Protein-protein docking: progress in CAPRI rounds 6-12 using a combination of methods: the introduction of steered solvated molecular dynamics

IN SILICO SCREENING OF PROTEIN–PROTEIN INTERACTIONS WITH ALL-TO-ALL RIGID DOCKING AND CLUSTERING: AN APPLICATION TO PATHWAY ANALYSIS

We propose a computational screening system of protein–protein interactions using tertiary structure data. Our system combines all-to-all protein docking and clustering to find interacting protein pairs. We tuned our prediction system by applying various parameters and clustering algorithms and succeeded in outperforming previous methods. This method was also applied to a biological pathway estimation problem to show its use in network level analysis. The structural data were collected from the Protein Data Bank, PDB. Then all-to-all docking among target protein structures was conducted using a conventional protein–protein docking software package, ZDOCK. The highest-ranked 2000 decoys were clustered based on structural similarity among the predicted docking forms. The features of generated clusters were analyzed to estimate the biological relevance of protein–protein interactions. Our system achieves a best F-measure value of 0.43 when applied to a subset of general protein–protein docking benchmark data. The same system was applied to protein data in a bacterial chemotaxis pathway, utilizing essentially the same parameter set as the benchmark data. We obtained 0.45 for the F-measure value. The proposed approach to computational PPI detection is a promising methodology for mediating between structural studies and systems biology by utilizing cumulative protein structure data for pathway analysis.

Analysis and prediction of protein quaternary structure

The quaternary structure (QS) of a protein is determined by measuring its molecular weight in solution. The data have to be extracted from the literature, and they may be missing even for proteins that have a crystal structure reported in the Protein Data Bank (PDB). The PDB and other databases derived from it report QS information that either was obtained from the depositors or is based on an analysis of the contacts between polypeptide chains in the crystal, and this frequently differs from the QS determined in solution.The QS of a protein can be predicted from its sequence using either homology or threading methods. However, a majority of the proteins with less than 30% sequence identity have different QSs. A model of the QS can also be derived by docking the subunits when their 3D structure is independently known, but the model is likely to be incorrect if large conformation changes take place when the oligomer assembles.

Classification of heterodimer interfaces using docking models and construction of scoring functions for the complex structure prediction

Protein–protein docking simulations can provide the predicted complex structural models. In a docking simulation, several putative structural models are selected by scoring functions from an ensemble of many complex models. Scoring functions based on statistical analyses of heterodimers are usually designed to select the complex model with the most abundant interaction mode found among the known complexes, as the correct model. However, because the formation schemes of heterodimers are extremely diverse, a single scoring function does not seem to be sufficient to describe the fitness of the predicted models other than the most abundant interaction mode. Thus, it is necessary to classify the heterodimers in terms of their individual interaction modes, and then to construct multiple scoring functions for each heterodimer type. In this study, we constructed the classification method of heterodimers based on the discriminative characters between near-native and decoy models, which were found in the comparison of the interfaces in terms of the complementarities for the hydrophobicity, the electrostatic potential and the shape. Consequently, we found four heterodimer clusters, and then constructed the multiple scoring functions, each of which was optimized for each cluster. Our multiple scoring functions were applied to the predictions in the unbound docking.

Protein-protein docking and analysis reveal that two homologous bacterial adenylyl cyclase toxins interact with calmodulin differently

Calmodulin (CaM), a eukaryotic calcium sensor that regulates diverse biological activities, consists of N- and C-terminal globular domains (N-CaM and C-CaM, respectively). CaM serves as the activator of CyaA, a 188-kDa adenylyl cyclase toxin secreted by Bordetella pertussis, which is the etiologic agent for whooping cough. Upon insertion of the N-terminal adenylyl cyclase domain (ACD) of CyaA to its targeted eukaryotic cells, CaM binds to this domain tightly ( approximately 200 pm affinity). This interaction activates the adenylyl cyclase activity of CyaA, leading to a rise in intracellular cAMP levels to disrupt normal cellular signaling. We recently solved the structure of CyaA-ACD in complex with C-CaM to elucidate the mechanism of catalytic activation. However, the structure of the interface between N-CaM and CyaA, the formation of which contributes a 400-fold increase of binding affinity between CyaA and CaM, remains elusive. Here, we used site-directed mutations and molecular dynamic simulations to generate several working models of CaM-bound CyaA-ACD. The validity of these models was evaluated by disulfide bond cross-linking, point mutations, and fluorescence resonance energy transfer experiments. Our study reveals that a beta-hairpin region (amino acids 259-273) of CyaA-ACD likely makes contacts with the second calcium binding motif of the extended CaM. This mode of interaction differs from the interaction of N-CaM with anthrax edema factor, which binds N-CaM via its helical domain. Thus, two structurally conserved, bacterial adenylyl cyclase toxins have evolved to utilize distinct binding surfaces and modes of activation in their interaction with CaM, a highly conserved eukaryotic signaling protein.

The RosettaDock server for local protein-protein docking

The RosettaDock server (http://rosettadock.graylab.jhu.edu) identifies low-energy conformations of a protein–protein interaction near a given starting configuration by optimizing rigid-body orientation and side-chain conformations. The server requires two protein structures as inputs and a starting location for the search. RosettaDock generates 1000 independent structures, and the server returns pictures, coordinate files and detailed scoring information for the 10 top-scoring models. A plot of the total energy of each of the 1000 models created shows the presence or absence of an energetic binding funnel. RosettaDock has been validated on the docking benchmark set and through the Critical Assessment of PRedicted Interactions blind prediction challenge.