The immune microenvironment of ovarian cancer, as influenced by glycolysis and glutaminolysis regulation

Abstract

Despite the current treatment approaches, new treatments are urgently needed to improve the clinical outcomes of ovarian cancer patients. Immune therapies have shown promising treatment effects in ovarian cancer. Clarification of the immune suppressive microenvironment in tumors would facilitate the discovery of new targets for treating ovarian cancer. Increased expression of genes responsible for glycolysis or glutaminolysis in ovarian cancer promotes proliferation, metastasis, and resistance to chemotherapies. Moreover, immune cells adjust their preference for different metabolic pathways under different statues. Thus, the metabolic inhibition in cancer might provide ideal targets for use together with immune therapies. In this study, I found that the intra-tumor infiltration of CD8+ T cells was positively correlated with advanced grade and stage in ovarian cancer, and the numbers of stromal infiltrated CD8+ T cells were higher in tissues with advanced stages and metastatic tumors. Moreover, ovarian cancer cells show increased migration and invasion ability along with enhanced expression of MMP-9, uPA, VEGF, bFGF, IL-8, IL-10, and PD-L1 after co-culture with CD8+ T cells. Moreover, FoxP3+ Tregs cell infiltration is positively correlated with histological grade. In addition, PD-L1 predicts poor clinical outcome in ovarian cancer according to the TCGA dataset. Moreover, I found that the glycolysis gate-keeping enzyme PDK1 could modulate PD-L1 expression via the c-jun-NH2-kinase (JNK)-c-Jun pathway in ovarian cancer cells. Elevated PDK1 expression has been found to be correlated with PD-L1 both in the TCGA ovarian cancer dataset and in ovarian cancer tissue arrays. Functional studies have revealed that the binding of PD1/PD-L1 suppresses the secretion of IFN-γ and increases apoptosis of CD8+ T cells after co-culture with ovarian cancer cells with overexpression of PDK1. However, knockdown of PDK1 relieves these effects on CD8+ T cells. Besides that, in vivo experiment showed the synergistic improved overall survival and increased inhibition on tumor growth of Dichloroacetate, a PDK inhibitor, with PD-L1 antibody accompanying with increased IFN-γ secretion by monocytes infiltrated into tumor islets. Additionally, an inverse correlation has been found between PDK1 expression and CD8+ T cell infiltration in ovarian cancer tissues. Additionally, I have elucidated that the glutaminase GLS predicts poor survival and correlates with the immunosuppressive microenvironment of ovarian cancer, based on the analyzing of TCGA ovarian cancer dataset. The GLS inhibitor compound 968 inhibits the proliferation of ovarian cancer cells but enhances Granzyme B secretion by CD8+ T cells. Furthermore, 968 enhances the apoptosis-inducing ability of CD8+ T cells toward cancer cells and has a synergistic effect with anti-PD-L1 in treating ovarian cancer both in vitro and in vivo. 968 also increases Granzyme B secretion by monocytes isolated from an ovarian tumor xenograft model. Additionally, 968 increases the infiltration of CD3+CD4+ T cells and CD3+CD8+ T cells into tumors, possibly through enhancing the secretion of CXCL10 and CXCL11 by tumor cells. Together, these findings shed light on the immunosuppressive microenvironment in ovarian cancer and the influence of glycolysis and glutaminolysis on the antitumor response, thus suggesting possible targets for use in conjunction with immunotherapies to improve clinical outcomes in ovarian cancer.