Quantitative analysis of targeted lipidomics in the hippocampus of APP/PS1 mice employing the UHPLC-MS/MS method

Abstract

Background: Alzheimer's disease (AD) is marked by the pathological features of amyloid-β plaque accumulation, as well as intracellular neurofibrillary tangles formation in the patients' brain. Aberrant lipid metabolism is increasingly recognized as one of the important contributors to AD.

Purpose: The main goal of this research was to conduct quantitative detection of targeted lipidomics in hippocampal tissue of APPSwe/PS1dE9 mice in order to identify lipid metabolic biomarkers of early-onset AD mice.

Methods: Our approach departs from conventional lipid detection methods, employing a highly accurate quantificational Ultra High Performance Liquid Chromatography Tandem Mass Spectrometry (UHPLC-MS/MS) technique to analyze targeted lipid biomarkers. The innovative method was utilized to detect targeted lipids in the hippocampus of AD and wild-type mice. Statistical method was performed by Student's t-test and multivariate analysis. Differential metabolites were identified through fulfilling the standard of Variable Importance in Projection surpassing one and the significance probability lower than 0.05 thresholds.

Results: Both groups utilized identical methodologies and adhered strictly to standardized treatment protocols. Sphingolipids (SPs), Glycerophospholipids (GPs), Glycolipids, Glycerides (GLs), Sterol Lipids (STs), and Free Fatty Acid (FA) were identified as prominent lipids exhibiting alterations in the hippocampus of AD models. Regarding glycolipid and glycerolipid composition, monogalactosyldiacylglycerols (MGDGs) and Triacylglycerols (TGs) constituted a significant proportion (p < 0.05, VIP > 1). Among glycerophospholipids, phosphatidylethanolamines (PEs) and phosphatidylcholines (PCs) emerged as significant constituents (p < 0.05, VIP > 1). Furthermore, hexosylceramides (HexCers) and ceramides (Cers) in the AD model's hippocampus, the prominent sphingolipids in the hippocampus of AD mice, existed as the two primary changed lipid metabolites. The levels of some TGs in GLs and CEs in STs showed a significant elevation (p < 0.05, VIP > 1). In contrast, most kinds of MGDGs, HexCers, Cers, PEs and FA (18:2) demonstrated a notable decrease in the hippocampus of AD group (p < 0.05, VIP > 1).

Conclusion: The present research represents the important quantitative identification of distinct lipid biomarker profiles within the hippocampal portion of 7.5-month-aged AD mice. It encompasses glycolipid, GLs, GPs, SPs, STs, and FAs using a targeted HPLC-MS method for quantification. These findings suggest potential diagnostic lipid biomarkers in hippocampus of early-onset AD mice related to cellular membrane integrity, atherosclerosis, oxidative stress damage, and inflammation.

Keywords: APP/PS1 mice; Alzheimer’s disease; UHPLC-MS; hippocampus; targeted lipidomics.

Figure 1

(A) TIC diagram of all QC samples. (B) The TIC diagram of sample “QC-01” and blank sample.

Figure 2

(A) OPLS-DA score plot for group APP/PS1 versus WT control. The OPLS-DA score plot demonstrated that the samples were dispersed into two groups (n = 12). The data for the WT control group are shown in purple, and the data for the APP/PS1 group are shown in green. (B) OPLS-DA permutation test for APP/PS1 versus WT control group. The green dot stands for the R²Y value acquired from the permutation test, the blue square dot indicates the Q² value obtained from the permutation test, and the two dashed lines, respectively, represent the regression lines of R²Y and Q².

Figure 3

(A) Pie chart illustrating the distribution of lipid compositions by the superclass. (B) Donut chart for group APP/PS1 vs. WT control. (C) Volcano plot of lipid metabolites comparing group APP/PS1 with WT control. Red signifies lipids that are significantly up-regulated; blue indicates lipids that are significantly down-regulated; gray denotes lipids that showed no significant changes.

Figure 4

Boxplot analysis of (A) CE (22:6) and (B) CE (22:4) in the hippocampal tissue for group APP/PS1 vs WT control.

Figure 5

Boxplot analysis comparing the levels of various TAGs in hippocampal tissue between the APP/PS1 and WT control mice. (A) TG(56:4)_FA(20:4) (p = 0.011); (B) TG(54:4)_FA(20:4) (p = 0.012); (C) TG(54:6)_FA(22:6) (p = 0.022); (D) TG(56:7)_FA(22:6) (p = 0.028); (E) TG(44:0)_FA(18:0) (p = 0.037); (F) TG(58:8)_FA(22:6) (p = 0.040).

Figure 6

The boxplot analysis demonstrated a remarkable reduction in the levels of MGDGs in the hippocampus of APP/PS1 group in contrast to WT control, suggesting significant changes. (A) MGDG (16:1_18:0); (B) MGDG (16:0_16:1); (C) MGDG (14:0_16:0).

Figure 7

Boxplot analysis of CER and HCER existing the most difference in the hippocampus tissue for group APP/PS1 vs WT control. (A) Cer(d18:1/18:1); (B) Cer(d18:1/16:1); (C) Cer(d18:1/24:1); (D) HexCer(d18:0−/20:0); (E) HexCer(d18:0/18:0); (F) HexCer(d18:1/16:0).

Figure 8

Boxplot analysis of PCs in the hippocampus tissue for group APP/PS1 vs WT control. (A) PC (16:0_16:1 + AcO); (B) PC (18:1_16:1 + AcO).

Figure 9

Boxplot analysis of PEs in the hippocampus tissue for group APP/PS1 vs WT control. (A) PE(O-16:0_22:4); (B) PE(P-18:0_20:3); (C) PE(P-18:1_20:3); (D) PE(P-16:0_20:4).

Figure 10

Heatmap depicting the correlation analysis of potential lipid biomarkers in the hippocampus of 7.5-month-old APP/PS1 mice. The two-color heat map visually represents the association between lipids. x-axis and y-axis depict the variations in metabolites among the groups, whereas the color blocks at various locations signify the correlation coefficients between the respective metabolites. Red color denotes a positive association, while blue color signifies a negative association, and darker colors indicate stronger correlations. Significant correlations are marked with an asterisk (*).