IL-1Β enriched monocytes mount massive IL-6 responses to common inflammatory triggers among chronically HIV-1 infected adults on stable anti-retroviral therapy at risk for cardiovascular disease

Abstract

Chronic infection by HIV increases the risk of cardiovascular disease (CVD) despite effective antiretroviral therapy (ART). The mechanisms linking HIV to CVD have yet to be fully elucidated. High plasma levels of the pro-inflammatory cytokine IL-6, which may be triggered by IL-1β, is a biomarker of CVD risk in HIV-negative adults, and of all-cause mortality in HIV disease. Monocytes play a pivotal role in atherosclerosis, and may be major mediators of HIV-associated inflammation. We therefore hypothesized that monocytes from HIV-infected adults would display high inflammatory responses. Employing a 10-color flow cytometry intracellular cytokine staining assay, we directly assessed cytokine and chemokine responses of monocytes from the cryopreserved peripheral blood of 33 chronically HIV-1 infected subjects. Participants were 45 years or older, on virologically suppressive ART and at risk for CVD. This group was compared to 14 HIV-negative subjects matched for age and gender, with similar CVD risk. We simultaneously detected intracellular expression of IL-1β, IL-6, IL-8 and TNF in blood monocytes in the basal state and after stimulation by triggers commonly found in the blood of treated, chronically HIV-infected subjects: lipopolysaccharide (LPS) and oxidized low-density lipoprotein (oxLDL). In the absence of stimulation, monocytes from treated HIV-infected subjects displayed a high frequency of cells producing IL-1β (median 19.5%), compared to low levels in HIV-uninfected persons (0.9% p<0.0001). IL-8, which is induced by IL-1β, was also highly expressed in the HIV-infected group in the absence of stimulation, 43.7% compared to 1.9% in HIV-uninfected subjects, p<0.0001. Strikingly, high basal expression of IL-1β by monocytes predicted high IL-6 levels in the plasma, and high monocyte IL-6 responses in HIV-infected subjects. Hyper-inflammatory IL-1β enriched monocytes may be a major source of IL-6 production and systemic inflammation in HIV-infected adults, and may contribute to the risk for all-cause mortality and cardiovascular disease in treated HIV infection.

Figure 1. Gating strategy for identification of total monocytes, monocyte subsets and detection of cytokine expression.

Gating Strategy. Panel A. Identification of total monocytes from peripheral blood mononuclear cells by exclusion of doublets, dead cells, CD3, CD56, CD19, CD20 and low HLA-DR expressing cells. Double-negative CD14-CD16- cells were also excluded. Forward scatter (FSC) vs. side scatter (SSC) plots are shown comparing total PBMC (doublet and dead cells excluded) and total monocytes. Panel B. Monocyte subsets were identified based on the expression of CD14 and CD16 (left column). The diagram at the bottom of the left column is a visual guide for the terminology of monocyte subsets employed in this report. Intracellular cytokines (IL-1β, IL-8/CXCL8, IL-6 and TNF) produced in total monocytes were detected in response to no stimulus, oxidized low density lipoprotein (oxLDL) or lipopolysaccharide (LPS). Fluorescence minus one control condition, in which the antibody conjugate in question is omitted to guide creation of the gate that defines positive expression of that target, is shown on the bottom row. The subject presented is HIV-infected and displays high but representative responses to stimuli.

Figure 2. Monocyte production of pro-inflammatory cytokines in the basal state and upon stimulation.

HIV-infected subjects are shown in red circles and HIV-uninfected subjects are shown in blue triangles. In the no stimulation condition (basal state), HIV-infected subjects showed higher levels of IL-1β (upper left panel) and IL-8 (lower left panel). Upon stimulation with either oxLDL or LPS, HIV-infected subjects exhibited higher levels of IL-1β, IL-8 and IL-6 (upper right panel) compared to the HIV-uninfected subjects. While HIV-1 infected subjects did tend to have higher TNF (lower right panel) responses upon stimulation, these differences were not significant. **** p < 0.0001, *** p < 0.001, ** p < 0.01.

Figure 3. Induction of broad, high frequency, polyfunctional monocyte responses to stimuli.

HIV-infected subjects are shown in circles, HIV-uninfected are shown in triangles. Top Panel: Unstimulated (basal) state, Middle Panel: After oxLDL stimulation, Bottom Panel: After LPS stimulation. To determine the frequency of cells producing one or more cytokines, we performed a sum of the median responses. Mann Whitney U was used for comparison tests, and all comparisons were adjusted for multiple comparisons (p = 0.05/16 comparisons per stimulation condition = p = 0.003), and only p-values meeting or exceeding the p-value threshold are shown. **** p < 0.0001, *** p < 0.001, ** p < 0.003.

Figure 4. HIV-1 infected subjects with well-controlled viremia have a greater proportion of classical (Mono1, CD14++CD16-) monocytes and a lower frequency of intermediate (Mono2, CD14++CD16+) monocytes, which correlates with production of IL-1β.

HIV-infected subjects are shown in red circles, and HIV-uninfected subjects are shown in blue triangles. Panel A. HIV-infected subjects had a greater fraction of monocytes that fall into the Mono1 (CD14++CD16- classical) subset, and a lower fraction that fall into the Mono2 (CD14++CD16+ intermediate) subsets. Upper right diagram represents monocyte gating scheme (Mono 1-4). Panel B. A lower proportion of monocytes in the Mono2 subset was associated with higher basal IL-1β production. **** p < 0.0001, *** p < 0.001,.

Figure 5. High basal IL-1β and IL-8 predict high plasma IL-6 levels and a high IL-6 response to oxLDL and LPS.

HIV-infected subjects are shown in red circles, and HIV-uninfected subjects are shown in blue triangles. Panel A. In HIV-infected subjects, high basal IL-1β (upper row) or high basal IL-8 (lower row) in total monocytes was associated with higher plasma IL-6 (left column). Higher levels of IL-1β and IL-8 production in the basal state predicted high levels of IL-6 response after stimulation by either oxLDL (middle column) or LPS (right column). Exact p-values shown, NS indicates higher than the Bonferroni adjusted p-value threshold of p = 0.0133 (0.05/3 comparisons per cytokine). Panel B. Association between basal production of IL-1β and basal production of IL-8 of total monocytes.

Figure 6. Per cell expression levels of IL-1β and TNF by each monocyte subset before and after stimulation.

Geometric mean fluorescence (GMF) intensities are shown by monocyte subset for IL-1β (left column) and TNF (right column) in the basal (top row) state, and after oxLDL (middle row) or LPS (bottom row) stimulation. In HIV-infected subjects, IL-1β was expressed at highest levels by the Mono1 (CD14++CD16-) and Mono2 (CD14++CD16+) populations. TNF expression was observed in all subsets after stimuli, with the brightest monocyte subsets being Mono2 (CD14++CD16+) and Mono3 (CD14+ CD16+). Statistical significance was adjusted for multiple comparisons (p = 0.05/4 comparisons per stimulation condition: p = 0.0125). **** p < 0.0001, *** p < 0.001, ** p < 0.01, * p < 0.0125.