It is generally accepted that cell shape plays a pivotal role in many aspects of cell behavior including proliferation, differentiation, and gene expression. Although previous reports have shown that implant-surface topography can alter cell shape in vitro, in vivo evidence for such an effect is largely based on intuitive interpretation of two-dimensional histological sections. The objective of this in vivo study was to develop a method to reconstruct in three dimensions the shape of epithelial cells attached to smooth and micromachined implant surfaces. Titanium-coated epoxy replicas of smooth and 10-mu m-deep micromachined grooved surfaces were implanted percutaneously in the parietal region of rats. After 7 days the implants and attached tissue were removed and processed for light and electron microscopy. One-micrometer-thick serial histological sections were used to trace and digitize cells and their nuclei into a Hewlett Packard computer. Three-dimensional images were reconstructed and rotated to measure length, width, height, area, orientation index and the angle cells or nuclei formed with the long axis of the grooves (XY angle), and the angle cells or nuclei formed with the long axis of the implant (XZ angle). Epithelial cells attached to the smooth surfaces were found to be significantly (P < .05) flatter and more spread than were the cells attached to the grooved surfaces. Cells on the smooth surfaces were aligned parallel with the long axis of the implant, whereas an the grooved surfaces cells were oriented obliquely with the implant. There was a strong correlation (r = 0.9) between measurements of parameters of cells and those of nuclei on the smooth surfaces, but little correlation was found for the cells and their nuclei on grooved surfaces. These measurements indicated that cell shape is altered by the topography of the implant surface in vivo, but that one cannot necessarily predict the shape of a cell from the shape of its nucleus. (C) 1995 John Wiley & Sons, Inc.
