Transcription factors perform a key role in regulating gene expression, but our ability to interpret regulatory DNA by identifying transcription factor binding sites or by predicting regulatory function is lacking. One generally unappreciated contributing factor is that each transcription factor has the potential to interact with cofactors, sometimes in structurally distinct ways, thus adding to the ability of transcription factor complexes to bind distinct DNA sequences. Perhaps the best illustration of this idea is the Hox family of proteins, which pattern the anterior/posterior axis in all metazoans. When bound to the cofactor Extradenticle (Exd), Hox proteins exhibit latent differences in DNA-binding specificity necessary to carry out their unique activities. In addition, they exhibit paralogue specific binding by recognizing low-affinity DNA-binding sites, illustrating a fundamental tradeoff between DNA binding specificity and affinity. In this proposal, I will extend these observations by studying the role of protein-protein interaction (PPI) motifs that mediate the interaction between Hox proteins and their cofactors. The physical interaction between Hox and Exd, mediated by a canonical motif present in all Hox proteins, has been particularly-well studied. However, some Hox proteins contain additional motifs. These motifs may promote alternate Hox-Exd-DNA conformations, a potential mechanism underlying paralogue-specific activity. To test this hypothesis, combinations of Exd interaction motifs will be mutated in the endogenous loci of three Hox genes: Ultrabithorax (Ubx), abdominal-A (abd-A) and Abdominal-B (Abd-B). ChIP-seq (Chromatin ImmunoPrecipitation) will be used to assess differences in in vivo DNA-binding profiles to identify motif-redundant and motif-specific binding sites. Next, differences in DNA- binding affinity between mutant Hox-Exd complexes will be determined by utilizing a version of SELEX-seq (Systematic Evolution of Ligands by Exponential Enrichment) modified to capture low affinity DNA-binding. Comparing the ChIP-seq and SELEX-seq datasets will generate a list of motif-specific sites driven to bind only by the interaction between Hox and Exd, and a list of binding events caused by other mechanisms.
The second aim of this study is to evaluate the role of Exd interaction motifs in regulating the transcriptional potential of Hox proteins as Hox proteins can both activate and repress genes expression. To address this question, RNA-seq, ATAC-seq and epigenetic mark specific ChIP-seq will be performed on tissues that express Hox proteins with mutant Exd interaction motifs.
The third aim of this study is to study how different isoforms of Homothorax (Hth), another Hox cofactor that is required for the nuclear localization of Exd, regulate Hox activity. Studying the mechanisms guiding Hox paralogue specificity and regulatory activity is important because of (1) the deep conservation of Hox genes in higher organisms, (2) other transcription factors exhibit the same phenomenon of low affinity DNA binding and (3) studying how transcription factors interact with DNA is important to interpreting the increasing number of regulatory loci associated with disease.
Recent advances in genome sequencing have revealed that many disease-causing variations lie within regulatory DNA. Transcription factors play a key role in interpreting these sequences to regulate gene expression, but knowledge of their activity is stymied by a general inability to predict in vivo binding events and activity. This project aims to gain fundamental insights into how the DNA-binding and regulatory specificity of transcription factors are governed, a necessary goal to toward understanding the role of regulatory DNA in disease.
