Strong scientific evidence has evolved over the past 30-40 years showing that any of a variety of fusion proteins resulting from chromosomal translocations can drive the formation of deadly forms of cancer. Some fusions, including BCR-ABL and translocations involving ALK, have been successfully targeted by anticancer therapeutics. In Ewing sarcoma in the cancer-causing EWSR1-FLI1 fusion does not have an obviously druggable entity. EWSR1 is a member of the FET (FUS, EWSR1, TAF-15) family of RNA binding proteins composed of a combination of structured RNA binding domains and intrinsically disordered protein (IDP) sequences of largely unknown function. It is the latter sequences, housed within the N-terminal 200 residues of FET proteins, that are translocated to form cancer-causing fusion proteins. These sequences are defined as being of low complexity (LC) due to the fact that they are dominated by representation of only a subset of the 20 amino acids. Despite this paucity of chemical diversity, we have found that the LC domains of FET proteins can indeed adopt a distinct molecular conformation in the form of labile, cross-? interactions. Our observations over the past 5 years, including our recently published structural studies of the FUS LC domain, may have begun to reveal the mechanism of action of the IDP sequences associated with FET proteins. These molecular studies offer immediate ideas as to how FUS polymers might bind repetitive proteins such as the C-terminal domain of RNA polymerase II, thus suggesting the mechanism of transcriptional activation by FET proteins. CTD binding favors regions of the LC domain extending in an unstructured conformation on the C-terminal side of the polymer core. We propose that a Velcro-like interaction is established between the iterative, heptad repeats of the CTD and the ?bottle brush? conformation of FUS polymers. Individual points of CTD:FUS LC interaction must be inherently weak, yet when multimerized both by the repetitive nature of the CTD and the polymeric structure of its target (the LC domains of FET proteins), substantive binding affinity is afforded. We propose structural and mechanistic studies of the LC domains of FET proteins aimed at discovery of the weak, ?druggable? point along the regulatory pathway by which the FET fusion proteins exert their oncogenic activity.
Aim 1 will use a combination of biochemistry and solid state NMR to resolve the molecular structure of labile, cross-? polymers formed from the low complexity domain of EWSR1.
Aim 2 will identify cellular proteins and RNAs that interact selectively with LC domain polymers of FET proteins.
In Aim 3 we will complete a high- throughput screen in search of chemicals that modulate polymerization of FET LC domains and/or recruitment of the CTD of RNA polymerase II. This screen may lead to the discovery of chemicals that will help us better understand how the LC domains of FET proteins actually work in living cells.
