FasL+PD‐L2+ Identifies a Novel Immunosuppressive Neutrophil Population in Human Gastric Cancer That Promotes Disease Progression

Abstract Neutrophils constitute abundant cellular components in human gastric cancer (GC) tissues, but their protumorigenic subset in pathogenesis of GC progression is unclear. Here, it is found that patients with GC show significantly higher neutrophil infiltration in tumors that is regulated by CXCL12‐CXCR4 chemotaxis. These tumor‐infiltrating neutrophils express high level immunosuppressive molecules FasL and PD‐L2, and this FasL+PD‐L2+ neutrophil subset with a unique phenotype constitutes at least 20% of all neutrophils in advanced GC and predicts poor patient survival. Tumor induces neutrophils to express FasL and PD‐L2 proteins with similar phenotype to those in GC tumors in both time‐dependent and dose‐dependent manners. Mechanistically, Th17 cell‐derived IL‐17A and tumor cell‐derived G‐CSF can significantly induce neutrophil FasL and PD‐L2 expression via activating ERK‐NF‐κB and JAK‐STAT3 signaling pathway, respectively. Importantly, upon over‐expressing FasL and PD‐L2, neutrophils acquire immunosuppressive functions on tumor‐specific CD8+ T‐cells and promote the growth and progression of human GC tumors in vitro and in vivo, which can be reversed by blocking FasL and PD‐L2 on these neutrophils. Thus, the work identifies a novel protumorigenic FasL+PD‐L2+ neutrophil subset in GC and provides new insights for human cancer immunosuppression and anti‐cancer therapies targeting these pathogenic cells.

Results were expressed as FasL + neutrophil percentage in total neutrophils or FasL + neutrophil number per million total cells and IL-17A expression in tumor tissues. Data are mean ± SEM and analyzed by Student t test, Mann-Whitney U test and 1-way ANOVA. *P<0.05, **P<0.01, n.s P>0.05 for groups connected by horizontal lines. Figure S7. Figure S7. IL-17A induces neutrophil FasL expression via activating ERK-NF-κB signaling pathway. (A) Statistical analysis of the expression of FasL on neutrophils exposed to TTCS with anti-IL-17A antibody or NTCS with IL-17A for 12 h (n=3). (B) Expression of FasL on neutrophils exposed to 50% TTCS with or without AG490 (a JAK inhibitor), SP600125 (a JNK inhibitor), FLLL32 (an STAT3 inhibitor), Wortmannin (a PI3K inhibitor), SB203580 (an MAPK inhibitor), or GSK-3β inhibitor for 12 h. (C) Expression of FasL on neutrophils exposed to TTCS or IL-17A with or without U0126 and/or BAY 11-7082 for 12 h. (D) Statistical analysis of the expression of FasL on neutrophils exposed to TTCS or IL-17A with or without U0126 and/or BAY 11-7082 for 12 h (n=3). Data are mean ± SEM and analyzed by Student t test, Mann-Whitney U test and 1-way ANOVA. *P<0.05, **P<0.01, n.s P>0.05 for groups connected by horizontal lines. Figure S8. Figure S8. Th17 cell-derived IL-17A induces neutrophil FasL expression via activating ERK-NF-κB signaling pathway. (A) Statistical analysis of the expression of FasL on neutrophils exposed to Th17 sup, non-Th17 sup or medium control for 12 h, or exposed to Th17 sup for 3, 6, 12 h, or exposed to Th17 sup (20%, 40%, or 80%) for 12 h (n=3). (B) Expression of FasL on neutrophils exposed to Th17 sup with or without U0126 and/or BAY 11-7082 for 12 h. (C) Statistical analysis of the expression of FasL on neutrophils exposed to Th17 sup with or without U0126 and/or BAY 11-7082 for 12 h (n=3). (D) Statistical analysis of the expression of FasL on neutrophils exposed to Th17 sup with anti-IL-17A antibody or non-Th17 sup with IL-17A for 12 h (n=3). (E) Representative image of CD15 + neutrophils and IL-17A + cells in tumor tissues of GC patients by immunohistochemical staining. Data are mean ± SEM and analyzed by Student t test, Mann-Whitney U test and 1-way ANOVA. *P<0.05, **P<0.01, n.s P>0.05 for groups connected by horizontal lines. Th17 sup: Th17 cell culture supernatants; non-Th17 sup: non-Th17 cell culture supernatants. Figure S9. on neutrophils exposed to G-CSF (100 ng/ml) or medium control for 12 h, or exposed to G-CSF (100 ng/ml) for 3, 6, 12 h, or exposed to G-CSF (25, 50, or 100 ng/ml) for 12 h (n=3). (C) G-CSF expression between autologous tumor and non-tumor tissues (n=51) was analyzed. (D) The correlations between PD-L2 + neutrophils and G-CSF in GC tumors were analyzed. Results were expressed as PD-L2 + neutrophil percentage in total neutrophils or PD-L2 + neutrophil number per million total cells and G-CSF expression in tumor tissues. Data are mean ± SEM and analyzed by Student t test, Mann-Whitney U test and 1-way ANOVA. *P<0.05, **P<0.01, n.s P>0.05 for groups connected by horizontal lines. Figure S10.
Figure S10. G-CSF induces neutrophil PD-L2 expression via activating JAK-STAT3 signaling pathway. (A) Statistical analysis of the expression of PD-L2 on neutrophils exposed to TTCS with anti-G-CSF antibody or NTCS with G-CSF for 12 h (n=3). (B) Expression of PD-L2 on neutrophils exposed to 50% TTCS with or without BAY 11-7082 (an IκBα inhibitor), SP600125 (a JNK inhibitor), U0126 (an ERK inhibitor), Wortmannin (a PI3K inhibitor), SB203580 (an MAPK inhibitor), or GSK-3β inhibitor for 12 h. (C) Expression of PD-L2 on neutrophils exposed to TTCS or G-CSF with or without AG490 and/or FLLL32 for 12 h. (D) Statistical analysis of the expression of PD-L2 on neutrophils exposed to TTCS or G-CSF with or without AG490 and/or FLLL32 for 12 h (n=3). Data are mean ± SEM and analyzed by Student t test, Mann-Whitney U test and 1-way ANOVA. *P<0.05, **P<0.01, n.s P>0.05 for groups connected by horizontal lines. Figure S11. Figure S11. Tumor cell-derived G-CSF induces neutrophil PD-L2 expression via activating JAK-STAT3 signaling pathway. (A) Statistical analysis of the expression of PD-L2 on neutrophils exposed to tumor cell sup, non-tumor cell sup or medium control for 12 h, or exposed to tumor cell sup for 3, 6, 12 h, or exposed to tumor cell sup (20%, 40%, or 80%) for 12 h (n=3). (B) Expression of PD-L2 on neutrophils exposed to tumor cell sup with or without AG490 and/or FLLL32 for 12 h.
(C) Statistical analysis of the expression of PD-L2 on neutrophils exposed to tumor cell sup with or without AG490 and/or FLLL32 for 12 h (n=3). (D) Statistical analysis of the expression of PD-L2 on neutrophils exposed to tumor cell sup with anti-G-CSF antibody or non-tumor cell sup with G-CSF for 12 h (n=3). (E) Representative data and statistical analysis of the expression of FasL on neutrophils exposed to TTCS with anti-G-CSF antibody for 12 h (n=3). (F) Representative data and statistical analysis of the expression of PD-L2 on neutrophils exposed to TTCS with anti-IL-17A antibody for 12 h (n=3). Data are mean ± SEM and analyzed by Student t test, Mann-Whitney U test and 1-way ANOVA. *P<0.05, **P<0.01, n.s P>0.05 for groups connected by horizontal lines.
Tumor cell sup: tumor cell culture supernatants; Non-tumor cell sup: non-tumor cell culture supernatants. Figure S12.    Cox proportional hazards regression model. Variables used in multivariate analysis were adopted by univariate analysis.
aNeutrophil number was acquired by immunohistochemical staining and counting and was expressed as number per field. CEA, carcinoembryonic antigen; H.pylori Ab, Helicobacter pylori antibody; HR, hazard ratio; CI, confidence interval; NA, not adopted. Cox proportional hazards regression model. Variables used in multivariate analysis were adopted by univariate analysis.
aNeutrophil number was acquired by immunohistochemical staining and counting and was expressed as number per field. CEA, carcinoembryonic antigen; H.pylori Ab, Helicobacter pylori antibody; HR, hazard ratio; CI, confidence interval; NA, not adopted.