Using electrostatic potentials to predict DNA-binding sites on DNA-binding proteins

A method to detect DNA-binding sites on the surface of a protein structure is important for functional annotation. This work describes the analysis of residue patches on the surface of DNA-binding proteins and the development of a method of predicting DNA-binding sites using a single feature of these surface patches. Surface patches and the DNA-binding sites were initially analysed for accessibility, electrostatic potential, residue propensity, hydrophobicity and residue conservation. From this, it was observed that the DNA-binding sites were, in general, amongst the top 10% of patches with the largest positive electrostatic scores. This knowledge led to the development of a prediction method in which patches of surface residues were selected such that they excluded residues with negative electrostatic scores. This method was used to make predictions for a data set of 56 non-homologous DNA-binding proteins. Correct predictions made for 68% of the data set.     

Cytoplasmic DNA-binding proteins

Cytoplasmic DNA-binding proteins were isolated from Chinese hamster liver, kidney and tissue culture cells by DNA-polyacrylamide chromatography. With homologous Chinese hamster DNA, and with calf thymus DNA, 1.4% of the proteins were bound to the column. With single-stranded DNA and with heterologous Micrococcus lysodeikticus DNA there was only 0.3% binding, suggesting the proteins preferentially bind to double-stranded DNA and show some sequence specificity. By a nitrocellulose filter assay the bound proteins had at least a 4- to 7-fold greater affinity for DNA than bulk cytoplasmic protein. SDS gel electrophoresis showed that specific proteins were being markedly concentrated by the column and it was primarily the high molecular weight proteins of 65 000 D and over which showed sequence specificity. Some proteins appeared in common with different organs, others were unique. These studies thus define a group of high molecular weight, cytoplasmic proteins which bind to native DNA with a degree of sequence specificity. Their possible relationship to gene regulation is discussed.     

DNA-binding studies with 6BT and 5I: implications for DNA-binding/carcinogenicity and DNA-binding/mutagenicity correlations

The divergent activities of a reported carcinogen/noncarcinogen pair of monoazo dyes related to the hepatocarcinogen Butter Yellow (DAB) are currently under investigation in our laboratories. As part of these studies we have determined (a) target organ distribution after oral dosing to rats and (b) covalent binding of 14C-labelled compound to DNA. In DNA-binding studies, 3 rat liver-metabolising systems were employed: in vivo (whole liver), isolated intact hepatocytes, and liver subcellular fractions. Distribution studies revealed that comparable levels of both compounds were detected in the liver at similar times after dosing, and these in vivo tissue concentrations were used for in vitro DNA-binding studies. At this 'in vivo equivalent dose', the carcinogen was consistently bound to DNA more effectively, and the difference (ratio of DNA binding) between the 2 compounds was far greater in vivo. In subsequent studies, covalent DNA binding to bacterial (Salmonella) DNA was assessed at the in vivo equivalent dose. In contrast to the afore-mentioned findings in mammalian systems, the carcinogen was bound less effectively to DNA, and gave fewer revertant counts/plate when the 2 compounds were bound to an equivalent extent. These data are discussed in view of their implications for DNA-binding/carcinogenicity correlations, and with respect to the relationship between DNA binding and mutagenicity in the Salmonella assay.     

Residue-level prediction of DNA-binding sites and its application on DNA-binding protein predictions

Protein-DNA interactions are crucial to many cellular activities such as expression-control and DNA-repair. These interactions between amino acids and nucleotides are highly specific and any aberrance at the binding site can render the interaction completely incompetent. In this study, we have three aims focusing on DNA-binding residues on the protein surface: to develop an automated approach for fast and reliable recognition of DNA-binding sites; to improve the prediction by distance-dependent refinement; use these predictions to identify DNA-binding proteins. We use a support vector machines (SVM)-based approach to harness the features of the DNA-binding residues to distinguish them from non-binding residues. Features used for distinction include the residue's identity, charge, solvent accessibility, average potential, the secondary structure it is embedded in, neighboring residues, and location in a cationic patch. These features collected from 50 proteins are used to train SVM. Testing is then performed on another set of 37 proteins, much larger than any testing set used in previous studies. The testing set has no more than 20% sequence identity not only among its pairs, but also with the proteins in the training set, thus removing any undesired redundancy due to homology. This set also has proteins with an unseen DNA-binding structural class not present in the training set. With the above features, an accuracy of 66% with balanced sensitivity and specificity is achieved without relying on homology or evolutionary information. We then develop a post-processing scheme to improve the prediction using the relative location of the predicted residues. Balanced success is then achieved with average sensitivity, specificity and accuracy pegged at 71.3%, 69.3% and 70.5%, respectively. Average net prediction is also around 70%. Finally, we show that the number of predicted DNA-binding residues can be used to differentiate DNA-binding proteins from non-DNA-binding proteins with an accuracy of 78%. Results presented here demonstrate that machine-learning can be applied to automated identification of DNA-binding residues and that the success rate can be ameliorated as more features are added. Such functional site prediction protocols can be useful in guiding consequent works such as site-directed mutagenesis and macromolecular docking.     

Nonintercalative DNA-binding antitumour compounds

Summary A family of compounds which appear to bind reversibly to double stranded DNA without intercalation between DNA base pairs has been defined. Methods are described by which this non-intercalative binding can be characterised using ultraviolet spectrometry, fluorimetry with ethidium as a probe, viscometry and other hydrodynamic techniques, circular dichroism and nuclear magnetic resonance spectrometry. Antibiotics which fall into this family include the antibiotics distamycin A, netropsin, mithramycin, chromomycin and olivomycin. Synthetic antitumour agents include diarylamidines such as berenil, phthalanilides, aromatic bisguanylhydrazones and bisquaternary ammonium heterocycles. A survey has been made of the general requirements of this family of compounds for DNA binding and biological activity. Binding of drugs to the minor groove of the DNA double helix appears to be the most likely mechanism for the antitumour action of these compounds.     

Novel DNA-binding element within the C-terminal extension of the nuclear receptor DNA-binding domain

The heterodimer of the ecdysone receptor (EcR) and ultraspiracle (Usp), members of the nuclear receptors superfamily, is considered as the functional receptor for ecdysteroids initiating molting and metamorphosis in insects. Here we report the 1.95Å structure of the complex formed by the DNA-binding domains (DBDs) the EcR and the Usp, bound to the natural pseudopalindromic response element. Comparison of the structure with that obtained previously, using an idealized response element, shows how the EcRDBD, which has been previously reported to possess extraordinary flexibility, accommodates DNA-induced structural changes. Part of the C-terminal extension (CTE) of the EcRDBD folds into an α-helix whose location in the minor groove does not match any of the locations previously observed for nuclear receptors. Mutational analyses suggest that the α-helix is a component of EcR-box, a novel element indispensable for DNA-binding and located within the nuclear receptor CTE. This element seems to be a general feature of all known EcRs.     

DP-Bind: a web server for sequence-based prediction of DNA-binding residues in DNA-binding proteins

This article describes DP-Bind, a web server for predicting DNA-binding sites in a DNA-binding protein from its amino acid sequence. The web server implements three machine learning methods: support vector machine, kernel logistic regression and penalized logistic regression. Prediction can be performed using either the input sequence alone or an automatically generated profile of evolutionary conservation of the input sequence in the form of PSI-BLAST position-specific scoring matrix (PSSM). PSSM-based kernel logistic regression achieves the accuracy of 77.2%, sensitivity of 76.4% and specificity of 76.6%. The outputs of all three individual methods are combined into a consensus prediction to help identify positions predicted with high level of confidence. AVAILABILITY: Freely available at http://lcg.rit.albany.edu/dp-bind. SUPPLEMENTARY INFORMATION: http://lcg.rit.albany.edu/dp-bind/dpbind_supplement.html.     

DBSI: DNA-binding site identifier

In this study, we present the DNA-Binding Site Identifier (DBSI), a new structure-based method for predicting protein interaction sites for DNA binding. DBSI was trained and validated on a data set of 263 proteins (TRAIN-263), tested on an independent set of protein-DNA complexes (TEST-206) and data sets of 29 unbound (APO-29) and 30 bound (HOLO-30) protein structures distinct from the training data. We computed 480 candidate features for identifying protein residues that bind DNA, including new features that capture the electrostatic microenvironment within shells near the protein surface. Our iterative feature selection process identified features important in other models, as well as features unique to the DBSI model, such as a banded electrostatic feature with spatial separation comparable with the canonical width of the DNA minor groove. Validations and comparisons with established methods using a range of performance metrics clearly demonstrate the predictive advantage of DBSI, and its comparable performance on unbound (APO-29) and bound (HOLO-30) conformations demonstrates robustness to binding-induced protein conformational changes. Finally, we offer our feature data table to others for integration into their own models or for testing improved feature selection and model training strategies based on DBSI.     

A novel DNA-binding motif in the nuclear matrix attachment DNA-binding protein SATB1.

The nuclear matrix attachment DNA (MAR) binding protein SATB1 is a sequence context-specific binding protein that binds in the minor groove, making virtually no contact with the DNA bases. The SATB1 binding sites consist of a special AT-rich sequence context in which one strand is well-mixed A's, T's, and C's, excluding G's (ATC sequences), which is typically found in clusters within different MARs. To determine the extent of conservation of the SATB1 gene among different species, we cloned a mouse homolog of the human STAB1 cDNA from a cDNA expression library of the mouse thymus, the tissue in which this protein is predominantly expressed. This mouse cDNA encodes a 764-amino-acid protein with a 98% homology in amino acid sequence to the human SATB1 originally cloned from testis. To characterize the DNA binding domain of this novel class of protein, we used the mouse SATB1 cDNA and delineated a 150-amino-acid polypeptide as the binding domain. This region confers full DNA binding activity, recognizes the specific sequence context, and makes direct contact with DNA at the same nucleotides as the whole protein. This DNA binding domain contains a novel DNA binding motif: when no more than 21 amino acids at either the N- or C-terminal end of the binding domain are deleted, the majority of the DNA binding activity is lost. The concomitant presence of both terminal sequences is mandatory for binding. These two terminal regions consist of hydrophilic amino acids and share homologous sequences that are different from those of any known DNA binding motifs. We propose that the DNA binding region of SATB1 extends its two terminal regions toward DNA to make direct contact with DNA.     

DNA-binding studies of fluoxetine antidepressant

Fluoxetine is a selective serotonin reuptake inhibitor (SSRI) antidepressant that is widely prescribed. The DNA-binding behavior of fluoxetine antidepressant and calf thymus DNA was investigated in Tris-HCl buffer at physiological pH 7.4 with a series of techniques, including UV-Vis and circular dichroism spectroscopies, competitive study with Hoechst 33258, viscometry, and cyclic voltammetry. Fluoxetine molecules bind to DNA via groove mode as illustrated by hypochromism with no red shift in the UV absorption band of fluoxetine, decrease in Hoechst-DNA solution fluorescence, and no significant changes in viscosity of DNA. The CD spectra of DNA molecules show a little change in stacking mode of base pair but no modification changes in DNA conformation, for example, from B-DNA to A or C-DNA. The binding constant (K(b)) of DNA with fluoxetine was calculated to be 6.7 × 10(4) M(-1), which is in the range of reported and known groove binders, such as distamycin. All results showed the groove-binding mode of interaction of fluoxetine with DNA.     

Comparative footprinting of DNA-binding proteins

MOTIVATION: Comparative modelling is a computational method used to tackle a variety of problems in molecular biology and biotechnology. Traditionally it has been applied to model the structure of proteins on their own or bound to small ligands, although more recently it has also been used to model protein-protein interfaces. This work is the first to systematically analyze whether comparative models of protein-DNA complexes could be built and be useful for predicting DNA binding sites. RESULTS: First, we describe the structural and evolutionary conservation of protein-DNA interfaces, and the limits they impose on modelling accuracy. Second, we find that side-chains from contacting residues can be reasonably modeled and therefore used to identify contacting nucleotides. Third, the DNASITE protocol is implemented and different parameters are benchmarked on a set of 85 regulators from Escherichia coli. Results show that comparative footprinting can make useful predictions based solely on structural data, depending primarily on the interface identity with respect to the template used. AVAILABILITY: DNASITE code available on request from the authors.     

Methylated DNA-binding proteins from Arabidopsis

The 5-methylcytosines (m5C) play a critical role in epigenetic control, often being recognized by proteins containing a methyl-CpG-binding domain (MBD). Database screening has identified at least 12 putative methyl-CpG-binding proteins from Arabidopsis; we have isolated corresponding cDNAs for seven of them. Despite variation in size and amino acid sequence, all seven proteins exclusively migrate into the nucleus as revealed by green fluorescent protein fusion protein assay, suggesting a relationship with chromatin structure. However, DNA-binding assays using bacterially expressed proteins and synthetic oligonucleotides containing m5C in CpGs showed only one to specifically bind, designated AtMBD5. Further analysis showed that AtMBD5 efficiently binds to m5C in CpNpN (N is A, T, or C) but not in CpNpG sequences, both frequently found in plant DNA. The other six proteins showed either nonspecific DNA binding or no ability to recognize m5C. RNA-blot hybridization and immunoblot analysis indicated AtMBD5 to be present essentially in all tissues. Using green fluorescent protein driven by the authentic promoter, AtMBD5 was found to be actively expressed in pistils and root tips. Because m5Cs in CpG and CpNpN are considered to function in gene expression and gene silencing, respectively, the present results suggest that AtMBD5 may have distinct functions in regulation and/or self defense of genes in actively proliferating cells.     

DNA-binding antibodies and parasitic diseases

DNA-binding antibodies are produced during the course of many parasitic infections, including malaria, leprosy and schistosomiasis. These antibodies are also present in certain autoimmune diseases, such as systemic lupus erythematosus, and much information is available about their properties and contribution to related disease processes Here, Anne Wozencraft and Norman Staines consider how DNA-binding antibodies might arise during parasitic infection and discuss how an increased knowledge of their properties and functions could lead to a greater understanding of mechanisms of immuno-pathology in these diseases.