Implications of Immune Checkpoint Expression During Aging in HIV-Infected People on Antiretroviral Therapy

Abstract

Immune checkpoint molecules (ICMs) regulate T cell responses. In chronic viral infections and cancer, where antigens can persistently stimulate the immune system, ICMs can serve as a barrier to effective immune responses. The role of ICMs in the setting of systemic low-grade inflammation as in aging and antiretroviral therapy (ART)-suppressed HIV infection is not known. In this study, we made use of stored samples from the FLORAH cohort of HIV-infected ART-suppressed adults (age range 19-77 years.) and age-matched HIV-uninfected controls. We measured the expression levels of ICMs: PD-1, LAG-3, TIGIT, TIM-3, and 2B4 on resting CD4 and CD8 T cells and maturation subsets. To determine how expression of these molecules can affect T cell function, we stimulated peripheral blood mononuclear cell with HIV Gag or p09/H1N1 antigen and performed intracellular cytokine staining by multiparameter flow cytometry. ICMs were expressed at higher levels in CD8 compared with CD4. PD-1 was the only molecule that remained significantly higher in HIV-infected individuals compared with controls. LAG-3 expression increased with age in CD4 and CD8 T cells. 2B4 expression on CD8 T cells was negatively associated with IL-2 production but showed no effect on CD4 T cell function. TIM-3 expression was negatively associated with IL-21 production in CD4 and CD8 T cells and also negatively correlated with flu vaccine responses in HIV-negative individuals. Taken altogether, this study demonstrates the marked variation in ICM expression in T cells among adults and sheds light on the biology of these molecules and their effects on antigen-specific T cell functions. Overall, our results point to TIM-3 as a potential biomarker for immune function in HIV⁺ individuals on ART.

Keywords: aging; immune checkpoint molecules; virus suppression.

FIG. 1.

ICM expression on total CD4 and CD8 T cells. Representative flow plots showing expression of each marker on CD4 and CD8 T cells with no antibody controls (top). Negative control indicates a sample that was stained only with CD3, CD4, CD8, and live/dead antibodies to define gating for positive expression of ICM. Summary data for all study participants showing cell frequencies for each marker separated between HIV-infected participants (red) and HIV-uninfected HCs (black) (bottom). Each dot represents one individual. Student's t-test was performed to determine significant differences between the groups, p-values shown on graphs or indicated not significant (ns). ICM, immune checkpoint molecule.

FIG. 2.

T cell subset distribution and relationship in HCs and HIV⁺. (A) Representative flow plot showing T cell maturation subsets using CD45RO and CCR7 expression and corresponding pie charts showing the overall distribution using the mean of each subset in CD4 T cells (top) and CD8 T cells (bottom) from HC participants. (B) Summary data showing the comparison of T cell maturation distribution in HC and HIV⁺ participants in CD4 T cells (top) and CD8 T cells (bottom). Solid black lines indicate significant comparisons between two subsets within HC or HIV⁺ participant groups and asterisks (*) indicate comparison between HC and HIV⁺ for the indicated subset, significance was determined as p < .05 using two-way ANOVA statistical test adjusted for multiple comparisons. (C) Line graph showing the relationship of maturation subsets with age in CD4 and CD8 T cells. Lines represent linear regression for the cell frequency and the biological age. HC, healthy control.

FIG. 3.

Comparison of ICM expression by T cell subset in HCs. Pie charts showing the overall distribution using the mean expression of each ICM within the T cell maturation subsets in CD4 T cells (top row) and CD8 T cells (bottom row) for HC participants.

FIG. 4.

Markers of immune senescence. (A) Flow plots showing gating for immune senescent cells based on CD28 and CD57 expression. (B) Summary data for all study participants showing cell frequencies for immune senescent cells (CD57⁺CD28⁻) separated between HIV-infected participants (red) and HIV-uninfected HC (black). Student's t-test was performed to determine significant differences between the groups, p-values shown on graphs or indicated not significant (ns). (C) Pie charts showing the distribution of expression for CD57⁺CD28⁻ cells within the T cell maturation subsets in CD4 T cells (top row) and CD8 T cells (bottom row) for HC and HIV⁺ participants.

FIG. 5.

Correlation analysis of ICM expression and antigen-induced cytokine production. Heatmaps showing the spearman correlation coefficient (r) for each comparison in the matrix grouped by cytokine: IL-2, IFNγ, IL-21, or TNF. Antigens used for stimulation of PBMC from HC participants in top two rows and HIV participants in bottom three rows. (A) CD4 T cells and (B) CD8 T cells. Only significant correlations are shown using colored squares, p < 0.05. PBMC, peripheral blood mononuclear cell.

FIG. 6.

Correlation analysis of ICM expression and in vivo flu vaccine response. X–Y plots showing the correlation of ICM expression versus H1N1 titer fold change T2/T0 as a measure of response to the influenza vaccine. (A) CD8⁺2B4⁺ cell frequencies in HC and HIV and (B) CD4⁺TIM-3⁺ cell frequencies in HC and HIV. Spearman correlation analysis was performed; p < .05 is considered significant.

FIG. 7.

Correlation analysis of ICM expression and total HIV DNA. X–Y plots showing the correlation of ICM expression versus total HIV DNA levels in PBMC. (A) Indicated CD4 parameters and (B) indicated CD8 parameters. Spearman correlation analysis was performed; p < .05 is considered significant.