Neovasculature induced by vascular endothelial growth factor is fenestrated

WG Roberts, GE Palade - Cancer research, 1997 - AACR
WG Roberts, GE Palade
Cancer research, 1997AACR
We have reported previously that topical administration of vascular endothelial growth
factor165 (VEGF) to a microvascular bed supplied with a continuous endothelium can
rapidly induce the formation of endothelial fenestrations (WG Roberts and GE Palade, J. Cell
Sci., 108: 2369–2379, 1995). From these results, we hypothesized that tumor vasculature, in
general, may also be fenestrated because it has been reported that tumor secretion of VEGF
causes the surrounding host vasculature to invade and feed the growing tumor. Using …
Abstract
We have reported previously that topical administration of vascular endothelial growth factor165 (VEGF) to a microvascular bed supplied with a continuous endothelium can rapidly induce the formation of endothelial fenestrations (W. G. Roberts and G. E. Palade, J. Cell Sci., 108: 2369–2379, 1995). From these results, we hypothesized that tumor vasculature, in general, may also be fenestrated because it has been reported that tumor secretion of VEGF causes the surrounding host vasculature to invade and feed the growing tumor. Using electron microscopy to characterize the endothelial cell morphology in tumor vessels from either the periphery or the core of the tumor and immunoblotting to detect secreted VEGF, we analyzed the vasculature of human and murine neoplastic tumors grown s.c. in male nude mice. To clarify the role of VEGF165, two models were used: (a) Chinese hamster ovary (CHO) cells stably transfected with hu VEGF165 and injected into mice (VEGF:CHO tumors); and (b) slow-release pellets containing purified VEGF or basic fibroblast growth factor implanted on the rat cremaster muscle. All tumors had vessels with fenestrated endothelium, open interendothelial junctions, and clustered fused caveolae. From all of the peripheral tumor vessels observed, fenestrated endothelium was observed in 41% from EMT, 35% from M1S, 37% from U87, and 56% from VEGF:CHO tumors, whereas surrounding skin and muscle, from which tumor vessels were derived, had fenestrated endothelium in 2 and 0% of all vessels, respectively. Additionally, further analysis revealed a substantial decrease in the anionic glycocalyx on the luminal face of the fenestral diaphragms in endothelium from tumors (especially VEGF:CHO) when compared to intestine or pancreas. Because the host tissue microvascular endothelium which supplies the tumor is not fenestrated, tumors can transform nonproliferating, nonfenestrated vessels into proliferating vessels, many of which have fenestrated endothelium. These data provide evidence that chronic VEGF exposure can induce fenestrations in nonfenestrated endothelium similar to the fenestrated endothelium found in tumor vessels.
AACR