Design of novel dihydroxynaphthoic acid inhibitors of Plasmodium falciparum lactate dehydrogenase.

We have studied inhibition of Plasmodium falciparum lactate dehydrogenase (pfLDH) by dihydroxynaphthoic acid (DHNA) analogues derivatives of hemigossypol-sesquiterpene found in cottonseed known to exhibit antimalarial activity. Molecular models of pfLDH-DHNA complexes were prepared from high-resolution crystal structures containing DHNA and azole inhibitors and binding affinities of the inhibitors were computed by molecular mechanics - polarizable continuum model of solvation (MM-PCM) approach. The 3D structures of the pfLDH-DHNA complexes were validated by a QSAR model, which confirmed consistency between the computed binding affinities and experimental inhibition constants for a training set and validation set of twelve DHNA inhibitors obtained from literature. Novel more potent DHNA analogs were identified by structure-based molecular design and predicted to inhibit pfLDH in the low nanomolar concentration range. In addition, the designed DHNA analogs displayed favorable predicted ADME-related profiles and an elevated selectivity for the pfLDH over the human isoform.

QSAR study of natural estrogen-like isoflavonoids and diphenolics from Thai medicinal plants.

Two species of Thai medicinal plants, Dalbergia parviflora R. (Leguminosae) and Belamcanda chinensis L. (Iridaceae), used traditionally for the regulation of menstrual disorders, have been found to contain a large number of potential estrogen-like compounds. A set of some 55 isolated isoflavonoids and diphenolics showed a wide range of estrogen activity as determined in breast cancer MCF-7 and T47D cell proliferation assays. This set of compounds was studied by means of computational techniques including quantitative structure-activity relationships (QSAR) and molecular modeling. It was found that the estrogenic potencies of the studied compounds depend mainly upon the presence/absence of hydroxyl groups attached to 3' and 5' positions of B ring of the isoflavone scaffold and the inter-atomic distance between the hydroxyl groups attached to the outer terminal positions 7 of A ring and 4' of B ring. In a QSAR model employing ligand-receptor interaction energy descriptors, the LigScore scoring function of Cerius(2) virtual screening module, which describes the receptor affinities of simultaneous binding to estrogenic receptors α and ß (ER(α) and ER(ß)), led to the best correlation between the observed estrogenic activities and computed descriptors. Consideration of independent binding to ER(α) and ER(ß) did not result in statistically significant QSAR models. It was thus concluded that simultaneous and possibly competitive interaction of the compounds with the ER(α) and ER(ß) receptors, in which the presence of hydroxyl groups at the abovementioned positions of the isoflavonoids and diphenolics molecular scaffold plays a dominant role, may determine the estrogenic potency of the considered phytochemicals.

Design of oseltamivir analogs inhibiting neuraminidase of avian influenza virus H5N1.

Neuraminidase is an important target for design of antiviral agents in the prophylaxis and treatment of avian influenza virus infections. We have shown the applicability of computer-assisted combinatorial techniques in the design, focusing and in silico screening of a virtual library of analogs of oseltamivir (Tamiflu) with the goal to find potent inhibitors of influenza A neuraminidase N1 that fill the cavity found adjacent to the active site. Crystal structure of oseltamivir-N1 complex was used in the structure-based focusing and virtual screening of the designed library. A target-specific Piecewise Linear Potential type 1 scoring function fitted for a training set of 14 carbocyclic inhibitors and validated for three other inhibitors was used to select virtual hits with predicted inhibitory activities in the subnanomolar range. The results of this computational study are useful as a rational guide for synthetic and medicinal chemists who are developing new drugs against the avian influenza virus H5N1.

Design and in silico screening of combinatorial library of antimalarial analogs of triclosan inhibiting Plasmodium falciparum enoyl-acyl carrier protein reductase.

Enoyl-acyl carrier protein reductase of Plasmodium falciparum (PfENR) is an important target for antimalarial agents that interfere with the FAS-II pathway of lipid synthesis, which is specific for the parasite. Recent studies showed that substituted analogs of triclosan (TCL) inhibit the purified PfENR enzyme with IC(50) values below 200 nM when the suboptimal 5-chloro group was replaced by larger hydrophobic moieties. We have used computer-assisted combinatorial techniques to design, focus and in silico screen a virtual library of TCL analogs substituted at positions 5, 4' and 2'. Our study can thus direct synthetic chemists working on the antimalarial FAS-II inhibitors towards the explored subset of the chemical space, which is predicted to contain compounds with PfENR inhibition potencies in the low nanomolar range and favorable ADME properties.