Application of stable isotopes to investigate the metabolism of fatty acids, glycerophospholipid and sphingolipid species

Progress in Lipid Research 54 (2014) 14–31

Contents lists available at ScienceDirect

Progress in Lipid Research journal homepage: www.elsevier.com/locate/plipres

Review

Application of stable isotopes to investigate the metabolism of fatty acids, glycerophospholipid and sphingolipid species Josef Ecker, Gerhard Liebisch ⇑ Institute of Clinical Chemistry and Laboratory Medicine, University of Regensburg, Germany

a r t i c l e

i n f o

Article history: Received 5 April 2013 Received in revised form 30 December 2013 Accepted 7 January 2014 Available online 24 January 2014 Keywords: Lipid biosynthesis Lipid metabolism Metabolic profiling Stable isotope labeling Mass spectrometry Lipidomics

a b s t r a c t Nature provides an enormous diversity of lipid molecules that originate from various pathways. To gain insight into the metabolism and dynamics of lipid species, the application of stable isotope-labeled tracers combined with mass spectrometric analysis represents a perfect tool. This review provides an overview of strategies to track fatty acid, glycerophospholipid, and sphingolipid metabolism. In particular, the selection of stable isotope-labeled precursors and their mass spectrometric analysis is discussed. Furthermore, examples of metabolic studies that were performed in cell culture, animal and clinical experiments are presented. Ó 2014 Elsevier Ltd. All rights reserved.

Contents 1. 2.

3.

4.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fatty acids [FA]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Biosynthesis and metabolism of FA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Mass isotopomer distribution analysis (MIDA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Studies of FA metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4. Comparison of labeling strategies and their analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Glycerophospholipids [GP] . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Biosynthesis and metabolism of GP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Studies of GP metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Comparison of labeling strategies and their analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sphingolipids [SL]. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Biosynthesis and metabolism of SL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Studies of SL metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Comparison of labeling strategies and their analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

15 15 15 16 16 19 20 20 20 23 25 25 26 27

Abbreviations: BMP, bis(monoacylglycero)phosphate; Cer, ceramide; CoA, Coenzyme A; Des, desaturation; DG, diacylglycerol; DihCer, dihydroceramide; DiSPH, dihydrosphingosine (sphinganine); DMDS, dimethyl disulfide; ESI-MS/MS, electrospray ionization tandem mass spectrometry; FA, fatty acid; FADS, fatty acid desaturase; FAME, fatty acid methyl ester; FASN, fatty acid synthase; FT-ICR, Fourier transform ion cyclotron resonance; GC–IRMS, gas chromatography isotope ratio mass spectrometry; GlcCer, glucosylceramide; GP, glycerophospholipid; HILIC, hydrophilic interaction chromatography; IRMS, isotope ratio mass spectrometry; IS, internal standard; LacCer, lactosylceramides; LC, liquid chromatography; LOD, limit of detection; LPA, lysophosphatidic acid; LPA, lysophosphatidic acid; MIDA, mass isotopomer distribution analysis; MTAD, 4-methyl-1,2,4-triazoline-3,5-dione; NL, neutral loss; NP, normal phase; PA, phosphatidic acid; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PEMT, phosphatidylethanolamine-N-methyltransferase; PFB, pentafluorobenzyl; PG, phosphatidylglycerol; PGP, phosphatidylglycerolphosphate; PI, phosphatidylinositol; PIS, precursor ion scan; PS, phosphatidylserine; RP, reversed phase; S1P, sphingosine-1-phosphate; SCD, stearoyl-CoA desaturase; SL, sphingolipid; SM, sphingomyelin; SMase, sphingomyelinase; SPC, sphingosylphosphorylcholine; SPH, sphingosine; SPP, S1P phosphatase; SPT, serine palmitoyl transferase. ⇑ Corresponding author. Address: Institute of Clinical Chemistry and Laboratory Medicine, University of Regensburg, D-93042 Regensburg, Germany. Tel.: +49 941 944 6240; fax: +49 941 944 6202. E-mail address: [email protected] (G. Liebisch). http://dx.doi.org/10.1016/j.plipres.2014.01.002 0163-7827/Ó 2014 Elsevier Ltd. All rights reserved.

J. Ecker, G. Liebisch / Progress in Lipid Research 54 (2014) 14–31

5.

6.

General considerations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1. Selection of label . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. Consideration of isotopic overlap and data analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3. Nomenclature of lipid species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions and outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction Traditionally, radioactive precursors have been applied to investigate lipid metabolism. During the past decade, numerous studies have demonstrated that stable isotopes can replace radioactive reagents for safety reasons and because mass spectrometric analysis can be used to provide data regarding the lipid species. Today, lipidomic technologies are able to cover almost the full lipidome [1–4]. A combination of these methods with stable isotope labeling is perfectly suitable for analyzing the metabolism of lipid species. This knowledge is of great importance because the reasons why nature provides such an incredible diversity of lipid species are not well understood [5]. In this review, we present an overview of studies that used stable isotopes to analyze the metabolism of fatty acids (FA), glycerophospholipids (GP) and sphingolipids (SL). Labeling strategies and mass spectrometric analysis are discussed. 2. Fatty acids [FA] 2.1. Biosynthesis and metabolism of FA FAs are key modules for various lipids, including cell membrane lipids such as GPs or triacylglycerides, which are the major components of lipid droplets [4]. Excess lipids or defects in lipid storage are associated with diseases such as metabolic syndrome. Moreover, FAs are the precursors of eicosanoids, which are potent

15

28 28 28 28 28 29

signaling molecules with inflammatory and anti-inflammatory effects [6]. For numerous cellular processes, such as cell growth or differentiation, as well as for key cellular functions, FAs must be synthesized de novo, primarily as building blocks for cell membrane generation [7]. As indicated in Fig. 1, in mammals, FAs are synthesized de novo from acetyl-CoA by FA synthase (FASN) to yield palmitate (FA 16:0) [8]. Palmitate can either be desaturated to palmitoleate (FA 16:1n-9) by stearoyl-CoA desaturase (SCD1) or elongated by an elongase (ELOVL6) to stearate (FA 18:0). Stearate can be further desaturated to oleate (FA 18:1n-9) by SCD1 [9,10]; palmitoleate can be elongated by ELOVL6 to yield vaccinate (FA 18:1n-7). The precursor for polyunsaturated FAs (PUFA) are linoleic acid (FA 18:2n-6) and alpha-linolenic acid (FA 18:3n-3) [6]. Both are essential FAs obtained from the diet in mammals. Unlike plants, mammals do not contain delta-12 and delta-15 desaturases, which are necessary to desaturate FA 18:1 and FA 18:2 [11]. Linoleic acid (n-6) metabolites include arachidonic acid (FA 20:4n-6) (Fig. 1), whereas alpha-linolenic metabolites (n-3) include eicosapentaenoic acid (EPA, FA 20:5n-3) and docosahexaenoic acid (DHA, FA 22:6n-3). Importantly, the metabolism of n-3 and n-6 PUFAs shares the same series of enzymes. In addition to endogenous metabolism, arachidonic acid, as well as EPA and DHA, can also be obtained directly from the diet. Fish is a typical source of n-3 FAs, such as EPA and DHA, whereas westernized diets primarily contain n-6 FAs, including linoleic acid and arachidonic acid [12]. Whereas arachidonic acid is the key precursor for inflammatory eicosanoids,

Fig. 1. Labeling of fatty acid synthesis and metabolism. The pathways of human endogenous fatty acid synthesis and PUFA metabolism are displayed. The stable isotopelabeled precursors (compare Table 1) are indicated in green, red and pink squares. The fatty acids are indicated in blue squares. CoA, Coenzyme A; Des., desaturation; Elo., elongation; ELOVL, Fatty acid elongase; FA, fatty acid; FADS, fatty acid desaturase; FASN, fatty acid synthase; SCD, Stearoyl-CoA desaturase.

16

J. Ecker, G. Liebisch / Progress in Lipid Research 54 (2014) 14–31

EPA and DHA are precursors for anti-inflammatory lipid mediators, including resolvins [6,13,14]. In mammals, FA synthesis and metabolism are tightly controlled by the nuclear transcription factor sterol regulatory element binding protein (SREBP) 1c, which regulates various genes that are involved in FA synthesis and metabolism, such as FASN and SCD 1 [15,16]. 2.2. Mass isotopomer distribution analysis (MIDA) FA synthesis can be determined by the incorporation of stable isotope-labeled polar precursors into FA 16:0 (Fig. 1). The label enrichment in palmitic acid is typically analyzed by mass isotopomer distribution analysis (MIDA) (see [17,18] for reviews). Mass isotopomers (Fig. 2, M0-M8) share the same nominal mass (independent from the isotope atoms, e.g., D or 13C, and position) and can be differentiated by mass spectrometry (usually by a quadrupole mass analyzer). Notably, ultra-high resolution mass spectrometry may resolve mass isotopomers that are derived from D and 13C (see Section 5.2). In a simplified view, MIDA regards FA synthesis as the polymerization of C2-units (acetyl-CoA) (Fig. 2). For instance, when 1-13Cacetate is supplied to cells, a maximum of eight labels can be incorporated into palmitate (FA 16:0). The probability of label incorporation depends on the proportion of the label in the substrate pool. Therefore, the mass isotopomer profile of the newly synthesized polymer is determined by the label enrichment of the substrate pool. Higher enrichment will result in more highly labeled products and vice versa. It is important to realize that the newly synthesized FA is diluted by preexisting unlabeled FA. To analyze the enrichment of the substrate pool, it is necessary to correct the measured mass isotopomer profile by the natural abundance. The enrichment of the precursor pool can be calculated from the corrected mass isotopomer profile. Another prerequisite for MIDA is the incorporation of at least two precursor molecules into the analyte; otherwise, it is not possible to assess the likelihood of labeling. In addition to FA synthesis, MIDA has been applied for methyl incorporation in the PEMT pathway [19] or fully labeled FA incorporation into triglycerides [20]. 2.3. Studies of FA metabolism Selected applications of stable isotopes to analyze FA synthesis and metabolism are summarized in Table 1. (a) Using polar precursors A well-established approach to investigate de novo FA synthesis is the administration of stable isotope-labeled polar precursors, such as 13C-labeled acetate, glucose or heavy water (D2O), and the analysis of label enrichment in FA 16:0 (Fig. 1). When acetate enters the cell, acetate is conjugated with CoA by acetyl-CoA synthase to yield acetyl-CoA, which is the basic substrate for FA synthesis [21]. Glucose enters the glycolytic pathway to generate pyruvate and ATP in the cytoplasm of cells [22]. Pyruvate can be further converted in mitochondria to acetyl-CoA by pyruvate dehydrogenase, which enters the citric acid cycle to yield citrate. Citrate can be fully oxidized to generate ATP by oxidative phosphorylation or can be transported back to the cytoplasm, where ATP citrate lyase catalyzes the formation of acetyl-CoA, which enters FA synthesis. Heavy water labels NADPH, which can be incorporated into de novo synthesized lipids [23]. The application of D2O as a probe to investigate the effects of dietary changes on FA metabolism in mice showed that, upon 70% calorie restriction, adipose whole body FA oxidation is dramatically increased to maintain health and survival [24]. The increase in FA degradation is balanced by induced mRNA expression of FASN, which leads to enhanced de novo FA synthesis. The fraction of newly synthesized FAs upon D2O labeling was measured by GC–MS of

FA methyl esters (FAMEs) and MIDA. Paradoxically, high fat feeding was also reported to increase the expression of SREBP-1c and target genes, such as FASN, SCD and ELOVL6 (see Fig. 1 for explanation), which suggests enhanced FA synthesis and metabolism [25,26]. For this reason, it was hypothesized that in addition to lipids that were obtained from a high fat diet, oleic acid in particular is synthesized de novo in the liver with high fat feeding. However, the infusion of mice with labeled acetate and the subsequent analysis of FAs as pentafluorobenzyl (PFB) derivatives showed that elevated hepatic oleic acid levels are due to increased hepatic elongation and to the desaturation of pre-existing palmitate rather than to de novo synthesized palmitate [25]. Enhanced FA elongation and desaturation that were induced by a high fat diet could be reversed by supplementation of the animals with fish oil. We previously found that macrophage colony-stimulating factor (MCSF)-dependent differentiation of primary human monocytes strongly induces SREBP-1c and its target genes [27]. The supplementation of cells with 1-13C-labeled acetate led to a massive label enrichment in palmitate after four days of cell differentiation. 13 C-enrichment in FA 16:0 was investigated by the analysis of mass isotopomers using GC–MS after FA methylation [28]. When FASN or SREBP-1c expression was suppressed using inhibitors or RNAi, 13 C-enrichment in palmitate and, accordingly, FA synthesis were decreased. With additional studies, we could show that the suppression of FA synthesis prevents phagocytosis, which represents the central macrophage function, as well as the development of filopodia and cellular organelles, which include the endoplasmic reticulum and Golgi network. In an elegant approach, Collins and colleagues demonstrated that the application of different stable isotope precursors allows a detailed investigation into FA synthesis precursor pools [29]. Adipocytes were supplied with 13C-labeled acetate, glucose, pyruvate or glutamine to determine the major precursors for de novo lipid synthesis during adipocyte differentiation. It is well-known that lipogenesis is upregulated during adipocyte differentiation and that triacylglycerols (TG), particularly those TGs that contain palmitate and oleate, accumulate. Glucose was found to be the major source for FA synthesis; however, other sources, including amino acids such as glutamine, incorporate into TG-palmitate. FAMEs were measured using GC–MS after the separation of the TG fraction, and tracer incorporation was analyzed by MIDA. In addition to studies on mammals or mammalian cells, labeling strategies to profile lipid metabolism have also been applied to other species, such as nematodes. Perez et al. [30] fed a mixture of unlabeled and 13C-labeled Escherichia coli and determined the isotope pattern of FA 16:0 using GC–MS of FAMEs. These authors could show that the majority of palmitate in Caenorhabditis elegans was derived from the uptake of bacterial palmitate because the major FA 16:0 fraction was either unlabeled or fully labeled from labeled E. coli (M+16). De novo FA synthesis accounted only for less than 10% of total palmitic acid. In addition to GC–MS, liquid chromatography (LC) that is coupled to high-resolution mass spectrometry is a powerful approach for lipid metabolic studies. A recent publication describes an LC method that uses an Orbitrap to track the FA metabolism of mouse kidney epithelial cells that are incubated with [U–13C]-glucose and [U–13C]-glutamine [31]. The authors found that activated Ras in immortalized baby mouse kidney cells leads to the increased generation of saturated and mono-unsaturated VLCFA (26 carbons and more). Importantly, the activation of the Ras signaling pathway is a common event in various aggressive cancer types. In general, FASN specific activity is seen as an indicator of aggressive tumors and poor patient outcome [22,32]. Therefore, Rudolph and colleagues have described an assay to quantify the activity of tissue FASN by tracing the formation of [U–13C]-palmitate as the primary product of labeled acetyl- and malonyl-CoA by GC–MS of PFB

J. Ecker, G. Liebisch / Progress in Lipid Research 54 (2014) 14–31

17

Fig. 2. Mass isotopomer distribution analysis (MIDA). The top left panel shows the natural abundance of mass isotopomers of both the acetate pool and palmitic acid that were analyzed as fatty acid methyl ester (FAME 16:0) by GC–MS. Mass isotopomers, which are displayed in the top right panel, are synthesized from 8 monomers, for example, palmitic acid synthesized from acetate monomers. The unlabeled monomers (grey) contain no isotopes, and the labeled monomers (dark green) contain a single labeled atom, e.g., 13C or deuterium. Mass isotopomers (M0–M8) represent a group of molecules with the same nominal mass independent from the isotopic atom and from the position of the isotope. Label enrichment in the precursor pool of 10% (middle panel) and 50% (lower panel) generates a time-dependent shift of mass isotopomer profiles. The labeling times t1, t2 and t3 represent labeling efficiencies of 10%, 50% and 90% of newly synthesized palmitic acid, respectively. Figure modified according to [18].

derivatives [33]. This assay might be useful for the staging and prognosis of human cancers. (b) Using labeled FA The application of stable isotope-labeled FAs as precursors is a common strategy to track the metabolism of longer chain FAs

(>16 carbons). For instance, the dietary supplementation of human volunteers with stable isotope-labeled FAs is often used to follow and elucidate pathways of postprandial fat deposition or to determine FA flux in plasma [34–36]l . Although many stable isotope-labeled FA species are commercially available, labeled long-chain PUFAs are often expensive when used for in vivo

18

J. Ecker, G. Liebisch / Progress in Lipid Research 54 (2014) 14–31

Table 1 Studies of fatty acid metabolism. Abbreviations for derivatives: DMDS dimethyl disulfide; FAME, fatty acid methyl ester; MTAD, 4-methyl-1,2,4-triazoline-3,5-dione; PFB, pentafluorobenzyl. Analytes

Authors

Experimental system and matrix

Objective

Label

Methods

[U–13C]-FA 16:0, TG-[U–13C]-16:0, D2-FA 16:0, TG-D2-FA 16:0

Bickerton et al. [36]

Human volunteers, plasma

[U–13C]-FA 16:0, D2-FA 16:0

GC–MS, FAME

D-enriched total FA 16:0, FA 18:0, FA 16:1n 9, FA 18:1n 9 13 C-labeled TG-fatty acids, mainly TG-FA 16:0

Bruss et al. [24] Collins et al. [29]

D2O

GC–MS, FAME GC–MS, FAME

13

C-enriched FA 16:0, D3-FA 16:0, D3-FA 16:1, D3-FA 18:0

Ecker, et al. [27]

D3-FA 16:0, D3-FA 16:1, D3-FA 18:0, D3-FA 18:1n 9, D3-FA C18:1n 7

Ecker et al. [42]

13

C-enriched FA 16:0, D3-FA 16:1, D3-FA 18:0 D5/D7- FA 16:0, D5/D7-FA 16:1, D5/D7-FA 18:0, D5/D7-FA 18:1 13 C enriched total VLCFAs (<24 carbons)

Ecker et al. [28] Gagne et al. [44] Kamphorst et al. [31]

Mice, liver and adipose tissue, serum Human primary preadipocytes and differentiated adipocytes Primary monocytes and differentiated macrophages Macrophages, dendritic cells derived from human primary monocytes Primary human macrophages Rats, plasma

Postprandial FA uptake in adipose tissue and skeletal muscle Analysis of fFA oxidation and synthesis De novo lipogenesis during adipocyte differentiation

[U–13C]-PUFAs, mainly [U–13C]-FA 22:6n 3

Le et al. [37]

[U–13C]-FA 18:2n 6, [U–13C]-FA 18:3n 3, D5-FA 18:2n 6, D5-FA 18:3n 3, 13Cand their D5-labeled metabolites 13 C-enriched FA 16:0, FA 18:0, FA 16:1, FA 18:1

Lin et al. [54] Oosterveer et al. [25]

Mice, liver tissue

[1-13C]-FA 18:1n 7 [1-13C]-c9,t11-CLA (C18:2)

Mosley et al. [46]

Cows, plasma and milk

[1-13C]-FA 18:1n 7 [1-13C]-c9,t11-CLA (C18:2)

Mosley et al. [47]

[1-13C]-FA 14:0, [1-13C]-FA C14:1, [1-13C]FA 16:0, [1-13C]-FA 16:1, [1-13C]-FA 18:0, [1-13C]-FA 18:1 D5-FA 18:3 (n 3), D5-FA 20:5 (n 3), D5FA 22:5 (n 3), D5-FA 22:6 (n 3), D5-FA 18:3 (n 3), D5-FA 20:5 (n 3), D5FA 22:5 (n 3), D5-FA 22:6 (n 3), 13 C enriched total FAs, TG-FAs, PL-FAs [U–13C]-FA 16:0 13

[U– C]-FA 16:0 13

C-enriched fatty acids, mainly FA 16:0 to FA 24:0

13

C-enriched and labeled FA 16:0, FA 16:1n 9, FA 18:1n 9

Immortalized baby mouse kidney epithelial cells

FA synthesis and metabolism during phagocytic differentiation of monocytes SCD activity

Method for FA analysis (labeled and unlabeled) Desaturation and oxidation of stearate Effects of Ras activity on VLCFA metabolism

[1-13C]-acetate, [U–13C]glucose, [U–13C]-pyruvate, [U–13C]-glutamine [2-13C]-acetate, D3-FA 16:0

D3-FA 16:0

GC–MS, FAME

[2-13C]-acetate, D3-FA 16:0

GC–MS, FAME LC-ESI-MS

D5-FA 18:0, D7-FA 18:0 [U–13C]-glucose [U–13C]-glutamine

LC-ESI-MS (High resolution) GC–MS, FAME

Biosynthesis of [U–13C]PUFAs

[U–13C]-glucose

Comparison of [U–13C]- and D5-labeled FA analysis and metabolism De novo Lipogenesis, FA elongation and desaturation upon high fat feeding Delta-9 desaturation of vaccenic acid, incorporation into milk fat

[U–13C]-FA 18:2n 6, [U–13C]-FA 18:3n 3, D5-FA 18:2n 6, D5-FA 18:3n 3 [1-13C]-acetate

[1-13C]-FA 18:1n

7

Lactating woman, milk and serum

Delta-9 desaturation of vaccenic acid, incorporation into milk fat

[1-13C]-FA 18:1n

7

Mosley et al. [45]

Cows, milk, plasma and fecal samples

Delta-9 desaturation of FAs, incorporation into milk fat

[1-13C]-FA 14:0, [1-13C]-FA 16:0, [1-13C]-FA 18:0

Pawlosky et al. [52] Pawlosky et al. [51] Perez et al. [30] Persson et al. [35] Rudolph et al. [33] Wong et al. [151]

Human volunteers, plasma Human volunteers, plasma C. elegans

Effects of beef- and fish-based diets on n 3 FA metabolism n 3 FA metabolism in smokers Effects of insulin signaling on lipid synthesis Method development, palmitate flux in plasma FASN activity

D5-FA 18:3 (n

3)

D5-FA 18:3 (n

3)

Yee et al. [43]

HepG2 cells

Marine heterotrophic microorganisms, Hyalochlorella marina Oral administration to rats, plasma

Human volunteers, plasma Recombinant FASN, crude tissue lysate HepG2 cells

labeling. An alternative source has been presented by Le and colleagues [37], who supplied the marine protist Hyalochlorella marina with [U–13C]-glucose to generate 13C-labeled PUFAs. After preparative HPLC, [U–13C]-DHA was obtained at gram-levels with >99% chemical purity and with >90% 13C-isotope purity. Commercially available D3-labeled saturated FAs are used in many studies to analyze the activity of desaturases. In particular, SCD1 (Fig. 1) is of major interest because SCD1 has been associated with many diseases. The inhibition of SCD1 leads to severe atherosclerosis, despite protection from obesity and hepatosteatosis

Peroxysomal beta-oxidation and its contribution to FA synthesis and elongation Effects of SCD1 inhibition on metabolism of LCFA (16 and more carbons)

GC–MS, FAME

GC–MS, PFB GC–MS, PFB

13 C-enriched E. coli (all carbons labeled) [U–13C]-FA 16:0

GC–MS, FAME, DMDS, MTAD GC–MS, FAME, DMDS, MTAD GC–MS, FAME, DMDS, MTAD GC–MS, PFB GC–MS, PFB GC–MS, FAME LC-ESI-MS

[1,2-13C2]-acetyl-CoA, [1,2,3-13C3]-malonyl-CoA [U–13C]-FA 18:0, [U–13C]-FA 16:0

GC–MS, PFB GC–MS, FAME

[U–13C]-FA 18:0, [1,2-13C2]acetate, [U–13C]-FA 16:0

GC–MS, FAME

[38,39]. Enhanced SCD1 expression and elevated levels of phospholipids that contain mono-unsaturated fatty acids have been reported to protect and attenuate ER stress [40,41]. We previously found that macrophages and dendritic cells, which are cell types that originate from a common myeloid precursor, significantly differ in their mono-unsaturated FA content [42]. To investigate whether these different proportions are due to differential FA synthesis capacities, macrophages and dendritic cells that were derived from primary human monocytes were supplied with D3-FA 16:0, and its metabolism was analyzed using GC–MS of FAMEs.

J. Ecker, G. Liebisch / Progress in Lipid Research 54 (2014) 14–31

We found decreased desaturation activity (D3-FA 16:1 generation from D3-FA 16:0) and reduced mRNA and protein expression of SCD1 in dendritic cells compared with macrophages [42]. To explore the effects of different SCD1 inhibitors on the FA metabolism of HepG2 cells, Yee et al. [43] supplied [U–13C]-FA 16:0, [U–13C]-FA 18:0 and [1,2-13C2]-acetate. SCD1 inhibitors differentially affected FA 16:0 and FA 18:0 desaturation, which depended on the origin of the saturated FA (e.g., FA 16:0 that was derived from b-oxidized FA 18:0 was differentially desaturated compared with exogenously supplied FA 16:0). These findings suggest that separate compartments exist for FA desaturation, elongation and chain shortening. Gagne and colleagues developed a fast LC–MS-based method to profile SCD activity and b-oxidation in rats [44]. When D7-FA 18:0 was orally administered to rats, significant levels of the parent compound, as well as D7-FA 18:1 and D7-FA 16:0, were found in plasma for up to 72 h. Using GC–MS, Mosley et al. [45] performed several stable isotope labeling studies with cows to investigate delta-9 desaturation and incorporation into milk lipids. A mixture of 13C-labeled FA 14:0, 16:0 and 18:0 was continuously infused into the duodenum for up to 24 h. After 8 h, the 13C-labeled parent compounds plus remarkable amounts of delta-9-desaturase products were observed in milk fat. When 13C-vaccenate (FA 18:1-t11) was infused for 4 h, the parent compound and its delta-9-desaturation product, 13 C-c9,t11-CLA, were found in milk fat [46]. Similar results were obtained when lactating woman were supplemented with 13Cvaccenate in the diet [47]. After 12 h, 13C-vaccenate and 13Cc9,t11-CLA could be detected in milk fat. c9,t11-CLA is of major interest because this molecule is a bioactive FA that has been reported to exert various beneficial health effects and to inhibit the progression of many diseases, such as cardiovascular and inflammatory diseases or cancer in animal models [48–50]. Similarly, several n 3-PUFAs, such as FA 20:5n 3 (EPA) and FA 22:6n 3 (DHA), are nutrient compounds with beneficial health effects (see the Section 1). In a recent study, the effects of smoking on n 3 FA metabolism were analyzed by the administration of yogurt that was blended with D5-FA 18:3n 3 ethyl ester to human volunteers [51]. FAs were analyzed by GC–MS after PFB derivatization. Smoking increased the amount of D5-FA 18:3 that was transferred into plasma and its metabolism to D5-FA 22:6n 3. Therefore, it was hypothesized that smoking enhances n 3 PUFA synthesis to compensate for the loss from lipid oxidation that was initiated by smoking. In another study from the same group, D5-FA 22:5n 3 to D5-FA 22:6n 3 conversion was much lower in volunteers after consuming a fish-based diet for 3 weeks compared with a beef-based diet [52]. These data suggest a feedback control mechanism that is responsive to the plasma concentration of FA 22:6n 3. 2.4. Comparison of labeling strategies and their analysis The most commonly applied methodology for FA analysis is gas chromatography of FA methyl esters (FAME) [53]. GC, particularly with polar columns, is suitable for the separation of complex FA mixtures (e.g., FA 18:1: cis-9-FA 18:1, trans-9-FA 18:1 cis-11-FA 18:1; 18:3n 3 or n 6). However, a drawback of polar columns is that the temperature limit does not usually permit the analysis of FAMEs with more than 30 carbon atoms. Another limitation of GC–MS with electron impact ionization (EI) (the most frequently used ionization technique) is the generation of fragment-rich spectra with low intensities of molecular ions. Intact molecular ions or, at least, large fragments that contain the label are essential for tracing FA metabolism or for performing MIDA (see Section 2.2). As an alternative, chemical ionization (CI) (e.g., negative chemical ionization of FA-pentafluorobenzyl (PFB) derivatives) leads to less fragmentation and, thereby, higher inten-

19

sities of molecular ions compared with EI. However, a disadvantage might be that CI is a less robust ionization method compared with EI. Concerning signal intensity, high endogenous FA levels lead to signal suppression in GC–MS analysis. Lin et al. [54] analyzed PFB derivatives of FA 18:2n 6 and FA 18:3n 3 by GC-NCI-MS, either D5- or [U–13C]-labeled, with and without a plasma matrix. Signal suppression by high endogenous FA pools was detected for both labels (with higher suppression for 13C-labed FAs). A similar suppressive effect by endogenous FAs has been observed by Patterson and colleagues when GC-EI-MS was applied [55,56]. Recently, a LC electrospray ionization high resolution mass spectrometry (LC–ESI-MS) method was introduced for the analysis of FA metabolism [31]. In contrast to EI and CI, ESI does not require the derivatization of FAs and generates intact molecular ions as a soft ionization technique. This technique is particularly advantageous for the analysis of extremely long chain fatty acids (VLCFA) that contain up to 36 carbon atoms because these species are not usually accessible for GC analysis with polar columns. In combination with high resolution MS, this method provides high specificity to profile FAs, including their isotope profiles. Importantly, LC–MS without high mass resolution may be biased by interfering matrix components, as observed by Gagne et al. [44]. These authors found substantial interference with D5-FA 18:0 and, therefore, changed their label to D7-FA 18:0. Concerning chromatographic resolution, LC is usually inferior compared with GC. To compensate for this limitation, FAs may be analyzed by tandem MS (MS/MS) to gain specificity. However, the collision-induced dissociation (CID) of FAs does not generate significant fragments that are suitable for sensitive and specific MS/MS analysis. As an alternative, ozone-induced [57] or radical-directed dissociation [58] might be applied as promising novel approaches to identify the structure of FAs, including the position(s) of double bonds. A great advantage of polar labels such as D2O or 13C-glucose is that these labels can be easily administered and are safe for use in human volunteers. D2O rapidly equilibrates with the body water, whereas glucose is subject to regulated transport. Moreover, low concentrations of D2O do not influence metabolism. Similar to acetate, which is the actual precursor of FA synthesis, these labels are only incorporated into endogenously synthesized FAs but not in PUFAs (in mammals). Therefore, these tracers are ideal for monitoring lipogenesis. Most commonly, two approaches were used to analyze FAs that were labeled by polar precursors; one is the analysis by GC–MS. As explained in Section 2.2, all these labels generate mass isotopomer profiles in FAs that may by analyzed by GC–MS. Using this technique, it is possible to obtain information regarding the precursor pool labeling by MIDA. The second approach is GC-isotope-ratio mass spectrometry (GC-IRMS), which allows the precise measurement of isotopic enrichment in FAs by combustion, i.e., 13C to 12C or D to H ratios. However, GCIRMS does not permit the profiling of mass isotopomer profiles. Consequently, an assessment of precursor pool labeling by MIDA is not possible but requires an analysis of label enrichment in the precursor pool. Labeled FAs are perfectly suitable as direct tracers of FA transport and metabolism. Moreover, the metabolism of PUFAs can be monitored only by the administration of labeled n 3 and n 6 FA because these FAs are not derived from endogenous precursors (Fig. 1). In contrast to polar labels, FAs must be provided with vehicles such as albumin or dissolved in lipophilic agents [20]. Notably, one must be aware that exogenous FA may influence lipid metabolism. In contrast to 13C-labeling, the positions of D-labels in tracers must be selected carefully because D-labels may be lost during desaturation. Concerning their metabolism and absorption, 13Cand D5-labeled FAs do not show significant differences [54,56].

20

J. Ecker, G. Liebisch / Progress in Lipid Research 54 (2014) 14–31

3. Glycerophospholipids [GP]

3.2. Studies of GP metabolism

3.1. Biosynthesis and metabolism of GP

Selected applications of stable isotopes to analyze the GP synthesis and metabolism are summarized in Table 2. (a) Using polar precursors Because FAs are essential parts of all GP classes, the labeling approaches that were described in the previous chapter may also be applied to trace GP metabolism (Fig. 3). As components of lung surfactant, PC and PG are vital for lung function. To study the kinetics of phospholipids in alveolar surfactant, the Cogo and Carnielli group applied polar precursors, such as D2O or [U–13C]-glucose, for in vivo labeling. Infants received the label intravenously, and tracheal aspirates were analyzed after the purification of PC and PG fractions by GC-IRMS [66–68]. Moreover, labeled FA [69] and PC (label administered together with surfactant [70]), were used to determined lung surfactant synthesis and pool size in infants (see [71] for a review). The major findings of these studies include that de novo synthesis and turnover rates of surfactant are low in infants that suffer from respiratory distress syndrome (RDS) independent of the labeled precursor (see also Table 2). Additionally, these infants had a small pool size and increased half-life when compared with studies in animals. A more selective approach compared with FA labeling is the introduction of labeled head groups into GPs (Fig. 3). This approach is currently the most frequently applied strategy to profile GP metabolism. Pioneering work was performed by DeLong and colleagues [72]. D9-choline and D4-ethanolamine were used to differentiate the molecular species that were derived from the Kennedy and PE-N-methyltransferase (PEMT) pathways in primary rat hepatocytes and rat hepatoma cells. PC and PE species were

Glycerophospholipids (GP) compose the main lipid category of mammalian cell membranes [59]. In addition to their function as building blocks of the lipid bilayer, these lipids play a vital role in cellular functions, including the regulation of transport processes, protein function and signal transduction [59–61]. Additionally, GPs are essential components of lipoproteins and influence their function and metabolism [62,63]. Nature provides enormous structural diversity of GP species [5]. Although the specific functions of these species are not well understood thus far, their particular structures define their biophysical characteristics and, thus, the properties of membranes [7]. The first step of GP biosynthesis (Fig. 3) [64,65] is the transfer of a FA to glycerol-3-phosphate to form lysophosphatidic acid (LPA), which is further acylated to phosphatidic acid (PA). PA, after conversion to CDP-diacylglycerol (CDP-DG), provides the precursor for phosphatidylinositol (PI) and phosphatidylglycerol (PG). Alternatively, diacylglycerol (DG), which is generated by the dephosphorylation of PA, fuels phosphatidylcholine (PC), phosphatidylethanolamine (PE), and phosphatidylserine (PS) synthesis. In the so-called Kennedy pathway, choline/ethanolamine are phosphorylated and converted to CDP-choline/ethanolamine, which are used for head group transfer to form PC and PE, respectively. Both PC and PE are substrates for PS synthesis, which may be converted to PE by decarboxylation in the mitochondria. In hepatocytes, PE can be derived additionally from N-methylation of PC.

Fig. 3. Labeling of glycerophospholipid metabolism. The major pathways of glycerophospholipid (GP) metabolism in mammalian cells are displayed. The stable isotopelabeled precursors (compare Table 2) are indicated in grey squares. GPs are indicated in blue squares: C, choline; DG, diacylglycerol; E, ethanolamine; FA, fatty acid, LPA, lysophosphatidic acid; PA, phosphatidic acid; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PG, phosphatidylglycerol; PGP, phosphatidylglycerolphosphate; PI, phosphatidylinositol; PS, phosphatidylserine. Enzymes in green: CDS, CDP-diacylglycerol synthase; CK, choline kinase; CPT, CDP-choline:1,2-diacylglycerol cholinephosphotransferase; CT, CTP:phosphocholine cytidylyltransferase; DAGK, diacylglycerol kinase; EK, ethanolamine kinase; EPT, CDP-ethanolamine:1,2-diacylglycerol ethanolaminephosphotransferase; ET, CTP:phosphoethanolamine cytidylyltransferase; GPAT, glycerol-3-phosphate acyltransferase; LPAAT, lysophosphatidic acid acyltransferase; PAP, phosphatidic acid phosphatase; PEMT, phosphatidylethanolamine N-methyltransferase; PGPP, phosphatidylglycerophosphate phosphatase; PGPS, phosphatidylglycerophosphate synthase; PIS, phosphatidylinositol synthase; PSD, phosphatidylserine decarboxylase; PSS, phosphatidylserine synthase.

21

J. Ecker, G. Liebisch / Progress in Lipid Research 54 (2014) 14–31 Table 2 Studies of glycerophospholipid metabolism. Analytes

Authors

Experimental system

Objective

Label

Methods

PC

Bernhard et al. [76]

Surfactant metabolism

D9-choline

PC

Besnard et al. [74]

Human – induced sputum Mice

Synthesis of lung PC

D9-choline

PC, PE, PS

Binder et al. [111]

Primary human skin fibroblasts

Effect of cholesterol loading on glycerophospholipid metabolism

D9-choline D4-ethanolamine 13 C3-serine

PC, PE, methylated PE

Boumann et al. [87]

Yeast Microsomes

Selectivity of PEmethylation

D3-methyl-methionine D4-ethanolamine

PC

Boumann et al. [86]

Yeast

D3-methyl-methionine D13-choline

PC, PE

Boumann et al. [88]

Yeast

PC

Bunt et al. [68]

LPC, PC (TG, CE) PC

Castro-Perez et al. [152] Cogo et al. [69]

Infants – tracheal aspirates Mice –plasma

Differentiation of CDPcholine and PEMT derived PC synthesis Species specificity of CDPcholine and –ethanolamine derived PC and PE synthesis Synthesis of lung surfactant in vivo Lipogenesis

ESI–MS/MS D9-PC PIS + 193 ESI–MS/MS D9-PC PIS + 193 ESI–MS/MS D9-PC PIS + 193 D4-PE NL + 145 13 C2-PE NL + 143 13 C3-PS NL + 188 ESI–MS/MS D9-PC PIS + 193 D6-PC PIS + 190 D3-PC PIS + 187 D4-PE NL + 145 D3-methyl-PE NL + 158 ESI–MS/MS D9-PC PIS + 193 D13-PC PIS + 197 ESI–MS/MS D13-PC PIS + 197 D4-PE NL + 145 GC-IRMS of purified PC fractions LC-(Q)TOF-MS

saturated PC

Cogo et al. [66]

PC, PE

DeLong et al. [72]

PC

PC, PE, PS

Infants – tracheal aspirates Infants – tracheal aspirates rat primary hepatocytes and rat hepatoma cells

Synthesis of lung surfactant in vivo Synthesis of lung surfactant in vivo Molecular species of PC and PE synthesis pathways

DeLong et al. [73]

rat primary hepatocytes and rat hepatoma cells

Choline as methyl group donor for the PEMT pathway

Ecker et al. [27]

Primary human monocytes and macrophages

Dynamics of glycerophospholipid species synthesis during macrophage differentiation

D13-choline D4-ethanolamine [U–13C]-glucose D2 O [U-13C]-FA 16:0 [U-13C]-FA 18:2 D2 O D9-choline D4-ethanolamine

D9-choline D4-ethanolamine

D9-choline D4-ethanolamine 13 C3-serine

PC, PE, PS

Ecker et al. [84]

Primary human monocytes derived macrophages

Effect of conjugated linoleic acids on GP metabolism

D9-choline D4-ethanolamine 13 C3-serine

PC (only presented) PC, PE, PI

Lane et al. [114]

MCF7-LCC2 cells

Lipid biosynthesis

[U–13C]-glucose

Hunt et al. [79–81]

Cultured cells

Dynamics and specificity of the endonuclear GP synthesis

D9-choline D4-ethanolamine D6-myo-inositol

PC

Hunt et al. [83]

CHO cells

D9-choline

PC, PE, PS

Kainu et al. [96]

HeLa cells

Species specificity of CEPT1 mediated PC synthesis (overexpression of CEPT1) Trafficking of PE and PS species to mitochondria

PE

Kainu et al. [95]

BHK21 cells

PS

Remodeling of PE and PS species

D9-labeled PC species D4-labeled PE species [D315N]-labeled PS species D9-choline D4-ethanolamine D3-serine D4-labeled PE species D3-labeled PS species

GC-IRMS of purified PC fractions GC-IRMS of purified saturated PC fractions ESI–MS/MS D9-PC PIS + 193 D4-PC PIS + 188 D4-PE NL + 145 ESI–MS/MS only D9-choline: D9-PC PIS + 193 D6-PC PIS + 190 D3-PC PIS + 187 D4-ethanolamine: D4-PC PIS + 188 D9-choline and D4ethanolamine D13-PC PIS + 197 D10-PC PIS + 194 D7-PC PIS + 191 ESI–MS/MS D9-PC PIS + 193 D4-PE NL + 145 13 C2-PE NL + 143 13 C3-PS NL + 188 ESI–MS/MS D9-PC PIS + 193 D4-PE NL + 145 13 C2-PE NL + 143 13 C3-PS NL + 188 FT-ICR-MS ESI–MS/MS D9-PC PIS + 193 D4-PE NL + 145 D6-PI PIS- 247 ESI–MS/MS D9-PC PIS + 193 LC–MS/MS D9-PC PIS + 193 D4-PE NL + 145 D3-PE NL + 144 D3-PS NL + 188 D315N-PS NL + 189 ESI–MS/MS, crude extracts and LC-separated fractions D4-PE NL + 145 D3-PS NL- 90 (continued on next page)

22

J. Ecker, G. Liebisch / Progress in Lipid Research 54 (2014) 14–31

Table 2 (continued) Analytes

Authors

Experimental system

Objective

Label

Methods

PC, PE, PS

Kainu et al. [94]

Cultured cells

Cyclodextrin mediated transfer of labeled GP species

D9-labeled PC species D4-labeled PE species [D315N]-labeled PS species

PC (TG, CE)

McLaren et al. [20,93]

Mice and primates – plasma

[U–13C]-FA 18:1

PI

Postle et al. [78]

PC

Postle et al. [75] Pynn et al. [19]

Kinetics of lung/surfactant PC PC synthesis in vivo, specificity and rate of CDPcholine and PEMT pathway

D9-choline

PC

Cultured cells Mice Mice – lung, broncho alveolar lavage fluid Mice (liver, lung, plasma) and human (plasma)

Lipid assembly and tracking of lipids in lipoprotein fractions PI synthesis

ESI–MS/MS D9-PC PIS + 193 D4-PE NL + 145 D315N-PS NL-91 LC–MS/MS PC PIS + 184

PC (SM, FC, CE)

Schifferer et al. [90]

HDL3-mediated cellular lipid efflux

D9-choline (13C3–FC)

PC, PI, PS

Sewell et al. [153]

Primary human monocyte-derived macrophages Primary human monocytes derived macrophages

GP synthesis in control and Crohn’s disease macrophages

D9-choline D6-myo-inositol D3-serine

PC

Torresin et al. [70] Tserng et al. [91,92] Vedovelli et al. [67]

Kinetics of exogenously administered surfactant Synthesis of DG and PC Phospholipid turnover in alveolar surfactant PS synthesis and decarboxylation

PC [U–13C]-16:0/[U–13C]-16:0

PC (DG) Saturated PC, PG PS

Infants – tracheal aspirates HL60 cells Infants – tracheal aspirates Microsomes of rat brain cortex; liver and brain cortex mitochondria

Wen et al. [97] Kevala et al. [98] Kim et al. [99]

analyzed by direct infusion electrospray ionization tandem mass spectrometry (ESI-MS/MS) of crude lipid extracts. These authors detected more saturated and shorter chain lengths for the PC species that were derived from CDP-choline (Kennedy pathway) compared with longer, more unsaturated species that were derived from PE conversion by PEMT. In a second study, DeLong et al. [73] showed that choline acts as a methyl group donor for the PEMT pathway in primary rat hepatocytes. Interestingly, PE-methylation was disrupted in rat hepatoma cells. These principles of head group labeling were also applied by the Postle group to thoroughly investigate the regulation of lung surfactant phospholipid synthesis (mainly PC) in various mouse models [74,75] and in human volunteers [76] (for a recent review, see [77]). Similarly, this group used D6-myo-inositol to profile the molecular specificity of PI synthesis in cultured cells and in a mouse model by intraperitoneal injection [78]. Mammalian cells were dominated by one species, PI 18:0/20:4, although newly synthesized PI showed a broad range of species. The analysis of synthesis kinetics suggested selective transport and/or hydrolysis of selected species as probable mechanisms for the enrichment of PI 18:0/20:4. Alan Hunt, from the same group, investigated endonuclear phospholipid metabolism in cultured cells [79–81]. He found that endonuclear PC synthesis generates high proportions of saturated and monounsaturated PC compared with whole cell PC species, whereas PUFA-PC species are depleted from endonuclear pools (see [82] for a review). In another study, he showed that the overexpression of 1,2-diacylglycerol-CDP:choline cholinephosphotransferase (CEPT1) in CHO cells leads to increased production of arachidonate-containing PC species [83]. Recently, the Postle group studied PC synthesis pathways in the liver in vivo using a mouse model and human volunteers [19]. The PEMT pathway was shown to preferentially synthesize PUFA-PC species, which is consistent with the data from the DeLong group in cultured hepatocytes [73].

D6-myo-inositol

D9-choline

[13C4]-FA 16:0 D2 O D35-FA 18:0 labeled PC, PE and PS species

ESI–MS/MS D6-PI PIS- 247 ESI–MS/MS D9-PC PIS + 193 ESI–MS/MS D9-PC PIS + 193 D6-PC PIS + 190 D3-PC PIS + 187 ESI–MS/MS D9-PC/SM PIS + 193 (13C3-FC/CE PIS + 372) ESI–MS/MS D9-PC PIS + 193 D6-PI PIS- 247 D3-PS NL-90 GC-IRMS of purified PC fractions GC–MS of diglycerides GC-IRMS of purified PG and saturated PC fractions LC–MS SIM+ of DG fragments (in source fragmentation) of the labeled products

As discussed in Section 2.3, we showed that the induction of FA synthesis is essential for phagocytic differentiation of primary human macrophages [27]. Using head group labeling and direct flow injection ESI-MS/MS, we could observe a tight coupling of FA synthesis to PC and PE synthesis. In another study, we found that the structure of exogenously administered CLA isomers determines not only the pathway these isomers enter but also the kinetics of macrophage GP metabolism [84]. Although c9,t11-CLA incorporates into PC and PE, t9,t11-CLA is preferentially bound to neutral lipids. Additionally, t9,t11-CLA, which is structurally similar to a straight chain FA, did not induce PC synthesis, whereas c9,t11-CLA induced PC synthesis to an even greater degree than linoleic acid. These data fit well with the model that CTP:phosphocholine cytidylyltransferase (CT), which is the key enzyme of PC synthesis, is activated by membrane curvature stress [85]. A series of elegant studies was performed by the de Kroon group to investigate the details of phospholipid metabolism in yeast [86–88]. D3-methyl-methionine, D13-choline and D4-ethanolamine were applied to explore the species specificity of yeast enzymes and their regulation by direct infusion ESI-MS/MS (see [89] for a review). These studies identified PC as a key player in determining the biophysical properties of membranes by the regulation of biosynthesis, turnover, and acyl chain remodeling. Apart from investigations into GP metabolism, stable isotope labeling may be applied to investigate cellular lipid efflux. Our group used D9-choline to quantify the export of cellular PC (and sphingomyelin; see Section 4.2a) species toward apoA-I and HDL3 in primary human monocyte-derived macrophages [90]. The labeling of cellular species is essential for such experiments because HDL3 contains a high concentration of phospholipids and only the labeling of cellular species permits accurate differentiation from HDL phospholipids. These experiments demonstrated a preferential efflux of monounsaturated PC species and suggested different cellular pools for lipid efflux toward apoA-I and HDL3.

J. Ecker, G. Liebisch / Progress in Lipid Research 54 (2014) 14–31

(b) Using labeled FAs Only a few studies have applied labeled FAs to monitor GP synthesis because tracing is limited to the species that contain these FAs or their metabolites. Tserng et al. studied the turnover of PC and DG (and sphingomyelin; see Section 4.2b) in HL60 cells that were supplied with [13C4]-palmitate [91,92]. Labeled palmitate was delivered to the cells as a complex with serum albumin. PC fractions were separated, converted to DG by SMase treatment and isotope incorporation was analyzed by GC–MS analysis of DG trimethylsilyl (TMS) derivatives. McLaren et al. [20,93] studied the incorporation of [U–13C]-FA 18:1 into triglycerides (TG), cholesteryl ester (CE) and PC by UPLC-MS/MS. The kinetics of different label administrations were tested in a study with primates, which included the oral uptake of FAs that were dissolved in oil or that were delivered intravenously as a bolus that was mixed with a fat emulsion (Intralipid) [20]. In another study, the incorporation into lipoprotein fractions was investigated in a mouse model [93]. (c) Using labeled GPs Stable isotope-labeled GPs are excellent tools to study phospholipid remodeling, as presented by Kainu et al. from the Somerharju group. These authors introduced phospholipid species that were labeled on the head group into cultured cells via mediation by methyl-b-cyclodextrin [94] and followed their remodeling by direct infusion ESI-MS/MS [95]. In contrast to major endogenous species, atypical PE and PS species were rapidly remodeled. In another

23

study, these tools were applied together with choline, ethanolamine and serine labeling to investigate the import of PS into mitochondria and the export of its decarboxylation product PE from mitochondria [96]. Interestingly, the efficient import of PS 38:4, together with the slow export of PE 38:4, explained the enrichment of these species in mitochondria. The Kim group presented a set of studies with PC, PE and PS species that were labeled with D35-FA 18:0 to follow either PS synthesis in microsomes [97] [98] or PS decarboxylation in mitochondria [99]. The analysis was performed by LC–MS of DG fragments that were generated by the in-source fragmentation of the labeled PS or PE. 3.3. Comparison of labeling strategies and their analysis Electrospray ionization tandem mass spectrometry (ESI-MS/ MS) has emerged as the method of choice for the characterization, identification and quantification of GP species from biological samples. There are two main approaches that exist, i.e., the direct infusion of crude lipid extracts, which is also termed shotgun lipidomics, and liquid chromatography that is coupled to mass spectrometry (LC–MS/MS) [1,4,13,100–102]. Using ESI, the suppression of ionization by matrix components has to be considered, particularly for direct MS approaches. Therefore, the application of internal standards (ISs) is mandatory for quantification (also for

Fig. 4. Labeling and analysis of PC, PE, PS synthesis and interconversion by ESI-MS/MS. The incorporation of labeled choline, ethanolamine and serine into PC, PE, PS and their tandem mass spectrometric analysis are displayed. The diagnostic head group scans of unlabeled lipids are indicated in grey squares. The monitoring of different labels is indicated by the following colored squares: Bright blue – serine labeling of PS and decarboxylation into PE; red – ethanolamine labeling of PE and methylation to PC; green – choline incorporation into PC; dark blue – monitoring of mono- and dimethyl-PE intermediates of the PEMT reaction. PEMT, phosphatidylethanolamine N-methyltransferase; PSD, phosphatidylserine decarboxylase; NL, neutral loss; PIS, precursor ion scan.

24

J. Ecker, G. Liebisch / Progress in Lipid Research 54 (2014) 14–31

relative comparison) [103,104]. Frequently, non-naturally occurring lipid species are used as ISs. The addition of two ISs for each lipid class, e.g., a low mass and a high mass species, allows a response control of individual samples by the calculation of the ratio of both ISs [105,106]. In the direct infusion analysis, the sensitivity for low abundance species may be limited by matrix effects due to signal suppression at high sample concentrations [107]. Therefore, species with low abundance are often analyzed by interfacing MS/MS with liquid chromatography (LC–MS/MS). LC reduces the matrix effects and increases species identification by the separation of isobaric or quasi-isobaric interference. In our opinion, polar stationary phases are favorable because these phases provide the co-elution of the lipid species of one lipid class and, consequently, also co-elution with their IS [13]. It is well-known that ESI is highly susceptible to matrix effects and that only the co-elution of analytes and ISs ensures accurate compensation for such interference. In contrast, reversed phase chromatography usually separates lipid species by chain length (including analytes and IS) and, thus, may be affected by matrix effects. Both normal phase [108,109] and hydrophilic interaction chromatography [110] are able to differentiate bond types (ester/ether) and fatty acid combinations or isobaric compounds, such as bis(monoacylglycero)phosphate (BMP) and phosphatidylglycerol (PG), respectively. Another approach to differentiate quasi-isobaric species, such as ester and ether species, is high resolution mass spectrometry (see also chapter 5.2) [3]. To quantify stable isotope-labeled species, calibration lines that are generated from unlabeled species may be applied because these species show a similar analytical response [111]. The standard approach to tracing the major pathways of GP metabolism is the introduction of labeled polar head groups (Fig. 4). The analysis by ESI-MS/MS provides diagnostic fragments that result from the phospholipid head groups for both unlabeled

and labeled species. For instance, PC that is labeled by the incorporation of D9-choline generates a fragment of m/z 193 upon CID and can be differentiated from unlabeled PC, which releases a phosphocholine fragment of m/z 184. Similarly, labeled and unlabeled PI exhibit different head group fragments, whereas PE and PS lose their head groups during CID as neutral fragments (accessible by a neutral loss scan). The interconversion of lipid classes is easily traceable over multiple steps. For example, after the incorporation of D4-ethanolamine into PE, PC species that result from that pathway are characterized by a fragment of m/z 188. Moreover, methylated intermediates of the PEMT pathway [87,112] may be labeled by either methionine or methyl group donation from choline [73]. Another strategy to quantify mono- and dimethyl-PE is chemical derivatization with deuterated methyliodide (CD3I) to form different mass-tagged PCs. This approach could be particularly useful for multiple precursor ion scanning on a hybrid quadrupole time-offlight mass spectrometer [113]. The labeling of polar head groups has many advantages compared with other approaches (Fig. 5). The administration of the water soluble GP head group precursor is easy and safe for in vivo studies. The incorporation of head groups as intact labels allows the tracing of all species of a phospholipid class as well as the combination of different labels in one experiment. A disadvantage of head group labeling may result from a small mass shift, e.g., seen for D3-serine. Such labels may be affected by an isotopic overlap from unlabeled species. For example, unlabeled PS 36:1 exhibits a +3 isotope peak with more than 2% of the molecule peak. In particular, for low labeling efficiencies, e.g., after short-term labeling, such inference can be significant, and the correction of the isotopic overlap should be considered. The use of high resolution MS enables the mass spectrometric resolution of labeled and unlabeled species without fragmentation and could be of great advantage for such studies [3,114] (see also Section 5.2).

Fig. 5. Comparison of labeling strategies for the glycerophospholipid metabolism. The main advantages and disadvantages of different label strategies for the GP metabolism are displayed.

J. Ecker, G. Liebisch / Progress in Lipid Research 54 (2014) 14–31

In addition to head group labeling, labeled FAs may be incorporated into GP. The labels and their pros and cons are described in Section 2.4 (Fig. 5). However, their application does not provide a comprehensive picture of the synthesis of one GP class; rather, their application traces the incorporation of FAs into various GP classes. When FAs are labeled using polar precursors, one has to consider that FA labeling generates a complex label pattern in the newly synthesized GPs. Such labeling patterns are difficult to analyze by ESI-MS/MS using head group-specific fragments as described above. Therefore, these experiments usually require laborious approaches, including the purification of lipid classes with subsequent GC–MS or IRMS after the liberation of the FAs. The administration of labeled FAs has the advantage that a defined mass shift is introduced by the label. When selecting the label, it has to be considered that the mass of the labeled GP species may overlap with endogenous species. Therefore, the use of a fully labeled FA, which generates a substantial mass shift, could be advantageous. However, the metabolism of the FA label by elongation, desaturation and degradation may substantially complicate the ESI-MS/MS analysis. Generally, one has to consider that newly synthesized GP species can be rapidly modified by acyl chain remodeling. To gain insight into remodeling processes, it is necessary to monitor shortterm changes in species profiles with high temporal resolution. It is difficult to follow the remodeling by labeling of the fatty acyls of GP species because these labels may be lost during remodeling and because several labeled FAs are required to obtain a comprehensive picture. Although their application requires vehicles such as cyclodextrin-mediated transfer, synthetic GP species that are labeled in the head group are perfect tracers to study acyl chain remodeling [94]. Such labeled GP species remain labeled during FA remodeling and are easily traceable by head group specific

25

scans [95]. Moreover, these tracers can be applied to study the transport of GP species and their conversion into other GP classes (except head group transfer) [96]. 4. Sphingolipids [SL] 4.1. Biosynthesis and metabolism of SL SLs are a diverse class of lipids that are composed of free sphingoid bases and their phosphates, ceramides, and sphingomyelins, as well as complex glycosphingolipids [115,116]. Many of these lipids are well-known signaling molecules that have been implicated in various diseases, including metabolic disorders [117], cancer [118] and neurological diseases [119,120]. These lipids exhibit unique biophysical properties [121]that explain their involvement in the formation of membrane domains [122]. De novo SL synthesis begins with the condensation of serine and palmitoyl-CoA, which is catalyzed by serine palmitoyltransferase (SPT) (Fig. 6). The condensation product, 3-keto-dihydrosphingosine, is reduced to dihydrosphingosine (sphinganine), which provides the backbone for a set of dihydro-sphingolipids that result from N-acylation by various ceramide synthases (CerS) [115]. The introduction of a 4,5-trans double bond generates ceramides (Cer). Sphingomyelin (SM) synthesis represents an interconnection with GP metabolism by the transfer of phosphocholine from PC to Cer. The synthesis of glycosphingolipids is initiated by the addition of glucose to Cer by glycosylceramide synthase (GCS). Sphingomyelinase (SMase) hydrolyzes SM and releases Cer, which may be further degraded by ceramidase (CDase) to sphingosine (SPH). SPH that is liberated from endolysosomal degradation fuels the salvage pathway, which synthesizes Cer by reacylation. SL degradation involves SPH phosphorylation to sphingosine-1-phosphate (S1P),

Fig. 6. Labeling of sphingolipid metabolism. The major pathways of SL metabolism in mammalian cells are displayed. Stable isotope-labeled precursors (compare Table 3) are indicated in grey squares. SLs are indicated in blue squares: Cer, ceramide; DihCer, dihydroceramide; DiSPH, dihydrosphingosine (sphinganine); GlcCer, glucosylceramide; LacCer, lactosylceramide; S1P, sphingosine-1-phosphate; SM, sphingomyelin; SPH, sphingosine. Enzymes in green: CDase, ceramidase; CerS, ceramide synthase; DES, dihydroceramide desaturase; GCS, glucosylceramide synthase; KDS, 3-keto-dihydrosphingosine reductase; LCS, lactosylceramide synthase; LCS, lactosylceramide synthase; SK, sphingosine kinase; SMase, sphingomyelinase; SMS, sphingomyelin synthase; SPP, S1P phosphatase; SPT, serine palmitoyltransferase.

26

J. Ecker, G. Liebisch / Progress in Lipid Research 54 (2014) 14–31

which is an important ligand for G-protein-coupled receptors [123], followed by cleavage, which is mediated by S1P lyase.

4.2. Studies of SL metabolism Selected applications of stable isotopes to analyze SL synthesis and metabolism are summarized in Table 3. (a) Using polar precursors A straightforward approach for monitoring SL biosynthesis is the administration of labeled serine. Berdyshev et al. used human pulmonary artery endothelial cells (HPAEC) to study the effect of sphingosine kinase 1 (SK1) overexpression on SL biosynthesis. By applying [U–13C,15N]-serine and LC–MS/MS analysis, these authors found that the enhanced activity of SK1 leads to a dramatic increase in sphinganine-1-phosphate, which is generated by the phosphorylation of de novo synthesized sphinganine [124]. In another study, the same experimental system was used to demonstrate that the immunomodulating drug FTY720 inhibits CerS, similar to fumonisin B1, but with a less inhibitory effect on the synthesis of Cer 16:0 [125]. Our group has applied D9-choline labeling to investigate the cellular efflux of SM to HDL3 in monocyte-derived macrophages by direct infusion ESI-MS/MS [90] (see Section 3.2a for PC efflux). We have shown that medium chain SM species are preferentially transferred to HDL3. (b) Using labeled FAs Instead of serine, labeled palmitate may be used to analyze the biosynthesis of SL. Haynes et al. [126] labeled HEK293 cells for up to 6 h with fully 13C-labeled palmitate that was bound to BSA at a concentration of 100 lM. The analysis of Cer, HexCer and SM was performed by LC–MS/MS. Palmitate may be incorporated both in the sphingoid base backbone and as N-linked FAs. However, it is possible to determine the label positions by mass spectrometry (see Section 4.3). The analysis of isotopic enrichment in the palmitoyl-CoA pool permitted the calculation of the rate of de novo SL synthesis.

Similar to the technique that was previously described for GP, Tserng et al. [92] used [13C4]-palmitate to profile Cer and SM synthesis in HL60 cells. Lipid classes were fractionated by column chromatography, and SM was converted to Cer before analysis by SMase treatment. Cer was analyzed after derivatization by GC– MS. The EI mass spectra of the Cer-derivatives displayed fragment ions that were specific for the backbone and for the N-acyl labeling. An assay to investigate the in vivo labeling of Cer in muscle biopsies was presented by Blanchnio-Zabielska et al. [127]. Volunteers received a 6 h infusion of [U–13C]-palmitate at a concentration of 2 nmol per kg fat-free mass per min. The analysis was performed by LC–MS/MS, which focused on labeled [13C16]-Cer 16:0. For the analysis of total labeling, these authors used a mass transition of 554 > 536, whereas labeling of N-linked FAs was monitored by 554 > 264 (due to low label enrichment, double labeling was not considered as described by Haynes et al. [126]). Stable isotope enrichment in Cer 16:0 was detected after 2 h in most of the biopsies. After 6 h of infusion, label incorporation in muscle Cer could be detected in all biopsies. (c) Using labeled SL Labeled SL species may be applied to monitor SL transport and metabolism. Tserng et al. [128] used Cer d18:1/[13C4]-16:0 to investigate the uptake and metabolism of this long chain Cer in HL60 cells. The analysis of purified lipid fractions was performed by GC–MS, as described above. These authors could not find enhanced cellular uptake of Cer after the addition of hydrocarbons (as described previously). Additionally, the data clearly indicated that exogenous Cer was degraded and that the liberated [13C4]FA 16:0 incorporated both the sphingoid backbone and N-acyl of newly synthesized Cer. Fukami et al. [129] synthesized [13C12-17]-Cer d18:0/16:0_OH by incubating bacteria with 13C-labeled acetic acid. Next, the labeled dihydro-Cer species were administered orally in a taurocholate suspension to mice. Different tissues were analyzed for isotopic enrichment in sphingoid bases by GC–MS after hydrolysis. After 12 days, labeling in epidermis and liver samples showed

Table 3 Studies of sphingolipid metabolism. Analytes

Authors

Experimental system

Objective

Label

Methods

Cer, Sph, S1P

Berdyshev et al. [124,125]

Human pulmonary artery endothelial cells (HPAEC)

[U–13C,15N]-serine

LC–MS/MS [M + 3] precursor and product ions

Cer

Blanchnio-Zabielska et al. [127]

Human volunteers

SL de novo synthesis upon SK overexpression or pharmacological inhibition Incorporation of palmitate into muscle ceramide

[U–13C]-FA 16:0

Total Sph content

Fukami et al. [129]

Mice

Incorporation of labeled Cer

13

Cer, HexCer, SM

Haynes et al. [126]

HEK293 cells

SL biosynthesis

C12–17–Cer d18:0/ 16:0_OH (bacterial synthesis) [U–13C]-FA 16:0

SM (PC, FC, CE)

Schifferer et al. [90]

HDL3-mediated cellular lipid efflux

D9-choline (13C3–FC)

Cer, SM

Tserng et al. [92]

Primary human monocyte-derived macrophages HL60 cells

FA turnover into SLs

[13C4]-FA 16:0

Cer, SM

Tserng et al. [128]

HL60 cells

Metabolism of exogenous long chain Cer

13

LC–MS/MS [13C16]-Cer 554 > 536, 554 > 264 GC–MS of TMS derivatives of sphingoid bases after hydrolysis LC–MS/MS SM [+16/+32] PIS + 184 SM (diff. of backbone) – conversion to Cer-1-P Backbone [M + H + 16]+ > 280 N-FA specific [M + H + 16]+ > 264 Double labeling [M + H + 32]+ > 264 for Cer, HexCer, Cer-1-P as SM product ESI–MS/MS D9-SM (PC) PIS + 193 (13C3–FC/CE PIS + 372) GC–MS of Cer (also derived from SM) GC–MS of Cer (also derived from SM)

C4-Cer d18:1/16:0

J. Ecker, G. Liebisch / Progress in Lipid Research 54 (2014) 14–31

approximately 4% isotopic enrichment in sphinganine. Sphingosine was not found to be labeled in the epidermis samples but was labeled in the liver samples (approximately 1% of total sphingosine). 4.3. Comparison of labeling strategies and their analysis Similar to those methods that were described for GP, the methods for the comprehensive analysis of SL species almost exclusively use either shotgun – [106,130–132] or LC–MS/MS – [133–137] based approaches (see Section 3.3 for a general comparison of these strategies). In principle, these methods are also applicable to tracing stable isotope-labeled SL species. However, the tracing of minor species requires extremely high sensitivity because the analysis of unlabeled samples is frequently performed close to the limit of detection. Therefore, studies of SL metabolism usually applied LC–MS/MS. Additionally, GC–MS can be used for the analysis of labeled Cer and SM (after SMase treatment). However, this approach is laborious and limited to single analytes. Both palmitic acid and serine label SL metabolism biosynthetically (Fig. 7). The general advantages and disadvantages of both precursors were discussed in the previous sections for FA and polar labels (see Sections 2.4 and 3.3). When labeling SL metabolism, special attention must be given to the mass shift that is introduced by the label. Sphingosine and dihydrosphingosine-based species exhibit only a delta-mass of 2 Da. Such a small mass shift may not allow for clear differentiation of unlabeled sphinganine and labeled sphingosine species. Although fully labeled palmitic acid generates a high mass shift, serine must be labeled with more than one isotope to generate a significant mass shift. Serine enters only the biosynthetic pathway, whereas palmitic acid may be incorporated by SPT and CerS as an acyl chain, which results in the double labeling of sphingolipid species (Fig. 6). However, the differentia-

27

tion of label incorporation into both positions is possible by tandem mass spectrometry, as shown by Haynes et al. [126] for [U–13C]-FA 16:0. In the positive ion mode, sphingosine-based Cer, HexCer and LacCer that are labeled in the sphingoid backbone generate a fragment of m/z 280 compared with m/z 264 for unlabeled species. The labeling of N-linked FAs only shifts the precursor mass by 16 Da but provides a fragment ion of m/z 264, which is diagnostic for the sphingoid backbone. Double labeling results in a mass shift of 32 Da for the precursor and in a sphingoid base fragment of m/z 280. SM generates only a phosphocholine fragment in the positive ion mode, which does not permit the differentiation of label incorporation. To differentiate backbone and N-linked FA labeling, Haynes et al. converted SM by phospholipase D treatment to Cer-1-P, which shows a fragment of the sphingoid backbone that is similar to Cer upon CID. The efficient labeling of SL biosynthesis requires sufficient label enrichment in the precursor pool. Serine originates in high quantities from endogenous sources and impairs high label enrichment in the precursor pool by dilution (our own unpublished observations in cultured fibroblasts). Similarly, high concentrations may be necessary to achieve reasonable enrichment in the palmitic acid precursor pool. However, this high concentration could perturb lipid metabolism or induce SL biosynthesis. In addition to introducing a label into the biosynthetic pathway, specific metabolic reactions may be monitored. To follow the CerS reaction, both labeled sphingoid bases and FAs are conceivable, although the application of a sphingoid base label is favorable compared with FA tracers. Although multiple FA labels would be necessary to profile of the substrate preference of CerS, a sphingoid base label represents the direct precursor of all Cer species. Moreover, with this approach, it is straightforward to compare the metabolism of different sphingoid bases. SMS action can be monitored

Fig. 7. Labeling of sphingolipid biosynthesis. The main advantages and disadvantages of palmitic acid and serine as labels for the SL biosynthesis are displayed. It is important to consider that sphingosine and dihydrosphingosine-based SLs exhibit only a mass difference of 2 Da.

28

J. Ecker, G. Liebisch / Progress in Lipid Research 54 (2014) 14–31

by choline transfer from labeled PC. Because the introduction of labeled PC requires vehicles, PC may be labeled by choline administration. The isotope enrichment of PC must be considered in calculations of SM synthesis rates. Other potential labels for evaluating single enzymatic steps are oxygen-labeled phosphate for the SK reaction or glucose to monitor glycosphingolipid synthesis. Because it is well-known that long-chain Cer hardly enters cultured cells when exogenously applied, vehicles such as cyclodextrin, as described for GPs, may be necessary for introducing intact sphingolipids [94]. 5. General considerations 5.1. Selection of label The most commonly applied tracers are deuterated or 13C-labeled precursors. Both isotopes are naturally occurring. Whereas deuterium represents only 0.015% of total hydrogen, 13C occurs as 1.108% of total carbon [138]. Although isotopes are chemically identical, these molecules may behave differently in biochemical reactions. Deuterium exhibits slightly higher bond strengths compared with hydrogen [138]. This difference may cause a so-called isotope effect, for example, when deuterium is directly involved in the biological reaction. However, a comparison of 13C- and D-labeled FAs did not show significant differences in their metabolism (see Section 2.4) [54,56]. Methodologically, labeling with deuterium exhibits a more pronounced effect on chromatographic separation compared with 13C labeling. Although deuterium-labeled FAs may show a clearly visible shift in retention times, 13C-labeled FAs elute at rather similar retention times as unlabeled FAs in GC [28,54]. It is known that the mass isotopomer distribution of FAMEs that are analyzed by GC-EIMS is influenced by gas-phase chemistry (specifically proton transfer from fragment ions to molecules) [139]. Consequently, deuterium labeling of FAMEs can significantly increase this effect due to deuteron transfer. Finally, one must be aware that some deuterium labels could be lost upon desaturation of a labeled FA, which should be considered by the proper selection of label positions. 5.2. Consideration of isotopic overlap and data analysis Lipid species frequently differ only in the number of double bonds, i.e., a mass difference of 2 Da. Therefore, the presence of naturally abundant isotopes, such as 13C, leads to the overlap of isotopic peaks. For example, PC 34:2 displays an M+2 isotope peak with an intensity of 13% of the monoisotopic peak (M+H analyzed by MS scan), which overlaps the monoisotopic peak of PC 34:1. The correction of such isotopic overlap has been described for various methods [104,106,110]. When labels are introduced, further corrections may be necessary as, for example, the amendment of an overlap of unlabeled PC with D3-PC [19] or, similarly, for D3-serine labeling (see Section 3.3). In contrast to conventional mass spectrometers (operation at unit resolution permits the separation of mass isotopomers), high resolution mass spectrometry is able to resolve isotopic overlap that results from a different number of double bonds [3]. For lipid species that belong to the same lipid class, the second isotopic peak of a lipid species with an additional double bond shows a mass difference of 0.009 Da compared with the monoisotopic peak of the more saturated species (e.g., PC 36:2 second isotope peak m/z 788.6077 compared with PC 36:1 monoisotopic peak m/z 788.6169). Therefore, a mass resolution of 100,000 is sufficient to differentiate these peaks [3,114]. Similarly, labeled lipid species may be analyzed by high resolution MS, as demonstrated by Lane and colleagues. These authors used a Fourier transform ion cyclo-

tron resonance (FT-ICR) MS and monitored the incorporation of [U–13C]-glucose into PC by analyzing the changes in the isotope peak pattern. Recently, algorithms were presented for high resolution mass spectrometry to correct the detected isotope pattern after stable isotope labeling for naturally abundant isotopes [140,141]. In general, for lipidomics experiments, specialized software is required for processing the raw data (e.g., for peak integration and picking, de-isotoping). Apart from the first commercial software packages, many software packages are available for the data analysis of unlabeled lipid species from shotgun and LC–MS based approaches [104,112,142–147]. In principle, these programs can be modified accordingly to also identify labeled species, for instance, from characteristic head group fragments (see Section 3.3).

5.3. Nomenclature of lipid species The nomenclature of labeled species may be based on a recently presented shorthand notation system [148]. Different annotation levels cover the structural information that is provided by the mass spectrometric analysis. Labeled species may be indicated by the isotope and by the number of labeled atoms in front of the lipid class identifier, for example, D9-PC 34:1. A standardized nomenclature could be of great advantage for the exchange of data as well as for their integration into databases, for instance, the recently presented LipidHome database [149].

6. Conclusions and outlook Stable isotope labeling, in combination with mass spectrometric analysis, has become the most important and powerful tool for investigating the metabolism of lipid species. In principle, the methods that are used for lipidomics analysis of unlabeled samples are also applicable for tracing labeled lipid species. A great challenge is the bioinformatic processing of complex data sets that are generated by stable isotope labeling. Usually, both labeled and unlabeled species are acquired at various incubation times. To handle such data, novel bioinformatics approaches may be necessary, as demonstrated by Zhang et al. [150] for the remodeling of PE species. Computational models to follow the dynamics of lipid species metabolism should be able to integrate data sets from different approaches, for example, a combination of GP species data from head group scan analyses (providing the sum composition of FAs) and data from FA scans (providing acyl chain combinations). Only the integration of complimentary data sets will generate a comprehensive picture and help to reveal the complexity of lipid species metabolism. A promising tool to investigate the dynamics of the lipidome is high resolution mass spectrometry, which has been shown to be an excellent approach for shotgun lipidomics [3] (see Section 5.2). As demonstrated by Lane and coworkers, the identification of labeled species is possible by high resolution and by exact mass [114]. Consequently, shotgun lipidomics that use high resolution mass spectrometry should also be able to generate a comprehensive picture of the dynamics of the main part of the lipidome. However, the ultimate sensitivity of LC–MS/MS (triple quadrupole instruments) is necessary to allow the analysis of both short-term labeling and minor species. High sensitivity is also necessary to follow metabolism in subcellular compartments. Therefore, these studies should benefit from the ongoing development of more sensitive tandem mass spectrometers. The investigation of the dynamics and species specificity at the compartment level will greatly improve the understanding of lipid species functions. In addition to monitoring metabolism, stable isotope labels are also applicable to track

J. Ecker, G. Liebisch / Progress in Lipid Research 54 (2014) 14–31

transport processes, such as intracellular transport, cellular export or whole body transport by lipoproteins. In summary, the application of stable isotopes to study the dynamics (metabolism and transport) of lipid species has great potential to improve our knowledge of the function of lipid species.

References [1] Blanksby SJ, Mitchell TW. Advances in mass spectrometry for lipidomics. Annu. Rev. Anal. Chem. (Palo. Alto. Calif.) 2010;3:433–65. [2] Ejsing CS, Sampaio JL, Surendranath V, Duchoslav E, Ekroos K, Klemm RW, et al. Global analysis of the yeast lipidome by quantitative shotgun mass spectrometry. Proc. Natl. Acad. Sci. USA 2009;106:2136–41. [3] Schwudke D, Schuhmann K, Herzog R, Bornstein SR, Shevchenko A. Shotgun lipidomics on high resolution mass spectrometers. Cold Spring Harb. Perspect. Biol. 2011;3:a004614. [4] Wenk MR. Lipidomics: new tools and applications. Cell 2010;143:888–95. [5] Shevchenko A, Simons K. Lipidomics: coming to grips with lipid diversity. Nat. Rev. Mol. Cell Biol. 2010;11:593–8. [6] Schmitz G, Ecker J. The opposing effects of n 3 and n 6 fatty acids. Prog. Lipid Res. 2008;47:147–55. [7] van Meer G, Voelker DR, Feigenson GW. Membrane lipids: where they are and how they behave. Nat. Rev. Mol. Cell Biol. 2008;9:112–24. [8] Smith S, Witkowski A, Joshi AK. Structural and functional organization of the animal fatty acid synthase. Prog. Lipid Res. 2003;42:289–317. [9] Matsuzaka T, Shimano H, Yahagi N, Kato T, Atsumi A, Yamamoto T, et al. Crucial role of a long-chain fatty acid elongase, Elovl6, in obesity-induced insulin resistance. Nat. Med. 2007;13:1193–202. [10] Miyazaki M, Ntambi JM. Role of stearoyl-coenzyme A desaturase in lipid metabolism. Prostaglandins Leukot. Essent. Fatty Acids 2003;68:113–21. [11] Pereira SL, Leonard AE, Mukerji P. Recent advances in the study of fatty acid desaturases from animals and lower eukaryotes. Prostaglandins Leukot. Essent. Fatty Acids 2003;68:97–106. [12] Marszalek JR, Lodish HF. Docosahexaenoic acid, fatty acid-interacting proteins, and neuronal function: breastmilk and fish are good for you. Annu. Rev. Cell Dev. Biol. 2005;21:633–57. [13] Ecker J. Profiling eicosanoids and phospholipids using LC–MS/MS: principles and recent applications. J. Sep. Sci. 2012;35:1227–35. [14] Serhan CN, Petasis NA. Resolvins and protectins in inflammation resolution. Chem. Rev. 2011;111:5922–43. [15] Jeon TI, Osborne TF. SREBPs: metabolic integrators in physiology and metabolism. Trends Endocrinol. Metab 2012;23:65–72. [16] Horton JD, Goldstein JL, Brown MS. SREBPs: activators of the complete program of cholesterol and fatty acid synthesis in the liver. J. Clin. Invest 2002;109:1125–31. [17] Lee WN. Stable isotopes and mass isotopomer study of fatty acid and cholesterol synthesis. A review of the MIDA approach. Adv. Exp. Med. Biol. 1996;399:95–114. [18] Hellerstein MK, Neese RA. Mass isotopomer distribution analysis at eight years: theoretical, analytic, and experimental considerations. Am. J. Physiol 1999;276:E1146–70. [19] Pynn CJ, Henderson NG, Clark H, Koster G, Bernhard W, Postle AD. Specificity and rate of human and mouse liver and plasma phosphatidylcholine synthesis analyzed in vivo. J. Lipid Res. 2011;52:399–407. [20] McLaren DG, He T, Wang SP, Mendoza V, Rosa R, Gagen K, et al. The use of stable-isotopically labeled oleic acid to interrogate lipid assembly in vivo: assessing pharmacological effects in preclinical species. J. Lipid Res. 2011;52:1150–61. [21] Starai VJ, Escalante-Semerena JC. Acetyl-coenzyme A synthetase (AMP forming). Cell Mol. Life Sci. 2004;61:2020–30. [22] Menendez JA, Lupu R. Fatty acid synthase and the lipogenic phenotype in cancer pathogenesis. Nat. Rev. Cancer 2007;7:763–77. [23] Parks EJ, Hellerstein MK. Thematic review series: patient-oriented research. Recent advances in liver triacylglycerol and fatty acid metabolism using stable isotope labeling techniques. J. Lipid Res. 2006;47:1651–60. [24] Bruss MD, Khambatta CF, Ruby MA, Aggarwal I, Hellerstein MK. Calorie restriction increases fatty acid synthesis and whole body fat oxidation rates. Am. J. Physiol Endocrinol. Metab 2010;298:E108–16. [25] Oosterveer MH, van Dijk TH, Tietge UJ, Boer T, Havinga R, Stellaard F, et al. High fat feeding induces hepatic fatty acid elongation in mice. PLoS. ONE. 2009;4:e6066. [26] Lin J, Yang R, Tarr PT, Wu PH, Handschin C, Li S, et al. Hyperlipidemic effects of dietary saturated fats mediated through PGC-1beta coactivation of SREBP. Cell 2005;120:261–73. [27] Ecker J, Liebisch G, Englmaier M, Grandl M, Robenek H, Schmitz G. Induction of fatty acid synthesis is a key requirement for phagocytic differentiation of human monocytes. Proc. Natl. Acad. Sci. USA 2010;107:7817–22. [28] Ecker J, Scherer M, Schmitz G, Liebisch G. A rapid GC–MS method for quantification of positional and geometric isomers of fatty acid methyl esters. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2012;897:98–104. [29] Collins JM, Neville MJ, Pinnick KE, Hodson L, Ruyter B, van Dijk TH, et al. De novo lipogenesis in the differentiating human adipocyte can provide all fatty acids necessary for maturation. J. Lipid Res. 2011;52:1683–92.

29

[30] Perez CL, Van Gilst MR. A 13C isotope labeling strategy reveals the influence of insulin signaling on lipogenesis in C. elegans. Cell Metab. 2008;8:266–74. [31] Kamphorst JJ, Fan J, Lu W, White E, Rabinowitz JD. Liquid chromatographyhigh resolution mass spectrometry analysis of fatty acid metabolism. Anal. Chem. 2011;83:9114–22. [32] Menendez JA. Fine-tuning the lipogenic/lipolytic balance to optimize the metabolic requirements of cancer cell growth: molecular mechanisms and therapeutic perspectives. Biochim. Biophys. Acta 1801;2010:381–91. [33] Rudolph MC, Karl MN, Wellberg EA, Johnson CA, Murphy RC, Anderson SM. Mammalian fatty acid synthase activity from crude tissue lysates tracing 13Clabeled substrates using gas chromatography-mass spectrometry. Anal. Biochem. 2012;428:158–66. [34] Fielding B. Tracing the fate of dietary fatty acids: metabolic studies of postprandial lipaemia in human subjects. Proc. Nutr. Soc. 2011;70:342–50. [35] Persson XM, Blachnio-Zabielska AU, Jensen MD. Rapid measurement of plasma free fatty acid concentration and isotopic enrichment using LC/MS. J. Lipid Res. 2010;51:2761–5. [36] Bickerton AS, Roberts R, Fielding BA, Hodson L, Blaak EE, Wagenmakers AJ, et al. Preferential uptake of dietary Fatty acids in adipose tissue and muscle in the postprandial period. Diabetes 2007;56:168–76. [37] Le PM, Fraser C, Gardner G, Liang WW, Kralovec JA, Cunnane SC, et al. Biosynthetic production of universally (13)C-labelled polyunsaturated fatty acids as reference materials for natural health product research. Anal. Bioanal. Chem. 2007;389:241–9. [38] Hodson L, Fielding BA. Stearoyl-CoA desaturase: rogue or innocent bystander? Prog. Lipid Res. 2013;52:15–42. [39] Brown JM, Chung S, Sawyer JK, Degirolamo C, Alger HM, Nguyen T, et al. Inhibition of stearoyl-coenzyme A desaturase 1 dissociates insulin resistance and obesity from atherosclerosis. Circulation 2008;118:1467–75. [40] Erbay E, Babaev VR, Mayers JR, Makowski L, Charles KN, Snitow ME, et al. Reducing endoplasmic reticulum stress through a macrophage lipid chaperone alleviates atherosclerosis. Nat. Med. 2009;15:1383–91. [41] Cao H, Gerhold K, Mayers JR, Wiest MM, Watkins SM, Hotamisligil GS. Identification of a lipokine, a lipid hormone linking adipose tissue to systemic metabolism. Cell 2008;134:933–44. [42] Ecker J, Liebisch G, Grandl M, Schmitz G. Lower SCD expression in dendritic cells compared to macrophages leads to membrane lipids with less monounsaturated fatty acids. Immunobiology 2010;215:748–55. [43] Yee JK, Mao CS, Hummel HS, Lim S, Sugano S, Rehan VK, et al. Compartmentalization of stearoyl-coenzyme A desaturase 1 activity in HepG2 cells. J. Lipid Res. 2008;49:2124–34. [44] Gagne S, Crane S, Huang Z, Li CS, Bateman KP, Levesque JF. Rapid measurement of deuterium-labeled long-chain fatty acids in plasma by HPLC-ESI-MS. J. Lipid Res. 2007;48:252–9. [45] Mosley EE, McGuire MA. Methodology for the in vivo measurement of the delta9-desaturation of myristic, palmitic, and stearic acids in lactating dairy cattle. Lipids 2007;42:939–45. [46] Mosley EE, Shafii DB, Moate PJ, McGuire MA. Cis-9, trans-11 conjugated linoleic acid is synthesized directly from vaccenic acid in lactating dairy cattle. J. Nutr. 2006;136:570–5. [47] Mosley EE, McGuire MK, Williams JE, McGuire MA. Cis-9, trans-11 conjugated linoleic acid is synthesized from vaccenic acid in lactating women. J. Nutr. 2006;136:2297–301. [48] Mitchell PL, McLeod RS. Conjugated linoleic acid and atherosclerosis: studies in animal models. Biochem. Cell Biol. 2008;86:293–301. [49] Kelley NS, Hubbard NE, Erickson KL. Conjugated linoleic acid isomers and cancer. J. Nutr. 2007;137:2599–607. [50] Bassaganya-Riera J, Hontecillas R. CLA and n 3 PUFA differentially modulate clinical activity and colonic PPAR-responsive gene expression in a pig model of experimental IBD. Clin. Nutr. 2006;25:454–65. [51] Pawlosky RJ, Hibbeln JR, Salem Jr N. Compartmental analyses of plasma n 3 essential fatty acids among male and female smokers and nonsmokers. J. Lipid Res. 2007;48:935–43. [52] Pawlosky RJ, Hibbeln JR, Lin Y, Goodson S, Riggs P, Sebring N, et al. Effects of beef- and fish-based diets on the kinetics of n 3 fatty acid metabolism in human subjects. Am. J. Clin. Nutr. 2003;77:565–72. [53] Quehenberger O, Armando AM, Dennis EA. High sensitivity quantitative lipidomics analysis of fatty acids in biological samples by gas chromatography-mass spectrometry. Biochim. Biophys. Acta 1811;2011:648–56. [54] Lin YH, Pawlosky RJ, Salem Jr N. Simultaneous quantitative determination of deuterium- and carbon-13-labeled essential fatty acids in rat plasma. J. Lipid Res. 2005;46:1974–82. [55] Patterson BW, Zhao G, Klein S. Improved accuracy and precision of gas chromatography/mass spectrometry measurements for metabolic tracers. Metabolism 1998;47:706–12. [56] Patterson BW, Wolfe RR. Concentration dependence of methyl palmitate isotope ratios by electron impact ionization gas chromatography/mass spectrometry. Biol. Mass Spectrom. 1993;22:481–6. [57] Brown SH, Mitchell TW, Blanksby SJ. Analysis of unsaturated lipids by ozoneinduced dissociation. Biochim. Biophys. Acta 1811;2011:807–17. [58] Pham HT, Trevitt AJ, Mitchell TW, Blanksby SJ. Rapid differentiation of isomeric lipids by photodissociation mass spectrometry of fatty acid derivatives. Rapid Commun. Mass Spectrom. 2013;27:805–15. [59] van Meer G, de Kroon AI. Lipid map of the mammalian cell. J. Cell Sci. 2011;124:5–8.

30

J. Ecker, G. Liebisch / Progress in Lipid Research 54 (2014) 14–31

[60] Bohdanowicz M, Grinstein S. Role of phospholipids in endocytosis, phagocytosis, and macropinocytosis. Physiol. Rev. 2013;93:69–106. [61] Wymann MP, Schneiter R. Lipid signalling in disease. Nat. Rev. Mol. Cell Biol. 2008;9:162–76. [62] Scherer M, Bottcher A, Liebisch G. Lipid profiling of lipoproteins by electrospray ionization tandem mass spectrometry. Biochim. Biophys. Acta 1811;2011:918–24. [63] Wiesner P, Leidl K, Boettcher A, Schmitz G, Liebisch G. Lipid profiling of FPLCseparated lipoprotein fractions by electrospray ionization tandem mass spectrometry. J. Lipid Res. 2009;50:574–85. [64] Hermansson M, Hokynar K, Somerharju P. Mechanisms of glycerophospholipid homeostasis in mammalian cells. Prog. Lipid Res. 2011;50:240–57. [65] Vance JE, Vance DE. Phospholipid biosynthesis in mammalian cells. Biochem. Cell Biol. 2004;82:113–28. [66] Cogo PE, Gucciardi A, Traldi U, Hilkert AW, Verlato G, Carnielli V. Measurement of pulmonary surfactant disaturated-phosphatidylcholine synthesis in human infants using deuterium incorporation from body water. J. Mass Spectrom. 2005;40:876–81. [67] Vedovelli L, Baritussio A, Carnielli VP, Simonato M, Giusti P, Cogo PE. Simultaneous measurement of phosphatidylglycerol and disaturatedphosphatidylcholine palmitate kinetics from alveolar surfactant. Study in infants with stable isotope tracer, coupled with isotope ratio mass spectrometry. J. Mass Spectrom. 2011;46:986–92. [68] Bunt JE, Zimmermann LJ, Wattimena JL, van Beek RH, Sauer PJ, Carnielli VP. Endogenous surfactant turnover in preterm infants measured with stable isotopes. Am. J. Respir. Crit Care Med. 1998;157:810–4. [69] Cogo PE, Carnielli VP, Bunt JE, Badon T, Giordano G, Zacchello F, et al. Endogenous surfactant metabolism in critically ill infants measured with stable isotope labeled fatty acids. Pediatr. Res. 1999;45:242–6. [70] Torresin M, Zimmermann LJ, Cogo PE, Cavicchioli P, Badon T, Giordano G, et al. Exogenous surfactant kinetics in infant respiratory distress syndrome: A novel method with stable isotopes. Am. J. Respir. Crit Care Med. 2000;161:1584–9. [71] Carnielli VP, Zimmermann LJ, Hamvas A, Cogo PE. Pulmonary surfactant kinetics of the newborn infant: novel insights from studies with stable isotopes. J. Perinatol. 2009;29(Suppl. 2):S29–37. [72] DeLong CJ, Shen YJ, Thomas MJ, Cui Z. Molecular distinction of phosphatidylcholine synthesis between the CDP-choline pathway and phosphatidylethanolamine methylation pathway. J. Biol. Chem. 1999;274:29683–8. [73] DeLong CJ, Hicks AM, Cui Z. Disruption of choline methyl group donation for phosphatidylethanolamine methylation in hepatocarcinoma cells. J. Biol. Chem. 2002;277:17217–25. [74] Besnard V, Matsuzaki Y, Clark J, Xu Y, Wert SE, Ikegami M, et al. Conditional deletion of Abca3 in alveolar type II cells alters surfactant homeostasis in newborn and adult mice. Am. J. Physiol. Lung Cell Mol. Physiol. 2010;298:L646–59. [75] Postle AD, Henderson NG, Koster G, Clark HW, Hunt AN. Analysis of lung surfactant phosphatidylcholine metabolism in transgenic mice using stable isotopes. Chem. Phys. Lipids 2011;164:549–55. [76] Bernhard W, Pynn CJ, Jaworski A, Rau GA, Hohlfeld JM, Freihorst J, et al. Mass spectrometric analysis of surfactant metabolism in human volunteers using deuteriated choline. Am. J. Respir. Crit Care Med. 2004;170:54–8. [77] Goss V, Hunt AN, Postle AD. Regulation of lung surfactant phospholipid synthesis and metabolism. Biochim. Biophys. Acta 1831;2013:448–58. [78] Postle AD, Dombrowsky H, Clarke H, Pynn CJ, Koster G, Hunt AN. Mass spectroscopic analysis of phosphatidylinositol synthesis using 6-deuteriatedmyo-inositol: comparison of the molecular specificities and acyl remodelling mechanisms in mouse tissues and cultured cells. Biochem. Soc. Trans. 2004;32:1057–9. [79] Hunt AN, Postle AD. Mass spectrometry determination of endonuclear phospholipid composition and dynamics. Methods 2006;39:104–11. [80] Hunt AN, Clark GT, Attard GS, Postle AD. Highly saturated endonuclear phosphatidylcholine is synthesized in situ and colocated with CDP-choline pathway enzymes. J. Biol. Chem. 2001;276:8492–9. [81] Hunt AN, Clark GT, Neale JR, Postle AD. A comparison of the molecular specificities of whole cell and endonuclear phosphatidylcholine synthesis. FEBS Lett. 2002;530:89–93. [82] Hunt AN. Completing the cycles; the dynamics of endonuclear lipidomics. Biochim. Biophys. Acta 2006;1761:577–87. [83] Hunt AN, Fenn HC, Clark GT, Wright MM, Postle AD, McMaster CR. Lipidomic analysis of the molecular specificity of a cholinephosphotransferase in situ. Biochem. Soc. Trans. 2004;32:1060–2. [84] Ecker J, Liebisch G, Scherer M, Schmitz G. Differential effects of conjugated linoleic acid isomers on macrophage glycerophospholipid metabolism. J. Lipid Res. 2010;51:2686–94. [85] Attard GS, Templer RH, Smith WS, Hunt AN, Jackowski S. Modulation of CTP:phosphocholine cytidylyltransferase by membrane curvature elastic stress. Proc. Natl. Acad. Sci. USA 2000;97:9032–6. [86] Boumann HA, Damen MJ, Versluis C, Heck AJ, De KB, de Kroon AI. The two biosynthetic routes leading to phosphatidylcholine in yeast produce different sets of molecular species. Evidence for lipid remodeling. Biochemistry 2003;42:3054–9. [87] Boumann HA, Chin PT, Heck AJ, De KB, De Kroon AI. The yeast phospholipid Nmethyltransferases catalyzing the synthesis of phosphatidylcholine

[88]

[89]

[90]

[91]

[92] [93]

[94] [95]

[96]

[97]

[98]

[99]

[100] [101]

[102]

[103]

[104]

[105]

[106]

[107]

[108]

[109]

[110]

[111]

[112]

[113]

preferentially convert di-C16:1 substrates both in vivo and in vitro. J. Biol. Chem. 2004;279:40314–9. Boumann HA, De KB, Heck AJ, De Kroon AI. The selective utilization of substrates in vivo by the phosphatidylethanolamine and phosphatidylcholine biosynthetic enzymes Ept1p and Cpt1p in yeast. FEBS Lett. 2004;569:173–7. de Kroon AI. Metabolism of phosphatidylcholine and its implications for lipid acyl chain composition in Saccharomyces cerevisiae. Biochim. Biophys. Acta 2007;1771:343–52. Schifferer R, Liebisch G, Bandulik S, Langmann T, Dada A, Schmitz G. ApoA-I induces a preferential efflux of monounsaturated phosphatidylcholine and medium chain sphingomyelin species from a cellular pool distinct from HDL(3) mediated phospholipid efflux. Biochim. Biophys. Acta 2007;1771:853–63. Tserng KY, Griffin RL. Phosphatidylcholine de novo synthesis and modification are carried out sequentially in HL60 cells: evidence from mass isotopomer distribution analysis. Biochemistry 2004;43:8125–35. Tserng KY, Griffin R. Studies of lipid turnover in cells with stable isotope and gas chromatograph-mass spectrometry. Anal. Biochem. 2004;325:344–53. McLaren DG, Wang SP, Stout SJ, Xie D, Miller PL, Mendoza V, et al. Tracking fatty acid kinetics in distinct lipoprotein fractions in vivo: a novel highthroughput approach for studying dyslipidemia in rodent models. J. Lipid Res. 2013;54:276–81. Kainu V, Hermansson M, Somerharju P. Introduction of phospholipids to cultured cells with cyclodextrin. J. Lipid Res. 2010;51:3533–41. Kainu V, Hermansson M, Somerharju P. Electrospray ionization mass spectrometry and exogenous heavy isotope-labeled lipid species provide detailed information on aminophospholipid acyl chain remodeling. J. Biol. Chem. 2008;283:3676–87. Kainu V, Hermansson M, Hanninen S, Hokynar K, Somerharju P. Import of phosphatidylserine to and export of phosphatidylethanolamine molecular species from mitochondria. Biochim. Biophys. Acta 1831;2013:429–37. Wen Z, Kim HY. Inhibition of phosphatidylserine biosynthesis in developing rat brain by maternal exposure to ethanol. J. Neurosci. Res. 2007;85:1568–78. Kevala JH, Kim HY. Determination of substrate preference in phosphatidylserine decarboxylation by liquid chromatography-electrospray ionization mass spectrometry. Anal. Biochem. 2001;292:130–8. Kim HY, Bigelow J, Kevala JH. Substrate preference in phosphatidylserine biosynthesis for docosahexaenoic acid containing species. Biochemistry 2004;43:1030–6. Pulfer M, Murphy RC. Electrospray mass spectrometry of phospholipids. Mass Spectrom. Rev. 2003;22:332–64. Han X, Yang K, Gross RW. Multi-dimensional mass spectrometry-based shotgun lipidomics and novel strategies for lipidomic analyses. Mass Spectrom. Rev. 2012;31:134–78. Brouwers JF. Liquid chromatographic-mass spectrometric analysis of phospholipids. Chromatography, ionization and quantification. Biochim. Biophys. Acta 1811;2011:763–75. Brugger B, Erben G, Sandhoff R, Wieland FT, Lehmann WD. Quantitative analysis of biological membrane lipids at the low picomole level by nanoelectrospray ionization tandem mass spectrometry. Proc. Natl. Acad. Sci. USA 1997;94:2339–44. Ejsing CS, Duchoslav E, Sampaio J, Simons K, Bonner R, Thiele C, et al. Automated identification and quantification of glycerophospholipid molecular species by multiple precursor ion scanning. Anal. Chem. 2006;78:6202–14. Liebisch G, Drobnik W, Lieser B, Schmitz G. High-throughput quantification of lysophosphatidylcholine by electrospray ionization tandem mass spectrometry. Clin. Chem. 2002;48:2217–24. Liebisch G, Lieser B, Rathenberg J, Drobnik W, Schmitz G. High-throughput quantification of phosphatidylcholine and sphingomyelin by electrospray ionization tandem mass spectrometry coupled with isotope correction algorithm. Biochim. Biophys. Acta 2004;1686:108–17. Koivusalo M, Haimi P, Heikinheimo L, Kostiainen R, Somerharju P. Quantitative determination of phospholipid compositions by ESI-MS: effects of acyl chain length, unsaturation, and lipid concentration on instrument response. J. Lipid Res. 2001;42:663–72. Hermansson M, Uphoff A, Kakela R, Somerharju P. Automated quantitative analysis of complex lipidomes by liquid chromatography/mass spectrometry. Anal. Chem. 2005;77:2166–75. Retra K, Bleijerveld OB, van Gestel RA, Tielens AG, van Hellemond JJ, Brouwers JF. A simple and universal method for the separation and identification of phospholipid molecular species. Rapid Commun. Mass Spectrom. 2008;22:1853–62. Scherer M, Schmitz G, Liebisch G. Simultaneous Quantification of Cardiolipin, Bis(monoacylglycero)phosphate and their Precursors by Hydrophilic Interaction LC–MS/MS Including Correction of Isotopic Overlap. Anal. Chem. 2010;82:8794–9. Binder M, Liebisch G, Langmann T, Schmitz G. Metabolic profiling of glycerophospholipid synthesis in fibroblasts loaded with free cholesterol and modified low density lipoproteins. J. Biol. Chem. 2006;281:21869–77. Schwudke D, Oegema J, Burton L, Entchev E, Hannich JT, Ejsing CS, et al. Lipid profiling by multiple precursor and neutral loss scanning driven by the datadependent acquisition. Anal. Chem. 2006;78:585–95. Bilgin M, Markgraf DF, Duchoslav E, Knudsen J, Jensen ON, de Kroon AI, et al. Quantitative profiling of PE, MMPE, DMPE, and PC lipid species by multiple

J. Ecker, G. Liebisch / Progress in Lipid Research 54 (2014) 14–31

[114]

[115] [116] [117] [118] [119] [120] [121] [122] [123] [124]

[125]

[126]

[127]

[128] [129]

[130]

[131]

[132]

[133]

[134]

precursor ion scanning: a tool for monitoring PE metabolism. Biochim. Biophys. Acta 1811;2011:1081–9. Lane AN, Fan TW, Xie Z, Moseley HN, Higashi RM. Isotopomer analysis of lipid biosynthesis by high resolution mass spectrometry and NMR. Anal. Chim. Acta 2009;651:201–8. Mullen TD, Hannun YA, Obeid LM. Ceramide synthases at the centre of sphingolipid metabolism and biology. Biochem. J. 2012;441:789–802. Merrill Jr AH. Sphingolipid and glycosphingolipid metabolic pathways in the era of sphingolipidomics. Chem. Rev. 2011;111:6387–422. Hla T, Dannenberg AJ. Sphingolipid signaling in metabolic disorders. Cell Metab. 2012;16:420–34. Morad SA, Cabot MC. Ceramide-orchestrated signalling in cancer cells. Nat. Rev. Cancer 2013;13:51–65. van Echten-Deckert G, Walter J. Sphingolipids: critical players in Alzheimer’s disease. Prog. Lipid Res. 2012;51:378–93. Horres CR, Hannun YA. The roles of neutral sphingomyelinases in neurological pathologies. Neurochem. Res. 2012;37:1137–49. Goni FM, Alonso A. Effects of ceramide and other simple sphingolipids on membrane lateral structure. Biochim. Biophys. Acta 2009;1788:169–77. Simons K, Gerl MJ. Revitalizing membrane rafts: new tools and insights. Nat. Rev. Mol. Cell Biol. 2010;11:688–99. Hla T, Brinkmann V. Sphingosine 1-phosphate (S1P): Physiology and the effects of S1P receptor modulation. Neurology 2011;76:S3–8. Berdyshev EV, Gorshkova IA, Usatyuk P, Zhao Y, Saatian B, Hubbard W, et al. De novo biosynthesis of dihydrosphingosine-1-phosphate by sphingosine kinase 1 in mammalian cells. Cell Signal. 2006;18:1779–92. Berdyshev EV, Gorshkova I, Skobeleva A, Bittman R, Lu X, Dudek SM, et al. FTY720 inhibits ceramide synthases and up-regulates dihydrosphingosine 1phosphate formation in human lung endothelial cells. J. Biol. Chem. 2009;284:5467–77. Haynes CA, Allegood JC, Wang EW, Kelly SL, Sullards MC, Merrill Jr AH. Factors to consider in using [U-C]palmitate for analysis of sphingolipid biosynthesis by tandem mass spectrometry. J. Lipid Res. 2011;52:1583–94. Blachnio-Zabielska AU, Persson XM, Koutsari C, Zabielski P, Jensen MD. A liquid chromatography/tandem mass spectrometry method for measuring the in vivo incorporation of plasma free fatty acids into intramyocellular ceramides in humans. Rapid Commun. Mass Spectrom. 2012;26:1134–40. Tserng KY, Griffin RL. Ceramide metabolite, not intact ceramide molecule, may be responsible for cellular toxicity. Biochem. J. 2004;380:715–22. Fukami H, Tachimoto H, Kishi M, Kaga T, Waki H, Iwamoto M, et al. Preparation of (13)C-labeled ceramide by acetic acid bacteria and its incorporation in mice. J. Lipid Res. 2010;51:3389–95. Han X. Characterization and direct quantitation of ceramide molecular species from lipid extracts of biological samples by electrospray ionization tandem mass spectrometry. Anal. Biochem. 2002;302:199–212. Liebisch G, Drobnik W, Reil M, Trumbach B, Arnecke R, Olgemoller B, et al. Quantitative measurement of different ceramide species from crude cellular extracts by electrospray ionization tandem mass spectrometry (ESI-MS/MS). J. Lipid Res. 1999;40:1539–46. Schuhmann K, Almeida R, Baumert M, Herzog R, Bornstein SR, Shevchenko A. Shotgun lipidomics on a LTQ Orbitrap mass spectrometer by successive switching between acquisition polarity modes. J. Mass Spectrom. 2012;47:96–104. Shaner RL, Allegood JC, Park H, Wang E, Kelly S, Haynes CA, et al. Quantitative analysis of sphingolipids for lipidomics using triple quadrupole and quadrupole linear ion trap mass spectrometers. J. Lipid Res. 2009;50:1692–707. Scherer M, Schmitz G, Liebisch G. High-throughput analysis of sphingosine 1phosphate, sphinganine 1-phosphate, and lysophosphatidic acid in plasma

[135]

[136]

[137]

[138] [139]

[140] [141]

[142] [143]

[144]

[145] [146] [147]

[148]

[149]

[150]

[151]

[152]

[153]

31

samples by liquid chromatography-tandem mass spectrometry. Clin. Chem. 2009;55:1218–22. Scherer M, Leuthauser-Jaschinski K, Ecker J, Schmitz G, Liebisch G. A rapid and quantitative LC–MS/MS method to profile sphingolipids. J. Lipid Res. 2010;51:2001–11. Bielawski J, Pierce JS, Snider J, Rembiesa B, Szulc ZM, Bielawska A. Comprehensive quantitative analysis of bioactive sphingolipids by highperformance liquid chromatography-tandem mass spectrometry. Methods Mol. Biol. 2009;579:443–67. Farwanah H, Wirtz J, Kolter T, Raith K, Neubert RH, Sandhoff K. Normal phase liquid chromatography coupled to quadrupole time of flight atmospheric pressure chemical ionization mass spectrometry for separation, detection and mass spectrometric profiling of neutral sphingolipids and cholesterol. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2009;877:2976–82. Rohwedder WK. Mass spectrometry of lipids labeled with stable isotopes. Prog. Lipid Res. 1985;24:1–18. Fagerquist CK, Neese RA, Hellerstein MK. Molecular ion fragmentation and its effects on mass isotopomer abundances of fatty acid methyl esters ionized by electron impact. J Am. Soc. Mass Spectrom. 1999;10:430–9. Millard P, Letisse F, Sokol S, Portais JC. IsoCor: correcting MS data in isotope labeling experiments. Bioinformatics 2012;28:1294–6. Moseley HN. Correcting for the effects of natural abundance in stable isotope resolved metabolomics experiments involving ultra-high resolution mass spectrometry. BMC. Bioinform. 2010;11:139. Haimi P, Uphoff A, Hermansson M, Somerharju P. Software tools for analysis of mass spectrometric lipidome data. Anal. Chem. 2006;78:8324–31. Hartler J, Trotzmuller M, Chitraju C, Spener F, Kofeler HC, Thallinger GG. Lipid Data Analyzer: unattended identification and quantitation of lipids in LC–MS data. Bioinformatics 2011;27:572–7. Herzog R, Schuhmann K, Schwudke D, Sampaio JL, Bornstein SR, Schroeder M, et al. LipidXplorer: a software for consensual cross-platform lipidomics. PLoS One 2012;7:e29851. Hubner G, Crone C, Lindner B. LipID – a software tool for automated assignment of lipids in mass spectra. J. Mass Spectrom. 2009;44:1676–83. Leavell MD, Leary JA. Fatty acid analysis tool (FAAT): An FT-ICR MS lipid analysis algorithm. Anal. Chem. 2006;78:5497–503. Song H, Hsu FF, Ladenson J, Turk J. Algorithm for processing raw mass spectrometric data to identify and quantitate complex lipid molecular species in mixtures by data-dependent scanning and fragment ion database searching. J. Am. Soc. Mass Spectrom. 2007;18:1848–58. Liebisch G, Vizcaino JA, Kofeler H, Trotzmuller M, Griffiths WJ, Schmitz G, et al. Shorthand notation for lipid structures derived from mass spectrometry. J. Lipid Res. 2013;54:1523–30. Foster JM, Moreno P, Fabregat A, Hermjakob H, Steinbeck C, Apweiler R, et al. LipidHome: a database of theoretical lipids optimized for high throughput mass spectrometry lipidomics. PLoS One 2013;8:e61951. Zhang L, Az-Diaz N, Zarringhalam K, Hermansson M, Somerharju P, Chuang J. Dynamics of the ethanolamine glycerophospholipid remodeling network. PLoS One 2012;7:50858. Wong DA, Bassilian S, Lim S, Paul WN. Lee, Coordination of peroxisomal betaoxidation and fatty acid elongation in HepG2 cells. J. Biol. Chem. 2004;279:41302–9. Castro-Perez JM, Roddy TP, Shah V, McLaren DG, Wang SP, Jensen K, et al. Identifying static and kinetic lipid phenotypes by high resolution UPLC-MS: unraveling diet-induced changes in lipid homeostasis by coupling metabolomics and fluxomics. J. Proteome. Res. 2011;10:4281–90. Sewell GW, Hannun YA, Han X, Koster G, Bielawski J, Goss V, et al. Lipidomic profiling in Crohn’s disease: abnormalities in phosphatidylinositols, with preservation of ceramide, phosphatidylcholine and phosphatidylserine composition. Int. J. Biochem. Cell Biol. 2012;44:1839–46.