Knee osteoarthritis (KOA) diagnosis is based on symptoms, assessed through questionnaires such as the WOMAC. However, the inconsistency of pain recording and the discrepancy between joint phenotype and symptoms highlight the need for objective biomarkers in KOA diagnosis. To this end, we study relationships among clinical and molecular data in a cohort of women (n=51) with Kellgren-Lawrence grade 2–3 KOA through Support Vector Machine (SVM) and a regulation network model (RNM). Clinical descriptors (i.e., pain catastrophism (CA); depression (DE); functionality (FU); joint pain (JP); rigidity (RI); sensitization (SE); synovitis (SY)) are used to classify patients. A Youden's test is performed for each classifier to determine optimal binarization thresholds for the descriptors. Thresholds are tested against patient stratification according to baseline WOMAC data from the Osteoarthritis Initiative, and the mean accuracy is 0.97. For our cohort, the data used as SVM inputs are KOA descriptors, synovial fluid (SL) proteomic measurements (n=25), and transcription factors (TF) activation obtained from RNM [2] stimulated with the SL measurements. The relative weights after classification reflect input importance. The performance of each classifier is evaluated through AUC-ROC analysis. The best classifier with clinical data is CA (AUC = 0.9), highly influenced by FU and SE, suggesting that kinesophobia is involved in pain perception. With SL input, leptin strongly influences every classifier, suggesting the importance of low-grade inflammation. When TF are used, the mean AUC is limited to 0.608, which can be related to the pleomorphic behaviour of osteoarthritic chondrocytes. Nevertheless, FU has an AUC of 0.7 with strong importance of FOXO downregulation. Though larger and longitudinal cohorts are needed, this unique combination of SVM and RNM shall help to map objectively KOA descriptors.
Osteoarthritis (OA) leads to articular cartilage degradation, following complex dysregulation of chondrocyte's metabolism towards a catabolic state. Mechanical and biochemical signals are involved and need to be considered to understand the condition. Regulatory network-based models (RNM) successfully simulated the biological activity of the chondrocyte and the transduction of mechanical signals at the molecular and cell levels. However, the knowledge gap between single-cell regulation and intercellular communication in tissue volumes hinders the interpretability of such models at larger scales. Accordingly, a novel tissue-level biochemical model is proposed. We hypothesise that it is possible to simulate interacting network effects through the transport of diluted species in a finite-element model, to grasp relevant dynamics of cell and tissue regulation in OA. Chondrocyte RNM equations were translated into a reaction term of 18 multi-species diffusion model (e.g., 3 anti-inflammatory and 8 pro-inflammatory interleukins, 3 pro-anabolic and 1 pro-catabolic growth factors, 2 nociceptive factors and 2 pro-inflammatory cytokines). Elements with RNM reaction terms represented the chondrocytes and were distributed randomly through the model, according to known cellular density in the knee cartilage, and could both react to and produce diffusive entities through the pericellular matrix, associated with reduced diffusion coefficients. The model was constructed over a 2D square of 0.47 mm sides considered to be in the middle of the cartilage, so boundary conditions were settled as periodic. Different simulations were initialised with initial concentrations of either healthy or pro-OA mediators. Preliminary results showed that, independently of the initial conditions, the chondrocytes successfully evolved into anabolic states, in absence of sustained pro-catabolic external stimulations, in contrast to single-cell RNM [2]. Our intercellular model suggests that paracrine communication may increase robustness towards cartilage maintenance, and future tests shall reveal new OA dynamics.
Intervertebral disc (IVD) degeneration (IDD) involves imbalance between the anabolic and the catabolic processes that regulate the extracellular matrix of its tissues. These processes are complex, and improved integration of knowledge is needed. Accordingly, we present a nucleus pulposus cell (NPC) regulatory network model (RNM) that integrates critical biochemical interactions in IVD regulation and can replicate experimental results. The RNM was built from a curated corpus of 130 specialized journal articles. Proteins were represented as nodes that interact through activation and inhibition edges. Semi-quantitative steady states (SS) of node activations were calculated. Then, a full factorial sensitivity analysis (SA) identified which out of the RNM 15 cytokines, and 4 growth factors affected most the structural proteins and degrading enzymes. The RNM was further evaluated against metabolic events measured in non-healthy human NP explant cultures, after 2 days of 1ng/ml IL-1B catabolic induction. The RNM represented successfully an anabolic basal SS, as expected in normal IVD. IL-1B was able to increase catabolic markers and angiogenic factors and decrease matrix proteins. Such activity was confirmed by the explant culture measurements. The SA identified TGF-β and IL1RA as the two most powerful rescue mediators. Accordingly, TGFβ signaling-based IDD treatments have been proposed and IL-1RA gene therapy diminished the expression of proteases. It resulted challenging to simulate rescue strategies by IL-10, but interestingly, IL-1B could not induce IL-10 expression in the explant cultures. Our RNM was confronted to independent in vitro measurements and stands for a unique model, to integrate soluble protein signaling and explore IDD.
