HOST-PARASITE COEVOLUTION
HOST-PARASITE COEVOLUTION: EVIDENCE FOR RARE ADVANTAGE AND TIME-LAGGED SELECTION IN A NATURAL POPULATION


Mark F. Dybdahl and Curtis M. Lively

Department of Biology, Indiana University, Bloomington, IN 47405 USA




Summary

In theory, parasites can create time-lagged, frequency-dependent selection in their hosts, resulting in oscillatory gene-frequency dynamics in both the host and the parasite (the Red Queen hypothesis). Oscillatory dynamics, however, have not been observed in natural populations. In the present study, we evaluated the dynamics of asexual clones of a New Zealand snail, Potamopyrgus antipodarum, and its trematode parasites over a five-year period. During the summer of each year, we determined host-clone frequencies in random samples of the snail in order to track genetic changes in the snail population. Similarly, we monitored changes in the parasite population, focusing on the dominant parasite, Microphallus sp., by calculating the frequency of clones in samples of infected individuals from the same collections. We then compared these results to the results of a computer model, which was designed to examine clone frequency dynamics for various levels of parasite virulence. Consistent with these simulations and with ideas regarding dynamic coevolution, parasites responded to common clones in a time-lagged fashion (Figure 1 and 2).




Figure 1. Estimated Microphallus infection rate (circles and dashed line), total infection rate (squares and solid line), and frequency in the random sample (bar graph) of the four common clones. Significant over-infection by Microphallus is shown by asterisks: *, P < 0.05 and ***, P < 0.0001.








Figure 2. Changes in the frequencies of common host clones plotted against (A) the time-lagged changes in infection frequencies of these clones, and (B) the contemporaneous (non-lagged) changes in infection frequencies. Changes in host-clone frequencies (x axis) are the between-year differences in clone frequencies in the random samples of the population. The changes in infection frequencies were estimated from the between-year differences in the host-clone frequencies in the Microphallus "infected sample." (A) shows the correlation between changes in host-clone frequencies (years y and y+1) in the random sample and the time-lagged changes (years y+1 and y+2) in the infection frequencies. (B) shows the correlation between changes in host-clone frequencies (years y and y+1) in the random sample and the contemporaneous changes (for the same years y and y+1) in the infection frequencies. Solid lines show the best fit least-squares regressions. The lack of a significant relationship in (B) indicates that Microphallus is not simply infecting snail clones in the proportions that they exist. The significant relationship in (A) is consistent with frequency-dependent, time-lagged Red Queen dynamics. The combined results are consistent with the predictions from our coevolutionary model.



























Finally, in a laboratory experiment, we found that clones that had been rare during the previous five years were significantly less infectible by Microphallus when compared to the common clones (Figure 3).Taken together, these results confirm that rare host genotypes are more likely to escape infection by parasites; they also show that host-parasite interactions produce, in a natural population, some of the dynamics anticipated by the Red Queen hypothesis.







Figure 3. The prevalence of experimental Microphallus sp. infections in each of the four recently common clones and in 40 other clones grouped as "Rare" clones. The line indicates the average prevalence of experimental infections in the common clones, and the sample size for each clonal group is on the right of the symbol. One binomial standard error is shown for each group.