Osteoblast reaction at the interface between surface-active materials and bone in vivo: a study using in situ hybridization.

Osteoblast activation after implantation of two kinds of surface-active material in bone was investigated chronologically using in situ hybridization with digoxygenin-labeled procollagen alpha 1(I) complementary RNA probe. The bioactive materials used were hydroxyapatite (HA) and apatite- and wollastonite-containing glass-ceramic (A-W GC). A hole was drilled bilaterally in the distal epiphysis of rabbit femurs with subsequent implantation of HA or A-W GC cylinders in a press-fit manner. Specimens were collected at 3, 7, 14, and 28 days after operation and decalcified. Then the undecalcified implant cores were pushed out of the hole without causing damage to the bony side of the interface. In situ hybridization documented no qualitative differences in the expression of procollagen alpha 1(I) RNA between HA and A-W GC. Few osteoblasts at the bone-material interface showed a specific signal at day 3, whereas many osteoblasts were positive around the materials at days 7 and 14, indicative of active new bone formation. The positive osteoblasts seemed to originate from preexisting trabeculae and lined the trabeculae, newly formed bone, and material surface. At day 28, many osteoblasts lining material-surrounding bone were negative, whereas those in remodeling canals were positive, suggesting that the bone was in the remodeling stage after bone formation. These findings were comparable to those with beta-tricalcium phosphate in a previous study, thus suggesting osteoconductive bone formation on HA and A-W GC.

Temporal and spatial patterns of osteoblast activation following implantation of beta-TCP particles into bone.

Temporal and spatial patterns of osteoblast activation around beta-TCP particles implanted into bone were analyzed by in situ hybridization with digoxygenin-labeled procollagen alpha 1(I) RNA probes. beta-TCP particles (150-300 microns in diameter) were implanted into rat tibiae, and specimens were collected 3, 5, 7, 14, and 28 days after operation. Activated osteoblasts displayed intense procollagen alpha 1(I) RNA specific labeling. At day 3, osteoblasts lining pre-existing trabeculae in places showed a specific signal. Additionally, scattered activated cells compatible with preosteoblasts also were observed in the vicinity of the trabeculae among red blood cells that filled the space between beta-TCP particles. Osteoblast activation on the surface of beta-TCP rarely was observed. At days 5 and 7, osteoblast activation and bone formation advanced centripetally. At the forefront of bone formation positive cells were scattered in the blood cell clots, and some of the positive cells colonized forming new bone matrix. Formation of new bone did not always begin at the surface of beta-TCP. At day 14, most of the beta-TCP particles were tightly associated with newly formed bone, and the number of positive osteoblasts was reduced. At day 28, absorption of the newly formed bone and the beta-TCP by multinuclear cells was sporadically demonstrated. Such cells often were accompanied by active osteoblasts, suggesting early bone remodeling. In conclusion, in situ hybridization with procollagen alpha 1(I) was employed to demonstrate precisely the mode of recruitment of bone cell precursors. beta-TCP does not positively guide collagen I expressing bone cells along its surface. It has no apparent effects on bone regeneration.

Analysis of osteoblast activity at biomaterial-bone interfaces by in situ hybridization.

To investigate the effects of bioactive materials on bone formation in vivo, a new experimental model using in situ hybridization has been developed. A hole was drilled bilaterally in the distal epiphysis of rabbit femurs with subsequent implantations of beta-tricalcium phosphate (beta-TCP) cylinders in a press-fit manner. Specimens were collected at 3, 7, 14, and 28 days after operation. Femurs with empty drilling holes, and normal distal femurs without operation were used as controls. All specimens were decalcified and hybridized with a procollagen alpha 1(I) complementary RNA probe labeled with digoxygenin. In normal-bone sections, procollagen alpha 1(I) RNA was clearly demonstrated in periosteal osteoblasts, in osteoblasts in the mineralizing zone adjacent to growth plates, and in osteoblasts lining remodeling canals. As for beta-TCP, labeled osteoblasts around the material were not found at day 3, whereas they were most intensively observed at day 7 and a little less at day 14, in accordance with new-bone formation around the material. Weaker signals were also detected in fibroblasts at day 7. At day 28, osteoblasts lining the surface of newly formed bone were mainly negative, whereas those adjacent to the resorption sites of the beta-TCP showed positive signals, demonstrating an active remodeling at the material surface. The temporal expression of procollagen alpha 1(I) RNA in the beta-TCP specimens was fundamentally the same as that in the empty-hole specimens, suggesting no remarkable acceleration or suppression of bone-forming activity of osteoblasts by beta-TCP, which is consistent with osteoconductive bone formation. This in situ hybridization method was suggested to be a powerful tool in analyzing the biological effects of bioactive materials.