Comparison of Rosetta flexible-backbone computational protein design methods on binding interactions

Computational design of binding sites in proteins remains difficult, in part due to limitations in our current ability to sample backbone conformations that enable precise and accurate geometric positioning of side chains during sequence design. Here we present a benchmark framework for comparison between flexible-backbone design methods applied to binding interactions. We quantify the ability of different flexible backbone design methods in the widely used protein design software Rosetta to recapitulate observed protein sequence profiles assumed to represent functional protein/protein and protein/small molecule binding interactions. The CoupledMoves method, which combines backbone flexibility and sequence exploration into a single acceptance step during the sampling trajectory, better recapitulates observed sequence profiles than the BackrubEnsemble and FastDesign methods, which separate backbone flexibility and sequence design into separate acceptance steps during the sampling trajectory. Flexible-backbone design with the CoupledMoves method is a powerful strategy for reducing sequence space to generate targeted libraries for experimental screening and selection.     

High-throughput screening and characterization of clones selected from phage display libraries.

BACKGROUND: Clones from phage display libraries are generally selected by a number of rounds of panning and regrowth, followed by primary screening to identify hits and secondary characterization to identify clones with optimal affinity and specificity. Because functional screening for binding or other activity can be material-, time-, and labor-intensive, sequencing is often used to identify the emergence of a consensus sequence prior functional characterization. However, the consensus sequence is not always the optimal one because factors such as phage growth rates, nonspecific binding, and other selection pressures can bias the selection process. METHODS: To improve function-based phage display library screening and characterization, we developed a multiplexed approach employing optically-encoded microsphere arrays and flow cytometry. RESULTS: We show that capture of phage from crude culture supernatants enables the efficient screening of binding activity and the evaluation of binding avidity. The approach uses small volumes and a homogeneous no-wash format that minimizes reagent consumption and sample handling. The use of optically-encoded microspheres allows many phage to be screened simultaneously, greatly increasing throughput. CONCLUSIONS: This approach is flexible, supporting primary and secondary screening for a range of functional assays, and scalable, potentially supporting the screening of thousands to hundreds of thousands of clones per hour.     

Early developability screen of therapeutic antibody candidates using Taylor dispersion analysis and UV area imaging detection

Therapeutic antibodies represent one of the fastest growing segments in the pharmaceutical market. They are used in a broad range of disease fields, such as autoimmune diseases, cancer, inflammation and infectious diseases. The growth of the segment has necessitated development of new analytical platforms for faster and better antibody selection and characterization. Early quality control and risk assessment of biophysical parameters help prevent failure in later stages of antibody development, and thus can reduce costs and save time. Critical parameters such as aggregation, conformational stability, colloidal stability and hydrophilicity, are measured during the early phase of antibody generation and guide the selection process of the best lead candidates in terms of technical developability. We report on the use of a novel instrument (ActiPix/Viscosizer) for measuring both the hydrodynamic radius and the absolute viscosity of antibodies based on Taylor dispersion analysis and UV area imaging. The looped microcapillary-based method combines low sample consumption, fast throughput and high precision compared to other conventional methods. From a random panel of 130 antibodies in the early selection process, we identified some with large hydrodynamic radius outside the normal distribution and others with non-Gaussian Taylor dispersion profiles. The antibodies with such abnormal properties were confirmed later in the selection process to show poor developability profiles. Moreover, combining these results with those of the viscosity measurements at high antibody concentrations allows screening, with limited amounts of materials, candidates with potential issues in pre-formulation development.     

Early developability screen of therapeutic antibody candidates using Taylor dispersion analysis and UV area imaging detection

Therapeutic antibodies represent one of the fastest growing segments in the pharmaceutical market. They are used in a broad range of disease fields, such as autoimmune diseases, cancer, inflammation and infectious diseases. The growth of the segment has necessitated development of new analytical platforms for faster and better antibody selection and characterization. Early quality control and risk assessment of biophysical parameters help prevent failure in later stages of antibody development, and thus can reduce costs and save time. Critical parameters such as aggregation, conformational stability, colloidal stability and hydrophilicity, are measured during the early phase of antibody generation and guide the selection process of the best lead candidates in terms of technical developability. We report on the use of a novel instrument (ActiPix/Viscosizer) for measuring both the hydrodynamic radius and the absolute viscosity of antibodies based on Taylor dispersion analysis and UV area imaging. The looped microcapillary-based method combines low sample consumption, fast throughput and high precision compared to other conventional methods. From a random panel of 130 antibodies in the early selection process, we identified some with large hydrodynamic radius outside the normal distribution and others with non-Gaussian Taylor dispersion profiles. The antibodies with such abnormal properties were confirmed later in the selection process to show poor developability profiles. Moreover, combining these results with those of the viscosity measurements at high antibody concentrations allows screening, with limited amounts of materials, candidates with potential issues in pre-formulation development.     

Natural glycan microarrays

Glycan microarrays are emerging as increasingly used screening tools with a high potential for unraveling protein–carbohydrate interactions: probing hundreds or even thousands of glycans in parallel, they provide the researcher with a vast amount of data in a short time-frame, while using relatively small amounts of analytes. Natural glycan microarrays focus on the glycans’ repertoire of natural sources, including both well-defined structures as well as still-unknown ones. This article compares different natural glycan microarray strategies. Glycan probes may comprise oligosaccharides from glycoproteins as well as glycolipids and polysaccharides. Oligosaccharides may be purified from scarce biological samples that are of particular relevance for the carbohydrate-binding protein to be studied. We give an overview of strategies for glycan isolation, derivatization, fractionation, immobilization and structural characterization. Detection methods such as fluorescence analysis and surface plasmon resonance are summarized. The importance of glycan density and multivalency is discussed. Furthermore, some applications of natural glycan microarrays for studying lectin and antibody binding are presented.     

Natural glycan microarrays

Glycan microarrays are emerging as increasingly used screening tools with a high potential for unraveling protein-carbohydrate interactions: probing hundreds or even thousands of glycans in parallel, they provide the researcher with a vast amount of data in a short time-frame, while using relatively small amounts of analytes. Natural glycan microarrays focus on the glycans' repertoire of natural sources, including both well-defined structures as well as still-unknown ones. This article compares different natural glycan microarray strategies. Glycan probes may comprise oligosaccharides from glycoproteins as well as glycolipids and polysaccharides. Oligosaccharides may be purified from scarce biological samples that are of particular relevance for the carbohydrate-binding protein to be studied. We give an overview of strategies for glycan isolation, derivatization, fractionation, immobilization and structural characterization. Detection methods such as fluorescence analysis and surface plasmon resonance are summarized. The importance of glycan density and multivalency is discussed. Furthermore, some applications of natural glycan microarrays for studying lectin and antibody binding are presented.     

Mixed mode of ligand-DNA binding results in S-shaped binding curves

S-shaped binding curves often characterize interactions of ligands with nucleic acid molecules as analyzed by different physico-chemical and biophysical techniques. S-shaped experimental binding curves are usually interpreted as indicative of the positive cooperative interactions between the bound ligand molecules. This paper demonstrates that S-shaped binding curves may occur as a result of the "mixed mode" of DNA binding by the same ligand molecule. Mixed mode of the ligand-DNA binding can occur, for example, due to 1) isomerization or dimerization of the ligands in solution or on the DNA lattice, 2) their ability to intercalate the DNA and to bind it within the minor groove in different orientations. DNA-ligand complexes are characterized by the length of the ligand binding site on the DNA lattice (so-called "multiple-contact" model). We show here that if two or more complexes with different lengths of the ligand binding sites could be produced by the same ligand, the dependence of the concentration of the complex with the shorter length of binding site on the total concentration of ligand should be S-shaped. Our theoretical model is confirmed by comparison of the calculated and experimental CD binding curves for bis-netropsin binding to poly(dA-dT) poly(dA-dT). Bis-netropsin forms two types of DNA complexes due to its ability to interact with the DNA as monomers and trimers. Experimental S-shaped bis-netropsin-DNA binding curve is shown to be in good correlation with those calculated on the basis of our theoretical model. The present work provides new insight into the analysis of ligand-DNA binding curves.     

Surface plasmon resonance study of the molecular recognition between polymerase and DNA containing various mismatches and conformational changes of DNA-protein complexes

Although surface plasmon resonance (SPR) biosensor technique has been used to study protein-protein interactions and to detect conformational changes of proteins, it has not been shown whether the SPR biosensor can be used to study a complex kinetic system such as the protein-DNA binding, which sometimes involves several binding steps as well as dynamic conformational changes of the complexes. In this study, we have used SPR biosensor and T7 polymerase as the model system to study the interactions of the polymerase with a series of DNA template-primer duplexes containing different number of mismatches and GC contents at various positions near the primer 3'-end. In general, the binding constants measured by the SPR are several magnitudes smaller than those determined in solution, indicating the limitation of the surface-based technique for measuring solution-based interactions. However, the distinct polymerase binding profiles obtained for DNA duplexes differed by as low as a single mismatch suggest that the SPR data can be used for relative comparison purpose among a set of experiments carried out under identical conditions. The successful fitting of the binding profiles using the established translocation model also demonstrated that SPR can be used to monitor conformational changes, as well as to derive relative kinetic values, within a complicated DNA-protein interaction system. The results also demonstrated that SPR biosensor may be used as a sensitive technique for studying molecular recognition events, such as single-base discrimination involved in protein-DNA interactions.     

Fluorescent labeling agents change binding profiles of glycan-binding proteins

Interactions of glycan-binding proteins (GBPs) with glycans are essential in cell adhesion, bacterial/viral infection, and cellular signaling pathways. Experimental characterization of these interactions based on glycan microarrays typically involves (1) labeling GBPs directly with fluorescent reagents before incubation with the microarrays, or (2) labeling GBPs with biotin before the incubation and detecting the captured GBPs after the incubation using fluorescently labeled streptavidin, or (3) detecting the captured GBPs after the incubation using fluorescently labeled antibodies raised against the GBPs. The fluorescent signal is mostly measured ex situ after excess fluorescent materials are washed off. In this study, by using a label-free optical scanner for glycan microarray detection, we measured binding curves of 7 plant lectins to 24 glycans: four β1-4-linked galactosides, three β1-3-linked galactosides, one β-linked galactoside, one α-linked N-acetylgalactosaminide, eight α2-3-linked sialosides, and seven α2-6-linked sialosides. From association and dissociation constants deduced by global-fitting the binding curves, we found that (1) labeling lectins directly with fluorescent agents change binding profiles of lectins, in some cases by orders of magnitude; (2) those lectin-glycan binding reactions characterized with large dissociation rates, though biologically relevant, are easily missed or deemed insignificant in ex situ fluorescence-based assays as most captured lectins are washed off before detection. This study highlights the importance of label-free real-time detection of protein-ligand interactions and the potential pitfall in interpreting fluorescence-based assays for characterization of protein-glycan interactions.     

A cell-based approach to characterize antimicrobial compounds through kinetic dose response

The rapid spread of antibiotic resistance has created a pressing need for the development of novel drug screening platforms. Herein, we report on the use of cell-based kinetic dose response curves for small molecule characterization in antibiotic discovery efforts. Kinetically monitoring bacterial growth at sub-inhibitory concentrations of antimicrobial small molecules generates unique dose response profiles. We show that clustering of profiles by growth characteristics can classify antibiotics by mechanism of action. Furthermore, changes in growth kinetics have the potential to offer insight into the mechanistic action of novel molecules and can be used to predict off-target effects generated through structure–activity relationship studies. Kinetic dose response also allows for detection of unstable compounds early in the lead development process. We propose that this kinetic approach is a rapid and cost-effective means to gather critical information on antimicrobial small molecules during the hit selection and lead development pipeline.     

A high efficiency strategy for binding property characterization of peptide-binding domains

A large proportion of protein-protein interactions is mediated by families of peptide-binding domains. Comprehensive characterization of each of these domains is critical for understanding the mechanisms and networks of protein interaction at the domain level. However, existing methods are all based on large scale screenings for each domain that are inefficient to deal with hundreds of members in major domain families. We developed a systematic strategy for efficient binding property characterization of peptide-binding domains based on high throughput validation screening of a specialized candidate ligand library using yeast two-hybrid mating array. Its outstanding feature is that the overall efficiency is dramatically improved compared with that of traditional screening, and it will be higher as the system cycles. PDZ domain family was first used to test the strategy. Five PDZ domains were rapidly characterized. Broader binding properties were identified compared with other methods, including novel recognition specificities that provided the basis for major revision of conventional PDZ classification. Several novel interactions were discovered, serving as significant clues for further functional investigation. This strategy can be easily extended to a variety of peptide-binding domains as a powerful tool for comprehensive analysis of domain binding property in proteomic scale.     

Genetic approaches to the identification of interactions between membrane proteins in yeast

The recent sequencing of entire eukaryotic genomes has renewed the interest in identifying and characterizing all gene products that are expressed in a given organism. The characterization of unknown gene products is facilitated by the knowledge of its binding partners. Thus, a novel protein may be classified by identifying previously characterized proteins that interact with it. If such an approach is carried out on a large scale, it may allow the rapid characterization of the thousands of predicted open reading frames identified by recent sequencing projects. Currently, the yeast two-hybrid system is the most widely used genetic assay for the detection of protein-protein interactions. The yeast two-hybrid system has become popular because it requires little individual optimization and because, as compared to conventional biochemical methods, the identification and characterization of protein-protein interactions can be completed in a relatively short time span. In this review, we briefly discuss the yeast two-hybrid system and its application to large scale screening studies that aim at deciphering all protein-protein interactions taking place in a given cell type or organism. We then focus on a class of proteins that is unsuitable for conventional yeast two-hybrid systems, namely integral membrane proteins and membrane-associated proteins, and describe several novel genetic systems that combine the advantages of the yeast two-hybrid system with the potential to identify interaction partners of membrane-associated proteins in their natural setting.     

Comprehensive analysis of lectin-glycan interactions reveals determinants of lectin specificity

Lectin-glycan interactions facilitate inter- and intracellular communication in many processes including protein trafficking, host-pathogen recognition, and tumorigenesis promotion. Specific recognition of glycans by lectins is also the basis for a wide range of applications in areas including glycobiology research, cancer screening, and antiviral therapeutics. To provide a better understanding of the determinants of lectin-glycan interaction specificity and support such applications, this study comprehensively investigates specificity-conferring features of all available lectin-glycan complex structures. Systematic characterization, comparison, and predictive modeling of a set of 221 complementary physicochemical and geometric features representing these interactions highlighted specificity-conferring features with potential mechanistic insight. Univariable comparative analyses with weighted Wilcoxon-Mann-Whitney tests revealed strong statistical associations between binding site features and specificity that are conserved across unrelated lectin binding sites. Multivariable modeling with random forests demonstrated the utility of these features for predicting the identity of bound glycans based on generalized patterns learned from non-homologous lectins. These analyses revealed global determinants of lectin specificity, such as sialic acid glycan recognition in deep, concave binding sites enriched for positively charged residues, in contrast to high mannose glycan recognition in fairly shallow but well-defined pockets enriched for non-polar residues. Focused fine specificity analysis of hemagglutinin interactions with human-like and avian-like glycans uncovered features representing both known and novel mutations related to shifts in influenza tropism from avian to human tissues. As the approach presented here relies on co-crystallized lectin-glycan pairs for studying specificity, it is limited in its inferences by the quantity, quality, and diversity of the structural data available. Regardless, the systematic characterization of lectin binding sites presented here provides a novel approach to studying lectin specificity and is a step towards confidently predicting new lectin-glycan interactions.     

Mixed Mode of Ligand-DNA Binding Results in S-Shaped Binding Curves

Abstract

S-shaped binding curves often characterize interactions of ligands with nucleic acid molecules as analyzed by different physicochemical and biophysical techniques. S-shaped experimental binding curves are usually interpreted as indicative of the positive cooperative interactions between the bound ligand molecules. This paper demonstrates that S-shaped binding curves may occur as a result of the “mixed mode” of DNA binding by the same ligand molecule. Mixed mode of the ligand-DNA binding can occur, for example, due to 1) isomerization or dimerization of the ligands in solution or on the DNA lattice, 2) their ability to intercalate the DNA and to bind it within the minor groove in different orientations. DNA- ligand complexes are characterized by the length of the ligand binding site on the DNA lattice (so-called “multiple-contact” model). We show here that if two or more complexes with different lengths of the ligand binding sites could be produced by the same ligand, the dependence of the concentration of the complex with the shorter length of binding site on the total concentration of ligand should be S-shaped. Our theoretical model is confirmed by comparison of the calculated and experimental CD binding curves for bis-netropsin binding to poly(dA-dT) poly(dA-dT). Bis-netropsin forms two types of DNA complexes due to its ability to interact with the DNA as monomers and trimers. Experimental S-shaped bis-netropsin-DNA binding curve is shown to be in good correlation with those calculated on the basis of our theoretical model. The present work provides new insight into the analysis of ligand-DNA binding curves.     

Evidence of Conformational Selection Driving the Formation of Ligand Binding Sites in Protein-Protein Interfaces

Many protein-protein interactions (PPIs) are compelling targets for drug discovery, and in a number of cases can be disrupted by small molecules. The main goal of this study is to examine the mechanism of binding site formation in the interface region of proteins that are PPI targets by comparing ligand-free and ligand-bound structures. To avoid any potential bias, we focus on ensembles of ligand-free protein conformations obtained by nuclear magnetic resonance (NMR) techniques and deposited in the Protein Data Bank, rather than on ensembles specifically generated for this study. The measures used for structure comparison are based on detecting binding hot spots, i.e., protein regions that are major contributors to the binding free energy. The main tool of the analysis is computational solvent mapping, which explores the surface of proteins by docking a large number of small “probe” molecules. Although we consider conformational ensembles obtained by NMR techniques, the analysis is independent of the method used for generating the structures. Finding the energetically most important regions, mapping can identify binding site residues using ligand-free models based on NMR data. In addition, the method selects conformations that are similar to some peptide-bound or ligand-bound structure in terms of the properties of the binding site. This agrees with the conformational selection model of molecular recognition, which assumes such pre-existing conformations. The analysis also shows the maximum level of similarity between unbound and bound states that is achieved without any influence from a ligand. Further shift toward the bound structure assumes protein-peptide or protein-ligand interactions, either selecting higher energy conformations that are not part of the NMR ensemble, or leading to induced fit. Thus, forming the sites in protein-protein interfaces that bind peptides and can be targeted by small ligands always includes conformational selection, although other recognition mechanisms may also be involved.     

A Bayesian genome screening of maximum number of drinks as an alcoholism phenotype with the new Haseman-Elston method

Common human disorders, such as alcoholism, may be the result of interactions of many genes as well as environmental risk factors. Therefore, it is important to incorporate gene x gene and gene x environment interactions in complex disease gene mapping. In this study, we applied a robust Bayesian genome screening method that can incorporate interaction effects to map genes underlying alcoholism through its application to the data of the Collaborative Studies on Genetics of Alcoholism provided by Genetic Analysis Workshop 14. Our Bayesian genome screening method uses the regression-based stochastic variable selection, coupled with the new Haseman-Elston method to identify markers linked to phenotypes of interest. Compared to traditional linkage methods based on single-gene disease models, our method allows for multilocus disease models for simultaneous screening including both main and interaction (epistatic) effects. It is conceptually simple and computationally efficient through the use of Gibbs sampler. We conducted genome-wide analysis and comparison between scans based on microsatellites and single-nucleotide polymorphisms. A total of 328 microsatellites and 11,560 single-nucleotide polymorphisms (by Affymetrix) on 22 autosomal chromosomes and sex chromosome were used.     

The binding of a ligand to an acceptor undergoing indefinite self-association

Explicit expressions are derived which describe the binding of a univalent ligand to equivalent and independent sites on each state of an acceptor undergoing indefinite self-association that is governed by an isodesmic equilibrium constant KI. From considerations of systems in which the same site-binding constant kA applies to all acceptor-ligand interactions, the general forms of binding curves and Scatchard plots are deduced for situations in which binding sites are either created or lost at each monomer-monomer interface. Greater generality is then introduced into the model by allowing ligand interactions with polymeric acceptor states to be governed by a site-binding constant kp that differs in magnitude from that for monomeric acceptor kA. Finally, experimental results with the glutamate dehydrogenase-GTP and lysozyme-saccharide systems are used to illustrate ways in which the present quantitative expressions may be applied to the characterization of inteactions between a ligand and an indefinitely self-associating acceptor.     

Screening and characterization of human monoglyceride lipase active site inhibitors using orthogonal binding and functional assays

Endocannabinoids such as 2-arachidonylglycerol (2-AG) are ligands for cannabinoid receptors that contribute to the transmission and modulation of pain signals. The antinociceptive effect of exogenous 2-AG suggests that inhibition of monoglyceride lipase (MGLL), the enzyme responsible for degrading 2-AG and arresting signaling, may be a target for pain modulation. Here we describe the characterization of MGLL ligands following a high-throughput screening campaign. Ligands were discovered using ThermoFluor, a label-free affinity-based screening tool that measures ligand binding via modulation of protein thermal stability. A kinetic fluorescent assay using the substrate 4-methylcoumarin butyrate was used to counterscreen confirmed HTS positives. A comparison of results from binding and inhibition assays allowed elucidation of compound mechanism of action. We demonstrate the limit of each technology and the benefits of using orthogonal assay techniques in profiling compounds.     

Profiling of toxicity and identification of distinct apoptosis profiles using a 384-well high-throughput flow cytometry screening platform

Methods and techniques used to detect apoptosis have benefited from advances in technologies such as flow cytometry. With a large arsenal of lasers, fluorescent labels, and readily accessible biological targets, it is possible to detect multiple targets with unique combinations of fluorescent spectral signatures from a single sample. Traditional flow cytometry has been limited as a screening tool as the sample throughput has been low, whereas the data analysis and generation of screening relevant results have been complex. The HTFC Screening System running ForeCyt software is an instrument platform designed to perform high-throughput, multiplexed screening with seamless transformation of flow cytometry data into screening hits. We report the results of a screen that simultaneously quantified caspase 3/7 activation, annexin V binding, cell viability, and mitochondrial integrity. Assay performance over 5 days demonstrated robustness, reliability, and performance of the assay. This system is high throughput in that a 384-well plate can be read and fully analyzed within 30 min and is sensitive with an assay window of at least 10-fold for all parameters and a Z' factor of ≥0.75 for all endpoints and time points. From a screen of 231 compounds, 11 representative toxicity profiles highlighting differential activation of apoptotic pathways were identified.     

Virtual screening for novel COX-2 inhibitors using the ZINC database

Cyclooxygenase-2 (COX-2) enzyme binds to arachidonic acid and releases metabolites that are used to induce pain and inflammation. COX-2 selective inhibitors such as celecoxib, rofecoxib and valdecoxib are currently used to reduce inflammatory response. However, they lack anti-thrombotic activity and hence lead to cardiovascular and renal liabilities apart from gastrointestinal irritation. Therefore, there is still a need to develop more potent COX-2 inhibitors. In this paper, we report the screening of various compounds from the ZINC database (contains 4.6 million small molecule compounds) using the eHiTS (electronic High Throughput Screening) software tool against the COX-2 protein. The strategy employed can be conveniently split into two categories, viz. screening and docking, respectively. Screening was performed using molecular constraints tool to filter compounds with physico-chemical properties similar to the 6COX bound ligand SC-558. The analysis resulted in 1042 Lipinski compliant hits which are docked and scored to identify structurally novel ligands that make similar interactions to those of known ligands or may have different interactions with other parts of the binding site. Our screening approach identified two molecules ZINC00663976 (eHITS score of -7.135 kcal/mol) and ZINC02062094 (eHITS score of -7.242 kcal/mol) from the ZINC database. Their energy scores are better than the 6COX bound co-crystallized ligand SC-558 with an eHiTS score of -6.559 kcal/mol. Both the ligands were docked within the binding pocket forming interactions with Leu352, Phe518, Met522, Val523, Ala527 and Ser353. Visual inspection suggested similar orientation and binding mode for ZINC02062094 with SC-558 ligand. The NH group of the ligand formed hydrogen bond interactions with the backbone NH of Ala527.