Negative feedback between PTH1R and IGF1 through the Hedgehog pathway in mediating craniofacial bone remodeling

Abstract

Regeneration of orofacial bone defects caused by inflammation-related diseases or trauma remains an unmet challenge. Parathyroid hormone 1 receptor (PTH1R) signaling is a key mediator of bone remodeling whereas the regulatory mechanisms of PTH1R signaling in oral bone under homeostatic or inflammatory conditions have not been demonstrated by direct genetic evidence. Here, we observed that deletion of PTH1R in Gli1+ progenitors led to increased osteogenesis and osteoclastogenesis. Single-cell and bulk RNA-Seq analysis revealed that PTH1R suppressed the osteogenic potential of Gli1+ progenitors during inflammation. Moreover, we identified upregulated IGF1 expression upon PTH1R deletion. Dual deletion of IGF1 and PTH1R ameliorated the bone-remodeling phenotypes in PTH1R-deficient mice. Furthermore, in vivo evidence revealed an inverse relationship between PTH1R and Hedgehog signaling, which was responsible for the upregulated IGF1 production. Our work underscored the negative feedback between PTH1R and IGF1 in craniofacial bone turnover and revealed mechanisms modulating orofacial bone remodeling.

Keywords: Bone biology; Bone development; Development; G protein–coupled receptors; Osteoclast/osteoblast biology.

Figure 1. PTH1R deletion causes decreased alveolar bone volume and PDL malformation.

(A–F) Immunofluorescence staining of PTH1R of Gli1^CreER PTH1R^fl/+ Rosa26^Ai14 and Gli1^CreER PTH1R^fl/fl Rosa26^Ai14 female mice at P42. (B, C, E, and F) Higher magnification of boxed regions, respectively. n = 3. (G) Three-dimensional micro-CT reconstruction of Gli1^CreER PTH1R^fl/+ Rosa26^Ai14 and Gli1^CreER PTH1R^fl/fl Rosa26^Ai14 mandibles at P42. n = 5 for male and n = 6 for female Gli1^CreER PTH1R^fl/+ Rosa26^Ai14 mice. n = 4 for male and n = 8 for female Gli1^CreER PTH1R^fl/fl Rosa26^Ai14 mice. (H) Quantitative micro-CT analysis of BV/TV, Tb.Th, Tb.Sp, and Tb.N of both sexes in each genotype. (I and L) HE staining of Gli1^CreER PTH1R^fl/+ Rosa26^Ai14 and Gli1^CreER PTH1R^fl/fl Rosa26^Ai14 female mice at P42. (J and M) Enlarged boxed areas of alveolar root furcation of mandibular first molar showed substantially reduced bone volume in PTH1R-cKO mice. (K and N) Higher magnification of PDL region showed the PDL space was replaced by bony tissue in PTH1R-cKO mice. n = 3. (O) Two-dimensional micro-CT images of coronal sections showed narrowed PDL space in PTH1R-cKO mice. n = 6. (P) Quantitative analysis of PDL width. n = 6. Female mice were used. (Q) Immunofluorescence staining of periostin (POSTN) of Gli1^CreER PTH1R^fl/+ Rosa26^Ai14 and Gli1^CreER PTH1R^fl/fl Rosa26^Ai14 female mice at P42. Boxed areas are shown at higher magnification. n = 3. Scale bar = 200 μm (A, B, I, and L), 50 μm (B, C, J, and K), 500 μm (G and O), and 100 μm (Q). Significance is determined using unpaired 2-sided Student’s t tests in H or using 2-way ANOVA with Tukey’s correction for multiple comparisons in D. Data are mean ± SEM. **P < 0.01, ****P < 0.0001.

Figure 2. Deletion of PTH1R in Gli1⁺ lineage cells stimulates bone remodeling.

(A) Double calcein labeling in the alveolar bone region of control and PTH1R-cKO mice at P42. n = 6. (B) Histomorphometry analysis of dynamic bone formation parameters at P42. n = 6. (C–H) Immunofluorescence staining of SP7 (C), RUNX2 (D), and COL1A1 (E) and quantification of SP7⁺Gli1⁺ (F), RUNX2⁺Gli1⁺ (G), and COL1A1⁺Gli1⁺ (H) cell number in PTH1R-cKO mice at P42. Boxed areas are shown at higher magnification. n = 5. (I) RT-qPCR analysis of osteogenesis-related gene expression (Sp7, Runx2, Alp, Spp1) in mandibles of PTH1R-cKO mice and control littermates at P42. n = 5 for control and n = 4 for PTH1R-cKO. (J) TRAP staining revealed increased TRAP⁺ osteoclast number in PTH1R-cKO mice at P42. Boxed areas are shown at higher magnification. n = 5. (K) Quantification of TRAP⁺ osteoclast number/bone surface. n = 5. (L) RT-qPCR analysis of osteoclastogenesis-related gene expression (Ctsk, Mmp9, Nfatc1, Atp6v0d2, Tnfrsf11a, Tnfsf11, Tnfrsf11b, Tnfsf11/Tnfrsf11b) in mandibles of PTH1R-cKO mice and control littermates at P42. n = 5 for control and n = 4 for PTH1R-cKO. Scale bar = 25 μm (A), 50 μm (C–E and J). Data were all obtained in female mice. Significance is determined using unpaired 2-sided Student’s t tests. Data are mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 3. PTH1R is a key regulator in inflammation-related bone disease.

(A) Uniform manifold approximation and projection (UMAP) visualization of aligned gene expression data showing 15 distinct clusters and cellular origin. The library consisted of cells extracted from mandibles of control mice (n = 8,340) and apical periodontitis (AP) mice (n = 6,808). (B) Violin plots of the expression of Pth1r and Gli1 in all clusters. UMI, unique molecular identifier. (C) UMAP visualization of Pth1r in all clusters. (D) UMAP visualization of 4 MSC subclusters. The cell numbers in each MSC subcluster were as follows: control: 102 and AP: 129 for the MSC population, control: 42 and AP: 52 for MSC_OLCs, control: 30 and AP: 45 for MSC_inflammatory cells, control: 22 and AP: 29 for MSC_endothelial cells, control: 8 and AP: 3 for MSC_neurological cells. (E) The expression level of Pth1r in reclustered MSC population. (F) Violin plot of the expression of Pth1r in 4 MSC subclusters. (G) Violin plot of the expression of Pth1r in alveolar bone using bulk RNA-Seq analysis under control and AP conditions. (H) Lineage-tracing results of Gli1⁺ cells in control mice under both homeostasis and AP conditions at P77. Yellow dashed line indicates interface between tooth root and PDL. n = 3. (I) Immunofluorescence staining of PTH1R in healthy or inflammatory microenvironment of control mice. Yellow dashed line indicates interface between tooth root and PDL. White arrows depict PTH1R⁺Gli1⁺ cells. n = 3. (J) Quantification of PTH1R⁺Gli1⁺ cell numbers in periapical bone of Gli1^CreER Rosa26^Ai14 mice under homeostasis and AP conditions. n = 3. Scale bar = 100 μm. Data were all obtained in male mice. Data are mean ± SEM. **P < 0.01, ***P < 0.001.

Figure 4. Gli1^CreER PTH1R^fl/fl mice exhibit restricted periapical lesion because of activated bone turnover.

(A) Two-dimensional micro-CT images of the mandibles from Gli1^CreER PTH1R^fl/+ Rosa26^Ai14 and Gli1^CreER PTH1R^fl/fl Rosa26^Ai14 mice in sham and AP groups. Red arrowheads depict the region of periapical lesion. n = 6 for sham and n = 5 for AP of Gli1^CreER PTH1R^fl/+ Rosa26^Ai14 mice. n = 6 for sham and n = 7 for AP of Gli1^CreER PTH1R^fl/fl Rosa26^Ai14 mice. (B) Quantitative analysis of trabecular bone parameters including BV/TV (%), Tb.Sp (mm), and Tb.Th (mm). (C) HE staining of the distal root of mandibular first molar showed the periapical lesion induced by AP. n = 3. (D and E) TRAP staining and quantification of TRAP⁺ cells/bone surface. n = 5. (F and G) Immunofluorescence staining and quantification showed increasing Runx2⁺Gli1⁺ cell numbers in periapical bone of Gli1^CreER PTH1R^fl/fl Rosa26^Ai14 mice under both homeostasis and AP conditions. Yellow dashed lines depict the region of distal root of the mandibular first molar. Boxed areas are shown at higher magnification. n = 4. (H and I) Immunofluorescence staining and quantification showed increased Sp7⁺Gli1⁺ cell numbers in inflammatory periapical bone of Gli1^CreER PTH1R^fl/fl Rosa26^Ai14 mice. Yellow dashed lines depict the region of distal root of the mandibular first molar. Boxed areas are shown at higher magnification. n = 4. Scale bar = 500 μm (A) and 100 μm (C, D, F, and H). Data were all obtained in male mice. Significance is determined using 2-way ANOVA with Tukey’s correction for multiple comparisons. Data are mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 5. IGF1 plays an important role in regulating bone turnover under both homeostasis and inflammatory microenvironment.

(A) Schematic diagram of the experimental design. (B) Venn diagram showing coexpressed genes among Control_Sham, Control_AP, PTH1R-cKO_Sham, PTH1R-cKO_AP (FPKM > 1). Male mice were used. FPKM, fragments per kilobase million. (C) Heatmap of representative genes associated with osteogenesis, osteoclastogenesis, and IGF signaling. n = 2 for each group. (D) GO analysis of Control_Sham versus PTH1R-cKO_Sham enriched GO terms related to insulin-like growth factor, bone formation, and bone resorption. (E) RT-qPCR of Igf1 and Igf2 expression in mandibles under control and inflammation. n = 6 for control and n = 4 for PTH1R-cKO. Male mice were used. (F) Violin plot of the expression of Igf1 and Igf2 in all clusters. (G) The expression level of Igf1 in reclustered MSC population. (H) The expression level of Igf1 and Igf2 in 4 MSC subclusters presented in violin plot. (I and J) Immunofluorescence staining and quantification showed upregulated IGF1⁺Gli1⁺ cells in periapical bone of Gli1^CreER PTH1R^fl/fl Rosa26^Ai14 male mice under homeostasis and AP conditions. Yellow dashed lines depict the region of distal root of the mandibular first molar. Boxed areas are shown at higher magnification. n = 4. (K) HE staining of alveolar bone of healthy individuals and patients with AP. (L and N) Immunofluorescence double staining of PTH1R and IGF1 and quantification of IGF1⁺PTH1R⁺ in healthy and inflammatory alveolar bone of human samples. Boxed areas are shown at higher magnification. n = 3. (M) PTH1R and IGF1 gene expression of human healthy and inflammatory alveolar bone tissues. n = 4 in healthy individuals and n = 9 in patients with AP. Scale bar = 100 μm (J and K), 25 μm (L). Significance is determined using unpaired 2-sided Student’s t tests between 2 groups and 2-way ANOVA with Tukey’s correction for multiple comparisons. Data are mean ± SEM. *P < 0.05, **P < 0.01, ****P < 0.0001.

Figure 6. Lack of IGF1 in PTH1R-cKO mice reverses increased bone formation and bone resorption activities.

(A) HE staining of control, Gli1^CreER PTH1R^fl/+ IGF1^fl/+, Gli1^CreER PTH1R^fl/fl IGF1^+/+, and Gli1^CreER PTH1R^fl/fl IGF1^fl/+ mice at P14. Boxed areas are shown at higher magnification. n = 3. (B) Immunofluorescence staining of COL1A1 of each group. n = 3. (C and D) Immunofluorescence staining and quantification of Runx2 showed no difference between control and Gli1^CreER PTH1R^fl/+ IGF1^fl/+ mice. PTH1R-cKO mice displayed increased Runx2⁺ cell number, the trend of which was reversed in Gli1^CreER PTH1R^fl/fl IGF1^fl/+ mice. Boxed areas are shown at higher magnification. n = 4. (E and F) TRAP staining and quantification exhibited no difference between control and Gli1^CreER PTH1R^fl/+ IGF1^fl/+ mice. PTH1R-cKO mice had higher TRAP⁺ osteoclast numbers. The number of osteoclasts is in a downregulated trend in Gli1^CreER PTH1R^fl/fl IGF1^fl/+ mice compared with PTH1R-cKO mice. n = 6. (G) RT-qPCR results of Igf1, Runx2, Col1a1, and Tnfsf11 expression in control and IGF1-knockdown OMSCs cultured in osteogenic media. n = 3. (H) RT-qPCR results of Igf1, Runx2, and Col1a1 expression in control and IGF1-knockdown OMSCs cultured in TNF-α (10 ng/mL, MilliporeSigma) added to osteogenic media. n = 3. Scale bar = 200 μm (A), 50 μm (B, C, and E). Male mice were used. Significance is determined using 1-way ANOVA with Tukey’s correction in D and F and 2-way ANOVA with Tukey’s correction for multiple comparisons in G and H. Data are mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 7. Activated Hedgehog signaling contributes to the increased IGF1 in PTH1R-cKO mice.

(A and B) Immunofluorescence staining and quantification showed increased IGF1⁺Gli1⁺ cell number in root furcation area of Gli1^CreER PTH1R^fl/fl Rosa26^Ai14 female mice at P42. n = 5. (C) RT-qPCR results showed upregulated Igf1 in PTH1R-cKO OMSCs. n = 3. (D) Heatmap depicting the expression of Hh signaling–related genes analyzed by RNA-Seq in control and PTH1R-cKO alveolar bone samples. n = 2. (E) RT-qPCR results showed upregulated Hh signaling–related genes (Ptch1, Smo, Gli2, Hip1) in PTH1R-cKO OMSCs. n = 3. (F and H) Immunofluorescence staining of Ptch1 and Smo in Gli1⁺ progenitors of Gli1^CreER PTH1R^fl/+ Rosa26^Ai14 and Gli1^CreER PTH1R^fl/fl Rosa26^Ai14 mice at P42. n = 3–5. (G) Immunofluorescence staining of Ptch1 showed no difference between control and Gli1^CreER PTH1R^fl/+ IGF1^fl/+ mice. PTH1R-cKO mice had activated Ptch1 expression, which was downregulated in Gli1^CreER PTH1R^fl/fl IGF1^fl/+ compared with PTH1R-cKO mice. n = 3. (I) Schematic representation of the experimental design for cellular studies using siRNA. (J) Gene expression profile of Hh signaling–related markers and Igf1 in OMSCs of control and PTH1R-cKO after siRNA treatment. n = 6. Scale bar = 100 μm. Significance is determined using unpaired 2-sided Student’s t tests between 2 groups and 2-way ANOVA with Tukey’s correction for multiple comparisons. Data are mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.