A Statistical Method to Evaluate the Effect of Shear Stress on Endothelial Gene Expression in vivo

Abstract

Objective To determine the effect of hemodynamic shear stress on endothelial gene expression in vivo. Methods Computational fluid dynamics (CFD) simulations were performed in three realistic models of the porcine abdominal aorta and its iliac artery branches. Using these calculations, a shear stress probability map in the iliac arteries was generated. This map showed the likelihood that a particular location would experience high, medium, or low time average shear stress. Based on this map, three anatomical regions likely to experience high, medium, and low shear stress levels were identified. The median time-average shear stress in these regions were 32.1, 15.2, and 9.7 dyn/cm2, respectively. Subsequently, endothelial cell RNA was collected from these regions in five additional porcine iliac arteries. Gene expression of select genes was measured in these samples using quantitative real time PCR. Results for the high and low shear regions are reported relative to the medium shear region. Results Expression of the genes tubulin, beta-catenin, c-jun, MCP-1, and VCAM was elevated in both the low and high shear stress regions relative to the medium shear region. ICAM expression was downregulated in both the low and high shear regions relative to the medium shear region. These effects were significant in nearly all cases. Conclusions This statistical method permits evaluation of phenotypic differences that arise due to hemodynamic forces in in vivo specimens. Arterial regions subjected to distinct levels of time average shear stress exhibit differences in endothelial gene expression. Supported by NIH grant H50442.