Self-initiated restrictive feeding: a new mouse model of restrictive eating disorders

Abstract

Restrictive eating disorders are serious psychiatric illnesses carrying the highest morality of any mental health condition, yet their neurobiology remains poorly understood. Progress is constrained by a shortage of animal models with convincing construct validity. In particular, anorexia nervosa (AN), a self-initiated restriction of caloric intake, is notoriously difficult to replicate in mice, which would otherwise offer an ideal pathway to understanding the neural mechanisms of this strongly gene-linked disorder, similar to advances in other neuropsychiatric conditions. Here, we introduce a model of self-initiated restrictive feeding (SIRF), using open-source behavioural devices. We show that SIRF specifically affects female mice, persists in the absence of the initial triggering conditions, and correlates with social rank - features that parallel AN. Meal pattern analysis shows that SIRF arises alongside decreased meal size and duration, with meal frequency unaltered in the long-term. Chemogenetic inhibition of the insular cortex, a brain region consistently implicated in AN and interoceptive processing, reduced SIRF. In sum, this proof-of-principle study suggests that SIRF provides a tractable model of self-initiated restriction as a platform for dissecting the neural circuits that initiate and sustain restrictive eating.