Generalist Eimeria species in rodents: Multilocus analyses indicate inadequate resolution of established markers

Intracellular parasites of the genus Eimeria are described as tissue/host‐specific. Phylogenetic classification of rodent Eimeria suggested that some species have a broader host range than previously assumed. We explore whether Eimeria spp. infecting house mice are misclassified by the most widely used molecular markers due to a lack of resolution, or whether, instead, these parasite species are indeed infecting multiple host species. With the commonly used markers (18S/COI), we recovered monophyletic clades of E. falciformis and E. vermiformis from Mus that included E. apionodes identified in other rodent host species (Apodemus spp., Myodes glareolus, and Microtus arvalis). A lack of internal resolution in these clades could suggest the existence of a species complex with a wide host range infecting murid and cricetid rodents. We question, however, the power of COI and 18S markers to provide adequate resolution for assessing host specificity. In addition to the rarely used marker ORF470 from the apicoplast genome, we present multilocus genotyping as an alternative approach. Phylogenetic analysis of 35 nuclear markers differentiated E. falciformis from house mice from isolates from Apodemus hosts. Isolates of E. vermiformis from Mus are still found in clusters interspersed with non‐Mus isolates, even with this high‐resolution data. In conclusion, we show that species‐level resolution should not be assumed for COI and 18S markers in coccidia. Host–parasite cospeciation at shallow phylogenetic nodes, as well as contemporary coccidian host ranges more generally, is still open questions that need to be addressed using novel genetic markers with higher resolution. Eimeria is described as tissue and host‐specific. For species infecting rodents, however, a broader host range has been reported based on phylogenetic inference with 18S and COI markers. We developed a multilocus genotyping approach to increase the resolution for particular parasite groups and show that the approach is able to differentiate a genetic cluster with specialized host usage in a case unresolved by 18S and COI. In another case, the approach recovers a cluster of isolates interleaved with regards to their host usage, and question whether failure to resolve parasite clusters with specialist host ranges should be interpreted as evidence for host generalism.