Preserved immune functionality and high CMV-specific T-cell responses in HIV-infected individuals with poor CD4+ T-cell immune recovery

Abstract

Poor CD4⁺ T-cell recovery after cART has been associated with skewed T-cell maturation, inflammation and immunosenescence; however, T-cell functionality in those individuals has not been fully characterized. In the present study, we assessed T-cell function by assessing cytokine production after polyclonal, CMV and HIV stimulations of T-cells from ART-suppressed HIV-infected individuals with CD4⁺ T-cell counts >350 cells/μL (immunoconcordants) or <350 cells/μL (immunodiscordants). A group of HIV-uninfected individuals were also included as controls. Since CMV co-infection significantly affected T-cell maturation and polyfunctionality, only CMV⁺ individuals were analyzed. Despite their reduced and skewed CD4⁺ T-cell compartment, immunodiscordant individuals showed preserved polyclonal and HIV-specific responses. However, CMV response in immunodiscordant participants was significantly different from immunoconcordant or HIV-seronegative individuals. In immunodiscordant subjects, the magnitude of IFN-γ⁺ CD8⁺ and IL-2⁺ CD4⁺ T-cells in response to CMV was higher and differently associated with the CD4⁺ T-cell maturation profile., showing an increased frequency of naïve, central memory and EMRA CMV-specific CD4⁺ T-cells. In conclusion, CD4⁺ and CD8⁺ T-cell polyfunctionality was not reduced in immunodiscordant individuals, although heightened CMV-specific immune responses, likely related to subclinical CMV reactivations, may be contributing to the skewed T-cell maturation and the higher risk of clinical progression observed in those individuals.

Figure 1

Phenotypic and functional changes induced by CMV infection. (A) CD4⁺ and CD8⁺ T-cell differentiation profile in HIV-uninfected CMV-seropositive and -seronegative individuals. The differential expression of CD45RA, CCR7 and CD27 by CD4⁺ and CD8⁺ T-cells was analyzed by boolean gating. Pie charts illustrate relative proportions of each of the different subsets in CMV-seropositive and CMV-seronegative. (B) CD4⁺ and CD8⁺ T-cell responses to PMA/ionomycin stimulation in HIV-uninfected individuals stratified by CMV serostatus. Polyfunctional profiles of CD4⁺ and CD8⁺ T-cells responding to PMA/ionomycin are showed. Detail of the 3-function combinations (IFN-γ, IL-2 and TNF-α) of CD4⁺ and CD8⁺ T-cells responding to PMA/ionomycin is depicted on the bottom of the figure. Statistical comparisons of the profiles were performed by partial permutation tests, using Spice software (*p < 0.05, **p < 0.01). HIV⁻: HIV-uninfected individuals; CMV⁻: CMV-seronegative individuals; CMV⁺: CMV-seropositive individuals.

Figure 2

CD4⁺ and CD8⁺ T-cell responses to PMA/ionomycin stimulation. (A) The frequency of CD4⁺ and CD8⁺ T-cells positive for each cytokine (IFN-γ, IL-2 and TNF-α) is represented (mean and SEM). (B) Polifunctionality index (PI) of HIV-uninfected individuals, immunoconcordant and immunodiscordant individuals (Mean and SEM are represented). (C) Pie charts representing the polyfunctional profile of CD4⁺ and CD8⁺ T-cells from HIV-uninfected individuals, immunoconcordant and immunodiscordant individuals. Arcs depict cytokine makeup within pie slice. (D) Dot graphs indicate percentage of responding cells based on different combinations of cytokine function. The combination of functions studied (pie slice) is indicated in the graph below (Median and IQR are represented). Nonparametric Mann-Whitney U test was used to analyze differences between groups (*p < 0.05, **p < 0.01). HIV⁻: CMV⁺ HIV-uninfected individuals; HIV⁺: CMV⁺ HIV-infected ART-treated individuals; IC: CMV⁺ immunoconcordant individuals; ID: CMV⁺ immunodiscordant individuals.

Figure 3

Functional CD4⁺ and CD8⁺ T-cell responses to CMV and HIV antigens. PBMCs were stimulated with different CMV antigenic preparation and the HIV p24 recombinant protein. The specific CD4⁺ and CD8⁺ T-cell responses were measured by flow cytometry in CMV⁺ HIV-uninfected individuals (blue bars, n = 12), CMV⁺ immunoconcordant individuals (green bars, n = 17 and n = 6 for CMV antigens and p24, respectively) and CMV⁺ immunodiscordant individuals (red bars, n = 12 and n = 8 for CMV antigens and p24, respectively). Response magnitudes are reported as the percentages of CD4⁺ and CD8⁺ T-cells producing IFN-γ and IL-2 after background subtraction. Bars indicate mean values ± SEM. Individual data of all subjects are represented by dots. Differences were tested using Mann-Whitney U nonparametric test (*p < 0.05).

Figure 4

Functionality of CMV and HIV-specific memory CD4⁺ and CD8⁺ T-cell subsets. The differential expression of CD45RA, CCR7 and CD27 by CD4⁺ and CD8⁺ T-cells was analyzed by boolean gating. Based on the expression of these surface markers we were able to discriminate eight different subpopulations expressing each possible combination of markers: naïve (T_N, CD45RA⁺CCR7⁺CD27⁺), central memory (T_CM, CD45RA⁻CCR7⁺CD27⁺), early and late effector memory (T_{EM early}, CD45RA⁻CCR7^-CD27⁺ and T_{EM late}, CD45RA⁻CCR7⁻CD27⁻), and early and late effector memory re-expressing CD45RA T-cells (T_{EMRA early}, CD45RA⁺CCR7⁻CD27⁺ and T_{EMRA late}, CD45RA⁺CCR7⁻CD27⁻). Other intermediate phenotypes (CD45RA⁺CCR7⁺CD27⁻ and CD45RA⁻CCR7⁺CD27⁻), which can not as yet be ascribed to a specific subpopulation or to a functionally unique subset, observed in low percentages are not shown. The frequency of IFN-γ⁺ and IL-2⁺ CMV and HIV-specific T-cells across distinct subsets are shown. CMV⁺ HIV-uninfected individuals (blue bars, n = 12), CMV⁺ immunoconcordant individuals (green bars, n = 16 and n = 6 for pp65 peptide pool and p24, respectively) and CMV⁺ immunodiscordant individuals (red bars, n = 11 and n = 8 for pp65 peptide pool and p24, respectively). The median and tukey ranges are shown for each group. Differences were tested using Mann-Whitney U nonparametric test (*p < 0.05, **p < 0.01).

Figure 5

Associations between total CMV-specific CD4⁺ T-cell response and CD4⁺ T-cell subsets. Correlations between total CMV-specific response (Median of total response IFN-γ and IL-2 production of all CMV antigens used) and CD4⁺ T-cell subsets are represented. CMV⁺ HIV-uninfected individuals (blue dots, n = 13), CMV⁺ immunoconcordant individuals (green dots, n = 18) and CMV⁺ immunodiscordant individuals (red dots, n = 13). Linear correlation (Spearman) r and p-values are shown.

Figure 6

Phenotypic profile of IFN-γ ⁺ virus-specific CD4⁺ and CD8⁺ T-cells. IFN-γ⁺ CMV- and HIV-specific CD4⁺ and CMV-specific CD8⁺ T-cells were determined after stimulation with different CMV antigens and the HIV p24 recombinant protein. The memory maturation profile of the IFN-γ⁺ specific cells was examined for CD45RA, CCR7 and CD27 expression by using a boolean gated strategy. The distribution of CMV- and HIV-specific IFN-γ⁺ response among different subsets was represented using SPICE software. The phenotypic patterns are color-coded and indicated. Significant differences between antigens are depicted in different colors: blue for CMV⁺ HIV⁻ individuals, green for CMV⁺ immunoconcordant (IC) and red for CMV⁺ immunodiscordant (ID). Significant differences between groups (intra-antigen) are also represented (black lines). Statistical testing by permutation performed with SPICE software (*p < 0.05, **p < 0.01, ***p < 0.001).