Effects of nucleotide analogues on human immunodeficiency virus type 1 integrase.

We extended our previous study with 3'-azido-3'-deoxythymidine nucleotides [Proc. Natl. Acad. Sci. USA 91:5771-5775 (1994)] and examined the effects on human immunodeficiency virus type 1 (HIV-1) integrase of the nucleotides of three nucleoside analogues currently under evaluation in clinical trials: beta-D-2',3'-didehydro-3'-deoxythymidine, beta-D-2'-ara-fluoro-2', 3'-dideoxyadenosine, and beta-L-2',3'-dideoxy-3'-thiacytidine. Beta-D-2',3'-Didehydro-3'-deoxythymidine and beta-D-2'-ara-fluoro-2',3'-dideoxyadenosine nucleotides had IC50 values for strand transfer of 100 and 200 microM, respectively, whereas the corresponding 2',3'-dideoxynucleoside triphosphates, ddT triphosphate and ddA triphosphate, did not inhibit the integrase at 800 and 200 microM, respectively. Beta-L-2',3'-Dideoxy-3'-thiacytidine triphosphate had no effect up to 500 microM. The L-enantiomers of 5-fluoro-2',3'-dideoxycytidine monophosphate and triphosphate had IC50 values of approximately 40 microM, whereas their D-enantiomer isomers showed no inhibition at 200 microM. NAD, pyridoxal phosphate, and coumermycin A1, which exhibit no antiviral activity but are typically used to probe nucleotide binding sites, were also tested. NAD was inactive, and its etheno derivative exhibited activity at 1 mM. In contrast, pyridoxal phosphate (IC50 = 18 microM and coumermycin A1 (IC50 = 5 microM were potent inhibitors. None of the coumermycin monomeric derivatives were active integrase inhibitors. The physiological ribonucleotides ATP and GTP inhibited HIV-1 integrase at or near cellular concentrations, suggesting that they may regulate HIV-1 integrase activity in cells. In general, the active nucleotides tested inhibited binding of HIV-1 integrase to its substrate DNA an inhibited an integrase deletion mutant containing only amino acids 50-212, indicating that nucleotides bind to the enzyme catalytic core. Consisently, the choice of nucleophile in the 3'-processing reaction was blocked to the same extent regardless of the nucleotide used (water, glycerol, or the viral DNA hydroxyl) by the enzyme. These observations suggest new strategies for antiviral drug development that could be based on nucleotide analogues as inhibitors of HIV-1 integrase.