Abstract
Purpose: :
Fibrosis is an enigmatic consequence of many ocular blinding diseases and it especially has devastating effects on the refractive property of the transparent cornea. The goal of this study was to validate that vimentin is a key target for corneal fibrosis.
Methods: :
We used the corneal model of combined alkali burn with epithelial scrape injury to elicit corneal fibrosis. Mice were also subjected to the vimentin inhibitor withaferin A (WFA; 2 mg/kg/day; intrapertoneal injection) for 7-14 days.
Results: :
Injured wild type mice sustain critical hallmarks of corneal fibrosis that concede from conjunctivalization, neovascularization, inflammatory infiltration and myofibroblast activation, which become progressively insidious by 14 days post injury. Injured corneas of vimentin deficient (Vim KO) mice have delayed myofibroblast activation, downregulated TGF-beta expression by 7 days post injury, and by day 14 show decreased CD11b-positive infiltrates and recover expression of the abundant corneal crystallin tissue transketolase. We show that wild type mice do not recover injury-downregulated expression of the critical cell cycle inhibitor p27kip1 and maintain the conjunctival phenotype with globlet cell expression, whereas, Vim KO mice engage a regenerative repair mechanism that is evident by day 7 with re-expression of p27kip1 that is mainated through day 14 when corneal clarity is fully recovered. This in vivo mechanism of epithelial cell growth arrest mediated by genetic deficiency of vimentin is mimicked by treatment of wild-type mice with WFA in vivo that also causes recovery of p27kip1 expression due to potent downregulation of its cognate ubiquitin conjugating E3 ligase Skp2. Moreover, exploiting WFA to cause conditional knockout of vimentin in vitro, we show that G2/M cell cycle arrest exerted by WFA is completely abrogated in cultured cells from mice deficient in p27kip1 or Skp2.
Conclusions: :
These findings for the first time illuminate a previously unrecognized important role for vimentin that acts as a central target for corneal fibrosis resulting from its broad expression in stromal, inflammatory, vascular and limbal compartments of the diseased cornea. We additionally demonstrate that this type III intermediate filament is druggable in vivo and afford WFA as a drug lead for therapeutic development.
Keywords: cytoskeleton • wound healing • proliferation