Mycobacterium tuberculosis resisters despite HIV exhibit activated T cells and macrophages in their pulmonary alveoli

Abstract

BACKGROUNDNatural resistance to Mycobacterium tuberculosis (Mtb) infection in some people with HIV (PWH) is unexplained.METHODSWe performed single cell RNA-sequencing of bronchoalveolar lavage cells, unstimulated or ex vivo stimulated with Mtb, for 7 PWH who were tuberculin skin test (TST) and IFN-γ release assay (IGRA) positive (called LTBI) and 6 who were persistently TST and IGRA negative (called resisters).RESULTSAlveolar macrophages (AM) from resisters displayed a baseline M1 macrophage phenotype while AM from LTBI did not. Resisters displayed alveolar lymphocytosis, with enrichment of all T cell subpopulations including IFNG-expressing cells. In both groups, mycobactericidal granulysin was expressed almost exclusively by a T cell subtype that coexpressed granzyme B, perforin and NK cell receptors. These poly-cytotoxic T lymphocytes (poly-CTL) overexpressed activating NK cell receptors and were increased in resister BAL. Following challenge with Mtb, only intraepithelial lymphocyte-like cells from LTBI participants responded with increased transcription of IFNG. AM from resisters responded with a stronger TNF signature at 6 hours after infection while at 24 hours after infection, AM from LTBI displayed a stronger IFN-γ signature. Conversely, at 24 hours after infection, only AM from resisters displayed an upregulation of MHC class I polypeptide-related sequence A (MICA) transcripts, which encode an activating ligand for poly-CTL.CONCLUSIONThese results suggest that poly-CTL and M1-like pre-activated AM mediate the resister phenotype in PWH.FUNDINGNational Institutes of Health. Canadian Institutes of Health Research. Digital Research Alliance of Canada. French National Research Agency. French National Agency for Research on AIDS and Viral Hepatitis. St. Giles Foundation. General Atlantic Foundation. South African Medical Research Council Centre for Tuberculosis Research.

Keywords: Infectious disease; Macrophages; T cells; Tuberculosis.

Figure 1. Resisters have higher lymphocyte proportions in cells obtained by BAL compared with LTBI.

(A) Schematic representation of the study design. BAL cells were obtained from all study participants and scRNA-Seq was conducted at 6 hours and 24 hours in the presence and absence of Mtb infection. Gene-expression data were derived both for uninfected (defined as baseline) and infected BAL cells. Analysis of scRNA-Seq data was used to estimate BAL cell identities and proportions and to perform DE analysis. Created with BioRender.com. (B) UMAP of the scRNA-Seq data from the BAL cells of all subjects identified myeloid cells and lymphocytes as 2 main populations. (C) Gene expression of canonical markers for macrophages (LYZ and CD68), DCs (LAMP3), leukocytes (PTPRC [CD45]), T cells (CD3D), and B cells (MS4A1). Higher expressions are shown by darker colors in the UMAP. (D) Density of cells obtained from LTBI and resister participants. Dashed-line circles indicate the BAL lymphocytes in the 2 groups. Yellow and dark blue colors indicate the highest and lowest density of cells in the UMAP, respectively. UMAPs included samples irrespective of infection status and incubation time point. (E) Box plot of lymphocyte proportions (%) in BAL cells obtained from resister and LTBI participants. Each dot represents the average lymphocyte percentage obtained from the scRNA-Seq libraries per subject. (F) Lymphocyte proportion (%) in PBMCs for the same resister and LTBI participants.

Figure 2. Annotation of myeloid cell subpopulations in BAL.

(A) UMAP of the myeloid subset with 12 clusters and their annotations. TR, tissue resident. (B) UMAP showing the gene expression of selected canonical markers used for the annotation of the myeloid cell subpopulations. Higher expressions are denoted by darker colors. UMAPs included data from all samples and all conditions. (C) Top 5 genes with highest DE compared with the remaining myeloid cells for each cluster. Color and size correspond to the scaled expression and the percentage of cells expressing the gene by cluster, respectively. Data from noninfected samples. (D) Cluster proportions relative to the total myeloid population from resister and LTBI BAL samples. Black dots represent the outliers from the ±1.5× interquartile range. Data from 6-hour noninfected samples.

Figure 3. Gene-expression differences in the absence of Mtb between resister and LTBI myeloid cell subpopulations in the absence of Mtb.

(A) Volcano plot for differences in gene expression between resister and LTBI samples for subpopulation AM.0. Volcano plots of remaining clusters are shown in Supplemental Figure 3. Dashed lines correspond to the log₂FC thresholds of –0.2 and 0.2. Total numbers of DEG higher (red) or lower (blue) expressed in resister samples are indicated in the top corners. (B) Numbers of DEG across all myeloid cell subpopulations. Purple and light blue indicate DEG with higher or lower expression in cells from resisters. (C) Selected Hallmark pathways enriched for genes with higher expression in resister compared with LTBI cells. (D) Differential TF activity in resister and LTBI BAL samples for 6-hour noninfected cells. The heatmap shows the top 10 TF displaying the largest mean differential activity per myeloid cell subpopulation, except for clusters with less than 10 significant TF. The mean TF activity scores for cells in each cluster are shown for the LTBI and resister cells. Positive and negative scores indicate stronger or weaker/inactive TF activity, respectively. Nonsignificant (FDR > 0.01) or not tested TFs are shown in gray. (E) Gene expression of selected M1 genes IL6, CCL3, and IL1B, and the M2 gene CD163 in 6-hour noninfected cells. Color and size correspond to the scaled expression and the percentage of cells expressing the gene by cluster, respectively.

Figure 4. Annotation of lymphocyte subpopulations in BAL.

(A) UMAP of the lymphocyte subset showing 19 clusters and their annotations. Data from both groups with samples from all conditions. (B) Gene expression of selected canonical markers used for the annotation of lymphocyte clusters. Higher expression is reflected by darker colors. (C) Top 5 genes with higher expression for each cluster compared with the remaining lymphocytes. Color and size correspond to the scaled expression and the percentage of cells expressing the gene by cluster, respectively. Data from noninfected samples. (D) Lymphocyte cluster proportions relative to total alveolar lymphocytes from resister and LTBI samples. Black dots represent the outliers from the ±1.5× interquartile range. Data from 6-hour noninfected samples. Two clusters presented nominal P < 0.05 using a 2-sided Wilcoxon’s test (L.7 P = 0.008 and L.11 P = 0.045) but failed to pass multiple test correction (Bonferroni’s threshold: P < 0.0026). (E and F) Gene expression of antimycobacterial mediators (E) IFNG, (F) GNLY, GZMB, and PRF1 in 6-hour noninfected lymphocytes from LTBI and resisters. Each dot in the violin plots represents a cell. The color and size legend of the circles on the right of each violin as detailed in panel C. The asterisks indicate a significant gene expression difference between resister and LTBI clusters (Wilcoxon’s P < 0.05). The L.18 cluster showing less than 10 cells in LTBI is not plotted. In F, only the clusters presenting 25% or more of positive cells in at least one of the groups are shown and the only cluster coexpressing the 3 genes is indicated with a box (L.8).

Figure 5. AM and DC responses to ex vivo infection with Mtb in resister and LTBI cells.

(A) Volcano plots of differential gene expression in response to Mtb challenge by group and time after infection for subpopulation AM.0. Dashed lines correspond to the log₂FC thresholds of –0.2 and 0.2. Total numbers of up- and downregulated DEG are indicated in the top corners (FDR < 0.2). (B) Proportions of DEG per cluster in response to Mtb challenge by group and time point. (C) GSEA results for selected Hallmark pathways that display enrichment for genes with changed expression in response to Mtb across clusters. All significant pathways presented in this figure were enriched for upregulated genes. Nonsignificant results (FDR > 0.05) are shown in white. (D) Log₂FC of expression for cluster AM.0 genes that respond significantly differently to Mtb in resister (x axis) and LTBI (y axis) cells. The coordinate lines correspond to log₂FC = 0. For each section of the plot, the total number of DEG is presented. (E) GSEA for Hallmark pathways based on the significantly differential Mtb response between resisters and LTBI at 6 hours and 24 hours. Genes were ranked according to overresponse in resister samples. Hence, positive and negative NES correspond to enrichment of genes with higher and lower log₂FC in resister compared with LTBI cells, respectively. Hallmark pathways as in panel C.

Figure 6. Alveolar lymphocyte responses to ex vivo infection with Mtb.

(A) Kinetics of IFNG transcription at 6 hours and 24 hours in presence and absence of Mtb in LTBI and resister lymphocyte clusters. The violin plots present the density and distribution of the IFNG log₂ expressions. The size of the circles in the dot plots indicates the percentage of cells expressing IFNG. Circle colors indicate IFNG scaled expression of the IFNG-positive cells. Each dot in the violin plots represents a cell. (B) Expression of antimicrobial effector molecules GNLY, GZMB, and PRF1 and NK receptors KLRD1, KLRC1, KLRC2, and KLRK1 at 6 hours and 24 hours of in vitro culture in presence and absence of Mtb in LTBI and resister lymphocyte clusters. The dot plot breakout insert focuses on 7 key cytotoxicity genes in cluster L.8. To the right of the breakout is a box plot of estimates of the L.8 cell frequencies in BAL samples including Wilcoxon’s P value for sample difference (Supplemental Figure 5). (C) MICA transcript expression at 24 hours of in vitro culture in presence and absence of Mtb in selected AM clusters from resister and LTBI participants.

Figure 7. Proposed model for Mtb infection resistance in the sample of PWH investigated in our study.

(A) At baseline, resisters display alveolar lymphocytosis resulting in increased constitutive levels of IFN-γ which pushes resister AM toward an M1-like state. (B) The AM preactivated state leads to increased TNF signalling, cellular stress, and upregulation of MICA after infection with Mtb. MICA is the ligand recognized by the activating NKG2D receptor expressed by poly-CTLs. (C) Poly-CTL coexpressing granzyme B, granulysin, and perforin are present in resister alveoli at more than 26 times higher numbers, leading to improved killing of infected AM and intracellular Mtb in alveoli of resisters. Created with BioRender.com. Poly-CTL, poly-cytotoxic (GNLY/GZMB/PRF1⁺) T lymphocytes (CD8⁺ T and γδ T).