Fig 1.
Treg frequencies are not significantly different across clinical categories.
A sequential gating strategy (A) was used to define Tregs as CD3+CD4+CD45RA-CD25hi (B), CD3+CD4+CD45RA-CD127loCD25hi (C) and CD3+CD4+CD45RA-CD127loCD25hiFoxP3+ (D) Frequencies of all three Treg populations were calculated as frequencies of total CD3+CD4+ T cells. Data shown is median frequency with range from multiple donors (IGRA-ve N = 15, IGRA+ve N = 14, PTB N = 37, ATT 6 months N = 9) in each clinical category. P value was determined by non-parametric One-Way ANOVA Kruskal–Wallis test.
Table 1.
Summary of subjects recruited under each clinical category.
Fig 2.
Treg mediated suppression is lower in polyclonally activated CD4+ Teff cells in PTB.
Teff and Treg populations were isolated either by magnetic beads or flow cytometry cell sorting. Data shown in A–C is from experiments carried out using Teff and Treg cells isolated by magnetic beads and that shown in D–E is from experiments carried out using Teff and Treg cells isolated by flow cytometry sorting. CFSE labelled Teff cells were co-cultured with Treg cells at a 1:0.5 or 1:1 (C) or 1:1 (E) ratio. Cells were activated with mitogenic anti-CD3/anti-CD28 beads at a bead:Teff cell ratio of 2:1 (A–C) or 1:1 (D–E). Proliferation was measured by CFSE dilution after 4 days of culture. Representative FACS plots of Teff cell CFSE dilution in presence or absence of Treg cells from different clinical categories are shown in (A). Percentage suppression was calculated as described in Materials and Methods and is compared between different categories (C, E). Proliferation of Teff cells in absence of Tregs was compared among all clinical categories used and is shown in B and D. Data shown in B and D is mean + SEM and in C and E is median frequency/range. All data shown is from multiple donors in each category: IGRA-ve N = 12 (B, C) and 8 (D, E); IGRA+ve N = 4 (B, C) and 7 (D, E); PTB N = 10 (B, C) and 17 (D, E); ATT 6 months N = 5 (D, E). P value was determined by non-parametric One-Way ANOVA Kruskal–Wallis test and Dunn’s multiple comparisons test. ***p < 0.001, **p < 0.01, *p < 0.05.
Fig 3.
Teff cells from PTB patients are resistant to suppression whilst their Treg cells preserve suppressive potential.
A diagrammatic representation of a criss-cross assay (A). A fixed number of CFSE labelled CD25- effector cells isolated by magnetic beads were cultured with different ratios of either autologous or heterologous CD25+ suppressors also isolated by magnetic beads. Cells were activated with anti-CD3/anti-CD28 activator beads at a Teff cell:bead ratio of 2:1 for a period of 4 days. After this CFSE dilution was measured by flow cytometry. Proliferation of IGRA-ve, PTB and ATT Teff cells in absence of suppressors was compared (B). Percentage suppression was calculated as described in Materials and Methods and is compared between autologous and heterologous criss-cross cultures between Teff and Treg cells from different clinical categories (C). Cells from 6 IGRA-ve, 8 PTB and 4 ATT donors were used to set up autologous or heterologous criss-cross assays. Data shown in (B) is mean + SEM and in (C) is median and range from multiple donors. P value was determined by non-parametric One-Way ANOVA Kruskal–Wallis test and Dunn’s multiple comparisons test. ***p < 0.001, **p < 0.01, *p < 0.05, ns- not significant.
Fig 4.
Frequency of HLA-DR+ Teff cells is elevated in PTB.
PBMC were stained with Avid, anti-CD3, anti-CD4, anti-CD45RA, anti-CD127, anti-CD25 and anti-HLA-DR, anti-CD38 and anti-PD-1. Stained samples were acquired on a FACS Aria Fusion after using appropriate single color compensation controls. A sequential gating strategy was employed to arrive at live CD3+CD4+CD45RA-CD127hiCD25lo Teff cells. Expression of HLA-DR, CD38 and PD-1 was studied in the Teff fraction. Representative HLA-DR staining on Teff cells from different clinical categories is shown (A). Frequency of HLA-DR+ Teff cells from different clinical categories was plotted (B). SPICE analyses of Boolean gating data of HLA-DR, CD38 and PD-1 expression on Teff cells derived from FlowJo was carried out and frequencies of Teff cells expressing different combinations of markers is shown (C). Data shown is median frequency with range from multiple donors (IGRA-ve N = 9, IGRA+ve N = 11, PTB N = 27, ATT 6 months N = 8) in each clinical category. P value for (B) was determined by non-parametric One-Way ANOVA Kruskal–Wallis test and Dunn’s multiple comparisons test. P value for (C) was determined by Mann Whitney test. ***p < 0.001, **p < 0.01, *p < 0.05.
Fig 5.
Depletion of HLA-DR+ fraction restores sensitivity to Treg suppression in HLA-DR- Teff population from PTB subjects.
A representative sequential gating strategy used for sorting pure total Teff, HLA-DR-, CD38-, PD-1- Teff and Treg cells from PTB patients (A). Proliferation of Teff cells in absence of Tregs in all cell populations studied was compared (B). CFSE labeled total Teff, HLA-DR-, CD38- and PD-1- Teff cells were co-cultured with autologous Treg cells at a ratio of 1:1 in the presence of anti-CD3/anti-CD28 activator beads at a Teff:bead ratio of 1:1. Proliferation was measured by CFSE dilution after 4 days of culture and percentage suppression was calculated (C). Suppression observed in control IGRA-ve total and HLA-DR- Teff cells was used for comparison (C). Total Teff and HLA-DR- Teff cells were collected from 15 PTB donors and CD38- and PD-1- Teff cells were collected from 3 PTB donors. Data for IGRA-ve suppression was from 5 donors. Data shown in (B) is mean + SEM. P value for (B) was determined by non-parametric One-Way ANOVA Kruskal–Wallis test and Wilcoxon matched-pairs signed rank with Bonferroni correction was performed for (C). *** p < 0.001.
Fig 6.
Transcriptome analysis reveals changes between total Teff and HLA-DR- Teff cell subsets from PTB subjects.
Total and HLA-DR- Teff cells from PTB subjects were isolated by FACS and activated with anti-CD3/anti-CD28 activator beads (Teff:bead ratio of 1:1) for 2, 24 and 96 hrs. Unstimulated cells were used as 0 hr control. RNA was isolated at each time and subjected to RNA sequencing. A list of up-regulated DEGs (cut-off of log2 FC > 2.5, P < 0.05) was obtained by comparing gene expression post activation to that at 0 hr in both total and HLA-DR- Teff cells. (A) Venn diagrams were drawn to identify unique and common genes up-regulated upon activation in the two cellular subsets of total and HLA-DR- Teff cells. (B) Pathway analysis of unique genes over-expressed at 2 hrs post activation was done using DAVID Functional Annotation Bioinformatics Microarray Analysis tool. A heatmap was constructed using FPKM values of genes in enriched pathways using R software. As FPKM values are unevenly distributed, z score was calculated per row i.e. per gene for proper visualization. The colour key was adjusted accordingly. Mean fold change in expression of selected genes from enriched pathways at 2, 24 and 96 hrs post activation in total (C) and HLA-DR- (D) Teff cells was plotted. A total of N = 5 at each time point for total and HLA-DR- Teff cells was used for obtaining all data.
Fig 7.
HLA-DR+ and HLA-DR- Teff cells in PTB differ in their capacity to secrete pro-inflammatory cytokines.
PBMC from PTB and IGRA-ve subjects were activated with either PHA or Mtb whole cell lysate. Brefeldin and monensin were added to cultures to prevent cytokine secretion. After 16 hrs of activation, cells were fixed, permeabilised and stained with an antibody cocktail comprising Avid, anti-CD3, anti-CD4, anti-CD45RA, anti-CD127, anti-CD25, anti-HLA-DR, anti-IFNγ anti-IL-2, anti-IL-17A and anti-IL-22. Expression of cytokines was measured in the Avid-CD3+CD4+CD45RA-CD127hiCD25loHLA-DR+ and Avid-CD3+CD4+CD45RA-CD127hiCD25loHLA-DR- Teff compartments. A representative FACS plot of PHA activated PBMC from a PTB donor with cytokine positive cells in the HLA-DR+ and HLA-DR- Teff fractions is shown (A). IFNγ+, IL-2+, IL-17A+ and IL-22+ cells in response to stimulation were measured as a frequency of HLA-DR+ and HLA-DR- Teff cells (B). A total of 9–11 PTB and 6 IGRA-ve donors were used. Paired Wilcoxon matched-pairs signed rank test with Bonferroni correction was used to determine P value. *p ≤ 0.013.
Fig 8.
PTB HLA-DR- but not HLA-DR+ total Teff are susceptible to Treg suppression via CCR5 and PD-L1.
(A, E) RNA was isolated from sorted PTB total and HLA-DR- Teff cells activated with anti-CD3/anti-CD28 for 2 hrs and converted to cDNA. Expression of CCL3L3, CCL4 and PD-L1 was measured by qRT-PCR. Expression is shown as relative copy number relative to GAPDH. (B) CCL3, CCL4 and CCL5 levels were measured in supernatants from cultures of PTB total and HLA-DR- cells activated with anti-CD3/anti-CD28 for 24 hrs by ELISA. CCR5 (C) and PD-L1 (F) expression was measured on unstimulated and anti-CD3/anti-CD28 activated PTB total and HLA-DR- Teff cells after 24 hrs and MFI of CCR5 (C) and PD-L1 (F) from multiple PTB donors was plotted for comparison between total and HLA-DR- Teff cells. CFSE labelled sorted PTB and IGRA-ve total and HLA-DR- Teff cells were co-cultured with autologous Treg cells at a ratio of 1:1 in the presence of either 5 μg/ml anti-CCR5 (D) or 5 μg/ml anti-PD-L1 (G). Appropriate 5 μg/ml isotype antibody was used as control in both cases and cells were activated with anti-CD3/anti-CD28 beads (beads:cell ratio of 1:1). Proliferation was measured by CFSE dilution after 4 days and percentage suppression was calculated (D and G). A total of 6–8 PTB and 6 IGRA-ve donors were used. Upper box shows data pertaining to CCR5 and lower box shows data pertaining to PD-L1. Paired Wilcoxon matched-pairs signed rank test with Bonferroni correction was used to determine P value. *p ≤ 0.03.
Fig 9.
PTB Teff containing HLA-DR+ subset are compromised in CCR5 and PD-L1 mediated signalling.
Sorted PTB total and HLA-DR- Teff cells were treated with 50 ng/ml each of CCL3 and CCL4 (A), 5 μg/ml anti-CCR5 (A), 5 μg/ml anti-PD-L1 (B) or 10 μg/ml recombinant PD-L1 (B). Cells were activated with anti-CD3/anti-CD28 beads (beads:cell ratio of 1:1). Anti-CCR5, anti-PD-L1 and recombinant PD-L1 were added 20 minutes prior to addition of mitogenic anti-CD3/anti-CD28. Unstimulated cells were used as control. NFκB activation was measured by ELISA after 180 minutes of stimulation and was expressed as absorbance at 450 nm. (A) shows data pertaining to CCR5 mediated signaling and (B) shows data pertaining to PD-L1 mediated signaling. Data shows median and maximum and minimum values from N = 6 PTB donors. Paired Wilcoxon matched-pairs signed rank test with Bonferroni correction was used to determine P value. *p ≤ 0.03.