The gastrointestinal nematodes of plains and Grevy's zebras: Phylogenetic relationships and host specificity

Abstract

Equids are chronically infected with parasitic strongyle nematodes. There is a rich literature on horse strongyles, but they are difficult to identify morphologically and genetic studies on strongyles infecting other equid species are few, hampering studies of host specificity. We sequenced expelled worms from two sympatric zebra species in central Kenya to expand the strongyle phylogeny and used DNA metabarcoding on faecal samples to genetically characterize zebra nemabiomes for the first time. We generated sequences for several species new to public genetic reference databases, all of which are typical strongyles in wild zebras (i.e., the three species of Cylindropharynx and Cyathostomum montgomeryi), and identified their closest relatives. We also discovered an apparent fungus infecting a quarter of the expelled Crossocephalus viviparus worms, a hyperabundant nematode species in the family Atractidae, hinting at the possibility that zebra host-parasite dynamics may involve a zebra-fungus mutualism. The two zebra species had similar nemabiomes; we found a complete overlap in the list of nematode species they carry and very similar prevalence (i.e., proportion of hosts infected) for the different nematode species. Our study suggests limited host-specificity in zebra strongyles and high potential for transmission between the plains zebra and the endangered Grevy's zebra.

Keywords: DNA metabarcoding; Equid nemabiome; Equid parasitology; Strongyle phylogeny; Zebra nematodes.

Graphical abstract

Fig. 1

Phylogeny of all sequenced Strongylidae with species-level taxon assignments from GenBank and the present study, coloured by genus (the same colour code is used across all figures in this paper). Branch tips were pruned such that only one tip was kept of all immediate sister taxa with identical taxon assignments, and the number of samples merged in each tip is indicated in parentheses. Bootstrap percentages over 50% are displayed in bold to highlight nodes with high support. Branch lengths represent the number of base substitutions per site, estimated with the Tamura 3-parameter model assuming gamma-distributed substitution rate variation.

Fig. 2

Nematode prevalence in plains vs. Grevy's zebras in (a) a bipartite graph, where edge widths indicate prevalence in each zebra species, and (b) a linear regression, shown in black with shading representing standard error and the one-to-one line indicated by the dashed red line. Only sequences comprising >1% of total reads were used and they were clustered into mOTUs by 98% similarity. Taxon labels followed by a letter signify species-level matches (>98% similarity) to reference worms identified only to genus (see SI2), while those followed by a number represent sequences that matched a reference only to the genus level (>95% similarity). Black boxes/points are taxa without a match of >95% to any identified sequences. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 3

Crossocephalus viviparus extracted from fresh zebra faeces were often infected with an apparent fungus. Hyphae emerging from the head (left, right) and from the tail (centre) of infected worms and stained with lactophenol blue (right). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)