Early postnatal allergic airway inflammation induces dystrophic microglia leading to excitatory postsynaptic surplus and autism-like behavior

Microglia play key roles in synaptic pruning, which primarily occurs from the postnatal period to adolescence. Synaptic pruning is essential for normal brain development and its impairment is implicated in neuropsychiatric developmental diseases such as autism spectrum disorders (ASD). Recent epidemiological surveys reported a strong link between ASD and atopic/allergic diseases. However, few studies have experimentally investigated the relationship between allergy and ASD-like manifestations, particularly in the early postnatal period, when allergic disorders occur frequently. Therefore, we aimed to characterize how allergic inflammation in the early postnatal period influences microglia and behavior using mouse models of short- and long-term airway allergy. Male mice were immunized by an intraperitoneal injection of aluminum hydroxide and ovalbumin (OVA) or phosphate-buffered saline (control) on postnatal days (P) 3, 7, and 11, followed by intranasal challenge with OVA or phosphate-buffered saline solution twice a week until P30 or P70. In the hippocampus, Iba-1-positive areas, the size of Iba-1-positive microglial cell bodies, and the ramification index of microglia by Sholl analysis were significantly smaller in the OVA group than in the control group on P30 and P70, although Iba-1-positive microglia numbers did not differ significantly between the two groups. In Iba-1-positive cells, postsynaptic density protein 95 (PSD95)-occupied areas and CD68-occupied areas were significantly decreased on P30 and P70, respectively, in the OVA group compared with the control group. Immunoblotting using hippocampal tissues demonstrated that amounts of PSD95, α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor 2, and N-methyl-D-aspartate (NMDA) receptor 2B were significantly increased in the OVA group compared with the control group on P70, and a similar increasing trend for PSD95 was observed on P30. Neurogenesis was not significantly different between the two groups on P30 or P70 by doublecortin immunohistochemistry. The social preference index was significantly lower in the three chamber test and the number of buried marbles was significantly higher in the OVA group than in the control group on P70 but not on P30, whereas locomotion and anxiety were not different between the two groups. Compared with the control group, serum basal corticosterone levels were significantly elevated and hippocampal glucocorticoid receptor (GR) amounts and nuclear GR translocation in microglia, but not in neurons or astrocytes, were significantly decreased in the OVA group on P70 but not on P30. Gene set enrichment analysis of isolated microglia revealed that genes related to immune responses including Toll-like receptor signaling and chemokine signaling pathways, senescence, and glucocorticoid signaling were significantly upregulated in the OVA group compared with the control group on P30 and P70. These findings suggest that early postnatal allergic airway inflammation induces dystrophic microglia that exhibit defective synaptic pruning upon short- and long-term allergen exposure. Furthermore, long-term allergen exposure induced excitatory postsynaptic surplus and ASD-like behavior. Hypothalamo-pituitary-adrenal axis activation and the compensatory downregulation of microglial GR during long-term allergic airway inflammation may also facilitate these changes.