Interrelationship between HIV-1 fitness and mutation rate

Abstract

Differences in replication fidelity, as well as mutator and antimutator strains, suggest that virus mutation rates are heritable and prone to natural selection. Human immunodeficiency virus type 1 (HIV-1) has many distinct advantages for the study of mutation rate optimization given the wealth of structural and biochemical data on HIV-1 reverse transcriptase (RT) and mutants. In this study, we conducted parallel analyses of mutation rate and viral fitness. In particular, a panel of 10 RT mutants-most having drug resistance phenotypes-was analyzed for their effects on viral fidelity and fitness. Fidelity differences were measured using single-cycle vector assays, while fitness differences were identified using ex vivo head-to-head competition assays. As anticipated, virus mutants possessing either higher or lower fidelity had a corresponding loss in fitness. While the virus panel was not chosen randomly, it is interesting that it included more viruses possessing a mutator phenotype rather than viruses possessing an antimutator phenotype. These observations provide the first description of an interrelationship between HIV-1 fitness and mutation rate and support the conclusion that mutator and antimutator phenotypes correlate with reduced viral fitness. In addition, the findings here help support a model in which fidelity comes at a cost of replication kinetics and may help explain why retroviruses like HIV-1 and RNA viruses maintain replication fidelity near the extinction threshold.

Figure 1. Experimental approach

Measurement of A) viral fidelity and B) viral fitness, among panel of 10 HIV-1 RT mutants. A) An HIV-1 vector (pHIG) sensitive to changes in fidelity was manipulated to consist of NL4-3 wt or one of ten RT mutants. Each of these vectors was co-transfected with a VSV-G envelope expression plasmid into a producer cell line. Cell culture supernatants were collected, filtered to remove cellular debris, and titered for downstream application. Viral equivalents were used to transduce CEM target cells to calculate relative mutant frequencies by flow cytometry analysis. B) The same 10 RT mutants were introduced into the NL4-3 molecular clone. Together with wt, these clones were generated and titered, as described above. Each of the mutant viruses was competed against wt by infecting 5 × 10⁵ CEM-GFP target cells at an equal MOI of 0.0001 (i.e., 50 infected cells). Virus was passaged every other day, for ten days, and then relative amounts of viral nucleic acid were quantified by duplex qPCR. The RT point mutations listed in bold text are nucleoside RT inhibitor mutations. Abbreviations: HSA, murine heat stable antigen; IRES, internal ribosome entry site; GFP, green fluorescence protein; VSV-G, vesicular stomatitis virus-glycoprotein; FL1-H, fluorescence channel 1, height of intensity; FL2-H, fluorescence channel 2, height of intensity; pol, HIV-1 gene consisting of protease, RT, and integrase; MOI = multiplicity of infection; qPCR, quantitative PCR.

Figure 2. HIV-1 RT mutants influence virus mutant frequencies

Each of the 10 RT mutants was measured for differences in fidelity relative to the wt reference strain. Mutants are displayed in ascending order of their RT position. Mutant frequencies were calculated relative to wt and are depicted on log-scale. The actual measurements are listed in the adjacent column. Values represent at least 4 biological replicates that were experimentally repeated 4 to 8 times. SD is standard deviation and represented by error bars. * P value < 0.05; ** P value < 0.01; *** P value < 0.001.

Figure 3. Impact of HIV-1 RT mutants on viral fitness

Each of the 10 RT mutants, and a wt control, was passaged with a wt reference strain in a head-to-head competition assay. Mutants are assembled by RT position, and respective fitness differences (W_D) are displayed on a log-scale. Asterisk symbols denote level of significance with numerical values along horizontal line showing fold difference from wt. Error bars represent SEM, standard error of the mean. * P value < 0.05; ** P value < 0.01; *** P value < 0.001.

Figure 4. Correlative influence of HIV-1 RT variants on both virus mutant frequency and fitness

Assembly of RT mutants that have A) higher, or B) lower, fidelity relative to wt and the corresponding fitness difference measurements. L74V and wt are included in both panel A) and B). Lines represent the linear regression calculated from best-fit values. Dashed lines are 95% confidence intervals of best-fit line. Error bars represent SEM, standard error of the mean.

Figure 5. Locations of amino acid residues in HIV-1 reverse transcriptase (RT) associated with drug resistance and/or enzyme fidelity

A ribbon diagram of HIV-1 RT is shown in side-view (A) or head-on (B) view of the catalytic domain, with interlaid primer:template DNAs, including a space-filling of amino acid residues investigated in this study. HIV-1 RT subdomains are indicated, and amino acid residues investigated in this study are represented as space-filled and are color-coded to indicate phenotype. Red denotes primary drug-resistant mutation site; blue denotes secondary drug-resistant mutation site; green denotes non-drug-resistant site. The yellow molecule depicts a dideoxynucleotide at the polymerase active site. Image adapted from of Huang, H. et al. and created with Protein Workshop software . Image obtained from the Research Collaboratory for Structural Bioinformatics (RCSB) Protein Data Bank (PDB) ; IPDB ID: 1RTD.