Distribution and complementarity of hydropathy in multisubunit proteins

Effectiveness and Efficiency: Label-Aware Hierarchical Subgraph Learning for Protein-Protein Interaction

The study of protein-protein interactions (PPIs) holds immense significance in understanding various biological activities, as well as in drug discovery and disease diagnosis. Existing deep learning methods for PPI prediction, including graph neural networks (GNNs), have been widely employed as the solutions, while they often experience a decline in performance in the real world. We claim that the topological shortcut is one of the key problems contributing negatively to the performance, according to our analysis. By modeling the PPIs as a graph with protein as nodes and interactions as edge types, the prevailing models tend to learn the pattern of nodes' degrees rather than intrinsic sequence-structure profiles, leading to the problem termed topological shortcut. The huge data growth of PPI leads to intensive computational costs and challenges computing devices, causing infeasibility in practice. To address the discussed problems, we propose a label-aware hierarchical subgraph learning method (laruGL-PPI) that can effectively infer PPIs while being interpretable. Specifically, we introduced edge-based subgraph sampling to effectively alleviate the problems of topological shortcuts and high computing costs. Besides, the inner-outer connections of PPIs are modeled as a hierarchical graph, together with the dependencies between interaction types constructed by a label graph. Extensive experiments conducted across various scales of PPI datasets have conclusively demonstrated that the laruGL-PPI method surpasses the most advanced PPI prediction techniques currently available, particularly in the testing of unseen proteins. Also, our model can recognize crucial sites of proteins, such as surface sites for binding and active sites for catalysis.

Hierarchical graph learning for protein-protein interaction

Protein-Protein Interactions (PPIs) are fundamental means of functions and signalings in biological systems. The massive growth in demand and cost associated with experimental PPI studies calls for computational tools for automated prediction and understanding of PPIs. Despite recent progress, in silico methods remain inadequate in modeling the natural PPI hierarchy. Here we present a double-viewed hierarchical graph learning model, HIGH-PPI, to predict PPIs and extrapolate the molecular details involved. In this model, we create a hierarchical graph, in which a node in the PPI network (top outside-of-protein view) is a protein graph (bottom inside-of-protein view). In the bottom view, a group of chemically relevant descriptors, instead of the protein sequences, are used to better capture the structure-function relationship of the protein. HIGH-PPI examines both outside-of-protein and inside-of-protein of the human interactome to establish a robust machine understanding of PPIs. This model demonstrates high accuracy and robustness in predicting PPIs. Moreover, HIGH-PPI can interpret the modes of action of PPIs by identifying important binding and catalytic sites precisely. Overall, "HIGH-PPI [ https://github.com/zqgao22/HIGH-PPI ]" is a domain-knowledge-driven and interpretable framework for PPI prediction studies.

DockNet: high-throughput protein-protein interface contact prediction

MOTIVATION

Over 300 000 protein-protein interaction (PPI) pairs have been identified in the human proteome and targeting these is fast becoming the next frontier in drug design. Predicting PPI sites, however, is a challenging task that traditionally requires computationally expensive and time-consuming docking simulations. A major weakness of modern protein docking algorithms is the inability to account for protein flexibility, which ultimately leads to relatively poor results.

RESULTS

Here, we propose DockNet, an efficient Siamese graph-based neural network method which predicts contact residues between two interacting proteins. Unlike other methods that only utilize a protein's surface or treat the protein structure as a rigid body, DockNet incorporates the entire protein structure and places no limits on protein flexibility during an interaction. Predictions are modeled at the residue level, based on a diverse set of input node features including residue type, surface accessibility, residue depth, secondary structure, pharmacophore and torsional angles. DockNet is comparable to current state-of-the-art methods, achieving an area under the curve (AUC) value of up to 0.84 on an independent test set (DB5), can be applied to a variety of different protein structures and can be utilized in situations where accurate unbound protein structures cannot be obtained.

AVAILABILITY AND IMPLEMENTATION

DockNet is available at https://github.com/npwilliams09/docknet and an easy-to-use webserver at https://biosig.lab.uq.edu.au/docknet. All other data underlying this article are available in the article and in its online supplementary material.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

The Spo7 sequence LLI is required for Nem1-Spo7/Pah1 phosphatase cascade function in yeast lipid metabolism

The Nem1-Spo7 complex in the yeast Saccharomyces cerevisiae is a protein phosphatase that catalyzes the dephosphory-lation of Pah1 phosphatidate phosphatase, required for its translocation to the nuclear/endoplasmic reticulum membrane. The Nem1-Spo7/Pah1 phosphatase cascade plays a major role in triacylglycerol synthesis and in the regulation of phospholipid synthesis. In this work, we examined Spo7, a regulatory subunit required for Nem1 catalytic function, to identify residues that govern formation of the Nem1-Spo7 complex. By deletion analysis of Spo7, we identified a hydrophobic Leu-Leu-Ile (LLI) sequence comprising residues 54-56 as being required for the protein to complement the temperature-sensitive phenotype of an spo7Δ mutant strain. Mutational analysis of the LLI sequence with alanine and arginine substitutions showed that its overall hydrophobicity is crucial for the formation of the Nem1-Spo7 complex as well as for the Nem1 catalytic function on its substrate, Pah1, in vivo Consistent with the role of the Nem1-Spo7 complex in activating the function of Pah1, we found that the mutational effects of the Spo7 LLI sequence were on the Nem1-Spo7/Pah1 axis that controls lipid synthesis and related cellular processes (e.g. triacylglycerol/phospholipid synthesis, lipid droplet formation, nuclear/endoplasmic reticulum membrane morphology, vacuole fusion, and growth on glycerol medium). These findings advance the understanding of Nem1-Spo7 complex formation and its role in the phosphatase cascade that regulates the function of Pah1 phosphatidate phosphatase.

Small protein-protein interfaces rich in electrostatic are often linked to regulatory function

Protein-protein interaction (PPI) is critical for several biological functions in living cells through the formation of an interface. Therefore, it is of interest to characterize protein-protein interfaces using an updated non-redundant structural dataset of 2557 homo (identical subunits) and 393 hetero (different subunits) dimer protein complexes determined by X-ray crystallography. We analyzed the interfaces using van der Waals (vdW), hydrogen bonding and electrostatic energies. Results show that on average homo and hetero interfaces are similar. Hence, we further grouped the 2950 interfaces based on percentage vdW to total energies into dominant (≥60%) and sub-dominant (<60%) vdW interfaces. Majority (92%) of interfaces have dominant vdW energy with large interface size (146 ± 87 (homo) and 137 ± 76 (hetero) residues) and interface area (1622 ± 1135 Ų (homo) and 1579 ± 1060 Ų (hetero)). However, a proportion (8%) of interfaces have sub-dominant vdW energy with small interface size (85 ± 46 (homo) and 88 ± 36 (hetero) residues) and interface area (823 ± 538 Ų (homo) and 881 ± 377 Ų (hetero)). It is found that large interfaces have two-fold more interface area and interface size than small interfaces with increasing hydrogen bonding energy to interface size. However, small interfaces have three-fold more electrostatics energy than large interfaces with increasing electrostatics to interface size. Thus, 8% of complexes having small interfaces with limited interface area and sub-dominant vdW energy are rich in electrostatics. It is interesting to observe that complexes having small interfaces are often associated with regulatory function. Hence, the observed structural features with known molecular function provide insights for the better understanding of PPI.Communicated by Ramaswamy H. Sarma.

A Hydrophobic Network: Intersubunit and Intercapsomer Interactions Stabilizing the Bacteriophage P22 Capsid

Double-stranded DNA (dsDNA) tailed phages and herpesviruses assemble their capsids using coat proteins that have the ubiquitous HK97 fold. Though this fold is common, we do not have a thorough understanding of the different ways viruses adapt it to maintain stability in various environments. The HK97-fold E-loop, which connects adjacent subunits at the outer periphery of capsomers, has been implicated in capsid stability. Here, we show that in bacteriophage P22, residue W61 at the tip of the E-loop plays a role in stabilizing procapsids and in maturation. We hypothesize that a hydrophobic pocket is formed by residues I366 and W410 in the P domain of a neighboring subunit within a capsomer, into which W61 fits like a peg. In addition, W61 likely bridges to residues A91 and L401 in P-domain loops of an adjacent capsomer, thereby linking the entire capsid together with a network of hydrophobic interactions. There is conservation of this hydrophobic network in the distantly related P22-like phages, indicating that this structural feature is likely important for stabilizing this family of phages. Thus, our data shed light on one of the varied elegant mechanisms used in nature to consistently build stable viral genome containers through subtle adaptation of the HK97 fold.IMPORTANCE Similarities in assembly reactions and coat protein structures of the dsDNA tailed phages and herpesviruses make phages ideal models to understand capsid assembly and identify potential targets for antiviral drug discovery. The coat protein E-loops of these viruses are involved in both intra- and intercapsomer interactions. In phage P22, hydrophobic interactions peg the coat protein subunits together within a capsomer, where the E-loop hydrophobic residue W61 of one subunit packs into a pocket of hydrophobic residues I366 and W410 of the adjacent subunit. W61 also makes hydrophobic interactions with A91 and L401 of a subunit in an adjacent capsomer. We show these intra- and intercapsomer hydrophobic interactions form a network crucial to capsid stability and proper assembly.

Predicting Protein-Protein Interaction Sites with a Novel Membership Based Fuzzy SVM Classifier

Predicting residues that participate in protein-protein interactions (PPI) helps to identify, which amino acids are located at the interface. In this paper, we show that the performance of the classical support vector machine (SVM) algorithm can further be improved with the use of a custom-designed fuzzy membership function, for the partner-specific PPI interface prediction problem. We evaluated the performances of both classical SVM and fuzzy SVM (F-SVM) on the PPI databases of three different model proteomes of Homo sapiens, Escherichia coli and Saccharomyces Cerevisiae and calculated the statistical significance of the developed F-SVM over classical SVM algorithm. We also compared our performance with the available state-of-the-art fuzzy methods in this domain and observed significant performance improvements. To predict interaction sites in protein complexes, local composition of amino acids together with their physico-chemical characteristics are used, where the F-SVM based prediction method exploits the membership function for each pair of sequence fragments. The average F-SVM performance (area under ROC curve) on the test samples in 10-fold cross validation experiment are measured as 77.07, 78.39, and 74.91 percent for the aforementioned organisms respectively. Performances on independent test sets are obtained as 72.09, 73.24 and 82.74 percent respectively. The software is available for free download from http://code.google.com/p/cmater-bioinfo.

Effective discrimination between biologically relevant contacts and crystal packing contacts using new determinants

In the structural models determined by X-ray crystallography, contacts between molecules can be divided into two categories: biologically relevant contacts and crystal packing contacts. With the growth in the number and quality of available large crystal packing contacts structures, distinguishing crystal packing contacts from biologically relevant contacts remains a difficult task, which can lead to wrong interpretation of structural models. In this study, we performed a systematic analysis on the biologically relevant contacts and crystal packing contacts. The analysis results reveal that biologically contacts are more tightly packed than crystal packing contacts. This property of biologically contacts may contribute to the formation of their interfacial core region. Meanwhile, the differences between the core and surface region of biologically contacts in amino acid composition and evolutionary measure are more dramatic than crystal packing contacts and these differences appear to be useful in distinguishing these two categories of contacts. On the basis of the features derived from our analysis, we developed a random forest model to classify biological relevant contacts and crystal packing contacts. Our method can achieve a high receiver operating curve of 0.923 in the 5-fold cross-validation and accuracies of 91.4% and 91.7% for two different test sets. Moreover, in a comparison study, our model outperforms other existing methods, such as DiMoVo, Pita, Pisa, and Eppic. We believe that this study will provide useful help in the validation of oligomeric proteins and protein complexes. The model and all data used in this paper are freely available at http://cic.scu.edu.cn/bioinformatics/bio-cry.zip.

PPIcons: identification of protein-protein interaction sites in selected organisms

The physico-chemical properties of interaction interfaces have a crucial role in characterization of protein-protein interactions (PPI). In silico prediction of participating amino acids helps to identify interface residues for further experimental verification using mutational analysis, or inhibition studies by screening library of ligands against given protein. Given the unbound structure of a protein and the fact that it forms a complex with another known protein, the objective of this work is to identify the residues that are involved in the interaction. We attempt to predict interaction sites in protein complexes using local composition of amino acids together with their physico-chemical characteristics. The local sequence segments (LSS) are dissected from the protein sequences using a sliding window of 21 amino acids. The list of LSSs is passed to the support vector machine (SVM) predictor, which identifies interacting residue pairs considering their inter-atom distances. We have analyzed three different model organisms of Escherichia coli, Saccharomyces Cerevisiae and Homo sapiens, where the numbers of considered hetero-complexes are equal to 40, 123 and 33 respectively. Moreover, the unified multi-organism PPI meta-predictor is also developed under the current work by combining the training databases of above organisms. The PPIcons interface residues prediction method is measured by the area under ROC curve (AUC) equal to 0.82, 0.75, 0.72 and 0.76 for the aforementioned organisms and the meta-predictor respectively.

Deciphering the shape and deformation of secondary structures through local conformation analysis

Background

Protein deformation has been extensively analysed through global methods based on RMSD, torsion angles and Principal Components Analysis calculations. Here we use a local approach, able to distinguish among the different backbone conformations within loops, α-helices and β-strands, to address the question of secondary structures' shape variation within proteins and deformation at interface upon complexation.

Results

Using a structural alphabet, we translated the 3 D structures of large sets of protein-protein complexes into sequences of structural letters. The shape of the secondary structures can be assessed by the structural letters that modeled them in the structural sequences. The distribution analysis of the structural letters in the three protein compartments (surface, core and interface) reveals that secondary structures tend to adopt preferential conformations that differ among the compartments. The local description of secondary structures highlights that curved conformations are preferred on the surface while straight ones are preferred in the core. Interfaces display a mixture of local conformations either preferred in core or surface. The analysis of the structural letters transition occurring between protein-bound and unbound conformations shows that the deformation of secondary structure is tightly linked to the compartment preference of the local conformations.

Conclusion

The conformation of secondary structures can be further analysed and detailed thanks to a structural alphabet which allows a better description of protein surface, core and interface in terms of secondary structures' shape and deformation. Induced-fit modification tendencies described here should be valuable information to identify and characterize regions under strong structural constraints for functional reasons.

Progress and challenges in predicting protein-protein interaction sites

The identification of protein-protein interaction sites is an essential intermediate step for mutant design and the prediction of protein networks. In recent years a significant number of methods have been developed to predict these interface residues and here we review the current status of the field. Progress in this area requires a clear view of the methodology applied, the data sets used for training and testing the systems, and the evaluation procedures. We have analysed the impact of a representative set of features and algorithms and highlighted the problems inherent in generating reliable protein data sets and in the posterior analysis of the results. Although it is clear that there have been some improvements in methods for predicting interacting sites, several major bottlenecks remain. Proteins in complexes are still under-represented in the structural databases and in particular many proteins involved in transient complexes are still to be crystallized. We provide suggestions for effective feature selection, and make it clear that community standards for testing, training and performance measures are necessary for progress in the field.

A feature-based approach to modeling protein-protein interaction hot spots

Identifying features that effectively represent the energetic contribution of an individual interface residue to the interactions between proteins remains problematic. Here, we present several new features and show that they are more effective than conventional features. By combining the proposed features with conventional features, we develop a predictive model for interaction hot spots. Initially, 54 multifaceted features, composed of different levels of information including structure, sequence and molecular interaction information, are quantified. Then, to identify the best subset of features for predicting hot spots, feature selection is performed using a decision tree. Based on the selected features, a predictive model for hot spots is created using support vector machine (SVM) and tested on an independent test set. Our model shows better overall predictive accuracy than previous methods such as the alanine scanning methods Robetta and FOLDEF, and the knowledge-based method KFC. Subsequent analysis yields several findings about hot spots. As expected, hot spots have a larger relative surface area burial and are more hydrophobic than other residues. Unexpectedly, however, residue conservation displays a rather complicated tendency depending on the types of protein complexes, indicating that this feature is not good for identifying hot spots. Of the selected features, the weighted atomic packing density, relative surface area burial and weighted hydrophobicity are the top 3, with the weighted atomic packing density proving to be the most effective feature for predicting hot spots. Notably, we find that hot spots are closely related to π–related interactions, especially π · · · π interactions.

A survey of available tools and web servers for analysis of protein-protein interactions and interfaces

The unanimous agreement that cellular processes are (largely) governed by interactions between proteins has led to enormous community efforts culminating in overwhelming information relating to these proteins; to the regulation of their interactions, to the way in which they interact and to the function which is determined by these interactions. These data have been organized in databases and servers. However, to make these really useful, it is essential not only to be aware of these, but in particular to have a working knowledge of which tools to use for a given problem; what are the tool advantages and drawbacks; and no less important how to combine these for a particular goal since usually it is not one tool, but some combination of tool-modules that is needed. This is the goal of this review.

A peptide containing residues 26-44 of tau protein impairs mitochondrial oxidative phosphorylation acting at the level of the adenine nucleotide translocator

Having confirmed that adenovirus-mediated overexpression of NH(2)-tau fragment lacking the first 25 aminoacids evokes a potent neurotoxic effect, sustained by protracted stimulation of NMDA receptors, in primary neuronal cultures we investigated whether and how chemically synthesized NH(2)-derived tau peptides, i.e. NH(2)-26-44 and NH(2)-1-25 fragments, affect mitochondrial function. We tested both fragments on each step of the processes leading to ATP synthesis via oxidative phosphorylation: i) electron flow via the respiratory chain from physiological substrates to oxygen with the activity of each individual complex of the respiratory chain investigated in some detail, ii) membrane potential generation arising from externally added succinate and iii) the activity of both the adenine nucleotide translocator and iv) ATP synthase. Oxidative phosphorylation is not affected by NH(2)-1-25 tau fragment, but dramatically impaired by NH(2)-26-44 tau fragment. Both cytochrome c oxidase and the adenine nucleotide translocator are targets of NH(2)-26-44 tau fragment, but adenine nucleotide translocator is the unique mitochondrial target responsible for impairment of oxidative phosphorylation by the NH(2)-26-44 tau fragment, which then exerts deleterious effects on cellular availability of ATP synthesized into mitochondria.

Dynamic alpha-helices: conformations that do not conform

Structural transitions are important for the stability and function of proteins, but these phenomena are poorly understood. An extensive analysis of Protein Data Bank entries reveals 103 regions in proteins with a tendency to transform from helical to nonhelical conformation and vice versa. We find that these dynamic helices, unlike other helices, are depleted in hydrophobic residues. Furthermore, the dynamic helices have higher surface accessibility and conformational mobility (P-value = 3.35e-07) than the rigid helices. Contact analyses show that these transitions result from protein-ligand, protein-nucleic acid, and crystal-contacts. The immediate structural environment differs quantitatively (P-value = 0.003) as well as qualitatively in the two alternate conformations. Often, dynamic helix experiences more contacts in its helical conformation than in the nonhelical counterpart (P-value = 0.001). There is differential preference for the type of short contacts observed in two conformational states. We also demonstrate that the regions in protein that can undergo such large conformational transitions can be predicted with a reasonable accuracy using logistic regression model of supervised learning. Our findings have implications in understanding the molecular basis of structural transitions that are coupled with binding and are important for the function and stability of the protein. Based on our observations, we propose that several functionally relevant regions on the protein surface can switch over their conformation from coil to helix and vice-versa, to regulate the recognition and binding of their partner and hence these may work as "molecular switches" in the proteins to regulate certain biological process. Our results supports the idea that protein structure-function paradigm should transform from static to a highly dynamic one.

Computational prediction of protein-protein interactions

Recently a number of computational approaches have been developed for the prediction of protein-protein interactions. Complete genome sequencing projects have provided the vast amount of information needed for these analyses. These methods utilize the structural, genomic, and biological context of proteins and genes in complete genomes to predict protein interaction networks and functional linkages between proteins. Given that experimental techniques remain expensive, time-consuming, and labor-intensive, these methods represent an important advance in proteomics. Some of these approaches utilize sequence data alone to predict interactions, while others combine multiple computational and experimental datasets to accurately build protein interaction maps for complete genomes. These methods represent a complementary approach to current high-throughput projects whose aim is to delineate protein interaction maps in complete genomes. We will describe a number of computational protocols for protein interaction prediction based on the structural, genomic, and biological context of proteins in complete genomes, and detail methods for protein interaction network visualization and analysis.

Hot spots-A review of the protein-protein interface determinant amino-acid residues

Proteins tendency to bind to one another in a highly specific manner forming stable complexes is fundamental to all biological processes. A better understanding of complex formation has many practical applications, which include the rational design of new therapeutic agents, and the analysis of metabolic and signal transduction networks. Alanine-scanning mutagenesis made possible the detection of the functional epitopes, and demonstrated that most of the protein-protein binding energy is related only to a group of few amino acids at intermolecular protein interfaces: the hot spots. The scope of this review is to summarize all the available information regarding hot spots for a better atomic understanding of their structure and function. The ultimate objective is to improve the rational design of complexes of high affinity and specificity as well as that of small molecules, which can mimic the functional epitopes of the proteic complexes.

Characterization of molecular recognition features, MoRFs, and their binding partners

Molecular Recognition Features (MoRFs) are short, interaction-prone segments of protein disorder that undergo disorder-to-order transitions upon specific binding, representing a specific class of intrinsically disordered regions that exhibit molecular recognition and binding functions. MoRFs are common in various proteomes and occupy a unique structural and functional niche in which function is a direct consequence of intrinsic disorder. Example MoRFs collected from the Protein Data Bank (PDB) have been divided into three subtypes according to their structures in the bound state: alpha-MoRFs form alpha-helices, beta-MoRFs form beta-strands, and iota-MoRFs form structures without a regular pattern of backbone hydrogen bonds. These example MoRFs were indicated to be intrinsically disordered in the absence of their binding partners by several criteria. In this study, we used several geometric and physiochemical criteria to examine the properties of 62 alpha-, 20 beta-, and 176 iota-MoRF complex structures. Interface residues were examined by calculating differences in accessible surface area between the complex and isolated monomers. The compositions and physiochemical properties of MoRF and MoRF partner interface residues were compared to the interface residues of homodimers, heterodimers, and antigen-antibody complexes. Our analysis indicates that there are significant differences in residue composition and several geometric and physicochemical properties that can be used to discriminate, with a high degree of accuracy, between various interfaces in protein interaction data sets. Implications of these findings for the development of MoRF-partner interaction predictors are discussed. In addition, structural changes upon MoRF-to-partner complex formation were examined for several illustrative examples.

Comprehensive statistical analysis of residues interaction specificity at protein-protein interfaces

We calculated interchain contacts on the atomic level for nonredundant set of 4602 protein-protein interfaces using an unbiased Voronoi-Delaune tessellation method, and made 20x20 residue contact matrixes both for homodimers and heterocomplexes. The area of contacts and the distance distribution for these contacts were calculated on both the residue and the atomic levels. We analyzed residue area distribution and showed the existence of two types of interresidue contacts: stochastic and specific. We also derived formulas describing the distribution of contact area for stochastic and specific interactions in parametric form. Maximum pairing preference index was found for Cys-Cys contacts and for oppositely charged interactions. A significant difference in residue contacts was observed between homodimers and heterocomplexes. Interfaces in homodimers were enriched with contacts between residues of the same type due to the effects of structure symmetry.

Interaction geometry involving planar groups in protein-protein interfaces

The geometry of interactions of planar residues is nonrandom in protein tertiary structures and gives rise to conventional, as well as nonconventional (X--H...pi, X--H...O, where X = C, N, or O) hydrogen bonds. Whether a similar geometry is maintained when the interaction is across the protein-protein interface is addressed here. The relative geometries of interactions involving planar residues, and the percentage of contacts giving rise to different types of hydrogen bonds are quite similar in protein structures and the biological interfaces formed by protein chains in homodimers and protein-protein heterocomplexes--thus pointing to the similarity of chemical interactions that occurs during protein folding and binding. However, the percentage is considerably smaller in the nonspecific and nonphysiological interfaces that are formed in crystal lattices of monomeric proteins. The C--H...O interaction linking the aromatic and the peptide groups is quite common in protein structures as well as the three types of interfaces. However, as the interfaces formed by crystal contacts are depleted in aromatic residues, the weaker hydrogen bond interactions would contribute less toward their stability.

Specificity of molecular interactions in transient protein-protein interaction interfaces

In this study, we investigate what types of interactions are specific to their biological function, and what types of interactions are persistent regardless of their functional category in transient protein-protein heterocomplexes. This is the first approach to analyze protein-protein interfaces systematically at the molecular interaction level in the context of protein functions. We perform systematic analysis at the molecular interaction level using classification and feature subset selection technique prevalent in the field of pattern recognition. To represent the physicochemical properties of protein-protein interfaces, we design 18 molecular interaction types using canonical and noncanonical interactions. Then, we construct input vector using the frequency of each interaction type in protein-protein interface. We analyze the 131 interfaces of transient protein-protein heterocomplexes in PDB: 33 protease-inhibitors, 52 antibody-antigens, 46 signaling proteins including 4 cyclin dependent kinase and 26 G-protein. Using kNN classification and feature subset selection technique, we show that there are specific interaction types based on their functional category, and such interaction types are conserved through the common binding mechanism, rather than through the sequence or structure conservation. The extracted interaction types are C(alpha)-- H...O==C interaction, cation...anion interaction, amine...amine interaction, and amine...cation interaction. With these four interaction types, we achieve the classification success rate up to 83.2% with leave-one-out cross-validation at k = 15. Of these four interaction types, C(alpha)--H...O==C shows binding specificity for protease-inhibitor complexes, while cation-anion interaction is predominant in signaling complexes. The amine ... amine and amine...cation interaction give a minor contribution to the classification accuracy. When combined with these two interactions, they increase the accuracy by 3.8%. In the case of antibody-antigen complexes, the sign is somewhat ambiguous. From the evolutionary perspective, while protease-inhibitors and sig-naling proteins have optimized their interfaces to suit their biological functions, antibody-antigen interactions are the happenstance, implying that antibody-antigen complexes do not show distinctive interaction types. Persistent interaction types such as pi...pi, amide-carbonyl, and hydroxyl-carbonyl interaction, are also investigated. Analyzing the structural orientations of the pi...pi stacking interactions, we find that herringbone shape is a major configuration in transient protein-protein interfaces. This result is different from that of protein core, where parallel-displaced configurations are the major configuration. We also analyze overall trend of amide-carbonyl and hydroxyl-carbonyl interactions. It is noticeable that nearly 82% of the interfaces have at least one hydroxyl-carbonyl interactions.

An algorithm for predicting protein-protein interaction sites: Abnormally exposed amino acid residues and secondary structure elements

Multiprotein systems mediate most regulatory processes in living organisms. Although the structures of the individual proteins are often defined, less is known of the structures of multiprotein systems. Computational methods for predicting interfaces, using evolutionary conservation and/or physicochemical data, have been developed. Here we consider the use of solvent accessibility, residue propensity, and hydrophobicity, in conjunction with secondary structure data, as prediction parameters. We analyze the influence of residue type and secondary structure on solvent accessibility and define a measure of "relative exposedness." Clustering abnormally high scoring residues provides a basis for predicting interaction sites. The analysis is extended to investigate abnormally exposed secondary structure elements, particularly beta-sheet strands. We show that surface-exposed beta-strands lacking protective features are more likely to be found at protein-protein interfaces, allowing us to create an algorithm with approximately 68% and approximately 75% accuracy in differentiating between interacting and edge strands in isolated beta-strands and beta-sheet strands, respectively. These methods of identifying abnormally exposed surface regions are combined in an algorithm, which, on a data set of 77 unbound and disjoint (single chain extracted from complex) structures, predicts 79% of the protein-protein interfaces correctly. If enzyme-inhibitor complexes, where the inhibitor mimics a nonprotein substrate, are excluded, the accuracy increases to 85%.

Statistical analysis and prediction of protein-protein interfaces

Predicting protein-protein interfaces from a three-dimensional structure is a key task of computational structural proteomics. In contrast to geometrically distinct small molecule binding sites, protein-protein interface are notoriously difficult to predict. We generated a large nonredundant data set of 1494 true protein-protein interfaces using biological symmetry annotation where necessary. The data set was carefully analyzed and a Support Vector Machine was trained on a combination of a new robust evolutionary conservation signal with the local surface properties to predict protein-protein interfaces. Fivefold cross validation verifies the high sensitivity and selectivity of the model. As much as 97% of the predicted patches had an overlap with the true interface patch while only 22% of the surface residues were included in an average predicted patch. The model allowed the identification of potential new interfaces and the correction of mislabeled oligomeric states.

Statistical analysis of predominantly transient protein-protein interfaces

A non-redundant set of 170 protein-protein interfaces of known structure was statistically analyzed for residue and secondary-structure compositions, pairing preferences and side-chain-backbone interaction frequencies. By focussing mainly on transient protein-protein interfaces, the results underline previous findings for protein-protein interfaces but also show some new interesting aspects of transient interfaces. The residue compositions at interfaces found in this study correlate well with the results of other studies. On average, contacts between pairs of hydrophobic and polar residues were unfavorable, and the charged residues tended to pair subject to charge complementarity. Secondary structure composition analysis shows that neither helices nor beta-sheets are dominantly populated at interfaces. Analyzing the pairing preferences of the secondary structure elements revealed a higher affinity within the same elements and alludes to tight packings. In addition, the results for the side-chain and backbone interaction frequencies, which were measured under more stringent conditions, showed a high occurrence of side-chain-backbone interactions. Taking a closer look at the helix and beta-sheet binding frequencies for a given side-chain and backbone interaction underlined the relevance of tight packings. The polarity of interfaces increased with decreasing interface size. These types of information may be useful for scoring complexes in protein-protein docking studies or for prediction of protein-protein interfaces from the sequences alone.

Morphological aspects of oligomeric protein structures

Features of multimeric proteins are reviewed to shed light on the formation of protein assemblies from a structural perspective. The features comprise biochemical and geometric properties. They are compiled on new low-redundancy sets of crystal structures of homomeric proteins with different symmetry and subunit multiplicity, as well as on a set of heteromeric proteins. Crystal structures of likely monomers provide a control group.

Protein–protein interactions: organization, cooperativity and mapping in a bottom-up Systems Biology approach

Understanding and ultimately predicting protein associations is immensely important for functional genomics and drug design. Here, we propose that binding sites have preferred organizations. First, the hot spots cluster within densely packed 'hot regions'. Within these regions, they form networks of interactions. Thus, hot spots located within a hot region contribute cooperatively to the stability of the complex. However, the contributions of separate, independent hot regions are additive. Moreover, hot spots are often already pre-organized in the unbound (free) protein states. Describing a binding site through independent local hot regions has implications for binding site definition, design and parametrization for prediction. The compactness and cooperativity emphasize the similarity between binding and folding. This proposition is grounded in computation and experiment. It explains why summation of the interactions may over-estimate the stability of the complex. Furthermore, statistically, charge-charge coupling of the hot spots is disfavored. However, since within the highly packed regions the solvent is screened, the electrostatic contributions are strengthened. Thus, we propose a new description of protein binding sites: a site consists of (one or a few) self-contained cooperative regions. Since the residue hot spots are those conserved by evolution, proteins binding multiple partners at the same sites are expected to use all or some combination of these regions.

Hot Regions in Protein–Protein Interactions: The Organization and Contribution of Structurally Conserved Hot Spot Residues

Structurally conserved residues at protein-protein interfaces correlate with the experimental alanine-scanning hot spots. Here, we investigate the organization of these conserved, computational hot spots and their contribution to the stability of protein associations. We find that computational hot spots are not homogeneously distributed along the protein interfaces; rather they are clustered within locally tightly packed regions. Within the dense clusters, they form a network of interactions and consequently their contributions to the stability of the complex are cooperative; however the contributions of independent clusters are additive. This suggests that the binding free energy is not a simple summation of the single hot spot residue contributions. As expected, around the hot spot residues we observe moderately conserved residues, further highlighting the crucial role of the conserved interactions in the local densely packed environment. The conserved occurrence of these organizations suggests that they are advantageous for protein-protein associations. Interestingly, the total number of hydrogen bonds and salt bridges contributed by hot spots is as expected. Thus, H-bond forming residues may use a "hot spot for water exclusion" mechanism. Since conserved residues are located within highly packed regions, water molecules are easily removed upon binding, strengthening electrostatic contributions of charge-charge interactions. Hence, the picture that emerges is that protein-protein associations are optimized locally, with the clustered, networked, highly packed structurally conserved residues contributing dominantly and cooperatively to the stability of the complex. When addressing the crucial question of "what are the preferred ways of proteins to associate", these findings point toward a critical involvement of hot regions in protein-protein interactions.

Protein-protein interactions; coupling of structurally conserved residues and of hot spots across interfaces. Implications for docking

Hot spot residues contribute dominantly to protein-protein interactions. Statistically, conserved residues correlate with hot spots, and their occurrence can distinguish between binding sites and the remainder of the protein surface. The hot spot and conservation analyses have been carried out on one side of the interface. Here, we show that both experimental hot spots and conserved residues tend to couple across two-chain interfaces. Intriguingly, the local packing density around both hot spots and conserved residues is higher than expected. We further observe a correlation between local packing density and experimental deltadeltaG. Favorable conserved pairs include Gly coupled with aromatics, charged and polar residues, as well as aromatic residue coupling. Remarkably, charged residue couples are underrepresented. Overall, protein-protein interactions appear to consist of regions of high and low packing density, with the hot spots organized in the former. The high local packing density in binding interfaces is reminiscent of protein cores.

Are protein-protein interfaces more conserved in sequence than the rest of the protein surface?

Protein interfaces are thought to be distinguishable from the rest of the protein surface by their greater degree of residue conservation. We test the validity of this approach on an expanded set of 64 protein-protein interfaces using conservation scores derived from two multiple sequence alignment types, one of close homologs/orthologs and one of diverse homologs/paralogs. Overall, we find that the interface is slightly more conserved than the rest of the protein surface when using either alignment type, with alignments of diverse homologs showing marginally better discrimination. However, using a novel surface-patch definition, we find that the interface is rarely significantly more conserved than other surface patches when using either alignment type. When an interface is among the most conserved surface patches, it tends to be part of an enzyme active site. The most conserved surface patch overlaps with 39% (+/- 28%) and 36% (+/- 28%) of the actual interface for diverse and close homologs, respectively. Contrary to results obtained from smaller data sets, this work indicates that residue conservation is rarely sufficient for complete and accurate prediction of protein interfaces. Finally, we find that obligate interfaces differ from transient interfaces in that the former have significantly fewer alignment gaps at the interface than the rest of the protein surface, as well as having buried interface residues that are more conserved than partially buried interface residues.

MIAX: a new paradigm for modeling biomacromolecular interactions and complex formation in condensed phases

A new paradigm is proposed for modeling biomacromolecular interactions and complex formation in solution (protein-protein interactions so far in this report) that constitutes the scaffold of the automatic system MIAX (acronym for Macromolecular Interaction Assessment X). It combines in a rational way a series of computational methodologies, the goal being the prediction of the most native-like protein complex that may be formed when two isolated (unbound) protein monomers interact in a liquid environment. The overall strategy consists of first inferring putative precomplex structures by identification of binding sites or epitopes on the proteins surfaces and a simultaneous rigid-body docking process using geometric instances alone. Precomplex configurations are defined here as all those decoys the interfaces of which comply substantially with the inferred binding sites and whose free energy values are lower. Retaining all those precomplex configurations with low energies leads to a reasonable number of decoys for which a flexible treatment is amenable. A novel algorithm is introduced here for automatically inferring binding sites in proteins given their 3-D structure. The procedure combines an unsupervised learning algorithm based on the self-organizing map or Kohonen network with a 2-D Fourier spectral analysis. To model interaction, the potential function proposed here plays a central role in the system and is constituted by empirical terms expressing well-characterized factors influencing biomacromolecular interaction processes, essentially electrostatic, van der Waals, and hydrophobic. Each of these procedures is validated by comparing results with observed instances. Finally, the more demanding process of flexible docking is performed in MIAX embedding the potential function in a simulated annealing optimization procedure. Whereas search of the entire configuration hyperspace is a major factor precluding hitherto systems from efficiently modeling macromolecular interaction modes and complex structures, the paradigm presented here may constitute a step forward in the field because it is shown that a rational treatment of the information available from the 3-D structure of the interacting monomers combined with conveniently selected computational techniques can assist to elude search of regions of low probability in configuration space and indeed lead to a highly efficient system oriented to solve this intriguing and fundamental biologic problem.

Dissecting protein-protein recognition sites

The recognition sites in 70 pairwise protein-protein complexes of known three-dimensional structure are dissected in a set of surface patches by clustering atoms at the interface. When the interface buries <2000 A2 of protein surface, the recognition sites usually form a single patch on the surface of each component protein. In contrast, larger interfaces are generally multipatch, with at least one pair of patches that are equivalent in size to a single-patch interface. Each recognition site, or patch within a site, contains a core made of buried interface atoms, surrounded by a rim of atoms that remain accessible to solvent in the complex. A simple geometric model reproduces the number and distribution of atoms within a patch. The rim is similar in composition to the rest of the protein surface, but the core has a distinctive amino acid composition, which may help in identifying potential protein recognition sites on single proteins of known structures.

Molecular mechanisms of protein-protein recognition: whether the surface placed charged residues determine the recognition process?

We studied the structure and composition of contact areas in 812 different kind dimeric protein-protein complexes from Brookhaven data base (PDB ) in order to reveal their pecularities with regard to protein-protein recognition. We have found, that the large portion of complexes (approximately 70%) have oppositely charged residues in the contact areas (interfaces) on the subunits surfaces, which form electrostatic contacts - R:E, R:D, K:E, K:D, H:E, H:D. These results are consistent with the current view that high rate complex formation may be driven by the long-range electrostatic interaction between charged AA residues of subunits surfaces. However, there are many complexes among the studied ones (approximately 30%), which have no electrostatic contacts at all in their contact area. Thus a question arises: what forces account for high complex formation rates (i.e. for the distant orienting of subunits before encounter) by forming complexes where the surface contact areas lack electrostatic contacts? We believe that the long-range orienting electrostatic interaction of subunits may account for all cases of efficient complex formation if one drops the traditional view that protein subunits interact mainly through their surfaces. We suggest that the distant orienting being due to the electrostatic interaction between the whole aggregates of partial electric charges of atoms of each complex subunits. Our preliminary model calculations (unpublished) made for ribonuclease dimer (does not have electrostatic contacts) conform this suggestion.

Dimerisation mutants of Lac repressor. II. A single amino acid substitution, D278L, changes the specificity of dimerisation

Assembly of the lactose repressor tetramer involves two subunit interfaces, the C-terminal heptad repeats, and the monomer-monomer interface. Dimerisation between two monomers of Lac repressor of Escherichia coli lacking the two C-terminal heptad repeats occurs through the interactions between three alpha-helices of each monomer, which form a highly hydrophobic interface. Residues possibly involved in specific dimer formation are known from X-ray studies and from the phenotypes of more than 4000 single amino acid substitutions. During the examination of numerous mutants within the dimerisation interface of Lac repressor, we found that substitution of one amino acid, D278 to leucine, is sufficient to change the specificity of dimerisation. Analysis of this single substitution indicates that D278L mutant Lac repressor represses like wild-type. However, it no longer forms heterodimers with wild-type Lac repressor.

The evolutionary transition from monomeric to oligomeric proteins: tools, the environment, hypotheses

Recently, renewed interest in the evolution of oligomeric proteins has seen new approaches explored and new hypotheses proposed. The model systems chosen are generally made up of pairs of homologous proteins, each composed of a monomer and a dimeric counterpart, but the question has been also approached by comparing statistically significant structural patterns in sets of monomeric and oligomeric proteins. Here the tools of genetics and chemistry potentially available to the evolution of oligomeric proteins are discussed, as well as the possible effects of environments on the early attempts to oligomerization. Traces of an ancestral monomeric status of oligomers may be detected in the significant presence of polar and charged residues at intersubunit interfaces, and by the recognition that, besides the hydrophobic effect, a 'hydrophilic' effect has also had a role in the construction of these interfaces. The traditional 'mutation' model is described and found to be based on a hierarchy of mutations, crowned by a 'primary' mutation, one that could prime oligomerization by irreversibly altering the structure of an ancestral monomer. The mechanism of oligomerization based on the exchange or 'swap' of structural elements between monomers is discussed. The possibility is also discussed that the main steps in the folding pathway of an oligomeric protein reiterate the main steps in its evolution.

Variability in quaternary association of proteins with the same tertiary fold: a case study and rationalization involving legume lectins

Legume lectins constitute a family of proteins in which small alterations arising from sequence variations in essentially the same tertiary structure lead to large changes in quaternary association. All of them are dimers or tetramers made up of dimers. Dimerization involves side-by-side or back-to-back association of the flat six-membered beta-sheets in the protomers. Variations within these modes of dimerization can be satisfactorily described in terms of angles defining the mutual disposition of the two subunits. In all tetrameric lectins, except peanut lectin, oligomerization involves the back-to-back association of side-by-side dimers. An attempt has been made to rationalize the observed modes of oligomerization in terms of hydrophobic surface area buried on association, interaction energy and shape complementarity, by constructing energy minimised models in each of which the subunit of one legume lectin is fitted in the quaternary structure of another. The results indicate that all the three indices favor and, thus, provide a rationale for the observed arrangements. However, the discrimination provided by buried hydrophobic surface area is marginal in a few instances. The same is true, to a lesser extent, about that provided by shape complementarity. The relative values of interaction energy turns out to be a still better discriminator than the other two indices. Variability in the quaternary association of homologous proteins is a widely observed phenomenon and the present study is relevant to the general problem of protein folding.

Protein folding via binding and vice versa

The terms intermolecular and intramolecular recognition are often used when referring to binding and folding, highlighting the common ground between the two processes. Most studies, however, are aimed at either one process or the other. Here, we show how knowledge from binding can aid in understanding folding and vice versa.

Morphology of protein-protein interfaces

BACKGROUND

Most soluble proteins are active as low-number oligomers. Statistical surveys of oligomeric proteins have defined the roles of hydrophobicity and complementarity in the stability of protein interfaces, but tend to average structural features over a diverse set of protein-protein interfaces, blurring information on how individual interfaces are stabilized.

RESULTS

We report a visual survey of 136 homodimeric proteins from the Brookhaven Protein Data Bank, with images that highlight the major structural features of each protein-protein interaction surface. Nearly all of these proteins have interfaces formed between two globular subunits. Surprisingly, the pattern of hydrophilicity over the surface of these interfaces is quite variable. Approximately one-third of the interfaces show a recognizable hydrophobic core, with a single large, contiguous, hydrophobic patch surrounded by a ring of intersubunit polar interactions. The remaining two-thirds of the proteins show a varied mixture of small hydrophobic patches, polar interactions and water molecules scattered over the entire interfacial area. Ten proteins in the survey have intertwined interfaces formed by extensive interdigitation of the two subunit chains. These interfaces are very hydrophobic and are associated with proteins that require both stability and internal symmetry.

CONCLUSIONS

The archetypal protein interface, with a defined hydrophobic core, is present in only a minority of the surveyed homodimeric proteins. Most homodimeric proteins are stabilized by a combination of small hydrophobic patches, polar interactions and a considerable number of bridging water molecules. The presence or absence of a hydrophobic core within these interfaces does not correlate with specific protein functions.