Mechanical and kinetic factors drive sorting of F-actin crosslinkers on bundles

In cells, actin binding proteins (ABPs) sort to different regions in order to establish F-actin networks with diverse functions, including filopodia used for cell migration, or contractile rings required for cell division. Recent experimental work uncovered a passive mechanism that may facilitate spatial localization of ABPs: binding of a short crosslinker protein to two actin filaments promotes the binding of other short crosslinkers and inhibits the binding of longer crosslinkers (and vice versa). We hypothesize this sorting arises because F-actin is semiflexible and cannot bend over short distances. We develop a mathematical theory and a kinetic Monte Carlo simulation encompassing the most important physical parameters for this process, and use simulations of a coarse-grained but molecularly explicit model to characterize and test our predictions about the interplay of mechanical and kinetic parameters. Our theory and data predict an explicit dependence of crosslinker separation on bundle polymerization rate. We perform experiments that confirm a dependence on polymerization rate, but in an unanticipated non-monotonic manner. We use simulations to show that this non-monotonic behavior can arise in situations where crosslinkers have equal bundling affinity at equilibrium, but differing microscopic binding rates to filaments. This dependence of sorting on actin polymerization rate is a non-equilibrium effect, qualitatively similar to non-equilibrium domain formation in materials growth. Thus our results reveal an avenue by which cells can organize molecular material to drive biological processes, and can also guide the choice and design of crosslinkers for engineered protein-based materials.