HLA-E/ Mtb specific CD4+ and CD8+ T cells have a memory phenotype in individuals with TB infection

Abstract

Introduction: Tuberculosis (TB) is the deadliest infectious disease worldwide and novel vaccines are urgently needed. HLA-E is a virtually monomorphic antigen presentation molecule and is not downregulated upon HIV co-infection. HLA-E restricted Mtb specific CD8⁺ T cells are present in the circulation of individuals with active TB (aTB) and Mtb infection (TBI) with or without HIV co-infection, making HLA-E restricted T cells interesting vaccination targets for TB.

Methods: Here, we performed in-depth phenotyping of HLA-E/Mtb specific and total T cell populations in individuals with TBI and in individuals with aTB or TBI and HIV using HLA-E/Mtb tetramers.

Results and discussion: We show that HIV co-infection is the main driver in changing the memory distribution of HLA-E/Mtb specific CD4⁺ and CD8⁺ T cell subsets. HLA-E/Mtb specific CD4⁺ and CD8⁺ T cells were found to circulate with comparable frequencies in all individuals and displayed expression of KLRG1, PD-1 and 2B4 similar to that of total T cells. The presence of HLA-E/Mtb specific T cells in individuals with aTB and TBI highlights the potential of HLA-E as a vaccine target for TB.

Keywords: HLA-E; T cells; Tuberculosis; immunophenotyping; spectral flow cytometry.

Figure 1

Individuals with TBI and HIV (n=13), aTB and HIV (n=11) or TBI (n=15) have differently distributed immune profiles. (A) UMAP visualization of 25 immune subsets identified using FlowSOM clustering on the total live PBMC population in all individuals combined. (B) UMAP density plots in individuals with TBI (top), TBI and HIV (middle) and aTB and HIV (bottom). (C) Volcano plot showing subsets that are significantly different between individuals with TBI and individuals with TBI and HIV in green and non-significant subsets in black. Significance was determined with EdgeR (p<0.01). (D) Bar graphs showing the frequency of live PBMC (Y-axis) in the subsets that are significantly different (X-axis) between individuals with TBI (blue circles) or TBI and HIV (orange circles). Bars show the median frequency and error bars the 95% confidence interval in each subset. Significance was tested using Multiple Mann-Whitney tests with multiple comparison correction (*p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001). (E) Same as (C), comparing individuals with TBI or aTB with HIV. (F) Same as (D), comparing individuals with TBI and HIV (blue circles) and aTB and HIV (orange circles).

Figure 2

HLA-E/Mtb CD3⁺, CD4⁺ and CD8⁺ T cells are present in the circulation of individuals with TBI and HIV (n=17), aTB with HIV (n=14) or TBI (n=20). (A) Density plots for HLA-E p44, Mtb pool 1 and 2 CD3⁺, CD4⁺ and CD8⁺ T cells in one representative individual with aTB and HIV. HLA-E TM (PE) is shown on the X-axis and CD3, CD4 and CD8 on the Y-axis (B) Frequency of HLA-E p44, Mtb pool 1 and Mtb pool 2 CD3⁺ (top), CD4⁺ (middle) and CD8⁺ (bottom) T cells in the circulation of all individuals combined. Bars represent the median frequency and error bars the 95% confidence interval. Significance was tested using the Friedman test with multiple comparison correction (*p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001). (C) Frequency of HLA-E CD3⁺, CD4⁺ and CD8⁺ T cells recognizing Mtb pool 1 and 2 in individuals with TBI (left) and in individuals with TBI or aTB and HIV (right). Bars represent the median frequency and error bars the 95% confidence interval. Significance was tested using the Friedman test (left) and Multiple Mann-Whitney test (right) with multiple comparison correction (****p <0.0001). (D) Correlation plots showing the frequency of HLA-E CD4⁺ T cells on the X-axis and CD8⁺ T cells on the Y-axis recognizing Mtb pool 1 and 2 in individuals with TBI (left), TBI and HIV (middle) or aTB and HIV (right). Significance was tested with Spearman’s rank correlation (****p < 0.0001).

Figure 3

HLA-E/Mtb specific immune profile mainly consists of memory CD4⁺ and CD8⁺ T cells in individuals with TBI and HIV (n=13), aTB and HIV (n=11) or TBI (n=15). (A) UMAP visualization of 10 subsets identified using FlowSOM clustering on the total live HLA-E TM⁺ population recognizing Mtb pool 1 and pool 2 in all individuals combined. (B) Stacked bar graphs showing the mean frequency of HLA-E TM⁺ cells in each subset, separated for Mtb pool 1 and 2, all individuals combined (n=39). (C) Stacked bar graphs showing the mean frequency of HLA-E TM⁺ cells in each subset, separated per cohort. (D) Bar graphs showing the frequency of HLA-E TM⁺ cells recognizing Mtb pool 1 and 2 (Y-axis) in the 10 HLA-E/Mtb specific subsets (X-axis), separated for individuals with TBI (blue circles) or TBI and HIV (orange circles). Bars show the median frequency and error bars the 95% confidence interval. Significance was tested using Multiple Mann-Whitney tests with multiple comparison correction. *p < 0.05; **p < 0.01; ****p < 0.0001. (E) Same as (D), comparing individuals with TBI and HIV (blue circles) and aTB and HIV (orange circles).

Figure 4

The exhaustion markers KLRG1, PD-1 and 2B4 are similarly expressed on HLA-E and the total T cells with differential expression between CD4⁺ and CD8⁺ T cells. (A) Density plots showing the expression of KLRG1, PD-1 and 2B4 on CD4⁺ (left) and CD8⁺ (right) T cells, comparing the total T cell population (top) and T cells recognizing HLA-E/Mtb pool 1 (bottom). Data is from one representative individual with TBI. (B) Stacked bar graphs showing the (co-)expression of KLRG1, 2B4 and PD-1 on HLA-E T cells recognizing pCMV (n=13), Mtb pool 1 (n=20) and 2 (n=19), compared to the total T cell population, in individuals with TBI. Stacks represent the mean expression frequency within CD4⁺ (left) and CD8⁺ (right) T cells. (C) Same as (B), for individuals with TBI and HIV for Mtb pool 1 and 2 (n=27) and pCMV (n=7).