Abstract

Background

ulcerative colitis (UC) is a chronic inflammatory disorder of the gastrointestinal tract. The great heterogeneity that characterises this disease leads to the urgent need of a molecular biomarker able to stratify UC patients according to their disease severity and therapeutic needs. We have previously shown that UC patients exhibit a deficiency of a specific sugar chain (branched N-glycans) on intestinal T cells due to a significant reduction of the transcription of MGAT5 glycogene. This alteration was associated with mucosal T-cell hyperactivation and disease severity. Herein, we have evaluated whether MGAT5 genetic variants (SNPs) that were previously associated with increased susceptibility to other autoimmune diseases, can be associated with UC severity and response to therapy.

Methods

Three SNPs of the intronic region of MGAT5 were genotyped in 929 UC patients from two European cohorts, Portugal and Belgium, using TaqMan SNP genotyping assays. Single SNP and haplotype associations were determined for UC clinical characteristics with the “SNPasssoc” and “haplo.stats” packages in R, and their functional effect was evaluated measuring the impact of MGAT5 genetic variants on the MGAT5 mRNA levels in peripheral blood T cells from UC patients.

Results

All 3 SNPs and a specific haplotype (GGT) were associated with non-response to immunosuppressors (SNP1–A carrier (OR = 2.21; p = 0.032); SNP2–G carrier (OR = 3.11; p = 0.004); SNP3–T carrier (OR = 2.29; p = 0.030); GGT (OR = 2.83; p = 0.007)) and with bad disease course (patients with more than 5 years of diagnosis that needed step-up to biologics; SNP2–GG genotype (OR = 3.55; p = 0.003); SNP3–TT genotype (OR = 3.03; p = 0.004); GGT (OR = 2.24; p = 0.007)). The same GGT haplotype but in Belgian cohort was further associated with severe disease course (need of biologics in the first 2 years after diagnosis—OR = 2.11; p = 0.037; SNP2–G carrier (OR = 2.21; p = 0.027) and SNP3–T carrier (OR = 2.51; p = 0.012)), as well as with disease extension (OR = 7.75; p = 0.007; G (SNP2) and T (SNP3) carriers—OR = 2.17; p = 0.041 and OR = 2.81; p = 0.007, respectively). In addition, the haplotype ACA appeared to be associated with family history (OR = 2.39; p = 0.021; SNP1 AA genotype—OR=3.59;p = 0.021) in Belgian cohort. Interestingly, HIGH-risk MGAT5 genotypes were associated with low levels of MGAT5 gene expression on blood T cells.

Conclusions

Altogether, we show that genetic variants in the MGAT5 glycogene may be used to stratify HIGH vs. LOW risk UC patients serving as a potential biomarker that may help the process of therapy-decision-making.

This content is only available as a PDF.
You do not currently have access to this article.