High-parameter phenotypic characterization reveals a subset of human Th17 cells that preferentially produce IL-17 against M. tuberculosis antigen

Abstract

Background: Interleukin-17-producing CD4 T cells contribute to the control of Mycobacterium tuberculosis (Mtb) infection in humans; whether infection with human immunodeficiency virus (HIV) disproportionately affects distinct Th17-cell subsets that respond to Mtb is incompletely defined.

Methods: We performed high-definition characterization of circulating Mtb-specific Th17 cells by spectral flow cytometry in people with latent TB and treated HIV (HIV-ART). We also measured kynurenine pathway activity by liquid chromatography-mass spectrometry (LC/MS) on plasma and tested the hypothesis that tryptophan catabolism influences Th17-cell frequencies in this context.

Results: We identified two subsets of Th17 cells: subset 1 defined as CD4⁺Vα7.2^-CD161⁺CD26⁺and subset 2 defined as CD4⁺Vα7.2^-CCR6⁺CXCR3^-cells of which subset 1 was significantly reduced in latent tuberculosis infection (LTBI) with HIV-ART, yet Mtb-responsive IL-17-producing CD4 T cells were preserved; we found that IL-17-producing CD4 T cells dominate the response to Mtb antigen but not cytomegalovirus (CMV) antigen or staphylococcal enterotoxin B (SEB), and tryptophan catabolism negatively correlates with both subset 1 and subset 2 Th17-cell frequencies.

Conclusions: We found differential effects of ART-suppressed HIV on distinct subsets of Th17 cells, that IL-17-producing CD4 T cells dominate responses to Mtb but not CMV antigen or SEB, and that kynurenine pathway activity is associated with decreases of circulating Th17 cells that may contribute to tuberculosis immunity.

Keywords: ART-suppressed; CD4 T-cells; HIV; antigen-responsive; immunity; interleukin-17; kynurenine pathway; tuberculosis.

Figure 1

Different definitions of T helper 17 cells reveal high but incomplete concordance. (A) Representative flow cytometry plots and gating schemes for definition of Th17 cells by published marker criteria independent of cytokine expression. Right panel, chemokine receptor CCR6 and CXCR3 expression used to distinguish subset 2, Th1^*, and Th1 non-MAIT CD4 T cells. Left panel: Use of surface expression of CD26 and CD161 to identify subset 1 non-MAIT CD4 T cells. Cells in both panels were stimulated with the MTB300 antigenic peptide pool. (B) High correlation of subset 1 (CD26⁺CD161⁺) and subset 2 (CCR6⁺CXCR3^-) non-MAIT CD4 T cells in cell populations from participants with varying TB and HIV status. Spearman correlation p and r values are shown for results pooled from all participants. (C) t- stochastic neighbor embedding (t-SNE) analysis reveals incomplete concordance of transcription factor (RORγt and T-bet) and cytokine (IFNγ and IL17) production and surface phenotypes of non-MAIT CD4⁺T cells. In this analysis, protein transport inhibitors (golgi stop and golgi plug) were not added to PBMC during antigen stimulation thus cytokine detection was not optimal. Prior to TST⁺= participants were considered Mtb unexposed due to a negative TST test; INH = isoniazid therapy. Statistic: Spearman correlation.

Figure 2

ART treated HIV infection is associated with disproportionate depletion of distinct subsets of T helper cells. (A) Reduced frequency of bulk CD4 T cells in HIV⁺vs HIV^-participants in Cohort 1. (B) ART-treated HIV is not associated with differential distribution of CD4 T cells in memory T cell subsets. Naive: CD45RA⁺CCR7⁺; Central Memory (CM): CD45RA-CCR7⁺; Effector Memory: CD45RA^-CCR7^-; TEMRA: CD45RA⁺CCR7^-. (C) Disproportionate reduction in frequencies of subset 1 and Th1^* but not subset 2 or Th1 cells in ART-treated HIV. Subset 1 = CD4⁺Vα7.2^-CD26⁺CD161⁺; Th1^* (also termed Th1Th17): CD4⁺Vα7.2^-CCR6⁺CXCR3⁺; subset 2: CD4⁺Vα7.2^-CCR6⁺CXCR3^-; and Th1: CD4⁺Vα7.2^-CCR6^-CXCR3⁺. Statistical comparisons were by Mann-Whitney test.

Figure 3

In contrast to responses to other stimuli, Mtb antigen responsive IL17 expression is enriched in subset 1 (CD26⁺CD161⁺) rather than subset 2 (CCR6⁺CXCR3^-) Th17 cells. (A) Left panel (Cohort 1): CD4 T cells that produce IL17 in response to MTB300 are predominantly in subset 1. Right panel (Cohort 2): MTB300-stimulated IL17 production predominates in subset 1 rather than subset 2 cells regardless of TST status and is not altered by INH preventive therapy. (B) Left panel (Cohort 1): In contrast to MTB300-responsive cells, IL17 production predominates in subset 2 cells in response to SEB stimulation. Right panel (Cohort 2): INH preventive therapy does not alter the pattern of IL17 production by SEB-stimulated subset 2 or subset 1 Th17 cells. (C) Left panel (Cohort 1): In ART-treated HIV participants, CD4 T cells that produce IFNγ in response to MTB300 stimulation are present at low frequencies. Right panel (Cohort 2): (D) Left panel (Cohort 1): CMV pp56 antigen stimulation induces higher frequencies of IFNγ than IL17-producing CD4 T cells. Right panel (Cohort 2): INH treatment does not alter CD4 T cell responses to CMV pp65 antigen stimulation. Statistics: Wilcoxon test.

Figure 4

In TST⁺participants, HIV infection is associated with increased indoleamine-2,3-dioxygenase (IDO) activity, which negatively correlates with Th17 cell subsets. (A) Plasma tryptophan concentrations; (B) Plasma kynurenine concentrations; (C) Plasma Kynurenine (K)/Tryptophan (T) ratios as a reflection of IDO activity. (D) Left panel: correlation of subset 1 Th17 cell frequencies with plasma K/T ratios; Middle panel: Correlation of regulatory T cells and plasma K/T ratios; Right panel: Correlation of the ratio of subset 1/Treg ratios and plasma K/T ratios. (E) Left panel: correlation of subset 2 Th17 cell frequencies with plasma K/T ratio; Right panel: Correlation of the ratio of subset 2/Treg ratios and plasma K/T ratios. Statistical analyses were by Mann-Whitney (A–C) or Spearman correlation (D, E).