When activated, the Arp2/3 complex initiates the formation of new actin filaments by a process called nucleation - new daughter filaments emerge from existing mother filaments in a y-branch configuration with a regular branch angle.

The generation of force and cell movement depends on cycles of motile F-actin network assembly and disassembly. The actin-related protein-2/3 (Arp2/3) complex is a nucleation factor that assembles new actin filaments from the sides of pre-existing ones to generate branched networks. The assembly and disassembly cycles rely on the activity of the Arp2/3) complex and other factors, including the capping protein (CP). CP caps the barbed ends of actin filaments and terminates filament elongation, but nevertheless promotes motility. As reported in Cell, Akin and Mullins solve this apparent paradox and show that CP increases the rate of motility by promoting Arp2/3 complex-mediated actin filament nucleation, thus highlighting the importance of network architecture versus assembly kinetics.

The authors used an actin-based motility system that was reconstituted using purified components. Arp2/3-activating factor-coated beads were mixed with the Arp2/3 complex, CP, cofilin, profilin and cytoplasmic actin. The beads first assembled spherically symmetrical actin shells and then broke symmetry and moved, producing polarized comet tails.

Variations in the concentration of CP and Arp2/3 had opposite effects on motility; more CP increased motility, whereas more Arp2/3 had an inhibitory effect. The authors analysed tail-growth rate and found that it increased at high CP concentrations and decreased at high Arp2/3 concentrations. However, neither CP nor Arp2/3 affected the overall rate at which F-actin was assembled, suggesting that CP and Arp2/3 altered the architecture of the network rather than the actin-assembly kinetics.

During the growth of actin shells, the total amount of actin remained constant even with varying concentrations of CP and Arp2/3, showing that shell size was dependent on the density of the network. The total number of filaments, and thus the rate of nucleation, was dependent on the concentration of CP, whereas Arp2/3 levels affected the orientation of actin filaments.

Interestingly, high levels of CP did not increase the concentration of soluble actin and had no effect on either the number of free barbed ends or the rate of polymer assembly, meaning that CP does not change the rate of filament elongation. Instead, high CP concentrations led to an increase in nucleation rate.

Available actin monomers can either interact with elongating barbed-ends or the Arp2/3 complex. The authors propose that by capping the barbed ends near the bead surface, CP causes actin monomers to favour participation in the nucleation reaction over filament elongation. Frequent rounds of nucleation produce many short and stiff filaments that may be more efficient at generating force, thus increasing motility. This model contradicts the previously accepted 'Funnelling Hypothesis' whereby CP increases actin monomer concentration by capping fast-growing filaments, thus allowing the remaining uncapped filaments to grow faster and drive higher motility.

Author: Kim Baumann