Feed your microbes to deal with stress: a psychobiotic diet impacts microbial stability and perceived stress in a healthy adult population

Abstract

The impact of diet on the microbiota composition and the role of diet in supporting optimal mental health have received much attention in the last decade. However, whether whole dietary approaches can exert psychobiotic effects is largely understudied. Thus, we investigated the influence of a psychobiotic diet (high in prebiotic and fermented foods) on the microbial profile and function as well as on mental health outcomes in a healthy human population. Forty-five adults were randomized into either a psychobiotic (n = 24) or control (n = 21) diet for 4 weeks. Fecal microbiota composition and function was characterized using shotgun sequencing. Stress, overall health and diet were assessed using validated questionnaires. Metabolic profiling of plasma, urine and fecal samples was performed. Intervention with a psychobiotic diet resulted in reductions of perceived stress (32% in diet vs. 17% in control group), but not between groups. Similarly, biological marker of stress were not affected. Additionally, higher adherence to the diet resulted in stronger decreases in perceived stress. While the dietary intervention elicited only subtle changes in microbial composition and function, significant changes in the level of 40 specific fecal lipids and urinary tryptophan metabolites were observed. Lastly, microbial volatility was linked to greater changes in perceived stress scores in those on the psychobiotic diet. These results highlight that dietary approaches can be used to reduce perceived stress in a human cohort. Using microbiota-targeted diets to positively modulate gut-brain communication holds possibilities for the reduction of stress and stress-associated disorders, but additional research is warranted to investigate underlying mechanisms, including the role of the microbiota.

Fig. 1. Perceived stress was reduced by dietary intervention and correlated with greater adherence.

A The diet intervention decreased perceived stress in healthy adult volunteers. Changes in PSS scores in each intervention group are shown. A lower score corresponds to less stress. Data expressed as mean ± SEM; **p ≤ 0.01. B Greater adherence to DIET resulted in greater reductions in stress. Adherence to dietary intervention correlated with changes in PSS score. PSS score changes were calculated as absolute number changes between pre- and post-intervention. More negative numbers correspond to greater changes.

Fig. 2. Microbiota composition and function that were affected by dietary intervention.

A Relative abundance of microbial taxa differentially affected by dietary treatment. V2 refers to pre-intervention abundance, V3 to post-intervention; n (CONT V2) = 21; n (CONT V3) = 21; n (DIET V2) = 24; n (DIET V3) = 24). B Changes in microbial functions in DIET and CONT group throughout dietary intervention. V2 refers to pre-intervention levels, V3 to post-intervention; n (CONT V2) = 21; n (CONT V3) = 21; n (DIET V2) = 24; n (DIET V3) = 24).

Fig. 3. Volatility is associated with greater changes in PSS scores in dietary intervention group.

Lower volatility (more stable microbiota) correlated with greater changes in PSS score. PSS score changes were calculated as absolute number changes between pre- and post-intervention. More negative numbers correspond to greater changes.

Fig. 4. Alterations in fecal, urine and plasma metabolites.

A Volcano plot showing fecal lipid metabolites statistically significantly altered by dietary intervention (A) and in control group (B). B Changes in urinary metabolites after dietary intervention. Only metabolites that changed statistically significantly are shown. V2 refers to pre-intervention levels, V3 to post-intervention; n (CONT V2) = 21; n (CONT V3) = 21; n (DIET V2) = 24; n (DIET V3) = 24). C Changes in plasma metabolites after dietary intervention. 5-hydroxyindole-3-acetic acid and 3-hydroxykynurenine statistically significantly decreased in the diet group; V2 refers to pre-intervention levels, V3 to post-intervention; n (CONT V2) = 21; n (CONT V3) = 21; n (DIET V2) = 24; n (DIET V3) = 24).