The binding site distance test score: a robust method for the assessment of predicted protein binding sites

We propose a novel method for scoring the accuracy of protein binding site predictions – the Binding-site Distance Test (BDT) score. Recently, the Matthews Correlation Coefficient (MCC) has been used to evaluate binding site predictions, both by developers of new methods and by the assessors for the community wide prediction experiment – CASP8. Whilst being a rigorous scoring method, the MCC does not take into account the actual 3D location of the predicted residues from the observed binding site. Thus, an incorrectly predicted site that is nevertheless close to the observed binding site will obtain an identical score to the same number of nonbinding residues predicted at random. The MCC is somewhat affected by the subjectivity of determining observed binding residues and the ambiguity of choosing distance cutoffs. By contrast the BDT method produces continuous scores ranging between 0 and 1, relating to the distance between the predicted and observed residues. Residues predicted close to the binding site will score higher than those more distant, providing a better reflection of the true accuracy of predictions. The CASP8 function predictions were evaluated using both the MCC and BDT methods and the scores were compared. The BDT was found to strongly correlate with the MCC scores whilst also being less susceptible to the subjectivity of defining binding residues. We therefore suggest that this new simple score is a potentially more robust method for future evaluations of protein-ligand binding site predictions.

Pharmacological characterisation of the agonist radioligand binding site of 5-HT2A, 5-HT2B and 5-HT2C receptors

In the present study we compared the affinity of various drugs for the high affinity "agonist-preferring" binding site of human recombinant 5-HT2A, 5-HT2B and 5-HT2C receptors stably expressed in monoclonal mammalian cell lines. To ensure that the "agonist-preferring" conformation of the receptor was preferentially labelled in competition binding experiments, saturation analysis was conducted using antagonist and agonist radiolabels at each receptor. Antagonist radiolabels ([H-3]-ketanserin for 5-HT2A receptor and [H-3]-mesulergine for 5-HT2B and 5-HT2C receptor) bound to a larger population of receptors in each preparation than the corresponding agonist radiolabel ([I-125]-DOI for 5-HT2A receptor binding and [H-3]-5-HT for 5-HT2B and 5-HT2C receptor binding). Competition experiments were subsequently conducted against appropriate concentrations of the agonist radiolabels bound to the "agonist-preferring" subset of receptors in each preparation. These studies confirmed that there are a number of highly selective antagonists available to investigate 5-HT2 receptor subtype function (for example, MDL 100907, RS-127445 and RS-102221 for 5-HT2A, 5-HT2B and 5-HT2C receptors respectively). There remains, however, a lack of highly selective agonists. (-)DOI is potent and moderately selective for 5-HT2A receptors, BW723C86 has poor selectivity for human 5-HT2B receptors, while Org 37684 and VER-3323 display some selectivity for the 5-HT2C receptor. We report for the first time in a single study, the selectivity of numerous serotonergic drugs for 5-HT2 receptors from the same species, in mammalian cell lines and using, exclusively, agonist radiolabels. The results indicate the importance of defining the selectivity of pharmacological tools, which may have been over-estimated in the past, and highlights the need to find more selective agonists to investigate 5-HT2 receptor pharmacology.

Predicting metal-binding site residues in low-resolution structural models

The accurate prediction of the biochemical function of a protein is becoming increasingly important, given the unprecedented growth of both structural and sequence databanks. Consequently, computational methods are required to analyse such data in an automated manner to ensure genomes are annotated accurately. Protein structure prediction methods, for example, are capable of generating approximate structural models on a genome-wide scale. However, the detection of functionally important regions in such crude models, as well as structural genomics targets, remains an extremely important problem. The method described in the current study, MetSite, represents a fully automatic approach for the detection of metal-binding residue clusters applicable to protein models of moderate quality. The method involves using sequence profile information in combination with approximate structural data. Several neural network classifiers are shown to be able to distinguish metal sites from non-sites with a mean accuracy of 94.5%. The method was demonstrated to identify metal-binding sites correctly in LiveBench targets where no obvious metal-binding sequence motifs were detectable using InterPro. Accurate detection of metal sites was shown to be feasible for low-resolution predicted structures generated using mGenTHREADER where no side-chain information was available. High-scoring predictions were observed for a recently solved hypothetical protein from Haemophilus influenzae, indicating a putative metal-binding site.

FunFOLDQA: a quality assessment tool for protein-ligand binding site residue predictions

The estimation of prediction quality is important because without quality measures, it is difficult to determine the usefulness of a prediction. Currently, methods for ligand binding site residue predictions are assessed in the function prediction category of the biennial Critical Assessment of Techniques for Protein Structure Prediction (CASP) experiment, utilizing the Matthews Correlation Coefficient (MCC) and Binding-site Distance Test (BDT) metrics. However, the assessment of ligand binding site predictions using such metrics requires the availability of solved structures with bound ligands. Thus, we have developed a ligand binding site quality assessment tool, FunFOLDQA, which utilizes protein feature analysis to predict ligand binding site quality prior to the experimental solution of the protein structures and their ligand interactions. The FunFOLDQA feature scores were combined using: simple linear combinations, multiple linear regression and a neural network. The neural network produced significantly better results for correlations to both the MCC and BDT scores, according to Kendall’s τ, Spearman’s ρ and Pearson’s r correlation coefficients, when tested on both the CASP8 and CASP9 datasets. The neural network also produced the largest Area Under the Curve score (AUC) when Receiver Operator Characteristic (ROC) analysis was undertaken for the CASP8 dataset. Furthermore, the FunFOLDQA algorithm incorporating the neural network, is shown to add value to FunFOLD, when both methods are employed in combination. This results in a statistically significant improvement over all of the best server methods, the FunFOLD method (6.43%), and one of the top manual groups (FN293) tested on the CASP8 dataset. The FunFOLDQA method was also found to be competitive with the top server methods when tested on the CASP9 dataset. To the best of our knowledge, FunFOLDQA is the first attempt to develop a method that can be used to assess ligand binding site prediction quality, in the absence of experimental data.

Proteins and their interacting partners: an introduction to protein–ligand binding site prediction methods

Elucidating the biological and biochemical roles of proteins, and subsequently determining their interacting partners, can be difficult and time consuming using in vitro and/or in vivo methods, and consequently the majority of newly sequenced proteins will have unknown structures and functions. However, in silico methods for predicting protein–ligand binding sites and protein biochemical functions offer an alternative practical solution. The characterisation of protein–ligand binding sites is essential for investigating new functional roles, which can impact the major biological research spheres of health, food, and energy security. In this review we discuss the role in silico methods play in 3D modelling of protein–ligand binding sites, along with their role in predicting biochemical functionality. In addition, we describe in detail some of the key alternative in silico prediction approaches that are available, as well as discussing the Critical Assessment of Techniques for Protein Structure Prediction (CASP) and the Continuous Automated Model EvaluatiOn (CAMEO) projects, and their impact on developments in the field. Furthermore, we discuss the importance of protein function prediction methods for tackling 21st century problems.

A proposed EPR approach to evaluating agonist binding site of a peptide receptor

Angiotensin II (Ang II) and its transmembrane AT(1) receptor were selected in order to test an innovative strategy that might allow the assessment of the agonist binding site in the receptor molecule. With the use of the 2,2,6,6-tetramethylpiperidine-1-oxyl-4-amino-4-carboxylic acid (TOAC) paramagnetic probe, a biologically active agonist (TOAC(1)-Ang II), as well as an inactive control (TOAC(4)-Ang II) analogs were mixed in solution with various synthesized AT(1) fragments. Comparative intermolecular interactions, as estimated by analyzing the EPR spectra of solutions, suggested the existence of an agonist binding site containing a sequence composed of portions of the N-terminal (13-17) and the third extracellular loop (266-278) fragments of the AT(1) molecule. Therefore, this combined EPR-TOAC approach shows promise as an alternative for use also in other applications related to specific intermolecular association processes.

The role in the substrate specificity and catalysis of residues forming the substrate aglycone-binding site of a beta-glycosidase

The relative contributions to the specificity and catalysis of aglycone, of residues E190, E194, K201 and M453 that form the aglycone-binding site of a beta-glycosidase from Spodoptera frugiperda (EC 3.2.1.21), were investigated through site-directed mutagenesis and enzyme kinetic experiments. The results showed that E190 favors the binding of the initial portion of alkyl-type aglycones (up to the sixth methylene group) and also the first glucose unit of oligosaccharidic aglycones, whereas a balance between interactions with E194 and K201 determines the preference for glucose units versus alkyl moieties. E194 favors the binding of alkyl moieties, whereas K201 is more relevant for the binding of glucose units, in spite of its favorable interaction with alkyl moieties. The three residues E190, E194 and K201 reduce the affinity for phenyl moieties. In addition, M453 favors the binding of the second glucose unit of oligosaccharidic aglycones and also of the initial portion of alkyl-type aglycones. None of the residues investigated interacted with the terminal portion of alkyl-type aglycones. It was also demonstrated that E190, E194, K201 and M453 similarly contribute to stabilize ES double dagger. Their interactions with aglycone are individually weaker than those formed by residues interacting with glycone, but their joint catalytic effects are similar. Finally, these interactions with aglycone do not influence glycone binding.

H19-DMR allele-specific methylation analysis reveals epigenetic heterogeneity of CTCF binding site 6 but not of site 5 in head-and-neck carcinomas: A pilot case-control analysis

Aberrant methylation of seven potential binding sites of the CTCF factor in the differentially methylated region upstream of the H19 gene (H19-DMR) has been suggested as critical for the regulation of IGF2 and H19 imprinted genes. In this study, we analyzed the allele-specific methylation pattern of CTCF binding sites 5 and 6 using methylationsensitive restriction enzyme PCR followed by RFLP analysis in matched tumoral and lymphocyte DNA from head-and-neck squamous cell carcinoma (HNSCC) patients, as well as in lymphocyte DNA from control individuals who were cancer-free. The monoallelic methylation pattern was maintained in CTCF binding site 5 in 22 heterozygous out of 91 samples analyzed. Nevertheless, a biallelic methylation pattern was detected in CTCF binding site 6 in a subgroup of HNSCC patients as a somatic acquired feature of tumor cells. An atypical biallelic methylation was also observed in both tumor and lymphocyte DNA from two patients, and at a high frequency in the control group (29 out of 64 informative controls). Additionally, we found that the C/T transition detected by HhaI RFLP suppressed one dinucleotide CpG in critical CTCF binding site 6, of a mutation showing polymorphic frequencies. Although a heterogeneous methylation pattern was observed after DNA sequencing modified by sodium bisulfite, the biallelic methylation pattern was confirmed in 9 out of 10 HNSCCs. These findings are likely to be relevant in the epigenetic regulation of the DMR, especially in pathological conditions in which the imprinting of IGF2 and H19 genes is disrupted.

DNA methylation in the CTCF-binding site I and the expression pattern of the H19 gene: Does positive expression predict poor prognosis in early stage head and neck carcinomas?

Loss of allele-specific expression by the imprinted genes IGF2 and H19 has been correlated with a differentially methylated region (DMR) upstream to the H19 gene. The H19-DMR contains seven potential CCCTC-binding factor (CTCF) binding sites. CTCF is a chromatin insulator and a multifunctional transcription factor whose binding to the H19-DMR is suppressed by DNA methylation. Our study included a group of 41 head and neck squamous cell carcinoma (HNSCC) samples. The imprinting status of the H19 gene was analyzed in 11 out of 35 positive cases for H19 gene expression, and only 1 of them showed loss of imprinting. We detected a significant correlation (P=0.041, Fisher's exact test) between H19 expression and tumor recurrence. Among H19 positive cases, six were T2, in which five developed recurrence and/or metastasis. Inversely, in the group of tumors that showed no H19 gene expression, 5 out of 24 were T2 and only I presented regional recurrence. These data support the hypothesis that H19 expression could be used as a prognostic marker to indicate recurrence in early stage tumors. We also examined the methylation of the CTCF binding site 1 in a subgroup of these samples. The H19 gene silencing and loss of imprinting were not correlated with the methylation pattern of the CTCF binding site 1. However, the significant correlation between H19 expression and tumor recurrence suggest that this transcript could be a marker for the progression of HNSCC. (c) 2005 Wiley-Liss, Inc.

Probing the Specificity of Binding to the Major Nuclear Localization Sequence-binding Site of Importin-alpha Using Oriented Peptide Library Screening

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Characterization of L-type Calcium Channel Binding-Site of a new class of Calcium modulators by a Multidisciplinary approach

Many potential diltiazem related L-VDCC blockers were developed using a multidisciplinary approach. This current study was to investigate and compare diltiazem with to the newly developed compounds by mouse Langendorff-perfused heart, Ca2+-transient and on recombinant L-VDCC. Twenty particular compounds were selected by the ligand-based virtual screening procedure (LBVS). From these compounds, five of them (5b, M2, M7, M8 and P1) showed a potent and selective inotropic activity on guinea-pig left atria driven 1 Hz. Further assays displayed an interesting negative inotropic effect of M2, M8, P1 and M7 on guinea pig isolated left papillary muscle driven at 1 Hz, a relevant vasorelaxant activity of 5b, M2, M7, M8 and P1 on K+-depolarized guinea-pig ileum longitudinal smooth muscle and a significant inhibition of contraction of 5b, M2, M8 and P1 on carbachol stimulated ileum longitudinal smooth muscle. Wild-type human heart and rabbit lung α1 subunits were expressed (combined with the regulatory α2δ and β3 subunits) in Xenopus Leavis oocytes using a two-electrode voltage clamp technique. Diltiazem is a benzothiazepine Ca2+ channel blocker used clinically for its antihypertensive and antiarrhythmic effects. Previous radioligand binding assays revealed a complex interaction with the benzothiazepine binding site for M2, M7 and M8. (Carosati E. et al. J. Med Chem. 2006, 49; 5206). In agreement with this findings, the relative order of increased rates of contraction and relaxation at lower concentrations s(≤10-6M) in unpaced hearts was M7>M2>M8>P1. Similar increases in Ca2+ transient were observed in cardiomyocytes. Diltiazem showed negative inotropic effects whereas 5b had no significant effect. Diltiazem blocks Ca2+current in a use-dependent manner and facilitates the channel by accelerating the inactivation and decelerating the recovery from inactivation. In contrast to diltiazem, the new analogs had no pronounced use-dependence. Application of 100 μM M8, M2 showed ~ 10% tonic block; in addition, M8, M2 and P1 shifted the steady state inactivation in hyperpolarized direction and the current inactivation time was significantly decreased compared with control (219.6 ± 11.5 ms, 226 ± 14.5 vs. 269 ± 12.9 vs. 199.28 ± 8.19 ms). Contrary to diltiazem, the recovery from the block by M8 and M2 was comparable to control. Only P1 showed a significantly decrease of the time for the recovery from inactivation. All of the compounds displayed the same sensitivity on the Ca2+ channel rabbit lung α1 except P1. Taken together, these findings suggest that M8, M2 and P1 might directly decrease the binding affinity or allow rapid dissociation from the benzothiazepine binding site.

PU.1 is regulated by NF-kappaB through a novel binding site in a 17 kb upstream enhancer element

The majority of patients with acute myeloid leukemia (AML) still die of their disease, and novel therapeutic concepts are needed. Timely expression of the hematopoietic master regulator PU.1 is crucial for normal development of myeloid and lymphoid cells. Targeted disruption of an upstream regulatory element (URE) located several kb upstream in the PU.1 promoter decreases PU.1 expression thereby inducing AML in mice. In addition, suppression of PU.1 has been observed in specific subtypes of human AML. Here, we identified nuclear factor-kappaB (NF-kappaB) to activate PU.1 expression through a novel site within the URE. We found sequence variations of this particular NF-kappaB site in 4 of 120 AML patients. These variant NF-kappaB sequences failed to mediate activation of PU.1. Moreover, the synergistic activation of PU.1 together with CEBPB through these variant sequences was also lost. Finally, AML patients with such variant sequences had suppressed PU.1 mRNA expression. This study suggests that changes of a single base pair in a distal element critically affect the regulation of the tumor suppressor gene PU.1 thereby contributing to the development of AML.