The masking, unmasking, and re-masking of food memories: neuronal ensemble mechanisms

Alarming estimates indicate that 2 out of 3 UK adults are overweight or obese, as a result of overeating, with an increased risk for diseases such as diabetes. Exposure to sensory stimuli or cues (e.g., the sight of food) linked with food recovers 'excitatory' memories about food and can elicit food cravings, which leads to excessive eating in certain individuals. Interestingly, cue exposure therapy (CET) reduces such desires to eat. It borrows well-established and elegant concepts from 'extinction learning' that Pavlov examined in his famous classical conditioning experiment nearly 100 years ago. He noticed that a dog presented with a bell associated with food would salivate, but when the bell was repeatedly presented without any food, the salivation response stopped or went 'extinct'. Likewise, in CET a food cue is repeatedly presented without food and reduces the desire to eat by creating a new 'inhibitory' memory. However, CET's success is limited and responses to food cues eventually return on their own or show 'spontaneous recovery', as memories about extinction learning are not properly retrieved.

Promisingly, animal studies from we and others have revealed that reminders or 'retrieval cues', such as a flashing light, associated with extinction learning help retrieve inhibitory extinction memories and dampen spontaneous recovery. To date, the brain mechanisms of spontaneous recovery and how it is suppressed are poorly understood. However, our recent studies in mice show that specialised groups of activated neurons called 'neuronal ensembles' help retrieve excitatory memories about food rewards when they are exposed to food cues and show food seeking behaviours. Moreover, we and others have shown that following extinction learning, food cues activate smaller neuronal ensembles and there are also physiological changes in the electrical properties of neurons and connections between neurons called synapses, which 'rewire' brain circuits. As such, might spontaneous recovery occur by reversing these changes in activity patterns and physiological properties of neurons related to extinction? And could extinction retrieval cues restore these properties back to what was seen during extinction learning?

To answer these important questions, we will investigate the activity of neuronal ensembles in mice during the emergence and suppression of spontaneous recovery of food seeking that is triggered by food cues and extinction retrieval cues, respectively. We will focus on the nucleus accumbens, a brain area that controls rewarding actions such as eating. We will first use devices that can probe the electrical and synaptic properties of individual neurons and reveal why these neurons show certain activity patterns during spontaneous recovery and its suppression. Next, we will use biochemical approaches to determine which specific neurons are activated during spontaneous recovery and whether extinction retrieval cues will switch off these neurons. In another experiment, we will label or 'tag' neurons that are activated during food seeking and extinction with a special chemical activity sensor called 'GCaMP', and then live record their fast neuronal activity during the emergence and suppression of spontaneous recovery using a technique called 'fibre photometry'. Using this tool, we link the neural activity with the behaviour of the mice in response to cues associated with food seeking and extinction experiences. Finally, we will rapidly manipulate the activity of these neurons using a light-based approach called 'optogenetics' to see if they actually control spontaneous recovery and its suppression. Taken together, our studies will help us better understand how to make CET more robust and long-lasting by strengthening inhibitory extinction memories that will better suppress our reactions to food cues.

Exposure to food cues retrieve 'excitatory' memories about food and elicits food cravings in humans and food seeking in mice. Cue exposure therapy (CET) reduces these cravings by using principles from extinction learning where food cues are repeatedly presented in the absence of food, creating a new 'inhibitory' memory. Unfortunately, its efficacy is limited and impeded with the passage of time due to the 'spontaneous recovery' of food cravings, which is proposed to be due to a failure to retrieve extinction memories. Encouragingly, retrieval cues associated with extinction learning, such as a flashing light, attenuate spontaneous recovery of food seeking in laboratory animals. Extinction learning modulates the recruitment of sparse sets of activated neurons or 'neuronal ensembles' that express the activity marker 'Fos' in the nucleus accumbens, a brain area implicated in reward. Also, extinction induces physiological adaptations (e.g. excitability changes) in these neurons. Here we will test the hypothesis that spontaneous recovery and its suppression of cue-evoked food seeking are modulated by the reversal and restoration, respectively, of extinction-associated adaptations and activity patterns in Fos-expressing neuronal ensembles in the nucleus accumbens. We will examine precise changes in synaptic physiology and intrinsic excitability of accumbens ensembles using ex vivo electrophysiology and neuronal morphology analysis. Real-time activity patterns of GCaMP-tagged accumbens ensembles will be assessed using in vivo fibre photometry. Finally, we will use optogenetics to interrogate whether these accumbens ensembles play a causal role in behaviour. This study will provide novel insights into the little-understood brain mechanisms driving spontaneous recovery and its suppression and has implications for understanding how to improve CET's efficacy.