A computational model of the inhibition of Mycobacterium tuberculosis ATPase by a new drug candidate R207910.

Diarylquinolines (DARQs) are a new class of potent inhibitors of the ATPase of Mycobacterium tuberculosis. We have created a homology model of a binding site for this class of compounds located on the contact area of the a-subunit (gene atpB) and c-subunits (gene atpE) of Mycobacterium tuberculosis ATPase. The binding pocket that was identified from the analysis of the homology model is formed by 4 helices of three c-subunits and 2 helices of the a-subunit. The lead compound of the DARQ series, R207910, was docked into the pocket using a simulated annealing, multiple conformer, docking algorithm. Different stereoisomers were treated separately. The best docking pose for each stereoisomer was optimized by molecular dynamics simulation on the 5300 atoms of the binding region and ligand. The interaction energies in the computed complexes enable us to rank the different stereoisomers in order of interaction strength with the ATPase binding pockets. We propose that the activity of R207910 against Mycobacterium tuberculosis is based on interference of the compound with the escapement geometry of the proton transfer chain. Upon binding the compound mimics the conserved Arg-186 residue of the a-subunit and interacts in its place with the conserved acidic residue Glu-61 of the c-subunit. This mode of action is corroborated by the good agreement between the computed interaction energies and the observed pattern of stereo-specificity in the model of the binding region.

An ant algorithm for the conformational analysis of flexible molecules.

Originally, the ant system was developed for optimization in discrete search spaces such as the traveling salesman problem. We detail our adaptation of the algorithm to optimization in the continuous search space of conformational analysis. The parameters of the algorithm were tuned using a simple test molecule, undecane, and a drug molecule, imatinib. The algorithm is further tested on four more drug or drug-like molecules, on vitamin A and on alanine tetrapeptide.

In search of a novel anti-HIV drug: multidisciplinary coordination in the discovery of 4-[[4-[[4-[(1E)-2-cyanoethenyl]-2,6-dimethylphenyl]amino]-2- pyrimidinyl]amino]benzonitrile (R278474, rilpivirine).

Ideally, an anti-HIV drug should (1) be highly active against wild-type and mutant HIV without allowing breakthrough; (2) have high oral bioavailability and long elimination half-life, allowing once-daily oral treatment at low doses; (3) have minimal adverse effects; and (4) be easy to synthesize and formulate. R278474, a new diarylpyrimidine (DAPY) non-nucleoside reverse transcriptase inhibitor (NNRTI), appears to meet these criteria and to be suitable for high compliance oral treatment of HIV-1 infection. The discovery of R278474 was the result of a coordinated multidisciplinary effort involving medicinal chemists, virologists, crystallographers, molecular modelers, toxicologists, analytical chemists, pharmacists, and many others.

Correlations between factors determining the pharmacokinetics and antiviral activity of HIV-1 non-nucleoside reverse transcriptase inhibitors of the diaryltriazine and diarylpyrimidine classes of compounds.

OBJECTIVE: To investigate the important factors that determine the bioavailability and the antiviral activity of the diaryltriazine (DATA) and diarylpyrimidine (DAPY) non-nucleoside reverse transcriptase inhibitors (NNRTIs) of HIV-1 in animal species and humans using cell-based assays, physicochemical and computed parameters. METHODS: This naturalistic study included 15 parameters ranging from molecular mechanics calculations to phase I clinical trials. The calculated parameters were solvent-accessible surface area (SASA), polar surface area and Gibbs free energy of solvation. Physicochemical parameters comprised lipophilicity (octanol/water partition coefficient [cLogP]), ionisation constant (pKa), solubility and aggregate radius. Cell-based assays included human colonic adenocarcinoma cell (Caco-2) permeability (transepithelial transport), drug metabolism and antiviral activity (negative logarithm of the molar effective concentration inhibiting viral replication by 50% [pEC50]). Exposure was tested in rats, dogs and human volunteers. RESULTS: Of the 15 parameters, eight correlated consistently among one another. Exposure (area under the plasma concentration-time curve [AUC]) in humans correlated positively with that in rats (r = 1.00), with transepithelial transport (r = 0.83), lipophilicity (r = 0.60), ionisability (r = 0.89), hydrodynamic radius of aggregates (r = 0.66) and with antiviral activity (r = 0.61). Exposure in humans was also seen to correlate negatively with SASA (r = -0.89). No consistent correlation was found between exposure in dogs and the eight parameters. Of the 14 DATA/DAPY molecules, 11 form aggregates with radii between 34 and 100 nm. CONCLUSIONS: We observed correlations between exposure in humans with exposure in rats, transepithelial transport (Caco-2 cells), ionisability, lipophilicity, aggregate radius and SASA in the class of DATA/DAPY NNRTI compounds. The lipophilic DATA/DAPY compounds form aggregates. It can be assumed that absorption in the intestinal tract and endocytosis in infected cells of these lipophilic compounds are governed by the common phenomenon of aggregate formation. As the lymphatic system offers a pathway for intestinal uptake of aggregates, this may offer a therapeutic advantage in the treatment of HIV-1 infection. Although it was not the objective of the study, we found that the rat was a better in vivo model than the dog for the prediction of systemic exposure in this particular set of compounds.

A pharmacophore docking algorithm and its application to the cross-docking of 18 HIV-NNRTI's in their binding pockets.

The docking of small molecules into the binding site of a target protein is an important but difficult step in structure-based drug design. The performance of a docking algorithm is usually evaluated by re-docking ligands into their native binding sites. We have explored the cross-docking of 18 HIV-NNRTIs (non-nucleoside inhibitors of HIV reverse transcriptase) of which the ligand-protein structure has been determined: each of the 18 ligands was docked into each of the 18 binding sites. The docking algorithms studied are an energy-based simulated annealing algorithm and a novel pharmacophore docking algorithm. It turns out that the energy-based docking of the ligands into non-native pockets is far less successful than the docking into their native pockets. The results can be improved by using explicit pharmacophore information, and by docking a ligand into a panel of protein structures and selecting the ligand-protein combination with the lowest interaction energy as the final result.

On the detection of multiple-binding modes of ligands to proteins, from biological, structural, and modeling data.

There are several indications that a given compound or a set of related compounds can bind in different modes to a specific binding site of a protein. This is especially evident from X-ray crystallographic structures of ligand-protein complexes. The availability of multiple binding modes of a ligand in a binding site may present an advantage in drug design when simultaneously optimizing several criteria. In the case of the design of anti-HIV compounds we observed that the more active compounds that are also resilient against mutation of the non-nucleoside binding site of HIV1-reverse transcriptase make use of more binding modes than the less active and resilient compounds.

Contribution of mutations in the cytochrome P450 14alpha-demethylase (Erg11p, Cyp51p) to azole resistance in Candida albicans.

The cytochrome P450 14alpha-demethylase, encoded by the ERG11 (CYP51) gene, is the primary target for the azole class of antifungals. Changes in the azole affinity of this enzyme caused by amino acid substitutions have been reported as a resistance mechanism. Nine Candida albicans strains were used in this study. The ERG11 base sequence of seven isolates, of which only two were azole-sensitive, were determined. The ERG11 base sequences of the other two strains have been published previously. In these seven isolates, 12 different amino acid substitutions were identified, of which six have not been described previously (A149V, D153E, E165Y, S279F, V452A and G4655). In addition, 16 silent mutations were found. Two different biochemical assays, subcellular sterol biosynthesis and CO binding to reduced microsomal fractions, were used to evaluate the sensitivity of the cytochromes for fluconazole and itraconazole. Enzyme preparations from four isolates showed reduced itraconazole susceptibility, whereas more pronounced resistance to fluconazole was observed in five isolates. A three-dimensional model of C. albicans Cyp51p was used to position all 29 reported substitutions, 98 in total identified in 53 sequences. These 29 substitutions were not randomly distributed over the sequence but clustered in three regions from amino acids 105 to 165, from 266 to 287 and from 405 to 488, suggesting the existence of hotspot regions. Of the mutations found in the two N-terminal regions only Y132H was demonstrated to be of importance for azole resistance. In the C-terminal region three mutations are associated with resistance, suggesting that the non-characterized substitutions found in this region should be prioritized for further analysis.