A neural and behavioural investigation of aberrant feeding in animal models of autism spectrum disorder

Abstract

Autism spectrum disorder (ASD), a neurodevelopmental disorder affecting approximately 1.85% of the population, is underpinned by complex interplay between genetics and environment. Thus far, ASD research has focused predominantly on abnormal social communication, anxiety, and cognitive impairments. It omits, however, dysregulated food intake, which is the symptom present in over 60% of ASD individuals. Data suggest that feeding abnormalities include heightened food selectivity, neophobia and food refusal, as well as pica (consumption of non-food items). The current knowledge of the scope and magnitude of abnormal appetite in ASD as well as neuropathology underlying this phenomenon is extremely limited. Thus, the overarching goal of this thesis was to provide systematic evidence pertaining to aberrant food intake in ASD and identify the relevant mechanisms, by utilizing two key animal models of this disorder (one in which ASD stems from a genetic mutation, and the other, in which ASD is induced pharmacologically). In Specific Aim 1 of this thesis, I investigated the hypothesis that ASD mutant male mice overconsume palatable foods beyond the level observed in neurotypical controls. In this animal model, ASD stems from the deletion of the contactin-associated protein like 2 (CNTNAP2) gene encoding a transmembrane neuronal protein which clusters K+ channels in the nodes of Ranvier. My studies show for the first time that Cntnap2-/- males drink more of sucrose, saccharin and intralipid solutions compared to wild-type (WT) controls. A c-Fos immunoreactivity analysis revealed that at the end of a sucrose solution meal, Cntnap2-/- mice have blunted activation in key regions associated with satiety and reward. Specifically in the central amygdala (CEA), dorsomedial hypothalamus (DMH), ventromedial hypothalamus (VMH) and the nucleus accumbens (Acb). Excessive consumption palatable sucrose sucrose was was ameliorated in mutants by intraperitoneal administration of oxytocin (OT), a satiety promoting neuropeptide whose dysregulated release is also associated with the broad symptomology of ASD, from anxiety to sociality impairments. In line with the injection experiment, the double staining for c-Fos and OT showed blunted OT neuronal activation in the hypothalamus of the mutants after a meal of sucrose. Overall, the data indicate that ASD Cntnap2-/-male mice show excessive intake of palatable foods, and it is associated with atypical neural activation (including the OT system) in response to a palatable meal. In Specific Aim 2, I examined whether the overconsumption of sucrose observed in male Cntnap2-/- mice also occurs in females. Research on ASD animal models often focuses on males due to concerns about oestrus driven behavioural variability and the subtler presentation of ASD-related behaviours in females. However, feeding issues in ASD affect both sexes. To address this, I evaluated sucrose intake in female Cntnap2-/- mice and compared their consummatory behaviour across different stages of the oestrous cycle. Female Cntnap2-/- mice exhibited overconsumption of sucrose, which remained consistent across all oestrous phases. This study suggests that palatability-driven overconsumption of sugar occurs in female Cntnap2-/- mice and does not appear to be influenced by oestrus phase. In Specific Aim 3, I investigated the hypothesis that Cntnap2-/- mice exhibit heightened hyponeophagia, a common feeding issue associated with ASD. I assessed hyponeophagia in Cntnap2-/- mice both in a novel environment and in their home cage. In both contexts, Cntnap2-/- mice displayed a significantly greater latency to approach novel food compared to WTs. Notably, in the home cage more instances of presentation of that food were required for the latency in Cntnap2-/- mice to decrease to match that of WTs. The analysis of c-Fos immunoreactivity revealed different densities of activation in key brain regions associated with reward, feeding, and fear processing between Cntnap2-/- and WT mice in response to novel food presentation. These regions included the ventral tegmental area (VTA), CEA, and arcuate nucleus (ARC). These findings suggest potential impairments in neural processing of food-related novelty cues, which may contribute to the exaggerated neophobic behaviours observed. Valproate (VPA), an anti-epileptic drug and known teratogen, induces an ASD phenotype in offspring when administered during the period of neural tube closure. Specific Aim 4 tested whether rats prenatally exposed to VPA exhibit impaired acquisition of the conditioned taste aversion (CTA), a protective learning mechanism against ingestion of foods inducing nausea/malaise. Feeding issues in ASD, such as pica, may reflect disruptions in aversion-based learning. Additionally, OT has been implicated in CTA acquisition, and VPA rats exhibit OT signaling abnormalities. Using a standard CTA paradigm with lithium chloride (LiCl) as a nausea-inducing agent, I found that VPA rats failed to acquire a CTA response. Gene expression analysis via qPCR revealed altered expression of catecholamine metabolism-, reward-, aversion-, and satiety-related genes in the CEA in LiCl-treated VPA rats compared to controls. These findings suggest that prenatal VPA exposure disrupts neural mechanisms critical for aversion-based learning, with the CEA playing a key role. In summary, this thesis provides novel insights into the neuromolecular mechanisms underlying feeding behaviours in ASD, highlighting distinct anomalies in palatable food intake, food neophobia, and aversive learning. Using Cntnap2-/- mice and VPA-exposed rats, the studies demonstrate significant overconsumption of palatable foods in both sexes, prolonged neophobic responses, and disrupted conditioned taste aversion acquisition, all underpinned by aberrant brain activation and gene expression.