In vitro large-wound re-endothelialization. Inhibition of centrosome redistribution by transient inhibition of transcription after wounding prevents rapid repair.

Rapid, efficient re-endothelialization of large wounds is characterized by a specific sequence of cytoskeletal events that occur after wounding. Wounds 1.5 mm wide were created down the middle of confluent porcine aortic endothelial monolayers to study regulation of repair. The wounded cultures were incubated for short periods with cycloheximide or actinomycin D to test the hypothesis that transient inhibition of translation and transcription at the time of wounding disrupts rapid repair by interfering with centrosome redistribution to the front of the cell, an early event associated with cell migration. Although centrosome reorientation did not occur when protein synthesis was inhibited with 20 micrograms/mL cycloheximide for 1 hour before and for up to 4 hours after wounding, reorientation did occur by 2 hours after cycloheximide was washed out. The times taken for the wound to close for cycloheximide-treated and control cells did not differ (60 +/- 1.1 vs 60 +/- 0.8 hours). When transcription was inhibited with 0.25 micrograms/mL actinomycin D for 1 hour before and for 1 hour after wounding, re-endothelialization was dramatically reduced. The time taken for the wound to close was almost five times longer (288 +/- 5.3 hours) than for control cells. The cells moved very slowly, maintaining a flattened, spread-out shape, as opposed to being elongated. The centrosomes did not reorient to the front of the cell throughout the entire period. However, addition of actinomycin D for 2 hours when centrosomes had already moved to the front of the cells (4 hours after wounding) did not reduce subsequent wound repair (60 +/- 1.3 hours). This study supports our hypothesis that centrosome redistribution is essential for efficient wound repair and suggests that redistribution is regulated by transcription of essential gene(s) that is induced immediately after wounding by an unknown short-lived signal. Two possible signals are the loss of cell contact and/or a soluble substance released from the cells at the time of wounding. When the signal is unable to induce transcription, dysfunctional repair occurs by a very slow centrosome-independent process.