Aryl hydrocarbon receptor (AhR)-mediated signaling as a critical regulator of skeletal cell biology

Abstract

The aryl hydrocarbon receptor (AhR) has been implicated in regulating skeletal progenitor cells and the activity of bone forming osteoblasts and bone resorbing osteoclasts, thereby impacting bone mass and the risk of skeletal fractures. The AhR also plays an important role in the immune system within the skeletal niche and the differentiation of mesenchymal stem cells into other cell lineages including chondrocytes and adipocytes. This transcription factor responds to environmental pollutants which can act as AhR ligands, initiating or interfering with various signaling cascades to mediate downstream effects, and also responds to endogenous ligands including tryptophan metabolites. This review comprehensively describes the reported roles of the AhR in skeletal cell biology, focusing on mesenchymal stem cells, osteoblasts, and osteoclasts, and discusses how AhR exhibits sexually dimorphic effects in bone. The molecular mechanisms mediating AhR’s downstream effects are highlighted to emphasize the potential importance of targeting this signaling cascade in skeletal disorders.

1. Introduction

Bone is a critical and specialized organ in vertebrates (Su, Yang et al. 2019) that plays key roles in movement, internal organ protection, hematopoiesis, and storage of minerals. The function of bone is maintained through highly controlled mechanisms that regulate the activity of bone forming cells (osteoblasts) and bone resorbing cells (osteoclasts). Bone modeling describes the process of bone growth achieved through the independent action of osteoblasts and osteoclasts (Langdahl, Ferrari et al. 2016). This process begins during fetal development and is a major contributor to both peak bone mass and overall bone shape (Siddiqui and Partridge 2016). Bone also undergoes constant remodeling activity where osteoblasts and osteoclasts work together to renew, repair, and adapt the tissue as needed in response to various stimuli and signals such as hormones, cytokines, growth factors, and biomechanical forces (Siddiqui and Partridge 2016, Eisa, Reddy et al. 2020). Skeletal diseases can reflect the effects of inadequate balance between bone formation and resorption activity. In osteoporosis, osteoclast-mediated bone resorption becomes unbalanced with osteoblast-mediated bone formation, leading to a net decrease in bone mass and a corresponding increase in risk of bone fracture.

Aging is often associated with bone loss and increased skeletal fragility that adversely impact patients’ quality of life (Kim, Hamrick et al. 2019). While many causative factors for osteoporosis exist, one proposed contributor to bone loss with age is activation of a specialized nuclear hormone receptor called the aryl hydrocarbon receptor (AhR) (Refaey, McGee-Lawrence et al. 2017, Eisa, Reddy et al. 2020). This receptor acts as a transcription factor that responds to environmental pollutants and has been implicated in regulatory mechanisms influencing the immune system, liver homeostasis, and metabolic diseases as well as bone (Wright, De Castro et al. 2017, Eisa, Reddy et al. 2020, Kondrikov, Elmansi et al. 2020). AhR functions through canonical and non-canonical pathways to regulate these biological processes. Canonical signaling involves the binding of AhR to a nucleotide sequence (5’-GCGTG-3’) called the xenobiotic response element (XRE), also referred to as dioxin response elements (DRE), via molecular mechanisms that have been studied and reviewed extensively in several previous papers (McMillan and Bradfield 2007, Beischlag, Luis Morales et al. 2008). However, there are relatively fewer studies describing the process of non-canonical AhR signaling (Jackson, Joshi et al. 2015, Wright, De Castro et al. 2017). The latter pathway has been described after failing to observe a readily identifiable XRE sequence in genes responsive to AhR activation (Huang and Elferink 2012, Wright, De Castro et al. 2017). In fact, AhR has been shown to form complexes with other proteins and interact with a variety of response elements, exhibiting a plethora of effects (Huang and Elferink 2012, Wilson, Joshi et al. 2013, Jackson, Li et al. 2014, Jackson, Joshi et al. 2015, Wright, De Castro et al. 2017).

Tryptophan metabolites such as kynurenine are well-known endogenous ligands to the AhR (Lanis, Alexeev et al. 2017). Since these compounds are polar in nature, they gain access to the intracellular compartment through a transporter known as solute carrier transporter 7a5 (SLC7A5), also referred to as large amino acid transporter 1 (LAT1) (Napolitano, Scalise et al. 2015, Scalise, Galluccio et al. 2018, Sinclair, Neyens et al. 2018). Once inside the cell, these ligands can exhibit their biological effects either through canonical or non-canonical AhR signaling cascades. Because AhR mediates diverse endogenous functions, its effect on the skeletal system have not been fully described. Therefore, understanding how AhR could affect the balance between bone formation and bone resorption activity paves the way for new therapeutics in the field. The purpose of this review is to explore the effects of signaling through AhR in mechanisms of bone cell activity.

2. Bone cells

Mesenchymal stem cells (MSCs) are multipotent progenitor cells that can differentiate into bone-forming osteoblasts (Stein and Lian 1993). MSC undergo an initial proliferation phase where genes involved in cell cycle and cell growth (e.g., c-myc, c-jun), are highly expressed (Stein and Lian 1993). This phase is accompanied by the initial stages of extracellular matrix production through the expression of proteins including type I collagen, transforming growth factor beta (TGF-β), and fibronectin (Stein and Lian 1993). During the early stages of differentiation, these cells begin to express genes associated with an osteoblastic bone cell phenotype including runt-related transcription factor 2 (RUNX2) (Stein and Lian 1993, Komori, Yagi et al. 1997, Korkalainen, Kallio et al. 2009). The expression of type I collagen persists during differentiation, although extracellular matrix composition changes considerably as the cells mature (Assis-Ribas, Forni et al. 2018). Furthermore, during the differentiation phase, proteins related to osteoblast phenotype including alkaline phosphatase, osteocalcin, and osteopontin are synthesized and released in part to promote matrix mineralization (Stein and Lian 1993), providing bone with its unique biochemical properties (Murshed 2018).

Osteoclasts, which resorb bone, are multinucleated cells that arise from hematopoietic stem cells (Boyle, Simonet et al. 2003). These cells solubilize the organic component of bone matrix by secreting proteases such as cathepsin K, matrix metalloproteinases (MMPs), and tartrate resistant acid phosphatase (TRAP) (Boyle, Simonet et al. 2003). They also solubilize the mineral component of bone matrix through the release of acidic hydrogen ions in the form of hydrochloric acid (Roodman 1999). Osteoclast differentiation is initiated by two fundamental molecules; macrophage colony stimulating factor (M-CSF) and receptor activator of nuclear factor kappa-B ligand (RANKL) (Boyle, Simonet et al. 2003). M-CSF, which is produced by a variety of cells including osteoblasts and stromal cells (Mun, Park et al. 2020) drives pre-osteoclast survival, proliferation, and surface expression of RANK (Korkalainen, Kallio et al. 2009). The binding of RANKL to RANK triggers a series of signaling cascades that results in the fusion of pre-osteoclasts to form polykaryons, which then mature to give rise to functional, bone-resorbing osteoclasts (Boyle, Simonet et al. 2003). Osteoblast-lineage cells can also secrete osteoprotegerin (OPG) which serves as a decoy receptor for RANKL, thus blocking RANKL-RANK signaling and suppresses osteoclastogenesis (Boyle, Simonet et al. 2003).

The coupled action of osteoblasts and osteoclasts is key to bone homeostasis. In addition to being regulated by endocrine factors, exogenous factors such as environmental chemicals and pollutants influence bone remodeling balance (Iqbal, Sun et al. 2013). For instance, cigarette smoke is known to impair bone formation and increase osteoclastic bone resorption (Iqbal, Sun et al. 2013). In fact, long term smoking has been associated with elevated risk of osteoporosis and bone fractures (Ward and Klesges 2001, Al-Bashaireh, Haddad et al. 2018). Smoke toxicants contain molecules like 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD; dioxin) that adversely affect the bone (Iqbal, Sun et al. 2013). Interestingly, TCDD mediates its effect at least in part through activating the AhR which in turn drives the expression of genes involved in bone resorption (Iqbal, Sun et al. 2013). The role of AhR in bone health has been investigated by several research groups (Herlin, Finnila et al. 2013, Tong, Niu et al. 2017, Eisa, Reddy et al. 2020), but its exact action remains elusive due to the diversity of cell type specific effects.

3. AhR signaling and regulation

AhR is a ligand-activated transcription factor involved in the response to environmental pollutants and chemicals such as aromatic hydrocarbons (Swedenborg and Pongratz 2010). In addition, it plays a role in a variety of biological processes like drug metabolism (Ramadoss, Marcus et al. 2005), immune regulation (Stevens, Mezrich et al. 2009), and cardiovascular activity (Zhu, Meng et al. 2019). In the absence of a ligand, AhR is present in the cytoplasm in a chaperone complex consisting of heat shock protein 90 (Hsp90), P23, aryl hydrocarbon receptor associated 9 (ARA9), and c-SRC (Carambia and Schuran 2021). This complex ensures that AhR is properly folded and maintained in the correct three-dimensional confirmation that allows ligand interaction (Stevens, Mezrich et al. 2009). Upon agonist binding to AhR, a conformational change is induced which exposes the nuclear localization signal and allows the AhR-chaperone complex to translocate to the nucleus (Stevens, Mezrich et al. 2009). In the nucleus, the chaperone complex disassembles and AhR forms a heterodimer with aryl hydrocarbon receptor nuclear translocator (ARNT) to generate an activated transcription factor (Stevens, Mezrich et al. 2009). Several co-activators can be recruited to the new complex facilitating its binding to an XRE sequence (Stevens, Mezrich et al. 2009). This binding triggers the expression of AhR target genes such as cytochrome P450 family 1 subfamily A member 1 (CYP1A1), and subfamily B member 1 (CYP1B1) (Zhu, Meng et al. 2019). The process described above is referred to as canonical AhR signaling (Figure 1).

Figure 1: Canonical and non-canonical AhR signaling pathways.

Upon binding to an agonist, such as kynurenine (KYN), AhR leaves its chaperone complex and translocates to the nucleus. In the canonical pathway, AhR interacts with ARNT and binds the xenobiotic response element (XRE) nucleotide sequence to drive the expression of target genes including CYP1A1 and CYP1B1. The non-canonical pathway, in contrast, involves the interaction of AhR and KLF6 binding to a non-consensus XRE (NC-XRE) nucleotide sequence to drive expression of target genes including PAI-1 and p21^cip1. Figure created with BioRender.com.

Another pathway for AhR-mediated effects is through non-canonical signaling, where AhR forms a complex with other molecules allowing it to bind to genes lacking XRE sequences (Wright, De Castro et al. 2017) (Figure 1). A non-consensus xenobiotic response element (NC-XRE) was first described by Huang and Elferink in 2012 (Huang and Elferink 2012), consisting of a 5'-GGGA-3’ tetranucleotide repeat. Subsequent studies demonstrated that this site utilizes Kruppel-like factor 6 (KLF6) as a binding partner rather than ARNT (Huang and Elferink 2012, Wilson, Joshi et al. 2013, Jackson, Joshi et al. 2015), and that target genes regulated through this mechanism include PAI-1 (Huang and Elferink 2012) and p21^cip1 (Jackson, Li et al. 2014).

The AhR-ARNT dimer can directly interact with estrogen receptor (ER) to create a transcriptionally active complex that binds the estrogen response element (ERE). (Ohtake, Takeyama et al. 2003, Jackson, Joshi et al. 2015). AhR can also regulate the cell cycle after complexing with S phase proteins including E2F, p300, pRb (Puga, Barnes et al. 2000, Marlowe, Knudsen et al. 2004). Accumulating evidence also demonstrates that AhR directly interacts with RelB to regulate NF-κB target genes (Vogel, Sciullo et al. 2007). In addition to the genomic effects induced by AhR, it can act through non-genomic manner to control biological processes (Carambia and Schuran 2021). For example, ligand binding triggers the dissociation of the chaperone complex and releases c-SRC, allowing c-SRC to act as a protein kinase that phosphorylates downstream proteins (Carambia and Schuran 2021). One effect of this c-SRC activity is increased focal adhesion kinase activation that can lead to increased cellular adhesion and decreased migration (Larigot, Juricek et al. 2018). Ligand binding to AhR can also increase intracellular calcium, which when coupled with its c-SRC activation ability, promotes cyclooxygenase 2 (COX2) and arachidonic acid production leading to inflammation (Larigot, Juricek et al. 2018). In addition, AhR-ligand-binding can promote the degradation of Indolamine-2,3-dioxygenase 1 (IDO1) by generating E3 ubiquitin ligase complex (Pallotta, Fallarino et al. 2014). IDO1 is responsible for the conversion of tryptophan to kynurenine and has been implicated in inflammatory processes (Pallotta, Fallarino et al. 2014).

AhR can be activated by a wide range of endogenous and exogenous ligands (Denison and Nagy 2003), and it is important to note that the downstream signaling mediated by AhR upon binding these molecules is both ligand- and cell-type dependent (Safe, Jin et al. 2020). Smoke toxicants such as TCDD and benzo [a] pyrene (BaP) are common AhR agonists that have been reported to influence the skeletal and immune systems (Table 1) (Iqbal, Sun et al. 2013). However, to our knowledge, less is known about the skeletal effects of non-classical AhR agonists such as thiabendazole (Seidel, Winters et al. 2001, Iqbal, Sun et al. 2013). Furthermore, proton pump inhibitors like omeprazole have been described as AhR agonists in breast cancer cells, but their influence on bone also occurs through non-AhR mediated effects such as blocking acid secretion and hindering calcium absorption (Hyun, Chun et al. 2010, Jin, Lee et al. 2014).

Table 1:

Known AhR ligands affecting the skeletal and immune systems

Compound	Reference
Agonists
TCDD	(Herlin, Finnila et al. 2013, Iqbal, Sun et al. 2013, Tong, Niu et al. 2017, Fader, Nault et al. 2018)
BaP	(Iqbal, Sun et al. 2013, Zhou, Jiang et al. 2017, An, Shi et al. 2020)
Kynurenine	(Kim, Hamrick et al. 2019, Eisa, Reddy et al. 2020)
Kynurenic acid	(DiNatale, Murray et al. 2010)
FICZ	(Oberg, Bergander et al. 2005)
3MC	(Naruse, Otsuka et al. 2004, Yu, Kondo et al. 2014, Yu, Pang et al. 2015)
Tetrandrine	(Jia, Tao et al. 2019)
Norisoboldine	(Wei, Lv et al. 2015, Lv, Wang et al. 2018)
PCB126	(Lee and Yang 2012)
Lipoxin A4	(Schaldach, Riby et al. 1999, Liu, Guan et al. 2017)
Leflunomide	(Liang, Li et al. 2020)
Serotonin	(Nieto, Rayo et al. 2020)
4-Hydroxytamoxifen	(DuSell, Nelson et al. 2010)
Sinomenine	(Tong, Yuan et al. 2016)
Antagonists
BAY 2416964	(Lu, Liu et al. 2022)
CH-223191	(Cui, Feng et al. 2020)
3’4’DMF	(Kido, Fujihara et al. 2014)
GNF351	(Watson, Nordberg et al. 2019)
α-naphthoflavone	(Cedervall, Lind et al. 2015)

AhR activity can be downregulated by negative feedback and proteasomal degradation (Davarinos and Pollenz 1999, Mimura, Ema et al. 1999). When the AhR/ARNT heterodimer binds an XRE, it induces the expression of AhR repressor (AhRR), which is a polypeptide structurally similar to AhR but carries a potent transcriptional repressor domain (Mimura, Ema et al. 1999). AhRR can dimerize with ARNT and reduce AhR transcriptional activity by competing for XRE binding (Mimura, Ema et al. 1999). Similarly, CYP1A1 and CYP1A2 are directly activated by AhR to promote the degradation of AhR ligands and limit the activity of this pathway (Carambia and Schuran 2021). Another mechanism to reduce AhR activity is through proteasomal degradation of the ligand-activated AhR after being exported from the nuclear compartment (Davarinos and Pollenz 1999). In addition to the processes described above, AhR-mediated signaling activity can also be epigenetically regulated via mechanisms including changes in DNA methylation, microRNAs, and histone deacetylases (Gomez-Duran, Ballestar et al. 2008, Amenya, Tohyama et al. 2016, Elmansi, Hussein et al. 2020, Tantoh, Wu et al. 2020, Disner, Lopes-Ferreira et al. 2021).

Growing evidence indicates that AhR signaling interacts with critical cellular pathways including vascular endothelial growth factor (VEGF) (Takeuchi, Takeuchi et al. 2009), WNT (Arinze, Yin et al. 2022), fibroblast growth factor (FGF) (Girer, Murray et al. 2016), TGF-β (Gramatzki, Pantazis et al. 2009), toll-like receptor (TLR) (Kado, Chang et al. 2017), the CXCL12/CXCR4 axis (Elmansi, Hussein et al. 2020), and ER (Vert and Chory 2011) signaling, among others. While a detailed summary of each of these mechanisms is beyond the scope of this review, clearly these pathways are all well-established regulators of myriad MSC, chondrocyte, osteoblast, and osteoclast functions as reviewed elsewhere (Plotkin and Bruzzaniti 2019, Hallett, Ono et al. 2021, Yahara, Nguyen et al. 2022), and accordingly AhR crosstalk with these pathways may drive downstream cellular responses. However, the relative impact of signaling directly mediated by AhR as compared to its modulation of other pathways remains to be elucidated in discrete skeletal cell populations.

4. Role of AhR in MSCs

MSCs constitute a heterogenous population of cells that can differentiate into several mesodermal cell lineages such as chondrocytes, adipocytes, and osteoblasts (Abney and Galipeau 2021). The AhR is strongly expressed in MSCs (Huang, Wang et al. 2022), and many independent reports indicate that AhR-mediated signaling inhibits MSC proliferation (Zhou, Jiang et al. 2017, Jia, Zhao et al. 2021, Huang, Wang et al. 2022). Human bone marrow MSC proliferation was inhibited by AhR ligand kynurenine and stimulated by AhR inhibitor SR1 in a concentration and time-dependent manner (Jia, Zhao et al. 2021); proliferation was also stimulated through siRNA-mediated knockdown of AhR expression, which protected against the inhibitory effects of kynurenine (Jia, Zhao et al. 2021). The cigarette smoke toxicant BaP decreased the proliferation and self-renewal of rat MSCs, comparable to the reduced proliferation rate seen in MSCs isolated from human smokers as compared to non-smokers (Zhou, Jiang et al. 2017). However, some disagreement exists, as the AhR ligand FICZ did not affect rat bone marrow MSC proliferation (Huang, Wang et al. 2022). Other cellular processes relating to cell survival are also affected by AhR signaling in MSC, for instance kynurenine disrupted autophagy and promoted senescence in murine BMSC in an AhR-dependent manner (Kondrikov, Elmansi et al. 2020).

MSCs play a role in a variety of biological processes including inflammation (Abney and Galipeau 2021), and the AhR is implicated in the anti-inflammatory effect of MSCs (de Almeida, Evangelista et al. 2017, Abney and Galipeau 2021). Upon tissue injury, MSCs migrate to the affected site to modulate inflammatory response (Ou-Yang, Huang et al. 2011). Xu et al. demonstrated that AhR knockout mice display heightened lung inflammation following allergen exposure due to impaired migratory potential of MSCs (Xu, Zhou et al. 2015). Mechanistically, after MSCs arrive to the site of injury, they induce a switch in macrophage phenotype from the proinflammatory M1 to the anti-inflammatory M2 (Cui, Feng et al. 2020) and pharmacological inhibition of AhR by CH223191 reduced the polarization of macrophages from M1 to M2 phenotype and exacerbated inflammation (Cui, Feng et al. 2020). Moreover, AhR signaling affected the ability of MSCs to produce soluble cytokines (Jensen, Leeman et al. 2003). This was revealed when an AhR-expressing bone marrow stromal cell line (BMS2) was treated with the AhR agonists 7,12-dimethylbenz [a]anthracene (DMBA) and TCDD, both of which reduced NF-κB levels (Jensen, Leeman et al. 2003). Since the interleukin-6 (IL-6) promoter features NF-κB binding sites, the reduction of NF-κB reduced the expression of IL-6, and lipopolysaccharide-induced increases in IL-6 levels were blunted in the presence of DMBA and TCDD (Jensen, Leeman et al. 2003). In addition, the AhR antagonists DMF and CH-223191 inhibited kynurenine-induced downregulation of gene and protein levels of the inflammatory and multifunctional cytokine CXCL12 in human and murine BMSCs (Elmansi, Hussein et al. 2020). Collectively, these effects indicate that AhR stimulation directly affects the inflammatory responses of MSCs.

The well-regulated expression of transcription factors is critical for MSCs differentiation into different cell lineage including osteoblasts (which will be described in the next section) and adipocytes, and disruptions in the expression of transcription factors can therefore compromise MSCs differentiation capacity. With regards to adipocyte commitment of MSCs, the AhR ligand BaP was administered to human bone marrow derived MSCs and suppressed adipogenic differentiation (Podechard, Fardel et al. 2009) by reducing mRNA expression of CCAAT/enhancer-binding protein beta (CEBPβ) and peroxisome proliferator-activated receptor γ (PPARγ), key transcription factors for adipogenesis (Podechard, Fardel et al. 2009). BaP also repressed expression of adipocyte markers like fatty acid binding protein-4 (Fabp4), whereas the AhR antagonist α-naphthoflavone successfully counteracted BaP’s inhibitory effects on adipogenesis (Podechard, Fardel et al. 2009). A similar study reported that treating canine MSCs with BaP inhibited adipocyte differentiation by activating AhR and downregulating PPARγ (Rathore and Cekanova 2015). Intriguingly, in vivo administration of the AhR ligand TCDD also decreased bone marrow adiposity (Fader, Nault et al. 2018).

5. Role of AhR in osteoblasts

Several studies suggest that activation of AhR by environmental toxicants like TCDD and BaP impedes the differentiation of MSCs into osteoblasts (Figure 2) (Korkalainen, Kallio et al. 2009, Tong, Niu et al. 2017, Zhou, Jiang et al. 2017, Watson, Nordberg et al. 2019). For instance, TCDD-treated MSCs showed a dose dependent decrease in mRNA levels of osteoblastic markers such as RUNX2, osteocalcin and alkaline phosphatase, mediated through inhibiting β-catenin expression (Tong, Niu et al. 2017). Zhou et al. verified that BaP inhibited alkaline phosphatase and matrix mineralization of MSCs and showed that in vitro treatment with BaP inhibited TGF-β1/SMAD4 and TGF-β1/ERK/AKT signaling pathways (Zhou, Jiang et al. 2017). Korkalainen et al. treated mice and rats with TCDD and isolated MSCs from bone marrow of tibia and femur, monitoring the progress of osteoblastic differentiation in these cells (Korkalainen, Kallio et al. 2009). Their quantitative real time-polymerase chain reaction (RT-PCR) results showed reduced mRNA levels of RUNX2, osteocalcin and alkaline phosphatase, suggesting impaired osteoblastic differentiation (Korkalainen, Kallio et al. 2009). To directly test the role of the AhR in these mechanisms, the authors repeated several experiments in AhR-knockout mice and found that the impaired osteoblastic differentiation attributed to TCDD was rescued in the knockout strain (Korkalainen, Kallio et al. 2009). In another study, skeletal changes including decreased cortical thickness and elevated cortical bone porosity were evident after 10 weeks of TCDD treatment (Herlin, Finnila et al. 2013). Bone matrix was harder after TCDD treatment, suggesting increased matrix brittleness in TCDD treated mice comparable to what is seen in older bone with impaired osteogenesis (Herlin, Finnila et al. 2013). A suppressive role of AhR activation on osteoblast differentiation was also evident in human derived MSCs (Watson, Nordberg et al. 2019), where TCDD treatment suppressed alkaline phosphatase activity, matrix mineralization, and expression of the transcription factor DLX5 and the osteogenic markers osteopontin and integrin-binding sialoprotein (Watson, Nordberg et al. 2019). Fibroblast growth factor 9 (FGF9) and FGF18 expression, known to inhibit MSC differentiation into osteoblasts, were upregulated by TCDD treatment (Watson, Nordberg et al. 2019). Blocking AhR with GNF351 attenuated TCDD-induced effects on matrix mineralization and rescued expression of genes related to extracellular matrix, osteogenic regulation and maintenance of multipotency (Watson, Nordberg et al. 2019).

Figure 2: Reported effects of AhR signaling in bone cells.

AhR has been reported to exhibit a diverse array of effects in cells of the MSC and HSC lineages, with many effects occurring in a ligand- and cell type-dependent fashion. Figure created with BioRender.com.

Surprisingly however, at least two studies have shown a net benefit of TCDD treatment on bone (Herlin, Finnila et al. 2013, Fader, Nault et al. 2018), resulting in part from increasing osteoblasts (Fader, Nault et al. 2018) (Figure 2). Likewise, treatment of MC3T3-E1 pre-osteoblasts with the endogenous AhR ligand 6-Formylindolo[3,2-b] carbazole (FICZ) accelerated osteoblastic differentiation and expression of osteoblastic genes including alkaline phosphatase, osteocalcin, and type 1 collagen (Yoshikawa, Izawa et al. 2021). Moreover, in vivo administration of FICZ increased bone formation rate and mineral apposition rate in wildtype mice (Yoshikawa, Izawa et al. 2021). Similarly, a recent report by Huang et al. demonstrated that FICZ treatment promoted osteogenic differentiation of rat bone marrow stromal cells (Huang, Wang et al. 2022). However, kynurenine (another endogenous AhR ligand) inhibited differentiation of BMSC-derived osteoblasts, reducing mineralized matrix production by these cells, and also decreased dynamic indices of bone mineralization activity (Refaey, McGee-Lawrence et al. 2017). These studies highlight the critical point that AhR-mediated signaling can be ligand-dependent, with different ligands capable of producing opposite phenotypic responses both in vivo and in vitro in a cell type-specific manner, as previously reviewed (Safe, Jin et al. 2020) (Figure 2). Many studies have also demonstrated that the effect of AhR can be species dependent (Unkila, Ruotsalainen et al. 1995, Karchner, Franks et al. 2006, Xu, Zhang et al. 2021). One such example is the distinctive sensitivity to TCDD among mammals, fish, toads and birds (Xu, Zhang et al. 2021). Among all vertebrates, the most sensitive species to TCDD are fish, while toads and frogs are considered the least sensitive (Peterson, Theobald et al. 1993, Xu, Zhang et al. 2021). Intraspecies variability is also observed, with 120-fold difference in the response to TCDD between two fish species; bull trout (Salvelinus confluentus) and zebrafish (Danio rerio) (King-Heiden, Mehta et al. 2012). Mammals exhibit differential responses to TCDD as well, for example, rats and hamsters demonstrate teratogenic effects upon exposure to this agent, but this impact is undetectable in guinea pigs (Kransler, McGarrigle et al. 2007).

Discrepancies in the sensitivity of different species to AhR-mediated signaling could be impacted by differing primary structures of the AhR ligand binding domain among species, where small amino acid variations in this domain greatly alter the extent of AhR activation (Xu, Zhang et al. 2021). Supporting this idea is a site directed mutagenesis study demonstrating that TCDD binding affinity was reduced upon replacing Ala375 with Val (Poland, Palen et al. 1994). Changes in the primary structure of AhR affect ligand and coregulator interactions, likely contributing to varied downstream responses (Pandini, Denison et al. 2007). Many different transcription factors and coregulators can bind to AhR depending on its microenvironment (Gargaro, Scalisi et al. 2021). For instance, AhR can either directly interact with signal transducers and activator of transcription (STAT) to regulate proinflammatory responses, or it can form a complex with STAT which then binds to NF-κB, suppressing the transcription of inflammatory mediators (Kimura, Naka et al. 2009). It has also been reported that a ligand-bound AhR/ARNT complex interacts with hypo-phosphorylated retinoblastoma protein which inhibits the cell cycle transition from G1 to S phase (Puga, Barnes et al. 2000). Through this interaction, cell cycle progression and cellular proliferation are slowed down by AhR in humans and rats (Gargaro, Scalisi et al. 2021). Taken together, these findings suggest that different coregulators of AhR present in different species could promote signaling through various AhR pathways to drive the expression of distinct gene sets.

Tissue-targeted genetic models used to directly investigate the role of AhR in osteoblasts are sparsely reported in the literature. AhR floxed mice crossed with α1(I)-Collagen-Cre were developed to investigate the effects of osteoblast-targeted AhR knockout, but surprisingly these mice presented with a bone phenotype that was comparable to their wildtype littermates when examined at young ages (Yu, Kondo et al. 2014). However, it is worth noting that the detrimental effects of the AhR ligand kynurenine on bone mass are attributed to an aging skeletal phenotype (Refaey, McGee-Lawrence et al. 2017), raising the possibility that the role of AhR in osteoblasts is age dependent. It is surprising that, to our knowledge, the role of AhR in osteocytes (which are terminally differentiated osteoblasts) has not yet been directly tested. Given the importance of osteocytes in the regulation of skeletal mechanobiology, mineral homeostasis, and the regulation of bone remodeling activity, such studies are likely warranted to better understand the overall role of AhR in skeletal biology

6. Role of AhR in osteoclasts

Osteoclasts are derived from hematopoietic stem cells, and several reports have found that genetic deletion of AhR promotes HSC proliferation (Singh, Bennett et al. 2014, Bennett, Singh et al. 2015, Unnisa, Singh et al. 2016) (Figure 2). The literature presents conflicting findings for the role for AhR in osteoclasts (Figure 2). Some studies show that AhR stimulation drives osteoclastogenesis and increases bone resorption (Iqbal, Sun et al. 2013, Eisa, Reddy et al. 2020), while others demonstrate that AhR inhibits osteoclastogenesis (Voronov, Heersche et al. 2005, Jia, Tao et al. 2019). Kynurenine is an endogenous metabolite that activates AhR signaling pathway (Opitz, Litzenburger et al. 2011), and in fact, AhR is the only known receptor for kynurenine thus far, responsible for mediating many of kynurenine’s effects (Eisa, Reddy et al. 2020). Therefore, kynurenine-centric studies may be informative regarding understanding AhR’s effects in osteoclasts. However, careful interpretation of these studies is critical as kynurenine has been shown to modulate cell bioenergetics (Pierce, Roberts et al. 2020), and it is not yet known whether this impact is AhR-dependent. A case-control study on elderly patients investigated relationships between bone marrow kynurenine levels, age, and osteoporosis-related phenotypes (Kim, Hamrick et al. 2019). Interestingly, the results revealed that kynurenine accumulates with age, and that patients with fragility hip fractures had 39.7% higher bone barrow kynurenine levels than patients without hip fractures (Kim, Hamrick et al. 2019). Relevant to osteoclasts, the researchers also analyzed bone marrow samples for markers of osteoclast activity like TRAP-5b and RANKL, finding elevated plasma level of these biochemical molecules and reduced bone mass density at the total femur with increasing age (Kim, Hamrick et al. 2019). These results suggest that elevated kynurenine levels (and by extension activating the AhR) could contribute to increased bone resorption and higher fragility observed within the elderly population (Kim, Hamrick et al. 2019). The negative influence of kynurenine on bone density and architecture was also noted in mice fed kynurenine or intraperitoneally injected with the agent (Refaey, McGee-Lawrence et al. 2017). Markers of osteoclastic activity such as RANKL and pyridinoline cross-links (PYD) were elevated in the serum of kynurenine treated mice, and histological markers of osteoclastic activity and bone marrow adiposity were also increased in the treated animals (Refaey, McGee-Lawrence et al. 2017). These results suggested that kynurenine signaling (likely mediated through the AhR) is involved in promoting osteoclastic resorption and skeletal aging.

On a mechanistic level, Eisa et al. investigated AhR’s role in the molecular pathway for kynurenine’s effect in osteoclasts in vitro, showing that kynurenine acted through the AhR pathway to enhance RANKL-dependent osteoclastogenesis and bone resorption (Eisa, Reddy et al. 2020). Cotreatment of Raw 264.7 cells with kynurenine and RANKL increased the mRNA and protein levels of c-fos and NFATc1, which are critical transcription factors that induce osteoclast differentiation (Eisa, Reddy et al. 2020). The number of TRAP positive cells was also elevated after kynurenine/RANKL treatment (Eisa, Reddy et al. 2020). Pharmacological and genetic blockade of AhR pathway attenuated kynurenine’s effect on osteoclasts, which confirmed the involvement of this cascade in osteoclast biology (Eisa, Reddy et al. 2020). The function of AhR in osteoclasts has been explored in vivo by using systemic and tissue specific AhR knockout mice (Yu, Kondo et al. 2014). When AhR was knocked out systemically (body-wide), mice exhibited greater bone mineral density, trabecular bone connectivity density, trabecular bone volume, and trabecular number compared to wild-type mice (Yu, Kondo et al. 2014). Both males and females with whole-body AhR knockout had fewer osteoclasts than wild-type mice with no changes in mineral apposition rate and bone formation rate, suggesting that skeletal effects of body-wide AhR deletion were primarily mediated through reduced bone resorption (Yu, Kondo et al. 2014). AhR floxed mice were then crossed with Cathepsin K-Cre mice to drive AhR conditional knockout in cells of the osteoclast lineage (Yu, Kondo et al. 2014), which showed that osteoclast-targeted knockout mice had higher bone mass with lower bone resorption (Yu, Kondo et al. 2014).

There is also contrasting evidence in the literature suggesting that AhR signaling inhibits osteoclast activity. For example, at least one report demonstrates that TCDD treatment of juvenile rodents reduced osteoclastic bone resorption (Fader, Nault et al. 2018). Through stimulating the AhR pathway, 3-methylcholanthrene (3MC) repressed the differentiation of osteoclasts via a reducing expression of the pro-osteoclastogenic molecule RANKL expression in osteoblast-lineage ST2 progenitor cells (Naruse, Otsuka et al. 2004). These studies suggested that the AhR pathway might reduce osteoclast differentiation indirectly by affecting osteoclast supporting cells rather than osteoclast precursors (Naruse, Otsuka et al. 2004). However, literature support also exists for the idea that AhR-mediated signaling can directly inhibit osteoclasts. For example, the endogenous AhR ligand FICZ dose-dependently inhibited actin ring formation and pit formation by bone marrow macrophage-derived osteoclasts (Yoshikawa, Izawa et al. 2021). Voronov et al. demonstrated that the AhR agonist BaP also inhibited osteoclast differentiation and bone resorption (Voronov, Heersche et al. 2005, Voronov, Li et al. 2008), attributed to crosstalk between AhR and NF-kB pathways (Voronov, Li et al. 2008). NF-kB is a transcription factor that resides in the cytoplasm in complex with IkBα, p65 and p50 (Oeckinghaus and Ghosh 2009). When NF-kB pathway is stimulated, IKK phosphorylates IkBα and targets it for degradation, allowing NF-kB to translocate to the nucleus and drive gene expression (Oeckinghaus and Ghosh 2009). NF-kB is a common transcription factor for AhR and RANKL signaling cascades, therefore, stimulating AhR creates a competition for NF-kB and limits RANKL induced osteoclastogenesis (Voronov, Li et al. 2008). Although the exact mechanism for the crosstalk between AhR and RANKL is not clearly understood, it was proposed that AhR can complex with p65 and IkBα in the cytoplasm (Voronov, Li et al. 2008). Upon AhR stimulation, AhR could act as a ligand-dependent E3 ubiquitin ligase and activate the IKK complex, driving the translocation of NF-kB to the nucleus and subsequent gene expression (Voronov, Li et al. 2008). All together, these events reduce the availability of NF-kB for RANKL mediated signaling, limiting osteoclastogenesis (Voronov, Li et al. 2008).

7. Role of AhR in cartilage and endochondral ossification

Endochondral bone development begins with a cartilaginous template which osteoblasts and osteoclasts then replace with bone tissue. This process is regulated by various endocrine factors including growth hormone and insulin-like growth factor-1 (IGF-1) (Cedervall, Lind et al. 2015). The AhR has been reported to play a role in cartilage development and subsequent bone growth. For example, AhR expression was suppressed in cartilage templates, but activation of AhR by the endogenous ligand FICZ positively promoted later processes of endochondral ossification (Huang, Wang et al. 2022). Cedervall et al. investigated how AhR acts in growth plate chondrocytes to regulate bone elongation, beginning with the finding that AhR was broadly expressed in the growth plate cartilage of human subjects and more highly expressed in hypertrophic as compared to resting chondrocytes (Cedervall, Lind et al. 2015). For mechanistic studies, fetal rat metatarsal bones were cultured and monitored for longitudinal growth after exposure to AhR modulators (Cedervall, Lind et al. 2015). TCDD did not affect bone growth at the concentrations tested (1 pM −10 nM), but higher dose of the AhR antagonist, α-naphthoflavone, increased chondrocyte apoptosis and suppressed bone growth (Cedervall, Lind et al. 2015). It is worth noting that at lower doses (10 pM and 10 nM), α-naphthoflavone did not affect bone growth and cytotoxic effects may have occurred through AhR independent mechanisms (Cedervall, Lind et al. 2015). The authors therefore concluded that AhR activation may not directly affect endochondral bone growth, and that AhR activation in the rat growth plate may require an unknown co-factor to affect cartilage formation (Cedervall, Lind et al. 2015). Yang et al. demonstrated that rabbit chondrocytes were exposed to TCDD at a concentration of 10 nM developed elevated levels of reactive oxygen species and nitric oxide (Yang and Lee 2010). This created cellular stress which eventually triggered chondrocyte apoptosis (Yang and Lee 2010). Inhibiting AhR using 10 μM of α-naphthoflavone attenuated these harmful effects of TCDD (Yang and Lee 2010), raising the possibility that apoptosis of chondrocytes driven by AhR activation could lead to cartilage damage and arthritis development (Yang and Lee 2010). In a parallel study, treating Japanese rice fish (medaka) with TCDD (ppt-ppb concentrations) impaired chondrogenesis and osteogenesis (Dong, Hinton et al. 2012). The recruitment of MSCs to hypural cartilage anlage was attenuated following TCDD treatment, and chondrocyte proliferation and differentiation were also impaired in a dose-dependent manner (Dong, Hinton et al. 2012). On a molecular level, the expression of type 2 collagen (a major cartilage extracellular matrix protein) was markedly blunted after TCDD treatment (Dong, Hinton et al. 2012). This effect was mediated through downregulating the key chondrogenic transcription factor, SRY-box transcription factor 9 (SOX9) (Dong, Hinton et al. 2012). With regards to subsequent bone formation, TCDD treatment attenuated Osterix expression and reduced perichondral ossification within the fish hypural anlage (Dong, Hinton et al. 2012). Mechanisms of endochondral ossification are initiated postnatally during fracture repair, and, when administered to rats with tibial fractures, the AhR ligand BaP delayed healing and resulted in a less mineralized callus (Zhou, Jiang et al. 2017). Together, these results imply that AhR activation may negatively influence cartilage and endochondral bone formation.

In addition to AhR’s role in cartilage during developmental and repair-mediated processes of endochondral ossification, AhR may also play a role in diseases affecting cartilage like osteoarthritis. Expression of AhR was up-regulated in human osteoarthritic cartilage as compared to intact controls (Klinger, Beyer et al. 2013), and was also highly expressed in meniscus and subchondral bone from osteoarthritic patients (Chang, Yao et al. 2021). The AhR signaling pathway was reported to be more active in osteoarthritic synovial tissues as compared to those from rheumatoid arthritis (Ogando, Tardaguila et al. 2016). As with other cell types described above, the effects of AhR-mediated signaling in chondrocytes with respect to osteoarthritis phenotypes are likely ligand dependent. Exposure to the exogenous AhR ligand BaP led to the development of osteoarthritis-like lesions, loss of proteoglycans, and reduced expression of aggrecan, type 2 collagen, and Sox-9 in the articular cartilage of wildtype (but not AhR-knockout) mice (Yoshikawa, Izawa et al. 2021). Interestingly, however, administration of the endogenous AhR ligand FICZ was protective against the development of mechanically induced osteoarthritis in mice (Yoshikawa, Izawa et al. 2021). Mechanistically, BaP promoted expression of the active forms of caspase-3 and caspase-9 in ATCD5 chondrocyte progenitor cells, and increased expression of cleaved caspase-3 in the mandibular chondrocytes of wildtype (but not AhR-knockout) mice, but administration of FICZ reduced metrics of apoptosis (Yoshikawa, Izawa et al. 2021). The beneficial effects of FICZ described above are consistent with a recent bioRxiv report, where intra-articular delivery of an indoleamine 2,3-dioxygenase and galectin-3 fusion protein (IDO-Gal3), intended to drive local expression of the endogenous AhR ligand kynurenine, improved functional outcomes in a rat surgical model of osteoarthritis (Partain, Bracho-Sanchez et al. 2021).

8. Sexually dimorphic effects of AhR in bone

Crosstalk between signaling cascades is a common phenomenon driving biological events (Vert and Chory 2011). Some evidence of sexually dimorphic effects of AhR have been reported in bone, which may be due in part to the interaction between AhR and estrogen receptors (ER) (Wejheden, Brunnberg et al. 2010, Tarnow, Bross et al. 2017). Wejhedem et al. designed a study to explore the long-term effect of exposing mice to AhR ligands (Wejheden, Brunnberg et al. 2010). As their experimental model, they studied three-month-old transgenic mice expressing constitutively active AhR. Results revealed that continuous AhR stimulation reduced bone formation in transgenic female mice but not in males (Wejheden, Brunnberg et al. 2010). Bone dimensions, density, and content were notably altered in female mice, while males were minimally affected (Wejheden, Brunnberg et al. 2010). Osteoclastic markers such as carboxy-terminal cross-linked telopeptide of type 1 collagen (CTX-1) and osteoclast volume density were increased by more than 60% and the resorption index of CTX-1/TRAP 5b was increased by 90% in females but not in males (Wejheden, Brunnberg et al. 2010). On an mRNA level, female but not male mice demonstrated altered expression of osteoclastic and osteoblastic markers in these studies (Wejheden, Brunnberg et al. 2010). Similarly, treatment with the endogenous AhR ligand kynurenine was reported to decrease osteoblasts in young female but not young male mice, and likewise impaired cellular energetics in BMSC-derived osteoblasts from females but not males (Pierce, Roberts et al. 2020). Another report demonstrated that TCDD treatment exhibited gender dependent effect in C57BL/6J mice (Herlin, Finnila et al. 2013). For instance, bone biomechanical properties were decreased in females following TCDD treatment, but males were not affected (Herlin, Finnila et al. 2013). Also, unlike males, TCDD treated females demonstrated unbalanced bone remodeling activity as indicated by lower PINP/CTX-1 ratio (Herlin, Finnila et al. 2013). In the same report, bone phenotype of both genders was compared between AhR knockout mice and wild-type mice, showing that AhR deletion resulted in higher trabecular number and trabecular bone volume fraction in both males and females (Herlin, Finnila et al. 2013). However, female knockout mice had lower cortical area, cortical thickness and bone mineral density compared to wild-type females (Herlin, Finnila et al. 2013), whereas male knockout mice displayed lower plasticity and harder cortical bone matrix than their wild-type controls (Herlin, Finnila et al. 2013). In another report, eight-week-old osteoclast-targeted AhR knockout mice were orchiectomized or ovariectomized then monitored for four weeks, showing that the osteoclast-targeted AhR knockout mice were resistant to bone loss induced by gonadectomy (Yu, Kondo et al. 2014). This implied that sex hormones might influence bone remodeling through AhR.

Multiple mechanisms have been proposed for the crosstalk between AhR and ER (Tarnow, Bross et al. 2017). Firstly, AhR could directly interact with ERα to form a complex which binds to AhR target genes (Matthews and Gustafsson 2006). Alternatively, the complex might bind to ERE and inhibit ER target gene expression (Jackson, Joshi et al. 2015) . The competition for shared co-factors between the two signaling cascades also influences the outcome of AhR activation (Swedenborg and Pongratz 2010). Moreover, AhR could increase proteasomal degradation of ER through upregulating CYP expression and estrogen metabolism (Wormke, Stoner et al. 2003). Estrogen has also been reported to alter kynurenine pathway by inhibiting kynurenine aminotransferase enzymes (Jayawickrama, Nematollahi et al. 2017). All together, these mechanisms suggest that AhR stimulation might exert antiestrogenic effects in bone (Wejheden, Brunnberg et al. 2010). Since estrogen loss usually increases bone resorption, the crosstalk between AhR and ER could contribute to the higher sensitivity of female mice to AhR activation (Wejheden, Brunnberg et al. 2010).

9. Conclusions

Taken together, the literature suggests that signaling through AhR likely plays an important role in skeletal homeostasis. Several reports have shown that activating AhR reduces osteoblast differentiation and activity (Korkalainen, Kallio et al. 2009, Tong, Niu et al. 2017, Zhou, Jiang et al. 2017, Watson, Nordberg et al. 2019), and that treating MSCs with AhR agonists suppressed osteoblastic markers (Siddiqui and Partridge 2016, Refaey, McGee-Lawrence et al. 2017, Tong, Niu et al. 2017). Additionally, AhR stimulation increased bone porosity and reduced matrix mineralization in vivo (Herlin, Finnila et al. 2013). However, as mentioned above, some AhR ligands like FICZ appear to have a beneficial effect on osteoblasts (Yoshikawa, Izawa et al. 2021). Similarly, the role of AhR in osteoclasts is still unclear, as some support exists for the idea that AhR activation drives osteoclastogenesis (Iqbal, Sun et al. 2013, Kim, Hamrick et al. 2019, Eisa, Reddy et al. 2020), whereas other research groups showed that AhR stimulation inhibits osteoclastogenesis (Naruse, Otsuka et al. 2004, Voronov, Li et al. 2008). The inconsistent findings suggest that the resultant effect of stimulating AhR likely depends on the specific ligand used, the ligand’s concentration, and duration of treatment (Park, Madhavaram et al. 2020, Safe, Jin et al. 2020). Indeed, these factors could act through different AhR signaling pathways (e.g., canonical, non-canonical, or non-genomic) to influence distinct downstream signaling proteins (Eisa, Reddy et al. 2020).

While the effect of AhR on muscle is beyond the scope of this review, the importance of bone-muscle crosstalk has recently come to light and is an area of increasing interest; in this context, it is important to note that AhR activation can also lead to muscle atrophy (Kaiser, Yu et al. 2019, He, He et al. 2020, Thome, Miguez et al. 2022). In fact, a recent publication from Thome et al demonstrated that muscle atrophy due to overexpression of AhR could be rescued via AhR antagonism (Thome, Miguez et al. 2022). However, the role of AhR in mechanisms of bone-muscle crosstalk has yet to be clearly defined, and since AhR modulation exhibits context-dependent effects, therapeutically targeting this pathway to combat musculoskeletal frailty in aging might require personalizing the treatment plan according to a variety of patient-specific and environmental factors such as gender, age, and lifestyle habits.

While much has been learned about the role of AhR in the skeleton, much remains to be explored. Many of the studies described above have utilized pharmacological inhibitors of AhR that are likely to have off-targeted effects, and many of the ligand-based studies have employed environmental pollutant AhR ligands like TCDD and BaP. To address this issue, the specific role of AhR in bone cells should be tested in future in vitro studies employing knockdown mediated by sequence-specific small interfering RNA (siRNA), small hairpin RNA (shRNA), or CRISPR/Cas9-mediated knockouts of AhR, along with creation of novel cell type-specific in vivo knockout mouse models of the AhR. Moreover, it will be critical to ascertain the contributions of endogenous AhR ligands (like kynurenine and other tryptophan metabolites) to skeletal biology, particularly in the context of aging. Although the literature demonstrates that kynurenine activates AhR, the precise mechanism is not yet understood. For example, it is not clear whether kynurenine and/or other tryptophan metabolites activate canonical, non-canonical or non-genomic modes of AhR signaling, and the specific signaling molecules and enzymes involved in the effect of kynurenine and other tryptophan metabolites on bone cells are not well defined. It has been recognized that the AhR cascade undergoes important crosstalk with other nuclear hormone receptors such as ER and the glucocorticoid receptor (Widerak, Ghoneim et al. 2006, Monostory, Pascussi et al. 2009, Denison, Soshilov et al. 2011). Among those, the crosstalk between AhR and ER is the best understood, while other interactions are less studied (Denison, Soshilov et al. 2011). Future studies focused on elucidating such molecular interactions would foster better understanding of the role of AhR in musculoskeletal disorders, ultimately promoting the discovery of new, effective therapeutic skeletal agents

Abbreviations

3’4’DMF

3',4'-dimethoxyflavone

3MC

3-methylcholanthrene

AhR

Aryl hydrocarbon receptor

AHRR

Aryl hydrocarbon receptor repressor

ARA9

Aryl hydrocarbon receptor associated 9

ARNT

Aryl hydrocarbon receptor nuclear translocator

BaP

Benzo(a)pyrene

CEBPβ

CCAAT/enhancer-binding protein beta

COX2

Cyclooxygenase 2

CTX-1

Carboxy-terminal cross-linked telopeptide of type 1 collagen

CYP1A1

Cytochrome P450 family 1 subfamily A member 1

CYP1B1

Cytochrome P450 family 1 subfamily B member 1

DMBA

7,12-dimethylbenz [a]anthracene

ER

Estrogen receptor

ERE

Estrogen response element

FGF9

Fibroblast growth factor 9

FICZ

6-Formylindolo[3,2-b] carbazole

HSC

Hematopoietic stem cells

Hsp90

Heat shock protein 90

IDO

Indolamine-2,3-dioxygenase

IGF-1

Insulin-like growth factor-1

IL-6

Interleukin-6

KLF6

Kruppel-like factor 6

KYN

Kynurenine

LAT1

Large amino acid transporter 1

m-CSF

Macrophage colony stimulating factor

MMP

Matrix metalloproteinases

MSC

Mesenchymal stem cells

NC-XRE

Non-consensus xenobiotic response element

OPG

Osteoprotegerin

PCB126

3,3′,4,4′,5-Pentachlorobiphenyl

PPARγ

Peroxisome proliferator-activated receptor γ

RANKL

Receptor activator of nuclear factor kappa-B ligand

RUNX2

Runt-related transcription factor 2

SLC7A5

Solute carrier transporter 7a5

SOX9

SRY-Box Transcription Factor 9

STAT

Signal transducers and activator of transcription

TCDD

2,3,7,8-tetrachlorodibenzo-p-dioxin

TGF-β

Transforming growth factor-β

TLR

Toll-like receptor

TRAP

Tartrate resistant acid phosphatase

Trp

Tryptophan

VEGF

Vascular endothelial growth factor

XRE

Xenobiotic response element