Hydrophobic complementarity in protein-protein docking

Low-Entropy Hydration Shells at the Spike RBD's Binding Site May Reveal the Contagiousness of SARS-CoV-2 Variants

The infectivity of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is primarily determined by the binding affinity between the receptor-binding domain (RBD) of the spike protein and the angiotensin-converting enzyme 2 (ACE2) receptor. Here, through screening off pseudo hydrophilic groups on protein surfaces, the distribution of low-entropy regions on hydration shells of the ACE2 receptor and the RBDs of multiple SARS-CoV-2 variants was demonstrated. Shape matching between the low-entropy hydration shells of multiple SARS-CoV-2 variants and the ACE2 receptor has been identified as a mechanism that drives hydrophobic attraction between the RBDs and the ACE2 receptor, which estimates the binding affinity. Low-entropy regions of the hydration shells, which play important roles in determining the binding of other viruses and their receptors, are demonstrated. The RBD-ACE2 binding is thus found to be guided by hydrophobic collapse between the shape-matched low-entropy regions of the hydration shells of the proteins. A measure of the low-entropy status of the hydration shells can be estimated by calculating genuine hydrophilic groups within the binding sites. An important indicator of the contagiousness of SARS-CoV-2 variants is the low-entropy level of its hydration shells at the spike protein binding site.

Spatial Layouts of Low-Entropy Hydration Shells Guide Protein Binding

Protein-protein binding enables orderly biological self-organization and is therefore considered a miracle of nature. Protein‒protein binding is driven by electrostatic forces, hydrogen bonding, van der Waals force, and hydrophobic interactions. Among these physical forces, only hydrophobic interactions can be considered long-range intermolecular attractions between proteins due to the electrostatic shielding of surrounding water molecules. Low-entropy hydration shells around proteins drive hydrophobic attraction among them that essentially coordinate protein‒protein binding. Here, an innovative method is developed for identifying low-entropy regions of hydration shells of proteins by screening off pseudohydrophilic groups on protein surfaces and revealing that large low-entropy regions of the hydration shells typically cover the binding sites of individual proteins. According to an analysis of determined protein complex structures, shape matching between a large low-entropy hydration shell region of a protein and that of its partner at the binding sites is revealed as a universal law. Protein‒protein binding is thus found to be mainly guided by hydrophobic collapse between the shape-matched low-entropy hydration shells that is verified by bioinformatics analyses of hundreds of structures of protein complexes, which cover four test systems. A simple algorithm is proposed to accurately predict protein binding sites.

MetaScore: A Novel Machine-Learning-Based Approach to Improve Traditional Scoring Functions for Scoring Protein-Protein Docking Conformations

Protein-protein interactions play a ubiquitous role in biological function. Knowledge of the three-dimensional (3D) structures of the complexes they form is essential for understanding the structural basis of those interactions and how they orchestrate key cellular processes. Computational docking has become an indispensable alternative to the expensive and time-consuming experimental approaches for determining the 3D structures of protein complexes. Despite recent progress, identifying near-native models from a large set of conformations sampled by docking-the so-called scoring problem-still has considerable room for improvement. We present MetaScore, a new machine-learning-based approach to improve the scoring of docked conformations. MetaScore utilizes a random forest (RF) classifier trained to distinguish near-native from non-native conformations using their protein-protein interfacial features. The features include physicochemical properties, energy terms, interaction-propensity-based features, geometric properties, interface topology features, evolutionary conservation, and also scores produced by traditional scoring functions (SFs). MetaScore scores docked conformations by simply averaging the score produced by the RF classifier with that produced by any traditional SF. We demonstrate that (i) MetaScore consistently outperforms each of the nine traditional SFs included in this work in terms of success rate and hit rate evaluated over conformations ranked among the top 10; (ii) an ensemble method, MetaScore-Ensemble, that combines 10 variants of MetaScore obtained by combining the RF score with each of the traditional SFs outperforms each of the MetaScore variants. We conclude that the performance of traditional SFs can be improved upon by using machine learning to judiciously leverage protein-protein interfacial features and by using ensemble methods to combine multiple scoring functions.

SARS-CoV-2 Variants, RBD Mutations, Binding Affinity, and Antibody Escape

Since 2020, the receptor-binding domain (RBD) of the spike protein of the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has been constantly mutating, producing most of the notable missense mutations in the context of "variants of concern", probably in response to the vaccine-driven alteration of immune profiles of the human population. The Delta variant, in particular, has become the most prevalent variant of the epidemic, and it is spreading in countries with the highest vaccination rates, causing the world to face the risk of a new wave of the contagion. Understanding the physical mechanism responsible for the mutation-induced changes in the RBD's binding affinity, its transmissibility, and its capacity to escape vaccine-induced immunity is the "urgent challenge" in the development of preventive measures, vaccines, and therapeutic antibodies against the coronavirus disease 2019 (COVID-19) pandemic. In this study, entropy-enthalpy compensation and the Gibbs free energy change were used to analyze the impact of the RBD mutations on the binding affinity of SARS-CoV-2 variants with the receptor angiotensin converting enzyme 2 (ACE2) and existing antibodies. Through the analysis, we found that the existing mutations have already covered almost all possible detrimental mutations that could result in an increase of transmissibility, and that a possible mutation in amino-acid position 498 of the RBD can potentially enhance its binding affinity. A new calculation method for the binding energies of protein-protein complexes is proposed based on the entropy-enthalpy compensation rule. All known structures of RBD-antibody complexes and the RBD-ACE2 complex comply with the entropy-enthalpy compensation rule in providing the driving force behind the spontaneous protein-protein docking. The variant-induced risk of breakthrough infections in vaccinated people is attributed to the L452R mutation's reduction of the binding affinity of many antibodies. Mutations reversing the hydrophobic or hydrophilic performance of residues in the spike RBD potentially cause breakthrough infections of coronaviruses due to the changes in geometric complementarity in the entropy-enthalpy compensations between antibodies and the virus at the binding sites.

Conformation-Specific Inhibitory Anti-MMP-7 Monoclonal Antibody Sensitizes Pancreatic Ductal Adenocarcinoma Cells to Chemotherapeutic Cell Kill

Matrix metalloproteases (MMPs) undergo post-translational modifications including pro-domain shedding. The activated forms of these enzymes are effective drug targets, but generating potent biological inhibitors against them remains challenging. We report the generation of anti-MMP-7 inhibitory monoclonal antibody (GSM-192), using an alternating immunization strategy with an active site mimicry antigen and the activated enzyme. Our protocol yielded highly selective anti-MMP-7 monoclonal antibody, which specifically inhibits MMP-7's enzyme activity with high affinity (IC₅₀ = 132 ± 10 nM). The atomic model of the MMP-7-GSM-192 Fab complex exhibited antibody binding to unique epitopes at the rim of the enzyme active site, sterically preventing entry of substrates into the catalytic cleft. In human PDAC biopsies, tissue staining with GSM-192 showed characteristic spatial distribution of activated MMP-7. Treatment with GSM-192 in vitro induced apoptosis via stabilization of cell surface Fas ligand and retarded cell migration. Co-treatment with GSM-192 and chemotherapeutics, gemcitabine and oxaliplatin elicited a synergistic effect. Our data illustrate the advantage of precisely targeting catalytic MMP-7 mediated disease specific activity.

Exploring designability of electrostatic complementarity at an antigen-antibody interface directed by mutagenesis, biophysical analysis, and molecular dynamics simulations

Antibodies protect organisms from a huge variety of foreign antigens. Antibody diversity originates from both genetic and structural levels. Antigen recognition relies on complementarity between antigen-antibody interfaces. Recent methodological advances in structural biology and the accompanying rapid increase of the number of crystal structures of proteins have enabled atomic-level manipulation of protein structures to effect alterations in function. In this study, we explored the designability of electrostatic complementarity at an antigen-antibody interface on the basis of a crystal structure of the complex. We designed several variants with altered charged residues at the interface and characterized the designed variants by surface plasmon resonance, circular dichroism, differential scanning calorimetry, and molecular dynamics simulations. Both successes and failures of the structure-based design are discussed. The variants that compensate electrostatic interactions can restore the interface complementarity, enabling the cognate antigen-antibody binding. Retrospectively, we also show that these mutational effects could be predicted by the simulations. Our study demonstrates the importance of charged residues on the physical properties of this antigen-antibody interaction and suggests that computational approaches can facilitate design of antibodies that recognize a weakly immunogenic antigen.

Protein-Protein Docking in Drug Design and Discovery

Protein-protein interactions (PPIs) are responsible for a number of key physiological processes in the living cells and underlie the pathomechanism of many diseases. Nowadays, along with the concept of so-called "hot spots" in protein-protein interactions, which are well-defined interface regions responsible for most of the binding energy, these interfaces can be targeted with modulators. In order to apply structure-based design techniques to design PPIs modulators, a three-dimensional structure of protein complex has to be available. In this context in silico approaches, in particular protein-protein docking, are a valuable complement to experimental methods for elucidating 3D structure of protein complexes. Protein-protein docking is easy to use and does not require significant computer resources and time (in contrast to molecular dynamics) and it results in 3D structure of a protein complex (in contrast to sequence-based methods of predicting binding interfaces). However, protein-protein docking cannot address all the aspects of protein dynamics, in particular the global conformational changes during protein complex formation. In spite of this fact, protein-protein docking is widely used to model complexes of water-soluble proteins and less commonly to predict structures of transmembrane protein assemblies, including dimers and oligomers of G protein-coupled receptors (GPCRs). In this chapter we review the principles of protein-protein docking, available algorithms and software and discuss the recent examples, benefits, and drawbacks of protein-protein docking application to water-soluble proteins, membrane anchoring and transmembrane proteins, including GPCRs.

Locating Large, Flexible Ligands on Proteins

Many biologically important ligands of proteins are large, flexible, and in many cases charged molecules that bind to extended regions on the protein surface. It is infeasible or expensive to locate such ligands on proteins with standard methods such as docking or molecular dynamics (MD) simulation. The alternative approach proposed here is scanning of a spatial and angular grid around the protein with smaller fragments of the large ligand. Energy values for complete grids can be computed efficiently with a well-known fast Fourier transform-accelerated algorithm and a physically meaningful interaction model. We show that the approach can readily incorporate flexibility of the protein and ligand. The energy grids (EGs) resulting from the ligand fragment scans can be transformed into probability distributions and then directly compared to probability distributions estimated from MD simulations and experimental structural data. We test the approach on a diverse set of complexes between proteins and large, flexible ligands, including a complex of sonic hedgehog protein and heparin, three heparin sulfate substrates or nonsubstrates of an epimerase, a multibranched supramolecular ligand that stabilizes a protein-peptide complex, a flexible zwitterionic ligand that binds to a surface basin of a Kringle domain, and binding of ATP to a flexible site of an ion channel. In all cases, the EG approach gives results that are in good agreement with experimental data or MD simulations.

Molecular Mechanisms and Kinetic Effects of FXYD1 and Phosphomimetic Mutants on Purified Human Na,K-ATPase

Phospholemman (FXYD1) is a single-transmembrane protein regulator of Na,K-ATPase, expressed strongly in heart, skeletal muscle, and brain and phosphorylated by protein kinases A and C at Ser-68 and Ser-63, respectively. Binding of FXYD1 reduces Na,K-ATPase activity, and phosphorylation at Ser-68 or Ser-63 relieves the inhibition. Despite the accumulated information on physiological effects, whole cell studies provide only limited information on molecular mechanisms. As a complementary approach, we utilized purified human Na,K-ATPase (α1β1 and α2β1) reconstituted with FXYD1 or mutants S63E, S68E, and S63E,S68E that mimic phosphorylation at Ser-63 and Ser-68. Compared with control α1β1, FXYD1 reduces Vmax and turnover rate and raises K0.5Na. The phosphomimetic mutants reverse these effects and reduce K0.5Na below control K0.5Na. Effects on α2β1 are similar but smaller. Experiments in proteoliposomes reconstituted with α1β1 show analogous effects of FXYD1 on K0.5Na, which are abolished by phosphomimetic mutants and also by increasing mole fractions of DOPS in the proteoliposomes. Stopped-flow experiments using the dye RH421 show that FXYD1 slows the conformational transition E2(2K)ATP → E1(3Na)ATP but does not affect 3NaE1P → E2P3Na. This regulatory effect is explained simply by molecular modeling, which indicates that a cytoplasmic helix (residues 60-70) docks between the αN and αP domains in the E2 conformation, but docking is weaker in E1 (also for phosphomimetic mutants). Taken together with previous work showing that FXYD1 also raises binding affinity for the Na(+)-selective site III, these results provide a rather comprehensive picture of the regulatory mechanism of FXYD1 that complements the physiological studies.

Systemic lupus erythematosus-associated neutrophil cytosolic factor 2 mutation affects the structure of NADPH oxidase complex

In a case-control association study with 3716 North Americans of Hispanic descent and 4867 North Americans of European descent, we show that the associations of rs17849502 (NCF2 His-389 → Gln) and rs13306575 (NCF2 Arg-395 → Trp) with systemic lupus erythematosus are independent. We have shown that His-389 → Gln disrupts the binding of NCF2 to the ZF domain of VAV1, resulting in decreased NADPH oxidase activity. With respect to Arg-395 → Trp, using protein docking and structure analyses, we provide a model for the involvement of this mutation in the structure and function of the NADPH oxidase complex. This model assigns a central role to Arg-395 in the structure and stability of the quaternary NCF2/NCF4/VAV1/RAC1 NADPH oxidase complex. Arg-395 stabilizes the C-terminal tail of NCF4 and the conformation of NCF2 loop 395-402, which in turn stabilize the evolutionarily conserved interactions of NCF2/NCF4 with the DH domain of VAV1 and RAC1 region 120-137. Our findings are consistent with the high levels of conservation of all of the residues involved in these interactions.

The autophagy protein Atg12 associates with antiapoptotic Bcl-2 family members to promote mitochondrial apoptosis

Autophagy and apoptosis constitute important determinants of cell fate and engage in a complex interplay in both physiological and pathological settings. The molecular basis of this crosstalk is poorly understood and relies, in part, on "dual-function" proteins that operate in both processes. Here, we identify the essential autophagy protein Atg12 as a positive mediator of mitochondrial apoptosis and show that Atg12 directly regulates the apoptotic pathway by binding and inactivating prosurvival Bcl-2 family members, including Bcl-2 and Mcl-1. The binding occurs independently of Atg5 or Atg3 and requires a unique BH3-like motif in Atg12, characterized by interaction studies and computational docking. In apoptotic cells, knockdown of Atg12 inhibited Bax activation and cytochrome c release, while ectopic expression of Atg12 antagonized the antiapoptotic activity of Mcl-1. The interaction between Atg12 and Bcl-2 family members may thus constitute an important point of convergence between autophagy and apoptosis in response to specific signals.

Lupus-associated causal mutation in neutrophil cytosolic factor 2 (NCF2) brings unique insights to the structure and function of NADPH oxidase

Systemic lupus erythematosus (SLE), the prototypic systemic autoimmune disease, is a debilitating multisystem autoimmune disorder characterized by chronic inflammation and extensive immune dysregulation in multiple organ systems, resulting in significant morbidity and mortality. Here, we present a multidisciplinary approach resulting in the identification of neutrophil cytosolic factor 2 (NCF2) as an important risk factor for SLE and the detailed characterization of its causal variant. We show that NCF2 is strongly associated with increased SLE risk in two independent populations: childhood-onset SLE and adult-onset SLE. The association between NCF2 and SLE can be attributed to a single nonsynonymous coding mutation in exon 12, the effect of which is the substitution of histidine-389 with glutamine (H389Q) in the PB1 domain of the NCF2 protein, with glutamine being the risk allele. Computational modeling suggests that the NCF2 H389Q mutation reduces the binding efficiency of NCF2 with the guanine nucleotide exchange factor Vav1. The model predicts that NCF2/H389 residue interacts with Vav1 residues E509, N510, E556, and G559 in the ZF domain of Vav1. Furthermore, replacing H389 with Q results in 1.5 kcal/mol weaker binding. To examine the effect of the NCF2 H389Q mutation on NADPH oxidase function, site-specific mutations at the 389 position in NCF2 were tested. Results show that an H389Q mutation causes a twofold decrease in reactive oxygen species production induced by the activation of the Vav-dependent Fcγ receptor-elicited NADPH oxidase activity. Our study completes the chain of evidence from genetic association to specific molecular function.

Antibodies targeting the catalytic zinc complex of activated matrix metalloproteinases show therapeutic potential

Endogenous tissue inhibitors of metalloproteinases (TIMPs) have key roles in regulating physiological and pathological cellular processes. Imitating the inhibitory molecular mechanisms of TIMPs while increasing selectivity has been a challenging but desired approach for antibody-based therapy. TIMPs use hybrid protein-protein interactions to form an energetic bond with the catalytic metal ion, as well as with enzyme surface residues. We used an innovative immunization strategy that exploits aspects of molecular mimicry to produce inhibitory antibodies that show TIMP-like binding mechanisms toward the activated forms of gelatinases (matrix metalloproteinases 2 and 9). Specifically, we immunized mice with a synthetic molecule that mimics the conserved structure of the metalloenzyme catalytic zinc-histidine complex residing within the enzyme active site. This immunization procedure yielded selective function-blocking monoclonal antibodies directed against the catalytic zinc-protein complex and enzyme surface conformational epitopes of endogenous gelatinases. The therapeutic potential of these antibodies has been demonstrated with relevant mouse models of inflammatory bowel disease. Here we propose a general experimental strategy for generating inhibitory antibodies that effectively target the in vivo activity of dysregulated metalloproteinases by mimicking the mechanism employed by TIMPs.

Insights into the mechanism of enhanced mobilization of hematopoietic progenitor cells and release of CXCL12 by a combination of AMD3100 and aminoglycoside-polyarginine conjugates

Mobilization of hematopoietic stem and progenitor cells (HSPCs) from the bone marrow to the peripheral blood is utilized in clinical HSPC transplantation protocols. Retention of HSPCs in the bone marrow is determined by relationships between the chemokine chemokine (C-X-C motif) ligand 12 (CXCL12) and its major receptor C-X-C chemokine receptor type 4 (CXCR4), and disruption of this retention by CXCR4 antagonists such as AMD3100 induces rapid HSPC mobilization. Here, we report that aminoglycoside-polyarginine conjugates (APACs) and N-α-acetyl-nona-D-arginine (r9) induce mobilization of white blood cells and, preferentially, immature hematopoietic progenitor cells (HPCs) in mice, similarly to AMD3100. Remarkably, administration of AMD3100 with each one of the APACs or r9 caused additional HPC mobilization. The mobilizing activity of APACs and r9 was accompanied by a significant elevation in plasma CXCL12 levels. To further understand how APACs, r9 and their combinations with AMD3100 compete with CXCL12 binding to CXCR4, as well with antibody against CXCR4 for CXCR4 binding, we have undertaken an approach combining experimental validation and docking to determine plausible binding modes for these ligands. On the basis of our biological and docking findings, and recently published NMR data, we suggest that combination of pairs of compounds such as APACs (or r9) with AMD3100 induces more efficient disruption of the CXCL12-CXCR4 interaction than AMD3100 alone, resulting in enhanced HPC mobilization.

SDOCK: a global protein-protein docking program using stepwise force-field potentials

Fast Fourier transform (FFT) method limits the forms of scoring functions in global protein-protein docking. On the other hand, force field potentials can effectively describe the energy hyper surface of biological macromolecules. In this study, we developed a new protein-protein docking program, SDOCK, that incorporates van der Waals attractive potential, geometric collision, screened electrostatic potential, and Lazaridis-Karplus desolvation energy into the scoring function in the global searching process. Stepwise potentials were generated from the corresponding continuous forms to treat the structure flexibility. After optimization of the atom solvation parameters and the weights of different potential terms based on a new docking test set that contains 142 cases with small or moderate conformational changes upon binding, SDOCK slightly outperformed the well-known FFT based global docking program ZDOCK3.0. Among the 142 cases tested, 52.8% gave at least one near-native solutions in the top 100 solutions. SDOCK was also tested on six blind testing cases in Critical Assessment of Predicted Interactions rounds 13 to 18. In all six cases, the near-native solutions could be found within the top 350 solutions. Because the SDOCK approach performs global docking based on force-field potentials, one of its advantages is that it provides global binding free energy surface profiles for further analysis. The efficiency of the program is also comparable with that of other FFT based protein-protein docking programs. SDOCK is available for noncommercial applications at http://mdl.ipc.pku.edu.cn/cgi-bin/down.cgi.

CAPRI targets T29-T42: proving ground for new docking procedures

The critical assessment of protein interactions (CAPRI) experiment provides a unique opportunity for unbiased assessment of docking procedures. The recent CAPRI targets T29-T42 entailed docking of bound, unbound, and modeled structures, presenting a wide range of prediction difficulty. We submitted accurate predictions for targets T40, T41, and T42, a good prediction for T32 and acceptable predictions for T29 and T34. The accuracy of our docking results generally matched the prediction difficulty; hence, docking of modeled proteins produced less accurate results. However, there were interesting exceptions: an accurate prediction was submitted for the dimer of modeled tetratricopeptide repeat (T42) and only an acceptable prediction for the bound/unbound case T29. The ensembles of docking models produced in the scans included an acceptable or better prediction for every target. We show here that our recently developed postscan reevaluation procedure, which tests propensity and solvation measures of the whole interface and the interface core, successfully distinguished these predictions from false docking models. For enzyme-inhibitor targets, we show that the distance of the interface from the enzyme's centroid ranked high native like docking models. Also, for one case we demonstrate that docking of an ensemble of conformers produced by normal modes analysis can improve the accuracy of the prediction.

Three-dimensional modeling of protein interactions and complexes is going 'omics

High-throughput interaction discovery initiatives have revealed the existence of hundreds of multiprotein complexes whose functions are regulated through thousands of protein-protein interactions (PPIs). However, the structural details of these interactions, often necessary to understand their function, are only available for a tiny fraction, and the experimental difficulties surrounding complex structure determination make computational modeling techniques paramount. In this manuscript, we critically review some of the most recent developments in the field of structural bioinformatics applied to the modeling of protein interactions and complexes, from large macromolecular machines to domain-domain and peptide-mediated interactions. In particular, we place a special emphasis on those methods that can be applied in a proteome-wide manner, and discuss how they will help in the ultimate objective of building 3D interactome networks.

ASPDock: protein-protein docking algorithm using atomic solvation parameters model

Background

Atomic Solvation Parameters (ASP) model has been proven to be a very successful method of calculating the binding free energy of protein complexes. This suggests that incorporating it into docking algorithms should improve the accuracy of prediction. In this paper we propose an FFT-based algorithm to calculate ASP scores of protein complexes and develop an ASP-based protein-protein docking method (ASPDock).

Results

The ASPDock is first tested on the 21 complexes whose binding free energies have been determined experimentally. The results show that the calculated ASP scores have stronger correlation (r ≈ 0.69) with the binding free energies than the pure shape complementarity scores (r ≈ 0.48). The ASPDock is further tested on a large dataset, the benchmark 3.0, which contain 124 complexes and also shows better performance than pure shape complementarity method in docking prediction. Comparisons with other state-of-the-art docking algorithms showed that ASP score indeed gives higher success rate than the pure shape complementarity score of FTDock but lower success rate than Zdock3.0. We also developed a softly restricting method to add the information of predicted binding sites into our docking algorithm. The ASP-based docking method performed well in CAPRI rounds 18 and 19.

Conclusions

ASP may be more accurate and physical than the pure shape complementarity in describing the feature of protein docking.

Exploring the molecular design of protein interaction sites with molecular dynamics simulations and free energy calculations

The significant work that has been invested toward understanding protein-protein interaction has not translated into significant advances in structure-based predictions. In particular redesigning protein surfaces to bind to unrelated receptors remains a challenge, partly due to receptor flexibility, which is often neglected in these efforts. In this work, we computationally graft the binding epitope of various small proteins obtained from the RCSB database to bind to barnase, lysozyme, and trypsin using a previously derived and validated algorithm. In an effort to probe the protein complexes in a realistic environment, all native and designer complexes were subjected to a total of nearly 400 ns of explicit-solvent molecular dynamics (MD) simulation. The MD data led to an unexpected observation: some of the designer complexes were highly unstable and decomposed during the trajectories. In contrast, the native and a number of designer complexes remained consistently stable. The unstable conformers provided us with a unique opportunity to define the structural and energetic factors that lead to unproductive protein-protein complexes. To that end we used free energy calculations following the MM-PBSA approach to determine the role of nonpolar effects, electrostatics and entropy in binding. Remarkably, we found that a majority of unstable complexes exhibited more favorable electrostatics than native or stable designer complexes, suggesting that favorable electrostatic interactions are not prerequisite for complex formation between proteins. However, nonpolar effects remained consistently more favorable in native and stable designer complexes reinforcing the importance of hydrophobic effects in protein-protein binding. While entropy systematically opposed binding in all cases, there was no observed trend in the entropy difference between native and designer complexes. A series of alanine scanning mutations of hot-spot residues at the interface of native and designer complexes showed less than optimal contacts of hot-spot residues with their surroundings in the unstable conformers, resulting in more favorable entropy for these complexes. Finally, disorder predictions revealed that secondary structures at the interface of unstable complexes exhibited greater disorder than the stable complexes.

Analysis of oligomeric proteins during unfolding by pH and temperature

During thermal transition and variation of pH, structural properties of 35 proteins and their complexes (bound with substrate and co-factor) were analyzed in detail. During pH alteration, these proteins were shown to have substantial differences in conformations. pH conformers were analyzed in detail. Free energy and other energy parameters were also estimated for these proteins at various pH and temperatures. Detailed structural analysis and binding interfaces of various substrates, inhibitors and cofactor of these proteins were also investigated using docking and molecular dynamic simulation.

Determination of ligand-binding sites on proteins using long-range hydrophobic potential

Here we developed a new program, HydrophoBicity On a Protein (HBOP), to find the ligand-binding site of a protein using the long-range hydrophobic-potential function estimated from the experimental data of Israelachvili and Pashley. We calculated the hydrophobic-potential energies at each grid point of a lattice around a protein using the potential function. The hydrophobic potential was evaluated using the carbon atoms of the hydrophobic residues, with the exception of those of the amide groups. We tested HBOP on 26 types of protein (72 protein-ligand complexes), the three-dimensional structures of which were determined experimentally. Although only one hydrophobic function was used, HBOP could successfully identify the binding sites in all of the proteins tested. Moreover, in 24 of the proteins, the binding sites were located in the most hydrophobic region. Surprisingly, the binding sites on sugar binding proteins were the most hydrophobic sites. It implies that the hydrophobic interaction plays an important role in the formation of protein-ligand complexes.

Bacterial RNase P RNA is a drug target for aminoglycoside-arginine conjugates

The ribonuclease P (RNase P) holoenzymes are RNPs composed of RNase P RNA (PRNA) and a variable number of P protein subunits. Primary differences in structure and function between bacterial and eukaryotic RNase P and its indispensability for cell viability make the bacterial enzyme an attractive drug target. On the basis of our previous studies, aminoglycoside-arginine conjugates (AACs) bind to HIV-1 TAR and Rev responsive element (RRE) RNAs significantly more efficiently than neomycin B. Their specific inhibition of bacterial rRNA as well as the findings that the hexa-arginine neomycin derivative (NeoR6) is 500-fold more potent than neomycin B in inhibiting bacterial RNase P, led us to explore the structure-function relationships of AACs in comparison to a new set of aminoglycoside-polyarginine conjugates (APACs). We here present predicted binding modes of AACs and APACs to PRNA. We used a multistep docking approach comprising rigid docking full scans and final refinement of the obtained complexes. Our docking results suggest three possible mechanisms of RNase P inhibition by AACs and APACs: competition with the P protein and pre-tRNA on binding to P1-P4 multihelix junction and to J19/4 region (probably including displacement of Mg2+ ions from the P4 helix) of PRNA; competition with Mg2+ ions near the P15 loop; and competition with the P protein and/or pre-tRNA near the P15 helix and interfering with interactions between the P protein and pre-tRNA at this region. The APACs revealed about 10-fold lower intermolecular energy than AACs, indicating stronger interactions of APACs than AACs with PRNA.

Efficient comprehensive scoring of docked protein complexes using probabilistic support vector machines

Biological systems and processes rely on a complex network of molecular interactions. While the association of biological macromolecules is a fundamental biochemical phenomenon crucial for the understanding of complex living systems, protein-protein docking methods aim for the computational prediction of protein complexes from individual subunits. Docking algorithms generally produce large numbers of putative protein complexes with only few of these conformations resembling the native complex structure within an acceptable degree of structural similarity. A major challenge in the field of docking is to extract near-native structure(s) out of the large pool of solutions, the so called scoring or ranking problem. A series of structural, chemical, biological and physical properties are used in this work to classify docked protein-protein complexes. These properties include specialized energy functions, evolutionary relationship, class specific residue interface propensities, gap volume, buried surface area, empiric pair potentials on residue and atom level as well as measures for the tightness of fit. Efficient comprehensive scoring functions have been developed using probabilistic Support Vector Machines in combination with this array of properties on the largest currently available protein-protein docking benchmark. The established classifiers are shown to be specific for certain types of protein-protein complexes and are able to detect near-native complex conformations from large sets of decoys with high sensitivity. Using classification probabilities the ranking of near-native structures was drastically improved, leading to a significant enrichment of near-native complex conformations within the top ranks. It could be shown that the developed schemes outperform five other previously published scoring functions.

Characterization of protein-protein interfaces

We analyze the characteristics of protein-protein interfaces using the largest datasets available from the Protein Data Bank (PDB). We start with a comparison of interfaces with protein cores and non-interface surfaces. The results show that interfaces differ from protein cores and non-interface surfaces in residue composition, sequence entropy, and secondary structure. Since interfaces, protein cores, and non-interface surfaces have different solvent accessibilities, it is important to investigate whether the observed differences are due to the differences in solvent accessibility or differences in functionality. We separate out the effect of solvent accessibility by comparing interfaces with a set of residues having the same solvent accessibility as the interfaces. This strategy reveals residue distribution propensities that are not observable by comparing interfaces with protein cores and non-interface surfaces. Our conclusions are that there are larger numbers of hydrophobic residues, particularly aromatic residues, in interfaces, and the interactions apparently favored in interfaces include the opposite charge pairs and hydrophobic pairs. Surprisingly, Pro-Trp pairs are over represented in interfaces, presumably because of favorable geometries. The analysis is repeated using three datasets having different constraints on sequence similarity and structure quality. Consistent results are obtained across these datasets. We have also investigated separately the characteristics of heteromeric interfaces and homomeric interfaces.

Study of recombinant antibody fragments and PAI-1 complexes combining protein-protein docking and results from site-directed mutagenesis

Elevated plasma levels of plasminogen activator inhibitor-1 (PAI-1) have been correlated with cardiovascular diseases such as myocardial infarction and venous thrombosis. PAI-1 has also been shown to play an important role in tumor development, diabetes, and obesitas. Monoclonal antibodies MA-8H9D4 and MA-56A7C10, and their single-chain variable fragments (scFv), exhibit PAI-1-neutralizing properties. In this study, a rigid-body docking approach is used to predict the binding geometry of two distinct conformations of PAI-1 (active and latent) in complex with these antibody fragments. Resulting models were initially refined by using the dead-end elimination algorithm. Different filtering criteria based on the mutagenesis studies and structural considerations were applied to select the final models. These were refined by using the slow-cooling torsion-angle dynamic annealing protocol. The docked structures reveal the respective epitopes and paratopes and their potential interactions. This study provides crucial information that is necessary for the rational development of low-molecular weight PAI-1 inhibitors.

Molecular modeling of the binding of pheromone biosynthesis activating neuropeptide to its receptor

Moth sex-pheromone biosynthesis follows a circadian cycle, which is cued by the release of the neurohormone pheromone biosynthesis activating neuropeptide (PBAN) to the hemolymph. PBAN binds to a G protein-coupled receptor (GPCR), in pheromone glands, (PG) initially identified by us in Helicoverpa zea moths (HezPBAN-R). In this study, the sequences of the seven transmembrane helices of HezPBAN-R were identified, built, packed and oriented correctly after multiple sequence alignment of the HezPBAN-R and several other GPCRs using the X-ray structure of rhodopsin as a template. Molecular dynamics simulations were run on three different beta-turn types of the C-terminal hexapeptide of PBAN and the results clustered into 12 structurally distinct groups. The lowest energy conformation from each group was used for computer-simulated docking with the model of the HezPBAN-R. Highest scoring complexes were examined and putative binding sites were identified. Experimental studies, using in vitro PG, revealed lower levels of pheromonotropic activity when challenged with pyrokinin-like peptides than with HezPBAN as ligand. Thus, the Drosophila melanogaster pyrokinin-1 receptor (CG9918) was chosen to create chimera receptors by exchanging between the three extracellular loops of the HezPBAN-R and the CG9918 for in silico mutagenesis experiments. The predicted docking model was validated with experimental data obtained from expressed chimera receptors in Sf9 cells.

Prediction of physical protein-protein interactions

Many essential cellular processes such as signal transduction, transport, cellular motion and most regulatory mechanisms are mediated by protein-protein interactions. In recent years, new experimental techniques have been developed to discover the protein-protein interaction networks of several organisms. However, the accuracy and coverage of these techniques have proven to be limited, and computational approaches remain essential both to assist in the design and validation of experimental studies and for the prediction of interaction partners and detailed structures of protein complexes. Here, we provide a critical overview of existing structure-independent and structure-based computational methods. Although these techniques have significantly advanced in the past few years, we find that most of them are still in their infancy. We also provide an overview of experimental techniques for the detection of protein-protein interactions. Although the developments are promising, false positive and false negative results are common, and reliable detection is possible only by taking a consensus of different experimental approaches. The shortcomings of experimental techniques affect both the further development and the fair evaluation of computational prediction methods. For an adequate comparative evaluation of prediction and high-throughput experimental methods, an appropriately large benchmark set of biophysically characterized protein complexes would be needed, but is sorely lacking.

Prediction of HIV-1 entry inhibitors neomycin-arginine conjugates interaction with the CD4-gp120 binding site by molecular modeling and multistep docking procedure

Developing of multi-target HIV-1 entry inhibitors represents an important avenue of drug therapy. Two such inhibitors are hexa-arginine-neomycin-conjugate (NeoR6) and nona-d-arginine-neomycin-conjugate (Neo-r9). Our findings that NeoR6-resistant mutations appear in the gp120 constant regions; and NeoR6 is not CCR5 antagonist, but inhibits CXCR4 and CCR5 HIV-1 using isolates, led us to suggest that NeoR6 may inhibit HIV-1 entry by interfering with the CD4-gp120 binding. To support this notion, we constructed a homology model of unliganded HIV-1(IIIB) gp120 and docked NeoR6 and Neo-r9 to it, using a multistep docking procedure: geometric-electrostatic docking by MolFit; flexible ligand docking by Autodock3 and final refinement of the obtained complexes by Discover3. Binding free energies were calculated by MM-PBSA methodology. The model predicts competitive inhibition of CD4-gp120 binding by NeoR6 and Neo-r9. We determined plausible binding sites between constructed CD4-bound gp120 trimer and homology modeled membranal CXCR4, and tested NeoR6 and Neo-r9 interfering with this interaction. These models support our notion that another mechanism of anti-HIV-1 activity of NeoR6 is inhibition of gp120-CXCR4 binding. These structural models and interaction of NeoR6 and Neo-r9 with gp120 and CXCR4 provide a powerful approach for structural based drug design for selective targeting of HIV-1 entry and/or for inhibition of other retroviruses with similar mechanism of entry.

A study of the structural correlates of affinity maturation: Antibody affinity as a function of chemical interactions, structural plasticity and stability

Mutations introduced in an antibody germline sequence as a result of somatic hypermutation could cause its derivatives to have an altered affinity for its target. Affinity maturation favors the selection of the antibodies which exhibit increased affinity. The mutations in 80 high affinity anti-thyroid peroxidase sequences derived from six germlines were analysed in terms of the physicochemical properties of the replacement residues, namely hydrophilicity, size and polarizability, and charge and polarity, in the context of its position and probable solvent accessibility. The effects of these substitutions were evaluated in terms of the resultant increased chemical interactivity potential of the affinity-matured antibodies relative to the germline. The results of the analysis would be useful in the rational design of antibodies and of other proteins for improved binding properties.

A protein-specifically adapted scoring function for the reranking of docking solutions

In this work, we developed a protein-specifically adapted scoring function and applied it to the reranking of protein-protein docking solutions generated with a conventional docking program. The approach was validated using experimentally determined structures of the bacterial HPr-protein in complex with four structurally nonhomologous binding partners as an example. A sufficiently large data basis for the generation of protein-specifically adapted pair potentials was generated by modeling all orthologous complexes for each type of interaction resulting in a total of 224 complexes. The parameters for potential generation were systematically varied and resulted in a total of 66,132 different scoring functions that were tested for their ability of successful reranking of 1000 docking solutions generated from modeled structures of the unbound binding partners. Parameters that proved critical for the generation of good scoring functions were the distance cutoff used for the generation of the pair potential, and an additional cutoff that allows a proper weighting of conserved and nonconserved contacts in the interface. Compared to the original scoring function, application of this novel type of scoring functions resulted in a significant accumulation of acceptable docking solutions within the first 10 ranks. Depending on the type of complex investigated one to five acceptable complex geometries are found among the 10 highest-ranked solutions and for three of the four systems tested, an acceptable solution was placed on the first rank.

Inherent limitations in protein-protein docking procedures

MOTIVATION

The limited success rate of protein-protein docking procedures is generally attributed to structure differences between the bound and unbound states of the molecules. Herein we analyze a large dataset of protein-protein docking results and identify additional parameters that affect the performance of docking procedures.

RESULTS

We find that the distinction between nearly correct models (NCMs) and decoys depends on the size of the interface to be predicted thus setting a limit to the prediction ability of docking procedures, particularly those in which the geometric complementarity descriptor is dominant. The geometric complementarity score in grid-based docking carries a large statistical error which further reduces the distinction between NCMs and decoys. We propose a method for correcting the statistical error and show that the distinction is improved when the docking models are ranked by statistically equivalent scores.

AVAILABILITY

MolFit can be downloaded from our website http://www.weizmann.ac.il/Chemical_Research_Support/molfit.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Structural model of the mAb 806-EGFR complex using computational docking followed by computational and experimental mutagenesis

In this work, we combined computational protein-protein docking with computational and experimental mutagenesis to predict the structure of the complex formed by monoclonal antibody 806 (mAb 806) and the epidermal growth factor receptor (EGFR). We docked mAb 806, an antitumor antibody, to its epitope of EGFR residues 287-302. Potential mAb 806-EGFR orientations were generated, and computational mutagenesis was used to filter them according to their agreement with experimental mutagenesis data. Further computational mutagenesis suggested additional mutations, which were tested to arrive at a final structure that was most consistent with experimental mutagenesis data. We propose that this is the EGFR-mAb 806 structure, in which mAb 806 binds to an untethered form of the receptor, consistent with published experimental results. The steric hindrance created by the antibody near the EGFR dimer interface interferes with receptor dimerization, and we postulate this as the structural origin for the antitumor effect of mAb 806.

Assessment of CAPRI predictions in rounds 3-5 shows progress in docking procedures

The current status of docking procedures for predicting protein-protein interactions starting from their three-dimensional (3D) structure is reassessed by evaluating blind predictions, performed during 2003-2004 as part of Rounds 3-5 of the community-wide experiment on Critical Assessment of PRedicted Interactions (CAPRI). Ten newly determined structures of protein-protein complexes were used as targets for these rounds. They comprised 2 enzyme-inhibitor complexes, 2 antigen-antibody complexes, 2 complexes involved in cellular signaling, 2 homo-oligomers, and a complex between 2 components of the bacterial cellulosome. For most targets, the predictors were given the experimental structures of 1 unbound and 1 bound component, with the latter in a random orientation. For some, the structure of the free component was derived from that of a related protein, requiring the use of homology modeling. In some of the targets, significant differences in conformation were displayed between the bound and unbound components, representing a major challenge for the docking procedures. For 1 target, predictions could not go to completion. In total, 1866 predictions submitted by 30 groups were evaluated. Over one-third of these groups applied completely novel docking algorithms and scoring functions, with several of them specifically addressing the challenge of dealing with side-chain and backbone flexibility. The quality of the predicted interactions was evaluated by comparison to the experimental structures of the targets, made available for the evaluation, using the well-agreed-upon criteria used previously. Twenty-four groups, which for the first time included an automatic Web server, produced predictions ranking from acceptable to highly accurate for all targets, including those where the structures of the bound and unbound forms differed substantially. These results and a brief survey of the methods used by participants of CAPRI Rounds 3-5 suggest that genuine progress in the performance of docking methods is being achieved, with CAPRI acting as the catalyst.

High-resolution protein-protein docking

The high-resolution prediction of protein-protein docking can now create structures with atomic-level accuracy. This progress arises from both improvements in the rapid sampling of conformations and increased accuracy of binding free energy calculations. Consequently, the quality of models submitted to the blind prediction challenge CAPRI (Critical Assessment of PRedicted Interactions) has steadily increased, including complexes predicted from homology structures of one binding partner and complexes with atomic accuracy at the interface. By exploiting experimental information, docking has created model structures for real applications, even when confronted with challenges such as moving backbones and uncertain monomer structures. Work remains to be done in docking large or flexible proteins, ranking models consistently, and producing models accurate enough to allow computational design of higher affinities or specificities.

Looking at enzymes from the inside out: the proximity of catalytic residues to the molecular centroid can be used for detection of active sites and enzyme-ligand interfaces

Analysis of the distances of the exposed residues in 175 enzymes from the centroids of the molecules indicates that catalytic residues are very often found among the 5% of residues closest to the enzyme centroid. This property of catalytic residues is implemented in a new prediction algorithm (named EnSite) for locating the active sites of enzymes and in a new scheme for re-ranking enzyme-ligand docking solutions. EnSite examines only 5% of the molecular surface (represented by surface dots) that is closest to the centroid, identifying continuous surface segments and ranking them by their area size. EnSite ranks the correct prediction 1-4 in 97% of the cases in a dataset of 65 monomeric enzymes (rank 1 for 89% of the cases) and in 86% of the cases in a dataset of 176 monomeric and multimeric enzymes from all six top-level enzyme classifications (rank 1 in 74% of the cases). Importantly, identification of buried or flat active sites is straightforward because EnSite "looks" at the molecular surface from the inside out. Detailed examination of the results indicates that the proximity of the catalytic residues to the centroid is a property of the functional unit, defined as the assembly of domains or chains that form the active site (in most cases the functional unit corresponds to a single whole polypeptide chain). Using the functional unit in the prediction further improves the results. The new property of active sites is also used for re-evaluating enzyme-inhibitor unbound docking results. Sorting the docking solutions by the distance of the interface to the centroid of the enzyme improves remarkably the ranks of nearly correct solutions compared to ranks based on geometric-electrostatic-hydrophobic complementarity scores.

Adaptation of a fast Fourier transform-based docking algorithm for protein design

Designing proteins with novel protein/protein binding properties can be achieved by combining the tools that have been developed independently for protein docking and protein design. We describe here the sequence-independent generation of protein dimer orientations by protein docking for use as scaffolds in protein sequence design algorithms. To dock monomers into sequence-independent dimer conformations, we use a reduced representation in which the side chains are approximated by spheres with atomic radii derived from known C2 symmetry-related homodimers. The interfaces of C2-related homodimers are usually more hydrophobic and protein core-like than the interfaces of heterodimers; we parameterize the radii for docking against this feature to capture and recreate the spatial characteristics of a hydrophobic interface. A fast Fourier transform-based geometric recognition algorithm is used for docking the reduced representation protein models. The resulting docking algorithm successfully predicted the wild-type homodimer orientations in 65 out of 121 dimer test cases. The success rate increases to approximately 70% for the subset of molecules with large surface area burial in the interface relative to their chain length. Forty-five of the predictions exhibited less than 1 A C(alpha) RMSD compared to the native X-ray structures. The reduced protein representation therefore appears to be a reasonable approximation and can be used to position protein backbones in plausible orientations for homodimer design.