Modularity of gene expression in butterflies

SNIC 2017/7-236


SNAC Small

Principal Investigator:

Sören Nylin


Stockholms universitet

Start Date:


End Date:


Primary Classification:

10615: Evolutionary Biology



The present proposal follows up on the previous research done in collaboration with SciLifeLab-WABI bioinformatics support (project: “RNA-Seq for EcoEvo”- RNA-Sequencing for Ecology and Evolution). In that collaboration we tested whether the similarities of gene-expression response in herbivorous caterpillars (Vanessa cardui) reflect the evolutionary history of adaptation to plants in the Vanessa and related genera, against the null hypothesis of gene-expression responses reflecting the phylogenetic relatedness of the plants in the study (as a proxy for chemical similarity). We found two different sets of gene expression, one associated with the caterpillars feeding on a group of plants forming an ancestral host repertoire for the taxon and one associated with the caterpillars feeding on new hosts, regardless of plant chemistry (Celorio-Mancera, et al. 2016 BMC Evolutionary Biology). These findings suggest that butterflies may retain an ancestral capability to consume plants to which their relatives have adapted through evolutionary time. In order to test this hypothesis, we will compare the expression profiles among different butterfly species within a related butterfly taxon (the Polygonia-Nymphalis group) spanning 12 My of evolution, which share at least one common plant-host i. e. Urtica dioica (nettle). The samples ready for submission consist of RNA extracted from the gut of caterpillars of the following species: Polygonia satyrus, Nymphalis xanthomelas, P. faunus and P. c-album exposed to different diet treatments. P. satyrus is a specialist on U. dioica and N. xanthomelas is a specialist on Salix sp (e.g. sallow) but the caterpillar of N. xanthomelas is also able to feed on U. dioica. P. faunus and P. c-album can consume U. dioica, as mentioned before, but also Salix sp and Ribes sp (e.g. gooseberry). Thus, we have 9 combinations of butterfly and host plant which we replicated sampling guts from 3 individuals from 3 butterfly families per combination, for a total of 81 RNA samples. We hypothesize that gene expression clusters will be found according to specifically adaptive modules, in this case by host regardless of butterfly species. That is, gene expression patterns on the same host plant in the same tissue will be more similar even in distantly related species reflecting the adaptation to those hosts ancestrally shared. Thus we predict that gene expression on U. dioica, on Ribes sp and on Salix sp respectively, will cluster together across butterfly species. Moreover, we hypothesize that gene expression on U. dioica will cluster together even across the 30 My separating P. c-album from V. cardui (using here data generated in our previous collaboration), because based on phylogenetic reconstructions of host-plant use in these butterflies and their relatives, the ancestral host for the entire tribe Nymphalini is also U. dioica.