Hypoxia induces vascular endothelial growth factor in cultured human endothelial cells

Smooth muscle cells, macrophages, glial cells, keratinocytes, and transformed cells have been established as synthesis sites for vascular endothelial growth factor (VEGF). The modulating effects of VEGF are essentially limited to endothelial cells (ECs), the only cell type consistently shown to express VEGF receptors. VEGF has thus been considered to act exclusively via a paracrine pathway. We sought to determine whether the role of human ECs might, under selected conditions, extend beyond that of a target to involve contingency synthesis of VEGF. In both unstimulated human umbilical vein ECs (HUVECs) and human derma-derived microvascular ECs (HMECs), Northern analysis detected no VEGF transcripts. Phorbol-12-myristate 13-acetate (10(-7) M) treatment, however, induced VEGF mRNA expression in both HUVECs and HMECs, peaking at 3 and 6 h, respectively, and returning to undetectable levels by 12 h. In vitro exposure of HUVECs to a hypoxic environment (pO2 = 35 mm of mercury) for 12, 24, and 48 h and exposure of HMECs for 6, 12, 24, and 48 h induced VEGF mRNA in a time-dependent fashion. Re-exposure to normoxia (pO2 = 150 mm of mercury) for 24 h after 24 h of hypoxia returned VEGF mRNA transcripts to undetectable levels in HUVECs. Cobalt chloride and nickel chloride treatment each induced VEGF mRNA in ECs. Cycloheximide treatment further augmented expression of VEGF mRNA induced by cobalt chloride, nickel chloride, and hypoxia in HUVECs. VEGF protein production in hypoxia HUVECs was demonstrated immunohistochemically. Conditioned media from hypoxic HUVECs caused a 2-fold increase in the incorporation of tritiated thymidine. Finally, immune precipitates of anti-KDR probed with anti-Tyr(P) antibodies demonstrated evidence of receptor autophosphorylation in hypoxic but not normoxic HUVECs. These findings thus establish the potential for an autocrine pathway that may augment and/or amplify the paracrine effects of VEGF in stimulating angiogenesis.