APOE genotype influences the gut microbiome structure and function in humans and mice: relevance for Alzheimer's disease pathophysiology

Abstract

Apolipoprotein E (APOE) genotype is the strongest prevalent genetic risk factor for Alzheimer's disease (AD). Numerous studies have provided insights into the pathologic mechanisms. However, a comprehensive understanding of the impact of APOE genotype on microflora speciation and metabolism is completely lacking. In this study, we investigated the association between APOE genotype and the gut microbiome composition in human and APOE-targeted replacement (TR) transgenic mice. Fecal microbiota amplicon sequencing from matched individuals with different APOE genotypes revealed no significant differences in overall microbiota diversity in group-aggregated human APOE genotypes. However, several bacterial taxa showed significantly different relative abundance between APOE genotypes. Notably, we detected an association of Prevotellaceae and Ruminococcaceae and several butyrate-producing genera abundances with APOE genotypes. These findings were confirmed by comparing the gut microbiota of APOE-TR mice. Furthermore, metabolomic analysis of murine fecal water detected significant differences in microbe-associated amino acids and short-chain fatty acids between APOE genotypes. Together, these findings indicate that APOE genotype is associated with specific gut microbiome profiles in both humans and APOE-TR mice. This suggests that the gut microbiome is worth further investigation as a potential target to mitigate the deleterious impact of the APOE4 allele on cognitive decline and the prevention of AD.-Tran, T. T. T., Corsini, S., Kellingray, L., Hegarty, C., Le Gall, G., Narbad, A., Müller, M., Tejera, N., O'Toole, P. W., Minihane, A.-M., Vauzour, D. APOE genotype influences the gut microbiome structure and function in humans and mice: relevance for Alzheimer's disease pathophysiology.

Keywords: SCFAs; apolipoprotein E; butyrate; gut microbiota; metabolomics.

Figure 1

Box plot of the relative abundance distribution of selected taxa at phylum level (A), order level (B), family level (C), and Clostridium cluster or genus level (D). Significant difference was observed in selected bacterial taxon abundance between human APOE genotypes. Significance values were calculated by the Kruskal-Wallis H test for all genotypes, followed by Dunn’s multiple comparisons and adjusted for false discovery rate using the Benjamini-Hochberg correction. *P < 0.05.

Figure 2

Differences in gut microbiome composition between APOE3-TR and APOE4-TR mice. A) Principal coordinates analysis (PCoA) based on unweighted and weighted UniFrac distances of partial sequences of bacterial 16S rRNA genes showing gut microbiota β diversity grouped by age and APOE genotypes. Samples are projected onto the first (PC1) and second (PC2) principal coordinate axes. β diversity analysis reveals significant gut microbiota differences between APOE3 and APOE4 genotype transgenic mice. Significant differences between groups were calculated by PERMANOVA tests. B) Comparison of relative abundance taxa between APOE3 and APOE4 in young mice samples, old mice samples, and both age groups combined were represented by log₂ fold changes. Significant differences in relative abundance of gut microbiota at the family and genus levels between APOE3-TR and APOE4-TR mice were detected in young mice, old mice, and in combined analysis of young and old mice. Statistical significances were determined by the Mann-Whitney U test and were corrected for the multiple comparison using the Benjamini-Hochberg adjustment. E3Y, APOE3 young mice; E4Y, APOE4 young mice; E3O, APOE3 old mice; E4O, APOE4 old mice. *P < 0.05.

Figure 3

Fecal metabolome analysis of APOE3-TR and APOE4-TR mice. A) Heatmap and cluster analysis of 2-way ANOVA of significantly differentially abundant metabolites grouped by age and APOE genotype. Four clusters within significantly differentially abundant metabolites showed distinct APOE genotype and age correlations. Clustering was obtained following similarity analysis using the Ward hierarchical algorithm and Euclidean distance metrics. B) COIA of the association between metabolites and microbiota composition in the gut. The left panel shows the COIA of the microbiota principal component analysis (solid circle) at OTU level and the principal component analysis of metabolomics (empty circle); length of arrow indicates the divergence between 2 data sets. The right panel shows coinertia of metabolome and microbiota data, represented by arrow length between the 2 data points per sample, grouped according to APOE genotype or age. A high overall similarity was found in the structure between the 2 data sets, which was statistically significant, and a higher concordance was found between microbiota composition and metabolites of APOE4 mice compared with APOE3 mice. Length of arrow was estimated using Euclidean distance measurement. E3Y, APOE3 young mice; E4Y, APOE4 young mice; E3O, APOE3 old mice; E4O, APOE4 old mice. *P < 0.05.