proteins STRUCTURE O FUNCTION O BIOINFORMATICS Similar chemistry, but different bond preferences in inter versus intra-protein interactions Mati Cohen, y Dana Reichmann, y Hani Neuvirth, and Gideon Schreiber * Department of Biological Chemistry, Weizmann Institute of Science, Rehovot 76100, Israel INTRODUCTION The interest in the mechanisms of binding and networking of protein–protein interactions has grown increasingly over the past years stemming from the realization that much of the cellular complexity is related to such interactions. The ability of certain proteins to form specific stable protein–protein complexes is fun- damental for biological processes, including events such as signal transduction, cell cycle regulation, immune response, and so forth. Along with protein folding, understanding molecular recogni- tion is one of the major biological challenges. Recognition must occur between the correct partners within a dense and complex cellular environment, and as the result of a subtle free energy bal- ance between the correct and wrong partners. Significant progress has been made in many aspects of computational biology, includ- ing docking, modeling affinity calculations, sensitivity to environ- mental changes, and point mutations, resulting in successful mo- lecular engineering. Despite this progress, many challenges lie ahead and a full understanding of the forces acting within or between proteins has not been reached. 1–5 The architecture and composition of protein–protein interfaces were studied intensively through the statistical analysis of databases of protein–protein interfaces, 6 including comparison between tran- sient and permanent protein–protein interfaces, 7 homodimers and heterodimers, 8 specific and nonspecific protein–protein interfaces 9 and protein–protein interfaces and oligomers. 6,10–12 These studies focused on the comparison of physico-chemical properties, such as the interface size, amino acid composition, residue–residue preferen- ces, solvation, accessible surface area (ASA), shape, and the occur- rence of polar and nonpolar contacts. Ofran and Rost found signifi- cant differences in amino acid composition and residue–residue The Supplementary Material referred to in this article can be found online at http://www. interscience.wiley.com/jpages/0887-3585/suppmat/ Abbreviations: ACM, aaromatic center of mass; ASA, accessible surface area; Mc–Mc, main chaina – main chain; Mc–Sc, main chain – side chain; PDB, protein data bank; Sc–Sc, side chain–side chain; SOD, sum of dots. Grant sponsor: Israel Ministry of Science and Technology; Grant number: 263; Grant sponsor: MINERVA; Grant number: 8525 y Mati Cohen and Dana Reichmann contributed equally to this work. *Correspondence to: Gideon Schreiber, Department of Biological Chemistry, Weizmann Insti- tute of Science, Rehovot 76100, Israel. E-mail: gideon.schreiber@weizmann.ac.il Received 13 May 2007; Revised 12 November 2007; Accepted 10 December 2007 Published online 7 February 2008 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/prot.21960 ABSTRACT Proteins fold into a well-defined structure as a result of the collapse of the polypeptide chain, while transient protein-complex formation mainly is a result of binding of two folded individual mono- mers. Therefore, a protein–protein interface does not resemble the core of monomeric proteins, but has a more polar nature. Here, we address the ques- tion of whether the physico-chemical characteristics of intraprotein versus interprotein bonds differ, or whether interfaces are different from folded mono- mers only in the preference for certain types of interactions. To address this question we assembled a high resolution, nonredundant, protein–protein interaction database consisting of 1374 homodimer and 572 heterodimer complexes, and compared the physico-chemical properties of these interactions between protein interfaces and monomers. We per- formed extensive statistical analysis of geometrical properties of interatomic interactions of different types: hydrogen bonds, electrostatic interactions, and aromatic interactions. Our study clearly shows that there is no significant difference in the chemis- try, geometry, or packing density of individual interactions between interfaces and monomeric structures. However, the distribution of different bonds differs. For example, side-chain–side-chain interactions constitute over 62% of all interprotein interactions, while they make up only 36% of the bonds stabilizing a protein structure. As on average, properties of backbone interactions are different from those of side chains, a quantitative difference is observed. Our findings clearly show that the same knowledge-based potential can be used for protein-binding sites as for protein structures. How- ever, one has to keep in mind the different architec- ture of the interfaces and their unique bond prefer- ence. Proteins 2008; 72:741–753. V V C 2008 Wiley-Liss, Inc. Key words: protein–protein interface; protein com- plex database; interaction geometry; interface sym- metry; protein packing. V V C 2008 WILEY-LISS, INC. PROTEINS 741