Ex vivo polarized pro-inflammatory vs. homeostatic monocytes/macrophages elicit differential responses within a human osteoarthritic joint explant model

Abstract

Purpose: Infiltrating pro-inflammatory monocytes/macrophages (MΦs, as a heterogenous population) are the most prevalent immune cells within the osteoarthritic knee joint and play a key role in osteoarthritis (OA) pathogenesis. MΦs exist on a functional spectrum from pro-inflammatory to homeostatic to promote inflammation and its subsequent resolution, as well as tissue remodeling in wound healing. This balance becomes dysregulated in disease states such as knee OA. However, the broad functional capacity of MΦs also allows them to be ex vivo polarized towards homeostatic or regulatory phenotypes to promote repair and inflammation resolution; polarized MΦ-based cell therapy has been escalated to early phase clinical trials for indications such as spinal cord injury, liver cirrhosis, and stroke. Our aim is to generate proof-of-concept data for a MΦ-based therapy for knee OA while investigating the disease contribution of pro-inflammatory and homeostatic MΦs in OA progression. Methods: An in-house human explant culture was optimized using full-depth biopsy punch cartilage and minced synovium from late stage knee OA arthroplasties for replicable responses in cartilage degradation and inflammation: qRT-PCR for gene expression, dimethylmethylene blue, Safranin-O, picrosirius red histology for extracellular matrix (ECM) loss, haemotoxylin and eosin histology for synovium structure, and multiplex immunoassays for secreted protein levels of cytokines and matrix proteases. Ex vivo polarized MΦs were generated using 48h cytokine-based protocols (IFN-γ+LPS, pro-inflammatory; IL-10+TGF-β, homeostatic) from healthy donor CD14+ peripheral blood mononuclear cells, validated by flow cytometry for surface markers (CD86/HLA-DR/CD163/CD206) and function (dextran endocytosis). Ex vivo polarized MΦs were then co-cultured with explant cartilage and synovium; tissue and conditioned medium were harvested at 2- and 7-day time points for analysis. Gene expression was analyzed by ΔΔCTand principle component analysis, while protein levels were evaluated by ANOVA. Results: Ex vivo polarized pro-inflammatory MΦs promote pro-inflammatory and catabolic cartilage gene expression, whereas homeostatic MΦs upregulate anabolic ECM gene expression and secretion of protease inhibitors (N=6, Figure 1); notably, PRG4 and TIMP-1 increase with homeostatic MΦ treatment by greater than 2-fold (gene, Figure 2A) or 1.5-fold (protein, Figure 2B). No significant change in proteoglycan loss was observed with MΦ treatment, but histology demonstrates chondrocytes retain physiological morphology over the course of explant culture. Synovium gene expression was variable between patients (N=4), but pro-inflammatory MΦ treatment trended to upregulate pro-inflammatory and chemotactic genes. Ex vivo polarized MΦs maintained their functional phenotype after co-culture with knee OA tissue. Conclusions: Ex vivo polarized pro-inflammatory versus homeostatic MΦs have differential effects on knee OA explant cartilage in co-culture with synovium, where homeostatic MΦs promote anabolic ECM and protease inhibitor genes. Our in-house explant model optimized for evaluating diverse short-term output on inflammation and cartilage degradation will allow further evaluation of MΦ roles in OA, as well as the testing of other novel OA therapies within human conditions. Our concurrent studies with murine knee OA models treated with ex vivo polarized MΦs will confirm our observations and lend credence to developing MΦ-based therapies for OA. View Large Image Figure ViewerDownload Hi-res image Download (PPT)