A green and efficient pseudotargeted lipidomics method for the study of depression based on ultra-high performance supercritical fluid chromatography-tandem mass spectrometry

Identification of multiple plasma lipids as diagnostic biomarkers of hypercholesterolemia and the underlying mechanisms based on pseudo-targeted lipidomics

RATIONALE

Hypercholesterolemia is an important risk factor for cardiovascular diseases and death. This study performed pseudo-targeted lipidomics to identify differentially expressed plasma lipids in hypercholesterolemia, to provide a scientific basis for the diagnosis and pathogenesis of hypercholesterolemia.

METHODS

Pseudo-targeted lipidomic analyses of plasma lipids from 20 patients with hypercholesterolemia and 20 normal control subjects were performed using liquid chromatography-mass spectrometry. Differentially expressed lipids were identified by principal component analysis and orthogonal partial least squares discriminant analysis. Receiver operating characteristic curves were used to identify differentially expressed lipids with high diagnostic value. The Kyoto Encyclopedia of Genes and Genomes pathway database was used to identify enriched metabolic pathways.

RESULTS

We identified 13 differentially expressed lipids in hypercholesterolemia using variable importance of projection > 1 and p < 0.05 as threshold parameters. The levels of eight sphingomyelins and cholesterol sulfate were higher and those of three triacylglycerols and lysophosphatidylcholine were reduced in hypercholesterolemia. Seven differentially expressed plasma lipids showed high diagnostic value for hypercholesterolemia. Functional enrichment analyses showed that pathways related to necroptosis, sphingolipid signaling, sphingolipid metabolism, and steroid hormone biosynthesis were enriched.

CONCLUSIONS

This pseudo-targeted lipidomics study demonstrated that multiple sphingomyelins and cholesterol sulfate were differentially expressed in the plasma of patients with hypercholesterolemia. We also identified seven plasma lipids, including six sphingomyelins and cholesterol sulfate, with high diagnostic value.

Recent Progress in Mass Spectrometry-Based Metabolomics in Major Depressive Disorder Research

Major depressive disorder (MDD) is a serious mental illness with a heavy social burden, but its underlying molecular mechanisms remain unclear. Mass spectrometry (MS)-based metabolomics is providing new insights into the heterogeneous pathophysiology, diagnosis, treatment, and prognosis of MDD by revealing multi-parametric biomarker signatures at the metabolite level. In this comprehensive review, recent developments of MS-based metabolomics in MDD research are summarized from the perspective of analytical platforms (liquid chromatography-MS, gas chromatography-MS, supercritical fluid chromatography-MS, etc.), strategies (untargeted, targeted, and pseudotargeted metabolomics), key metabolite changes (monoamine neurotransmitters, amino acids, lipids, etc.), and antidepressant treatments (both western and traditional Chinese medicines). Depression sub-phenotypes, comorbid depression, and multi-omics approaches are also highlighted to stimulate further advances in MS-based metabolomics in the field of MDD research.

Development and application of lipidomics for food research

Lipidomics is an emerging and promising omics derived from metabolomics to comprehensively analyze all of lipid molecules in biological matrices. The purpose of this chapter is to introduce the development and application of lipidomics for food research. First, three aspects of sample preparation are introduced: food sampling, lipid extraction, and transportation and storage. Second, five types of instruments for data acquisition are summarized: direct infusion-mass spectrometry (MS), chromatographic separation-MS, ion mobility-MS, MS imaging, and nuclear magnetic resonance spectroscopy. Third, data acquisition and analysis software are described for the lipidomics software development. Fourth, the application of lipidomics for food research is discussed such as food origin and adulteration analysis, food processing research, food preservation research, and food nutrition and health research. All the contents suggest that lipidomics is a powerful tool for food research based on its ability of lipid component profile analysis.

Gut bacterial species in late trimester of pregnant sows influence the occurrence of stillborn piglet through pro-inflammation response

Maternal gut microbiota is an important regulator for the metabolism and immunity of the fetus during pregnancy. Recent studies have indicated that maternal intestinal microbiota is closely linked to the development of fetus and infant health. Some bacterial metabolites are considered to be directly involved in immunoregulation of fetus during pregnancy. However, the detailed mechanisms are largely unknown. In this study, we exploited the potential correlation between the gut microbiota of pregnant sows and the occurrence of stillborn piglets by combining the 16S rRNA gene and metagenomic sequencing data, and fecal metabolome in different cohorts. The results showed that several bacterial species from Bacteroides, potential pathogens, and LPS-producing bacteria exhibited significantly higher abundances in the gut of sows giving birth to stillborn piglets. Especially, Bacteroides fragilis stood out as the key driver in both tested cohorts and showed the most significant association with the occurrence of stillborn piglets in the DN1 cohort. However, several species producing short-chain fatty acids (SCFAs), such as Prevotella copri, Clostridium butyricum and Faecalibacterium prausnitzii were enriched in the gut of normal sows. Functional capacity analysis of gut microbiome revealed that the pathways associated with infectious diseases and immune diseases were enriched in sows giving birth to stillborn piglets. However, energy metabolism had higher abundance in normal sows. Fecal metabolome profiling analysis found that Lysophosphatidylethanolamine and phosphatidylethanolamine which are the main components of cell membrane of Gram-negative bacteria showed significantly higher concentration in stillbirth sows, while SCFAs had higher concentration in normal sows. These metabolites were significantly associated with the stillborn-associated bacterial species including Bacteroides fragilis. Lipopolysaccharide (LPS), IL-1β, IL-6, FABP2, and zonulin had higher concentration in the serum of stillbirth sows, indicating increased intestinal permeability and pro-inflammatory response. The results from this study suggested that certain sow gut bacterial species in late trimester of pregnancy, e.g., an excess abundance of Bacteroides fragilis, produced high concentration of LPS which induced sow pro-inflammatory response and might cause the death of the relatively weak piglets in a farrow. This study provided novel evidences about the effect of maternal gut microbiota on the fetus development and health.

Untargeted UHPLC-ESI-QTOF-MS/MS analysis with targeted feature extraction at precursor and fragment level for profiling of the platelet lipidome with ex vivo thrombin-activation

Lipids play a major role in platelet signaling and activation. In this study, we analyzed the platelet lipidome in an untargeted manner by reversed-phase UHPLC for lipid species separation coupled to high-resolution QTOF-MS/MS in data-independent acquisition (DIA) mode with sequential window acquisition of all theoretical fragment ion mass spectra (SWATH) for compound detection. Lipid identification and peak picking was supported by the characteristic regular elution pattern of lipids differing in carbon and double bond numbers. It was primarily based on post-acquisition targeted feature extraction from the SWATH data. Multiple extracted ion chromatograms (EICs) from SWATH data of diagnostic ions on MS1 and MS2 level from both positive and negative ion mode allowed to distinguish between poorly resolved isomeric lipids based on their distinct fragment ions, which were used for relative quantification at a molecular lipid species level. It supports assay specificity for relative lipid quantitation via multiple quantifiably ions unlike to data-dependent acquisition methods which rely on precursor ions only. This approach was used to analyze human platelet samples. 457 lipids were annotated. Concentrations of lipids were estimated by stable isotope-labelled lipid class-specific internal standards as surrogate calibrants. Heatmaps of lipid concentrations in dependence on carbon and double bond numbers for the distinct lipid classes revealed a snapshot of the platelet lipidome in the resting state with lipid species distributions within classes supporting some functional interpretations. As expected, activation of the platelets by thrombin has led to significant alterations in the platelet lipidome as proven by univariate (volcano plot) and multivariate (PLS-DA) statistics. Several lipids were significantly up-regulated (lysophosphatidylinositols, oxylipins such as thromboxane B2 (TXB2), hydroxyheptadecatrienoic acid (HHT), hydroxyeicosatetraenoic acid (HETE), hydroxyoctadecadienoic acid (HODE), sphingoid-bases, (very) long chain saturated fatty acids) or down-regulated (lysophosphatidylethanolamines, polyunsaturated fatty acids, phosphatidylinositols). Several of them are well known as biomarkers of platelet activation while others may provide some further insights into pathways of platelet activation and platelet metabolism.

Determination of inflammation-related lipids in depressive rats by on-line supercritical fluid extraction-supercritical fluid chromatography-tandem mass spectrometry

An on-line supercritical fluid extraction coupled with supercritical fluid chromatography-quadrupole tandem mass spectrometry method was developed to determine lipids related to inflammation in brain tissues of depressed rats. The analysis of 23 lipids from extraction to separation and detection only took 15 min and required 1 mg of brain tissue powder. The matrix effect of the on-line method for endogenous lipids was systematically investigated, and targeted lipids were quantified by matrix effect corrected calibration curves in this study. The on-line method was comprehensively optimized and evaluated. All calibration curves for lipids showed good linearity (correlation coefficient >0.99). The limits of detection and the limits of quantification were in the range of 0.0261-0.396 pg and 0.0791-1.20 pg. The recoveries and the matrix effect were in the range of 85.3-117.5% and 51.9-176.6%, respectively. The relative standard deviations of precision ranged from 2.7 to 14.2%, with accuracies higher than 87.2%. Compared with liquid-liquid extraction coupled with liquid chromatography-tandem mass spectrometry method, the on-line method obtained higher recovery and sensitivity with significantly reduced analytical time, manual operations, and sample amounts. Finally, this on-line method was applied to analyses of brain tissues of depressed rats. Six pro-inflammatory lipids increased in depressed rats, while six anti-inflammatory lipids decreased. Liquiritin and fluoxetine were presumed to promote a similar synthesis of anti-inflammatory lipids. Based on the results, this on-line method showed great promise in analyzing lipids in complex biological samples.

A high coverage pseudotargeted lipidomics method based on three-phase liquid extraction and segment data-dependent acquisition using UHPLC-MS/MS with application to a study of depression rats

Pseudotargeted analysis combines the advantages of untargeted and targeted lipidomics methods based on chromatography-mass spectrometry (MS). This study proposed a comprehensive pseudotargeted lipidomics method based on three-phase liquid extraction (3PLE) and segment data-dependent acquisition (SDDA). We used a 3PLE method to extract the lipids with extensive coverage from biological matrixes. 3PLE was composed of one aqueous and two organic phases. The upper and middle organic phases enriched neutral lipids and glycerophospholipids, respectively, combined and detected together. Besides, the SDDA strategy improved the detection of co-elution ions in the lipidomics analysis. A total of 554 potential lipids were detected by the developed approach in both positive and negative modes using ultra-high-performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS). Compared with the conventional liquid-liquid extraction (LLE) approaches, including methyl tert-butyl ether (MTBE) and Bligh-Dyer (BD) methods, 3PLE combined with SDDA significantly increased the lipid coverage 87.2% and 89.7%, respectively. Also, the proposed pseudotargeted lipidomics approach exhibited higher sensitivity and better repeatability than the untargeted approach. Finally, we applied the established pseudotargeted method to the plasma lipid profiling from the depressed rats and screened 61 differential variables. The results demonstrated that the pseudotargeted method based on 3PLE and SDDA broadened lipid coverage and improved the detection of co-elution ions with excellent sensitivity and precision, indicating significant potential for the lipidomics analysis.