Modeling protein-protein,protein-peptide,and protein-oligosaccharide complexes: CAPRI 7th edition

We present the seventh report on the performance of methods for predicting the atomic resolution structures of protein complexes offered as targets to the community-wide initiative on the Critical Assessment of Predicted Interactions. Performance was evaluated on the basis of 36 114 models of protein complexes submitted by 57 groups—including 13 automatic servers—in prediction rounds held during the years 2016 to 2019 for eight protein-protein, three protein-peptide, and five protein-oligosaccharide targets with different length ligands. Six of the protein-protein targets represented challenging hetero-complexes, due to factors such as availability of distantly related templates for the individual subunits, or for the full complex, inter-domain flexibility, conformational adjustments at the binding region, or the multi-component nature of the complex. The main challenge for the protein-peptide and protein-oligosaccharide complexes was to accurately model the ligand conformation and its interactions at the interface. Encouragingly, models of acceptable quality, or better, were obtained for a total of six protein-protein complexes, which included four of the challenging hetero-complexes and a homo-decamer. But fewer of these targets were predicted with medium or higher accuracy. High accuracy models were obtained for two of the three protein-peptide targets, and for one of the protein-oligosaccharide targets. The remaining protein-sugar targets were predicted with medium accuracy. Our analysis indicates that progress in predicting increasingly challenging and diverse types of targets is due to closer integration of template-based modeling techniques with docking, scoring, and model refinement procedures, and to significant incremental improvements in the underlying methodologies.     

Structural modeling of protein complexes: Current capabilities and challenges

Proteins frequently interact with each other, and the knowledge of structures of the corresponding protein complexes is necessary to understand how they function. Computational methods are increasingly used to provide structural models of protein complexes. Not surprisingly, community-wide Critical Assessment of protein Structure Prediction (CASP) experiments have recently started monitoring the progress in this research area. We participated in CASP13 with the aim to evaluate our current capabilities in modeling of protein complexes and to gain a better understanding of factors that exert the largest impact on these capabilities. To model protein complexes in CASP13, we applied template-based modeling, free docking and hybrid techniques that enabled us to generate models of the topmost quality for 27 of 42 multimers. If templates for protein complexes could be identified, we modeled the structures with reasonable accuracy by straightforward homology modeling. If only partial templates were available, it was nevertheless possible to predict the interaction interfaces correctly or to generate acceptable models for protein complexes by combining template-based modeling with docking. If no templates were available, we used rigid-body docking with limited success. However, in some free docking models, despite the incorrect subunit orientation and missed interface contacts, the approximate location of protein binding sites was identified correctly. Apparently, our overall performance in docking was limited by the quality of monomer models and by the imperfection of scoring methods. The impact of human intervention on our results in modeling of protein complexes was significant indicating the need for improvements of automatic methods.     

I-TASSER server for protein 3D structure prediction

Background  

Prediction of 3-dimensional protein structures from amino acid sequences represents one of the most important problems in computational structural biology. The community-wide Critical Assessment of Structure Prediction (CASP) experiments have been designed to obtain an objective assessment of the state-of-the-art of the field, where I-TASSER was ranked as the best method in the server section of the recent 7th CASP experiment. Our laboratory has since then received numerous requests about the public availability of the I-TASSER algorithm and the usage of the I-TASSER predictions.     

Assessing predictions of protein-protein interaction: the CAPRI experiment

The Critical Assessment of PRedicted Interactions (CAPRI) experiment was designed in 2000 to test protein docking algorithms in blind predictions of the structure of protein-protein complexes. In four years, 17 complexes offered by crystallographers as targets prior to publication, have been subjected to structure prediction by docking their two components. Models of these complexes were submitted by predictor groups and assessed by comparing their geometry to the X-ray structure and by evaluating the quality of the prediction of the regions of interaction and of the pair wise residue contacts. Prediction was successful on 12 of the 17 targets, most of the failures being due to large conformation changes that the algorithms could not cope with. Progress in the prediction quality observed in four years indicates that the experiment is a powerful incentive to develop new procedures that allow for flexibility during docking and incorporate nonstructural information. We therefore call upon structural biologists who study protein-protein complexes to provide targets for further rounds of CAPRI predictions.     

Electrostatic Interactions in the Binding Pathway of a Transient Protein Complex Studied by NMR and Isothermal Titration Calorimetry

Much of our knowledge of protein binding pathways is derived from extremely stable complexes that interact very tightly, with lifetimes of hours to days. Much less is known about weaker interactions and transient complexes because these are challenging to characterize experimentally. Nevertheless, these types of interactions are ubiquitous in living systems. The combination of NMR relaxation dispersion Carr–Purcell–Meiboom–Gill (CPMG) experiments and isothermal titration calorimetry allows the quantification of rapid binding kinetics for complexes with submillisecond lifetimes that are difficult to study using conventional techniques. We have used this approach to investigate the binding pathway of the Src homology 3 (SH3) domain from the Fyn tyrosine kinase, which forms complexes with peptide targets whose lifetimes are on the order of about a millisecond. Long range electrostatic interactions have been shown to play a critical role in the binding pathways of tightly binding complexes. The role of electrostatics in the binding pathways of transient complexes is less well understood. Similarly to previously studied tight complexes, we find that SH3 domain association rates are enhanced by long range electrostatics, whereas short range interactions are formed late in the docking process. However, the extent of electrostatic association rate enhancement is several orders of magnitudes less, whereas the electrostatic-free basal association rate is significantly greater. Thus, the SH3 domain is far less reliant on electrostatic enhancement to achieve rapid association kinetics than are previously studied systems. This suggests that there may be overall differences in the role played by electrostatics in the binding pathways of extremely stable versus transient complexes.     

Docking proteins and peptides under evolutionary constraints in Critical Assessment of PRediction of Interactions rounds 38 to 45

Computational structural prediction of macromolecular interactions is a fundamental tool toward the global understanding of cellular processes. The Critical Assessment of PRediction of Interactions (CAPRI) community-wide experiment provides excellent opportunities for blind testing computational docking methods and includes original targets, thus widening the range of docking applications. Our participation in CAPRI rounds 38 to 45 enabled us to expand the way we include evolutionary information in structural predictions beyond our standard free docking InterEvDock pipeline. InterEvDock integrates a coarse-grained potential that accounts for interface coevolution based on joint multiple sequence alignments of two protein partners (co-alignments). However, even though such co-alignments could be built for none of the CAPRI targets in rounds 38 to 45, including host-pathogen and protein-oligosaccharide complexes and a redesigned interface, we identified multiple strategies that can be used to incorporate evolutionary constraints, which helped us to identify the most likely macromolecular binding modes. These strategies include template-based modeling where only local adjustments should be applied when query-template sequence identity is above 30% and larger perturbations are needed below this threshold; covariation-based structure prediction for individual protein partners; and the identification of evolutionarily conserved and structurally recurrent anchoring interface motifs. Overall, we submitted correct predictions among the top 5 models for 12 out of 19 interface challenges, including four High- and five Medium-quality predictions. Our top 20 models included correct predictions for three out of the five targets we missed in the top 5, including two targets for which misleading biological data led us to downgrade correct free docking models.     

Scoring a diverse set of high-quality docked conformations: a metascore based on electrostatic and desolvation interactions

Predicting protein-protein interactions involves sampling and scoring docked conformations. Barring some large structural rearrangement, rapidly sampling the space of docked conformations is now a real possibility, and the limiting step for the successful prediction of protein interactions is the scoring function used to reduce the space of conformations from billions to a few, and eventually one high affinity complex. An atomic level free-energy scoring function that estimates in units of kcal/mol both electrostatic and desolvation interactions (plus van der Waals if appropriate) of protein-protein docked conformations is used to rerank the blind predictions (860 in total) submitted for six targets to the community-wide Critical Assessment of PRediction of Interactions (CAPRI; http://capri.ebi.ac.uk). We found that native-like models often have varying intermolecular contacts and atom clashes, making unlikely that one can construct a universal function that would rank all these models as native-like. Nevertheless, our scoring function is able to consistently identify the native-like complexes as those with the lowest free energy for the individual models of 16 (out of 17) human predictors for five of the targets, while at the same time the modelers failed to do so in more than half of the cases. The scoring of high-quality models developed by a wide variety of methods and force fields confirms that electrostatic and desolvation forces are the dominant interactions determining the bound structure. The CAPRI experiment has shown that modelers can predict valuable models of protein-protein complexes, and improvements in scoring functions should soon solve the docking problem for complexes whose backbones do not change much upon binding. A scoring server and programs are available at http://structure.pitt.edu.     

ClusPro: an automated docking and discrimination method for the prediction of protein complexes

MOTIVATION: Predicting protein interactions is one of the most challenging problems in functional genomics. Given two proteins known to interact, current docking methods evaluate billions of docked conformations by simple scoring functions, and in addition to near-native structures yield many false positives, i.e. structures with good surface complementarity but far from the native. RESULTS: We have developed a fast algorithm for filtering docked conformations with good surface complementarity, and ranking them based on their clustering properties. The free energy filters select complexes with lowest desolvation and electrostatic energies. Clustering is then used to smooth the local minima and to select the ones with the broadest energy wells-a property associated with the free energy at the binding site. The robustness of the method was tested on sets of 2000 docked conformations generated for 48 pairs of interacting proteins. In 31 of these cases, the top 10 predictions include at least one near-native complex, with an average RMSD of 5 A from the native structure. The docking and discrimination method also provides good results for a number of complexes that were used as targets in the Critical Assessment of PRedictions of Interactions experiment. AVAILABILITY: The fully automated docking and discrimination server ClusPro can be found at http://structure.bu.edu     

ATP binding enables broad antibiotic selectivity of aminoglycoside phosphotransferase(3')-IIIa: an elastic network analysis

The bacterial enzyme aminoglycoside phosphotransferase(3′)-IIIa (APH) confers resistance against a wide range of aminoglycoside antibiotics. In this study, we use the Gaussian network model to investigate how the binding of nucleotides and antibiotics influences the dynamics and thereby the ligand binding properties of APH. Interestingly, in NMR experiments, the dynamics differ significantly in various APH complexes, although crystallographic studies indicate that no larger conformational changes occur upon ligand binding. Isothermal titration calorimetry also shows different thermodynamic contributions to ligand binding. Formation of aminoglycoside-APH complexes is enthalpically driven, while the enthalpic change upon aminoglycoside binding to the nucleotide-APH complex is much smaller. The differential effects of nucleotide binding and antibiotic binding to APH can be explained theoretically by single-residue fluctuations and correlated motions of the enzyme. The surprising destabilization of β-sheet residues upon nucleotide binding, as seen in hydrogen/deuterium exchange experiments, shows that the number of closest neighbors does not fully explain residue flexibility. Additionally, we must consider correlated motions of dynamic protein domains, which show that not only connectivity but also the overall protein architecture is important for protein dynamics.     

Energy-based graph convolutional networks for scoring protein docking models

Structural information about protein-protein interactions, often missing at the interactome scale, is important for mechanistic understanding of cells and rational discovery of therapeutics. Protein docking provides a computational alternative for such information. However, ranking near-native docked models high among a large number of candidates, often known as the scoring problem, remains a critical challenge. Moreover, estimating model quality, also known as the quality assessment problem, is rarely addressed in protein docking. In this study, the two challenging problems in protein docking are regarded as relative and absolute scoring, respectively, and addressed in one physics-inspired deep learning framework. We represent protein and complex structures as intra- and inter-molecular residue contact graphs with atom-resolution node and edge features. And we propose a novel graph convolutional kernel that aggregates interacting nodes’ features through edges so that generalized interaction energies can be learned directly from 3D data. The resulting energy-based graph convolutional networks (EGCN) with multihead attention are trained to predict intra- and inter-molecular energies, binding affinities, and quality measures (interface RMSD) for encounter complexes. Compared to a state-of-the-art scoring function for model ranking, EGCN significantly improves ranking for a critical assessment of predicted interactions (CAPRI) test set involving homology docking; and is comparable or slightly better for Score_set, a CAPRI benchmark set generated by diverse community-wide docking protocols not known to training data. For Score_set quality assessment, EGCN shows about 27% improvement to our previous efforts. Directly learning from 3D structure data in graph representation, EGCN represents the first successful development of graph convolutional networks for protein docking.     

Solubilization and delivery by GroEL of megadalton complexes of the lambda holin

GroEL can solubilize membrane proteins by binding them in its hydrophobic cavity when detergent is removed by dialysis. The best-studied example is bacteriorhodopsin, which can bind in the GroEL chaperonin at two molecules per tetradecamer. Applying this approach to the holin and antiholin proteins of phage lambda, we find that both proteins are solubilized by GroEL, in an ATP-sensitive mode, but to vastly different extents. The antiholin product, S107, saturates the chaperonin at six molecules per tetradecameric complex, whereas the holin, S105, which is missing the two N-terminal residues of S107, forms a hyper-solubilization complex with up to 350 holin molecules per GroEL, or approximately 4 MDa of protein per 0.8 MDa tetradecamer. Gel filtration chromatography and immunoprecipitation experiments confirmed the existence of complexes of the predicted masses for both S105 and S107 solubilization. For S105, negatively stained electron microscopic images show structures consistent with protein shells of the holin assembled around the chaperonin tetradecamer. Importantly, S105 can be delivered rapidly and efficiently to artificial liposomes from these complexes. In these delivery experiments, the holin exhibits efficient membrane-permeabilizing activity. The S107 antiholin can block formation of the hypersolubilization complexes, suggesting that their formation is related to an oligomerization step intrinsic to holin function.     

i3Drefine Software for Protein 3D Structure Refinement and Its Assessment in CASP10

Protein structure refinement refers to the process of improving the qualities of protein structures during structure modeling processes to bring them closer to their native states. Structure refinement has been drawing increasing attention in the community-wide Critical Assessment of techniques for Protein Structure prediction (CASP) experiments since its addition in 8^th CASP experiment. During the 9^th and recently concluded 10^th CASP experiments, a consistent growth in number of refinement targets and participating groups has been witnessed. Yet, protein structure refinement still remains a largely unsolved problem with majority of participating groups in CASP refinement category failed to consistently improve the quality of structures issued for refinement. In order to alleviate this need, we developed a completely automated and computationally efficient protein 3D structure refinement method, i3Drefine, based on an iterative and highly convergent energy minimization algorithm with a powerful all-atom composite physics and knowledge-based force fields and hydrogen bonding (HB) network optimization technique. In the recent community-wide blind experiment, CASP10, i3Drefine (as ‘MULTICOM-CONSTRUCT’) was ranked as the best method in the server section as per the official assessment of CASP10 experiment. Here we provide the community with free access to i3Drefine software and systematically analyse the performance of i3Drefine in strict blind mode on the refinement targets issued in CASP10 refinement category and compare with other state-of-the-art refinement methods participating in CASP10. Our analysis demonstrates that i3Drefine is only fully-automated server participating in CASP10 exhibiting consistent improvement over the initial structures in both global and local structural quality metrics. Executable version of i3Drefine is freely available at http://protein.rnet.missouri.edu/i3drefine/.     

A genome-wide Ras-effector interaction network

Here using structural information and protein design tools we have drawn the network of interactions between 20 Ras subfamily proteins with 50 putative Ras binding domains. To validate this network we have cloned six poorly characterized Ras binding domains (RBD) and two Ras proteins (RERG, DiRas1). These, together with previously described RBD domains, Ras and Rap proteins have been analyzed in 70 pull-down experiments. Comparing our interaction network with these and previous pull-down experiments (total of 150 cases) shows a very high accuracy for distinguishing between binders and non-binders ( approximately 0.80). Bioinformatics information was integrated to distinguish those in vitro interactions that are more likely to be relevant in vivo. We proposed several new interactions between Ras family members and effector domains that are of relevance in understanding the physiological role of these proteins. More broadly our results demonstrate that (domain-domain) interaction specificities between members of protein families can be accurately predicted using structural information.     

Identifying specific protein interaction partners using quantitative mass spectrometry and bead proteomes

The identification of interaction partners in protein complexes is a major goal in cell biology. Here we present a reliable affinity purification strategy to identify specific interactors that combines quantitative SILAC-based mass spectrometry with characterization of common contaminants binding to affinity matrices (bead proteomes). This strategy can be applied to affinity purification of either tagged fusion protein complexes or endogenous protein complexes, illustrated here using the well-characterized SMN complex as a model. GFP is used as the tag of choice because it shows minimal nonspecific binding to mammalian cell proteins, can be quantitatively depleted from cell extracts, and allows the integration of biochemical protein interaction data with in vivo measurements using fluorescence microscopy. Proteins binding nonspecifically to the most commonly used affinity matrices were determined using quantitative mass spectrometry, revealing important differences that affect experimental design. These data provide a specificity filter to distinguish specific protein binding partners in both quantitative and nonquantitative pull-down and immunoprecipitation experiments.     

Rigidification of a flexible protease inhibitor variant upon binding to trypsin

The Tyr35-->Gly replacement in bovine pancreatic trypsin inhibitor (BPTI) has previously been shown to dramatically enhance the flexibility of the trypsin-binding region of the free inhibitor and to destabilize the interaction with the protease by about 3 kcal/mol. The effects of this replacement on the enzyme-inhibitor interaction were further studied here by X-ray crystallography and isothermal titration calorimetry (ITC). The co-crystal structure of Y35G BPTI bound to trypsin was determined using 1.65 A resolution X-ray diffraction data collected from cryopreserved crystals, and a new structure of the complex with wild-type BPTI under the same conditions was determined using 1.62 A data. These structures reveal that, in contrast to the free protein, Y35G BPTI adopts a conformation nearly identical with that of the wild-type protein, with a water-filled cavity in place of the missing Tyr side-chain. The crystallographic temperature factors for the two complexes indicate that the mutant inhibitor is nearly as rigid as the wild-type protein when bound to trypsin. Calorimetric measurements show that the change in enthalpy upon dissociation of the complex is 2.5 kcal/mol less favorable for the complex containing Y35G BPTI than for the complex with the wild-type inhibitor. Thus, the destabilization of the complex resulting from the Y35G replacement is due to a more favorable change in entropy upon dissociation. The heat capacity changes for dissociation of the mutant and wild-type complexes were very similar, suggesting that the entropic effects probably do not arise from solvation effects, but are more likely due to an increase in protein conformational entropy upon dissociation of the mutant inhibitor. These results define the biophysical role of a highly conserved core residue located outside of a protein-binding interface, demonstrating that Tyr35 has little impact on the trypsin-bound BPTI structure and acts primarily to define the structure of the free protein so as to maximize binding affinity.     

Template-based modeling of diverse protein interactions in CAPRI rounds 38-45

Structures of proteins complexed with other proteins, peptides, or ligands are essential for investigation of molecular mechanisms. However, the experimental structures of protein complexes of interest are often not available. Therefore, computational methods are widely used to predict these structures, and, of those methods, template-based modeling is the most successful. In the rounds 38-45 of the Critical Assessment of PRediction of Interactions (CAPRI), we applied template-based modeling for 9 of 11 protein-protein and protein-peptide interaction targets, resulting in medium and high-quality models for six targets. For the protein-oligosaccharide docking targets, we used constraints derived from template structures, and generated models of at least acceptable quality for most of the targets. Apparently, high flexibility of oligosaccharide molecules was the main cause preventing us from obtaining models of higher quality. We also participated in the CAPRI scoring challenge, the goal of which was to identify the highest quality models from a large pool of decoys. In this experiment, we tested VoroMQA, a scoring method based on interatomic contact areas. The results showed VoroMQA to be quite effective in scoring strongly binding and obligatory protein complexes, but less successful in the case of transient interactions. We extensively used manual intervention in both CAPRI modeling and scoring experiments. This oftentimes allowed us to select the correct templates from available alternatives and to limit the search space during the model scoring.     

Discriminating physiological from non-physiological interfaces in structures of protein complexes: A community-wide study

Reliably scoring and ranking candidate models of protein complexes and assigning their oligomeric state from the structure of the crystal lattice represent outstanding challenges. A community-wide effort was launched to tackle these challenges. The latest resources on protein complexes and interfaces were exploited to derive a benchmark dataset consisting of 1677 homodimer protein crystal structures, including a balanced mix of physiological and non-physiological complexes. The non-physiological complexes in the benchmark were selected to bury a similar or larger interface area than their physiological counterparts, making it more difficult for scoring functions to differentiate between them. Next, 252 functions for scoring protein-protein interfaces previously developed by 13 groups were collected and evaluated for their ability to discriminate between physiological and non-physiological complexes. A simple consensus score generated using the best performing score of each of the 13 groups, and a cross-validated Random Forest (RF) classifier were created. Both approaches showed excellent performance, with an area under the Receiver Operating Characteristic (ROC) curve of 0.93 and 0.94, respectively, outperforming individual scores developed by different groups. Additionally, AlphaFold2 engines recalled the physiological dimers with significantly higher accuracy than the non-physiological set, lending support to the reliability of our benchmark dataset annotations. Optimizing the combined power of interface scoring functions and evaluating it on challenging benchmark datasets appears to be a promising strategy.     

Docking and scoring protein complexes: CAPRI 3rd Edition

The performance of methods for predicting protein-protein interactions at the atomic scale is assessed by evaluating blind predictions performed during 2005-2007 as part of Rounds 6-12 of the community-wide experiment on Critical Assessment of PRedicted Interactions (CAPRI). These Rounds also included a new scoring experiment, where a larger set of models contributed by the predictors was made available to groups developing scoring functions. These groups scored the uploaded set and submitted their own best models for assessment. The structures of nine protein complexes including one homodimer were used as targets. These targets represent biologically relevant interactions involved in gene expression, signal transduction, RNA, or protein processing and membrane maintenance. For all the targets except one, predictions started from the experimentally determined structures of the free (unbound) components or from models derived by homology, making it mandatory for docking methods to model the conformational changes that often accompany association. In total, 63 groups and eight automatic servers, a substantial increase from previous years, submitted docking predictions, of which 1994 were evaluated here. Fifteen groups submitted 305 models for five targets in the scoring experiment. Assessment of the predictions reveals that 31 different groups produced models of acceptable and medium accuracy-but only one high accuracy submission-for all the targets, except the homodimer. In the latter, none of the docking procedures reproduced the large conformational adjustment required for correct assembly, underscoring yet again that handling protein flexibility remains a major challenge. In the scoring experiment, a large fraction of the groups attained the set goal of singling out the correct association modes from incorrect solutions in the limited ensembles of contributed models. But in general they seemed unable to identify the best models, indicating that current scoring methods are probably not sensitive enough. With the increased focus on protein assemblies, in particular by structural genomics efforts, the growing community of CAPRI predictors is engaged more actively than ever in the development of better scoring functions and means of modeling conformational flexibility, which hold promise for much progress in the future.     

The interaction of PTP-BL PDZ domains with RIL: An enigmatic role for the RIL LIM domain

PDZ domains are protein-protein interaction modules that are crucial for the assembly of structural and signaling complexes. PDZ domains specifically bind short carboxyl-terminal peptides and occasionally internal sequences that structurally resemble peptide termini. Previously, using yeast two-hybrid methodology, we studied the interaction of two PDZ domains present in the large submembranous protein tyrosine phosphatase PTP-BL with the C-terminal half of the LIM domain-containing protein RIL. Deletion of the extreme RIL C-terminus did not eliminate binding, suggesting the presence of a PDZ binding site within the RIL LIM moiety. We have now performed experiments in mammalian cell lysates and found that the RIL C-terminus proper, but not the RIL LIM domain, can interact with PTP-BL, albeit very weakly. However, this interaction with PTP-BL PDZ domains is greatly enhanced when the combined RIL LIM domain and C-terminus is used, pointing to synergistic effects. NMR titration experiments and site-directed mutagenesis indicate that this result is not dependent on specific interactions that require surface exposed residues on the RIL LIM domain, suggesting a stabilizing role in the association with PTP-BL.