Table 1.
Viral kinetics parameters for H1N1pdm09 WT and MUT-I223V*.
Fig 1.
Experimental data and extracted parameters characterizing virus replication fitness.
Measured H1N1pdm09 WT-I223 (blue circles) and MUT-I223V (green triangles) viral load concentrations for the (A) multiple-cycle, (B) single-cycle, and (C) mock-yield assays. The points represent the geometric mean and their error bars correspond to the standard deviation of the experimental data points collected in triplicate at each time point. The lines were produced using Eq (1) with the median parameter values presented in Table 1 solved as described in Methods. (D) Probability density functions of the key virus replication parameters with statistically significant differences indicated with a (⋆).
Fig 2.
Simulated competition between the WT-I223 and MUT-I223V strains.
The infectious (PFU/mL) and total (RNA/mL) viral load (top), and the fraction of infectious cells infected by each strain (bottom) are shown. The curves are produced using the median values of the extracted MCMC distributions. (A) In the absence of therapy, the mutant is disadvantaged due to its longer eclipse phase (τE), and shorter infectious lifespan (τI). (B) In the presence of oseltamivir, the mutant gains the fitness advantage. The quantitative gains or losses of each strain along with statistical significance are provided in the Appendix.
Table 2.
Comparative impact of the H275Y and I223V mutations*.
Fig 3.
Impact of the H275Y and I223V NA mutations in the H1N1pdm09 background.
The probability density functions of the key viral replication parameters characterizing the fitness of the WT-H275 and MUT-H275Y H1N1pdm09 strains (from earlier work, [18]), and that of the WT-I223 and MUT-I223V strains are presented side-by-side for comparison. To facilitate the comparison across the two separate experiments, the WT distributions have been centred at one (100) and the H275Y and I223V mutants’ parameters are shown relative to their respective WT parameters.
Fig 4.
Simulated competition between the wild-type, MUT-H275Y and MUT-I223V strains.
The infectious (PFU/mL) and total (RNA/mL) viral load (top), and the fraction of infectious cells infected by each strain (bottom) are shown. (A) Without therapy, the fitness of the WT and MUT-H275Y strains is comparable. (B) In the presence of oseltamivir, the WT has a replicative disadvantage over the MUT-H275Y. (C) Without therapy, the MUT-H275Y strain already has a fitness advantage over the MUT-I223V strain. (D) In the presence of oseltamivir, this advantage is increased.
Table 3.
Inter-experimental changes in the viral kinetic parameters for WT H1N1pdm09*.
Fig 5.
Differences between inter-experimental replicates.
(A) Infections with the same H1N1pdm09 wild-type strain illustrate that viral titer growth rates, peak values, and decay rates can vary noticeably between three experiments performed on different dates. (B) Infections with either the MUT-H275Y or the MUT-I223V single mutants over two separate experiments. The peak viral titers are more similar to the wild-type performed in the same experiment (comparing to left graph), than within the same strain in different experiments. (C) Recorded RNA titers either from the current study or previously presented (Pinilla, 2012 in [18], Pizzorno, 2011 in [34]) and reproduced here for comparative purposes. (D) Probability density functions of parameters for the WT and mutant pair of strains in either the current (blue, green) or previous (black, red) experiment illustrate that some infection parameters are clearly more experiment-specific than strain-specific, pointing to a genuine impact of inter-experimental variability on the actual viral replication parameters.
Table 4.
Calculated oseltamivir efficacies as a function of dose*.
Table 5.
Viral kinetics parameters for H1N1pdm09 WT and MUT-H275Y*.
Table 6.
Simulated competition experiments*.