Despite potent antiretroviral therapy (ART) HIV-1-infected (HIV+) individuals have an approximately two-fold increased risk of atherosclerotic cardiovascular disease (CVD), but the mechanisms are unclear . High-density lipoprotein (HDL) levels are a powerful independent negative predictor of CVD; however, HDL structure and function rather than absolute level may more accurately predict atherosclerosis . HIV+ ART-treated individuals have a higher prevalence of dyslipidemia, low HDL levels , impaired lipoprotein metabolism, and increased HDL lipid hydroperoxide content and redox activity (HDLox)  that are associated with CVD in some, but not all studies [3–5]. In HIV-uninfected (HIV−) persons, abnormal HDL function strongly correlates with CVD , but there is limited evidence in HIV infection.
Cell-based assays such as cholesterol efflux assays  have been used to determine HDL function, but have numerous drawbacks including heterogeneity with regards to cells used, reported readout, and standardization [3,4]. Cell-free assays may give more robust measures of HDL function compared with cell-based assays . We have developed a novel cell-free fluorometric method that measures HDL-associated lipid peroxidation (HDLox) and hence reduction of HDL antioxidant function [3,4]. In HIV+ ART-treated participants with low CVD risk, the readout from our assay correlated with coronary artery calcium score, carotid intima media thickness (surrogate measures of subclinical atherosclerosis) and macrophage activation [3–5]. Furthermore, impaired HDL antioxidant function (as measured by higher HDLox) was associated with impaired HDL remodeling  (a CVD-relevant measure of HDL function ) using a cell-free assay that measures the ability of HDL to release apolipoprotein A-I (ApoA-I) [HDL-ApoA-I exchange (HAE)] [9,10]. Thus, use of two independent cell-free assays of HDL activity may more accurately reflect HDL dysfunction in treated HIV+ persons.
Monocytes play a central role in the development of atherosclerosis, and HIV-related chronic monocyte activation may contribute to increased CVD, but the mechanisms are unknown. Using an established in vitro model of early atherosclerotic events, we have shown that monocytes from HIV+ individuals have a greater potential to develop into lipid-laden foam cells (which are associated with plaque progression) following transendothelial migration than monocytes from age-matched HIV− controls . Furthermore, we found that serum factors from ART-treated HIV+ individuals further promote monocyte-derived foam cell formation (MDFCF) . We have shown that dysfunctional HDL persists in virologically suppressed HIV+ individuals [3,4] on potent ART and may be the circulating factor that directly promotes MDFCF, a key event in HIV-related CVD. This research question can be reliably studied in a large prospective study of HIV-1 infected persons on chronic ART and well matched uninfected controls, where complex effects of confounders on MDFCF need to be dissected. Given the above complexities, known limitations of all HDL function assays, and lack of any preliminary data that support the role of dysfunctional HDL in promoting MDFCF, it would be difficult to design such as prospective study. Thus, to address this question, we conducted an exploratory study using a subset of samples from an established cohort of HIV+ individuals on potent ART with low CVD risk specifically selected to be matched for relevant factors. We isolated native HDL from a homogeneous group of HIV+ persons and incubated it directly with monocytes from healthy participants to dissect effects of HIV(+)HDL, independent to the documented effect of HIV on MDFCF  in an established ex-vivo model of early atherosclerotic events.
Study design and participants
The work utilized samples from the Center for Clinical AIDS Research and Education HDL function study; a cross-sectional study investigating impaired HDL function among viremic and virologically suppressed HIV+ men on stable ART . Sociodemographic and clinical characteristics, and measures of HDL function (HDLox, HAE %) have been described . To avoid confounding effects from sex , inflammation (other than HIV), traditional CVD risk factors and ART on atherogenesis, we selected HIV+ persons with the following inclusion criteria: men, no known inflammatory comorbidities other than HIV, no known CVD or risk factors for CVD (e.g., metabolic syndrome, diabetes, dyslipidemia, use of lipid lowering medication, hypertension, family history of CVD, Framingham Risk Score ≥6%), on stable and identical ART more than 6 months prior to visit, and with dysfunctional HDL as defined by two independent cell-free assays of HDL function. In the setting of our exploratory study, we selected 10 HIV-1-infected persons from this group who had the highest HDLox and lowest HAE to explore the question whether dysfunctional HDL in the setting of low CVD risk directly upregulates MDFCF. Five healthy men matched by age and race to the HIV+ group were also included. All individuals enrolled in the study provided written informed consent and the study was approved by the local Institutional Review Boards.
Lipoprotein preparation, oxidation, and determination of high-density lipoprotein function
HDL was purchased (EMD Millipore, Billerica, Massachusetts, USA) or isolated from plasma using ultracentrifugation and lipoproteins were oxidized in vitro as previously [9,10]. HDLox was quantified using a validated fluorometric assay that measures HDL lipid peroxidation based on the oxidation of the fluorochrome Amplex Red (Thermo Fisher Scientific, Waltham, Massachusetts) [9,10]. HAE assays were performed as previously described . Different aliquots of identical HDL were used for the HDLox, HAE, and the foam cell assays.
In-vitro model of monocyte migration and foam cell formation
The in-vitro MDFCF assay was performed using monocytes purified via negative selection (Miltenyi Biotec, Cologne, Germany) from freshly isolated peripheral blood mononuclear cells from healthy donors . Monocytes were preincubated with 20 μg/ml donor or commercial HDL [unmodified or oxidized with copper(II) sulfate (CuSO4) or 13(S)-hydroperoxy-9Z, 11E-octadecadienoic acid (13(S)-HPODE)] in serum-free media for 1 hr. Monocytes were then added to tumor necrosis factor-activated human umbilical vein endothelial cells monolayers on type I fibrous collagen gels to transmigrate and form foam cells as previously described (Supplemental Figure 1, http://links.lww.com/QAD/B159). Lipids were maintained in the media throughout the experiment.
Detailed methods are described in the supplemental material, http://links.lww.com/QAD/B159
Baseline characteristics of the study participants are shown in Table 1. HIV+ participants (median age: 42 years; n = 10) had suppressed viremia (<50 copies/ml) and the group overall had a low CVD risk (Framingham Risk Score <6%). All HIV+ persons were on stable ART more than 6 months prior to visit with efavirenz/emtricitabine/tenofovir disoproxil fumarate. HIV+ and control participants had comparable lipid profiles, Framingham Risk Score, and rates of cytomegalovirus seropositivity.
Dysfunctional high-density lipoprotein present in virologically suppressed, antiretroviral therapy-treated HIV-1 infection directly increases monocyte-derived foam cell formation
To investigate whether HDL dysfunction (as indicated by cell-free assays) was associated with proatherogenic effects, we used isolated HDL, assessed to be dysfunctional, from ART-treated HIV+ individuals [(HIV(+)HDL, n = 10)] and compared their influence on MDFCF in vitro to HDL isolated from healthy controls (HIV(−)HDL, n = 5). The decreased function of the isolated HDL assessed using two independent assays is presented in Fig. 1. Median HDLox of HIV+ persons was ∼50% higher compared with uninfected participants (Fig. 1a, P < 0.001). ART-treated HIV+ study participants had ∼35% lower median %HAE compared with HIV− participants (P < 0.001, Fig. 1b, Table 1).
Next, isolated donor HDL were preincubated with monocytes from HIV− individuals prior to (and maintained in culture throughout) analysis in our in-vitro foam cell model (Supplemental Figure 1, http://links.lww.com/QAD/B159). When media-containing dysfunctional HIV(+)HDL was added to tumor necrosis factor-activated human umbilical vein endothelial cell, a significantly increased proportion of monocytes differentiated into foam cells as compared with the same monocytes exposed to media-containing HDL from HIV− persons (median foam cells 33 vs. 26.2%, respectively, P = 0.015, Fig. 1c; from n = 3 replicate experiments using monocytes from three individual HIV− donors).
In-vitro oxidized high-density lipoprotein directly increases monocyte-derived foam cell formation
Given that oxidation of HDL leads to abnormal HDL function, we next compared the magnitude of the proatherogenic effects of HIV(+)HDL with chemically derived HDLox. In-vitro oxidation of lipoproteins using CuSO4, although widely used, has uncertain physiological relevance as this type of oxidation does not occur in vivo. Therefore, we also used 13(S)-HPODE, a potent in-vivo-generated lipid that renders HDL dysfunctional  as a comparator. Both forms of HDLox were associated with a significant increase in foam cell formation by healthy monocytes following transendothelial migration as compared with monocytes incubated with unmodified HDL (P = 0.004 and 0.0006 for CuSO4 and 13(S)-HPODE HDLox vs. unmodified HDL, respectively; Fig. 1d). Indeed, the direct stimulation of foam cell formation with HDLox generated with the more physiological reagent 13(S)-HPODE (Fig. 1) was comparable with the effects of HIV(+)HDL (Fig. 1c).
In this exploratory study, we found that HIV(+)HDL from HIV+ individuals on effective ART had abnormal HDL function based on two independent cell-free assays and increased MDFCF in an established in-vitro model of atherogenesis. Notably, the selected HIV+ participants were all receiving a nucleoside reverse transcriptase inhibitors/nonnucleoside reverse transcriptase inhibitor regimen that is associated with a more favorable lipid profile compared with protease inhibitor-containing regimens, and had low overall CVD risk. Given our previous finding that monocytes isolated from HIV+ donors have heightened MDFCF potential , it is possible that dysfunctional HDL from HIV+ persons may synergize with these proatherogenic monocytes, resulting in even more prominent effects on foam cell formation. We also showed that treatment of monocytes with HDL oxidized in vitro by two different mechanisms enhanced MDFCF. Given our prior data that dysfunctional HDL in chronic-treated HIV is oxidized, this suggests that oxidation underlies the atherogenic properties of HIV(+)HDL. Our study is among the first that provides important mechanistic insight into the role of dysfunctional HDL as a possible driver of CVD in chronic-treated HIV infection. Our hypothesis is shown in Supplemental Figure 2, http://links.lww.com/QAD/B159. Elucidating the mechanisms driving HDL-induced foam cells may identify oxidized HDL as a potential therapeutic target to reduce HIV-related CVD risk.
In-vitro studies indicate that oxidative modification of HDL may impair cholesterol efflux activity and inhibit HDL remodeling/exchange of apoA-I , a CVD-relevant measure of HDL function . We previously showed that dysfunctional HDL with impaired antioxidant HDL function (as measured by higher HDLox) in HIV+ individuals on long-term ART and without clinical CVD was associated with in-vivo progression of CVD , may stimulate endothelial cells to induce monocyte/macrophage chemotaxis [3,4], was positively correlated with noncalcified coronary atherosclerotic plaque , independently correlated with several markers of inflammation and immune activation , and was associated with impaired HDL remodeling . Here, we show that foam cell formation is enhanced when monocytes were exposed to chemically oxidized HDL, thus, linking HDL oxidation and impaired HDL function to processes which have a central role in HIV-related CVD.
The strengths of our study are the careful covariate phenotyping of our selected study population, including novel measures of HDL function and established physiologically relevant model of atherogenesis. Of note, MDFCF is not considered a function of HDL and, thus, the hypothesis that two abnormal HDL functions lead to a third impaired HDL function cannot explain our finding that dysfunctional HDL directly promotes MDFCF. However, all assays of HDL function have limitations  and HDL proteomics and cell-based cholesterol efflux assays were not performed. Other important limitations of our study include the small sample size and the inclusion of men only. Larger studies are needed to elucidate the differential effects of dysfunctional HDL on MDCF given the complex and unclear underlying physiological modulators of lipid metabolism responsible for the differences between men and women . For our in-vitro assay, we used monocytes from HIV(−) donors only and, as mentioned above, it is possible that foam cell formation in response to HIV(+) HDL might be further enhanced in monocytes from ART-treated and/or HIV-infected individuals. Here, we preselected HIV+ persons with evidence of dysfunctional HDL from a retrospective study, to specifically assess whether this dysfunction was associated with altered MDFCF. Such an approach is reasonable in the setting of a hypothesis-generating study, and we were able to demonstrate direct ex-vivo proatherogenic effects of dysfunctional HDL. These findings will inform larger studies to elucidate the mechanisms responsible for heightened MDFCF, and determine whether these effects may be more prominent in older HIV-1 infected persons and/or those with higher CVD risk.
In conclusion, our data provide important mechanistic insights into the role of dysfunctional HDL as possible driver of increased atherosclerotic risk in this population which warrants further investigation in larger prospective studies of subclinical atherosclerosis and CVD clinical events.
We thank our study participants and physicians who referred participants to this study. The authors gratefully acknowledge the contribution to this work of the Victorian Operational Infrastructure Support Program received by the Burnet Institute.
The work was supported by NIH grants NIH Grant #K08AI08272, NIH/NCATS Grant # UL1TR000124, and NHMRC grant #1108792 to A.C.H. and A.J. M.S.B. was supported by American Heart Association Western States Affiliate postdoctoral fellowship 14POST1833018. M.N.O. is funded by the California Tobacco Related Disease Research Program 21RT-0125.
M.N.O. is a founder of and owns a significant stake in Seer Biologics, Inc. but the content of this manuscript provides no benefit to Seer BioLogics, Inc. Potential benefit in no way influenced the thoroughness, stringency, interpretation, and presentation of this manuscripts content.
Part of this data were presented at the Annual Conference on Retroviruses and Opportunistic Infections (CROI) 2017, Seattle, Washington, USA, 13–16 February 2017.
Conflicts of interest
There are no conflicts of interest.
1. Kelesidis T, Currier JS. Dyslipidemia and cardiovascular risk in human immunodeficiency virus infection
. Endocrinol Metab Clin North Am
2. Gordon DJ, Rifkind BM. High-density lipoprotein: the clinical implications of recent studies
. N Engl J Med
3. Kelesidis T, Currier JS, Huynh D, Meriwether D, Charles-Schoeman C, Reddy ST, et al. A biochemical fluorometric method for assessing the oxidative properties of HDL
. J Lipid Res
4. Kelesidis T, Roberts CK, Huynh D, Martinez-Maza O, Currier JS, Reddy ST, Yang OO. A high throughput biochemical fluorometric method for measuring lipid peroxidation in HDL
. PLoS One
5. Zanni MV, Kelesidis T, Fitzgerald ML, Lo J, Abbara S, Wai B, et al. HDL redox activity is increased in HIV-infected men in association with macrophage activation and noncalcified coronary atherosclerotic plaque
. Antivir Ther
6. Navab M, Reddy ST, Van Lenten BJ, Fogelman AM. HDL and cardiovascular disease: atherogenic and atheroprotective mechanisms
. Nat Rev Cardiol
7. Rohatgi A, Khera A, Berry JD, Givens EG, Ayers CR, Wedin KE, et al. HDL cholesterol efflux capacity and incident cardiovascular events
. N Engl J Med
8. Kelesidis T, Oda MN, Borja MS, Yee Y, Ng KF, Huynh D, et al. Predictors of impaired HDL function in HIV-1 infected compared to uninfected individuals
. J Acquir Immune Defic Syndr
9. Borja MS, Zhao L, Hammerson B, Tang C, Yang R, Carson N, et al. HDL-apoA-I exchange: rapid detection and association with atherosclerosis
. PLoS One
10. Cavigiolio G, Geier EG, Shao B, Heinecke JW, Oda MN. Exchange of apolipoprotein A-I between lipid-associated and lipid-free states: a potential target for oxidative generation of dysfunctional high density lipoproteins
. J Biol Chem
11. Maisa A, Hearps AC, Angelovich TA, Pereira CF, Zhou J, Shi MD, et al. Monocytes from HIV-infected individuals show impaired cholesterol efflux and increased foam cell formation after transendothelial migration
12. Wang X, Magkos F, Mittendorfer B. Sex differences in lipid and lipoprotein metabolism: it's not just about sex hormones
. J Clin Endocrinol Metab