Central amygdala neuronal ensembles contribute to food-related conflict in a novel eating disorder model

Background

Anorexia nervosa (AN) is a severe eating disorder characterised by a fear of weight gain, low body weight, and excessive physical activity [1]. It presents one of the highest mortality rates among adolescent psychiatric conditions and remains challenging to treat effectively. Despite the disorder’s clinical severity, the neurobiological mechanisms underlying AN are still poorly understood [2]. Current treatment approaches primarily target behavioural symptoms and lack efficacy in addressing underlying neurobiological drivers. In this study we developed a novel mouse model to investigate the food-related conflict characteristic of patients with AN, which has not been adequately represented in animal models, by combining two established approaches (Activity-based anorexia (ABA) and conditioned taste avoidance (CTA)) to trigger strong avoidance behaviour even under severe hunger.

Methods

In this study, we used female mice used to characterise food intake, hyperactivity, and anxiety-like behaviour in mice undergoing the Food Approach-Avoidance (FAAC) model. TRAP2 mice (Targeted Recombination in Active Populations), a transgenic line in which a 2A-iCreER^T2 cassette is inserted into the Fos locus, allowing for cue-dependent, tamoxifen-inducible labelling of neurons. This system was employed to investigate neurons active during food approach avoidance conflicts. Our primary objective was to identify and subsequently manipulate neurons active during food-related conflict to allow for investigation of the underlying neurocircuitry. Injection of viral vectors for chemogenetic targeting of labelled neurons in the Central Amygdala (CeA) was employed to test whether these neurons were sufficient to drive behavioural changes.

Results

Mice exposed to the FAAC model showed no changes in hyperactivity, nor general food intake. While not displaying differences in general anxiety tests (Elevated Plus Maze, Light Dark Box, Marble Burying Test), mice undergoing the FAAC model showed changes in food related anxiety (Chocolate Burying Test) compared to control 4 groups including FAAC, ABA, CTA, running control, sedentary control, early CTA), n = 6 per group, p < 0.001, using a one-way ANOVA followed by Dunnett’s post-hoc test) and extinction of conditioned food avoidance (4 groups including FAAC, ABA, CTA, negative control), n = 7-8 per group, p < 0.001, using a one-way ANOVA followed by Dunnett’s post-hoc test) pointing towards differences in the processing of food-related conditioned learning under AN-like conditions. Chemogenetic modulation (3 groups (TRAP2 mice with either activating DREADDs, inhibitory DREADDs, or control virus specifically expressed in ensembles active during ABA) of CeA neurons active during the FAAC paradigm allowed us to dissect their functional involvement in mediating AN-like behaviour. Activation of these ensembles, resulted in significantly longer extinction time of conditioned taste avoidance (3 groups including activating DREADDs, inhibitory DREADDs, or control virus), n = 8 per group, p =0.0221, using a log-rank test), indicating a potential role in food-related decision making under AN-like conditions.

Conclusion. We propose a novel rodent model for food-related conflict thereby creating a platform to further investigate the underlying neurobiology driving this behaviour. Our results highlight the CeA as player in the neurocircuit mediating food approach-avoidance conflict.

References

Battle D. E. (2013). Diagnostic and Statistical Manual of Mental Disorders (DSM). CoDAS, 25(2), 191–192. https://doi.org/10.1590/s2317-17822013000200017
Fichter, M. M., & Quadflieg, N. (2016). Mortality in eating disorders - results of a large prospective clinical longitudinal study. The International journal of eating disorders, 49(4), 391–401. https://doi.org/10.1002/eat.22501