Human and Bovine Milk Oligosaccharides Elicit Improved Recognition Memory Concurrent With Alterations in Regional Brain Volumes and Hippocampal mRNA Expression

Abstract

Human milk contains a unique profile of oligosaccharides (OS) and preliminary evidence suggests they impact brain development. The objective of this study was to assess the impact of bovine and/or human milk oligosaccharides (HMO) (2'-fucosyllactose and Lacto-N-neotetraose) on cognition, brain development, and hippocampal gene expression. Beginning on postnatal day (PND) 2, male pigs received one of four milk replacers containing bovine milk oligosaccharides (BMOS), HMO, both (BMOS + HMO), or neither. Pigs were tested on the novel object recognition task using delays of 1- or 48-h at PND 22. At PND 32-33, magnetic resonance imaging procedures were used to assess structural brain development and hippocampal tissue was collected for analysis of mRNA expression. Pigs consuming only HMO exhibited recognition memory after a 1-h delay and those consuming BMOS + HMO exhibited recognition memory after a 48-h delay. Both absolute and relative volumes of cortical and subcortical brain regions were altered by diet. Hippocampal mRNA expression of GABRB2, SLC1A7, CHRM3, and GLRA4 were most strongly affected by diet. HMO and BMOS had distinct effects on brain structure and cognitive performance. These data suggest different mechanisms underlie their influence on brain development.

Keywords: brain; cognition; human milk oligosaccharides (HMOs); milk; neuroimaging; nutrition; pig; prebiotics.

FIGURE 1

Study timeline. Pigs were reared artificially from postnatal days 2–33. Starting on postnatal day 22, pigs were tested on the novel object recognition test twice using delays of 1 h or 2 days, with task order randomized and counterbalanced between groups. On PND 32–33, pigs were subjected to magnetic resonance imaging and on PND 33 brain tissue was collected for quantification of hippocampal gene expression. Pigs were weighed daily to track growth.

FIGURE 2

Body weight (BW) during the trial. No differences in average daily body weight gain were observed between groups (P = 0.15).

FIGURE 3

Performance during the novel object recognition task assessed using a two-way analysis of variance including BMOS, HMO, and their interaction. (A) An interaction effect was observed for total distance moved during the first habituation trial. (B) During the first habituation trial, pigs fed diets containing BMOS showed a lesser increase in distance moved per minute than those not fed BMOS. (C) During the sample phase pigs fed HMO in any diet spent more time investigating objects than those fed diets without HMO. (D) Pigs fed the HMO-only diet were able to display a recognition index greater than 0.5 after a 1-h delay, however only those fed both BMOS and HMO were able to display a recognition index greater than 0.5 after a 48-h delay, as indicated by the asterisks (measured by a one-sample t-test for a recognition index greater than 0.5).

FIGURE 4

Representative 3D surface rendering of the brain from the pig brain atlas. Only brain regions altered by the diet are shown. Although affected by diet, the cortices and gray matter are not highlighted to allow visualization of subcortical structures.

FIGURE 5

(A–E) Absolute volume of regions affected by diet assessed using a two-way analysis of variance including BMOS, HMO, and their interaction. In each case, (except C) an interaction effect was observed wherein pigs fed the CON or BMOS + HMO diet displayed similar absolute volumes compared to pigs fed either BMOS or HMO. (F–K) Relative volumes of regions affected by diet. (F,G) A main effect of HMO was seen for both the left and right cortices, pigs fed any diet containing HMO demonstrating larger relative cortices. (H) Pigs fed diets containing HMO had a larger relative corpus callosum, whereas pigs fed diets containing BMO had a smaller relative corpus callosum. (I–K) Similar to the effects seen for absolute volumes, interactions effects of diet for relative volumes of the caudate, lateral ventricle, and pons were observed, wherein pigs fed the CON or BMOS + HMO diet displayed similar relative volumes compared to pigs fed either BMOS or HMO.

FIGURE 6

(A) Hippocampal tissue was assessed for the mRNA expression of 93 genes. Figure depicts standardized data (mean = 0, standard deviation = 1) centered by control group. Values below zero indicate decreased expression compared to control, whereas values above zero indicate increased expression. Bars show mean ± standard error, genes significantly impacted by diet are denoted by an asterisk. Accession numbers for each gene can be found in Supplementary Table 6. (B) Genes were sorted in descending order by Z-score for each diet, visualizing the trend where the HMO group exhibited greater downregulation of mRNA compared to the BMOS or BMOS + HMO group. CON, control group; BMOS, pigs fed a mixture of bovine milk oligosaccharides; HMO, Pigs fed 2’-fucosyllactose + Lacto-N-neotetraose; BMOS + HMO, pigs fed both human and bovine milk oligosaccharides.

FIGURE 7

Of the variables affected by diet, the genes (A–D) GABRR1, GRIN1, HDAC5, and SLC1A7 were shown to correlate with the recognition index after a 1-h delay. No relationships were found between MRI or gene expression outcomes affected by diet and the recognition index after a 48-h delay. The relationships observed were not present equally across all diets, however, study-wide all were negatively related to the recognition index.