G Model
ARTICLE IN PRESS
CHROMA-356196; No. of Pages 9
Journal of Chromatography A, xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
A fast Liquid Chromatography-Mass Spectrometry methodology for membrane lipid profiling through Hydrophilic Interaction Liquid chromatography Andrea Anesi a,∗ , Graziano Guella a,b a b
Bioorganic Chemistry Lab, Department of Physics, University of Trento, Via Sommarive 14, 38123 Povo (Trento), Italy Biophysical Institute, CNR, Via alla Cascata 56/C, 38123 Povo (Trento), Italy
a r t i c l e
i n f o
Article history: Received 1 September 2014 Received in revised form 12 January 2015 Accepted 13 January 2015 Available online xxx Keywords: Hydrophilic-interaction liquid chromatography (HILIC)–diol column–lipid separation–mass spectrometry
a b s t r a c t In this paper, we report the development of a new method based on HILIC-ESI-MS for the separation of several different membrane lipid classes and their detection on a triple quadrupole mass spectrometer using Precursor Ion (PIS) and Neutral Loss (NL) scanning in positive ion mode. Four different columns were tested for their ability to separate, under different conditions, a mixture of 14 lipid standards containing 7 glycerophospholipids (GPL), 2 glycosphyngolipids (GSL), 3 glycolipids (GL) and 2 betaine lipids (BL). The best separation was obtained using a Lichrosphere DIOL column as stationary phase and water (10 mm ammonium acetate)/acetonitrile gradient elution as mobile phase which allows the separation of the 14 lipid classes within 35 min runtime. Our method was successfully tested for the separation and analysis of crude lipid extracts obtained from a green alga (Jaoa bullata), a dinoflagellate (Peridinium cinctum) and a plant (Vitis vinifera cv. Corvina) © 2015 Elsevier B.V. All rights reserved.
1. Introduction Lipidomics, or “the system-level analysis and characterization of lipids and factors that interacts with them” [1,2], is an emerging area within the field of “omics” sciences [3]. The term “lipids” defines a vast array of chemically diverse substances, which are mostly water-insoluble and contain fatty acids or their derivatives [4]. Lipids have been categorized into eight major classes: fatty acids, glycerolipids, glycerophospholipids, polyketides, prenyl lipids, saccharolipids, sphingolipids and sterols [5]. Lipids are involved in many cellular processes and aberrant lipid metabolism is a determinant of the onset of some important diseases, such as diabetes, atherosclerosis, obesity, Alzheimer’s disease and some cancer types [1]. It is thus clear why the comprehensive analysis of the lipidome, the complete set of lipids within a cell, tissue or organism, is of great interest. Various analytical methodologies have been applied for lipid analysis, such as thin-layer-chromatography (TLC), highperformance liquid chromatography (HPLC) coupled with mass spectrometry (HPLC-MS), diode array (HPLC-DAD) or evaporative light scattering detector (HPLC-ELSD), nuclear magnetic resonance
∗ Corresponding author. Tel.: +39 0461281547. E-mail address: [email protected] (A. Anesi).
(NMR) and gas-chromatography (GC) coupled to mass spectrometry and/or flame ionization detector (GC-MS/FID). The use of HPLC has been widely exploited as it is possible to achieve a fast and consistent separation of many components of a mixture, and given the large sample loading compared to TLC, the ease of automation and the hyphenation with a wide array of detectors are advantageous [1,5,6]. Separation of lipids can be achieved either by normal phase (NP) or reverse-phase-liquid chromatography (RP-LC), as depicted in Fig. 1. NP methods exploit silica, diol or amine-bonded stationary phases in combination with mixtures of hexane/isopropanol/water as mobile phases; under NP-LC lipids are separated according to different polarity of the heads [5,7–9]. On the other hand, RP-based separation on octasilyl (C8)- and octadecylsilyl (C18)- derivatized stationary phases is achieved in combination with solvent systems usually composed of mixtures of methanol, acetonitrile and water. Under RP-based conditions, the separation of membrane lipids mainly relies on the general chemical features of their acyl (or alkyl in plasmalogens) chains such as length, number and position of unsaturation [5,9–13]. For this reason, RP-and NP-based separations are considered complementary because the former is particularly suitable to establish the diversity of molecular lipid species within the same lipid class (intra-class differentiation) whilst the latter can be exploited to establish the total lipid species belonging to different membrane lipids classes in a given sample
http://dx.doi.org/10.1016/j.chroma.2015.01.035 0021-9673/© 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: A. Anesi, G. Guella, A fast Liquid Chromatography-Mass Spectrometry methodology for membrane lipid profiling through Hydrophilic Interaction Liquid chromatography, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.01.035
G Model CHROMA-356196; No. of Pages 9 2
ARTICLE IN PRESS A. Anesi, G. Guella / J. Chromatogr. A xxx (2015) xxx–xxx
phosphatidylcholine (PC) and lyso-PC, phosphatidylethanolamine (PE), phosphatidylglycerol (PG), phosphatidylinositol (PI), phosphatidylserine (PS)), 2 GSL (sphingomyelin (SM) and ceramide CER), 3 GLs (MGDG, DGDG and SQDG) and 2 BLs (DGTS and DGCC). The separation of lipids was tested by using four different columns with acetonitrile, water and ammonium acetate as mobile phases and then further optimized on the column giving the best chromatographic results. Lipid classes were separated within a 35 min-run time, showing a good repeatability for most of lipid classes. The method was further tested on complex lipid extracts, such as the dinoflagellate Peridinium cinctum, the green alga Jaoa bullata and the berries of Vitis vinifera. 2. Materials and methods Fig. 1. Chromatographic methods exploited in lipid analysis: lipids can be separated in NP and HILIC according to the different polarity of their heads or in RP according to the hydrophobicity of their acyl chains.
(inter-classes differentiation) [9]. Of course, since the partition processes determining the chromatographic separations are multiple and complex, using an RP-column is often possible to roughly separate (especially when the lipid classes have quite different polarity) not only the lipid species belonging to a given class but also lipid species belonging to different lipid classes [13–15]. In recent years, lipid analysis has benefitted from the development of hydrophilic interaction liquid chromatography (HILIC), which allows one to overcome (i) the scarce reproducibility observed in NP-LC due to the low miscibility of hexane/isopropanol/water solvent systems and (ii) the low ESI compatibility of NP-based mobile phases. The principle of HILIC lipid separation is the same as for NP-LC but with the difference that mobile phases are similar to those used for RP-LC. An HILIC stationary phase is hydrophilic and/or charged, with solvent systems consisting of acetonitrile/methanol/water (by eventual addition of an ion-pairing reagent as volatile buffer). Much of the lipid separation using HILIC chromatography has been done for the separation of glycerophospholipids (GPLs) and glycerosphingolipids (GSLs) classes from various matrices, as tissues, protozoans or milk [16–19]. Bacteria, eukaryotic microorganisms and higher plants are sources of other lipid classes such as the glycolipids (GL) mono- and digalactosyldiacylglycerols (MGDG and DGDG) and sulfoquinovosyldiacylglycerol (SQDG), and the betaine (BL) etherlinked glycerolipids diacylglyceryl-hydroxymethoxy-trimethyl-alanine (DGTA), diacylglyceryl-trimethyl-homoserine (DGTS) and diacylglyceryl-carboxymethylcholine (DGCC) [20–26]. These lipids are co-extracted with GPL using common extraction procedures [22,27,28]. In few studies, betaine lipids and glycolipids were separated from GPL using HILIC chromatography in combination with hexane, isopropanol and water solvent systems [21,24] while in plant lipidomics, glycolipids were separated from phospholipids using mixtures of methanol, acetonitrile and water [23,29]. In a recent work, Kind and co-workers developed an HILIC method based on acetonitrile, water, ammonium acetate and acetic acid for the qualitative determination by high resolution-MS of lipids secreted by algae [22]. Lipid classes, mainly GPLs phosphatidylcholine and phosphatidylinositol, the glycolipid SQDG, the betaine DGTS, and triacylglycerols, were separated within 20 minrun time but many species were still coeluted and, therefore, a longer run time would be needed for a clearer separation. Here we present a simple but detailed diol-based HILICESI-MS method for the separation and detection of fourteen different lipid classes that constitute cellular membranes of various micro- and macrorganisms: 7 GPLs (phosphatidic acid (PA),
2.1. Chemicals Analytical grade acetonitrile, HPLC grade chloroform and reagent grade methanol were purchased from VWR (VWR International PBI, Milan, Italy); deionized water, filtered at 0.2 m, was obtained from Elix Water Purification System (Merck Millipore, Billerica, MA, USA). Reagent grade ammonium acetate was purchased from Rudi Pont (Chimica Rudi Pont, Torino, Italy); reagent grade ammonium formate was purchased from Carlo Erba (Carlo Erba Reagents, Milan, Italy); reagent grade acetic acid was purchased from Merck (Merck KGaA, Darmstadt, Germany) and LC-MS grade formic acid was purchased from Fisher Scientific (Fisher Scientific, Illkirch, France). 2.2. Commercial standards 1,2-Dioleoyl-sn-glycero-3-phospho-L-serine (PS sodium salt), 1,2-Dioleoyl-sn-glycero-3-phosphate (PA sodium salt), L-␣-phosphatidylinositol mixture extracted from soy (PI sodium salt), L-␣-phosphatidylethanolamine mixture extracted from soy (PE sodium salt), 1,2-Distearoyl-sn-glycero-3phospho-(1 -rac-glycerol) (PG sodium salt), 1,2-Dilinolenoylsn-glycero-3-phosphocholine (PC), 1-Myristoyl-2-hydroxy-snglycero-3-phosphocholine (lyso-PC), N-palmytoyl-d-erythrosphingosylphosphorylcholine (SM), N-lignoceroyl-D-erythrosphingosine (CER) and 1,2-Dipalmitoyl-sn-glycero-3-O-4 -(N,N,Ntrimethyl)-homoserine (DGTS) were purchased from Avanti Polar Lipids (Avanti Polar Lipids Inc., Alabaster, AL, USA). Saturated monogalactosyldiacylglycerols- (MGDG) and digalactosyldiacylglycerols (DGDG) were purchased from Matreya (Matreya LLC, Pleasant Gap, PA, USA). 2.3. Extraction of non-commercially available standard compounds Sulfoquinovosyldiacylglycerols (SQDG) were isolated from Codium bursa raw extract by Solid Phase Extraction (SPE) on a 2 g RP 18 cartridge (Phenomenex, Torrance, CA, USA)using methanolwater 85:15. The DGCC were isolated from Peridinium aciculiferum raw extract by SPE on a 2 g silica cartridge (Phenomenex, Torrance, CA, USA). A gradient elution of hexane/ethyl acetate was used to remove non-polar lipids, as sterols and triacylglycerols, and pigments; then polar lipids were eluted using 45 ml of methanol acidified with 0.1% of formic acid. The fraction was dried under rotary evaporation (Büchi Labortechnik AG, Flawil, Swiss), re-dissolved into chloroform and subsequently injected in HPLC-HILIC-UV system to collect single molecular species. The 1-myristoyl-2-docosahexaenoyl-sn-glycero-3-Odiacylglyceryl-carboxymethylcholine was selected as a standard.
Please cite this article in press as: A. Anesi, G. Guella, A fast Liquid Chromatography-Mass Spectrometry methodology for membrane lipid profiling through Hydrophilic Interaction Liquid chromatography, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.01.035
G Model CHROMA-356196; No. of Pages 9
ARTICLE IN PRESS A. Anesi, G. Guella / J. Chromatogr. A xxx (2015) xxx–xxx
3
2.4. Optimization of HILIC separation
2.6. Sample preparation for method application
For the optimization of HILIC separation, 4 different columns were tested: Luna 3 m HILIC (150 × 3 mm I.D., particle size: ˚ and a Luna 5 m HILIC (250 × 4.6 mm 3 m, pore diameter: 200 A) ˚ both purchased I.D., particle size: 5 m, pore diameter: 200 A) from Phenomenex (Phenomenex, Torrance, CA, USA), SeQuant® ZIC® -pHILIC (150 × 4.6 mm I.D., particle size: 5 m, pore diame˚ and LiChrospher® DIOL (250 × 4.6 mm I.D., particle size: ter: 100 A) ˚ both purchased from Merck (Merck 5 m, pore diameter: 100 A) KGaA, Darmstadt, Germany). Four combinations of solvent A (acetonitrile: water 95:5) were tested: neat, containing 10 mM ammonium acetate, containing 0.01% formic acid and containing both 10 mM ammonium acetate and 0.01% formic acid. Three different solvent B combinations were tested: water containing 10 mM ammonium acetate, water containing 10 mM ammonium acetate and 0.05% of acetic acid; water containing 10 mM ammonium formate and 0.125% formic acid. Both isocratic and gradient elution were tested.
To test our method, crude lipid extracts from various organisms were prepared. About 500 mg of a) cell culture of the dinoflagellate Peridinium cinctum, b) green alga Jaoa bullata and c) de-seeded grape berries (ripening stage) of Vitis vinifera cv. Corvina, powdered in liquid nitrogen were extracted according to [27]. Briefly, 10 mL of chloroform/methanol 2:1 (v/v) were added and samples were sonicated for 15 min in an ultrasonic bath (Sonorex Super, Bandelin electronics, Berlin, Germany). The samples were then centrifuged at 3000 × g for 10 min at room temperature and the organic phase (bottom layer) was collected into a new glass tube. The extraction procedure was repeated three times. Finally, the organic phase was filtered by using glass filters under vacuum and reduced to dryness on a rotary evaporation (Büchi Labortechnik AG, Flawil, Swiss). Total lipids were re-suspended in 200 L of chloroform and 10 L were used for MS and MS/MS analysis. Lastly, a mixture containing equal amounts of P. cinctum, J. bullata and V. vinifera cv. Corvina raw extracts was also analyzed to test our method in a sample where complexity derives from lipid classes and lipid species.
2.5. High performance liquid chromatography-electrospray ionization-triple quadrupole-mass spectrometry (HPLC-ESI-QqQ-MS) The LC analyses were carried out on a Shimadzu Prominence LC-20A system controlled by LCSolution 1.23 A software (Shimadzu Corporation, Milan, Italy). A LiChrospher® DIOL (250 × 4.6 mm, par˚ purchased from Merck ticle size: 5 m, pore diameter: 100 A) (Merck KGaA, Darmstadt, Germany) was used for lipid separation. Solvent A consisted of acetonitrile: water 95:5 containing 10 mM ammonium acetate, solvent B consisted of water containing 10 mM ammonium acetate. The separation was carried out at room temperature with a flow of 0.9 mL/min; injection volume was 10 L of standard dissolved in chloroform. The system was maintained at 100% A for 2 min; then B was increased to 5% over 23 min, then to 10% B over 5 min and then kept to 10% B for 5 min. The column was re-equilibrated to starting conditions for 10 min prior to new injection. The MS acquisitions were performed on an API 3000 triple quadrupole controlled by Analyst software 1.4.2 (ABI Sciex, Framingham, MA, USA) and equipped with an electrospray ionization source operated in positive ion mode. The following parameters were used: nebulizer gas: 8 psi, curtain gas: 8 psi, ion spray voltage: +5 kV, temperature: 300 ◦ C; declustering potential: 40 V; focusing potential: 300 V; entrance potential: 15 V; mass range: 450–1000 m/z. Specific Precursor Ion (PIS) and Neutral Loss (NL) scanning in positive ion mode were also conducted using the parameters reported in Table 1. Parameters for PIS and NL scanning were taken from standard compound analysis.
3. Results and discussion 3.1. Optimization of HILIC separation Separation of fourteen lipid species was tested with 4 different columns different conditions. The Phenomenex Luna 5 HILIC (250 × 4.6 mm) and Merck LiChrospher® DIOL (250 × 4.6 mm) columns showed similar performances but PS peak shape was better on Merck LiChrospher® column, as shown in Fig. 2; for this reason, the column was selected for standard analysis. Under isocratic elution in 100%A, lipids eluted completely after 60 min (data not shown); therefore, we opted for a gradient elution, which usually ensures better repeatability and resolution, more narrow peaks and faster analysis time [18]. Gradient elution was modified to ensure rapid and complete separation of standards: starting conditions (100%A) were maintained isocratically for 2 min and then %B was first increased to 5% over 23 min, then to 10% over 5 min and kept at 10% B for 5 min. A total of 15% of water was sufficient for the complete elution of all species and no lipid carryover was detected also in real samples (data not shown). 10 min equilibration time at starting conditions resulted to be enough to maintain repeatability. The mobile phase composition is very important for HILIC separation as highlighted by Lísa et al., [2]. The use of hexane/isopropanol/water mixtures provides the best results in terms of resolution but the reproducibility is scarce, due to low
Table 1 Mass spectrometry conditions in positive ion mode for molecular-ion analysis of fourteen classes of polar lipids using a HILIC-ESI-QqQ-MS.
Phospholipids
Sphingolipids Betaine lipids Glycolipids
Lipid class
Molecular ion
Scan type
Mass of fragment ion/neutral
Molecular formula of fragment ion/neutral
PC, Lyso PC PE PS PA PG PI SM CER DGCC DGTS/A MGDG DGDG SQDG
[M + H]+ [M + H]+ [M + H]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + H]+ [M + H]+ [M + H]+ [M + H]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+
PIS NL NL NL NL NL PIS PIS PIS PIS NL NL NL
184 141 185 115 189 277 184 264 104 236 179 341 261
[C5 H15 NO4 P]+ [C2 H8 NO4 P] [C3 H8 NO6 P] [H6 NO4 P] [C3 H12 NO6 P] [C6 H16 NO9 P] [C5 H15 NO4 P]+ [C18 H34 N]+ [C5 H14 NO]+ [C10 H22 NO5 ]+ [C6 H13 NO5 ] [C12 H23 NO10 ] [C6 H15 NO8 S]
Please cite this article in press as: A. Anesi, G. Guella, A fast Liquid Chromatography-Mass Spectrometry methodology for membrane lipid profiling through Hydrophilic Interaction Liquid chromatography, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.01.035
G Model CHROMA-356196; No. of Pages 9 4
ARTICLE IN PRESS A. Anesi, G. Guella / J. Chromatogr. A xxx (2015) xxx–xxx
Fig. 2. Extracted Ion Chromatogram (EIC) of PS standard (m/z 788.5) separated on Merck LiChrospher® 250 × 4.6 mm column (left panel) and on Phenomenex Luna 5 HILIC 250 × 4.6 mm (right panel). Injection volume: 10 L of same standard solution.
miscibility of the solvent and the increased background noise in MS [2]. Common solvent systems consist of a mixture of acetonitrile, water and/or methanol with the addition of buffers as ammonium acetate or formate [2,16–19,23]. Water is essential for HILIC separation process: water is adsorbed on the surface of silica particles and creates a waterenriched layer where the partitioning of solutes is achieved. For this reason, a low amount of water, at least 3-5%, is required to ensure HILIC separation. Therefore, we added 5% of water to acetonitrile, also to ensure a rapid and complete dissolution of ammonium acetate. The use of neat solvent A affected the retention and separation of many lipid classes, particularly of those negatively charged. The phospholipids PA, PG, PI and PS and glycolipid SQDG and DGDG were less retained, and their peaks were broadened by their nonlinear isotherms of distribution. Zwitterionic PC, Lyso PC, PE, SM,
DGTS and DGCC were less affected keeping a symmetric Gaussian shape (data not shown). The addition of 10 mM ammonium acetate increased the retention of negatively charged lipids and improved peak shapes, while retention times of zwitterionic species were slightly reduced. The addition of 0.01% formic acid further increased the retention of negatively charged species and peaks of PA, PI and PS were broadened. The simultaneous addition of 10 mM ammonium acetate and 0.01% formic acid increased retention of negatively charged lipids similarly to the addition of 0.01% formic acid, while zwitterionic species were retained halfway between the pure addition of ammonium acetate and formic acid. Three commonly used solvents B mixtures were tested: water containing 10 mM ammonium acetate; water containing 10 mM ammonium acetate and 0.05% of acetic acid; water containing 10 mM ammonium formate and 0.125% formic acid. The effect of
Fig. 3. BPC in positive ion mode of standard mixture; mass range: 450–1000 m/z; injection volume: 10 L.
Please cite this article in press as: A. Anesi, G. Guella, A fast Liquid Chromatography-Mass Spectrometry methodology for membrane lipid profiling through Hydrophilic Interaction Liquid chromatography, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.01.035
G Model CHROMA-356196; No. of Pages 9
ARTICLE IN PRESS A. Anesi, G. Guella / J. Chromatogr. A xxx (2015) xxx–xxx
5
Table 2 Elution times of lipid standard on Merck LiChrospher® column under HILIC conditions reported in Section 2.4. Average retention time and standard deviation were calculated on three consecutive runs.
Fig. 4. Structures of betaine lipid DGCC (above) and PC (below).
solvent B composition on retention of standard lipid species was low compared to that of solvent A. Given the results and ease of use, solvent A consisted of acetonitrile: water 95:5 containing 10 mM ammonium acetate and solvent B was water containing 10 mM ammonium acetate. 3.2. Analysis of a standard lipid mixture A standard lipid mixture containing fourteen different lipid classes was separated under gradient elution on a Merck LiChrospher® DIOL column (250 × 4.6 mm). Baseline separation of most lipid classes was achieved within 35 min of analysis. Fig. 3 shows the base peak chromatogram (BPC) recorded in positive ion mode where lipids were detected as [M + H]+ or [M + NH4 ]+ ions. The GSL ceramide (CER) was poorly retained under HILIC conditions and eluted at the beginning of elution (tR = 2.78 min), MGDG eluted after 3.72 min, followed by SQDG at 7.48 min, DGTS at 7.71 min, PG at 12.08 min, and DGDG at 13.18 min. The polar head of betaine DGCC is structurally similar to that of PC but lacks the phosphate group (Fig. 4); under our chromatographic setup, PC eluted at 15.80 min, almost one min before DGCC (16.49 min). As no DGCC commercial standards were available, we isolated three different molecular species of DGCC from dinoflagellate P. aciculiferum by using HILIC-UV. DGCC and PC molecular species from P. aciculiferum were recently characterized by our research group [13]: the DGCC fraction was predominantly made up by 36:6 (m/z 772.5), 38:6 (m/z 800.5) and 44:12 (m/z 872.5) molecular species and PC fraction is made up 36:6 (m/z 778.5), 40:11 (m/z 824.5) and 44:12 (m/z 878.5) molecular species. PE eluted between 21 and 22 min, followed by SM at 24.94 min. EIC of lyso-PC 14:0 (m/z 468.5, [M + H]+ ) showed two peaks, a minor one eluting at 27.23 min followed by principal peak eluting at 29.18 min. We assigned the major one to 1-myristoyl-2-hydroxy-sn-glycero-3-PC and the minor one to 1hydroxy-2-myristoyl-sn-glycero-3-PC. This isomerization can be attributed to 1,2 migration of acyl group on the glycerol backbone. This chemical process is well described to occur both in chemical and biochemical conditions [30,31] in several lipid classes among which lyso-galactolipids [30] and lyso-PC [31]. Lastly, PI eluted between 31 and 32 min, followed by PA at 32.23 min and PS at 32.69 min. The list of retention times for each lipid standard is reported in Table 2. For most lipid classes, retention times are reproducible (SD ca 2%) over several runs and only PG and DGDG showed significant variations from run to run (SD ca 4%).
Lipid class
Molecular species
m/z
Ion type
RT (min) ± SD
CER MGDG MGDG SQDG SQDG SQDG SQDG SQDG SQDG SQDG DGTS PG DGDG PC DGCC PE PE PE PE SM Lyso PC Lyso PC PI PI PI PI PI PI PA PS
D18:1/24:0 34:0 36:0 28:0 30:0 32:0 34:3 34:2 36:5 36:4 36:0 36:0 36:0 36:6 36:6 34:2 34:1 36:2 36:1 d18:1/16.0 14:0 14:0 34:3 34:2 34:1 36:4 36:3 36:2 36:2 36:2
651.5 776.5 804.5 756.5 784.5 812.5 834.5 836.5 858.5 860.5 712.5 796.5 966.5 778.5 772.5 716.5 718.5 744.5 746.5 703.5 468.5 468.5 850.5 852.5 854.5 876.5 878.5 880.5 718.5 788.5
[M + H]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + H]+ [M + NH4 ]+ [M + NH4 ]+ [M + H]+ [M + H]+ [M + H]+ [M + H]+ [M + H]+ [M + H]+ [M + H]+ [M + H]+ [M + H]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + H]+
2.78 (± 0.03) 3.72 (± 0.02) 3.59 (± 0.08) 8.30 (± 0.18) 8.03 (± 0.13) 7.78 (± 0.15) 7.52 (± 0.07) 7.51 (± 0.09) 7.34 (± 0.12) 7.22 (± 0.07) 7.71 (± 0.07) 12.08 (± 0.44) 13.18 (± 0.25) 15.80 (± 0.13) 16.49 (± 0.11) 21.61 (± 0.17) 21.53 (± 0.11) 21.18 (± 0.11) 21.10 (± 0.15) 24.94 (± 0.12) 27.23 (± 0.16) 29.18 (± 0.18) 32.03 (± 0.12) 31.95 (± 0.14) 31.93 (± 0.13) 31.73 (± 0.14) 31.68 (± 0.11) 31.65 (± 0.13) 32.23 (± 0.13) 32.69 (± 0.07)
3.3. Intra-class separation of lipid molecular species The partial separation, within a given lipid class, leading to individual molecular species has been reported in HILIC conditions (16, 18;); the effect was due to the different length of fatty acyl chains and their degree of unsaturation. However, the resolution of intraclass separation is low if compared to that of inter-class, which rules the behavior of lipids in HILIC. Fig. 5A shows the NL scanning of 141 Da for PE. Standard L-␣phosphatidylethanolamine extracted from soy bean contained a mixture of different molecular species, mainly PEs 34:2 (m/z 716.5), 34:1 (m/z 718.5), 36:2 (m/z 744.5) and 36:1 (m/z 746.5). Under our experimental setup, the intra-class separation is consistent with [16,18]; the retention time is lowered by the increase in acyl chain length and the decrease in the unsaturation degree. The length of the acyl chains with a given unsaturation degree plays a dominant role on intra-class separation with respect to the unsaturation number within a given chain length. As depicted in Fig. 5B, which represents EIC of single molecular species from Fig. 2, 38:4 PE (RT:20.26 ) and 38:5 PE (20.35 ) elute before 36:1 PE and 36:2 PE (RT: 21.24 and 21.30 respectively) and before 34:1 PE and 34:2 PE (RT: 21.41 and 21.52 respectively). The effect due to length of acyl chains was predominant on that of unsaturation degree. 36:1 PE eluted before 36:2 PE and, similarly, 34:1 PE eluted before 34:2 PE (Fig. 5B). The same effect was observed for other lipid classes. In Fig. 6 are shown EIC of 34:3 PI (m/z 850.5; RT: 31.91 min), 34:2 PI (m/z 852.5; RT: 31.82 min), 34:1 PI (m/z 854.5; RT: 31.81 min), 36:4 PI (m/z 876.5; RT: 31.51 min), 36:3 PI (m/z 878.5; RT: 31.57 min) and 36:2 PI (m/z 880.5; RT: 31.51 min), where the effect of chain length and degree of unsaturation are clear.
Please cite this article in press as: A. Anesi, G. Guella, A fast Liquid Chromatography-Mass Spectrometry methodology for membrane lipid profiling through Hydrophilic Interaction Liquid chromatography, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.01.035
G Model CHROMA-356196; No. of Pages 9 6
ARTICLE IN PRESS A. Anesi, G. Guella / J. Chromatogr. A xxx (2015) xxx–xxx
Fig. 5. A: NL scanning of 141 Da for PE in the range 600-1000 m/z. B: EIC of 34:1 PE (m/z 718.5), 34:2 PE (m/z 716.5), 36:1 PE (m/z 746.5), 36:2 PE (m/z 744.5), 38:4 PE (m/z 768.5) and 38:5 PE (m/z 766.5) contained in soy L-␣-phosphatidylethanolamine from Avanti Polar Lipids.
Fig. 6. EIC of 34:3 PI (m/z 850.5), 34:2 PI (m/z 852.5), 34:1 PI (m/z 854.5), 36:4 PI (m/z 876.5), 36:3 PI (m/z 878.5) and 36:2 PI (m/z 880.5) from soy L-␣-phosphatidylinositol from Avanti Polar Lipids.
Please cite this article in press as: A. Anesi, G. Guella, A fast Liquid Chromatography-Mass Spectrometry methodology for membrane lipid profiling through Hydrophilic Interaction Liquid chromatography, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.01.035
G Model CHROMA-356196; No. of Pages 9
ARTICLE IN PRESS A. Anesi, G. Guella / J. Chromatogr. A xxx (2015) xxx–xxx
Fig. 7. A: BPC of Peridinium cinctum raw extract; B: BPC of Jaoa bullata; C: BPC of Vitis vinifera cv. Corvina (ripening stage).
Please cite this article in press as: A. Anesi, G. Guella, A fast Liquid Chromatography-Mass Spectrometry methodology for membrane lipid profiling through Hydrophilic Interaction Liquid chromatography, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.01.035
7
G Model CHROMA-356196; No. of Pages 9
ARTICLE IN PRESS A. Anesi, G. Guella / J. Chromatogr. A xxx (2015) xxx–xxx
8
Fig. 8. BPC of mixture of Peridinium cinctum, Jaoa bullata and Vitis vinifera cv. Corvina raw extracts.
3.4. Method application
4. Conclusion
To test the new method, lipids of P. cinctum, J. bullata and V. vinifera cv. Corvina were analyzed. As no pre-purification step was applied, raw extracts may contain chlorophylls, carotenoids, neutral acylglycerols lipids (TAG and DAG) and sterols. As our method is particularly focused on membrane lipids, which are more interesting from a phylogenetic point of view than acylglycerols, all these bioorganic compounds were not followed in this work. Furthermore, HILIC is surely not the best method for their analysis, as there are poorly retained (data not shown). In Fig. 7, the base peak chromatograms of samples detected in positive ion mode are shown. P. cinctum contained mainly MGDG, SQDG, DGDG, PC, DGCC and traces of PE and PI (Fig. 7A); the lipid profile of P. cinctum is very similar to that of its closely related species P. aciculiferum recently reported by Flaim and co-workers [13]. The green algae J. bullata contained mainly MGDG, SQDG, DGTS/A, DGDG and traces of PC (Fig. 7B) as also depicted by [25]. Vitis vinifera cv. Corvina (ripening stage) contained MGDG, DGDG, PC, PE, PI and PA (Fig. 7C). The identity of each class is confirmed by specific NL and PIS experiments that are reported in Supplementary Materials. In Fig. 8 is shown the BPC of mixture containing similar amount of P. cinctum, J. bullata and V. vinifera cv. Corvina raw extracts; despite the sample complexity, lipid classes are almost base peakseparated. Only a partial overlap of some PG and DGDG and PC and DGCC is observed, which is the result of intra-class separation of lipids due to wide range of different lipids, with diverse acyl chain length and unsaturation degree (see Supplementary Materials).
Hydrophilic interaction liquid chromatography (HILIC), commonly used for separation of polar metabolites is here used for separation of lipid classes, generally obtained by normal-phase chromatography. The new HILIC method provides a fast and comprehensive tool for the baseline separation of fourteen different classes of membrane lipids on a diol column and is fully compatible with full scan and NL/PIS ESI-MS analysis. Given the broad diversity of lipid classes covered in this survey, this method can be applied for the preliminary investigation of the lipidomes from different sources, such as archea, prokaryotes, micro-and macro-eukaryotic organisms such as algae, fungi, higher and animals. In terms of chemotaxonomic significance, the methodology here described is much more powerful than the classical membrane lipid analysis based on RP chromatography. In fact, it is much more valuable to establish the number and nature of lipid classes (inter-class analysis) present in an environmental sample than the detailed distribution within a given lipid class (intra-class profile). Acknowledgments We want to thanks a) Dr. Giovanna Flaim from Edmund Mach Foundation (San Michele all’Adige, Italy) for Peridinium aciculiferum and Peridinium cinctum cell cultures; b) Dr. Abdelkader Bensalem for providing pure SQDG fraction extracted from Codium bursa; c) Dr. Marco Cantonati and Dr. Daniel Spitale for Jaoa bullata samples; Prof. Mario Pezzotti and Dr. Sara Zenoni for providing Vitis vinifera
Please cite this article in press as: A. Anesi, G. Guella, A fast Liquid Chromatography-Mass Spectrometry methodology for membrane lipid profiling through Hydrophilic Interaction Liquid chromatography, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.01.035
G Model CHROMA-356196; No. of Pages 9
ARTICLE IN PRESS A. Anesi, G. Guella / J. Chromatogr. A xxx (2015) xxx–xxx
cv. Corvina samples. The authors also acknowledge support from MIUR (PRIN 2010-2011 grant). Appendix A. Supplementary data
[16]
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.chroma. 2015.01.035.
[17]
References
[18]
[1] M.R. Wenk, The emerging field of lipidomics, Nat. Rev. Drug Discov. 4 (2005) 594–610, http://dx.doi.org/10.1038/nrd1776. [2] M. Lísa, E. Cífková, M. Holˇcapek, Lipidomic profiling of biological tissues using off-line two-dimensional high-performance liquid chromatography-mass spectrometry, J. Chromatogr. A. 1218 (2011) 5146–5156, http://dx.doi.org/ 10.1016/j.chroma.2011.05.081. [3] R. Taguchi, T. Houjou, H. Nakanishi, T. Yamazaki, M. Ishida, M. Imagawa, T. Shimizu, Focused lipidomics by tandem mass spectrometry, J. Chromatogr. B. 823 (2005) 26–36, http://dx.doi.org/10.1016/j.jchromb.2005.06.005. [4] Christie W.W. 2011. http://lipidlibrary.aocs.org/index.html. American Oil Chemists Society. [5] W.W. Christie, Chromatographic analysis of phospholipids, Z. Lebensm. Unters Forsch. 185 (1985) 171–182. [6] S. Milne, P. Ivanova, J. Forrester, H.A. Brown, Lipidomics: an analysis of cellular lipids by ESI-MS, Methods 39 (2006) 92–103. [7] S. Mawatari, Y. Okuma, T. Fujino, Separation of intact plasmalogen and all other phospholipids by a single run of high-performance liquid chromatography, Anal. Biochem. 370 (2007) 54–59. [8] L.-Q. Pang, Q.-L. Liang, Y-M. Wang, L. Ping, G.A. Luo, Simultaneous determination and quantification of seven major phospholipid classes in human blood using normal-phase liquid chromatography coupled with electrospray mass spectrometry and the application in diabetes nephropathy, J. Chromatogr. B 869 (1–2) (2008) 118–125. [9] H. Nie, R. Liu, Y. Yang, Y. Bai, Y. Guan, D. Qian, T. Wang, H. Liu, Lipid profiling of rat peritoneal surface layers by online normal- and reversed phase 2D LC QToF-MS, J. Lipid Res. 51 (2010), 2833-2344. [10] G. Guella, R. Frassanito, I. Mancini, A new solution for an old problem: the regiochemical distribution of the acyl chains in galactolipids can be established by electrospray ionization tandem mass spectrometry, Rapid Commun. Mass Spectrom. 17 (2003) 1982–1994. [11] G. Flaim, U. Obertegger, G. Guella, Changes in galactolipid composition of the cold freshwater dinoflagellate Borghiella dodgei in response to temperature, Hydrobiologia 698 (2012) 285–293, http://dx.doi.org/10.1007/ s10750-012-1070-8. [12] C.W.N. Damen, G. Isaac, J. Langridge, T. Hankmeier, R.J. Vreeken, Enhanced lipid isomer separation in human plasma using reversed-phase UPLC with ion mobility /high-resolution MS detection, J. Lipid Res. 55 (8) (2014) 1772–1783. [13] G. Flaim, U. Obertegger, A. Anesi, G. Guella, Temperature-induced changes in lipid biomarkers and mycosporine-like amino acids in the psychrophilic dinoflagellate Peridinium aciculiferum, Freshwater Biol. (2014), http://dx.doi. org/10.1111/fwb.12321. [14] J. Hummel, S. Segu, Y. Li, S. Irgang, J. Jueppner, P. Giavalisco, Ultra performance liquid chromatography and high resolution mass spectrometry for the analysis of plant lipids, Front. Plant Sci. 2 (2011) 54. [15] X. Gao, Q. Zhang, D. Meng, G. Issac, R. Zhao, T.L. Fillmore, R.K. Chu, J. Zhou, K. Tang, Z. Hu, R.J. Moore, R.D. Smith, M.G. Katze, T.O. Metz, A
[19]
[20] [21]
[22]
[23]
[24]
[25]
[26]
[27]
[28] [29]
[30]
[31]
9
reverse-phase capillary ultra-performance liquid chromatography-mass spectrometry (UPLC-MS) method for comprehensive top-down/bottom-up lipid profiling, Anal. Bioanal. Chem. 402 (9) (2012) 2923–2933, http://dx.doi.org/ 10.1007/s00216-012-5773-5. M. Schwalbe-Herrmann, J. Willmann, D. Leibfritz, Separation of phospholipid classes by hydrophilic interaction chromatography detected by electrospray ionization mass spectrometry, J. Chromatogr. A 1217 (32) (2010) 5179–5183. L. Zheng, R. T’Kind, S. Decuypere, S.J. von Freyend, G.H. Coombs, D.G. Watson, Profiling of lipids in Leishmania donovani using hydrophilic interaction chromatography in combination with Fourier transform mass spectrometry, Rapid Commun. Mass Spectrom. 24 (2010) 2074–2082. C. Zhu, A. Dane, G. Spijksma, M. Wang, J. van der Greef, G. Luo, T. Hankermeier, R.J. Vreeken, An efficient hydrophilic interaction liquid chromatography separation of 7 phospholipid classes based on a diol column, J. Chromatogr. A 1220 (2012) (2011) 26–34, http://dx.doi.org/10.1016/j.chroma.2011.11.034. P. Donato, F. Cacciola, F. Cichello, M. Russo, P. Dugo, L. Mondello, Determination of phospholipids in milk samples by means of hydrophilic interaction liquid chromatography coupled with evaporative light scattering and mass spectrometry detection, J. Chromatogr. A 1218 (2011) 6476–6482. V.M. Dembitsky, Betaine ether-linked glycerolipids: chemistry and biology, Prog. Lipid Res. 35 (1) (1996) 1–51. J. Brandsma, E.C. Hopmans, C.P.D. Brussard, H.J. Witte, S. Schouten, J.S. Sinninghe Damsté, Spatial distribution of intact polar lipids in North Sea surface waters: relationship with environmental conditions and microbial community composition, Limnol. Oceanogr. 57 (4) (2012) 959–973. T. Kind, j.K. Meissen, D. Yang, F. Nocito, A. Vaniya, Y.-S. Cheng, J.S. VanderGheynst, O. Fiehn, Qualitative analysis of algal secretions with multiple mass spectrometric platforms, J. Chromatogr. A 1244 (2012) 139–147, http://dx.doi.org/10.1016/j.chroma.2012.04.074. Y. Okazaki, H. Otsuki, T. Narisawa, M. Kobayashi, S. Sawai, Y. Kamide, M. Kusano, T. Aoki, M.Y. Hirai, K. Saito, A new class of plant lipid is essential for protection against phosphorus depletion, Nat. Commun. 4 (2013(a)) 1510, http://dx.doi.org/10.1038/ncomms2512. K.J. Popendorf, H.F. Fredricks, B.A.S. Van Mooy, Molecular ion-independent quantification of polar glycerolipids classes in marine plankton using triple quadrupole MS, Lipids 48 (2013) 185–195. J. Mares, M. Cantonati, G. Guella, D. Spitale, The benthic chlorophyte genus Jaoa (Ulvales), a putative China endemic, in Lake Garda, Italy: ecology, taxonomy, and molecular analyses, Freshwater Sci. 33 (2) (2014), http://dx.doi.org/10.1086/675860. S. Pilati, D. Brazzale, G. Guella, A. Milli, C. Ruberti, F. Biasioli, M. Zottini, C. Moser, The onset of grapevine berry ripening is characterized by ROS accumulation and lipoxygenase-mediated membrane perosidation in the skin, BMC Plant Biol. 14 (2014) 87. J. Folch, M. Lees, G.H. Sloane Stanley, A simple method for the isolation and purification of total lipids from animal tissues, J. Biol. Chem. 226 (1957) 497–509. E.G. Bligh, W.J. Dyer, A rapid method of total lipid extraction and purification, Can. J. Biochem. Physiol. 37 (1959) 911–917. Y. Okazaki, Y. Kamide, M.Y. Hirai, K. Saito, Plant lipidomics based on hydrophilic interaction chromatography coupled to ion trap time-of-flight mass spectrometry, Metabolomics 9 (2013) 121–131. I. Mancini, G. Guella, A. Defant, M.L. Candenas, C.P. Armesto, D. Depentori, F. Pietra, Polar metabolites of the tropical green seaweed Caulerpa taxifolia which is spreading in the Mediterranean sea: glycoglycerolipids and stable enols (␣keto esters), Helv. Chim. Acta 81 (1998) 1681–1691. M. Croset, N. Brossard, A. Polette, M. Lagarde, Characterization of plasma unsaturated lysophosphatidylcholines in human and rat, Biochem. J. 345 (2000) 61–67.
Please cite this article in press as: A. Anesi, G. Guella, A fast Liquid Chromatography-Mass Spectrometry methodology for membrane lipid profiling through Hydrophilic Interaction Liquid chromatography, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.01.035
ARTICLE IN PRESS
CHROMA-356196; No. of Pages 9
Journal of Chromatography A, xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma
A fast Liquid Chromatography-Mass Spectrometry methodology for membrane lipid profiling through Hydrophilic Interaction Liquid chromatography Andrea Anesi a,∗ , Graziano Guella a,b a b
Bioorganic Chemistry Lab, Department of Physics, University of Trento, Via Sommarive 14, 38123 Povo (Trento), Italy Biophysical Institute, CNR, Via alla Cascata 56/C, 38123 Povo (Trento), Italy
a r t i c l e
i n f o
Article history: Received 1 September 2014 Received in revised form 12 January 2015 Accepted 13 January 2015 Available online xxx Keywords: Hydrophilic-interaction liquid chromatography (HILIC)–diol column–lipid separation–mass spectrometry
a b s t r a c t In this paper, we report the development of a new method based on HILIC-ESI-MS for the separation of several different membrane lipid classes and their detection on a triple quadrupole mass spectrometer using Precursor Ion (PIS) and Neutral Loss (NL) scanning in positive ion mode. Four different columns were tested for their ability to separate, under different conditions, a mixture of 14 lipid standards containing 7 glycerophospholipids (GPL), 2 glycosphyngolipids (GSL), 3 glycolipids (GL) and 2 betaine lipids (BL). The best separation was obtained using a Lichrosphere DIOL column as stationary phase and water (10 mm ammonium acetate)/acetonitrile gradient elution as mobile phase which allows the separation of the 14 lipid classes within 35 min runtime. Our method was successfully tested for the separation and analysis of crude lipid extracts obtained from a green alga (Jaoa bullata), a dinoflagellate (Peridinium cinctum) and a plant (Vitis vinifera cv. Corvina) © 2015 Elsevier B.V. All rights reserved.
1. Introduction Lipidomics, or “the system-level analysis and characterization of lipids and factors that interacts with them” [1,2], is an emerging area within the field of “omics” sciences [3]. The term “lipids” defines a vast array of chemically diverse substances, which are mostly water-insoluble and contain fatty acids or their derivatives [4]. Lipids have been categorized into eight major classes: fatty acids, glycerolipids, glycerophospholipids, polyketides, prenyl lipids, saccharolipids, sphingolipids and sterols [5]. Lipids are involved in many cellular processes and aberrant lipid metabolism is a determinant of the onset of some important diseases, such as diabetes, atherosclerosis, obesity, Alzheimer’s disease and some cancer types [1]. It is thus clear why the comprehensive analysis of the lipidome, the complete set of lipids within a cell, tissue or organism, is of great interest. Various analytical methodologies have been applied for lipid analysis, such as thin-layer-chromatography (TLC), highperformance liquid chromatography (HPLC) coupled with mass spectrometry (HPLC-MS), diode array (HPLC-DAD) or evaporative light scattering detector (HPLC-ELSD), nuclear magnetic resonance
∗ Corresponding author. Tel.: +39 0461281547. E-mail address: [email protected] (A. Anesi).
(NMR) and gas-chromatography (GC) coupled to mass spectrometry and/or flame ionization detector (GC-MS/FID). The use of HPLC has been widely exploited as it is possible to achieve a fast and consistent separation of many components of a mixture, and given the large sample loading compared to TLC, the ease of automation and the hyphenation with a wide array of detectors are advantageous [1,5,6]. Separation of lipids can be achieved either by normal phase (NP) or reverse-phase-liquid chromatography (RP-LC), as depicted in Fig. 1. NP methods exploit silica, diol or amine-bonded stationary phases in combination with mixtures of hexane/isopropanol/water as mobile phases; under NP-LC lipids are separated according to different polarity of the heads [5,7–9]. On the other hand, RP-based separation on octasilyl (C8)- and octadecylsilyl (C18)- derivatized stationary phases is achieved in combination with solvent systems usually composed of mixtures of methanol, acetonitrile and water. Under RP-based conditions, the separation of membrane lipids mainly relies on the general chemical features of their acyl (or alkyl in plasmalogens) chains such as length, number and position of unsaturation [5,9–13]. For this reason, RP-and NP-based separations are considered complementary because the former is particularly suitable to establish the diversity of molecular lipid species within the same lipid class (intra-class differentiation) whilst the latter can be exploited to establish the total lipid species belonging to different membrane lipids classes in a given sample
http://dx.doi.org/10.1016/j.chroma.2015.01.035 0021-9673/© 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: A. Anesi, G. Guella, A fast Liquid Chromatography-Mass Spectrometry methodology for membrane lipid profiling through Hydrophilic Interaction Liquid chromatography, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.01.035
G Model CHROMA-356196; No. of Pages 9 2
ARTICLE IN PRESS A. Anesi, G. Guella / J. Chromatogr. A xxx (2015) xxx–xxx
phosphatidylcholine (PC) and lyso-PC, phosphatidylethanolamine (PE), phosphatidylglycerol (PG), phosphatidylinositol (PI), phosphatidylserine (PS)), 2 GSL (sphingomyelin (SM) and ceramide CER), 3 GLs (MGDG, DGDG and SQDG) and 2 BLs (DGTS and DGCC). The separation of lipids was tested by using four different columns with acetonitrile, water and ammonium acetate as mobile phases and then further optimized on the column giving the best chromatographic results. Lipid classes were separated within a 35 min-run time, showing a good repeatability for most of lipid classes. The method was further tested on complex lipid extracts, such as the dinoflagellate Peridinium cinctum, the green alga Jaoa bullata and the berries of Vitis vinifera. 2. Materials and methods Fig. 1. Chromatographic methods exploited in lipid analysis: lipids can be separated in NP and HILIC according to the different polarity of their heads or in RP according to the hydrophobicity of their acyl chains.
(inter-classes differentiation) [9]. Of course, since the partition processes determining the chromatographic separations are multiple and complex, using an RP-column is often possible to roughly separate (especially when the lipid classes have quite different polarity) not only the lipid species belonging to a given class but also lipid species belonging to different lipid classes [13–15]. In recent years, lipid analysis has benefitted from the development of hydrophilic interaction liquid chromatography (HILIC), which allows one to overcome (i) the scarce reproducibility observed in NP-LC due to the low miscibility of hexane/isopropanol/water solvent systems and (ii) the low ESI compatibility of NP-based mobile phases. The principle of HILIC lipid separation is the same as for NP-LC but with the difference that mobile phases are similar to those used for RP-LC. An HILIC stationary phase is hydrophilic and/or charged, with solvent systems consisting of acetonitrile/methanol/water (by eventual addition of an ion-pairing reagent as volatile buffer). Much of the lipid separation using HILIC chromatography has been done for the separation of glycerophospholipids (GPLs) and glycerosphingolipids (GSLs) classes from various matrices, as tissues, protozoans or milk [16–19]. Bacteria, eukaryotic microorganisms and higher plants are sources of other lipid classes such as the glycolipids (GL) mono- and digalactosyldiacylglycerols (MGDG and DGDG) and sulfoquinovosyldiacylglycerol (SQDG), and the betaine (BL) etherlinked glycerolipids diacylglyceryl-hydroxymethoxy-trimethyl-alanine (DGTA), diacylglyceryl-trimethyl-homoserine (DGTS) and diacylglyceryl-carboxymethylcholine (DGCC) [20–26]. These lipids are co-extracted with GPL using common extraction procedures [22,27,28]. In few studies, betaine lipids and glycolipids were separated from GPL using HILIC chromatography in combination with hexane, isopropanol and water solvent systems [21,24] while in plant lipidomics, glycolipids were separated from phospholipids using mixtures of methanol, acetonitrile and water [23,29]. In a recent work, Kind and co-workers developed an HILIC method based on acetonitrile, water, ammonium acetate and acetic acid for the qualitative determination by high resolution-MS of lipids secreted by algae [22]. Lipid classes, mainly GPLs phosphatidylcholine and phosphatidylinositol, the glycolipid SQDG, the betaine DGTS, and triacylglycerols, were separated within 20 minrun time but many species were still coeluted and, therefore, a longer run time would be needed for a clearer separation. Here we present a simple but detailed diol-based HILICESI-MS method for the separation and detection of fourteen different lipid classes that constitute cellular membranes of various micro- and macrorganisms: 7 GPLs (phosphatidic acid (PA),
2.1. Chemicals Analytical grade acetonitrile, HPLC grade chloroform and reagent grade methanol were purchased from VWR (VWR International PBI, Milan, Italy); deionized water, filtered at 0.2 m, was obtained from Elix Water Purification System (Merck Millipore, Billerica, MA, USA). Reagent grade ammonium acetate was purchased from Rudi Pont (Chimica Rudi Pont, Torino, Italy); reagent grade ammonium formate was purchased from Carlo Erba (Carlo Erba Reagents, Milan, Italy); reagent grade acetic acid was purchased from Merck (Merck KGaA, Darmstadt, Germany) and LC-MS grade formic acid was purchased from Fisher Scientific (Fisher Scientific, Illkirch, France). 2.2. Commercial standards 1,2-Dioleoyl-sn-glycero-3-phospho-L-serine (PS sodium salt), 1,2-Dioleoyl-sn-glycero-3-phosphate (PA sodium salt), L-␣-phosphatidylinositol mixture extracted from soy (PI sodium salt), L-␣-phosphatidylethanolamine mixture extracted from soy (PE sodium salt), 1,2-Distearoyl-sn-glycero-3phospho-(1 -rac-glycerol) (PG sodium salt), 1,2-Dilinolenoylsn-glycero-3-phosphocholine (PC), 1-Myristoyl-2-hydroxy-snglycero-3-phosphocholine (lyso-PC), N-palmytoyl-d-erythrosphingosylphosphorylcholine (SM), N-lignoceroyl-D-erythrosphingosine (CER) and 1,2-Dipalmitoyl-sn-glycero-3-O-4 -(N,N,Ntrimethyl)-homoserine (DGTS) were purchased from Avanti Polar Lipids (Avanti Polar Lipids Inc., Alabaster, AL, USA). Saturated monogalactosyldiacylglycerols- (MGDG) and digalactosyldiacylglycerols (DGDG) were purchased from Matreya (Matreya LLC, Pleasant Gap, PA, USA). 2.3. Extraction of non-commercially available standard compounds Sulfoquinovosyldiacylglycerols (SQDG) were isolated from Codium bursa raw extract by Solid Phase Extraction (SPE) on a 2 g RP 18 cartridge (Phenomenex, Torrance, CA, USA)using methanolwater 85:15. The DGCC were isolated from Peridinium aciculiferum raw extract by SPE on a 2 g silica cartridge (Phenomenex, Torrance, CA, USA). A gradient elution of hexane/ethyl acetate was used to remove non-polar lipids, as sterols and triacylglycerols, and pigments; then polar lipids were eluted using 45 ml of methanol acidified with 0.1% of formic acid. The fraction was dried under rotary evaporation (Büchi Labortechnik AG, Flawil, Swiss), re-dissolved into chloroform and subsequently injected in HPLC-HILIC-UV system to collect single molecular species. The 1-myristoyl-2-docosahexaenoyl-sn-glycero-3-Odiacylglyceryl-carboxymethylcholine was selected as a standard.
Please cite this article in press as: A. Anesi, G. Guella, A fast Liquid Chromatography-Mass Spectrometry methodology for membrane lipid profiling through Hydrophilic Interaction Liquid chromatography, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.01.035
G Model CHROMA-356196; No. of Pages 9
ARTICLE IN PRESS A. Anesi, G. Guella / J. Chromatogr. A xxx (2015) xxx–xxx
3
2.4. Optimization of HILIC separation
2.6. Sample preparation for method application
For the optimization of HILIC separation, 4 different columns were tested: Luna 3 m HILIC (150 × 3 mm I.D., particle size: ˚ and a Luna 5 m HILIC (250 × 4.6 mm 3 m, pore diameter: 200 A) ˚ both purchased I.D., particle size: 5 m, pore diameter: 200 A) from Phenomenex (Phenomenex, Torrance, CA, USA), SeQuant® ZIC® -pHILIC (150 × 4.6 mm I.D., particle size: 5 m, pore diame˚ and LiChrospher® DIOL (250 × 4.6 mm I.D., particle size: ter: 100 A) ˚ both purchased from Merck (Merck 5 m, pore diameter: 100 A) KGaA, Darmstadt, Germany). Four combinations of solvent A (acetonitrile: water 95:5) were tested: neat, containing 10 mM ammonium acetate, containing 0.01% formic acid and containing both 10 mM ammonium acetate and 0.01% formic acid. Three different solvent B combinations were tested: water containing 10 mM ammonium acetate, water containing 10 mM ammonium acetate and 0.05% of acetic acid; water containing 10 mM ammonium formate and 0.125% formic acid. Both isocratic and gradient elution were tested.
To test our method, crude lipid extracts from various organisms were prepared. About 500 mg of a) cell culture of the dinoflagellate Peridinium cinctum, b) green alga Jaoa bullata and c) de-seeded grape berries (ripening stage) of Vitis vinifera cv. Corvina, powdered in liquid nitrogen were extracted according to [27]. Briefly, 10 mL of chloroform/methanol 2:1 (v/v) were added and samples were sonicated for 15 min in an ultrasonic bath (Sonorex Super, Bandelin electronics, Berlin, Germany). The samples were then centrifuged at 3000 × g for 10 min at room temperature and the organic phase (bottom layer) was collected into a new glass tube. The extraction procedure was repeated three times. Finally, the organic phase was filtered by using glass filters under vacuum and reduced to dryness on a rotary evaporation (Büchi Labortechnik AG, Flawil, Swiss). Total lipids were re-suspended in 200 L of chloroform and 10 L were used for MS and MS/MS analysis. Lastly, a mixture containing equal amounts of P. cinctum, J. bullata and V. vinifera cv. Corvina raw extracts was also analyzed to test our method in a sample where complexity derives from lipid classes and lipid species.
2.5. High performance liquid chromatography-electrospray ionization-triple quadrupole-mass spectrometry (HPLC-ESI-QqQ-MS) The LC analyses were carried out on a Shimadzu Prominence LC-20A system controlled by LCSolution 1.23 A software (Shimadzu Corporation, Milan, Italy). A LiChrospher® DIOL (250 × 4.6 mm, par˚ purchased from Merck ticle size: 5 m, pore diameter: 100 A) (Merck KGaA, Darmstadt, Germany) was used for lipid separation. Solvent A consisted of acetonitrile: water 95:5 containing 10 mM ammonium acetate, solvent B consisted of water containing 10 mM ammonium acetate. The separation was carried out at room temperature with a flow of 0.9 mL/min; injection volume was 10 L of standard dissolved in chloroform. The system was maintained at 100% A for 2 min; then B was increased to 5% over 23 min, then to 10% B over 5 min and then kept to 10% B for 5 min. The column was re-equilibrated to starting conditions for 10 min prior to new injection. The MS acquisitions were performed on an API 3000 triple quadrupole controlled by Analyst software 1.4.2 (ABI Sciex, Framingham, MA, USA) and equipped with an electrospray ionization source operated in positive ion mode. The following parameters were used: nebulizer gas: 8 psi, curtain gas: 8 psi, ion spray voltage: +5 kV, temperature: 300 ◦ C; declustering potential: 40 V; focusing potential: 300 V; entrance potential: 15 V; mass range: 450–1000 m/z. Specific Precursor Ion (PIS) and Neutral Loss (NL) scanning in positive ion mode were also conducted using the parameters reported in Table 1. Parameters for PIS and NL scanning were taken from standard compound analysis.
3. Results and discussion 3.1. Optimization of HILIC separation Separation of fourteen lipid species was tested with 4 different columns different conditions. The Phenomenex Luna 5 HILIC (250 × 4.6 mm) and Merck LiChrospher® DIOL (250 × 4.6 mm) columns showed similar performances but PS peak shape was better on Merck LiChrospher® column, as shown in Fig. 2; for this reason, the column was selected for standard analysis. Under isocratic elution in 100%A, lipids eluted completely after 60 min (data not shown); therefore, we opted for a gradient elution, which usually ensures better repeatability and resolution, more narrow peaks and faster analysis time [18]. Gradient elution was modified to ensure rapid and complete separation of standards: starting conditions (100%A) were maintained isocratically for 2 min and then %B was first increased to 5% over 23 min, then to 10% over 5 min and kept at 10% B for 5 min. A total of 15% of water was sufficient for the complete elution of all species and no lipid carryover was detected also in real samples (data not shown). 10 min equilibration time at starting conditions resulted to be enough to maintain repeatability. The mobile phase composition is very important for HILIC separation as highlighted by Lísa et al., [2]. The use of hexane/isopropanol/water mixtures provides the best results in terms of resolution but the reproducibility is scarce, due to low
Table 1 Mass spectrometry conditions in positive ion mode for molecular-ion analysis of fourteen classes of polar lipids using a HILIC-ESI-QqQ-MS.
Phospholipids
Sphingolipids Betaine lipids Glycolipids
Lipid class
Molecular ion
Scan type
Mass of fragment ion/neutral
Molecular formula of fragment ion/neutral
PC, Lyso PC PE PS PA PG PI SM CER DGCC DGTS/A MGDG DGDG SQDG
[M + H]+ [M + H]+ [M + H]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + H]+ [M + H]+ [M + H]+ [M + H]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+
PIS NL NL NL NL NL PIS PIS PIS PIS NL NL NL
184 141 185 115 189 277 184 264 104 236 179 341 261
[C5 H15 NO4 P]+ [C2 H8 NO4 P] [C3 H8 NO6 P] [H6 NO4 P] [C3 H12 NO6 P] [C6 H16 NO9 P] [C5 H15 NO4 P]+ [C18 H34 N]+ [C5 H14 NO]+ [C10 H22 NO5 ]+ [C6 H13 NO5 ] [C12 H23 NO10 ] [C6 H15 NO8 S]
Please cite this article in press as: A. Anesi, G. Guella, A fast Liquid Chromatography-Mass Spectrometry methodology for membrane lipid profiling through Hydrophilic Interaction Liquid chromatography, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.01.035
G Model CHROMA-356196; No. of Pages 9 4
ARTICLE IN PRESS A. Anesi, G. Guella / J. Chromatogr. A xxx (2015) xxx–xxx
Fig. 2. Extracted Ion Chromatogram (EIC) of PS standard (m/z 788.5) separated on Merck LiChrospher® 250 × 4.6 mm column (left panel) and on Phenomenex Luna 5 HILIC 250 × 4.6 mm (right panel). Injection volume: 10 L of same standard solution.
miscibility of the solvent and the increased background noise in MS [2]. Common solvent systems consist of a mixture of acetonitrile, water and/or methanol with the addition of buffers as ammonium acetate or formate [2,16–19,23]. Water is essential for HILIC separation process: water is adsorbed on the surface of silica particles and creates a waterenriched layer where the partitioning of solutes is achieved. For this reason, a low amount of water, at least 3-5%, is required to ensure HILIC separation. Therefore, we added 5% of water to acetonitrile, also to ensure a rapid and complete dissolution of ammonium acetate. The use of neat solvent A affected the retention and separation of many lipid classes, particularly of those negatively charged. The phospholipids PA, PG, PI and PS and glycolipid SQDG and DGDG were less retained, and their peaks were broadened by their nonlinear isotherms of distribution. Zwitterionic PC, Lyso PC, PE, SM,
DGTS and DGCC were less affected keeping a symmetric Gaussian shape (data not shown). The addition of 10 mM ammonium acetate increased the retention of negatively charged lipids and improved peak shapes, while retention times of zwitterionic species were slightly reduced. The addition of 0.01% formic acid further increased the retention of negatively charged species and peaks of PA, PI and PS were broadened. The simultaneous addition of 10 mM ammonium acetate and 0.01% formic acid increased retention of negatively charged lipids similarly to the addition of 0.01% formic acid, while zwitterionic species were retained halfway between the pure addition of ammonium acetate and formic acid. Three commonly used solvents B mixtures were tested: water containing 10 mM ammonium acetate; water containing 10 mM ammonium acetate and 0.05% of acetic acid; water containing 10 mM ammonium formate and 0.125% formic acid. The effect of
Fig. 3. BPC in positive ion mode of standard mixture; mass range: 450–1000 m/z; injection volume: 10 L.
Please cite this article in press as: A. Anesi, G. Guella, A fast Liquid Chromatography-Mass Spectrometry methodology for membrane lipid profiling through Hydrophilic Interaction Liquid chromatography, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.01.035
G Model CHROMA-356196; No. of Pages 9
ARTICLE IN PRESS A. Anesi, G. Guella / J. Chromatogr. A xxx (2015) xxx–xxx
5
Table 2 Elution times of lipid standard on Merck LiChrospher® column under HILIC conditions reported in Section 2.4. Average retention time and standard deviation were calculated on three consecutive runs.
Fig. 4. Structures of betaine lipid DGCC (above) and PC (below).
solvent B composition on retention of standard lipid species was low compared to that of solvent A. Given the results and ease of use, solvent A consisted of acetonitrile: water 95:5 containing 10 mM ammonium acetate and solvent B was water containing 10 mM ammonium acetate. 3.2. Analysis of a standard lipid mixture A standard lipid mixture containing fourteen different lipid classes was separated under gradient elution on a Merck LiChrospher® DIOL column (250 × 4.6 mm). Baseline separation of most lipid classes was achieved within 35 min of analysis. Fig. 3 shows the base peak chromatogram (BPC) recorded in positive ion mode where lipids were detected as [M + H]+ or [M + NH4 ]+ ions. The GSL ceramide (CER) was poorly retained under HILIC conditions and eluted at the beginning of elution (tR = 2.78 min), MGDG eluted after 3.72 min, followed by SQDG at 7.48 min, DGTS at 7.71 min, PG at 12.08 min, and DGDG at 13.18 min. The polar head of betaine DGCC is structurally similar to that of PC but lacks the phosphate group (Fig. 4); under our chromatographic setup, PC eluted at 15.80 min, almost one min before DGCC (16.49 min). As no DGCC commercial standards were available, we isolated three different molecular species of DGCC from dinoflagellate P. aciculiferum by using HILIC-UV. DGCC and PC molecular species from P. aciculiferum were recently characterized by our research group [13]: the DGCC fraction was predominantly made up by 36:6 (m/z 772.5), 38:6 (m/z 800.5) and 44:12 (m/z 872.5) molecular species and PC fraction is made up 36:6 (m/z 778.5), 40:11 (m/z 824.5) and 44:12 (m/z 878.5) molecular species. PE eluted between 21 and 22 min, followed by SM at 24.94 min. EIC of lyso-PC 14:0 (m/z 468.5, [M + H]+ ) showed two peaks, a minor one eluting at 27.23 min followed by principal peak eluting at 29.18 min. We assigned the major one to 1-myristoyl-2-hydroxy-sn-glycero-3-PC and the minor one to 1hydroxy-2-myristoyl-sn-glycero-3-PC. This isomerization can be attributed to 1,2 migration of acyl group on the glycerol backbone. This chemical process is well described to occur both in chemical and biochemical conditions [30,31] in several lipid classes among which lyso-galactolipids [30] and lyso-PC [31]. Lastly, PI eluted between 31 and 32 min, followed by PA at 32.23 min and PS at 32.69 min. The list of retention times for each lipid standard is reported in Table 2. For most lipid classes, retention times are reproducible (SD ca 2%) over several runs and only PG and DGDG showed significant variations from run to run (SD ca 4%).
Lipid class
Molecular species
m/z
Ion type
RT (min) ± SD
CER MGDG MGDG SQDG SQDG SQDG SQDG SQDG SQDG SQDG DGTS PG DGDG PC DGCC PE PE PE PE SM Lyso PC Lyso PC PI PI PI PI PI PI PA PS
D18:1/24:0 34:0 36:0 28:0 30:0 32:0 34:3 34:2 36:5 36:4 36:0 36:0 36:0 36:6 36:6 34:2 34:1 36:2 36:1 d18:1/16.0 14:0 14:0 34:3 34:2 34:1 36:4 36:3 36:2 36:2 36:2
651.5 776.5 804.5 756.5 784.5 812.5 834.5 836.5 858.5 860.5 712.5 796.5 966.5 778.5 772.5 716.5 718.5 744.5 746.5 703.5 468.5 468.5 850.5 852.5 854.5 876.5 878.5 880.5 718.5 788.5
[M + H]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + H]+ [M + NH4 ]+ [M + NH4 ]+ [M + H]+ [M + H]+ [M + H]+ [M + H]+ [M + H]+ [M + H]+ [M + H]+ [M + H]+ [M + H]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + NH4 ]+ [M + H]+
2.78 (± 0.03) 3.72 (± 0.02) 3.59 (± 0.08) 8.30 (± 0.18) 8.03 (± 0.13) 7.78 (± 0.15) 7.52 (± 0.07) 7.51 (± 0.09) 7.34 (± 0.12) 7.22 (± 0.07) 7.71 (± 0.07) 12.08 (± 0.44) 13.18 (± 0.25) 15.80 (± 0.13) 16.49 (± 0.11) 21.61 (± 0.17) 21.53 (± 0.11) 21.18 (± 0.11) 21.10 (± 0.15) 24.94 (± 0.12) 27.23 (± 0.16) 29.18 (± 0.18) 32.03 (± 0.12) 31.95 (± 0.14) 31.93 (± 0.13) 31.73 (± 0.14) 31.68 (± 0.11) 31.65 (± 0.13) 32.23 (± 0.13) 32.69 (± 0.07)
3.3. Intra-class separation of lipid molecular species The partial separation, within a given lipid class, leading to individual molecular species has been reported in HILIC conditions (16, 18;); the effect was due to the different length of fatty acyl chains and their degree of unsaturation. However, the resolution of intraclass separation is low if compared to that of inter-class, which rules the behavior of lipids in HILIC. Fig. 5A shows the NL scanning of 141 Da for PE. Standard L-␣phosphatidylethanolamine extracted from soy bean contained a mixture of different molecular species, mainly PEs 34:2 (m/z 716.5), 34:1 (m/z 718.5), 36:2 (m/z 744.5) and 36:1 (m/z 746.5). Under our experimental setup, the intra-class separation is consistent with [16,18]; the retention time is lowered by the increase in acyl chain length and the decrease in the unsaturation degree. The length of the acyl chains with a given unsaturation degree plays a dominant role on intra-class separation with respect to the unsaturation number within a given chain length. As depicted in Fig. 5B, which represents EIC of single molecular species from Fig. 2, 38:4 PE (RT:20.26 ) and 38:5 PE (20.35 ) elute before 36:1 PE and 36:2 PE (RT: 21.24 and 21.30 respectively) and before 34:1 PE and 34:2 PE (RT: 21.41 and 21.52 respectively). The effect due to length of acyl chains was predominant on that of unsaturation degree. 36:1 PE eluted before 36:2 PE and, similarly, 34:1 PE eluted before 34:2 PE (Fig. 5B). The same effect was observed for other lipid classes. In Fig. 6 are shown EIC of 34:3 PI (m/z 850.5; RT: 31.91 min), 34:2 PI (m/z 852.5; RT: 31.82 min), 34:1 PI (m/z 854.5; RT: 31.81 min), 36:4 PI (m/z 876.5; RT: 31.51 min), 36:3 PI (m/z 878.5; RT: 31.57 min) and 36:2 PI (m/z 880.5; RT: 31.51 min), where the effect of chain length and degree of unsaturation are clear.
Please cite this article in press as: A. Anesi, G. Guella, A fast Liquid Chromatography-Mass Spectrometry methodology for membrane lipid profiling through Hydrophilic Interaction Liquid chromatography, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.01.035
G Model CHROMA-356196; No. of Pages 9 6
ARTICLE IN PRESS A. Anesi, G. Guella / J. Chromatogr. A xxx (2015) xxx–xxx
Fig. 5. A: NL scanning of 141 Da for PE in the range 600-1000 m/z. B: EIC of 34:1 PE (m/z 718.5), 34:2 PE (m/z 716.5), 36:1 PE (m/z 746.5), 36:2 PE (m/z 744.5), 38:4 PE (m/z 768.5) and 38:5 PE (m/z 766.5) contained in soy L-␣-phosphatidylethanolamine from Avanti Polar Lipids.
Fig. 6. EIC of 34:3 PI (m/z 850.5), 34:2 PI (m/z 852.5), 34:1 PI (m/z 854.5), 36:4 PI (m/z 876.5), 36:3 PI (m/z 878.5) and 36:2 PI (m/z 880.5) from soy L-␣-phosphatidylinositol from Avanti Polar Lipids.
Please cite this article in press as: A. Anesi, G. Guella, A fast Liquid Chromatography-Mass Spectrometry methodology for membrane lipid profiling through Hydrophilic Interaction Liquid chromatography, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.01.035
G Model CHROMA-356196; No. of Pages 9
ARTICLE IN PRESS A. Anesi, G. Guella / J. Chromatogr. A xxx (2015) xxx–xxx
Fig. 7. A: BPC of Peridinium cinctum raw extract; B: BPC of Jaoa bullata; C: BPC of Vitis vinifera cv. Corvina (ripening stage).
Please cite this article in press as: A. Anesi, G. Guella, A fast Liquid Chromatography-Mass Spectrometry methodology for membrane lipid profiling through Hydrophilic Interaction Liquid chromatography, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.01.035
7
G Model CHROMA-356196; No. of Pages 9
ARTICLE IN PRESS A. Anesi, G. Guella / J. Chromatogr. A xxx (2015) xxx–xxx
8
Fig. 8. BPC of mixture of Peridinium cinctum, Jaoa bullata and Vitis vinifera cv. Corvina raw extracts.
3.4. Method application
4. Conclusion
To test the new method, lipids of P. cinctum, J. bullata and V. vinifera cv. Corvina were analyzed. As no pre-purification step was applied, raw extracts may contain chlorophylls, carotenoids, neutral acylglycerols lipids (TAG and DAG) and sterols. As our method is particularly focused on membrane lipids, which are more interesting from a phylogenetic point of view than acylglycerols, all these bioorganic compounds were not followed in this work. Furthermore, HILIC is surely not the best method for their analysis, as there are poorly retained (data not shown). In Fig. 7, the base peak chromatograms of samples detected in positive ion mode are shown. P. cinctum contained mainly MGDG, SQDG, DGDG, PC, DGCC and traces of PE and PI (Fig. 7A); the lipid profile of P. cinctum is very similar to that of its closely related species P. aciculiferum recently reported by Flaim and co-workers [13]. The green algae J. bullata contained mainly MGDG, SQDG, DGTS/A, DGDG and traces of PC (Fig. 7B) as also depicted by [25]. Vitis vinifera cv. Corvina (ripening stage) contained MGDG, DGDG, PC, PE, PI and PA (Fig. 7C). The identity of each class is confirmed by specific NL and PIS experiments that are reported in Supplementary Materials. In Fig. 8 is shown the BPC of mixture containing similar amount of P. cinctum, J. bullata and V. vinifera cv. Corvina raw extracts; despite the sample complexity, lipid classes are almost base peakseparated. Only a partial overlap of some PG and DGDG and PC and DGCC is observed, which is the result of intra-class separation of lipids due to wide range of different lipids, with diverse acyl chain length and unsaturation degree (see Supplementary Materials).
Hydrophilic interaction liquid chromatography (HILIC), commonly used for separation of polar metabolites is here used for separation of lipid classes, generally obtained by normal-phase chromatography. The new HILIC method provides a fast and comprehensive tool for the baseline separation of fourteen different classes of membrane lipids on a diol column and is fully compatible with full scan and NL/PIS ESI-MS analysis. Given the broad diversity of lipid classes covered in this survey, this method can be applied for the preliminary investigation of the lipidomes from different sources, such as archea, prokaryotes, micro-and macro-eukaryotic organisms such as algae, fungi, higher and animals. In terms of chemotaxonomic significance, the methodology here described is much more powerful than the classical membrane lipid analysis based on RP chromatography. In fact, it is much more valuable to establish the number and nature of lipid classes (inter-class analysis) present in an environmental sample than the detailed distribution within a given lipid class (intra-class profile). Acknowledgments We want to thanks a) Dr. Giovanna Flaim from Edmund Mach Foundation (San Michele all’Adige, Italy) for Peridinium aciculiferum and Peridinium cinctum cell cultures; b) Dr. Abdelkader Bensalem for providing pure SQDG fraction extracted from Codium bursa; c) Dr. Marco Cantonati and Dr. Daniel Spitale for Jaoa bullata samples; Prof. Mario Pezzotti and Dr. Sara Zenoni for providing Vitis vinifera
Please cite this article in press as: A. Anesi, G. Guella, A fast Liquid Chromatography-Mass Spectrometry methodology for membrane lipid profiling through Hydrophilic Interaction Liquid chromatography, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.01.035
G Model CHROMA-356196; No. of Pages 9
ARTICLE IN PRESS A. Anesi, G. Guella / J. Chromatogr. A xxx (2015) xxx–xxx
cv. Corvina samples. The authors also acknowledge support from MIUR (PRIN 2010-2011 grant). Appendix A. Supplementary data
[16]
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.chroma. 2015.01.035.
[17]
References
[18]
[1] M.R. Wenk, The emerging field of lipidomics, Nat. Rev. Drug Discov. 4 (2005) 594–610, http://dx.doi.org/10.1038/nrd1776. [2] M. Lísa, E. Cífková, M. Holˇcapek, Lipidomic profiling of biological tissues using off-line two-dimensional high-performance liquid chromatography-mass spectrometry, J. Chromatogr. A. 1218 (2011) 5146–5156, http://dx.doi.org/ 10.1016/j.chroma.2011.05.081. [3] R. Taguchi, T. Houjou, H. Nakanishi, T. Yamazaki, M. Ishida, M. Imagawa, T. Shimizu, Focused lipidomics by tandem mass spectrometry, J. Chromatogr. B. 823 (2005) 26–36, http://dx.doi.org/10.1016/j.jchromb.2005.06.005. [4] Christie W.W. 2011. http://lipidlibrary.aocs.org/index.html. American Oil Chemists Society. [5] W.W. Christie, Chromatographic analysis of phospholipids, Z. Lebensm. Unters Forsch. 185 (1985) 171–182. [6] S. Milne, P. Ivanova, J. Forrester, H.A. Brown, Lipidomics: an analysis of cellular lipids by ESI-MS, Methods 39 (2006) 92–103. [7] S. Mawatari, Y. Okuma, T. Fujino, Separation of intact plasmalogen and all other phospholipids by a single run of high-performance liquid chromatography, Anal. Biochem. 370 (2007) 54–59. [8] L.-Q. Pang, Q.-L. Liang, Y-M. Wang, L. Ping, G.A. Luo, Simultaneous determination and quantification of seven major phospholipid classes in human blood using normal-phase liquid chromatography coupled with electrospray mass spectrometry and the application in diabetes nephropathy, J. Chromatogr. B 869 (1–2) (2008) 118–125. [9] H. Nie, R. Liu, Y. Yang, Y. Bai, Y. Guan, D. Qian, T. Wang, H. Liu, Lipid profiling of rat peritoneal surface layers by online normal- and reversed phase 2D LC QToF-MS, J. Lipid Res. 51 (2010), 2833-2344. [10] G. Guella, R. Frassanito, I. Mancini, A new solution for an old problem: the regiochemical distribution of the acyl chains in galactolipids can be established by electrospray ionization tandem mass spectrometry, Rapid Commun. Mass Spectrom. 17 (2003) 1982–1994. [11] G. Flaim, U. Obertegger, G. Guella, Changes in galactolipid composition of the cold freshwater dinoflagellate Borghiella dodgei in response to temperature, Hydrobiologia 698 (2012) 285–293, http://dx.doi.org/10.1007/ s10750-012-1070-8. [12] C.W.N. Damen, G. Isaac, J. Langridge, T. Hankmeier, R.J. Vreeken, Enhanced lipid isomer separation in human plasma using reversed-phase UPLC with ion mobility /high-resolution MS detection, J. Lipid Res. 55 (8) (2014) 1772–1783. [13] G. Flaim, U. Obertegger, A. Anesi, G. Guella, Temperature-induced changes in lipid biomarkers and mycosporine-like amino acids in the psychrophilic dinoflagellate Peridinium aciculiferum, Freshwater Biol. (2014), http://dx.doi. org/10.1111/fwb.12321. [14] J. Hummel, S. Segu, Y. Li, S. Irgang, J. Jueppner, P. Giavalisco, Ultra performance liquid chromatography and high resolution mass spectrometry for the analysis of plant lipids, Front. Plant Sci. 2 (2011) 54. [15] X. Gao, Q. Zhang, D. Meng, G. Issac, R. Zhao, T.L. Fillmore, R.K. Chu, J. Zhou, K. Tang, Z. Hu, R.J. Moore, R.D. Smith, M.G. Katze, T.O. Metz, A
[19]
[20] [21]
[22]
[23]
[24]
[25]
[26]
[27]
[28] [29]
[30]
[31]
9
reverse-phase capillary ultra-performance liquid chromatography-mass spectrometry (UPLC-MS) method for comprehensive top-down/bottom-up lipid profiling, Anal. Bioanal. Chem. 402 (9) (2012) 2923–2933, http://dx.doi.org/ 10.1007/s00216-012-5773-5. M. Schwalbe-Herrmann, J. Willmann, D. Leibfritz, Separation of phospholipid classes by hydrophilic interaction chromatography detected by electrospray ionization mass spectrometry, J. Chromatogr. A 1217 (32) (2010) 5179–5183. L. Zheng, R. T’Kind, S. Decuypere, S.J. von Freyend, G.H. Coombs, D.G. Watson, Profiling of lipids in Leishmania donovani using hydrophilic interaction chromatography in combination with Fourier transform mass spectrometry, Rapid Commun. Mass Spectrom. 24 (2010) 2074–2082. C. Zhu, A. Dane, G. Spijksma, M. Wang, J. van der Greef, G. Luo, T. Hankermeier, R.J. Vreeken, An efficient hydrophilic interaction liquid chromatography separation of 7 phospholipid classes based on a diol column, J. Chromatogr. A 1220 (2012) (2011) 26–34, http://dx.doi.org/10.1016/j.chroma.2011.11.034. P. Donato, F. Cacciola, F. Cichello, M. Russo, P. Dugo, L. Mondello, Determination of phospholipids in milk samples by means of hydrophilic interaction liquid chromatography coupled with evaporative light scattering and mass spectrometry detection, J. Chromatogr. A 1218 (2011) 6476–6482. V.M. Dembitsky, Betaine ether-linked glycerolipids: chemistry and biology, Prog. Lipid Res. 35 (1) (1996) 1–51. J. Brandsma, E.C. Hopmans, C.P.D. Brussard, H.J. Witte, S. Schouten, J.S. Sinninghe Damsté, Spatial distribution of intact polar lipids in North Sea surface waters: relationship with environmental conditions and microbial community composition, Limnol. Oceanogr. 57 (4) (2012) 959–973. T. Kind, j.K. Meissen, D. Yang, F. Nocito, A. Vaniya, Y.-S. Cheng, J.S. VanderGheynst, O. Fiehn, Qualitative analysis of algal secretions with multiple mass spectrometric platforms, J. Chromatogr. A 1244 (2012) 139–147, http://dx.doi.org/10.1016/j.chroma.2012.04.074. Y. Okazaki, H. Otsuki, T. Narisawa, M. Kobayashi, S. Sawai, Y. Kamide, M. Kusano, T. Aoki, M.Y. Hirai, K. Saito, A new class of plant lipid is essential for protection against phosphorus depletion, Nat. Commun. 4 (2013(a)) 1510, http://dx.doi.org/10.1038/ncomms2512. K.J. Popendorf, H.F. Fredricks, B.A.S. Van Mooy, Molecular ion-independent quantification of polar glycerolipids classes in marine plankton using triple quadrupole MS, Lipids 48 (2013) 185–195. J. Mares, M. Cantonati, G. Guella, D. Spitale, The benthic chlorophyte genus Jaoa (Ulvales), a putative China endemic, in Lake Garda, Italy: ecology, taxonomy, and molecular analyses, Freshwater Sci. 33 (2) (2014), http://dx.doi.org/10.1086/675860. S. Pilati, D. Brazzale, G. Guella, A. Milli, C. Ruberti, F. Biasioli, M. Zottini, C. Moser, The onset of grapevine berry ripening is characterized by ROS accumulation and lipoxygenase-mediated membrane perosidation in the skin, BMC Plant Biol. 14 (2014) 87. J. Folch, M. Lees, G.H. Sloane Stanley, A simple method for the isolation and purification of total lipids from animal tissues, J. Biol. Chem. 226 (1957) 497–509. E.G. Bligh, W.J. Dyer, A rapid method of total lipid extraction and purification, Can. J. Biochem. Physiol. 37 (1959) 911–917. Y. Okazaki, Y. Kamide, M.Y. Hirai, K. Saito, Plant lipidomics based on hydrophilic interaction chromatography coupled to ion trap time-of-flight mass spectrometry, Metabolomics 9 (2013) 121–131. I. Mancini, G. Guella, A. Defant, M.L. Candenas, C.P. Armesto, D. Depentori, F. Pietra, Polar metabolites of the tropical green seaweed Caulerpa taxifolia which is spreading in the Mediterranean sea: glycoglycerolipids and stable enols (␣keto esters), Helv. Chim. Acta 81 (1998) 1681–1691. M. Croset, N. Brossard, A. Polette, M. Lagarde, Characterization of plasma unsaturated lysophosphatidylcholines in human and rat, Biochem. J. 345 (2000) 61–67.
Please cite this article in press as: A. Anesi, G. Guella, A fast Liquid Chromatography-Mass Spectrometry methodology for membrane lipid profiling through Hydrophilic Interaction Liquid chromatography, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.01.035
