No abstract

Exposure of spreading anchorage-dependent cells to laminar flow is a common technique to measure the strength of cell adhesion. Since cells protrude into the flow stream, the force exerted by the fluid on the cells is a function of cell shape. To assess the relationship between cell shape and the hydrodynamic force on adherent cells, we obtained numerical solutions of the velocity and stress fields around bovine aortic endothelial cells during various stages of spreading and calculated the force required to detach the cells. Morphometric parameters were obtained from light and scanning electron microscopy measurements. Cells were assumed to have a constant volume, but the surface area increased during spreading until the membrane was stretched taut. Two-dimensional models of steady flow were generated using the software packages ANSYS (mesh generation) and FIDAP (problem solution). The validity of the numerical results was tested by comparison with published results for a semicircle in contact with the surface. The drag force and torque were greatest for round cells making initial contact with the surface. During spreading, the drag force and torque declined by factors of 2 and 20, respectively. The calculated forces and moments were used in adhesion models to predict the wall shear stress at which the cells detached. Based upon published values for the bond force and receptor number, round cells should detach at shear stresses between 2.5 and 6 dyn/cm(2), whereas substantially higher stresses are needed to detach spreading and fully spread cells. Results from the simulations indicate that (1) the drag force varies little with cell shape whereas the torque is very sensitive to cell shape, and (2) the increase in the strength of adhesion during spreading is due to increased contact area and receptor densities within the contact area.

In order to simulate regions of flow separation observed in vivo, a conventional parallel plate flow chamber was modified to produce an asymmetric sudden expansion. The flow field was visualized using light reflecting particles and the size of the recirculation zone was measured by image analysis of the particles. Finite element numerical solutions of the two and three-dimensional forms of the Navier-Stokes equation were used to determine the wall shear stress distribution and predict the location of reattachment. For two different size expansions, numerical estimates of the reattachment point along the centerline of the flow chamber agreed well with experimental values for Reynolds numbers below 473. Even at a Reynolds number of 473, the flow could be approximated as two-dimensional for 80 percent of the chamber width. Peak shear stresses in the recirculation zone as high as 80 dyne/cm2 and shear stress gradients of 2500 (dyne/cm2)/cm were produced. As an application of this flow chamber, subconfluent bovine aortic endothelial cell shape and orientation were examined in the zone of recirculation during a 24 h exposure to flow at a Reynolds number of 267. After 24 h, gradients in cell orientation and shape were observed within the recirculation zone. At the location of reattachment, where the wall shear stress was zero but the shear stress gradients were large, cells plated at low density were still aligned with the direction of flow. No preferred orientation was observed at the gasket edge where the wall shear stress and shear stress gradients were zero. At higher cell densities, no alignment was observed at the separation point.(ABSTRACT TRUNCATED AT 250 WORDS)

The manner in which fluid stresses are transmitted from the apical to the basal surface of the endothelium will influence the dynamics of cell/substrate contacts. Such dynamics could be important in the design of synthetic vascular grafts to promote endothelial cell adhesion. To examine whether the initial response of cell/substrate contact sites to flow depends on the magnitude of the applied shear stress, subconfluent monolayers of endothelial cells were exposed to flow at 10, 20, and 30 dyn cm-2 wall shear stresses for 20 min. Cell/substrate contact sites were visualized with total internal reflection fluorescence microscopy. Flow induced a rapid fluctuation in the membrane topography, which was reflected in dynamic changes in cell/substrate contacts. Exposure to flow caused marked changes in contact area. Contact movement occurred normal and parallel to the direction of flow. Contact sites demonstrated significant variability in contact area at 30 dyn cm-2 during the experiment but no significant movement of the contact sites in flow direction after 20 min of flow. Mean square displacements of the contact center of mass were described in terms of a directed diffusion model. Prior to onset of flow, contact movement was random. Flow induced a significant convective component to contact movement for 300-600 s, followed by reestablishment of diffusive growth and movement of contacts. These results suggest that fluid stresses are rapidly transmitted from the apical to the basal surface of the cell via the cytoskeleton.