Adipose-derived mesenchymal stromal cells specifically modulated CXCL10/IP10 chemokine detected in osteoarthritic milieu

Abstract

Purpose: Osteoarthritic (OA) synovial fluid (SF) has been shown to contain cytokines (i.e., IL1β, IL6, TNFα) and chemokines (i.e., CXCL8/IL8, CXCL10/IP10, CXCL12/SDF-1, CCL3/MIP1α) mediators. Among them, CXCL10/IP10 chemokine levels has been reported to correlate with osteoarthritis severity by playing a role in leukocyte homing to inflamed tissues which may then lead to the perpetuation of chronic inflammation and resulting tissue damage. Adipose-derived mesenchymal stromal cells (ASC) are commonly utilized in various cell therapies and recently in phase I clinical trials to counteract the development of knee osteoarthritis. Administered-ASC in the knee joints enter in close contact with the OA inflammatory milieu and hypoxic environment, which could modulate their characteristics. The objective of this study was to investigate the OA milieu effect on good manufactured practice (GMP)-ASC, both in normoxic and hypoxic conditions and evaluated their migration properties and cytokine receptors. Moreover, we analyzed CXCL10/IP10 localization in OA synovium. Methods: Two different OA milieu: OA synovial fluid (SF) and OAconditioned medium (CM) from synoviocytes were used to treat GMP-ASC both in normoxia or hypoxia (1,5%O2) . These milieu adequately represent the OA environment for ASC injection and the milieu containing all the soluble factors released by the target tissue of ASC, respectively. GMP-ASC were tested for cell migration, proliferation, cytokine receptors expression (CXCR1, CXCR2, CXCR3, CXCR4, CXCR7, CCR1, CCR2, CCR3, CCR5, IL6R), and cytokines (CXCL8/IL8, CXCL10/IP10, CXCL12/SDF-1, CCL2/MCP1, CCL3/MIP1α, CCL4/MIP1β, CCL5/RANTES, IL6) release. Healthy SF was used as control. OA synovium was used to immunolocalize CXCL10/IP10. Results: In OA-SF we detected higher amount of CXCL10/IP10 than in OA-CM while CCL2/MCP1 and CCL4/MIP1β were higher in OA-CM compared to OA-SF. All the other cytokines were detected in the same amounts in both OA milieu. We demonstrated that GMP-ASC show an increase in proliferation, migration, no changes in the amount of cytokines released and modulation of CXCR1, CXCR3, CCR1 and CCR5 receptors in hypoxic condition. Moreover, GMP-ASC migration increased 15-fold when treated either with OA-SF or OA-CM compared to healthy SF, both in normoxic and hypoxic conditions. GMP-ASC treated in OA milieu showed an increase of CXCR3, CCR3 and IL6R and a decrease of CCR1 receptors. Blocking experiments demonstrate that CXCL10/IP10 was the only chemokine of the OA milieu down-modulated after treatment with GMP-ASC. We confirmed that CXCL10/IP10 was mainly positive on synovial macrophages. Conclusions: In conclusion, our study demonstrates a novel specific effect of OA milieu on both GMP-ASC migration and cytokine receptor expression that were strictly dependent on an inflammatory environment and hypoxic condition. Therefore, only the use of a specific culture OA environment allows to specifically define the real therapeutic effect of GMP-ASC, suggesting that their specific recruitment to synovial macrophages is partially dependent on CXCL10/IP10-CXCR3 axis. This is the first study that focuses on an adequate GMP-ASC injection environment, presenting novel indications that contribute to understand the role of OA milieu on GMP-ASC .