The epidermal growth factor receptor (EGFR) is overexpressed in many human cancers and cancer cell lines due to gene amplification and/or increased gene transcription. We have continued our efforts to identify and characterize transcription factors that regulate EGFR gene expression. In our studies of EGFR gene regulation, we have documented the interaction of transcription factors such as Sp1, AP1,AP2 and p53 with the promoter region. Recent reports that p53-related molecules, p63/p51/p73L/p40/KET, can tranactivate p53 target promoters prompted us to examine whether these molecules are involved in EGFR gene regulation. TAp63/p51A represses basal EGFR promoter activity in contrast to the activation by p53. Additionally, dNp63/p73L, a dominant-negative effector of TAp63/p51A, rescues the repression by dNp63/p73L. The repression by TAp63/p51A is not mediated via the p53 binding site and appears to involve an inhibition of AP-1 activity. We have also shown the nuclear factor kappaB, NF-kappaB, can bind to several sites in the EGFR promoter but does not transactivate. These results add to the complex nature of EGFR gene regulation and its expression in cancer cells. Another factor that we have shown to interact with the EGFR promoter region is the transcriptional repressor GCF2. Genomic DNA from this clone was labeled and used in FISH analysis to determine chromosomal location. The GCF2 gene was localized to chromosome 3 q27 region. This is a region that is frequently overrepresented in malignant lymphomas. This is consistent with our finding of high level GCF2 expression in six Burkitts lymphoma cell lines and in one cell line derived from a T-cell lymphoma. We have recently determined that GCF2 can repress both basal and TAT-mediated activation of the HIV-LTR. GCF2 is able to bind to TAR RNA and other double-stranded RNAs. There does not appear to be interaction between GCF2 and TAT. GCF2 repression of the HIV-LTR requires both amino-terminal and carboxy-terminal regions of GCF2. It is unclear as to whether other transcription factors are involved in the GCF2-mediated repression of the HIV-LTR. In collaboration with Levon Khachigan, University of New South Wales, Sidney Australia, we also showed that the platelet-derived growth factor A-chain (PDGF-A) promoter was repressed by GCF2. The binding site was determined by DNase I footprinting analysis and found to be identical in sequence to the site in the EGFR promoter. The GCF2 binding site in the PDGF-A promoter overlaps binding sites for Sp1, Sp3 and Egr-1. GCF2 was shown to compete with these three factors for binding to this site. The EGFR promoter is also regulated by Egr-1 but in a cell dependent manner. In collaboration with Xu-Wen Liu, NICHD, we have isolated and cloned the rat EGFR promoter region. The rat EGFR promoter is also GC-rich and contains 64% homology to the human promoter. The TCCTCCTCC repeats as well as Sp1 binding and multiple transcription initiation sites are features of both promoters. We have shown that the TCCTCCTCC repeats are required for nerve growth factor downregulation of EGFR gene expression. We have identified the Wilms Tumor protein, WT1, as a factor that binds to this region. WT1 binds to this promoter region and enhances promoter activity. Upon nerve growth factor treatment, the binding of WT1 is decreased. Thus, a possible mechanism for nerve growth factor downregulation of rat EGFR expression may be due to loss of WT1 activity. To clarify earlier reports on a factor termed GCF, we collaborated with Masato Takimoto, Hokkaido University, Hokkaido Japan, to revise the GCF sequence. The revised sequence has an alternative 5' region that does not have DNA binding or repression activity. The revised GCF does not appear to function as a transcription factor.
