The relationship between foraging efficiency, neural structure and learning in bumblebees (Bombus terrestris)

In recent decades, social bee populations have been facing pressure from anthropogenic stressors such as habitat fragmentation and exposure to pesticides. In order to help inform policies to protect the ecosystem services that these species provide it is important to better understand the main behavioural driver of this service: foraging, and the proximate mechanisms that underlie it. Here I explore the topic of learning in foraging bumblebees Bombus terrestris, focusing on the relationship between foraging efficiency and neural plasticity, and how stressors such as pesticides can affect this relationship.
In Chapter 2, I begin by developing immunostaining methodology for brain tissues of B. terrestris, based on modification of existing protocols that were not reproducible. This technique allowed me to visualise and quantify the density of synaptic boutons (also known as microglomeruli; MG) within a neuropile of the insect central nervous system associated with learning, the mushroom bodies.
I then use this method to explore the relationship between foraging and MG density in Chapter 3, as learning abilities correlate with MG density and better learners are expected to be more efficient foragers. I measured the foraging efficiency of workers in 6 colonies during summer 2019 and quantified the MG of these foragers. While no significant relationship was detected, this chapter highlights the gaps in knowledge regarding foraging efficiency and offers routes to attempt to close them.
In the two following chapters (4 and 5) I present a case study on a novel neurotoxic insecticide, sulfoxaflor, that has documented negative sublethal effects on B. terrestris individuals and colonies. I investigated how this pesticide could influence the relationship between neural structure, learning and foraging strategies. I first tested the effect of chronic exposure to sulfoxaflor on the synaptic density of the main centre for stimulus integration and important for foraging in bumblebees, the mushroom bodies (MB). No significant effect of the compound was detected either on MG density or volume of the MB. I then focused on testing short-term memory, which is less likely to have a direct neural correlate, using a radial-arm maze (RAM) mimicking within-patch foraging and for which I present a validation experiment of the paradigm. Once again, I found no significant effect of the insecticide on the foragers’ performance in the RAM.
The results of this thesis show that the relationship between foraging, learning and neuroanatomy is less straightforward than traditionally believed and, additionally, that sulfoxaflor is unlikely to be causing negative effects to bumblebee colonies through a disruption of this relationship.