Actin's protein toolbox revealed

Cells often have partially redundant mechanisms to perform cellular functions, which can be an obstacle in deciphering molecular functions in a single mutant background. In this study, we were interested in determining how yeast cells control the elongation of actin filaments in endocytic actin networks and we successfully identified three partially overlapping mechanisms.

HFSP Long-Term Fellow Alphée Michelot and colleagues
authored on Mon, 15 April 2013

Dense networks of actin filaments are assembled at sites of clathrin-mediated endocytosis. Their function is to provide forces necessary for the invagination of the membrane, and to participate in the scission of the vesicle at the end of the process. Models describing how the actin filaments are generated and regulated from a biochemical point of view have been proposed in the past decade, and have enabled the community to define a limited set of essential proteins. Among these proteins, inhibitors of actin filament elongation are believed to be crucial, because they limit the polymerization of actin filaments exclusively to areas where filaments have just been assembled, offering cells a way to control precisely where forces are exerted.

We challenged current models of actin network regulation with the observation that in yeast, the absence of the only factor known as an inhibitor of actin filament elongation (the heterodimer of capping protein), only causes mild endocytic phenotypes. Moreover, the dynamic properties of the actin networks do not appear to be greatly affected in these mutants. These results clearly contradict current models, and indicate either that these models need to be corrected, or most probably that other unknown factors in yeast are able to perform functions similar to the function of capping protein. In order to identify additional candidate proteins involved in inhibition of actin filament elongation, we used the power of yeast genetics and generated double mutants of a capping protein gene with mutants of other yeast genes on a genome-wide scale. We evaluated genetic interactions based on colony sizes, hypothesizing that if a protein has a partial overlap of function with capping protein, the corresponding double mutants should display a negative genetic interaction.

Using this strategy, we identified three distinct pathways leading to the inhibition of actin filament elongation. Interestingly, we demonstrated that these different factors work on different sub-populations of actin filaments, showing that although all of these molecules are able to inhibit filament elongation, they also have slightly different functions in vivo. Individually, absence of any of these three pathways leads to minor defects, because loss of function of any of the three pathways can be at least partially compensated by another. However, double mutants display major phenotypes, and reconcile cellular biology observations with the current models.

In conclusion, this study provided a new opportunity to apply the unique power of yeast for dissecting functional pathways. Because of the simplicity of its genome and the facility of designing mutants, it offers great possibilities to unveil functions of proteins in a complex cellular background.


Actin filament elongation in Arp2/3-derived actin networks is controlled by three distinct mechanisms. Michelot A, Grassart A, Okreglak V, Costanzo M, Boone C, Drubin DG. Dev Cell. 2013 Jan 28;24(2):182-95. doi: 10.1016/j.devcel.2012.12.008.

Pubmed link