Improving the iterative Linear Interaction Energy approach using automated recognition of configurational transitions

Recently an iterative method was proposed to enhance the accuracy and efficiency of ligand-protein binding affinity prediction through linear interaction energy (LIE) theory. For ligand binding to flexible Cytochrome P450s (CYPs), this method was shown to decrease the root-mean-square error and standard deviation of error prediction by combining interaction energies of simulations starting from different conformations. Thereby, different parts of protein-ligand conformational space are sampled in parallel simulations. The iterative LIE framework relies on the assumption that separate simulations explore different local parts of phase space, and do not show transitions to other parts of configurational space that are already covered in parallel simulations. In this work, a method is proposed to (automatically) detect such transitions during the simulations that are performed to construct LIE models and to predict binding affinities. Using noise-canceling techniques and splines to fit time series of the raw data for the interaction energies, transitions during simulation between different parts of phase space are identified. Boolean selection criteria are then applied to determine which parts of the interaction energy trajectories are to be used as input for the LIE calculations. Here we show that this filtering approach benefits the predictive quality of our previous CYP 2D6-aryloxypropanolamine LIE model. In addition, an analysis is performed of the gain in computational efficiency that can be obtained from monitoring simulations using the proposed filtering method and by prematurely terminating simulations accordingly.     

Impairment of human CYP1A2-mediated xenobiotic metabolism by Antley-Bixler syndrome variants of cytochrome P450 oxidoreductase

Y459H and V492E mutations of cytochrome P450 reductase (CYPOR) cause Antley-Bixler syndrome due to diminished binding of the FAD cofactor. To address whether these mutations impaired the interaction with drug-metabolizing CYPs, a bacterial model of human liver expression of CYP1A2 and CYPOR was implemented. Four models were generated: POR^null, POR^wt, POR^YH, and POR^VE, for which equivalent CYP1A2 and CYPOR levels were confirmed, except for POR^null, not containing any CYPOR. The mutant CYPORs were unable to catalyze cytochrome c and MTT reduction, and were unable to support EROD and MROD activities. Activity was restored by the addition of FAD, with V492E having a higher apparent FAD affinity than Y459H. The CYP1A2-activated procarcinogens, 2-aminoanthracene, 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone, and 2-amino-3-methylimidazo(4,5-f)quinoline, were significantly less mutagenic in POR^YH and POR^VE models than in POR^wt, indicating that CYP1A2, and likely other drug-metabolizing CYPs, are impaired by ABS-related POR mutations as observed in the steroidogenic CYPs.     

The transition of bovine trypsinogen to a trypsin-like state upon strong ligand binding. II. The binding of the pancreatic trypsin inhibitor and of isoleucine-valine and of sequentially related peptides to trypsinogen and to p-guanidinobenzoate-trypsinogen.

p-Guanidinobenzoate-trypsinogen is transformed into a trypsin-like conformation upon binding of Ile-Val as evidenced by specific changes in its circular dichroism spectrum. By means of this signal the association constants for the binding of a variety of peptides sequentially analogous to either the bovine trypsin N-terminus or to the N-terminal activation peptide sequences of several trypsinogens have been determined at different Ca²⁺ concentrations. Ile-Val and Ile-Val-Gly exhibit the strongest binding affinity of all peptides investigated. Replacement of the first isoleucine or of the second valine residue by other amino acids considerably reduces the peptide affinity. Discussion of these is based on the known spatial arrangement of the Ile16-Val17-Gly18 N-terminus and of the Ile-Val dipeptide in the Ile16 cleft (crystal structures of bovine trypsin and of the trypsinogen-PTI³-Ile-Val complex; Bode et al., 1978). The free energies of binding of the first and of the second peptide residue are almost additive indicating independency between both subsites. The third residue, glycine, does not significantly contribute to binding. The peptide analogues of various trypsinogen N-termini exhibit no measurable affinity for the Ile 16 cleft.The equilibrium constant for the binding of PTI to trypsinogen and the affinity of Ile-Val for the resulting binary complex have been determined in the presence and absence of Ca²⁺, using the competitive PTI-binding to α-chymotrypsin. These competition experiments allow the estimation of the standard free-energy changes due to the conformational transition of trypsinogen into a trypsin-like state (+43 kJ mol^?1, 20 °C; stabilization of the “activation domain”; Fehlhammer et al., 1977), due to the binding of the trypsin N-terminus (—55 kJ mol^?1) and of the peptide analogues (e.g. Ile-Val; ?28 kJ mol^?1) into the preformed Ile 16 cleft, and due to the specific burying of the covalently linked pGB group in the fixed specificity pocket (— 39 kJ mol^?1). This pocket is co-operatively linked with the Ile 16 cleft according to a free-energy change coupling of —43 kJ mol^?1.     

A calorimetric study of the binding of 2'-deoxyuridine-5'-phosphate and 5-fluoro-2'-deoxyuridine-5'-phosphate to thymidylate synthetase.

The thermodynamic parameters, ΔH′, ΔG′, and ΔS′, and the stoichiometry for the binding of the substrate 2′-deoxyuridine-5′-phosphate (dUMP) and the inhibitor 5-fluoro-2′-deoxyuridine-5′-phosphate (FdUMP) to Lactobacillus casei thymidylate synthetase (TSase) have been investigated using both direct calorimetric methods and gel filtration methods. The data obtained show that two ligand binding sites are available but that the binding of the second mole of dUMP is extremely weak. Binding of the first mole of dUMP can best be illustrated by dUMP + TSase + H⁺?(dUMP-TSase-H⁺). [1] The enthalpy, ΔH₁′, for reaction [1] was measured directly on a flow modification of a Beckman Model 190B microcalorimeter. Experiments in two different buffers (I = 0.10 m) show that ΔH₁′ = ?28 kJ mol^?1 and that 0.87 mol of protons enters into the reaction. Analysis of thermal titrations for reaction [1] indicates a free energy change of ΔG₁′ = ?30 kJ mol^?1 (K₁ = 1.7 × 10⁵ m^?1). From these parameters, ΔS₁′ was calculated to be +5 J mol^?1 degree^?1, showing that the reaction is almost totally driven by enthalpy changes. Gel filtration experiments show that at very high substrate concentrations, binding to a second site can be observed. Gel filtration experiments performed at low ionic strength (I = 0.05 m) reveal a stronger binding, with ΔG₁′ = ?35 kJ mol^?1 (K₁ = 1.2 × 10⁶ m^?1), suggesting that the forces driving the interaction are, in part, electrostatic. Addition of 2-mercaptoethanol (0.10 m) had the effect of slightly increasing the dUMP binding constant. Binding of FdUMP to TSase is best illustrated by 2FdUMP + TSase + n_HH⁺?FdUMP₂ ? TSase ? (H⁺)_nH. [2] The enthalpy for this reaction, ΔH₂, was also measured calorimetrically and found to be ?30 kJ mol^?1 with n_H = 1.24 at pH 7.4 Assuming two FdUMP binding sites per dimer as established by Galivan et al. [Biochemistry15, 356–362 (1976)] our calorimetric results indicate different binding energies for each site. Based on the binding data, a thermodynamic model is presented which serves to rationalize much of the confusing physical and chemical data characterizing thymidylate synthetase.     

Binding of simple carbohydrates and some N-acetyllactosamine-containing oligosaccharides to Erythrina cristagalli agglutinin as followed with a fluorescent indicator ligand

Erythrina cristagalli agglutinin, a dimeric lectin [J. L. Iglesias, et al. (1982) Eur. J. Biochem.123, 247–252] was shown by equilibrium dialysis to be bivalent for 4-methylumbelliferyl-β-d-galactoside. Upon binding to the lectin, this ligand showed a difference absorption spectrum with two maxima (at 322 and 336 nm) of equal intensity (Δ? = 1.2 × 10³m^?1 cm^?1). A similar spectrum with a comparable value of Δ? was obtained with 4-methylumbelliferyl-N-acetyl-β-d-galactosaminide. Binding of methyl-α-d-galactoside, lactose, and N-acetyllactosamine all produced small but equally intense protein difference spectra with a maximum (Δ? = 2.8 × 10² M^?1 cm^?1) at 291.6 nm. Upon binding of N-dansyl-d-galactosamine to the lectin, there was a fivefold increase in fluorescence intensity of this ligand. The association constant for N-dansyl-d-galactosamine was caused by a very favorable ΔS° of the dansyl group without affecting the strictly carbohydrate-specific character of binding. N-Dansyl-d-galactosamine was employed as a fluorescent indicator ligand in substitution titrations. This involved the use of simple carbohydrates, N-acetyllactosamine, and oligosaccharides which occur in the carbohydrate units of N-glycoproteins; the latter were Gal(β → 4)GlcNAc(β1 → 2)Man, Gal(β1 → 4)GlcNAc(β1 → 6)Man, and Gal(β1 → 4)GlcNAc(β1 → 6)[Gal(β1 → 4)GlcNAc(β1 → 2)]Man. The titrations were performed at two temperatures to determine the thermodynamic parameters. In the series N-acetyl-d-galactosamine, methyl-α-d-galactoside, and lactose, ?ΔH° increased from 24 to 41 kJ mol^?1; it increased further for N-acetyllactosamine and then remained unchanged for the N-acetyllactosamine-containing oligosaccharides (55 ± 1 kJ mol^?1). This indicated that the site specifically accommodated the disaccharide structure with an important contribution of the 2-acetamido group in the penultimate sugar. Beyond this, no additional contacts seemed to be formed. This conclusion also followed from considerations of ΔS° values which became more unfavorable in the above series (?23 to ?101 ± 4 J mol^?1 K^?1); the most negative value of ΔS° was observed with N-acetyllactosamine and the three N-acetyllactosamine-containing oligosaccharides.     

Characterization of bi-functional CYP154 from Nocardia farcinica IFM10152 in the O-dealkylation and ortho-hydroxylation of formononetin

Among the 27 cytochrome P450s (CYPs) of Nocardia farcinica IFM10152, three CYPs have been identified as having O-dealkylation catalytic activity. Of the two that encode CYP154 subfamilies, the one encoded by the nfa22930 gene showed distinct O-dealkylation and subsequent hydroxylation of formononetin. Firstly, formononetin was O-dealkylated into daidzein, which was subsequently mono-hydroxylated at the 3′-position of the B-ring into ortho-dihydroxy-isoflavone. Apparent k_cat/K_m values of CYP154 for the O-dealkylation of formononetin and the hydroxylation of daidzein were 3.57 and 1.84 μM⁻¹ min⁻¹, respectively. The dissociation constants of CYP154 based on spectral changes upon binding to each substrate were 5.16 and 3.11 μM, respectively. Homology modeling and docking simulation found that Thr247 is responsible for the 3′-position hydroxylation reaction by forming a hydrogen bond with the 4′-hydroxyl group of daidzein that forces the proton at the 3′-position to face the heme center. Site-directed mutagenesis of Thr247 to alanine drastically decreased the binding affinity for daidzein (9.73 μM) as well as 3′-position hydroxylation catalytic activity by 3 fold (0.48 μM⁻¹ min⁻¹).     

An efficient computational method for calculating ligand binding affinities

Virtual compound screening using molecular docking is widely used in the discovery of new lead compounds for drug design. However, the docking scores are not sufficiently precise to represent the protein-ligand binding affinity. Here, we developed an efficient computational method for calculating protein-ligand binding affinity, which is based on molecular mechanics generalized Born/surface area (MM-GBSA) calculations and Jarzynski identity. Jarzynski identity is an exact relation between free energy differences and the work done through non-equilibrium process, and MM-GBSA is a semimacroscopic approach to calculate the potential energy. To calculate the work distribution when a ligand is pulled out of its binding site, multiple protein-ligand conformations are randomly generated as an alternative to performing an explicit single-molecule pulling simulation. We assessed the new method, multiple random conformation/MM-GBSA (MRC-MMGBSA), by evaluating ligand-binding affinities (scores) for four target proteins, and comparing these scores with experimental data. The calculated scores were qualitatively in good agreement with the experimental binding affinities, and the optimal docking structure could be determined by ranking the scores of the multiple docking poses obtained by the molecular docking process. Furthermore, the scores showed a strong linear response to experimental binding free energies, so that the free energy difference of the ligand binding (ΔΔG) could be calculated by linear scaling of the scores. The error of calculated ΔΔG was within ≈±1.5 kcal•mol⁻¹ of the experimental values. Particularly, in the case of flexible target proteins, the MRC-MMGBSA scores were more effective in ranking ligands than those generated by the MM-GBSA method using a single protein-ligand conformation. The results suggest that, owing to its lower computational costs and greater accuracy, the MRC-MMGBSA offers efficient means to rank the ligands, in the post-docking process, according to their binding affinities, and to compare these directly with the experimental values.     

Heterologous expression of CYP102A5 variant from Bacillus cereus CYPPB-1: Validation of model for predicting drug metabolism of human P450 probe substrates

CYP102A5 variant (ADL27534) from isolated Bacillus cereus CYPPB-1 was heterologously expressed in Escherichia coli Top 10 cells. Comparative sequence analysis of purified CYP102A5 variant with respect to reported CYP102A5 (AAP10153) from Bacillus cereus ATCC 14579 revealed amino acid sequence changes at positions P245S and M318I of heme domain. The binding affinities of 15 selected human P450 probe substrates towards isolated CYP102A5 were analyzed in silico using a homology model together with molecular docking techniques to predict the human drug metabolism. In vitro analysis suggested that the purified CYP102A5 metabolizes typical substrates of human CYP2C9, CYP2D6, CYP2E1, and CYP3A4, such as coumarin, propranolol, aniline, chlorzoxazone, p-nitrophenol, and nifedipine. The calculated K _M values for propranolol, chloroxazone, coumarin, aniline, and 4-nitrophenol were calculated to be 0.962?±?0.041, 1.254?±?0.057, 2.859?±?0.083, 2.732?±?0.106, and 2.528?±?0.11 mM, respectively. Importantly, taking a ChemScore cutoff value of ?31 kJ/mol, substrate binding at active site and in vitro activity as the distinguishing lines between “substrates” and “nonsubstrates” revealed one false-positive and one false-negative results out of the 15 compounds examined. This is the first report on validation of CYP102A family homology model for in silico prediction of human drug metabolism.     

Binding of sodium n-dodecyl sulphate and protons to catalase

The binding of sodium n-dodecyl sulphate to catalase has been measured by equilibrium dialysis in the pH range 3.2 to 10.0. On the acid side of the isoelectric point (pH 5.4) the surfactant anions initially bind to cationic sites on the protein and subsequent binding is cooperative. At high pH on the alkaline side of the isoelectric point only cooperative binding is observed. The binding data have been combined with protein titration curves to calculate the Gibbs energies of formation of protein titration curves to calculate the Gibbs energies of formation of protein surfactant proton complexes. Contributions to the Gibbs energies of complex formation by surfactant and protein binding have been estimated. The average Gibbs energies of surfactant binding to specific cationic sites are ca. 28 kJ mol^?1 and for cooperative binding ca. 15 kJ mol^?1.     

Exchange rates of pyridine on ferro- and ferriprotoporphyrin (IX) dimethylester in chloroform.

Nuclear magnetic resonance line-widths data have been used to determine the rate of solvent exchange from the first coordination sphere of ferro-and ferriprotoporphyrin(IX) dimethylester (Fe-PPD) in pyridine/chloroform. The average values of kinetic parameters for pyridine (PY) exchange indicate an SN2 mechanism tor Fe(III)-PPD(ΔH^&;#; = 36 kJ · mol⁻¹ ; ΔS^&;#; = −53 J·mol⁻¹K⁻¹; T_M(298 K) = 0.07 msec) and an SNI mechanism for Fe(II)-PPD (ΔH^&;#; = 67 kJ·mol⁻¹; ΔS^&;#; = 42 J · mol⁻¹K⁻¹; T_M(298 K) = 0.06 msec). Parallel to the accelerated ligand exchange rate at rising temperatures a redistribution of the electrons causing a transition of the metal porphyrin from the low-spin state to the high-spin state is observed. Enthalpy and entropy of the thermodynamic equilibrium between low- and high-spin Fe-PPD have been determined from experimental values of the average magnetic moment. A mean lifetime of low-spin Fe(III)-PPD was estimated from line. widths changes (T_L→H(298 K)≈ 20 msec) and the corresponding activation parameters have been obtained (ΔH^&;#;_L→H(298 K) = 26 kJ · mol⁻¹; ΔS^&;#;_L→H(298K) = −125 J · mol⁻¹K⁻¹).     

Interaction of cyproheptadine hydrochloride with human serum albumin using spectroscopy and molecular modeling methods

The interaction between cyproheptadine hydrochloride (CYP) and human serum albumin (HSA) was investigated by fluorescence spectroscopy, UV–vis absorption spectroscopy, Fourier transform infrared spectroscopy (FT‐IR) and molecular modeling at a physiological pH (7.40). Fluorescence of HSA was quenched remarkably by CYP and the quenching mechanism was considered as static quenching since it formed a complex. The association constants K_a and number of binding sites n were calculated at different temperatures. According to Förster's theory of non‐radiation energy transfer, the distance r between donor (human serum albumin) and acceptor (cyproheptadine hydrochloride) was obtained. The effect of common ions on the binding constant was also investigated. The effect of CYP on the conformation of HSA was analyzed using FT‐IR, synchronous fluorescence spectroscopy and 3D fluorescence spectra. The thermodynamic parameters ΔH and ΔS were calculated to be ?14.37 kJ mol^?1 and 38.03 J mol^?1 K^?1, respectively, which suggested that hydrophobic forces played a major role in stabilizing the HSA‐CYP complex. In addition, examination of molecular modeling indicated that CYP could bind to site I of HSA and that hydrophobic interaction was the major acting force, which was in agreement with binding mode studies. Copyright © 2012 John Wiley & Sons, Ltd.     

Thermodynamics of base interaction in (A)n and (A.U)n

Using precision scanning microcalorimetry we studied (A)_n and (A·U)_n melting in highly diluted solutions (0.3 to 5.0 mm) with different Na⁺ activity. This permitted us to determine directly the thermodynamic functions of stacking interaction in (A)_n and base-pairing in (A·U)_n. For (A-A) stacking at (A)_n melting temperature we obtained ΔH^(A)n_m = 12.6 kJ mol^?1; ΔS^(A)n_m = 41 J K^?1 mol^?1. For A·U base-pairing at a standard temperature of 298 K and 0.1 m-Na⁺ we have: ΔH^(A·U) = 34 kJ mol^?1; ΔS^(A·U) = 102 J K^?1 mol^?1ΔG^(A·U) = ?3.5 kJ mol^?1.     

<Superscript>13</Superscript>C-Methyl isocyanide as an NMR probe for cytochrome P450 active sites

The cytochromes P450 (CYPs) play a central role in many biologically important oxidation reactions, including the metabolism of drugs and other xenobiotic compounds. Because they are often assayed as both drug targets and anti-targets, any tools that provide: (a) confirmation of active site binding and (b) structural data, would be of great utility, especially if data could be obtained in reasonably high throughput. To this end, we have developed an analog of the promiscuous heme ligand, cyanide, with a ¹³CH₃-reporter attached. This ¹³C-methyl isocyanide ligand binds to bacterial (P450cam) and membrane-bound mammalian (CYP2B4) CYPs. It can be used in a rapid 1D experiment to identify binders, and provides a qualitative measure of structural changes in the active site. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

C2-symmetric inhibitors of Plasmodium falciparum plasmepsin II: synthesis and theoretical predictions

A series of C(2)-symmetric compounds with a mannitol-based scaffold has been investigated, both theoretically and experimentally, as Plm II inhibitors. Four different stereoisomers with either benzyloxy or allyloxy P1/P1' side chains were studied. Computational ranking of the binding affinities of the eight compounds was carried out using the linear interaction energy (LIE) method relying on a complex previously determined by crystallography. Within both series of isomers the theoretical binding energies were in agreement with the enzymatic measurements, illustrating the power of the LIE method for the prediction of ligand affinities prior to synthesis. The structural models of the enzyme-inhibitor complexes obtained from the MD simulations provided a basis for interpretation of further structure-activity relationships. Hence, the affinity of a structurally similar ligand, but with a different P2/P2' substituent was examined using the same procedure. The predicted improvement in binding constant agreed well with experimental results.     

Calorimetric studies of oxyhemoglobin dissociation. I. Reaction of sodium dithionite with oxygen

Calorimetric studies of the reduction of free oxygen in solution by sodium dithionite are in agreement with a stoichiometry of 2 moles Na₂S₂O₄ per mole of oxygen. The reaction is biphasic with ΔH_t - 118±7 kcal mol^?1 (?494 ± 29 kJ mol^?1). The initial phase of the reaction proceeds with an enthalpy change of ca ?20 kcal (?84 kJ) and occurs when 0.5 moles of dithionite have been added per mole dioxygen present. This could be interpreted as the enthalpy change for the addition of a single electron to form the superoxide anion. Further reduction of the oxygen to water by one or more additional steps is accompanied by an enthalpy change of ca ?100 kcal (?418. 5 kJ). Neither of these reductive phases is consistent with the formation of hydrogen peroxide as an intermediate. The reduction of hydrogen peroxide by dithionite in 0.1 M phosphate buffer, pH 7.15, is a much slower process and with an enthalpy change of ca ? 74 kcal mol^?1 (?314 kJ mol^?1). Dissociation of oxyhemoglobin induced by the reduction of free oxygen tension with dithionite also shows a stoichiometry of 2 moles dithionite per mole oxygen present and an enthalpy change of ca. ?101 ±9 kcal mol^?1 (?423± 38 kJ mol^?1). The difference in the observed enthalpies (reduction of dioxygen vs. oxyhemoglobin) has been attributed to the dissociation of oxyhemoglobin, which is 17 kcal mol^?1 (71 kJ mol^?1).     

Determination of glucose-fructose oxidoreductase activity in whole cells of Zymomonas mobilis

A method for the determination of glucose-fructose oxidoreductase (GFOR) activity in whole cells of Zymomonas mobilis is described. The K _m and the theoretical V _max for GFOR were 192 g glucose.l^-1 and 17 g gluconic acid.g^-1 cell.h^-1, respectively. The changes in enthalpy (31.1 kJ.mol^-1), entropy (0.41 kJ.K^-1), and Gibbs free energy (-97.5 kJ.mol^-1) related to glucose to gluconic acid conversion were also determined.