Gene regulation is fundamental for every biological process involved in animal development and disease. We focus on transcription, the first step of DNA-based gene expression in which a particular segment of DNA is copied into RNA. Numerous proteins are involved in mammalian transcription. Coordination of the large number of proteins toward precise transcriptional regulation requires a complex network of thousands of specific protein-protein interactions. Many of the functionally critical interactions are mediated by intrinsically disordered regions (IDRs) in transcription proteins, which do not fold into well-defined 3D structures. This fact challenges the classical structure-function paradigm, which postulates that unique protein structures determine their functions. How IDRs successfully execute functions and specificity in transcription has been an enigma for decades. A large part of the transcriptional regulation puzzle thus remains missing.
We strive to understand the interaction behaviors of these IDRs and elucidate their roles in regulating transcription under normal and disease conditions, with an initial focus on the mechanisms of transcription control by chimeric oncogenic transcription factors. IDRs are ubiquitous in eukaryotic proteomes, e.g., they constitute nearly 50% of the human proteome. They are referred to as the “dark proteome” and perform functions in numerous essential cellular processes besides transcription with unclear mechanisms. Further putting a spotlight on IDRs are many recent discoveries on functionally relevant liquid-liquid phase separation in the cell, which is often driven by IDR-mediated biomolecular interactions. By investigating the IDRs involved in transcription, we aim to ultimately reach a predictive understanding of sequence-interaction-function relationships of the entire dark proteome. Our research will also shed new light on the molecular mechanisms of different cancers and suggest new therapeutic strategies.
Complex biomolecular transactions, such as those involved in transcriptional regulation, are often highly dynamic, reversible, and heterogeneous. An in-depth mechanistic understanding of these transactions often requires visualizing biomolecules one at a time with high spatiotemporal resolution. To this end, we develop novel single-molecule imaging methods and combine them with genome editing, biochemical, genomic, and proteomic approaches in our research.