TissueGene-C promotes an anti-inflammatory micro-environment in a rat monoiodoacetate model of osteoarthritis via polarization of M2 macrophages leading to pain relief and structural improvement

Osteoarthritis (OA) is a progressive, degenerative, low-grade inflammatory disease characterized by cartilage extracellular matrix (ECM) degradation, synovitis, and structural modification of the subchondral bone, most commonly in the knee and hip joints. The main symptom of OA is pain, which leads to functional limitations that severely impact the quality of life of patients (Felson 2006; Hunter and Bierma-Zeinstra 2019; Loeser et al. 2012). Currently, OA affects more than 250 million people globally. Aging and obesity are known risk factors for OA, so its prevalence is expected to rapidly increase as the average life expectancy of the population increases and lifestyles become increasingly sedentary, thereby promoting obesity (Lee et al. 2013; Sharma et al. 2006; Wallace et al. 2017). Despite the high prevalence of OA, there are no effective disease-modifying osteoarthritis drugs (DMOADs) that can relieve pain, repair damaged cartilage, and support the healing of the surrounding joint tissues (Oldershaw 2012; Perera et al. 2012). Analgesia is the traditional treatment for OA, but it is insufficient in treating the structural modifications or altering the progress of the disease as analgesic drugs cannot reduce inflammation or halt/reverse cartilage damage (Appleton 2018). To overcome these limitations and develop more effective therapeutics, recent strategies have focused on the employment of chondrogenic growth or anti-inflammatory factors (Li et al. 2017).

OA is conventionally considered a “wear and tear” disease. However, recent research has demonstrated the importance of low-grade inflammation and chronic synovial inflammation in the development and symptomatic progression of OA. It has been shown that mechanically driven inflammation precedes detectable cartilage degeneration in OA (Roemer et al. 2011; Sokolove and Lepus 2013) and that activated synovial macrophages appear to play a role in OA pathogenesis (Bondeson et al. 2006; Goldring 1999; Pelletier et al. 1991, 2001). Thus, there are a number of ongoing efforts to treat OA with anti-inflammatory strategies designed to block pro-inflammatory cytokines or modify macrophage phenotype and function (Bondeson 2010; Bondeson et al. 2010; Mosser and Edwards 2008). Activated macrophages may exhibit 2 phenotypes: M1 macrophages that have pro-inflammatory properties and M2 macrophages that exhibit immunosuppressive properties (Mosser and Edwards 2008). The induction of M2 macrophages improves the micro-environment supporting tissue repair and regeneration in inflammatory diseases, such as OA. Transforming growth factor-β1 (TGF-β1) has been shown to possess potent immunosuppressive and anti-inflammatory properties through the regulation of the ability of monocytes/macrophages to release inflammatory cytokines (Becker et al. 2006; Zhang et al. 2016). Additionally, it positively regulates chondrocyte proliferation, differentiation, and ECM synthesis and deposition (Blaney Davidson et al. 2007; van Beuningen et al. 1994; Yang et al. 2001). Currently, there are no therapeutic strategies that incorporate cell and gene therapy with TGF-β1 as the key biological component.

TissueGene-C (TG-C) is a novel gene and cell therapy consisting of human allogeneic chondrocytes and irradiated GP2-293 cells overexpressing TGF-β1 mixed in a 3:1 ratio (Noh et al. 2010; Yoon et al. 2015). Clinical trials have demonstrated that TG-C treatment improves pain and function in patients with knee OA (Cherian et al. 2015; Ha et al. 2015). Moreover, TG-C reduces inflammation while boosting the intrinsic anabolic function of chondrocytes. These observations support TG-C as a DMOAD candidate. However, despite the remarkable results of these clinical trials, the molecular mechanisms underlying the effect of TG-C remain elusive. The aim of this study was to verify the potential of TG-C as a DMOAD and elucidate its mechanism of action, particularly its impact on macrophage polarization, in a well-established preclinical model of OA. In this study, we investigated whether TG-C can induce pain relief and improve the cartilage structure in arthritic knee joints by stimulating the polarization of M2 macrophages using a rat monoiodoacetate (MIA) OA model.

In this study, we examined the mechanism underlying the action of TG-C to determine whether it is a novel DMOAD that targets the inflammatory micro-environment in the OA synovial joint. We found that TG-C provided pain relief from day 7 post-treatment until the end of the observation period (Fig. 2a) as demonstrated using the von Frey filament test. Moreover, TG-C generated hyaline cartilage, which is predominantly composed of type II collagen and is the main constituent of normal joints, rather than type I collagen which is found in other connective tissues and fibrocartilage (Fig. 2b–d). This is important for improved biomechanical joint function and serves as a clinical endpoint.

After evaluating the long-term efficacy, we observed that the expression of IL-10 in the synovium was increased 4 days after treatment of TG-C. Since OA pathogenesis involves the upregulation of inflammatory responses, it is possible that IL-10 is the source by which TG-C combats the MIA-induced OA micro-environment. Interestingly, IL-10 correlated with the increased CD68, a marker of macrophages, in the TG-C group (Fig. 3). Macrophages can be categorized into two distinct subsets (M1 and M2) depending on their activation phenotypes (Ferrante and Leibovich 2012; Jiang et al. 2014). M1 macrophages express high levels of TNF-α, IL-12, IL-23, and low levels of IL-10; this leads to Th1 and Th17 immune responses, which can cause tissue damage (Dai et al. 2015). In contrast, M2 macrophages express high levels of IL-10 and low levels of IL-12 and promote tissue remodeling by reducing inflammation (He and Marneros 2013; Mankin 1982; Zhang et al. 2016). Combinations of IL-4, IL-10 and TGF-β have been shown to induce M2 macrophages (Cao et al. 2010; Mia et al. 2014). In this study, M1 and M2 macrophages were both present in the MIA-induced OA model; however, the level of M1 macrophages was higher (Fig. 4b). In contrast, treatment with TG-C increased the expression of M2 macrophage specific markers, including CD163, Arg1, IL-10RA and Hmox-1 (Fig. 5). This is important, as the M2 dominant micro-environment may play an important role in the improvement of OA symptoms. The imbalance of the M1/M2 macrophage ratio is associated with disease progression in various inflammatory diseases (Funes et al. 2018; Hristodorov et al. 2015; Parisi et al. 2018). In knee OA patients, the imbalanced ratio of M1/M2 macrophages was observed and the degree of this imbalance was associated with the severity of the knee OA. Therefore, the re-balancing of the M1/M2 macrophage ratio has been suggested as a novel therapeutic approach, particularly in the inflammatory and metabolic phenotypes of knee OA (Liu et al. 2018; Mobasheri et al. 2017; Van Spil et al. 2019). In this regard, the change toward the M2-dominant micro-environment by TG-C treatment reflects a reduction in inflammation and indicates the possibility of joint recovery, structure restoration and genuine disease modification. For instance, arginine metabolism is critical for the regulation of mammalian immune responses (Munder 2009) and expression of Arg1 and the resulting depletion of l-arginine is a strong immunosuppressive pathway of the immune system (Bronte and Zanovello 2005). Several studies have demonstrated that l-arginine deficiency leads to an impaired T-cell immune response. Boutard et al. demonstrated that TGF-β upregulates arginase activity in rat peritoneal macrophages and Durante et al. has also demonstrated that TGF-β stimulates l-arginine metabolism by inducing both Arg1 mRNA and arginase activity (Boutard et al. 1995; Durante et al. 2001). These studies support our findings as the changes induced by TG-C promotes an anti-inflammatory micro-environment in the OA knee joint.

Our analysis of macrophage phenotypic markers suggests polarization of macrophages from M1-dominant to M2-dominant following the TG-C injection (Figs. 4, 5). Importantly, neutralization studies indicate that both IL-10 (Fig. 6) and TGF-β1 (Fig. 7) are required for this effect, as well as for the reduction in pain and histological features of OA. This is consistent with our observations as neither HC nor TC alone showed independent OA treatment potential in the MIA model (Fig. 2a). Moreover, the neutralization studies strongly suggest that the induction of an M2 macrophage-dominant synovial micro-environment is key to the efficacy of TG-C in the arthritic knee joint.

The limitations of this study are as follows. Firstly, although we broadly classify macrophages into M1 and M2 types, we do recognize that M2 macrophages can be further divided into several phenotypic subtypes and the function of each subtype is slightly different. Subsequent studies on the population of M2 macrophages induced by TG-C and the long-term stability of their anti-inflammatory phenotype will be needed. In addition, even though CD86 is predominantly expressed in M1 macrophages, it is also expressed in dendritic cells and lymphocytes. Therefore, to accurately investigate the immune-regulating micro-environment induced by TG-C, further cellular characterization using flow cytometry and quantification of cell surface markers is necessary. Secondly, we found that the effect of TG-C was linked to M2 macrophages, but it is not certain that the therapeutic effect is directly induced by M2 macrophages or whether M2 macrophages function to restore homeostasis within an inflamed joint, which subsequently creates a more favorable micro-environment for the release and action of another factor or a combination of other factors that may synergistically mediate anti-inflammatory effects and symptom relief. Despite these limitations, our results support TG-C as a DMOAD candidate, as it provides pain relief and structural improvement in OA joints, which are key requirements of a novel DMOAD.

In summary, the present study indicates that TG-C can significantly reduce both pain and cartilage degeneration in an MIA-induced OA rat model by promoting M2 macrophage polarization through IL-10 and TGF-β1. After application of TG-C to the knee joint cavity, M2 macrophages promoted an anti-inflammatory micro-environment in the knee joints and contributed to structural improvement and analgesic effects. Although further studies are needed to clarify the prevailing mechanisms and identify any other synergistic factors, our study together with the previous clinical trial results demonstrate that TG-C is a DMOAD candidate.