Glycoproteoforms of osteoarthritis-associated lubricin in plasma and synovial fluid.

Abstract

Lubricin/proteoglycan-4 (PRG-4) is a mucinous glycoprotein that lubricates cartilage and maintains normal tissue function and cell homeostasis. Altered O-glycoproteforms of lubricin have been found in osteoarthritis (OA) synovial fluid (SF), which could ostensibly be used to diagnose early onset OA. However, SF is invasive to obtain and generally would not be surveyed from otherwise healthy individuals. Thus, a plasma-based OA screening tool focused on lubricin glycosylation could be a less invasive method to aid in early-stage OA diagnosis. In this report, we used glycomics and glycoproteomics to characterize glycoproteoforms of OA lubricin in SF and plasma. We obtained near-complete sequence coverage of lubricin's mucin domain and its glycosylation using matched SF and plasma from OA patients (N=5). From SF lubricin we observed a spectrum of O-glycans ranging from a single GalNAcα1-Ser/Thr monosaccharide up to branched pentasaccharides. In contrast, plasma based lubricin was predominantly decorated with sialylated Galβ1-3GalNAcα1-Ser/Thr (Sialyl T). To explain the glycosylation differences observed between SF and plasma lubricin, we present splice variant-specific peptides found within the non-glycosylated region, revealing that that the longest spliceoform of lubricin was present exclusively in SF, while additional shorter splice variants could only be detected in plasma. Based on our glycoproteomic data, we developed and validated a lectin assay for lubricin, and applied this on a larger cohort of matched SF/plasma (N=19) to confirm the glycosylation differences between SF and plasma proteoforms. Next, we leveraged our assay to screen over 100 OA patient samples (OA patients N=108/controls N=38) to probe plasma lubricin as an OA biomarker. Here, we detected a decrease in α2,6 linked sialic acid in OA patients and further show that the extent of α2,6 and α2,3 sialylation on plasma-associated lubricin correlated with patient characteristics, especially Body Mass Index (BMI).