Purchase this article with an account.
J.J. Michon; Transcription Factor Binding Site Clustering and Tentative Identification of cis–Regulatory Modules in Human Retinal Pigment Epithelium (RPE) Genes . Invest. Ophthalmol. Vis. Sci. 2005;46(13):3080.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose:To use bioinformatics techniques including phylogenetic footprinting to test the hypothesis that both general and tissue–specific transcription factor binding sites can be identified adjacent to genes expressed in the RPE, among the most cell–type specific transcripts of any human tissue, and that these sequences are organized into putative cis–regulatory modules. Methods: Twenty–two transcripts expressed exclusively or predominantly in human RPE cells were identified using serial analysis of gene expression (SAGE) tags as a starting material. Other source material included ESTs and full length cDNAs in public databases. These were mapped to human genomic sequences using the UC Santa Cruz Genome Server and regions from upstream within 4 kb or downstream within 1 kb of the transcription start site were isolated. Orthologous sequences in vertebrate species including mouse, rat, chicken and zebrafish were identified where available and multiple alignments were performed. Sequence motifs were identified using a Gibbs sampling strategy. Results: Binding sites corresponding to known basal transcription factors were identified and confirmed by comparison with experimentally verified sites using the TRANSFAC database. Sequence comparisons were made between genes known to be expressed in both non–RPE cells and RPE (such as transthyretin, cathepsin D, ribosomal protein S2) versus those with predominantly RPE expression. Non–coding sequences subject to positive selection across species boundaries were found to be organized into clusters within several hundred base pairs of each other. Conserved sequences potentially mediating tissue specificity were detected adjacent to basal transcription factor binding sites with a much higher likelihood than background would predict. Conclusions: The identification of functionally important noncoding regions that control transcription is one of the major challenges in understanding the genome. Transcriptional regulation in RPE cells may be controlled in a modular fashion, with basal transcription factor binding sites clustering with tissue–specific regulatory sites. This evidence supports a model of cooperative binding of transcription factors that may account for the tissue specificity exhibited by many genes active in the RPE.
This PDF is available to Subscribers Only