Design of some nucleic acid antimetabolites: expectations and reality.

3'-Azido-2'3'-dideoxythymidine (AZT), a thymidine analogue with potent antiretroviral activity against human immunodeficiency virus (HIV) in vitro, has been shown to confer a clinical benefit in patients with advanced acquired immune deficiency syndrome (AIDS). Other 2',3'-dideoxynucleosides, e.g., 2',3'-dideoxy-cytidine and 2',3'-dideoxyadenosine, block the infectivity of HIV against helper/inducer T cells in vitro and protect the cell against the cytopathic effect of the virus. The majority of these antiretroviral agents were synthesized more than twenty years ago as analogues of physiologically important deoxynucleosides in the quest of a more effective cancer chemotherapy. None of the synthetic analogues manifested significant activity when screened against L1210 leukemia in BDF mice despite the fact that the phosphorylation reactions, crucial to the activation of the 2',3'-dideoxynucleosides, are catalyzed by kinases of appropriate target cells. However, the 2',3'-dideoxynucleoside triphosphates, relative to corresponding deoxynucleotides, have a reduced affinity for DNA polymerase alpha, an enzyme that has key DNA synthetic and repair functions in the life of a cell. In contrast, HIV reverse transcriptase, like host cellular DNA polymerase beta (a repair enzyme) and gamma (a mitochondrial enzyme), is much more susceptible to the inhibitory effects of the dideoxynucleotides. This would explain the activity of the fradulent nucleotides at low concentrations against pathogenic retroviruses vis à vis the low cytoxic activity observed with these agents as anti-leukemia drugs.