Correlation of VEGF expression by leukocytes with the growth and regression of blood vessels in the rat cornea

PURPOSE. TO determine the temporal and spatial relationships between neovas cularization and basic fibroblast growth factor (bFGF) and vascular endothe lial growth factor (VEGF) mRNA and protein expression in the rat cornea aft er cautery with silver nitrate. METHODS. In female Sprague-Dawley rats, a silver nitrate applicator was pla ced on the central cornea to elicit circumferential angiogenesis, and blood vessel growth was quantified by digital image analysis of corneal flat-mou nts. Total RNA or protein was extracted from whole corneas until 1 week aft er cautery, and bFGF and VEGF mRNA and protein levels were determined by re verse transcription-polymerase chain reaction (RT-PCR) and enzyme-linked im munosorbent assay (ELISA). To localize VEGF mRNA and protein, paraformaldeh yde-fixed and paraffin-embedded histologic cross sections of corneas were e xamined by in situ hybridization and immunohistochemistry. Macrophages were identified by ED2 immunohistochemistry. To examine the regulation of VEGF, rats were treated with dexamethasone (0.5 mg/kg per day) and hyperoxia (70 % O-2). RESULTS. The neovascular response progresses in three phases: (1) a nonprol iferative phase preceding vessel growth (less than or equal to 48 hours aft er cautery); (2) a proliferative phase with maximal growth rate between 3 a nd 4 days; and (3) a regressive phase (day 7) with a decrease in vessel den sity accompanying the completion of vessel elongation. In corneas after cau tery, bFGF mRNA expression was unchanged, and bFGF protein concentration de creaseed by 97% after 24 hours and returned to control levels by day 7. In contrast, VEGF(164) and VEGF(188) mRNA splice variants and protein peaked 4 8 hours after cautery, remained elevated 4 days after cautery, and decrease d to near baseline by day 7. The peak concentration of VEGF in the cornea a t 48 hours was calculated to be 720 pM, which is sufficient to evoke a func tional response. In situ hybridization and immunohistochemistry showed VEGF expressed initially in neutrophils (24-48 hours) and subsequently in macro phages (4 days) adjacent to the cautery site. Treatment with either dexamet hasone or systemic hyperoxia inhibited both neovascularization and the incr ease in VEGF expression. Dexamethasone inhibited 27% of cautery-induced VEG F upregulation at 24 hours and 23% at 48 hours, hyperoxia inhibited 32% at 24 hours and 43% at 48 hours, and combined treatment with both dexamethason e and hyperoxia had an additive effect (56% inhibition at 24 hours). CONCLUSIONS. VEGF production by leukocytes correlates temporally and spatia lly with cautery-induced angiogenesis in the rat cornea. Both inflammatory products and hypoxia appear to sufficiently increase VEGF expression near t he cautery lesion to increase vascular permeability of limbal vessels and i nduce endothelial cell migration and proliferation.