In a series of experiments (summary in Table ) using the crustacean host D. magna and P. ramosa, its bacterial parasite, we investigated if parasites exposed to male versus female host individuals differ in the likelihood of successful infection (Experiments 1 and 2), in parasite fitness, in the rate of proliferation within the host (Experiments 2 and 3), and in induced disease symptoms (Experiments 2, 4 and 5).
Infection rate and parasite proliferation in male versus female host individuals
P. ramosa had higher infection rates in females when three-day-old individuals were exposed to parasite spores for 11 days (Experiment 1, Linear mixed model, factor "Sex", df = 1, deviance = 27.4, P < 0.00001, Figure ). The infection rate increased with the dose of parasite spores (factor "Dose", df = 3, deviance = 34.9, P < 0.00001, Figure ) but the sex difference did not vary with dose (interaction "Sex × Dose", df = 1, deviance = 1.35, P = 0.45, Figure ). However, when we exposed one-day-old hosts, before the sexual dimorphism becomes apparent, for a short period (48 hours; Experiment 2), we did not observe a difference in the proportion of infected females versus males (Linear mixed model, factor "Sex", df = 1, deviance = 1.6, P = 0.21, Figure ).
Figure 1 Proportion of infected male versus female D. magna hosts after exposure to P. ramosa spores. (a) The data correspond to Experiment 1, with long exposure time (for 11 days) of sexually dimorphic hosts (three-day-old). The proportion of infected increased (more ...)
Spore counts (Figure ) and spore densities (Figure ) at Day 20 of the experiment were higher in females than in males (Experiment 3; with a short period of exposure of young, sexually immature host individuals). This suggests that the rate of spore production in females was higher than in males in the first 20 days of the experiment. As animals were exposed to the parasite before sex differentiation, the differences in spore counts and densities are unlikely to be caused by differences in the number of spores ingested (that is, differences in the initial inocula). Between Days 20 and 27 of the experiment, the rate of spore production (slope in Figure ) no longer differs significantly between the sexes (Two-way ANOVA (log(spore number)): n = 142; factor "Sex" df = 1, F = 289.37, P < 0.00001; factor "Day" df = 1, F = 31.96, P < 0.00001; interaction "Sex × Day" df = 1, F = 1.62, P = 0.2). Parasite density increased in males during the later phase of parasite proliferation (Welch's t-test: df = 61.67, t = -3.23, P = 0.002; Figure ), but did not in females (Welch's t-test: df = 59, 03, t = -0.29, P = 0.77; Figure ).
Number and density of parasite spores in male and female hosts, 20 and 27 days post-exposure. The P. ramosa spore number (A) and the density (B) were higher in females D. magna than in males. Error bars represent the 95% confidence intervals.
At parasite induced host death, females harbored many more spores than males (Experiment 2, Clone Kela-08-10: Kruskal-Wallis rank test, df = 1, χ2 = 6.2, P = 0.01; Clone Kela-20-13: df = 1, χ2 = 32.1, P < 0.00001, Figure ).
Figure 3 Number of spores at death in male and female hosts. P. ramosa had higher fitness in females than in males in both Daphnia (Experiment 2, Kruskal-Wallis rank test, Clone Kela-08-10: df = 1, χ2 = 6.2, P = 0.01; Clone Kela-20-13: df = 1, χ (more ...)
Effects of parasites on males versus female hosts
Over all the experiments, we found no significant difference in mortality before Day 14 (the earliest it is possible to reliably check for infection status) between male and female hosts, and between host clones (Two-way ANOVA (log(number of dead individuals before Day 14)): "Host clone", df = 10, F = 2.5, P = 0.08, "Sex", df = 1, F = 1.41, P = 0.26). Individuals dead before Day 14 were excluded from further analysis.
We monitored lifespan of infected versus control male and female hosts (Experiment 2) and showed that control Daphnia of both sexes lived longer than their infected counterparts (females: Log-rank test: n = 232, df = 1, χ2 = 111, P < 0.00001; males: n = 260, χ2 = 190, df = 1, P < 0.00001, Figure ). We did not detect a significant difference in cost of infection on survival between male and female hosts (Coxph: factor "Infection status", Exp(coef) = 15.61, Z = 10.8, P < 0.00001, factor "Sex", Exp(coef) = 7.41, Z = 7.67, P < 0.00001, factor "Infection status × Sex", Exp(coef) = 1.37, Z = 1.035, P = 0.3). The median lifespan was reduced by about 50% in both sexes (Figure ).
Figure 4 Survival of control (solid lines) versus infected (dotted lines) of male and female hosts. Control D. magna live about twice longer than infected ones. P. ramosa reduces the lifespan of both female and male Daphnia. We did not find a statistically significant (more ...)
We tested whether P. ramosa induces gigantism in its hosts in two experiments (Experiments 2 and 4). We found that infected females were larger than uninfected females (Figure , Table ), while body size of infected males was not significantly different from uninfected males (Figure , Table ).
Figure 5 Body length of infected versus uninfected male and female hosts, 21 days post-exposure. P. ramosa induced gigantism in female D. magna but not in males (see Table 2 for details). Error bars show 95% confidence intervals. Stars in the legend represent (more ...)
Summary of differences in body length between infected and uninfected female and male hosts
As seen previously during infection with P. ramosa, infected female hosts in our experiments did not produce eggs. We tested whether infected males showed signs of castration (Experiment 5) by looking for and counting spermatozoa. All adult males had spermatozoa, but infected males had significantly lower counts (linear regression controlling for variance due to the factor "Age"; factor "Infection status", df = 1, F = 25.2, P < 0.001, Figure ). Spermatozoa counts increased with age for uninfected individuals (linear regression with quadratic term, factor ("Age")^2, df = 1, F = 10.35, P = 0.001 and factor "Age" df = 1, F = 3.39, P = 0.07, Figure , left panel), but not for infected individuals (linear regression, factor "Age", df = 1, F = 0.05, P = 0.82, Figure , right panel).
Figure 6 Spermatozoa counts in uninfected (left) and infected (right) male hosts. The counting was performed over the period where most of adult infected male Daphnia are expected to have spermatozoa and have survived infection (based on Figure 4). Infected males (more ...)