In SilicoIdentification of Functional Protein Interfaces

Rachel E Bell,Nir Ben-Tal

doi:10.1002/cfg.309

Abstract

Proteins perform many of their biological roles through protein–protein, protein–DNA or protein–ligand interfaces. The identification of the amino acids comprising these interfaces often enhances our understanding of the biological function of the proteins. Many methods for the detection of functional interfaces have been developed, and large-scale analyses have provided assessments of their accuracy. Among them are those that consider the size of the protein interface, its amino acid composition and its physicochemical and geometrical properties. Other methods to this effect use statistical potential functions of pairwise interactions, and evolutionary information. The rationale of the evolutionary approach is that functional and structural constraints impose selective pressure; hence, biologically important interfaces often evolve at a slower pace than do other external regions of the protein. Recently, an algorithm, Rate4Site, and a web-server, ConSurf (http://consurf.tau.ac.il/), for the identification of functional interfaces based on the evolutionary relations among homologous proteins as reflected in phylogenetic trees, were developed in our laboratory. The explicit use of the tree topology and branch lengths makes the method remarkably accurate and sensitive. Here we demonstrate its potency in the identification of the functional interfaces of a hypothetical protein, the structure of which was determined as part of the international structural genomics effort. Finally, we propose to combine complementary procedures, in order to enhance the overall performance of methods for the identification of functional interfaces in proteins.

Highlights

The discrimination of functional oligomeric protein–protein interfaces observed in threedimensional (3D) structures from contacts that are artifacts of crystallization is a challenging question
Many criteria have previously been used to identify and characterize protein–protein interfaces. These rely on considerations of the solvent accessible surface area buried upon association [14,13]; free energy changes upon alanine-scanning mutations [31]; experimental [35] and in silico [25] twohybrid systems; scoring functions based on statistical potentials [21,26]; and physicochemical and geometric properties of the surface, such as electrostatics [28], hydrophobicity [32], amino acid composition [22] and shape complementarity or planarity [18,15,12]
The most rudimentary approach for the identification of biological inter-protein interfaces uses the solvent accessibility of the protein surface area (ASA)

Summary

Introduction

The discrimination of functional oligomeric protein–protein interfaces observed in threedimensional (3D) structures from contacts that are artifacts of crystallization is a challenging question. Other approaches are based on scoring evolutionary conservation. The most rudimentary approach for the identification of biological inter-protein interfaces uses the solvent accessibility of the protein surface area (ASA).

Results

Conclusion