Student Theses and Dissertations

Date of Award

2024

Document Type

Thesis

Degree Name

Doctor of Philosophy (PhD)

RU Laboratory

Alushin Laboratory

Abstract

The actin cytoskeleton integrates physical and chemical cues from a cell’s environment to instruct its behavior, coordinating processes including cell migration, organelle dynamics, and endocytosis. Actin-binding proteins (ABPs) specify actin filament (F-actin) organization to build micron-scale networks which power these cell dynamics. Interactions between ABPs and F-actin are regulated by both biochemical processes and mechanical forces. The mechanisms mediating this interplay remain unclear at the protein structural level. Newly polymerized F-actin hydrolyzes bound ATP to produce metastable ADP-Pi–F-actin, which persists for minutes before phosphate release to yield long-lived ADP–F-actin. I determined cryo-EM structures of both ADP–F-actin and ADP-Pi–F-actin with sufficient resolution (~2.4 Å) to confidently build atomic models including solvent, which showed the two states to be essentially identical at the protein backbone level. This led me to hypothesize that differences in bound small molecules and solvent could alter the energetic landscape of filament bending, producing structural differences in bent states which could be discriminated by ABPs. By developing a neural-network–based particle picking approach and using methods to handle flexible heterogeneity in cryo-electron microscopy (cryo-EM) maps, I determined structures of bent ADP–F-actin and ADP-Pi–F-actin, revealing that individual actin subunits were distorted in nucleotide-state dependent patterns. Although ADP–F-actin and ADP-Pi–F-actin have essentially identical protein structures in their ground states, the release of phosphate and associated changes in the solvent content of ADP–F-actin’s nucleotide cleft renders its subunits more deformable. This decreased rigidity at the subunit level allows ADP–F-actin to bend more readily. I summarized these findings in a “steric boundaries” model for mechanical regulation: Lattice architectural rearrangements remodel the physical space available for a protomer to occupy, thereby inducing the protomer to deform to minimize steric clashes. I continued my explorations of the mechanobiology of F-actin by investigating the effects of tethered myosin motor systems on F-actin structure. Previous members of the Alushin lab had identified nanoscale domains of oscillating curvature, but these structures were refractory to traditional averaging approaches. I updated the tools developed to investigate the nucleotide state dependence of F-actin bending mechanics to these structures. These tools enabled 3Dreconstruction of these structures. From these reconstructions, the myosin motor-evoked filament lattice deformations were elucidated. The remodeled filament lattice is detected byα-catenin, which cooperatively binds along individual strands, preferentially engaging interfaces featuring extended inter-subunit distances while simultaneously suppressing rotational deviations to regularize the lattice. Collectively, we found that myosin forces can deform F-actin, generating a conformational landscape that is detected and reciprocally modulated by α-catenin, providing a direct structural glimpse at force transduction through the cytoskeleton. I next studied how ABPs integrate actin filaments into functional ensembles, focusing on the structural basis for F-actin bundling by ABPs in filopodia. Filopodia are actin-based cellular protrusions localized at the leading edge of migrating cells, which play a prominent role in the dissemination of metastatic cancer cells. I first focused on T-plastin, an ABP that loosely bundles F-actin at the roots of filipodia. By adapting my neural-network–based picking scheme to identify bundled filaments while excluding single filaments, I determined structures of T-plastin bridging filaments in two configurations: parallel and antiparallel. By comparing these structures with a structure of T- plastin bound to a single filament, I was able to reconstruct a sequential bundling pathway, in which T-plastin’s conformation changes upon initial binding to a single actin filament, priming it to engage another filament in either one of these two distinct configurations. Most recently, I have focused on fascin, the protein primarily responsible for tightly bundling F-actin in filopodia. Our structure of fascin bridging filaments explains the mechanism of action of a fascin inhibitor currently in phase 3 clinical trials as a cancer therapeutic, and heterogeneity analysis revealed substantial crossbridge flexibility which mediates crosslinking helical actin filaments into hexagonal bundles with mismatched symmetry. I also adapted my neural network denoising approach to analyze cryo-electron tomograms of fascin-bundled F-actin. This enabled me to localize tens of thousands of individual proteins across dozens of bundled filaments with unprecedented accuracy, allowing unambiguous assignment of fascin positions and orientations without averaging. From these data, I discovered geometric rules which produce emergent patterns of fascin decoration within F-actin bundles and impose constraints on bundle growth. Altogether, I elucidated how individual fascin crossbridge flexibility at the nanoscale supports large flexible bundle construction at the mesoscale. The research presented in this thesis lays the groundwork for future work that will integrate these computational approaches for visualizing F-actin networks and force-induced deformations of filaments. Specifically, branched actin networks at the leading edges of migrating cells and bundled actin networks such as those at focal adhesions are force responsive and good targets for these studies. These investigations will probe the structural basis for the cellular mechano-response through the actin cytoskeleton in its native context.

Comments

A Thesis Presented to the Faculty of The Rockefeller University in Partial Fulfillment of the Requirements for the degree of Doctor of Philosophy

Available for download on Tuesday, March 04, 2025

Included in

Life Sciences Commons

Share

COinS