Self-organization of a propulsive actin network as an evolutionary process

The leading edge of motile cells is propelled by polymerization of actin filaments according to a dendritic nucleation/array treadmilling mechanism. However, little attention has been given to the origin and maintenance of the dendritic array. Here we develop and test a population–kinetics model that explains the organization of actin filaments in terms of the reproduction of dendritic units. The life cycle of an actin filament consists of dendritic nucleation on another filament (birth), elongation by addition of actin subunits and, finally, termination of filament growth by capping protein (death). The regularity of branch angle between daughter and mother filaments endows filaments with heredity of their orientation. Fluctuations of branch angle that become fixed in the actin network create errors of orientation (mutations) that may be inherited. In our model, birth and death rates depend on filament orientation, which then becomes a selectable trait. Differential reproduction and elimination of filaments, or natural selection, leads to the evolution of a filament pattern with a characteristic distribution of filament orientations. We develop a procedure based on the Radon transform for quantitatively analyzing actin networks in situ and show that the experimental results are in agreement with the distribution of filament orientations predicted by our model. We conclude that the propulsive actin network can be understood as a self-organizing supramolecular ensemble shaped by the evolution of dendritic lineages through natural selection of their orientation.