Macrophages Instruct Aberrant Glycosylation in Colon Cancer by Chemokine and Cytokine Signals

Patients affected by inflammatory bowel diseases, such as ulcerative colitis and Crohn disease, have an increased risk of developing colorectal cancer. Chronic inflammation alters the recruitment and function of leukocytes in the colon. Tumor-associated myeloid cells, specifically macrophages, can exert protumor functions by supporting cancer stem cells, progression to less differentiated cells, angiogenesis, local and distant migration, and negative regulation of antitumor adaptive immunity (1). Despite abundant evidence of cross-talk between tumor-associated macrophages (TAM) and cancer cells, the role of TAMs in colorectal cancer remains unclear, because different studies associate TAM infiltration with either antitumor activity and improved disease-free survival or protumor activity and disease progression.

Colon cancer cells accumulate mutations and oncosuppressor genes. They also rewire their metabolism, via aberrant protein glycosylation, during tumorigenesis. The mechanisms by which altered glycosylation contributes to tumor progression are still poorly defined. Kvorjak and colleagues investigate the contribution of human macrophages to the glycosylation of the transmembrane glycoprotein Mucin 1 (MUC-1) under clinically relevant inflammatory conditions, that is, ulcerative colitis and colitis-associated colon cancer. Coupling genetic and IHC analyses of a limited set of genes and markers, the authors find that macrophages in inflamed areas of colon are adjacent to epithelial cells and share a phenotype consistent with both M1 and M2 polarization states (2). In vitro polarized M1 and M2 human macrophages induce distinctive sets of genes that intervene in glycosylation when cocultured with colon cell lines. Specifically, ST6GALNAC1, a glycosyltransferase that catalyzes the transfer of a sialic acid to N-acetylgalactosamine residues on mucins to form the cancer-associated sialyl-Tn (sTn) antigen, is under the control of signals generated by M2 macrophages (2). In ulcerative colitis and colitis-associated colon cancer, but not normal mucosa, ST6GALNAC1-overexpressing epithelial cells and MUC-1-sTn–expressing cancer cells are in the proximity of macrophages. Sialylated MUC-1 is detectable by immunoprecipitation and is enriched in colon cancer lines cocultured with M2 macrophages, indicating that the ST6GALNAC1 gene product is enzymatically active.

Through cytokine and chemokine profiling, a peculiar axis in ST6GALNAC1 regulation was identified. Neutralizing antibodies to either IL13 or CCL17 decrease ST6GALNAC1 mRNA in colon cancer cells, whereas the addition of exogenous IL13 and CCL17 increases MUC-1-sTn and ST6GALNAC1 protein and mRNA. Despite this convergence, IL13 and CCL17 act through distinct intracellular signaling pathways. IL13 induces phosphorylation of the STAT6 transcription factor, with phosphoSTAT6 associating with the ST6GALNAC1 promoter region on putative regulatory sequences. CCL17 activates the phosphorylation of two components in the NF-κB signaling pathways, IKB-α and p65, leading to p65 localization on the ST6GALNAC1 promoter region. Using a computational model that integrates these newly discovered intracellular signal perturbations that regulate ST6GALNAC1 during the evolution of colitis and colon cancer, the authors predict how dynamic and temporal pharmacologic intervention can influence glycosylation-dependent tumorigenesis (2).

Results of this article and the use of the computational model offer an alternative strategy to targeting TAMs, aside from the conventional use of antibodies or small molecules targeting CSF-1R (3), focusing on reprogramming pathways that fuel the tumor-promoting and immunosuppressive functions of M2 cells, including IL4 and IL13 cytokines. Despite the inherent limitation of analyzing a single-gene pathway, algorithms predicting signaling factors that regulate aberrant glycosylation might help to inform future treatment combinations and when to consider immunotherapy targeting altered MUC-1, via vaccination, therapies reliant upon MUC-1–specific antibodies, or the use of chimeric antigen receptor T cells (4). Applied to multiple genes and biochemical pathways, computational systems may contribute to the next wave of rational and personalized immunotherapy combinations.

V. Bronte is a consultant/advisory board member for Xios Therapeutics and Codiak BioSciences and has ownership interest (including patents) in Xios Therapeutics, Codiak BioSciences, BioNTech AG, and IO Biotech ApS. No other potential conflicts of interest were disclosed.