Fighting HIV-1 and related diseases using rational medicinal chemistry approaches: from computational chemistry to biological evaluation

HIV-1 is undoubtedly one of the viruses that have characterized the last half century in human history. This virus, found to be the AIDS etiological cause, is still one of the biggest challenges for drug therapy, since there are still neither a vaccine nor a definitive cure. Therefore, despite more than 25 marketed drugs administered in several combinations in HAART (Highly Active Anti-Retroviral Therapy), there is still the need of molecules active through new mechanisms, in order to overcome the many resistances developed by the virus to counter the pharmacological treatment. The main purpose of this thesis is to identify new active molecules able to inhibit viral replication with innovative mechanisms. Also, considering that the AIDS patients easily develop secondary infections, another part of the work is focused on the study of antifungal agents previously developed in our laboratories. The first chapter contains the research carried out directly on the HIV-1 enzyme Integrase. This one results to be particularly interesting because against it, at the moment, only three marketed drugs with the same mechanism of action are available. Moreover, it has not a homologue protein in human cells, eliminating the problem of poor selectivity. Starting with the study of Integrase, after computational analysis and biological tests, we discovered a very interesting dual-inhibitor with unique activity profile and active in vitro with no toxic effects at the maximum tested concentration. The second chapter is focused on the study of molecules with antifungal activity able to block candida infections, which easily occur in immunocompromised patients. This study was carried out on macrocyclic compounds developed in our laboratory, revealing a very unusual mechanism of entrance into fungal cells. Specific biological experiments were performed to investigate in depth the behaviour of these molecules, leading to a potential new treatment to overcome the problem of antifungal resistance. The third and last chapter is focused on the development of a new part of code for the program KNIME, a data-pipelining software used in CADDD (Computer-Aided Drug Design and Development). Leaning on LigandScout, new code (nodes) was programmed, and the existing one was implemented. In particular, we focused on two topics: chirality and fingerprints. Two nodes were programmed to address chirality issues (Standard Property and Chirality Enumerator node) and four based on fingerprints (Fingerprint Calculator node, Fingerprint Similarity node, Fingerprint Clustering node and the Diversity Picker node).