Background Our previous studies demonstrated increased levels of S100A4 protein in sera, synovial fluid and synovial membrane of patients with rheumatoid arthritis (RA) compared to osteoarthritis. S100A4 regulates apoptosis and induces production of matrix metalloproteinases by synovial fibroblasts. Furthermore, S100A4 stimulates synthesis of tumour necrosis factor (TNF)-α by mononuclear cells. The aim of this study was to investigate the effect of loss of S1004 in induction of experimental arthritis in the human TNF transgenic (hTNFtg) mouse model.
Methods We crossed the heterozygous hTNFtg mice with S100A4 knockout (S100A4-/-) mice. Mice were clinically assessed for paw swelling, grip strength and body weight weekly from 6th to 14th week of age in a blinded manner. Sections of hind paws and tibias were histologically analysed for synovial inflammation, cartilage loss, bone erosions, osteoclast numbers and bone formation parameters with the OsteoMeasure image analysis system.
Results In the group of hTNFtg; S100A4-/- mice, paw swelling, grip strength and body weight were significantly improved compared to hTNFtg; S100A4+/+ (p < 0.01 for all parameters). Consistent with the clinical observations, histological analysis of the tarsal joints of hTNFtg; S100A4-/- mice showed reduced pannus formation (area of inflammation decreased by 66 ± 3%, p < 0.01) and cartilage destruction (cartilage loss decreased by 63 ± 6%, p < 0.01) compared to hTNFtg; S100A4+/+ mice. Similarly, osteoclast numbers were decreased by 84 ± 3% (p < 0.01) and bone erosions were less severe (area of bone erosion decreased by 81 ± 4%, p < 0.01) in hTNFtg; S100A4-/- mice. Furthermore, hTNFtg; S100A4-/- mice were protected from systemic bone loss. Absence of S100A4 completely reversed increased osteoclast formation and bone resorption in hTNFtg mice. hTNFtg; S100A4-/- mice had an increased bone volume per total volume (BV/TV) by 78 ± 20% (p < 0.05) and a decrease in trabecular separation by 39 ± 4% (p < 0.05), decreased numbers of osteoclasts per bone perimeter (NOc/BPm decreased by 43 ± 2%, p < 0.01), decreased bone surface covered by osteoclasts (Oc.S/BS decreased by 52 ± 3%, p < 0.01), increased numbers of osteoblasts per bone perimeter (NOb/BPm increased by 129 ± 20%, p < 0.05) and increased bone formation rate per bone surface (BFR/BS increased by 112 ± 18%, p < 0.05) compared to hTNFtg; S100A4+/+.
Conclusions These results suggest that inhibition of S100A4 effectively prevents induction of experimental arthritis via protecting against TNF-induced synovial inflammation, cartilage and bone destruction, and systemic bone loss. Our results support the role of S100A4 in the pathogenesis of RA where an increased S100A4 protein in circulation and locally at sites of inflammation may be linked to the process of aggressive fibroblast behaviour. Thus, S100A4 might represent a novel therapeutic target in RA.