Osteoarthritis (OA) has always been an uncertain condition in terms of inflammation even though the name itself denotes that it is an inflammatory process. Previously it was believed that OA was caused due to biomechanical causes and was employed as a negative control for inflammation during comparisons to rheumatoid arthritis (RA) ^[1][16]. It has now been brought to light that OA is much more than just an injury caused due to overuse of the joint. OA is a complex biological response as a result of its interaction with tissue resident cells and their mediators which amplifies physical stress incapacitating the normal function of ligaments, muscles and menisci ^[2][3][17,18]. Research has identified the process of inflammation as the initial step along the negative chain of events that leads to early OA.

Although the inflammatory response in OA is not as pronounced as RA, several authors have confirmed there is low-grade inflammation in OA ^[4][19]. The presence of inflammation in OA has been studied extensively using various techniques, in the early 1980′s Goldenberg et al. exhibited that majority of the inflammation is present in the synovium of the OA patients through histopathological analysis ^[5][20]. Additionally, the histological evidence was validated when correlated to the levels of serum C-reactive protein (CRP) and the levels of the inflammatory marker interleukin-6 (IL-6) in the synovial fluid in patients undergoing total hip or knee arthroplasty ^[6][21]. A comparative study between synovial tissues of patients with early and late OA revealed increased infiltration of mononuclear cells and inflammatory cytokines in patients with early knee OA to late knee OA ^[7][22]. Later, due to sensitive imaging techniques like magnetic resonance imaging (MRI) and direct arthroscopic visualization, it was confirmed that inflammation is visible in the synovium at the early stages of OA even before there is visible articular cartilage damage ^[8][23]. A recent study revealed that synovitis is one of the key factors in identifying early OA which was confirmed through the analysis of serum matrix metalloproteinase-3 (MMP-3) concentration, effusion-synovitis volume and synovial score ^[9][24]. All these studies validate the significant role of synovitis at any (early or late) stage of OA.

In the past decade, researchers have also been investigating the connection between low-grade synovitis and the manifestation of OA’s clinical symptoms. Synovitis has been linked to more severe symptoms including pain and joint dysfunction and may generate a more rapid deterioration of cartilage ^[10][4]. Synovitis has been linked to symptoms such as discomfort in people with knee OA. The correlation between pain and synovitis on MRI, found that changes in pain levels over time corresponded with changes in synovitis, lending credence to the idea that the two are causally connected ^[11][25]. Recently, a similar relationship between pain and synovitis was described using contrast-enhanced MRI wherein the likelihood of experiencing painful knee OA was found to rise 9-fold with increasing synovitis severity ^[12][26]. Ayral et al., published a study that established a link between synovitis and the progression of cartilage degeneration. The presence or absence of synovitis and the overall health of the cartilage surfaces were easily discerned during the initial arthroscopy. The rate of cartilage degeneration was measured by an arthroscopic examination performed 12 months after the original surgery. The presence of synovitis was related with more severe chondropathy at baseline and was present in around 50% of patients. Moreover, at one year, patients with synovitis were more likely to have advanced cartilage pathology than those without the inflammation ^[13][27].

During OA, the macrophages fail to keep up their stability and are activated through various ways mainly when macrophages are stimulated by damage associated molecular patterns (DAMPs) or pathogen-associated molecular patterns (PAMPs) upon interaction with germline-encoded surface pattern recognition receptors (PRRs) on macrophages; they activate the nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) pathway, causing the cells to release an increased amount of inflammatory mediators ^[14][29]. Another key signaling channel is the NOD-like receptor family, pyrin domain-containing 3 (NLRP3) inflammasome-mediated pathway. Both pathways can activate the macrophages during OA and trigger the production of two of the most extensively studied pro-inflammatory cytokines, interleukin-1β (IL-1β) and tumor necrosis factor (TNF-α).

IL-1β released by macrophages stimulates chondrocytes to synthesize MMPs, especially collagenase-1 (MMP-1), stromelysin (MMP-3), collagenase-3 (MMP-13), and ADAMTS-4 and 5, which are known to cause cartilage degradation and synovial damage ^[15][16][30,31]. IL-1β increases the production of other cytokines, such as IL-6, and IL-8 and chemokines, such as CCL2 (monocyte chemoattractant protein-1, MCP-1), macrophage inflammatory protein-1 alpha (MIP-1α/CCL3), and C–C motif chemokine ligand 5 (CCL5) in chondrocytes via a paracrine mechanism ^[17][32]. These mediators attract new macrophages to the joint, where they continue to release IL-1β, thereby prolonging the inflammatory cycle ^[18][33]. Additionally, IL-1β promotes the release of a variety of pro-inflammatory mediators, such as prostaglandin E2 (PGE2), nitric oxide (NO), and cyclooxygenase-2 (COX-2), which stimulate the extracellular signal-regulated kinases (ERKs) pathway. Activation of the ERK pathway inhibits type II collagen and aggrecan formation, as well as ECM synthesis ^[19][34]. Likewise, the c-Jun N-terminal kinases (JNKs) pathway also inhibits collagen II synthesis by inhibiting the SOX-9 gene. The mitogen-activated protein kinase (MAPK) signaling cascades regulates MMP-1, MMP-13, and ADAMTS-4, whereas MMP-3 and ADAMTS-5 are exclusively regulated by the ERK and JNK pathways, respectively ^[20][35].

Similarly, TNF-α also exhibits comparable effects on chondrocytes, increases IL-6, IL-8 and IL-18, suppresses the production of proteoglycans and type II collagen, and stimulates chondrocytes to generate MMPs and ADAMTS for ECM degradation ^[21][22][36,37]. Alternate to secreting inflammatory mediators, activated macrophages also produce growth factors. Vascular endothelial growth factor (VEGF) contributes to the severity and inflammation of OA. The articular cartilage, serum, and synovium of patients with late-stage OA show elevated VEGF expression. An increase in angiogenesis and VEGF production is the root cause of synovitis. VEGF has been shown to increase symptoms in patients with OA by stimulating inflammation-promoting macrophages to migrate throughout the inflamed tissue while also delivering nutrition and oxygen ^[23][38]. Other growth factors, including bone morphogenetic protein-2 (BMP-2), bone morphogenetic protein-7 (BMP-7), and transforming growth factor beta (TGF-β), contribute to the production of osteophytes and synovial fibrosis. Apart from osteophyte production, macrophages induce the formation of osteoclasts, which disintegrate the underlying bone and further degrade the cartilage and bone beneath by prompting osteoblasts to release a variety of cytokines and MMPs ^[24][39]. In addition to chondrocytes, activated macrophages stimulate neighboring fibroblast like synoviocyte (FLS) to produce a variety of inflammatory cytokines and chemokines, as well as pro-inflammatory mediators resulting in synovial hyperplasia, joint swelling, inflammation, and pain. FLS secrete a clear, viscid, lubricating fluid known as synovial fluid ^[25][40]. Macrophages also impair the primary role of FLS of maintaining cartilage homeostasis and shielding the cartilage surface from friction and deterioration ^[26][41]. The role of macrophages in inflammation is owed to their secretion of inflammatory cytokines. The majority of the cells within the joint interact with the cytokines released by macrophages, and these interactions influence the production of cytokines, other inflammatory mediators, and enzymes by these cells via intracellular signal transduction pathways, which plays a crucial role in the pathogenesis of OA.

Macrophages’ role in inflammation emphasizes that they have a critical role in OA immunopathogenesis and are not just a consequence of it. Therefore, depletion of macrophages can be a potential intervention that promote tissue repair and remodeling. Blom et al. demonstrated that removal of macrophages from the synovial lining decreased the expression of MMP-3 and MMP-9 in the synovium but not in the cartilage, and also reduced osteophyte formation in the collagenase induced OA (CIOA) mouse model ^[27][42]. Bondeson et al., showed that depleting synovial macrophages with anti-CD14-conjugated magnetic beads reduced inflammatory cytokines ^[28][43]. Correspondingly, the pro-inflammatory cytokines produced by FLS cease to secrete these cytokines and halts ECM degradation ^[29][44]. Wu et al. sought to determine whether the same is true for obesity-related OA and discovered that short-term macrophage depletion elevated synovitis and T-cell and neutrophil infiltration into the operated joint. These researchers concluded that macrophages are essential regulators of the responses of other immune cells and macrophage depletion cannot be employed to reduce inflammation in obese arthritic patients ^[30][45]. However, these results oppose those of Sun et al., who demonstrated that clodronate-liposome-mediated macrophage depletion and resolution of inflammation using a pro-resolving lipid mediator, resolvin D1, reduce pro-inflammatory gene expression and enhance anti-inflammatory gene expression in a similar obesity-associated mouse model of OA ^[31][46].

Crucial components to consider include the number of rounds of macrophage depletion and the type of depletion, which explains the reduction in OA severity observed in the clodronate-liposome-mediated depletion model, which was subjected to local depletion and frequent injections ^[31][46]. Although macrophage depletion is effective in reducing the severity of OA and inflammation, there are a few disadvantages associated with this approach. Macrophages are not only agents of destruction, but also play critical defensive and reparative roles in the host. Therefore, their depletion may have unintended consequences. When inflammation is localized to a single organ, systemic depletion of macrophages will have a major impact on the ability of macrophages to maintain homeostasis in all healthy tissues, which is clearly not a promising therapeutic approach ^[32][47]. Finally, macrophage depletion impairs the host immune system, which should be avoided.

Often, during an inflammatory response, macrophages can exhibit a spectrum of phenotypes; however, the two most frequent phenotypes that define macrophages are classically stimulated M1 and alternatively stimulated M2 macrophages ^[33][48]. M1 macrophages (CD80⁺, CD86⁺) mainly exert pro-inflammatory effects; they are formed owing to numerous stimuli, such as TNF-α, interferon gamma (IFN-γ), or lipopolysaccharide (LPS), leading to the release of large amounts of pro-inflammatory cytokines (IL-1β, IL-6, IL-8, and IL-18) ^[34][49]. M2 macrophage phenotype (CD163⁺ and CD206⁺) pertains to tissue repair and downregulation of inflammation and secrete anti-inflammatory cytokines (IL-4, IL-10, and IL-13) ^[34][49]. The presence of both M1 and M2 macrophages in the synovium has been described by Liu et al.; they conducted a study which analyzed the ratio of M1/M2 macrophages in human normal vs. OA knee samples and concluded that the increase in M1/M2 ratio positively corresponded to the severity of OA classified through the level of Kellgren Lawrence grade of OA in the knee ^[35][50]. Furthermore, the ratio of M1/M2 was also studied in canine ^[36][51] and equine models ^[37][52] wherein the synovial fluid samples showed higher M1/M2 ratio compared to normal groups. The presence of inflammatory macrophages in equine model was also confirmed through coculture of osteochondral-synovial explant ex vivo OA model wherein the ratio of NO (µM)/urea (µM) increased over time, suggesting that macrophages in the synovium gradually underwent a shift to M1 phenotype ^[38][53]. Another study correlated the radiographic OA intensity and symptoms with the quantity of activated macrophages present in the OA knee joints detected utilizing the imaging agent 99mTc-EC20 (etarfolatide), which specifically binds to folate receptor β (FR β) on activated, but not resting, macrophages or other immune cells ^[39][54]. Zhang et al. showed that predominantly M1 macrophages accumulated in human and mouse OA synovial tissue and not M2 macrophages. Further confirmation was provided with the help of a transgenic mouse model having enhanced M1 or M2 macrophages, the M1 macrophages in the synovium aggravated CIOA whereas the presence of M2 macrophage downregulated the development of OA ^[40][55]. Studies carried out with the help of anterior cruciate ligament transection (ACLT) rodent model and destabilization of medial meniscus (DMM) murine model, two well established animal models of OA also reported higher number of F4/80⁺ CD86⁺ and nitric oxide synthase 2 (NOS2⁺) M1 macrophages in the synovium ^[40][55]. However, data collected using the in-silico method CIBERSORT using publicly accessible transcriptome information revealed an abundance of M2 macrophages (30.1%), resting T-cells (23.9%), and activated NK cells (16.2%) in the synovial tissue of OA patients. While these statistics differ somewhat from the immunological profile of normal synovium (26.8%, 24.1%, 15.0%, respectively), the increase in M2 macrophages was statistically significant ^[41][56]. However, further research is needed to determine the importance of these alterations and the processes by which they may develop.

Currently, the options for treating OA are very limited. The already existing conventional therapeutic approaches such as physiotherapy, pharmacological drugs and surgery are not adequate as they are not significant enough to modify the prevailing course of the disease or prevent the process of cartilage degeneration. There is a significant need for disease modifying therapeutic intervention for OA. Recently, researchers have focused their attention on targeting macrophages due to their high plasticity and ability to perform distinct biological functions based on the signals received within the tissue microenvironment. In spite of the hypothesized detrimental role of activated macrophages in OA, their systemic depletion was found to be fatal. Instead, reprogramming macrophages may be a future therapeutic strategy ^[42][57]. Polarizing macrophages to an anti-inflammatory phenotype holds great promise for the treatment of inflammation in OA.

5. Macrophage Polarization by Mesenchymal Stem Cells

All of the aforementioned ways for polarizing macrophages are capable of reducing inflammation and pain, but they cannot repair the cartilage. Further cartilage deterioration requires complete knee arthroplasty ^[43][44][72,73]. OA treatment has the potential to be revolutionized by stem cell treatment. MSCs have piqued the interest of many researchers because of their experimental applicability, and the ability to differentiate into many lineages such as bone, muscle, fat, and cartilage ^[45][12]. MSCs can be extracted from bone marrow, synovium, adipose tissue, umbilical cord, blood, dental pulp and endometrium ^[46][74]. The International Society for Cellular Therapy (ISCT) has established three baseline requirements that all MSCs, regardless of their origin, must satisfy. Initial attachment to the plastic surface is required for MSCs to proliferate under standard circumstances. Additional requirements for MSCs include the expression of the surface markers CD73, CD90, and CD105. Furthermore, MSCs are required to differentiate into osteoblasts, chondrocytes, and adipocytes under certain in vitro conditions ^[47][75]. One of the most compelling arguments for making MSCs a standard treatment for OA is that they can repair cartilage, allowing damaged cartilage to regenerate ^[48][76]. MSCs are self-renewing stromal cells that can develop into a variety of cell types ^[49][77]. Although Friedenstein was the first to effectively isolate bone-forming cells from a guinea pig, Owen provided this field of research a much-needed boost by extending it to rats ^[50][78]. In 1992, it was announced that human bone marrow MSCs (BM-MSCs) had been extracted and cultivated to increase in number; by 1995, they were being pumped into patients ^[51][79]. Over the past quarter century, infusion approaches have demonstrated such a high level of safety that the Food and Drug Administration (FDA) now lists more than 950 clinical trials involving MSCs. MSCs have been used to treat a variety of orthopedic disorders, including OA, due to their tissue regeneration and immunomodulatory properties. MSCs have been shown to be effective in treating OA in phase 1 trials over the last few years and a number of unpublished Phase 2 trials, notably ADIPOA2 ^[52][53][80,81]. MSCs have an effect on immune cells, such as macrophages, dendritic cells, T lymphocytes, and natural killer (NK) cells ^[54][82]. MSCs play a pivotal role in controlling different functions of macrophages, such as differentiation of naive macrophages, modulation of their phagocytic ability, enhancement of their bactericidal effect, and manipulation of the plasticity and polarity of macrophages. MSCs have been reported to possess the property of immune evasiveness due to their close and reciprocal interaction with immune cells and their immunomodulatory properties ^[55][83]. This suggests that MSCs may be immune-protected when injected into an allogenic environment, preventing detection and rejection by the immune system. Nevertheless, MSCs do not need to remain in the body for an extended period of time in order to exert a therapeutic effect. A brief presence can permanently alter tissue cell behavior under certain pathological conditions. Thus, it is necessary to comprehend how the host immune system reacts to allogenic MSCs and how this may influence the therapeutic efficacy of MSCs in various inflammation models. MSC-based therapies have been developed in the context of inflammation observed in numerous disease models, such as graft versus host disease (GVHD) ^[56][84], inflammatory bowel disease (IBD) ^[57][85], diabetic cardiomyopathy ^[58][86], and many others.

6. Priming Enhances Mesenchymal Stem Cell Immunomodulation

6.1. Proinflammatory Cytokines

Priming MSCs with proinflammatory cytokines such as TNF-α has been demonstrated to upregulate key paracrine mediators such as IDO, PGE2, and hepatocyte growth factor (HGF), but to a lesser extent than IFN-γ. MSCs primed with IFN-γ have been reported to release adhesion proteins VCAM1 and ICAM1, as well as the chemokine ligands CXCL9, CXCL10, and CXCL11 and high levels of HLA-G and IDO. Co-culturing IFN-γ primed MSCs with activated PBMCs increased the frequency of CD4⁺CD25⁺CD127^dim/- T-cells, IL-10 and IL-6, while decreasing the frequency of Th17 cells, IFN-γ and TNF-α production ^[59][168]. MSCs stimulated with IFN-γ enhanced production of programmed cell death-1 ligands (PDL-1) to inhibit T-cell effector function and have also been demonstrated to inhibit NK cell cytotoxicity ^[60][169]. Similarly, BM-MSCs preconditioned with IL-17 reduced Th1 secretion of cytokines such as TNF-α, IFN-γ, IL-2 and enhanced iTreg cell formation. Furthermore, genes such as MMP1, MMP13, and CXCL6 that are predominantly correlated with migration and chemostatic responses were identified ^[61][170]. In light of the fact that MSCs from varying donors and sources exhibit varying cytokine priming responses ^[62][171], it may be necessary to combine cytokine priming in order to maintain a significant and consistent effect. Compared to priming MSCs with a single agent, combined IFN-γ and TNF-α priming significantly reduced donor-specific variability in MSC immunomodulatory potency. Chenyang Liu et al. demonstrated that supernatant from MSCs that have been pretreated with IFN-γ and TNF-α has been shown to switch macrophages to the M2-type, which in turn promotes cutaneous wound recovery with minimal scarring by stimulating the IL-6-dependent signaling pathway ^[63][172]. Likewise, in another study MSCs isolated from menstrual blood and stimulated with IFN-γ and TNF-α showed elevated levels of IDO1, EV release, and differential expression of miRNAs related to the immune response and inflammation ^[64][173]. Besides IFN-γ and TNF-α, MSCs primed with different combinatorial cytokine cocktail like LPS/ TNF-α also exhibited polarization of macrophages to the M2 phenotype expressing high amounts of PGE2, Arg1, and CD206, and displayed improved alkaline phosphate activity and bone mineralization potential ^[65][174]. The miRNA expression profile of foreskin MSCs is drastically altered after treatment with a cytokine cocktail containing IL-1β, TNF-α, IFN-γ, and IFN-α, with 13 miRNAs being downregulated and 3 others being upregulated. These miRNAs with altered expression levels are speculated to target multiple potential signaling pathways that control cellular activity in response to inflammatory cues. Several pro-inflammatory cytokine mixes have been used to alter the expression of immune mediators and miRNAs by MSCs in culture ^[66][175]. One of the major drawbacks of this strategy is the cost of recombinant cytokines.

6.2. Chemical Agents

In order to reduce the costs associated with recombinant cytokines, MSCs have been primed with a variety of pharmacological chemicals and small molecules in order to increase their therapeutic efficacy. The use of chemical agents like all-trans retinoic acid (ATRA) has shown to inhibit PBMC production of pro-inflammatory cytokines. Priming MSCs with ATRA improves wound-healing capacities in vivo. The gene expression of COX-2, VEGF, CCR2, HIF-1α, CXCR4, angiopoietin-2 (Ang-2), and angiopoietin-4 (Ang-4) is elevated by preconditioning rat BM-MSCs with ATRA ^[67][176]. Matteo Haupt et al. showed that the therapeutic potential of MSC EVs preconditioned with lithium is higher than that of EVs from native MSCs. Increased levels of miR-1906, a new regulator of toll-like receptor 4 (TLR4) signaling, were found in MSC EVs after treatment with lithium, which led to reduced cerebral inflammation and rapid neuroprotection in mice with stroke ^[68][177]. The histone deacetylase inhibitor valproic acid (VPA) and the bioactive lipid sphingosine-1-phosphate (S1P) have similar anti-inflammatory and proliferative actions. Priming MSCs with valproic acid and lithium before intranasal infusion improved neuropathological characteristics and function in a mouse model of Huntington’s disease ^[69][178]. Priming MSCs with cytokines and chemicals added to MSC culture media facilitates their ex vivo growth for therapeutic applications. This has been examined in relation to the preparation of xeno-free and serum-free media. In a recent study, Jin et al. created a hypoxic, calcium-rich environment for stem cells to grow while preserving them in a xeno-free, chemically defined cryopreservation media. The paracrine factor PTX-3 generated by these stem cells was shown to remodel M1 macrophages into their anti-inflammatory M2 phenotype in a rat OA model ^[70][179]. Small molecules are also being investigated as a method of priming MSCs due to their unique qualities such as low cost, minuscule size, robust stability, and non-immunogenicity. Oren Levy et al. showed a decrease in the expression of TNF-α at the site of inflammation after pre-treatment of MSCs with a kinase inhibitor (Ro-31–8425). Similar to Ro-31-8425, priming of MSCs with the small molecule tetrandrine boosted PGE2 synthesis via the NF-κB/COX-2 signaling pathway, which reduced TNF-α production in RAW264.7 during co-culture ^[71][180].

6.3. Hypoxia

Under hypoxic growth conditions, when the oxygen level is between 0 and 10%, MSCs can secrete more immunomodulatory molecules. It is well documented that hypoxic preconditioning can stimulate the production of immunomodulatory molecules in MSCs, such as IDO, IL-10 and PGE2 ^[72][181]. Hypoxia-exposed MSCs drive bone marrow-derived macrophage polarization to the M2 phenotype via the TGF-1/Smad3 signaling pathway, ameliorating ischemic stroke conditions by reducing apoptotic cells and fibrosis and promoting neovascularization in the infarcted region ^[73][182]. According to recent studies, EV density and load might also be modified by employing hypoxic preconditioning. However, hypoxia seemed to have no effect on the mean size, morphology, or surface biomarkers of MSC-derived EVs ^[74][183]. In response to hypoxia and serum deprivation, primed MSCs produced more dipeptides, suggesting that hypoxic MSCs augment their pool of free amino acids to meet energy requirements that cannot be properly met by the glycolytic process. Subsequently, it was also established that there are 21 different metabolites in primed MSC derived exosomes that have been linked to immunoregulation. The activation of regulatory T-cells, the polarization of macrophages toward the M2 state, and the regulation of anti-inflammatory responses are all directly influenced by these molecules ^[75][184]. Despite evidence indicating that MSCs grown under hypoxic conditions can result in the production of EVs, the real situation is still unclear. The variation may be attributed to the degree of hypoxia, as minute variations in oxygen concentration and exposure time can have a significant impact. In addition, it is important to note that while some studies have shown that hypoxia may promote cellular longevity, others have shown that cells may die ^[76][185].

6.4. Biophysical Stimulation

Another strategy that has been investigated is biophysical stimulation of MSCs. Priming approaches, including altering the texture and rigidity of culture surfaces, may influence cytokine release by MSCs ^[77][186]; however, this method has limited scalability. In an effort to create an environment that is analogous to that of the MSC niche, researchers have investigated the use of a variety of 3D based cell culture approaches ^[78][187]. When maintained in a three-dimensional environment, MSCs tend to produce more immunomodulatory factors. Spheroid creation is the most popular approach for MSC 3D cultivation ^[79][188]. Under these conditions, less oxygen may diffuse into the inner layer of cells, creating a hypoxic environment that enhances cell-cell interactions and modifies the release of immunomodulatory molecules. MSCs secreted more TSG6, HGF, and PGE2 when cultured in 3D spheroids; IDO activity and the ability to limit T-cell proliferation were both found to be attenuated when MSCs were cultured in aggregates ^[80][81][189,190]. Hydrogel encapsulation of MSCs is one of the most exciting approaches for producing a 3D-MSC-secretome. Hydrogels permit the change of the mechanical properties such as rigidity and firmness and the inclusion of patterns unique to the natural ECM, both of which increase the secretome’s complexity. Recent attention has been drawn to biopolymer hydrogels due to their capacity to alter the paracrine actions of MSCs ^[82][83][191,192]. The field of cell engineering is expanding fast, making all these methods particularly attractive.

7. Effect of Macrophages on Mesenchymal Stem Cells

Guihard et al. demonstrated that conditioned media from human monocytes activated with LPS or TLR ligands promoted bone formation by human BM-MSCs ^[84][193]. M1 macrophages promote osteogenesis in MSCs via stimulating the COX-2-PGE pathway ^[85][86][87][194,195,196]. Regardless of their polarization status (M0, M1, or M2), human ADSCs can be blocked from transforming into adipocytes in vitro by macrophage derived supernatants ^[88][197]. According to previous studies, M2-type macrophages enhance MSC proliferation and migration, but M1-type macrophages cause MSC apoptosis ^[89][90][198,199]. According to de Witte et al., the phagocytosis of MSCs by monocytes is essential for the immunological regulation of MSCs ^[91][200]. Li et al. discovered that enhanced synthesis of TSG-6 in response to contact with pro-inflammatory macrophages improves MSCs’ inhibitory control of T-cells and macrophages ^[92][88]. Mouse BM-MSCs cocultured with macrophages enhanced IL-10 release in response to LPS stimulation via a PGE2-dependent mechanism. MSCs cannot secrete PGE2 under coculture conditions unless activated by TNF-α and iNOS generated by macrophages ^[93][201]. In response to pro-inflammatory cytokines produced by macrophages, MSCs produce immune modulators such as PGE2 and IL-1RA ^[94][202]. According to the aforementioned research, macrophages produce cytokines that activate MSCs after being activated by pro-inflammatory mediators MSCs respond to the activation of macrophages by modulating the immune response. There is a feedback loop between macrophages and MSCs within the disease microenvironment. MSCs and macrophages work together to keep the inflammatory environment in balance.