We had three major objectives with the overall aim of advancing our understanding of emotional states in honey bees, focusing on the neural mechanisms underlying these states and their phylogenetic link to those of mammals, including humans. We specifically selected the emotion of fear for this study, as fear responses are well-characterized in vertebrates. Our central question was whether bees react to fear-inducing stimuli in a purely "reflexive" manner or whether there is a subjective component accompanying these reactions. To address this, we outlined three work packages. In Objective 1, we characterized the behavioural and physiological correlates of fear in honey bees. The bees were subjected to a nociceptive stimulus (electric shock) within a classical conditioning protocol that allowed them to form expectations about that stimulus. We developed and refined a protocol (originally used at the host institution, ICARUS) that pairs a colour signal with an electric shock in a two-chamber compartment. Our results showed that bees not only learned the association between a light and the negative event but also retained this negative association for up to one hour. Additionally, we found that the intensity of the shock (5, 20, and 60 µA) modulated their behaviour (slowing or speeding the learning rate of individuals). This behavioural characterization was made possible through a custom-made Python script that allowed us to track and subsequently extract fine movements of the bees under testing conditions. Furthermore, we discovered that bees that learned the association exhibited higher CO2 emissions and thorax temperatures compared to control groups, i.e. bees that did not receive a shock when approaching the light or bees that received a shock that was not temporally contingent with the light. Overall, across six experiments that combined behavioural and physiological measures, the observed patterns strongly suggest that the bees' reactions were not simply reflexive. Subsequently, in Objective 2, we aimed to identify the molecular cascades of fear. We performed brain dissection on the samples collected during Objective 1 by dividing the brain in two main regions: the central brain and the optical lobes. Using High-Performance Liquid Chromatography (HPLC) we quantified the amount of Dopamine (DA), Serotonin (5HT) and Octopamine (OA) in the brain of 60 honey bees. We found that bees that learned the association between the light and the negative event have elevated amount of Serotonin and Octopamine compared to the control groups, indicating the potential roles of these neurotransmitters in the emotional processing. We further explored the role OA and 5HT in the emotional process of fear by conducting extensive pharmacological experiments. Depending on the experiment, we topically treated the bees’ brain with a specific drug, before testing (as described in Objective 1). Based on the HPLC analysis, we decided to focus on three main drugs, known to specifically block the octopaminergic and serotonergic systems. In detail, we conducted an initial experiment in which bees were topically treated with a drop of 1 µL of drugs on the cuticle, 30 minutes before conditioning. Bees were divided into nine groups: a No Shock group, Unpaired and Paired groups treated only with the solvent (Dimethylformamide) as controls, Paired groups treated with Epinastine (OA antagonist) at two different concentrations (1 and 10 nM), and Paired groups treated with a cocktail of Ketanserin and Methiotepin (5-HT antagonists) at two different concentrations (1 and 10 nM); finally, a Paired group treated only with Methiotepin at two different concentrations (1 and 10 nM). We found that bees treated with the Serotonin antagonist failed to learn the task. The involvement of Serotonin, a neurotransmitter widely known to be involved in mammalian emotions, highlights the phylogenetic link between invertebrate and vertebrate neurobiology.
