Angiogenesis Is Suppressed by Junctional Adhesion Molecule A Through Transcription Factors HIF-1a- and Id1-Dependent Modulation of VEGF/VEGFR2 Expression

Abstract 618

The process of angiogenesis is associated with a number of human pathologies. Vascular endothelial growth factor (VEGF) is a potent angiogenic growth factor. VEGF expression is upregulated during ischemia and tumor growth. VEGF is also expressed in healthy adult vascular tissue. However, VEGF-dependent angiogenesis is normally suppressed in healthy adults. The mechanism of this suppression is not known. Here we show that junctional adhesion molecule A (JAM-A), a tight junction protein, is an endogenous suppressor of VEGF-induced angiogenesis. Using gene-targeted Jam-A null mice, we found that murine melanoma tumor growth and associated angiogenesis were significantly augmented (P<0.001) in Jam-A null mice compared to wild-type mice. Additionally, Jam-A null mice showed significantly enhanced (P<0.004) vascular permeability as assessed by a Miles assay using Evans blue dye. Furthermore, VEGF-, but not FGF-2-dependent angiogenesis is significantly augmented (P<0.001) in the absence of Jam-A as assessed by Matrigel plug and aortic ring sprouting, two well-accepted in vivo and ex vivo angiogenic assays, respectively. Vascular endothelial cells isolated from Jam-A null mouse aorta showed significantly enhanced (P<0.05) cell migration and tube-like structure formation (P<0.05) in response to VEGF. Additionally, we found the plasma levels of VEGF in Jam-A null mice is significantly (P<0.000001) and age-dependently increased compared to WT mice. Furthermore, both mRNA and protein levels of VEGF and its receptor VEGFR2 were significantly increased (P<0.001) in Jam-A null endothelial cells, suggesting that the VEGF/VEGFR2 signaling axis is enhanced in the absence of Jam-A. To further confirm this finding, we injected anti-VEGFR2 (DC101) into the Jam-A null mice inoculated with murine melanoma (B16F0) cells. Inhibition of VEGF/VEGFR-2 signaling by DC101 significantly reduced (P<0.00003) tumor growth and associated angiogenesis in Jam-A null mice. Additionally, vascular permeability observed in Jam-A null mice was completely abrogated upon DC101 treatment. When tested if the expression of soluble Flt (sFlt), which is known to trap VEGF, is downregulated in the absence of Jam-A, we found no significant difference in sFlt mRNA expression in Jam-A null endothelial cells compared to WT. In order to determine the mechanism of this upregulation of the VEGF signaling axis, we tested the expression of hypoxia inducible factor 1a (HIF1-a) and inhibitor of DNA binding 1 (Id1), two transcription factors known to upregulate VEGF and VEGFR2 gene expression. Interestingly, mRNA and protein levels of both HIF1-a and Id1 were significantly augmented (P<0.02) in endothelial cells lacking Jam-A. Consistent with this finding, the overexpression of JAM-A in HUVECs attenuated the levels of Id1. These results suggest that JAM-A suppresses VEGF/VEGFR2 expression in endothelial cells by attenuating HIF1-a and Id1 expression, thus suppressing adult angiogenesis. During pathological conditions such as ischemia and tumor growth, it is possible that JAM-A levels are downregulated, thus supporting pathological angiogenesis.