The most important finding of this study was that the M2Φ in the transplanted SVF could act directly on intra-articular tissues to secrete TGF-β and IL-10, which may have contributed to the therapeutic effects of the SVF on OA. To the best of our knowledge, this is the first study to demonstrate a possible direct effect of the M2Φ within the SVF on OA treatment. These findings suggest that SVF transplantation may be an effective option for regenerative therapy in cases of OA.
In this study, to evaluate the differences in therapeutic mechanisms between the SVF and ADSCs, macroscopic, histological, and immunohistochemical analyses were performed in the in vivo experiments via the xenotransplantation of human SVF and ADSCs into immunocompromised rat OA models. Macroscopic and histological analyses revealed that the SVF group showed superior suppression of OA progression and synovitis than the ADSC group. Immunohistochemical staining showed that collagen II and SOX9 expression was higher in the SVF group than in the ADSC group; in contrast, MMP-13 and ADAMTS-5 expression was lower in the SVF group than in the ADSC group. SOX9, a chondrogenic transcription factor, plays an important role in chondrogenesis, particularly by activating collagen II coexpression[41, 42]. MMP-13 and ADAMTS-5 affect the degradation of collagen II and aggrecan, respectively, causing degenerative changes in cartilage, and have been used as indicators of therapeutic efficacy in OA models[43, 44]. In summary, the in vivo experiments revealed that increased anabolic effects and decreased catabolic effects on OA cartilage were more effective in the SVF group than in the ADSC group.
Next, macrophages in the synovium were investigated in the in vivo experiments to understand the mechanisms underlying the OA-suppressing effects of the SVF and ADSCs. Previous studies have indicated that synovial macrophages play an important role in the pathogenesis of OA[45–47]. M1 macrophages in the synovium exacerbate experimental OA, while M2Φ reduce its progression, and in particular, an increased M1/M2 ratio is associated with increased expression of inflammatory cytokines, resulting in OA[48, 49]. Previous studies have reported that ADSCs exert anti-inflammatory effects by polarizing macrophages to the M2 phenotype via the PGE2 pathway[27, 28]. Another report showed that intra-articular injection of the SVF increased the number of M2Φ in the synovium of a mouse OA model[29]. In this study, the SVF group showed markedly superior suppression of synovitis and a decreased M1/M2 macrophage ratio than the ADSC group. Thus, the SVF group, which contained the same number of stem cells as the ADSC group, showed stronger polarization toward M2Φ than the ADSC group, suggesting that the immunomodulatory effect of the SVF was partly influenced by heterogeneous cells other than ADSCs. In addition, immunofluorescence staining revealed hNA-positive nuclei in F4/80- and CD163-positive cells in rat synovial tissue in the SVF group. However, in the ADSC group, very few hNA-positive nuclei were found in the F4/80- and CD163-positive cells in the rat synovium. Thus, human-derived M2Φ in the transplanted SVF can directly affect the synovial tissues of recipient rats, and the paracrine effect of the SVF on OA may be due to the synergistic effects of ADSCs and M2Φ in the SVF acting directly and contributing to the maintenance of the synovial membrane and articular cartilage homeostasis.
Furthermore, the in vitro experiments were performed to confirm these in vivo findings. To assess the effect of M2Φ in the SVF, flow cytometry was performed to detect the percentage of CD163-positive cells in the SVF and sort M2Φ. Previous studies have reported that macrophages constitute approximately 20% of the SVF[24]. Of the macrophages in the SVF, approximately 70% were positive for M2 anti-inflammatory markers, supporting our results[25, 26]. M2Φ secrete TGF-β and IL-10, induce collagen II expression in chondrocytes, and decrease MMP-13 expression[50]. Previous studies have reported that co-culture of the SVF or ADSCs with isolated pellets produces notably larger pellets in the SVF group than in the ADSC group[30, 32]. Our results support these findings: moreover, the M2Φ group showed a larger pellet size than the ADSC group. In ELISAs, the levels of both TGF-β and IL-10 were highest in the SVF group, followed by the M2Φ, ADSC, and control groups. In contrast, the levels of MMP-13 were the lowest in the SVF group, followed by the M2Φ, ADSC, and control groups. Therefore, M2Φ were more effective than ADSCs in promoting the secretion of anti-inflammatory cytokines such as TGF-β and IL-10, inducing collagen II expression, enhancing anabolic effects on chondrocytes, and decreasing the expression of catabolic markers such as MMP-13. In summary, the paracrine effect of the SVF was influenced more by M2Φ than ADSCs, and the synergistic effect of M2Φ and ADSCs markedly increased anabolism and decreased the catabolic effect on chondrocytes.
Our study had several limitations. First, we administered fixed doses of the SVF and ADSCs via a single injection. However, the optimal dose in a rat OA model has not yet been determined. Second, we used a rat model of this experiment. Because rats are quadrupeds, the established OA model may not necessarily translate to a human model owing to differences in postural and gait dynamics between rats and humans. Third, the normal lifespan of a rat is approximately 2 years; however, the rats in this study were only 9 weeks old. Therefore, they may not correspond to changes in OA in middle-aged humans. Fourth, we used immunocompromised rats as a model because they are commonly used in similar studies on xenotransplantation to prevent the rejection of transplanted human tissues and cells[51–53]. However, several studies have reported that specific immune response signals inhibit OA progression[54, 55]. Thus, immunodeficiency may interfere with the therapeutic effects of the SVF. Finally, histological and immunohistochemical evaluations were performed using semi-quantitative methods. Therefore, more objective and accurate quantitative methods should be considered in future studies.