556. Distinct neural signatures of heightened anxiety in autism-related and trait anxiety models

Abstract

Abstract Background Anxiety disorders represent a major clinical burden across neurodevelopmental and affective conditions, yet treatment response remains highly variable. Anxiety is highly prevalent among individuals with autism spectrum disorder (ASD) and is often more difficult to treat than normotypic pathological anxiety, suggesting that the underlying neural substrates may differ from those mediating classical trait anxiety. However, the extent to which these circuits overlap remains unclear. ASH1L haploinsufficiency is a high-confidence ASD risk factor associated with some evidence of elevated anxiety-related behaviour while mice selectively bred for high anxiety-related behaviour (HAB) mice represent a validated model of pathological trait anxiety. Understanding whether these phenotypes converge or diverge at the circuit level is critical for developing mechanism-informed therapeutic strategies. Aims & Objectives This study aimed to characterize anxiety-related whole-brain neuronal activation patterns in ASH1L+/− and HAB mice following open arm (OA) exposure as compared with their wildtype/normotypic controls, and to compare activation patterns across models to identify shared versus model-specific neurobiological correlates relevant to clinical anxiety subtypes. Method ASH1L+/− mice and wildtype controls, as well as HAB and normal anxiety (NAB) counterparts were exposed to the blocked OA of the elevated plus maze. Anxiety-like behaviour was quantified, and stimulus-evoked neuronal activation was mapped using c-Fos immunohistochemistry across cortical, amygdalar, hypothalamic, and hindbrain regions. Results Male HAB mice and male ASH1L+/− mice displayed increased anxiety-like behaviour during OA exposure as indicated by the reduced time spent in the distal part of the OA as compared with their controls. HAB mice, compared to NABs, showed strong OA-induced hyperactivation of classical stress- and anxiety-associated areas including limbic, hypothalamic and brainstem regions. Compared with wild-type controls, male ASH1L+/− mice showed a tendency toward increased c-Fos expression in frontal and somatosensory cortices following OA exposure, but either unchanged or even reduced c-Fos expression in regions implicated in anxiety regulation and cognitive appraisal including basal amygdala subnuclei, prefrontal, motor, and insular cortices. Notably, both the HAB and ASH1L+/- models showed a reduced c-Fos response in the cingulate cortex compared with their respective control group. Discussion & Conclusions These findings demonstrate that similar anxiety phenotypes can arise from fundamentally distinct neural mechanisms. While trait anxiety in HAB mice is characterized by hyper-reactive stress and/anxiety circuitries, anxiety associated with ASD-related ASH1L haploinsufficiency reflects in part impaired recruitment of cortical–amygdalar networks critical for threat evaluation and regulatory control. This distinction mirrors clinical heterogeneity in anxiety disorders where normotypic anxiety patients exhibit exaggerated stress reactivity while patients with ASD-associated anxiety show deficits in cognitive-emotional integration. Together, this work highlights the need for circuit-informed stratification of anxiety phenotypes to guide personalized neuropsychopharmacological treatment approaches. Supported by the Austrian Science Fund FWF FG 18-B.