Tissue-Specific Expression of TIGIT, PD-1, TIM-3, and CD39 by γδ T Cells in Ovarian Cancer

Abstract

Phenotypic characterization of γδ T cells in the MALs (malignant ascites lymphocytes), TILs (tumor infiltrating lymphocytes), and PBLs (peripheral blood lymphocytes) of ovarian cancer (OvCA) patients is lacking. Therefore, we quantified γδ T cell prevalence in MAL, TIL, and PBL specimens from n = 18 OvCA patients and PBL from age-matched healthy donors (HD, n = 14). Multicolor flow cytometry was performed to evaluate the expression of inhibitory receptors (TIGIT, PD-1 and TIM-3), stimulatory receptors (Ox40), and purinergic ectoenzymes (CD39 and CD73) on γδ T cell subsets. We identified an abundant infiltration of Vδ1 T cells in the MALs and TILs. These cells varied in their differentiation: The majority of Vδ1 TILs displayed an effector memory (EM) phenotype, whereas Vδ1 MALs had a more mature phenotype of terminally differentiated effector memory cells (TEMRA) with high CD45RA expression. TIGIT and TIM-3 were abundantly expressed in both MALs and PBLs, whereas Vδ1 TILs exhibited the highest levels of PD-1, CD39, and Ox40. We also observed specific clusters on mature differentiation stages for the analyzed molecules. Regarding co-expression, Vδ1 TILs showed the highest levels of cells co-expressing TIGIT with PD-1 or CD39 compared to MALs and PBLs. In conclusion, the Vδ1 T cell population showed a high prevalence in the MALs and primary tumors of OvCA patients. Due to their (co-)expression of targetable immune receptors, in particular TIGIT with PD-1 and CD39 in TILs, Vδ1 T cell-based approaches combined with the inhibition of these targets might represent a promising strategy for OvCA.

Keywords: CD39; PD-1; TIGIT; TIL; ascites; co-expression; differentiation; ovarian cancer; γδ T cells.

Figure 1

Enriched infiltration of Vδ1 T cells in malignant ascites- (MAL) and tumor-infiltrating lymphocytes (TIL) of ovarian cancer (OvCA) patients. Flow cytometric analysis regarding the co-expression of γδ TCR and the Vδ1 and Vδ2 receptor on CD3⁺ T cells was performed for peripheral blood- (PBL, triangles, n = 17), malignant ascites- (MAL, squares, n = 18), and tumor-infiltrating lymphocyte (TIL, diamonds, n = 9) samples from patients with high-grade serous ovarian cancer (OvCA), and PBL from healthy donors (HD, open circles, n = 14). (A) Representative flow cytometry data show the gating of γδ T cells (upper row); Vδ1 and Vδ2 T cells (lower row) for HD PBL (left); and OvCA PBL (second to left), MAL (second to right), and TIL (right) within CD3⁺ T cells. (B) Summary data depict the frequency of γδ T cells in HD PBL and OvCA PBL, MAL, and TIL. p-values were obtained by the Mann–Whitney test. * p < 0.05, ** p < 0.01. (C) T-distributed stochastic neighbor embedding (tSNE) analysis illustrates the distribution of Vδ1 and Vδ2 T cells within all γδ T cells in PBL from n = 4 HDs (left), and PBL (second to left), MAL (second to right), and TIL (right) from n = 4 OvCA patients, respectively. (D) Summary data show the frequency of Vδ1 (red) and Vδ2 (black) T cell subpopulations. Note that the number of samples for MAL and TIL was decreased to n = 16 and n = 8, respectively, as specimens with near to no Vδ1⁺ and Vδ2⁺ cells were excluded from this analysis. p-values were obtained by the Mann–Whitney test and Wilcoxon matched-pairs signed-rank test. * p < 0.05, *** p < 0.001.

Figure 2

OvCA-derived Vδ1 T cells show a shift toward increased TEMRA and EM differentiation. γδ T cell differentiation was analyzed in Vδ1 (red) and Vδ2 (black) T cells via expression of CD45RA and CD27 for HD PBL (open circles, n = 14), and OvCA PBL (triangles, n = 17), MAL (squares, n = 16), and TIL (diamonds, n = 8). CD27⁻CD45RA⁻: effector memory (EM); CD27⁻CD45RA^high: terminally differentiated (TEMRA) with high CD45RA expression. (A) Correlative analysis of the infiltration of Vδ1 T cells and the differentiation into CD27⁻CD45RA^high or CD27⁻CD45RA⁻ phenotypes, respectively, was performed for PBL (upper panel), MAL (middle panel), and TIL (lower panel) of OvCA patients. Pearson’s test was used to test for correlations. (B) Summary data depict the differentiation of Vδ1 vs. Vδ2 T cells into TEMRA CD27⁻CD45RA^high cells in HD PBL vs. OvCA PBL (left panel) vs. OvCA PBL vs. MAL vs. TIL (right panel). p-values were obtained by the Mann–Whitney test and Wilcoxon matched-pairs signed-rank test. ** p < 0.01, *** p < 0.001. (C) Summary data show the differentiation of Vδ1 vs. Vδ2 T cells into EM CD27⁻CD45RA⁻ cells in OvCA PBL vs. MAL vs. TIL. p-values were obtained by the Wilcoxon matched-pairs signed-rank test. * p < 0.05, ** p < 0.01, **** p < 0.0001. (D) Representative flow cytometry data show the differentiation of Vδ1 (upper panels) and Vδ2 (lower panels) T cells in HD PBL (left column), as well as PBL (second to left column), MAL (second to right column), and TIL (right column) from OvCA patients. The red squares indicate the gating of CD27⁻CD45RA^high cells.

Figure 3

TIGIT, PD-1, CD39, and Ox40 emerge as co-regulatory receptors of interest in OvCA tissues and are preferentially expressed on Vδ1 T cells. The surface expression of TIGIT, PD-1, TIM-3, CD39, CD73, and Ox40 was compared between Vδ1 (red) and Vδ2 (black) T cells in HD PBL (open circles, n = 14), and OvCA PBL (triangles, n = 17), MAL (squares, n = 16), and TIL (diamonds, n = 8). (A) Summary data present the frequency of co-regulatory receptor (CRR)⁺ cells in Vδ1 and Vδ2 T cells. p-values were obtained by the Mann–Whitney test and Wilcoxon matched-pairs signed-rank test. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. (B) tSNE analysis illustrates the distribution of CRR⁺ T cells in the PBL from n = 4 HD (first row), and PBL (second row), MAL (third row) and TIL (fourth row) from n = 4 OvCA patients, respectively.

Figure 4

Co-regulatory molecules show specific clustering patterns on mature Vδ1 differentiation stages and are co-expressed in OvCA. The surface expression of TIGIT, PD-1, and CD39 on Vδ1 T cells was compared between differentiation stages based on CD27 and CD45RA expression in OvCA PBL (triangles, n = 17), MAL (squares, n = 16), and TIL (diamonds, n = 8). CD27⁺CD45RA⁺: naïve (NA); CD27⁺CD45RA⁻: central memory (CM); CD27⁻CD45RA⁻: effector memory (EM); CD27⁻CD45RA⁺: terminally differentiated effector memory (TEMRA); CD27⁻CD45RA^high: TEMRA subpopulation with high CD45RA expression. (A) Summary data show the frequency of CRR⁺ cells in the respective differentiation stage vs. all other stages taken together. Upper row: TIGIT⁺ cells in all TEMRA (light blue, left panel) and TEMRA^high Vδ1 T cells (dark blue, right panel) vs. other (black; for TEMRA: NA, CM, and EM; for TEMRA^high: NA, CM, EM, and TEMRA^low/int). Lower row: PD-1⁺ cells in CM (green, left panel) and CD39⁺ cells in EM Vδ1 T cells (red, right panel) vs. other (black; for CM: NA, EM, and TEMRA; for EM: NA, CM, and TEMRA). p-values were obtained by the Wilcoxon matched-pairs signed-rank test. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. (B) Representative flow cytometry data illustrate the expression of TIGIT (left column), PD-1 (middle column), and CD39 (right column) in Vδ1 T cells based on the expression of CD27 and CD45RA in OvCA PBL (first row), MAL (second row), and TIL (third row), respectively. (C) Summary data depict the expression of PD-1, TIM-3, and CD39 on TIGIT⁺ (orange) vs. TIGIT⁻ (black) cells on Vδ1 T cells. p-values were obtained by the Wilcoxon matched-pairs signed-rank test. * p < 0.05, ** p < 0.01, *** p < 0.001. (D) Representative flow cytometry data illustrate the (co-)expression of TIGIT and PD-1 (first row), TIM-3 (second row), and CD39 (third row) in Vδ1 T cells in OvCA PBL (left column), MAL (middle column), and TIL (right column), respectively.