Virtual screening against highly charged active sites: identifying substrates of alpha-beta barrel enzymes

We have developed a virtual ligand screening method designed to help assign enzymatic function for alpha-beta barrel proteins. We dock a library of approximately 19,000 known metabolites against the active site and attempt to identify the relevant substrate based on predicted relative binding free energies. These energies are computed using a physics-based energy function based on an all-atom force field (OPLS-AA) and a generalized Born implicit solvent model. We evaluate the ability of this method to identify the known substrates of several members of the enolase superfamily of enzymes, including both holo and apo structures (11 total). The active sites of these enzymes contain numerous charged groups (lysines, carboxylates, histidines, and one or more metal ions) and thus provide a challenge for most docking scoring functions, which treat electrostatics and solvation in a highly approximate manner. Using the physics-based scoring procedure, the known substrate is ranked within the top 6% of the database in all cases, and in 8 of 11 cases, it is ranked within the top 1%. Moreover, the top-ranked ligands are strongly enriched in compounds with high chemical similarity to the substrate (e.g., different substitution patterns on a similar scaffold). These results suggest that our method can be used, in conjunction with other information including genomic context and known metabolic pathways, to suggest possible substrates or classes of substrates for experimental testing. More broadly, the physics-based scoring method performs well on highly charged binding sites and is likely to be useful in inhibitor docking against polar binding sites as well. The method is fast (<1 min per ligand), due largely to an efficient minimization algorithm based on the truncated Newton method, and thus, it can be applied to thousands of ligands within a few hours on a small Linux cluster.     

A modified immunoblot method to identify substrates of protein kinases

While protein kinases are key components in multiple cellular processes, efficient identification of cognate in vivo substrates remains challenging. Here we describe a powerful method to screen potential substrates of protein kinases by partial transfer of proteins from a 2D-PAGE gel to a Western blot membrane. This approach allowed precise pinpointing of candidate substrate spots in the 2D gel, and identifying physiological substrates of protein kinases in Mycobacterium tuberculosis.     

A ligand-centric analysis of the diversity and evolution of protein-ligand relationships in E.coli

As enzymes evolve and diverge from common ancestor sequences, they often keep their overall reaction chemistry but specialize in the binding of different cognate ligands. This study borrows methods for the computational assessment of 2D similarity of small molecules from the field of chemoinformatics, to examine the extent of structure conservation of cognate ligands binding to similar proteins. Proteins from 87 structural superfamilies from Escherichia coli form the core dataset, which is extended using homologues with functional assignments from any organism. We find that correlation of the substrate similarity with protein similarity (measured by either sequence-based or structure-based scores) can only be clearly established for very similar proteins. At low sequence identities, the superfamily to which a protein belongs can give helpful clues to its function, and more importantly, the confidence attached to such clues is superfamily-dependent. Our data indicate that only a few superfamilies show great substrate diversity, and that most exhibit conservation of at least part of the structural scaffold of the substrate.     

Identification of DNA-binding Proteins Using Structural, Electrostatic and Evolutionary Features

DNA-binding proteins (DBPs) participate in various crucial processes in the life-cycle of the cells, and the identification and characterization of these proteins is of great importance. We present here a random forests classifier for identifying DBPs among proteins with known 3D structures. First, clusters of evolutionarily conserved regions (patches) on the surface of proteins were detected using the PatchFinder algorithm; earlier studies showed that these regions are typically the functionally important regions of proteins. Next, we trained a classifier using features like the electrostatic potential, cluster-based amino acid conservation patterns and the secondary structure content of the patches, as well as features of the whole protein, including its dipole moment. Using 10-fold cross-validation on a dataset of 138 DBPs and 110 proteins that do not bind DNA, the classifier achieved a sensitivity and a specificity of 0.90, which is overall better than the performance of published methods. Furthermore, when we tested five different methods on 11 new DBPs that did not appear in the original dataset, only our method annotated all correctly.The resulting classifier was applied to a collection of 757 proteins of known structure and unknown function. Of these proteins, 218 were predicted to bind DNA, and we anticipate that some of them interact with DNA using new structural motifs. The use of complementary computational tools supports the notion that at least some of them do bind DNA.     

Discovery of binding proteins for a protein target using protein–protein docking‐based virtual screening

Target structure‐based virtual screening, which employs protein‐small molecule docking to identify potential ligands, has been widely used in small‐molecule drug discovery. In the present study, we used a protein–protein docking program to identify proteins that bind to a specific target protein. In the testing phase, an all‐to‐all protein–protein docking run on a large dataset was performed. The three‐dimensional rigid docking program SDOCK was used to examine protein–protein docking on all protein pairs in the dataset. Both the binding affinity and features of the binding energy landscape were considered in the scoring function in order to distinguish positive binding pairs from negative binding pairs. Thus, the lowest docking score, the average Z‐score, and convergency of the low‐score solutions were incorporated in the analysis. The hybrid scoring function was optimized in the all‐to‐all docking test. The docking method and the hybrid scoring function were then used to screen for proteins that bind to tumor necrosis factor‐α (TNFα), which is a well‐known therapeutic target for rheumatoid arthritis and other autoimmune diseases. A protein library containing 677 proteins was used for the screen. Proteins with scores among the top 20% were further examined. Sixteen proteins from the top‐ranking 67 proteins were selected for experimental study. Two of these proteins showed significant binding to TNFα in an in vitro binding study. The results of the present study demonstrate the power and potential application of protein–protein docking for the discovery of novel binding proteins for specific protein targets. Proteins 2014; 82:2472–2482. © 2014 Wiley Periodicals, Inc.     

Specific Chaperones for the Type VII Protein Secretion Pathway

Mycobacteria use the dedicated type VII protein secretion systems ESX-1 and ESX-5 to secrete virulence factors across their highly hydrophobic cell envelope. The substrates of these systems include the large mycobacterial PE and PPE protein families, which are named after their characteristic Pro-Glu and Pro-Pro-Glu motifs. Pathogenic mycobacteria secrete large numbers of PE/PPE proteins via the major export pathway, ESX-5. In addition, a few PE/PPE proteins have been shown to be exported by ESX-1. It is not known how ESX-1 and ESX-5 recognize their cognate PE/PPE substrates. In this work, we investigated the function of the cytosolic protein EspG(5), which is essential for ESX-5-mediated secretion in Mycobacterium marinum, but for which the role in secretion is not known. By performing protein co-purifications, we show that EspG(5) interacts with several PPE proteins and a PE/PPE complex that is secreted by ESX-5, but not with the unrelated ESX-5 substrate EsxN or with PE/PPE proteins secreted by ESX-1. Conversely, the ESX-1 paralogue EspG(1) interacted with a PE/PPE couple secreted by ESX-1, but not with PE/PPE substrates of ESX-5. Furthermore, structural analysis of the complex formed by EspG(5) and PE/PPE indicates that these proteins interact in a 1:1:1 ratio. In conclusion, our study shows that EspG(5) and EspG(1) interact specifically with PE/PPE proteins that are secreted via their own ESX systems and suggests that EspG proteins are specific chaperones for the type VII pathway.     

Sequence Alignment Reveals Possible MAPK Docking Motifs on HIV Proteins

Over the course of HIV infection, virus replication is facilitated by the phosphorylation of HIV proteins by human ERK1 and ERK2 mitogen-activated protein kinases (MAPKs). MAPKs are known to phosphorylate their substrates by first binding with them at a docking site. Docking site interactions could be viable drug targets because the sequences guiding them are more specific than phosphorylation consensus sites. In this study we use multiple bioinformatics tools to discover candidate MAPK docking site motifs on HIV proteins known to be phosphorylated by MAPKs, and we discuss the possibility of targeting docking sites with drugs. Using sequence alignments of HIV proteins of different subtypes, we show that MAPK docking patterns previously described for human proteins appear on the HIV matrix, Tat, and Vif proteins in a strain dependent manner, but are absent from HIV Rev and appear on all HIV Nef strains. We revise the regular expressions of previously annotated MAPK docking patterns in order to provide a subtype independent motif that annotates all HIV proteins. One revision is based on a documented human variant of one of the substrate docking motifs, and the other reduces the number of required basic amino acids in the standard docking motifs from two to one. The proposed patterns are shown to be consistent with in silico docking between ERK1 and the HIV matrix protein. The motif usage on HIV proteins is sufficiently different from human proteins in amino acid sequence similarity to allow for HIV specific targeting using small-molecule drugs.     

Aza-β-amino acid containing peptidomimetics as cAMP-dependent protein kinase substrates

Peptidomimetic analogs of the peptide RRASVA, known as the “minimal substrate” of the catalytic subunit of the cAMP-dependent protein kinase (PKA), were synthesized by consecutive replacement of natural amino acids by their aza-β³ analogs. The peptidomimetics were tested as PKA substrates and the kinetic parameters of the phosphorylation reaction were determined. It was found that the interaction of these peptidomimetics with the enzyme active center was sensitive to the location of the backbone modification, while the maximal rate of the reaction was practically not affected by the structure of substrates. The pattern of molecular recognition of peptidomimetics was in agreement with the results of structure modeling and also with the results of computational docking study of peptide and peptidomimetic substrates with the active center of PKA. It was concluded that the specificity determining factors which govern substrate recognition by the enzyme should be grouped along the phosphorylatable substrate, and such clustering might open new perspectives for pharmacophore design of peptides and peptide-like ligands.     

Histone H2A and H4 N-terminal Tails Are Positioned by the MEP50 WD Repeat Protein for Efficient Methylation by the PRMT5 Arginine Methyltransferase

The protein arginine methyltransferase PRMT5 is complexed with the WD repeat protein MEP50 (also known as Wdr77 or androgen coactivator p44) in vertebrates in a tetramer of heterodimers. MEP50 is hypothesized to be required for protein substrate recruitment to the catalytic domain of PRMT5. Here we demonstrate that the cross-dimer MEP50 is paired with its cognate PRMT5 molecule to promote histone methylation. We employed qualitative methylation assays and a novel ultrasensitive continuous assay to measure enzyme kinetics. We demonstrate that neither full-length human PRMT5 nor the Xenopus laevis PRMT5 catalytic domain has appreciable protein methyltransferase activity. We show that histones H4 and H3 bind PRMT5-MEP50 more efficiently compared with histone H2A(1–20) and H4(1–20) peptides. Histone binding is mediated through histone fold interactions as determined by competition experiments and by high density histone peptide array interaction studies. Nucleosomes are not a substrate for PRMT5-MEP50, consistent with the primary mode of interaction via the histone fold of H3-H4, obscured by DNA in the nucleosome. Mutation of a conserved arginine (Arg-42) on the MEP50 insertion loop impaired the PRMT5-MEP50 enzymatic efficiency by increasing its histone substrate K_m, comparable with that of Caenorhabditis elegans PRMT5. We show that PRMT5-MEP50 prefers unmethylated substrates, consistent with a distributive model for dimethylation and suggesting discrete biological roles for mono- and dimethylarginine-modified proteins. We propose a model in which MEP50 and PRMT5 simultaneously engage the protein substrate, orienting its targeted arginine to the catalytic site.     

A structure-based anatomy of the E.coli metabolome

The Escherichia coli metabolome has been characterised using the two-dimensional structures of 745 metabolites, obtained from the EcoCyc and KEGG databases. Physicochemical properties of the metabolome have been calculated to provide an overview of this set of cognate ligands. A library of fragments commonly found among these molecules has been employed to reveal the main constituents of metabolites, and to assist a broad classification of the metabolome into biochemically relevant classes. Fragment-based fingerprints reveal the metabolome as a continuum in the two-dimensional structural space, where clusters of molecules sharing similar scaffolds can be identified, but are generally overlapping. Nucleotide, carbohydrate and amino acid-like molecules are the most prominent, but at high levels of similarity, a more detailed classification is possible. Classification schemes for the metabolome are a promising tool for understanding the chemical diversity of the metabolome. When used in conjunction with existing classifications of the proteome, they can help to elucidate the binding preferences and promiscuity of proteins and their cognate substrates.     

The Structure of an NDR/LATS Kinase–Mob Complex Reveals a Novel Kinase–Coactivator System and Substrate Docking Mechanism

Eukaryotic cells commonly use protein kinases in signaling systems that relay information and control a wide range of processes. These enzymes have a fundamentally similar structure, but achieve functional diversity through variable regions that determine how the catalytic core is activated and recruited to phosphorylation targets. “Hippo” pathways are ancient protein kinase signaling systems that control cell proliferation and morphogenesis; the NDR/LATS family protein kinases, which associate with “Mob” coactivator proteins, are central but incompletely understood components of these pathways. Here we describe the crystal structure of budding yeast Cbk1–Mob2, to our knowledge the first of an NDR/LATS kinase–Mob complex. It shows a novel coactivator-organized activation region that may be unique to NDR/LATS kinases, in which a key regulatory motif apparently shifts from an inactive binding mode to an active one upon phosphorylation. We also provide a structural basis for a substrate docking mechanism previously unknown in AGC family kinases, and show that docking interaction provides robustness to Cbk1’s regulation of its two known in vivo substrates. Co-evolution of docking motifs and phosphorylation consensus sites strongly indicates that a protein is an in vivo regulatory target of this hippo pathway, and predicts a new group of high-confidence Cbk1 substrates that function at sites of cytokinesis and cell growth. Moreover, docking peptides arise in unstructured regions of proteins that are probably already kinase substrates, suggesting a broad sequential model for adaptive acquisition of kinase docking in rapidly evolving intrinsically disordered polypeptides.     

An ATP analog-sensitive version of the tomato cell death suppressor protein kinase Adi3 for use in substrate identification

Adi3 is a protein kinase from tomato that functions as a cell death suppressor and its substrates are not well defined. As a step toward identifying Adi3 substrates we developed an ATP analog-sensitive version of Adi3 in which the ATP-binding pocket is mutated to allow use of bulky ATP analogs. Met385 was identified as the “gatekeeper” residue and the M385G mutation allows for the use of two bulky ATP analogs. Adi3^M385G can also specifically utilize N⁶-benzyl-ATP to phosphorylate a known substrate and provides a tool for identifying Adi3 substrates.     

Novel Isoform-Specific Interfaces Revealed by PKA RIIβ Holoenzyme Structures

The cAMP-dependent protein kinase catalytic (C) subunit is inhibited by two classes of functionally nonredundant regulatory (R) subunits, RI and RII. Unlike RI subunits, RII subunits are both substrates and inhibitors. Because RIIβ knockout mice have important disease phenotypes, the RIIβ holoenzyme is a target for developing isoform-specific agonists and/or antagonists. We also know little about the linker region that connects the inhibitor site to the N-terminal dimerization domain, although this linker determines the unique globular architecture of the RIIβ holoenzyme. To understand how RIIβ functions as both an inhibitor and a substrate and to elucidate the structural role of the linker, we engineered different RIIβ constructs. In the absence of nucleotide, RIIβ(108-268), which contains a single cyclic nucleotide binding domain, bound C subunit poorly, whereas with AMP-PNP, a non-hydrolyzable ATP analog, the affinity was 11 nM. The RIIβ(108-268) holoenzyme structure (1.62 Å) with AMP-PNP/Mn²⁺ showed that we trapped the RIIβ subunit in an enzyme:substrate complex with the C subunit in a closed conformation. The enhanced affinity afforded by AMP-PNP/Mn²⁺ may be a useful strategy for increasing affinity and trapping other protein substrates with their cognate protein kinase. Because mutagenesis predicted that the region N-terminal to the inhibitor site might dock differently to RI and RII, we also engineered RIIβ(102-265), which contained six additional linker residues. The additional linker residues in RIIβ(102-265) increased the affinity to 1.6 nM, suggesting that docking to this surface may also enhance catalytic efficiency. In the corresponding holoenzyme structure, this linker docks as an extended strand onto the surface of the large lobe. This hydrophobic pocket, formed by the αF-αG loop and conserved in many protein kinases, also provides a docking site for the amphipathic helix of PKI. This novel orientation of the linker peptide provides the first clues as to how this region contributes to the unique organization of the RIIβ holoenzyme.     

Revealing the allosterome: systematic identification of metabolite-protein interactions

Small molecule allostery modifies protein function but is not easily discovered. We introduce mass spectrometry integrated with equilibrium dialysis for the discovery of allostery systematically (MIDAS), a method for identifying physiologically relevant, low-affinity metabolite-protein interactions using unmodified proteins and complex mixtures of unmodified metabolites. In a pilot experiment using five proteins, we identified 16 known and 13 novel interactions. The known interactions included substrates, products, intermediates, and allosteric regulators of their protein partners. MIDAS does not depend upon enzymatic measurements, but most of the new interactions affect the enzymatic activity of the protein partner. We found that the fatty acid palmitate interacts with both glucokinase and glycogen phosphorylase. Further characterization revealed that palmitate inhibited both enzymes, possibly providing a mechanism for sparing carbohydrate catabolism when fatty acids are abundant.     

Efficient molecular docking of NMR structures: application to HIV-1 protease

Docking ligands into an ensemble of NMR conformers is essential to structure-based drug discovery if only NMR structures are available for the target. However, sequentially docking ligands into each NMR conformer through standard single-receptor-structure docking, referred to as sequential docking, is computationally expensive for large-scale database screening because of the large number of NMR conformers involved. Recently, we developed an efficient ensemble docking algorithm to consider protein structural variations in ligand binding. The algorithm simultaneously docks ligands into an ensemble of protein structures and achieves comparable performance to sequential docking without significant increase in computational time over single-structure docking. Here, we applied this algorithm to docking with NMR structures. The HIV-1 protease was used for validation in terms of docking accuracy and virtual screening. Ensemble docking of the NMR structures identified 91% of the known inhibitors under the criterion of RMSD < 2.0 A for the best-scored conformation, higher than the average success rate of single docking of individual crystal structures (66%). In the virtual screening test, on average, ensemble docking of the NMR structures obtained higher enrichments than single-structure docking of the crystal structures. In contrast, docking of either the NMR minimized average structure or a single NMR conformer performed less satisfactorily on both binding mode prediction and virtual screening, indicating that a single NMR structure may not be suitable for docking calculations. The success of ensemble docking of the NMR structures suggests an efficient alternative method for standard single docking of crystal structures and for considering protein flexibility.     

Extensive in vivo metabolite-protein interactions revealed by large-scale systematic analyses

Natural small compounds comprise most cellular molecules and bind proteins as substrates, products, cofactors, and ligands. However, a large-scale investigation of in?vivo protein-small metabolite interactions has not been performed. We developed a mass spectrometry assay for the large-scale identification of in?vivo protein-hydrophobic small metabolite interactions in yeast and analyzed compounds that bind ergosterol biosynthetic proteins and protein kinases. Many of these proteins bind small metabolites; a few interactions were previously known, but the vast majority are new. Importantly, many key regulatory proteins such as protein kinases bind metabolites. Ergosterol was found to bind many proteins and may function as a general regulator. It is required for the activity of Ypk1, a mammalian AKT/SGK kinase homolog. Our study defines potential key regulatory steps in lipid biosynthetic pathways and suggests that small metabolites may play a more general role as regulators of protein activity and function than previously appreciated.     

Crystallographic analysis of CaaX prenyltransferases complexed with substrates defines rules of protein substrate selectivity

Post-translational modifications are essential for the proper function of many proteins in the cell. The attachment of an isoprenoid lipid (a process termed prenylation) by protein farnesyltransferase (FTase) or geranylgeranyltransferase type I (GGTase-I) is essential for the function of many signal transduction proteins involved in growth, differentiation, and oncogenesis. FTase and GGTase-I (also called the CaaX prenyltransferases) recognize protein substrates with a C-terminal tetrapeptide recognition motif called the Ca1a2X box. These enzymes possess distinct but overlapping protein substrate specificity that is determined primarily by the sequence identity of the Ca1a2X motif. To determine how the identity of the Ca1a2X motif residues and sequence upstream of this motif affect substrate binding, we have solved crystal structures of FTase and GGTase-I complexed with a total of eight cognate and cross-reactive substrate peptides, including those derived from the C termini of the oncoproteins K-Ras4B, H-Ras and TC21. These structures suggest that all peptide substrates adopt a common binding mode in the FTase and GGTase-I active site. Unexpectedly, while the X residue of the Ca1a2X motif binds in the same location for all GGTase-I substrates, the X residue of FTase substrates can bind in one of two different sites. Together, these structures outline a series of rules that govern substrate peptide selectivity; these rules were utilized to classify known protein substrates of CaaX prenyltransferases and to generate a list of hypothetical substrates within the human genome.     

Cognate ligand domain mapping for enzymes

Here, we present an automatic assignment of potential cognate ligands to domains of enzymes in the CATH and SCOP protein domain classifications on the basis of structural data available in the wwPDB. This procedure involves two steps; firstly, we assign the binding of particular ligands to particular domains; secondly, we compare the chemical similarity of the PDB ligands to ligands in KEGG in order to assign cognate ligands. We find that use of the Enzyme Commission (EC) numbers is necessary to enable efficient and accurate cognate ligand assignment. The PROCOGNATE database currently has cognate ligand mapping for 3277 (4118) protein structures and 351 (302) superfamilies, as described by the CATH and (SCOP) databases, respectively. We find that just under half of all ligands are only and always bound by a single domain, with 16% bound by more than one domain and the remainder of the ligands showing a variety of binding modes. This finding has implications for domain recombination and the evolution of new protein functions. Domain architecture or context is also found to affect substrate specificity of particular domains, and we discuss example cases. The most popular PDB ligands are all found to be generic components of crystallisation buffers, highlighting the non-cognate ligand problem inherent in the PDB. In contrast, the most popular cognate ligands are all found to be universal cellular currencies of reducing power and energy such as NADH, FADH2 and ATP, respectively, reflecting the fact that the vast majority of enzymatic reactions utilise one of these popular co-factors. These ligands all share a common adenine ribonucleotide moiety, suggesting that many different domain superfamilies have converged to bind this chemical framework.     

Energetics-based discovery of protein-ligand interactions on a proteomic scale

Biochemical functions of proteins in cells frequently involve interactions with various ligands. Proteomic methods for the identification of proteins that interact with specific ligands such as metabolites, signaling molecules, and drugs are valuable in investigating the regulatory mechanisms of cellular metabolism, annotating proteins with unknown functions, and elucidating pharmacological mechanisms. Here we report an energetics-based target identification method in which target proteins in a cell lysate are identified by exploiting the effect of ligand binding on their stabilities. Urea-induced unfolding of proteins in cell lysates is probed by a short pulse of proteolysis, and the effect of a ligand on the amount of folded protein remaining is monitored on a proteomic scale. As proof of principle, we identified proteins that interact with ATP in the Escherichia coli proteome. Literature and database mining confirmed that a majority of the identified proteins are indeed ATP-binding proteins. Four identified proteins that were previously not known to interact with ATP were cloned and expressed to validate the result. Except for one protein, the effects of ATP on urea-induced unfolding were confirmed. Analyses of the protein sequences and structure models were also employed to predict potential ATP binding sites in the identified proteins. Our results demonstrate that this energetics-based target identification approach is a facile method to identify proteins that interact with specific ligands on a proteomic scale.     

An efficient method for protein phosphorylation using the artificially introduced of cognate-binding modules into kinases and substrates

Protein phosphorylation is a major post-translational modification that regulates cellular signal transduction. The phosphorylation of substrate proteins by kinases requires cognate pairs of substrates and kinases. In addition, phosphorylation is mediated through both indirect and direct interaction between these kinases and substrates, which makes it difficult to effectively prepare large quantities of recombinant phosphorylated proteins. Here, we report a novel protein phosphorylation method involving the artificial introduction of cognate-binding modules into substrates and enzymes. This enhances the local concentration of substrates around enzymes so that the enzymatic reaction proceeds more efficiently. We prepared substrate proteins containing an SH3 domain at their N-terminus, and a kinase containing an SH3-binding motif at its C-terminus. This method was successfully applied to the phosphorylation of CrkII and the Vav DH domain, and we prepared (15)N-labelled phosphorylated CrkII for NMR analysis.