Modeling environmental risk factors of autism in mice induces IBD-related gut microbial dysbiosis and hyperserotonemia

Abstract

Autism spectrum disorder (ASD) is a range of neurodevelopmental conditions that are sharply increasing in prevalence worldwide. Intriguingly, ASD is often accompanied by an array of systemic aberrations including (1) increased serotonin, (2) various modes of gastrointestinal disorders, and (3) inflammatory bowel disease (IBD), albeit the underlying cause for such comorbidities remains uncertain. Also, accumulating number of studies report that the gut microbial composition is significantly altered in children with ASD or patients with IBD. Surprisingly, when we analyzed the gut microbiota of poly I:C and VPA-induced mouse models of ASD, we found a distinct pattern of microbial dysbiosis that highly recapitulated those reported in clinical cases of ASD and IBD. Moreover, we report that such microbial dysbiosis led to notable perturbations in microbial metabolic pathways that are known to negatively affect the host, especially with regards to the pathogenesis of ASD and IBD. Lastly, we found that serum level of serotonin is significantly increased in both poly I:C and VPA mice, and that it correlates with increases of a bacterial genus and a metabolic pathway that are implicated in stimulation of host serotonin production. Our results using animal model identify prenatal environmental risk factors of autism as possible causative agents of IBD-related gut microbial dysbiosis in ASD, and suggest a multifaceted role of gut microbiota in the systemic pathogenesis of ASD and hyperserotonemia.

Fig. 1

Gut microbial composition and diversity indexes in ASD mice. a Gut microbial composition comparison at the phylum level. b PD index and c Shannon index of alpha-diversity in CTL and ASD mice. *p < 0.05 vs. CTL. Beta-analysis of CTL and ASD mice shown by d timing of feces sampling or e drug treatment. n = 9–14 in all experiments

Fig. 2

Relative abundances of gut microbial taxa in ASD mice. Relative abundances of microbial taxa in CTL and ASD mice. The data are divided with respect to the pattern of dysbiosis reported in previous clinical studies on either children with ASD or patients with IBD. a Taxa reported to have been significantly a increased in children with ASD or b patients with IBD. c Taxa that are decreased in patients with IBD. d Taxa that are increased (Desulfovibrio) or decreased (Prevotella, F. prausnitzii, Oscillospira) in both ASD and IBD. Asterisks next to taxa names denote the statistical significance of the difference between CTL and each ASD experimental group. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 vs. CTL. n = 9-14 in all experiments

Fig. 3

Prevotellaceae family is significantly and independently altered in ASD mice. a Relative abundance of Prevotellaceae in ASD mice compared to CTL mice. P values were calculated against CTL group. b Unweighted co-occurrence network analysis of gut microbial families in ASD mice. c Co-occurrence network weighted with linkage strength. Red and blue lines denote positive and negative correlation, respectively

Fig. 4

Possible metabolic perturbations caused by microbial dysbiosis in ASD mice. a KEGG pathway analysis based on 16S data drawn out in heatmap. All pathways except “Protein digestion and absorption” have significant (p < 0.05) differences between CTL and each ASD experimental group. In “Protein digestion and absorption”, the p-values of CTL vs. VPA and CTL vs. Poly I:C are 0.045 and 0.055, respectively. Pathways in the red box have been reported to be perturbed in children with ASD, and those in the blue box have been implicated in patients with IBD. Purple box denotes that ASD and IBD showed the same manner of alteration in those pathways. b Relative abundance analysis of Sporanaerobacter in CTL and ASD mice. *p < 0.05 vs. CTL. c Serotonin level in serum detected by ELISA in CTL and ASD mice. *p < 0.05 vs. CTL. n = 9-14 in all experiments