[en] Objectives: This study aimed to develop a targeted LC–MS/MS method for the quantitation of a broad panel of plasma-derived fragments originating from joint-related collagens (types I, II, and III). The optimised method will subsequently be used to stratify osteoarthritis (OA) patients according to molecular endotypes using a clustering-based statistical workflow, with the long-term goal of supporting personalised medicine approaches in OA.
Materials and methods: In silico multi-enzymatic digestions of collagens type I–III were performed, and MRM transitions were designed using Skyline. The transition list was implemented on a triple quadrupole mass spectrometer (TQ6500, Sciex) coupled to a UPLC system (Shimadzu). To validate the transitions, bone, cartilage, and synovial membrane tissues were digested using trypsin, chymotrypsin, and Glu-C, followed by purification and size exclusion chromatography fractionation. Samples were analysed by LC–MS/MS to confirm the detectability of collagen-derived peptides.
Results: Bone tissue analysis enabled the detection of 132 peptides from the α1 and α2 chains of type I collagen, corresponding to nearly 50% sequence coverage. Cartilage tissue analysis yielded 74 peptides from type II collagen, with a coverage of 48%, while synovial membrane analysis resulted in the detection of 55 peptides from type III collagen, corresponding to a coverage of 40%. Regions with lower coverage mainly corresponded to telopeptides, likely due to cross-linking and post-translational modifications.
Conclusion: The developed LC–MS/MS method enabled the detection of 261 peptides derived from joint-related collagens. From an initial set of 14,350 theoretical transitions, 950, 293, and 232 transitions were retained for collagens type I, II, and III, respectively. This validated transition panel provides a robust analytical foundation for adaptation to plasma samples and for the molecular stratification of OA patients, supporting future personalised medicine strategies.
