Molecuar Docking Applied to the search and rational design of new carbonic anhidrase VII inhibitors

GANTNER MELISA; VILLALBA MARIA LUISA; GAVERNET LUCIANA

virtual

Congreso; 1st Congress of Women in Bioinformatics and Data Science Latin America.; 2020

Carbonic anhydrase isoform VII (CAVII) is a metalloenzyme that catalyzes the carbon dioxide to bicarbonate reversible hydration. In mammals, it is expressed mainly in the brain and it has been validated as a new molecular target for the treatment of epilepsy and neuropathic pain. The aim of this work was the development and application of a molecular docking model to the search and rational design of new enzyme inhibitors.Experimental structures of human CAVII were downloaded from the Protein Data Bank. Additionally, a test set of 295 inhibitors and 211 non-inhibitors was generated by a manually curated database search. Different docking programs and parameters were evaluated. A library of decoys (DUD-E) was generated for validation purposes. The ROC and BEDROC curves, and the enrichment factor were calculated to evaluate the performance of the models. AutoDock4Zn with the 3MDZ experimental structure was the model that best reproduces the experimental complexes and that better discriminate between inhibitors and non-inhibitors. The model was then applied in a virtual screening campaign on a subset of 5661 compounds that contain sulfur in their structure, extracted from ZINC15 database. In addition, the model was used to guide the selection of sulfamides and their acid derivatives to be synthesized as potential CAVII inhibitors, by analyzing the interactions predicted by the docking model of the new proposed structures within the enzyme. Finally, in silico predictions of drug-like characteristics, ADME and toxicity profiles were performed.Five candidates from virtual screening were commercially purchased and eleven compounds designed were synthesized. The in vitro inhibitory activity against human CAVII is being evaluated. The model constitutes a valuable tool for virtual screening and rational design of new CAVII inhibitors. It is the first molecular docking model with the ability to correctly discriminate between inhibitors and non-inhibitors demonstrated by in silico validations.