Automatic Actin Filament Quantification of Osteoblasts and Their Morphometric Analysis on Microtextured Silicon-Titanium Arrays

Abstract

Microtexturing of implant surfaces is of major relevance in the endeavor to improve biorelevant implant designs. In order to elucidate the role of biomaterial’s topography on cell physiology, obtaining quantitative correlations between cellular behavior and distinct microarchitectural properties is in great demand. Until now, the microscopically observed reorganization of the cytoskeleton on structured biomaterials has been difficult to convert into data. We used geometrically microtextured silicon-titanium arrays as a model system. Samples were prepared by deep reactive-ion etching of silicon wafers, resulting in rectangular grooves (width and height: 2 µm) and cubic pillars (pillar dimensions: 2 × 2 × 5 and 5 × 5 × 5 µm); finally sputter-coated with 100 nm titanium. We focused on the morphometric analysis of MG-63 osteoblasts, including a quantification of the actin cytoskeleton. By means of our novel software FilaQuant, especially developed for automatic actin filament recognition, we were first able to quantify the alterations of the actin network dependent on the microtexture of a material surface. The cells’ actin fibers were significantly reduced in length on the pillared surfaces <em>versus</em> the grooved array (4–5 fold) and completely reorganized on the micropillars, but without altering the orientation of cells. Our morpho-functional approach opens new possibilities for the data correlation of cell-material interactions.

Keywords

• actin cytoskeleton
• MG-63 osteoblasts
• microtexturing
• ridge detection
• image processing