Page 88 - Physico-Chemical Niche Conditions for Bone Cells

PFF affects pre-osteoblast behavior
Cell vertical displacement and center of mass
The video of a live bone cell was performed by an SP8 lightning confocal microscope (Leica, Solms, Gemany). Prior to 4 h of PFF treatment, the cells were stained by Sir-actin (Spirochrome, Stein-am-Rhein, Switzerland) for F-actin at 37°C. The flow apparatus with a parallel-plate flow chamber (1.4×1.4×0.02 cm3) was connected to the confocal microscope. The video of the live bone cell was made for 109 sec (before PFF treatment: 0-11 sec, PFF treatment: 12-109 sec). The video data was analyzed by Image J software (https://imagej.net/Downloads). The quantification of cell vertical displacement was performed as follows: (1) Select a rectangle region as the landmark for alignment. (2) Matching method: Normalized correlation coefficient. (3) Search area (pixels around a region of interest (ROI)): 0 (Template will be searched on the whole image if search area = 0). (4) Select subpixel registration (Interpolation method for subpixel translation: Bilinear). (5) Select show align coordinates in results table. Thereafter, the bottom of the cell was selected in the confocal image using the rectangular selection tool for image matching. (6) Measure the center of mass (coM) displacement. “Center of mass” was selected in “Set Measurements” via “Analyse” in Image J software. The whole stack or frames (video) was measured by using “Measure Stack”. In addition, the 60rd frame was equal to 11th sec, i.e. 1 frame = 0.18 sec.
Computational fluid dynamics modeling
Geometry reconstructions
Z-stack images of a pre-osteoblast at different time points were imported into a solid modeling
software (CATIA V5R21, Dassault Sytemes, Veilizy-Villacoublay, France) to convert the
volume, 1.2×10 mm ; length, 83 μm; height, 7 μm) and a glass slide (length×width×height, 24×24×0.15 mm; total volume, 86.4 mm3) in the parallel-plate flow chamber were constructed using commercial finite element software (COMSOL Multiphysics v5.4, Stockholm, Sweden; Fig. 1B) at the same z-coordinates intervals as in the experiments. PFF was initiated by allowing the fluid flow to enter at the left surface, and flow from the left to the right side of the chamber (Fig. 1B).
Model assumptions
To model the fluid dynamics on an adherent pre-osteoblast in a parallel-plate flow chamber, some assumptions were made about boundary and initial conditions to simplify the FE modeling. The cell was assumed to be attached to the bottom surface of the chamber during the modeling period. The chamber and the attached pre-osteoblast were assumed to be rigid and not affected by the fluid flow, i.e. incompressible and impermeable for fluid. The cell
geometries to solid objects (Fig. 1B). A three-dimensional (3D)-model of a pre-osteoblast (total -5 3
86