Progress in FY2018 was in the following areas: FIBRIL FORMATION BY THE LOW-COMPLEXITY DOMAIN OF FUS: In collaboration with Prof. Steven McKnight (UT Southwestern Medical Center) and his colleagues, we completed and published a complete molecular structural model for fibrils formed by the 214-residue low-complexity domain of the protein FUS (Murray et al., Cell 2017). Surprisingly, a specific segment of the FUS LC domain (residues 39-95) forms the structurally ordered core of the fibrils, while much of the rest of the LC domain remains dynamically disordered. The amino acid composition of this core-forming segment is not significantly different from the overall amino acid composition of the LC domain. The core-forming domain also lacks hydrophobic residues (as does the overall LC domain). We are now working to clarify the nature of inter-residue interactions that drive FUS-LC self-assembly, and to clarify why residues 39-95 form the fibril core. To this end, we have performed additional solid state NMR measurements, on 2H,15N,13C-labeled FUS-LC fibrils, that promise to provide new information about sidechain-sidechain interactions involving polar sidechains of Ser, Thr, Gln, and Tyr residues. These data are currently being analyzed. We have also recorded 2D and 3D solid state NMR spectra of fibrils formed by residues 1-108 and residues 111-214 of the FUS-LC. Spectra of fibrils formed by residues 1-108 match spectra of full-length FUS-LC fibrils, indicating that the core structure of full-length FUS-LC does not depend on interactions with the C-terminal half. We are currently analyzing data for fibrils formed by residues 111-214 to determine how this structure compares with the full-length FUS-LC fibril core and to determine why the N-terminal core structure is more stable. A manuscript describing this work is currently under review. AFFECT OF ASP-TO-VAL MUTATION ON HNRNPA2 FIBRIL FORMATION: Previous work by other labs has shown that an Asp-to-Val mutation at residue 290 of the low-complexity domain of hnRNPA2 leads to amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD) by facilitating intracellular hnRNPA2 aggregation. To understand the structural basis for this phenomenon, we have performed solid state NMR and electron microscopy measurements on fibrils formed by wild-type hnRNPA2-LC and by the D290V mutant. The data indicate that residue 290 is near the end of a beta-strand that forms parallel beta-sheets in the wild-type fibrils, and that Asp290 is negatively charged in the fibrils near neutral pH. This suggests that replacement of Asp290 by Val eliminates intermolecular electrostatic repulsions that make wild-type hnRNPA2-LC fibrils less stable, and creates new hydrophobic interactions that may stabilize the D290V mutant fibrils. In fact, from guanidine denaturation experiments, we find that wild-type fibrils are indeed less stable than mutant fibrils by approximately 1.2 kJ/mol. This work
Showing the most recent 10 out of 18 publications
