Abstract
BACKGROUND: Osteoarthritis (OA) is increasingly thought to be a multifactorial disease in which sustained gut inflammation serves as a continued source of inflammatory mediators driving degenerative processes at distant sites such as joints. The objective of this study was to use the equine model of naturally occurring obesity associated OA to compare the fecal microbiome in OA and health and correlate those findings to differential gene expression synovial fluid (SF) cells, circulating leukocytes and cytokine levels (plasma, SF) towards improved understanding of the interplay between microbiome and immune transcriptome in OA pathophysiology. METHODS: Feces, peripheral blood mononuclear cells (PBMCs), and SF cells were isolated from healthy skeletally mature horses (n=12; 6 males, 6 females) and those with OA (n=6, 2 females, 4 males). Horses were determined to have OA via lameness evaluation, response to intra-articular (IA) diagnostic analgesia, and radiographic and arthroscopic evidence. Horses were excluded who had received medications or joint injections within 2 months. Cytokine analyses of plasma and SF were performed via multiplex immunoassay. Fecal bacterial microbial 16s DNA sequencing was performed and correlated to bulk RNA sequencing of SF cells and PBMC performed using an Illumina based platform. RESULTS: Horses with OA had higher body condition scores (P=0.009). Cytokines were elevated in plasma [interleukin (IL)-2, IL-6, IL-18, interferon gamma (IFN-γ), interferon gamma inducible protein 10 (CXCL10 or IP-10), granulocyte colony-stimulating factor (G-CSF)] and SF (IL-1β, IL-6, IL-17A, IL-18, IP-10, G-CSF) in OA. Microbial principal coordinate analysis (PCoA) using Bray-Curtis dissimilarity for β-diversity demonstrated distinct grouping of samples from OA versus healthy horses (P=0.003). Faith alpha diversity was reduced in OA (P=0.02). Analysis of microbiome composition showed differential relative abundance of taxa on multiple levels in OA. Specific phyla (Firmicutes, Verrucomicrobia, Tenericutes, Fibrobacteres), correlated to transcriptomic differences related to cell structure, extracellular matrix, collagen, laminin, migration, and motility, or immune response to inflammation in OA. CONCLUSIONS: These findings provide compelling evidence for a link between obesity, gut microbiome dysbiosis and differential gene expression in distant joint sites associated with development of OA in a relevant large animal model, establishing a connection here that provides a platform from which development of therapeutic interventions targeting the gut microbiome can build.