Staphylococcus aureus-specific TIGIT+ Treg are present in the blood of healthy subjects - a hurdle for vaccination?

Abstract

Staphylococcus aureus poses an enormous burden of morbidity and mortality worldwide. Making an efficacious vaccine has however proven extremely challenging. Due to colonizing interactions, pre-existing S. aureus-specific CD4⁺ T cells are often found in the human population and yet a detailed characterization of their phenotypes and how they might in turn impact vaccine efficacy are thus far unknown. Using an activation induced marker assay to sort for S. aureus-specific CD4⁺ T cells in an effector function-independent manner, single cell transcriptomic analysis was conducted. Remarkably, S. aureus-specific CD4⁺ T cells consisted not only of a broader spectrum of conventional T cells (Tcon) than previously described but also of regulatory T cells (Treg). As compared to polyclonally-activated CD4⁺ T cells, S. aureus-specific Tcon were enriched for the expression of the Th17-type cytokine genes IL17A, IL22 and IL26, while higher percentages of S. aureus-specific Treg expressed the T Cell Immunoreceptor with Ig and ITIM domains (TIGIT), a pleiotropic immune checkpoint. Notably, the antagonistic anti-TIGIT mAb Tiragolumab increased IL-1β production in response to S. aureus in vitro. Therefore, these results uncover the presence of S. aureus-specific TIGIT⁺ Treg in the blood of healthy subjects that could blunt responses to vaccination and indicate TIGIT as a potential targetable biomarker to overcome pre-exposure-induced immunosuppression.

Keywords: Staphylococcus aureus; TIGIT; Th17; Treg; colonization; host-pathogen interactions; vaccines.

Figure 1

AIM assay identified CD4⁺ T cells specific for S. aureus that are enriched for the skin-homing marker CLA in the blood of healthy subjects. PBMCs from healthy donors were enriched via negative selection into CD4⁺ T cells (CD4⁺) and CD4⁺ T cell-depleted (CD4^-) fractions. CD4^- fraction was irradiated and mixed at a 1:1 ratio with the CD4⁺ fraction. Cells were then stimulated for 24 h with HK S. aureus, or α-CD3/CD28 as positive control, or left untreated (Medium) as negative control. (A) Flow-cytometric plots showing co-expression of OX40 and CD137 (AIM⁺) on CD4⁺ T cells from one representative donor after stimulation with HK S. aureus (S. aureus-specific) or α-CD3/CD28 antibodies (polyclonally-activated). See Supplementary Figure 1 for gating strategy. (B) Percentages of AIM⁺ CD4⁺ T cells for 16 healthy donors analyzed. Each circle represents a donor, colored circles represent the 6 donors that were further analyzed for CLA expression in D and by SORT-seq. Mean values ± 95% CI for each culture condition are also shown. Dotted line represents the arbitrary cut-off (0.16% AIM⁺ CD4⁺ T cells, set doubling the highest percentage of AIM⁺ CD4⁺ T cells observed without stimulation) used to identify responders to S. aureus (12 donors out of 16). (C) Flow-cytometric plots showing CLA expression on S. aureus-specific and polyclonally-activated CD4⁺ T cells from one representative donor. (D) Donor-matched percentages of S. aureus-specific or polyclonally-activated CLA⁺CD4⁺ T cells from 6 donors (indicated by colored circles in B). Statistical analysis was performed using a paired t-test. P ≤ **0.01.

Figure 2

SORT-Seq analysis of S. aureus-specific and polyclonally-activated CD4⁺ T cells revealed 3 cell clusters: Activated T cells (Tact), Trm-like cells and Treg. PBMCs isolated from 6 healthy donors (coloured circles in Figures 1B, D ) were separated via negative selection into CD4⁺ and CD4^- fractions. After irradiation of CD4^- cells, both fractions were re-mixed at a 1:1 ratio and stimulated with HK S. aureus or polyclonally with α-CD3/CD28. CD4⁺ T cells co-expressing OX40 and CD137 (AIM⁺) after 24 h stimulation were sorted via flow-cytometry and analyzed by SORT-seq. (A) UMAP representation of scRNA-seq merged data of S. aureus-specific and polyclonally-activated CD4⁺ T cells from 6 donors (4,690 cells in total). Merged datasets were divided into 3 clusters via Leiden clustering and annotated using different colors. (B) Heatmap of unsupervised clustering analysis showing the fold upregulation of the top 20 genes used to differentiate each of the 3 clusters shown in (A, C) Trajectory analysis starting from CCR7-expressing CD4⁺ T cells (black dot in the center of the UMAP). Heatmap showing that cells along the trajectory displayed signature differences increasingly associated with different cell function of Tact, Trm-like and Treg cells. (D) Gene set enrichment analysis (GSEA) reporting the Normalized Enrichment Score (NES) for each of the 3 clusters using the top 50 cluster-defining genes compared against gene lists obtained from published literature. Circle size indicates proportion of genes present in a specific gene set whereas the color scale refers to P value. (E) Proportional representations in each of the 3 clusters for each of the 6 donors analyzed per stimulation condition. No significative difference by two-tailed t-test among S. aureus-specific and polyclonally-activated CD4⁺ T cells.

Figure 3

The cytokine signature of S. aureus-specific CD4⁺ T cells from healthy donors consists of IL17A, IL22, and IL26. (A-C) PBMCs isolated from 6 healthy donors were sorted via negative selection into CD4⁺ and CD4^- fractions. CD4^- cells were irradiated and both fractions were re-mixed at a ratio of 1:1 and stimulated for 24 h with HK S. aureus or α-CD3/CD28. AIM⁺ CD4⁺ T cells were then FACS-sorted and scRNA-seq analysis performed. (A) UMAPs showing the distinct cell localizations for each stimulation condition: S. aureus-specific and polyclonally-activated CD4⁺ T cells in red and blue, respectively. The distribution and co-localization of cytokine transcription was assessed by mapping areas of expression (represented by colored contours) for the reported cytokine-encoding genes. (B) Heatmap representing the percentages of S. aureus-specific and polyclonally-activated CD4⁺ T cells transcribing a panel of 14 cytokines calculated for each donor. Cytokines were listed from the one expressed by the highest to the one expressed by the lowest percentage of S. aureus-specific CD4⁺ T cells. Cytokine genes expressed by statistically higher percentages of S. aureus-specific vs. polyclonally-activated CD4⁺ T cells are indicated by red asterisks while those expressed by higher percentages of polyclonally-activated cells are indicated by blue asterisks. (C) Each symbol represents the percentage of S. aureus-specific or polyclonally-activated CD4⁺ T cells transcribing CCR6 or CXCR3 from 1 out of 6 donors analyzed. Mean values ± SEM are also shown. (D-F) Three distinct cell populations were FACS-sorted from 4 donors: after HK-S. aureus stimulation OX40^-CD137^- (AIM^-) and AIM⁺ CD4⁺ T cells, and after α-CD3/CD28 stimulation AIM⁺ CD4⁺ T cells. Cells were rested overnight, stimulated with PMA/Ionomycin for 4 h and stained intracellularly for IL-17A and IFN-γ. (D) Dot plots from a representative donor gating on live, CD4⁺ T cells. (E) Percentages and (F) Boolean gates of IFN-γ- and/or IL-17A-producing CD4⁺ T cells from 4 donors. Statistical analysis was performed using paired students t-tests (B, C) or a one-way ANOVA with Dunnett’s test for multiple comparisons (E). *P ≤ 0.05, **P ≤ 0.01, P ≤ ***0.001.

Figure 4

Skin-tropic S. aureus-specific CD4⁺ T cells are further enriched for IL17A and IL26 gene expression. PBMCs isolated from 6 healthy donors were sorted via negative selection into CD4⁺ and CD4^- fractions. CD4^- cells were irradiated and both fractions were re-mixed at a ratio of 1:1. Cells were then stimulated for 24 h with HK S. aureus or α-CD3/CD28 or left untreated. CD4⁺ T cells expressing the activation markers OX40 and CD137 (AIM⁺ CD4⁺ T cells) were index-sorted based on CLA expression via flow-cytometry and scRNA-Seq analysis was performed. (A) The transcription of the 14 cytokine-encoding genes analyzed in Figure 3B was quantified in S. aureus-specific CD4⁺ T cells separated based on CLA expression. Only IL17A and IL26 were transcribed at statistically significantly higher levels in CLA⁺ as compared to CLA^- cells, as assessed by paired Student’s t-test. *P ≤ 0.05. To confirm transcriptional results at the protein level, AIM⁺ CD4⁺ T cells after stimulation with HK S. aureus were FACS-sorted, rested overnight and then re-stimulated with PMA and Ionomycin for 3 h and stained on surface for CLA and intracellularly for IL-17A and IFN-γ. (B) Dot plots from a representative donor are shown. (C) Percentages of IL-17A⁺ or IFN-g⁺ cells in CLA⁺ and CLA^- cells of 4 donors are shown. Statistical analysis was performed using paired Student’s t-test. *P ≤ 0.05.

Figure 5

S. aureus-specific Treg present in the blood of healthy subjects express the co-inhibitory receptor TIGIT. (A) Selecting cells from the Treg cluster (in orange in the UMAP), the differentially expressed genes between S. aureus-specific (in red) and polyclonally-activated (in blue) Treg are reported. (B) Violin plot showing on a cell per cell basis the expression of TIGIT in S. aureus-specific and polyclonally-activated Treg. Percentages of cells expressing TIGIT in each group are reported. (C, D) Surface expression of TIGIT on S. aureus-specific and polyclonally-activated Treg identified as CD40L^-CD127^low CD4⁺ T cells. (C) Representative dot plots gating on AIM⁺ cells after S. aureus or α-CD3/CD28 stimulation. (D) Percentages of AIM⁺ Treg expressing TIGIT in response to each stimulus for 6 donors. (E, F) Percentages of AIM⁺TIGIT⁺ Treg after stimulation with (E) HK microorganisms: S. aureus (n = 6 donors), Candida albicans (C. albicans, n = 6 donors), Klebsiella pneumoniae (K. pneumoniae, n = 4 donors), or (F) SARS-Cov2 Spike Protein, Influvac Tetra vaccine (n = 5 donors/stimulus). Polyclonal stimulation with α-CD3/CD28 was done for comparison (n = 6 donors). (G) CD4⁺ fraction mixed with irradiated CD4^- fraction from PBMCs of healthy subjects were stimulated with HK S. aureus as described for AIM assay for 3 d in presence of the antagonist anti-TIGIT mAb Tiragolumab, isotype control mAb or medium alone. Concentrations of cytokines present in cultures supernatants collected at day 3 were measured and expressed as fold induction versus concentrations found in cultures where no antibodies were added (medium, dotted line, n = 4-5 donors, 3 independent experiments). Statistical analysis was performed using a paired t-test (D), or a mixed-effects analysis followed by Dunnet’s multiple comparison test (E, F) or a 2-way ANOVA followed by Sidak’s test for multiple comparisons (G). *P ≤ 0.05, P ≤ **0.01.