Search

Nutlin-3a is a reversible inhibitor of the p53/MDM2 interaction. We have synthesized the derivative Nutlin-3a-aa bearing an additional exocyclic methylene group in the piperazinone moiety. Nutlin-3a-aa is more active than Nutlin-3a against purified wild-type MDM2, and is more effective at increasing p53 levels and releasing transcription of p53 target genes from MDM2-induced repression. X-ray analysis of wild-type MDM2-bound Nutlin-3a-aa indicated that the orientation of its modified piperazinone ring was altered in comparison to the piperazinone ring of MDM2-bound Nutlin-3a, with the exocyclic methylene group of Nutlin-3a-aa pointing away from the protein surface. Our data point to the introduction of exocyclic methylene groups as a useful approach by which to tailor the conformation of bioactive molecules for improved biological activity.

Recent breakthroughs in protein structure prediction demarcate the start of a new era in structural bioinformatics. Combined with various advances in experimental structure determination and the uninterrupted pace at which new structures are published, this promises an age in which protein structure information is as prevalent and ubiquitous as sequence. Machine learning in protein bioinformatics has been dominated by sequence-based methods, but this is now changing to make use of the deluge of rich structural information as input. Machine learning methods making use of structures are scattered across literature and cover a number of different applications and scopes; while some try to address questions and tasks within a single protein family, others aim to capture characteristics across all available proteins. In this review, we look at the variety of structure-based machine learning approaches, how structures can be used as input, and typical applications of these approaches in protein biology. We also discuss current challenges and opportunities in this all-important and increasingly popular field.

Free-electron lasers could enable X-ray imaging of single biological macro-molecules and the study of protein dynamics, paving the way for a powerful new imaging tool in structural biology, but a low signal-to-noise ratio and missing regions in the detectors, colloquially termed 'masks', affect data collection and hamper real-time evaluation of experimental data. In this article, the challenges posed by noise and masks are tackled by introducing a neural network pipeline that aims to restore diffraction intensities. For training and testing of the model, a data set of diffraction patterns was simulated from 10 900 different proteins with molecular weights within the range of 10-100 kDa and collected at a photon energy of 8 keV. The method is compared with a simple low-pass filtering algorithm based on autocorrelation constraints. The results show an improvement in the mean-squared error of roughly two orders of magnitude in the presence of masks compared with the noisy data. The algorithm was also tested at increasing mask width, leading to the conclusion that demasking can achieve good results when the mask is smaller than half of the central speckle of the pattern. The results highlight the competitiveness of this model for data processing and the feasibility of restoring diffraction intensities from unknown structures in real time using deep learning methods. Finally, an example is shown of this preprocessing making orientation recovery more reliable, especially for data sets containing very few patterns, using the expansion-maximization-compression algorithm.

Free-electron lasers could enable X-ray imaging of single biological macro-molecules and the study of protein dynamics, paving the way for a powerful new imaging tool in structural biology, but a low signal-to-noise ratio and missing regions in the detectors, colloquially termed 'masks', affect data collection and hamper real-time evaluation of experimental data. In this article, the challenges posed by noise and masks are tackled by introducing a neural network pipeline that aims to restore diffraction intensities. For training and testing of the model, a data set of diffraction patterns was simulated from 10 900 different proteins with molecular weights within the range of 10-100 kDa and collected at a photon energy of 8 keV. The method is compared with a simple low-pass filtering algorithm based on autocorrelation constraints. The results show an improvement in the mean-squared error of roughly two orders of magnitude in the presence of masks compared with the noisy data. The algorithm was also tested at increasing mask width, leading to the conclusion that demasking can achieve good results when the mask is smaller than half of the central speckle of the pattern. The results highlight the competitiveness of this model for data processing and the feasibility of restoring diffraction intensities from unknown structures in real time using deep learning methods. Finally, an example is shown of this preprocessing making orientation recovery more reliable, especially for data sets containing very few patterns, using the expansion-maximization-compression algorithm.

Free-electron lasers could enable X-ray imaging of single biological macro-molecules and the study of protein dynamics, paving the way for a powerful new imaging tool in structural biology, but a low signal-to-noise ratio and missing regions in the detectors, colloquially termed 'masks', affect data collection and hamper real-time evaluation of experimental data. In this article, the challenges posed by noise and masks are tackled by introducing a neural network pipeline that aims to restore diffraction intensities. For training and testing of the model, a data set of diffraction patterns was simulated from 10 900 different proteins with molecular weights within the range of 10-100 kDa and collected at a photon energy of 8 keV. The method is compared with a simple low-pass filtering algorithm based on autocorrelation constraints. The results show an improvement in the mean-squared error of roughly two orders of magnitude in the presence of masks compared with the noisy data. The algorithm was also tested at increasing mask width, leading to the conclusion that demasking can achieve good results when the mask is smaller than half of the central speckle of the pattern. The results highlight the competitiveness of this model for data processing and the feasibility of restoring diffraction intensities from unknown structures in real time using deep learning methods. Finally, an example is shown of this preprocessing making orientation recovery more reliable, especially for data sets containing very few patterns, using the expansion-maximization-compression algorithm.

The protein structure field is experiencing a revolution. From the increased throughput of techniques to determine experimental structures, to developments such as cryo-EM that allow us to find the structures of large protein complexes or, more recently, the development of artificial intelligence tools, such as AlphaFold, that can predict with high accuracy the folding of proteins for which the availability of homology templates is limited. Here we quantify the effect of the recently released AlphaFold database of protein structural models in our knowledge on human proteins. Our results indicate that our current baseline for structural coverage of 48%, considering experimentally-derived or template-based homology models, elevates up to 76% when including AlphaFold predictions. At the same time the fraction of dark proteome is reduced from 26% to just 10% when AlphaFold models are considered. Furthermore, although the coverage of disease-associated genes and mutations was near complete before AlphaFold release (69% of Clinvar pathogenic mutations and 88% of oncogenic mutations), AlphaFold models still provide an additional coverage of 3% to 13% of these critically important sets of biomedical genes and mutations. Finally, we show how the contribution of AlphaFold models to the structural coverage of non-human organisms, including important pathogenic bacteria, is significantly larger than that of the human proteome. Overall, our results show that the sequence-structure gap of human proteins has almost disappeared, an outstanding success of direct consequences for the knowledge on the human genome and the derived medical applications., E.P-P and V.R-S are supported by the La Caixa Junior Leader Fellowship LCF/BQ/PI18/11630003 from La Caixa Foundation (https://fundacionlacaixa.org/ca/home). E.P-P is also supported by a Ramon y Cajal fellowship from the Spanish Ministry of Science (RYC2019-026415-I). A.V. is supported by Institució Catalana de Recerca Avançada (ICREA - https://www.icrea.cat) The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript., Peer Reviewed, Postprint (published version)

Free-electron lasers could enable X-ray imaging of single biological macro-molecules and the study of protein dynamics, paving the way for a powerful new imaging tool in structural biology, but a low signal-to-noise ratio and missing regions in the detectors, colloquially termed 'masks', affect data collection and hamper real-time evaluation of experimental data. In this article, the challenges posed by noise and masks are tackled by introducing a neural network pipeline that aims to restore diffraction intensities. For training and testing of the model, a data set of diffraction patterns was simulated from 10 900 different proteins with molecular weights within the range of 10-100 kDa and collected at a photon energy of 8 keV. The method is compared with a simple low-pass filtering algorithm based on autocorrelation constraints. The results show an improvement in the mean-squared error of roughly two orders of magnitude in the presence of masks compared with the noisy data. The algorithm was also tested at increasing mask width, leading to the conclusion that demasking can achieve good results when the mask is smaller than half of the central speckle of the pattern. The results highlight the competitiveness of this model for data processing and the feasibility of restoring diffraction intensities from unknown structures in real time using deep learning methods. Finally, an example is shown of this preprocessing making orientation recovery more reliable, especially for data sets containing very few patterns, using the expansion-maximization-compression algorithm.

Malgré les nouvelles méthodes expérimentales à haut-débit et l'incroyable quantité de données qu'elles génèrent, il reste difficile de mettre en évidence le lien entre une maladie héréditaire, ses causes génétiques et l'impact des variants impliqués sur la structure tridimensionnelle d'une protéine, sa stabilité et sa fonction. En effet, bien qu'il existe de nombreux outils bioinformatiques capables de prédire quels variants dans l'exome humain sont responsables de maladies, peu prédisent la raison moléculaire de cet effet. Or cette information est nécessaire au développement rationnel de médicaments atténuant les symptômes des maladies génétiques et sont une étape importante dans le développement de traitements personnalisés.Le but de cette thèse est de combler ce manque en présentant deux nouveaux outils, SNPMuSiC et FuncMuSiC, qui exploitent la structure tridimensionnelle des protéines pour prédire l'effet délétère de variants dans l'exome humain. Tous deux prédisent si un variant est délétère ou neutre sur base d'une caractéristique biophysique: SNPMuSiC sur base de la variation de stabilité d'une protéine causée par la mutation et FuncMuSiC sur base de la proximité entre le résidu muté et les sites fonctionnels annotés. Par construction, ils fournissent donc une interprétation du phénotype au niveau moléculaire.Nous avons également analysé la sphingomyéline phosphodiestérase (SMPD1), une enzyme clé dans le métabolisme des sphingolipides qui joue un rôle central dans le maintien des membranes plasmiques eucaryotes. Plusieurs mutations dans cette enzyme sont liées à la maladie de Niemann-Pick, qui se manifeste par une série de phénotypes allant de symptômes graves impactant le système nerveux central (type A) à des symptômes n'impactant que les viscères (type B).En combinant des informations évolutionnaires, structurales et contextuelles, nous avons développé l'algorithme SMPD1-ZooM. Ce dernier prédit si un variant au sein de SMPD1 est neutre, associé au type A o, Doctorat en Sciences, info:eu-repo/semantics/nonPublished

Free-electron lasers could enable X-ray imaging of single biological macro-molecules and the study of protein dynamics, paving the way for a powerful new imaging tool in structural biology, but a low signal-to-noise ratio and missing regions in the detectors, colloquially termed 'masks', affect data collection and hamper real-time evaluation of experimental data. In this article, the challenges posed by noise and masks are tackled by introducing a neural network pipeline that aims to restore diffraction intensities. For training and testing of the model, a data set of diffraction patterns was simulated from 10 900 different proteins with molecular weights within the range of 10-100 kDa and collected at a photon energy of 8 keV. The method is compared with a simple low-pass filtering algorithm based on autocorrelation constraints. The results show an improvement in the mean-squared error of roughly two orders of magnitude in the presence of masks compared with the noisy data. The algorithm was also tested at increasing mask width, leading to the conclusion that demasking can achieve good results when the mask is smaller than half of the central speckle of the pattern. The results highlight the competitiveness of this model for data processing and the feasibility of restoring diffraction intensities from unknown structures in real time using deep learning methods. Finally, an example is shown of this preprocessing making orientation recovery more reliable, especially for data sets containing very few patterns, using the expansion-maximization-compression algorithm.

Identification of Mycobacterium tuberculosis (Mtb) increasingly involves characterising large sections of genetic material, such as through whole genome sequencing. While some mutations identified through these techniques are well characterised and strongly associated with anti-tuberculous drug resistance, such molecular methods frequently identify mutations with unknown significance or limited understanding of associated functional biological pathways. In this PhD, I have developed computational protein structural tools and mathematical models of TB transmission, that use genomic data to understand the impact of genomic changes and predict the consequences with regards to transmissibility and drug susceptibility of Mtb. Drug resistant mutations often carry both a selective advantage and a fitness cost, which can be reflected by the changes in protein structure and function. I developed a pipeline that captured the molecular consequences of coding mutations on protein stability, dynamics and interactions. Using my pipeline to evaluate the mechanistic consequences of mutations, I applied it to the real-time genomic analysis of a Victorian tuberculosis patient. The analysis led to identification of a novel resistant strain and altered patient treatment – the first reported use of structural information to guide clinical resistance detection. The information was then used to inform a compartmental epidemiological model of Mtb transmission in order to understand the rise of drug resistance in two high TB-incidence setting. Using a adaptive metropolis algorithm, I estimated drug resistance amplification proportions for two first-line anti-tuberculosis drugs, and explored how structural changes may alter the fitness landscape and transmission dynamics. The work highlighted the power of combining genomic, epidemiological and structural information in the fight against tuberculosis, and presents examples of application across the spectrum from laboratory, clinical and progr

This detailed book collects modern and established computer-based methods aimed at addressing the drug discovery challenge from disparate perspectives by exploiting information on ligand-protein recognition. Beginning with methods that allow for the exploration of specific areas of chemical space and the designing of virtual libraries, the volume continues with sections on methods based on docking, quantitative models, and molecular dynamics simulations, which are employed for ligand discovery or development, as well as methods exploiting an ensemble of protein structures for the identification of potential protein targets. Written for the highly successful Methods in Molecular Biology series, chapters include introductions to their respective topics, lists of the necessary materials, step-by-step, readily reproducible laboratory protocols, and tips on troubleshooting and avoiding known pitfalls. Authoritative and cutting-edge, Protein-Ligand Interactions and Drug Design provides detailed practical procedures of solid computer-aided drug design methodologies employed to rationalize and optimize protein-ligand interactions, for experienced researchers and novices alike.

Photosystem II (PSII) catalyzes light-induced water oxidation through an S-i-state cycle, leading to the generation of di-oxygen, protons and electrons. Pumpprobe time-resolved serial femtosecond crystallography (TR-SFX) has been used to capture structural dynamics of light-sensitive proteins. In this approach, it is crucial to avoid light contamination in the samples when analyzing a particular reaction intermediate. Here, a method for determining a condition that avoids light contamination of the PSII microcrystals while minimizing sample consumption in TR-SFX is described. By swapping the pump and probe pulses with a very short delay between them, the structural changes that occur during the S-1-to-S-2 transition were examined and a boundary of the excitation region was accurately determined. With the sample flow rate and concomitant illumination conditions determined, the S-2-state structure of PSII could be analyzed at room temperature, revealing the structural changes that occur during the S-1-to-S-2 transition at ambient temperature. Though the structure of the manganese cluster was similar to previous studies, the behaviors of the water molecules in the two channels (O1 and O4 channels) were found to be different. By comparing with the previous studies performed at low temperature or with a different delay time, the possible channels for water inlet and structural changes important for the water-splitting reaction were revealed.

Producción Científica, Coronavirus disease 19, or COVID-19, is an infection associated with an unprecedented worldwide pandemic caused by the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), which has led to more than 215 million infected people and more than 4.5 million deaths worldwide. SARS-CoV-2 cell infection is initiated by a densely glycosylated spike (S) protein, a fusion protein, binding human angiotensin converting enzyme 2 (hACE2), that acts as the functional receptor through the receptor binding domain (RBD). In this article, the interaction of hACE2 with the RBD and how fusion is initiated after recognition are explored, as well as how mutations influence infectivity and immune response. Thus, we focused on all structures available in the Protein Data Bank for the interaction between SARS-CoV-2 S protein and hACE2. Specifically, the Delta variant carries particular mutations associated with increased viral fitness through decreased antibody binding, increased RBD affinity and altered protein dynamics. Combining both existing mutations and mutagenesis studies, new potential SARS-CoV-2 variants, harboring advantageous S protein mutations, may be predicted. These include mutations S13I and W152C, decreasing antibody binding, N460K, increasing RDB affinity, or Q498R, positively affecting both properties.

As a problem in data science the inverse Ising (or Potts) problem is to infer the parameters of a Gibbs-Boltzmann distributions of an Ising (or Potts) model from samples drawn from that distribution. The algorithmic and computational interest stems from the fact that this inference task cannot be carried out efficiently by the maximum likelihood criterion, since the normalizing constant of the distribution (the partition function) cannot be calculated exactly and efficiently. The practical interest on the other hand flows from several outstanding applications, of which the most well known has been predicting spatial contacts in protein structures from tables of homologous protein sequences. Most applications to date have been to data that has been produced by a dynamical process which, as far as it is known, cannot be expected to satisfy detailed balance. There is therefore no a priori reason to expect the distribution to be of the Gibbs-Boltzmann type, and no a priori reason to expect that inverse Ising (or Potts) techniques should yield useful information. In this review we discuss two types of problems where progress nevertheless can be made. We find that depending on model parameters there are phases where, in fact, the distribution is close to Gibbs-Boltzmann distribution, a non-equilibrium nature of the under-lying dynamics notwithstanding. We also discuss the relation between inferred Ising model parameters and parameters of the underlying dynamics., QC 20201217

To mimic a hypothetical pathway for protein evolution, we previously tailored a monomeric protein (cyt cb562 ) for metal-mediated self-assembly, followed by re-design of the resulting oligomers for enhanced stability and metal-based functions. We show that a single hydrophobic mutation on the cyt cb562 surface drastically alters the outcome of metal-directed oligomerization to yield a new trimeric architecture, (TriCyt1)3. This nascent trimer was redesigned into second and third-generation variants (TriCyt2)3 and (TriCyt3)3 with increased structural stability and preorganization for metal coordination. The three TriCyt variants combined furnish a unique platform to 1) provide tunable coupling between protein quaternary structure and metal coordination, 2) enable the construction of metal/pH-switchable protein oligomerization motifs, and 3) generate a robust metal coordination site that can coordinate all mid-to-late first-row transition-metal ions with high affinity.

At the center of many complex biosynthetic pathways, the acyl carrier protein (ACP) shuttles substrates to appropriate enzymatic partners to produce fatty acids and polyketides. Carrier proteins covalently tether their cargo via a thioester linkage to a phosphopantetheine cofactor. Due to the labile nature of this linkage, chemoenzymatic methods have been developed that involve replacement of the thioester with a more stable amide or ester bond. We explored the importance of the thioester bond to the structure of the carrier protein by using solution NMR spectroscopy and molecular dynamics simulations. Remarkably, the replacement of sulfur with other heteroatoms results in significant structural changes, thus suggesting more rigorous selections of isosteric substitutes is needed.

At the center of many complex biosynthetic pathways, the acyl carrier protein (ACP) shuttles substrates to appropriate enzymatic partners to produce fatty acids and polyketides. Carrier proteins covalently tether their cargo via a thioester linkage to a phosphopantetheine cofactor. Due to the labile nature of this linkage, chemoenzymatic methods have been developed that involve replacement of the thioester with a more stable amide or ester bond. We explored the importance of the thioester bond to the structure of the carrier protein by using solution NMR spectroscopy and molecular dynamics simulations. Remarkably, the replacement of sulfur with other heteroatoms results in significant structural changes, thus suggesting more rigorous selections of isosteric substitutes is needed.

The possibility of imaging single proteins constitutes an exciting challenge for x-ray lasers. Despite encouraging results on large particles, imaging small particles has proven to be difficult for two reasons: not quite high enough pulse intensity from currently available x-ray lasers and, as we demonstrate here, contamination of the aerosolized molecules by nonvolatile contaminants in the solution. The amount of contamination on the sample depends on the initial droplet size during aerosolization. Here, we show that, with our electrospray injector, we can decrease the size of aerosol droplets and demonstrate virtually contaminant-free sample delivery of organelles, small virions, and proteins. The results presented here, together with the increased performance of next-generation x-ray lasers, constitute an important stepping stone toward the ultimate goal of protein structure determination from imaging at room temperature and high temporal resolution. © 2019 The Authors.

A common interpretation of Anfinsen's hypothesis states that one amino acid sequence should fold into a single, native, ordered state, or a highly similar set thereof, coinciding with the global minimum in the folding-energy landscape, which, in turn, is responsible for the function of the protein. However, this classical view is challenged by many proteins and peptide sequences, which can adopt exchangeable, significantly dissimilar conformations that even fulfill different biological roles. The similarities and differences of concepts related to these proteins, mainly chameleon sequences, metamorphic proteins, and switch peptides, which are all denoted herein “turncoat” polypeptides, are reviewed. As well as adding a twist to the conventional view of protein folding, the lack of structural definition adds clear versatility to the activity of proteins and can be used as a tool for protein design and further application in biotechnology and biomedicine.

The development of accurate epitope prediction tools is important in facilitating disease diagnostics, treatment and vaccine development. The advent of new approaches making use of antibody and TCR sequence information to predict receptor-specific epitopes have the potential to transform the epitope prediction field. Development and validation of these new generation of epitope prediction methods would benefit from regularly updated high-quality receptor-antigen complex datasets. To address the need for high-quality datasets to benchmark performance of these new generation of receptor-specific epitope prediction tools, a webserver called SCEptRe (Structural Complexes of Epitope-Receptor) was created. SCEptRe extracts weekly updated 3D complexes of antibody-antigen, TCR-pMHC and MHC-ligand from the Immune Epitope Database and clusters them based on antigen, receptor and epitope features to generate benchmark datasets. SCEptRe also provides annotated information such as CDR sequences and VDJ genes on the receptors. Users can generate custom datasets based by selecting thresholds for structural quality and clustering parameters (e.g. resolution, R-free factor, antigen or epitope sequence identity) based on their need. SCEptRe provides weekly updated, user-customized comprehensive benchmark datasets of immune receptor-epitope structural complexes. These datasets can be used to develop and benchmark performance of receptor-specific epitope prediction tools in the future. SCEptRe is freely accessible at http://tools.iedb.org/sceptre .

Highly flexible proteins present a special challenge for structure determination because they are multi-structured yet not disordered, so their conformational ensembles are essential for understanding function. Because spectroscopic measurements of multiple conformational populations often provide sparse data, experiment selection is a limiting factor in conformational refinement. We have developed an approach using molecular simulations and information theory to select which experiments best refine conformational ensembles. We test this approach on three flexible proteins. For proteins where a clear mechanistic hypothesis exists, we systematically identify experiments that test this hypothesis. When available data do not yield such mechanistic hypotheses, we identify experiments that significantly outperform structure-guided approaches in conformational refinement. Our approach offers a particular advantage when refining challenging, underdetermined protein conformational ensembles.

Highly flexible proteins present a special challenge for structure determination because they are multi-structured yet not disordered, so their conformational ensembles are essential for understanding function. Because spectroscopic measurements of multiple conformational populations often provide sparse data, experiment selection is a limiting factor in conformational refinement. We have developed an approach using molecular simulations and information theory to select which experiments best refine conformational ensembles. We test this approach on three flexible proteins. For proteins where a clear mechanistic hypothesis exists, we systematically identify experiments that test this hypothesis. When available data do not yield such mechanistic hypotheses, we identify experiments that significantly outperform structure-guided approaches in conformational refinement. Our approach offers a particular advantage when refining challenging, underdetermined protein conformational ensembles.

Highly flexible proteins present a special challenge for structure determination because they are multi-structured yet not disordered, so their conformational ensembles are essential for understanding function. Because spectroscopic measurements of multiple conformational populations often provide sparse data, experiment selection is a limiting factor in conformational refinement. We have developed an approach using molecular simulations and information theory to select which experiments best refine conformational ensembles. We test this approach on three flexible proteins. For proteins where a clear mechanistic hypothesis exists, we systematically identify experiments that test this hypothesis. When available data do not yield such mechanistic hypotheses, we identify experiments that significantly outperform structure-guided approaches in conformational refinement. Our approach offers a particular advantage when refining challenging, underdetermined protein conformational ensembles.

The electron-rich isonitrile is an important functionality in bioactive natural products, but its biosynthesis has been restricted to the IsnA family of isonitrile synthases. We herein provide the first structural and biochemical evidence of an alternative mechanism for isonitrile formation. ScoE, a putative non-heme iron(II)-dependent enzyme from Streptomyces coeruleorubidus, was shown to catalyze the conversion of (R)-3-((carboxymethyl)amino)butanoic acid to (R)-3-isocyanobutanoic acid through an oxidative decarboxylation mechanism. This work further provides a revised scheme for the biosynthesis of a unique class of isonitrile lipopeptides, of which several members are critical for the virulence of pathogenic mycobacteria.

Highly flexible proteins present a special challenge for structure determination because they are multi-structured yet not disordered, so their conformational ensembles are essential for understanding function. Because spectroscopic measurements of multiple conformational populations often provide sparse data, experiment selection is a limiting factor in conformational refinement. We have developed an approach using molecular simulations and information theory to select which experiments best refine conformational ensembles. We test this approach on three flexible proteins. For proteins where a clear mechanistic hypothesis exists, we systematically identify experiments that test this hypothesis. When available data do not yield such mechanistic hypotheses, we identify experiments that significantly outperform structure-guided approaches in conformational refinement. Our approach offers a particular advantage when refining challenging, underdetermined protein conformational ensembles.

Highly flexible proteins present a special challenge for structure determination because they are multi-structured yet not disordered, so their conformational ensembles are essential for understanding function. Because spectroscopic measurements of multiple conformational populations often provide sparse data, experiment selection is a limiting factor in conformational refinement. We have developed an approach using molecular simulations and information theory to select which experiments best refine conformational ensembles. We test this approach on three flexible proteins. For proteins where a clear mechanistic hypothesis exists, we systematically identify experiments that test this hypothesis. When available data do not yield such mechanistic hypotheses, we identify experiments that significantly outperform structure-guided approaches in conformational refinement. Our approach offers a particular advantage when refining challenging, underdetermined protein conformational ensembles.

A new family of tandem facial glucosides/maltosides (TFGs/TFMs) for membrane protein manipulation was prepared. The best detergent varied depending on the hydrophobic thickness of the target protein, but ether-based TFMs (TFM-C0E, TFM-C3E, and TFM-C5E) were notable for their ability to confer higher membrane protein stability than the previously developed amide-based TFA-1 (P. S. Chae, K. Gotfryd, J. Pacyna, L. J. W. Miercke, S. G. F. Rasmussen, R. A. Robbins, R. R. Rana, C. J. Loland, B. Kobilka, R. Stroud, B. Byrne, U. Gether, S. H. Gellman, J. Am. Chem. Soc. 2010, 132, 16750–16752). Thus, this study not only introduces novel agents with the potential to be used in membrane protein research but also highlights the importance of both the hydrophobic length and linker functionality of the detergent in stabilizing membrane proteins.

The catalytic activity of the allosteric enzyme pyruvate decarboxylase from yeast is strictly controlled by its own substrate pyruvate via covalent binding at a separate regulatory site. Kinetic studies, chemical modifications, cross-linking, small-angle X-ray scattering, and crystal structure analyses have led to a detailed understanding of the substrate activation mechanism at an atomic level with C221 as the core moiety of the regulatory site. To characterize the individual role of the residues adjacent to C221, we generated variants H92F, H225F, H310F, A287G, S311A, and C221A/C222A. The integrity of the protein structure of the variants was established by small-angle X-ray scattering measurements. The analyses of both steady state and transient kinetic data allowed the identification of the individual roles of the exchanged side chains during allosteric enzyme activation. In each case, the kinetic pattern of activation was modulated but not completely abolished. Despite the crucial role of C221, the covalent binding of pyruvate is not obligate for enzyme activation but is a requirement for a kinetically efficient transition from the inactive to the active state. Moreover, only one of the three histidines guiding the activator molecule to the binding pocket, H310, specifically interacts with C221. H310 stabilizes the thiolate form of C221, ensuring a rapid nucleophilic attack of the thiolate sulfur on C2 of the regulatory pyruvate, thus forming a regulatory dyad. The influence of the other two histidines is less pronounced. Substrate activation is slightly weakened for A287G and significantly retarded for S311A., QC 20170607

Viral capsids ensure viral genome integrity by protecting the enclosed nucleic acids. Interactions between the genome and capsid and between individual capsid proteins (i.e., capsid architecture) are intimate and are expected to be characterized by strong evolutionary conservation. For this reason, a capsid structure-based viral classification has been proposed as a way to bring order to the viral universe. The seeming lack of sufficient sequence similarity to reproduce this classification has made it difficult to reject structural convergence as the basis for the classification. We reinvestigate whether the structure-based classification for viral coat proteins making icosahedral virus capsids is in fact supported by previously undetected sequence similarity. Since codon choices can influence nascent protein folding cotranslationally, we searched for both amino acid and nucleotide sequence similarity. To demonstrate the sensitivity of the approach, we identify a candidate gene for the pandoravirus capsid protein. We show that the structure-based classification is strongly supported by amino acid and also nucleotide sequence similarities, suggesting that the similarities are due to common descent. The correspondence between structure-based and sequence-based analyses of the same proteins shown here allow them to be used in future analyses of the relationship between linear sequence information and macromolecular function, as well as between linear sequence and protein folds. IMPORTANCE Viral capsids protect nucleic acid genomes, which in turn encode capsid proteins. This tight coupling of protein shell and nucleic acids, together with strong functional constraints on capsid protein folding and architecture, leads to the hypothesis that capsid protein-coding nucleotide sequences may retain signatures of ancient viral evolution. We have been able to show that this is indeed the case, using the major capsid proteins of viruses forming icosahedral capsids. Importantly

Protein evolution at the sequence level confers more subtle changes at the secondary and tertiary structure levels. These types of molecular information are crucial to reveal the evolutionary history of proteins. Phylogenetic trees are generally constructed by comparison of DNA, RNA or protein sequences, a process which is very effective, particularly on similar sequences. Sequence-based phylogenies are widely used to infer evolutionary relationships and as sequences are one-dimensional data, they have the advantage of being easy to handle. However comparison of homologous proteins with no apparent sequence similarities can result in polytomies - regions of the tree where branch order cannot be distinguished. As protein structures are more highly conserved than their sequences, protein structures are therefore an untapped source of potentially deep evolutionary information. For this reason, both types of information may be combined to infer potentially improved phylogenies. In this thesis, we aim to develop methods to integrate both protein sequence and structural information for a better understanding of the evolutionary history of proteins. Firstly, we identify a method suitable for inferring protein similarity from structures for the purpose of developing structural phylogeny. Different structure-based alignment methods for evaluating structural similarity of proteins are examined. A comparison between structure-based phylogenetic trees and the SCOP classification is performed to measure how well the methods reflect the manual protein classification. Secondly, we evaluate both sequence-based and structure-based phylogeny methods for a better understanding of protein family evolution. We demonstrate that the trees inferred from sequence and structural information have their own strength and limitations. Neither sequence nor structural data alone produced a tree of similar quality. During the evaluation of trees, different aspects of proteins similarity are covered

Enzyme catalysis relies on conformational plasticity, but structural information on transient intermediates is difficult to obtain. We show that the three-dimensional (3D) structure of an unstable, low-abundance enzymatic intermediate can be determined by nuclear magnetic resonance (NMR) spectroscopy. The approach is demonstrated for Staphylococcus aureus sortase A (SrtA), which is an established drug target and biotechnological reagent. SrtA is a transpeptidase that converts an amide bond of a substrate peptide into a thioester. By measuring pseudocontact shifts (PCSs) generated by a site-specific cysteine-reactive paramagnetic tag that does not react with the active-site residue Cys184, a sufficient number of restraints were collected to determine the 3D structure of the unstable thioester intermediate of SrtA that is present only as a minor species under non-equilibrium conditions. The 3D structure reveals structural changes that protect the thioester intermediate against hydrolysis.

For more than a decade, the Joint Center for Structural Genomics (JCSG; www.jcsg.org) worked toward increased three-dimensional structure coverage of the protein universe. This coordinated quest was one of the main goals of the four high-throughput (HT) structure determination centers of the Protein Structure Initiative (PSI; www.nigms.nih.gov/Research/specificareas/PSI). To achieve the goals of the PSI, the JCSG made use of the complementarity of structure determination by X-ray crystallography and nuclear magnetic resonance (NMR) spectroscopy to increase and diversify the range of targets entering the HT structure determination pipeline. The overall strategy, for both techniques, was to determine atomic resolution structures for representatives of large protein families, as defined by the Pfam database, which had no structural coverage and could make significant contributions to biological and biomedical research. Furthermore, the experimental structures could be leveraged by homology modeling to further expand the structural coverage of the protein universe and increase biological insights. Here, we describe what could be achieved by this structural genomics approach, using as an illustration the contributions from 20 NMR structure determinations out of a total of 98 JCSG NMR structures, which were selected because they are the first three-dimensional structure representations of the respective Pfam protein families. The information from this small sample is representative for the overall results from crystal and NMR structure determination in the JCSG. There are five new folds, which were classified as domains of unknown functions (DUF), three of the proteins could be functionally annotated based on three-dimensional structure similarity with previously characterized proteins, and 12 proteins showed only limited similarity with previous deposits in the Protein Data Bank (PDB) and were classified as DUFs.

Mutations in the protein transthyretin can cause as well as protect individuals from transthyretin amyloidosis, an incurable fatal inherited disease. Little is known, however, about the structural basis of pathogenic and clinically protective transthyretin mutants. Here we determined the solution structure of a transthyretin monomer that carries the clinically important T119M mutation. The structure displays a non-native arrangement that is distinct from all known structures of transthyretin and highlights the importance of high-resolution studies in solution for understanding molecular processes that lead to amyloid diseases.

For enzyme activity, an exact structural and motional orchestration of the active site and its surroundings is believed to be key. In order to reveal such possible phenomena at atomic resolution on the basis of experimental evidence, an experimental restraint driven two-state ensemble of the prototypical enzyme cyclophilin was determined by using a recently introduced exact NOE approach. The ensemble description reveals the presence of an open and a closed state of cyclophilin, which is indicative of large-scale correlated motion. In the open state, the catalytic site is preorganized for catalysis, thus suggesting the mechanism of action to be conformational sampling, while the ligand-binding loop appears to act through an induced fit mechanism. This finding is supported by affinity measurements of a cyclophilin designed to be more open. Overall, more than 60-70 % of the side-chain conformations of cyclophilin appear to be correlated.

CalE6 is a previously uncharacterized protein involved in the biosynthesis of calicheamicins in Micromonospora echinospora. It is a pyridoxal-5'-phosphate-dependent enzyme and exhibits high sequence homology to cystathionine gamma-lyases and cystathionine g-synthases. However, it was found to be active towards methionine and to convert this amino acid into a-ketobutyrate, ammonium, and methanethiol. The crystal structure of the cofactor-bound holoenzyme was resolved at 2.0 (A) over cap; it contains two active site residues, Gly105 and Val322, specific for methionine gamma-lyases. Modeling of methionine into the active site allows identification of the active site residues responsible for substrate recognition and catalysis. These findings support that CalE6 is a putative methionine gamma-lyase producing methanethiol as a building block in biosynthesis of calicheamicins.

Peroxisome-proliferator activated receptor-γ (PPARγ) is a nuclear hormone receptor that forms a heterodimeric complex with retinoid X receptor-α (RXRα) to regulate transcription of genes involved in fatty acid storage and glucose metabolism. PPARγ is a target for pharmaceutical intervention in type 2 diabetes, and insight into interactions between PPARγ, RXRα, and DNA is of interest in understanding the function and regulation of this complex. Phosphorylation of PPARγ by cyclin-dependent kinase 5 (Cdk5) has been shown to dysregulate the expression of metabolic regulation genes, an effect that is counteracted by PPARγ ligands. We applied molecular dynamics (MD) simulations to study the relationship between the ligand-binding domains of PPARγ and RXRα with their respective DNA-binding domains. Our results reveal that phosphorylation alters collective motions within the PPARγ-RXRα complex that affect the LBD-LBD dimerization interface and the AF-2 coactivator binding region of PPARγ.

Peroxisome-proliferator activated receptor-γ (PPARγ) is a nuclear hormone receptor that forms a heterodimeric complex with retinoid X receptor-α (RXRα) to regulate transcription of genes involved in fatty acid storage and glucose metabolism. PPARγ is a target for pharmaceutical intervention in type 2 diabetes, and insight into interactions between PPARγ, RXRα, and DNA is of interest in understanding the function and regulation of this complex. Phosphorylation of PPARγ by cyclin-dependent kinase 5 (Cdk5) has been shown to dysregulate the expression of metabolic regulation genes, an effect that is counteracted by PPARγ ligands. We applied molecular dynamics (MD) simulations to study the relationship between the ligand-binding domains of PPARγ and RXRα with their respective DNA-binding domains. Our results reveal that phosphorylation alters collective motions within the PPARγ-RXRα complex that affect the LBD-LBD dimerization interface and the AF-2 coactivator binding region of PPARγ.

CalE6 is a previously uncharacterized protein involved in the biosynthesis of calicheamicins in Micromonospora echinospora. It is a pyridoxal-5'-phosphate-dependent enzyme and exhibits high sequence homology to cystathionine gamma-lyases and cystathionine g-synthases. However, it was found to be active towards methionine and to convert this amino acid into a-ketobutyrate, ammonium, and methanethiol. The crystal structure of the cofactor-bound holoenzyme was resolved at 2.0 (A) over cap; it contains two active site residues, Gly105 and Val322, specific for methionine gamma-lyases. Modeling of methionine into the active site allows identification of the active site residues responsible for substrate recognition and catalysis. These findings support that CalE6 is a putative methionine gamma-lyase producing methanethiol as a building block in biosynthesis of calicheamicins.

CalE6 is a previously uncharacterized protein involved in the biosynthesis of calicheamicins in Micromonospora echinospora. It is a pyridoxal-5'-phosphate-dependent enzyme and exhibits high sequence homology to cystathionine gamma-lyases and cystathionine g-synthases. However, it was found to be active towards methionine and to convert this amino acid into a-ketobutyrate, ammonium, and methanethiol. The crystal structure of the cofactor-bound holoenzyme was resolved at 2.0 (A) over cap; it contains two active site residues, Gly105 and Val322, specific for methionine gamma-lyases. Modeling of methionine into the active site allows identification of the active site residues responsible for substrate recognition and catalysis. These findings support that CalE6 is a putative methionine gamma-lyase producing methanethiol as a building block in biosynthesis of calicheamicins.

The cytotoxic complex formed between alpha-lactalbumin and oleic acid (OA) has inspired many studies on protein-fatty acid complexes, but structural insight remains sparse. After having used small-angle X-ray scattering (SAXS) to obtain structural information, we present a new, generic structural model of cytotoxic protein-oleic acid complexes, which we have termed liprotides (lipids and partially denatured proteins). Twelve liprotides formed from seven structurally unrelated proteins and prepared by different procedures all displayed core-shell structures, each with a micellar OA core and a shell consisting of flexible, partially unfolded protein, which stabilizes the OA micelle. The common structure explains similar effects exerted on cells by different liprotides and is consistent with a cargo off-loading of the OA into cell membranes.

The rapid growth of the number of protein sequences that can be inferred from sequenced genomes presents challenges for function assignment, because only a small fraction (currently <1%) has been experimentally characterized. Bioinformatics tools are commonly used to predict functions of uncharacterized proteins. Recently, there has been significant progress in using protein structures as an additional source of information to infer aspects of enzyme function, which is the focus of this review. Successful application of these approaches has led to the identification of novel metabolites, enzyme activities, and biochemical pathways. We discuss opportunities to elucidate systematically protein domains of unknown function, orphan enzyme activities, dead-end metabolites, and pathways in secondary metabolism.

Peanut allergens can trigger a potent and sometimes dangerous immune response in an increasing number of people. The molecular structures of these allergens form the basis for understanding this response. This review describes the currently known peanut allergen structures and discusses how modifications both enzymatic and non-enzymatic affect digestion, innate immune recognition, and IgE interactions. The allergen structures help explain cross-reactivity among allergens from different sources, which is useful in improving patient diagnostics. Surprisingly, it was recently noted that similar short peptide sequences among unrelated peanut allergens could also be a source of cross-reactivity. The molecular features of peanut allergens continue to inform predictions and provide new research directions in the study of allergic disease.

We present a method based on the maximum entropy principle that can re-weight an ensemble of protein structures based on data from residual dipolar couplings (RDCs). The RDCs of intrinsically disordered proteins (IDPs) provide information on the secondary structure elements present in an ensemble; however even two sets of RDCs are not enough to fully determine the distribution of conformations, and the force field used to generate the structures has a pervasive influence on the refined ensemble. Two physics-based coarse-grained force fields, Profasi and Campari, are able to predict the secondary structure elements present in an IDP, but even after including the RDC data, the re-weighted ensembles differ between both force fields. Thus the spread of IDP ensembles highlights the need for better force fields. We distribute our algorithm in an open-source Python code.

© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. Molecular plasticity controls enzymatic activity: the native fold of a protein in a given environment is normally unique and at a global free-energy minimum. Some proteins, however, spontaneously undergo substantial fold switching to reversibly transit between defined conformers, the >metamorphic> proteins. Here, we present a minimal metamorphic, selective, and specific caseinolytic metallopeptidase, selecase, which reversibly transits between several different states of defined three-dimensional structure, which are associated with loss of enzymatic activity due to autoinhibition. The latter is triggered by sequestering the competent conformation in incompetent but structured dimers, tetramers, and octamers. This system, which is compatible with a discrete multifunnel energy landscape, affords a switch that provides a reversible mechanism of control of catalytic activity unique in nature. Shape shifting: A minimal metamorphic, selective, and specific caseinolytic metallopeptidase, selecase, reversibly transits between several different states of defined three-dimensional structure (monomer and tetramer represented in picture). The competent conformation is sequestered in incompetent but structured dimers, tetramers, and octamers, which are associated with loss of enzymatic activity due to autoinhibition.

QC 20130207

AuthorCount:10

Background Predicting protein subnuclear localization is a challenging problem. Some previous works based on non-sequence information including Gene Ontology annotations and kernel fusion have respective limitations. The aim of this work is twofold: one is to propose a novel individual feature extraction method; another is to develop an ensemble method to improve prediction performance using comprehensive information represented in the form of high dimensional feature vector obtained by 11 feature extraction methods. Methodology/Principal Findings A novel two-stage multiclass support vector machine is proposed to predict protein subnuclear localizations. It only considers those feature extraction methods based on amino acid classifications and physicochemical properties. In order to speed up our system, an automatic search method for the kernel parameter is used. The prediction performance of our method is evaluated on four datasets: Lei dataset, multi-localization dataset, SNL9 dataset and a new independent dataset. The overall accuracy of prediction for 6 localizations on Lei dataset is 75.2% and that for 9 localizations on SNL9 dataset is 72.1% in the leave-one-out cross validation, 71.7% for the multi-localization dataset and 69.8% for the new independent dataset, respectively. Comparisons with those existing methods show that our method performs better for both single-localization and multi-localization proteins and achieves more balanced sensitivities and specificities on large-size and small-size subcellular localizations. The overall accuracy improvements are 4.0% and 4.7% for single-localization proteins and 6.5% for multi-localization proteins. The reliability and stability of our classification model are further confirmed by permutation analysis. Conclusions It can be concluded that our method is effective and valuable for predicting protein subnuclear localizations. A web server has been designed to implement the proposed method. It is freely available a

Allosteric interactions of the molecular chaperone Hsp90 with a large cohort of cochaperones and client proteins allow for molecular communication and event coupling in signal transduction networks. The integration of cochaperones into the Hsp90 system is driven by the regulatory mechanisms that modulate the progression of the ATPase cycle and control the recruitment of the Hsp90 clientele. In this work, we report the results of computational modeling of allosteric regulation in the Hsp90 complexes with the cochaperones p23 and Aha1. By integrating protein docking, biophysical simulations, modeling of allosteric communications, protein structure network analysis and the energy landscape theory we have investigated dynamics and stability of the Hsp90-p23 and Hsp90-Aha1 interactions in direct comparison with the extensive body of structural and functional experiments. The results have revealed that functional dynamics and allosteric interactions of Hsp90 can be selectively modulated by these cochaperones via specific targeting of the regulatory hinge regions that could restrict collective motions and stabilize specific chaperone conformations. The protein structure network parameters have quantified the effects of cochaperones on conformational stability of the Hsp90 complexes and identified dynamically stable communities of residues that can contribute to the strengthening of allosteric interactions. According to our results, p23-mediated changes in the Hsp90 interactions may provide "molecular brakes'' that could slow down an efficient transmission of the inter-domain allosteric signals, consistent with the functional role of p23 in partially inhibiting the ATPase cycle. Unlike p23, Aha1-mediated acceleration of the Hsp90-ATPase cycle may be achieved via modulation of the equilibrium motions that facilitate allosteric changes favoring a closed dimerized form of Hsp90. The results of our study have shown that Aha1 and p23 can modulate the Hsp90-ATPase activity and di

Background: Employing methods to assess the quality of modeled protein structures is now standard practice in bioinformatics. In a broad sense, the techniques can be divided into methods relying on consensus prediction on the one hand, and single-model methods on the other. Consensus methods frequently perform very well when there is a clear consensus, but this is not always the case. In particular, they frequently fail in selecting the best possible model in the hard cases (lacking consensus) or in the easy cases where models are very similar. In contrast, single-model methods do not suffer from these drawbacks and could potentially be applied on any protein of interest to assess quality or as a scoring function for sampling-based refinement.Results: Here, we present a new single-model method, ProQ2, based on ideas from its predecessor, ProQ. ProQ2 is a model quality assessment algorithm that uses support vector machines to predict local as well as global quality of protein models. Improved performance is obtained by combining previously used features with updated structural and predicted features. The most important contribution can be attributed to the use of profile weighting of the residue specific features and the use features averaged over the whole model even though the prediction is still local.Conclusions: ProQ2 is significantly better than its predecessors at detecting high quality models, improving the sum of Z-scores for the selected first-ranked models by 20% and 32% compared to the second-best single-model method in CASP8 and CASP9, respectively. The absolute quality assessment of the models at both local and global level is also improved. The Pearson's correlation between the correct and local predicted score is improved from 0.59 to 0.70 on CASP8 and from 0.62 to 0.68 on CASP9; for global score to the correct GDT_TS from 0.75 to 0.80 and from 0.77 to 0.80 again compared to the second-best single methods in CASP8 and CASP9, respectively. ProQ2 is ava, QC 20131016