Structural and functional analysis of multi-interface domains

Abstract

A multi-interface domain is a domain that can shape multiple and distinctive binding sites to contact with many other domains, forming a hub in domain-domain interaction networks. The functions played by the multiple interfaces are usually different, but there is no strict bijection between the functions and interfaces as some subsets of the interfaces play the same function. This work applies graph theory and algorithms to discover fingerprints for the multiple interfaces of a domain and to establish associations between the interfaces and functions, based on a huge set of multi-interface proteins from PDB. We found that about 40% of proteins have the multi-interface property, however the involved multi-interface domains account for only a tiny fraction (1.8%) of the total number of domains. The interfaces of these domains are distinguishable in terms of their fingerprints, indicating the functional specificity of the multiple interfaces in a domain. Furthermore, we observed that both cooperative and distinctive structural patterns, which will be useful for protein engineering, exist in the multiple interfaces of a domain.

Figure 1. Multi-interface domain illustration.

Domain a of protein A interacting with domain b of protein B produces interface a–b on domain a, and domain a binding to protein C generates interface a–c on domain a. Interfaces a–b and a–c are distinguishable on domain a, thus domain a is a multi-interface domain.

Figure 2. Multiple interfaces in the catalytic domain of plasmin.

The interfaces are colored in limegreen (46 residues), marine (6 residues), and red (7 residues) for plasmin interacting with a streptokinase, a protein inhibitor, and another plasmin symmetric unit, respectively. The two overlapping residues are colored orange, and the five molecular functions of this domain retrieved from GO are shown at the top right corner.

Figure 3. Flowchart of multi-interface protein data set construction and data analysis.

Figure 4. Multi-interface domain distribution at different SCOP levels.

Length of lines represents the normalized number of multi-interface proteins at each classification level, and multiple lines under the same level of one cluster represent different sub-clusters. The clusters are organized as a rooted tree structure from a higher level to a lower level, and the clusters at the same level are plotted in the clockwise descendant order. Here a represents and b represents .

Figure 5. Amino acid distribution of the antigen-binding interface and the protein-binding interface in the Ig VH domain.

Figure 6. Amino acid distribution of the six binding sites in the proteasome beta subunit domain.

bs means binding site.

Figure 7. An example of a fingerprint in the proteasome beta subunit domain.

(a) is a diagram of the given fingerprint, where the filled circles represent the interface residues, and the lines represent the contacts. Color and color shade represent residue hydrophobicity index defined by Kyte and Doolittle . (b) is the real structure of (a) presented in PDB entry 3NZX. The dashed lines represent the contacts determined by Delaunay triangulation, and the highlighted lines are the contacts shown in (a).

Figure 8. An example of a pair of fingerprints from multiple interfaces within the Ig VH domain.

The dash line separated structures are fingerprints, and the solid orange lines between these fingerprints represent cooperative relations between them. The filled circles represent interface residues and the solid green lines represent residue contacts. Color of residues indicates their hydrophobicity.

Figure 9. Fingerprints assortativity of the Ig VH domain and Ig VL-kappa domain.

Figure 10. An example of a cross-domain interface in a multi-interface domain.

The two interfaces are colored by orange and forest, respectively. The forest colored interface is a cross-domain interface formed by the interaction between chain H (rendered by surface) and chain G (rendered by cartoon). Chain H has two domains, which are Hemochromatosis protein Hfe -1 and 2 domain (the left part of chain H) and Hemochromatosis protein Hfe -3 domain (the right part of chain H).

Figure 11. Distribution of cross-domain interfaces at various SCOP class levels.

A line between two classes means a cross-domain interface in these two classes. The weight on each line indicates the number of instances that have cross-domain interfaces in our data set. Cross-domain multi-interfaces with very small numbers are not shown here.