The genomics of mutualistic dependence in the plant-associated ant genus Philidris

Abstract

Mutualisms are ubiquitous, ecologically and evolutionary important and highly diverse. One element
of mutualism diversity that is relatively poorly studied is mutualistic dependence, the level to which a
species relies on its mutualistic partner. Mutualistic dependence is important to many aspects of
mutualism evolution, particularly stability and has even been found to affect the evolution of traits not
involved in the mutualism. The effect of mutualism on genome evolution is also understudied with the
exception of bacterial endosymbionts. Here, I investigated genome evolution in the ant genus Philidris
(Dolichoderinae). Philidris is most well-known for the Fijian farming ant Philidris nagasau, which has
evolved an obligate, highly specialised farming mutualism with six species of the epiphytic plant
genus Squamellaria (Hydnophytinae, Rubiaceae). In addition to the highly specialised mutualism,
several Philidris species engage in a facultative generalist “ant-garden” mutualism. Additionally, there
are some Philidris populations that do not engage with plant mutualism at all. This variability in
mutualism strategy makes Philidris an ideal system to study mutualistic dependence.

First, I establish the extent of plant-cultivation by ants by performing a systematic review and creating
a database of ant-farmed plants and plant-farming ants. I assess the diversity of these mutualisms in
relation to taxonomic diversity, presence of domatia and food rewards, specificity, and dependence
and finally assess how these systems compare with the conceptualisation of “true” agriculture. I
conclude that this framework is overly restrictive and propose a new inclusive eco-evolutionary
framework encompassing all farming systems.

Next in Chapter 3, I assemble and annotate two genomes (of the species P. nagasau and P.
laevigata) as well as 38 genomes from high coverage Illumina resequencing data representing 9
operational taxonomic units (OTUs). I use these genomes to construct a genome-wide phylogeny
from 2,100 genes. This comparative genomic and phylogenomic framework is the foundation of the
next two chapters.

In Chapter 4 I investigate gene family expansion and contractions within Philidris and find that P.
nagasau has the most significantly expanding and contracting gene families. Within these expanding
and contracting gene families are olfactory receptors (ORs), gustatory receptors (GRs) and genes
relating to the nervous system, especially the neurotransmitter glutamate. I investigate signatures of
selection within the OR repertoire of P. nagasau and find that 11% have significant evidence for
diversifying selection.

In Chapter 5, I investigate signatures of selection within Philidris single copy orthologues to
investigate genome-wide signatures evolution relating to the shift to obligate specialist mutualism.
Philidris nagasau has a higher average ratio of non-synonymous to synonymous substitutions (dN/dS)
which appears to be a result of both positive selection on some genes and relaxed purifying selection
on others. Philidris nagasau has a higher nucleotide substitution rate compared to a sister clade,
indicating that the shift to obligate specialist mutualism evolves with Red Queen dynamics.

These results suggest that the obligate specialist mutualism between P. nagasau and Squamellaria is
likely mediated by olfactory cues. We detect signatures relaxation and intensification in the genome,
mirroring the losses and gains in behaviours in P. nagasau.