| Abstract
| - Malaria has been one of the most significant public health problems for centuries. It affects many tropicaland subtropical regions of the world. The increasing resistance of Plasmodium spp. to existing therapies hasheightened alarms about malaria in the international health community. Nowadays, there is a pressing needfor identifying and developing new drug-based antimalarial therapies. In an effort to overcome this problem,the main purpose of this study is to develop simple linear discriminant-based quantitative structure−activityrelationship (QSAR) models for the classification and prediction of antimalarial activity using some of theTOMOCOMD−CARDD (TOpological MOlecular COMputer Design−Computer Aided “Rational” DrugDesign) fingerprints, so as to enable computational screening from virtual combinatorial datasets. In thissense, a database of 1562 organic chemicals having great structural variability, 597 of them antimalarialagents and 965 compounds having other clinical uses, was analyzed and presented as a helpful tool, notonly for theoretical chemists but also for other researchers in this area. This series of compounds was processedby a k-means cluster analysis in order to design training and predicting sets. Afterward, two linear classificationfunctions were derived in order to discriminate between antimalarial and nonantimalarial compounds. Themodels (including nonstochastic and stochastic indices) correctly classify more than 93% of the compoundset, in both training and external prediction datasets. They showed high Matthews' correlation coefficients,0.889 and 0.866 for the training set and 0.855 and 0.857 for the test one. The models' predictivity was alsoassessed and validated by the random removal of 10% of the compounds to form a new test set, for whichpredictions were made using the models. The overall means of the correct classification for this process(leave group 10% full-out cross validation) using the equations with nonstochastic and stochastic atom-based quadratic fingerprints were 93.93% and 92.77%, respectively. The quadratic maps-based TOMOCOMD−CARDD approach implemented in this work was successfully compared with four of the mostuseful models for antimalarials selection reported to date. The developed models were then used in a simulationof a virtual search for Ras FTase (FTase = farnesyltransferase) inhibitors with antimalarial activity; 70%and 100% of the 10 inhibitors used in this virtual search were correctly classified, showing the ability of themodels to identify new lead antimalarials. Finally, these two QSAR models were used in the identificationof previously unknown antimalarials. In this sense, three synthetic intermediaries of quinolinic compoundswere evaluated as active/inactive ones using the developed models. The synthesis and biological evaluationof these chemicals against two malaria strains, using chloroquine as a reference, was performed. An accuracyof 100% with the theoretical predictions was observed. Compound 3 showed antimalarial activity, beingthe first report of an arylaminomethylenemalonate having such behavior. This result opens a door to avirtual study considering a higher variability of the structural core already evaluated, as well as of otherchemicals not included in this study. We conclude that the approach described here seems to be a promisingQSAR tool for the molecular discovery of novel classes of antimalarial drugs, which may meet the dualchallenges posed by drug-resistant parasites and the rapid progression of malaria illnesses.
|
