Maternal omega-3 fatty acids regulate offspring obesity through persistent modulation of gut microbiota

Abstract

Background: The early-life gut microbiota plays a critical role in host metabolism in later life. However, little is known about how the fatty acid profile of the maternal diet during gestation and lactation influences the development of the offspring gut microbiota and subsequent metabolic health outcomes.

Results: Here, using a unique transgenic model, we report that maternal endogenous n-3 polyunsaturated fatty acid (PUFA) production during gestation or lactation significantly reduces weight gain and markers of metabolic disruption in male murine offspring fed a high-fat diet. However, maternal fatty acid status appeared to have no significant effect on weight gain in female offspring. The metabolic phenotypes in male offspring appeared to be mediated by comprehensive restructuring of gut microbiota composition. Reduced maternal n-3 PUFA exposure led to significantly depleted Epsilonproteobacteria, Bacteroides, and Akkermansia and higher relative abundance of Clostridia. Interestingly, offspring metabolism and microbiota composition were more profoundly influenced by the maternal fatty acid profile during lactation than in utero. Furthermore, the maternal fatty acid profile appeared to have a long-lasting effect on offspring microbiota composition and function that persisted into adulthood after life-long high-fat diet feeding.

Conclusions: Our data provide novel evidence that weight gain and metabolic dysfunction in adulthood is mediated by maternal fatty acid status through long-lasting restructuring of the gut microbiota. These results have important implications for understanding the interaction between modern Western diets, metabolic health, and the intestinal microbiome.

Keywords: Maternal diet; Microbiome; Microbiota; Obesity; n-3 PUFA.

Fig. 1

Experimental design and tail fatty acid profiles. a fat-1 (n = 15) and WT mothers (n = 9) were mated while on a high-n-6 PUFA diet (10% corn oil). At birth, WT offspring were cross-fostered to different mothers for the period of lactation (4 weeks) to produce four experimental groups based on the mothers’ genotype (biological mother’s genotype/foster mother’s genotype): fat-1/WT (n = 8 males, n = 7 females), WT/fat-1 (n = 9 males, n = 10 females), fat-1/fat-1 (n = 9 males, n = 5 females), and WT/WT (n = 10 males, n = 10 females). During lactation, mothers were continued on a high-n-6 PUFA diet. After 4 weeks of lactation, offspring were weaned onto a high-fat diet (HFD) (60% kcal from fat) for 3 months during which body weights were assessed along with a number of other parameters. b WT mothers displayed a significantly greater tail n-6/n-3 ratio compared with fat-1 mothers. c Following lactation for 4 weeks and prior to HFD, WT/WT male offspring had significantly greater tail n-6/n-3 ratio compared with all other groups. d However, after 3 months on a HFD, differences in tail n-6/n-3 fatty acid ratio were eliminated and there were no significant differences between groups. Data shown as mean ± SEM. n = 5–15 per group, n = 1–4 per cage. Bars with different letters are significantly different

Fig. 2

Weight gain and metabolic markers on a HFD. a, b There were no differences in weight gain between female groups. c, d WT/WT male offspring gained significantly more weight on a HFD than all other groups (repeated-measures two-way ANOVA (time and group) with Tukey’s post-hoc test). e Fat-1/fat-1 displayed the lowest circulating LBP and intestinal permeability (f). Data shown as mean ± SEM. fat-1/WT: n = 8 males, n = 7 females; WT/fat-1: n = 9 males, n = 10 females; fat-1/fat-1: n = 9 males, n = 5 females; WT/WT: n = 10 males, n = 10 females. n = 1–4 animals per cage. Bars with different letters are significantly different. *p < 0.05 and **p < 0.01, WT/WT vs. WT/fat-1; #p < 0.05 and ##p < 0.01 WT/WT vs fat-1/WT; ^&p < 0.05, WT/WT vs. fat-1/fat-1

Fig. 3

Phylum compositional and diversity differences in fecal microbiota between offspring of fat-1 and WT mothers. a Analysis of beta diversity identified significant clustering of offspring groups by foster maternal genotype. b α-diversity as measured by the Shannon index did not differ between post-HFD groups, however, appeared to be reduced in WT/WT compared with WT/fat-1 before HFD feeding. c The Firmicutes:Bacteroidetes ratio was lowest in fat-1/fat-1 offspring. d Fecal microbiota composition differed significantly between offspring groups at phylum level and appeared dependent on foster mother genotype. Significant differences were determined by non-parametric analysis using the Kruskall-Wallis test followed by Mann-Whitney test. n = 5–10 per group, n = 1–4 per cage. Groups with different letters are significantly different

Fig. 4

Genus level distribution of fecal microbiota between offspring of fat-1 and WT mothers. a Least discriminant analysis of effect size (LEfSe) on clustered groups based on maternal genotype revealed that Firmicutes were more abundant in offspring of WT foster mothers whereas Bacteroidetes and Proteobacteria were more abundant in offspring of fat-1 foster mothers. b Normalized data of fecal microbiota relative abundance revealed distinct differences between offspring groups. Microbiota composition appeared similar in groups of the same foster mother genotype. Data represents OTUs with significantly different relevant abundances between treatment groups (as determined by Kruskall-Wallis testing and Benjamani-Hocherg multiple correction testing). Each row represents an OTU labeled by lowest taxonomic description and OTU ID, normalized to the row maximum. Data normalized per taxonomic read. c Clostridia displayed significantly higher relative abundance in offspring of WT foster mothers. Bacteroides displayed higher relative abundance in offspring of fat-1 foster mothers. Akkermansia displayed the highest relative abundance in fat-1/fat-1 offspring. Epsilonproteobacteria were almost entirely depleted (< 0.1%) in offspring fostered to WT mothers. n = 5–10 per group, n = 1–4 per cage. Significant differences were determined by non-parametric analysis using the Kruskall-Wallis test followed by Mann-Whitney test and FDR correction by Benjamani-Hochburg testing. Groups with different letters are significantly different

Fig. 5

Network analyses of microbiome and host metabolic phenotype interactions. Host-microbiota interaction network built from Spearman’s non-parametric rank correlation coefficient (P < 0.05) between host parameters (mother and offspring pre-HFD n-6/n-3 ratio, body weight, IP, and LBP) and microbial parameters (pre- and post-HFD OTUs with FDR-corrected p values < 0.05, FIR/BAC ratio, and Shannon ADI) for a pre-HFD and b post-HFD. Each node was colored according to the modularity score and nodes were grouped as three (a) or four (b) modules. Lines represent statistically significant correlations and are colored red for positive and green for negative correlations. c Partial least square (PLS)-regression loading score plot illustrating the association between host parameters (dependent variables—Y) and microbial parameters (explanatory variables—X; red dots). Explanatory variables of interest with variable importance in the projection (VIP) scores 1 or > 1 were labeled with circles on the red dots. Samples from four different groups (fat-1/WT, WT/fat-1, fat-1/fat-1, WT/WT) were observed (green dots) and grouped using circles based on where they clustered on the plot. Leave-one-out cross-validation (LOO-CV) was applied. d Multiple factor analysis (MFA) using Spearman type principal component analysis of host and microbiota data. One end of the each connecting line for an observation indicates the host parameters (differently colored to indicate the groups) and another end (red) indicates the microbiota