Untargeted lipidomics and metagenomics reveal the mechanism of aspirin eugenol ester relieving hyperlipidemia in ApoE-/- mice

Abstract

Hyperlipidemia is induced by abnormal lipid metabolism, which can cause the occurrence of cardiovascular diseases and lead to grievous injury to health. Studies showed that AEE had a significant therapeutic effect on hyperlipidemia and is likely to be associated with the up-regulation of cholesterol 7-alpha hydroxylase (CYP7A1), the key enzyme for cholesterol conversion to bile acids, but no research confirmed whether the effect of AEE on hyperlipidemia was related to the gut microbiota and liver lipids. At the same time, more and more studies have shown that gut microbiota and lipids are closely related to hyperlipidemia. Hence, in this study, we investigated the effects of AEE on liver lipids through LC-MS-based untargeted lipidomics and the effects of AEE on gut microbiota based on cecal contents metagenomics by Illumina sequencing in HFD-induced hyperlipidemia ApoE-/- mice at the overall level. The results of lipidomics showed that AEE relieved hyperlipidemia by decreasing the concentration of 10 PEs and 12 SMs in the liver and regulating the pathways of glycerophospholipid metabolic pathway, sphingolipid signaling pathway, and NF-kB signaling pathway. The results of metagenomics concluded that AEE treatment changed the composition of gut microbiota and regulated the functions of lipid transport and metabolism, as well as the metabolism of bile acids and secondary bile acids. The results of the joint analysis between lipidomics and metagenomics showed that the abundance of Verrucomicrobia, Verrucomicrobiales, Candidatus_Gastranaerophilales, and Candidatus_Melainabacteria was significantly positively correlated with the concentration of SM (d18:1/18:0) and PE (16:0/18:1) in the process of AEE alleviating hyperlipidemia in mice. In conclusion, these results suggested that the effect of AEE on hyperlipidemia was closely related to the gut microbiota by the change of bile acids and liver lipids.

Keywords: ApoE-/- mice; aspirin eugenol ester; hyperlipidemia; lipidomics; metagenomics.

FIGURE 1

The number of lipid species in lipid classes. Different colors represent different lipid classes and the numbers represent the number of lipid species.

FIGURE 2

Lipidomics analysis of mice liver samples. (A) PCA score plots of liver lipid profiling on all samples. (B–D) The concentration of different lipid compositions in the liver. Data are expressed as the means ± SD (n = 6). Differences were assessed by ANOVA and denoted as follows: **P < 0.01, *P < 0.05 compared with the ND group; ^##P < 0.01, ^#P < 0.05 compared with the HFD group.

FIGURE 3

OPLS-DA and volcano plot analysis of lipid in the liver. (A) OPLS-DA analysis in HFD vs. ND. (B) OPLS-DA analysis in AEE vs. HFD. (C) OPLS-DA displacement test of HFD vs. ND (Q2 = –0.1). (D) OPLS-DA displacement test of AEE vs. HFD (Q2 = –0.182). (E) Volcano plot of differentially lipids in HFD vs. ND. (F) Volcano plot of differentially lipids in AEE vs. HFD. Red dots represent significantly upregulated lipids and blue dots represent significantly downregulated lipids. Lipids that are not significant are represented by gray dots.

FIGURE 4

AEE altered the important differential lipid species in the liver. (A) Different lipids in HFD vs. ND. (B) Different lipids in AEE vs. HFD. (C–F) The important hepatic differential lipid species in the HFD and AEE groups. Data are expressed as the means ± SD (n = 6). Differences were assessed by ANOVA and denoted as follows: **P < 0.01, *P < 0.05 compared with the ND group; ^##P < 0.01, ^#P < 0.05 compared with the HFD group.

FIGURE 5

Lipid metabolism pathway analysis based on KEGG enrichment map for significantly different lipid species in HFD vs. ND (A) and AEE vs. HFD (B). In the metabolic pathway P-value was the significance of the metabolic pathway, and the significance-enriched pathway was selected for bubble mapping. The ordinate was the name of the metabolic pathway and the abscissa was the Rich factor. The larger the Rich Factor was, the greater the enrichment would be. The color from green to red indicated that P-value decreased in turn, and the lower P-value represents the more significant degree of enrichment. A larger dot indicates a greater number of metabolites were enriched on this pathway.

FIGURE 6

Microbiota compositions at the Phylum level. (A) PCA results of the similarities in bacterial community structures at the Phylum level. (B) Analysis of Anosim based on Phylum level. (C,D) Histogram of relative abundance at the Phylum level. (E–N) Relative content of Candidatus_Melainabacteria, Candidatus_Saccharibacteria, Cyanobacteria, Deferribacteres, Euryarchaeota, Firmicutes, Fusobacteria, Spirochetes, Tenericutes, and Verrucomicrobia at the Phylum level. Values are presented as mean ± SD (n = 6). Differences were assessed by ANOVA and denoted as follows: ^**P < 0.01, *P < 0.05 compared with the ND group; ^◆P < 0.05 compared with the HFD group.

FIGURE 7

Microbiota compositions at the Order level. (A) PCA results of the similarities in bacterial community structures at the Order levels. (B) Analysis of Anosim based on Order level. (C,D) Histogram of relative abundance at the Order level. (E–J) Relative content of Acholeplasmatales, Candidatus_Gastranaerophilales, Coriobacteriales, Deferribacterales, Rhodospirillales, and Verrucomicrobiales at the Order level. Values are presented as mean ± SD (n = 6). Differences were assessed by ANOVA and denoted as follows: ^**P < 0.01, *P < 0.05 compared with the ND group; ^{◆ ◆}P < 0.01, ^◆P < 0.05 compared with the HFD group.

FIGURE 8

Microbiota compositions at the Species level. (A) PCA results of the similarities in bacterial community structures at the Species level. (B) Analysis of Anosim based on Species level. (C,D) Histogram of relative abundance at the Species level. (E–K) Relative content of bacterium_1XD8-92, Bacteroidaceae_bacterium, Bacteroidales_bacterium, Clostridia_bacterium, Muribaculaceae_bacterium, Muribaculaceae_bacterium_Isolate-004_(NCI), and Ruminococcaceae_bacterium at the Species level. Values are presented as mean ± SD (n = 6). Differences were assessed by ANOVA and denoted as follows: ^**P < 0.01, *P < 0.05 compared with the ND group; ^{◆ ◆}P < 0.01 compared with the HFD group.

FIGURE 9

The analysis of linear discriminant analysis coupled with effect size measurements (LEfSe). (A) The linear discriminant analysis (LDA). Different colors indicated different groups, the blue, green, and red bars indicated species with relatively high abundance in the ND, HFD, and AEE groups, respectively (LDA). (B) Differential species annotation branch diagram. Differential species annotation branch illustration. Different colors indicated different subgroups. The blue, green, and red nodes indicated differentially significant species with relatively high abundance in the ND, HFD, and AEE groups, and the yellow nodes indicated species with no significant difference in the comparison between the two groups (HFD vs. ND; AEE vs. HFD). The node diameter was in direct proportion to the relative abundance. The nodes in each layer represented Phylum, Order, and Species, respectively, from inside to outside. The notes for species markers in each layer represented Phylum, Order, and Species from outside. The lettered species names are shown in the legend on the right.

FIGURE 10

The significant difference functions in eggNOG. (A–J) The significant difference functions in eggNOG are DNA primase activity, unsaturated fatty acids biosynthesis, lipid A biosynthesis acyltransferase, two component, sigma 54 specific, transcriptional regulator, fis family, type site-specific deoxyribonuclease, the biosynthesis of lipid A, succinate dehydrogenase, bile acid, lipid transport and transposase. The horizontal axis is the sample grouping; Vertical is the relative abundance of the corresponding function. P < 0.05 indicated significant difference between the two groups, and P < 0.01 indicated extremely significant difference between the two groups.

FIGURE 11

The significant difference functions in KEGG. (A–M) The significant difference functions in KEGG are renin-angiotensin system, inositol phosphate metabolism, phosphotransferase system, secondary bile acid biosynthesis, insulin signaling pathway, Valine, leucine and isoleucine biosynthesis, base excision repair, lysosome, DNA replication, glycosphingolipid biosynthesis, tropane, piperidine and pyridine alkaloid biosynthesis, protein processing in endoplasmic reticulum, and biosynthesis of unsaturated fatty acids. The horizontal axis is the sample grouping; Vertical is the relative abundance of the corresponding function. P < 0.05 indicated significant difference between the two groups, and P < 0.01 indicated extremely significant difference between the two groups.

FIGURE 12

Correlation heat map of metagenomic and lipidomics. (A) Correlation analysis at the Phylum level. (B) Correlation analysis at the Order level. (C) Correlation analysis at the species level. For each species/gene that behaves differently, each column is listed as the corresponding metabolite. The orange color showed a positive correlation, while the blue color showed a negative correlation. The deeper the color was, the greater the correlation would be. The closer the color was to white, the closer the correlation would be to zero. ***P < 0.001; **P < 0.01; *P < 0.05.