2. MSCs’ Molecular Mechanisms of Action
Results of a PubMed (MEDLINE) search of studies related to MSCs in knee OA were analyzed. The searches were from the beginning of the search engine until 31 October 2022 using the keywords “Knee osteoarthritis MSCs”. Only the studies on MSCs in knee OA that the authors considered to be of most interest were included. PubMed found 391 articles, of which 77 were selected.
MSCs play significant roles in the repair and regeneration process. These include the reduction of cell death to continually replace lost cells, the secretion of trophic factors that stabilize the extracellular matrix, and the suppression of immune cell activation to prevent inflammation
[10].
In one study, exosomes were harvested from ESC-MSCs in conditioned culture media by a sequential centrifugation process. Then, ESC-MSCs or their exosomes were intraarticularly injected. An in vitro model with primary mouse chondrocytes stimulated with interleukin-1 beta (IL-1β) was used to assess the impact of the conditioned medium with or without exosomes and titrated doses of isolated exosomes for two days, before immunocytochemistry or western blot analysis. Destabilization of the medial meniscus (DMM) surgery on knees of C57BL/6 J mice was used as an OA model
[11]. The results revealed that exosomes from human embryonic stem cell-induced MSCs (ESC-MSCs) had a beneficial therapeutic impact against OA by balancing the production and decomposition of the extracellular matrix (ECM) of chondrocytes
[11].
Using a Sprague–Dawley rat design of collagenase II and IL-1β-induced OA chondrocytes, long non-coding ribonucleic acid malat-1 from human MSC (hMSC)-delivered extracellular vesicles (EVs) promoted chondrocyte proliferation, alleviated chondrocyte inflammation and cartilage degeneration, and enhanced chondrocyte repair
[12].
A study by Yang et al. used tropoelastin as the injection medium and compared it with classic media, such as normal saline, hyaluronic acid, and platelet-rich plasma (PRP), in intraarticular MSC injection. The authors found that tropoelastin promoted the emigration of infrapatellar fat pad MSCs (IPFP-MSCs) and protected knee cartilage from OA damage by enhancing cell adhesion and activating the integrin beta-1/extracellular signal-regulated protein kinase 1/2/vinculin pathway. These findings provided new insights into intraarticular injections of MSCs for the treatment of OA
[13].
In one study, cultured MSCs originating from three types of tissues (bone marrow, adipose tissue, and synovium) were treated with IL-1β and tumor necrosis factor-alpha or not to attain conditioned media. Each conditioned medium was used to analyze paracrine factors related to cartilage restoration by liquid chromatography-tandem mass spectrometry. MSCs from these tissues expressed 93 proteins under normal circumstances and 105 proteins under inflammatory circumstances
[14]. The differentially expressed proteins might contribute to the regeneration of damaged cartilage.
In a suspended synovium culture model, MSCs were released from the synovial membrane across a medium into a non-contacting culture dish
[15]. The authors found that, in knees with OA, endogenous MSCs were possibly similarly mobilized from the synovium through the synovial fluid, acting in a protective manner. In the natural course of OA, however, the number of mobilized MSCs is limited, leading to OA progression. In a rat OA model, it was also noticed that injections of synovial MSCs inhibited the progression of cartilage degeneration. The injected synovial MSCs emigrated into the synovial membrane, maintained their MSC characteristics, and augmented gene expressions of TSG-6, PRG-4, and bone morphogenetic protein-2. That is, exogenous synovial MSCs could facilitate anti-inflammation, lubrication, and formation of cartilage matrix in osteoarthritic knees
[15].
It has been shown that chronic swelling results in excess Ca
2+ transfer from the endoplasmic reticulum to the mitochondria, leading to mitochondrial calcium overload and further mitochondrial harm
[16]. Moreover, under conditions of chronic inflammation, injured mitochondria accumulate over time in MSCs due to mitophagy by activation of the Wnt/β-catenin pathway, which impairs MSC differentiation. Zhai et al. isolated tissue-specific MSCs in the periodontal ligament, termed periodontal ligament stem cells, from healthy patients and patients with periodontitis. Based on the mechanistic invention, intracellular microenvironment (esterase and low pH)-responsive nanoparticles were constructed to capture Ca
2+ surrounding mitochondria in MSCs to control MSC mitochondrial calcium flux against mitochondrial dysfunction. The nanoparticles were able to liberate siRNA from MSCs to restrain the Wnt/β-catenin pathway and control mitophagy of the initially dysfunctional mitochondria. The aforementioned nanoparticles (“nanorepairers”) physiologically reestablished the activity of mitochondria and MSCs, which could be a new effective therapy against OA
[16].
3. Experimental Studies
A systematic review of animal studies concluded that intraarticular injections of MSCs could not be recommended for knee OA clinical trials. They also stated that, based on the internal and external validity of animal studies, high-quality experimental studies and greater efforts to translate preclinical studies to clinical trials were required at that time (in 2018)
[17].
The safety and efficacy of intraarticular allogenic MSC injections was investigated in a pig OA experiment after bilateral medial knee meniscectomy
[18]. Bone marrow-originated MDCs (BM-MSCs) were labelled with superparamagnetic iron oxide (SPIO) nanoparticles to permit cell tracing by magnetic resonance imaging (MRI). At a concentration of ≤20 µg/mL, SPIO nanoparticles were not toxic to BM-MSCs. Four weeks following surgery, OA lesions were noticed on the MRI. Between 8 and 24 h after the injections, the SPIO-marked BM-MSCs were displaced into the damaged portion of the cartilage. In addition, histological and immunohistochemistry analysis found no significant difference between the right knee (treatment group) and the left knee (control group). The appropriate concentration of SPIO nanoparticles for labelling BM-MSCs was 20 µg/mL, while allogenic MSCs were able to move into the impaired cartilage and accumulate around it. No significant differences were found between the treatment group and the control group
[18].
One study compared the efficacy of extracorporeal shockwave therapy (ESWT), Wharton’s jelly MSCs (WJ-MSCs), and a combination of ESWT and WJ-MSCs in early knee OA in rats. The results of the pathological study, micro-computed tomography, and immunohistochemistry stain demonstrated that all three options significantly improved early knee OA. The combined therapy group augmented the bone volume and trabecular thickness, as well as diminished the synovitis more than ESWT or WJ-MSCs alone. However, there was no significant difference in the combined ESWT and WJ-MSCs, as demonstrated in the expressions of IGF-1 and transforming growth factor (TGF)-β1, and the decrease in the TUNEL activity in OA knees. Moreover, WJ-MSC therapy significantly augmented the expression of type II collagen compared with ESWT and ESWT combined with WJ-MSCs in OA knees. In the mechanistic factors analysis, a synergistic effect was observed by ESWT combined with WJ-MSCs in the expression of RUNX-2, SOX-9, and collagen Xα1 on knee OA
[19].
It has been observed that IPFP-MSC-derived exosomes protected articular cartilaginous tissue from injury and improved gait anomaly in OA mice while keeping cartilage homeostasis, a mechanism that could be related to miR100-5p-regulated constraint of the mTOR-autophagy pathway. Wu et al. investigated the role and underlying mechanisms of IPFP-MSC-derived exosomes on OA in vitro and in vivo. At that time (2019), those authors stated that IPFP-MSC-derived exosomes could have utility in the treatment of knee OA, given that it was relatively easy in clinical practice to obtain human IPFP from patients with OA by arthroscopic surgery
[20].
One study found that intraarticular injection of human umbilical cord MSCs expressing miR-140-5p induced cartilage self-repair in rat OA, highlighting the potential therapeutic utility of such injections in OA treatment
[21]. In another study of adult male albino rats, it was observed that intraarticularly-injected umbilical cord blood MSCs cured knee OA better than when they were intravenously injected
[22].
In a rat OA model, it was found that selective administration of kartogenin to synovial fluid-derived MSCs (SF-MSCs) by engineered exosomes produced a uniform dissemination of kartogenin in the cytosol, increased its effective concentration in the cell, and strongly promoted chondrogenesis of SF-MSCs in vitro and in vivo
[23]. Using a rat OA model, other authors observed that exosomes from human bone marrow MSCs (BM-MSCs) had a beneficial therapeutic impact on OA by diminishing senescence and the death of chondrocytes. This result suggested that MSC-derived exosomes could have therapeutic value in OA
[24].
In a rat knee OA model, it was demonstrated that low-intensity pulsed ultrasound (LIPUS) improved the therapeutic effectiveness of MSCs in cartilage reconstruction by increased autophagy-mediated exosome liberation
[25]. In MSCs isolated from rat bone marrow in vitro, results showed that LIPUS facilitated exosome liberation from MSCs by triggering autophagy. The in vivo results showed that LIPUS substantially potentiated the positive impact of MSCs in OA cartilaginous tissue. This impact was substantially reduced by GW4869, an inhibitor of exosome liberation
[25].
In a murine OA model, mouse ADSCs were acquired from adipose tissue and transfected with modified RNA. The results of histological and immunohistochemical analyses of knee joints harvested at 4 and 8 weeks after OA induction indicated that insulin-like growth factor 1 (IGF-1)-adipose derived stem cells (ADSCs) had a better therapeutic effect than native ADSCs. This outcome was shown by an inferior histological Osteoarthritis Research Society International score and less ECM loss. Such results supported the potential therapeutic utility of IGF-1-ADSCs for the treatment of OA and cartilage repair in clinical practice
[26].
In one study, articular cartilage defects were created in the intertrochlear groove of articular cartilage in rabbit femurs. Integrin α10-MSCs were labeled with SPIO nanoparticles co-conjugated with rhodamine B to allow visualization by both MRI and fluorescence microscopy. The results showed emigration and homing of human integrin α10β1-selected MSCs to cartilage defects in the rabbit knees following intraarticular administration, as well as chondrogenic differentiation of MSCs in regenerated cartilage tissue
[27].
A recent publication evaluated the effectiveness of BM-MSC management in cartilage repair, utilizing a rat experiment of monoiodoacetate-induced AO of the knee joint. OA was induced in the knee joint of rats by an intracapsular injection of monoiodoacetate (2 mg/50
μL) on day zero. The authors concluded that BM-MSCs could be an effective treatment for inflamed knees, and that their effect could be mediated by their anti-inflammatory and antioxidant potential
[28].
Ai et al. found that MSCs and MSC-EVs reduced OA pain through direct action on peripheral sensory neurons
[29]. In their study, the authors elicited knee OA in adult male C57BL/6J mice by DMM surgery. The DMM mice treated with MSCs and MSC-EVs did not show the pain-related behavioral changes (i.e., locomotion, digging, and sleep) that the untreated DMM mice did. The lack of pain-related behaviors in the MSC/MSC-EV-treated mice was not due to diminished joint damage, but rather to the knee-innervating sensory neuron hyperexcitability that was observed in the untreated DMM mice. Moreover, they found that NGF-induced sensory neuron hyperexcitability was averted with MSC-EV management
[29].
In a medial meniscal transection pre-clinical model of OA, sodium alginate microencapsulation of human MSCs modulated the paracrine signaling response and improved the efficacy of OA treatment. Three weeks post-surgery, after OA was established, intraarticular injections of encapsulated hMSCs or nonencapsulated hMSCs were administered. Six weeks post-surgery, microstructural changes in the knee joint were quantified with micro-computed tomography. Encapsulated hMSCs diminished articular cartilage degeneration and subchondral bone remodeling
[30].
In a systematic review of animal models and cell doses, it was observed that rats were the most frequently employed species for modeling knee OA, and that anterior cruciate ligament transection was the most frequent approach used for producing OA
[31]. A correlation was found between the cell doses and the body weight of the animals. In clinical trials, there was a great disparity in the dose of MSCs used to manage knee OA, ranging from 1 × 106 to 200 × 106 cells, with a mean of 37.91 × 106 cells. It was also found that in preclinical and clinical studies on knee OA, MSCs have significant potential for pain relief and tissue protection
[31].