Osteoinductive hydroxyapatite-coated titanium implants.

Previous studies have shown that heterotopic induction of bone formation by calcium phosphate-based macroporous constructs is set into motion by the geometry of the implanted substrata, i.e. a sequence of repetitive concavities assembled within the macroporous spaces. The aim of this study was to construct osteoinductive titanium implants that per se, and without the exogenous application of the osteogenic soluble molecular signals of the transforming growth factor-ß supergene family, would initiate the induction of bone formation. To generate intrinsically osteoinductive titanium implants for translation in clinical contexts, titanium grade Ti-6A1-4V cylinders of 15 mm in length and 3.85 mm in diameter, with or without concavities, were plasma sprayed with crystalline hydroxyapatite resulting in a uniform layer of 30 µm in thickness. Before coating, experimental titanium implants were prepared with a sequence of 36 repetitive concavities 1600 µm in diameter and 800 µm in depth, spaced a distance of 1000 µm apart. Mandibular molars and premolars were extracted to prepare edentulous mandibular ridges for later implantation. Planar and geometric hydroxyapatite-coated titanium constructs were implanted in the left and right edentulized hemi-mandibles, respectively, after a healing period of 7-8 months, 3 per hemi-mandible. Three planar and three geometric implants were implanted in the left and right tibiae, respectively; additionally, planar and geometric constructs were also inserted in the rectus abdominis muscle. Six animals were euthanized at 30 and 90 days after implantation; one animal had to be euthanized 5 days after surgery and the remaining animal was euthanized 31 months after implantation. Undecalcified longitudinal sections were precision-sawed, ground and polished to 40-60 µm; all sections were stained with a modified Goldner's trichrome. Undecalcified specimen block preparation was performed using the EXAKT precision cutting and grinding system. Histomorphometric analyses of bone in contact (BIC) showed that on day 30 there was no difference between the geometric vs. planar control implants; on day 90, the ratio of BIC to surface within the geometric implants was greater than on the standard planar implants in both mandibular and tibial sites; 31 months after implantation, selected concavities cut into the geometric implants harvested from the rectus abdominis muscle showed the spontaneous induction of bone formation with mineralized bone surfaced by osteoid seams. These data in non-human primates indicate that geometrically-constructed plasma-sprayed titanium implants are per se osteogenic, the concavities providing a unique microenvironment to initiate bone differentiation by induction.

Synergistic induction of bone formation by hOP-1, hTGF-beta3 and inhibition by zoledronate in macroporous coral-derived hydroxyapatites.

Thirty coral-derived calcium carbonate-based macroporous constructs with limited hydrothermal conversion to hydroxyapatite (7% HA/CC) were implanted in the rectus abdominis of three adult non-human primate Papio ursinus to investigate the intrinsic induction of bone formation. Macroporous constructs with 125 microg human recombinant osteogenic protein-1 (hOP-1) or 125 microg human recombinant transforming growth factor-beta(3) (hTGF-beta(3)) were also implanted. The potential synergistic interaction between morphogens was tested by implanting binary applications of hOP-1 and hTGF-beta(3) 5:1 by weight, respectively. To evaluate the role of osteoclastic activity on the implanted macroporous surfaces, coral-derived constructs were pre-loaded with 0.24 mg of bisphosphonate zoledronate (Zometa). To correlate the morphology of tissue induction with osteogenic gene expression and activation, harvested specimens on day 90 were analyzed for changes in OP-1 and TGF-beta(3) mRNA synthesis by quantitative real-time polymerase chain reaction (qRT-PCR). The induction of bone formation in 7% HA/CC solo correlated with OP-1 expression. Massive bone induction formed by binary applications of the recombinant morphogens. Single applications of hOP-1 and hTGF-beta(3) also resulted in substantial bone formation, not comparable however to synergistic binary applications. Zoledronate-treated macroporous constructs showed limited bone formation and in two specimens bone formation was altogether absent; qRT-PCR showed a prominent reduction of OP-1 gene expression whilst TGF-beta(3) expression was far greater than OP-1. The lack of bone formation by zoledronate-treated specimens indicates that osteoclastic activity on the implanted coral-derived constructs is critical for the spontaneous induction of bone formation. Indirectly, zoledronate-treated samples showing lack of OP-1 gene expression and absent or very limited bone formation by induction confirm that the spontaneous induction of bone formation by coral-derived macroporous constructs is initiated by secreted BMPs/OPs, in context the OP-1 isoform.