Discussion
Myxozoans are a group of endoparasitic cnidarians with complex life cycles, exploiting invertebrates and vertebrates (usually fish) as definitive and intermediate hosts, respectively. It is now known that covert myxozoan infections are vertically transmitted to dormant overwintering stages (statoblasts) of bryozoan hosts. After statoblast dormancy is broken overt infection development triggered in newly hatched bryozoans enables transmission to fish hosts via infectious spores. In at least one system such vertical transmission is substantial - up to 45% of statoblasts produced by the bryozoan Fredericella sultana are infected by the myxozoan Tetracapsuloides bryosalmonae. Here we investigate whether myxozoan infections are vertically transmitted to statoblasts in a contrasting system along with potential drivers of variation in infection prevalence.
The bryozoan
Cristatella mucedo produces hooked statoblasts that float upon release, unlike the adherent statoblasts produced by
F. sultana. Previous research has demonstrated metapopulation dynamics with local
C. mucedo populations undergoing extinction and (re)colonization. Infections by the myxozoan
Buddenbrockia bryozoides have been hypothesized to contribute to local extinctions of
C. mucedo while colonisation has been strongly linked with transport of statoblasts by waterfowl. We collected 1349
C. mucedo statoblasts from 48 sites in four regions of the United Kingdom representing lowland agricultural (Norfolk and Northern Ireland), upland (Cumbria), and urbanized (Greater Glasgow) landscapes. In each region we collected samples from hydrologically connected, semi-connected and isolated sites. Following DNA extraction, statoblasts were screened for myxozoan DNA using targeted primers for a portion of the 18S rRNA gene. We also genotyped each statoblast using 10 microsatellite markers in order to characterise patterns of infection in highly clonal bryozoan populations. Our results provide evidence for vertical transmission with infected statoblasts detected in 37 of 48 populations. For infected populations the mean infection prevalence ≤ 0.20 (SD ≤ 0.12, range ≤ 0.3 - 43%; n ≤ 37). Such vertical transmission may enable myxozoan infections to be spread to new sites by waterfowl vectors along with their bryozoan hosts. Infection prevalence was significantly greater in populations from Norfolk and Northern Ireland suggesting that agricultural environments may favour parasites as a result of nutrient- enhanced primary production and hence more food for suspension-feeding bryozoan hosts. Hydrological connectivity exerted no apparent effect on infection prevalence, thus the spread of infection may be achieved by other processes (e.g. waterfowl movements). Finally, we obtained evidence for disproportionate infections of common host clones. This result suggests that Red Queen dynamics may often characterise even relatively ephemeral local populations. There was no correlation between the number of host clones and infection prevalence. Our investigations provide evidence that vertical transmission of myxozoans appears to be a common strategy that creates a reservoir for fish disease, and they identify both landscape and invertebrate host genotype as drivers of infection prevalence.
