Two mutants of HIV-1 reverse transcriptase (RT) associated with high-level resistance of the virus to AZT (RT-AZT: D67N, K70R, T215Y, K219Q, and M41L) or 3-TC (RT-3TC: M184V) were expressed in Escherichia coli and purified. None of these mutants showed significant changes in the affinity and kinetics of binding to a DNA/DNA primer/template. RT-AZT was investigated in detail with respect to its kinetics of incorporation of nucleotides. No change in the relative rates of TMP and AZTMP incorporation could be detected for RT-AZT with respect to wild type RT. These results imply that there is no increased discrimination against AZTTP in the mutant. This was found for DNA/DNA and DNA/RNA primer/template. Additionally, rapid kinetics of incorporation of 3'-amino-3'-deoxythymidine 5'-monophosphate (a possible metabolite of AZT) were investigated and compared with TMP incorporation, but no difference in its relative rates of incorporation between wild type RT and RT-AZT was detected. In contrast, the already very slow rate of incorporation of 3-TCMP seen with wild type enzyme was drastically reduced (by a factor of 23 and 36 with DNA/DNA primer/template and DNA/RNA primer/template, respectively) for RT-3TC, showing a clear correlation between in vitro and in vivo effects. The affinity of 3-TCTP to the RT-3TC-primer/template complex was not affected by the mutation M184V. A 1.6-fold cross-resistance to ddATP, the converted form of the prodrug ddI, could also be shown for RT-3TC, but no cross-resistance to ddCTP was detected. Additionally, rapid kinetics of AZTMP incorporation by RT-3TC were investigated. There was an indication of a slightly higher rate of incorporation of AZTMP by RT-3TC than wild type RT.

The dissociation of dimeric reverse transcriptase (RT) of the human immunodeficiency virus (HIV) types 1 and 2 has been investigated using acetonitrile as a dissociating agent. The equilibrium transitions were monitored by combining different approaches (fluorescence spectroscopy, polymerase activity assay, and size-exclusion HPLC). The dissociation of RT induced a complete loss of polymerase activity and a 25% increase of the intrinsic fluorescence. It is fully reversible, and the midpoints of the equilibrium transition curves are dependent on the concentration of the enzyme used, suggesting a two-state transition model for the dissociation of RT in which dimers are in equilibrium with folded monomers. For both RTs, the heterodimeric form is more stable against dissociating agents and different pH than the corresponding homodimeric form. Moreover, heterodimeric HIV-2 RT exhibits a higher stability than HIV-1 RT, with a free energy of dissociation of 12.1 kcal/mol at pH 6.5 and 25 degrees C, instead of 10 kcal/mol for HIV-1 RT. The binding of a primer/template induces a marked conformational change in both RTs, shown by the lower accessibility of the tryptophans to quenchers and the increase in tryptophan heterogeneity, and stabilized the dimeric form of both RTs (10-100-fold). The central role of hydrophobic interactions in dimer formation has been revealed by the 30% increase of exposure of the tryptophan cluster to quenchers upon dissociation of RT and the binding of 4 equiv of 1-anilino-8-naphthalenesulfonate to the dissociated enzymes.

Intrinsic protein fluorescence of reverse transcriptases from HIV-1 and HIV-2 provides a sensitive signal for monitoring the interaction of the enzymes with primer/template duplex molecules. Kd values for 18/36-mer DNA/DNA duplexes were found to be in the range of a few nanomolar (about 3 times higher for the enzyme from HIV-2 than for that from HIV-1). The quenching of protein fluorescence induced on binding primer/template, together with an increase in extrinsic fluorescence on interaction with primer/template containing a fluorescent nucleotide at the 3'-end of the primer, was used to investigate the kinetics of interaction with reverse transcriptase from HIV-1. The results can be explained in terms of a two-step binding model, with a rapid diffusion-limited initial association (k(ass) = ca. 5 x 10(8) M-1 s-1) followed by a slow isomerization step (k = ca. 0.5 s-1). These (forward) rate constants are increased in the presence of a non-nucleoside inhibitor (S-TIBO) of HIV-1 reverse transcriptase, while the reverse rate constant for the second step is decreased, leading to an increase in affinity between the enzyme and primer/template by a factor of at least 10 when S-TIBO is bound. The results are discussed in terms of present knowledge of the structure of reverse transcriptase.