Behavioural and neural alterations in offspring of mouse dams with ulcerative colitis induced during mid-to-late gestation

Abstract

Rodent models have been pivotal in linking maternal immune activation (MIA) during pregnancy to schizophrenia- and autism-related behaviours in offspring, typically mimicking viral or bacterial infections. However, whether these findings extend to other maternal inflammatory conditions, such as ulcerative colitis (UC), a common inflammatory bowel disease (IBD) in women of childbearing age, remains unclear. This study establishes a chemical-induced UC model in pregnant C57BL/6J mice to investigate its impact on offspring psychological functioning from juvenile (postnatal days 31-40) to adult (postnatal days 73-105) stages. UC-like pathology in the mouse was induced by administering 2.5% dextran sulphate sodium (DSS) in drinking water from gestational days (GD) 9 to 12, resulting in UC pathology by GD13, persisting through GD17, with elevated maternal inflammatory cytokines (TNF-α and IL-1β). Increased IL-1β in the foetal brain at GD13 indicated foetal brain inflammation, aligning with MIA mechanisms. Offspring behavioural assessments revealed both similarities and differences compared to viral/bacterial MIA models. The gestational UC model impaired prepulse inhibition (PPI) in adult, but not juvenile, offspring, mirroring the age-dependent onset of schizophrenia-like sensory gating deficits and supporting a link between maternal immune activation and psychopathology. Spatial memory was impaired in juvenile and adult offspring, affecting familiarity discrimination but sparing working memory, suggesting a less severe hippocampal impact than in viral/bacterial MIA models. One plausible explanation for this discrepancy is that immune activation in the UC model may be weaker or more localised than in viral and bacterial infection models. This may also account for the lack of overt depression-like effect or social dysfunction in my model. Unexpectedly, the UC model induced abnormal persistent latent inhibition (LI), contrasting with deficits seen in infectious MIA models, where disrupted LI is linked to schizophrenia's positive symptoms. The persistent latent inhibition may indicate cognitive rigidity, associated with negative symptoms, possibly tied to glutamatergic hypofunction via NMDA receptors. However, the offspring showed an attenuated psychostimulant response to the NMDA receptor antagonist MK-801, unlike the heightened response in viral/bacterial MIA models, highlighting distinct neurodevelopmental outcomes. Post-mortem analyses found no increase in NMDA receptor synthesis but revealed elevated hippocampal brain-derived neurotrophic factor (BDNF) from juvenile to adult stages and increased synaptophysin expression in foetal brains at GD13 and GD17. These findings suggest early synaptic integrity changes and sustained BDNF elevation, potentially conferring resilience to NMDA receptor antagonism by supporting synaptic plasticity and mitigating excitotoxicity. This contrasts with synaptic protein deficits in MIA models, further distinguishing the impact of gestational UC. This study establishes a gestational UC model, demonstrating its distinct impact on offspring neurodevelopment compared to infectious MIA models. It highlights synaptic and behavioural adaptations, challenging assumptions about uniform MIA effects and emphasising the variability of immune-driven psychiatric risks. The lack of social impairment and opposing effects on latent inhibition and NMDA sensitivity underscore UC's specific influence. Future research should explore mechanistic links between gestational UC and neurodevelopmental changes. To conclude, my study stresses the need of exploring diverse maternal inflammatory conditions to advance understanding of their impact on neurodevelopmental disorders and psychopathology.