Multiple sclerosis (MS) is an inflammatory autoimmune disorder affecting the central nervous system (CNS), with unresolved aetiology. Previous studies have implicated N-glycosylation, a highly regulated enzymatic attachment of complex sugars to targeted proteins, in MS pathogenesis. We investigated individual variation in N-glycosylation of the total plasma proteome and of IgG in MS. Both plasma protein and IgG N-glycans were chromatographically profiled and quantified in 83 MS cases and 88 age- and sex-matched controls. Comparing levels of glycosylation features between MS cases and controls revealed that core fucosylation (p = 6.96 × 10⁻³) and abundance of high-mannose structures (p = 1.48 × 10⁻²) were the most prominently altered IgG glycosylation traits. Significant changes in plasma protein N-glycome composition were observed for antennary fucosylated, tri- and tetrasialylated, tri- and tetragalactosylated, high-branched N-glycans (p-value range 1.66 × 10⁻²–4.28 × 10⁻²). Classification performance of N-glycans was examined by ROC curve analysis, resulting in an AUC of 0.852 for the total plasma N-glycome and 0.798 for IgG N-glycome prediction models. Our results indicate that multiple aspects of protein glycosylation are altered in MS, showing increased proinflammatory potential. N-glycan alterations showed substantial value in classification of the disease status, nonetheless, additional studies are warranted to explore their exact role in MS development and utility as biomarkers.

Human protein glycosylation is a complex process, and its in vivo regulation is poorly understood. Changes in glycosylation patterns are associated with many human diseases and conditions. Understanding the biological determinants of protein glycome provides a basis for future diagnostic and therapeutic applications. Genome-wide association studies (GWAS) allow to study biology via a hypothesis-free search of loci and genetic variants associated with a trait of interest. Sixteen loci were identified by three previous GWAS of human plasma proteome N-glycosylation. However, the possibility that some of these loci are false positives needs to be eliminated by replication studies, which have been limited so far. Here, we use the largest set of samples so far (4802 individuals) to replicate the previously identified loci. For all but one locus, the expected replication power exceeded 95%. Of the 16 loci reported previously, 15 were replicated in our study. For the remaining locus (near the KREMEN1 gene), the replication power was low, and hence, replication results were inconclusive. The very high replication rate highlights the general robustness of the GWAS findings as well as the high standards adopted by the community that studies genetic regulation of protein glycosylation. The 15 replicated loci present a good target for further functional studies. Among these, eight loci contain genes encoding glycosyltransferases: MGAT5, B3GAT1, FUT8, FUT6, ST6GAL1, B4GALT1, ST3GAL4 and MGAT3. The remaining seven loci offer starting points for further functional follow-up investigation into molecules and mechanisms that regulate human protein N-glycosylation in vivo.

Aims/hypothesis: We previously demonstrated that N-glycosylation of plasma proteins and IgGs is different in children with recent-onset type 1 diabetes compared with their healthy siblings. To search for genetic variants contributing to these changes, we undertook a genetic association study of the plasma protein and IgG N-glycome in type 1 diabetes. Methods: A total of 1105 recent-onset type 1 diabetes patients from the Danish Registry of Childhood and Adolescent Diabetes were genotyped at 183,546 genetic markers, testing these for genetic association with variable levels of 24 IgG and 39 plasma protein N-glycan traits. In the follow-up study, significant associations were validated in 455 samples. Results: This study confirmed previously known plasma protein and/or IgG N-glycosylation loci (candidate genes MGAT3, MGAT5 and ST6GAL1, encoding beta-1,4-mannosyl-glycoprotein 4-beta-N-acetylglucosaminyltransferase, alpha-1,6-mannosylglycoprotein 6-beta-N-acetylglucosaminyltransferase and ST6 beta-galactoside alpha-2,6-sialyltransferase 1 gene, respectively) and identified novel associations that were not previously reported for the general European population. First, novel genetic associations of IgG-bound glycans were found with SNPs on chromosome 22 residing in two genomic intervals close to candidate gene MGAT3; these include core fucosylated digalactosylated disialylated IgG N-glycan with bisecting N-acetylglucosamine (GlcNAc) (p_discovery=7.65 × 10⁻¹², p_replication=8.33 × 10⁻⁶ for the top associated SNP rs5757680) and core fucosylated digalactosylated glycan with bisecting GlcNAc (p_discovery=2.88 × 10⁻¹⁰, p_replication=3.03 × 10⁻³ for the top associated SNP rs137702). The most significant genetic associations of IgG-bound glycans were those with MGAT3. Second, two SNPs in high linkage disequilibrium (missense rs1047286 and synonymous rs2230203) located on chromosome 19 within the protein coding region of the complement C3 gene (C3) showed association with the oligomannose plasma protein N-glycan (p_discovery=2.43 × 10⁻¹¹, p_replication=8.66 × 10⁻⁴ for the top associated SNP rs1047286). Conclusions/interpretation: This study identified novel genetic associations driving the distinct N-glycosylation of plasma proteins and IgGs identified previously at type 1 diabetes onset. Our results highlight the importance of further exploring the potential role of N-glycosylation and its influence on complement activation and type 1 diabetes susceptibility. Graphical

Glycomics measurements, like all other high-throughput technologies, are subject to technical variation due to fluctuations in the experimental conditions. The removal of this non-biological signal from the data is referred to as normalization. Contrary to other omics data types, a systematic evaluation of normalization options for glycomics data has not been published so far. In this paper, we assess the quality of different normalization strategies for glycomics data with an innovative approach. It has been shown previously that Gaussian Graphical Models (GGMs) inferred from glycomics data are able to identify enzymatic steps in the glycan synthesis pathways in a data-driven fashion. Based on this finding, here, we quantify the quality of a given normalization method according to how well a GGM inferred from the respective normalized data reconstructs known synthesis reactions in the glycosylation pathway. The method therefore exploits a biological measure of goodness. We analyzed 23 different normalization combinations applied to six large-scale glycomics cohorts across three experimental platforms: Liquid Chromatography-ElectroSpray Ionization-Mass Spectrometry (LC-ESI-MS), Ultra High Performance Liquid Chromatography with Fluorescence Detection (UHPLC-FLD), and Matrix Assisted Laser Desorption Ionization-Furier Transform Ion Cyclotron Resonance-Mass Spectrometry (MALDI-FTICR-MS). Based on our results, we recommend normalizing glycan data using the ‘Probabilistic Quotient’ method followed by log-transformation, irrespective of the measurement platform. This recommendation is further supported by an additional analysis, where we ranked normalization methods based on their statistical associations with age, a factor known to associate with glycomics measurements.