Hypoxia inducible factor HIF-1 promotes myeloid-derived suppressor cells accumulation through ENTPD2/CD39L1 in hepatocellular carcinoma.

Myeloid-derived suppressor cells (MDSCs) possess immunosuppressive activities, which allow cancers to escape immune surveillance and become non-responsive to immune checkpoints blockade. Here we report hypoxia as a cause of MDSC accumulation. Using hepatocellular carcinoma (HCC) as a cancer model, we show that hypoxia, through stabilization of hypoxia-inducible factor-1 (HIF-1), induces ectoenzyme, ectonucleoside triphosphate diphosphohydrolase 2 (ENTPD2/CD39L1), in cancer cells, causing its overexpression in HCC clinical specimens. Overexpression of ENTPD2 is found as a poor prognostic indicator for HCC. Mechanistically, we demonstrate that ENTPD2 converts extracellular ATP to 5'-AMP, which prevents the differentiation of MDSCs and therefore promotes the maintenance of MDSCs. We further find that ENTPD2 inhibition is able to mitigate cancer growth and enhance the efficiency and efficacy of immune checkpoint inhibitors. Our data suggest that ENTPD2 may be a good prognostic marker and therapeutic target for cancer patients, especially those receiving immune therapy.Myeloid-derived suppressor cells (MDSCs) promote tumor immune escape. Here, the authors show that in hepatocellular carcinoma, hypoxia induces the expression of ENTPD2 on cancer cells leading to elevated extracellular 5'-AMP, which in turn promote the maintenance of MDSCs by preventing their differentiation.

Hypoxia induces myeloid-derived suppressor cell recruitment to hepatocellular carcinoma through chemokine (C-C motif) ligand 26.

UNLABELLED: A population of stromal cells, myeloid-derived suppressor cells (MDSCs), is present in tumors. Though studies have gradually revealed the protumorigenic functions of MDSCs, the molecular mechanisms guiding MDSC recruitment remain largely elusive. Hypoxia, O2 deprivation, is an important factor in the tumor microenvironment of solid cancers, whose growth often exceeds the growth of functional blood vessels. Here, using hepatocellular carcinoma as the cancer model, we show that hypoxia is an important driver of MDSC recruitment. We observed that MDSCs preferentially infiltrate into hypoxic regions in human hepatocellular carcinoma tissues and that hypoxia-induced MDSC infiltration is dependent on hypoxia-inducible factors. We further found that hypoxia-inducible factors activate the transcription of chemokine (C-C motif) ligand 26 in cancer cells to recruit chemokine (C-X3-C motif) receptor 1-expressing MDSCs to the primary tumor. Knockdown of chemokine (C-C motif) ligand 26 in cancer cells profoundly reduces MDSC recruitment, angiogenesis, and tumor growth. Therapeutically, blockade of chemokine (C-C motif) ligand 26 production in cancer cells by the hypoxia-inducible factor inhibitor digoxin or blockade of chemokine (C-X3-C motif) receptor 1 in MDSCs by chemokine (C-X3-C motif) receptor 1 neutralizing antibody could substantially suppress MDSC recruitment and tumor growth. CONCLUSION: This study unprecedentedly reveals a novel molecular mechanism by which cancer cells direct MDSC homing to primary tumor and suggests that targeting MDSC recruitment represents an attractive therapeutic approach against solid cancers. (Hepatology 2016;64:797-813).