Monocytes from HIV-infected individuals show impaired cholesterol efflux and increased foam cell formation after transendothelial migration

Abstract

Design: HIV-infected (HIV+) individuals have an increased risk of atherosclerosis and cardiovascular disease which is independent of antiretroviral therapy and traditional risk factors. Monocytes play a central role in the development of atherosclerosis, and HIV-related chronic inflammation and monocyte activation may contribute to increased atherosclerosis, but the mechanisms are unknown.

Methods: Using an in-vitro model of atherosclerotic plaque formation, we measured the transendothelial migration of purified monocytes from age-matched HIV+ and uninfected donors and examined their differentiation into foam cells. Cholesterol efflux and the expression of cholesterol metabolism genes were also assessed.

Results: Monocytes from HIV+ individuals showed increased foam cell formation compared with controls (18.9 vs. 0%, respectively, P = 0.004) and serum from virologically suppressed HIV+ individuals potentiated foam cell formation by monocytes from both uninfected and HIV+ donors. Plasma tumour necrosis factor (TNF) levels were increased in HIV+ vs. control donors (5.9 vs. 3.5 pg/ml, P = 0.02) and foam cell formation was inhibited by blocking antibodies to TNF receptors, suggesting a direct effect on monocyte differentiation to foam cells. Monocytes from virologically suppressed HIV+ donors showed impaired cholesterol efflux and decreased expression of key genes regulating cholesterol metabolism, including the cholesterol transporter ABCA1 (P = 0.02).

Conclusion: Monocytes from HIV+ individuals show impaired cholesterol efflux and are primed for foam cell formation following transendothelial migration. Factors present in HIV+ serum, including elevated TNF levels, further enhance foam cell formation. The proatherogenic phenotype of monocytes persists in virologically suppressed HIV+ individuals and may contribute mechanistically to increased atherosclerosis in this population.

Figure 1. Monocytes from HIV+ subjects show increased foam cell formation which is enhanced by HIV+ serum

Percentage of foam cells present within collagen were measured following Oil Red O staining, as described in Methods, 48 hours after monocyte migration for seven HIV− controls and eight HIV+ male donors using (A) non-activated and (B) TNF-activated endothelium (HUVEC) in the presence of either pooled human control serum (pHS – identical batch used for all samples) or autologous serum (AS) from each individual donor. Bars represent median and interquartile range of data collected from 11 independent experiments. (C) Percentage of foam cells present within collagen was measured using monocytes prepared from six different HIV-negative donors added to TNF-activated HUVEC, and incubated with either pHS or serum from five different HIV+ donors receiving cART and with viral load <20 copies/ml. Box and whisker plots show median, IQR and 1.5xIQR (Tukey method). Paired two-tailed Wilcoxon rank sum test was used for within group comparisons and unpaired two-tailed Mann Whitney U test for between group comparisons. *p<0.05, **p<0.01.

Figure 2. Tracking of monocyte migration via live cell imaging

Images were obtained from a representative experiment measuring migration over 48 hours of PKH26-labeled monocytes from a control donor (A) and HIV+ donor (B) in the presence of pooled human serum.. Cells were incubated at 37°C in a humidified temperature controlled chamber coupled to a DeltaVision microscope, images were captured using a CCD camera through a 20x 0.45 NA lens and cell movement was tracked over 48 hours. The color gradients represent time with red as the earliest time (closest to 0 hours) and yellow as latest time (closest to 48 hours; see also Fig 2mov1 and Fig 2mov2). (C) Tracks from live cell imaging experiments that could be assigned unambiguously to individual cells were identified and their displacement lengths computed using Imaris software. Frequency histogram of migration track lengths for monocytes added to unactivated HUVEC from HIV− controls in the presence of pooled human serum (pHS) (ctrl; black bars, 64 tracks combined from four independent experiments) or monocytes from HIV+ donors in the presence of pHS (green bars, 51 tracks/3 experiments) or autologous serum (AS; red bars, 35 tracks/3 experiments). Histograms are nudged by 10 data points. Average monocyte migration speed recorded in the first hour (D) or last 10 hours (E; hours 38–48) of migration for monocytes from HIV− controls in the presence of pHS and HIV+ donors in the presence of pHS or AS was computed using Imaris software. Box and whisker plots show median, IQR and 1.5xIQR (Tukey method; dots represent points outside these ranges). Significance was tested using an unpaired two-tailed Mann Whitney U test, **** p<0.0001.

Figure 3. CXCL10 and TNF are elevated in HIV+ individuals and TNF acts at a post-migration stage to enhance foam cell formation

Concentration of (A) CXCL10 and (B) TNF in plasma from all subjects used in Figure 1 was measured by ELISA. Bars indicate median and interquartile range. * p<0.05, as determined by Mann-Whitney U test. (C) Percentage of foam cells within collagen gels was measured 48 hours after adding monocytes to TNF-activated HUVEC in the presence of the indicated amounts of antibodies to TNF receptors I and II (α-TNF), or isotype control (Iso) antibodies. Data are from monocytes isolated from six HIV-negative individuals incubated with pooled human serum (pHS) (left columns, n=6) or serum from virologically suppressed HIV+ donors (right, hatched columns; monocytes from each donor incubated with three different HIV+ sera, n=18). Differences between isotype and α-TNF treated samples were detected using paired non-parametric two-tailed Wilcoxon rank sum test and between group differences detected using unpaired non-parametric two-tailed Mann Whitney U test, *p<0.05, ****p<0.0001.

Figure 4. Flow cytometry analysis and sorting of transmigrated monocytes extracted from collagen gels

(A) Upper panels: Monocytes from a representative HIV− donor were extracted from collagen gels 48 hours after transmigration, stained with BODIPY 493/503 and anti-CD14-APC, then analyzed by flow cytometry. Left panel: FSC vs. SSC plot; Right panel: CD14-APC vs. BODIPY-FITC scatter plot. The two indicated populations were isolated by cell sorting. Bottom panel: FACS-purified cells from the gates shown were counterstained with Hoechst 33258, centrifuged onto glass slides and mounted in Fluoromount-G. Images were captured on a charge-coupled device (CCD) camera (AxioCamMRm Rev. 3, Carl Zeiss, Germany) through a 100x 1.3 numerical aperture oil immersion lens on a Zeiss Axio Observer.Z1 inverted microscope. (B–F). Cells extracted from collagen gels after 48 hours were analyzed by flow cytometry for the following markers: CD14 (B), BODIPY (C), CD36 (D), CD11b (E) and intracellular TNF (F). Box and whisker plots show median, IQR and 1.5 standard deviations (Tukey plot). *p<0.05; **p<0.01 as determined by paired non-parametric two-tailed Wilcoxon rank sum test

Figure 5. Monocytes from virologically-suppressed HIV+ donors show altered expression of cholesterol metabolism genes and impaired cholesterol efflux

PBMC from controls and virologically-suppressed HIV+ individuals (n=8 for each) were loaded with BODIPY-cholesterol and subsequent cholesterol efflux from monocytes was measured via flow cytometry. The percentage cholesterol efflux after 30 minutes (as a proportion of total loaded) is shown for experiments conducted in the absence (A) or presence (B) of the cholesterol accepter methyl-β-cyclodextrin. Gene expression of the cholesterol metabolism genes ABCA1 (C), ABCG1 (D), HMG-CoA reductase (E) and ACAT (F) was determined via qPCR analysis of mRNA isolated from purified monocytes from controls and virologically-suppressed HIV+ donors (n=11 in each group) and standardized to GAPDH expression. Box and whisker plots show median, IQR and 1.5 standard deviations (Tukey plot). Significance was tested using an unpaired two-tailed Mann Whitney U test, *p<0.05.