Bone is a complex connective tissue which constitutes the skeletal system of the body and is primarily involved in the structural functions and basic movements of the body. Bone also regulates metabolic functions of the body by acting as a primary site for hematopoiesis. Being a site for hematopoiesis, it is in reciprocal interaction with different immune cells (Lorenzo et al., 2008). This interaction of bone and immune cells is termed as ‘osteoimmunology’ and plays critical role in bone metabolism as well as bone formation during bone defect repair (Ono and Takayanagi, 2017). Bone constantly undergoes the process of remodeling in which the old woven bone is replaced by new bone by a complex interplay between bone resorbing cells osteoclasts and bone forming cells osteoblasts. As such, bone possesses a unique ability to regenerate itself, especially in events of microfracture (Clarke, 2008). Severe tissue loss and degeneration due to road accidents, excessive sports practice and certain genetic abnormalities compromise the self-regenerative ability of the bone, and limit the structural and functional recovery of the tissue. Thus, although most of the bone defects can be healed spontaneously, 5–10% are complicated by delayed union and non-union, and require external fixation and surgical aid. The past decade has seen substantial advancement in the field of bone tissue engineering which involves the use of undifferentiated or osteogenically differentiated mesenchymal stem cells (MSCs) along with a suitable scaffold to treat massive bone defects (Amini et al., 2012).

Mesenchymal stem cells (MSCs) are mesoderm derived multipotent cells with the ability to self-renew and differentiate into multiple lineages. First identified in the bone marrow in late 1960s, MSCs have been isolated from a wide array of sources which include adipose tissue, dental tissue, synovial membrane, menstrual blood, placenta and amniotic fluid (Friedenstein et al., 1968, Brown et al., 2019, Berebichez-Fridman and Montero-Olvera, 2018). MSCs have been shown to differentiate into not only cells of mesodermal lineage such as osteoblasts, chondrocytes and adipocytes but also into cells of ectodermal (such as neurons) and endodermal (such as hepatocytes, islet cells) lineages (Kode et al., 2009). MSCs do not express class II MHC molecules and any co-stimulatory molecules such as B7–1, B7–2, CD80, CD86, CD40 and CD40L, and thus do not trigger any immune response and are considered as immunologically inert (Klyushnenkova et al., 2005). This property of MSCs has allowed them to be transplanted even in allogenic settings. MSCs have found potential applications in cell replacement and tissue regeneration therapies including bone regeneration. Following a bone fracture, various immune cells are recruited to the fractured site which secrete pro-inflammatory cytokines resulting in an inflammatory cascade (Dimitriou et al., 2005). Immune cells such as macrophages also secrete various chemotactic signals such as SDF-1, C1q, PDGF, bFGF, CCL5, TNF-α, IL-1β and certain TLR (toll-like receptors) ligands to name a few (Kallmeyer and Pepper, 2015, English, 2013). Both tissue resident and adoptively transferred stem cells are homed towards the inflamed bone tissue region in response to these signals. MSCs initiate bone repair by various mechanisms including secretion of anti-inflammatory molecules, in vivo differentiation into osteoblasts, recruitment of osteoprogenitor cells to the site of injury and secretion of alkaline phosphatase and other bioactive factors for synthesis and mineralization of bone matrix (Ma et al., 2023). Additionally, the interplay between donor MSCs and host immune cells at the bone defect site also plays a substantial role in determining the overall success of bone repair. While undifferentiated MSCs are known to be immunomodulatory, these undergo osteogenic differentiation during fracture repair and therefore, it is critical to determine the nature of interactions between MSCs derived osteoblasts and immune cells. Accordingly, several studies have demonstrated the immune-regulatory effects of MSCs derived osteoblasts and other lineage cells. The present review summarizes the current literature available on immunomodulation by osteogenically differentiated MSCs as well as the role of this immunomodulation in repairing bone defects.

While MSCs were discovered in 1960 s, their immunomodulatory capability was recognized almost 30 years later when it was reported that MSCs could inhibit T-cell proliferation in vitro and also prolong skin graft survival in vivo (Bartholomew et al., 2002). Since then, MSCs have also emerged as promising therapeutic options for autoimmune disorders, sepsis and transplant surgeries. MSCs influence both innate and adaptive immune response by interfering with complement components, TLRs, monocytes/macrophages, dendritic cells (DCs), natural killer (NK) cells, and lymphocytes particularly T-helper (Th) cells and regulatory T-cells (Tregs) (Fig. 1) (English, 2013). What’s interesting is that the immune-regulatory functions of MSCs are retained even if they are metabolically inactivated or fragmented (apoptosed MSCs) (Weiss and Dahlke, 2019). It is now evident that endogenous as well as exogenously transplanted MSCs can home to the site of inflamed tissue in response to the biochemical signals produced locally by the tissue (Lin et al., 2017a). For example, MSCs can respond to complement component C1q, C3a and C5a, IL1α/1β, SDF-1, PDGFA, PDGFB, EGF, HGF, TNFα and certain TLR ligands which are present in the microenvironment of injured or inflamed tissue. These signaling molecules allow MSCs to become polarized towards anti-inflammatory phenotype (English, 2013). It should be noted that MSCs require prior activation or licensing to become immunosuppressive in nature. Although MSCs require direct contact with T-cells to exert their effects, a few studies have also shown that MSCs can also act from a distance. For example, a study showed that MSCs entrapped in lungs still managed to suppress early immune response in case of corneal injury (Roddy et al., 2011). This distant action of MSCs is mediated by secretion of various immunosuppressive molecules such as IDO, TSG-6, PGE-2, TGF-β and NO (Ren et al., 2008, English et al., 2007). For contact mediated immunosuppression, MSCs secrete various chemotactic factors such as CXCL-9, CXCL-10, CCL-2 which recruit effector T cells. MSCs then result in T-cell inhibition by NO or FAS/FASL induced apoptosis (Ren et al., 2008, Akiyama et al., 2012). In addition to directly inhibiting T-cells, MSCs also induce immune tolerance by altering the balance between Th1/Th2/Th17 subtypes, inducing polarization of M1 macrophages (pro-inflammatory phenotype) to M2 macrophages (anti-inflammatory phenotype), inhibiting DC maturation, inducing tolerogenic DC phenotype and promoting Tregs activation and function (Gao et al., 2016). A summary of MSCs effect on immune cells is presented in Table 1.

Although the data which support that MSCs possess anti-inflammatory phenotype and can inhibit immune cell function is quite overwhelming, there are several contrasting reports which suggest otherwise. For example, in case of pathogenic or sterile inflammation, it is logical to presume that MSCs orchestrate the clearance of foreign or necrotic cells by supporting survival and function of immune cells. In this context, Thakur et al. (2013) showed that MSCs reduce anti-inflammatory cytokine production and accumulation of T-regs in case of malarial infection. One of the ways MSCs carry out these functions is by responding to local biochemical signals which alter their expression of MHC molecules and co-stimulatory signals. Although MHC class II molecules are not expressed by MSCs in general, the expression can be altered depending upon the physical location and chemical cues present in the microenvironment (Yagi et al., 2010). In this context, several studies have elucidated the factors governing the immune regulatory phenotype of MSCs. In particular, IFNγ plays a crucial role in regulating the immune switch in MSCs as well as MHC expression kinetics (Chan et al., 2008). Previous studies have shown that low levels of IFNγ result in increased expression of MHC II molecules and reduced expression of negative co-stimulatory molecule PD-L1 on MSCs surface while high IFNγ levels are associated with decreased MHC II and increased PD-L1 expression (Sheng et al., 2008, Chan et al., 2006). When MSCs encounter any pathogenic antigen, increased MHC II expression allows these cells to act as antigen presenting cells. Antigen presentation to T-cells results in their activation. As activated T-cells produce more IFNγ, MHC II expression is reduced and PD-L1 expression is enhanced on MSCs surface which result in inhibition of T-cells (Sheng et al., 2008). In addition to IFNγ, a few studies have suggested the role of TLR signaling in regulating the immune modulation response of MSCs. Activation of TLR4 signaling in MSCs induces a pro-inflammatory phenotype and secretion of IL-6 and IL-8 which can be enhanced by stimulation with IFNγ (English, 2013, Romieu-Mourez et al., 2009). Moreover, TLR4 activated MSCs have also been shown to activate T-lymphocyte activation in vitro (Waterman et al., 2010). While MSCs have been shown to directly modulate immune cell functions, the evidence for direct modulation of MSCs by immune cells is still lacking. It is proposed that immune cells exert their effects on MSCs by secretion of various cytokines, particularly IFN-γ. IFN-γ acts synergistically with other cytokines like TNFα and IL-1ꞵ to modulate MSCs immunoregulatory function. Table 2 presents a summary of effects of immune cytokines on MSCs. Therefore, it can be concluded that there exists bidirectional interaction between MSCs and immune cells/cytokines which determine their immunostimulatory or immunosuppressive phenotype and the outcome of MSCs mediated repair of tissue injury (Yagi et al., 2010).

The differentiation of MSCs to osteoblasts is the most critical part of fracture repair and bone formation. The process of osteogenic differentiation of MSCs is regulated by two key pathways: BMP signaling and Wnt signaling. BMPs induce osteogenesis by activating Smad proteins and RUNX2 transcription factor which is the master regulator of transcription of various osteogenic genes (Cao and Chen, 2005). The canonical Wnt/β-catenin signaling can be both osteostimulatory and osteosuppressive. For osteostimulation, it suppresses PPARγ and CCAAT/enhancer binding protein alpha expression to inhibit adipogenesis while inducing the expression of Runx2, Dlx5 and Osterix (Kim et al., 2013). Immune cells and their cytokines regulate osteoblast formation generally by these two pathways. Recently it was shown that immune cells act via soluble factors to modulate MSCs osteogenic differentiation (Khokhani et al., 2022). The study demonstrated that stimulation of human peripheral blood mononuclear cells with immune modulatory agents resulted in increased ALP activity and osteogenesis in MSCs. Macrophages are the most studied and well described immune cells for their effect on MSCs to drive osteogenesis (Fig. 2). It has been suggested that all phenotypes of macrophages upregulate ALP activity, matrix mineralization and osteogenic differentiation of MSCs (Chen et al., 2022). Macrophages produce BMP-2, BMP-6 and TGFβ1 to stimulate osteogenesis specific genes such as Runx2 and ALP (Champagne et al., 2002). Guihard et al. (2012) showed that unpolarized as well as activated M1 and M2 macrophages produce Oncostatin M (OSM) which induce osteoblast differentiation program by activating the expression of Cbfa1 and ALP in MSCs. OSM is reported to have pro-osteogenic and anti-adipogenic effects by activating STAT3, JAK2, JAK3 and MEK-ERK pathways in MSCs (Chen et al., 2022, Song et al., 2007). As M1 macrophages are pro-inflammatory, several studies have also indicated their anti-osteogenic role although these have a more defined ‘osteoclastogenic’ role. M1 macrophages secrete TNFα, IL-1β and IFNγ which inhibit Runx2 and ALP resulting in decreased osteogenesis (Gong et al., 2016). M1 macrophage inhibition of osteogenesis has also been implicated as an indirect cause of bone loss in osteoporosis (Wang et al., 2022). M1 macrophages are mainly associated with secretion of different pro-inflammatory cytokines which are shown to affect osteoblast formation in a concentration dependent manner. TNFα and IL-1β are key molecules secreted by M1 macrophages which affect osteogenic differentiation of MSCs via different signaling mechanisms. High levels of TNFα activate Wnt/β-catenin pathway which cause downregulation of Runx2, Osx, ALP, OPN and OCN, thus inhibiting osteogenic differentiation of MSCs (Qin et al., 2015). TNFα also activates NFκB which affects osteogenic differentiation by inhibiting BMP/Smad pathway or promoting β-catenin degradation via induction of Smurf1 and Smurf2 (Yamazaki et al., 2009, Chang et al., 2013). The anti-osteogenic effect of TNFα is also implicated in postmenopausal osteoporosis where it inhibits osteogenic differentiation by upregulating P2Y2 receptor via ERK and JNK signaling (Du et al., 2018). On the contrary, lower doses of TNFα can result in positive regulation of osteogenic differentiation. Low level of TNFα results in IKK2/NFκB induced BMP2 activation which in turn enhances Runx2 and Osx expression, ALP secretion and matrix mineralization (Mo et al., 2022, Hess et al., 2009). Increasing evidence shows that IL-1β also exerts both positive and negative effects on osteogenic differentiation of MSCs. IL-1β downregulates Wnt/β-catenin signaling via FoxD3/miR496 and reduces the expression of Runx2, ALP, Osx and Ocn as well as formation of mineralization nodules (Huang and Chen, 2017). However, it can also promote osteogenic differentiation by activating BMP/Smad signaling at lower doses (Wang et al., 2020).

While M1 macrophages and pro-inflammatory cytokines seem to have dual role in osteogenic differentiation, the positive osteogenic effect of M2 phenotype is unquestionable. M2 macrophages directly promote osteogenic differentiation by activating Smad1 signaling and secreting BMP-2 and VEGF to increase osteoblast formation and angiogenesis which have synergistic effect on bone healing (Chen et al., 2022). It is possible that M1 macrophages are required for recruitment of MSCs as well as osteoprogenitor cells and their differentiation into osteoblasts at fracture healing sites. A subsequent switch from M1 to M2 macrophage phenotype supports the later stages of bone repair (Pajarinen et al., 2019). In addition to macrophages, B and T cells also affect osteogenic differentiation, although their role cannot be generalized as these lymphocytes consist of different subtypes which have differing role in osteoblast formation.