Innate immune pathway activation to modulate mesenchymal stromal cell (MSC) interactions with synovium and cartilage

Mesenchymal stromal cells (MSCs) have been evaluated as a local therapeutic option to treat osteoarthritis (OA) with conflicting clinical results. Our previous studies have evaluated immune licensing of MSC through activation of Toll-like receptor and cytosolic cGAS-STING pathways, with demonstrated improvement in functional and structural outcomes in a rodent model of OA. The objective of this study was to investigate impact of MSC activation on their interaction with relevant joint target cells to better understand the mechanisms by which pre-activation improves MSC activity for treatment of osteoarthritis.

Equine bone-marrow-derived MSCs (passage 2-3) from 3 healthy donors were stimulated with a TLR3-pathway agonist (polyinosinic:polycytidylic acid) or STING pathway agonist (2'3'-cGAMP) (10 μg/mL, 2 h, 2 × 10⁶ cells/mL in suspension). Cells were plated (100,000 cells/well, 24-well plates) and conditioned media (CM) collected at 24 h. Equine monocyte-derived macrophages, synovial cells, and chondrocytes were stimulated with IL-1ß/TNF-α (20 ng/mL, 2 h), washed and treated 24 h with MSC-CM, TLR-MSC-CM or STING-MSC-CM, washed and cultured 24 h. CM was examined for cytokine secretion by multiplex immunoassay and ELISA (25 cytokines). Bulk RNA sequencing was performed on MSC and joint cell lines via an Illumina based platform.

TLR-MSC-CM decreased IL-1β (p = 0.02), IL-6 (p = 0.02) secretion by synoviocytes and IL-18 secretion by activated chondrocytes (p = 0.002). STING-MSC-CM decreased IL-6, IL-8 secretion (p = 0.08) by synoviocytes, decreased IL-8 (p = 0.05) by activated chondrocytes, increased G-CSF (p = 0.01), IL-4 (p = 0.01) and decreased IL-5 (p = 0.01) by activated macrophages. Transcriptomic analyses indicated differential gene expression in each cell line following CM treatment varied by cell line. STING-MSC-CM vs TLR-MSC-CM induced 38 significantly altered DEGs in synoviocytes, 20 in chondrocytes, and 47 in macrophages.

These findings indicate that joint cells respond differently to factors secreted by TLR or STING pathway activated MSC. The pathways altered were different for each target cell type and no clear pattern of responses was apparent. These results indicate that in vitro modeling of target cell responses to "licensed" MSC can provide new information on the MSC and target cell interactions, though ultimately the functional impacts of activated MSC need to be evaluated using in vivo models.