Targeting tumor-intrinsic hexosamine biosynthesis sensitizes pancreatic cancer to anti-PD1 therapy.

Pancreatic ductal adenocarcinoma (PDAC) is considered to be a highly immunosuppressive and heterogenous neoplasm. Despite improved knowledge regarding the genetic background of the tumor and better understanding of the tumor microenvironment, immune checkpoint inhibitor therapy (targeting CTLA4, PD1, PDL1) has not been very successful against PDAC. The robust desmoplastic stroma, along with an extensive extracellular matrix (ECM) that is rich in hyaluronan, plays an integral role in this immune evasion. Hexosamine biosynthesis pathway (HBP), a shunt pathway of glycolysis, is a metabolic node in cancer cells that can promote survival pathways on the one hand and influence the hyaluronan synthesis in the ECM on the other. The rate-limiting enzyme of the pathway, glutamine-fructose amidotransferase 1 (GFAT1), uses glutamine and fructose 6-phosphate to eventually synthesize uridine diphosphate N-acetylglucosamine (UDP-GlcNAc). In the current manuscript, we targeted this glutamine-utilizing enzyme by a small molecule glutamine analog (6-diazo-5-oxo-l-norleucine [DON]). Our results showed that DON decreased the self-renewal potential and metastatic ability of tumor cells. Further, treatment with DON decreased hyaluronan and collagen in the tumor microenvironment, leading to an extensive remodeling of the ECM and an increased infiltration of CD8+ T cells. Additionally, treatment with DON sensitized pancreatic tumors to anti-PD1 therapy, resulting in tumor regression and prolonged survival.

p53 Dimers associate with a head-to-tail response element to repress cyclin B transcription.

DNA damage induced by the topoisomerase I inhibitor SN38 activates cell cycle checkpoints which promote cell cycle arrest. This arrest can be abrogated in p53-defective cells by the Chk1 inhibitor 7-hydroxystaurosporine (UCN-01). Previously, we compared p53 wild-type MCF10A cells with derivatives whose p53 function was inhibited by over-expression of the tetramerization domain (MCF10A/OD) or expression of shRNA against p53 (MCF10A/Δp53). Treatment of SN38-arrested MCF10A/OD cells with UCN-01 abrogated S, but not G2 arrest, while the MCF10A/Δp53 cells abrogated both S and G2 arrest. The MCF10A/OD cells had reduced levels of cyclin B, suggesting that tetramerization of p53 is not required for repression of cyclin B gene expression. In the present study, we analyzed p53 oligomerization status using glutaraldehyde cross-linking. Following SN38 treatment, MCF10A cells contained oligomeric forms of p53 with molecular weights approximating monomers, dimers, trimers, and tetramers. However, MCF10A/OD cells possessed only monomers and dimers suggesting that these complexes may be involved in repression of cyclin B. While genes transcriptionally activated by p53 contain a consensus sequence with elements repeated in a head-to-head orientation, the cyclin B promoter contains similar elements oriented head-to-tail. Chromatin immunoprecipitation (ChIP) assays revealed that p53 associates with this head-to-tail element in both MCF10A and MCF10A/OD. Electrophoretic mobility shift assays (EMSA) using a biotin-labeled probe containing the head-to-tail element showed a shift in mobility consistent with the molecular weight of tetramers and dimers in MCF10A nuclear extract, but only the dimer in MCF10A/OD nuclear extract. Taken together, these results suggest a novel mechanism whereby p53 dimers associate with the head-to-tail element to repress cyclin B transcription.