Automated docking of monosaccharide substrates and analogues and methyl alpha-acarviosinide in the glucoamylase active site

Hydro-alcoholic extract of Morus nigra reduces fasting blood glucose and HbA1c% in diabetic patients, probably via competitive and allosteric interaction with alpha-glucosidase enzyme; a clinical trial and in silico analysis

OBJECTIVES

1-Deoxynojirimycin (1-DNJ), the main active component found in Morus nigra (black mulberry) is reported to be effective in controlling diabetes. We have evaluated the effect of hydro-alcoholic extract of M. nigra leaves on the fasting blood glucose (FBS) and hemoglobin A1c% (HbA1c%) in diabetic patients. Furthermore, we compared the interaction of 1-DNJ and glucose molecules with the alpha-glucosidase enzyme, which has a critical role in the lysis of glucose-based polymers in human cells.

METHODS

4% hydro-alcoholic extract was prepared from black mulberry leaves. Patients in treatment (n=50) and control (n=50) groups received 3 mL extract or placebo in water, respectively, and three times a day. Fasting blood glucose and HbA1c% were evaluated before and after three months of evaluation. Potential binding sites of 1-DNJ or glucose on the enzyme glucosidase found by docking study. Docking scores were obtained using an energy minimization method by Molegro Virtual Docker software. The Mean ± SD of each variable was compared between groups at the 95% significant level.

RESULTS

Age mean ± SD was equal to 54.79 ± 9.203 (38-69) years. There was no significant difference between intervention and placebo groups considering FBS (p=0.633) but was for HbA1c% (p=0.0011), before treatment. After three months, both FBS and HbA1c% were significantly reduced in patients under mulberry leaves extract-treatment. FBS changed was from 182.23 ± 38.65 to 161.23 ± 22.14 mg/dL in treatment group (p<0.001) and from 178.45 ± 39.46 to 166.23 ± 29.64 mg/dL in control group (p<0.001). HbA1c was changed from 7.23 ± 0.25 to 6.13 ± 0.61% in treatment group (p<0.001) and from 7.65 ± 0.85 to 7.12 ± 0.33% in control group (p=0.854). Docking results showed that 1-DNJ binds more efficiently, and with a significant score than glucose, to human alpha-glucosidase.

CONCLUSIONS

This clinical trial and virtual analysis showed that a hydro-alcoholic extract of black mulberry (M. nigra) leaf may be efficient in reducing the blood glucose and HbA1c% in diabetic patients. Furthermore, docking studies propose a competitive and allosteric regulation for herbal ingredients. Drug-development could be based on the presented idea in this report.

Synthesis, enzymatic evaluation, and docking studies of fluorogenic caspase 8 tetrapeptide substrates

The synthesis, enzymatic evaluation, and molecular modeling studies of new fluorogenic tetrapeptide-based substrates selective for caspase 8, having the general structure Ac-IETD-AXX, are described. Various fluorescent reporter groups (AXX), i.e., 3- and 4-substituted coumarins and quinolin-2(1H)-ones were synthesized by von Pechmann condensation. They were subsequently coupled with the caspase-8-selective tetrapeptide Ac-IETD-OH under newly developed synthetic conditions to give the desired substrates in good yields and in high enantiomeric purity. Based on K(M) and V(max) values, the new compounds proved to be excellent substrates for recombinant human caspase 8. In contrast, the K(M) values for the same compounds as substrates for human caspase 3 were approximately 10-20-fold higher. Molecular modeling studies based on the X-ray crystal structures of both human caspases 3 and 8 revealed that there is sufficient room within both active sites to accommodate substrates with moderately bulky substituents in the 3- and 4-positions of the fluorogenic coumarins and quinolin-2(1H)-ones. Automated docking of the substrates into the active sites of both human caspases 3 and 8 with the program AutoDock 3 gave structures similar to the published crystallographic structures for the same tetrapeptide bound to caspase 8 in the form of an irreversible inhibitor. The calculated binding energies for the new substrates to either caspase 3 or 8 showed little difference between the substrates, consistent with the K(M) data. In addition, the calculated binding energies (DeltaG) to caspase 8 were considerably more negative than those to caspase 3, also consistent with the K(M) data. A possible molecular interaction that might explain the selectivity of the IETD tetrapeptide motif for caspase 8 over caspase 3 is discussed.

Twisting of glycosidic bonds by hydrolases

Patterns of scissile bond twisting have been found in crystal structures of glycoside hydrolases (GHs) that are complexed with substrates and inhibitors. To estimate the increased potential energy in the substrates that results from this twisting, we have plotted torsion angles for the scissile bonds on hybrid Quantum Mechanics::Molecular Mechanics energy surfaces. Eight such maps were constructed, including one for alpha-maltose and three for different forms of methyl alpha-acarviosinide to provide energies for twisting of alpha-(1,4) glycosidic bonds. Maps were also made for beta-thiocellobiose and for three beta-cellobiose conformers having different glycon ring shapes to model distortions of beta-(1,4) glycosidic bonds. Different GH families twist scissile glycosidic bonds differently, increasing their potential energies from 0.5 to 9.5 kcal/mol. In general, the direction of twisting of the glycosidic bond away from the conformation of lowest intramolecular energy correlates with the position (syn or anti) of the proton donor with respect to the glycon's ring oxygen atom. This correlation suggests that glycosidic bond distortion is important for the optimal orientation of one of the glycosidic oxygen lone pairs toward the enzyme's proton donor.

Comparing programs for rigid-body multiple structural superposition of proteins

Different programs and methods were employed to superimpose protein structures, using members of four very different protein families as test subjects, and the results of these efforts were compared. Algorithms based on human identification of key amino acid residues on which to base the superpositions were nearly always more successful than programs that used automated techniques to identify key residues. Among those programs automatically identifying key residues, MASS could not superimpose all members of some families, but was very efficient with other families. MODELLER, MultiProt, and STAMP had varying levels of success. A genetic algorithm program written for this project did not improve superpositions when results from neighbor-joining and pseudostar algorithms were used as its starting cases, but it always improved superpositions obained by MODELLER and STAMP. A program entitled PyMSS is presented that includes three superposition algorithms featuring human interaction.

Modeling protein recognition of carbohydrates

We have a limited understanding of the details of molecular recognition of carbohydrates by proteins, which is critical to a multitude of biological processes. Furthermore, carbohydrate-modifying proteins such as glycosyl hydrolases and phosphorylases are of growing importance as potential drug targets. Interactions between proteins and carbohydrates have complex thermodynamics, and in general the specific positioning of only a few hydroxyl groups determines their binding affinities. A thorough understanding of both carbohydrate and protein structures is thus essential to predict these interactions. An atomic-level view of carbohydrate recognition through structures of carbohydrate-active enzymes complexed with transition-state inhibitors reveals some of the distinctive molecular features unique to protein-carbohydrate complexes. However, the inherent flexibility of carbohydrates and their often water-mediated hydrogen bonding to proteins makes simulation of their complexes difficult. Nonetheless, recent developments such as the parameterization of specific force fields and docking scoring functions have greatly improved our ability to predict protein-carbohydrate interactions. We review protein-carbohydrate complexes having defined molecular requirements for specific carbohydrate recognition by proteins, providing an overview of the different computational techniques available to model them.

Force calculations in automated docking: Enzyme-substrate interactions in Fusarium oxysporum Cel7B

AutoDock is a small-molecule docking program that uses an energy function to score docked ligands. Here AutoDock's grid-based method for energy evaluation was exploited to evaluate the force exerted by Fusarium oxysporum Cel7B on the atoms of docked cellooligosaccharides and a thiooligosaccharide substrate analog. Coupled with the interaction energies evaluated for each docked ligand, these forces give insight into the dynamics of the ligand in the active site, and help to elucidate the relative importance of specific enzyme-substrate interactions in stabilizing the substrate transition-state conformation. The processive force on the docked substrate in the F. oxysporum Cel7B active site is less than half of that on the docked substrate in the Hypocrea jecorina Cel7A active site. Hydrogen bonding interactions of the enzyme with the C2 hydroxyl group of the glucosyl residue in subsite -2 and with the C3 hydroxyl group of the glucosyl residue in subsite +1 are the most significant in stabilizing the distorted14B transition-state conformation of the glucosyl residue in subsite -1. The force calculations also help to elucidate the mechanism that prevents the active site from fouling.

Hypocrea jecorina (Trichoderma reesei) Cel7A as a molecular machine: A docking study

Hypocrea jecorina (formerly Trichoderma reesei) Cel7A has a catalytic domain (CD) and a cellulose-binding domain (CBD) separated by a highly glycosylated linker. Very little is known of how the 2 domains interact to degrade crystalline cellulose. Based on the interaction energies and forces on cello-oligosaccharides computationally docked to the CD and CBD, we propose a molecular machine model, where the CBD wedges itself under a free chain end on the crystalline cellulose surface and feeds it to the CD active site tunnel. Enzyme-substrate interactions produce the forces required to pull cellulose chains from the surface and also to help the enzyme move on the cellulose chain for processive hydrolysis. The energy to generate these forces is ultimately derived from the chemical energy of glycosidic bond breakage.

Visualizing complexes of phospholipids with Streptomyces phospholipase D by automated docking

The automated docking program AutoDock was used to dock nine phosphatidic acids (PAs), six phosphatidylcholines, five phosphatidylethanolamines, four phosphatidylglycerols, one phosphatidylinositol and two phosphatidylserines, which have two identical saturated fatty acid residues with an even numbers of carbon atoms, onto the active site of Streptomyces sp. PMF phospholipase D (PLD). Two PAs with one double bond on the fatty acid chain linked to the C2 of the glycerol residue were also docked. In general, binding energies become progressively more negative as fatty acid residues become longer. When these residues are of sufficient length, one is coiled against a hydrophobic cliff in a well that also holds the glycerol and phosphate residues and the head group, while the other generally is bound by a hydrophobic surface outside the well. Phosphatidylcholines have the only head group that is firmly bound by the active site, giving a possible structural explanation for the low selectivity of Streptomyces PLD for other phospholipid substrates.

Automated docking of ligands to an artificial active site: augmenting crystallographic analysis with computer modeling

The W191G cavity of cytochrome c peroxidase is useful as a model system for introducing small molecule oxidation in an artificially created cavity. A set of small, cyclic, organic cations was previously shown to bind in the buried, solvent-filled pocket created by the W191G mutation. We docked these ligands and a set of non-binders in the W191G cavity using AutoDock 3.0. For the ligands, we compared docking predictions with experimentally determined binding energies and X-ray crystal structure complexes. For the ligands, predicted binding energies differed from measured values by +/- 0.8 kcal/mol. For most ligands, the docking simulation clearly predicted a single binding mode that matched the crystallographic binding mode within 1.0 A RMSD. For 2 ligands, where the docking procedure yielded an ambiguous result, solutions matching the crystallographic result could be obtained by including an additional crystallographically observed water molecule in the protein model. For the remaining 2 ligands, docking indicated multiple binding modes, consistent with the original electron density, suggesting disordered binding of these ligands. Visual inspection of the atomic affinity grid maps used in docking calculations revealed two patches of high affinity for hydrogen bond donating groups. Multiple solutions are predicted as these two sites compete for polar hydrogens in the ligand during the docking simulation. Ligands could be distinguished, to some extent, from non-binders using a combination of two trends: predicted binding energy and level of clustering. In summary, AutoDock 3.0 appears to be useful in predicting key structural and energetic features of ligand binding in the W191G cavity.

Specific empirical free energy function for automated docking of carbohydrates to proteins

We present an automated docking protocol specifically optimized to predict the structure and affinity of a protein-carbohydrate complex. A scoring function was developed based on a training set of 30 protein-carbohydrate complexes of known structure and affinity. Combinations of several models for hydrogen bonding, torsional entropy loss, and solvation were tested for their ability to fit the training set data, and the best model was used with AutoDock. The electrostatic empirical coefficient is larger than in a previously obtained model using a training set comprised of various types of protein-ligand complexes, indicating that electrostatic interactions play a more important role in determining the affinity between a carbohydrate and a protein. The differences in the relative weighting of the empirical coefficients in the model yields predicted free energies for the training set with a standard error of 1.403 kcal/mol. The new scoring function was tested on 17 Aspergillus niger glucoamylase inhibitors for which binding energies had been determined experimentally. Free energies of complex formation were predicted with a residual standard error of 1.101 kcal/mol. The new scoring function therefore provides a robust method for predicting free energies of formation and optimal conformations of carbohydrate-protein complexes.

Understanding protein structure-function relationships in Family 47 alpha-1,2-mannosidases through computational docking of ligands

Family 47 alpha-1,2-mannosidases are crucial enzymes involved in N-glycan maturation in the endoplasmic reticula and Golgi apparati of eukaryotic cells. High-resolution crystal structures of the human and yeast endoplasmic reticulum alpha-1,2-mannosidases have been recently determined, the former complexed with the inhibitors 1-deoxymannojirimycin and kifunensine, both of which bind in its active site in the unusual 1C4 conformation. However, unambiguous identification of the catalytic proton donor and nucleophile involved in glycoside bond hydrolysis was not possible from this structural information. In this work, alpha-D-galactose, alpha-D-glucose, and alpha-D-mannose were computationally docked in the active site in the energetically stable 4C1 conformation as well as in the 1C4 conformation to compare their interaction energetics. From these docked structures, a model for substrate and conformer selectivity based on the dimensions of the active site was proposed. Alpha-D-galactopyranosyl-(1-->2)-alpha-D-mannopyranose, alpha-D-glucopyranosyl-(1-->2)-alpha-D-mannopyranose, and alpha-D-mannopyranosyl-(1-->2)-alpha-D-mannopyranose were also docked into the active site with their nonreducing-end residues in the 1C4 and E4 (representing the transition state) conformations. Based on the docked structure of alpha-D-mannopyranosyl-E4-(1-->2)-alpha-D-mannopyranose, the catalytic acid and base are Glu132 and Glu435, respectively.

Automated docking of alpha-(1-->4)- and alpha-(1-->6)-linked glucosyl trisaccharides and maltopentaose into the soybean beta-amylase active site

The Lamarckian genetic algorithm of AutoDock 3.0 was used to dock alpha-maltotriose, methyl alpha-panoside, methyl alpha-isopanoside, methyl alpha-isomaltotrioside, methyl alpha-(6(1)-alpha-glucopyranosyl)-maltoside, and alpha-maltopentaose into the closed and, except for alpha-maltopentaose, into the open conformation of the soybean beta-amylase active site. In the closed conformation, the hinged flap at the mouth of the active site closes over the substrate. The nonreducing end of alpha-maltotriose docks preferentially to subsites -2 or +1, the latter yielding nonproductive binding. Some ligands dock into less optimal conformations with the nonreducing end at subsite -1. The reducing-end glucosyl residue of nonproductively-bound alpha-maltotriose is close to residue Gln194, which likely contributes to binding to subsite +3. In the open conformation, the substrate hydrogen-bonds with several residues of the open flap. When the flap closes, the substrate productively docks if the nonreducing end is near subsites -2 or -1. Trisaccharides with alpha-(1-->6) bonds do not successfully dock except for methyl alpha-isopanoside, whose first and second glucosyl rings dock exceptionally well into subsites -2 and -1. The alpha-(1-->6) bond between the second and third glucosyl units causes the latter to be improperly positioned into subsite +1; the fact that isopanose is not a substrate of beta-amylase indicates that binding to this subsite is critical for hydrolysis.

Automated docking of ligands to antibodies: methods and applications

Many approaches to studying protein-ligand interactions by computational docking are currently available. Given the structures of a protein and a ligand, the ultimate goal of all docking methods is to predict the structure of the resulting complex. This requires a suitable representation of molecular structures and properties, search algorithms to efficiently scan the configuration space for favorable interaction geometries, and accurate scoring functions to evaluate and rank the generated orientations. For many of the available methods, tests on experimentally known antibody-antigen or antibody-hapten complexes have appeared in the literature. In addition, some of them have been used in predictive studies on antibody-ligand interactions to provide structural insights where adequate experimental information is missing. The AutoDock program is presented as example of a method for flexibly docking ligands to antibodies. Applying parameters of the second-generation AMBER force field, three antibody-hapten complexes (AN02, DB3, NC6.8) are used as new test cases to analyze the ability of the method to reproduce experimental findings. The X-ray structures could be reconstituted and the corresponding solutions were ranked with best energy score in all cases. Docking to the free instead of the complexed NC6.8 structure indicated the limits of the rigid protein treatment, although fairly good guesses about the location of the binding site and the contact residues could still be obtained if conformational flexibility was allowed at least in the ligand.

Automated docking of maltose, 2-deoxymaltose, and maltotetraose into the soybean beta-amylase active site

In this study, products and substrates were docked into the active site of beta-amylase using the simulated annealing algorithm AutoDock. Lowest-energy conformers reproduced known crystallographic atom positions within 0.4 to 0.8 A rmsd. Docking studies were carried out with both open and closed configurations of the beta-amylase mobile flap, a loop comprising residues 96 to 103. Ligands with two rings docked within the cleft near the active site when the flap was open, but those with four rings did not. The flap must be closed for alpha-maltotetraose to adopt a conformation allowing it to dock near the crystallographically determined subsites. The closed flap is necessary for productive but not for nonproductive binding, and therefore it plays a essential role in catalysis. The gain in total binding energy upon closing of the flap for alpha-maltose docked to subsites -2, -1 and +1, +2 is about 22 kcal/mol, indicating more favorable interactions are possible with the flap closed. Larger intermolecular interaction energies are observed for two alpha-maltose molecules docked to subsites -2, -1 and +1, +2 than for one alpha-maltotetraose molecule docked from subsites -2 to +2, suggesting that it is only upon cleavage of the alpha-1,4 linkage that optimal closed-flap binding can occur with the crytallographically determined enzyme structure.

Steady-state inhibitory kinetic studies on the ligand binding modes of Aspergillus niger glucoamylase

Inhibitory activities of 1-deoxynojirimycin and gluconolactone on Aspergillus niger glucoamylase were studied in relation to the subsite structure of the enzyme. Although both of these inhibitors are considered to bind at subsite 1 of the enzyme active site, 1-deoxynojirimycin showed competitive type inhibition but gluconolactone was a mixed type (or noncompetitive type) inhibitor for the hydrolysis of p-nitrophenyl alpha-D-glucoside. The former type of inhibition suggested that the main binding mode of the substrate was productive, but the latter, nonproductive. A possible way of explaining these apparent inconsistent results is to assume that the main binding mode of the substrate is productive and gluconolactone forms a nonproductive ternary complex with the enzyme and the substrate.

Ligand binding by antibody IgE Lb4: assessment of binding site preferences using microcalorimetry, docking, and free energy simulations

Antibody IgE Lb4 interacts favorably with a large number of different compounds. To improve the current understanding of the structural basis of this vast cross-reactivity, the binding of three dinitrophenyl (DNP) amino acids (DNP-alanine, DNP-glycine, and DNP-serine) is investigated in detail by means of docking and molecular dynamics free energy simulations. Experimental binding energies obtained by isothermal titration microcalorimetry are used to judge the results of the computational studies. For all three ligands, the docking procedure proposes two plausible subsites within the binding region formed by the antibody CDR loops. By subsequent molecular dynamics simulations and calculations of relative free energies of binding, one of these subsites, a tyrosine-surrounded pocket, is revealed as the preferred point of complexation. For this subsite, results consistent with experimental observations are obtained; DNP-glycine is found to bind better than DNP-serine, and this, in turn, is found to bind better than DNP-alanine. The suggested binding mode makes it possible to explain both the moderate binding affinity and the differences in binding energy among the three ligands.

A minimized Fc binding peptide from protein A induces immunocyte proliferation and evokes Th1-type response in mice

It is now well established that PA is a potent biological response modifier, showing simultaneously antitumor, antitoxic, anticarcinogenic, antifungal, antiparasitic and immunomodulatory properties. Since PA is a foreign protein, it is quite logical to assume that it may be cleaved into smaller peptide fragments in vivo which may be responsible for biological activities of whole PA molecule. The present study was undertaken to dissect out the structural entities of PA responsible for its biological properties. Protein A (PA) of Staphylococcus aureus has a unique property of binding with immunoglobulins. On the basis of molecular modeling and energy minimization studies a 20-mer tryptic fragment (theoretical) was predicted to retain IgG binding capacity which has been verified by immunoblot. This peptide sequence was selected to carry out experimental studies to show its functional mimicry of PA. We observed in the sera of 20-mer peptide treated mice that the concentrations of IFNgamma, TNFalpha and IL1alpha increase to a peak level by 4 h; on the other hand, there was a decrease in IL4, IL6 and IL10 concentrations at the same time (4 h). The ratio of IFNgamma to IL4 showed Th1 type of response with the peptide as well as with that of PA. The nitric oxide concentration in sera also increases and the peak increase was in 6 h with both the peptide and PA. Cell cycle analysis using FACS shows that 20 micrograms dose of peptide was non-toxic to thymocytes and spleenocytes; on the other hand, it was immunoproliferative, shifting the thymocytes and spleenocytes from G0/G1 to S phase of the cell cycle. Further studies are in progress to evaluate other biological properties of the peptide, to evaluate if this peptide could be used as a substitute of PA to mimic at least some of its biological activities.

Molecular modeling and experimental approaches toward designing a minimalist protein having Fc-binding activity of Staphylococcal protein A

Protein A (PA), a cell wall constituent of Staphylococcus aureus, has got the unique property of binding with the Fc fragment of IgG from various species. The sequence data indicate five highly homologous Fc-binding regions in protein A. Computer sequence analysis provided the tryptic and chymotryptic fragments of IgG-binding domains of protein A. Molecular modeling in conjunction with molecular mechanical calculation has been used to search for the smallest possible proteolytic fragments of PA, still retaining Fc-binding activity. A 20-residue peptide (typtic fragment) and a 16-residue peptide (chymotryptic fragment) have been indicated, by molecular modeling studies, to possess IgG-binding affinity comparable to that of the B domain of Protein A. Binding of a 20-residue peptide has been substantiated experimentally by immunoprecipitation, capillary electrophoresis, and circular dichroism spectroscopic analyses.