Synthesis, Biological Evaluation and QSAR Studies of Novel Antimalarial Agents

Abstract

Malaria remains as one of the most devastating diseases of the developing world concentrated mainly in tropical regions. Despite the huge advances in our understanding of the disease, it continues to be one of the greatest causes of serious illness and death in the world. Chloroquine (CQ) and other quinoline antimalarials have been the mainstay drugs in the prevention and treatment of malaria because of their low cost, safety and efficacy. However, the indiscriminate use of these drugs has been seriously eroded their efficacy in recent years, mainly because of the development and spread of resistance. This wide spread of resistance has severely limited the choice of available antimalarial drugs, which clearly highlights the urgent need of novel chemotherapeutic agents for the treatment of malaria. Covering these aspects the first chapter of this thesis gives an overview of 4- aminoquinoline derivatives as antimalarial agents. The Chapter 2 (Synthesis and Antimalarial Activity of 4-Aminoquinoline Derivatives) devoted to the synthetic strategies adopted in the thesis and the antimalarial activity of the designed compounds. The scheme 1 in the first section of this chapter outlines the synthetic strategy of 7-trifluoromethyl-4-aminoquinolones having thiazolidin-4-ones in lateral side chain. In the second section, the schemes 2 and 3 outline the synthesis of designed amino acid conjugates of 4-aminoquinolines. These conjugates suggest the requirement of basicity and hydrophobicity for antimalarial activity, heme binding and inhibition of hemozoin formation. The chapter 3 provides introduction to QSAR methodologies including feature selection approaches. This chapter also describes 2D- / 3D- QSAR models generated in the thesis for piperazinoquinolines (Chloroquine related) and anilinoquinolines (Amodaiquine related). As a part of defined objective of the thesis, the datasets are exclusively selected for investigating structure activity relationships in these compounds. Attempts have been made to integrate the QSAR findings with the designing of amino acid conjugates synthesized (Chapter 2). The Chapter 4 expounds 3D-QSAR relations in antimalarial agents for relatively new target, Plasmodium falciparum Protein farnesyltransferase (Pf-PFT). The Pf-PFT is a relatively new validated target for antimalarial agents. The benzodiazepine BMS-414662 is a known inhibitor of mammalian-PFT. The successful translation of BMS-414662 to tetrahydroquinolines (THQ) and ethylenediamines (ED) for Pf-PFT inhibition has generated interest to explore the structure-activity relations in these analogues. Thus the CoMFA/ CoMSIA 3D-QSAR analysis of THQ and ED analogues have been carried out and the results are presented in sections one and two respectively. The contour maps obtained from the models has indicated the steric and electrostatic characteristics of D- and E-ring regions for the activity. The study has offered insight for designing new compounds against Pf-PFT. Incorporating these features a few potential analogues are proposed. The Chapter 5 reports a perplexing observation made during an ANN investigation. In modeling approaches, ANN has a special place to address the nonlinear phenomena or curved manifold. It is routinely believed that input variables of ANN have to be made available through one or other feature selection approaches. Thus, often one or other feature selection approach is used prior to ANN to feed the input variables for its models. Contrary to this belief, we observed that prior feature selection is not essential for ANN and it is a desirable option for meaningful outputs in terms of the rationale behind the inputs. In order to verify our observation a large number of feature sets, objectively selected as well as arbitrarily chosen, from different databases namely, thiazolidinones, anilinoquinolines and piperazinoquinolines have been analyzed using ANN. In this investigation, datasets created from random numbers are used as control. As part of this investigation, for the first time three metrics namely External distances (ED), Internal distances (ID) and Dissimilarity of subspace (DS) have been introduced to characterize descriptor and regression space of the feature sets. The findings have been rigorously debated and critically reviewed by field experts of specialized journal QSAR & Combinatorial Science. These results position ANN as a powerful tool to identify the patterns in the data. The amino acid conjugates (Chapter 2) synthesized as part of this thesis were analyzed in ANN and found to yield good predictive models. The fundamental algorithms and programs used in the most of the data analysis of this thesis are developed by Dr. Y. S. Prabhakar.