Difference in penetration of horseradish peroxidase tracer as a foreign substance into the peri-implant or junctional epithelium of rat gingivae.

Horseradish peroxidase (HRP) tracer was applied to the gingival sulcus of implants or natural teeth at 5, 25, or 50 mg/ml to investigate the sealing capacities of the peri-implant epithelium (PIE) and junctional epithelium (JE); the extent of HRP penetration was observed under electron microscopy. A Ti-6Al-4V implant was inserted either immediately (immediate implantation) or 2 weeks (delayed implantation) after extraction of the maxillary left first molar of rats. The JE of the right molar was used as a control. Although the whole PIE of undecalcified sections appeared to be attached to the implant surface, electron microscopically, the internal basement lamina (IBL) and hemidesmosomes were deficient in the coronal-middle region of the PIE. There were extensive extracellular deposits of HRP in the intercellular spaces between PIE cells, and HRP was blocked to some extent by the lamina lucida and lamina densa of the external basal lamina and basal cell junction. HRP was detected in the connective tissue under the PIE, but was not found in the connective tissue under the JE. Intracellularly, HRP was found in the vesicles and granules of PIE cells and JE cells. These were fewer in number in PIE cells than in JE cells. There were no differences between the findings for immediate and delayed implantation. The results indicate that a deficiency in the IBL permitted penetration of HRP from the gingival sulcus into the connective tissue under the PIE, and suggest that the endocytotic capacity of PIE cells is inferior to that of JE cells.