Structure activity relationship towards design of cryptosporidium specific thymidylate synthase inhibitors.

Cryptosporidiosis is a human gastrointestinal disease caused by protozoans of the genus Cryptosporidium, which can be fatal in immunocompromised individuals. The essential enzyme, thymidylate synthase (TS), is responsible for de novo synthesis of deoxythymidine monophosphate. The TS active site is relatively conserved between Cryptosporidium and human enzymes. In previous work, we identified compound 1, (2-amino-4-oxo-4,7-dihydro-pyrrolo[2,3-d]pyrimidin-methyl-phenyl-l-glutamic acid), as a promising selective Cryptosporidium hominis TS (ChTS) inhibitor. In the present study, we explore the structure-activity relationship around 1 glutamate moiety by synthesizing and biochemically evaluating the inhibitory activity of analogues against ChTS and human TS (hTS). X-Ray crystal structures were obtained for compounds bound to both ChTS and hTS. We establish the importance of the 2-phenylacetic acid moiety methylene linker in optimally positioning compounds 23, 24, and 25 within the active site. Moreover, through the comparison of structural data for 5, 14, 15, and 23 bound in both ChTS and hTS identified that active site rigidity is a driving force in determining inhibitor selectivity.

Computational approaches to molecular recognition.

Recent advances in the computation of free energies have facilitated the understanding of host-guest and protein-ligand recognition. Rigorous perturbation methods have been assessed and expanded, and more approximate techniques have been developed that allow faster treatment of diverse systems.

Investigations of neurotrophic inhibitors of FK506 binding protein via Monte Carlo simulations.

The binding and solution-phase properties of six inhibitors of FK506 binding protein (FKBP12) were investigated using free energy perturbation techniques in Monte Carlo statistical mechanics simulations. These nonimmunosuppressive molecules are of current interest for their neurotrophic activity when bound to FKBP12 as well as for their potential as building blocks for chemical inducers of protein dimerization. Relative binding affinities were computed and analyzed for ligands differing by a phenyl ring, an external phenyl or pyridyl substituent, and a pipecolyl or prolyl ring. Such results are, in general, valuable for inhibitor optimization and, in the present case, bring into question some of the previously reported binding data.

Estimation of binding affinities for selective thrombin inhibitors via Monte Carlo simulations.

Monte Carlo simulations have been performed on a series of 20 active-site-directed thrombin inhibitors to determine the interactions and energetics associated with the binding of these compounds. Physicochemical descriptors of potential value in the prediction of binding affinities were averaged during simulations of each inhibitor unbound in water and bound to thrombin. Regression equations based on 3-5 descriptors are able to reproduce the experimental binding affinities, which cover a 7 kcal/mol range, with rms errors of 1.0-1.3 kcal/mol, and yield correlation coefficients, r(2), of 0.7-0.8. On the basis of these results, the quantities most important in determining the binding affinities are: (1) the enhancement of van der Waals interactions in going from solution to the bound state, (2) the intramolecular strain induced in the inhibitor upon binding, (3) the number of hydrogen bonds lost in the binding process, and (4) the number of rotatable bonds in the inhibitor. The descriptors are physically reasonable and, in combination with the insights gained from analysis of the simulation structures, suggest directions for the development of improved thrombin inhibitors.

Binding affinities for sulfonamide inhibitors with human thrombin using Monte Carlo simulations with a linear response method.

The binding of sulfonamide inhibitors to human thrombin is examined to evaluate the viability of calculating free energies of binding, deltaGb, utilizing Monte Carlo (MC) statistical mechanics with a linear response approach. Coulombic and van der Waals energy components determined from MC simulations of the bound and unbound inhibitors solvated in water plus a solvent-accessible surface area term, as an index for cavity formation, were correlated with the free energies of binding for the inhibitor MD-805 and six derivatives. The best correlations yield an average error of 0.8 kcal/mol for the seven binding affinities, which cover an observed range of 6.0 kcal/mol. The MC simulations also provided insights into the interactions occurring in the active site and the origins of variations in deltaGb. Equatorial placement of the carboxylate group at C2 in the piperidine ring of the inhibitors causes electrostatic destabilization with the side chain of Glu-H192, while axial disposition of the C4-methyl group reduces favorable hydrophobic interactions in the P-pocket of the enzyme.

Estimation of binding affinities for HEPT and nevirapine analogues with HIV-1 reverse transcriptase via Monte Carlo simulations.

The interactions and energetics associated with the binding of 20 HEPT and 20 nevirapine nonnucleoside inhibitors of HIV-1 reverse transcriptase (RT) have been explored in an effort to establish simulation protocols and methods that can be used in the development of more effective anti-HIV drugs. Using crystallographic structures as starting points, all 40 inhibitors were modeled in the bound and unbound states via Monte Carlo (MC) statistical mechanics methods. Potentially useful descriptors of binding affinity were configurationally averaged for each inhibitor during the MC simulations, and correlations were sought with reported experimental activities. A viable regression equation was obtained using only four descriptors to correlate the 40 experimental activities with an r(2)() of 0.75 and cross-validated q(2)() of 0.69. The computed activities show a rmsd of 0.94 kcal/mol in comparison with experiment and an average unsigned error of 0.69 kcal/mol. The MC results reveal three physically reasonable parameters that control the binding affinities: (1) loss of hydrogen bonds with the inhibitor is unfavorable, (2) burial of hydrophobic surface area is favorable, and (3) a good geometrical fit without steric clashes is needed for the protein-inhibitor complex. It is gratifying that the corresponding descriptors are statistically the most important quantities for determining the anti-HIVRT activity for the 40 compounds. Representative examples are also given in which structural and thermodynamic information from the MC simulations is used to help understand binding differences for related compounds. A key pi-type hydrogen bond has been identified between secondary-amide nevirapine analogues and Tyr188A of HIVRT that explains their otherwise surprising activity and the ineffectiveness of nevirapine against the Y188C mutant.

Prediction of binding affinities for TIBO inhibitors of HIV-1 reverse transcriptase using Monte Carlo simulations in a linear response method.

Monte Carlo (MC) simulations in combination with a linear response approach were used to estimate the free energies of binding for a series of 12 TIBO nonnucleoside inhibitors of HIV-1 reverse transcriptase. Separate correlations were made for the R6 and S6 absolute conformations of the inhibitors, as well as for the analogous N6-monoprotonated species. Models based upon the neutral unbound inhibitors produced overall better fits to experimental values than did those using the protonated unbound inhibitors, with only slight differences between the neutral R6 and S6 cases. The best results were obtained with a three-parameter linear response equation containing van der Waals (alpha), electrostatic (beta), and solvent accessible surface area (SASA, gamma) terms. The averaged (R6 and S6) rms error was approximately 0.88 kcal/mol for the observed range of 4.06 kcal/mol in inhibitor activities. The averaged values of alpha, beta, and gamma were -0.150, 0.114, and 0. 0286, respectively. Omission of the alpha term gave beta 0.152 and gamma 0.022 with a rms of 0.92. The unweighted van der Waals components were found to be highly attractive but failed to correlate well across the series of inhibitors. Contrastingly, while the electrostatic components are all repulsive, they show a direct correlation with inhibitor activity as measured by DeltaGbinding. The role of gamma is primarily to produce an overall negative binding energy, and it can effectively be replaced with a negative constant. During the MC simulations of the unbound solvated inhibitors, the R6 and S6 absolute conformations do not interconvert due to the formation of a favorable hydrogen bond to solvent. In the complex, however, interconversion of these conformations of the inhibitor is observed during the course of the simulations, a phenomenon which is apparently not observed in the crystalline state of the complex. Hydrogen bonding of the inhibitor to the backbone NH of K101 and the lack of such an interaction with the C=O of K101 or with solvent correlate with enhanced activity, as does the ability to assume a number of different orientations of the inhibitor dimethylallyl moiety with respect to residues Y181 and Y188 while retaining contact with W229. Overall, the use of a combination of MC simulation with a linear response method shows promise as a relatively rapid means of estimating inhibitor activities. This approach should be useful in the preliminary evaluation of potential modifications to known inhibitors to enhance activity.

Developing a dynamic pharmacophore model for HIV-1 integrase.

We present the first receptor-based pharmacophore model for HIV-1 integrase. The development of "dynamic" pharmacophore models is a new method that accounts for the inherent flexibility of the active site and aims to reduce the entropic penalties associated with binding a ligand. Furthermore, this new drug discovery method overcomes the limitation of an incomplete crystal structure of the target protein. A molecular dynamics (MD) simulation describes the flexibility of the uncomplexed protein. Many conformational models of the protein are saved from the MD simulations and used in a series of multi-unit search for interacting conformers (MUSIC) simulations. MUSIC is a multiple-copy minimization method, available in the BOSS program; it is used to determine binding regions for probe molecules containing functional groups that complement the active site. All protein conformations from the MD are overlaid, and conserved binding regions for the probe molecules are identified. Those conserved binding regions define the dynamic pharmacophore model. Here, the dynamic model is compared to known inhibitors of the integrase as well as a three-point, ligand-based pharmacophore model from the literature. Also, a "static" pharmacophore model was determined in the standard fashion, using a single crystal structure. Inhibitors thought to bind in the active site of HIV-1 integrase fit the dynamic model but not the static model. Finally, we have identified a set of compounds from the Available Chemicals Directory that fit the dynamic pharmacophore model, and experimental testing of the compounds has confirmed several new inhibitors.

Computer simulations of organic reactions in solution.

Quantum and statistical mechanics have been used to determine energy profiles for the SN2 reaction of Cl- + CH3Cl in the gas phase, in aqueous solution, and in liquid DMF. The energy profile in the gas phase has the characteristic double-well form featuring unsymmetrical ion-dipole complexes as minima and a symmetrical transition state. Hydration causes the reaction surface to become almost unimodal and increases the barrier significantly. The reaction profile in DMF is intermediate between those for the gas phase and aqueous solution. The ion-dipole complexes are still free energy minima in DMF. Thus, the reaction in DMF involves initial formation of the complex before the rate-determining step. The computed results are shown to be in good accord with experimental free energies of activation. The same technique has been applied to the addition reaction of OH- + H2C = O in the gas phase and aqueous solution. Ab initio 6-31 + G* calculations indicate that the reaction proceeds essentially without activation in the gas phase. Hydration introduces a substantial energy barrier. The transition state in water has been located at a C-O separation of roughly 2 A. A key finding for both reactions is that the activation barriers induced by hydration result primarily from change in strengths rather than in numbers of solute-water hydrogen bonds along the reaction paths.

Molecular dynamics simulations of the unfolding of an alpha-helical analogue of ribonuclease A S-peptide in water.

Molecular dynamics simulations of the S-peptide analogue AETAAAKFLREHMDS have been conducted in aqueous solution for 300 ps at 278 K and for 500 ps in two different runs at 358 K. The results show agreement with experimental observations in that at low temperature, 5 degrees C, the helix is stable, while unfolding is observed at 85 degrees C. In the low-temperature simulation a solvent-separated ion pair was formed between Glu-2 and Arg-10, and the side chain of His-12 reoriented toward the C-terminal end of the alpha-helix. Detailed analyses of the unfolding pathways at high temperature have also revealed that the formation or disappearance of main-chain helical hydrogen bonds occurs frequently through an alpha in equilibrium with 3(10) in equilibrium with no hydrogen bond sequence.

Prediction of drug solubility from Monte Carlo simulations.

Monte Carlo statistical mechanics simulations have been carried out for 150 organic solutes in water. Physically significant descriptors such as the solvent-accessible surface area, numbers of hydrogen bonds, and indices for cohesive interactions in solids are correlated with pharmacologically important properties including octanol/water partition coefficient (log P) and aqueous solubility (log S). The regression equation for log S only requires five descriptors to provide a correlation coefficient, r2, of 0.9 and rms error of 0.7 for the 150 solutes. The descriptors can form a basis for structural modifications to guide an analogue's properties into desired ranges.

Rationale for the observed COX-2/COX-1 selectivity of celecoxib from Monte Carlo simulations.

Computational studies have yielded an analysis of the contributions to the free energy difference between the binding of celecoxib to COX-1 and to COX-2. The energetic and structural results point to the Ile to Val mutation at residue 523 as the key contributor to COX-2 selectivity; unfavorable steric contact between a sulfonamide oxygen and the delta methyl group of Ile523 destabilizes the complex with COX-1. The His to Arg change at residue 513 is less significant.

Computational binding studies of human pp60c-src SH2 domain with a series of nonpeptide, phosphophenyl-containing ligands.

Monte Carlo/free energy perturbation (MC/FEP) simulations were performed on a series of nonpeptide ligands of the human pp60c-src SH2 domain in order to calculate relative free energies of binding for each compound and to understand the structural requirements for high affinity binding. The amido compound, exhibiting the highest experimental affinity, takes advantage of an interaction with a previously unobserved structural water.

Improved convergence of binding affinities with free energy perturbation: application to nonpeptide ligands with pp60src SH2 domain.

Free Energy Perturbations (FEP) in the context of Monte Carlo (MC) simulations were conducted to predict the relative free energies of binding for a series of human Src SH2 domain ligands. Two procedures for disappearing atoms during a single-topology FEP are investigated and dramatic differences in free energy convergence behavior are seen. Comparison of these two protocols suggests that the coupling of the removal of angular constraints with the disappearance of an atom may significantly slow free energy convergence. The series of ligands under investigation here cover a range of modifications at the 3-position of 4-([[4-(cyclohexyl methoxy)benzyl]amino]carbonyl) phenyl phosphate. Unlike any other compound in this study, the 3-amide analog can form two hydrogen bonds within the region of the perturbation, one to a backbone amide hydrogen and one to a highly coordinated water molecule. Agreement with experimental trends in binding affinity is seen, although the computed relative free energy of binding of the amido compound is underestimated. These results are reconciled by examination of the hydration energies of model systems, which predict primary amides as too hydrophilic.

Modeling the complexation of substituted benzenes by a cyclophane host in water.

Monte Carlo statistical mechanics simulations have been used to study the complexation of disubstituted benzenes by Diederich's octamethoxy tetraoxaparacyclophane host. Relative free energies of binding were obtained in water at 25 degrees C for benzene, p-xylene, p-cresol, p-dicyanobenzene, and hydroquinone from statistical perturbation theory. The computed results agree well with experimental data, including the binding affinity of benzene, which was determined after the calculations were completed. The computed structures for the complexes reveal details that are important for understanding the order of binding affinities. It is found that hydroquinone protrudes from one side of the complex and participates in hydrogen bonds between one hydroxyl group and two water molecules and in an intracomplex hydrogen bond between the other hydroxyl group and ether oxygens. The calculations also show a clear preference for binding p-cresol with the hydroxyl group hydrated rather than inside the host's cavity.

Molecular dynamics simulations of the unfolding of apomyoglobin in water.

Molecular dynamics simulations of apomyoglobin have been conducted in aqueous solution for 350 ps at 25 degrees C and for 500 ps in two different runs at 85 degrees C. The structures obtained at the higher temperature display properties similar to those of molten globules. Close agreement is obtained between the computed structural models and experimental data on the helical content of both native apomyoglobin and the low-pH unfolding intermediate. The results also suggest explanations for the surprising observations on the effects of mutations at the interface of the A, G, and H helices. Detailed analyses of the final structures and the unfolding pathways at high temperature clearly show that the most stable alpha-helical regions are those in contact with other helices.

Molecular dynamics simulations of the unfolding of barnase in water and 8 M aqueous urea.

Molecular dynamics simulations of barnase have been conducted both in water and in 8 M urea solution for 500 ps at 25 degrees C and for 2000 ps at 85 degrees C. The final structure of the aqueous simulation at room temperature matches closely the structure obtained by NMR and the experimentally observed protections from isotopic exchange. The comparison of the structures generated by the aqueous simulation at 85 degrees C reveals a trajectory composed of groups of geometrically related structures separated by narrow regions of rapid change in structure. The first of these regions displays changes in backbone rmsd to the crystal structure and solvent-accessible area suggestive of a transition state, while the properties observed during the final 300 ps of the simulation are consistent with a stable intermediate. These assignments were confirmed by calculation of the "progress along the reaction coordinate" phi-values using an empirical equation based on a linear response method. The pathway of unfolding defined in this fashion agrees well with the experimental results of site-directed mutagenesis in terms of secondary structure content of the transition state and the intermediate and reproduces the relative stability of the different elements of secondary structure. The results of the simulations in urea suggest a mechanism at the molecular level for its well-known enhancement of the denaturation of proteins. The analysis of radial distribution functions shows that the first solvation shell of the protein is enriched in urea relative to the bulk solvent. The displacement of water molecules allows greater exposure of hydrophobic side chains, as witnessed particularly in the analysis of solvent-accessible surface areas at the higher temperature. Almost all urea molecules in the first shell form at least one hydrogen bond with the protein. They provide a more favorable environment for accommodation of the remaining water molecules, and they facilitate the separation of secondary structure elements by acting as a bridge between groups previously forming intraprotein hydrogen bonds.

Estimation of the binding affinities of FKBP12 inhibitors using a linear response method.

A series of non-immunosuppressive inhibitors of FK506 binding protein (FKBP12) are investigated using Monte Carlo statistical mechanics simulations. These small molecules may serve as scaffolds for chemical inducers of protein dimerization, and have recently been found to have FKBP12-dependent neurotrophic activity. A linear response model was developed for estimation of absolute binding free energies based on changes in electrostatic and van der Waals energies and solvent-accessible surface areas, which are accumulated during simulations of bound and unbound ligands. With average errors of 0.5 kcal/mol, this method provides a relatively rapid way to screen the binding of ligands while retaining the structural information content of more rigorous free energy calculations.

Mechanism for the rotamase activity of FK506 binding protein from molecular dynamics simulations.

Molecular dynamics (MD) and free energy perturbation (FEP) methods are used to study the binding and mechanism of isomerization of a tetrapeptide (AcAAPFNMe) by FK506 binding protein (FKBP). Detailed structures are predicted for the complexes of FKBP with the peptide in both ground-state and transition-state forms. The results support a mechanism of catalysis by distortion, where a large number of nonbonded interactions act together to stabilize preferentially the twisted transition state. The two most important groups for the catalysis are suggested to be Trp59 and Asp37, but several other groups are identified as directly or indirectly involved in the binding and catalysis. However, the structural results do not support the notion that the keto oxygen of the immunosuppressive agents FK506 and rapamycin mimics the oxygen for the twisted peptide bond in the FKBP-transition-state complex.