Control of aberrant TDP-43 phase transitions by the selective autophagy machinery
Loading...
Date issued
Authors
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Reuse License
Abstract
Neurodegenerative diseases such as amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD) converge on the failure of neuronal proteostasis. A central hallmark of these diseases is the cytoplasmic aggregation of TAR DNA binding protein of 43 kDa (TDP-43), a ubiquitously expressed RNA-binding protein whose physiological role in RNA processing and stress responses is essential for neuronal health. How cells normally surveil and clear aberrant TDP-43 assemblies, and how this fails in disease, remains insufficiently understood.
In the first part of this thesis (Part A), I investigated how proteasome dysfunction precipitates TDP-43 pathology and how the cell responds to such assemblies. Using stable cell lines expressing GFP-tagged TDP-43 variants, I found that acute proteasome inhibition rapidly drives the condensation of cytosolic TDP-43, in particular C-terminal fragments (CTFs) of TDP-43, into liquid-like puncta that progressively “age” into undynamic aggregate-like structures that become partially detergent-insoluble and acquire hallmarks of pathological TDP-43 inclusions, such as ubiquitination and S409/410 phosphorylation. TDP-43 CTF condensates induced by proteasome inhibition were distinct from stress granules or aggresomes, indicating a unique condensation pathway intrinsic to TDP-43 CTFs. Live-cell experiments revealed that these assemblies are not irreversible endpoints but can be cleared upon washout of the proteasome inhibitor. Their clearance required both Hsp70 chaperone activity, which likely fragments and remodels TDP-43 aggregates, and the autophagy–lysosome system, which engulfs and degrades them. Functional perturbations uncovered distinct contributions: p62/SQSTM1 promotes TDP-43 CTF condensate formation by clustering ubiquitylated TDP-43 species, whereas WIPI2 is essential for phagophore initiation and TDP-43 CTF clearance. Proximity proteomics mapped this sequence of events at molecular resolution, revealing a state-dependent hand-off: diffuse TDP-43 CTF species engage the proteasome, while condensed intermediates recruit autophagy factors including p62, NBR1, WIPI2, and VCP. Immunohistochemistry in FTLD-TDP brain tissue confirmed p62 co-localizes with TDP-43 inclusions and revealed NBR1 as a new aggregating protein in FTLD-TDP patients, underscoring the disease relevance of this pathway. Together, Part A establishes that proteasome failure is sufficient to trigger a regulated condensation of TDP-43, and that successful resolution depends on a coordinated chaperone–autophagy program.
In the second part of this thesis (Part B), I examined the role of UBQLN2, a proteostasis factor genetically linked to ALS/FTD. I optimized its purification and demonstrated that UBQLN2 enhances RNA binding of both TDP-43 and FUS and modulates FUS phase separation behavior. These findings suggest that UBQLN2 might have a dual protective role: stabilizing the RNA-bound state of RNA-binding proteins to raise the threshold for aberrant condensation, and routing ubiquitylated clients toward degradation once condensates form. Disease-linked UBQLN2 mutations may impair both protective mechanisms, thereby accelerating the accumulation of pathological TDP-43 and FUS assemblies.
Taken together, this work outlines a mechanistic framework that helps to explain how TDP-43 pathology emerges from proteasome dysfunction and progresses through a liquid-to-amorphous trajectory that can still be cleared if chaperones and autophagy engage efficiently. It also positions UBQLN2 as a critical modulator at the intersection of phase behavior and proteostasis routing. These insights highlight key molecular checkpoints: proteasomal throughput, chaperone licensing, and autophagic capture, where therapeutic interventions could possibly reinforce proteostasis mechanisms and thus delay disease progression in ALS and FTD