Inhibition of Midkine Augments Osteoporotic Fracture Healing

Abstract

The heparin-binding growth and differentiation factor midkine (Mdk) is proposed to negatively regulate osteoblast activity and bone formation in the adult skeleton. As Mdk-deficient mice were protected from ovariectomy (OVX)-induced bone loss, this factor may also play a role in the pathogenesis of postmenopausal osteoporosis. We have previously demonstrated that Mdk negatively influences bone regeneration during fracture healing. Here, we investigated whether the inhibition of Mdk using an Mdk-antibody (Mdk-Ab) improves compromised bone healing in osteoporotic OVX-mice. Using a standardized femur osteotomy model, we demonstrated that Mdk serum levels were significantly enhanced after fracture in both non-OVX and OVX-mice, however, the increase was considerably greater in osteoporotic mice. Systemic treatment with the Mdk-Ab significantly improved bone healing in osteoporotic mice by increasing bone formation in the fracture callus. On the molecular level, we demonstrated that the OVX-induced reduction of the osteoanabolic beta-catenin signaling in the bony callus was abolished by Mdk-Ab treatment. Furthermore, the injection of the Mdk-Ab increased trabecular bone mass in the skeleton of the osteoporotic mice. These results implicate that antagonizing Mdk may be useful for the therapy of osteoporosis and osteoporotic fracture-healing complications.

Fig 1. Midkine-antibody (Mdk-Ab) treatment accelerated osteoporotic fracture healing.

Biomechanical, micro-computed tomography (μCT) and histomorphometric analysis of the fractured femurs at day 23. Biomechanical testing: A) relative flexural rigidity of the fractured femur in comparison with intact femur determined by biomechanical testing. Parameters determined by μCT analysis (volume of interest 2): B) bone volume to tissue volume ratio and C) tissue volume (n = 6–7 per group). Parameters determined by histomorphometric analysis of the whole fracture callus at day 10 (white bars) and day 23 (grey bars): D) bone area to tissue area ratio, E) cartilage area to tissue area ratio and F) fibrous tissue area to tissue area ratio. *Significantly different from sham+vehicle or OVX+vehicle group (p<0.05) by Kruskal-Wallis test. (n = 5–7 per group.) G) Representative images of sections from undecalcified femurs at day 23, stained using Giemsa; scale bar: 250 μm.

Fig 2. Beta-catenin-positive area was decreased after ovariectomy (OVX) and increased after midkine-antibody (Mdk-Ab) treatment.

A) Sections of fractured femurs from four mice of each group were stained for beta-catenin and counterstained using hematoxylin. Representative images of the periosteal callus proximal to the osteotomy gap at day 10 after fracture are shown. Scale bar left column: 100 μm. Scale bar right column: 50 μm. B) Quantification of the beta-catenin-positive area in %. (n = 4 per group.) *Significantly different from sham+vehicle or OVX+vehicle group (p>0.05). C) Sections of fractured femurs from four mice of each group were stained for active beta-catenin and counterstained using hematoxylin. Representative images of the periosteal callus proximal to the osteotomy gap at day 10 after fracture are shown. Scale bar: 100 μm. C = cortex; HC = hypertrophic chondrocytes; OB = osteoblasts; PC = proliferating chondrocytes. Upper row: vehicle treatment; lower row: Mdk-Ab treatment.

Fig 3. Midkine-antibody (Mdk-Ab) treatment increased the bone content in the intact femur of ovariectomized (OVX) mice after short-time treatment.

Micro-computed tomography (μCT) analysis of the cortical bone at the midshaft of the intact femur volume of interest 2 (VOI 2): A) cortical tissue mineral density (TMD) and B) cortical thickness. μCT analysis of the distal part of the intact femur (VOI 3): C) trabecular TMD, D) bone volume to tissue volume ratio, E) trabecular thickness and F) trabecular number. *Significantly different from sham+vehicle or OVX+vehicle group (p<0.05) by Kruskal-Wallis test. (n = 6–7 per group.) G) Three-dimensional reconstructions of the trabecular region of the distal intact femur (VOI 3), representative images are shown.

Fig 4. Midkine-antibody (Mdk-Ab) treatment increased the bone content in vertebral bodies of ovariectomized (OVX) mice after short-time treatment.

Micro-computed tomography (μCT) analysis of the vertebral bodies (volume of interest 4): A) two-dimensional images of the second caudal vertebral body, representative images are shown. B) Trabecular tissue mineral density, C) trabecular bone volume to tissue volume ratio, D) trabecular thickness and E) trabecular number. *Significantly different from sham+vehicle or OVX+vehicle group (p<0.05) by Kruskal-Wallis test. (n = 6−7 per group).