Regulation of the integrin subunit alpha5 gene promoter by the transcription factors Sp1/Sp3 is influenced by the cell density in rabbit corneal epithelial cells.

PURPOSE: Expression of the alpha5beta1 fibronectin (Fn) integrin is well recognized in the corneal epithelium and has been postulated to increase during wound healing. In the present study, the regulatory influence of the positive transcription factors Sp1/Sp3 on the activity directed by the promoter of the alpha5 gene was examined in rabbit corneal epithelial cells (RCECs) primary cultured at various cell densities.

METHODS: Expression of the alpha5 subunit was assessed at the transcriptional level by semiquantitative RT-PCR analyses. The regulatory elements necessary to direct expression of the alpha5 gene were identified by transfecting RCECs with recombinant plasmids bearing various lengths from the alpha5 gene promoter fused to the CAT reporter gene. Binding of Sp1/Sp3 to the alpha5 promoter was assessed by both electrophoretic mobility shift assays (EMSAs) and DNaseI footprinting. Endogenous levels of Sp1/Sp3 were determined by Western blot and supershift analyses. The regulatory influence exerted by Sp1/Sp3 on the alpha5 promoter was evaluated both by site-directed mutagenesis and cotransfection in Sp1-deficient Drosophila SL-2 Schneider cells.

RESULTS: Subconfluent RCECs expressed nearly five times more alpha5 transcript than 48-hour postconfluent RCECs. The activity directed by the alpha5 promoter was found to be affected by cell density. Strong promoter activity was observed in subconfluent RCECs, whereas a dramatic repression was measured in postconfluent cells. EMSA and DNaseI footprinting provided evidence for the binding of Sp1 to both a proximal site located within the previously reported alpha5 fibronectin responsive element (FRE), and a distal site located between positions -117 and -101. Cotransfection experiments in Schneider cells, as well as transfection of RCECs with recombinant constructs bearing mutations into the distal Sp1 site, confirmed the positive regulatory influence of Sp1 on both the -42/-92 and -92/-132 alpha5 promoter segments. Most of all, EMSA and Western blot analyses demonstrated the expression of substantial amounts of Sp1/Sp3 in subconfluent but not postconfluent RCECs.

CONCLUSIONS: These results provide support to the hypothesis that the strong reduction in the activity of the alpha5 promoter when RCECs reach a high cell density is the consequence of a reduced expression of Sp1/Sp3 under such cell culture conditions.