Microbiota alteration is associated with the development of stress-induced despair behavior

Abstract

Depressive disorders often run in families, which, in addition to the genetic component, may point to the microbiome as a causative agent. Here, we employed a combination of behavioral, molecular and computational techniques to test the role of the microbiota in mediating despair behavior. In chronically stressed mice displaying despair behavior, we found that the microbiota composition and the metabolic signature dramatically change. Specifically, we observed reduced Lactobacillus and increased circulating kynurenine levels as the most prominent changes in stressed mice. Restoring intestinal Lactobacillus levels was sufficient to improve the metabolic alterations and behavioral abnormalities. Mechanistically, we identified that Lactobacillus-derived reactive oxygen species may suppress host kynurenine metabolism, by inhibiting the expression of the metabolizing enzyme, IDO1, in the intestine. Moreover, maintaining elevated kynurenine levels during Lactobacillus supplementation diminished the treatment benefits. Collectively, our data provide a mechanistic scenario for how a microbiota player (Lactobacillus) may contribute to regulating metabolism and resilience during stress.

Conflict of interest statement

The authors declare no competing financial interests.

Figures

Figure 1. Unpredictable chronic mild stress (UCMS) induces despair behavior and microbiota dysregulation.

(a) Experimental design. (b) Quantification of escape behavior in the forced swim test (n = 11 naïve and 12 stressed; representative of 3 independent experiments; two-tailed t-test with Welch’s correction, **p < 0.01; mean ± s.e.m.). (c) Total bacterial load quantification by qRT-PCR of 16S rRNA (n = 17 samples per group; two-tailed t-test; mean ± s.e.m.). (d) Principal coordinate analysis of microbiome communities in naïve and stressed mice. Analysis based on 2 UCMS experiments (n = 12 naïve and 16 stressed; representative of 2 sequencing experiments) (e) Representative graphs of bacterial class distribution in individual subjects show a decrease in bacilli (yellow) (n = 5 naïve and 6 stressed; representative of 2 sequencing experiments).

Figure 2. Lactobacillus levels correlate with depressive behavior.

(a) Lactobacillus quantification in fecal samples by qRT-PCR, relative to 16S rRNA (n = 10 naïve and 11 stressed; representative of 3 independent experiments; two-tailed t-test with Welch’s correction, ***p < 0.001; mean ± s.e.m.). (b) Correlation analysis between Lactobacillus levels and escape behavior (n = 22 pairs, two-tailed Spearman r, **p = 0.01, line of best fit with 95% CI). (c) Correlation analysis between Lactobacillus levels and escape behavior in C57BL/6J (Jax), BALB/cJ (Jax), and C57BL/6N (Taconic), naive and stressed (n = 45 pairs, two-tailed Spearman r, **p = 0.01, line of best fit with 95% CI). (c’) Dashed insert with expanded X-axis for better resolution.

Figure 3. Treatment with probiotic L. reuteri ameliorates the escape behavior induced by chronic stress.

(a) Experimental design of L. reuteri supplementation regimen. (b) qRT-PCR quantification of Lactobacillus levels in fecal samples of L. reuteri or broth-control treated mice, relative to 16S rRNA (n = 5; representative of 2 independent experiments; two-tailed t-test, *p < 0.05; mean ± s.e.m.). (c) Forced swim test quantification of escape behavior of naïve and stressed mice treated with either L. reuteri or bacteria-free broth (n = 9 per group; representative of 2 independent experiments; 2-way ANOVA followed by Bonferroni post-hoc, **p < 0.01; mean ± s.e.m.). (d) Nestlet shredding test quantification of escape behavior of naïve and stressed mice treated with either L. reuteri or bacteria-free broth (n = 9 per group; representative of 2 independent experiments; 2-way ANOVA followed by Bonferroni post-hoc, **p < 0.01; mean ± s.e.m.). (e) Principal component analyses of serum metabolite composition after untargeted metabolomics assay (n = 5 mice per group) showing two (i) or three (ii) group comparisons; shaded areas represent 95% CI. (f) Normalized MS peaks of tryptophan - kynurenine pathway metabolites in the sera of naïve, stressed and L. reuteri treated stressed mice (n = 3–5 per group; 2-way ANOVA, ***p < 0.001; mean ± s.e.m.).

Figure 4. Lactobacillus supplementation improves behavior by moderating kynurenine metabolism.

(a) Representation of tryptophan-kynurenine pathway, depicting H2O2 inhibition of pathway-initiating enzyme IDO1. (b) Production of H2O2 by E. coli and L. reuteri (n = 3 wells per group; representative of 2 independent experiments; two-tailed t-test, ***p < 0.001; mean ± s.e.m.). (c) Production of H2O2 by endogenous Lactobacillus species compared to L. reuteri (n = 7–8 colonies per group; one-way ANOVA with Dunnet’s multiple comparisons, ***p < 0.001; mean ± s.e.m.). (d) Fecal H2O2 levels in naïve, stressed and L. reuteri treated stressed mice (n = 10 naïve, 11 stressed and 7 stressed + L.reuteri mice per group; one-way ANOVA with Dunnet’s multiple comparisons, *p < 0.05, ***p < 0.001; mean ± s.e.m.). (e) Correlation between Lactobacillus and H2O2 levels in naïve and stressed mice (n = 5 mice per group; Spearman r test, **p = 0.01). (f) qRT-PCR quantification of ido1 expression in the intestines of naïve, stressed mice and L. reuteri treated stressed mice, relative to GAPDH. (n = 4 naïve, 7 stressed and stressed + L.reuteri mice per group, 1-way ANOVA followed by Dunnett’s post-hoc, p < 0.01; mean ± s.e.m.). (g) Experimental design of L. reuteri and/or kynurenine administration. (h) Forced swim test quantification of escape behavior of naïve mice treated with L-kynurenine or saline control (n = 10 saline and 8 L-kynurenine mice per group; representative of 2 independent experiments; two-tailed t test with Welch’s correction, *p < 0.05; mean ± s.e.m.). (i) Forced swim test quantification of escape behavior of stressed mice treated with either L. reuteri alone or L. reuteri and L-kynurenine (n = 7 per group; one-way ANOVA followed by Dunnett’s post-hoc, **p < 0.01; mean ± s.e.m.).