Corneal Vascular Regression by Forced Epithelial Cell Movements

Abstract

Purpose: : We found previously that loss of homeostatic epithelial cell movements was closely related to pathological vascular growth in the cornea, and we hypothesized that there is a causal relationship between them. We tested the hypothesis by examining: 1) if cell motility inhibitors cause corneal neovascularization, and 2) if restoration of centripetal epithelial cell movements triggers regression of established central corneal blood vessels.

Methods: : Ubiquitous GFP mice, Dstn^corn1, and a cross of these (GFP-Dstn^corn1) were used. To impede homeostatic cell movements, cytoskeletal inhibitors including cytochalasin D, taxol and ML-9 were applied topically. In Dstn^corn1 mice where corneal epithelial cells were stationary, centripetal cell movements were created by repeated mechanical scraping of central epithelium twice a week. Epithelial cell movements were determined by in vivo time-lapse microscopy, tracking changes of epithelial GFP. Corneal vascular growth was monitored in vivo by angiography with sulforhodamine 101.

Results: : Similar to what we found previously with colchicine, topical taxol application abolished centripetal epithelial cell movements and triggered irreversible neovascularization. After ML-9 application, some corneas developed peripheral new vessels that regressed later on, and others developed blood vessels that reached the center. Cytochalasin D also triggered vascularization, but the effect was reversible when the drug treatment was stopped. In an adult Dstn^corn1 mouse cornea, epithelial cells are generally immobile and central corneal vasculature was stable. Repeated central epithelial scraping in these eyes forced cells to move centripetally, and as a consequence radial epithelial GFP stripes were observed. Coincidentally, central corneal vessels regressed completely within two weeks, and the central cornea remained free from blood vessels for at least a month, some for more than two months. Central vessels grew back soon after restoration of stationary epithelium by way of stopping the epithelial scraping. In a small number of eyes, epithelium failed to heal properly within a few days after epithelial scraping, thus failing to trigger constant cell movements; robust vascular regression was not obvious in these eyes.

Conclusions: : Some cell motility inhibitors triggered corneal vascularization in a normal eye. Forced epithelial cell movements triggered regression of vasculature in the central cornea of Dstn^corn1 mice. These results are consistent with a hypothesis that loss of homeostatic epithelial cell movements is a cause of pathological vascular growth in the cornea.

Keywords: cornea: epithelium • microscopy: light/fluorescence/immunohistochemistry • neovascularization