Notch inhibits osteoblast differentiation and causes osteopenia

Abstract

Notch receptors are determinants of cell fate decisions. To define the role of Notch in the adult skeleton, we created transgenic mice overexpressing the Notch intracellular domain (NICD) under the control of the type I collagen promoter. First-generation transgenics were small and osteopenic. Bone histomorphometry revealed that NICD caused a decrease in bone volume, secondary to a reduction in trabecular number; osteoblast and osteoclast number were decreased. Low fertility of founder mice and lethality of young pups did not allow the complete establishment of transgenic lines. To characterize the effect of Notch overexpression in vitro, NICD was induced in osteoblasts and stromal cells from Rosa(notch) mice, in which a STOP cassette flanked by lox(P) sites is upstream of NICD, by transduction with an adenoviral vector expressing Cre recombinase (Cre) under the control of the cytomegalovirus (CMV) promoter (Ad-CMV-Cre). NICD impaired osteoblastogenesis and inhibited Wnt/beta-catenin signaling. To determine the effects of notch1 deletion in vivo, mice in which notch1 was flanked by lox(P) sequences (notch1(loxP/loxP)) were mated with mice expressing Cre recombinase under the control of the osteocalcin promoter. Conditional null notch1 mice had no obvious skeletal phenotype, possibly because of rescue by notch2; however, 1-month-old females exhibited a modest increase in osteoclast surface and eroded surface. Osteoblasts from notch1(loxP/loxP) mice, transduced with Ad-CMV-Cre and transfected with Notch2 small interfering RNA, displayed increased alkaline phosphatase activity. In conclusion, Notch signaling in osteoblasts causes osteopenia and impairs osteo-blastogenesis by inhibiting the Wnt/beta-catenin pathway.

Figure 1

Phenotype of NICD-overexpressing mice. A, Representative skeletal radiograph of a 4-wk-old WT control and a NICD transgenic mouse. B and C, Representative bone histomorphometry performed on femurs from 4-wk-old WT control and NICD transgenic mice stained with Von Kossa (B) and toluidine blue (C). Final magnifications, ×40 (B) and ×400 (C). Bars, 400 μm (B) and 40 μm (C). Arrows point to osteoblasts.

Figure 2

Effect of NICD overexpression on osteoblastic differentiation in bone marrow stromal cells harvested from Rosa^notch mice and transduced with Ad-CMV-GFP or Ad-CMV-Cre. A, Stromal cells were cultured for 3 d after confluence and examined for lox^P recombination by Southern blot analysis. B, Stromal cells were cultured to confluence, exposed to control medium (Veh), BMP-2 at 3 nm, or Wnt 3a at 2.7 nm for 72 h and APA quantified in extracts from cells transduced with control Ad-CMV-GFP (white bars) or Ad-CMV-Cre (black bars). APA is expressed as nanomoles of p-nitrophenol per minute per microgram of total protein. Bars, means ± sem for six observations. *, Significantly different from control cells, P < 0.05. C, Stromal cells transduced with control Ad-CMV-GFP or Ad-CMV-Cre were cultured to confluence; total RNA was reverse transcribed and amplified by real-time RT-PCR in the presence of specific primers. Data are expressed as notch1, hes1, alkaline phosphatase (Ap), and osteocalcin copy number, determined by real-time RT-PCR, corrected for gapdh expression. Values are means ± sem, n = 3. *, Significantly different from control cells, P < 0.05. D, Stromal cells were cultured to confluence, serum deprived overnight, and treated 16 h with Wnt 3a at 2.7 nm (+) or control vehicle (−) for the determination of cytosolic β-catenin levels. Cytosolic extracts were resolved by gel electrophoresis and transferred to Immobilon P membranes, which were incubated with antibodies to β-catenin (β-cat) or actin.

Figure 3

Effect of NICD overexpression on osteoblastic differentiation in calvarial osteoblasts harvested from Rosa^notch mice and transduced with Ad-CMV-GFP or Ad-CMV-Cre. A, Calvarial osteoblasts were cultured for 3 d after confluence and examined for lox^P recombination by Southern blot analysis. B, Calvarial osteoblasts were cultured to 70% confluence and transfected with 12xCSL-Luc, HES1-Luc, and a CMV/β-galactosidase expression vector. After 16 h, cells were switched to fresh DMEM for 24 h and harvested. Data shown represent luciferase/β-galactosidase activity for control Ad-CMV-GFP cells (white bars) or Ad-CMV-Cre (black bars). Bars, means ± sem for six observations. *, Significantly different from control cells, P < 0.05. C, Calvarial osteoblasts were cultured to confluence; exposed to control medium (Veh), BMP-2 at 3 nm, or Wnt 3a at 2.7 nm for 72 h; and APA quantified in extracts from cells transduced with Ad-CMV-GFP cells (white bars) or Ad-CMV-Cre (black bars). APA is expressed as nanomoles of p-nitrophenol per minute per microgram of total protein. Bars, Means ± sem for six observations. *, Significantly different from control cells, P < 0.05. D, Calvarial osteoblasts transduced with Ad-CMV-GFP or Ad-CMV-Cre were cultured to confluence; total RNA was reverse transcribed, and amplified by real-time RT-PCR in the presence of specific primers. Data are expressed as notch1 and osteocalcin copy number, determined by real-time RT-PCR, and corrected for rpl38 expression. Values are means ± sem, n = 4. *, Significantly different from control cells, P < 0.05.

Figure 4

Phenotype of conditional null notch1^Δ/Δ 4-wk-old female mice. A, Representative PCR analysis of calvarial DNA from conditional null notch1^Δ/Δ 4-wk-old mice and notch1^loxP/loxP littermate control mice. B, Von Kossa staining performed on femurs from 4-wk-old notch1^Δ/Δ female mice and notch1^loxP/loxP female littermate controls. Final magnification, ×40. Bar, 400 μm.

Figure 5

Effect of Notch signaling inhibition on osteoblastic differentiation in calvarial osteoblasts harvested from notch1^loxP/loxP mice transduced with Ad-CMV-GFP or Ad-CMV-Cre and transfected with Notch2 siRNA. Transduced osteoblasts were cultured to 70% confluence, and transfected with Notch2 (siNotch2) or control scrambled siRNA (siSCRAM). A, Deletion of Notch1 and down-regulation of Notch2 mRNA were documented in parallel cultures by real-time RT-PCR at the completion of the experiment. Data are expressed as notch1 (N1) and notch2 (N2) copy number, determined by real-time RT-PCR, corrected for gapdh expression. Values represent means ± sem for three observations. *, Significantly different from control cells, P < 0.05. B, Ad-CMV-GFP-transduced cells were transiently transfected with scrambled siRNA (white bars) and used as controls, whereas Ad-CMV-Cre-transduced cells were transiently transfected with notch2 siRNA (black bars). Twenty-four hours after transfection, cells were exposed to control medium (Veh), BMP-2 at 3 nm, or Wnt 3a at 2.7 nm and APA quantified in extracts from cells cultured for 72 h. APA is expressed as nanomoles of p-nitrophenol per minute per microgram of total protein. Bars, means ± sem for six observations. *, Significantly different from control cells, P < 0.05.