Bone marrow contribution to synovial hyperplasia following joint surface injury

Background: Joint surface injury, a known risk factor for osteoarthritis, triggers synovial hyperplasia, which involves proliferation of mesenchymal stromal/stem cells (MSCs). Whether these proliferative MSCs are resident synovial cells or move into the tissue from elsewhere is not known. The aim of this study was to determine the contribution of bone marrow-derived cells to synovial hyperplasia following joint surface injury. Methods: Lethally irradiated mice were transplanted with green fluorescent protein (GFP)-labelled bone marrow, and MSC chimerism was determined by the colony-forming unit fibroblast (CFU-F) assay and phenotypic analysis. To label host slow-cycling cells prior to bone marrow transplant, mice received iododeoxyuridine for 3 weeks. Mice then were subjected to GFP(+) bone marrow transplant, underwent joint surface injury and received chlorodeoxyuridine (CldU) for 7 days to label cells proliferating after injury. GFP- and nucleoside-labelled cells in normal and injured knee joint synovium were quantified in situ by immunofluorescence staining of paraffin-embedded tissue sections. The phenotype of GFP-labelled cells was determined by co-staining for the haematopoietic marker CD16/CD32 and the MSC/fibroblast marker platelet-derived growth factor receptor a (Pdgfra). Results: CFU-F assay and phenotypic analysis demonstrated successful bone marrow mesenchymal lineage chimerism in mice that underwent transplants. Bone marrow reconstitution preceded the detection of GFP-labelled cells in synovium. The percentage of GFP(+) cells in synovium increased significantly in response to injury, while the proportion of GFP+ cells that were labelled with the proliferation marker CldU did not increase, suggesting that the expansion of the GFP+ cell population in synovium was due mainly to bone marrow cell infiltration. In contrast, proliferation of host slow-cycling cells was significantly increased in the hyperplastic synovium. In both control and injured knee joints, the majority of marrow-derived GFP(+) cells in the synovium were haematopoietic (CD16/32(+)), while a minority of cells expressed the pan-fibroblast/MSC marker Pdgfra. Conclusions: Our findings indicate that synovial hyperplasia following joint surface injury involves proliferation of resident slow-cycling cells, with a contribution from infiltrating bone marrow-derived cells. Understanding the process of synovial hyperplasia may reveal ways to restore homeostasis in injured joints and prevent secondary osteoarthritis.