Nematode endoparasites do not codiversify with their stick insect hosts

Abstract

Host-parasite coevolution stems from reciprocal selection on host resistance and parasite infectivity, and can generate some of the strongest selective pressures known in nature. It is widely seen as a major driver of diversification, the most extreme case being parallel speciation in hosts and their associated parasites. Here, we report on endoparasitic nematodes, most likely members of the mermithid family, infecting different Timema stick insect species throughout California. The nematodes develop in the hemolymph of their insect host and kill it upon emergence, completely impeding host reproduction. Given the direct exposure of the endoparasites to the host's immune system in the hemolymph, and the consequences of infection on host fitness, we predicted that divergence among hosts may drive parallel divergence in the endoparasites. Our phylogenetic analyses suggested the presence of two differentiated endoparasite lineages. However, independently of whether the two lineages were considered separately or jointly, we found a complete lack of codivergence between the endoparasitic nematodes and their hosts in spite of extensive genetic variation among hosts and among parasites. Instead, there was strong isolation by distance among the endoparasitic nematodes, indicating that geography plays a more important role than host-related adaptations in driving parasite diversification in this system. The accumulating evidence for lack of codiversification between parasites and their hosts at macroevolutionary scales contrasts with the overwhelming evidence for coevolution within populations, and calls for studies linking micro- versus macroevolutionary dynamics in host-parasite interactions.

Keywords: Codiversification; cophylogeny; endoparasite; host–parasite interaction.

Figure 1

(A) Locations of the endoparasitic nematodes sampled in this study. Numbers in brackets indicate the number of nematodes per host species and location. Please note that the large number of nematodes collected from the T. cristinae host is explained by T. cristinae being the most intensively sampled host species (not by this species being more infected than others). (B) Picture of an endoparasitic nematode after it exited and killed its Timema host.

Figure 2

Phylogenetic placement of endoparasitic nematodes from Timema within the Nematoda phylum. (A) Maximum‐likelihood phylogeny based on the 18S rRNA sequence of 57 nematodes. The highlighted group corresponds to Clade I, which comprises the 24 Timema endoparasitic nematode sequences (see Fig. 3 for details of this clade). Numbers associated with branches indicate bootstrap support (1000 replicates). (B) Nematode orders described in each clade and (C) their trophic ecologies. Information indicated in (B) and (C) are from Blaxter et al. (1998).

Figure 3

Tanglegrams (generated with TreeMap 3.0β) comparing the nematode endoparasite phylogeny (right) to the Timema host phylogeny (left) with gray lines indicating host–parasite associations. The two endoparasitic nematode sublineages are combined in (A) and treated separately in (B) and (C).

Figure 4

Pairwise genetic distances between endoparasitic nematodes as a function of geographic distances (km) (A) Pairwise distances between sequences from all endoparasitic nematodes (B) Pairwise distances within lineages 1 and 2 (distances between sequences from different lineages are not included).