Development of improved methodologies for the discovery of novel bioactive compounds and their application to problems of biomedical relevance.

Abstract

Chemoinformatics is a discipline that has positioned itself both in the world of aca- demic research and in the industrial field as a fundamental tool given its high effi- ciency/cost ratio. However, it also has certain limitations that have been addressed in the development of this industrial thesis.

The principal objective on which this work is based is fundamentally to apply methodological improvements in chemoinformatic techniques for the discovery of new bioactive compounds against different therapeutic targets, both in an academic and industrial context.

The computational techniques used are based on both ligand structure (phar- macophore modeling, QSAR, three-dimensional similarity) and protein structure (docking and molecular dynamics). Some of the therapeutic targets studied have been Zika virus protease NS2B-NS3, acetyl and butyrylcholinesterase enzymes, the α-galactosidase enzyme and various proteins involved in diabetes.

The first contribution of this thesis (which is presented as a compendium of publications) investigated the discovery of a pharmacological chaperone for the enzyme α-galactosidase with the ability to bind to the enzyme in a region other than the active site of the enzyme so that it would stabilize its three-dimensional structure and recover its functionality without inhibiting it at high concentrations as Migalastat (the only drug approved by the Food and Drug Administration for the treatment of Fabry disease) does. Using a combined strategy of virtual screening based on the three-dimensional shape of the chemical compounds and the modeling of the molecular coupling we discovered a compound, 2,6-ditiopurine, which after laboratory tests was found to be able to stabilize and promote the maturation of the enzyme.

In the following three contributions, docking techniques were used to describe the potential mechanism of action as acetyl and butyrylcholinesterase inhibitors of N-acetyl-tryptophan compounds, different classes of tanshinones, rosmarinic acid, hyperforin and hyguanin C. In all these cases, our simulation results predict that more stable binding of these compounds with enzymes takes place in the active site of the same, so that the compounds would block the entry of the substrate, which would explain their activity as inhibitors.

In the penultimate contribution we screened Drugbankn˜(Law et al., 2013), a public database containing FDA-approved drugs and compounds used in clinical research) for searching compounds suitable for the treatment of Zika virus infection. In this case, protease NS2B-NS3 was selected as a therapeutic target for which 8 enzyme-inhibiting compounds were found by means of structure based virtual screening. Of these, the Novobiocin compound proved to inhibit both the protein in vitro and in vivo, obtaining 100 % survival of the animal models used for the experiment.

An on-line platform was finally developed for the identification of new anti- diabetic compounds. This thesis presents a work derived from the use of this platform to carry out an inverse screening and identify the potential therapeutic target of different natural compounds. We have found bibliographic evidences of experimental validation for eight of the compounds for which the computational method predicted a possible therapeutic target confirming the the have certain activity.