Abstract
Cofilin is an essential actin filament severing protein that functions in the dynamic remodeling of the actin cytoskeleton. Filament severing activity is most efficient at sub-stoichiometric cofilin binding densities (i.e. <1 cofilin per actin filament subunit), and peaks when the number density of boundaries (i.e. junctions) between bare and cofilin-decorated segments is maximal. A model in which local topological and mechanical discontinuities lead to preferential fragmentation at boundaries accounts for available experimental data, including direct visualization of cofilin and actin during real-time severing events. The boundary-severing model predicts that ligands (e.g. other actin-binding proteins) that compete with cofilin for actin filament binding and modulate cofilin occupancy on filaments will alter the bare-decorated segment boundary density, and thus, the filament severing activity of cofilin. Here, we directly test this model prediction by evaluating the effects of phalloidin and myosin, two ligands that compete with cofilin for filament binding, on the actin filament binding and severing activities of cofilin. Our experiments demonstrate that competitive displacement of cofilin lowers cofilin occupancy and promotes severing when initial cofilin occupancy is high (i.e. >50%). Even in the presence of competitive ligands, maximum severing activity occurs when cofilin-decorated boundary density is highest, consistent with preferential fragmentation at boundaries. We propose a general "severodyne" framework for the modulation of cofilin-mediated actin filament severing by small molecule or actin-binding protein ligands that compete with cofilin for actin filament binding.