Studies on the fidelity and error specificity of human immunodeficiency virus type 1 reverse transcriptase

Abstract

Nucleoside analog-resistant RT variants have been shown to display an increased fidelity of nucleotide insertion and mispair extension. Since the earlier studies measured the influence of nucleoside analog resistance mutations on specific types of errors, we have addressed the influence of such mutations on the overall fidelity and error specificity of RT by using the M13-based forward mutation assay. We first examined the fidelity of the multi-drug resistant RTs containing either the Q151M alteration alone or that with multiple mutations A62V, V75I, F77L, F116Y, and Q151M. Our studies revealed no significant differences in overall fidelity, however interesting differences in error specificity were noted. Specifically, the spectrum of mutations generated Q151M mutated enzyme contained three base substitution hotspots that had not been previously observed for HIV-1 RT, suggesting that a single amino acid substitution can significantly affect the sequence context in which the errors are made. Next, we examined the overall fidelity and error specificity of the 3TC-resistant M184I variant HIV-1 RT alongside the M184V variant and the wild type control HIV-1HXB2 RT. While the M184V variant showed no increase in overall fidelity, the M184I variant displayed a 4-fold increase in overall fidelity when compared to wild-type, and a 2.5-fold increase when compared to the M184V variant.;Next, we examined the mutational specificity of RTs with known differences in processivity. While these enzymes showed no change in overall fidelity, one of them makes significantly fewer frameshifts than wild type, and both enzymes displayed significant changes in the number of errors at two sites along the template. The data in this thesis suggests that one cannot make generalizations about the relationship between processivity and fidelity.;Our data suggests that the in vitro error rate accurately reflects the viral mutation rate. Previous studies have suggested that the in vitro error rate of HIV-1 RT is 20-fold lower than the in vivo mutation rate. However, when we compared the error rate and mutation rate from the same strain (HIV-1 NL4-3), we found the difference to be merely two-fold, suggesting that the in vitro determined rate is close to the overall mutation rate.