Core fucosylated (CF) glycoproteins are closely associated with the pathogenesis of pancreatic ductal adenocarcinoma (PDAC). However, how core fucosylation alterations connect to critical signaling pathways in PDAC, such as dense extracellular matrix (ECM) remodeling and KRAS–MAPK signaling, remains largely unknown. Here, we re-analyzed core fucosylation profiles of six matched PDAC tumor and adjacent normal tissues from a published dataset and identified 54 significantly up-regulated core fucosylation events, which were predominantly associated with ECM remodeling. Leveraging these events, we developed a PDAC prognostic signature comprising three CF glycoproteins, including LRP1, FN1, and LAMC2. This signature was further validated using the tumor-normal comparison in Clinical Proteomic Tumor Analysis Consortium (CPTAC) PDAC cohort. Then, unbiased screening across 1,387 pathways using the Cancer Genome Atlas Program Pancreatic Adenocarcinoma Database (TCGA-PAAD) identified cell–ECM interactions as the dominant association with the signature. Moreover, it was confirmed using tumor-normal comparison across three independent Gene Expression Omnibus (GEO) cohorts (AUC 0.888–0.941) and was associated with adverse prognosis in TCGA-PAAD. Together, these works proposed a core fucosylation-derived prognostic signature linking glycosylation alterations to matrix remodeling and poor outcome in PDAC.

Multiple Omicron subvariants of SARS-CoV-2 have emerged as dominant global concerns. Although the spike (S) proteins of Omicron subvariants harbor more than 30 mutations compared to the original wild type (WT), N-glycosylation sites within these S proteins are highly conserved. Site-specific glycosylation of S proteins from Omicron subvariants, particularly in the receptor binding domain (RBD) involved in binding to neutralizing antibodies, remain largely unexplored. Here, we purified recombinant S proteins and their corresponding RBDs from two Omicron subvariants (BA.5 and XBB.1) as well as the WT, and characterized site specific glycosylation of these proteins. Our glycoproteomic analysis revealed smaller glycans with mono-fucosylation at the site N331 in the RBD region of trimeric S proteins of Omicron subvariants relative to WT, which might reduce steric constraint for antibody binding to this region. Besides, higher levels of multi-fucosylation and sialylation at the site N331 were detected in monomeric RBDs compared to corresponding trimeric S proteins, suggesting more susceptible of RBDs to modification mediated by the glycan processing enzymes. We believe that the glycosylation profiles of Omicron subvariants will facilitate our understanding of the increased infectivity and transmissibility of Omicron subvariants, and thus assist the diagnosis, prevention, and treatment of COVID-19 infection.