Journal of Molecular Biology
Volume 285, Issue 5, 5 February 1999, Pages 2177-2198
Journal home page for Journal of Molecular Biology

Regular article
The atomic structure of protein-protein recognition sites1

https://doi.org/10.1006/jmbi.1998.2439Get rights and content

Abstract

The non-covalent assembly of proteins that fold separately is central to many biological processes, and differs from the permanent macromolecular assembly of protein subunits in oligomeric proteins. We performed an analysis of the atomic structure of the recognition sites seen in 75 protein-protein complexes of known three-dimensional structure: 24 protease-inhibitor, 19 antibody-antigen and 32 other complexes, including nine enzyme-inhibitor and 11 that are involved in signal transduction.

The size of the recognition site is related to the conformational changes that occur upon association. Of the 75 complexes, 52 have “standard-size” interfaces in which the total area buried by the components in the recognition site is 1600 (±400) Å2. In these complexes, association involves only small changes of conformation. Twenty complexes have “large” interfaces burying 2000 to 4660 Å2, and large conformational changes are seen to occur in those cases where we can compare the structure of complexed and free components. The average interface has approximately the same non-polar character as the protein surface as a whole, and carries somewhat fewer charged groups. However, some interfaces are significantly more polar and others more non-polar than the average.

Of the atoms that lose accessibility upon association, half make contacts across the interface and one-third become fully inaccessible to the solvent. In the latter case, the Voronoi volume was calculated and compared with that of atoms buried inside proteins. The ratio of the two volumes was 1.01 (±0.03) in all but 11 complexes, which shows that atoms buried at protein-protein interfaces are close-packed like the protein interior. This conclusion could be extended to the majority of interface atoms by including solvent positions determined in high-resolution X-ray structures in the calculation of Voronoi volumes. Thus, water molecules contribute to the close-packing of atoms that insure complementarity between the two protein surfaces, as well as providing polar interactions between the two proteins.

Introduction

Many biological processes are carried out, or regulated, through the interactions between preformed proteins. The importance of such interactions in biology has made the protein recognition process an area of considerable interest. Here, we describe aspects of the atomic structure of the recognition sites seen in protein-protein complexes of known structure. The protein-protein complexes whose structures we analyse here are non-covalent assemblies of proteins that fold separately to carry out independent functions before they associate, as opposed to permanent macromolecular assemblies, such as oligomeric proteins, virus capsids or muscle fibres. Though the two types of complexes have features in common, they also have many that differ, and are best treated separately.

General analyses of structural aspects of protein-protein interaction have been carried out Chothia and Janin 1975, Argos 1988, Janin and Chothia 1990, Janin 1995, Janin 1996, Jones and Thornton 1995, Jones and Thornton 1996, Jones and Thornton 1997, Tsai et al 1996, Chothia 1997. Recently, however, there has been a large increase in the number of known three-dimensional structures that contain protein-protein recognition sites. These structures cover a much broader range of activities than earlier ones, which were almost exclusively protease-inhibitor and antibody-antigen complexes. They allow us to determine the extent to which rules based on the few structures that were first available, can be generalised. We also find features that were missed in previous work.

The aspects of structure with which we are concerned here are those related to the stabilization of protein association: the size and chemical character of the protein surface that is buried at interfaces; the packing density of atoms that make contacts across the interface, which expresses complementarity; and polar interactions through hydrogen bonds and interface water molecules. Each of these aspects can be described at the level of the individual atom.

From the Protein Databank (PDB; Bernstein et al., 1977), we selected 72 non-redundant files that contain X-ray structures of protein-protein complexes determined at a resolution of 3.1 Å or better. When more than one complex was present in the asymmetric unit, only one copy was retained. The atomic co-ordinates of another three complexes were given to us by their authors.

Table 1 lists the 75 structures: 24 are complexes of proteases with protein inhibitors, 19 are complexes of antibody Fab or Fv fragments with cognate protein antigens, and 32 complexes of other kinds. Among the latter, there are nine complexes between enzymes and protein inhibitors or substrates, and 11 that involve GTP/GDP-binding proteins (G-proteins) or other components of the cellular cycle or signal transduction pathways. Although nearly all these complexes are binary in the sense that they have only two protein components, a few also contain cofactors or other non-protein components.

Section snippets

The interface areas of protein-protein recognition sites

Our basic tool for measuring the extent of a protein-protein recognition site is the interface area: the area of the accessible surface on both partners that becomes inaccessible to solvent due to protein-protein contacts. This area is the sum of the solvent-accessible surface areas (ASA) of the isolated components less that of the complex.

Solvent accessible surface areas were determined here using the Lee & Richards (1971) algorithm. The computer program is based on that originally produced by

Conformation changes

For many of the complexes listed in Table 1, atomic structures are known not just for the complex but also for one or both of the free components. Comparison of the free and complexed structures can show the changes that take place upon association. With one exception, the proteases, inhibitors, antigens and antibodies that form complexes with standard size interfaces undergo only small changes in conformation. These include shifts in surface loops or movements of short segments of polypeptide

Classes of atoms at protein-protein recognition sites

In describing the atomic structure of recognition sites it is useful to distinguish between three classes of atoms (Figure 2). These are (i) interface atoms (types A, B, C): all atoms that lose solvent-accessible surface on the formation of the complex. Interface atoms on different components are within the sum of the Van der Waals radii plus the diameter of the water probe (2.8 Å). Not all interface atoms make actual Van der Waals contact across the interface. Those which do will be called:

The chemical character of the interfaces

Chemical groups at the protein surface may be allotted to one of three types: non-polar (all groups containing aliphatic and aromatic carbons); neutral polar groups (all groups containing non-carbon atoms, except those carrying a net electric charge), and charged groups. Their relative contribution to the interface area is listed in Table 6 (see page 2186). The average composition of the solvent-accessible surface of small globular proteins is 57 % non-polar, 24 % neutral polar and 19 % charged

The character of amino acids at interfaces

Table 4 describes the contribution of each of the 20 amino acid types to the interfaces and to the protein surface that remains accessible to solvent in the complexes. The two surfaces have amino acid compositions that are significantly different. Interfaces are much richer in aromatic residues His, Tyr, Phe and Trp than the average protein surface (21 % versus 8 %), and somewhat richer in aliphatic residues Leu, Ile, Val and Met (17 % versus 11 %). They are depleted in the charged residues

Polar interactions: hydrogen bonds and interface water molecules

Polar interactions at interfaces were determined using the program HBPLUS (McDonald et al., 1994) and standard geometrical parameters. The number of such bonds, Nhb, in each complex is listed in Table 1. The average interface contains ten hydrogen bonds, but Nhb can be as low as 1 between cytochrome c and cytochrome peroxidase (2pcc) or as high as 34 between Gtβγ and phosducin (2trc). Standard-size interfaces in protease-inhibitor and antibody-antigen complexes interfaces contain 9 (±5)

Atomic packing at interfaces

The volumes occupied by atoms inside proteins or buried in their interfaces can be determined by constructing Voronoi polyhedra around their atoms, and calculating their volume. The packing density at interfaces was determined by calculating the volume of the Voronoi polyhedron of each atom buried in the interface, summing the values to give a total volume V, and comparing V to a reference value Vo. To derive Vo, we used the mean volumes that the atomic groups of residues occupy in protein

Diversity of interfaces

The analysis of the atomic structure presented in the previous sections shows that recognition sites have some features that all, or almost all, share and other features that vary. To illustrate these general points at a more detailed level we describe and discuss here the recognition sites in two complexes that have standard size interfaces and two complexes with large interfaces. All four structures were determined at high resolution (2 Å or better). The recognition sites in these complexes

Discussion

Early analyses of protein-protein recognition sites were based on a small number of structures, nearly exclusively protease-inhibitor and antigen-lysozyme complexes Chothia and Janin 1975, Janin and Chothia 1990. They revealed a remarkable homogeneity in the size of the interfaces measured by the interface area, and suggested that regions of the protein surface that form recognition sites do not differ greatly in their chemical character from the rest of the solvent-accessible surface. The

Conclusion

The examples of antibody-antigen, protease-inhibitors and many other complexes show that a standard-size interface is sufficient for both stability and specificity. These standard recognition sites form stable protein-protein complexes irrespective of the size of the proteins, and no large conformational change is involved in their formation. On average, these interfaces bury 1600 Å2, have a chemical character that is close to the average protein surface, and involve nine hydrogen bonds and the

Acknowledgements

L.L. thanks IBM for an IBM Cooperative Fellowship and The Fondazione Cassa di Risparmio and the Compagnia di San Paolo of Torino for support. J.J. acknowledges financial support from EMBL-European Bioinformatics Institute (Hinxton, Cambridge, UK) and Université Paris-Sud (Orsay, France) during a sabbatical leave. We thank Dr S. Wodak for discussion, Dr N. Strynadka, S. Smerdon, M. Knossow, D. Fleury, N. Chinardet and H.K. Song for providing co-ordinates prior to deposition.

References (121)

  • J.P. Derrick et al.

    The third IgG-binding domain from streptococcal protein G

    J. Mol. Biol.

    (1994)
  • R.A. Engh et al.

    Crystal structures of catalytic subunit of cAMP-dependent protein kinase in complex with isoquinolinesulfonyl protein kinase inhibitor H7, H8 and H89. Structural implications for selectivity

    J. Biol. Chem.

    (1996)
  • F. Frigerio et al.

    Crystal and molecular structure of the bovine a-chymotrypsin-eglin C complex at 2.0 Å resolution

    J. Mol. Biol.

    (1992)
  • M. Fujinaga et al.

    Crystal and molecular structures of α-chymotrypsin with its inhibitor turkey ovomucoid third domain at 1.8 Å resolution

    J. Mol. Biol.

    (1987)
  • T.R. Gamble et al.

    Crystal structure of human cyclophilin A bound to the amino-terminal domain of HIV1 capsid

    Cell

    (1996)
  • R. Gaudet et al.

    Crystal structure at 2.4 Å resolution of the complex of transducin βγ and its regulator, phosducin

    Cell

    (1996)
  • M. Gerstein et al.

    The volume of atoms on the protein surface calculated from simulation using Voronoi polyhedra

    J. Mol. Biol.

    (1995)
  • J.P. Griffith et al.

    X-ray structure of calcineurin inhibited by the immunophilin-immunosuppressant FKBP12-FK506 complex

    Cell

    (1995)
  • M. Harel et al.

    Crystal structure of an acetylcholinesterase-fasciclin complexinteraction of a three-fingered toxin from snake venom with its target

    Structure

    (1995)
  • Y. Harpaz et al.

    Volume changes on protein folding

    Structure

    (1994)
  • Q. Huang et al.

    The refined 1.6 Å resolution crystal structure of the complex formed between porcine β-trypsin and MCTI-A, a trypsin inhibitor of the squash family

    J. Mol. Biol.

    (1993)
  • R. Huber et al.

    Structure of the complex formed by bovine trypsin and bovine pancreatic trypsin inhibitor. II Crystallographic refinement at 1.9 Å resolution

    J. Mol. Biol.

    (1974)
  • J. Janin

    Principles of protein-protein recognition from structure to thermodynamics

    Biochimie

    (1995)
  • J. Janin et al.

    Stability and specificity of protein-protein interactionsthe case of the trypsin-trypsin inhibitor complexes

    J. Mol. Biol.

    (1976)
  • J. Janin et al.

    The structure of protein-protein recognition sites

    J. Biol. Chem.

    (1990)
  • J. Janin et al.

    Surface, subunit interfaces and interior of oligomeric proteins

    J. Mol. Biol.

    (1988)
  • S. Jones et al.

    Analysis of protein-protein interaction sites using surface patches

    J. Mol. Biol.

    (1997)
  • M.C. Lawrence et al.

    Shape complementarity at protein/protein interfaces

    J. Mol. Biol.

    (1993)
  • J. Lescar et al.

    Crystal Structure of a cross reaction complex between Fab F9.13.7 and guinea fowl lysozyme

    J. Biol. Chem.

    (1995)
  • R.L. Malby et al.

    The structure of a complex between the NC10 antibody and influenza virus neuraminidase and comparison with the overlapping site of the NC41 antibody

    Structure

    (1994)
  • I.K. McDonald et al.

    Satisfying hydrogen bonding potential in proteins

    J. Mol. Biol.

    (1994)
  • S. Miller et al.

    Interior and surface of monomeric proteins

    J. Mol. Biol.

    (1987)
  • P.R. Mittl et al.

    A new structural class of serine protease inhibitors revealed by the structure of the hirustatin-kallikrein complex

    Structure

    (1997)
  • L. Prasad et al.

    Evaluation of mutagenesis for epitope mappingstructure of an antibody-protein antigen complex

    J. Biol. Chem.

    (1993)
  • D.C. Rees et al.

    Refined crystal structure of potato inhibitor complex of carboxypeptidase A at 2.5 Å resolution

    J. Mol. Biol.

    (1982)
  • F.M. Richards

    The interpretation of protein structurestotal volume, group volume distributions and packing density

    J. Mol. Biol.

    (1974)
  • T.J. Rydel et al.

    The refined crystal structure of the hirudin-thrombin complex

    J. Mol. Biol.

    (1991)
  • G. Schreiber et al.

    Energetics of protein-protein interactionsanalysis of the Barnase-Barstar interface by single mutations and double mutant cycles

    J. Mol. Biol.

    (1995)
  • H.K. Song et al.

    Kunitz-type soybean trypsin inhibitor revisitedrefined structure of its complex with porcine trypsin reveals an insight of the interaction between a homologous inhibitor from Erythrina caffra and tissue-type plasminogen activator

    J. Mol. Biol.

    (1998)
  • M. Stewart et al.

    Structural basis for molecular recognition between nuclear transport factor 2 (NTF2) and the GDP-bound form of the Ras-family GTPase Ran

    J. Mol. Biol.

    (1998)
  • P. Argos

    An investigation of protein subunit and domain interfaces

    Protein Eng.

    (1988)
  • N. Ban et al.

    Crystal structure of an idiotype-anti-idiotype Fab complex

    Proc. Natl Acad. Sci. USA

    (1994)
  • D.W. Banner et al.

    The crystal structure of the complex of blood coagulation factor VIIA with soluble tissue factor

    Nature

    (1996)
  • T.N. Bhat et al.

    Bound water molecules and conformational stabilization help mediate an antigen-antibody association

    Proc. Natl Acad. Sci. USA

    (1994)
  • J. Bigay et al.

    Roles of lipid modifications of transducin subunits in their GDP-dependent association and membrane binding

    Biochemistry

    (1994)
  • T.L. Bigler et al.

    Binding of aminoacid side chains to preformed cavitiesinteraction of serine proteases with turkey ovomucoid third domains with coded and noncoded P1 residues

    Protein Sci.

    (1993)
  • W. Bode et al.

    X-ray crystal structure of the complex of human leukocyte (Pmn elastase) and the third domain of the turkey ovomucoid inhibitor

    EMBO J.

    (1986)
  • W. Bode et al.

    The high resolution X-ray crystal structure of the complex formed between subtilsin Carlsberg and eglin C, an elastase inhibitor from the leech Hirudo medicinalis

    Eur. J. Biochem.

    (1987)
  • A.M. Buckle et al.

    Protein-protein recognitioncrystal structural analysis of a barnase-barstar complex at 2.0 Å resolution

    Biochemistry

    (1994)
  • M.J. Castro et al.

    Alanine point-mutations in the reactive region of bovine pancreatic trypsin inhibitoreffects on the kinetics and thermodynamics of binding to β-trypsin and a-chymotrypsin

    Biochemistry

    (1996)
  • Cited by (1804)

    View all citing articles on Scopus
    1

    Edited by A. R. Fersht

    View full text