Albany 2013: Book of Abstracts
June 11-15 2013
©Adenine Press (2012)
Genomic regions flanking E-box binding sites influence DNA binding specificity of bHLH transcription factors through DNA shape
Transcriptional regulation of gene expression is enacted mainly through binding of transcription factors (TFs) to specific, short DNA sites in cis regulatory regions of genes. Most TFs are members of protein families that share a common DNA binding domain and thus recognize similar DNA binding sequences. It is not well understood why paralogous TFs often bind different genomic target sites in vivo to effect different regulatory programs, despite apparently recognizing the same sequence motifs. Here, we designed custom protein binding microarrays (PBMs) to analyze the DNA binding specificities of two Saccharomyces cerevisiae basic helix-loop-helix (bHLH) proteins, Tye7 and Cbf1, as a model system. Our data reveal that E-box DNA binding sequences (CAnnTG), when tested in the context of their native genomic flanking sequences, are bound differently by Cbf1 and Tye7. Computational models of the PBM data indicate that DNA sequence features located in the genomic sequences outside the E-box contribute to DNA binding specificity in vitro. Our analyses suggest that these flanking regions affect DNA binding specificity indirectly by influencing the three-dimensional structure of the E-box binding sites. Finally, we show that these subtle differences in intrinsic sequence preferences of Cbf1 and Tye7 in vitro help to explain their differential DNA binding preferences in vivo. Our results provide further evidence that the local shape of DNA binding sites may be an important feature in distinguishing the DNA binding preferences among paralogous TFs and thus may play a widespread role in determining how transcriptional regulatory specificity within TF families is achieved.
1Departments of Biostatistics and Bioinformatics, Computer Science, and Molecular Genetics and Microbiology, Duke University, Durham, NC 27708
2Institute for Genome Sciences and Policy, Duke University, Durham, NC 27708
3Department of Pharmacology and Cancer Biology, Duke University, Durham, NC 27708
4Molecular and Computational Biology Program, Departments of Biological Sciences, Chemistry, and Physics and Astronomy, University of Southern California, 1050 Childs Way, Los Angeles, CA 90089
5Division of Genetics, Departments of Medicine and Pathology, Brigham and Women’s Hospital and Harvard Medical School, Boston, MA 02115
6Harvard-MIT Division of Health Sciences and Technology (HST), Harvard Medical School, Boston, MA 02115
7These authors contributed equally to this work.
Ph: (919) 667-8673