Aims: In a recent study, chemical microroughening of bioactive glass surface was shown to promote attachment of osteoblastic cells and osseointegration of porous bioactive glass implant. The current in vivostudy employed molecular biologic techniques to clarify the osteogenic effects of smooth and microrough glass surfaces. Methods:Using a rat model, a portion of the medullary canal in the proximal tibia was evacuated and filled with microroughened or smooth bioactive glass microspheres. The primary bone healing response and subsequent remodelling were analysed at 1, 2, and 8 weeks, respectively. The expression of various genes for the bone matrix components (type I collagen, osteocalcin, osteopontin, osteonectin) and proteolytic enzymes (cathepsin K, MMP-9) were determined by Northern analysis. Results: The microroughened bioactive glass microspheres were found to induce higher mRNA levels for osteopontin and lower levels for osteonectin at 2 weeks after operation when compared to smooth control micropheres. At 8 weeks, the MMP-9 expression levels were significantly higher with microroughened bioactive glass microspheres. Conclusion: Microroughening of the bioactive glass surface triggered temporal changes in the expression of specific genes.

Aims: The present study examined the effect of ade-novirus-mediated recombinant human BMP-2 (RAd-BMP-2) gene therapy combined with bioactive glass (BG) microspheres in promotion of new bone formation. Methods: Harlan Dawley female rats (n=72) underwent unilateral surgery of right or left tibia in a random order. A round cortical window (⁻ 2.8 mm) was drilled into the anteromedial cortex of the proximal tibia. A smaller unicortical hole (⁻ 1.0 mm) was drilled 5 mm distally. Bone marrow was removed and the medullary space between the cortical holes was þlled with BG microspheres. Adenoviral vectors RAdBMP-2 carrying the BMP-2 gene or RAdLacZ harbouring the E. coli LacZ reporter gene were injected locally into the medullary spaces. The control defects were þlled with BG microspheres only. Empty control defects were left to heal without any þlling. The rats were killed 4 days, 2 and 8 weeks after surgery and the tibias were harvested for analyses. At each time point, six animals were used for pQCT, radiography, BEI-SEM and histomorphometric analyses. Results: All BG-þlled defects showed a time-related increase of intramedullary new bone. At 8 weeks, there was signiþcantly more new bone in defects treated with BG and RAdBMP-2 gene than in defects left to heal without þlling (p=0.003) (BG + RAdBMP-2: 25.0 ± 6.0% and empty control defects: 12.3 ± 3.8%). Also defects þlled with BG only showed higher new bone formation than empty control defects, but this was not statistically signiþcant (p=0.10) (BG: 19.9 ± 7.3%). Conclusions: The current study showed that local BMP-2 gene therapy enhances new bone formation on bioactive glass microspheres.