Synthesis and computational studies of nucleoside analogues as antiviral agents

Resumo

Nucleoside analogues have been widely used in the antiviral therapies developed over the last three decades. Therapeutic deficiencies shown by the majority of the drugs currently in use have prompted the search for novel antiviral agents. In this context, carbocyclic nucleoside analogues have received extraordinary attention, since the absence of the N-glycosidic linkage in these derivatives results in an increased metabolic stability in vivo. L-nucleosides have been also broadly studied, since some of them feature good antiviral activity, higher metabolic stability and lower toxicity than their natural D-counterparts. On the other hand, molecular modelling techniques have been one of the cornerstones in the design and optimization of therapeutic compounds. In particular, molecular docking techniques, used to simulate the interaction between an organic molecule and a macromolecule, have proved to be a very useful tool. In the field of nucleoside analogues, which are commercialized as prodrugs, docking techniques may be applied to simulate both their activation process as well as their subsequent interaction with the therapeutic target, a viral polymerase. The present thesis has been focused on the study of carbocyclic nucleoside analogues as antiviral agents, both from the synthetic and theoretical points of view. In particular, five new cyclobutene and cyclobutane L-nucleoside analogues have been synthesized, and their mechanism of action has been investigated by means of molecular docking. The information gained in these studies has been subsequently used to perform a rational design of new nucleoside analogues. First, a family of five new cyclobutene and cyclobutane L-nucleoside analogues, which present purine and pyrimidine nucleobases, has been synthesized following a divergent synthetic pathway. The key step has been the [2+2] photocycloaddition of (S)-5-pivaloyloxymethyl-2(5H)-furanone to (Z)-1,2-dichloroethylene and subsequent Zn-promoted reductive elimination of the dichlorinated cycloadducts to construct the cyclobutene ring. The antiviral activity of these nucleosides has been evaluated against a wide range of viruses (e.g. herpes simplex virus, HSV) without showing significant activity. The antiviral activity of nucleosides depends both on their ability to inhibit the target viral enzyme as well as their activation process, which consists of three successive phosphoryl transfers to convert the nucleoside or analogue in its triphosphorylated derivative. Thus, the activation process and the subsequent interaction with HIV-1 reverse transcriptase of a series of cyclobutane-fused nucleosides previously synthesized in our research group were investigated by means of molecular docking. The theoretical studies have demonstrated that the presence of the cyclobutane fused to the 2' and 3' positions of the ribose moiety is detrimental to the adequate binding of these analogues to the viral DNA chain, which accounts for their lack of activity. A similar study has been carried out to rationalize the absence of activity against HSV of the five cyclobutene and cyclobutane L-nucleoside analogues previously prepared. The results derived from the study revealed that three of the nucleoside analogues cannot be activated to the required triphosphorylated form. Conversely, the antiviral activity of the other two analogues is not restricted by their activation process, indicating that a possible limiting factor could be their interaction with the HSV DNA polymerase. Finally, a rational design of five new cyclobutane and cyclobutene L-nucleoside analogues as anti-HSV agents has been performed, based on the data obtained in the previous studies. Similarly, five new cyclohexene and bicyclo[4.1.0]heptane nucleoside analogues have also been designed.