Time and flow-dependent changes in the p27kip1 gene network drive maladaptive vascular remodeling

Objective: Although clinical studies have identified that a single nucleotide polymorphism in the p27(kip1) gene is associated with success or failure after vein bypass grafting, the underlying mechanisms for this difference are not well defined. Using a high-throughput approach in a flow-dependent vein graft model, we explored the differences in p27(kip1)-related genes that drive the enhanced hyperplastic response under low-flow conditions. Methods: Bilateral rabbit carotid artery interposition grafts with jugular vein were placed with a unilateral distal outflow branch ligation to create differential flow states. Grafts were harvested at 2 hours and at 1, 3, 7, 14, and 28 days after implantation, measured for neointimal area, and assayed for cell proliferation. Whole-vessel messenger RNA was isolated and analyzed using an Affymetrix (Santa Clara, Calif) gene array platform. Ingenuity Pathway Analysis (Ingenuity, Redwood City, Calif) was used to identify the gene networks surrounding p27(kip1). This gene set was then analyzed for temporal expression changes after graft placement and for differential expression in the alternate flow conditions. Results: Outflow branch ligation resulted in an eightfold difference in mean flow rates throughout the 28-day perfusion period (P <.001). Flow reduction led to a robust hyperplastic response, resulting in a significant increase in intimal area by 7 days (0.13 +/- 0.04 mm(2) vs 0.014 +/- 0.006 mm(2); P <.005) and progressive growth to 28 days (1.37 +/- 0.05 mm(2) vs 0.39 +/- 0.06 mm(2); P <.001). At 7 days, low-flow grafts demonstrated a burst of actively dividing intimal cells (36.4 +/- 9.4 cells/mm(2) vs 11.5 +/- 1.9 cells/mm(2); P=. 04). Sixty-five unique genes within the microarray were identified as components of the p27(kip1) network. At a false discovery rate of 0.05, 26 genes demonstrated significant temporal changes, and two dominant patterns of expression were identified. Class comparison analysis identified differential expression of 11 genes at 2 hours and seven genes and 14 days between the high-flow and low-flow grafts (P <.05). At 2 hours, oncostatin M and cadherin 1 were the most differentially expressed. Cadherin 1 and protein kinase B exhibited the greatest differential expression at 14 days. Conclusions: Alterations in flow and shear stress result in divergent patterns of vein graft remodeling. Associated with the dramatic increase in neointimal expansion observed in low-flow vs high-flow grafts is a subset of differentially expressed p27(kip1)-associated genes that correlate with critical stages in the adaptive response. These represent potential biologic targets whose activity may be altered to augment maladaptive vascular remodeling.