Beauty is in the eye of the beholder: proteins can recognize binding sites of homologous proteins in more than one way

Abstract

Understanding the mechanisms of protein-protein interaction is a fundamental problem with many practical applications. The fact that different proteins can bind similar partners suggests that convergently evolved binding interfaces are reused in different complexes. A set of protein complexes composed of non-homologous domains interacting with homologous partners at equivalent binding sites was collected in 2006, offering an opportunity to investigate this point. We considered 433 pairs of protein-protein complexes from the ABAC database (AB and AC binary protein complexes sharing a homologous partner A) and analyzed the extent of physico-chemical similarity at the atomic and residue level at the protein-protein interface. Homologous partners of the complexes were superimposed using Multiprot, and similar atoms at the interface were quantified using a five class grouping scheme and a distance cut-off. We found that the number of interfacial atoms with similar properties is systematically lower in the non-homologous proteins than in the homologous ones. We assessed the significance of the similarity by bootstrapping the atomic properties at the interfaces. We found that the similarity of binding sites is very significant between homologous proteins, as expected, but generally insignificant between the non-homologous proteins that bind to homologous partners. Furthermore, evolutionarily conserved residues are not colocalized within the binding sites of non-homologous proteins. We could only identify a limited number of cases of structural mimicry at the interface, suggesting that this property is less generic than previously thought. Our results support the hypothesis that different proteins can interact with similar partners using alternate strategies, but do not support convergent evolution.

Figure 1. Schematic representation of the methodology.

(1) pairs of complexes in which homologous proteins A and A′ are seen in interaction with two unrelated proteins B and C are retrieved from the ABAC database; (2) homologous proteins A and A′ are superimposed using Multiprot; (3) the analysis is restricted to protein-protein interaction binding sites, and carried out separately for A/A′ and B/C sides; (4) the number of similar atoms is computed after superimposition of the binding sites: here, two different types of atoms are represented by squares and triangles; (5) random interfaces are created by randomizing the atom types, in order to obtain random distributions and to compute p-values.

Figure 2. The five categories of ABAC pairs.

For each pair of complexes, one structure is displayed in pink and the other in green, with the superimposed A/A′ domains on the left side and the B/C domains on the right side. Images are generated using Pymol . Structural mimicry, alternate loop conformations and residue insertion/deletion are highlighted by thicker representations. Hereafter, complexes are named by their PDB code (first four letters), combined with the identifiers of interacting chains (last two letters). A: category O, PDB structure 1dg1_HG (dimer of domain 2 of elongation factor Tu of E. coli) versus PDB structure 1g7c_AB (domain 2 of elongation factor eEF-1 alpha from S. cerevisiae complexed with guanine nucleotide exchange factor domain from elongation factor-1 beta), B: category M, PDB structure 1avw_AB (trypsin from pig complexed with soybean trypsin inhibitor) versus PDB structure 1fak_BD (human coagulation factor VIIa complexed with bovine pancreatic trypsin inhibitor), C: category E, PDB structure 1wq1_RG (human cH-p21 Ras protein complexed with p120GAP domain) versus PDB structure 1gzs_AB (human CDC42 complexed with GEF domain of SopE toxin from S. typhimurium), D: category I, PDB structure 1bui_AC (catalytic domain of human plasmin complexed with staphylokinase from S. aureus) versus PDB structure 1gl0_BA (bovine chymotrypsinogen complexed with protease inhibitor PMP-D2V from L. migratoria), E: category S, PDB structure 1p8j_HE (N-terminal domain of murine furin complexed with C-terminal domain of furin) versus PDB structure 1ic6_AB (dimer proteinase K from T. album).

Figure 3. Similarity at protein-protein interfaces in ABAC pairs.

First row: all-atom representations, second row: coarse-grain representations, third row: C representations. First column: number of superimposed elements on A/A′ versus B/C side, second column: number of similar elements on A/A′ versus B/C side, third column: fraction of similar elements on A/A′ versus B/C side.

Figure 4. Schematic illustration of ABAC pairs.

Domains A and A′, from the same SCOP family, interact with B and C from different SCOP superfamilies. The overlaps of binding sites, indicated by gray ellipses, are highlighted in red. The three figures illustrate three levels of spatial overlapping between binding sites. By construction, the size of the overlap on the A/A′ side is greater than on the B/C side.

Figure 5. Distribution of similarity P-value at protein-protein interfaces of ABAC pairs of the category O.

First row: all-atom representations, second row: coarse-grain representations, third row: C representations, first column: P-values of the A/A′ domains, second column: P-values of the B/C domains. White bars correspond to a number of similar elements equal to zero, which, by definition, yields a P-value equal to 1, since the random model cannot give a number of similar residues lower than zero.

Figure 6. Extent of overlap of interfaces in ABAC pairs.

Overlap of binding sites is highlighted in red. Residues involved in one of the binding sites but out of the overlap are highlighted in yellow. The four pairs of complexes belong to the O category.

Figure 7. Schematic view of promiscuous protein-protein binding at the family level.

Atoms/residues at the interfaces are symbolized by small squares and circles. The preferred atoms/residues in each complexes are highlighted in red, they are the key determinant of the complexes. A: different binding partners B and C interact at the same binding site of the similar proteins A and A′, but use their own set of atoms/residues. B: different binding partners B and C use atoms/residues out of the common binding site of A/A′. In both cases, binding sites of A and A′ are similar, but the alternate binding strategies can result in no similarity between B and C binding sites.