Parabacteroides distasonis induces depressive-like behavior in a mouse model of Crohn's disease

Abstract

Patients with inflammatory bowel disease (IBD) are particularly susceptible to behavioral diagnoses, and the microbiome has been repeatedly implicated in the pathogenesis of IBD. The intestinal microbiome's ability to affect behavior has become increasingly recognized and studied. The so-called 'psychobiome' has been linked to a plethora of neurological and psychological diagnoses, including autism and Parkinson's disease. Despite the ability of many bacterial species within the human intestinal microbiome to synthesize neurotransmitters, it has never been previously reported that a single bacterial species is sufficient to induce depression. Here, we demonstrate that our mouse model of Crohn's disease (CD)-like ileitis, the SAMP1/YitFc (SAMP1), does not exhibit baseline behavioral abnormalities. By comparison, SAMP6 mice develop depressive-like behavior that is associated with a rise in the GABA-producing bacterial genus Parabacteroides. We finally demonstrate that administration of Parabacteroides distasonis into our SAMP1 mice induces depressive-like behavior. Colonization with P. distasonis was not associated with increased intestinal inflammation or alterations in other measures of behavior. The intestinal environment of CD may be particularly conducive to colonization with P. distasonis and subsequent induction of depressive-like behavior. To our knowledge, this is the first report of a bacterial species specifically inducing depressive-like behavior.

Keywords: Crohn’s disease; Inflammatory bowel disease; Major depressive disorder; Microbiome; Microbiome-gut-brain axis; Parabacteroides distasonis; Psychobiome.

Figure 1.. SAMP6, but not SAMP1, develop depressive-like behavior at later time points without anxiety-like behavior, motor, or spatial memory deficits.

Anxiety-like behavior was measured using (a) open field and (b) elevated plus maze. (c) Short-term spatial memory was measured by novel arm exploration in Y-Maze. (d) Gross locomotor activity was measured using computer-defined line crossings during the open field test. (e) Motor coordination was assessed by measuring fall latency during rota-rod. (f) Depressive-like behavior was assessed by measuring immobility time during tail suspension. Each time point is representative of individual mice and behavioral assays were performed in a consistent order. Mice did not perform behavioral assays more than once. Data compared with two-way ANOVA (strain and age as independent variables). Symbols represent individual mice (n=30) from three technical replicates, bars indicate mean ± S.E.M. (n.s. p > 0.05, ****p < 0.0001).

Figure 2.. Depressive-like behavior in SAMP6 mice is associated with expansion of Parabacteroides.

16s rRNA analysis of the fecal microbiome revealed some changes in the richness and evenness of the samples as measured by the (a) Shannon and (b) Simpson diversity indices. (c) Principal component analysis (PCA) of the fecal microbiomes over time. Principal component 1 and 2 shown here accounting for ~39% of explained variance. (d) 16s rRNA measured relative abundance of Parabacteroides. Pairwise Wilcoxon Rank Sum tests performed to analyze data; bars indicate mean ± S.E.M. (*p < 0.05, **p < 0.01, and ****p < 0.0001).

Figure 3.. Inoculation with Parabacteroides distasonis induces depressive-like behavior in SAMP1 mice.

(a) Administration of P. distasonis did increase time immobile during tail suspension (b) but did not worsen ileitis. Alluvial plot of 16s rRNA data for (c) sham and (d) P. distasonis gavaged groups reveals largely similar microbiomes. (e) Relative abundance of Parabacteroides in sham versus inoculated group. Tail suspension was analyzed using student’s t-test, histology was analyzed with Mann-Whitney test, and relative abundance was measured using Wilcoxon Rank Sum test. Symbols represent individual mice (n=10) from two technical replicates, bars indicate mean ± S.E.M. (n.s. p > 0.05, *p < 0.05, **p < 0.01, ***p< 0.001 and ****p < 0.0001).