Since the role of AhR in immune cells has been extensively summarized in previous review articles, we will briefly describe the major functions of AhR in key immune cells here [2
]. The immunomodulatory effects of AhR occur primarily through effector T cells and regulatory T cells (Treg) via direct and indirect mechanisms such as modulation of antigen-presenting-cell (APC) activity. AhR activation leads to an increase in FoxP3+ Treg cells via various mechanisms, including modulation of FoxP3 expression and direct transactivation [5
]. Furthermore, AhR activation appears to influence the differentiation of a subset of precursor Th17 cells into either a regulatory T cell or a fully Th17 cell phenotype [7
]. More importantly, in conventional dendritic cells (DC), the consequences of AhR activation are more nuanced with ligand-dependent effects. For example, in the context of antigen presentation, the AhR agonist 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) decreases CD11c expression and increases MHC-II, CD86, IL-6, and TNF-κ in murine DCs [8
]. On the other hand, when murine DCs were treated with the AhR agonist 2-(1′H-indole-3′-carbonyl)-thiazole-4-carboxylic acid methyl ester (ITE), MHC-II and costimulatory molecules as well as Th1 and Th17 polarizing cytokine production were decreased [10
]. Furthermore, AhR activation inhibits the differentiation of human monocytes into Langerhans DCs in vitro, while AhR inhibition via StemRegenin 1 induces human CD34+ stem cell progenitor differentiation into mature DCs [11
]. Therefore, AhR activation has context-dependent effects on the function and differentiation of DCs.
Aryl-hydrocarbon receptor (AhR) is a ligand-activated transcription factor of the Pern-Arnt-Sim (PAS) superfamily initially implicated in xenobiotic metabolism of environmental pollutants; however, recent studies have also elucidated the role of AhR in immune regulation and bone remodeling [1
Tobacco smoke, which is a well-established risk factor in bone-remodeling disorders, including osteoporosis, contains numerous environmental toxins, including polycyclic aryl hydrocarbons (PAH) such as benzo[a]pyrene (BaP) and dioxins such as TCDD; importantly, both BaP and TCDD are well-established AhR agonists [12
Following activation of AhR, intracellular signaling occurs via either the genomic or non-genomic pathway. At steady state, AhR remains in a multi-protein complex that include heat shock protein 90 (HSP90) and c-src protein kinase localized in the cytoplasm. Upon ligand-binding, AhR undergoes nuclear translocation, forms a complex with AhR nuclear translocator (ARNT), then exerts transcriptional control over target genes harboring dioxin-response elements (DRE), or xenobiotic-responsive elements (XRE) such as the cytochrome P450 family 1 subfamily A member 1 (Cyp1a1) and cytochrome P450 family 1 subfamily B member 1 (Cyp1b1) [13
]. On the other hand, non-genomic AhR signaling pathways may also lead to gene expression regulation. For instance, the release of c-src from the multi-protein complex upon ligand binding leads to phosphorylation of target proteins [14
]. Furthermore, in addition to its function as a transcription factor, AhR also acts as an E3 ubiquitin ligase that leads to ubiquitination and targeting for degradation in the proteasome [15
] (Figure 1
Homeostasis in bone remodeling depends on relative levels of bone formation and resorption, which in turn depend primarily on the relative activities of osteoblasts (OB) and osteoclasts (OC). In this context, AhR signaling leads to modulation of the NF-κB, Wnt, and MAP kinase pathways; influences the function and differentiation of OB and OC; and results in altered bone remodeling [16
]. The overall effect of AhR activation in OB is suppressed cell differentiation. First, AhR activation by TCDD inhibits the differentiation of bone-marrow-derived stem cells into OB [21
]. In in vivo models, TCDD leads to bones with thin cortices, a harder matrix, and a greater trabecular fraction, resulting in overall mechanical weakness [22
]. Second, AhR inhibition by resveratrol leads to an increase in bone mineral density (BMD) and bone strength in murine models [23
]. Overall, AhR agonism has dose-dependent effects on OB in which hyperactivation and hypoactivation, respectively, inhibit and promote bone formation.
Indeed, recent studies provide a relatively unified description of the effects of AhR signaling in OB; however, the effects of AhR on OC function and differentiation remain unclear. Therefore, the aim of this review is to summarize the pertinent published findings on AhR signaling in OC, provide a consistent view of the effects of AhR activation on OC function and differentiation, and provide a more complete description of the role of AhR signaling in bone remodeling in health and disease.
2. Role of AhR in Osteoclast Differentiation and Function
AhR is well-known to be expressed in bone cells, including OB and OC [12
]. The activation of AhR by a representative uremic toxin and putative endogenous ligand indoxyl-sulfate (IS) affects osteoclastogenesis in a dose-dependent, exposure duration-dependent manner. Low doses of IS with short exposure duration (3 days) stimulated differentiation of osteoclast precursor cells. On the other hand, longer exposures (5 days) led to suppressed osteoclast differentiation [25
]. Thus, dose and duration of agonist exposure are important variables in AhR-mediated modulation of OC differentiation by IS.
Furthermore, the effects of AhR activation by PAH are also dependent on agonist dose as well as cell density. BaP is pro-osteoclastogenic in vitro when cell density is low and anti-osteoclastogenic when cell density is high. At low concentrations, BaP is highly pro-osteoclastogenic, whereas at high concentrations (2 μM) its effects on osteoclasts are nonexistent in vitro [26
]. Consistent with these findings, BaP at high concentrations (10−5
M) reduces osteoclastogenesis in the cell line RAW264.7 [27
]. Moreover, in addition to the TCDD-mediated dose-dependent suppression of osteoclastogenesis, short durations of exposure to TCDD do not alter the function of OC, suggesting that duration of exposure is an important variable in PAH-mediated AhR modulation [12
]. Thus, OC differentiation and function are modulated by PAH through AhR in a dose-dependent, cell-density-dependent, and duration-dependent manner.
AhR appears to also be required for basal receptor activator of nuclear factor kappa beta ligand (RANKL)-induced osteoclastogenesis. Osteoclastogenesis is suppressed in AhR null cell cultures, unlike its AhR proficient counterpart. These findings are recapitulated in vivo where mice genetically deficient in AhR or the downstream effector genes, Cyp1a1 or Cyp1b1, demonstrate reduced bone resorption and high bone mass. Consistently, BaP-mediated AhR activation in wild-type mice stimulates osteoclastogenesis and osteoclast function [26
]. In addition to mediating the proosteoclastogenic effects of BaP, AhR itself contributes at least in part to RANKL-induced osteoclastogenesis. Cyp1 gene expression is modulated by AhR as well as RANKL. RANKL strongly induces Cyp1a1 and Cyp1a2 expression in wild-type cells, but not in AhR null cells. Bone marrow cell cultures from Cyp1a1/1a2−/−
double knockout and Cyp1a1/1a2/1b1−/−
triple knockout mice demonstrate strongly suppressed osteoclastogenesis, albeit less marked than AhR−/−
mice. Furthermore, TCDD administration is unable to stimulate osteoclastogenesis in cell cultures from Cyp1a1/1a2−/−
double knockout and Cyp1a1/1a2/1b1−/−
triple knockout mice, unlike in cell cultures from wild-type mice. These results are recapitulated via Cyp1 inhibitors such as proadifen and tetramethylsilane (TMS), which suppress osteoclastogenesis in dose-dependent fashion [26
]. Taken together, these findings suggest that AhR-induced Cyp1 expression mediates at least in part the induction of osteoclastogenesis by BaP and RANKL.
In addition, in transgenic mice engineered to express constitutively active AhR (CA-AhR), total osteoclast resorptive activity is heightened in female mice but not in males. On the other hand, the indirect index for total number of OC in the entire skeleton remains unchanged and the RANKL/osteoprotegerin (OPG) is higher in female CA-AhR mice. Furthermore, there is increased osteoclast volume and serum levels of c-terminal telopeptide (CTX) and cathepsin K, suggesting augmented trabecular bone resorption in female CA-AhR mice [29
]. The reason for the sex-specific effects of dioxin AhR agonists such as TCDD as noted above depends on endogenous estrogen status. Supporting this finding is that in ovariectomized mice that demonstrate an estrogen-deficient phenotype, TCDD induces an estrogenic response in certain tissue types such as the uterus; in contrast, AhR agonists have estrogen antagonist effects in presence of estrogen [30
]. In addition, these gender-specific effects in osteoclastogenesis are not limited to AhR function. For example, osteoclastogenesis in the presence of M-CSF and RANKL differs between BMM derived from male versus female mice [32
]. Taken together, results indicate that AhR activation has gender-specific, in particular estrogen-specific, OC-augmenting effects.
The estrogen receptor (ER) ligands estradiol (E2), 4-hydroxy-TAM (4-OHT), and raloxifene suppress OC differentiation. Moreover, the suppressive effects of 4-OHT on OC differentiation were nullified via cotreatment with the AhR antagonist, α-naphthoflavone (ANF). Thus, the inhibitory effects of 4-OHT on OC differentiation are mediated at least in part by AhR. On the other hand, neither the synthetic ER ligand ICI 182,780 (ICI) nor raloxifene are significantly capable of activating AHR in OCs despite their ability to induce robust Cyp1a1 expression in breast cancer cells, suggesting that AhR agonist activity is both cell-type and agonist-type dependent [33
TCDD is associated with dose-dependent suppression of bone growth in animal models at doses relevant to environmental exposure and health risk [34
]. OC differentiation is mechanistically inseparable from OB differentiation, as OB produces key regulators of OC differentiation such as RANKL and M-CSF [36
]. TCDD as well as 3-MC suppress OB differentiation; however, whereas TCDD does not affect mature OB despite strong AhR expression, 3-MC downregulates the expression of RANKL, which is required for OC differentiation [12
]. In addition, BaP directly inhibits osteoclast formation and function, where this inhibitory activity is dependent on AhR-RANKL crosstalk [28
In the setting of co-culture of osteoclast precursor cells and bone marrow stromal cells, the PAH AhR agonists BaP and 7,12-dimethylbenz-[a]anthracene (DMBA) decrease the number of OC independently of agonist dose at high cell densities. In contrast, in the setting of RAW 264.7 cells cultured with RANKL, BaP and DMBA increased OC numbers at low cell densities (<100 TRAP+ cells/well). These findings suggest the possibility of stromal cells, OC, or OC precursors determining the effects of BaP and DMBA on osteoclastogenesis. Furthermore, resveratrol as well as high concentrations of RANKL reverses BaP-induced inhibition of OC function. The expression of Cyp1b1 is strongly induced by BaP (10−5
M) in OC cultures. While this BaP-mediated effect remained unaffected by low concentrations of RANKL (25 ng/mL), high concentrations (200 ng/mL) were able to reverse the effects [27
Taken together, AhR agonists modulate cell differentiation and function with the direction of overall effect depending on agonist dose, treatment duration, and cell density in OC. Furthermore, AhR is implicated in RANKL-induced osteoclastogenesis, wherein RANKL induces the expression of the AhR-related genes, Cyp1a1 and Cyp1a2. In turn, the pro-osteoclastogenic actions of RANKL and AhR agonists such as BaP are mediated by AhR-induced Cyp1 expression. The results of studies demonstrating the effects of specific AhR ligands on osteoclast differentiation and function have been summarized elsewhere (Table 1
Osteoclasts are multinucleated bone cells that are formed by fusion of mononuclear precursor cells in the presence of osteoclastogenic cytokines, including RANKL and M-CSF. RANKL binds to RANK to activate NF-κB and NFAT1c, which stimulate osteoclast differentiation. AhR, an intracellular ligand-activating transcription factor, is activated by various exogenous and endogenous ligands and it turn activates various signaling pathways that interplay with RANKL-RANK pathway and play a crucial role in modulating osteoclast differentiation and function. Among these pathways, AhR is essential in mediating RANKL-stimulated expression of c-Fos, suggesting that the Ahr-c-cFos pathway is important in Ahr-mediated osteoclast stimulation.
Understanding this mechanistic basis of AhR-AhR ligand signaling pathways in osteoclastogensis and the subsequent disruption of bone remodeling is important for understanding factors that contribute to pathological bone remodeling. Overall, the aim of this review is to provide insight into the role of AhR agonists in osteoclast differentiation. The inhibition/promotion mechanisms of osteoclastogenesis through AhR agonists and various signaling pathways is explored in a ligand-, duration-, and concentration-specific manner. Lastly, there is a paucity of research conducted in human OC in the context of AhR modulation. Notably, there are marked differences in the regulation of osteoclastogenesis in mice and humans. For instance, IL-27 acts as a potent direct inhibitor of human osteoclast precursor cells, whereas in mice, IL-27 mainly acts as an indirect modulator of osteoclast differentiation through cells other than osteoclast precursors. Furthermore, in humans, triggering receptor expressed on myeloid cells 2 (TREM-2) deficiency is associated with impaired osteoclast differentiation and bone remodeling as well as Nasu–Hakola disease whereas in mice, deficiency in TREM-2 has unclear effects in vivo and has osteoclastogenic effects in vitro. Therefore, further research on the effects and mechanisms of AhR activation in OC is warranted in human OCs. Elucidation of AhR function in human osteoclasts will introduce a potential therapeutic target for various human diseases in which osteoclasts are implicated in pathogenesis, including bone destructive diseases such as osteoporosis and cancer.