Figure 1.
Heatkilled vs. live BCG-activated expression of Treg-cell markers on CD4+ and CD8+ T cells.
A: Gating strategy: cells were gated on single cells, live lymphocytes, CD3+ and CD4+CD8− vs. CD4−CD8+. Demonstrated is the synchronized gating on the positive population of interest for CD4+CD8− and CD8+CD4− T cells; here the CD25-positive population. B: Heatkilled and live BCG activate CD25+Foxp3+ and LAG-3+CD39+ T cells. Expression of regulatory T cell markers on CD4+ and CD8+ T cells of in vitro PPD responders was analysed by flowcytometry six days after heatkilled or live BCG stimulation. For each donor gating was compared to samples not stimulated with BCG (demonstrated in Fig. S1). Data are representative of seven responders.
Figure 2.
Treg-cell marker frequency and density are increased on live BCG-activated CD8+ vs. CD4+ T cells.
A: BCG induces Treg-cell marker expression on CD4+ and CD8+ T cells; after live BCG stimulation the percentage of total CD8+ T cells expressing CD25, Foxp3, CD39, LAG-3 or CCL4 is significantly higher compared to CD4+ T cells, depicted here as frequency of CD8+ or CD4+ population. Differences in Treg marker expression between heatkilled BCG–activated CD8+ vs. CD4+ T cells were not significant, except for expression of CCL4; CCL4 expression was also significantly higher on CD8+ T cells compared to CD4+ T cells in samples not stimulated with BCG (Fig. S1) (*p < 0.05, Wilcoxon signed-ranks test). B: Mean fluorescence intensities (MFIs) of CD25 and CD39 are increased on live BCG-activated CD8+ T cells as compared to CD4+ T cells. Gating was performed as demonstrated in figure 1A. To assess differences in intrinsic intensity of expression on CD4+ and CD8+ T cells, respectively, MFIs of positive Treg marker populations in samples not stimulated with BCG were compared; this was similar on CD4+ and CD8+ T cells for MFIs of CD25, Foxp3 and CD39. Data are representative of seven in vitro PPD-responders six days after heatkilled or live BCG stimulation (*p < 0.05; Wilcoxon signed-ranks test).
Figure 3.
Co-expression of multiple Treg-cell markers enriches for CD8+, and not CD4+ T cells.
A. The percentage of total CD8+ T cells co-expressing CD39, LAG-3, CCL4, CD25 and/or Foxp3 in different combinations is significantly increased, compared to CD4+ T cells. Demonstrated is a combined analysis using Boolean gating of cells from ten donors six days after live BCG infection. Gating was performed as in figure 1A. Boxes: 25th to 75th percentiles; line at median; whiskers: minimum to maximum (*p < 0.05, **p < 0.01; Wilcoxon signed-ranks test). B. Combining Treg markers enriches for CD8+ T cells as opposed to CD4+ T cells. Boolean gating was performed on CD3+ T cells of ten donors; CD8 vs. CD4 gating is demonstrated (top) and the CD8+ proportion of these gated populations is demonstrated (bottom) for a selection of CD3+ Boolean gates. The CD8+ proportion increased significantly using a combination of Treg markers as compared to the complete CD3+ population. Boxes: minimum to maximum, line at median (Wilcoxon signed-ranks test).
Figure 4.
Suppressive activity resides predominantly in live BCG-activated CD8+ T cells.
Heatkilled BCG-activated and live BCG-activated T cell lines were expanded, and live BCG T-cell lines were enriched for CD4 or CD8 expression using magnetic beads. Suppressive capacity was tested in a co-culture assay by titrating these T cell lines onto a Th1 reporter clone that was stimulated with its cognate peptide [30], [31]. Proliferation was measured by (3H)TdR incorporation after three days. Proliferation was divided by Th1 reporter clone proliferation in the absence of Treg cells to obtain relative proliferation as described previously [30]–[32]. Dotted lines represent background proliferation of T cells, in the absence of reporter clone peptide, relative to Th1 clone proliferation. A. Suppressive activity was confined to live BCG-activated T cells, and could not be demonstrated for heat-killed BCG-specific T cells. Data are depicted as mean + SE of five different heat-killed BCG-activated T cell lines, and six live BCG-activated T cell lines. B. Suppressive activity resides predominantly in CD8+ T cells, and not in CD4+ T cells (mean + SE of CD4+ and CD8+ T cell lines of three donors, tested in at least two independent assays; Wilcoxon signed-ranks test, p < 0.001).