Article Figures & Data
Figures
- Figure 1.
Tumor-infiltrating CD4+FOXP3+ T cells are highly activated and express functional Treg markers. Comparison of expression of activation, memory, and Treg effector markers in CD4+ T-cell subsets of TILs and PBMC at the time of TUR-B. A, Percentage of FOXP3 expressing CD4+ T cells at the different locations as indicated. PBMC from UBC patients (n = 20) compared with PBMC control samples from healthy blood donors (age, 51–67 years old; white bar, n = 8) and UBC TILs (n = 13), respectively, using Mann–Whitney test. B, Histogram overlays of investigated markers in CD4+FOXP3+ T cells at the different locations as indicated. Plots show representative data from one TUR-B patient. C, Percentage of the CD4+FOXP3+ and CD4+FOXP3− T-cell subsets expressing CD69, HLA-DR, CD45RO, CD39, or CTLA-4 at the different locations. PBMC control (white bar, n = 8), PBMC UBC (light gray bar, n = 19), and TILs (dark gray bar, n = 13). Significance was determined using the Kruskal–Wallis test with Dunn multiple comparisons test. Bar graphs (A and C) show median, and error bars interquartile range. *, P < 0.05; **, P < 0.01; ***, P < 0.001.
- Figure 2.
Tumor-infiltrating CD4+FOXP3+ T cells exhibit Treg functions. Functional analysis of PBMC- versus TIL-derived CD4+FOXP3+ T cells from patients at the time of TUR-B. A, pSTAT5 phosphorylation after rhIL2 stimulation. B, pSTAT5 dose–response curves for the indicated populations. C, Intracellular flow cytometry demonstrating scarce production of IL2 and IFNγ in freshly isolated TIL-derived FOXP3+ cells of one representative patient at the time of TUR-B, dot plots gated on CD4+ cells (left). Mean cytokine expression in the investigated sites as indicated (n = 5; right). FOXP3+ and FOXP3− populations compared using paired t test. D, PBMC-derived CD4+CD25− Teff responder cells were cocultured with indicated ratios of Tregs derived from TILs or PBMCs, respectively. Responder proliferation was assessed by CFSE dilution. Flow cytometric data are representative of five (C) or two (A, B, D) independent experiments. Bar graphs indicate mean ± SD. *, P < 0.05; **, P < 0.01.
- Figure 3.
Tumor-infiltrating CD4+FOXP3+ T cells are epigenetically committed Tregs with stable FOXP3 expression. A, Results from bisulfite sequencing of FACS-sorted Treg and Teff cell populations. Pie charts show mean methylation percentages (black) at the indicated CpG sites of the FOXP3 promoter region (left) and the CNS2 region (right) in PBMCs and TILs. Data displayed represent mean methylation at each site of 3 to 5 patients for PBMCs and TILs from UBC patients. Methylation data from PBMC-derived populations from a healthy blood donor are displayed for comparison (PBMC control). B, Treg and Teff cell populations from one UBC patient were sorted according to their surface expression of CD4, CD25, and CD127. After sorting, cells were cultured for 14 days without stimulus other than 20 IU/mL IL2 every 3 days. FOXP3 expression in Treg (filled) and Teff cells (empty) from PBMCs (red) and UBC TILs (black) was analyzed ex vivo at day 0 (left), and at day 14 of culture (right).
- Figure 4.
Tumor-infiltrating Tregs regulate MMP2. A, Schematic illustration of a UBC tumor. During the TUR-B, tumor tissue samples were collected separately from the central part (CP) and invasive front (IF)/base of the tumor respectively. B, Flow cytometric data of TILs from the CP and IF reveals that CD4+ T cells from the IF of muscle-invasive bladder cancer (MIBC) have significantly lower FOXP3+ % than CD4+ T cells derived from CP or IF of non-MIBC tumors (n = 7 and 8 for superficial tumors CP and IF respectively, and correspondingly n = 8 and 4 for invasive tumors). CP were compared with IF samples using paired t test. Superficial tumor samples were compared with invasive using two-sided unpaired t test. Bars indicate mean (±SD). C, Kaplan–Meier analysis of patient survival after stratification according to Treg frequency at the IF. Low Treg frequency was defined as ≤median and high Treg frequency as >median. D, Correlation of MMP2 expression in tumor samples (both CP and IF) to FOXP3 % of CD4+ T cells in TILs (top) using linear regression analysis. Correlation of the difference in MMP2 expression between the CP and IF of the tumor and the difference in FOXP3 % of CD4+ T cells in TILs (bottom), n = 8. E, Immunohistochemistry of UBC tumor depicting MMP2 expression in brown/red and FOXP3 in blue. F–K, Treg-mediated suppression of MMP2 production in M2 macrophages (E–G) and the invasive bladder cancer cell line 5637 (I–K). Graphs illustrate MMP2 concentration in cell culture medium (F and I), dose-dependent change in MMP2 concentration in the cell culture medium (F and I) and relative decrease of MMP2 mRNA in the target population (H and K). Statistical methods used were repeated measures ANOVA (F), nonlinear regression analysis (G, H, K), unpaired two-tailed t test (I), and linear regression analysis (J). Representative data from one out of two independent experiments are shown (F–K). *, P < 0.05; **, P < 0.01; ***, P < 0.001.
Additional Files
Supplementary Data
- Tables S1, S2 and S3 - Table S1 shows patient characteristics. Table S2 summarizes the antibodies used. Table S3 shows primer sequences used.
- Supplemental figure S1-S7 - Supplemental figure 1 illustrates gating strategy appliead to FACS data. Supplemental figure 2 shows gene expression correlation to Treg frequency overlaid on KEGG bladder cancer pathway. Supplemental figure 3 shows expression of functional Treg markers in FOXP3+ and FOXP3- cells from PBMCs and TILs. Supplemental figure 4 compares the phenotype of TIL Tregs from MIBC and non-MIBC tumors. Supplemental figure 5 illustrates post-sort purity sorted populations. Supplemental figure 6 illustrates the correlation of Treg frequency at tumor sublocations to tumor stage and survival. Supplemental figure 7 shows expression level of different MMPs in UBC tumor tissue.
Volume 6, Issue 5
