T-cell branched glycosylation as a mediator of colitis-associated colorectal cancer progression: a potential new risk biomarker in inflammatory bowel disease

Abstract

Background and Aims: Inflammatory bowel disease (IBD) is a chronic inflammatory disorder of the gastrointestinal tract, established as a risk factor for colorectal cancer (CRC) development. Long-standing inflammation appears to play a central role in colitis-associated colorectal cancer (CAC). However, the molecular mechanism underlying CAC progression is still elusive. Previous evidence showed that levels of branched glycosylation regulate T-cell-mediated immune response associated with IBD severity. Here, we revealed that colonic T cells from IBD patients are dynamically regulated by branched N-glycosylation and associated with the risk of CAC development. Methods: We performed in silico analysis for glycome and immune profile of a publicly available human dataset of CAC patients. Additionally, in a well-characterized cohort of CAC patients, we evaluated the N-glycosylation profile of infiltrated colonic immune cells at different stages of carcinogenesis (colitis, dysplasia and cancer).In vivo studies were conducted in Mgat5 KO mice, using AOM/DSS model to induce CAC. Tumor development and colonic T cells glycoprofile were characterized during CAC development. Results: The combined analysis of human IBD and CAC clinical samples, together with glycoengineered mouse model susceptible to CAC, revealed a gradual and dynamic increase of branched N-glycans in T cells from colitis to dysplasia and cancer. This glycosylation switch was shown to impose inhibitory properties in T cells, precluding an effective antitumor immune response. Mechanistically, we demonstrated that the deletion of branched N-glycans in Mgat5 knockout mice led to CAC suppression due to increased infiltration of CD8+and γδ T cells, contributing to an effective antitumor immune response. From the clinical standpoint, we demonstrated that branched N-glycosylation levels detected in inflamed lesions from IBD patients predicted CAC progression with a sensitivity of 83.3% and specificity of 67.9% when assessed together with age at diagnosis. Conclusions: Overall, we here disclosed a new mechanism underlying CAC development, identifying a potential clinical biomarker plausible to improve the efficacy of cancer surveillance programs through the early identification of high-risk IBD patients, for preventive clinical and therapeutic strategies.