Journal Pre-proof Integration of lipidomics and metabolomics for in-depth understanding of cellular mechanism and disease progression Raoxu Wang, Bowen Li, Sin Man Lam, Guanghou Shui PII: S1673-8527(19)30200-0 DOI: https://doi.org/10.1016/j.jgg.2019.11.009 Reference: JGG 742 To appear in: Journal of Genetics and Genomics Received Date: 26 September 2019 Revised Date: 19 November 2019 Accepted Date: 25 November 2019 Please cite this article as: Wang, R., Li, B., Lam, S.M., Shui, G., Integration of lipidomics and metabolomics for in-depth understanding of cellular mechanism and disease progression, Journal of Genetics and Genomics, https://doi.org/10.1016/j.jgg.2019.11.009. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. Copyright © 2019, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, and Genetics Society of China. Published by Elsevier Limited and Science Press. All rights reserved.

Integration of lipidomics and metabolomics for in-depth understanding of cellular mechanism and disease progression Raoxu Wang1,2, Bowen Li3, Sin Man Lam*1,3, and Guanghou Shui*1,2 1. State Key Laboratory of Molecular Developmental Biology, Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing 100101, China. 2. University of Chinese Academy of Sciences, Beijing 100101, China. 3. Lipidall Technologies Company Limited, Changzhou 213000, China. *Correspondence to: Dr Guanghou Shui, ghshui@genetics.ac.cn Dr Sin Man Lam, smlam@genetics.ac.cn Abstract Mass spectrometry (MS)-based omics technologies are now widely used to profile small molecules in multiple matrices to confer comprehensive snapshots of cellular metabolic phenotypes. The metabolomes of cells, tissues and organisms comprise a variety of molecules including lipids, amino acids, sugars, organic acids, etc. Metabolomics mainly focus on the last three classes, while lipidomics has emerged as an independent omics owing to the complexities of the organismal lipidomes. The potential roles of lipids and small metabolites in disease pathogenesis have been widely investigated in various human diseases, but system-level understanding is largely lacking, which could be partly attributed to the insufficiency in terms of metabolite coverage and quantitation accuracy in current analytical technologies. While scientists are continuously striving to develop high-coverage omics approaches, integration of metabolomics and lipidomics is becoming an emerging approach to for mechanistic investigation. Integration of metabolome and lipidome offers a complete atlas of the metabolic landscape, enabling comprehensive network analysis to identify critical metabolic drivers in disease pathology, facilitating the study of interconnection between lipids and other metabolites in disease progression. In this review, we summarize omics-based findings on the roles of lipids and metabolites in the pathogenesis of selected major diseases threatening public health. We also discuss the advantages of integrating lipidomics and metabolomics for in- depth understanding of molecular mechanism in disease pathogenesis. Keywords Metabolomics, lipidomics, integration, disease progression 1. Introduction

Metabolomics is a relatively young branch of omics science that holds great potential in delivering a closest cellular snapshot of what has happened (Au, 2018). A major challenge in interpreting metabolomics data is that the molecules profiled may represent products at different stages of metabolic reactions and cellular processes, which could have distinct functions in different organisms or even in different cells of one organism. As metabolites are chemically heterogeneous with respect to structural diversity and a wide range of physical properties (i.e., polarity, stability, solubility, etc.) (Au et al., 2016), metabolomics poses an additional layer of challenge analytically relative to other omics. Technical development is still ongoing to achieve higher coverage of metabolites within a single analytical method. Lipidomics represents an emerging discipline that connects lipid biology, technology and medicine, which strives to build an all-inclusive atlas of the cellular/tissue lipidome (Lam et al., 2017a). Due to the structural and functional diversity of lipids, coupled with their high endogenous abundance, lipidomics, originally classified under the broader scope of \"metabolomics\", has become an independent field (Figure 1). According to LIPIDMAPS classification, lipids are further divided into eight major categories, including fatty acyls, glycerolipids, glycerophospholipids, sterol lipids, prenol lipids, sphingolipids, saccharolipids and polyketides (Fahy et al., 2005). Among them, the former six represent the main lipid classes in mammals (Havel et al., 1955). Various strategies have been developed for comprehensive analysis of lipids, including shotgun and chromatography-coupled MS-based lipidomics (Fahy et al., 2011; Lam and Shui, 2013; Lam et al., 2017a). In essence, metabolomics mainly focuses on polar metabolites (water soluble), for instance, sugars, amino acids, organic acids, and nucleotides (Liu and Xu, 2018). Lipidomics strives to comprehensively identify and quantify all kinds of lipid molecular species (Han and Gross, 2005; Schmelzer et al., 2007) Because of their hydrophobic nature, lipids need to be treated and analyzed separately (i.e., requiring different solvent systems) from small molecule metabolites (Smith et al., 2014). Currently over 40, 000 lipids have been documented, with only minimal overlap in terms of coverage with metabolomics. The overlap usually lies in lipids that are more hydrophilic, such as lysophospholipids, acyl-carnitines and free fatty acids (Zhang et al., 2019). It is nonetheless noteworthy that the extraction and analysis of these overlapping lipids in traditional metabolomics are usually limited and not as efficient as that in lipidomics (Nordström et al., 2006). MS analysis is the most widely used technique in both metabolomics and lipidomics analyses, partly due to its superior sensitivity over nuclear magnetic resonance (NMR) (Tian et al., 2016). As shown in Figure 2, the workflow of lipidomics is generally similar to that of metabolomics, except for extraction strategies. While extraction of lipids are quite complicated, which requires diverse organic solvent systems to target different lipid classes, satisfactory extraction of small molecule metabolites can be achieved using solvent mixtures containing methanol, acetonitrile and/or water (Lam et al., 2017a; Tian et al., 2016). In most cases, the lipidome or metabolome extracts can be directly used for liquid chromatography coupled to mass spectrometry (LC-MS) analysis, while derivatization, solid-phase extraction and enrichment might be required for compounds present in trace amounts or when dealing with complex chemical matrices.

Typically, lipidomics or metabolomics analysis can be broadly classified into non- targeted and targeted approaches, which differ in data acquisition method and subsequent data processing, each with its own advantages and disadvantages. Data dependent acquisition (DDA)/ information dependent acquisition (IDA) and data independent acquisition (DIA) are emerging techniques to provide rapid and comprehensive qualitative and quantitative data (Figure 2). Various modes of ionization techniques can be applied depending on sample nature as well as the metabolites of interest. The large amount of data acquired with different strategies will be handled with designated data processing software to transform analytical data into biologically meaningful metabolic maps (Figure 2). Lastly, statistical and bioinformatics analysis will be performed to obtain altered metabolites and pathways, as well as visualizations that will aid our understanding of the data and underlying biology (Lam et al., 2017b). Technical details of lipidomics and metabolomics approaches had been reviewed extensively elsewhere (Dettmer et al., 2007; Lam et al., 2017a). Quantitative profiling analysis of lipidome and metabolome will provide mechanistic clues to biological processes and functions, therefore improving our knowledge of disease development and progression. In particular, the integration of metabolomics and lipidomics is key to understanding the pathogenesis of complex metabolic diseases, given the high degree of inter-connectivity between lipid and non- lipid (i.e. amino acid and carbohydrate) metabolism in governing overall cellular energy homeostasis (Chao et al., 2019; Liu et al., 2001). In this review, we will outline existing knowledge on the respective roles of lipids and small molecule metabolites, as well as their synergistic functions, in essential cellular process and progression of selected metabolic diseases. We will further discuss the advantages of integrating lipidomics and metabolomics for in-depth understanding of the relationship and interconnection between lipid and other metabolites in disease progression. 2. Lipids and their functions in disease progression Lipidomics has been widely used for investigating lipid metabolism in various organismal models and human diseases, which has been reviewed elsewhere (Lam and Shui, 2013), and hence will not be the focus of this section. Instead, functions of individual lipid molecule or lipid class in selected diseases including metabolic syndrome, dementia, cardiovascular diseases and various forms of cancers, will be discussed herein to shed the light on the importance of accurate quantification and precise identification of lipids facilitated by lipidomics. 2.1 Fatty acyls Fatty acids function as an energy source and precursors of various bioactive lipid molecules and membrane lipids. High levels of plasma free fatty acids (FFAs) might be indications of expandability issues of white adipocytes, inefficient energy expenditure and/or perturbed β-oxidation (Beloribi-Djefaflia et al., 2016). With respect to oncology,

FFAs released from hydrolysis of triacylglycerides (TAGs) are taken up by metastatic ovarian cancer cells and used as energy source (Nieman et al., 2011). As precursors of various bioactive lipids, different free fatty acid species have been implicated in the initiation and progression of a plethora of diseases. For instance, free arachidonic acid was shown to be increased in lung tumors with high MYC activity. MYC overexpression rises the levels of arachidonic acid-derived eicosanoids via the lipoxygenase (LOX) and cycloxygenase (COX) pathways. Inhibiting the COX/5-LOX pathways reduces cell proliferation and thus alleviates tumor burden, highlighting the role of arachidonic acid and its derived eicosanoids in the cancer cell proliferation (Hall et al., 2016). Palmitic acid has been shown to trigger endoplasmic reticulum (ER) stress and promote apoptotic and necrotic cell death in the renal proximal tubular cell line (Katsoulieris et al., 2009). Cleavage of polyunsaturated fatty acids from membrane phospholipids releases polyunsaturated fatty acids that leads to significant up-regulation of eicosanoid biosynthesis and thus decreases life span of Caenorhabditis elegans dauer larva (Lam et al., 2017c). Changes in fatty acyl composition can severely alter membrane fluidity, thickness, packing, as well as the dynamics and functions of membrane proteins (Katsoulieris et al., 2009; Lam et al., 2017c; Martin et al, 2010; Rysman et al., 2010; Staubach, and Hanisch, 2011). For example, increased saturated phospholipids (PLs) level results in alteration of signal transduction in cancer cells, which protect cancer cells from oxidative injury and inhibit its uptake of chemotherapeutic drugs (Rysman et al., 2010; Staubach, and Hanisch, 2011). Lipid rafts isolated from the brains of Alzheimer’s disease (AD) patients displayed diminished levels of n-3 polyunsaturated fatty acids (particularly docosahexaenoic acid, DHA), monoenes (mainly oleic acid) and reduced peroxidizability indexes (Martin et al, 2010). Furthermore, differential fatty acyl composition in AD frontal and entorhinal cortices was shown to create a higher degree of membrane order and viscosity as validated via their greater resistance to artificial probes. The biological relevance of these membrane fatty acyl alterations to AD pathogenesis was further strengthened by observations that β-secretase specifically accumulates in these altered membrane microdomains of AD subjects even at mild stages of the disease, therefore providing a direct link between fatty acyl compositional changes in membrane microdomains and amyloidogenic processing of amyloid precursor protein (APP) (Diaz et al., 2015; Fabelo et al., 2014). On a similar note, decreased phospholipids containing DHAs, was observed to accompany increases in very-long-chain sphingomyelins (VLC-SMs) during normative ageing in the neural membranes of Rhesus macaques, suggesting that these membranes may elicit a time-dependent interchange of DHA-enriched versus raft (SM/cholesterol) microdomains as normative ageing progresses ( Lam et al., 2016). Those findings further highlighted the application of lipidomics in investigating effects of membrane fatty acyl compositions on cellular processes and signaling transduction in neurodegenerative diseases. 2.2 Glycerophospholipids Glycerophospholipids are main constituents of cellular membranes involved in cellular signal transduction. According to the nature of their polar head groups, phospholipid family is further categorized into phosphatidylcholine (PC),

phosphatidylethanolamine (PE), phosphatidylserine (PS), phosphatidylinositol (PI), phosphatidic acid (PA), cardiolipin (CL), etc (Figure 3). PC and PE are the most abundant phospholipid classes in mammalian cells, and showed distinct changes during the progression of a number of diseases (Beloribi- Djefaflia et al., 2016; Chen et al., 2018; Epand et al., 1996; Fabelo et al., 2014; Guo et al., 2014; Koeberle et al., 2013; Lam et al., 2016; Marien et al., 2015; Yu et al., 2011). Enhanced PC levels were observed in numerous cancers, which may suggest the positive association between endogenous PC levels and elevated proliferative rate in cancer cells (Guo et al., 2014). Lipidomics analysis demonstrated that 20:4-PC fluctuates during the cell cycle and slows down cell cycle progression by interfering with the protein serine/threonine kinase Akt membrane binding, thus contributing to cell cycle regulation in cancer (Koeberle et al., 2013). Similar to PC, PE also showed a consistent elevation in tumors (Chen et al., 2018). Overexpression of phosphatidylethanolamine-binding protein 4 (PEBP4) in lung cancer could modulate the development, invasion, and metastatic potential of tumors (Yu et al., 2011). It was thought that the increase in PEs may partially work as agonists of PEBP to mediate signaling transduction (Chen et al., 2018). Furthermore, lipidomic analysis of the sarcoplasmic reticulum (SR) showed that perturbed ratio of PC to PE is associated with decreased the sarcoplasmic reticulum calcium ATPase (SERCA) activity, leading to aberrations in calcium homeostasis that exacerbates heart failure (Spears et al., 2018). These findings emphasize the relevance of phospholipid-induced alterations in cell signaling in cancer development and progression. PIs and their phosphorylated forms phosphatidylinositol polyphosphates (PIPs) perform vital roles in cell signaling transduction and membrane trafficking. Upon stimulation, the production of PI(3,4,5)P3 by PI 3-kinase can subsequently serve to recruit and co-localize the pleckstrin homology (PH) domains of both Akt and PtdIns- dependent kinases (PDKs) to the plasma membrane. The membrane co-localization promotes Akt phosphorylation by PDK, which turns on the Akt-mediated pathway for cellular growth and survival (Wishart and Dixon, 2002). Ablation of PTEN, which cleaves phosphate group from PI(3,4,5)P3, is shown to promote tumor development by regulating Akt-dependent pathway (Maehama et al., 2001). PI and its phosphorylated metabolites, namely PI(4,5)P2 and PI(3,4,5)P3, were shown to be involved in delineating the cellular apical and basolateral membrane polarity respectively (Gassama-Diagne et al., 2006). In Huntington’s disease (HD), perturbed PI homeostasis may result from interaction of mutant huntingtin protein with membrane phospholipids that subsequently alters the normal morphology of ependymal cells especially in terms of their ciliary formation and function (Hunter et al., 2018). Lipidomics analysis demonstrated that PI (16:0-18:1) is a potential biomarker indicative of the malignancy of breast tumor (Yang et al., 2016). Almost exclusively, CL is distributed in mitochondrial membranes, involving in the maintenance of membrane integrity and mitochondrial functionality (Beloribi-Djefaflia et al., 2016). Patients with Barth syndrome have a primary defect in CL remodeling. The accelerated degradation of CL and reduced incorporation of linoleic acid into CL lead to aberrant cell function (Vreken et al., 2000). In another study, lipidomic analysis demonstrated the abnormal distribution of CL molecular species and reduced CL levels in brain tumor mitochondria, leading to irreversible respiratory injury that may impede the use of alternative energy sources to glucose (Kiebish et al., 2008).

Previous work had also revealed that drastic changes in CL profiles were associated with synaptic mitochondrial dysfunction in early stages of AD (Monteiro-Cardoso et al., 2015). The findings above indicated that individual lipids are specific (i.e. with precise head group and fatty acyl moieties) in their biological functions. Lipidomics quantitation that confer structural details of individual lipids is therefore essential to unveil the precise lipid targets implicated in disease pathogenesis. Plasmalogens are a subclass of glycerophospholipids that contain a vinyl ether fatty alcohol substituent at sn-1 position of the glycerol backbone. As oxidation of enol ether does not generate reactive species that propagate the oxidation of PUFAs, plasmalogen PE (pPE) is known to act as effective antioxidants in cellular membranes (Brosche and Platt, 1998). Deficiency in pPE in AD supports the oxidative stress hypothesis that oxidative stress may be a key risk factor in the disease induction and progression of AD (Naudi et al., 2015). As pPE and plasmalogen PC (pPC) are synthesized in peroxisomes, the decreased pPE and pPC content in AD have been attributed to peroxisomal dysfunction (Wallner and Schmitz, 2011). Lysophospholipids contain only one FA moiety in their structures; examples are lysophosphatidylcholine (lyso-PC), lysophospholipid acid (lyso-PA), and lyso phosphatidylinositol (lyso-PI). Lyso-PC acts as an inflammatory medium that modulates proliferation and apoptosis of endothelial cells, thereby affecting the development of atherosclerosis (Akerele and Cheema, 2015). Lipoprotein-associated phospholipase A2 (Lp-PLA2), responsible for the formation of various lyso-PLs, is a well-known risk factor for ischemic stroke (Li et al., 2017; Wang et al., 2017). Pathological overstimulation of PLA2 leads to high levels of circulating lyso-PCs in cerebrospinal fluid (CSF) of multiple sclerosis (MuS) patients (Pieragostino et al., 2015). Lyso-PA influences both the apoptosis of myocardial cells (Tsukahara et al., 2014) and the proliferation of fibroblasts (Abdel-Latif et al., 2015; Kostic et al., 2015), and was found to be play an important part in coronary heart disease development (Liu et al., 2018). 2.3 Sphingolipids Sphingolipids mainly include ceramides, sphingomyelin (SM), and glycosphingolipids (Fahy et al., 2005), and exhibit a wide spectrum of biological functions depending on cellular context (Futerman and Hannun, 2004). Perturbation in sphingolipid levels of plasma and tissues were shown to elevate the risk of cardiovascular complications in diabetics (Pinz et al., 2017). Besides, acyl chain length-specific alterations in brain sphingolipid profiles were also reported during aging (Tu et al., 2017). Ceramides levels, in general, are known to elevate in stable coronary artery disease (Meikle et al., 2014), AD (Naudi et al., 2015), hypertension (Spijkers et al., 2011), obesity, diabetes, insulin resistance (IR) (Sas et al., 2015). Low density lipoproteins enriched (LDLs) enriched in ceramides can lead to IR and promote inflammation in macrophages mediated by IL-6 (Boon et al., 2013; de Mello et al., 2009). Ceramides increase mitochondrial membrane permeability and facilitate the release of apoptogenic factors, which subsequently trigger the caspase cascade that constitutes the intrinsic pathway of apoptosis (Colombini et al., 2017). Interestingly, different ceramide species (with fatty acyl carbon atom numbers ranging from 14 to 26) are found to elicit disparate biological functions (Hla and Kolesnick, 2014). C16:0 ceramide was reported to be pro-

apoptotic and exacerbates obesity-related IR (Raichur et al., 2014; Turpin et al., 2014), while very-long-chain ceramides such as C24:0, being anti-apoptotic that promote cell proliferation, were reported to exert effects largely antagonistic to their short-chain counterparts. Comparable results were observed in the study of early liver IR in rats (Taltavull et al., 2016). SMs have been implicated in promoting cell proliferation in the brain. Higher levels of SMs in the myelin layer was previously noted in the subventricular zone (SVZ) of HD patients, which were postulated to enhance reception of stimuli from neighboring degenerating caudate to promote cell proliferation and neurogenesis in order to compensate for neuronal cell death (Curtis et al., 2003; Hunter et al., 2018). Gangliosides denote a class of lipids that comprises one or more sialic acids attached to a sugar chain. GM3, being the simplest member of this complex family of lipids, are known to function as important precursors for biosynthesis of the a- and b-series of gangliosides endogenously abundant in the brain (Proia, 2004). Higher plasma GM3 levels are also found to be associated with IR (Guo et al., 2014; Tagami et al., 2002), type 2 diabetes mellitus (T2DM) (Shui et al., 2013) and Parkinson’s disease (PD) (Chan et al., 2017). Lysosomal degradation of GM3 internalized via the endocytic pathway was observed to be diminished in PD (Alcalay et al., 2015; Chan et al., 2017; Murphy et al., 2014). GM3 was also found to speed up α-synuclein aggregation in vitro using artificial liposomes, as opposed remaining phospholipids that slowed down aggregation (Grey et al., 2015). Furthermore, lipidomics analyses have previously reported that disease-dependent alterations in both SM and GM3 were found to be fatty acyl-specific (Lam et al., 2014; Shui et al., 2013). 2.4 Neutral lipids Neutral lipids represents a group of hydrophobic molecules lacking charged groups, and mainly include triacylglycerols (TAG), diacylglycerols (DAG), cholesterol and its esters in mammals. TAG is thought to be implicated in various diseases including cardiovascular diseases, ischemic stroke and dyslipidemia (Labreuche et al., 2009; Labreuche et al., 2010; Stahlman et al., 2012; Stegemann et al., 2014; Yang et al., 2017). In dyslipidemic women, very-low-density lipoprotein (VLDL) particles containing TAG species enriched in palmitic acid promoted secretion of proinflammatory mediators from muscle cells (Stahlman et al., 2012). DAGs have been implicated in insulin resistance, AD and hypertension (Kulkarni et al., 2013; Li et al., 2004; Wood et al., 2015). Enhanced levels of cellular DAGs recruit various isoforms of protein kinase C to plasma membrane, responsible for their activation and subsequent inhibitory phosphorylation of IRS1, thereby inhibiting insulin signaling pathway and eventually glucose uptake (Erion and Shulman, 2010; Li et al., 2004; Perry et al., 2014). DAG 34:2 and DAG 36:2 levels were significantly increased in both plasma and the frontal cortex of AD patients, and displayed associations with Mini-Mental State Examination (MMSE) scores (Wood et al., 2015). Elevated DAGs were attributed to enhanced phospholipase-mediated degradation of PEs, the levels of which were reportedly reduced in AD (Gonzalez-Dominguez et al., 2014; Wood et al., 2015). As a potential marker for hypertension, DAG was significantly associated with both higher

systolic and diastolic blood pressure (Kulkarni et al., 2013), which might stem from its effect on TRPC6 channel mediating vasoconstriction (Wang et al., 2008). Cholesterol is one of the most abundant lipids in animal cells, which not only serves as basic building blocks of membranes but also a precursor for the biosynthesis of steroid hormones, bile acids and vitamin D. Cholesterol is implicated in various cellular functions including intracellular transport, cell signaling, and so on (Incardona and Eaton, 2000). Cholesterol is pivotal in the development and progression of atherosclerosis, since enhanced levels of free cholesterol and cholesteryl esters are known to give rise to unstable plaques (Davies, 1996). Besides, cholesterol also exerts a critical role in promoting plaque progression and inflammation underlying secondary plaque characteristics (Meikle et al., 2014). As a major component of lipid raft, cholesterol modulates amyloidogenic processing of APP in Alzheimer’s disease and regulates cell survival and cell death in cancer via cholesterol-enriched membrane microdomains (Di Paolo and Kim, 2011; Mollinedo and Gajate, 2015). 3. Small molecule metabolites and their implication in disease progression Metabolomics refers to the systematic study of the chemical fingerprints of cellular processes, primarily aims at quantitative and/or qualitative analysis of small polar molecules (metabolome) including amino acids, carbohydrates, organic, etc. As the metabolome denotes the end-product of cellular processes (Fiehn, 2002), its comprehensive analysis (metabolomics) provides the most relevant functional readout or physiological status of an organism. Amino acid disorders can arise from impaired synthesis or degradation of amino acids. Metabolomics has revealed that branched chain amino acids (BCAA) and related metabolites are more strongly associated with IR than lipids (Newgard, 2012). BCAA profiles could be used as biomarkers to predict incident diabetes and intervention/treatment consequence (Zhang et al., 2014). Elevated levels of BCAA were observed incident to T2DM manifestation and might reflect a state of IR (Batch et al., 2014). Increased levels of BCAA may be attributed to lower BCAA catabolism regulated by the branched-chain alpha-ketoacid dehydrogenase complex (BCKD). A decrease in the quantity and activity of the BCKD complex mediated by adiposity signals was observed to curtail BCAA catabolism and clearance (Nagata et al., 2013). In contrast to the increased levels of BCAA, glycine levels was decreased in both predictive studies and overt diabetes (Klein and Shearer, 2016). Glycine reduction was speculated to result from the increase of gluconeogenesis (Floegel et al., 2013), as well as glutathione consumption and decreased glutathione synthesis driven by increased oxidative stress (Sekhar et al., 2011). Similar to the observations in T2DM, different amino acids function in distinct pathways in the pathogenesis of stroke. Cysteine was reported to associate with ischemic brain damage. In patients with early stroke deterioration, elevated cysteine concentration was reported. Its deleterious effect may be mediated by its downstream product hydrogen sulfide (Wong et al., 2006). Oxidation of glutathione results in oxidative stress and possibly ischemic stroke via increased reactive oxygen species (Lien et al., 2017). Excessive release of glutamate promotes calcium influx into neurons, thus inducing neurotoxicity, neuronal cell death and brain damage (Lai et al., 2011; Lai et al.,

2014; Xing et al., 2012). Tryptophan metabolism was found to be altered upon stroke onset. Its oxidation via a major route, the kynurenine metabolic pathway, has been implicated in inflammatory response, oxidative stress, endothelial dysfunction, carotid plaque enlargement, and cerebral ischemia (Chen and Guillemin, 2009; Wang et al., 2015; Isabel Cuartero et al., 2016). Carbohydrate metabolism is significantly altered in obesity, diabetes, tumors and ischemia (Belanger et al., 2011; Berthet et al., 2012; Fiehn et al., 2010; Floegel et al., 2013; Gaglio et al., 2011; Gogna et al., 2015; Mergenthaler et al., 2013; Moore et al., 2014; Padberg et al., 2014; Ying et al., 2012). The increased levels of various carbohydrates including glucose, fructose, lactate, and gluconic acid indicate the perturbation of carbohydrates catabolism in obesity (Fiehn et al., 2010; Gogna et al., 2015; Moore et al., 2014). As obesity and diabetes are tightly related, similar changes in carbohydrates were observed in diabetes. A few metabolomics studies had also revealed that sugars, including glucose, mannose, fructose, and hexose, were increased even 3–7 years before clinical manifestation of T2DM (Floegel et al., 2013; Padberg et al., 2014). The enhanced uptake of glucose and its utilization in glycolysis leading to elevated lactate production is the first adaptive event in tumorigenesis (Gaglio et al., 2011; Ying et al., 2012). During episodes of ischemia, increased lactate production was also observed, which possibly denotes a protective effort by the brain to restore energy homeostasis and prevent neuronal apoptosis (Belanger et al., 2011; Berthet et al., 2012; Mergenthaler et al., 2013). The tricarboxylic acid (TCA) cycle represents a critical means for energy release in the form of adenosine triphosphates (ATPs) via oxidation of acetyl-CoAs generated from carbohydrates, lipids and amino acids common to all aerobic organisms. An existing model for obesity and T2DM pathogenesis proposes that elevated rates of β-oxidation to cope with metabolic overload in skeletal muscles may drain TCA cycle supporting intermediates as a result of incomplete β-oxidation, which consequently leads to IR (Koves et al., 2008; Jelenik and Roden, 2013). Using unbiased and targeted metabolomics approaches, it was discovered that the levels of key metabolites involved in the mitochondrial shuttle, such as citrate and malate, were appreciably altered in AD patients that may form the basis of mitochondrial dysfunction in AD brains (Paglia et al., 2016). 4. Integration of metabolomics and lipidomics 4.1 Integrated omics provides a deeper understanding of cellular process and pathophysiology Integrated studies of metabolomics and lipidomics serve to confer unbiased and comprehensive empirical datasets particularly useful in terms of systematically uncovering molecular mechanisms behind complex cellular processes. While lipidomics mainly provides a snapshot of water-insoluble molecules, metabolomics profiles water- soluble polar compounds. Previous studies have indicated that both lipids and polar metabolites were involved in the pathophysiology of diseases with a high degree of interconnectivity (Fahrmann et al., 2015; La Torre et al., 2013; Murfitt et al., 2018;

Oresic et al., 2008; Overgaard et al., 2016; Sysi-Aho et al., 2011; Zheng et al., 2014). As lipids and water-soluble polar metabolites are often interconnected in common biological pathways, synergistic application of lipidomics and metabolomics can serve to reaffirm pathway perturbations and identify key enzymes at precise steps of metabolic pathway responsible for biological aberrations underlying cellular processes and/or diseases. Integration of metabolomics and lipidomics is able to cover majority of small molecules and thus offer a global signature of cellular damage in diseases, allowing the discovery of pathway alterations in an unbiased manner. The discovery of a gut-to-cardiac axis in instigating atherosclerosis denotes a classic example of how combination of a series of metabolomics and lipidomics studies facilitated the elucidation of precise pathway alterations culminating in cardiovascular diseases (CVD). Metabolomics analyses first identified that plasma levels of three metabolites implicated in phosphatidylcholine metabolism, namely choline, betaine, and trimethylamine-N-oxide (TMAO) that emerged as effective prognostic indicators of CVD in clinical cohorts (Wang et al., 2011). Mechanistic investigation using mouse models revealed that dietary supplementation of choline enhanced the activity of multiple pro- atherosclerotic macrophage scavenger receptors that were lost in germ-free mice. In a separate study to elucidate the determinants of plasma TMAO in humans, lipidomics and metabolomics were simultaneously applied to monitor the levels of phosphatidylcholines, cholines and TMAO in human cohort categorized based on median concentrations of TMAO and choline (Obeid et al., 2016). In a step-wise regression model, it was found that plasma phosphatidylcholine denotes a negative determinant of TMAO, while trimethylamine and choline were positive determinants. These findings cumulatively validated the biological connectivity between phosphatidylcholines, choline, trimethylamine (TMA) and TMAO in terms of CVD progression, and reinforced the mechanistic proposal that choline released from dietary intake of phosphatidylcholines is metabolized first by host enzymes to form gaseous TMA, which is subsequently captured and converted by gut microbes into TMAO. Release of TMAO into the blood stream promotes cholesterol absorption by macrophages, and reduces bile acid production that represents a critical avenue of cholesterol excretion (Figure 4). These cellular processes lead to enhanced cholesterol accumulation that promotes the formation of cholesterol- laden macrophages that ultimately give rise to atherosclerotic plaques (Wang et al., 2011). In addition to unravelling disease pathology, the combination of metabolomics and lipidomics in this instance has also undoubtedly put forth several new treatment regimens for alleviation of CVD, such as dietary intervention or probiotics usage to modulate endogenous TMAO production. An integrated omics approach has also been utilized to monitor early stage disease progression in both animal models and human subjects with type 1 diabetes (T1DM) (Fahrmann et al., 2015; La Torre et al., 2013; Oresic et al., 2008; Overgaard et al., 2016; Sysi-Aho et al., 2011). Serum levels of succinate, PC and TAG at birth were identified as prognostic indicators of T1DM development later in life (Oresic et al., 2008). In an animal study comparing non-obese diabetic (NOD) and NOD-E (transgenic NOD mice that express the I-E heterodimer of the major histocompatibility complex II mice), researchers demonstrated that the development of T1DM in NOD mice is accompanied by changes in both lipid, purine, as well as amino acid (tryptophan) metabolism.

Integrated metabolomics and lipidomics could also provide small molecule signatures of disease etiology. In particular, a high degree of high intra-class correlation was observed specifically between metabolites implicated in lipid and amino acid metabolites, but not for that in carbohydrate metabolism, in individuals at risk of atherosclerosis (Zheng et al., 2014). A global snapshot of lipidome and metabolome might be critical to identify a more general cellular and biological process like membrane integrity maintenance and cell growth, highlighting their essential role in disease progression (Figure 5). Multiple Sclerosis (MuS) represents a chronic inflammatory disease related to central nervous system. CSF was analyzed with lipidomics coupled with metabolomics. Lyso-PC, lyso-PI and PI were found to be altered in MuS and correlated with clinic variables of the disease. Elevated level of glutamate was also observed in MuS. The concomitant elevations in CSF glutamate and lyso-PCs were interestingly found to be elicited via a connected pathway (Pieragostino et al., 2015). Glutamate release, calcium influx, and the subsequent activation of cellular PLA2 were critical in instigating membrane disintegration, a characteristic feature of neuronal degeneration in acute and chronic neurological disorders (Klein, 2000). Thyroid carcinoma is the most common cancer of endocrine system with a sharp rise in its incidence over the past decade. The altered metabolites of highest clinical importance included a mixture of lipids and small molecules, including glucose, galactose, lactate, inositol, citrate, cholesterol, etc. In serum and plasma samples, lipids, in particular, cholesterol and sphingomyelin, were the most distinct compounds between normal and cancerous groups. All these metabolites take part to meet the high energy demand and biosynthesis of building blocks (proteins, lipids and nucleotides) required for cell growth and division. Taken together, carbohydrate, lipid and nucleotide metabolism were emphasized as the most important pathways implicated in thyroid cancer pathogenesis (Farrokhi et al., 2017) (Figure 5). Patterns of changes in lipids and other metabolites may also differ in disease progression. For instance, sphingomyelins and ether-containing phosphatidylcholines were changed in preclinical biomarker-defined AD stages, while acylcarnitines, BCAAs valine and α-aminoadipic acid were altered only later in symptomatic stages (Toledo et al., 2017). This example has demonstrated how integration of metabolomics and lipidomics for metabolic network analysis can further improve initial diagnosis and monitoring of disease progression instead of looking at information from a single “omics” at a time (Figure 5). Global measurement of metabolites render possible an unbiased network analysis approach to sieve out key metabolic drivers in disease pathophysiology (Soltow et al., 2010) (Figure 5). Preliminary small-scale metabolomics or lipidomics studies in AD have underscored metabolic perturbations including PC (Mapstone et al., 2014; Simpson et al., 2016), ceramide–SM pathways (Han et al., 2011), pPE (Han et al., 2001; Wood et al., 2011), amines (Ibanez et al., 2012), and mitochondrial defects (T Trushina et al., 2013). Metabolic network analysis have associated changes in norepinephrine and purines with augmented levels of CSF tau, while tryptophan and methionine correlated with the levels of Aβ (Kaddurah-Daouk et al., 2013). Inverse correlations between brain volume changes

and cognition with metabolites including acylcarnitines, valine, and α-aminoadipic acid were observed, suggestive of a shift in energy substrate utilization in later stages of AD. The association between long-chain acylcarnitines and odd-chain acylcarnitines and amino acids with cognitive scores further reinforced the notion of changing fuel utilization from fatty acids to amino acids in neurodegeneration. Furthermore, co- clustering of short-chain acylcarnitines with amines, both segregated from PCs and SMs, indicated that the short-chain acylcarnitines were implicated in a common metabolic pathway with amino acids instead of lipids in AD patients. This novel finding derived from a synergistic approach of lipidomics cum metabolomics has unveiled a disease- associated transition in pathways related to utilization of energy substrates (Toledo et al., 2017). Aside from elucidation of disease pathology, an integrated omics-approach is also useful in delineating the precise metabolic pathway alterations underlying specific genetic modulation associated with interesting metabolic phenotypes. For example, a combined metabolomics and lipidomics approach revealed that resistance to high fat diet- induced obesity and diabetes in transgenic mice overexpressing CYP7A1 was attributed an interconnection between bile production and circulating lipid levels (Qi et al., 2015). In particular, it was found that CYP7A1 overexpression leads to elevated levels of tauro- β-muricholic acids that antagonizes intestinal farnesoid X receptor signaling, increasing overall pool of intestinal bile acids. Expanded intestinal bile acid pool reduces bile de- conjugation by gut microbes and thus lowers the levels of secondary bile such as deoxycholic acid, which was previously shown to activate acid sphingomyelinase that mediates SM to ceramide conversion (Gupta et al., 2004). In line with pathway alterations uncovered via metabolomics, lipidomics revealed diminished levels of ceramides in CYP7A1 transgenic mice, which may essentially confer the protection against inflammation and insulin resistance under a high-fat diet. Simultaneous application of lipidomics and metabolomics also allows unbiased interrogation of mechanisms of action of drugs, which is critical for evaluation of drug safety and usage. For example, omics analyses revealed that Qingkailing, a traditional Chinese medicine widely applied for treatment of fever, acts through parallel pathways implicating both lipid and amino acid metabolism (Qin et al., 2016). Lipidomics revealed that Qingkailing increases release of arachidonic acids from arachidonyl-PCs levels, which may be subsequently channeled for biosynthesis of prostaglandin E2 that reduces fever (Kita et al., 2015). On the other hand, metabolomics shown that Qingkailing administration increases plasma level of tyrosine, which was reported to cross the blood- brain barrier and serves as precursors for the biosynthesis of neurotransmitters modulating body temperature (Lipton and Clark, 1986). As such, can be seen that a single drug often has multiple arms of actions that induce changes in both the biological lipidome and metabolome, and that lipids and small metabolites are highly interconnected in functional networks within biological organisms. In light of this, we discuss some known pathway connections between lipids and critical small metabolites including glucose and amino acids that denote the fundamental constituents of the biological metabolic pool in the following section (Table 1). 4.2 Effort to analytically combine metabolomics and lipidomics

Attempts had been made to analyze both metabolomics and lipidomics with a single extraction procedure or a biphasic extraction method, this will allow the profiling of various classes of biomolecules from the same sample (Cicalini et al., 2019). A few new analytical techniques had facilitated simultaneous analysis of metabolome and lipidome in a single analytical run, which include SIMPLEX extraction approach (Coman et al., 2016), stacked injections of a biphasic extraction (Broeckling and Prenni, 2018), and two-dimensional liquid chromatography-mass spectrometry (Wang et al., 2017). With these approach, reasonable coverage and extraction efficiency had been achieved in a single analytical run (Schwaiger et al., 2019). Nonetheless, the differences in chemical polarities coupled with the requirements of separate metabolite libraries for structural identifications would mean that metabolomics and lipidomics would still remain largely separate in terms of analytical strategies, at least in the short run. Consolidating and unifying metabolomics and lipidomics data on a biological basis downstream of analytics data acquisition would remain a preferred approach for integration of metabolomics and lipidomics. 4.3 Strategies for integrating omics data Integrating metabolomics and lipidomics data compensates for missing information in the metabolic network, and multiple orthogonal sources of evidence supporting the same hypothesis reduce the chance of false discoveries. Data integration thus holds great potential to unravel complex relationships between two omes and provide a comprehensive view of the metabolic state. In general, integrative omics data analysis aims to answer three questions (1) which features from different omics data sets best explain the variable of interest, (2) how the omics data sets are related to each other and (3) what functional interpretation can be drawn by integrating omics data sets. Different methods are needed to address each of the questions, but a common strategy is dimension reduction (Figure 6). Dimension reduction is an important step in high- dimensional data analysis, especially in omics data analysis, because omics data are extremely high-dimensional, highly correlated and noisy. Not surprisingly, with the immense number of features and considerable complexity, dimension reduction is crucial in any integrative omics data analysis. Dimension reduction techniques for the integrative analysis of omics data are available for one omics data set (individual or concatenated lipidomics and metabolomics data sets) like principal component analysis (PCA), multiple factor analysis (MFA) and nonnegative matrix factorization (NMF). For pairs of omics data sets, approaches like canonical correlation analysis (CCA) and partial least square of K-tables (multi-block PLS) can be used, while generalized CCA is usually used for multiple omics datasets (Meng et al., 2016; Zeng and Lumley, 2018). Many of these statistical approaches, such as PCA, CCA and PLS are projection-based methods whereby samples are summarized by some unobserved latent variables or components which are linear combination of the original variables, such that samples can be classified and visualized in a small-dimensional space. In the linear combination, the weights of the original variables in a component are indicated by a loading vector and usually most of the weights are non-zero, which indicates that all the original variables are still considered. In order to achieve higher efficiency and interpretability, and to control for overfitting, variable selection is often jointly employed using regularization technique, for example sparse PCA (sPCA), regularized canonical correlation (rCCA), sparse PLS

(sPLS), etc. Implementation of these methods are available in various R packages, and a comprehensive summary had been presented elsewhere (Meng et al., 2016). Also notably, another R package mixOmics (Rohart et al., 2017) offers a range of multivariate analyses of multiple omics data sets focusing on data exploration, dimension reduction and visualization. Machine learning approaches, such as random forest (RF), has also been applied to integrate multiple omics data sets (Acharjee et al., 2016) (Figure 6). Random forest are ensembles of decision trees whereby each decision tree is trained on a random bootstrap sample of the original data over a random subset of variables. RF chooses the classification that gets most “votes” from all the trees in the forest. RF runs efficiently on large data sets and has built-in methods for imputation of continuous and categorical variables, which can effectively handle missing data in omics data, as well as associated clinical and phenotype data. Besides, RF can estimate variable importance as the difference between classification outcome of the original data and the one after permutating values of the variable, which can aid in the selection of candidate bio- markers. In addition, variable selection can be implemented by iteratively eliminating variables with lowest variable importance and evaluating the model performance on test data. Analyses of omics data and associated clinical and phenotype data always yield a wealth of information, including differential metabolites/lipids, fold change, as well as strength of association like correlation, partial correlation, odds ratio, etc. For researchers to better understand the molecular mechanism at the system level, biochemical relationships among metabolites and lipids, and their structural similarities also have to be taken into account. Network-based data integration approaches excel in this aspect for their great flexibility and scalability (Figure 6). Typically, in a network, nodes represent metabolites and edges represent relationships, in which properties of metabolites like P values, fold change, chemical class can be mapped to node color, size, shape, and properties of relationships can be mapped to edge width, line type, color, etc. R package igraph provides user-friendly methods for construction and manipulation of general network graphs, and algorithms for network analysis, including hub and module identification. Other R packages like WGCNA (Langfelder and Horvath, 2008), MEGENA (Song and Zhang, 2015) and MetaMapp (Barupal et al., 2012) provide methods for more specialized networks with different focuses, like network structure analysis and integration of biochemical and pathway information. Such analyses may be particularly useful in analyzing complex metabolic relationships that display a high level of interconnectivity, such as lipid networks, to reveal changes in metabolite correlations in different physiological states. For instance, MEGENA had shown that correlations between polyunsaturated pPEs and TAGs were altered in incident diabetes compared to control four years prior to disease onset, unveiling candidate pathways that may be responsible for the manifestation of overt dyslipidemia accompanying diabetes onset (Lu et al., 2019). 5. Concluding remarks and perspectives Metabolomics and lipidomics have emerged as the principal tools to address various basic biological questions as well as to delineate the molecular pathophysiology

of human diseases. Specific metabolites/lipids have unique and remarkable functional properties, and play specific roles in regulating cell proliferation, metabolism, organelle function, endocytosis, autophagy, stress responses, apoptosis, signal transduction, aging and so on. While more comprehensive information is urgently required for pathway analysis, scientists are continuously striving to develop high-coverage omics approaches, such as an integrated platform for extensive profiling of diverse sphingolipids (Lam et al., 2018), high-coverage global lipidomics approaches (Lam et al., 2014; Lu et al., 2019; Zhou et al., 2018), and high-coverage targeted metabolomics method (Li et al., 2017; Zha et al., 2018). In future, advances in chromatography and MS techniques will further aid to expand the coverage of metabolites and lipids. Although both metabolomics and lipidomics have significantly contributed to advancing our current understanding of cellular mechanism of disease progression, it has become apparent that a single-sided omics approach fails to take into account the high level of interconnectivity in the cellular metabolism of the analytes covered by either approach. While analytical feasibility has demarcated lipidomics from the more general field of metabolomics, an integration of metabolomics and lipidomics in approaching clinical and biological questions is clearly necessary for a complete metabolic overview prerequisite to answering various fundamental questions in cellular biology and disease pathology. A synergistic omics approach has several advantages, such as in providing a complete set of molecular changes and global signature, highlighting the common cellular mechanisms among lipids and other metabolites, enabling comprehensive network analysis to identify critical metabolic drivers in disease pathology, and facilitating the study of interconnection between lipids and other metabolites in disease progression. Despite considerable technological advancement within the respective fields of metabolomics and lipidomics, limitations and challenges still exist for an integrative approach. Aside from the analytical challenges imposed by the vast differences in chemical properties that may circumvent an integrated analytical approach, an integrated metabolomics/lipidomics approach also adds a sheer amount of complexity in terms of data analysis, processing as well as interpretation. In addition to continuously delving deeper into the respective fields of metabolomics and lipidomics, a convergent approach in terms of data interpretation on a biological basis may denote the final resolution to providing critical clues to cellular biology and disease pathology that have by far remained elusive. Acknowledgements This research was funded by National Key R\&D Program of China (2018YFA0800901, 2018YFA0506902), The Strategic Priority Research Program of the Chinese Academy of Sciences (XDA12030211), National Natural Science Foundation of China (31671226, 31871194). Art works used for illustrations in Figure 4 are royalty-free images legally purchased from Dreamstime.com; and credited to the following individual artwork creators ID 16071279 © Diamondimages; ID 19279074 © Alila07; ID 27780974 © Skypixel; ID 33479194 © Andegraund548 ; ID 73591906 © Burlesck.

Figure Legends Figure 1. Metabolomics and lipidomics are distinct fields of omics demarcated by their target analytes. The relative distribution of metabolites in human plasma is shown (Quehenberger and Dennis, 2011). Data were compiled from Lentner, Wishart et al., and Quehenberger et al. (Lentner, 1981; Quehenberger et al., 2010; Wishart et al., 2009). Figure 2. Typical workflow in metabolomics and lipidomics. Lipids and metabolites are extracted from the source material (i.e. tissue, cell and fluids) using corresponding solvent system. After proper sample processing of the crude extract, metabolites are analyzed with direct infusion (Shotgun) in MS or MS coupled with chromatography. Lipidomics/metabolomics analysis can be conducted using non-targeted or targeted approaches. Data-mining using bioinformatics and computational tools helps identify perturbed cellular processes and molecular pathways that are altered in experimental or disease condition. Figure 3. Schematic summary of phospholipid biosynthesis and the major phospholipid classes with their associated diseases. The abbreviated mechanisms are illustrated. CDP-DAG, CDP-diacylglycerol; PGP, Phosphatidylglycerophosphate; PG, Phosphatidylglycerol. Figure 4. Combination of a series of metabolomics and lipidomics studies facilitated the discovery of a gut-to-cardiac axis in instigating atherosclerosis. Choline is released from dietary intake of phosphatidylcholines. The trimethyl functional group of choline is then converted into TMA by gut micobes. Liver flavin-containing monooxygenase (FMO) subsequently captures TMA and converts it into TMAO. Release of TMAO into the blood stream promotes cholesterol absorption by macrophages, and reduces bile acid production that represents a critical avenue of cholesterol excretion. These cellular processes lead to enhanced cholesterol accumulation that promotes the formation of cholesterol-laden macrophages that ultimately give rise to atherosclerotic plaques. Figure 5. Advantages of integrating metabolomics and lipidomics. Integration of metabolomics and lipidomics brings about several advantages by providing a complete of molecular changes and global signature, enabling comprehensive network analysis to identify critical metabolic drivers in disease pathology, revealing changes in interconnected metabolic pathways and reinforcing biomarker panel for diagnosis and prognosis. Figure 6. A general schema showing the strategies for integrating omics data. Table 1. Inter-connections between lipids and other metabolites in various diseases. Table 2. Abbreviations.

Related diseases Connections Reference/s Lipids \& Atherosclerosis Elevated FFA inhibit the utilization of Abdel-aleem glucose Type 2 diabetes glucose to generate ATP. et al., 1996; Atherosclerosis Lipids \& Elevated levels of triacylglycerol-rich Lee et amino acids Type 2 diabetes VLDL is responsible for impaired al.,2017 glucose tolerance. Lichtenstein Lipids \& TCA CAD and Schwab, cycle Non-alcoholic fatty Enhanced myocardial SAA metabolism 2000; Wang was shown to culminate in chronic et al., 2012 liver disease oxidative stress mediated by CoA-SH Lee et al., Glioma activation of fatty acyls, which in turn 2017; afflicts overall cellular glucose tolerance. Stamatikos In obesity and over-nutrition, lipid and Paton, substrates inhibit amino acid catabolism in adipose tissue, while the resultant 2013 elevated circulating BCAA subsequently contributes to lower fatty acid oxidation Newgard et in skeletal muscle and liver. al., 2009 Because of, the transportation of fatty acids for oxidation is impeded. Increased Shah et al., FFA and acylcarnitines in patients due 2009 to impaired FA transportation to mitochondria Sunny et al., FFA surplus stimulates TCA cycle. 2011 Elevated hepatic TAG content was found to be associated with increasing flux of Yan et al., the TCA cycle. TCA cycle is not impeded 2009; Zhou for the oxidation of acetyl-CoA generated et al., 2019 from lipid oxidation. Mutations in IDH1 disrupt TCA cycle and produce 2-HG. Elevation in 2-HG and decrease in TAG was found in IDH mutant glioma tissue. Table 1. Inter-connections between lipids and other metabolites in various diseases.

AD Abbreviations APP ATPs Alzheimer’s disease BCAA Amyloid precursor protein BCKD Adenosine triphosphates CCA Branched chain amino acids CL Branched-chain alpha-ketoacid dehydrogenase complex COX Canonical correlation analysis CSF Cardiolipin CVD Cycloxygenase CYP7A1 Cerebrospinal fluid DAG Cardiovascular diseases DDA Cholesterol 7α-hydroxylase gene DHA Diacylglycerols DIA Data dependent acquisition ER Docosahexaenoic acid FA Data independent acquisition FFAs Endoplasmic reticulum HD Fatty acid HILIC Free fatty acids IDA Huntington’s disease IL-6 Hydrophilic interaction liquid chromatography IR Information dependent acquisition LC-MS Interleukin-6 LDL Insulin resistance LOX Liquid chromatography coupled to mass spectrometry Lp-PLA2 Low density lipoproteins Lyso-PA Lipoxygenase Lyso-PC Lipoprotein-associated phospholipase A2 Lyso-PI Lysophospholipid acid MFA Lysophosphatidylcholine MMSE Lyso phosphatidylinositol MS Multiple factor analysis MuS Mini-Mental State Examination NMF Mass spectrometry NMR Multiple sclerosis NOD Nonnegative matrix factorization PA Nuclear magnetic resonance PC Non-obese diabetic PCA Phosphatidic acid PD Phosphatidylcholine PDKs Principal component analysis PE Parkinson’s disease PEBP4 PtdIns-dependent kinases PH Phosphatidylethanolamine PI Phosphatidylethanolamine-binding protein 4 PIPs Pleckstrin homology PLs Phosphatidylinositol PLS Phosphatidylinositol polyphosphates pPC Phospholipids pPE Partial least square PS Plasmalogen phosphatidylcholine PUFA Plasmalogen phosphatidylethanolamine rCCA Phosphatidylserine RF Polyunsaturated fatty acids Regularized canonical correlation Random forest

RP Reversed-phase SERCA Sarcoplasmic reticulum calcium ATPase SM Sphingomyelin SR Sarcoplasmic reticulum sPCA Sparse principal component analysis sPLS Sparse partial least square SVZ Subventricular zone T1DM Type 1 diabetes mellitus T2DM Type 2 diabetes mellitus TAG Triacylglycerols TCA Tricarboxylic acid TMA Trimethylamine TMAO Trimethylamine-N-oxide TRPC6 Transient receptor potential cation channel 6 VLC-SMs Very-long-chain sphingomyelins VLDL Very-low-density lipoprotein Table 2. Abbreviations.

References Abdel-aleem, S., Nadab, M.A., Sayed-Ahmeda, M., Hendricksona, S.C., Louisa, J.S., Walthalla, H.P., Lowea, J.E., 1996. Regulation of fatty acid oxidation by acetyl-CoA generated from glucose utilization in isolated myocytes. Mol. Cell. Cardiol. 28, 825–833. Abdel-Latif, A., Heron, P.M., Morris, A.J., Smyth, S.S., 2015. Lysophospholipids in coronary artery and chronic ischemic heart disease. Curr. Opin. Lipidol. 26, 432-437. Acharjee, A., Ament, Z., West, J.A., Stanley, E., Griffin, J.L., 2016. Integration of metabolomics, lipidomics and clinical data using a machine learning method. BMC Bioinformatics 17. Akerele, O.A., Cheema, S.K., 2015. Fatty acyl composition of lysophosphatidylcholine is important in atherosclerosis. Med. Hypotheses 85, 754-760. Alcalay, R.N., Levy, O.A., Waters, C.C., Fahn, S., Ford, B., Kuo, S.H., Mazzoni, P., Pauciulo, M.W., Nichols, W.C., Gan-Or, Z., Rouleau, G., Chung, W., Wolf, P., Oliva, P., Keutzer, J., Zhang, X., 2015. Glucocerebrosidase activity in Parkinson's disease with and without GBA mutations. Brain 138, 2648-2658. Au, A., 2018. Metabolomics and lipidomics of ischemic stroke. Adv. Clin. Chem. 85, 31-69. Au, A., Cheng, K.K., Wei, L.K., 2016. Metabolomics, lipidomics and pharmacometabolomics of human hypertension. Adv. Exp. Med. Biol. 956, 599-613. Barupal, D.K., Haldiya, P.K., Wohlgemuth, G., Kind, T., Kothari, S.L., Pinkerton, K.E., Fiehn, O., 2012. MetaMapp: mapping and visualizing metabolomic data by integrating information from biochemical pathways and chemical and mass spectral similarity. BMC Bioinformatics 13, 99. Batch, B.C., Hyland, K., Svetkey, L.P., 2014. Branch chain amino acids: biomarkers of health and disease. Curr. Opin. Clin. Nutr. Metab. Care 17, 86-89. Belanger, M., Allaman, I., Magistretti, P.J., 2011. Brain energy metabolism: focus on astrocyteneuron metabolic cooperation. Cell Metab. 14, 724–738. Beloribi-Djefaflia, S., Vasseur, S., Guillaumond, F., 2016. Lipid metabolic reprogramming in cancer cells. Oncogenesis 5, e189. Berthet, C., Castillo, X., Magistretti, P.J., Hirt, L., 2012. New evidence of neuroprotection by lactate after transient focal cerebral ischaemia: extended benefit after intracerebroventricular injection and efficacy of intravenous administration. Cerebrovasc. Dis. 34 329–335. Boon, J., Hoy, A. J., Stark, R., Brown, R. D., Meex, R. C., Henstridge, D. C., Schenk, S., Meikle, P.J., Horowitz, J.F., Kingwell, B.A., Bruce, C.R., Wattet, M.J., 2013. Ceramides contained in LDL are elevated in type 2 diabetes and promote inflammation and skeletal muscle insulin resistance. Diabetes 62, 401–410. Broeckling, C.D., Prenni, J.E., 2018. Stacked injections of biphasic extractions for improved metabolomic coverage and sample throughput. Anal. Chem. 90, 1147-1153. Brosche, T., Platt,D., 1998. The biological significance of plasmalogens in defense against oxidative damage. Exp. Gerontol. 33, 363–369. Chao, H.W., Chao, S.W., Lin, H., Ku, H.C., Cheng, C.F., 2019. Homeostasis of glucose and lipid in non-alcoholic fatty liver disease. Int. J. Mol. Sci. 20, 298. Chen, Y., Guillemin, G.J., 2009. Kynurenine pathway metabolites in humans: disease and healthy states. Int. J. Tryptophan Res: IJTR 2, 1–19. Chen, Y., Ma, Z., Shen, X., Li, L., Zhong, J., Min, L.S., Xu, L., Li, H., Zhang, J., Dai, L., 2018. Serum lipidomics profiling to identify biomarkers for non-small cell lung cancer. Biomed. Res. Int. 2018, 5276240. Colombini, M., 2017. Ceramide channels and mitochondrial outer membrane permeability. J. Bioenerg. Biomembr. 49, 57-64. Cicalini, I., Rossi, C., Pieragostino, D., Agnifili, L., Mastropasqua, L., Di Ioia, M., De Luca, G., Onofrj, M., Federici, L., Del Boccio, P., 2019. Integrated lipidomics and metabolomics analysis of tears in multiple sclerosis: An insight into diagnostic potential of lacrimal fluid. Int. J. Mol. Sci. 20. Coman, C., Solari, F.A., Hentschel, A., 2016. Simultaneous metabolite, protein, lipid extraction (SIMPLEX): a combinatorial multimolecular omics approach for systems biology. Mol. Cell Proteomics 15, 1453-1466. Curtis, M.A., Penney, E.B., Pearson, A.G., van Roon-Mom, W.M., Butterworth, N.J., Dragunow, M., Connor, B., Faull, R.L., 2003. Increased cell proliferation and neurogenesis in the adult human Huntington's disease brain. Proc. Natl. Acad. Sci. USA 100, 9023-9027. Davies, M.J., 1996. Stability and instability: two faces of coronary atherosclerosis. The Paul Dudley White Lecture 1995. Circulation 94, 2013-2020. de Mello, V.D., Lankinen, M., Schwab, U., Kolehmainen, M., Lehto, S., Seppanen-Laakso, T., Oresic, M., Pulkkinen, L., Uusitupa, M., Erkkila, A.T., 2009. Link between plasma ceramides, inflammation and insulin resistance: association with serum IL-6 concentration in patients with coronary heart disease. Diabetologia 52, 2612-2615. Dettmer, K., Aronov, P.A., Hammock, B.D., 2007. Mass spectrometry-based metabolomics. Mass Spectrom. Rev. 26, 51-78. Di Paolo, G., Kim, T.W., 2011. Linking lipids to Alzheimer's disease: cholesterol and beyond. Nat. Rev. Neurosci. 12, 284-296.

Diaz, M., Fabelo, N., Martin, V., Ferrer, I., Gomez, T., Marin, R., 2015. Biophysical alterations in lipid rafts from human cerebral cortex associate with increased BACE1/AβPP interaction in early stages of Alzheimer's disease. J. Alzheimers Dis. 43, 1185-1198. Epand, R.M., Fuller, N., Rand, R. P., 1996. Role of the position of unsaturation on the phase behavior and intrinsic curvature of phosphatidylethanolamines. Biophys. J. 71, 1806–1810. Erion, D.M., Shulman, G.I., 2010. Diacylglycerol-mediated insulin resistance. Nat. Med. 16, 400-402. Fabelo, N., Martin, V., Marin, R., Moreno, D., Ferrer, I., Diaz, M., 2014. Altered lipid composition in cortical lipid rafts occurs at early stages of sporadic Alzheimer's disease and facilitates APP/BACE1 interactions. Neurobiol. Aging 35, 1801-1812. Fahrmann, J., Grapov, D., Yang, J., Hammock, B., Fiehn, O., Bell, G.I., Hara, M., 2015. Systemic alterations in the metabolome of diabetic NOD mice delineate increased oxidative stress accompanied by reduced inflammation and hypertriglyceremia. Am. J. Physiol. Endocrinol. Metab. 308, E978-989. Fahy, E., Cotter, D., Sud, M., 2011. Subramaniam, S. Lipid classification, structures and tools. Biochim. Biophys. Acta 1811, 637-647. Fahy, E., Subramaniam, S., Brown, H.A., Glass, C.K., Merrill, A.H., Jr., Murphy, R.C., Raetz, C.R., Russell, D.W., Seyama, Y., Shaw, W., Shimizu, T., Spener, F., van Meer, G., VanNieuwenhze, M.S., White, S.H., Witztum, J.L., Dennis, E.A., 2005. A comprehensive classification system for lipids. J. lipid Res. 46, 839-861. Farrokhi Yekta, R., Rezaie Tavirani, M., Arefi Oskouie, A., Mohajeri-Tehrani, M.R., Soroush, A.R., 2017. The metabolomics and lipidomics window into thyroid cancer research. Biomarkers 22, 595-603. Fiehn, O., 2002. Metabolomics - the link between genotypes and phenotypes. Plant Mol. Biol. 48, 155-171. Fiehn, O., Garvey, W.T., Newman, J.W., Lok, K.H., Hoppel, C.L., Adams, S.H., 2010. Plasma metabolomic profiles reflective of glucose homeostasis in non-diabetic and type 2 diabetic obese African-American women. PLoS One 5, e15234. Finkelstein, J.D., 1990. Methionine metabolism in mammals. J. Nutr. Biochem 1, 228–237. Floegel, A., Stefan, N., Yu, Z., Mühlenbruch, K., Drogan, D., Joost, H.-G., Fritsche, A., Häring, H.-U., de Angelis, M.H., Peters, A., Roden, M., Prehn, C., Wang-Sattler, R., Illig, T., Schulze, M.B., Jerzy Adamski, J., Boeing, H., Pischonet, T., 2013. Identifcation of serum metabolites associated with risk of type 2 diabetes using a targeted metabolomic approach. Diabetes 62, 639–648. Futerman, A.H., Hannun, Y.A., 2004. The complex life of simple sphingolipids. EMBO Rep. 5, 777-782. Gaglio, D., Metallo, C.M., Gameiro, P.A., Hiller, K., Danna, L.S, Balestrieri, C., Alberghina, L., Stephanopoulos, G., Chiaradonna, F., 2011. Oncogenic K-Ras decouples glucose and glutamine metabolism to support cancer cell growth. Mol. Syst. Biol. 7, 523. Gassama-Diagne, A., Yu, W., ter Beest, M., Martin-Belmonte, F., Kierbel, A., Engel, J., Mostov, K., 2006. Phosphatidylinositol-3,4,5-trisphosphate regulates the formation of the basolateral plasma membrane in epithelial cells. Nat. Cell Biol. 8, 963-970. Gogna, N., Krishna, M., Oommen, A.M., Dorai, K., 2015. Investigating correlations in the altered metabolic profiles of obese and diabetic subjects in a South Indian Asian population using an NMR-based metabolomic approach. Mol. BioSyst. 11, 595-606. Gonzalez-Dominguez, R., Garcia-Barrera, T., Gomez-Ariza, J.L., 2014. Metabolomic study of lipids in serum for biomarker discovery in Alzheimer's disease using direct infusion mass spectrometry. J. Pharm. Biomed. Anal. 98, 321- 326. Grey, M., Dunning, C.J., Gaspar, R., Grey, C., Brundin, P., Sparr, E., Linse, S., 2015. Acceleration of alpha-synuclein aggregation by exosomes. J. Biol. Chem. 290, 2969-2982. Guo, S., Wang, Y., Zhou, D., Li, Z., 2014. Significantly increased monounsaturated lipids relative to polyunsaturated lipids in six types of cancer microenvironment are observed by mass spectrometry imaging. Sci. Rep. 4, 5959. Gupta, S., Natarajan, R., Payne, S.G., Studer, E.J., Spiegel, S., Dent, P., Hylemon, P.B., 2004. Deoxycholic acid activates the c-Jun N-terminal kinase pathway via FAS receptor activation in primary hepatocytes. Role of acidic sphingomyelinase-mediated ceramide generation in FAS receptor activation. J. Biol. Chem. 279, 5821-5828. Hall, Z., Ament, Z., Wilson, C.H., Burkhart, D.L., Ashmore, T., Koulman, A., Littlewood, T., Evan, G.I., Griffin, J.L., 2016. Myc expression drives aberrant lipid metabolism in lung cancer. Cancer Res. 76, 4608-4618. Han, X., Gross, R.W., 2005. Shotgun lipidomics: Electrospray ionization mass spectrometric analysis and quantitation of cellular lipidomes directly from crude extracts of biological samples. Mass Spectrom. Rev. 24, 367–412. Han, X., Holtzman, D., McKeel, D.W.Jr., 2001. Plasmalogen deficiency in early Alzheimer's disease subjects and in animal models: molecular characterization using electrospray ionization mass spectrometry. J. Neurochem. 77, 1168– 1180. Han, X., Rozen, S., Boyle, S.H, Hellegers, C., Cheng, H., Burke, J.R., Welsh-Bohmer, K.A., Doraiswamy, P.M., Kaddurah-Daouk, R., 2011. Metabolomics in early Alzheimer's disease: identification of altered plasma sphingolipidome using shotgun lipidomics. PLoS One 6, e21643. Havel, R.J., Eder, H.A., Bragdon, J.H., 1955. The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J. Clin. Invest. 34, 1345-1353. Hunter, M., Demarais, N.J., Faull, R.L.M., Grey, A.C., Curtis, M.A., 2018. Subventricular zone lipidomic architecture loss in Huntington's disease. J. Neurochem. 146, 613-630.

Ibanez, C., Simo, C., Martin-Alvarez, P.J., Kivipelto, M., Winblad, B., Cedazo-Minguez, A., Cifuentes, A., 2012. Toward a predictive model of Alzheimer's disease progression using capillary electrophoresis-mass spectrometry metabolomics. Anal. Chem. 84, 8532–8540. Ilka Pinz, M.P.R., Sergey Ryzhov, Douglas B., 2017. Sawyer, and Calvin Vary. A novel lipidomic approach to understand human diabetic heart disease. The FASEB J. 31, No. 1_supplement. Incardona, J.P., Eaton, S., 2000. Cholesterol in signal transduction. Curr. Opin. Cell Biol. 12, 193–203. Isabel Cuartero, M., de la Parra, J., García-Culebras, A., Ballesteros, I., Lizasoain, I., Angeles Moro, M., 2016. The kynurenine pathway in the acute and chronic phases of cerebral ischemia. Curr. Pharm. Des. 22, 1060–1073. Jelenik, T., Roden, M., 2013. Mitochondrial plasticity in obesity and diabetes mellitus. Antioxid. Redox Signaling 19, 258-268. Kaddurah-Daouk, R., Zhu, H., Sharma, S., Bogdanov, M., Rozen, S.G., Matson, W., Oki, N.O., Motsinger-Reif, A.A., Churchill1, E., Lei, Z., Appleby, D., Kling, M.A., Trojanowski, J.Q., Doraiswamy, P.M., Arnold, S.E., Pharmacometabolomics Research Network, 2013. Alterations in metabolic pathways and networks in Alzheimer's disease. Transl. Psychiatry 3, e244. Katsoulieris, E., Mabley, J.G., Samai, M., Green, I.C., Chatterjee, P.K., 2009. alpha-Linolenic acid protects renal cells against palmitic acid lipotoxicity via inhibition of endoplasmic reticulum stress. Eur. J. Pharmacol. 623, 107-112. Kiebish, M.A., Han, X., Cheng, H., Chuang, J.H., Seyfried, T.N., 2008. Cardiolipin and electron transport chain abnormalities in mouse brain tumor mitochondria: lipidomic evidence supporting the Warburg theory of cancer. J. Lipid Res. 49, 2545-2556. Kita, Y., Yoshida, K., Tokuoka, S.M., Hamano, F., Yamazaki, M., Sakimura, K., Kano, M., Shimizu, T., 2015. Fever is mediated by conversion of endocannabinoid 2-arachidonoylglycerol to prostaglandin E2. PLoS One 10, e0133663. Klein, M.S., Shearer, J., 2016. Metabolomics and type 2 diabetes: Translating basic research into clinical application. J. Diabetes Res. 2016, 3898502. Klein, J., 2000. Membrane breakdown in acute and chronic neurodegeneration: focus on choline-containing phospholipids. J. Neural Transm. 107, pp1027–1063. Koeberle, A., Shindou, H., Koeberle, S. C., Laufer, S. A., Shimizu, T., Werz, O., 2013. Arachidonoyl- phosphatidylcholine oscillates during the cell cycle and counteracts proliferation by suppressing Akt membrane binding. Proc. Natl. Acad. Sci. USA 110, 2546–2551. Kostic, I., Fidalgo-Carvalho, I., Aday, S., Vazao, H., Carvalheiro, T., Graos, M., Duarte, A., Cardoso, C., Goncalves, L., Carvalho, L., Paiva, A., Ferreira, L., 2015. Lysophosphatidic acid enhances survival of human CD34 (+) cells in ischemic conditions. Sci. Rep. 5, 16406. Koves, T.R., Ussher, J.R., Noland, R.C., Slentz, D., Mosedale, M., Ilkayeva, O., Bain, J., Stevens, R., Dyck, J.R., Newgard, C.B., Lopaschuk, G.D., Muoio, D.M., 2008. Mitochondrial overload and incomplete fatty acid oxidation contribute to skeletal muscle insulin resistance. Cell Metab. 7, 45-56. Kulkarni, H., Meikle, P.J., Mamtani, M., Weir, J.M., Barlow, C.K., Jowett, J.B., Bellis, C., Dyer, T.D., Johnson, M.P., Rainwater, D.L., Almasy, L., Mahaney, M.C., Comuzzie, A.G., Blangero, J., Curran, J.E., 2013. Plasma lipidomic profile signature of hypertension in Mexican American families: specific role of diacylglycerols. Hypertension 62, 621- 626. La Torre, D., Seppänen-Laakso, T., Larsson, H.E., Hyötyläinen, T., Ivarsson, S.A., Lernmark, A., Oresic, M., the DiPiS study group, 2013. Decreased cord-blood phospholipids in young age at onset type 1 diabetes. Diabetes 62, 3951-3956. Labreuche, J., Deplanque, D., Touboul, P.J., Bruckert, E., Amarenco, P., 2010. Association between change in plasma triglyceride levels and risk of stroke and carotid atherosclerosis: systematic review and meta-regression analysis. Atherosclerosis 212, 9-15. Labreuche, J., Touboul, P.J., Amarenco, P., 2009. Plasma triglyceride levels and risk of stroke and carotid atherosclerosis: a systematic review of the epidemiological studies. Atherosclerosis 203, 331-345. Lai, T.W., Shyu, W.C., Wang, Y.T., 2011. Stroke intervention pathways: NMDA receptors and beyond. Trends Mol. Med. 17, 266-275. Lai, T.W., Zhang, S., Wang, Y.T., 2014. Excitotoxicity and stroke: identifying novel targets for neuroprotection. Prog. Neurobiol. 115, 157-188. Lam, S.M., Chua, G.H., Li, X.J., Su, B., Shui, G., 2016. Biological relevance of fatty acyl heterogeneity to the neural membrane dynamics of rhesus macaques during normative aging. Oncotarget 7, 55970-55989. Lam, S.M., Shui, G., 2013. Lipidomics as a principal tool for advancing biomedical research. J. Genet. Genomics 40, 375-390. Lam, S.M., Tian, H., Shui, G., 2017a. Lipidomics, en route to accurate quantitation. Biochim. Biophys. Acta, Mol. Cell Biol. Lipids 1862, 752-761. Lam, S.M., Tong, L., Duan, X., Petznick, A., Wenk, M.R., Shui, G., 2014. Extensive characterization of human tear fluid collected using different techniques unravels the presence of novel lipid amphiphiles. J. Lipid Res. 55, 289-298. Lam, S.M., Wang, R., Miao, H., Li, B., Shui, G., 2018. An integrated method for direct interrogation of sphingolipid homeostasis in the heart and brain tissues of mice through postnatal development up to reproductive senescence. Anal. Chim. Acta 1037, 152-158. Lam, S.M., Wang, Y., Duan, X., Wenk, M.R., Kalaria, R.N., Chen, C.P., Lai, M.K., Shui, G., 2014. Brain lipidomes of subcortical ischemic vascular dementia and mixed dementia. Neurobiol. Aging 35, 2369-2381.

Lam, S.M., Wang, Y., Li, B., Du, J., Shui, G., 2017b. Metabolomics through the lens of precision cardiovascular medicine. J. Genet. Genomics 44, 127-138. Lam, S.M., Wang, Z., Li, J., Huang, X., Shui, G., 2017c. Sequestration of polyunsaturated fatty acids in membrane phospholipids of Caenorhabditis elegans dauer larva attenuates eicosanoid biosynthesis for prolonged survival. Redox Bio. 12, 967-977. Langfelder, P., Horvath, S., 2008. WGCNA: An R package for weighted correlation network analysis. BMC Bioinformatics 9. Lee, J., Jung, S., Kim, N., Shin, M.J., Ryu, D.H., Hwang, G.S., 2017. Myocardial metabolic alterations in mice with diet-induced atherosclerosis: linking sulfur amino acid and lipid metabolism. Sci. Rep. 7, 13597. Lentner, C., 1981. Physical chemistry, composition of blood, hematology, somatometric data, in: Lentner, C. (Eds.), Geigy scientific tables. Ciba-Geigy, New Jersey, pp. 359. Li, D., Wei, W., Ran, X., Yu, J., Li, H., Zhao, L., Zeng, H., Cao, Y., Zeng, Z., Wan, Z., 2017. Lipoprotein-associated phospholipase A2 and risks of coronary heart disease and ischemic stroke in the general population: A systematic review and meta-analysis. Clin. Chim. Acta 471, 38-45. Li, L., Ren, W., Kong, H., Zhao, C., Zhao, X., Lin, X., Lu, X., Xu, G., 2017. An alignment algorithm for LC-MS-based metabolomics dataset assisted by MS/MS information. Anal. Chim. Acta. 990, 96-102. Li, Y., Soos, T.J., Li, X., Wu, J., Degennaro, M., Sun, X., Littman, D.R., Birnbaum, M.J., Polakiewicz, R.D., 2004. Protein kinase C Theta inhibits insulin signaling by phosphorylating IRS1 at Ser(1101). J. Biol. Chem. 279, 45304- 45307. Lichtenstein, A.H., Schwab, U.S., 2000. Relationship of dietary fat to glucose metabolism. Atherosclerosis 150, 227– 243. Lien, L.M., Chiou, H.Y., Yeh, H.L., Chiu, S.Y., Jeng, J.S., Lin, H.J., Hu, C.J., Hsieh, F.I., Wei, Y.H., 2017. Significant association between low mitochondrial DNA content in peripheral blood leukocytes and ischemic stroke. J. Am. Heart Assoc. 6, e006157. Lipton, J.M., Clark, W.G., 1986. Neurotransmitters in temperature control. Annu. Rev. Physiol. 48, 613-623. Liu, C., Zong, W., Zhang, A., Zhang, H., Luan, Y., Sun, H., Cao, H., Wang, X., 2018. Lipidomic characterisation discovery for coronary heart disease diagnosis based on high-throughput ultra-performance liquid chromatography and mass spectrometry. RSC Advances 8, 647-654. Liu, X., Xu, G., 2018. Recent advances in using mass spectrometry for mitochondrial metabolomics and lipidomics - A review. Anal. Chim. Acta 1037, 3–12. Liu, Y.Q., Montanya, E., Leahy, J., 2001. Increased islet DNA synthesis and glucose-derived lipid and amino acid production in association with beta-cell hyperproliferation in normoglycaemic 60% pancreatectomy rats. Diabetologia 44, 1026-1033. Lu, J., Lam, S.M., Wan, Q., Shi, L., Huo, Y., Chen, L., Tang, X., Li, B., Wu, X., Peng, K., Li, M., Wang, S., Xu, Y., Xu, M., Bi, Y., Ning, G., Shui, G., Wang, W., 2019. High-coverage targeted lipidomics reveals novel serum lipid predictors and lipid pathway dysregulation antecedent to type 2 diabetes onset in normoglycemic Chinese adults. Diabetes Care dc190100. Maehama,T., Taylor, G.S., Dixon, J. E., 2001. PTEN and myotubularin: novel phosphoinositide phosphatases. Annu. Rev. Biochem. 70, 247-279. Malmezat, T., Breuillé, D., Pouyet, C., Buffière, C., Denis, P., Mirand, P.P., Obled, C., 2000. Methionine transsulfuration is increased during sepsis in rats. Am. J. Physiol. Endocrinol. Metab. 279, E1391–1397. Mapstone, M., Cheema, A.K., Fiandaca, M.S., Zhong, X., Mhyre, T.R., MacArthur, L.H., Hall, W.J., Fisher, S.G., Peterson, D.R., Haley, J.M., Nazar, M.D., Rich, S.A., Berlau, D.J., Peltz, C.B., Tan, M.T., Kawas, C.H., Federoff, H.J., 2014. Plasma phospholipids identify antecedent memory impairment in older adults. Nat. Med. 20, 415–418. Marien, E., Meister, M., Muley, T., Fieuws, S., Bordel, S., Derua, R., Spraggins, J., Van de Plas, R., Dehairs, J., Wouters, J., Bagadi, M., Dienemann, H., Thomas, M., Schnabel, P.A., Caprioli, R.M., Waelkens, E., Swinnen, J.V., 2015. Non-small cell lung cancer is characterized by dramatic changes in phospholipid profiles. Int. J. Cancer 137, 1539-1548. Martin, V., Fabelo, N., Santpere, G., Puig, B., Marin, R., Ferrer, I., Diaz, M., 2010. Lipid alterations in lipid rafts from Alzheimer's disease human brain cortex. J. Alzheimers Dis. 19, 489-502. Meikle, P.J., Wong, G., Barlow, C.K., Kingwell, B.A., 2014. Lipidomics: potential role in risk prediction and therapeutic monitoring for diabetes and cardiovascular disease. Pharmacol. Ther. 143, 12-23. Meng, C., Zeleznik, O.A., Thallinger, G.G., Kuster, B., Gholami, A.M., Culhane, A.C., 2016. Dimension reduction techniques for the integrative analysis of multi-omics data. Briefings Bioinf. 17, 628-641. Mergenthaler, P., Lindauer, U., Dienel, G.A., Meisel, A., 2013. Sugar for the brain: the role of glucose in physiological and pathological brain function. Trends Neurosci. 36, 587–597. Mollinedo, F., Gajate, C., 2015. Lipid rafts as major platforms for signaling regulation in cancer. Adv. Biol. Regul. 57, 130-146. Monteiro-Cardoso, V.F., Oliveira, M.M., Melo, T., Domingues, M.R., Moreira, P.I., Ferreiro, E., Peixoto, F., Videira, R.A., 2015. Cardiolipin profile changes are associated to the early synaptic mitochondrial dysfunction in Alzheimer's disease. J. Alzheimer's Dis. 43, 1375-1392.

Moore, S.C., Matthews, C.E., Sampson, J.N., Stolzenberg-Solomon, R.Z., Zheng, W., Cai, Q., Tan, Y.T., Chow, W.H., Ji, B.T., Liu, D.K., Xiao, Q., Boca, S.M., Leitzmann, M.F., Yang, G., Xiang, Y.B., Sinha, R., Shu, X.O., Cross, A.J., 2014. Human metabolic correlates of body mass index. Metabolomics 10, 259-269. Murfitt, S.A., Zaccone, P., Wang, X., Acharjee, A., Sawyer, Y., Koulman, A., Roberts, L.D., Cooke, A., Griffin, J.L., 2018. Metabolomics and lipidomics study of mouse models of type 1 diabetes highlights divergent metabolism in purine and tryptophan metabolism prior to disease onset. J. Proteome Res. 17, 946-960. Murphy, K.E., Gysbers, A.M., Abbott, S.K., Tayebi, N., Kim, W.S., Sidransky, E., Cooper, A., Garner, B., Halliday, G.M., 2014. Reduced glucocerebrosidase is associated with increased alpha-synuclein in sporadic Parkinson's disease. Brain 137, 834-848. Nagata, C., Nakamura, K., Wada, K., Tsuji, M., Tamai, Y., Kawachi, T., 2013. Branched-chain amino acid intake and the risk of diabetes in a Japanese community: the Takayama study. Am. J. Epidemiol. 178, 1226-1232. Naudi, A., Cabre, R., Jove, M., Ayala, V., Gonzalo, H., Portero-Otin, M., Ferrer, I., Pamplona, R., 2015. Lipidomics of human brain aging and Alzheimer's disease pathology. Int. Rev. Neurobiol. 122, 133-189. Newgard, C.B., 2012. Interplay between lipids and branched-chain amino acids in development of insulin resistance. Cell Metab. 15, 606-614. Newgard,C.B., An, J., Bain, J.R., Muehlbauer, M.J., Stevens, R.D., Lien, L.F., Haqq, A.M., Shah, S.H., Arolotto, M., Slentz, C.A., Rochon, J., Gallup, D., Ilkayeva, O., Wenner, B.R., Yancy, W.E, Eisenson, H., Musante, G., Surwit, R., Millington, D.S., Butler, M.D., Svetkey, L.P., 2009. A branched-chain amino acid-related metabolic signature that differentiates obese and lean humans and contibutes to insulin resistance. Cell Metab. 9, 311–326. Nieman, K.M., Kenny, H.A., Penicka, C.V., Ladanyi, A., Buell-Gutbrod, R., Zillhardt, M.R., Romero, I.L., Carey, M.S., Mills, G.B., Hotamisligil, G.S., Yamada, S.D., Peter, M.E., Gwin, K., Lengyelet, E., 2011. Adipocytes promote ovarian cancer metastasis and provide energy for rapid tumor growth. Nat. Med. 17, 1498-1503. Obeid, R., Awwad, H.M., Rabagny, Y., Graeber, S., Herrmann, W., Geisel, J., 2016. Plasma trimethylamine N-oxide concentration is associated with choline, phospholipids, and methyl metabolism. Am. J. Clin. Nutr. 103, 703-711. Oresic, M., Simell, S., Sysi-Aho, M., Nanto-Salonen, K., Seppanen-Laakso, T., Parikka, V., Katajamaa, M., Hekkala, A., Mattila, I., Keskinen, P., Yetukuri, L., Reinikainen, A., Lähde, J., Suortti, T., Hakalax, J., Simell, T., Hyöty, H., Veijola, R., Ilonen, J., Lahesmaa, R., Knip, M., Simell, O., 2008. Dysregulation of lipid and amino acid metabolism precedes islet autoimmunity in children who later progress to type 1 diabetes. J. Exp. Med. 205, 2975-2984. Overgaard, A.J., Weir, J.M., De Souza, D.P., Tull, D., Haase, C., Meikle, P.J., Pociot, F., 2016. Lipidomic and metabolomic characterization of a genetically modified mouse model of the early stages of human type 1 diabetes pathogenesis. Metabolomics 12, 13. Padberg, I., Peter, E., Gonzalez-Maldonado, S., Witt, H., Mueller, M., Weis, T., Bethan, B., Liebenberg, V., Wiemer, J., Katus, H.A., Rein, D., Schatz, P., 2014. A new metabolomic signature in type-2 diabetes mellitus and its pathophysiology. PLoS One 9, e85082. Paglia, G., Stocchero, M., Cacciatore, S., Lai, S., Angel, P., Alam, M.T., Keller, M., Ralser, M., Astarita, G., 2016. Unbiased metabolomic investigation of Alzheimer's disease brain points to dysregulation of mitochondrial aspartate metabolism. J. Proteome Res. 15, 608-618. Perry, R.J., Samuel, V.T., Petersen, K.F., Shulman, G.I., 2014. The role of hepatic lipids in hepatic insulin resistance and type 2 diabetes. Nature 510, 84-91. Pieragostino, D., D'Alessandro, M., di Ioia, M., Rossi, C., Zucchelli, M., Urbani, A., Di Ilio, C., Lugaresi, A., Sacchetta, P., Del Boccio, P., 2015. An integrated metabolomics approach for the research of new cerebrospinal fluid biomarkers of multiple sclerosis. Mol. BioSyst. 11, 1563-1572. Qi, Y., Jiang, C., Cheng, J., Krausz, K.W., Li, T., Ferrell, J.M., Gonzalez, F.J., Chiang, J.Y., 2015. Bile acid signaling in lipid metabolism: metabolomic and lipidomic analysis of lipid and bile acid markers linked to anti-obesity and anti- diabetes in mice. Biochim. Biophys. Acta 1851, 19-29. Qin, L., Zhang, Z., Guo, M., Zhang, Q., Wang, Q., Lu, Z., Zhao, H., Liu, Y., Fu, S., Wang, M., Gao, X., 2016. Plasma metabolomics combined with lipidomics profiling reveals the potential antipyretic mechanisms of Qingkailing injection in a rat model. Chem. Biol. Interact. 254, 24-33. Proia, R.L., 2004. Gangliosides help stabilize the brain. Nat. Genet. 36, 1147–1148. Quehenberger, O., Dennis, E.A., 2011. The human plasma lipidome. N. Engl. J. Med. 365, 1812-1823. Quehenberger, O., Armando, A.M., Brown, A.H., Milne, S.B., Myers, D.S., Merrill, A.H., Bandyopadhyay, S., Jones, K.N, Kelly, S., Shaner, R.L, Sullards, C.M., Wang, E., Murphy, R.C., Barkley, R.M., Leiker, T.J., Raetz, C.R.H., Guan, Z., Laird, G.M., Six, D.A., Russell, D.W., McDonald, J.G., Subramaniam, S., Fahy, E., Dennis, E.A., 2010. Lipidomics reveals a remarkable diversity of lipids in human plasma. J. Lipid Res. 51, 3299-3305. Raichur, S., Wang, S.T., Chan, P.W., Li, Y., Ching, J., Chaurasia, B., Dogra, S., Ohman, M.K., Takeda, K., Sugii, Pewzner-Jung, Y., Futerman, A.H., Summers, S.A., 2014. CerS2 haploinsufficiency inhibits beta-oxidation and confers susceptibility to diet-induced steatohepatitis and insulin resistance. Cell Metab. 20, 687-695. Rodrigues, B., Cam, M.C., McNeill, J.H., 1998. Metabolic disturbances in diabetic cardiomyopathy. Mol. Cell. Biochem. 180, 53–57. Rohart, F., Gautier, B., Singh, A., Le Cao, K.A., 2017. mixOmics: An R package for 'omics feature selection and multiple data integration. PLoS Comput. Biol. 13, e1005752.

Rysman, E., Brusselmans, K., Scheys, K., Timmermans, L., Derua, R., Munck, S., Van Veldhoven, P.P., Waltregny, D., Daniels, V.W., Machiels, J., Vanderhoydonc, F., Smans, K., Waelkens, E., Verhoeven, G., Swinnen, J.V., 2010. De novo lipogenesis protects cancer cells from free radicals and chemotherapeutics by promoting membrane lipid saturation. Cancer Res. 70, 8117-8126. Sas, K.M., Nair, V., Byun, J., Kayampilly, P., Zhang, H., Saha, J., Brosius, F.C., 3rd, Kretzler, M., Pennathur, S., 2015. Targeted Lipidomic and Transcriptomic Analysis Identifies Dysregulated Renal Ceramide Metabolism in a Mouse Model of Diabetic Kidney Disease. J. Proteomics Bioinf. Suppl 14, 002. Schmelzer, K., Fahy, E., Subramaniam, S., Dennis, E.A., 2007. The Lipid Maps Initiative in Lipidomics. Methods Enzymol. 432, 171–183. Schwaiger, M., Schoeny, H., El Abiead, Y., Hermann, G., Rampler, E., Koellensperger, G., 2019. Merging metabolomics and lipidomics into one analytical run. Analyst 144, 220–229. Sekhar, R.V., McKay, S.V., Patel, S.G., Guthikonda, A.P., Reddy, V.T., Balasubramanyam, A., Jahoor, F., 2011. Glutathione synthesis is diminished in patients with uncontrolled diabetes and restored by dietary supplementation with cysteine and glycine. Diabetes Care 34, 162-167. Shah, S.H., Hauser, E.R., Bain, J.R., Muehlbauer, M.J., Haynes, C., Stevens, R.D., Wenner, B.R., Dowdy, Z.E., Granger, C.B., Ginsburg, G.S., Newgard, C.B., Krauset, W.E., 2009. High heritability of metabolomic profiles in families burdened with premature cardiovascular disease. Mol. Syst. Biol. 5, 258. Shui, G., Lam, S.M., Stebbins, J., Kusunoki, J., Duan, X., Li, B., Cheong, W.F., Soon, D., Kelly, R.P., Wenk, M.R., 2013. Polar lipid derangements in type 2 diabetes mellitus: potential pathological relevance of fatty acyl heterogeneity in sphingolipids. Metabolomics 9, 786-799. Simpson, B.N., Kim, M., Chuang, Y.F., Beason-Held, L., Kitner-Triolo, M., Kraut, M., Lirette, S.T., Windham, B.G., Griswold, M.E., Legido-Quigley, C., Thambisetty, M., 2016. Blood metabolite markers of cognitive performance and brain function in aging. J. Cerebr. Blood F. Met. 36, 1212–1223. Singh, S., Schwarz, K., Horowitz, J., Frenneaux, M., 2014. Cardiac energetic impairment in heart disease and the potential role of metabolic modulators: a review for clinicians. Circ. Cardiovasc. Genet. 7, 720–728. Smith, R., Mathis, A.D., Ventura, D., Prince, J.T., 2014. Proteomics, lipidomics, metabolomics: A mass spectrometry tutorial from a computer scientist’s point of view. BMC Bioinformatics 15. Soltow,Q.A., Jones,D., Promislow, D.E., 2010. A network perspective on metabolism and aging. Integr. Comp. Biol. 50, 844–854. Spears, L.D., Razani, B., Funai, K., Feng, C., Song, H.W., Semenkovich, C.F., 2018. Lipogenesis Regulates the Response of Cardiac Muscle to Ischemic Stress Through Sarcoplasmic Reticulum Calcium Atpase. Artherioscl. Throm. Vas. 37, A539. Spijkers, L.J., van den Akker, R.F., Janssen, B.J., Debets, J.J., De Mey, J.G., Stroes, E.S., van den Born, B.J., Wijesinghe, D.S., Chalfant, C.E., MacAleese, L., Eijkel, G.B., Heeren, R.M.A., Alewijnse, A.E., Peters, S.L.M., 2011. Hypertension is associated with marked alterations in sphingolipid biology: a potential role for ceramide. PLoS One 6, e21817. Stahlman, M., Pham, H.T., Adiels, M., Mitchell, T.W., Blanksby, S.J., Fagerberg, B., Ekroos, K., Boren, J., 2012. Clinical dyslipidaemia is associated with changes in the lipid composition and inflammatory properties of apolipoprotein-B-containing lipoproteins from women with type 2 diabetes. Diabetologia 55, pp1156–1166. Stamatikos, A.D., Paton, C.M., 2013. Role of stearoyl-CoA desaturase-1 in skeletal muscle function and metabolism. Am. J. Physiol. Endocrinol. Metab. 305, E767–775. Staubach, S., Hanisch, F.G., 2011. Lipid rafts: signaling and sorting platforms of cells and their roles in cancer. Expert Rev. Proteomics 8, 263-277. Stegemann, C., Pechlaner, R., Willeit, P., Langley, S.R., Mangino, M., Mayr, U., Menni, C., Moayyeri, A., Santer, P., Rungger, G., Spector, T.D., Willeit, J., Kiechl, S., Mayr, M., 2014. Lipidomics profiling and risk of cardiovascular disease in the prospective population-based Bruneck study. Circulation 129, 1821-1831. Sunny, N.E., Parks, E.J., Browning, J.D., Burgess, S.C., 2011. Excessive hepatic mitochondrial TCA cycle and gluconeogenesis in humans with nonalcoholic fatty liver disease. Cell Metab. 14, 804-810. Storch, K.J., Wagner, D. A., Young, V. R., 1991. Methionine kinetics in adult men: efects of dietary betaine on L- [2H3-methyl-1-13C] methionine. Am. J. Clin. Nutr. 54, 386–394. Sysi-Aho, M., Ermolov, A., Gopalacharyulu, P.V., Tripathi, A., Seppänen-Laakso, T., Maukonen, J., Mattila, I., Ruohonen, S.T., Vähätalo, L., Yetukuri, L., Härkönen, T., Lindfors, E., Nikkilä, J., Ilonen, J., Simell, O., Saarela, M., Knip, M., Kaski, S., Savontaus, E., Orešič, M., 2011. Metabolic regulation in progression to autoimmune diabetes. PLoS Comput. Biol. 7, e1002257. T Trushina, E., Dutta, T., Persson, X.M., Mielke, M.M., Petersen, R.C., 2013. Identification of altered metabolic pathways in plasma and CSF in mild cognitive impairment and Alzheimer's disease using metabolomics. PLoS One 8, e63644. Tagami, S., Inokuchi Ji, J., Kabayama, K., Yoshimura, H., Kitamura, F., Uemura, S., Ogawa, C., Ishii, A., Saito, M., Ohtsuka, Y., Sakaue, S., Igarashiet, Y., 2002. Ganglioside GM3 participates in the pathological conditions of insulin resistance. J. Biol. Chem. 277, 3085-3092.

Taltavull, N., Ras, R., Marine, S., Romeu, M., Giralt, M., Mendez, L., Medina, I., Ramos-Romero, S., Torres, J.L., Nogues, M.R., 2016. Protective effects of fish oil on pre-diabetes: a lipidomic analysis of liver ceramides in rats. Food Funct. 7, 3981-3988. Tian, H., Lam, S.M., Shui, G., 2016. Metabolomics, a powerful tool for agricultural research. Int. J. Mol. Sci. 17, 1871. Toledo, J.B., Arnold, M., Kastenmuller, G., Chang, R., Baillie, R.A., Han, X., Thambisetty, M., Tenenbaum, J.D., Suhre, K., Thompson, J.W., John-Williams, L.S., MahmoudianDehkordim, S., Rotroff, D.M., Jack, J.R., Motsinger- Reif, Risacherno, S.L., Blach, C., Lucasp, J.E., Massaro, T., Louie, G., Zhu, H., Dallmann, G., Klavins, K., Koal, T., Kim, S., Nho, K., Shen, L., Casanova, R., Varma, S., Legido-Quigley, C., Moseley, M.A., Zhu, K., Henrion, M.Y.R., van der Lee, S.J., Harm, A.C., Demirkan, A., Hankemeier,T., van Duijn, C.M., Trojanowski, J.Q., Shaw, L.M., Saykin, A.J., Weiner, M.W., Doraiswamy, P.M., Kaddurah-Daouk, R., 2017. Metabolic network failures in Alzheimer's disease: A biochemical road map. Alzheimer's Dementia 13, 965-984. Nordström, A., O’Maille, G., Qin, C., Siuzdak, G., 2006. Nonlinear data alignment for UPLC-MS and HPLC-MS based metabolomics: Quantitative analysis of endogenous and exogenous metabolites in human serum. Anal. Chem. 78, 3289–3295. Tsukahara, R., Haniu, H., Matsuda, Y., Tsukahara, T., 2014. Heart-type fatty-acid-binding protein (FABP3) is a lysophosphatidic acid-binding protein in human coronary artery endothelial cells. FEBS Open Bio. 4, 947-951. Tu, J., Yin, Y., Xu, M., Wang, R., Zhu, Z.-J., 2017. Absolute quantitative lipidomics reveals lipidome-wide alterations in aging brain. Metabolomics 14, 5. Turpin, S.M., Nicholls, H.T., Willmes, D.M., Mourier, A., Brodesser, S., Wunderlich, C.M., Mauer, J., Xu, E., Hammerschmidt, P., Bronneke, H.S., Trifunovic, A., LoSasso, G., Wunderlich, F.H., Kornfeld, J.-W., Blüher, M., Krönke, M., Brüninget, J.C., 2014. Obesity-induced CerS6-dependent C16:0 ceramide production promotes weight gain and glucose intolerance. Cell Metab. 20, 678-686. Vorkas, P.A., Shalhoub, J., Isaac, G., Want, E.J., Nicholson, J.K., Holmes, E., Davies, A.H., 2015. Metabolic phenotyping of atherosclerotic plaques reveals latent associations between free cholesterol and ceramide metabolism in atherogenesis. J. Proteome Res. 14, 1389-1399. Vreken, P., Valianpour, F., Nijtmans, L.G., Grivell, L.A., Plecko, B., Wanders, R.J., Barth, P.G., 2000. Defective remodeling of cardiolipin and phosphatidylglycerol in Barth syndrome. Biochem. Biophys. Res. Commun. 279, 378- 382. Wang, D., Kong, J., Wu, J., Wang, X., Lai, M., 2017. GC-MS-based metabolomics identifies an amino acid signature of acute ischemic stroke. Neurosci. Lett. 642, 7-13. Wang, J., Stancakova, A., Soininen, P., Kangas, A. J., Paananen, J., Kuusisto, J., Ala‐Korpela, M., Laaksoet, M., 2012. Lipoprotein subclass profiles in individuals with varying degrees of glucose tolerance: a population-based study of 9399 Finnish men. J. Intern. Med. 272, 562–572. Wang, Q., Liu, D., Song, P., Zou, M.H., 2015. Tryptophan-kynurenine pathway is dysregulated in inflammation, and immune activation. Front. Biosci 20, 1116–1143. Wang, S., Zhou, L., Wang, Z., Shi, X., Xu, G., 2017. Simultaneous metabolomics and lipidomics analysis based on novel heart-cutting two-dimensional liquid chromatography-mass spectrometry. Anal. Chim. Acta 966, 34-40. Wang, Y., Deng, X., Hewavitharana, T., Soboloff, J., Gill, D.L., 2008. Stim, ORAI and TRPC channels in the control of calcium entry signals in smooth muscle. Clin. Exp. Pharmacol. P. 35, 1127-1133. Wang, Z., Klipfell, E., Bennett, B.J., Koeth, R., Levison, B.S., Dugar, B., Feldstein, A.E., Britt, E.B., Fu, X., Chung, Y.M., Wu, Y., Schauer, P., Smith, J.D. Allayee, H., Tang, W.H.W., DiDonato, J.A., Lusis, A.J., Hazen, S.L., 2011. Gut flora metabolism of phosphatidylcholine promotes cardiovascular disease. Nature 472, 57-63. Wallner, S., Schmitz, G., 2011. Plasmalogens the neglected regulatory and scavenging lipid species. Chem. Phys. Lipids 164, 573-589. Wishart, D.S., Knox, C., Guo, A.C., Eisner, R., Young, N., Gautam, B., Hau, D.D., Psychogios, N., Dong, E., Bouatra, S., et al., 2009. HMDB: a knowledgebase for the human metabolome. Nucleic Acids Res. 37, D603-610. Wishart, M.J., Dixon, J.E., 2002. PTEN and myotubularin phosphatases: from 3-phosphoinositide dephosphorylation to disease. Trends Cell Biol. 12, 579-585. Wong, P.T., Qu, K., Chimon, G.N., Seah, A.B., Chang, H.M., Wong, M.C., Ng, Y.K., Rumpel, H., Halliwell, B., Chen, C.P., 2006. High plasma cyst (e) ine level may indicate poor clinical outcome in patients with acute stroke: possible involvement of hydrogen sulfide. J. Neuropathol. Exp. Neurol. 65, 109–115. Wood, P.L., Barnette, B.L., Kaye, J.A., Quinn, J.F., Woltjer, R.L., 2015. Non-targeted lipidomics of CSF and frontal cortex grey and white matter in control, mild cognitive impairment, and Alzheimer's disease subjects. Acta neuropsychiatrica 27, 270-278. Wood, P.L., Smith, T., Lane, N., Khan, M.A., Ehrmantraut, G., Goodenowe, D.B., 2011. Oral bioavailability of the ether lipid plasmalogen precursor, PPI-1011, in the rabbit: a new therapeutic strategy for Alzheimer's disease. Lipids Health Dis. 10, 227. Xing, C., Arai, K., Lo, E.H., Hommel, M., 2012. Pathophysiologic cascades in ischemic stroke. International journal of stroke 7, 378-385. Yan, H., Parsons, D.W., Jin, G., McLendon, R., Rasheed, B.A., Yuan, W., Kos, I., Batinic-Haberle, I., Jones, S., Riggins, G.J., Friedman, H., Friedman, A., Reardon, D., Herndon, J., Kinzler, K.W., Velculescu, V.E., Vogelstein, B., Bigner, D.D., 2009. IDH1 and IDH2 mutations in gliomas. N. Engl. J. Med. 360, 765-773.

Yang, L., Li, M., Shan, Y., Shen, S., Bai, Y., Liu, H., 2016. Recent advances in lipidomics for disease research. J. Sep. Sci. 39, 38-50. Yang, L., Lv, P., Ai, W., Li, L., Shen, S., Nie, H., Shan, Y., Bai, Y., Huang, Y., Liu, H., 2017. Lipidomic analysis of plasma in patients with lacunar infarction using normal-phase/reversed-phase two-dimensional liquid chromatography- quadrupole time-of-flight mass spectrometry. Anal. Bioanal. Chem. 409, 3211-3222. Ying, H., Kimmelman, A.C., Lyssiotis, C.A., Hua, S., Chu, G.C., Fletcher-Sananikone, E., Locasale, J.W., Son, J., Zhang, H., Coloff, J.L., Yan, H., Wang, W., Chen, S., Viale, A., Zheng, H., Paik, J., Lim, C., Guimaraes, A.R., Martin, E.S., Chang, J., Hezel, A.F., Perry, S.R., Hu, J., Gan, B., Xiao, Y., Asara, J.M., Weissleder, R., Wang, Y.A., Chin, L., Cantley, L.C., DePinho, R.A., 2012. Oncogenic Kras maintains pancreatic tumors through regulation of anabolic glucose metabolism. Cell Death Differ. 149, 656–670. Yu, G.P., Chen, G.Q., Wu, S., Shen, K., Ji, Y., 2011. The expression of PEBP4 protein in lung squamous cell carcinoma. Tumour biol. 32, 1257-1263. Zeng, I.S.L., Lumley, T., 2018. Review of Statistical Learning Methods in Integrated Omics Studies (An Integrated Information Science). Bioinf. Biol. Insights 12, 1-16. Zha, H., Cai, Y., Yin, Y., Wang, Z., Li, K., Zhu, Z.J., 2018. SWATH to MRM: Development of high-coverage targeted metabolomics method ssing SWATH technology for biomarker discovery. Anal. Chem. 90, 4062-4070. Zhang, A.H., Qiu, S., Xu, H.Y., Sun, H., Wang, X.J., 2014. Metabolomics in diabetes. Clin. Chim. Acta 429, 106-110. Zhang, H., Shao, X., Zhao, H., Li, X., Wei, J., Yang, C., Cai, Z., 2019. Integration of Metabolomics and Lipidomics Reveals Metabolic Mechanisms of Triclosan-Induced Toxicity in Human Hepatocytes. Environ. Sci. Technol. 53, 5406–5415. Zheng, Y., Yu, B., Alexander, D., Couper, D.J., Boerwinkle, E., 2014. Medium-term variability of the human serum metabolome in the Atherosclerosis Risk in Communities (ARIC) study. Omics 18, 364-373. Zhou, L., Wang, Z., Hu, C., Zhang, C., Kovatcheva-Datchary, P., Yu, D., Liu, S., Ren, F., Wang, X., Li, Y., Hou, X., Piao, H., Lu, X., Zhang, Y., Xu, G., 2019. Integrated metabolomics and lipidomics analyses reveal metabolic reprogramming in human glioma with IDH1 mutation. J. Proteome Res. 18, 960-969. Zhou, Z., Shen, X., Chen, X., Tu, J., Xiong, X., Zhu, Z.J., 2018. LipidIMMS Analyzer: Integrating multi-dimensional information to support lipid identification in ion mobility - mass spectrometry based lipidomics. Bioinformatics 35, 698–700.