PDB Structures Research Articles

Model building of protein complexes from intermediate-resolution cryo-EM maps with deep learning-guided automatic assembly

Advances in microscopy instruments and image processing algorithms have led to an increasing number of cryo-electron microscopy (cryo-EM) maps. However, building accurate models into intermediate-resolution EM maps remains challenging and labor-intensive. Here, we propose an automatic model building method of multi-chain protein complexes from intermediate-resolution cryo-EM maps, named EMBuild, by integrating AlphaFold structure prediction, FFT-based global fitting, domain-based semi-flexible refinement, and graph-based iterative assembling on the main-chain probability map predicted by a deep convolutional network. EMBuild is extensively evaluated on diverse test sets of 47 single-particle EM maps at 4.0–8.0 Å resolution and 16 subtomogram averaging maps of cryo-ET data at 3.7–9.3 Å resolution, and compared with state-of-the-art approaches. We demonstrate that EMBuild is able to build high-quality complex structures that are comparably accurate to the manually built PDB structures from the cryo-EM maps. These results demonstrate the accuracy and reliability of EMBuild in automatic model building.

Assessment of AlphaFold2 for Human Proteins via Residue Solvent Exposure.

As only 35% of human proteins feature (often partial) PDB structures, the protein structure prediction tool AlphaFold2 (AF2) could have massive impact on human biology and medicine fields, making independent benchmarks of interest. We studied AF2's ability to describe the backbone solvent exposure as a functionally important and easily interpretable "natural coordinate" of protein conformation, using human proteins as test case. After screening for appropriate comparative sets, we matched 1818 human proteins predicted by AF2 against 7585 unique experimental PDBs, and after curation for sequence overlap, we assessed 1264 comparative pairs comprising 115 unique AF2 structures and 652 unique experimental structures. AF2 performed markedly worse for multimers, whereas ligands, cofactors, and experimental resolution were interestingly not very important for performance. AF2 performed excellently for monomer proteins. Challenges relating to specific groups of residues and multimers were analyzed. We identified larger deviations for lower-confidence scores (pLDDT), and exposed residues and polar residues (e.g., Asp, Glu, Asn) being less accurately described than hydrophobic residues. Proline conformations were the hardest to predict, probably due to a common location in dynamic solvent-accessible parts. In summary, using solvent exposure as a metric, we quantified the performance of AF2 for human proteins and provided estimates of the expected agreement as a function of ligand presence, multimer/monomer status, local residue solvent exposure, pLDDT, and amino acid type. Overall performance was found to be excellent.

In silico molecular docking for assessing anti-fungal competency of hydroxychavicol, a phenolic compound of betel leaf (Piper betle L.) against COVID-19 associated maiming mycotic infections

Objective To investigate the inhibitory nature of hydroxychavicol against the COVID-19 associated mycotic infections, the present in silico study was performed in hydroxychavicol with the target Lanosterol 14 alpha demethylase and its competency was compared with four approved anti-fungal drugs. Significance The corona virus pandemic has drawn stark lines between rich nations and poor, and the occurrence of COVID-19 associated mycotic infections, mucormycosis epidemic stands as the latest manifestation. The increase in resistance in known fungal pathogens to the available anti-fungal drugs and side effects are the important demands that forced to search anti-fungal compounds from medicinal plants as therapeutic alternatives. During the fishing expedition, Piper betle L., gets tremendous attention for its rich source of medicinally important compounds. Among them, hydroxychavicol has the enormous supportive records against microbial growth. Methods Hydroxychavicol and the chosen drugs were retrieved from the Pubchem database and subjected to ADME analysis. The structure of the target of the chosen COVID-19 associated fungal pathogens was retrieved from PDB and unavailable protein structures were modeled using the Swiss Model and validated. Virtual screening (PyRx version 0.8) was performed and the interactions were visualized using BIOVIA Discovery Studio. Results ADME screening of hydroxychavicol was found to have clear reciprocity with the drug-likeliness nature and the subsequent molecular docking study revealed its good binding affinity toward the target protein suggesting its inhibitory nature. Conclusion This study offers the possibility of making use of the suppressive nature of hydroxychavicol in the treatment of mycotic infections either exclusively/in synergistic approach.

Searching for New Z-DNA/Z-RNA Binding Proteins Based on Structural Similarity to Experimentally Validated Zα Domain.

Z-DNA and Z-RNA are functionally important left-handed structures of nucleic acids, which play a significant role in several molecular and biological processes including DNA replication, gene expression regulation and viral nucleic acid sensing. Most proteins that have been proven to interact with Z-DNA/Z-RNA contain the so-called Zα domain, which is structurally well conserved. To date, only eight proteins with Zα domain have been described within a few organisms (including human, mouse, Danio rerio, Trypanosoma brucei and some viruses). Therefore, this paper aimed to search for new Z-DNA/Z-RNA binding proteins in the complete PDB structures database and from the AlphaFold2 protein models. A structure-based similarity search found 14 proteins with highly similar Zα domain structure in experimentally-defined proteins and 185 proteins with a putative Zα domain using the AlphaFold2 models. Structure-based alignment and molecular docking confirmed high functional conservation of amino acids involved in Z-DNA/Z-RNA, suggesting that Z-DNA/Z-RNA recognition may play an important role in a variety of cellular processes.

A computational methodology to diagnose sequence-variant dynamic perturbations by comparing atomic protein structures.

MotivationThe objective is to diagnose dynamics perturbations caused by amino acid mutations as prerequisite to assess protein functional health or drug failure, simply using network models of protein X-ray structures.ResultsWe find that the differences in the allocation of the atomic interactions of each amino acid to 1D, 2D, 3D, 4D structural levels between variants structurally robust, recover experimental dynamic perturbations. The allocation measure validated on two B-pentamers variants of AB5 toxins having 17 mutations, also distinguishes dynamic perturbations of pathogenic and non-pathogenic Transthyretin single-mutants. Finally, the main proteases of the coronaviruses SARS-CoV and SARS-CoV-2 exhibit changes in the allocation measure, raising the possibility of drug failure despite the main proteases structural similarity.AvailabilityThe Python code used for the production of the results is available at github.com/lorpac/protein_partitioning_atomic_contacts. The authors will run the analysis on any PDB structures of protein variants upon request.Supplementary informationSupplementary data are available at Bioinformatics online.

RR3DD: an RNA global structure-based RNA three-dimensional structural classification database

ABSTRACT The three-dimensional (3D) structure of RNA usually plays an important role in the recognition with RNA-binding protein. Along with the discovering of RNAs, several RNA databases are developed to study the functions of RNA based on sequence, secondary structure, local 3D structural motif and global structure. Based on RNA function and structure, different RNAs are classified and stored in SCOR and DARTS, respectively. The classification of RNA structures is useful in RNA structure prediction and function annotation. However, the SCOR and DARTS are not updated any more. In this study, we present an RNA classification database RR3DD based on RNA fold with the global 3D structural similarity. The RR3DD includes 13,601 RNA chains from PDB and mmCIF format structures which are classified into 780 RNA folds. The RNA chains from PDB and mmCIF format structures are aligned and clustered into 675 and 220 RNA folds, respectively. By analysing the RNA structure in RR3DD, we find that there are 11 clusters with more than 50 members. These clusters include rRNAs, riboswitches, tRNAs and so on. By mapping RR3DD into Rfam, we found that some RNAs without annotation by Rfam can be annotated through structural alignment. For example, we analysed tRNAs and found that tRNA were successfully grouped in RR3DD for which Rfam did not classify them into one family. Finally, we provide a web interface of RR3DD offering functions of browsing RR3DD, annotating RNA 3D structure and finding templates for RNA homology modelling.

Changes in Protein Structural Motifs upon Post-Translational Modification in Kidney Cancer.

Post-translational modification (PTM) leads to conformational changes in protein structure, modulates the biological function of proteins, and, consequently, changes the signature of metabolic transformations and the immune response in the body. Common PTMs are reversible and serve as a mechanism for modulating metabolic trans-formations in cells. It is likely that dysregulation of post-translational cellular signaling leads to abnormal proliferation and oncogenesis. We examined protein PTMs in the blood samples from patients with kidney cancer. Conformational changes in proteins after modification were analyzed. The proteins were analyzed using ultra-high resolution HPLC-MS/MS and structural analysis was performed with the AMBER and GROMACS software packages. Fifteen proteins containing PTMs were identified in blood samples from patients with kidney cancer. For proteins with PDB structures, a comparative analysis of the structural changes accompanying the modifications was performed. Results revealed that PTMs are localized in stable and compact space protein globule motifs that are exposed to a solvent. The phenomenon of modification is accompanied, as a rule, by an increase in the area available for the solvent of the modified amino acid residue and its active environment.

Fifty Years of Open Access to PDB Structures

Fifty Years of Open Access to PDB Structures

AlphaFold 2: Why It Works and Its Implications for Understanding the Relationships of Protein Sequence, Structure, and Function.

AlphaFold 2 (AF2) was the star of CASP14, the last biannual structure prediction experiment. Using novel deep learning, AF2 predicted the structures of many difficult protein targets at or near experimental resolution. Here, we present our perspective of why AF2 works and show that it is a very sophisticated fold recognition algorithm that exploits the completeness of the library of single domain PDB structures. It has also learned local side chain packing rearrangements that enable it to refine proteins to high resolution. The benefits and limitations of its ability to predict the structures of many more proteins at or close to atomic detail are discussed.

Support Vector Machine as a Supervised Learning for the Prioritization of Novel Potential SARS-CoV-2 Main Protease Inhibitors.

In the last year, the COVID-19 pandemic has highly affected the lifestyle of the world population, encouraging the scientific community towards a great effort on studying the infection molecular mechanisms. Several vaccine formulations are nowadays available and helping to reach immunity. Nevertheless, there is a growing interest towards the development of novel anti-covid drugs. In this scenario, the main protease (Mpro) represents an appealing target, being the enzyme responsible for the cleavage of polypeptides during the viral genome transcription. With the aim of sharing new insights for the design of novel Mpro inhibitors, our research group developed a machine learning approach using the support vector machine (SVM) classification. Starting from a dataset of two million commercially available compounds, the model was able to classify two hundred novel chemo-types as potentially active against the viral protease. The compounds labelled as actives by SVM were next evaluated through consensus docking studies on two PDB structures and their binding mode was compared to well-known protease inhibitors. The best five compounds selected by consensus docking were then submitted to molecular dynamics to deepen binding interactions stability. Of note, the compounds selected via SVM retrieved all the most important interactions known in the literature.

Abstract P-46: Structure of A. Baumannii Phage Tapaz, Revealed with Cryo-Electron Microscopy

Background: Acinetobacter baumannii is an opportunistic pathogen and one of the six most important multidrug resistant microorganisms in hospitals worldwide. Some of its strains are resistant to most of the antibiotics, A. baumannii is included into the Priority 1 part of Global Priority List of Antibiotic-resistant Bacteria. Phage therapy is considered to be an alternative strategy to antibiotic treatments. Methods: A. baumannii strain NIPH601 cells were grown till OD6000.4 and infected with the phage at MOI 10:1. After complete lysis took place cell debris was spined down and phage particles were precipitated with the PEG6000 (final concentration 10% PEG 6000, 0.5 NaCl). Virus particles were collected by centrifugation, resuspended at SM buffer and applied on CsCl step gradient. Gradient was spinned down for 2 hours at 40000g and the fraction containing phage particles was collected and dialyzed against SM buffer. Purified phage particles were applied to Quantifoil 1.2/1.3 grids and plunge-froze in Vitrobot Mark IV (TFS) Micrographs were collected in HKU, Shenzhen campus with Titan Krios cryoelectron microscope (TFS), equipped with Gatan K3 direct electron detector. The micrographs were acquired with 1.06 Å pixel size and 1.5 um average defocus value in counting mode with 50 frames and 1.2 e/Å2/frame dose rate. All image processing was performed with Relion3.0 software, except for the particle picking step performed with cryolo. Results: Lytic A. baumannii phage TaPaz belongs to the family Myoviridae. BLAST search over NCBI “nr” (non-redundant) database revealed close homology with previously published sequences of Acinetobacter phage vB_AbaM_B9 and Acinetobacter phage BS46. However, no structural information about any homologous proteins was found among the PDB structures. The cryo-EM map was reconstructed with single particle analysis independently for the capsid, tail and baseplate regions. The capsid was reconstructed at 3.9 Å resolution with I3 symmetry applied (Fig. 1A). The baseplate region of the phage was reconstructed at 3.5 Å resolution with C3 symmetry (Fig. 1B). The tail region was reconstructed at 2.6 Å resolution with helical symmetry (Rise 36.4 Å, Twist 25.7 deg). Initial atomic model for the tail region was built from sequence with Deeptracer and was further refined in coot (Fig. 1C). Conclusion: We successfully obtained the near-atomic resolution structural map of phage TaPaz. The data obtained contribute to enhancing knowledge of structural diversity of bacterial viruses infecting A. baumannii.

Spirostanol Sapogenins and Saponins from Convallaria majalis L. Structural Characterization by 2D NMR, Theoretical GIAO DFT Calculations and Molecular Modeling.

Two new spirostanol sapogenins (5β-spirost-25(27)-en-1β,2β,3β,5β-tetrol 3 and its 25,27-dihydro derivative, (25S)-spirostan-1β,2β,3β,5β-tetrol 4) and four new saponins were isolated from the roots and rhizomes of Convallaria majalis L. together with known sapogenins (isolated from Liliaceae): 5β-spirost-25(27)-en-1β,3β-diol 1, (25S)-spirostan-1β,3β-diol 2, 5β-spirost-25(27)-en-1β,3β,4β,5β-tetrol 5, (25S)-spirostan-1β,3β,4β,5β-tetrol 6, 5β-spirost-25(27)-en-1β,2β,3β,4β,5β-pentol 7 and (25S)-spirostan-1β,2β,3β,4β,5β-pentol 8. New steroidal saponins were found to be pentahydroxy 5-O-glycosides; 5β-spirost-25(27)-en-1β,2β,3β,4β,5β-pentol 5-O-β-galactopyranoside 9, 5β-spirost-25(27)-en-1β,2β,3β,4β,5β-pentol 5-O-β-arabinonoside 11, 5β-(25S)-spirostan-1β,2β,3β,4β,5β-pentol 5-O-galactoside 10 and 5β-(25S)-spirostan-1β,2β,3β,4β,5β-pentol 5-O-arabinoside 12 were isolated for the first time. The structures of those compounds were determined by NMR spectroscopy, including 2D COSY, HMBC, HSQC, NOESY, ROESY experiments, theoretical calculations of shielding constants by GIAO DFT, and mass spectrometry (FAB/LSI HR MS). An attempt was made to test biological activity, particularly as potential chemotherapeutic agents, using in silico methods. A set of 12 compounds was docked to the PDB structures of HER2 receptor and tubulin. The results indicated that diols have a higher affinity to the analyzed targets than tetrols and pentols. Two compounds (25S)-spirosten-1β,3β-diol 1 and 5β-spirost-25(27)-en-1β,2β,3β,4β,5β-pentol 5-O-galactoside 9 were selected for further evaluation of biological activity.

Region and Country Trends in the Determination of Coronavirus PDB Protein Structures

The Covid-19 pandemic provides a nearly unique opportunity to study the international dimensions of scientific research. This is a project that was conducted with undergraduates who used this opportunity to acquire essential research experiences that otherwise are difficult to acquire, given that college teaching laboratories are shut down during the pandemic. We describe an investigation of the national and regional publication patterns of coronavirus protein database (pdb) structures, beginning when the first coronavirus virion pdb structures were published in 2002 to the present. This period comprises three separate coronavirus-induced pandemics, first the Severe Acute Respiratory Syndrome (SARS) beginning in 2003, then the Middle East Respiratory Syndrome (MERS) beginning in 2012, then the current Covid-19 pandemic beginning in 2019. Each of these pandemics is reflected the deposition of pdb structure depositions from a variety of countries. This study investigates the sources of these depositions and relates them to demographic and economic metrics. The United States, the UK, and China lead the world in the number of depositions. Researchers in some countries deposit many more pdb structures than their country metrics would suggest.

Determining Interactions Between Leucine 320 and BHPI To Identify a Possible Binding Site On Estrogen Receptor Alpha

Excluding lung cancer, breast cancer death rates among women are higher than any other cancer. BHPI, a small molecule bio-modulator of estrogen receptor (ERα), could be vital in developing life-saving treatment for people living with breast cancer. BHPI activates an endoplasmic reticulum (EnR) stress-response pathway that maintains protein folding homeostasis and quality control. This eventually ends in necrotic cell death, initiated by strong and sustained activation of the anticipatory UPR, killing ERα positive cancer cells. BHPI possibly binds too tightly to ERα, so it has thus far been recalcitrant to crystallography and binding studies. In order to utilize BHPI in breast cancer treatment, it is important to identify its binding site on ERα. Three estrogen receptor PDB structures: 1a52, 3ert, and 5u2b, were studied through computational programs RefineD, PyMOL, and SwissDock. Interactions between a specific amino acid, Leucine 320, and BHPI were studied using these programs to identify and probe possible binding sites. BHPI and leucine appear to have hydrophobic interactions. Future studies are required to indicate additional interactions between Leucine 320 and BHPI.

P324A Point Mutation as a Probe to Determine the Binding Site of BHPI on Estrogen Receptor Alpha

BHPI (1,3-Dihydro-3,3-bis(4-hydroxyphenyl)-7-methyl-2H-indol-2-one) is a noncompetitive bio-modulator of estrogen receptor alpha (ERα) that induces necrotic cell death through ATP depletion and has shown promise as a preclinical anticancer drug. BHPI kills cancer cells through activating the anticipatory Unfolded Protein Response (aUPR). The UPR is a stress sensor within the endoplasmic reticulum that is normally a cytoprotective mechanism, but once hyperactivated by BHPI, leads to necrosis. Three ERα PDB structures 1a52, 3ert, and 5u2b were computationally analyzed with a Proline 324 to Alanine (P324A) point mutation to determine where and how BHPI binds to ERα. These studies are important for understanding the effects of amino acid mutations on protein-ligand interaction, as well as the binding affinity of ERα for BHPI.

An integrated deep learning and dynamic programming method for predicting tumor suppressor genes, oncogenes, and fusion from PDB structures

Mutations in proto-oncogenes (ONGO) and the loss of regulatory function of tumor suppression genes (TSG) are the common underlying mechanism for uncontrolled tumor growth. While cancer is a heterogeneous complex of distinct diseases, finding the potentiality of the genes related functionality to ONGO or TSG through computational studies can help develop drugs that target the disease. This paper proposes a classification method that starts with a preprocessing stage to extract the feature map sets from the input 3D protein structural information. The next stage is a deep convolutional neural network stage (DCNN) that outputs the probability of functional classification of genes. We explored and tested two approaches: in Approach 1, all filtered and cleaned 3D-protein-structures (PDB) are pooled together, whereas in Approach 2, the primary structures and their corresponding PDBs are separated according to the genes' primary structural information. Following the DCNN stage, a dynamic programming-based method is used to determine the final prediction of the primary structures' functionality. We validated our proposed method using the COSMIC online database. For the ONGO vs TSG classification problem the AUROC of the DCNN stage for Approach 1 and Approach 2 DCNN are 0.978 and 0.765, respectively. The AUROCs of the final genes' primary structure functionality classification for Approach 1 and Approach 2 are 0.989, and 0.879, respectively. For comparison, the current state-of-the-art reported AUROC is 0.924. Our results warrant further study to apply the deep learning models to humans’ (GRCh38) genes, for predicting their corresponding probabilities of functionality in the cancer drivers.

Functional and Therapeutic Relevance of Rho GTPases in Innate Immune Cell Migration and Function during Inflammation: An In Silico Perspective.

Systematic regulation of leukocyte migration to the site of infection is a vital step during immunological responses. Improper migration and localization of immune cells could be associated with disease pathology as seen in systemic inflammation. Rho GTPases act as molecular switches during inflammatory cell migration by cycling between Rho-GDP (inactive) to Rho-GTP (active) forms and play an essential role in the precise regulation of actin cytoskeletal dynamics as well as other immunological functions of leukocytes. Available reports suggest that the dysregulation of Rho GTPase signaling is associated with various inflammatory diseases ranging from mild to life-threatening conditions. Therefore, it is crucial to understand the step-by-step activation and inactivation of GTPases and the functioning of different Guanine Nucleotide Exchange Factors (GEFs) and GTPase-Activating Proteins (GAPs) that regulate the conversion of GDP to GTP and GTP to GDP exchange reactions, respectively. Here, we describe the molecular organization and activation of various domains of crucial elements associated with the activation of Rho GTPases using solved PDB structures. We will also present the latest evidence available on the relevance of Rho GTPases in the migration and function of innate immune cells during inflammation. This knowledge will help scientists design promising drug candidates against the Rho-GTPase-centric regulatory molecules regulating inflammatory cell migration.

Magnesium Coordination of ATP - Force Field Evaluation and Structure-Driven Correction

In the numerous molecular recognition and catalytic processes across biochemistry involving adenosine triphosphate (ATP), the common bioactive form is its magnesium chelate, ATP.Mg2+. In aqueous solution, two chelation geometries predominate, with Mg contacting either the terminal beta and gamma phosphate groups (bidentate) or all three (tridentate). These subforms are approximately isoenergetic, but separated by a high energy barrier. Force field-based atomistic simulation studies of this complex require an accurate representation of its structure and energetics. Here we focused on the energetics of ATP.Mg2+ coordination. With unbiased molecular dynamics simulation showing prohibitively slow interconversion, we devised an enhanced sampling scheme to calculate free energy differences between different Mg2+-phosphate configurations. We observed striking contradictions between Amber and CHARMM force field descriptions, most prominently in opposing predictions of the favoured coordination mode. Through further configurational free energy calculations, conducted against a diverse set of ATP.Mg2+-protein complex structures to supplement otherwise limited experimental data, we observed systematic biases for each force field. However, the force field calculations were strongly predictive of experimentally observed coordination modes, enabling additive corrections to the coordination free energy that deliver close agreement with experiment. Reassessing the applicability of the corrected force field descriptions of ATP.Mg2+ for biomolecular simulation, we observe that calculated properties broadly agree with experimental measurements of solution geometry and the distribution of ATP.Mg2+ structures found in the Protein Data Bank. However, while the corrected Amber force field appears to reproduce both bidentate and tridentate configurations, CHARMM displays an erroneous preference triphosphate overextension. This will affect the interpretation of simulations of bidentate ATP.Mg2+, which comprises the majority of PDB structures.

Biological halogen bonds in protein-ligand complexes: a combined QTAIM and NCIPlot study in four representative cases.

In this study, the PDB has been manually scrutinized by using a subset of all PDB entries containing organic iodinated ligands. Four structures exhibiting short IA halogen bonding (HaB) contacts (A stands for the σ-hole acceptor) have been selected and further analysed. In most hits, the sigma-hole acceptor corresponds to an O-atom of the amido group belonging to the protein backbone. In a minority of hits, the electron donors are O, S, Se or π-systems of the amino-acid side chains. A judicious selection of four PDB structures presenting all four types of HaB interactions (C-IA, A = O, S, Se, π) has been performed. For these selected structures, a comprehensive RI-MP2/def2-TZVP study has been carried out to evaluate the HaB energetically. Moreover, the interactions have been characterized by combining the quantum theory of "atoms-in-molecules" (QTAIM) and the noncovalent interaction plot (NCIPlot) and rationalized using the molecular electrostatic potential (MEP) surface.

Developing a Kinase-Specific Target Selection Method Using a Structure-Based Machine Learning Approach.

IntroductionDespite recent advances in the drug discovery field, developing selective kinase inhibitors remains a complicated issue for a number of reasons, one of which is that there are striking structural similarities in the ATP-binding pockets of kinases.ObjectiveTo address this problem, we have designed a machine learning model utilizing various structure-based and energy-based descriptors to better characterize protein–ligand interactions.MethodsIn this work, we use a dataset of 104 human kinases with available PDB structures and experimental activity data against 1202 small-molecule compounds from the PubChem BioAssay dataset “Navigating the Kinome”. We propose structure-based interaction descriptors to build activity predicting machine learning model.Results and DiscussionWe report a ligand-oriented computational method for accurate kinase target prioritizing. Our method shows high accuracy compared to similar structure-based activity prediction methods, and more importantly shows the same prediction accuracy when tested on the special set of structurally remote compounds, showing that it is unbiased to ligand structural similarity in the training set data. We hope that our approach will be useful for the development of novel highly selective kinase inhibitors.