Residue-by-residue details of FUS protein phase separation and aggregation

Project Summary RNA-binding proteins are essential components of numerous large complexes that carry out fundamental processes including transcription, splicing, and DNA repair. Many RNA-binding proteins possess regions predicted to be disordered based on low-complexity sequence characteristics that are critical to normal RNA- processing functions, but also drive aberrant protein assembly in various neurodegenerative disease and cancers. The molecular interactions and functional roles of these disordered domains remain incompletely characterized, especially in the context of disease. Fused in Sarcoma (FUS) is one of twenty-nine human RNA-binding proteins that contains both an essential disordered low-complexity domain (LC) with unusually low charged residue composition and a high frequency of aromatic amino acids as well as several RGG motif regions. Despite disorder, these domains are thought to facilitate interactions in normal RNA metabolism by forming dynamic associations, thereby enabling tunable, reversible spatial clustering. Yet, excessive self- association between FUS disordered domains is believed to result in the formation of pathological neuronal inclusions in sub-types of amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD), irreversible neurodegenerative diseases that lack effective treatments. Moreover, fusion of the FUS disordered domains to several DNA-binding domains through chromosomal translocations results in uncontrolled gene expression leading to a family of aggressive cancers. Though FUS has emerged as the primary model system for understanding biological phase separation, the contacts holding together FUS assemblies and their structures in physiology and disease are currently unknown because they are invisible to traditional techniques in structural biology. However, we have demonstrated that we can visualize dynamic assemblies of FUS with residue-level resolution. This project will apply advanced nuclear magnetic resonance spectroscopy, molecular simulation, and cell models of FUS function to 1) visualize the molecular contacts that mediate LLPS and in- cell interactions, 2) probe the molecular and cellular impact of LC and RGG mutations of FUS found in familial ALS, and 3) determine the functionally relevant atomic details of FUS complexes with RNA and the C-terminal domain (CTD) of RNA polymerase II associated with RNA processing and transcription. These studies of FUS assembly will provide necessary structure/function information on future pharmacological targets for inhibiting pathological protein associations in types of ALS, FTD, leukemia, and sarcoma. Furthermore, because FUS is only one of many essential RNA-binding proteins containing aggregation-prone low complexity domains, the results of the project will serve as a foundation for understanding an entire class of proteins and for correcting their dysfunctions in disease.