A flexible linker allows the two halves of an FH2 dimer (blue and tan) to hold onto actin while the filament grows.


Simultaneously holding onto and building something is no mean feat. Yingwu Xu, Michael Eck (Dana-Farber Cancer Institute and Harvard Medical School, Boston, MA), and colleagues find that the formin homology-2 (FH2) domain is ideally suited to the task. The apparently flexible structure of an FH2 dimer is consistent with a stair-stepping model of actin filament growth in which each half of the dimer alternately dissociates to make room for an incoming actin monomer even as the other half of the dimer maintains its grip on the actin filament.

The model did not come to Eck immediately. “We stared at the structure for about a year,” he says. But then the group tried to reconcile the perfect twofold symmetry of the FH2 crystal structure with the helical alignment of actin monomers in a filament. The only way the two would come together, they realized, was if the FH2 was flexible.

A prime suspect for the source of flexibility was a linker region of FH2. It was relatively unstructured in the crystal, could be clipped by proteases, and was the only significant tie between two parts of the dimer. Sure enough, a crystal of a mutant FH2 that was still functional had a completely different orientation of the linker and thus of the two parts of FH2.

Eck believes the two structures prove flexibility rather than demonstrating the two most important orientations of the FH2 regions. Eck hopes that eventually he will see oscillations in fluorescence resonance energy transfer (FRET) as the FH2 steps alternately close to and far away from the monomers of a growing actin filament. ▪


Xu, Y., et al.