Population genetics of multi-host parasites--the case for molecular epidemiological studies of Schistosoma japonicum using larval stages from naturally infected hosts.

Population genetics of multi-host pathogens offers great potential for the understanding of their complex epidemiology but care must be taken to ensure that the sampling procedure does not bias estimates of population indices. The transfer of material to laboratory passage, in particular, runs the risk of bottlenecking and imposing non-random host-induced selection pressures according to the hosts used in passage. We present a novel technique allowing single-locus microsatellite genotyping of the naturally sampled larval stages, enabling unbiased population genetic studies of the multi-host zoonotic parasite Schistosoma japonicum. The utility of these larval genotyping methods for molecular epidemiological studies are illustrated in results from 3 separate data sets. In the first data set, potential loss of alleles based on the definitive host species used for laboratory maintenance was identified by comparing adult worm populations derived from mice and rabbits infected with cercarial populations originating from the same set of snails. In the second data set, bottlenecking was demonstrated by the loss of alleles in adult worms derived within a single generation of laboratory maintenance compared to their parent field-collected cercarial samples. In the final data set, comparison of miracidia and adult worms recovered from naturally infected animals demonstrated that larval analyses can provide stage-specific epidemiological information and that population genetics of schistosomes can be well described by analysis of larval stages. Our results thus advocate the use of natural life-cycle stages to obtain an accurate and ethical representation of the population genetic structure of S. japonicum and other multi-host pathogens.