The polymerization of actin monomers into filaments and larger structures, e.g., bundles, is a key process in the machinery of the cell. In the last decade, experimental measurements of actin and actin-related proteins have greatly increased our understanding of both the mechanism of cell motility and the role of cellular components that regulate filament growth, i.e., capping proteins. In spite of these tremendous achievements, there are still numerous fundamental questions that remain unanswered. Chief among these questions is the role of the actin-bound nucleotide in the structure, properties and function of the actin cytoskeleton. This presentation describes an investigation of the folding of the alpha-helical domain in the DNase I-binding loop of actin.

Historically, the main challenges in applying the principles of molecular modeling to systems of biological relevance have been the poor scaling of these techniques and lack of computational power. Of particular relevance to the present study is the issue of sampling rare-events. For example, the folding or unfolding of an alpha helix such as the one described above requires the formation or breaking of many hydrogen bonds along the peptide backbone. Transformation between these two states requires the crossing of a free-energy barrier that is much larger than kBT, and therefore extremely unlikely to be observed during a classical molecular dynamics (MD) simulation, wherein the longest accessible timescales are on the order of 100 ns. To address the issue of sampling rare events, the method of metadynamics was introduced, wherein the simulation is biased along one or more collective variables that offer a reduced description of the transition of interest. Due to the long timescales under consideration, application of such a technique is essential for accurate modeling and simulation of the process of interest.

First, a systematic methodology for using metadynamics to obtain free-energy landscapes for folding alpha helices will be presented. Various strategies for selection of collective variables will be discussed with an emphasis on obtaining an efficient yet general procedure for studying the folding of the alpha helix. Next, free-energy landscapes for the folding of the alpha helix within the DNase I-binding loop will be presented. We have studied this folding process as a function of the bound nucleotide within actin, and differences between the landscapes will be discussed. Finally, the role of folding in the DNase I-binding loop will be discussed in the context of the supramolecular structures F-actin and the Arp2/3 complex.