Vasor: Accurate prediction of variant effects for amino acid substitutions in multidrug resistance protein 3.

The phosphatidylcholine floppase multidrug resistance protein 3 (MDR3) is an essential hepatobiliary transport protein. MDR3 dysfunction is associated with various liver diseases, ranging from severe progressive familial intrahepatic cholestasis to transient forms of intrahepatic cholestasis of pregnancy and familial gallstone disease. Single amino acid substitutions are often found as causative of dysfunction, but identifying the substitution effect in in vitro studies is time and cost intensive. We developed variant assessor of MDR3 (Vasor), a machine learning-based model to classify novel MDR3 missense variants into the categories benign or pathogenic. Vasor was trained on the largest data set to date that is specific for benign and pathogenic variants of MDR3 and uses general predictors, namely Evolutionary Models of Variant Effects (EVE), EVmutation, PolyPhen-2, I-Mutant2.0, MUpro, MAESTRO, and PON-P2 along with other variant properties, such as half-sphere exposure and posttranslational modification site, as input. Vasor consistently outperformed the integrated general predictors and the external prediction tool MutPred2, leading to the current best prediction performance for MDR3 single-site missense variants (on an external test set: F1-score, 0.90; Matthew's correlation coefficient, 0.80). Furthermore, Vasor predictions cover the entire sequence space of MDR3. Vasor is accessible as a webserver at https://cpclab.uni-duesseldorf.de/mdr3_predictor/ for users to rapidly obtain prediction results and a visualization of the substitution site within the MDR3 structure. The MDR3-specific prediction tool Vasor can provide reliable predictions of single-site amino acid substitutions, giving users a fast way to initially assess whether a variant is benign or pathogenic.

TopProperty: Robust Metaprediction of Transmembrane and Globular Protein Features Using Deep Neural Networks.

Transmembrane proteins (TMPs) are critical components of cellular life. However, due to experimental challenges, the number of experimentally resolved TMP structures is severely underrepresented in databases compared to their cellular abundance. Prediction of (per-residue) features such as transmembrane topology, membrane exposure, secondary structure, and solvent accessibility can be a useful starting point for experimental design or protein structure prediction but often requires different computational tools for different features or types of proteins. We present TopProperty, a metapredictor that predicts all of these features for TMPs or globular proteins. TopProperty is trained on datasets without bias toward a high number of sequence homologs, and the predictions are significantly better than the evaluated state-of-the-art primary predictors on all quality metrics. TopProperty eliminates the need for protein type- or feature-tailored tools, specifically for TMPs. TopProperty is freely available as a web server and standalone at https://cpclab.uni-duesseldorf.de/topsuite/.

TopDomain: Exhaustive Protein Domain Boundary Metaprediction Combining Multisource Information and Deep Learning.

Protein domains are independent, functional, and stable structural units of proteins. Accurate protein domain boundary prediction plays an important role in understanding protein structure and evolution, as well as for protein structure prediction. Current domain boundary prediction methods differ in terms of boundary definition, methodology, and training databases resulting in disparate performance for different proteins. We developed TopDomain, an exhaustive metapredictor, that uses deep neural networks to combine multisource information from sequence- and homology-based features of over 50 primary predictors. For this purpose, we developed a new domain boundary data set termed the TopDomain data set, in which the true annotations are informed by SCOPe annotations, structural domain parsers, human inspection, and deep learning. We benchmark TopDomain against 2484 targets with 3354 boundaries from the TopDomain test set and achieve F1 scores of 78.4% and 73.8% for multidomain boundary prediction within ±20 residues and ±10 residues of the true boundary, respectively. When examined on targets from CASP11-13 competitions, TopDomain achieves F1 scores of 47.5% and 42.8% for multidomain proteins. TopDomain significantly outperforms 15 widely used, state-of-the-art ab initio and homology-based domain boundary predictors. Finally, we implemented TopDomainTMC, which accurately predicts whether domain parsing is necessary for the target protein.

Promiscuous Esterases Counterintuitively Are Less Flexible than Specific Ones.

Understanding mechanisms of promiscuity is increasingly important from a fundamental and application point of view. As to enzyme structural dynamics, more promiscuous enzymes generally have been recognized to also be more flexible. However, examples for the opposite received much less attention. Here, we exploit comprehensive experimental information on the substrate promiscuity of 147 esterases tested against 96 esters together with computationally efficient rigidity analyses to understand the molecular origin of the observed promiscuity range. Unexpectedly, our data reveal that promiscuous esterases are significantly less flexible than specific ones, are significantly more thermostable, and have a significantly increased specific activity. These results may be reconciled with a model according to which structural flexibility in the case of specific esterases serves for conformational proofreading. Our results signify that an esterase sequence space can be screened by rigidity analyses for promiscuous esterases as starting points for further exploration in biotechnology and synthetic chemistry.

TopSuite Web Server: A Meta-Suite for Deep-Learning-Based Protein Structure and Quality Prediction.

Proteins carry out the most fundamental processes of life such as cellular metabolism, regulation, and communication. Understanding these processes at a molecular level requires knowledge of their three-dimensional structures. Experimental techniques such as X-ray crystallography, NMR spectroscopy, and cryogenic electron microscopy can resolve protein structures but are costly and time-consuming and do not work for all proteins. Computational protein structure prediction tries to overcome these problems by predicting the structure of a new protein using existing protein structures as a resource. Here we present TopSuite, a web server for protein model quality assessment (TopScore) and template-based protein structure prediction (TopModel). TopScore provides meta-predictions for global and residue-wise model quality estimation using deep neural networks. TopModel predicts protein structures using a top-down consensus approach to aid the template selection and subsequently uses TopScore to refine and assess the predicted structures. The TopSuite Web server is freely available at https://cpclab.uni-duesseldorf.de/topsuite/.

TopModel: Template-Based Protein Structure Prediction at Low Sequence Identity Using Top-Down Consensus and Deep Neural Networks.

Knowledge of protein structures is essential to understand proteins' functions, evolution, dynamics, stabilities, and interactions and for data-driven protein- or drug design. Yet, experimental structure determination rates are far exceeded by that of next-generation sequencing, resulting in less than 1/1000th of proteins having an experimentally known 3D structure. Computational structure prediction seeks to alleviate this problem, and the Critical Assessment of Protein Structure Prediction (CASP) has shown the value of consensus and meta-methods that utilize complementary algorithms. However, traditionally, such methods employ majority voting during template selection and model averaging during refinement, which can drive the model away from the native fold if it is underrepresented in the ensemble. Here, we present TopModel, a fully automated meta-method for protein structure prediction. In contrast to traditional consensus and meta-methods, TopModel uses top-down consensus and deep neural networks to select templates and identify and correct wrongly modeled regions. TopModel combines a broad range of state-of-the-art methods for threading, alignment, and model quality estimation and provides a versatile workflow and toolbox for template-based structure prediction. TopModel shows a superior template selection, alignment accuracy, and model quality for template-based structure prediction on the CASP10-12 datasets compared to 12 state-of-the-art stand-alone primary predictors. TopModel was validated by prospective predictions of the nisin resistance protein (NSR) protein from Streptococcus agalactiae and LipoP from Clostridium difficile, showing far better agreement with experimental data than any of its constituent primary predictors. These results, in general, demonstrate the utility of TopModel for protein structure prediction and, in particular, show how combining computational structure prediction with sparse or low-resolution experimental data can improve the final model.

TopScore: Using Deep Neural Networks and Large Diverse Data Sets for Accurate Protein Model Quality Assessment.

The value of protein models obtained with automated protein structure prediction depends primarily on their accuracy. Protein model quality assessment is thus critical to select the model that can best answer biologically relevant questions from an ensemble of predictions. However, despite many advances in the field, different methods capture different types of errors, begging the question of which method to use. We introduce TopScore, a meta Model Quality Assessment Program (meta-MQAP) that uses deep neural networks to combine scores from 15 different primary predictors to predict accurate residue-wise and whole-protein error estimates. The predictions on six large independent data sets are highly correlated to superposition-independent errors in the model, achieving a Pearson's Rall2 of 0.93 and 0.78 for whole-protein and residue-wise error predictions, respectively. This is a significant improvement over any of the investigated primary MQAPs, demonstrating that much can be gained by optimally combining different methods and using different and very large data sets.

Recognition motif and mechanism of ripening inhibitory peptides in plant hormone receptor ETR1.

Synthetic peptides derived from ethylene-insensitive protein 2 (EIN2), a central regulator of ethylene signalling, were recently shown to delay fruit ripening by interrupting protein-protein interactions in the ethylene signalling pathway. Here, we show that the inhibitory peptide NOP-1 binds to the GAF domain of ETR1 - the prototype of the plant ethylene receptor family. Site-directed mutagenesis and computational studies reveal the peptide interaction site and a plausible molecular mechanism for the ripening inhibition.

Determinants of FIV and HIV Vif sensitivity of feline APOBEC3 restriction factors.

BACKGROUND: Feline immunodeficiency virus (FIV) is a global pathogen of Felidae species and a model system for Human immunodeficiency virus (HIV)-induced AIDS. In felids such as the domestic cat (Felis catus), APOBEC3 (A3) genes encode for single-domain A3Z2s, A3Z3 and double-domain A3Z2Z3 anti-viral cytidine deaminases. The feline A3Z2Z3 is expressed following read-through transcription and alternative splicing, introducing a previously untranslated exon in frame, encoding a domain insertion called linker. Only A3Z3 and A3Z2Z3 inhibit Vif-deficient FIV. Feline A3s also are restriction factors for HIV and Simian immunodeficiency viruses (SIV). Surprisingly, HIV-2/SIV Vifs can counteract feline A3Z2Z3. RESULTS: To identify residues in feline A3s that Vifs need for interaction and degradation, chimeric human-feline A3s were tested. Here we describe the molecular direct interaction of feline A3s with Vif proteins from cat FIV and present the first structural A3 model locating these interaction regions. In the Z3 domain we have identified residues involved in binding of FIV Vif, and their mutation blocked Vif-induced A3Z3 degradation. We further identified additional essential residues for FIV Vif interaction in the A3Z2 domain, allowing the generation of FIV Vif resistant A3Z2Z3. Mutated feline A3s also showed resistance to the Vif of a lion-specific FIV, indicating an evolutionary conserved Vif-A3 binding. Comparative modelling of feline A3Z2Z3 suggests that the residues interacting with FIV Vif have, unlike Vif-interacting residues in human A3s, a unique location at the domain interface of Z2 and Z3 and that the linker forms a homeobox-like domain protruding of the Z2Z3 core. HIV-2/SIV Vifs efficiently degrade feline A3Z2Z3 by possible targeting the linker stretch connecting both Z-domains. CONCLUSIONS: Here we identified in feline A3s residues important for binding of FIV Vif and a unique protein domain insertion (linker). To understand Vif evolution, a structural model of the feline A3 was developed. Our results show that HIV Vif binds human A3s differently than FIV Vif feline A3s. The linker insertion is suggested to form a homeo-box domain, which is unique to A3s of cats and related species, and not found in human and mouse A3s. Together, these findings indicate a specific and different A3 evolution in cats and human.

Structural basis of lantibiotic recognition by the nisin resistance protein from Streptococcus agalactiae.

Lantibiotics are potent antimicrobial peptides. Nisin is the most prominent member and contains five crucial lanthionine rings. Some clinically relevant bacteria express membrane-associated resistance proteins that proteolytically inactivate nisin. However, substrate recognition and specificity of these proteins is unknown. Here, we report the first three-dimensional structure of a nisin resistance protein from Streptococcus agalactiae (SaNSR) at 2.2 Å resolution. It contains an N-terminal helical bundle, and protease cap and core domains. The latter harbors the highly conserved TASSAEM region, which lies in a hydrophobic tunnel formed by all domains. By integrative modeling, mutagenesis studies, and genetic engineering of nisin variants, a model of the SaNSR/nisin complex is generated, revealing that SaNSR recognizes the last C-terminally located lanthionine ring of nisin. This determines the substrate specificity of SaNSR and ensures the exact coordination of the nisin cleavage site at the TASSAEM region.

VisualCNA: a GUI for interactive constraint network analysis and protein engineering for improving thermostability.

UNLABELLED: Constraint network analysis (CNA) is a graph theory-based rigidity analysis approach for linking a biomolecule's structure, flexibility, (thermo)stability and function. Results from CNA are highly information-rich and require intuitive, synchronized and interactive visualization for a comprehensive analysis. We developed VisualCNA, an easy-to-use PyMOL plug-in that allows setup of CNA runs and analysis of CNA results linking plots with molecular graphics representations. From a practical viewpoint, the most striking feature of VisualCNA is that it facilitates interactive protein engineering aimed at improving thermostability. AVAILABILITY AND IMPLEMENTATION: VisualCNA and its dependencies (CNA and FIRST software) are available free of charge under GPL and academic licenses, respectively. VisualCNA and CNA are available at http://cpclab.uni-duesseldorf.de/software; FIRST is available at http://flexweb.asu.edu.

Molecular dynamics simulations and structure-guided mutagenesis provide insight into the architecture of the catalytic core of the ectoine hydroxylase.

Many bacteria amass compatible solutes to fend-off the detrimental effects of high osmolarity on cellular physiology and water content. These solutes also function as stabilizers of macromolecules, a property for which they are referred to as chemical chaperones. The tetrahydropyrimidine ectoine is such a compatible solute and is widely synthesized by members of the Bacteria. Many ectoine producers also synthesize the stress protectant 5-hydroxyectoine from the precursor ectoine, a process that is catalyzed by the ectoine hydroxylase (EctD). The EctD enzyme is a member of the non-heme-containing iron(II) and 2-oxoglutarate-dependent dioxygenase superfamily. A crystal structure of the EctD protein from the moderate halophile Virgibacillus salexigens has previously been reported and revealed the coordination of the iron catalyst, but it lacked the substrate ectoine and the co-substrate 2-oxoglutarate. Here we used this crystal structure as a template to assess the likely positioning of the ectoine and 2-oxoglutarate ligands within the active site by structural comparison, molecular dynamics simulations, and site-directed mutagenesis. Collectively, these approaches suggest the positioning of the iron, ectoine, and 2-oxoglutarate ligands in close proximity to each other and with a spatial orientation that will allow the region-selective and stereo-specific hydroxylation of (4S)-ectoine to (4S,5S)-5-hydroxyectoine. Our study thus provides a view into the catalytic core of the ectoine hydroxylase and suggests an intricate network of interactions between the three ligands and evolutionarily highly conserved residues in members of the EctD protein family.

Binding region of alanopine dehydrogenase predicted by unbiased molecular dynamics simulations of ligand diffusion.

Opine dehydrogenases catalyze the reductive condensation of pyruvate with L-amino acids. Biochemical characterization of alanopine dehydrogenase from Arenicola marina revealed that this enzyme is highly specific for L-alanine. Unbiased molecular dynamics simulations with a homology model of alanopine dehydrogenase captured the binding of L-alanine diffusing from solvent to a putative binding region near a distinct helix-kink-helix motif. These results and sequence comparisons reveal how mutations and insertions within this motif dictate the L-amino acid specificity.