A Diet Enriched with Lacticaseibacillus rhamnosus HN001 and Milk Fat Globule Membrane Alters the Gut Microbiota and Decreases Amygdala GABA a Receptor Expression in Stress-Sensitive Rats

Abstract

Brain signalling pathways involved in subclinical anxiety and depressed mood can be modulated via the gut brain axis (GBA), providing the potential for diet and dietary components to affect mood. We investigated behavioural, physiological and gut microbiome responses to the Lacticaseibacillus rhamnosus strain HN001 (LactoB HN001™), which has been shown to reduce postpartum anxiety and depression, and a milk fat globule membrane-enriched product, Lipid 70 (Surestart^TM MFGM Lipid 70), which has been implicated in memory in stress-susceptible Wistar Kyoto rats. We examined behaviour in the open field, elevated plus maze and novel object recognition tests in conjunction with the expression of host genes in neuro-signalling pathways, and we also assessed brain lipidomics. Treatment-induced alterations in the caecal microbiome and short-chain fatty acid (SCFA) profiles were also assessed. Neither ingredient induced behavioural changes or altered the brain lipidome (separately or when combined). However, with regard to brain gene expression, the L. rhamnosus HN001 + Lipid 70 combination produced a synergistic effect, reducing GABAA subunit expression in the amygdala (Gabre, Gat3, Gabrg1) and hippocampus (Gabrd). Treatment with L. rhamnosus HN001 alone altered expression of the metabotropic glutamate receptor (Grm4) in the amygdala but produced only minor changes in gut microbiota composition. In contrast, Lipid 70 alone did not alter brain gene expression but produced a significant shift in the gut microbiota profile. Under the conditions used, there was no observed effect on rat behaviour for the ingredient combination. However, the enhancement of brain gene expression by L. rhamnosus HN001 + Lipid 70 implicates synergistic actions on region-specific neural pathways associated with fear, anxiety, depression and memory. A significant shift in the gut microbiota profile also occurred that was mainly attributable to Lipid 70.

Keywords: anxiety; behaviour; depression; gut–brain axis; lipid; neurotransmitter; stress.

Figure 1

Volcano plots showing differential gene expression (DE) in brain-tissue regions of rats who received various treatments relative to control rats: (a) Amygdala: increased DE (red) of the Grm4 gene in the Lacticaseibacillus rhamnosus strain HN001 (LactoB HN001™) treatment group; (b) Amygdala: decreased DE (blue) of Gabre, Gat3 and Gabrg1 genes in the L. rhamnosus HN001 + Lipid 70 (Surestart^TM MFGM Lipid 70) group; (c). Hippocampus: decreased DE (blue) of the Gabrd gene in the L. rhamnosus HN001 + Lipid 70 group. Thresholds were based on an FDR correction of (p < 0.05) and a log₂fold change > 0.263 (FC < 1.2).

Figure 2

Caecal microbiome. Composition profiles at the phylum level for (a) individual rats and (b) group means. (c) Boxplot of caecal relative abundance of unclassified Enterobacteriaceae in rats provided with plain drinking water (non-HN001) or water supplemented with the Lacticaseibacillus rhamnosus strain HN001 (LactoB HN001™). Points show individual rats coloured by diet group (non-Lipid 70, yellow; Lipid 70 (Surestart^TM MFGM Lipid 70), blue).

Figure 3

Short-chain fatty acids (SCFA) in caecal content. Differences in SCFA among treatment groups are shown as a bar chart (a) comparing overall profiles, and (b–f) as box plots showing median SCFA levels (μmole acid per g caecal content) with 95% confidence intervals (p-values from ANOVA are shown). HN—Lacticaseibacillus rhamnosus strain HN001 (LactoB HN001™); MF—milk fat globule membrane (Surestart^TM MFGM Lipid 70).

Figure 4

Sparse partial least-squares (sPLS) canonical analysis of open-field behavioural test measurements and gene expression in the amygdala. Network showing variables with canonical correlations > |0.5|, with genes indicated by ellipses and behavioural measurements by rectangles. Edge colours indicate the direction of correlations between variables, with positive values shown in red and negative values in blue.