The Role of the Gut Microbiota in the Effects of Early-Life Stress and Dietary Fatty Acids on Later-Life Central and Metabolic Outcomes in Mice

Abstract

Early-life stress (ELS) leads to increased vulnerability for mental and metabolic disorders. We have previously shown that a low dietary ω-6/ω-3 polyunsaturated fatty acid (PUFA) ratio protects against ELS-induced cognitive impairments. Due to the importance of the gut microbiota as a determinant of long-term health, we here study the impact of ELS and dietary PUFAs on the gut microbiota and how this relates to the previously described cognitive, metabolic, and fatty acid profiles. Male mice were exposed to ELS via the limited bedding and nesting paradigm (postnatal day (P)2 to P9 and to an early diet (P2 to P42) with an either high (15) or low (1) ω-6 linoleic acid to ω-3 alpha-linolenic acid ratio. 16S rRNA was sequenced and analyzed from fecal samples at P21, P42, and P180. Age impacted α- and β-diversity. ELS and diet together predicted variance in microbiota composition and affected the relative abundance of bacterial groups at several taxonomic levels in the short and long term. For example, age increased the abundance of the phyla Bacteroidetes, while it decreased Actinobacteria and Verrucomicrobia; ELS reduced the genera RC9 gut group and Rikenella, and the low ω-6/ω-3 diet reduced the abundance of the Firmicutes Erysipelotrichia. At P42, species abundance correlated with body fat mass and circulating leptin (e.g., Bacteroidetes and Proteobacteria taxa) and fatty acid profiles (e.g., Firmicutes taxa). This study gives novel insights into the impact of age, ELS, and dietary PUFAs on microbiota composition, providing potential targets for noninvasive (nutritional) modulation of ELS-induced deficits. IMPORTANCE Early-life stress (ELS) leads to increased vulnerability to develop mental and metabolic disorders; however, the biological mechanisms leading to such programming are not fully clear. Increased attention has been given to the importance of the gut microbiota as a determinant of long-term health and as a potential target for noninvasive nutritional strategies to protect against the negative impact of ELS. Here, we give novel insights into the complex interaction between ELS, early dietary ω-3 availability, and the gut microbiota across ages and provide new potential targets for (nutritional) modulation of the long-term effects of the early-life environment via the microbiota.

Keywords: diet; early-life stress; gut-brain axis; interventions; microbiome; microbiota; polyunsaturated fatty acids.

FIG 1

Age impacts α- and β-diversity and ELS and dietary ω-6/ω-3 ratio affect β-diversity, dependent on each other. (A) Experimental timeline. (B to D) Chao1 (A), Shannon (B), and phylogenetic diversity (C) plots displaying an increase in α-diversity with age (GLMM at sequencing depth of 11,535; P < 0.0001;). (E) β-diversity at the genus level analyzed by PERMANOVA showing effect of age (P < 0.0001) with clustering of the four experimental groups at P42 (condition-diet interaction P = 0.0064) and P180 (condition-diet interaction P = 0.0066). (F and G) db-RDA of β-diversity aggregated at the genus level for both ages separately. The 10 genera explaining most variation in the principal-component analysis (PCA) and db-RDA were visualized; (F) db-RDA at P42, ANOVA-like permutation test for RDA (P = 0.018); (G) db-RDA at P180, ANOVA-like permutation test for RDA (P = 0.003). (H) Cladogram showing significant age-mediated changes in relative abundance of bacterial species at several taxonomic levels. GLMM, general linear mixed model; db-RDA, distance-based redundancy analysis; ANOVA, (analysis of variance).

FIG 2

Early-life stress and early dietary ω-6/ω-3 ratio affect the microbiota composition at P42 in interaction with each other. (A) Cladogram showing significant condition- and diet-mediated changes in the relative abundance of bacterial taxa at several taxonomic levels at P42. (B to J) Bar graphs of detected interaction effects (condition-diet) for bacterial taxa at P42 (GLMM P < 0.05 and q < 0.1). @, main effect of diet; &, interaction of condition-diet; ^, significant difference with Tukey post hoc test (P < 0.05); GLMM, general linear mixed model.

FIG 3

Early-life stress and early dietary ω-6/ω-3 ratio affect the microbiota composition at P180 in interaction with each other. (A) Cladogram showing significant condition- and diet-mediated changes in the relative abundance of bacterial taxa at several taxonomic levels at P180. (B to J) Bar graphs of detected interaction effects (condition-diet) for bacterial taxa at P180 (GLMM P < 0.05 and q < 0.1). #, main effect of condition; @, main effect of diet; &, interaction of condition-diet; ^, significant difference with Tukey post hoc test (P < 0.05); GLMM, general linear mixed model.

FIG 4

Bacterial taxa are correlated with several peripheral and central outcome parameters within the same mice. (A) Correlation heatmap between bacterial taxa at P42 and metabolic outcome parameters at P42; (B) correlation heatmap between bacterial taxa at P42 and P180 and fatty acid levels in the hippocampus at P42 and P180, respectively (–1 < Spearman’s rho < 1). (C to J) Correlation plots between relative abundance of selected bacterial taxa and behavioral, metabolic, and/or fatty acid outcomes. (C) P42 Odoribacter and adult OLT performance. (D) P42 Porphyromonadaceae and OLT performance. (E) Odoribacter and body weight. (F) P42 Peptostreptococcaceae and P42 ω-3 FA hippocampus. (G) P42 Peptostreptococcaceae and P42 liver ω-3 FA. (H) P42 Peptostreptococcaceae and P42 erythrocyte ω-3 FA. (I) Bifidobacterium and hippocampal LCPUFA levels. (J) Bifidobacterium and liver palmitic acid (C_16:0).