Unraveling the Assembly of Large Macromolecular Machines by Integrating Computational Techniques with Experimental Data

Abstract

Proteins often assemble in large macromolecular complexes to achieve a specific task. Unfortunately, owing to their size and complexity, the structure of these machines is difficult to be determined at atomistic resolution. Thus, the ability to reliably predict the conformation of multimeric assemblies is desirable. We present a new approach that uses a Particle Swarm Optimization search guided by experimental-based restraints to characterize protein quaternary structure. Moreover, the natural subunit flexibility as extracted from molecular dynamics simulations is explicitly included during model building. This scheme has been successfully used to model of the heptameric soluble and functional forms of pore-forming toxin aerolysin from Aeromonas hydrophila (see Figure). The model is based on the high-resolution X-ray structure of aerolysin monomer and the low-resolution cryo-EM map of the heptamer. The same strategy has been extended to determine the membrane-embedded basal body of the multi-MDa type III secretion system from Yersinia enterocolitica. The method is of general applicability and, coupled with accurate energy functions, can efficiently exploit the spatial restraints derived from a variety of experimental techniques to produce consistent models for the assembly of biological systems.View Large Image | View Hi-Res Image | Download PowerPoint Slide