Author response: PTH regulates osteogenesis and suppresses adipogenesis through Zfp467 in a feed-forward, PTH1R-cyclic AMP-dependent manner

Abstract

Full text Figures and data Side by side Abstract Editor's evaluation Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract Conditional deletion of the PTH1R in mesenchymal progenitors reduces osteoblast differentiation, enhances marrow adipogenesis, and increases zinc finger protein 467 (Zfp467) expression. In contrast, genetic loss of Zfp467 increased Pth1r expression and shifts mesenchymal progenitor cell fate toward osteogenesis and higher bone mass. PTH1R and ZFP467 could constitute a feedback loop that facilitates PTH-induced osteogenesis and that conditional deletion of Zfp467 in osteogenic precursors would lead to high bone mass in mice. Prrx1Cre; Zfp467fl/fl but not AdipoqCre; Zfp467fl/fl mice exhibit high bone mass and greater osteogenic differentiation similar to the Zfp467-/- mice. qPCR results revealed that PTH suppressed Zfp467 expression primarily via the cyclic AMP/PKA pathway. Not surprisingly, PKA activation inhibited the expression of Zfp467 and gene silencing of Pth1r caused an increase in Zfp467 mRNA transcription. Dual fluorescence reporter assays and confocal immunofluorescence demonstrated that genetic deletion of Zfp467 resulted in higher nuclear translocation of NFκB1 that binds to the P2 promoter of the Pth1r and increased its transcription. As expected, Zfp467-/- cells had enhanced production of cyclic AMP and increased glycolysis in response to exogenous PTH. Additionally, the osteogenic response to PTH was also enhanced in Zfp467-/- COBs, and the pro-osteogenic effect of Zfp467 deletion was blocked by gene silencing of Pth1r or a PKA inhibitor. In conclusion, our findings suggest that loss or PTH1R-mediated repression of Zfp467 results in a pathway that increases Pth1r transcription via NFκB1 and thus cellular responsiveness to PTH/PTHrP, ultimately leading to enhanced bone formation. Editor's evaluation The study provides evidence that the hormone PTH increases bone mass by, at least in part, regulating the factor Zfp467. In turn, Zfp67 controls expression of the receptor for PTH, thus creating a feedback loop that overall augments bone mass. The findings are novel and of great interest. The study is significant as it unveils a novel feedback loop involving PTH, a critical endocrine regulator of calcium, phosphate, and bone mass. https://doi.org/10.7554/eLife.83345.sa0 Decision letter Reviews on Sciety eLife's review process Introduction Intermittent administration of parathyroid hormone (PTH 1–34), teriparatide, is anabolic for bone and is a well-established treatment for osteoporosis (Estell and Rosen, 2021). Both PTH and PTHrP act through the PTH1R to drive bone formation, increase bone mass, and lower fracture risk (Goltzman, 2018). It has been established that one mechanism for PTH-driven osteogenesis is through induction of cyclic AMP and the PKA pathway, although PKC can also be activated by ligand binding to the PTH1R (Wein and Kronenberg, 2018). Downstream targets from PTH1R activation include IGF-1, FGF-2, Rankl, Sclerostin, Wnt signaling, salt-inducible kinases, and the BMPs (Wein and Kronenberg, 2018; Nishimori et al., 2019; Yu et al., 2012). Pth1r is expressed in chondrocytes, and the entire osteoblast lineage from early osteogenic progenitors to osteoblasts during which it is upregulated (Goltzman, 2018). Osteocytes and lining cells also express Pth1r (Goltzman, 2018; Wein and Kronenberg, 2018). Recent studies have noted that the PTH1R is also expressed in mature adipocytes and their immediate precursors, marrow adipocyte-like progenitors, or MALPs (Zhong et al., 2020). Intermittent PTH treatment increases bone formation both by enhancing the number of osteoblasts and their function, resulting in higher bone mass (Balani et al., 2017). Several reports have demonstrated that PTH-induced lineage allocation of skeletal stromal cells into osteoblasts is at the expense of adipogenesis (Fan et al., 2017). This is consistent with human studies which confirmed that PTH treatment reduces bone marrow adiposity principally through a shift in lineage allocation (Maridas et al., 2019). The transcriptional mechanisms whereby PTH drives a progenitor cell toward an osteoblast are multiple, complex, and redundant. Previous study reported that genetic deletion of the PTH1R in Prrx1+ cells resulted in low bone mass and a marked increase in bone marrow adiposity (Fan et al., 2017). A previous study for the mechanisms that drive adipogenesis that identified Zfp467 as a gene markedly upregulated in the absence of the PTH1R (Fan et al., 2017). Zinc finger proteins (ZFPs) are one of the largest classes of transcription factors in eukaryotic genomes and Zfp521 and Zfp423 have been identified as critical determinants of both adipogenesis and osteogenesis (Ganss and Jheon, 2004; Kiviranta et al., 2013). Zfp467 has been reported to enhance both Sost and Pparg expression in marrow stromal cells, and is highly expressed in both adipocyte and osteoblast progenitors (Quach et al., 2011). Furthermore, an earlier study identified Zfp467 as a transcriptional regulator of lineage allocation among mesenchymal progenitors cells in the marrow (You et al., 2012). In accordance with that report, global genetic deletion of Zfp467 resulted in high bone mass and a reduction in bone marrow adipose tissue (BMAT) and peripheral adipose depots (Le et al., 2021). Hence, the inverse relationship between PTH1R and Zfp467 suggested an additional pathway by which PTH could impact lineage allocation. To determine the molecular mechanism of this interaction, Zfp467-/- osteoblast progenitors were used for studying and a potential transcriptional modulator of the PTH1R, regulated by ZFP467, was identified. This report details both in vitro and in vivo studies, and points to a novel pathway that impacts skeletal formation through the PTH1R. Results Previous study showed that global deletion of Zfp467 increased trabecular bone volume and cortical bone thickness, compared to wild-type mice at the age of 16 weeks (Le et al., 2021). And histomorphometry results showed higher structural and dynamic formation parameters in Zfp467-/- mice vs. Zfp467+/+ (Le et al., 2021). To assess whether the effect on bone mass seen in the Zfp467 global knockout (KO) mice was cell-autonomous to mesenchymal cells, we generated mice with deletion of Zfp467 in the limb MSCs by crossing Zfp467fl/fl mice with the Prrx1Cre mice or AdipoqCre mice (Figure 1—figure supplement 1). Similar to the global Zfp467 null mice, Prrx1Cre Zfp467 mice have increased trabecular bone mass Body mass and body size were not significantly different between Prrx1Cre; Zfp467fl/fl and control littermates in both males and females (data not shown). Micro-computed tomography (μCT) analysis showed that Prrx1Cre; Zfp467fl/fl mice had higher trabecular bone volume fraction (Tb.BV/TV), greater connectivity density (Conn.D), and higher trabecular number (Tb.N) with a significant decrease in structural model index (SMI) and trabecular separation (Tb.Sp) in both males and females (Figure 1), indicating an increase in trabecular bone mass. Cortical thickness was increased, although not significantly in males and marginally decreased in females at this age. In contrast, tibial adipose tissue volume fraction in the marrow was significantly decreased in males (Figure 1B), and showed a similar non-significant trend in females (Figure 1D). Although the time point of examination differed (12 weeks here instead of 16 weeks in the global KO; Le et al., 2021), these results show that, overall, Prrx1Cre; Zfp467fl/fl mice showed a similar bone phenotype to the global Zfp467 KO mice. It is therefore conceivable that progenitor fate was indeed altered in the global Zfp467 KO mice, resulting in significant beneficial skeletal changes. Figure 1 with 2 supplements see all Download asset Open asset Prrx1Cre Zfp467 mice have more trabecular bone mass and less adipose tissue in bone marrow, recapitulating the global Zfp467 null mice. (A, B) Male and (C, D) female 12-week-old Prrx1Cre; Zfp467fl/fl mice and control mice were measured using trabecular and cortical bone of tibiae. Marrow adipose tissue volume (Ad.V) was quantified by osmium tetroxide staining and micro-computed tomography (μCT). Data shown as mean ± SD by unpaired Student’s t test, n=5–8 per group. To assess the cellular activities that contributed to increased bone mass in male Prrx1Cre;Zfp467fl/fl mice, static and dynamic histomorphometry was performed. Consistent with µCT, histomorphometric analysis showed that although the changes in BV/TV were not significant, the overall trend was clearly toward an increase, and Tb.N (p=0.0271) and number of osteoblasts/total area (N.Ob/T.Ar, p=0.0105) were markedly increased in Prrx1Cre;Zfp467fl/fl male mice compared to Zfp467fl/fl control (Table 1), confirming an increase in osteogenesis and bone mass. However, in contrast to global Zfp467 KO mice, the increased trabecular bone was observed in male Prrx1Cre;Zfp467fl/fl mice only, whereas no significant changes in the bone phenotype were found in female mice between groups. Table 1 Quantification of structural and cellular parameters in the left tibiae of 12-week-old Prrx1Cre;Zfp467fl/fl and control Zfp467fl/fl mice by histomorphometry. MaleFemaleZfp467fl/flPrrx1Cre; Zfp467fl/flp-ValueZfp467fl/flPrrx1Cre; Zfp467fl/flp-ValueBV/TV (%)9.1482±2.652312.570±2.96940.067.4561±1.94738.1315±3.85040.71Tb.Th (μm)35.961±3.777237.700±5.87600.6335.384±2.561134.748±6.58620.76Tb.Sp (μm)396.80±168.61274.83±62.8170.13484.18±119.44501.35±288.610.90Tb.N (n/μm)2.5294±0.60373.3260±0.49420.032.0813±0.41752.3440±1.16500.61OS/BS (%)21.767±11.72231.987±12.7040.1614.244±5.11598.8012±4.28100.07O.Th (μm)2.0084±0.44114.8227±3.43060.073.1650±0.45873.4638±1.27930.60Ob.S/BS (%)16.672±7.697325.817±8.58710.0721.776±6.888618.210±8.04280.43N.Ob/B.Pm (n/mm)12.333±5.400319.082±5.64690.0517.250±5.187014.051±4.71400.29Oc.S/BS (%)10.601±3.650812.975±4.23340.3118.439±4.774019.192±6.78880.83N.Oc/B.Pm (n/mm)5.3885±1.97016.3505±2.11710.428.7722±2.19359.6039±3.45460.63MS/BS (%)45.655±3.536643.990±4.34190.4745.333±5.400944.027±1.87330.68MAR (μm/day)1.2265±0.22381.1889±0.12660.711.8148±0.12941.6824±0.48160.63BFR/BS (μm3/μm3/day)0.5634±0.13730.5220±0.06480.490.8207±0.09330.7457±0.22920.59BFR/BV (%/day)3.2079±1.23392.6756±0.53730.324.6232±0.41234.5078±1.72350.85 Data are means ± SD (n=6–7). BV/TV, bone volume/total volume; Tb.Th, trabecular thickness; Tb.Sp, trabecular separation; Tb.N, trabecular number; OS/BS, osteoid surface/bone surface; O.Th, osteoid thickness; Ob.S/BS, osteoblast surface/bone surface; N.Ob/B.Pm, osteoblast number/bone perimeter; Oc.S/BS, osteoclast surface/bone surface; N.Oc/BS, osteoclast number/ bone surface; N.Oc/B.Pm, osteoclast number/bone perimeter; MS/BS, mineralizing surface/ bone surface; MAR, mineral apposition ratio; BFR/BS, bone formation ratio/ one surface; BFR/BV, bone formation rate/bone surface. AdipoqCre Zfp467 mice have similar cortical and trabecular bone mass to controls In order to determine the whole-body phenotype of the conditional knockout (cKO) of Zfp467 mice. AdipoqCre Zfp467fl/fl; mice have similar body weight, fat mass, lean mass, and femoral areal BMD at 12 weeks compared to control mice in both males and females (data not shown). µCT was performed and analyzed in the metaphysis and cortical bone at the tibial mid-diaphysis in 12-week-old AdipoqCre Zfp467fl/fl; mice and control mice. Not surprisingly, no significant difference was found regarding Tb.BV/TV, Conn.D, Tb.N, trabecular thickness (Tb.Th), Tb.Sp, and SMI (Figure 1—figure supplement 2A,B) between controls and AdipoqCre mice. In addition, no differences in cortical bone measurements including total area (Tt. Ar), cortical area/total area (Ct.Ar/Tt.Ar), cortical thickness (Ct.Th), marrow area (Ma.Ar), cortical porosity (Ct. Porosity), and cortical tissue mineral density (Ct.TMD) was observed between groups (Figure 1—figure supplement 2A,B). PTH suppressed the expression levels of Zfp467 via the PKA pathway To understand the mechanisms that underly the impact of mesenchymal deletion of Zfp467, the signaling pathway of PTH relative to Zfp467 in osteoblasts was examined by qPCR. Consistent with a previous study which showed that short-term PTH treatment suppressed the expression of Zfp467 (Quach et al., 2011), treating cells with 100 nM PTH for 10 min could significantly suppress Zfp467 expression in both COBs (Figure 2A) and bone marrow stromal cells (BMSCs) (Figure 2B). When pretreating COBs and BMSCs with PKA and PKC inhibitors (10 µM H89 and 5 µM Go6983, Selleck Chemicals, Houston, TX), respectively, for 2 hr prior to 10 min of 100 nM PTH treatment, significant rescue of Zfp467 suppression was seen in H89 group, but Go6983 had no effect (Figure 2C and D). Forskolin, a selective PKA pathway activator, was also found to significantly inhibit the expression level of Zfp467 in COBs (Figure 2E) and BMSCs (Figure 2F). Moreover, consistent with previous study that Prrx1Cre; Pth1rfl/fl mice showed higher expression level of Zfp467 in bone marrow, silenced Pth1r with siRNA in both COBs and BMSC, and found Pth1r knockdown significantly upregulated the expression level of Zfp467 (Figure 2G and H). These data suggested that PTH1R activation could downregulate Zfp467 expression via the PKA pathway, and that ZFP467 could be one of the important downstream targets of PTH signaling. Figure 2 Download asset Open asset Parathyroid hormone (PTH) suppressed the expression levels of Zfp467 via the PKA pathway. (A) PTH treatments significantly suppressed Zfp467 expression within 10 min of treatment in Zfp467+/+ calvarial osteoblasts (COBs). Data shown as mean ± SD by one-way ANOVA, n=3 independent experiments for each group. (B) PTH treatments significantly suppressed Zfp467 expression within 10 min of treatment in Zfp467+/+ bone marrow stromal cells (BMSCs). Data shown as mean ± SD by one-way ANOVA, n=3 independent experiments for each group. (C) qPCR results· of Zfp467+/+ in COBs with 2 hr PKA or PKC inhibitor treatment prior to 10 min of 100 nM PTH exposure, PKA but not PKC inhibitor was able to rescue the suppression of Zfp467 induced by PTH. Data shown as mean ± SD by one-way ANOVA, n=3 independent experiments for each group. (D) qPCR results of Zfp467+/+ in BMSCs with 2 hr PKA or PKC inhibitor treatment prior to 10 min of 100 nM PTH exposure. PKA but not PKC inhibitor was able to rescue the suppression of Zfp467 induced by PTH. Data shown as mean ± SD by one-way ANOVA, n=3 independent experiments for each group. Forskolin significantly suppressed Zfp467 expression within 1 hr of treatment in Zfp467+/+ COBs. Data shown as mean ± SD by unpaired Student’s t test, n=3 independent experiments for each group. (F) Forskolin significantly suppressed Zfp467 expression after 6 hr of treatment in Zfp467+/+ BMSCs. Data shown as mean ± SD by unpaired Student’s t test, n=3 independent experiments for each group. (G) Pth1r-siRNA treatment in Zfp467+/+ COBs led to an increase of Zfp467 expression in Zfp467+/+ COBs. Data shown as mean ± SD by unpaired Student’s t test, n=3 independent experiments for each group. (H) Pth1r-siRNA treatment in Zfp467+/+ COBs led to an increase of Zfp467 expression in Zfp467+/+ BMSCs. Data shown as mean ± SD by unpaired Student’s t test, n=3 independent experiments for each group. NC, negative control. Zfp467-/- cells have greater Pth1r transcriptional levels driven by both the P1 and P2 promoter Our previous study showed that the global absence of Zfp467 resulted in a significant increase in trabecular bone volume, a marked reduction in peripheral and marrow adipose tissue, and an ~40% increase in Pth1r gene expression in bone from the Zfp467-/- mice compared to littermate controls (Le et al., 2021). These data suggested the possibility of a positive feedback loop whereby the suppression of Zfp467 mediated by PTH leads to an increase of PTH1R. Consistent with this tenet, higher gene and protein expression level of PTH1R was found in Zfp467-/- COBs and BMSCs (Figure 3A and B). Three transcripts of Pth1r (NM_011199.2, NM_001083935.1, and NM_001083936.1) with different transcription starting sites (TSS) were reported based on NCBI database and UCSC Genome Browser on Mouse (Figure 3C). The three transcripts shared the same coding sequence and the only difference was located at the 5’ untranslated region. Based on the different 5’ untranslated regions, related primers were designed; by qPCR both the Pth1r-T1 and Pth1r-T2 transcripts were upregulated in Zfp467-/- cells (Figure 3D). However, the expression level of Pth1r-T2 was much higher than Pth1r-T1. Pth1r-T1 and Pth1r-T2 transcripts were driven by P1 and P2 promoters, respectively. As P1 is much longer than P2 and hadn’t been investigated before, a P1 promoter-driven dual-fluorescence reporter with four different length P1 promoters were designed to assess any change in the promoter activity of P1 in both COBs and BMSCs in the absence of 467 (Figure 3C). The 1.6 and 2.1 kb promoter-driven reporters are higher in activated Zfp46 -/- cells compared to Zfp467+/+ cells (Figure 3E), which indicated the binding site of Pth1r P1 promoter in Zfp467 -/- cells is between 0.6 and 1.1 kb ahead of P1 TSS. Figure 3 Download asset Open asset Zfp467-/- cells have greater Pth1r transcriptional levels driven by both the P1 and P2 promoter. (A) qPCR results of baseline calvarial osteoblasts (COBs) and bone marrow stromal cells (BMSCs). Higher expression level of Pth1r was found in both Zfp467-/- COBs and BMSCs. Data shown as mean ± SD by unpaired Student’s t test, n=5–7 independent experiments for each group. (B) Western blot analysis of baseline COBs and BMSCs. Higher expression level of PTH1R was found in both Zfp467-/- COBs and BMSCs. (C) A schematic of three different Pth1r transcripts and P1 promoter of Pth1r. Four different length P1 promoter constructs were designed and inserted into dual-fluorescence reporter vector. (D) qPCR results of three Pth1r transcripts and total Pth1r. Total Pth1r and Pth1r-T1, T2 but not Pth1r-T3 were upregulated in both Zfp467-/- COBs and BMSCs. Data shown as mean ± SD by unpaired Student’s t test, n=3 independent experiments for each group. (E) Dual-fluorescence assay using indicated four P1 reporter constructs. The 1.6 and 2.1 kb constructs-driven reporter is higher activated in Zfp467-/- cells compared to Zfp467+/+ cells. Data shown as mean ± SD by unpaired Student’s t test, n=3 independent experiments for each group. (F) A schematic of P1 and P2 promoter constructs of Pth1r. (G) Dual-fluorescence assay using indicated P1 and P2 reporter constructs. Both P1 and P2-2 were found significantly higher activated in Zfp467-/- cells. Data shown as mean ± SD by unpaired Student’s t test, n=3 independent experiments for each group. NC, negative control, TSS, transcription starting site. Figure 3—source data 1 Western blot for Figure 3B. https://cdn.elifesciences.org/articles/83345/elife-83345-fig3-data1-v2.zip Download elife-83345-fig3-data1-v2.zip Figure 3—source data 2 Western blot for Figure 3B PTH1R in calvarial osteoblasts (COBs). https://cdn.elifesciences.org/articles/83345/elife-83345-fig3-data2-v2.zip Download elife-83345-fig3-data2-v2.zip Figure 3—source data 3 Western blot for Figure 3B ACTIN in calvarial osteoblasts (COBs). https://cdn.elifesciences.org/articles/83345/elife-83345-fig3-data3-v2.zip Download elife-83345-fig3-data3-v2.zip Figure 3—source data 4 Western blot for Figure 3B PTH1R in bone marrow stromal cells (BMSCs). https://cdn.elifesciences.org/articles/83345/elife-83345-fig3-data4-v2.zip Download elife-83345-fig3-data4-v2.zip Figure 3—source data 5 Western blot for Figure 3B ACTIN in bone marrow stromal cells (BMSCs). https://cdn.elifesciences.org/articles/83345/elife-83345-fig3-data5-v2.zip Download elife-83345-fig3-data5-v2.zip Combined with the previously reported two potential transcription factor binding sites of P2 (Tohmonda et al., 2013), three P1- and P2-driven dual-fluorescence reporters were constructed (Figure 3F). Both P1 and P2 were activated in Zfp467-/- cells, but P2 showed much higher activity in BMSCs and COBs. P2-2 was also much more active in Zfp467-/- cells than Zfp467+/+ cells. P2-2 was therefore chosen for transcription factor prediction via JASPAR, Animal TFDB and PROMO databases; six transcription factors predicted by all three databases were noted, including CREB, EBF1, MYOD, cFOS, NFκB1, and GATA1. Zfp467-/- cells have higher NFκB1 and GATA1 nuclear translocation Using qPCR, no differences in the transcriptional levels of these transcription factors was observed (Figure 4—figure supplement 1). Therefore, those potential transcription factors that could bind to cryptic sites in the Pth1r P2-2 promoter in a pre-osteoblast cell line MC3T3-E1 were overexpressed subsequently. Only GATA1 and NFκB1 overexpression could significantly upregulate the expression level of Pth1r, especially Pth1r-T1 and -T2 (Figure 4A). Nuclear translocation level of NFκB1 and GATA1 were further detected using immunofluorescence, nuclear protein isolation, and western blot. NFκB1 and GATA1 were almost evenly distributed in the cytoplasm and nucleus of Zfp467+/+ cells but underwent partial translocation to the nucleus in Zfp467-/- cells (Figure 4B). Zfp467-/- cells also showed much higher nuclear protein level of both NFκB1 and GATA1 (Figure 4C and D). Figure 4 with 1 supplement see all Download asset Open asset Zfp467-/- cells have higher NFκB1 and GATA1 nuclear translocation. (A) qPCR results of overexpression of Ebf1, Myod, Myog, Gata1, and NFκB1 in MC3T3-E1 cell line. GATA1 and NFκB1 overexpression could significantly upregulate the expression level of Pth1r. Data shown as mean ± SD by unpaired Student’s t test, n=3 independent experiments for each group. (B) Representative confocal images of GATA1 and NFκB1 immunofluorescence in Zfp467+/+ and Zfp467-/- bone marrow stromal cells (BMSCs) and related quantification. (C) Nuclear protein level of GATA1 and NFκB1 in Zfp467+/+ and Zfp467-/- BMSCs. (D) Quantification analysis for nuclear protein level of GATA1 and NFκB1 in Zfp467+/+ and Zfp467-/- BMSCs. Data shown as mean ± SD, n=3 independent experiments for each group. Data shown as mean ± SD by unpaired Student’s t test, n=3 independent experiments for each group. Figure 4—source data 1 Western blot for Figure 3C. https://cdn.elifesciences.org/articles/83345/elife-83345-fig4-data1-v2.zip Download elife-83345-fig4-data1-v2.zip Figure 4—source data 2 Western blot for Figure 3C ACTIN. https://cdn.elifesciences.org/articles/83345/elife-83345-fig4-data2-v2.zip Download elife-83345-fig4-data2-v2.zip Figure 4—source data 3 Western blot for Figure 3C H3. https://cdn.elifesciences.org/articles/83345/elife-83345-fig4-data3-v2.zip Download elife-83345-fig4-data3-v2.zip Figure 4—source data 4 Western blot for Figure 3C p50 in cytoplasm. https://cdn.elifesciences.org/articles/83345/elife-83345-fig4-data4-v2.zip Download elife-83345-fig4-data4-v2.zip Figure 4—source data 5 Western blot for Figure 3C GATA1 in cytoplasm. https://cdn.elifesciences.org/articles/83345/elife-83345-fig4-data5-v2.zip Download elife-83345-fig4-data5-v2.zip Figure 4—source data 6 Western blot for Figure 3C GATA1 in nuclear. https://cdn.elifesciences.org/articles/83345/elife-83345-fig4-data6-v2.zip Download elife-83345-fig4-data6-v2.zip Figure 4—source data 7 Western blot for Figure 3C p50 in nuclear. https://cdn.elifesciences.org/articles/83345/elife-83345-fig4-data7-v2.zip Download elife-83345-fig4-data7-v2.zip An NFκB1-RelB heterodimer may drive greater Pth1r transcription in Zfp467-/- cells In order to confirm whether NFκB1 or GATA1 could activate the specific P1 or P2-2 Pth1r promoter, P1 and P2-2 dual-luciferase reporter and Gata1, Nfkb1 overexpression plasmids were co-transfected in MC3T3-E1 cells. Only the Nkfb1 overexpression group was able to significantly activate the P2-2 promoter (Figure 5A). Chromatin immunoprecipitation (ChIP) results showed that the DNA was properly sheared and IP was successfully conducted (Figure 5B). ChIP-qPCR results showed that the first two parts of P2 were properly enriched in our IP product (>0.5%) (Figure 5—figure supplement 1A), and the first part of P2 was approximately 20-fold more highly enriched in our NFκB1 IP product than IgG (Figure 5C, Figure 5—figure supplement 1B); this indicated that NFκB1 binds to the P2 promoter, especially at the first 200 bp site. Subsequently, COBs and BMSCs were treated with Nfkb1 siRNA and showed that Nfkb1 knockdown could significantly inhibit the expression of Pth1r in both Zfp467+/+ and Zfp467-/- COBs and BMSCs. Importantly, Nfkb1 knockdown in Zfp467-/- cells reverts the levels of Pth1r to the levels seen in Zfp467+/+ cells (Figure 5D–G). Figure 5 with 1 supplement see all Download asset Open asset NFκB1 was found to transactivate Pth1r expression in Zfp467-/- cells. (A) Reporter assays using the indicated P1 or P2 reporter construct and an expression vector bearing Gata1, Nfkb1, or a control empty vector. Data shown as mean ± SD by one-way ANOVA, n=3 independent experiments for each group. Data shown as mean ± SD by one-way ANOVA, n=3 independent experiments for each group. (B) Immunoblot assay using a control rabbit IgG antibody (IgG) or the anti-NFκB1 antibody during chromatin immunoprecipitation assay. (C) DNA enrichment of Pth1r P2 promoter, ratio between NFκB1 and IgG IP products, first part and last two parts of P2 were significantly enriched by NFκB1 antibody. Data shown as mean ± SD by unpaired Student’s t test, n=3 independent experiments for each group. (D, E) qPCR results of the expression levels of Nfkb1 and Pth1r in Nfkb1 siRNA-treated Zfp46 +/+ and Zfp467-/- calvarial osteoblasts (COBs) and bone marrow stromal cells (BMSCs). Data shown as mean ± SD by two-way ANOVA, n=3 independent experiments for each group. (F) Western blot analysis of Nfkb1 and Pth1r in Nfkb1 siRNA-treated Zfp467 +/+ and Zfp467-/- COBs and BMSCs. (G) Quantification for PTH1R protein level. Data shown as mean ± SD by two-way ANOVA, n=3 independent experiments for each group. NC, negative control. Figure 5—source data 1 Western blot for Figure 5B and F. https://cdn.elifesciences.org/articles/83345/elife-83345-fig5-data1-v2.zip Download elife-83345-fig5-data1-v2.zip Figure 5—source data 2 Western blot for Figure 5B p50 with IP samples. https://cdn.elifesciences.org/articles/83345/elife-83345-fig5-data2-v2.zip Download elife-83345-fig5-data2-v2.zip Figure 5—source data 3 Western blot for Figure 5C PTH1R. https://cdn.elifesciences.org/articles/83345/elife-83345-fig5-data3-v2.zip Download elife-83345-fig5-data3-v2.zip Figure 5—source data 4 Western blot for Figure 5C p50. https://cdn.elifesciences.org/articles/83345/elife-83345-fig5-data4-v2.zip Download elife-83345-fig5-data4-v2.zip Figure 5—source data 5 Western blot for Figure 5C ACTIN. https://cdn.elifesciences.org/articles/83345/elife-83345-fig5-data5-v2.zip Download elife-83345-fig5-data5-v2.zip In order to determine whether NFκB1 could bind to the Pth1r P2-2 promoter directly, DNA pulldown assay was performed using biotin-labeled Pth1r P2 promoter as a probe. As shown in Figure 6A, the biotin-Pth1rP2 group showed a specific band in both MC3T3-E1 nuclear extracts and purified NFκB1 protein, suggesting a direct physical interaction between NFκB1 and Pth1r P2 promoter. However, noticed that NFκB1 does not have a transcriptional activation domain, NFκB1 must heterodimerize with other transcription factors in order to increase gene transcription. Using String database and checking published studies, eight candidates that might heterodimerize with NFκB1 to regulate gene transcription were obtained: NFYC, NPAS1, Rel, AKAP8, RelA, RelB, ANKRD42, and HDAC1. Using siRNA to knock down all these potential NFκB1 partners (Figure 6B), the upregulated Pth1r induced by Nfkb1 overexpression could only be dampened by Npas1 and Relb siRNA (Figure 6C). Further co-immunoprecipitation (co-IP) results confirmed that NFκB1 could heterodimerize with RelB only (Figure 6D), which suggested that NFκB1-Relb heterodimers may drive greater Pth1r transcription in Zfp467-/- cells. Figure 6 with 1 supplement see all Download asset Open asset NFκB1 heterodimerize with RelB to transactivate the expression of Pth1r. (A) DNA pulldown assay with biotin-labeled Pth1r P2. MC3T3-E1 nuclear extracts or NFκB1 recombinant protein was probed with biotin-Pth1rP2 and then subjected to immunoblotting using NFκB1 antibody. (B) qPCR results of the expression levels of Nfyc, Npas1, Rel, Akap8, Rela, Ankrd42, Relb, and Hdac1 in related siRNA-treated MC3T3-E1 cells. Data shown as mean ± SD by one-way ANOVA, n=3 independent experiments for each group. NC, negative control. (C) qPCR results of the expression levels of Pth1r in Nfkb1 overexpression plasmid and Nfyc, Npas1, Rel, Akap8, Rela, Ankrd42, Relb, or Hdac1 siRNA co-transfected MC3T3-E1 cells. NC, negative control. (D) IP results using NFκB1 antibody in MC3T3-E1 protein extracts, IgG was used as a negative control. (E) Protein level of p-p105 and p50 in Zfp467 knockdown MC3T3-E1 cells. Data shown as mean ± SD by unpaired Student’s t test, n=3 independent experiments for each group. (F) Protein and mRNA level of NFκB-inducing kinase (NIK) in control and Zfp467 siRNA-treated MC3T3-E1 cells. Data shown as mean ± SD by unpaired Student’s t test, n=3 independent experiments for eac