Effects of potyvirus effective population size in inoculated leaves on viral accumulation and the onset of symptoms

Abstract

Effective population size (N(e)) is a key parameter for understanding evolutionary processes, but it is generally not considered in epidemiological studies or in studying infections of individual hosts. Whether N(e) has an effect on the onset of symptoms and viral accumulation in Tobacco etch virus (TEV) infection of Nicotiana tabacum plants is considered here. Using mixtures of TEV variants carrying fluorescent markers, the dose dependence of N(e) was confirmed, and the inoculation procedure was found to be the main source of variation in these experiments. Whereas the onset of symptoms was independent of N(e), there was less and more variable accumulation at 6 days postinoculation for small N(e) values (N(e) < 5). The observed variation in accumulation was not heritable, however, suggesting that this variation was not due to the fixation of deleterious mutations in the small founder populations. On the other hand, virus-induced fluorescence and accumulation in the inoculated leaf were strongly N(e) dependent. Systemic accumulation was independent of N(e), although removal of the inoculated leaf led to a small reduction in systemic accumulation for small N(e) values. For whole plants, N(e)-dependent effects on accumulation were no longer observed at 9 days postinoculation. Therefore, the effects of N(e) on accumulation are due mainly to limited expansion in the inoculated leaf and are transient. In this system, N(e)-dependent effects will be strongest at low doses and early in infection. We conclude that N(e) can have implications for epidemiology and infection at the individual host level, beyond determining the rate of mixed-genotype infection.

Fig 1

Comparison of IAH model predictions and data. Comparisons are shown for IAH model predictions and data for the dose-focus relationship (A) and for rates of infection and mixed-genotype infection (B). In panel A, the ln-transformed dose is plotted on the abscissa, and the ln-transformed number of foci plus 1 is plotted on the ordinate. In panel B, the value for log mean foci is plotted on the abscissa, while the frequency of infection (squares and solid line) or mixed-genotype infection (diamonds and dotted line) is plotted on the ordinate. For both panels, error bars represent 95% confidence intervals. None of the data points in panel B are significantly different from model predictions (Table 3).

Fig 2

Dose-symptomatology relationship at 6 dpi. The log virion dose is plotted on the abscissa, and the frequency of symptomatology or infection at 6 dpi is plotted on the ordinate. The symptomatology data are given by triangles (mean ± 95% confidence interval), whereas squares represent infection data for comparison. The HHS model was plotted for dose-symptomatology data (dotted line), as this model gave the lowest Akaike weight (Table 1). However, the HHS model does not have appreciably more support than the DA model. For the dose-infection curve, the IAH model was plotted, as this is the simplest model and neither of the other models has appreciably more support (Table 2). Note that the dose-symptomatology curve is much shallower than the dose-response curve and that the frequency of symptomatic plants is considerably lower than the frequency of infected plants.

Fig 3

Effects of dose on viral accumulation at 6 dpi. The virion dose is plotted on the abscissa, and viral accumulation at 6 dpi (number of genome copies per 100 ng of total RNA extracted from plants) is plotted on the ordinate. Note that both scales are logarithmic. The mean accumulation ± standard deviation is given and appears to increase with dose. A Jonckheere-Terpstra test confirmed that there is a significant increase in accumulation with dose.

Fig 4

Effects of N_e on accumulation at 6 dpi. N_e is plotted on the abscissa, and viral accumulation at 6 dpi (number of genome copies per 100 ng of total RNA extracted from plants) is plotted on the ordinate. Circles represent measurements for individual plants; error bars show standard deviations. A Jonckheere-Terpstra test demonstrated that there is a significant increase of accumulation with N_e. The line depicts the fitted geometric model, which confirms that the increase of accumulation with N_e occurs only for small N_e values.

Fig 5

Fluorescence in the inoculated leaf at high and low doses. GFP and mCherry fluorescence is shown for inoculated N. tabacum leaves at 6 dpi. For panels A to C, a high virion dose was given (1.44 × 10⁷), whereas for panels D to F a low virion dose was given (5.34 × 10⁵), and plants with a single focus of primary infection at day 3 were selected. RT-qPCR also demonstrated significantly less accumulation in the inoculated leaves of plants infected at only a single focus. Note the variation in the amount of fluorescence at low doses (D to F). In some cases, only the focus of primary infection and a faint trail along the vascular tissue are present (D), whereas in other cases the virus did achieve limited expansion into the inoculated leaf (F).

Fig 6

Effects of N_e on accumulation when the inoculated leaf is removed. Viral accumulation at 6 dpi is given for plants infected with a large or small N_e. The inoculated leaf was removed at 5 dpi, when all infected plants in the experiment had shown systemic fluorescence. Errors bars represent standard deviations.

Fig 7

Time course experiment on the effects of N_e on accumulation. The time of collection is plotted on the abscissa, and viral accumulation (number of genome copies per 100 ng of total RNA extracted from plants) is plotted on the ordinate. Circles represent measurements for individual plants (error bars show standard deviations) from the low-dose treatment group (1:270 virion dilution; N_e = 2.05 ± 1.28), whereas squares represent plants from the high-dose treatment group (1:10 virion dilution; N_e = 23.30 ± 10.57). Lines represent a logistic model fitted to the high-dose (solid line) and low-dose (dotted line) data. There were effects of N_e on accumulation only at low doses, whereas similar levels of accumulation were measured at high doses.