Infrapatellar fat pad from osteoarthritis patients displays a distinctive eicosanoid release profile compared to healthy donors

Purpose: Obesity, an important risk factor for osteoarthritis (OA), induces inflammatory changes in adipose tissue. The infrapatellar fat pad (IPFP) present in the knee joint is considered as an active osteoarthritis joint tissue contributing to local inflammatory and destructive processes. However, soluble mediators released by IPFP involved in these processes are largely unknown. Eicosanoids are essential mediators of the inflammatory response and contribute to the initiation and the resolution of inflammation. In this study we aimed to detect differences between healthy and OA derived IPFP based on their secretion of lipid mediators involved in the eicosanoid pathway. Methods: IPFP explants of 13 OA donors (joint replacement surgery) and 11 donors with macroscopically healthy knee joints (post-mortem) were cultured for 24 hours (50 mg IPFP explant/ml) and the supernatant was collected. Fat conditioned medium (FCM) samples were measured by an LC-MS/MS method for lipid mediators of inflammation (e.g. oxylipids, eicosanoids). Data were subjected to univariate and multivariate statistical analyses. Results: On an individual basis, lipid mediators involved in the eicosanoid pathway were not significantly different between OA and healthy FCM. However, when assessed with multivariate statistical analysis (PLS-DA) 14 out of 29 quantified eicosanoids were found to be important to distinguish OA from healthy FCM. Lipoxin A4, considered as an important anti-inflammatory oxylipid, was the most discriminative eicosanoid in the model with lower levels in OA than in healthy FCM. Prostaglandins (PG) derived from eicosapentaenoic acid (EPA) are generally regarded as anti-inflammatory and lower levels of PGD3 were found in OA FCM. In contrary, no changes in level of EPA were observed. Remarkably, the levels of docosahexaenoic acid (DHA) and its derived precursor for resolvin D1, generally regarded as anti-inflammatory, were higher in OA FCM. The level of the fatty acid arachidonic acid (AA), generally regarded as a precursor for pro-inflammatory PG, was higher in OA than in healthy FCM. This was supported by a higher level of PGF2alfa and PGD2 in OA FCM. Conversely, AA-derived PGB2 and PGF2beta levels were lower. Furthermore, OA patients showed increased concentrations of thromboxane A2. Linoleic acid-derived eicosanoids 9-HODE and 13-HODE, major components of ox-LDL and regarded as markers for lipid peroxidation, were lower in OA. Overall, the differences observed between healthy and OA FCM could not be attributed to basal metabolism changes, because downstream products of a certain pathway were differentially affected (e.g. PG derived from AA). Conclusions: OA patients can be distinguished from healthy persons based on their IPFP secretion of lipid mediators involved in the eicosanoid pathway. However, these differences are small, since none of the distinctive eicosanoids were significantly altered on an individual basis. Based on biological pathway analyses the eicosanoid profile associated with OA is likely to be specifically regulated by the disease process rather than due to basal metabolism changes or generic enzyme alterations. In general, pro-inflammatory eicosanoids appear to be higher and anti-inflammatory appear to be lower in OA than in healthy FCM which strengthens our hypothesis, but there are exceptions. These data provide new insight in mechanisms important for the pathogenesis of OA.