Q-Prep workstation (Beckman Coulter Immunotech) and analyzed by means of an EPICS XL flow cytometer (IL Coulter Milan, Italy). Five thousand monocytes gated on side scatter-PE parameters were evaluated for CD36 antigen expression. Because most CD14+ antibodies are CD36+, we measured the channel of mean fluorescence intensity (MFI) of the CD36+ peak on a logarithmic scale. Measurement of MFI is easily influenced by flow cytometer acquisition conditions. To minimize possible errors, instrument calibration was checked daily by means of Flow-set fluorospheres (Beckman Coulter Immunotech), which have a mean fluorescence intensity comparable to that of the monoclonal antibodies used for the study. If needed, photomultiplier tube (PMT) voltage was changed so that Flow-set fluorosphere MFI was always acquired in the same channel. Any PMT voltage changes required never exceeded 5 mV.
HIV RNA plasma levels were measured by means of branched-DNA signal amplification-based hybridization (Chiron Corporation, Emeryville, CA). CD4 cell count was measured by means of flow cytometry. Total cholesterol, triglycerides, and LDL cholesterol were measured in serum collected early in the morning with the patient in a fasting condition. All statistical analyses were performed by means of univariate and multivariate analyses of variance (ANOVA) models using SPSS 11.5 for Windows XP. In each of these models, CD36 expression was the dependent variable; that is, it was the parameter whose variability was to be explained through correlation with other independent factors, represented in our analysis by epidemiologic, virologic, immunologic, and metabolic parameters. Any correlation statistically significant in multivariate analysis is reported in our results with the P value and with the value of R2 (explained variance) of the model.
Epidemiologic and metabolic features of the whole study population are shown in Table 1. CD36 expression on circulating monocytes was significantly higher in HIV-positive patients compared with HCs (P < 0.0001), and triglyceride levels were significantly higher in HIV-positive patients compared with HCs (P < 0.01). On multivariate analysis, HIV infection was the only variable independently associated with CD36 MFI (R2 = 0.208, P < 0.0001); no statistically significant correlation was observed in whole study population between CD36 MFI and age, sex, body mass index (BMI), triglycerides, cholesterol, or LDL serum levels.
Table 2 shows immunovirologic and metabolic features of HIV-positive patients stratified according to antiretroviral regimens. Of the 165 HIV-positive patients, 38 were naive for antiretroviral therapy, 50 patients were on a PI-based regimen (9 patients were receiving nelfinavir, 7 were receiving atazanavir, 25 were receiving lopinavir/ritonavir, 8 were receiving indinavir/ritonavir, 2 were receiving amprenavir, and 2 were receiving saquinavir soft gel), 48 were on a nonnucleoside reverse transcriptase inhibitor (NNRTI)-based regimen (32 were receiving efavirenz and 16 were receiving nevirapine), and 27 were on a regimen containing only a nucleoside reverse transcriptase inhibitor (NRTI). Overall 56 patients were taking zidovudine, 33 were taking stavudine, 26 were taking tenofovir, 20 were taking didanosine, 79 were taking lamivudine, 12 were taking abacavir, and 1 was taking zalcitabine. On univariate analysis, the CD4 cell count was negatively correlated to CD36 MFI on monocytes (P < 0.05; Pearson coefficient, −0.179). On multivariate analysis, the CD4 cell count maintained a statistically significant correlation with CD36 levels (P < 0.05, R2 = 0.074). In HIV-positive patients, no correlation was observed between CD36 MFI and sex, age, BMI, HIV RNA levels, time on antiretroviral therapy, antiretroviral regimen, triglycerides, cholesterol, or LDL serum levels.
A further multivariate analysis considering triglyceride serum levels as a dependent variable instead of CD36 MFI displayed a significant correlation only between triglycerides and cholesterol serum levels (P < 0.001), whereas no correlation was found with the other immunologic and virologic parameters measured (not shown).
CD36 is a transmembrane glycoprotein whose wide distribution in human tissues is consistent with its role as a lipid receptor/transporter. Its engagement regulates a number of metabolic pathways, and the effects of its altered expression are potentially different depending on the cell subset. In adipose tissue as well as heart and skeletal muscle and other tissues in which lipids are important substrates for energy, CD36 mediates the uptake of LCFAs. Evidence generated in isolated cells and in rodent models lacking CD36 demonstrated that its deficiency results in insulin resistance syndromes.5 CD36 is also highly expressed on monocytes and macrophages, in which it functions as a high-affinity receptor for oxidatively modified lipids that are present in ox-LDL and in the membrane of apoptotic cells. Upregulation of CD36 on monocytes leads to cholesterol accumulation, foam cell formation, and the genesis of atherosclerotic lesions.9
Given the broad involvement of CD36 in lipid and glucose metabolism and the relevance of metabolic disorders associated with HIV treatment, we investigated the influence of HIV infection and antiretroviral therapy on CD36 homeostasis. We performed a cross-sectional study in a large population of HIV-infected patients who were naive for antiretroviral therapy or had a long exposure to HAART regimens, including or not including PIs.
In HIV-positive patients, we observed a significantly higher expression of CD36 than in HCs. On multivariate analysis, the only variable strongly related to increased expression of CD36 on peripheral monocytes was HIV infection, whereas no correlation was observed with lipid serum levels, HIV RNA levels, time on HAART, or type of regimen. Interestingly, we observed a significant inverse correlation between CD36 levels and CD4 absolute count.
In contrast to our results, a recent study described an in vivo early and marked decrease of CD36 levels on monocytes induced by PIs.10 The authors observed a reduction of CD36 expression in healthy volunteers receiving ritonavir alone for 7 days and in HIV-positive patients receiving combination therapy, including lopinavir/ritonavir, for 1 month.
Besides the difference in sample size, our study differs from that of Serghides et al10 in 2 main aspects. First, our data were obtained by means of a cross-sectional analysis; second, CD36 MFI levels are expressed as absolute numbers, and the statistic is performed by comparing mean values of CD36 levels between groups. Conversely, Serghides et al10 present a short-term longitudinal study describing, for each patient, the percent change in CD36 levels between baseline and the end of follow-up, whereas the difference in CD36 absolute levels between HIV-positive patients and HCs is not investigated. Moreover, our patients were exposed to combined antiretroviral regimens for a longer time frame than those in the study of Serghides and colleagues10; thus, the effects exerted by PIs in CD36 homeostasis might be evident in the short term, whereas they might be masked by conditions related to HIV infection along the natural history of the disease.
Regulation of CD36 expression is under the control of a multitude of factors, and several mechanisms may account for increased CD36 levels in HIV infection. Macrophage colony-stimulating factors (M-CSFs) released by HIV-infected cells12 and interleukin (IL)-4, whose production is increased in HIV infection,13 upregulate CD36 expression on monocytes. Furthermore, proinflammatory cytokines (IL-1 and tumor necrosis factor), whose release is increased in HV infection, can modulate CD36 levels, inducing or inhibiting its expression depending on the tissue considered.14 Lastly, CD36 levels increase after exposure to specific ligands such as ox-LDL expressed on apoptotic cell membrane. HIV infection is characterized by enhanced T-cell apoptosis, particularly in advanced disease15 and in patients with a low CD4 count, notwithstanding antiretroviral therapy and virologic success,16 that is consistent with our finding of high CD36 levels in patients with a low CD4 count.
Our data do not support the hypothesis of a role of altered expression of CD36 in the genesis of HAART-related insulin resistance and increase in blood lipid levels; rather, CD36 overexpression on monocytes may function as a proatherogenic condition. To date, it is still unclear whether HIV infection itself, antiretroviral therapies, or both predispose to accelerated atherosclerosis. Autopsy findings of coronary disease were described in young HIV-infected patients even before the advent of HAART, thus suggesting a direct role of HIV infection in the genesis of atherosclerotic lesions.17 Recent studies described an increased incidence of cardiovascular disease in HIV-positive patients that was not caused by HAART administration.18 Conversely, other investigators reported a correlation between the development of plaque within vessels and the use of PIs.19 The increase in CD36 levels that we observed on monocytes from HIV-positive patients, although statistically significant, is quite modest. Nevertheless, as related to HIV infection itself, it is a conceivably permanent condition in infected patients, resulting in the potential for continued lipid accumulation and foam cell formation. Thus, CD36 overexpression might be added to a number of proatherogenic conditions associated with HIV infection, such as inflammatory endothelial injury, glucose intolerance, and lipid disorders. To define the effect of CD36 overexpression in the genesis of atherosclerosis in HIV-infected patients better, it would be useful to correlate the levels of CD36 expression with the actual presence of atherosclerotic lesions by means of noninvasive examinations (eg, B-mode ultrasound imaging of peripheral arteries).
1. Dubé MP, Stein JH, Aberg JA, et al. Guidelines for the evaluation and management of dyslipidemia in human immunodeficiency virus (HIV)-infected adults receiving antiretroviral therapy: recommendations of the HIV Medical Association of the Infectious Disease Society of America and the Adult AIDS Clinical Trial Group. Clin Infect Dis. 2003;37:613-627.
2. Carr A. HIV lipodystrophy: risk factors, pathogenesis, diagnosis and management. AIDS. 2003;17:141-148.
3. Silverstein RL, Febbraio M. CD36 and atherosclerosis. Curr Opin Lipidol. 2000;11:483-491.
4. Ibrahimi A, Abumrad NA. Role of CD36 membrane transport of long-chain fatty acids. Curr Opin Clin Nutr Metab Care. 2002;5:139-145.
5. Aitman TJ, Glazier AM, Wallace CA, et al. Identification of CD36 (Fat) as an insulin-resistance gene causing defective fatty acid and glucose metabolism in hypertensive rats. Nat Genet. 1999;21:76-83.
6. Miyaoka K, Kuwasako T, Hirano K, et al. CD36 deficiency associated with insulin resistance. Lancet. 2001;357:686-687.
7. Tanaka T, Okamoto F, Sohmiya K. Lack of myocardial iodine-123 15-(p-iodiphenyl)-3-R,S-methylpentadecanoic acid (BMIPP) uptake and CD36 abnormality-CD36 deficiency and hypertrophic cardiomyopathy. Jpn Circ J. 1997;61:724-725.
8. Nicholson AC, Han J, Febbraio M, et al. Role of CD36, the macrophage class B scavenger receptor, in atherosclerosis. Ann NY Acad Sci. 2001;947:224-228.
9. Bouillier A, Bird DA, Chang M, et al. Scavenger receptors, oxidized LDL, and atherosclerosis. Ann NY Acad Sci. 2001;947:214-222.
10. Serghides L, Nathoo S, Walmsley S, et al. CD36 deficiency induced by antiretroviral therapy. AIDS. 2002;16:353-358.
11. Dressman J, Kincer J, Matveev SV, et al. HIV protease inhibitors promote atherosclerotic lesion formation independent of dyslipidemia by increasing CD36-dependent cholesteryl ester accumulation in macrophages. J Clin Invest. 2003;111:389-397.
12. Gruber MF, Weih KA, Boone EJ, et al. Endogenous macrophage CSF production is associated with viral replication in HIV-1-infected human monocyte-derived macrophages. J Immunol. 1995;154:5528-5535.
13. Clerici M, Shearer GM. The Th 1-Th2 hypothesis of HIV infection: new insights. Immunol Today. 1994;15:575-581.
14. Memon RA, Feingold KR, Moser AH, et al. Regulation of fatty acid transport protein and fatty acid translocase mRNA levels by endotoxin and cytokines. Am J Physiol. 1998;274:E210-E217.
15. Ameisen JC, Capron A. Cell dysfunction and depletion in AIDS: the programmed cell death hypothesis. Immunol Today. 1991;4:102-105.
16. Pitrak DL, Bolanos J, Hershow R, et al. Discordant CD4 T lymphocyte responses to antiretroviral therapy for HIV infection are associated with ex-vivo rate of apoptosis. AIDS. 2001;15:1317-1319.
17. Paton P, Tabib A, Loire R, et al. Coronary artery lesions and human immunodeficiency virus infection. Res Virol. 1993;144:225-231.
18. Mooser V. Atherosclerosis and HIV in the highly active antiretroviral therapy era: towards an epidemic of cardiovascular disease? AIDS. 2003;17 (Suppl):S65-S69.
19. Maggi P, Lillo A, Perilli F, et al. Colour-Doppler ultrasonography of carotid vessels in patients treated with antiretroviral therapy: a comparative study. AIDS. 2004;17:1023-1028.
© 2005 Lippincott Williams & Wilkins, Inc.