The Notch Ligand Jagged1 Regulates the Osteoblastic Lineage by Maintaining the Osteoprogenitor Pool

Abstract

Notch signaling is critical for osteoblastic differentiation; however, the specific contribution of individual Notch ligands is unknown. Parathyroid hormone (PTH) regulates the Notch ligand Jagged1 in osteoblastic cells. To determine if osteolineage Jagged1 contributes to bone homeostasis, selective deletion of Jagged1 in osteolineage cells was achieved through the presence of Prx1 promoter-driven Cre recombinase expression, targeting mesenchymal stem cells (MSCs) and their progeny (PJag1 mice). PJag1 mice were viable and fertile and did not exhibit any skeletal abnormalities at 2 weeks of age. At 2 months of age, however, PJag1 mice had increased trabecular bone mass compared to wild-type (WT) littermates. Dynamic histomorphometric analysis showed increased osteoblastic activity and increased mineral apposition rate. Immunohistochemical analysis showed increased numbers of osteocalcin-positive mature osteoblasts in PJag1 mice. Also increased phenotypically defined Lin^- /CD45^- /CD31^- /Sca1^- /CD51⁺ osteoblastic cells were measured by flow cytometric analysis. Surprisingly, phenotypically defined Lin^- /CD45^- /CD31^- /Sca1⁺ /CD51⁺ MSCs were unchanged in PJag1 mice as measured by flow cytometric analysis. However, functional osteoprogenitor (OP) cell frequency, measured by Von Kossa⁺ colony formation, was decreased, suggesting that osteolineage Jagged1 contributes to maintenance of the OP pool. The trabecular bone increases were not due to osteoclastic defects, because PJag1 mice had increased bone resorption. Because PTH increases osteoblastic Jagged1, we sought to understand if osteolineage Jagged1 modulates PTH-mediated bone anabolism. Intermittent PTH treatment resulted in a significantly greater increase in BV/TV in PJag1 hind limbs compared to WT. These findings demonstrate a critical role of osteolineage Jagged1 in bone homeostasis, where Jagged1 maintains the transition of OP to maturing osteoblasts. This novel role of Jagged1 not only identifies a regulatory loop maintaining appropriate populations of osteolineage cells, but also provides a novel approach to increase trabecular bone mass, particularly in combination with PTH, through modulation of Jagged1. © 2017 American Society for Bone and Mineral Research.

Keywords: GENETIC ANIMAL MODELS; NOTCH; OSTEOBLAST; PARATHYROID HORMONE; STROMAL/STEM CELL.

Fig. 1

Jagged1 expression in osteolineage cells increases with differentiation and is specifically depleted in Prx1CrexJag1 fl/fl (PJag1) mice. (A) In vitro culture of differentiating osteoblastic cells harvested from WT calvaria from day 5 to day 25, stained with alkaline phosphatase (pink) and von Kossa (black). (B,C) Real-time qPCR quantifying Osteocalcin (B) and Jagged1 (C) RNA expression. Gene expression levels relative to β-actin were calculated and normalized to day 0. *p < 0.05 compared to day 0 (n = 5). (D) Schematic representation of targeting achieved by the Prx1-Cre transgene. (E) Real-time qPCR quantifying Jagged1 gene expression levels relative to β-actin from freshly isolated OLBACs (n = 9). (F) Western blot analyses of Jagged1 in protein extracts from BMSCs of 2- to 3-month-old WT and PJag1 mice expanded in vitro for 10 days. (G) Representative immunohistochemical analysis of medial longitudinal sections of the femur of WT and PJag1 mice: Jagged1 protein presence is indicated by brown staining. Magnification ×40. Scale bar = 20 μm. *p < 0.05 compared to WT. MSC = mesenchymal stem cells; OP = osteoprogenitor; Pre-Ob = pre-osteoblast; Ob = osteoblast.

Fig. 2

Trabecular bone mass is increased in PJag1 mice at the trabecular region distal to the growth plate. (A) Representative μCT 3D image reconstructions of tibia and femur cross-sections in WT and PJag1 littermate mice. (B–E) Quantification of femur bone volume/total volume (BV/TV, in B), trabecular (Tb) number (C), Tb spacing (D), and Tb thickness (E). (F) Representative μCT image identifying region of the hindlimb (tibia) classified as proximal and distal with respect to the growth plate (white dashed line: growth plate). (G,H) μCT quantification of trabecular parameters of the proximal (G) and distal (H) regions of the femur and tibia, in WT and PJag1 mice. For each experimental group, n = 5 littermate mice. Data are presented as mean ± SE. *p < 0.05.

Fig. 3

Osteolineage Jagged1 deletion increases osteoblastic function. (A) Calcein double labeling shown in representative longitudinal sections of tibia trabecular region. (B–E) Histomorphometric quantification of mineral apposition rate (MAR, in B), double labeled surface/bone surface percent (Dls/Bs, in C), bone formation rate/bone surface (BFR/BS, in D), and BV/TV (E). WT: n = 4, PJag1: n = 6. All data are presented as mean ± SE. *p < 0.05.

Fig. 4

Osteolineage Jagged1 deletion increases the number of osteoblastic cells without changes in the MSCs but with depletion of osteoprogenitors. (A) Schematic representation highlighting the osteoblastic population thought to express Osteocalcin. (B) Expression analysis of osteocalcin gene by qPCR in OLBACs from long bones of WT and PJag1 mice (n = 9). (C) Representative immunohistochemical analysis of medial longitudinal sections of the femur of WT and PJag1 mice: Osteocalcin protein is indicated by brown staining. Magnification ×40. Scale bar = 20 μm. (D) Quantification of Osteocalcin-positive cells (n = 4). (E) Quantification of TUNEL-positive cells on longitudinal sections of the femur of WT and PJag1 mice (n = 3 per experimental group). (F) Real-time qPCR quantifying cyclin d1 RNA expression in OLBACs from long bones of WT and PJag1 mice (n = 3 per experimental group). Data are presented as mean ± SE. All analyses were performed with sex-matched littermates. (G) In vitro culture of osteoprogenitors (osteoblastic colony forming unit [CFU-OB]) using limiting dilution analysis of BMSCs (WT: n = 6. PJag1: n = 4). (H) Quantification of osteoprogenitor frequency by L-Calc software. (I) Quantification of functional MSCs by in vitro culture of fibroblastic colony forming unit (CFU-F) assay of BMSCs stained with methyl violet (WT: n = 6, PJag1: n = 6). (J) Gating strategy for flow cytometric analysis of MSC and osteoblastic cells (OBC). (K) Quantification of phenotypically defined Lin⁻CD45⁻CD31⁻Sca1⁺CD51⁺ MSCs (WT: n = 4, PJag1: n = 5). (L) Quantification of phenotypically defined Lin⁻CD45⁻CD31⁻Sca1⁻CD51⁺ OBCs (WT: n = 4, PJag1: n = 5) by flow cytometry. All data are presented as mean ± SE. *p < 0.05.

Fig. 5

Osteolineage Jagged1 modulates osteoclastogenesis and bone resorption. (A) Bone marrow serum CTX measured by ELISA. WT: n = 5, PJag1: n = 6. (B,C) Histomorphometric quantification of osteoclast surface normalized to bone surface (OC.S/BS), osteoclast number normalized to trabecular bone perimeter (#OC/BS). (D) Representative TRAP enzymatic staining (pink) of medial longitudinal sections through the femur in WT and PJag1 mice. Magnification ×60. Scale bar = 10 μm. (E) Percent of TRAP-positive cells organized by size of osteoclast surface covered by each individual osteoclast (WT: n = 5, PJag1: n = 6). (F–I) Histomorphometric quantification of osteoclast surface normalized to bone surface (OC.S/BS, in F and G), osteoclast number normalized to trabecular bone perimeter (#OC/BS, in H and I) in the proximal and distal regions of femur (F, H) and tibia (G, I) of WT and PJag1 mice. (J,K) Representative images (J) and quantification (K) of TRAP-stained cocultures of WT or PJag1 bone associated cells (BAC) and WT bone marrow. BAC numbers and dose of dehydroxy vitamin D (D3) are shown. n = 6 per experimental group. (L,M) Quantification of OPG (L) and RANK-L (M) gene expression by qRT-PCR in OLBACs collected from long bones of WT and PJag1 mice. WT: n = 9, PJag1: n = 9. All data are presented as mean ± SE. *p < 0.05.

Fig. 6

PTH enhances PJag1 trabecular mass in mice. (A) Representative μCT 3D reconstruction of trabecular bone cross-sections of the femur and tibia in WT and PJag1 mice treated with vehicle (VEH) or PTH. (B–E) Quantification of μCT parameters from femur (B,D) and tibia (C,E) in WT and PJag1 littermates treated with either vehicle or intermittent PTH (n = 5 mice per experimental group), focusing on the proximal (B,C) and distal (D,E) regions All data are presented as mean ± SE. *p < 0.05.

Fig. 7

Model of Jagged1 maintenance of osteolineage cells. While Notch acts to restrain the transition between MSCs and osteoprogenitors as well as that from osteoprogenitor to maturing osteoblasts, osteolineage Jagged1 provides restriction of osteoblast cell numbers while preserving osteoprogenitor cell pool.