BHLHE40 promotes osteoclastogenesis and abnormal bone resorption via c-Fos/NFATc1

Dysregulated osteoclast activity due to altered osteoclast differentiation causes multiple bone diseases. Osteoclasts are multinucleated giant cells derived from hematopoietic stem cells and play a major role in bone absorption. However, the mechanisms underlying the tight regulation of osteoclast differentiation in multiple pathophysiological status remain unknown. We showed that Bhlhe40 upregulation is tightly associated with osteoclast differentiation and osteoporosis. Functionally, Bhlhe40 promoted osteoclast differentiation in vitro, and Bhlhe40 deficiency led to increased bone mass and decreased osteoclast differentiation in vivo. Moreover, Bhlhe40 deficient mice resisted estrogen deficiency and aging-induced osteoporosis. Mechanism study showed that the increase in bone mass due to Bhlhe40 deficiency was a cell intrinsic defect in osteoclast differentiation in these mice. BHLHE40 upregulated the gene expression of Fos and Nfatc1 by directly binding to their promoter regions. Notably, inhibition of Fos/Nfatc1 abrogated the enhanced osteoclast differentiation induced by BHLHE40 overexpression. Our research reveals a novel Bhlhe40/c-Fos/Nfatc1 axis involved in regulating osteoclastogenesis and shows that osteoporosis caused by estrogen deficiency and aging can be rescued by regulating Bhlhe40 in mice. This may help in the development of a new strategy for the treatment of osteoporosis.


Background
Bone homeostasis is dynamically maintained by balancing bone formation and bone resorption, which are mainly mediated by osteoblasts and osteoclasts (OCs), respectively. Unbalanced osteoblastogenesis and osteoclastogenesis cause multiple bone diseases. For instance, defective osteoclasts cause osteosclerosis and bone marrow failure [1][2][3][4]. Overactivated osteoclasts lead to bone loss and osteoporosis [5], and abnormal bone resorption is mostly caused by osteoclasts [6]. Regulating osteoclast activity is known to be a good method for the clinical treatment of abnormal bone resorption diseases. Indeed, various osteoclast drugs reduce abnormal bone resorption [7,8]. However, these drugs also cause side effects including cardiovascular events and breast cancer risk [9]. In addition, these drugs usually fail to cure malignant osteoporosis and osteosclerosis [10,11]. Although bone marrow transplantation is considered to be an effective method to treat malignant osteoporosis and osteosclerosis [12,13], it is not available for most patients. There is an urgent need to study osteoclast differentiation to develop new strategies and identify potential targets.
OCs are multinucleated giant cells derived from hematopoietic stem cells (HSCs) and play a major role in bone absorption. Osteoclast differentiation is determined by macrophage colony stimulating factor (M-CSF) and receptor activator of NF-κB ligand (RANKL). The binding of RANKL and RANK recruits tumor necrosis factor receptor-associated factor 6 (TRAF6) and activates mitogen-activated protein kinase (MAPK) and NF-κB. MAPK ultimately stimulates the expression of nuclear factor of activated T cells (NFATc) [14] and activator protein-1 (AP-1) [15], which are two master transcription factors in osteoclast differentiation. Factors including inflammation, tumors, and estrogen deficiency may affect RANKL/RANK signaling and the expression of these transcription factors, leading to excessive activation of osteoclasts.
In this study, we found that Bhlhe40 was upregulated during osteoclastogenesis and was associated with osteoporosis. Bhlhe40 −/− mice exhibited increased bone mass and resistance to bone loss induced by estrogen deficiency. Bhlhe40 promoted osteoclastogenesis by upregulating Fos/Nfatc1 in vitro and in vivo. Our research reveals a novel Bhlhe40/Fos/Nfatc1 axis involved in osteoclastogenesis and may provide a theoretical basis and potential targets for the development of novel strategies for bone resorption diseases.

Bhlhe40 upregulation correlated with OC differentiation and osteoporosis in mice
To investigate the potential role of Bhlhe40 in OC differentiation, we analyzed the GSE54779 dataset from the GEO database and found that the mRNA level of Bhlhe40 was significantly upregulated in RANKL-treated mouse bone marrow macrophages (BMMs) during osteoclastogenesis (Fig. 1A, B). We further verified the upregulation of Bhlhe40 mRNA in mouse BMMs undergoing osteoclast differentiation in the presence of M-CSF and RANKL ( Fig. 1C and Additional file 1: Figure S1). Upregulation of BHLHE40 protein in BMMs and RAW264.7 cells was also observed by immunofluorescence during osteoclastogenesis ( Fig. 1D-G). Moreover, we labeled BMMs or OCs with CD11b or CTSK and found Bhlhe40 upregulation in BMMs and OCs from ovariectomized (OVX)-induced osteoporosis mice compared to those from sham-operated control mice (Fig. 1H-K). In addition, upregulation of Bhlhe40 was also observed in elderly osteoporotic mice (Eld) (Additional file 2: Figure  S2 A-D). These results suggested that BHLHE40 upregulation was tightly associated with OC differentiation and osteoporosis.

Bhlhe40 promoted OC differentiation in vitro
To investigate the potential role of Bhlhe40 in osteoclastogenesis, we downregulated Bhlhe40 by siRNA ( Fig. 2A). Bhlhe40 downregulation impaired the formation of osteoclasts ( Fig. 2B-D) and the expression of OC-specific genes (Fig. 2E). In contrast, Bhlhe40 overexpression induced the opposite phenotypes ( Fig. 2F-J). In addition, Bhlhe40 downregulation or upregulation did not seem to affect the viability of BMMs (Additional file 3: Figure S3). These results suggested that Bhlhe40 may play a positive role in osteoclastogenesis.

Bhlhe40 deficiency impaired bone resorption in vivo
To further investigate the role of Bhlhe40 in osteoclast formation in vivo, we took advantage of Bhlhe40 knockout (Bhlhe40 −/− ) mice. Bhlhe40 deletion was verified by PCR (Additional file 4: Figure S4). Femurs from 4-weekold Wt and Bhlhe40 −/− mice were separated to measure mechanical characteristics and bone mass. The mechanical strength of femurs from Bhlhe40 −/− mice was higher than that of femurs from Wt mice: they were able to endure more force before breaking than Wt mice (Fig. 3A, B). Quantitative microtomography (micro-CT) analysis showed that Bhlhe40 −/− mice exhibited lower values for trabecular spacing (Tb. Sp) as well as higher values for bone volume/tissue volume (BV/TV), number of trabeculae (Tb. N), trabecular thickness (Tb. Th), bone density (BMD) and cortical bone thickness (Ct. Th) than their Wt littermates (Fig. 3C, D). Consistently, H&E staining demonstrated an increased number of trabeculae and cortical thickness in Bhlhe40 −/− mice compared with Wt mice (Fig. 3E). TRAP staining showed that Bhlhe40 −/− mice had fewer osteoclasts (yellow arrow) on the bone surface than their Wt littermates (Fig. 3F). Furthermore, lower expression of RANK was observed in Bhlhe40 −/− mice than in their Wt littermates (Fig. 3G). In addition, BMMs isolated from 4-week-old Bhlhe40 −/− mice displayed less osteoclast differentiation, as evidenced by the reduced expression of OC specific genes (Fig. 3H) and decreased TRAP and phalloidin staining (Fig. 3I-K). We also isolated BMSCs from Wt and Bhlhe40 −/− mice and compared osteoblast differentiation. Although ALP activity C RNA expression of Bhlhe40 in BMMs during osteoclast differentiation (n = 3). D Immunofluorescence of BHLHE40 in BMMs during osteoclast differentiation (Scale bar, 50 μm). E Immunofluorescence of BHLHE40 in RAW264.7 during osteoclast differentiation (Scale bar, 50 μm). F Quantitative analysis (0D: n = 3; 6D: n = 4; Others: n = 5) of BHLHE40 in BMMs during osteoclast differentiation in D. G Quantitative analysis (6D: n = 4; Others: n = 5) of BHLHE40 in RAW264.7 during osteoclast differentiation in E. H, I Immunofluorescence and quantitative analysis (n = 3) of BHLHE40 (green) in BMMs between Sham and Osteoporosis mice. BMMs were stained with CD11b (red). Nucleus was stained with DAPI (blue) (Scale bar, 150 μm). J, K Immunofluorescence and quantitative analysis (n = 3) of BHLHE40 (green) in OCs between Sham and Osteoporosis mice. OCs were stained with CTSK (red). Nucleus was stained with DAPI (blue) (Scale bar, 150 μm). All data are mean ± SD; * P < 0.05, ** P < 0.01, *** P < 0.001. by unpaired (B, C, F, G), paired (I, K) Student's t test and one-way ANOVA followed by Tukey's post hoc test appeared to be reduced in BMSCs from Bhlhe40 −/− mice, no significant difference in Alizarin Red S staining (ARS) or Von Kossa staining was observed between Wt and Bhlhe40 −/− mice (Additional file 5: Figure S5A). Less collagen was observed in the trabeculae of 4-week-old Bhlhe40 −/− mice than in those of Wt mice (Masson staining), whereas no significant difference in mineralization was observed (Goldner staining) (Additional file 5: Figure  S5B, C). Notably, more ossification occurred in 3-day-old Bhlhe40 −/− mice than in their Wt littermates (Additional file 5: Figure S5D). Moreover, no significant difference in the number of mature osteoblasts expressing osteocalcin (OCN) was observed in Bhlhe40 −/− mice compared with their Wt littermates (Additional file 5: Figure S5E, F). These results suggested that Bhlhe40 deficiency led to increased bone mass in vivo mainly due to impaired osteoclastogenesis rather than osteoblastogenesis.

Osteosclerosis in Bhlhe40 −/− mice was dominated by osteoclasts instead of the microenvironment.
To verify that the increase in bone mass due to BHLHE40 deficiency was an intrinsic cell defect rather than other environmental factors, bone marrow transplantation was performed. Since adult osteoclasts are mainly derived from HSCs [6], this allows us to specifically study the effect of gene knockout in osteoclasts on bone mass, and excluding the influence of other components in the bone marrow. Bone marrow cells from Wt and Bhlhe40 −/− mice were engrafted into Wt and Bhlhe40 −/− recipient mice, respectively (Fig. 4A). Chimerism was verified by PCR (Additional file 6: Figure S6). We noticed that the phenotype of chimeric mice depended on the donor cells. For instance, the WT-to-WT group showed comparable values for bone mass, number of trabeculae, and number of osteoclasts compared to those of the WT-to-KO group. Consistently, the KO-to-WT group and KO-to-KO group exhibited comparable values for bone mass, number of trabeculae, and number of osteoclasts. Notably, chimeric mice receiving KO donor cells (the KO-to-WT and KO-to-KO groups) showed increased bone mass and number of trabeculae and a decreased number of osteoclasts compared with those of chimeric mice receiving Wt donor cells (Fig. 4B-E). Considering that the majority of osteoclasts are derived from hematopoietic cells, these observations demonstrated that the increase in bone mass in Bhlhe40 −/− mice was due to intrinsic defects in osteoclast differentiation.

Bhlhe40 directly regulated c-Fos and Nfatc1 during osteoclast differentiation
To dissect the mechanism by which Bhlhe40 promotes osteoclast differentiation, we performed RNA-Seq. Heatmap and volcano analyses identified 1157 differentially expressed genes (DEGs) between Wt and Bhlhe40 −/− mice, of which 541 genes were upregulated and 616 genes were downregulated (Fig. 5A, B). KEGG analysis of DEGs revealed a significant enrichment of genes related to cytokine-cytokine receptor interactions, endocrine resistance, osteoclast differentiation and the NF-κB signaling pathway (Fig. 5C, D). The enrichment of genes related to osteoclasts in vivo was further analyzed by GOChord. The enriched genes were assigned to the following categories: "osteoclast differentiation", "NF-κB signaling pathway", "rheumatoid arthritis", "Hippo signaling pathway", "endocrine resistance", and "hematopoietic cell lineage" (Fig. 5E). These observations suggest that Bhlhe40 may regulate the expression of osteoclast differentiation genes. We performed Venn diagram analysis by using the BHLHE40 target genes predicted by the Gene Transcription Regulation Database (list 1), the DEGs from our RNA-seq (list 2), key genes involved in osteoclast differentiation (list 3) and osteoclast differentiation-regulated DEGs from GEO databases (list 4) and 8 genes were identified as common genes including Nfatc1 (Fig. 5F). The expression changes in Wt and Bhlhe40 −/− mice were verified for seven genes (Clec2g could not be detected by qPCR due to its low abundance in cells) (Fig. 5G). However, only 4 genes were further confirmed in Bhlhe40 overexpressing cells, and no significant changes in Src and Ocstamp expression were observed (Fig. 5H). Although Fos was not identified as a DEG in RNA-Seq, we verified its downregulation in Bhlhe40 −/− mice and its upregulation in Bhlhe40 overexpressing cells (Fig. 5G, H). Taken together, these results suggested that Bhlhe40 may promote osteoclastogenesis by upregulating the master regulators Fos and Nfatc1.

Bhlhe40 directly targeted c-Fos and Nfatc1
To further investigate the regulatory role of Bhlhe40 in Nfatc1 and Fos expression, we first confirmed the correlation between BHLHE40 and NFATC1 expression  . All data are mean ± SD; * P < 0.05, ** P < 0.01, *** P < 0.001. by unpaired Student's t test (GSE51496) or FOS (GSE7158) in the GEO database ( Fig. 6A, B). Immunofluorescence showed that Bhlhe40 −/− mice (4 weeks old) expressed less NFATc1 and c-Fos in both BMMs (CD11b +) and OCs (CTSK +) than their Wt littermates ( Fig. 6C-F). Downregulation of NFATc1 and c-Fos in Bhlhe40 −/− mice was also detected by immunohistochemistry ( Fig. 6G-J). Consistently, mice receiving Wt donor cells had higher expression of NFATc1 and c-Fos than mice receiving Bhlhe40 −/− donor cells (Fig. 6K, L). These observations suggested that Bhlhe40 regulated Nfatc1 and Fos during osteoclast differentiation. To explore whether BHLHE40 directly regulates Fos and Nfatc1, 3 putative Bhlhe40-binding sites were identified in the Nfatc1 gene promoter region, and 2 sites in the Fos gene promoter region were identified (Fig. 6M, N) by JASPAR (http:// jaspar. gener eg. net). ChIP analysis in BMMs demonstrated that the amount of DNA in these putative binding sites was significantly increased by the BHLHE40 antibody compared to the preimmune control antibody (Fig. 6M, N). These findings demonstrated that BHLHE40 directly bound to the promoter regions of the Nfatc1 and Fos genes and regulated their expression.

c-Fos/Nfatc1 was required for Bhlhe40 function during osteoclast differentiation
To verify the role of Nfatc1 and Fos in Bhlhe40-mediated osteoclastogenesis, Nfatc1 and Fos were knocked down by siRNA (Additional file 7: Figure S7) in BHLHE40 overexpressing cells. Nfatc1 knockdown or Fos knockdown abrogated the enhanced osteoclastogenesis induced by Bhlhe40 overexpression, as evidenced by a decrease in TRAP + cells and osteoclast-specific gene expression ( Fig. 7A-F). Consistently, an NFATc1 inhibitor (NFAT Inhibitor) or a c-Fos inhibitor (T-5224) also generated similar results ( Fig. 7G-L). These results suggested that Nfatc1 and Fos were required for Bhlhe40 function in osteoclast differentiation.

BHLHE40 deficiency alleviated OVX and aging-induced osteoporosis in vivo
To examine whether BHLHE40 played a role in osteoporosis, an OVX-induced osteoporosis mouse model was established. Micro-CT results showed that OVX Wt mice exhibited significant bone mass loss and decreased values for BV/TV, bone surface area/tissue volume (BS/TV), and Tb. N compared with those of the Sham Wt group (Fig. 8A, B). In sharp contrast, OVX did not cause significant bone mass loss or significant changes in these values in Bhlhe40 −/− mice (Fig. 8A, B). Similarly, H&E staining showed that OVX did not significantly change bone mass or Tb. N in Bhlhe40 −/− mice but did so in Wt mice (Fig. 8C). There were more TRAP-positive cells (yellow arrow) in OVX Wt mice than in Sham Wt mice whereas no significant changes were observed between OVX and Sham Bhlhe40 −/− mice (Fig. 8D). Moreover, micro-CT results indicated that the senile Wt mice showed significant bone loss and decreased BV/TV, BS/TV and Tb. N values compared with those of the adult Wt group (bone mass decreased by approximately 75%), whereas Bhlhe40 deficiency partially rescued bone loss in senile mice (bone mass decreased by approximately 40%) (Fig. 8E, F). Furthermore, the values of indicators of osteoporosis (trabecular bone model factor (Tb. Pf) and structural model index (SMI)) increased in senile Wt mice, while aging did not cause similar changes in Bhlhe40 −/− mice (Fig. 8F). These findings suggested that BHLHE40 may play an important role in osteoporosis, especially OVX-induced osteoporosis.

Discussion
Osteoclasts are critical for bone resorption and abnormal activation of osteoclasts causes a variety of bone diseases including osteoporosis. OC differentiation is controlled by the combined action of multiple transcription factors [25]. Bhlhe40 is a helix-loop-helix transcription factor that plays a key role in various biological processes and is expressed in a variety of human tissues and cells, including resident macrophages [18,19]. However, the role of Bhlhe40 in osteoclastogenesis and its therapeutic effect on abnormal bone resorption-related diseases are still unclear. Here, we demonstrated that Bhlhe40 upregulation was associated with abnormal bone resorption diseases and that Bhlhe40 deficiency inhibited osteoclastogenesis in vitro and in vivo. In addition, bone marrow transplantation showed that osteoclasts were the culprit of osteosclerosis caused by Bhlhe40 deficiency. Through RNA-seq and chromatin immunoprecipitation, we identified a new Bhlhe40/Fos/Nfatc1 axis through which Bhlhe40 affects osteoclastogenesis. Finally, we established a mouse model of estrogen-deficient and senile osteoporosis and found a significant improvement in abnormal bone resorption in Bhlhe40-deficient mice, suggesting that Bhlhe40 is an effective target for the treatment of osteoporosis.
Although limited studies have shown that Bhlhe40 may promote PI3KCA/Akt/GSK3β-regulated osteoblastogenesis [26], the effect of Bhlhe40 on osteoclasts has not been proven, and maintaining bone mass balance is a complex process that requires the participation of osteoblasts and osteoclasts. In contrast, the lack of Bhlhe40 reduces cement loss caused by Porphyromonas gingivalis in a periodontitis model [20]. These studies seem to be contradictory and prompted us to further explore how Bhlhe40 regulates bone mass in abnormal bone resorption diseases. To this end, we isolated bone marrow mesenchymal cells from 4-week-old Wt and Bhlhe40 −/− mice for verification. Interestingly, Bhlhe40 knockout affected the number of ALP + cells but did not affect mineralization, as evidenced by comparable ARS staining at 21 days and Von Kossa staining at 28 days (Additional file 5: Figure S5A). Masson and Goldner staining of the femurs in Wt and Bhlhe40 −/− mice may support this possibility: Collagen reduction was observed whereas mineralization showed no significant change in Bhlhe40 −/− mice. OCN is characteristic of mature osteoblasts and is considered an indicator of mineralization [27]. We failed to find a significant difference in the number of OCN + cells by immunohistochemistry. These observations exclude the possibility that the increased bone mass in Bhlhe40 −/− mice is due to increased osteoblastogenesis. Indeed, bone marrow transplantation demonstrated that the phenotype of Wt or Bhlhe40 −/− recipient mice depended on the donor cells (Fig. 4). Many studies have shown that osteoclasts develop through the fusion of monocyte precursors derived from HSCs in the presence of M-CSF and RANKL [28,29]. In contrast, it is difficult for osteoblasts or bone cells to colonize the bone marrow through bone marrow transplantation [30]. If the bone mass changes in Bhlhe40 −/− mice are dominated by osteoblasts or other bone cells, the phenotype of the recipient mice should depend on themselves rather than the donors. Therefore, our experiment proved that the bone mass changes in Bhlhe40 −/− mice are dominated by osteoclasts. Taken together, our observations indicate that the changes in bone mass caused by Bhlhe40 deficiency are most likely dominated by osteoclasts rather than osteoblasts.
Our RNA-seq and subsequent experiments have shown that Bhlhe40 affects osteoclastogenesis by regulating the transcription of Nfatc1 and Fos. The Fos/Nfatc1 pathway plays a crucial role in the formation of osteoclasts. NFATc1 promotes the expression of a variety of osteoclast-specific genes [31]. c-Fos is an important part of transcriptional AP-1. BMMs lacking c-Fos cannot differentiate into osteoclasts, and mice lacking c-Fos develop osteosclerosis, indicating the key role of c-Fos in osteoclastogenesis [32,33]. Indeed, downregulation and inhibition of Fos or Nfatc1 significantly reversed the increase in osteoclastogenesis in Bhlhe40 overexpressing BMMs, which indicates that Bhlhe40-mediated regulation of osteoclastogenesis depends on Fos/Nfatc1. Our ChIP-PCR confirmed the direct binding of BHLHE40 to the promoter regions of the Nfatc1 and Fos genes. In addition, many studies have demonstrated that Bhlhe40 promotes inflammation [20,34,35], which causes bone loss by promoting osteoclast differentiation [36,37]. Our RNA-Seq also revealed NF-κB signaling and rheumatoid arthritis, which are tightly associated with inflammation. We do not exclude the possibility that Bhlhe40 promotes osteoclast differentiation by enhancing inflammation.
Many osteolytic diseases are caused by the excessive activation of OCs. As mentioned earlier, abnormal bone resorption is mainly inhibited clinically by regulating the function of OCs. However, most drugs have off-target effects and potential side effects [38,39]. In addition, these drugs usually cannot cure malignant osteoporosis and osteopetrosis [10,11]. Studies have shown that some refractory osteoporosis in humans and mice can be partially treated by bone marrow transplantation [40,41], but this approach is not available to most patients and causes substantial pain. Here, our research showed that Fig. 6 Bhlhe40 deficiency leads to a decrease in c-Fos/Nfatc1 and directly targets Fos and Nfatc1. A The correlation expression of Bhlhe40 and Nfatc1 (GSE51495 dataset) (n = 30). B The correlation expression of Bhlhe40 and Fos (GSE7158 dataset) (n = 26). C, D Immunofluorescence and quantitative (n = 5) of NFATc1 (green) in BMMs or OCs between Wt and Bhlhe40 −/− mice. BMMs were stained with CD11b (red). OCs were stained with CTSK (red). Nucleus was stained with DAPI (blue) (Scale bar, 150 μm). E, F Immunofluorescence and quantitative (n = 5) of c-Fos (green) in BMMs or OCs between Wt and Bhlhe40 −/− mice. BMMs were stained with CD11b (red). OCs were stained with CTSK (red). Nucleus was stained with DAPI (blue) (Scale bar, 150 μm). G, H Immunohistochemistry of NFATc1 and quantitative analysis (n = 5) in femurs from Wt and Bhlhe40 −/− mice (Scale bar, 50 μm). I, J Immunohistochemistry of c-Fos and quantitative analysis (n = 5) in femurs from Wt and Bhlhe40 −/− mice (Scale bar, 50 μm). K Immunohistochemistry of NFATc1 in femurs from each group of mice for bone marrow transplantation (Scale bar, 50 μm). L Immunohistochemistry of c-Fos in femurs from each group of mice for bone marrow transplantation (Scale bar, 50 μm). M Anti-BHLHE40 ChIP assay for the Nfatc1 promoter in BMMs. Predicted anti-BHLHE40 binding sites are indicated. Immunoprecipitated DNA was amplified by qPCR (n = 3). Normal rabbit IgG was used as a negative control. Results are presented as ChIP/Input. N Anti-BHLHE40 ChIP assay for the Fos promoter. Predicted Anti-BHLHE40 binding sites are indicated. Immunoprecipitated DNA was amplified by qPCR (n = 3). Normal rabbit IgG was used as a negative control. Results are presented as ChIP/Input. All data are mean ± SD; ns P > 0.05, * P < 0.05, ** P < 0.01, *** P < 0.001. by unpaired Student's t test, two-tailed and one-way ANOVA followed by Tukey's post hoc test  target for aging-induced osteoporosis. These results indicate that diseases related to abnormal bone resorption can be cured by downregulating Bhlhe40. In addition, Tcirg1, considered to be the main case of malignant infantile osteopetrosis (MIOP) [42], was observed to be downregulated in Bhlhe40 −/− mice by RNA-Seq. This result reveals the possibility of using Bhlhe40 as a potential therapeutic target for osteopetrosis. Since the lack of Bhlhe40 also partially affects osteogenesis, the inhibition of Bhlhe40 may have broad application prospects in OVX and inflammation-related osteolysis, which are believed to be dominated by osteoclasts. In senile osteoporosis, which is caused by multiple factors [43], the inhibition of Bhlhe40 should be combined with other treatments.

Conclusion
In summary, our research demonstrated that Bhlhe40 may regulate OC differentiation through Fos/Nfatc1 signaling. Bone mass and trabecular bone are important determinants of bone strength in humans. Loss of bone mass in osteoporosis can affect bone strength and may cause fractures. Our research showed that osteoporosis caused by estrogen deficiency and aging can be rescued by regulating Bhlhe40. Therefore, Bhlhe40 may be a promising target for the treatment of osteoporosis.

Cell culture and osteoclast differentiation
The mouse macrophage cell line RAW264.7 (ATCC, Cat# TIB-71 ™ ) and human kidney epithelial cell line 293T (ATCC, Cat# CRL-3216 ™ ) were cultured in DMEM (HyClone, Cat# SH30243.01B) supplemented with 10% fetal bovine serum (FBS) at 37 °C in a 5% CO 2 atmosphere. All cell lines were routinely screened for mycoplasma contamination using a Mycoplasma Stain Assay Kit (Beyotime, Cat# C0296). Bone marrow macrophages (BMMs) were isolated from 4-week-old male C57BL/6J mice as previously described [44,45]. In short, the femur was separated under aseptic conditions, and the bone marrow cavity was flushed with α-MEM (HyClone, Cat# SH30265.01B) containing 10% FBS and 1% penicillin-streptomycin. The cell suspension was cultured in complete α-MEM at 37 °C in a 5% CO 2 atmosphere overnight to filter off adherent mesenchymal stem cells. The supernatant was transferred to a new cell culture flask, M-CSF was added to a final concentration of 50 ng/ml, and the cells were cultured for another 3 days to allow BMMs to adhere and proliferate.

The proliferation and apoptosis assay
MTT was used to detect the proliferation of BMMs after knockdown or overexpression of BHLHE40. Inoculate treated BMMs (1 × 10 5 cells/ml) in a 96-well plate. Theer days later, 100ul of complete α-MEM containing 10% MTT buffer (Procell, Cat# PB180519) was added to each well and incubated at 37 °C for 4 h. After removing the supernatant, 100 µl of dimethyl sulfoxide (DMSO) was added to the wells, and a multifunctional microplate reader (PE Enspire) was used to detect the absorbance at a wavelength of 492 nm.
For the apoptosis assay, BMMs with knockdown or overexpressed of BHLHE40 and their controls were gently dissociated with trypsin without EDTA, and apoptosis was detected with an Annexin V-FITC/PI apoptosis kit (MultiSciences, Cat# 70-AP101-100) according to the manufacturer's instructions. CytoFlex S flow cytometer and Cytexpert analysis software (Beckman Coulter) were used for analysis.

Mice and bone marrow transplantation
Bhlhe40 −/− (BHLHE40 KO) mice were a gift from the Collaborative Innovation Center of Model Animal Wuhan University. Targeting of the Bhlhe40 locus was performed by standard techniques. The BssHII-BssHII genomic fragment of Bhlhe40 as previously described [47], which contains the promoter and coding region including the ATG in exon 1, was replaced with a Neo cassette. The resulting chimeric mice were backcrossed to C57BL/6 mice more than six to eight times.
For elderly animal models, 1-month-old or 12-monthold male mice (C57BL/6) were sacrificed, and the femurs were isolated and fixed with 4% paraformaldehyde for subsequent experiments.
To establish an osteoporosis mouse model [48], 6-weekold female mice (C57BL/6) were anesthetized with 1-4% isoflurane. A 0.5 cm single midline dorsal incision was made in the lower back through the skin. The connective tissue under the skin was gently freed, the ovary was positioned under the thin muscle layer, and a small incision was made on each side to enter the abdominal cavity of the skin. Expose fallopian tubes and ovaries. The ovaries were identified and placed back into the abdominal cavity (Sham). Ligation was performed around the fallopian tube and small sterile scissors were used to gently cut off the fallopian tube to remove the ovaries. The rest of the fallopian tube was placed back into the abdominal cavity (OVX) and sutured layer by layer.
Since adult osteoclasts are mainly derived from HSCs [6], this allows us to specifically study the effect of gene knockout in osteoclasts on bone mass. Bone marrow transplantation was performed as previously described [49] and grouped as shown in Fig. 4A. In brief, 6-weekold male recipient mice (Wt and Bhlhe40 −/− ) were lethally irradiated (8 Gy). After 4 h, the donor mice (4-week-old Wt and Bhlhe40 −/− ) were sacrificed and the femurs were separated. The bone marrow was collected from the femur by washing 3 times with PBS supplemented with 1% FBS and filtering the cells through a 200 mesh screen. The cells were washed in PBS and resuspended at a concentration of 10 7 cells/ml. Then, 10 6 donor cells (100 μl) were injected into the tail vein of each mouse in the recipient group, and the animals were kept for 7 weeks. Peripheral blood was collected (50 µl) to extract DNA for chimerism identification by PCR. The primer sequences used in the assay are listed in Additional file 8: Table S1. No statistical method was used to predetermine the sample size of animal studies. No mice were excluded from analyses except those with unsuccessful transplantation.

Knockdown and overexpression
To knock down Bhlhe40, Nfatc1 and Fos, a riboFECT CP Transfection Kit (RiboBio, Cat # R10035.7) was used for siRNA transfection for 24 h. The gene sequences used for the target siRNA are listed in Additional file 8: Table S1. The Bhlhe40 overexpression plasmid was purchased from Shanghai Genechem. Lentivirus packaging and infection were performed as previously described [50]. The plasmids pSPAX2, pMD2.G and overexpression plasmids were cotransfected into 293 T cells to package lentivirus. The supernatant was collected at 48 and 72 h and lentiviruses were concentrated (Millipore, Ultracel-100 regenerated cellulose membrane, 15 mL sample volume, Cat# UFC9100). Lentiviral infection (MOI = 100) of BMMs was performed in the presence of 10 µg/ml polybrene for 24 h. Cells were selected with 2 µg/ml puromycin for 72 h. The knockdown efficiency or overexpression level of Bhlhe40 was verified by qPCR. The knockdown efficiency levels of Nfatc1 and Fos were verified by qPCR.

Chromatin immunoprecipitation (ChIP)
The samples were prepared as described previously [51]. BMMs were cultured in 30 ng/ml M-CSF for 3 days and fixed in 1% formaldehyde for 10 min at room temperature. Cells were neutralized with glycine and collected to make cell lysates. Cells were resuspended in sonication buffer at 4 °C for 10 min, and then each sample was sonicated 6 times with 40% AMP for 15 s. After centrifugation at 8000 g for 30 s at 4 °C, chromatin was collected from the supernatant, and 50 μl aliquots of the sample were used to extract the input DNA. After incubating with 1 μl of 1 mg/ml RNase A (Sigma-Aldrich, Cat# R6513) and 2 μl of proteinase K (20 mg/ml) at 65 °C overnight, the input DNA was purified using a TIANGEN Common DNA Product Purification Kit (TIANGEN, Cat# DP204). The chromatin samples were incubated with 1 mg/ml BSA and 40 μl prebalanced protein A/G magnetic beads (MCE, Cat# HY-K0202) for 2 h at 4 °C. After the beads were removed, 1 μg of anti-BHLHE40 antibody was added, incubated overnight and then precipitated with protein A/G magnetic beads. The beads were washed, and DNA was eluted and purified with a TIANGEN Common DNA Product Purification Kit after RNase A treatment. The immunoprecipitated DNA fragments were collected for qPCR amplification. Normal rabbit IgG was used as a negative control. DNA samples were analysed by real-time PCR, and IgG was used for standardization. The nucleotide sequence of the oligonucleotide used in the assay is shown in Additional file 8: Table S1. The antibodies used were summarized in Additional file 9: Table S2.

RNA extraction and quantitative PCR
TRIzol (Thermo Fisher, Cat# 15596026) was used to extract total RNA from BMMs. The reverse transcriptase kit (Vazyme, Cat# R223) was used to reverse transcribe an aliquot of 200 ng total RNA into cDNA. SYBR green mixture (Vazyme, Cat# Q311) and Monad Real-Time PCR instrument (Monad q225) were used for quantitative PCR. The primers used for specific transcripts are listed in Additional file 8: Table S1. The mRNA level was normalized to that of GAPDH.

Micro-CT analysis
All mice were euthanized with CO 2 , and the femurs were separated and evaluated by the SkyScan 1176 high-resolution micro-CT imaging system (Bruker, Germany). According to the manufacturer's instructions [49], each femur was scanned separately at 55 kV and 200 μA using a 0.25-mm aluminium filter to obtain an isometric resolution of 7-μm. NRecon (Bruker, Germany) was used to reconstruct the image and CTAn (Bruker, Germany) was used for quantitative analysis. The volume of interest (VOI) contains most of the metaphysis and part of the diaphysis. CTVol software was used (Bruker Micro-CT, Germany) to adjust the 3D model. Trabecular or cortical bone parameters in an area from 0.2 mm to 2.3 mm or 4.0 mm to 5.4 mm below the growth plate of the femur were measured.

Mechanical testing
The soft tissue was removed from the femur to ensure consistency between the conditions used in the mechanical test. All 8 samples were mechanically tested in the vertical direction. Polymethyl methacrylate rigid base (PMMA, Aladdin, Cat# 9011-14-7) was used to fix both sides of the femur and contact the metal base [52]. The dynamic fatigue testing machine (BOSE ElectroForce 3220 225N) was brought in contact with the PMMA bases on both sides until nominal compression (< 10 N) was observed. Then, the mouse femur was compressed at a speed of 1 mm/s until the applied load dropped significantly (loss of 50% or more). The mechanical test was carried out at room temperature. All test data were collected at 20 Hz.

Skeletal preparation and staining
As mentioned earlier [53], mice were sacrificed by CO2 and skin and muscles were removed. After fixation in 95% ethanol overnight, remaining samples were transferred to acetone for 48 h. Cartilages were then stained in 0.3% Alcian Blue (Sigma-Aldrich, Cat# A5268) overnight at room temperature. The samples were incubated overnight in 95% ethanol to decolorize and incubated with 1% KOH solution at room temperature for 1 h to remove residual tissue. The KOH solution was removed and incubated with 0.1% ARS solution (Sigma-Aldrich, Cat# A5533) overnight at 4 °C for bone staining. Specimens were incubated in 1% KOH until the excess ARS and tissues were completely cleared, and then stored in glycerol for pictures by Nikon camera.
For frozen sections, femurs were fixed in 4% paraformaldehyde for 6 h and decalcified as previously described [54]. After the samples were washed, they were soaked in PBS containing 30% sucrose at 4 °C for 1-2 days, and embedded in a gel-filled disposable histological plastic mold. We used a Leica CM3050S cryostat to slice the samples (20 μm) at − 25 °C.
For immunohistochemistry, sections were deparaffinized and treated with citrate buffer solution (pH 6.0) 3 times at 95 °C, and treated with 3% H 2 O 2 for 20 min at room temperature. Samples were then blocked with 5% bovine serum albumin (BSA) at room temperature for 1 h, and incubated with the primary antibody overnight at 4 °C. After the samples were washed, the Polink-2 Plus polymer HRP detection system (ZSGB-BIO, Cat# PV6001) was used for the secondary antibody incubation of the samples, and DAB (ZSGB-BIO, Cat# ZLI-9017) was used for color development. Hematoxylin was used for nuclear staining. After dehydration and fixation, the slices were scanned by the Aperio VERSA 8 (Leica, Germany) scanner.

Immunofluorescence staining
BMMs and RAW264.7 cells cultured in complete α-MEM containing 30 ng/ml M-CSF and 50 ng/ml RANKL for 0, 2, 4, or 6 days were fixed with 4% paraformaldehyde for 20 min and permeabilized with PBS containing 0.5% Triton-X for 30 min. For anti-BHLHE40 immunofluorescence staining, cells were blocked with PBS containing 5% BSA after fixation and permeabilization, and cells were incubated with BHLHE40 primary antibodies overnight at 4 °C. Then, the cells were incubated with secondary antibodies for 1 h. The images were taken by confocal microscopy (Leica SP8, Germany) and analysed with Leica Application Suite X (LAS X).
For immunofluorescence, before staining with antibodies, the sections were allowed to equilibrate for 30 min at room temperature and rehydrated with PBS at room temperature 3 times for 5 min. After the samples were permeabilized or blocked with PBS containing 0.2% Triton-X 100 or 5% BSA for 1 h, the sections were incubated with the primary antibody overnight at 4 °C. After washing with PBS 3 times for 5 min, sections were incubated with fluorescently labeled secondary antibodies for 1 h. Images were captured with a confocal microscope (Leica SP8) and analysed with LAS X. Sections were also stained with integrin alpha-M (CD11b) (BMMs marker), CTSK (OCs marker) and DAPI (nuclear).
The primary and secondary antibodies used are listed in Additional file 9: Table S2.

Transcriptomic analysis
BMMs were isolated from Wt and Bhlhe40 −/− mice, cultured with 50 ng/ml M-CSF for 3 days, and harvested in a GenCatch TM Total RNA Extraction Kit (Epoch Life Sciences, Cat# 1660050) 48 h after OC differentiation. Two micrograms of RNA was used for cDNA library preparation and sequencing following the manufacturer's instructions (HiSeq 2500, Illumina). HISAT2 [55] was used to map sequencing reads to the mouse genome and SAMtools [56] was used to aggregate tag counts at the gene level allowing only one read per position per length. DESeq2 [57] was used to determine the differentially expressed genes. ClusterProfiler [58] was used to perform gene set enrichment analysis, including GO and KEGG analyses. GOplot [59] was used for the visualization of GO analysis. The RNA-seq data in this study are available in SRA, and the BioProject accession number is PRJNA763091 (http:// www. ncbi. nlm. nih. gov/ biopr oject/ 763091).

Bioinformatics analysis
To explore the potential correlation between Bhlhe40 and OC differentiation, GSE54779 (GPL6246 Affymetrix mouse gene 1.0 ST array) dataset from the Gene Expression Omnibus (GEO) repository at the National Centre of Biotechnology Information website (https:// www. ncbi. nlm. nih. gov/ geo/) was analysed. This dataset contains three BMM samples treated with RANKL and M-CSF and three BMM samples treated with M-CSF alone. All raw expression data were standardized by R and the Bioconductor (http:// www. bioco nduct or. org) Affymetrix package. The Linear Array Microarray Analysis (Limma) software package was used to identify the differentially expressed genes (DEGs) in BMM samples treated with M-CSF and RANKL compared with samples stimulated with M-CSF alone. Values of p < 0.05 and |log2FC|> 0.5 were considered thresholds for identifying DEGs.
In order to determine the correlation of Bhlhe40 and Fos or Nfatc1, GSE7158 or GSE51495 from the GEO repository at the National Center for Biotechnology Information website (https:// www. ncbi. nlm. nih. gov/ geo/) was used for analysis (GPL570 Affymetrix Human Genome U133 Plus 2.0 Array or GPL6104 Illumina humanRef-8 v2.0). GraphPad Prism 7.0 was used to analyse the correlation between Bhlhe40 and Fos or Nfatc1 expression in each sample.

Statistical analysis
Data were expressed as the mean ± SD. The results were considered statistically significant if the P value was less than 0.05. The random number method was used for random assignment. Researchers blinded to group assignment during the experiment. Individual data points were shown, and the number of samples or images analysed was indicated in the figure and/or legend. The data were analysed using the appropriate Student's t test when comparing two groups or one-way analysis of variance when comparing more than 2 groups. The data points correspond to independent experiments. The variances between the compared groups were similar. All statistical tests were performed using Prism 7.0 software.