Transient overexpression of sonic hedgehog alters the architecture and mechanical properties of trabecular bone

Abstract

Bone formation and remodeling involve coordinated interactions between osteoblasts and osteoclasts through signaling networks involving a variety of molecular pathways. We hypothesized that overexpression of Sonic hedgehog (Shh), a morphogen with a crucial role in skeletal development, would stimulate osteoblastogenesis and bone formation in adult animals in vivo. Systemic administration of adenovirus expressing the N-terminal form of Shh into adult mice resulted in a primary increase in osteoblasts and their precursors. Surprisingly, however, this was associated with altered trabecular morphology, decreased bone volume, and decreased compressive strength in the vertebrae. Whereas no change was detected in the number of osteoclast precursors, bone marrow stromal cells from Shh-treated mice showed enhanced osteoclastogenic potential in vitro. These effects were mediated by the PTH/PTH-related protein (PTHrP) pathway as evidenced by increased sensitivity to PTH stimulation and upregulation of the PTH/PTHrP receptor (PPR). Together, these data show that Shh has stimulatory effects on osteoprogenitors and osteoblasts in adult animals in vivo, which results in bone remodeling and reduced bone strength because of a secondary increase in osteoclastogenesis.

FIG. 1

Remodeling of vertebral trabecular bone induced by elevated levels of Shh. Vertebrae from C57BL/6 mice were harvested 18 days after intravenous administration of adenovirus encoding a soluble form of Shh (AdShhN), a control adenovirus without a transgene (AdNull), or PBS. Microscopic morphology of mouse vertebrae were examined in histologic sections stained with H&E by brightfield microscopy or by μCT. (A) H&E images of vertebrae. PBS (left; AdNull (middle); and AdShhN (right). AdShhN-treated vertebrae show remodeling of trabecular architecture. Bar = 200 μm. (B) Total trabecular area in H&E-stained sagittal sections (n = 5 for all groups). (C) Reconstructed μCT image of a 370-μm midline sagittal section through the lumbar level 6 (L₆) vertebra. (D) Tissue mineral density of L₆ vertebrae (n = 5 for all groups). For B and D, data shown as mean ± SE.

FIG. 2

Decreased biomechanical properties of AdShhN-treated vertebrae. Isolated lumbar level 6 (L₆) vertebra (n = 10 vertebrae/group) were oriented axially relative to the load and compressed at 0.05 mm/s to a maximum load of 75 N and then ashed after the mechanical testing. (A) Compressive stiffness. (B) Ultimate force. Data shown as mean ± SD.

FIG. 3

Osteoblasts and osteoclasts increase in vertebrae 18 days after AdShhN treatment. (A) Immunohistochemical staining for procollagen I, a cytoplasmic marker of osteoblasts: PBS (left); AdNull (middle); and AdShhN (right). Brown indicates positive staining. Bar = 40 μm. (B) Immunohistochemical staining for cathepin K, a marker of active osteoclasts: PBS (left); AdNull (middle); and AdShhN (right). Bar = 40 μm. (C–E) Quantification of osteoblast and osteoclast numbers (n = 5 for all groups). (C) Osteoblasts/mm of trabecular area. (D) Osteoclasts/mm of trabecular area. (E) Osteoblast/osteoclast ratio. For C–E, data shown as mean ± SE.

FIG. 4

Immature osteoblasts, osteoblast precursors, and mesenchymal stem cells increase with AdShhN treatment. (A) Vertebrae stained for runt-related transcription factor-2 (Runx-2), a marker of early osteoblasts 18 days after treatment: PBS (left); AdNull (middle); and AdShhN (right). Brown indicates positive staining. Bar = 40 μm. (B) Quantification of Runx-2–positive cells/mm of trabecular surface area (n = 5 for all groups). (C–E) Osteoblastogenesis and CFU-F assays were performed using cultured bone marrow cells from AdShhN and control mice. Alizarin red-S staining for mineralization after 3 wk of culture in osteogenic media: PBS (left); AdNull (middle); and AdShhN (right). Bar = 80 μm. (D) Alizarin red-S concentration. (E) CFU-F number per well after 14 days. For B, D, and E, data shown as mean ± SE.

FIG. 5

No change in the number of pre-osteoclasts, but enhanced osteoclastogenic potential of bone marrow stromal cells after AdShhN treatment. (A and B) Osteoclast precursors were harvested from bone marrow 18 days after AdShhN treatment and cultured with osteoclast-inducing medium for 6 days. (A) Examples of osteoclast cultures stained with TRACP (PBS, left; AdNull, middle; and AdShhN, right; bar = 100 μm) and (B) quantification of TRACP⁺ osteoclasts. Total number of TRACP⁺ cells with greater than three nuclei per cell were counted (n = 5, all groups). (C and D) Osteoclastogenesis was also measured in precursors harvested from naive mice and co-cultured on monolayers of bone marrow stroma established from AdShhN-treated or control mice. Co-cultures were supplemented with dihydroxyvitamin D₃ and PTH. (C) Examples of co-cultures stained with TRACP (PBS, left; AdNull, middle; and AdShhN, right; bar = 100 μm) and (D) quantification of osteoclasts in co-cultures. For B and D, data shown as mean ± SE. (E) Expression levels of OPG and RANKL expression measured in RNA extracted from femoral bone marrow cells 18 days after vector administration and quantified by TaqMan real-time PCR (n = 5, all groups). Data shown as mean ± SE.

FIG. 6

Enhanced osteoclastogenesis is associated with increased expression and activation of the PTH/PTHrP pathway. (A) Expression level of PPR measured in RNA extracted from femoral bone marrow cells 18 days after vector administration and quantified by TaqMan real-time PCR (n = 5, all groups). Data shown as mean ± SE. (B) Quantification of osteoclasts in co-cultures of marrow cells from animals treated with AdShhN or control and untreated osteoclast precursors cultured with (+) or without (−) PTHrP blocker. Cells were stained with TRACP on day 6 of culture and counted under brightfield microscopy (n = 5, all groups).