The choice for adjuvant chemotherapy in stage II colorectal cancer (CRC) is controversial as many patients are cured by surgery alone and it is difficult to identify patients with high-risk of recurrence of the disease. There is a need for better stratification of this group of patients. Mass spectrometry imaging could identify patients at risk. We report here the N-glycosylation signatures of the different cell populations in a group of stage II CRC tissue samples. The cancer cells, compared to normal epithelial cells, have increased levels of sialylation and high-mannose glycans, as well as decreased levels of fucosylation and highly branched N-glycans. When looking at the interface between cancer and its microenvironment, it seems that the cancer N-glycosylation signature spreads into the surrounding stroma at the invasive front of the tumor. This finding was more outspoken in patients with a worse outcome within this sample group.

The increase incidence of early colorectal cancer (T1 CRC) last years is mainly due to the introduction of population-based screening for CRC. T1 CRC staging based on histological criteria remains challenging and there is high variability among pathologists in the scoring of these criteria. It is crucial to unravel the biology behind the progression of adenoma into T1 CRC. Glycomic studies have reported extensively on alterations of the N-glycomic pattern in CRC; therefore, investigating these alterations may reveal new insights into the development of T1 CRC. We used matrix-assisted laser desorption ionization (MALDI) mass spectrometry imaging (MSI) to spatially profile the N-glycan species in a cohort of pT1 CRC using archival formalin-fixed and paraffin-embedded (FFPE) material. To generate structural information on the observed N-glycans, CE-ESI-MS/MS was used in conjunction with MALDI-MSI. Relative intensities and glycosylation traits were calculated based on a panel of 58 N-glycans. Our analysis showed pronounced differences between normal epithelium, dysplastic, and carcinoma regions. High-mannose-type N-glycans were higher in the dysplastic region than in carcinoma, which correlates to increased proliferation of the cells. We observed changes in the cancer invasive front, including higher expression of α2,3-linked sialic acids which followed the glycosylation pattern of the carcinoma region.

Objective. Changes in protein glycosylation are a hallmark of immune-mediated diseases. Glycans are master regulators of the inflammatory response and are important molecules in self-nonself discrimination. This study was undertaken to investigate whether lupus nephritis (LN) exhibits altered cellular glycosylation to identify a unique glycosignature that characterizes LN pathogenesis.Methods. A comprehensive tissue glycomics characterization was performed in kidney specimens from patients with systemic lupus erythematosus and biopsy-proven LN. A combination of advanced tissue mass spectrometry, in situ glyco-characterization, and ex vivo glycophenotyping was performed to structurally map the repertoire of N-glycans in LN tissue samples.Results. LN exhibited a unique glycan signature characterized by increased abundance and spatial distribution of unusual mannose-enriched glycans that are typically found in lower microorganisms. This glycosignature was specific for LN, as it was not observed in other kidney diseases. Exposure of mannosylated glycans in LN was shown to occur at the cell surface of kidney cells, promoting increased recognition by specific glycan-recognizing receptors expressed by immune cells. This abnormal glycosignature of LN was shown to be due to a deficient complex N-glycosylation pathway and a proficient O-mannosylation pathway. Moreover, mannosylation levels detected in kidney biopsy samples from patients with LN at the time of diagnosis were demonstrated to predict the development of chronic kidney disease (CKD) with 93% specificity.Conclusion. Cellular mannosylation is a marker of LN, predicting the development of CKD, and thus representing a potential glycobiomarker to be included in the diagnostic and prognostic algorithm of LN.