Expansion of CD8+ T cells lacking Sema4D/CD100 during HIV-1 infection identifies a subset of T cells with decreased functional capacity

Abstract

Sema4D, also known as CD100, is a constitutively expressed immune semaphorin on T cells and NK cells. CD100 has important immune regulatory functions that improve antigen-specific priming by antigen-presenting cells, and can also act as a costimulatory molecule on T cells. We investigated the consequence of HIV-1 infection on CD100 expression by T cells, and whether CD100 expression signifies functionally competent effector cells. CD100 expression on T cells from healthy individuals was compared with HIV-1-infected subjects including elite controllers, noncontrollers, and patients receiving antiretroviral therapy. The frequency and fluorescence intensity of CD100 on CD8(+) and CD4(+) T cells were decreased during HIV-1 infection. Furthermore, the absolute number of CD100-expressing CD8(+) T cells was positively associated with the magnitude of HIV-1-specific T-cell responses. CD8(+) T cells lacking CD100 expression were functionally impaired and present in increased numbers in HIV-1-infected individuals. The number of CD100(-)CD8(+) T cells positively correlated with T-cell immunosenescence, immune activation, and viral load. Loss of CD100 expression appears to result from direct antigen stimulation, as in vitro cytokine exposure and viral replication did not significantly impact CD100 expression. These data suggest that loss of CD100 expression probably plays an important role in dysfunctional immunity in HIV-1 infection.

Figure 1

CD100 T cell expression in HIV-1–infected subjects is altered compared with healthy controls. PBMCs from early HIV-1–infected patients (n = 12), elite controllers (EC, n = 20), viremic noncontrollers (NC, n = 20), and healthy controls (HCs, n = 15) were surface stained for CD100 expression. (A) Absolute numbers of CD8⁺ T cells and CD4⁺ T cells expressing CD100. (B) Absolute CD8⁺ and CD4⁺ T cells. (C) Percentage of CD8⁺ T cells or CD4⁺ T cells expressing CD100. (D) Geometric mean fluorescence intensity (gMFI) of CD100 on CD8⁺ T cells and CD4⁺ T cells. (E-G) CD100 expression on CD8⁺ T cells from early HIV-infected subjects at time points, (i) within the first year of HIV-1 infection (early), (ii) 1 to 5 years after infection (late), and (iii) within 3 months after starting ART (treatment). (E) Absolute numbers of CD100⁺CD8⁺ T cells, (F) percentage of CD100 expression, and (G) CD100 gMFI. Statistical analysis was performed using (A-D) Kruskal Wallis tests with Dunn posttest; (E-G) paired t tests (*P < .05, **P < .01, and ***P < .001).

Figure 2

CD100 frequency and gMFI is not significantly modified by sustained ART but reconstitutes CD100⁺ naive CD8⁺ T cells. (A) Representative dot plot depicting the different CD8⁺ T-cell memory populations (TCM indicates central memory T cell; TEM, effector memory T cell; and TEMRA, terminally differentiated effector memory T cell). (B) Absolute numbers of CD100⁺CD8⁺ T cells, (C) percentage of CD100 expression, and (D) CD100 gMFI within each memory subset in HCs, untreated HIV-1–infected subjects (HIV⁺ ART⁻), and treated HIV-1–infected subjects (HIV⁺ ART⁺). Statistical analysis was performed using Kruskal Wallis tests with Dunn posttest (*P < .05, **P < .01, and ***P < .001).

Figure 3

Antigen-specific CD8⁺ T-cell expression of CD100 is decreased and the absolute numbers of CD100 expressing total CD8⁺ T cells, naive cells and TEMRAS correlate with p24-specific T-cell responses in untreated, viremic chronically infected individuals. Representative dot plots of (A) CMV pentamer-specific T cells in HC and (B) HIV-1–specific CD8⁺ T cells in a chronically HIV-1–infected individual. (C) Histogram of CD100 expression by pentamer-specific CD8⁺ T cells. (D) Summary of CD100 expression on antigen-specific CD8⁺ T cells in HCs and HIV-1 patients. Comparison of the (E) percentage of CD100 expression and (F) CD100 gMFI in HIV-1 infected subjects with both CMV and HIV-1–specific CD8⁺ T cells within the same individual. Correlation of absolute numbers of CD100 expressing (G) CD8⁺ T cells, (H) naive CD8⁺ T cells, and (I) CD8⁺ TEMRA cells with p24-specific T-cell responses in untreated, viremic chronically infected HIV-1 patients (n = 53). (J) Correlation of CD100⁺ CD8⁺ T cells with CD4⁺ T-cell count in both untreated early and chronically infected patients (n = 84). Statistical analysis was performed using (D) Kruskal Wallis tests with Dunn posttest, and (E-F) 2-tailed paired Student t tests. (G-J) Correlations were determined by 2-tailed nonparametric Spearman correlations.

Figure 4

Expansion of CD100⁻CD8⁺ T cells coexpressing CD57 and PD-1 in HIV-1–infected individuals. Correlation of the number of CD100⁻CD8⁺ T cells with (A) immune activation as assessed by percentage of CD38⁺HLA-DR⁺CD8⁺ T cells of all untreated subject groups (n = 84), or (B) viral loads of all untreated subject groups (black line; n = 84) or only untreated viremic patients (gray line; n = 65). (C) Correlation of the number of CD100⁻CD8⁺ T cells during the first year of infection and the AUC for viral loads over time (n = 12). D) Frequency of CD100 coexpression with CD57 and PD-1 in the total CD8⁺ T cell population of ECs (n = 20) and untreated, chronically infected NCs (n = 60) compared with HCs. Shaded bars represent minimum-maximum range and the black middle bar indicates the median. (E) Frequency of CD100, CD57, and PD-1 coexpression in HIV (lightest gray) and CMV (medium gray) pentamer-specific CD8⁺ T-cell population of HIV-1–infected subjects and CMV pentamer-specific CD8⁺ T cell population of healthy controls (darkest gray). Open bars represent mean and symbols represent individual subjects. Statistical analysis: (A-C) Correlations were determined by 2-tailed nonparametric Spearman correlations; (D-E) Student t tests (*P < .05, **P < .01, and ***P < .001).

Figure 5

CD100⁻CD8⁺ T cells are functionally impaired after stimulation. (A) Representative ELISPOT wells (left column) illustrating IFN-γ production in response to PMA/ionomycin in FACS sorted CD100⁺ or CD100⁻ CD8⁺ T cells from HCs (n = 3). Control wells containing cells in media only are shown on the right column of wells. (B) Summary of 3 independent ELISPOT experiments. (C) Assessment of IFNγ, MIP-1β IL-2, TNF-α, perforin, and granzyme B production by Luminex assay after PMA/ionomycin stimulation representing cytokine production per 1000 cells (n = 9). D) Assessment of IL-2, IFN-γ, perforin production by Luminex assay after CMV antigen-specific stimulation with CMV peptide-pulsed autologous PBMCs representing cytokine production per 1000 antigen-specific T cells (n = 6). Statistical analysis was performed using (B) Student t tests (***P < .001), and (C-D) 2-tailed paired t tests.

Figure 6

CD100 expression is down-regulated on antigen-specific T cells after peptide stimulation but not after in vitro cytokine culture or HIV-1 infection. (A) CD100 gMFI after 7 days of culture in the absence or presence of NL4–3 infection. (B) CD100 gMFI after 24 hours of incubation in the presence of proinflammatory or homeostatic cytokines or anti-CD3 plus anti-CD28. (C) Representative dot plots illustrating IFN-γ expression in CD100⁻ CD8⁺ T cells after CEF or p24 peptide pool stimulation. (D) Percentage of CD100⁺ and CD100⁻ CD8⁺ T cells expressing IFN-γ after CEF or p24 stimulation. HCs were assessed for CD100 expression after CEF stimulation. HIV-1 infected subjects were assessed for CD100 expression after p24 or CEF stimulation. CEF and p24 responses in HIV⁺ subjects are not necessarily matched from the same individual. (E) CD100 frequency on CMV pentamer-specific CD8⁺ T cells (open circles) in HC and HIV-1 infected individuals and HIV-1 pentamer-specific CD8⁺ T cells (closed square) in HIV-1 infected individuals compared with IFN-γ producing cells after PBMC peptide-stimulation with the peptide corresponding to the pentamer within the same individual. Statistical analysis was performed using (D) Student t tests between CD100⁺ and CD100⁻ cells, and (E) 2-tailed, paired Student t tests (*P < .05 and **P < .01). CEF = pool of peptides from cytomegalovirus, Epstein-Barr, and influenza viruses.