The genetic and neural bases underlying predator aversion behavior in deer mice
Sensory and Behavioural Biology
Final Report Abstract
Our understanding of how behaviors evolve remains limited – especially in the context of natural populations of mammals. How does genetic variation act through neural circuits to translate to behavioral variation? In this project I combined population genomics with behavioral assay to investigate natural population of deer mice in the Californian Channel Islands that have evolved in the presence or absence of island foxes – the only mammalian top predator on these islands. While deer mice co-occur with island foxes on six islands, they have evolved in the absence of foxes on the two smallest during their recent evolutionary history. Previous trapping experiments suggested that mice on these two islands have lost their aversion to fox feces. Aversion to predator odors, including fox feces, is typically a non-conditioned innate behavior, making the apparent lack in two natural populations an attractive system to investigate its genetic and neural bases and evolution. By generating whole genome resequencing data for 68 mice from the eight islands and an outgroup from mainland southern California, I was able to infer the evolutionary relationships of all populations. My results show that mice on the two fox-free islands have evolved changes in this behavior twice independently. Genome scans revealed 101 and 175 genes in regions under positive selection in the two fox-free populations. These genes are enriched for gene ontology terms related to behavior and neuronal phenotypes (e.g., synapse, dendrite, neuron projection). Moreover, I detected 41 and 54 olfactory receptor genes with highly differentiated derived coding changes in the two fox-free populations, respectively; nine of these receptor genes are shared between the two populations. The identified genes under positive selection and olfactory receptor genes constitute interesting candidate genes for further investigation. Behavioral assays conducted on one fox-containing and one fox-free island in the field confirmed that mice on the fox-free island do not seem to show aversion to fox odor (2,5-dihydro-2,4,5-trimethylthiazoline; TMT), in contrast to mice on the fox-containing island that spent significantly more time away from fox odor than a no odor control. Interestingly, mice on the fox-containing island also avoided a nonpredator odor (Carvone), whereas mice on the fox-free island did not. This could indicate baseline differences in neophobia and boldness, but further behavioral assays are needed to test this hypothesis. Overall, this project laid the foundation and generated valuable resources for ongoing investigations of the genetic and neural bases underlying the repeated evolution of predator aversion in this fascinating natural system of closely related populations of deer mice.
