The chemistry of carbocyclic nucleosides

The background to the utility of conventional nucleosides for the treatment of viral diseases was given in the general introduction. A number of these compounds inhibit herpes viruses and the selectivity of these compounds in most cases depends on their preferential activation to their monophosphate esters by the viral enzyme, thymidine kinase (TK) [1]. On the other hand, normal purine ribo- or deoxyribonucleosides are substrates for purine nucleoside phosphorylases (PNPs), enzymes which cleave the N-glycosidic bond between the heterocycle moiety and the sugar [2]. It is in order to avoid such enzymatic degradation as well as to improve the anti-viral activity and selectivity of action of nucleosides, that a great number of structural modifications have been carried out on both the sugar and the heterocycle. One stratagem has been to replace the oxygen of the furanosyl ring by a methylene group, which gives rise to carbocyclic analogues of nucleosides. Many carbocyclic nucleosides have by now been synthesized that exhibit biological activity as well as resistance to degradation by phosphorylases. Examples are the carbocyclic analogues of BVDU (8), and IDU (10), namely C-BVDU (9), and C- IDU (11) (Figure 3.1). The latter two compounds, as expected, are not substrates for phosphorylases [3, 4], but they maintain their in vitro activities against HSV-1 [5]. Similarly, C-adenosine or aristeromycin (13), the cyclopentane analogue of adenosine (12), is far less susceptible to cleavage by S-adenosylhomocysteine hydrolase than adenosine itself.