The transactive response DNA-binding protein of 43kDa (TDP-43) is a DNA- and RNA-binding protein recognized to form pathological amyloid inclusions in the brains of patients with age-related neurodegenerative disorders, including Alzheimer’s disease, frontotemporal lobar degeneration, and cerebral age-related TDP-43 with sclerosis. Interestingly, these inclusions have also been found in subsets of clinically asymptomatic patients of the aging population, highlighting the need to understand the structural basis of disease development. Current evidence suggests that TDP-43 aggregation is facilitated by the production of C-terminal fragments of the protein that have a high propensity to form amyloid fibrils due to the presence of a low complexity domain (LCD)—a region with low amino acid diversity. In addition to forming amyloid fibrils, the TDP-43 LCD was recently identified to undergo liquid-liquid phase separation (LLPS), a phenomenon whereby protein-protein interactions induce dynamic, reversible coacervation. Prolonged or repetitive coacervation is thought to drive pathological deposition of aggregates; however, a direct connection between TDP-43 LLPS and aggregation that would provide a mechanistic basis for Alzheimer’s disease and related neurodegenerative disorders has not been established. To better understand the structural underpinnings of TDP-43 amyloid aggregates and the role of LLPS in their formation, LCD aggregation conditions were screened and LLPS was identified to regulate amyloid polymerization. Because a comprehensive structural model for TDP-43 fibrils has yet to be proposed, electron paramagnetic resonance (EPR) spectroscopy was subsequently utilized to identify immobilized residues of the LCD that likely constitute the fibrillar core, the structural building block of amyloids. Interestingly, EPR spectra of residues within the fibrillar core did not suggest that the TDP-43 LCD adopts the most commonly observed cross-β organization—a parallel, in-register organization of β-strands that is observed among other protein aggregates in Alzheimer’s disease. As such, we hypothesized that TDP-43 fibrils adopt a non-canonical structure. To test this hypothesis, we first propose a comprehensive EPR-based approach to (i) identify the full extent of residues involved in the fibrillar core, ascertain which residues participate in β-strands and (ii) identify whether a parallel, in-register motif is present. If this motif can be ruled out, (iii) fiber X-ray diffraction will be used to differentiate non-canonical structures (antiparallel, β-solenoid), and (iv) EPR-derived spin-spin distances will be used to determine the organization of residues at the intermolecular interface between fibrillar subunits. Preliminary data further suggest that LLPS might facilitate formation of an extended fibrillar core. Our second aim therefore will identify whether LLPS alters the underlying cross-β structure of TDP-43 amyloids and propagates a structural p...