Co-evolutionary analysis reveals insights into protein-protein interactions

Protein-protein interactions play crucial roles in biological processes. Experimental methods have been developed to survey the proteome for interacting partners and some computational approaches have been developed to extend the impact of these experimental methods. Computational methods are routinely applied to newly discovered genes to infer protein function and plausible protein-protein interactions. Here, we develop and extend a quantitative method that identifies interacting proteins based upon the correlated behavior of the evolutionary histories of protein ligands and their receptors. We have studied six families of ligand-receptor pairs including: the syntaxin/Unc-18 family, the GPCR/G-alpha's, the TGF-beta/TGF-beta receptor system, the immunity/colicin domain collection from bacteria, the chemokine/chemokine receptors, and the VEGF/VEGF receptor family. For correlation scores above a defined threshold, we were able to find an average of 79% of all known binding partners. We then applied this method to find plausible binding partners for proteins with uncharacterized binding specificities in the syntaxin/Unc-18 protein and TGF-beta/TGF-beta receptor families. Analysis of the results shows that co-evolutionary analysis of interacting protein families can reduce the search space for identifying binding partners by not only finding binding partners for uncharacterized proteins but also recognizing potentially new binding partners for previously characterized proteins. We believe that correlated evolutionary histories provide a route to exploit the wealth of whole genome sequences and recent systematic proteomic results to extend the impact of these studies and focus experimental efforts to categorize physiologically or pathologically relevant protein-protein interactions.     

Coupled folding and binding with alpha-helix-forming molecular recognition elements

Many protein-protein and protein-nucleic acid interactions involve coupled folding and binding of at least one of the partners. Here, we propose a protein structural element or feature that mediates the binding events of initially disordered regions. This element consists of a short region that undergoes coupled binding and folding within a longer region of disorder. We call these features "molecular recognition elements" (MoREs). Examples of MoREs bound to their partners can be found in the alpha-helix, beta-strand, polyproline II helix, or irregular secondary structure conformations, and in various mixtures of the four structural forms. Here we describe an algorithm that identifies regions having propensities to become alpha-helix-forming molecular recognition elements (alpha-MoREs) based on a discriminant function that indicates such regions while giving a low false-positive error rate on a large collection of structured proteins. Application of this algorithm to databases of genomics and functionally annotated proteins indicates that alpha-MoREs are likely to play important roles protein-protein interactions involved in signaling events.     

Microscale mesoarrays created by dip-pen nanolithography for screening of protein-protein interactions

Using microarrays to probe protein-protein interactions is becoming increasingly attractive due to their compatibility with highly sensitive detection techniques, selectivity of interaction, robustness and capacity for examining multiple proteins simultaneously. The major drawback to using this approach is the relatively large volumes and high concentrations necessary. Reducing the protein array spot size should allow for smaller volumes and lower concentrations to be used as well as opening the way for combination with more sensitive detection technologies. Dip-Pen Nanolithography (DPN) is a recently developed technique for structure creation on the nano to microscale with the capacity to create biological architectures. Here we describe the creation of miniaturised microarrays, 'mesoarrays', using DPN with protein spots 400× smaller by area compared to conventional microarrays. The mesoarrays were then used to probe the ERK2-KSR binding event of the Ras/Raf/MEK/ERK signalling pathway at a physical scale below that previously reported. Whilst the overall assay efficiency was determined to be low, the mesoarrays could detect KSR binding to ERK2 repeatedly and with low non-specific binding. This study serves as a first step towards an approach that can be used for analysis of proteins at a concentration level comparable to that found in the cellular environment.     

Prediction of protein-protein interactions using distant conservation of sequence patterns and structure relationships

MOTIVATION: Given that association and dissociation of protein molecules is crucial in most biological processes several in silico methods have been recently developed to predict protein-protein interactions. Structural evidence has shown that usually interacting pairs of close homologs (interologs) physically interact in the same way. Moreover, conservation of an interaction depends on the conservation of the interface between interacting partners. In this article we make use of both, structural similarities among domains of known interacting proteins found in the Database of Interacting Proteins (DIP) and conservation of pairs of sequence patches involved in protein-protein interfaces to predict putative protein interaction pairs. RESULTS: We have obtained a large amount of putative protein-protein interaction (approximately 130,000). The list is independent from other techniques both experimental and theoretical. We separated the list of predictions into three sets according to their relationship with known interacting proteins found in DIP. For each set, only a small fraction of the predicted protein pairs could be independently validated by cross checking with the Human Protein Reference Database (HPRD). The fraction of validated protein pairs was always larger than that expected by using random protein pairs. Furthermore, a correlation map of interacting protein pairs was calculated with respect to molecular function, as defined in the Gene Ontology database. It shows good consistency of the predicted interactions with data in the HPRD database. The intersection between the lists of interactions of other methods and ours produces a network of potentially high-confidence interactions.     

In vitro protein microarrays for detecting protein-protein interactions: application of a new method for fluorescence labeling of proteins

Protein microarrays or proteome chips are potentially powerful tools for comprehensive analysis of protein-protein interactions. In interaction analysis, a set of immobilized proteins is arrayed on slides and each slide is probed with a set of fluorescently labeled proteins. Here we have developed and tested an in vitro protein microarray, in which both arraying and probing proteins were prepared by cell-free translation. The in vitro synthesis of fluorescently labeled proteins was accomplished by a new method: a fluorophore-puromycin conjugate was incorporated into a protein at the C-terminus on the ribosome. The resulting fluorescently labeled proteins were confirmed to be useful for probing protein-protein interactions on protein microarrays in model experiments. Since the in vitro protein microarrays can easily be extended to a high-throughput format and also combined with in vitro display technologies such as the streptavidin-biotin linkage in emulsions method (Doi and Yanagawa, FEBS Lett. 1999, 457, 227-230), our method should be useful for large-scale analysis of protein-protein interactions.     

Structural principles for the multispecificity of small GTP-binding proteins

The functional diversity of small GTP-binding proteins (G proteins) and their ability to function as molecular switches are based on their interactions with many different proteins. A wealth of structural data has revealed that their partners are often unrelated to each other in sequence and structure, but their binding sites are in general overlapping, notably at the so-called switch regions, whose conformation is sensitive to the nature of the bound nucleotide. We termed "multispecificity" this unique property of G proteins and investigated its structural principles by a database-implemented comparison of their protein-protein interfaces. Multispecific residues were found to be distributed throughout the G protein surface, with the highest multiplicity at the switch regions, each engaging interactions with 50-80% of the bound partners. Remarkably, residues involved in multiple interactions do not define consensus binding sites where all partners have convergent interactions. Rather, they adapt to multiple stereochemical and structural environments by combining the composite nature of amino acids with structural plasticity. We propose that not only the nucleotide switch but also multispecificity is the hallmark of the G protein module. Thus, G proteins are representative of highly connected proteins located at nodes of protein interactomes, probably the best structurally characterized member of this emerging class of proteins to date. This central functional property is also their Achilles' heal, facilitating their hijacking by pathogens, but may constitute an unexplored advantage in designing or screening novel therapeutic molecules.     

Protein microarrays as a discovery tool for studying protein-protein interactions

Exploring the function of the genome and the encoded proteins has emerged as a new and exciting challenge in the postgenomic era. Novel technologies come into view that promise to be valuable for the investigation not only of single proteins, but of entire protein networks. Protein microarrays are the innovative assay platform for highly parallel in vitro studies of protein-protein interactions. Due to their flexibility and multiplexing capacity, protein microarrays benefit basic research, diagnosis and biomedicine. This review provides an overview on the basic principles of protein microarrays and their potential to multiplex protein-protein interaction studies.     

Identification of VCP/p97, carboxyl terminus of Hsp70-interacting protein (CHIP), and amphiphysin II interaction partners using membrane-based human proteome arrays

Proteins mediate their biological function through interactions with other proteins. Therefore, the systematic identification and characterization of protein-protein interactions have become a powerful proteomic strategy to understand protein function and comprehensive cellular regulatory networks. For the screening of valosin-containing protein, carboxyl terminus of Hsp70-interacting protein (CHIP), and amphiphysin II interaction partners, we utilized a membrane-based array technology that allows the identification of human protein-protein interactions with crude bacterial cell extracts. Many novel interaction pairs such as valosin-containing protein/autocrine motility factor receptor, CHIP/caytaxin, or amphiphysin II/DLP4 were identified and subsequently confirmed by pull-down, two-hybrid and co-immunoprecipitation experiments. In addition, assays were performed to validate the interactions functionally. CHIP e.g. was found to efficiently polyubiquitinate caytaxin in vitro, suggesting that it might influence caytaxin degradation in vivo. Using peptide arrays, we also identified the binding motifs in the proteins DLP4, XRCC4, and fructose-1,6-bisphosphatase, which are crucial for the association with the Src homology 3 domain of amphiphysin II. Together these studies indicate that our human proteome array technology permits the identification of protein-protein interactions that are functionally involved in neurodegenerative disease processes, the degradation of protein substrates, and the transport of membrane vesicles.     

The Role of Protein Interactions in Mediating Essentiality and Synthetic Lethality

Genes are characterized as essential if their knockout is associated with a lethal phenotype, and these “essential genes” play a central role in biological function. In addition, some genes are only essential when deleted in pairs, a phenomenon known as synthetic lethality. Here we consider genes displaying synthetic lethality as “essential pairs” of genes, and analyze the properties of yeast essential genes and synthetic lethal pairs together. As gene duplication initially produces an identical pair or sets of genes, it is often invoked as an explanation for synthetic lethality. However, we find that duplication explains only a minority of cases of synthetic lethality. Similarly, disruption of metabolic pathways leads to relatively few examples of synthetic lethality. By contrast, the vast majority of synthetic lethal gene pairs code for proteins with related functions that share interaction partners. We also find that essential genes and synthetic lethal pairs cluster in the protein-protein interaction network. These results suggest that synthetic lethality is strongly dependent on the formation of protein-protein interactions. Compensation by duplicates does not usually occur mainly because the genes involved are recent duplicates, but is more commonly due to functional similarity that permits preservation of essential protein complexes. This unified view, combining genes that are individually essential with those that form essential pairs, suggests that essentiality is a feature of physical interactions between proteins protein-protein interactions, rather than being inherent in gene and protein products themselves.     

Reverse interactomics: decoding protein-protein interactions with combinatorial peptide libraries

Identification of binding partners is the crucial first step towards understanding the biological function of a protein. Many protein-protein interactions occur via modular domains that recognize short peptide motifs in their target proteins. Here we describe a chemical/bioinformatics approach for predicting the binding partners of modular domains. The optimal binding motif(s) of a protein domain is identified by screening a combinatorial peptide library. The resulting consensus sequence is used to search protein and genomic databases for potential binding proteins, which are subsequently confirmed (or disproved) by conventional protein binding assays (e.g. pull-down and co-immunoprecipitation).     

In silico discovery and experimental validation of new protein-protein interactions

We introduce a framework for predicting novel protein-protein interactions (PPIs), based on Fisher's method for combining probabilities of predictions that are based on different data sources, such as the biomedical literature, protein domain and mRNA expression information. Our method compares favorably to our previous method based on text-mining alone and other methods such as STRING. We evaluated our algorithms through the prediction of experimentally found protein interactions underlying Muscular Dystrophy, Huntington's Disease and Polycystic Kidney Disease, which had not yet been recorded in protein-protein interaction databases. We found a 1.74-fold increase in the mean average prediction precision for dysferlin and a 3.09-fold for huntingtin when compared to STRING. The top 10 of predicted interaction partners of huntingtin were analysed in depth. Five were identified previously, and the other five were new potential interaction partners. The full matrix of human protein pairs and their prediction scores are available for download. Our framework can be extended to predict other types of relationships such as proteins in a complex, pathway or related disease mechanisms.     

Analysis of molecular recognition features (MoRFs)

Several proteomic studies in the last decade revealed that many proteins are either completely disordered or possess long structurally flexible regions. Many such regions were shown to be of functional importance, often allowing a protein to interact with a large number of diverse partners. Parallel to these findings, during the last five years structural bioinformatics has produced an explosion of results regarding protein-protein interactions and their importance for cell signaling. We studied the occurrence of relatively short (10-70 residues), loosely structured protein regions within longer, largely disordered sequences that were characterized as bound to larger proteins. We call these regions molecular recognition features (MoRFs, also known as molecular recognition elements, MoREs). Interestingly, upon binding to their partner(s), MoRFs undergo disorder-to-order transitions. Thus, in our interpretation, MoRFs represent a class of disordered region that exhibits molecular recognition and binding functions. This work extends previous research showing the importance of flexibility and disorder for molecular recognition. We describe the development of a database of MoRFs derived from the RCSB Protein Data Bank and present preliminary results of bioinformatics analyses of these sequences. Based on the structure adopted upon binding, at least three basic types of MoRFs are found: α-MoRFs, β-MoRFs, and ι-MoRFs, which form α-helices, β-strands, and irregular secondary structure when bound, respectively. Our data suggest that functionally significant residual structure can exist in MoRF regions prior to the actual binding event. The contribution of intrinsic protein disorder to the nature and function of MoRFs has also been addressed. The results of this study will advance the understanding of protein-protein interactions and help towards the future development of useful protein-protein binding site predictors.     

A de novo protein binding pair by computational design and directed evolution

The de novo design of protein-protein interfaces is a stringent test of our understanding of the principles underlying protein-protein interactions and would enable unique approaches to biological and medical challenges. Here we describe a motif-based method to computationally design protein-protein complexes with native-like interface composition and interaction density. Using this method we designed a pair of proteins, Prb and Pdar, that heterodimerize with a Kd of 130 nM, 1000-fold tighter than any previously designed de novo protein-protein complex. Directed evolution identified two point mutations that improve affinity to 180 pM. Crystal structures of an affinity-matured complex reveal binding is entirely through the designed interface residues. Surprisingly, in the in vitro evolved complex one of the partners is rotated 180° relative to the original design model, yet still maintains the central computationally designed hotspot interaction and preserves the character of many peripheral interactions. This work demonstrates that high-affinity protein interfaces can be created by designing complementary interaction surfaces on two noninteracting partners and underscores remaining challenges.     

Protein-protein interactions more conserved within species than across species

Experimental high-throughput studies of protein-protein interactions are beginning to provide enough data for comprehensive computational studies. Today, about ten large data sets, each with thousands of interacting pairs, coarsely sample the interactions in fly, human, worm, and yeast. Another about 55,000 pairs of interacting proteins have been identified by more careful, detailed biochemical experiments. Most interactions are experimentally observed in prokaryotes and simple eukaryotes; very few interactions are observed in higher eukaryotes such as mammals. It is commonly assumed that pathways in mammals can be inferred through homology to model organisms, e.g. the experimental observation that two yeast proteins interact is transferred to infer that the two corresponding proteins in human also interact. Two pairs for which the interaction is conserved are often described as interologs. The goal of this investigation was a large-scale comprehensive analysis of such inferences, i.e. of the evolutionary conservation of interologs. Here, we introduced a novel score for measuring the overlap between protein-protein interaction data sets. This measure appeared to reflect the overall quality of the data and was the basis for our two surprising results from our large-scale analysis. Firstly, homology-based inferences of physical protein-protein interactions appeared far less successful than expected. In fact, such inferences were accurate only for extremely high levels of sequence similarity. Secondly, and most surprisingly, the identification of interacting partners through sequence similarity was significantly more reliable for protein pairs within the same organism than for pairs between species. Our analysis underlined that the discrepancies between different datasets are large, even when using the same type of experiment on the same organism. This reality considerably constrains the power of homology-based transfer of interactions. In particular, the experimental probing of interactions in distant model organisms has to be undertaken with some caution. More comprehensive images of protein-protein networks will require the combination of many high-throughput methods, including in silico inferences and predictions. http://www.rostlab.org/results/2006/ppi_homology/     

Cellular Localization and Characterization of Cytosolic Binding Partners for Gla Domain-containing Proteins PRRG4 and PRRG2

The genes encoding a family of proteins termed proline-rich γ-carboxyglutamic acid (PRRG) proteins were identified and characterized more than a decade ago, but their functions remain unknown. These novel membrane proteins have an extracellular γ-carboxyglutamic acid (Gla) protein domain and cytosolic WW binding motifs. We screened WW domain arrays for cytosolic binding partners for PRRG4 and identified novel protein-protein interactions for the protein. We also uncovered a new WW binding motif in PRRG4 that is essential for these newly found protein-protein interactions. Several of the PRRG-interacting proteins we identified are essential for a variety of physiologic processes. Our findings indicate possible novel and previously unidentified functions for PRRG proteins.     

Mass spectrometry identification of NASP binding partners in HeLa cells

Nuclear autoantigenic sperm protein (NASP) is a linker histone binding protein that is cell-cycle regulated. Synchronized HeLa cells are delayed in progression through the G1/S border when transiently transfected to overexpress full-length NASP, but not the histone-binding site (HBS) deletion mutant (NASP-DeltaHBS). The purpose of the current study was to identify possible NASP-associated proteins in HeLa cell nuclei that could elucidate NASP's influence on the cell cycle and chromatin remodeling. For this purpose, we employed a new approach: mass spectrometry identification of initially cross-linked proteins after their separation in a second dimension by reducing SDS-polyacrylamide gel electrophoresis (SDS-PAGE). Of the twelve proteins identified, three appear to be relevant to NASP's function: heat shock protein 90 (HSP90), DNA-activated protein kinase, and ATP-dependent DNA helicase II (70-kDa subunit). Individual protein-protein interactions were tested by immunoprecipitation techniques. This new method can be used for expedited identification of binding partners of different proteins in enriched fractions and as a complementary or alternative strategy to the yeast two-hybrid system and immunoprecipitation methods.     

Monitoring protein-protein interactions in mammalian cells by trans-SUMOylation

Protein-protein interactions are essential for almost all cellular processes, hence understanding these processes mainly depends on the identification and characterization of the relevant protein-protein interactions. In the present paper, we introduce the concept of TRS (trans-SUMOylation), a new method developed to identify and verify protein-protein interactions in mammalian cells in vivo. TRS utilizes Ubc9-fusion proteins that trans-SUMOylate co-expressed interacting proteins. Using TRS, we analysed interactions of 65 protein pairs co-expressed in HEK (human embryonic kidney)-293 cells. We identified seven new and confirmed 16 known protein interactions, which were determined via endogenous SUMOylation sites of the binding partners or by using SUMOylation-site tags respectively. Four of the new protein interactions were confirmed by GST (glutathione transferase) pull-down and the p38α-Edr2 interaction was verified by co-localization analysis. Functionally, this p38α-Edr2 interaction could possibly be involved in the recruitment of p38α to the polycomb chromatin-remodelling complex to phosphorylate Bmi1. We also used TRS to characterize protein-interaction domains of the protein kinase pairs p38α-MK2 [MK is MAPK (mitogen-activated protein kinase)-activated protein kinase] and ERK3 (extracellular-signal-regulated kinase 3)-MK5 and of the p38α-p53 complex. The ability of TRS to monitor protein interactions in mammalian cells in vivo at levels similar to endogenous expression makes it an excellent new tool that can help in defining the protein interactome of mammalian cells.     

Predicting protein-protein interaction by searching evolutionary tree automorphism space

MOTIVATION: Uncovering the protein-protein interaction network is a fundamental step in the quest to understand the molecular machinery of a cell. This motivates the search for efficient computational methods for predicting such interactions. Among the available predictors are those that are based on the co-evolution hypothesis "evolutionary trees of protein families (that are known to interact) are expected to have similar topologies". Many of these methods are limited by the fact that they can handle only a small number of protein sequences. Also, details on evolutionary tree topology are missing as they use similarity matrices in lieu of the trees. RESULTS: We introduce MORPH, a new algorithm for predicting protein interaction partners between members of two protein families that are known to interact. Our approach can also be seen as a new method for searching the best superposition of the corresponding evolutionary trees based on tree automorphism group. We discuss relevant facts related to the predictability of protein-protein interaction based on their co-evolution. When compared with related computational approaches, our method reduces the search space by approximately 3 x 10(5)-fold and at the same time increases the accuracy of predicting correct binding partners.