Academic literature on the topic 'Quality assessment of protein-ligand crystal structures'

Our understanding of the structure–function relationships of biomolecules and thereby applying it to drug discovery programs are substantially dependent on the availability of the structural information of ligand–protein complexes. However, the correct interpretation of the electron density of a small molecule bound to a crystal structure of a macromolecule is not trivial. Our analysis involving quality assessment of ~0.28 million small molecule–protein binding site pairs derived from crystal structures corresponding to ~66,000 PDB entries indicates that the majority (65%) of the pairs might need little (54%) or no (11%) attention. Out of the remaining 35% of pairs that need attention, 11% of the pairs (including structures with high/moderate resolution) pose serious concerns. Unfortunately, most users of crystal structures lack the training to evaluate the quality of a crystal structure against its experimental data and, in general, rely on the resolution as a ‘gold standard’ quality metric. Our work aims to sensitize the non-crystallographers that resolution, which is a global quality metric, need not be an accurate indicator of local structural quality. In this article, we demonstrate the use of several freely available tools that quantify local structural quality and are easy to use from a non-crystallographer’s perspective. We further propose a few solutions for consideration by the scientific community to promote quality research in structural biology and applied areas.

The appearance at the end of 2019 of the new SARS-CoV-2 coronavirus led to an unprecedented response by the structural biology community, resulting in the rapid determination of many hundreds of structures of proteins encoded by the virus. As part of an effort to analyze and, if necessary, remediate these structures as deposited in the Protein Data Bank (PDB), this work presents a detailed analysis of 81 crystal structures of the main protease 3CLpro, an important target for the design of drugs against COVID-19. The structures of the unliganded enzyme and its complexes with a number of inhibitors were determined by multiple research groups using different experimental approaches and conditions; the resulting structures span 13 different polymorphs representing seven space groups. The structures of the enzyme itself, all determined by molecular replacement, are highly similar, with the exception of one polymorph with a different inter-domain orientation. However, a number of complexes with bound inhibitors were found to pose significant problems. Some of these could be traced to faulty definitions of geometrical restraints for ligands and to the general problem of a lack of such information in the PDB depositions. Several problems with ligand definition in the PDB itself were also noted. In several cases extensive corrections to the models were necessary to adhere to the evidence of the electron-density maps. Taken together, this analysis of a large number of structures of a single, medically important protein, all determined within less than a year using modern experimental tools, should be useful in future studies of other systems of high interest to the biomedical community.

Radiation damage remains one of the major limitations to accurate structure determination in protein crystallography (PX). Despite the use of cryo-cooling techniques, it is highly probable that a number of the structures deposited in the Protein Data Bank (PDB) have suffered substantial radiation damage as a result of the high flux densities of third generation synchrotron X-ray sources. Whereas the effects of global damage upon diffraction pattern reflection intensities are readily detectable, traditionally the (earlier onset) site-specific structural changes induced by radiation damage have proven difficult to identify within individual PX structures. More recently, however, development of theBDamagemetric has helped to address this problem.BDamageis a quantitative, per-atom metric identifies potential sites of specific damage by comparing the atomicB-factor values of atoms that occupy a similar local packing density environment in the structure. Building upon this past work, this article presents a program,RABDAM, to calculate theBDamagemetric for all selected atoms within any standard-format PDB or mmCIF file.RABDAMprovides several useful outputs to assess the extent of damage suffered by an input PX structure. This free and open-source software will allow assessment and improvement of the quality of PX structures both previously and newly deposited in the PDB.

e15036 Background: Cancer cells respond to increases in DNA damage by deploying their DNA damage response (DDR) pathways. We are building a platform for the discovery and development of target-specific DDR therapeutics, including small molecule inhibitors and targeted protein degradation warheads, founded on fragment- and structure-based drug discovery. Methods: Our DDR platform, which includes hit-to-lead, lead optimization and candidate selection, starts with hit generation from a new technology that uses high-throughput protein X-ray crystallography to directly screen compound libraries. Our hit generation process produces empirical evidence of direct target engagement. The elucidation of high-quality ligand-bound 3D structures reveals the location and pose of the ligand and details of the protein-ligand interactions. Thus we can predict the structure-function consequences of the hit molecule engagement, which sets the stage for rapid assessment of synthetic tractability and intellectual property. After hit identification, we apply a multi-pronged approach in hit-to-lead conversion and lead optimization using iterative biophysical and biochemical assays, coupled with crystallography. We are applying our approach to several new targets in DDR and will present some early progress in this space. Results: Apurinic/apyrimidinic endonuclease 1 (APE1) is the major repair enzyme for abasic sites in DNA and contributes to DNA strand break processing. Many studies have associated increased APE1 levels with enhanced growth, migration, and drug resistance in human tumor cells, and with decreased patient survival. APE1 has been implicated in over 20 human cancers, including glioblastoma, making the protein an attractive target for the development of anticancer therapeutics. Despite intensive effort, there are no clinical endonuclease inhibitors of APE1. We have identified 25 diverse small molecule fragments that bind to APE1 at two distinct sites, including the endonuclease site. Pol eta (or PolH) is a DNA polymerase implicated, among other things, in the development of cisplatin resistance in a subset of ovarian cancers. In our quest to develop PolH inhibitors, we have identified 5 diverse fragments that bind to two distinct sites in the polymerase including a new potential allosteric site. Our results on APE1 and PolH represent the first known cases of crystal structures of small molecules bound to these proteins. Flap endonuclease (FEN1) is implicated in several cancers including for example ER/tamoxifen-resistant breast cancer. We are developing a targeted protein degradation approach using PROTACs (Proteolysis-Targeting Chimeras) toward the development of novel therapeutics against FEN1. Conclusions: Our results will help us develop small molecule inhibitors and targeted protein degradation against DDR targets that may be effective as single therapies or be used to make existing therapies more effective.

A grand challenge for structural biology is to efficiently inform macromolecular functions that involve flexible or unstructured regions and require multiple conformational states including membrane proteins and protein-nucleic acid complexes. Small angle X-ray scattering (SAXS) can probe at resolutions sufficient to distinguish conformational states, characterize flexible macromolecules, and screen in high-throughput under most solution conditions. However, methods for analyzing SAXS data and models have restricted progress. We are therefore developing SAXS methods that provide high-throughput, quantitative and superposition-independent evaluation of solution-state conformations and quantitatively define flexibility and disorder. The SIBYLS beamline provides hardware and software to integrate SAXS and crystal structure results. Our SAXS methods aim to improve crystallization and interpretation of crystal structures and NMR structure quality. We have invented a statistically robust method for assessing model-data agreements (chi-square free) akin to cross-validation. We also developed a metric and method for rapid quantitative and comprehensive assessment of molecular similarity suitable to examine functionally important conformational changes. To extend SAXS analysis to low concentrations and complex mixtures, we are developing SAXS with gold nano-crystal labels to enable examination of protein-induced DNA distortions along pathways key to the DNA repair, replication, transcription, and packaging. Collective results suggest SAXS can provide accurate shapes, assembly states, and comprehensive conformations of flexible complexes in solution that inform biology in fundamental ways.

Fascin is an actin-bundling protein overexpressed in various invasive metastatic carcinomas through promoting cell migration and invasion. Therefore, blocking Fascin binding sites is considered a vital target for antimetastatic drugs. This inspired us to find new Fascin binding site blockers. First, we built an active compound set by collecting reported small molecules binding to Fascin’s binding site 2. Consequently, a high-quality decoys set was generated employing DEKOIS 2.0 protocol to be applied in conducting the benchmarking analysis against the selected Fascin structures. Four docking programs, MOE, AutoDock Vina, VinaXB, and PLANTS were evaluated in the benchmarking study. All tools indicated better-than-random performance reflected by their pROC-AUC values against the Fascin crystal structure (PDB: ID 6I18). Interestingly, PLANTS exhibited the best screening performance and recognized potent actives at early enrichment. Accordingly, PLANTS was utilized in the prospective virtual screening effort for repurposing FDA-approved drugs (DrugBank database) and natural products (NANPDB). Further assessment via molecular dynamics simulations for 100 ns endorsed Remdesivir (DrugBank) and NANPDB3 (NANPDB) as potential binders to Fascin binding site 2. In conclusion, this study delivers a model for implementing a customized DEKOIS 2.0 benchmark set to enhance the VS success rate against new potential targets for cancer therapies.

The binding of natural ligands and synthetic drugs to the P2Y12 receptor is of great interest because of its crucial role in platelets activation and the therapy of arterial thrombosis. Up to now, all computational studies of P2Y12 concentrated on the available crystal structures, while the role of intrinsic protein dynamics and the membrane environment in the functioning of P2Y12 was not clear. In this work, we performed all-atom molecular dynamics simulations of the full-length P2Y12 receptor in three different membrane environments and in two possible conformations derived from available crystal structures. The binding of ticagrelor, its two major metabolites, adenosine diphosphate (ADP) and 2-Methylthioadenosine diphosphate (2MeS-ADP) as agonist, and ethyl 6-[4-(benzylsulfonylcarbamoyl)piperidin-1-yl]-5-cyano-2-methylpyridine-3-carboxylate (AZD1283)as antagonist were assessed systematically by means of ensemble docking. It is shown that the binding of all ligands becomes systematically stronger with the increase of the membrane rigidity. Binding of all ligands to the agonist-bound-like conformations is systematically stronger in comparison to antagonist-bound-likes ones. This is dramatically opposite to the results obtained for static crystal structures. Our results show that accounting for internal protein dynamics, strongly modulated by its lipid environment, is crucial for correct assessment of the ligand binding to P2Y12.

High-throughput X-ray crystal structures of protein–ligand complexes are critical to pharmaceutical drug development. However, cryocooling of crystals and X-ray radiation damage may distort the observed ligand binding. Serial femtosecond crystallography (SFX) using X-ray free-electron lasers (XFELs) can produce radiation-damage-free room-temperature structures. Ligand-binding studies using SFX have received only modest attention, partly owing to limited beamtime availability and the large quantity of sample that is required per structure determination. Here, a high-throughput approach to determine room-temperature damage-free structures with excellent sample and time efficiency is demonstrated, allowing complexes to be characterized rapidly and without prohibitive sample requirements. This yields high-quality difference density maps allowing unambiguous ligand placement. Crucially, it is demonstrated that ligands similar in size or smaller than those used in fragment-based drug design may be clearly identified in data sets obtained from <1000 diffraction images. This efficiency in both sample and XFEL beamtime opens the door to true high-throughput screening of protein–ligand complexes using SFX.

The notable ability of human liver cytochrome P450 3A4 (CYP3A4) to metabolize diverse xenobiotics encourages researchers to explore in-depth the mechanism of enzyme action. Numerous CYP3A4 protein crystal structures have been deposited in protein data bank (PDB) and are majorly used in molecular docking analysis. The quality of the molecular docking results depends on the three-dimensional CYP3A4 protein crystal structures from the PDB. Present review endeavors to provide a brief outline of some technical parameters of CYP3A4 PDB entries as valuable information for molecular docking research. PDB entries between 22 April 2004 and 2 June 2021 were compiled and the active sites were thoroughly observed. The present review identified 76 deposited PDB entries and described basic information that includes CYP3A4 from human genetic, Escherichia coli (E. coli) use for protein expression, crystal structure obtained from X-ray diffraction method, taxonomy ID 9606, Uniprot ID P08684, ligand–protein structure description, co-crystal ligand, protein site deposit and resolution ranges between 1.7[Formula: see text]Å and 2.95[Formula: see text]Å. The observation of protein–ligand interactions showed the various residues on the active site depending on the ligand. The residues Ala305, Ser119, Ala370, Phe304, Phe108, Phe213 and Phe215 have been found to frequently interact with ligands from CYP3A4 PDB. Literature surveys of 17 co-crystal ligands reveal multiple mechanisms that include competitive inhibition, noncompetitive inhibition, mixed-mode inhibition, mechanism-based inhibition, substrate with metabolite, inducer, or combination modes of action. This overview may help researchers choose a trustworthy CYP3A4 protein structure from the PDB database to apply the protein in molecular docking analysis for drug discovery.

Using the theory of machine learning to assist the virtual screening (VS) has been an effective plan. However, the quality of the training set may reduce because of mixing with the wrong docking poses and it will affect the screening efficiencies. To solve this problem, we present a method using the ensemble learning to improve the support vector machine to process the generated protein-ligand interaction fingerprint (IFP). By combining multiple classifiers, ensemble learning is able to avoid the limitations of the single classifier’s performance and obtain better generalization. According to the research of virtual screening experiment with SRC and Cathepsin K as the target, the results show that the ensemble learning method can effectively reduce the error because the sample quality is not high and improve the effect of the whole virtual screening process.

Deviation from normal healthy conditions, termed as disease, can often be triggered due to the malfunctioning of proteins. Modulating the functions of proteins by administering therapeutic agents (drugs) may alleviate the disease conditions. The majority of the drugs currently available in the market are small organic molecules due to their pharmacological and commercial advantages. These small molecule drugs interact with the protein targets through specific sites on the surface of the protein structure (binding sites). Thus, the structural data of protein-small molecule complexes forms a crucial starting point for most drug discovery programs. The work reported in this thesis deals with understanding various aspects of protein-small molecule interactions. The thesis begins (Chapter 1) with a general introduction on the implication of proteins structural data in drug discovery programs. Chapter 2 provides a fundamental understanding of the general trend in local quality of protein-small molecule crystal complexes deposited in the Protein Data Bank (PDB). Our results suggest ‘seeing is not always believing’ and aims to sensitize the non-crystallographer user community that high-resolution need not always guarantee confident small molecule binding poses. The study indicates 35% of the inspected ~0.28 million protein-small molecule binding site pairs available from ~66000 PDB entries, need serious attention before using those as input in any important applications. Results reported in Chapter 3 suggest that the stereochemical quality of bound small molecules generally agrees well with their crystallographic quality. The findings from this work could be the stepping-stones for developing structure determination technique-independent ligand pose validation tools. The learning from Chapter 3 is extended to Chapter 4 to investigate the stereochemical quality of the small molecules bound to protein structures determined by cryo-EM. Our data shows that the stereochemical quality of small molecules bound to high-resolution protein structures determined by cryo-EM is comparable to high-quality small molecules bound to protein crystal structures. Chapter 5 presents a computational analysis aimed at providing insights into the molecular basis of the specificity of a novel anti-tubercular compound, NU-6027 (identified in a phenotypic screening by experimental collaborators), towards two out of the eleven known Serine-Threonine Protein Kinases in Mycobacterium tuberculosis (Mtb). Chapter 6 reports the development of a freely available web server that facilitates the identification of new uses of old drugs and aid in drug repurposing. In Chapter 7, the principles of ‘neighborhood behavior’ are exploited to identify potential known drugs that could be repurposed against the main protease of SARS-CoV-2. Chapter 8 discusses a virtual screening strategy to identify potential binders of a novel Mtb target, Rv1636 (or the Universal Stress Protein). Collaborators have experimentally validated some of the compounds shortlisted from the computational studies. Chapter 9 summarizes the findings from work reported in the entire thesis and future applications. Overall, this thesis inspects protein-small molecule complexes from a local perspective, aiding the design of rigorous computational experiments that can contribute to solving global unmet medical needs. Interested readers may contact the author directly for Supplementary data at "sohini@iisc.ac.in"
DST-INSPIRE fellowship