Mechanism and fidelity of HIV reverse transcriptase.

Abstract

We have examined the RNA-dependent and DNA-dependent polymerase and ribonuclease H catalytic activities of human immunodeficiency virus reverse transcriptase using rapid transient kinetic methods with defined synthetic 25/45-mer DNA/RNA and DNA/DNA primer/templates. The Kd value for interaction of the enzyme with duplex DNA was 4.7 nM, and the value for RNA/DNA heteroduplex was of similar magnitude. A pre-steady state burst of nucleoside triphosphate incorporation was observed for both DNA and RNA templates. Analysis of the dATP concentration dependence of the burst rate provided Kd values for dATP of 4 and 14 microM and maximum rates of single nucleotide incorporation, kpol, of 33 and 74 s-1, for DNA and RNA templates, respectively. Subsequent turnovers were limited by the rate of dissociation of the primer/template from the enzyme at rates of 0.18 and 0.06 s-1 for duplex DNA and RNA/DNA heteroduplex, respectively. Analysis of rates of DNA polymerization and RNA cleavage using the RNA template revealed that the two activities are independent of one another. The polymerization rate (4-70 s-1) was dependent on dATP concentration, whereas the RNA cleavage occurred at a constant rate of 10 s-1 over the 100-fold dATP concentration range (2-200 microM). Examination of the RNA cleavage products resulting from a single turnover indicates that the polymerase and ribonuclease domains of the enzyme are separated by a distance corresponding to 19 bases of RNA/DNA heteroduplex, consistent with the recently published crystal structure (Kohlstaedt, L. A., Wang, J., Friedman, J., Rice, P. A., and Steitz, T. A. (1992) Science 256, 1783-1790). Analysis of the kinetics of processive synthesis suggested that the initial binding of dNTP leads to a faster rate of dissociation of DNA from the enzyme. Further investigation supported a two-step dNTP binding mechanism with the formation of an initial E.DNA.dNTP complex followed by a more stable E'.DNA.dNTP complex. The Kd values for incorporation of incorrect nucleoside triphosphates opposite a DNA template thymidine were 1010 microM for dGTP, 1240 microM for dCTP, and 840 microM for dTTP. The corresponding maximum kpol rates were 4.8 s-1 for dGTP, 0.52 s-1 for dCTP, and 0.41 s-1 for dTTP. These values provide fidelity estimates of 1740 for discrimination against dGTP, 19,700 for dCTP, and 16,900 for dTTP misincorporations at this site.