Skip to main page content
U.S. flag

An official website of the United States government

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Comparative Study
. 2012 May 3:12:6.
doi: 10.1186/1472-6807-12-6.

Comparison of tertiary structures of proteins in protein-protein complexes with unbound forms suggests prevalence of allostery in signalling proteins

Lakshmipuram S Swapna  1 , Swapnil Mahajan, Alexandre G de Brevern, Narayanaswamy Srinivasan
Affiliations
Comparative Study

Comparison of tertiary structures of proteins in protein-protein complexes with unbound forms suggests prevalence of allostery in signalling proteins

Lakshmipuram S Swapna et al. BMC Struct Biol. .

Abstract

Background: Most signalling and regulatory proteins participate in transient protein-protein interactions during biological processes. They usually serve as key regulators of various cellular processes and are often stable in both protein-bound and unbound forms. Availability of high-resolution structures of their unbound and bound forms provides an opportunity to understand the molecular mechanisms involved. In this work, we have addressed the question "What is the nature, extent, location and functional significance of structural changes which are associated with formation of protein-protein complexes?"

Results: A database of 76 non-redundant sets of high resolution 3-D structures of protein-protein complexes, representing diverse functions, and corresponding unbound forms, has been used in this analysis. Structural changes associated with protein-protein complexation have been investigated using structural measures and Protein Blocks description. Our study highlights that significant structural rearrangement occurs on binding at the interface as well as at regions away from the interface to form a highly specific, stable and functional complex. Notably, predominantly unaltered interfaces interact mainly with interfaces undergoing substantial structural alterations, revealing the presence of at least one structural regulatory component in every complex.Interestingly, about one-half of the number of complexes, comprising largely of signalling proteins, show substantial localized structural change at surfaces away from the interface. Normal mode analysis and available information on functions on some of these complexes suggests that many of these changes are allosteric. This change is largely manifest in the proteins whose interfaces are altered upon binding, implicating structural change as the possible trigger of allosteric effect. Although large-scale studies of allostery induced by small-molecule effectors are available in literature, this is, to our knowledge, the first study indicating the prevalence of allostery induced by protein effectors.

Conclusions: The enrichment of allosteric sites in signalling proteins, whose mutations commonly lead to diseases such as cancer, provides support for the usage of allosteric modulators in combating these diseases.

PubMed Disclaimer

Figures

Figure 1
Figure 1
Distribution of parameters capturing structural change for Control and Test datasets. Distribution of values for the parameters A). Cα RMSD B). %PB changes and C). PB substitution scores calculated at a per-protein level for Control-Rigid, Control-Monomer and PPC datasets. Buried residues are indicated with filled boxes. Ires - interacting residues; NonIres - non-interacting residues; Core_Res (≤5% RSA) - buried residues; Surf10_RSA (>10% RSA) – surface residues. The figure shows that protein-protein complexes undergo significantly larger structural changes when compared with unliganded forms for all residues types. The p-values for all of the following comparisons performed using Mann–Whitney test indicates statistical significance (p-value < 0.0001) : M-All_Res vs. P-All_Res, M-Core_Res vs. P-Core_Res, and M-Surf10_RSA vs. P-Surf10_RSA, for all the 3 parameters. This trend is prominently captured by the parameters Cα RMSD and %PB changes.
Figure 2
Figure 2
Characteristics of different types of interfaces. The three kinds of interfaces are pre-made (blue color), induced-fit (brown) and others (green). A). A plot of Cα RMSD for the pair of interacting partners is shown. Completely pre-made interfaces are enclosed in a yellow square. B). The extent of conformational change for the three kinds of interfaces is shown. The graphs plot only the majority of the points (Cα RMSD ≤6 Å) for the sake of clarity. This figure illustrates that A). Pre-made interfaces largely bind to induced-fit interfaces, and B). Although pre-made interfaces show small magnitude of structural change, the extent of conformational change they undergo is comparable to that observed in induced-fit/other interfaces. For both sets, points corresponding to complexes solved at a resolution >2.5 Å are encircled.
Figure 3
Figure 3
Structural changes observed in interfaces. Protein undergoing change is shown as cartoon, with unbound form in light cyan and the bound form in blue, and its partner as a ribbon, with unbound form in light orange and bound form in magenta. Direction of movement is indicated as black arrow. A) Large (~10 Å Cα RMSD) moving out to avoid steric clash (alpha actin & BNI1 protein; 1Y64). B). Movement to optimise interaction with partner (GTP binding protein & Rho GTPase activating protein; 1GRN). C). Conformational change accompanied by movement mainly to avoid steric clashes with the partner (Glycoprotein Ib alpha & von Willebrand factor; 1 M10). In B) and C), the region of interest is colored green and red in the unbound and bound forms, respectively. D). An interface (actin & deoxyribonuclease I; 1ATN) where certain region moves away to avoid steric clash (colored in red), some region undergoes conformational change with movement to optimise an interaction (depicted in green for the unbound form and brown for the bound form) and another region undergoes rigid body movement to optimise its interaction (colored in lemon yellow in the unbound form and orange in the bound form). All the figures containing protein structures were generated using PyMOL [77].
Figure 4
Figure 4
Scatter plot of PBc for interacting residues vs. rest of surface residues for PPC dataset. Proteins showing higher proportion of PB changes at the interface are encircled in purple whereas proteins showing PB changes in the non-interacting surface region when the interacting region remains unaltered are encircled in green. This plot reveals the existence of several protein-protein complexes which exhibit substantial conformational changes at non-interacting surface regions even though the interface region is largely unmodified (shown in green circle).
Figure 5
Figure 5
Distribution of non-interacting residues with PB change. A). Histogram of “% of ‘residues nearby interface’ in a protein undergoing PB change” is plotted. B). Histogram of “% of non-interacting residues with PB change” which are near to interface is plotted. The upper-bound value for every range is indicated as the label on x-axis. This figure reveals that most of the conformational changes occurring in the non-interacting surface regions are not near the interface.
Figure 6
Figure 6
Normal mode analysis of structural changes in regions of low B-factor away from interface.The protein containing region of interest is depicted as cartoon and the interface of other protein in ribbon. Unbound and bound form of the protein of interest is coloured pale cyan and marine blue, respectively. The partner protein’s unbound and bound forms are coloured light orange and yellow, respectively. Interacting residues are coloured in red and non-interacting residues with PB change in green. All regions of interest are marked with a black circle, irrespective of whether they are intrinsically mobile or rigid. Regions identified to be intrinsically mobile according to NMA are coloured violet. Regions of interest occurring within the intrinsically mobile segments are coloured in dark green. The complexes shown are A). TolB – PAL complex (2HQS) B). Complement C3 and Epstein-Barr virus receptor C2 complex (1GHQ) C). Ran GTPase and Regulator of chromosome condensation (RCC1) complex (1I2M). The partner containing the region of interest is represented in italics. These figures show that noninteracting regions observed to undergo conformational changes upon complexation are usually intrinsically mobile, which is a characteristic of a functional site
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Reichmann D, Rahat O, Cohen M, Neuvirth H, Schreiber G. The molecular architecture of protein-protein binding sites. Curr Opin Struct Biol. 2007;17(1):67–76. - PubMed
    1. Levy ED, Pereira-Leal JB. Evolution and dynamics of protein interactions and networks. Curr Opin Struct Biol. 2008;18(3):349–357. - PubMed
    1. Vidal M, Cusick ME, Barabasi AL. Interactome networks and human disease. Cell. 2011;144(6):986–998. - PMC - PubMed
    1. Schreiber G, Keating AE. Protein binding specificity versus promiscuity. Curr Opin Struct Biol. 2011;21(1):50–61. - PMC - PubMed
    1. Janin J, Wodak SJ. Protein modules and protein-protein interaction. Introduction. Adv Protein Chem. 2002;61:1–8. - PubMed

Publication types

LinkOut - more resources