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

An anion exchange membrane sensor detects EGFR and its activity state in plasma CD63 extracellular vesicles from patients with glioblastoma

We present a quantitative sandwich immunoassay for CD63 Extracellular Vesicles (EVs) and a constituent surface cargo, EGFR and its activity state, that provides a sensitive, selective, fluorophore-free and rapid alternative to current EV-based diagnostic methods. Our sensing design utilizes a charge-gating strategy, with a hydrophilic anion exchange membrane functionalized with capture antibodies and a charged silica nanoparticle reporter functionalized with detection antibodies. With sensitivity and robustness enhancement by the ion-depletion action of the membrane, this hydrophilic design with charged reporters minimizes interference from dispersed proteins, thus enabling direct plasma analysis without the need for EV isolation or sensor blocking. With a LOD of 30 EVs/μL and a high relative sensitivity of 0.01% for targeted proteomic subfractions, our assay enables accurate quantification of the EV marker, CD63, with colocalized EGFR by an operator/sample insensitive universal normalized calibration. We analysed untreated clinical samples of Glioblastoma to demonstrate this new platform. Notably, we target both total and "active" EGFR on EVs; with a monoclonal antibody mAb806 that recognizes a normally hidden epitope on overexpressed or mutant variant III EGFR. Analysis of samples yielded an area-under-the-curve (AUC) value of 0.99 and a low p-value of 0.000033, surpassing the performance of existing assays and markers.

Detection of EGFR and its Activity State in Plasma CD63-EVs from Glioblastoma Patients: Rapid Profiling using an Anion Exchange Membrane Sensor

We present a novel quantitative immunoassay for CD63 EVs (extracellular vesicles) and a constituent surface cargo, EGFR and its activity state, that provides a sensitive, selective, fluorophore-free and rapid alternative to current EV-based diagnostic methods. Our sensing design utilizes a charge-gating strategy, with a hydrophilic anion exchange membrane and a charged silica nanoparticle reporter. With sensitivity and robustness enhancement by the ion-depletion action of the membrane, this hydrophilic design with charged reporters minimizes interference from dispersed proteins and fluorophore degradation, thus enabling direct plasma analysis. With a limit of detection of 30 EVs/μL and a high relative sensitivity of 0.01% for targeted proteomic subfractions, our assay enables accurate quantification of the EV marker, CD63, with colocalized EGFR by an operator/sample insensitive universal normalized calibration. Glioblastoma necessitates improved non-invasive diagnostic approaches for early detection and monitoring. Notably, we target both total and "active" EGFR on EVs; with a monoclonal antibody mAb806 that recognizes a normally hidden epitope on overexpressed or mutant variant III EGFR. This approach offers direct glioblastoma detection from untreated human patient samples. Analysis of glioblastoma clinical samples yielded an area-under-the-curve (AUC) value of 0.99 and low p-value of 0.000033, significantly surpassing the performance of existing assays and markers.

In silico docking analysis of beta-defensin 20 against cation channel sperm-associated protein 1-4 to predict its role in the sperm maturation

Beta-defensin 20 (DEFB20) is widely expressed in the epididymis with gene features involved in epididymal sperm maturation. However, the action mechanism and function of DEFB20 in sperm maturation are still unclear. One of the important roles of beta-defensin is the ion channel activity. The cation channel sperm-associated protein (CatSper) alpha is an ion channel protein found on the sperm surface. This study aimed to investigate the interaction between DEFB20 and CatSper1-4 protein in relation to the sperm maturation process. Protein sequences were obtained from the National Center for Biotechnology Information (NCBI). Protein modeling and validation were carried out by using the Robetta modeling server and the Ramachandran plot method. Rosetta web server was used for the docking analysis. The results revealed a natural interaction between DEFB20 and CatSper1-4. The interaction occurred at the cation channel (close to the casein kinase II), ion transport protein, and kinase c phosphorylation of the CatSper1-4 active site. The DEFB20 region interacting with CatSper2-4 was the beta-defensin domain, while with CatSper1 was the non-beta-defensin domain. Based on the analysis, DEFB20 may interact with CatSper α subunits, particularly CatsSper1, to affect ion channel activity during sperm maturation.

Computational Structure Prediction for Antibody-Antigen Complexes From Hydrogen-Deuterium Exchange Mass Spectrometry: Challenges and Outlook

Although computational structure prediction has had great successes in recent years, it regularly fails to predict the interactions of large protein complexes with residue-level accuracy, or even the correct orientation of the protein partners. The performance of computational docking can be notably enhanced by incorporating experimental data from structural biology techniques. A rapid method to probe protein-protein interactions is hydrogen-deuterium exchange mass spectrometry (HDX-MS). HDX-MS has been increasingly used for epitope-mapping of antibodies (Abs) to their respective antigens (Ags) in the past few years. In this paper, we review the current state of HDX-MS in studying protein interactions, specifically Ab-Ag interactions, and how it has been used to inform computational structure prediction calculations. Particularly, we address the limitations of HDX-MS in epitope mapping and techniques and protocols applied to overcome these barriers. Furthermore, we explore computational methods that leverage HDX-MS to aid structure prediction, including the computational simulation of HDX-MS data and the combination of HDX-MS and protein docking. We point out challenges in interpreting and incorporating HDX-MS data into Ab-Ag complex docking and highlight the opportunities they provide to build towards a more optimized hybrid method, allowing for more reliable, high throughput epitope identification.

The human antibody sequence space and structural design of the V, J regions, and CDRH3 with Rosetta

The human adaptive immune response enables the targeting of epitopes on pathogens with high specificity. Infection with a pathogen induces somatic hyper-mutation and B-cell selection processes that govern the shape and diversity of the antibody sequence landscape. To date, even the largest immunome repertoires of adaptive immune receptors acquired by next-generation sequencing cannot fully capture the vast antibody sequence space of a single individual, which is estimated to be at least 1012 potential sequences. Degeneracy of the genetic code means that the number of possible nucleotide triplets (64) is greater than the number of canonical amino acids (20), resulting in some amino acids being encoded by multiple triplets and different amino acids sharing the same nucleotide in 1 or 2 positions in the triplet. We hypothesize that the degeneracy of the genetic code can be used to statistically model an enlarged space of human antibody amino acid sequences, accommodating for the discrepancy between the observed and the hypothesized antibody sequence space. Facilitated by Bayesian statistics and immunome repertoire clustering, we calculated amino acid probabilities from single nucleotide frequencies to infer a human amino acid sequence space that is used to design human-like antibodies with Rosetta. We show that antibodies designed with our restraints are on average up to 16.6% more human-like in the V and J regions compared to the Rosetta designs produced without constraints. The human-likeness of the heavy-chain CDR3 region (CDRH3) could be increased for 8 of 27 antibodies compared to Rosetta designs with a similar number of mutations and could be successfully applied on Mus musculus antibodies to demonstrate humanization.

Receptor tyrosine kinases as druggable targets in glioblastoma: Do signaling pathways matter?

Glioblastoma (GBM) is the most malignant primary brain tumor without effective therapies. Since bevacizumab was FDA approved for targeting vascular endothelial growth factor receptor 2 (VEGFR2) in adult patients with recurrent GBM, targeted therapy against receptor tyrosine kinases (RTKs) has become a new avenue for GBM therapeutics. In addition to VEGFR, the epidermal growth factor receptor (EGFR), platelet-derived growth factor receptor (PDGFR), hepatocyte growth factor receptor (HGFR/MET), and fibroblast growth factor receptor (FGFR) are major RTK targets. However, results from clinical Phase II/III trials indicate that most RTK-targeting therapeutics including tyrosine kinase inhibitors (TKIs) and neutralizing antibodies lack clinical efficacy, either alone or in combination. The major challenge is to uncover the genetic RTK alterations driving GBM initiation and progression, as well as to elucidate the mechanisms toward therapeutic resistance. In this review, we will discuss the genetic alterations in these 5 commonly targeted RTKs, the clinical trial outcomes of the associated RTK-targeting therapeutics, and the potential mechanisms toward the resistance. We anticipate that future design of new clinical trials with combination strategies, based on the genetic alterations within an individual patient’s tumor and mechanisms contributing to therapeutic resistance after treatment, will achieve durable remissions and improve outcomes in GBM patients.

Modeling Immunity with Rosetta: Methods for Antibody and Antigen Design

Structure-based antibody and antigen design has advanced greatly in recent years, due not only to the increasing availability of experimentally determined structures but also to improved computational methods for both prediction and design. Constant improvements in performance within the Rosetta software suite for biomolecular modeling have given rise to a greater breadth of structure prediction, including docking and design application cases for antibody and antigen modeling. Here, we present an overview of current protocols for antibody and antigen modeling using Rosetta and exemplify those by detailed tutorials originally developed for a Rosetta workshop at Vanderbilt University. These tutorials cover antibody structure prediction, docking, and design and antigen design strategies, including the addition of glycans in Rosetta. We expect that these materials will allow novice users to apply Rosetta in their own projects for modeling antibodies and antigens.

Structure-function analysis of oncogenic EGFR Kinase Domain Duplication reveals insights into activation and a potential approach for therapeutic targeting

Mechanistic understanding of oncogenic variants facilitates the development and optimization of treatment strategies. We recently identified in-frame, tandem duplication of EGFR exons 18 - 25, which causes EGFR Kinase Domain Duplication (EGFR-KDD). Here, we characterize the prevalence of ERBB family KDDs across multiple human cancers and evaluate the functional biochemistry of EGFR-KDD as it relates to pathogenesis and potential therapeutic intervention. We provide computational and experimental evidence that EGFR-KDD functions by forming asymmetric EGF-independent intra-molecular and EGF-dependent inter-molecular dimers. Time-resolved fluorescence microscopy and co-immunoprecipitation reveals EGFR-KDD can form ligand-dependent inter-molecular homo- and hetero-dimers/multimers. Furthermore, we show that inhibition of EGFR-KDD activity is maximally achieved by blocking both intra- and inter-molecular dimerization. Collectively, our findings define a previously unrecognized model of EGFR dimerization, providing important insights for the understanding of EGFR activation mechanisms and informing personalized treatment of patients with tumors harboring EGFR-KDD. Finally, we establish ERBB KDDs as recurrent oncogenic events in multiple cancers.

An EGFR mutant with kinase domain duplication (EGFR-KDD) was previously identified in an index patient, but the functional and therapeutic implications remain unclear. Here, the authors show that KDD occurs in other ErbB receptors in multiple cancers, and characterize the mechanism and inhibition of EGFR-KDD.

Combining experimental techniques with molecular dynamics to investigate the impact of different enzymatic hydrolysis of β-lactoglobulin on the antigenicity reduction

β-Lactoglobulin (β-LG) is one of the major food allergens. Enzymatic hydrolysis is a promising strategy to reduce the antigenicity of β-LG in industrial production. The relationship between the cleavage sites of β-LG by protease and its antigenic active sites were explored in this study. Molecular docking and molecular dynamics (MD) were used to analyze the active sites and interaction force of β-LG and IgG antibody. Whey protein was hydrolyzed by four specific enzymes and the antigenicity of the hydrolysates were determined by ELISA. The results of MD showed that the amino acid residue Gln155 (-4.48 kcal mol^-1) played the most important roles in the process of binding. Hydrolysates produced by AY-10, which was the only one with specificity towards cleavage sites next to a Gln, had the lowest antigenicity at the same hydrolysis degree. Antigenicity decrease was related to the energy contribution of the cleavage site in the active sites.

Glycan-mediated functional assembly of IL-1RI: structural insights into completion of the current description for immune response

Interleukin 1 Receptor type I (IL-1RI) is a multi-domain transmembrane receptor that triggers the inflammatory response. Understanding its detailed mechanism of action is crucial for treating immune disorders. IL-1RI is activated upon formation of its functional assembly that occurs by binding of the IL-1 cytokine and the accessory protein (Il-1RAcP) to it. X-ray crystallography, small-Angle X-ray Scattering and molecular dynamics simulation studies showed that IL-1RI adopts two types of 'compact' and 'extended' conformational states in its dynamical pattern. Furthermore, glycosylation has shown to play a critical role in its activation process. Here, classical and accelerated atomistic molecular dynamics were carried out to examine the role of full glycosylation of IL-1RI and IL-1RAcP in arrangement of the functional assembly. Simulations showed that the 'compact' and 'extended' IL-1RI form two types of 'cytokine-inaccessible-non-signaling' and 'cytokine-accessible-signaling' assemblies with the IL-1RacP, respectively that are both abiding in the presence of glycans. Suggesting that the cytokine binding to IL-1RI is not required for the formation of IL-1RI-IL-1RAcP complex and the 'compact' complex could act as a down-regulatory mechanism. The 'extended' complex is maintained by formation of several persistent hydrogen bonds between the IL-1RI-IL-1RAcP inter-connected glycans. Taken together, it was shown that full glycosylation regulates formation of the IL-1RI functional assembly and play critical role in cytokine biding and triggering the IL-1RI involved downstream pathways in the cell.Communicated by Ramaswamy H. Sarma.

A Head Start: CAR-T Cell Therapy for Primary Malignant Brain Tumors

OPINION STATEMENT

Oncology is the midst of a therapeutic renaissance. The realization of immunotherapy as an efficacious and expanding treatment option has empowered physicians and patients alike. However, despite these remarkable advances, we have only just broached the potential immunotherapy has to offer and have yet to successfully expand these novel modalities to the field of neuro-oncology. In recent years, exciting results in preclinical studies of immune adjuvants, oncolytic viruses, or cell therapy have been met with only fleeting signs of response when taken to early phase trials. Although many have speculated why these innovative approaches result in impaired outcomes, we are left empty-handed in a field plagued by a drought of new therapies. Herein, we will review the recent advances across cellular therapy for primary malignant brain tumors, an approach that lends itself to overcoming the inherent resistance mechanisms which have impeded the success of prior treatment attempts.

Depatuxizumab Mafodotin (ABT-414)-induced Glioblastoma Cell Death Requires EGFR Overexpression, but not EGFRY1068 Phosphorylation

Glioblastomas commonly (40%) exhibit epidermal growth factor receptor (EGFR) amplification; half of these tumors carry the EGFRvIII deletion variant characterized by an in-frame deletion of exons 2-7, resulting in constitutive EGFR activation. Although EGFR tyrosine kinase inhibitors had only modest effects in glioblastoma, novel therapeutic agents targeting amplified EGFR or EGFRvIII continue to be developed.Depatuxizumab mafodotin (ABT-414) is an EGFR-targeting antibody-drug conjugate consisting of the mAb 806 and a toxic payload, monomethyl auristatin F. Because glioma cell lines and patient-derived glioma-initiating cell models expressed too little EGFR in vitro to be ABT-414-sensitive, we generated glioma sublines overexpressing EGFR or EGFRvIII to explore determinants of ABT-414-induced cell death.Overexpression of EGFRvIII induces sensitization to ABT-414 more readily than overexpression of EGFR in vitro and in vivo Exposure to ABT-414 in vivo eliminated EGFRvIII-expressing tumor cells, and recurrent tumors were devoid of EGFRvIII expression. There is no bystander killing of cells devoid of EGFR expression. Surprisingly, either exposure to EGF or to EGFR tyrosin kinase inhibitors reduce EGFR protein levels and are thus not strategies to promote ABT-414-induced cell killing. Furthermore, glioma cells overexpressing kinase-dead EGFR or EGFRvIII retain binding of mAb 806 and sensitivity to ABT-414, allowing to dissociate EGFR phosphorylation from the emergence of the "active" EGFR conformation required for ABT-414 binding.The combination of EGFR-targeting antibody-drug conjugates with EGFR tyrosine kinase inhibitors carries a high risk of failure. Promoting EGFR expression rather than phosphorylation should result in glioblastoma cell sensitization to ABT-414.

Implementation of an experimental and computational tool set to study protein-mAb interactions

This work focused on the development of a combined experimental and computational tool set to study protein-mAb interactions. A model protein library was first screened using cross interaction chromatography to identify proteins with the strongest retention. Fluorescence polarization was then employed to study the interactions and thermodynamics of the selected proteins-lactoferrin, pyruvate kinase, and ribonuclease B with the mAb. Binding affinities of lactoferrin and pyruvate kinase to the mAb were seen to be relatively salt insensitive in the range examined. Further, a strong entropic contribution was observed, suggesting the importance of hydrophobic interactions. On the other hand, ribonuclease B-mAb binding was seen to be enthalpically driven and salt sensitive, indicating the importance of electrostatic interactions. Protein-protein docking was then carried out and the results identified the CDR region on the mAb as an important binding site for all three proteins. The binding interfaces identified for the mAb-lactoferrin and mAb-pyruvate kinase systems were found to contain complementary hydrophobic and oppositely charged clusters on the interacting regions which were indicative of both hydrophobic and electrostatic interactions. On the other hand, the binding site on ribonuclease B was predominantly positively charged with minimal hydrophobicity. This resulted in an alignment with negatively charged clusters on the mAb, supporting the contention that these interactions were primarily electrostatic in nature. Importantly, these computational results were found to be consistent with the fluorescence polarization data and this combined approach may have utility in examining mAb-HCP interactions which can often complicate the downstream processing of biologics. © 2019 American Institute of Chemical Engineers.

Computational Modeling of Hepatitis C Virus Envelope Glycoprotein Structure and Recognition

Hepatitis C virus (HCV) is a major global health concern, and though therapeutic options have improved, no vaccine is available despite decades of research. As HCV can rapidly mutate to evade the immune response, an effective HCV vaccine must rely on identification and characterization of sites critical for broad immune protection and viral neutralization. This knowledge depends on structural and mechanistic insights of the E1 and E2 envelope glycoproteins, which assemble as a heterodimer on the surface of the virion, engage coreceptors during host cell entry, and are the primary targets of antibodies. Due to the challenges in determining experimental structures, structural information on E1 and E2 and their interaction is relatively limited, providing opportunities to model the structures, interactions, and dynamics of these proteins. This review highlights efforts to model the E2 glycoprotein structure, the assembly of the functional E1E2 heterodimer, the structure and binding of human coreceptors, and recognition by key neutralizing antibodies. We also discuss a comparison of recently described models of full E1E2 heterodimer structures, a simulation of the dynamics of key epitope sites, and modeling glycosylation. These modeling efforts provide useful mechanistic hypotheses for further experimental studies of HCV envelope assembly, recognition, and viral fitness, and underscore the benefit of combining experimental and computational modeling approaches to reveal new insights. Additionally, computational design approaches have produced promising candidates for epitope-based vaccine immunogens that specifically target key epitopes, providing a possible avenue to optimize HCV vaccines versus using native glycoproteins. Advancing knowledge of HCV envelope structure and immune recognition is highly applicable toward the development of an effective vaccine for HCV and can provide lessons and insights relevant to modeling and characterizing other viruses.

Modeling and docking of antibody structures with Rosetta

We describe Rosetta-based computational protocols for predicting the 3D structure of an antibody from sequence (RosettaAntibody) and then docking the antibody to protein antigens (SnugDock). Antibody modeling leverages canonical loop conformations to graft large segments from experimentally determined structures, as well as offering (i) energetic calculations to minimize loops, (ii) docking methodology to refine the V_L-V_H relative orientation and (iii) de novo prediction of the elusive complementarity determining region (CDR) H3 loop. To alleviate model uncertainty, antibody-antigen docking resamples CDR loop conformations and can use multiple models to represent an ensemble of conformations for the antibody, the antigen or both. These protocols can be run fully automated via the ROSIE web server (http://rosie.rosettacommons.org/) or manually on a computer with user control of individual steps. For best results, the protocol requires roughly 1,000 CPU-hours for antibody modeling and 250 CPU-hours for antibody-antigen docking. Tasks can be completed in under a day by using public supercomputers.

Robust Antibody-Antigen Complexes Prediction Generated by Combining Sequence Analyses, Mutagenesis, In Vitro Evolution, X-ray Crystallography and In Silico Docking

Hu 15C1 is a potent anti-human Toll-like receptor 4 (TLR4) neutralizing antibody. To better understand the molecular basis of its biological activity, we used a multidisciplinary approach to generate an accurate model of the Hu 15C1-TLR4 complex. By combining site-directed mutagenesis, in vitro antibody evolution, affinity measurements and X-ray crystallography of Fab fragments, we identified key interactions across the Hu 15C1-TLR4 interface. These contact points were used as restraints to predict the structure of the Fab region of Hu 15C1 bound to TLR4 using computational molecular docking. This model was further evaluated and validated by additional site-directed mutagenesis studies. The predicted structure of the Hu 15C1-TLR4 complex indicates that the antibody antagonizes the receptor dimerization necessary for its activation. This study exemplifies how iterative cycles of antibody engineering can facilitate the discovery of components of antibody-target interactions.

mAb806 binding to epidermal growth factor receptor: a computational study

The epidermal growth factor receptor (EGFR) is an important target in the treatment of cancer. A very potent antibody, mAb806, has been developed against overexpressed EGFR and was found to be particularly active in brain tumors. Structural studies reveal that it binds to an epitope on the extracellular region of the EGFR. However, this epitope is cryptic/buried in crystal structures of the active (untethered) and inactive (tethered) EGFR, and it is unclear as to how the antibody interacts with this region. To explore this interaction, we combined molecular docking, steered molecular dynamics, and equilibrium molecular dynamics simulations. Our computational models reveal that the antibody induces local unfolding around the epitope to form the antibody-EGFR complex. In addition, regions in the vicinity of the epitope also modulate the interaction, which are in accordance with several other known antibody-antigen interactions, and offers new possibilities for the design of antibodies with increased potency and specificity for this receptor.

Computational design of binding proteins to EGFR domain II

We developed a process to produce novel interactions between two previously unrelated proteins. This process selects protein scaffolds and designs protein interfaces that bind to a surface patch of interest on a target protein. Scaffolds with shapes complementary to the target surface patch were screened using an exhaustive computational search of the human proteome and optimized by directed evolution using phage display. This method was applied to successfully design scaffolds that bind to epidermal growth factor receptor (EGFR) domain II, the interface of EGFR dimerization, with high reactivity toward the target surface patch of EGFR domain II. One potential application of these tailor-made protein interactions is the development of therapeutic agents against specific protein targets.

Rice mitogen activated protein kinase kinase and mitogen activated protein kinase interaction network revealed by in-silico docking and yeast two-hybrid approaches

Protein-protein interaction is one of the crucial ways to decipher the functions of proteins and to understand their role in complex pathways at cellular level. Such a protein-protein interaction network in many crop plants remains poorly defined owing largely to the involvement of high costs, requirement for state of the art laboratory, time and labour intensive techniques. Here, we employed computational docking using ZDOCK and RDOCK programmes to identify interaction network between members of Oryza sativa mitogen activated protein kinase kinase (MAPKK) and mitogen activated protein kinase (MAPK). The 3-dimentional (3-D) structures of five MAPKKs and eleven MAPKs were determined by homology modelling and were further used as input for docking studies. With the help of the results obtained from ZDOCK and RDOCK programmes, top six possible interacting MAPK proteins were predicted for each MAPKK. In order to assess the reliability of the computational prediction, yeast two-hybrid (Y2H) analyses were performed using rice MAPKKs and MAPKs. A direct comparison of Y2H assay and computational prediction of protein interaction was made. With the exception of one, all the other MAPKK-MAPK pairs identified by Y2H screens were among the top predictions by computational dockings. Although, not all the predicted interacting partners could show interaction in Y2H, yet, the harmony between the two approaches suggests that the computational predictions in the present work are reliable. Moreover, the present Y2H analyses per se provide interaction network among MAPKKs and MAPKs which would shed more light on MAPK signalling network in rice.

Characterization of nuclear localization signal in the N terminus of integrin-linked kinase-associated phosphatase (ILKAP) and its essential role in the down-regulation of RSK2 protein signaling

Integrin-linked kinase-associated phosphatase (ILKAP) is a serine/threonine (S/T) phosphatase that belongs to the protein phosphatase 2C (PP2C) family. Many previous studies have demonstrated that ILKAP plays key roles in the regulation of cell survival and apoptosis. Researchers have thus far considered ILKAP a cytoplasmic protein that negatively regulates integrin signaling by interacting with and phosphorylating integrin-linked kinase 1 (ILK1). In this study, we found that both endogenous and tagged ILKAP mainly localize to the nucleus and that the nuclear transport of ILKAP is nuclear localization signal (NLS) importin-mediated. The ILKAP protein interacts directly with importin α1, α3, and α5. The NLS in ILKAP is located in the N-terminal region between amino acids 71 and 86, and the NLS-deleted ILKAP protein was distributed in the cytoplasm. In addition, we show that Lys-78 and Arg-79 are critical for the binding of ILKAP to importin α. We also found that nuclear ILKAP interacts with ribosomal protein S6 kinase-2 (RSK2) and induces apoptosis by inhibiting RSK2 activity and down-regulating the expression level of the RSK2 downstream substrate cyclin D1. These results indicate that ILKAP is a nuclear protein that regulates cell survival and apoptosis through the regulation of RSK2 signaling.

A method for integrative structure determination of protein-protein complexes

MOTIVATION

Structural characterization of protein interactions is necessary for understanding and modulating biological processes. On one hand, X-ray crystallography or NMR spectroscopy provide atomic resolution structures but the data collection process is typically long and the success rate is low. On the other hand, computational methods for modeling assembly structures from individual components frequently suffer from high false-positive rate, rarely resulting in a unique solution.

RESULTS

Here, we present a combined approach that computationally integrates data from a variety of fast and accessible experimental techniques for rapid and accurate structure determination of protein-protein complexes. The integrative method uses atomistic models of two interacting proteins and one or more datasets from five accessible experimental techniques: a small-angle X-ray scattering (SAXS) profile, 2D class average images from negative-stain electron microscopy micrographs (EM), a 3D density map from single-particle negative-stain EM, residue type content of the protein-protein interface from NMR spectroscopy and chemical cross-linking detected by mass spectrometry. The method is tested on a docking benchmark consisting of 176 known complex structures and simulated experimental data. The near-native model is the top scoring one for up to 61% of benchmark cases depending on the included experimental datasets; in comparison to 10% for standard computational docking. We also collected SAXS, 2D class average images and 3D density map from negative-stain EM to model the PCSK9 antigen-J16 Fab antibody complex, followed by validation of the model by a subsequently available X-ray crystallographic structure.

Computer-aided antibody design

Recent clinical trials using antibodies with low toxicity and high efficiency have raised expectations for the development of next-generation protein therapeutics. However, the process of obtaining therapeutic antibodies remains time consuming and empirical. This review summarizes recent progresses in the field of computer-aided antibody development mainly focusing on antibody modeling, which is divided essentially into two parts: (i) modeling the antigen-binding site, also called the complementarity determining regions (CDRs), and (ii) predicting the relative orientations of the variable heavy (V(H)) and light (V(L)) chains. Among the six CDR loops, the greatest challenge is predicting the conformation of CDR-H3, which is the most important in antigen recognition. Further computational methods could be used in drug development based on crystal structures or homology models, including antibody-antigen dockings and energy calculations with approximate potential functions. These methods should guide experimental studies to improve the affinities and physicochemical properties of antibodies. Finally, several successful examples of in silico structure-based antibody designs are reviewed. We also briefly review structure-based antigen or immunogen design, with application to rational vaccine development.

Benchmarking and analysis of protein docking performance in Rosetta v3.2

RosettaDock has been increasingly used in protein docking and design strategies in order to predict the structure of protein-protein interfaces. Here we test capabilities of RosettaDock 3.2, part of the newly developed Rosetta v3.2 modeling suite, against Docking Benchmark 3.0, and compare it with RosettaDock v2.3, the latest version of the previous Rosetta software package. The benchmark contains a diverse set of 116 docking targets including 22 antibody-antigen complexes, 33 enzyme-inhibitor complexes, and 60 ‘other’ complexes. These targets were further classified by expected docking difficulty into 84 rigid-body targets, 17 medium targets, and 14 difficult targets. We carried out local docking perturbations for each target, using the unbound structures when available, in both RosettaDock v2.3 and v3.2. Overall the performances of RosettaDock v2.3 and v3.2 were similar. RosettaDock v3.2 achieved 56 docking funnels, compared to 49 in v2.3. A breakdown of docking performance by protein complex type shows that RosettaDock v3.2 achieved docking funnels for 63% of antibody-antigen targets, 62% of enzyme-inhibitor targets, and 35% of ‘other’ targets. In terms of docking difficulty, RosettaDock v3.2 achieved funnels for 58% of rigid-body targets, 30% of medium targets, and 14% of difficult targets. For targets that failed, we carry out additional analyses to identify the cause of failure, which showed that binding-induced backbone conformation changes account for a majority of failures. We also present a bootstrap statistical analysis that quantifies the reliability of the stochastic docking results. Finally, we demonstrate the additional functionality available in RosettaDock v3.2 by incorporating small-molecules and non-protein co-factors in docking of a smaller target set. This study marks the most extensive benchmarking of the RosettaDock module to date and establishes a baseline for future research in protein interface modeling and structure prediction.

Structural model of the TRPP2/PKD1 C-terminal coiled-coil complex produced by a combined computational and experimental approach

Autosomal dominant polycystic kidney disease (ADPKD) is caused by mutations in TRPP2 and PKD1, which form an ion channel/receptor complex containing three TRPP2 and one PKD1. A TRPP2 C-terminal coiled-coil trimer, critical for the assembly of this complex, associates with a single PKD1 C-terminal coiled-coil. Many ADPKD pathogenic mutations result in the abolishment of the TRPP2/PKD1 coiled-coil complex. To gain molecular and functional insights into this heterotetrameric complex, we computationally constructed a structural model by using a two-step docking strategy, based on a known crystal structure of the TRPP2 coiled-coil trimer. The model shows that this tetrameric complex has a novel di-trimer configuration: An upstream trimer made of three TRPP2 helices and a downstream trimer made of two TRPP2 helices and one PKD1 helix. Mutagenesis and biochemical analysis identified critical TRPP2/PKD1 interface contacts essential for the heteromeric coiled-coil complex. Mutation of these interface positions in the full-length proteins showed that these interactions were critical for the assembly of the full-length complex in cells. Our results provide a means to specifically weaken the TRPP2 and PKD1 association, thus facilitating future in vitro and in vivo studies on the functional importance of this association.

Antibodies as a model system for comparative model refinement

Predicting the conformations of loops is a critical aspect of protein comparative (homology) modeling. Despite considerable advances in developing loop prediction algorithms, refining loops in homology models remains challenging. In this work, we use antibodies as a model system to investigate strategies for more robustly predicting loop conformations when the protein model contains errors in the conformations of side chains and protein backbone surrounding the loop in question. Specifically, our test system consists of partial models of antibodies in which the "scaffold" (i.e., the portion other than the complementarity determining region, CDR, loops) retains native backbone conformation, whereas the CDR loops are predicted using a combination of knowledge-based modeling (H1, H2, L1, L2, and L3) and ab initio loop prediction (H3). H3 is the most variable of the CDRs. Using a previously published method, a test set of 10 shorter H3 loops (5-7 residues) are predicted to an average backbone (N-C alpha-C-O) RMSD of 2.7 A while 11 longer loops (8-9 residues) are predicted to 5.1 A, thus recapitulating the difficulties in refining loops in models. By contrast, in control calculations predicting the same loops in crystal structures, the same method reconstructs the loops to an average of 0.5 and 1.4 A for the shorter and longer loops, respectively. We modify the loop prediction method to improve the ability to sample near-native loop conformations in the models, primarily by reducing the sensitivity of the sampling to the loop surroundings, and allowing the other CDR loops to optimize with the H3 loop. The new method improves the average accuracy significantly to 1.3 A RMSD and 3.1 A RMSD for the shorter and longer loops, respectively. Finally, we present results predicting 8-10 residue loops within complete comparative models of five nonantibody proteins. While anecdotal, these mixed, full-model results suggest our approach is a promising step toward more accurately predicting loops in homology models. Furthermore, while significant challenges remain, our method is a potentially useful tool for predicting antibody structures based on a known Fv scaffold.

RosettaBackrub--a web server for flexible backbone protein structure modeling and design

The RosettaBackrub server (http://kortemmelab.ucsf.edu/backrub) implements the Backrub method, derived from observations of alternative conformations in high-resolution protein crystal structures, for flexible backbone protein modeling. Backrub modeling is applied to three related applications using the Rosetta program for structure prediction and design: (I) modeling of structures of point mutations, (II) generating protein conformational ensembles and designing sequences consistent with these conformations and (III) predicting tolerated sequences at protein–protein interfaces. The three protocols have been validated on experimental data. Starting from a user-provided single input protein structure in PDB format, the server generates near-native conformational ensembles. The predicted conformations and sequences can be used for different applications, such as to guide mutagenesis experiments, for ensemble-docking approaches or to generate sequence libraries for protein design.

Practically useful: what the Rosetta protein modeling suite can do for you

The objective of this review is to enable researchers to use the software package Rosetta for biochemical and biomedicinal studies. We provide a brief review of the six most frequent research problems tackled with Rosetta. For each of these six tasks, we provide a tutorial that illustrates a basic Rosetta protocol. The Rosetta method was originally developed for de novo protein structure prediction and is regularly one of the best performers in the community-wide biennial Critical Assessment of Structure Prediction. Predictions for protein domains with fewer than 125 amino acids regularly have a backbone root-mean-square deviation of better than 5.0 Å. More impressively, there are several cases in which Rosetta has been used to predict structures with atomic level accuracy better than 2.5 Å. In addition to de novo structure prediction, Rosetta also has methods for molecular docking, homology modeling, determining protein structures from sparse experimental NMR or EPR data, and protein design. Rosetta has been used to accurately design a novel protein structure, predict the structure of protein−protein complexes, design altered specificity protein−protein and protein−DNA interactions, and stabilize proteins and protein complexes. Most recently, Rosetta has been used to solve the X-ray crystallographic phase problem.

A molecular model for cocaine binding by the immunotherapeutic human/mouse chimeric monoclonal antibody 2E2

Immunotherapy by cocaine-binding monoclonal antibodies (mAbs) has emerged as a promising strategy for the treatment of cocaine addiction. The human (gamma1 heavy chain)/murine (lambda light chain) chimeric mAb 2E2 has excellent affinity and specificity for cocaine and recent animal studies have demonstrated 2E2's ability in vivo to reduce cocaine levels in the brain as well as alter cocaine self-administration behavior in rats. In this study, we used mAb 2E2 amino acid sequence information to create a homology model for the 3-D structure of its Fv fragment. Subsequent computational docking studies revealed the intermolecular interactions potentially responsible for mAb 2E2's cocaine binding properties. The driving force of cocaine binding was identified as a combination of hydrophobic interactions and a single hydrogen bond between a light chain tyrosine residue and a carbonyl oxygen atom of cocaine. The model also allowed for an in silico evaluation of single/double residue mutations in the heavy and light chain variable regions that might further enhance mAb 2E2's cocaine binding properties.

Nimotuzumab, an antitumor antibody that targets the epidermal growth factor receptor, blocks ligand binding while permitting the active receptor conformation

Overexpression of the epidermal growth factor (EGF) receptor (EGFR) in cancer cells correlates with tumor malignancy and poor prognosis for cancer patients. For this reason, the EGFR has become one of the main targets of anticancer therapies. Structural data obtained in the last few years have revealed the molecular mechanism for ligand-induced EGFR dimerization and subsequent signal transduction, and also how this signal is blocked by either monoclonal antibodies or small molecules. Nimotuzumab (also known as h-R3) is a humanized antibody that targets the EGFR and has been successful in the clinics. In this work, we report the crystal structure of the Fab fragment of Nimotuzumab, revealing some unique structural features in the heavy variable domain. Furthermore, competition assays show that Nimotuzumab binds to domain III of the extracellular region of the EGFR, within an area that overlaps with both the surface patch recognized by Cetuximab (another anti-EGFR antibody) and the binding site for EGF. A computer model of the Nimotuzumab-EGFR complex, constructed by docking and molecular dynamics simulations and supported by mutagenesis studies, unveils a novel mechanism of action, with Nimotuzumab blocking EGF binding while still allowing the receptor to adopt its active conformation, hence warranting a basal level of signaling.

RosettaAntibody: antibody variable region homology modeling server

The RosettaAntibody server (http://antibody.graylab.jhu.edu) predicts the structure of an antibody variable region given the amino-acid sequences of the respective light and heavy chains. In an initial stage, the server identifies and displays the most sequence homologous template structures for the light and heavy framework regions and each of the complementarity determining region (CDR) loops. Subsequently, the most homologous templates are assembled into a side-chain optimized crude model, and the server returns a picture and coordinate file. For users requesting a high-resolution model, the server executes the full RosettaAntibody protocol which additionally models the hyper-variable CDR H3 loop. The high-resolution protocol also relieves steric clashes by optimizing the CDR backbone torsion angles and by simultaneously perturbing the relative orientation of the light and heavy chains. RosettaAntibody generates 2000 independent structures, and the server returns pictures, coordinate files, and detailed scoring information for the 10 top-scoring models. The 10 models enable users to use rational judgment in choosing the best model or to use the set as an ensemble for further studies such as docking. The high-resolution models generated by RosettaAntibody have been used for the successful prediction of antibody–antigen complex structures.

Antibodies specifically targeting a locally misfolded region of tumor associated EGFR

Epidermal Growth Factor Receptor (EGFR) is involved in stimulating the growth of many human tumors, but the success of therapeutic agents has been limited in part by interference from the EGFR on normal tissues. Previously, we reported an antibody (mab806) against a truncated form of EGFR found commonly in gliomas. Remarkably, it also recognizes full-length EGFR on tumor cells but not on normal cells. However, the mechanism for this activity was unclear. Crystallographic structures for Fab:EGFR(287-302) complexes of mAb806 (and a second, related antibody, mAb175) show that this peptide epitope adopts conformations similar to those found in the wtEGFR. However, in both conformations observed for wtEGFR, tethered and untethered, antibody binding would be prohibited by significant steric clashes with the CR1 domain. Thus, these antibodies must recognize a cryptic epitope in EGFR. Structurally, it appeared that breaking the disulfide bond preceding the epitope might allow the CR1 domain to open up sufficiently for antibody binding. The EGFR(C271A/C283A) mutant not only binds mAb806, but binds with 1:1 stoichiometry, which is significantly greater than wtEGFR binding. Although mAb806 and mAb175 decrease tumor growth in xenografts displaying mutant, overexpressed, or autocrine stimulated EGFR, neither antibody inhibits the in vitro growth of cells expressing wtEGFR. In contrast, mAb806 completely inhibits the ligand-associated stimulation of cells expressing EGFR(C271A/C283A). Clearly, the binding of mAb806 and mAb175 to the wtEGFR requires the epitope to be exposed either during receptor activation, mutation, or overexpression. This mechanism suggests the possibility of generating antibodies to target other wild-type receptors on tumor cells.

Toward high-resolution homology modeling of antibody Fv regions and application to antibody-antigen docking

High-resolution homology models are useful in structure-based protein engineering applications, especially when a crystallographic structure is unavailable. Here, we report the development and implementation of RosettaAntibody, a protocol for homology modeling of antibody variable regions. The protocol combines comparative modeling of canonical complementarity determining region (CDR) loop conformations and de novo loop modeling of CDR H3 conformation with simultaneous optimization of V(L)-V(H) rigid-body orientation and CDR backbone and side-chain conformations. The protocol was tested on a benchmark of 54 antibody crystal structures. The median root mean square deviation (rmsd) of the antigen binding pocket comprised of all the CDR residues was 1.5 A with 80% of the targets having an rmsd lower than 2.0 A. The median backbone heavy atom global rmsd of the CDR H3 loop prediction was 1.6, 1.9, 2.4, 3.1, and 6.0 A for very short (4-6 residues), short (7-9), medium (10-11), long (12-14) and very long (17-22) loops, respectively. When the set of ten top-scoring antibody homology models are used in local ensemble docking to antigen, a moderate-to-high accuracy docking prediction was achieved in seven of fifteen targets. This success in computational docking with high-resolution homology models is encouraging, but challenges still remain in modeling antibody structures for sequences with long H3 loops. This first large-scale antibody-antigen docking study using homology models reveals the level of "functional accuracy" of these structural models toward protein engineering applications.

Interaction of antibodies with ErbB receptor extracellular regions

Antibodies to the extracellular region of the ErbB receptors have played key roles in the development of a mechanistic understanding of this family of receptor tyrosine kinases. An extensively studied class of such antibodies inhibits activation of ErbB receptors, and these antibodies have been the focus of intense development as anti-cancer agents. In this review we consider the properties of ErbB receptors antibodies in light of the current structure-based model for ErbB receptor homo- and hetero-dimerization and activation. Crystal structures of the Fab fragments from five different inhibitory antibodies in complex with the extracellular regions of EGFR and ErbB2 have been determined. These structures highlight several different modes of binding and mechanisms of receptor inhibition. Information about antibody interactions with the structurally well-characterized soluble extracellular regions of ErbB receptors can be combined with the rich knowledge of the effects of these antibodies in cultured cells, and in vivo, to provide insights into the conformation and activation of ErbB receptors at the cell surface.

Conformer selection and induced fit in flexible backbone protein-protein docking using computational and NMR ensembles

Accommodating backbone flexibility continues to be the most difficult challenge in computational docking of protein-protein complexes. Towards that end, we simulate four distinct biophysical models of protein binding in RosettaDock, a multiscale Monte-Carlo-based algorithm that uses a quasi-kinetic search process to emulate the diffusional encounter of two proteins and to identify low-energy complexes. The four binding models are as follows: (1) key-lock (KL) model, using rigid-backbone docking; (2) conformer selection (CS) model, using a novel ensemble docking algorithm; (3) induced fit (IF) model, using energy-gradient-based backbone minimization; and (4) combined conformer selection/induced fit (CS/IF) model. Backbone flexibility was limited to the smaller partner of the complex, structural ensembles were generated using Rosetta refinement methods, and docking consisted of local perturbations around the complexed conformation using unbound component crystal structures for a set of 21 target complexes. The lowest-energy structure contained >30% of the native residue-residue contacts for 9, 13, 13, and 14 targets for KL, CS, IF, and CS/IF docking, respectively. When applied to 15 targets using nuclear magnetic resonance ensembles of the smaller protein, the lowest-energy structure recovered at least 30% native residue contacts in 3, 8, 4, and 8 targets for KL, CS, IF, and CS/IF docking, respectively. CS/IF docking of the nuclear magnetic resonance ensemble performed equally well or better than KL docking with the unbound crystal structure in 10 of 15 cases. The marked success of CS and CS/IF docking shows that ensemble docking can be a versatile and effective method for accommodating conformational plasticity in docking and serves as a demonstration for the CS theory--that binding-competent conformers exist in the unbound ensemble and can be selected based on their favorable binding energies.

The RosettaDock server for local protein-protein docking

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

Modeling the structure of mAb 14B7 bound to the anthrax protective antigen

The anthrax protective antigen (PA) is a key component of the tripartite anthrax toxin. Monoclonal antibody (mAb) 14B7 and its engineered, affinity-matured variants have been shown to be effective in blocking PA binding to cellular receptors and mitigating anthrax toxicity. Here, we perform computational structural modeling of the mAb 14B7-PA interaction. Our objectives are to determine the structure of the 14B7-PA complex, to deduce a structural explanation for the affinity maturation from the docking models, and to study the effect of inaccuracies in the antibody homology model on docking. We used the RosettaDock program to dock PA with the mAb 14B7 crystal structure or homology model. Our simulations generate two distinct binding orientations consistent with experimental residue mutations that diminish 14B7-PA binding. Furthermore, the models suggest new site-directed mutations to positively identify one of these two solutions as the correct 14B7-PA docking orientation. The models indicate that PA regions 648-660 and 712-720 may be important for 14B7 binding in addition to the known PA epitope, and the binding interfaces are similar to that seen in the PA complex with cellular receptor CMG2. Antibody residues involved in affinity maturation do not contact the antigen in the docking models, suggesting that affinity maturation in the 14B7 family does not result from direct enhancements of antibody-antigen contacts. Docking the homology model produces low-resolution representations of the crystal structure docking orientations, but homology model docking is frustrated by antibody H3 loop conformation errors. This work demonstrates the usefulness and limitations of computational structure prediction for the development of antibody therapeutics, and reemphasizes the need for flexible backbone docking algorithms to achieve high-resolution docking using homology models.

Incorporating biochemical information and backbone flexibility in RosettaDock for CAPRI rounds 6-12

In CAPRI rounds 6-12, RosettaDock successfully predicted 2 of 5 unbound-unbound targets to medium accuracy. Improvement over the previous method was achieved with computational mutagenesis to select decoys that match the energetics of experimentally determined hot spots. In the case of Target 21, Orc1/Sir1, this resulted in a successful docking prediction where RosettaDock alone or with simple site constraints failed. Experimental information also helped limit the interacting region of TolB/Pal, producing a successful prediction of Target 26. In addition, we docked multiple loop conformations for Target 20, and we developed a novel flexible docking algorithm to simultaneously optimize backbone conformation and rigid-body orientation to generate a wide diversity of conformations for Target 24. Continued challenges included docking of homology targets that differ substantially from their template (sequence identity <50%) and accounting for large conformational changes upon binding. Despite a larger number of unbound-unbound and homology model binding targets, Rounds 6-12 reinforced that RosettaDock is a powerful algorithm for predicting bound complex structures, especially when combined with experimental data.

Integrating diverse data for structure determination of macromolecular assemblies

To understand the cell, we need to determine the macromolecular assembly structures, which may consist of tens to hundreds of components. First, we review the varied experimental data that characterize the assemblies at several levels of resolution. We then describe computational methods for generating the structures using these data. To maximize completeness, resolution, accuracy, precision, and efficiency of the structure determination, a computational approach is required that uses spatial information from a variety of experimental methods. We propose such an approach, defined by its three main components: a hierarchical representation of the assembly, a scoring function consisting of spatial restraints derived from experimental data, and an optimization method that generates structures consistent with the data. This approach is illustrated by determining the configuration of the 456 proteins in the nuclear pore complex (NPC) from baker's yeast. With these tools, we are poised to integrate structural information gathered at multiple levels of the biological hierarchy--from atoms to cells--into a common framework.

Using the natural evolution of a rotavirus-specific human monoclonal antibody to predict the complex topography of a viral antigenic site

BACKGROUND

Understanding the interaction between viral proteins and neutralizing antibodies at atomic resolution is hindered by a lack of experimentally solved complexes. Progress in computational docking has led to the prediction of increasingly high-quality model antibody-antigen complexes. The accuracy of atomic-level docking predictions is improved when integrated with experimental information and expert knowledge.

METHODS

Binding affinity data associated with somatic mutations of a rotavirus-specific human adult antibody (RV6-26) are used to filter potential docking orientations of an antibody homology model with respect to the rotavirus VP6 crystal structure. The antibody structure is used to probe the VP6 trimer for candidate interface residues.

RESULTS

Three conformational epitopes are proposed. These epitopes are candidate antigenic regions for site-directed mutagenesis of VP6, which will help further elucidate antigenic function. A pseudo-atomic resolution RV6-26 antibody-VP6 complex is proposed consistent with current experimental information.

CONCLUSION

The use of mutagenesis constraints in docking calculations allows for the identification of a small number of alternative arrangements of the antigen-antibody interface. The mutagenesis information from the natural evolution of a neutralizing antibody can be used to discriminate between residue-scale models and create distance constraints for atomic-resolution docking. The integration of binding affinity data or other information with computation may be an advantageous approach to assist peptide engineering or therapeutic antibody design.

Targeted cancer therapies based on antibodies directed against epidermal growth factor receptor: status and perspectives

Compelling experimental and clinical evidence suggests that epidermal growth factor receptor (EGFR) plays an important role in the pathogenesis of a variety of human cancers; thus, providing a strong rationale for the development of receptor antagonists as effective and specific therapeutic strategies for the treatment of EGFR-expressing cancers. Monoclonal antibodies (mAb), owing to their high specificity towards a given target, represent a unique class of novel cancer therapeutics. A number of anti-EGFR mAb are currently being developed in our clinic, including two that have been approved by the United States Food and Drug Administration for the treatment of refractory metastatic colorectal cancer (mCRC) and squamous cell carcinomas of the head and neck (SCCHN). Cetuximab (Erbitux, IMC-C225), an IgG1 mAb, has demonstrated significant antitumor activity, both as a single agent and in combination with chemotherapeutics and radiation, in patients with refractory mCRC and SCCHN, respectively. Panitumumab (Vectibix), an IgG2 mAb, has been approved as a single agent for the treatment of patients with refractory mCRC. These mAb, via blocking ligand/receptor interactions, exert their biological activity via multiple mechanisms, including inhibition of cell cycle progression, potentiation of cell apoptosis, inhibition of DNA repair, inhibition of angiogenesis, tumor cell invasion and metastasis and, potentially, induction of immunological effector mechanisms. Anti-EGFR antibodies have demonstrated good safety profiles and potent anticancer activity in our clinic and may prove to be efficacious agents in the treatment of a variety of human malignancies.

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.