Background Osteoarthritis (OA) is characterized by the progressive loss of cartilage structural extracellular matrix (ECM) components, mainly collagenous and non-collagenous proteins. The release of these proteins from the tissue can vary according to the stage of the disease. Characterization of molecular differences between cartilage subtypes will provide a background for a better understanding of OA onset and progression.
Objectives The aim of this study was to perform a quantitative proteomics approach to identify and quantify those proteins released from normal (N) and OA human articular cartilages.
Methods Tissue explants were obtained from the dissection of 4 N and 4 OA cartilages, both from femoral heads and tibial condyles. Among the OA samples, we differentiated the wounded zones (WZ) from those corresponding to the area adjacent to the lesion, or unwounded zones (UZ). The study was approved by the local ethical committee. Cartilage shavings from each donor were cut into 6 mm discs and five discs/donor were placed into 96-well plates and incubated during 6 days (37°C/5% CO2). The conditioned media from each condition were collected and their proteins were digested with trypsin. Each peptide mixture was labelled with different isobaric tags using the iTRAQ reagents (ABSciex). Then, labelled peptides of the different conditions were mixed, desalted and separated by liquid chromatography (LC). The resulting fractions were grouped and resolved by reversed-phase nano-LC coupled to mass spectrometry (MS). The identification and relative quantification of the proteins was carried out with Protein Pilot 3.0 software.
Results Globally we were able to identify 402 proteins released from the cartilage explants. Measurement of the different iTRAQ tags intensities allowed the relative quantification of almost all of them (n=390). After statistical analysis we found 104 secreted proteins showing differences in abundance (0.8≤ ratio ≥1.2, p ≤0.05) between the different OA zones (WZ and UZ) and N samples. We classified them into 3 sets of proteins: a first group of 11 proteins modulated specifically in UZ sample; a second group of 38 proteins altered only in WZ samples, and finally a third group of 55 proteins modified in both OA zones. Proteins increased in UZ samples are classified as early OA biomarkers, while proteins increased only in WZ samples are defined as late OA biomarkers. Proteins showing a gradual increase among OA samples are indicated as progression biomarkers. Although some of these proteins had a previously reported putative biomarker value for OA, most of them are novel candidates of the disease onset (the first group) and progression (the second and third groups). Interestingly, when we compared these results with those obtained from cartilage tissue extracts we found that most of the proteins increased in OA cartilage secretomes (like HPLN1, TENA, and BIGH3) are decreased in cartilage proteomes and viceversa. Furthermore, some of these modulations confirmed the results obtained in OA serum in a previous work by our group (e.g. APOE, C3, and HPT).
Conclusions We describe a novel panel of cartilage secreted proteins with potential biomarker value. This panel is now being explored in biological fluids (synovial fluid and serum) for the development of early diagnosis and/or anti-OA therapy monitoring strategies.
Disclosure of Interest None declared