A new concept for multidimensional selection of ligand conformations (MultiSelect) and multidimensional scoring (MultiScore) of protein-ligand binding affinities.

In this work, eight different scoring functions have been combined with the aim of improving the prediction of protein-ligand binding conformations and affinities. The obtained scores were analyzed using multivariate statistical methods to generate expressions, with the ability (1) to select the best candidate between different docked conformations of an inhibitor (MultiSelect) and (2) to quantify the protein-ligand binding affinity (MultiScore). By use of the docking program GOLD, 40 different inhibitors were docked into the active site of three matrix metalloproteinases (MMP's), yielding a total of 120 enzyme-inhibitor complexes. For each complex, a single conformation of the inhibitor was selected using principal component analysis (PCA) for the scores obtained by the eight functions SCORE, LUDI, GRID, PMF_Score, D_Score, G_Score, ChemScore, and F_Score. Binding affinities were estimated based on partial least-squares projections onto latent structures (PLS) on the eight scores of each selected inhibitor conformation. By use of this procedure, R(2) = 0.78 and Q(2) = 0.78 were obtained when comparing experimental and calculated binding affinities. MultiSelect was evaluated by applying the same method for selecting docked conformations for 18 different protein-ligand complexes of known three-dimensional structure. In all cases, the selected ligand conformations were found to be very similar to the experimentally determined ligand conformations. A more general evaluation of MultiScore was performed using a set of 120 different protein-ligand complexes for which both the three-dimensional structures and the binding affinities were known. This approach allowed an evaluation of MultiScore independently of MultiSelect. The generality of the method was verified by obtaining R(2) = 0.68 and Q(2) = 0.67, when comparing calculated and experimental binding affinities for the 120 X-ray structures. In all cases, LUDI, SCORE, GRID, and F_Score were included as important functions, whereas the fifth function was PMF_Score and ChemScore for the MMP and X-ray models, respectively.

Structural differences of matrix metalloproteinases. Homology modeling and energy minimization of enzyme-substrate complexes.

Matrix metalloproteinases are extracellular enzymes taking part in the remodeling of extracellular matrix. The structures of the catalytic domain of MMP1, MMP3, MMP7 and MMP8 are known, but structures of enzymes belonging to this family still remain to be determined. A general approach to the homology modeling of matrix metalloproteinases, exemplified by the modeling of MMP2, MMP9, MMP12 and MMP14 is described. The models were refined using an energy minimization procedure developed for matrix metalloproteinases. This procedure includes incorporation of parameters for zinc and calcium ions in the AMBER 4.1 force field, applying a non-bonded approach and a full ion charge representation. Energy minimization of the apoenzymes yielded structures with distorted active sites, while reliable three-dimensional structures of the enzymes containing a substrate in active site were obtained. The structural differences between the eight enzyme-substrate complexes were studied with particular emphasis on the active site, and possible sites for obtaining selectivity among the MMP's are discussed. Differences in the P1' pocket are well-documented and have been extensively exploited in inhibitor design. The present work indicates that selectivity could be further improved by considering the P2 pocket as well.

Structure-property model for membrane partitioning of oligopeptides.

The aim of this study was to develop a structure-property model for membrane partitioning of oligopeptides using statistical design methods and multivariate data analysis. A set of 20 tetrapeptides with optional N-methylations at residues 2 and 4 was designed by a D-optimal design procedure. After synthesis and purification, the membrane partitioning abilities of the peptides were tested in two chromatographic systems with phospholipids as the stationary phase: immobilized artificial membrane chromatography (IAM) and immobilized liposome chromatography (ILC). The relationship between these measures and three different sets of calculated descriptors was analyzed by partial least-squares projection to latent structures (PLS). The descriptors used were the molecular surface area, Molsurf parameters, and Volsurf parameters. All three models were of good statistical quality and supported that a large hydrogen-bonding potential and the presence of a negative charge impair membrane partitioning, whereas hydrophobic parameters promote partitioning. The findings are in accordance with what has been found for absorption of known drugs and have implications for the design of peptide-like drugs with good oral bioavailability.

Towards an understanding of drug resistance in malaria: three-dimensional structure of Plasmodium falciparum dihydrofolate reductase by homology building.

A three-dimensional (3-D) model of dihydrofolate reductase (DHFR) from Plasmodium falciparum has been constructed by homology building. The model building has been based on a structural alignment of five X-ray structures of DHFR from different species. The 3-D model of the plasmodial DHFR was obtained by amino acid substitution in the human DHFR, which was chosen as template, modification of four loops (two insertions, two deletions) and subsequent energy minimization. The active site of P. falciparum DHFR was analyzed and compared to human DHFR with respect to sequence variations and structural differences. Based on this analysis the molecular consequences of point mutations known to be involved in drug resistance were discussed. The significance of the most important point mutation causing resistance, S108N, could be explained by the model, whereas the point mutations associated with enhanced resistance, N51I and C59R, seem to have a more indirect effect on inhibitor binding.

Predicting drug absorption from molecular surface properties based on molecular dynamics simulations.

PURPOSE: To develop an efficient method for generating representative conformations for calculation of the conformationally dependent molecular surface area, and to investigate the relation between this parameter and the permeability in Caco-2 cells. METHODS: High temperature molecular dynamics (MD) simulations were used to obtain 1000 conformations of six beta-blocking agents and their prodrugs. The Boltzmann averaged (B.a.) polar surface area of the 1000 conformations was correlated to the apparent permeability coefficients (Papp) for transport across filter-grown Caco-2 cells. RESULTS: Sampling of 1000 conformations during the MD simulations was sufficient for obtaining a representative set of conformations. The B.a. polar water accessible surface area (PWASA) yielded an excellent linear correlation with Papp for both series of compounds under study (R2 = 0.98). Thus, the improved permeability of the prodrugs could be explained by a reduced PWASA. The improvement of permeability after derivatization correlated positively with the size of the non-polar water accessible surface area-suggesting a synergistic effect of the cyclopropyl and the non-polar parts of the molecule to shield the polar parts from contact with water. CONCLUSIONS: An efficient method for generating the representative conformations for calculation of the B.a. polar surface area has been established. An excellent linear correlation explaining the improved permeability of the prodrugs was obtained.

Excitatory amino acid receptor ligands: resolution, absolute stereochemistry, and enantiopharmacology of 2-amino-3-(4-butyl-3-hydroxyisoxazol-5-yl)propionic acid.

(RS)-2-Amino-3-(4-butyl-3-hydroxyisoxazol-5-yl)propionic acid (Bu-HIBO, 6) has previously been shown to be an agonist at (RS)-2-amino-3-(3-hydroxy-5-methylisoxazol-4-yl)propionic acid (AMPA) receptors and an inhibitor of CaCl2-dependent [3H]-(S)-glutamic acid binding (J. Med. Chem. 1992, 35, 3512-3519). To elucidate the pharmacological significance of this latter binding affinity, which is also shown by quisqualic acid (3) but not by AMPA, we have now resolved Bu-HIBO via diastereomeric salt formation using the diprotected Bu-HIBO derivative 11 and the enantiomers of 1-phenylethylamine (PEA). The absolute stereochemistry of (S)-Bu-HIBO (7) (ee = 99.0%) and (R)-Bu-HIBO (8) (ee > 99.6%) were established by an X-ray crystallographic analysis of compound 15, a salt of (R)-PEA, and diprotected 8. Circular dichroism spectra of 7 and 8 were recorded. Whereas 7 (IC50 = 0.64 microM) and 8 (IC50 = 0.57 microM) were equipotent as inhibitors of CaCl2-dependent [3H]-(S)-glutamic acid binding, neither enantiomer showed significant affinity for the synaptosomal (S)-glutamic acid uptake system(s). AMPA receptor affinity (IC50 = 0.48 microM) and agonism (EC50 = 17 microM) were shown to reside exclusively in the S-enantiomer, 7. Compounds 7 and 8 did not interact detectably with kainic acid or N-methyl-D-aspartic acid (NMDA) receptor sites. Neither 7 nor 8 affected the function of the metabotropic (S)-glutamic acid receptors mGlu2 and mGlu4a, expressed in CHO cells. Compound 8 was shown also to be inactive at mGlu1 alpha, whereas 7 was determined to be a moderately potent antagonist at mGlu1 alpha (Ki = 110 microM) and mGlu5a (Ki = 97 microM). Using the rat cortical wedge preparation, the AMPA receptor agonist effect of 7 was markedly potentiated by coadministration of 8 at 21 degrees C, but not at 2-4 degrees C. These observations together indicate that the potentiation of the AMPA receptor agonism of 7 by 8 is not mediated by metabotropic (S)-glutamate receptors but rather by the CaCl2-dependent (S)-glutamic acid binding system, which shows the characteristics of a transport mechanism. After intravenous administration in mice, 7 (ED50 = 44 mumol/kg) was slightly more potent than AMPA (1) (ED50 = 55 mumol/kg) and twice as potent as Bu-HIBO (6) (ED50 = 94 mumol/kg) as a convulsant, whereas 8 was inactive. After subcutaneous administration in mice, Bu-HIBO (ED50 = 110 mumol/kg) was twice as potent as AMPA (ED50 = 220 mumol/kg) as a convulsant. Since 7 and Bu-HIBO (EC50 = 37 microM) are much weaker than AMPA (EC50 = 3.5 microM) as AMPA receptor agonists in vitro, the presence of a butyl group in the molecules of Bu-HIBO and 7 seems to facilitate the penetration of these compounds through the blood-brain barrier.

Conformational analysis of six- and twelve-membered ring compounds by molecular dynamics.

A molecular dynamics (MD)-based conformational analysis has been performed on a number of cycloalkanes in order to demonstrate the reliability and generality of MD as a tool for conformational analysis. MD simulations on cyclohexane and a series of methyl-substituted cyclohexanes were performed at temperatures between 400 and 1200 K. Depending on the simulation temperature, different types of interconversions (twist-boat-twist-boat, twist-boat-chair and chair-chair) could be observed, and the MD simulations demonstrated the expected correlation between simulation temperature and ring inversion barriers. A series of methyl-substituted 1,3-dioxanes were investigated at 1000 K, and the number of chair-chair interconversions could be quantitatively correlated to the experimentally determined ring inversion barrier. Similarly, the distribution of sampled minimum-energy conformations correlated with the energy-derived Boltzmann distribution. The macrocyclic ring system cyclododecane was subjected to an MD simulation at 1000 K and 71 different conformations could be sampled. These conformations were compared with the results of previously reported conformational analyses using stochastic search methods, and the MD method provided 19 out of the 20 most stable conformations found in the MM2 force field. Finally, the general performance of the MD method for conformational analysis is discussed.

Molecular mechanics calculations of proteins. Comparison of different energy minimization strategies.

A general strategy for performing energy minimization of proteins using the SYBYL molecular modelling program has been developed. The influence of several variables including energy minimization procedure, solvation, dielectric function and dielectric constant have been investigated in order to develop a general method, which is capable of producing high quality protein structures. Avian pancreatic polypeptide (APP) and bovine pancreatic phospholipase A2 (BP PLA2) were selected for the calculations, because high quality X-ray structures exist and because all classes of secondary structure are represented in the structures. The energy minimized structures were evaluated relative to the corresponding X-ray structures. The overall similarity was checked by calculating RMS distances for all atom positions. Backbone conformation was checked by Ramachandran plots and secondary structure elements evaluated by the length on hydrogen bonds. The dimensions of active site in BP PLA2 is very dependent on electrostatic interactions, due to the presence of the positively charged calcium ion. Thus, the distances between calcium and the calcium-coordinating groups were used as a quality index for this protein. Energy minimized structures of the trimeric PLA2 from Indian cobra (N.n.n. PLA2) were used for assessing the impact of protein-protein interactions. Based on the above mentioned criteria, it could be concluded that using the following conditions: Dielectric constant epsilon = 4 or 20; a distance dependent dielectric function and stepwise energy minimization, it is possible to reproduce X-ray structures very accurately without including explicit solvent molecules.

Synthesis and pharmacology of highly selective carboxy and phosphono isoxazole amino acid AMPA receptor antagonists.

(RS)-2-Amino-3-(3-hydroxy-5-methyl-4-isoxazolyl)propionic acid (AMPA, 5) and the selective AMPA receptor antagonist (RS)-2-amino-3-[3-(carboxymethoxy)-5-methyl-4-isoxazolyl]propionic acid (AMOA, 7) have been used as leads for the design and synthesis of a number of potential AMPA receptor antagonists. Two parallel series of AMOA analogs were synthesized, containing either a distal carboxylic acid (compounds 8b-g and 11b) or a phosphonic acid (compounds 9a-g, 10c, and 11c). Pharmacological characterization of the synthesized compounds was carried out using a series of receptor binding assays and by in vitro electrophysiological experiments using the rat cortical slice model. The two analogs with a tert-butyl substituent, (RS)-2-amino-3-[5-tert-butyl-3-(carboxymethoxy)-4-isoxazolyl]pr opi onic acid (ATOA, 8b) and the corresponding phosphonic acid analog ATPO (9b), were the most potent and selective AMPA antagonists within each series. ATOA and ATPO showed IC50 values of 150 and 28 microM, respectively, toward AMPA-induced depolarizations in the cortical slice model compared to IC50 = 320 microM for the parent compound, AMOA. These two new competitive AMPA antagonists were significantly more selective than AMOA, showing no antagonism (up to 1 mM) toward NMDA-induced responses, whereas AMOA (at 1mM) showed weak (19%) inhibition toward NMDA-induced responses. The structure-activity relationships for the two series of compounds revealed considerable differences with respect to the substituents effects, and the phosphonic acid analogs generally exhibited significantly higher potencies compared to the carboxylic acid analogs.

Three-dimensional structures of human phospholipase A2 from pancreas and synovial fluid by model building.

Three-dimensional structures of the enzyme phospholipase A2 (PLA2) from human pancreas and from human synovial fluid were constructed by model building based on high-resolution X-ray crystallographic structures and homology considerations. The structure of the human pancreatic PLA2 was based on the X-ray structure of the highly homologous bovine pancreatic PLA2 (Type I) by amino acid substitution and modification of the C-terminal part followed by geometry relaxation. The structure of the PLA2 from human synovial fluid was constructed from the X-ray structure of PLA2 from Crotalus atrox (Type II) by modification of the calcium binding loop, amino acid substitution, and insertion of two additional amino acid residues followed by geometry relaxation. The structure of the two human PLA2's have been compared and their use as targets for rational design of enzyme inhibitors discussed.

Excitatory amino acid receptor ligands. Synthesis and biological activity of 3-isoxazolol amino acids structurally related to homoibotenic acid.

The 3-isoxazolol amino acid (RS)-2-amino-3-(3-hydroxy-5-methylisoxazol-4-yl)propionic acid (AMPA, 2) and the isomeric compound (RS)-2-amino-3-(3-hydroxy-4-methylisoxazol-5-yl)propionic acid (4-methylhomoibotenic acid, 4a) are potent agonists at the AMPA subtype of central excitatory amino acid receptors. Using 4a as a lead structure, the amino acids 4c-e, in which the 4-methyl group of 4a is replaced by substituents of different size and polarity, were synthesized. Attempts to synthesize 4-(bromomethyl)homoibotenic acid (4f), a potential receptor alkylating agent, were unsuccessful. 4-Butylhomoibotenic acid (4c) and 4-(2-hydroxyethyl)homoibotenic acid (4e) were equipotent as inhibitors of [3H]AMPA binding (IC50 = 2 microM) and showed similar excitatory activity in the rat cortical slice preparation. 4d did not show significant affinity for AMPA receptor sites, but turned out to be a weak N-methyl-D-aspartic acid (NMDA) receptor antagonist. However, like 4c,e, 4d did not significantly affect the binding of the competitive NMDA antagonist, [3H]CPP, or the noncompetitive NMDA antagonist, [3H]MK-801. None of the amino acids 4c-e showed detectable affinity for [3H]kainic acid binding sites. Like the parent compound 4a (IC50 = 0.18 microM), 4c (IC50 = 0.18 microM), 4e (IC50 = 0.14 microM), and in particular 4d (IC50 = 0.02 microM) were effective inhibitors of calcium chloride-dependent [3H]glutamic acid binding, whereas AMPA is inactive (IC50 greater than 100 microM) in this binding assay. Thus, 4d is an effective and highly selective inhibitor of calcium chloride-dependent [3H]glutamic acid binding and may be a useful tool for studies of the physiological relevance and pharmacological importance of this binding affinity.