Document Detail

A versatile ultra-high performance LC-MS method for lipid profiling.
Jump to Full Text
MedLine Citation:
PMID:  24548922     Owner:  NLM     Status:  Publisher    
Abstract/OtherAbstract:
A new UPLC-based untargeted lipidomic approach using a qTOF hybrid mass spectrometer is introduced. The applied binary gradient enables separations of lipid species including constitutional isomeric compounds and low abundant lipid classes such as phosphatidic acid (PA). Addition of phosphoric acid to the solvents improves peak shapes for acidic phospholipids. MS(E) scans allow simultaneous acquisition of full scan data and collision induced fragmentation to improve identification of lipid classes and to obtain structural information. The method was used to investigate the lipidome of yeast.
Authors:
Oskar L Knittelfelder; Bernd P Weberhofer; Thomas O Eichmann; Sepp D Kohlwein; Gerald N Rechberger
Publication Detail:
Type:  JOURNAL ARTICLE     Date:  2014-1-29
Journal Detail:
Title:  Journal of chromatography. B, Analytical technologies in the biomedical and life sciences     Volume:  951-952C     ISSN:  1873-376X     ISO Abbreviation:  J. Chromatogr. B Analyt. Technol. Biomed. Life Sci.     Publication Date:  2014 Jan 
Date Detail:
Created Date:  2014-2-19     Completed Date:  -     Revised Date:  -    
Medline Journal Info:
Nlm Unique ID:  101139554     Medline TA:  J Chromatogr B Analyt Technol Biomed Life Sci     Country:  -    
Other Details:
Languages:  ENG     Pagination:  119-128     Citation Subset:  -    
Copyright Information:
Copyright © 2014 The Authors. Published by Elsevier B.V. All rights reserved.
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine

Full Text
Journal Information
Journal ID (nlm-ta): J Chromatogr B Analyt Technol Biomed Life Sci
Journal ID (iso-abbrev): J. Chromatogr. B Analyt. Technol. Biomed. Life Sci
ISSN: 1570-0232
ISSN: 1873-376X
Publisher: Elsevier
Article Information
© 2014 The Authors
License:
Received Day: 8 Month: 7 Year: 2013
Revision Received Day: 7 Month: 1 Year: 2014
Accepted Day: 8 Month: 1 Year: 2014
pmc-release publication date: Day: 01 Month: 3 Year: 2014
Print publication date: Day: 01 Month: 3 Year: 2014
Volume: 951-952 Issue: 100
First Page: 119 Last Page: 128
PubMed Id: 24548922
ID: 3946075
Publisher Id: S1570-0232(14)00015-4
DOI: 10.1016/j.jchromb.2014.01.011

A versatile ultra-high performance LC-MS method for lipid profiling
Oskar L. Knittelfeldera
Bernd P. Weberhofera
Thomas O. Eichmanna
Sepp D. Kohlweina
Gerald N. Rechbergerab Email: gerald.rechberger@uni-graz.at
aInstitute of Molecular Biosciences, University of Graz, Humboldtstraße 50/II, 8010 Graz, Austria
bOmics Center Graz, Austria
Corresponding author at: Institute of Molecular Biosciences (IMB), University of Graz, Humboldtstraße 50/II, A-8010 Graz, Austria. Tel.: +43 316/380 1933; fax: +43 316/380 9016. gerald.rechberger@uni-graz.at

Introduction

“Lipids” is the general term for a very large and complex group of substances. The combination of a large variety of fatty acids with diverse headgroups constitute a group of molecules with very different characteristics that play many important roles in biological systems. Non-polar triacylglycerols (TGs) are important storage lipids [1,2], amphiphilic glycerophospholipids (GPs) are the main building blocks of biological membranes and many lipid intermediates and degradation products act as signalling molecules in various biological pathways [3–5]. The diversity of lipids in all organisms is a challenge for the qualitative as well as quantitative lipid analysis, so called “lipidomics”, which, however has dramatically gained importance in many fields of biosciences. Lipid metabolism is linked to many severe diseases such as obesity, diabetes, atherosclerosis and cancer-related diseases [6,7]. As a consequence lipidomic research is an emerging discipline not only in the field of medicine, but also in fundamental research on model organisms [8].

The number of approaches towards qualitative and quantitative lipidomics is as complex as the lipids themselves and depends on the lipid source and the specific problem of interest. These methods comprise well-established ones such as thin layer chromatography and gas chromatography as well as mass spectrometric (MS) approaches [9,10].

In lipid mass spectrometry electrospray ionization (ESI) is the most commonly used ionization technique. The ESI-MS-based lipid analysis is generally split into two different basic principles – the shotgun and the liquid-chromatography (LC)-MS-approach. Shotgun lipidomics generally uses total lipid extracts typically without prior purification steps. Separation is achieved “in-source”, depending on the specific ionization and adduct formation properties of each lipid class [11,12]. The shotgun approach provides a large amount of lipidomic data for various tissues [13,14] and organisms such as yeast [15] and Drosophila[16]. In several cases, however a chromatographic pre-separation of lipid extracts provides advantages. The major challenges of all electrospray based ionization techniques are ion suppression and ion enhancement effects. Especially when some lipid classes are predominant, low abundant lipid species may be missed due to ion suppression [17].

Many different LC methods have been published, interestingly most of them focus either on neutral lipid or on glycerophospholipid species. Multiple methods for neutral lipid separation are described [18–25] but these methods either do not address other lipids in detail or require prior purification steps to remove them [26].

Vice versa many different methods for polar lipid analysis are published, but they do not address the separation of neutral lipids [27–35].

An isocratic HPLC method to analyze the cellular lipidome of yeast was published by Shui et al. [36]. Within 30 min run time the major lipid classes are analyzed by LC-MS, but the chromatographic resolution, especially of the glycerophospholipids, seems to be rather limited.

To separate polar and nonpolar lipids 2D-chromatography can be applied. Sommer et al. [37] show a good separation of many different lipid classes and even separation of lipid species within these classes. But the approach is quite complex, using a pre-separation between polar and nonpolar lipids on a silica column, followed by normal-phase HPLC with two different gradient systems depending on the polarity of the analytes. Both LC-runs use a ternary gradient and the eluents are mixtures of up to four solvents. Fractions collected from the normal-phase separation were further analyzed by reversed phase chromatography.

Another 2D-LC method is published by Nie et al. [38], who combines normal- and reversed-phase chromatography with a solvent evaporation interface after the normal phase column to enrich the analytes prior to reversed phase separation. The total run time of this 2D-approach is 150 min.

Methods that allow the separation of polar and nonpolar lipids in one chromatographic run often use complex gradients. Graeve et al. [39] describe a ternary gradient in a normal-phase LC system coupled to an evaporative light scattering detector that allows the separation of all major lipid classes within a single 35 min run. All three eluents are mixtures of two solvents, one of them containing ethanolamine as modifier. Especially in normal-phase chromatography modifiers like ethanolamine or triethylamine are necessary to improve peak shape and separation performance, but in combination with electrospray ionization mass spectrometry they can cause significant ion suppression effects. [40]. McLaren et al. [41] describe a modifier-free LC method for total lipid separation, but this method requires even a quarternary gradient. All four eluents consist of two solvents including dichloromethane and chloroform, which can be harmful for sealings in the LC system.

Both methods do not achieve a separation of lipid species within a certain lipid class.

A UPLC-MS based method published by Ikeda et al. [42] is capable of separating lipid species even within a lipid class, but the gradient time is up to 90 min.

The analysis described by Fauland et al. is promising [43], the gradient time is only 30 min, but the chromatographic resolution is not sufficient for low abundant lipids such as phosphatidic acid.

Our aim was to develop a versatile LC-MS method for general lipid profiling analyzing neutral lipids and glycerophospholipids in one run. This method is suitable for polar and non-polar lipid species and may be combined with ESI-MS in both positive and negative ionization mode. The gradient system is simple and robust, the hazard potential of the solvents is low both for the operator and the LC system. To achieve an optimal separation of different lipid classes/species we extended the total run time to 50 min. Thereby we could reduce the ion suppression of co-eluting analytes.

Another advantage of a chromatographic separation is the identification of isobaric substances. For example, bis(monoacylglycero)phosphate (BMP), an important signalling molecule, has the same elemental formula as the corresponding phosphatidylglycerol (PG), and even a very high resolution MS will not be able to distinguish between these two lipid classes without prior separation. With the described method the chromatographic peaks of BMP and the corresponding PG are clearly separated. Additionally, in source fragmentation can also “synthesize” lipids. When for example phosphatidylserine (PS) loses its headgroup during the ESI process, the resulting phosphatidic acid (PA) cannot be distinguished from PA originally present in the sample. By LC separation the retention times of the endogenous PA and PA derived from PS are different, and, therefore a correct determination of endogenous PA levels is possible. By adding phosphoric acid to the solvent system we could further improve the detection of PA.

We demonstrate the applicability for lipid profiling, using yeast lipid extracts as an example. To show the ability to separate isobaric lipid species we used murine tissue lipid extracts.


Materials and methods
2.1  Chemicals and reagents

All solvents were at least HPLC grade. Water, 2-propanol, and phosphoric acid were purchased from Roth (Karlsruhe, GER), acetonitrile and methanol from J.T.Baker (Austin, TX, USA) and chloroform, formic acid and leucine-enkephalin from Sigma (Vienna, AUT). Ammonium acetate was purchased from Merck (Darmstadt, GER), TG and DG standards (TG 45:0, TG 51:0, TG 57:0, DG 28:0) from Larodan (Malmö, SWE), GP standards (PC 34:0, PS 34:0, PA 34:0, PE 34:0) from Avanti Polar Lipids (Alabaster, AL, USA).

2.2  Preparation of standards

Internal standard solution: reference substances were combined to 1 mg/ml each in chloroform/methanol (2/1, v/v) (C/M) and prior to extraction this stock solution was diluted to 0.05 mg/ml.

Lock mass solution: Leucine-enkephalin (100 pg/μl) was dissolved in water/methanol (1/1, v/v) with 0.1% formic acid.

2.3  Yeast cultivation and strains

In this study wild type (WT) BY4742 (Euroscarf collection,[44]) was used to develop the LC-MS method. WT was either grown in rich medium (YPD: 20 g/l glucose, 20 g/l pepton, 10 g/l yeast extract) to stationary phase or cultivated in minimal medium. The standard minimal medium (MM) contained 20 g/l glucose and 6.7 g/l yeast nitrogen base containing ammonium sulphate. After autoclaving uracil (final conc. 20 mg/l), adenine (final conc. 29.3 mg/l), amino acids (according to [45]), 4-morpholineethanesulfonic acid (MES, 1 M, pH 6 with sodium hydroxid; final conc. 20 mM), vitamins, trace elements (according to [46]) and inositol (final conc. 75 μM) were added. WT was pre-cultivated for 16 h in MM at 30 °C and 180 rpm in a rotary shaker. For the final cultivation WT was inoculated to a density of OD600 0.1 in fresh MM. After 4.5 h 20 OD units were harvested, centrifuged and immediately shock frozen in liquid nitrogen. Pellets were stored at −80 °C until they were subjected to lipid extraction.

2.4  Mouse samples

Various tissue samples were surgically removed from sacrificed mice as described elsewhere [47] and stored at −80 °C until they were subjected to lipid extraction.

2.5  Sample preparation

Cell pellets were re-suspended in C/M and spiked with 50 μl of 0.05 mg/ml standard mix. Cell disruption was carried out with glass beads by shaking in a Heidolph Multi Reax test tube shaker (Schwabach, GER) at 4 °C for 30 min. Yeast and mouse lipids were extracted according to the Folch method [48]. Evaporated lipid extracts were re-suspended in 1 ml C/M and diluted 1:5 with isopropanol for measurement in positive ESI mode. For measurements in the negative ionization mode the solvent was changed to 90 vol% isopropanol and 10 vol% C/M without any dilution.

2.6  Instrumentation

Chromatographic separation was performed using an AQUITY-UPLC system (Waters, Manchester, UK) equipped with a BEH-C18-column, 2.1 × 150 mm, 1.7 μm (Waters). A binary gradient was applied. Solvent A consisted of water/methanol (1/1, v/v), solvent B was 2-propanol. Both solvents contained phosphoric acid (8 μM), ammonium acetate (10 mM) and formic acid (0.1 vol%). The linear gradient started at 45% solvent B and increased to 90% solvent B within 30 min. In the following 2 min solvent B percentage was increased to 100% and was kept at this level for 10 min. Starting conditions were achieved in 1 min and the column was re-equilibrated for 7 min, resulting in a total HPLC run time of 50 min. The column compartment was kept at 50 °C. A SYNAPT™G1 qTOF HD mass spectrometer (Waters) equipped with an ESI source was used for analysis. The following source parameters were used: capillary temperature 100 °C, desolvatization temperature: 400 °C, N2 as nebulizer gas. The capillary voltage was 2.6 kV in positive and 2.1 kV in negative ionization mode. For MSE[49] two alternating scan modes were defined in the MS setup. The first scan mode resembles the full scan (mass range: m/z 50–1800; scan time: 1 s; data collection: centroid). The second scan mode (mass range: m/z 50–1800; scan time: 1 s; data collection: centroid) applied a collision energy ramp (25–45 V) to fragment all generated ions. Both scan modes showed a resolution of around 9000 (FWHH). For MS/MS experiments the same settings for scan time and collision energy were used.

The lock spray was achieved by an external pump (L-6200, Hitachi) at a flow rate of 0.2 ml/min split in a 1:13 ratio. Leucine-enkephaline ([M + H]+: m/z 556.2771 and [M-H]: m/z 554.2615) was used as reference substance in the lock-spray. The lock mass was measured every 15 s independent of the other scan modes, which allowed a continuous mass correction. The mass error was always below 5 ppm.

Data acquisition was performed by the MassLynx 4.1 software (Waters), for lipid analysis the “Lipid Data Analyser” software was used [50]. The batch quantitation setup was as followed: retention time tolerance before/after: 1 min; relative base peak cut off: 0.1‰; retention time shift: 0.5 min; isotopic quantitation of 2 isotopes where 1 isotopic peak has to match.


Results and discussion
3.1  LC method

The method is based on a previously described gradient system using water and acetonitrile/2-propanol (5/2), both containing 1% 1 M NH4Ac and 0.1% HCOOH [51]. However, since the separation within the single lipid species was not satisfactory we tested various combinations of acetonitrile, acetone, methanol and 2-propanol. Best results were achieved by using water/methanol as solvent A and 2-propanol as solvent B, both containing ammonium acetate (10 mM) and formic acid (0.1 vol%) (Fig. 1).

An often mentioned advantage of UPLC systems is the short chromatographic run time and therefore the possibility for high sample throughput. We tested the separation performance at various gradient times (Fig. 2) and observed a baseline separation of a lipid species from the M + 2 isotopic peak of the corresponding one time more unsaturated species at a minimum gradient time of 12 min and a total run time of 30 min. Shorter run times caused co-elution of these species, which makes additional arithmetic inevitable to correct for the overlap in the mass spectra. There is a noticeable ion suppression effect in the baseline of the negative ESI TIC visible at the retention time of the TG species (Fig. 1 and supplementary figure 1). Comparable effects may also occur in the region of glycerophospholipids and they especially influence the signals of low abundant lipid species (see supplementary figure 2). Longer gradient times improve peak separation which is useful for more complex lipid samples, such as from mammalian tissue.

The resolving power of the described method allows to separate isobaric isomers such as PC 18:1_18:1 and PC 18:2_18:0 (Fig. 3) or TG 18:2_18:2_18:2 and TG 16:0_16:0_22:6 (Fig. 4) in a lipid extract of mouse muscle cells. Fig. 3 shows 2 peaks in the extracted ion chromatogram (XIC) of m/z 830.59, which corresponds to the formiate adduct of PC 36:2 in negative ion mode. Fragmentation of the first peak leads to a prominent ion at m/z 281 (carboxylate anion of C18:1 fatty acid), whereas the second peak fragments to m/z 279 (C18:2) and m/z 283 (C18:0). Furthermore there are minor peaks detectable which correspond to C16:0 and C20:2 carboxylate anion. In total PC 36:2 has three possible fatty acid compositions namely 18:1_18:1, 18:0_18:2 and 16:0_20:2. Fig. 4 shows the separation of the isomers of TG 54:6 as well as their verification with MS/MS experiments. The product ion spectra of the precursor m/z 896.77 show the characteristic neutral loss of C18:2 fatty acid at the first peak and the neutral loss of C22:6 and C16:0 fatty acids at the second peak. The considerably improved separation of lipid species within a certain lipid class makes the method suitable also for low resolution mass spectrometers.

The combination of good separation performance for TG, DG, PC, PE, PI, and LPC and a total run time of 50 min was acceptable, nevertheless the analysis of PA and PS was limited due to poor peak shape.

The problem of PA-peak-tailing has been described previously, however the addition of phosphoric acid (8 μM in both solvents) lead to a significant improvement of peak shape [52] (see supplementary Figure 3). After 2 years using the described method no negative effect caused by phosphoric acid could be observed, the system stability did not change over time and retention times varied by less than 5 seconds over 50 sample runs.

3.2  MS method

PC, PE, LPC, Cer, CL and NLs were analyzed in positive ESI mode, however all GPs could also be detected in negative ESI mode (Fig. 5). In Fig. 5 the separation of the most abundant species of different GPs is demonstrated, whereby PI, PS and PA exclusively ionized in the negative ESI mode. BMP and PG ionized in both ESI modes forming sodium and ammonium adducts additionally to the base peak proton adduct. For data analysis the MSE scan mode was applied [53]. By alternating full- and MS/MS scans without any precursor mass selection data were recorded as two separated traces. Data recorded in the full scan mode were used for peak integration and relative analysis. During the MS/MS experiment ions that were detected in the full scan were fragmented together, thus the MS/MS spectrum represents the combination of all the fragments of all co-eluting substances.

The advantage of this scanning technique compared to classical MS/MS approaches is that data analysis is not limited to pre-defined precursor- or neutral-loss-scans. Due to the fact that all potential precursors are fragmented it is not necessary to define all the characteristic fragments or neutral losses in advance. The undirected acquisition of all the full- and MS/MS data within one run allows permanent re-analysis of the data to identify additional lipids or fragments. The MSE scan in combination with a reliable chromatographic system and stable retention times further permits the detection of characteristic head groups and neutral losses and the assignment of the fragments to a specific precursor ion in the full scan trace. In Fig. 6 we show the characterization of TG 50:2 with m/z 848. The fragments with the same retention time as the non-fragmented TG are caused by the neutral losses of NH3 and C16:0, C16:1, C18:0 and C18:1 fatty acid. Several combinations of these fatty acids are possible to build TG 50:2, but other fatty acids can be excluded as building blocks of this specific TG species due to different retention times.

3.3  Glycerophospholipids

Yeast lipid extracts are not as complex as extracts from mammalian tissues, nevertheless the ion suppression of low abundant species by the predominant PC species has to be considered. Although all GP except the LGPs and CL elute between 8 and 24 min the peaks of the major PC species do not overlap significantly with other GPs. While in positive ion mode only PC and PE are analyzed, all glycerophospholipids can be detected in negative ion mode (Fig. 5). Most of the glycerophospholipids ionize as [M-H] adducts whereby PC builds adduct ions with formiate and also shows a strong [M-CH3] signal. Standards of PG and BMP can be measured in positive (proton, sodium and ammonium adducts) and negative mode and it is possible to separate the structural isomers with the applied gradient (Fig. 7).

It should be mentioned that under these conditions in-source fragmentation can occur. This may lead to multiple peaks for a certain m/z. As an example, PS may lose its headgroup to some extend resulting in the corresponding PA species. However different retention times clearly allow discrimination between PA as a sample component or decomposition product of PS.

Cardiolipin elutes close to TG at the end of the chromatographic run and can be detected in negative mode as the corresponding [M-H] ions as well as in positive ion mode as [M + H]+ ions.

3.4  Sphingolipids

A total yeast lipid extract prepared by the Folch method shows only weak sphingolipid signals whereby the most abundant phytoceramides [54] elute at retention times of about 25 min (Fig. 8) and are detectable both in positive and negative mode. Extraction of sphingolipids and inositol containing lipids can be improved by applying the extraction method of Angus et al. [55]. For detailed analysis of low abundant sphingolipids it may be advantageous to perform mild alkaline hydrolysis [54] of the lipid extract to remove all lipids containing ester-linked fatty acids. The remaining lipids contain only N-linked fatty acids and are also well separated with the described chromatographic method showing good ionization properties in the negative mode [56].

3.5  Neutral lipids
3.5.1  Di- and triacylglycerols

Diacylglycerols (DGs) partially co-elute with phospholipids but are easily detectable (Fig. 9). DGs form sodium adducts and show neutral loss of water caused by in-source fragmentation. TGs are mainly detected as ammonium- and sodium-adducts. The formation of these adducts varies depending on the chain length of the fatty acids. Shorter TG species ionize preferentially as [M + Na]+ adducts, whereas longer chain length leads to predominant [M + NH4]+ species. The TG species that differ just by one double bond are clearly baseline separated, as shown in Fig. 10.

The MSE scan in combination with good and stable chromatographic separation allows the assignment of fatty acids to the TGs, as shown in Fig. 6. Whereas the TGs in yeast contain mainly C16:0, C16:1, C18:0 and C18:1 fatty acids, the TG pattern in mammalian cells is much more complex. Fig. 4 illustrates the potential of the described method to separate the isobaric TG species in such samples.

3.5.2  Steryl esters

A drawback of the described method is the limited analysis of steryl esters (SEs). The chromatographic separation of TGs and SEs is poor and due to ion suppression effects during the ESI process the signals of steryl esters in a yeast total lipid extract are near the detection limit. A solid phase extraction that removes TGs [57] prior to UPLC separation, however makes steryl esters accessible for analysis.

3.5.3  Normalization of lipid data

For normalization a mix of internal standards was added prior to lipid extraction. The selection of internal standards is always a compromise between practicability and costs. The use of stable isotope labelled standards for each lipid species of interest would lead to even better results, but most of such standards are either not available or rather expensive. A compromise is the use of lipids with naturally not occurring fatty acid composition which elute close to the lipid species in the sample. Due to the rather simple yeast lipidome the use of only one GP internal standards for each lipid class with a retention time close to that of the corresponding GPs in the sample is sufficient to obtain a reliable picture of changes in the various lipid levels. For instance GP 34:0 species are absent from yeast samples, therefore these lipids were used as internal standards for GPs. Due to the lack of suitable PI standards we normalized the PI species to PE 34:0. The DGs were normalized to DG 28:0. For TG normalization we decided to use three different internal standards (TG 45:0, TG 51:0 and TG 57:0) to consider the wider range of retention times. Therefore, TG species TG 42:0 to TG 48:3 were normalized to TG 45:0, TG 50:0 to TG 54:3 were normalized to TG 51:0 and all the TGs with longer fatty acids were normalized to TG 57:0.

All these analyses are relative. For absolute quantification it would be necessary to establish the corresponding calibration curves for each analyte of interest and to use stable isotope labelled internal standards. Comparison with quantitative data obtained with thin layer chromatography and HPLC coupled to an evaporative light scattering detector confirms that trends in the changes of the amounts of lipid classes can also clearly be seen with this method. The major advantage of this method is the amount of information obtained about changes of the species pattern within a lipid class, even when the total amount of this lipid class does not change.

Ion suppression can never be totally excluded. The negative effects can be reduced or, compared to the shotgun approach, limited to co-eluting species. The chromatographic separation of the analyte peak from the M + 2 isotope peak of the corresponding more unsaturated lipid species facilitates correct peak integration. A triple quadrupole MS with unit mass resolution can be used for quantification, whereas a shotgun approach requires high resolution MS or additional arithmetic operations to distinguish between these isobars.

3.6  Application

The combination of MSE–MS and a simple and stable chromatographic method are feasible tools for lipid profiling. Major lipid classes are accessible with such a setup. For lipidomic studies we added a mixture of internal standards containing various TGs and GPs to the cell pellets prior to cell disruption and lipid extraction. The extracted ion chromatograms (XICs) of the lipid species of interest are integrated and normalized to the corresponding internal standard area using lipid data analyzer software [50]. WT yeast cells were grown to the exponential phase and lipids were extracted and measured. The relative abundances of the different GP species are shown in Fig. 11 and a GP class specific distribution is shown. While PC and PE contain more double unsaturated species, PI, PA and PS contain more mono unsaturated species. Also the pattern of TG species can easily be detected (Fig. 12). This method is used to compare the lipid profile of different yeast mutant strains as well as the changes in the lipid profile during cell growth and under different growth conditions.

In general this method has already been applied to different kinds of organisms like yeast (including Saccharomyces cerevisiae and Yarrowia lipolytica), Drosophila and various murine tissues [58]. The separation is powerful enough to handle lipids with higher variability in fatty acid composition and degree of saturation.

The quadrupole–time of flight high resolution mass spectrometer allows the determination of lipid elemental composition with an error below 5 ppm. By recording full and MS/MS scans identification of lipids is greatly simplified and no prior knowledge of the sample is required.


Conclusion

The UPLC method presented here is applicable for lipid profiling and allows the analysis of the major lipid classes within an acceptable run time. Depending on the sample complexity the gradient can be shortened or prolonged to achieve an applicable balance between chromatographic resolution and sample throughput. Due to a very simple solvent composition the method provides very stable retention times and allows good separation even within the species of one lipid class. Even isomeric structures such as PG and BMP can clearly be separated. To a certain extend the method allows the separation of isobaric lipid species. The good separation performance is achieved on a standard UHPLC-reversed-phase column and allows in-depth analysis of lipid extracts and helps to gain insight in changes of the lipidome of various organisms. Many MS based methods for lipidomic analysis require high resolution MS to identify single lipid species. The separation of the lipid species within a lipid class is sufficient enough so that even a low resolution-MS like a triple quadruple can be used for lipid analysis.

In this study, Saccharomyces cerevisiae was used as a model system. Although the yeast lipidome is not as complex as the mammalian system, the presented method is also suitable to perform lipid analysis of higher organisms.


Notes

This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-No Derivative Works License, which permits non-commercial use, distribution, and reproduction in any medium, provided the original author and source are credited.

References
1. Kohlwein S.D.. J. Biol. Chem.285Year: 20101566320231294
2. Lass A.,Zimmermann R.,Oberer M.,Zechner R.. Prog. Lipid Res.50Year: 20111421087632
3. Shevchenko A.,Simons K.. Nat. Rev. Mol. Cell Biol.11Year: 201059320606693
4. Wenk M.R.. Cell143Year: 201088821145456
5. Gaspar M.L.,Aregullin M.,Jesch L.R.,Stephen aNunez M.,Villa-García S.A.,Henry. Biochim. Biophys. Acta1771Year: 200724116920401
6. Zechner R.,Zimmermann R.,Eichmann T.O.,Kohlwein S.D.,Haemmerle G.,Lass A.,Madeo F.. Cell Metab.15Year: 201227922405066
7. Das S.K.,Eder S.,Schauer S.,Diwoky C.,Temmel H.,Guertl B.,Gorkiewicz G.,Tamilarasan K.P.,Kumari P.,Trauner M.,Zimmermann R.,Vesely P.,Haemmerle G.,Zechner R.,Hoefler G.. Science333Year: 201123321680814
8. Guan X.L.,Riezman I.,Wenk M.R.,Riezman H.. Methods Enzymol.470Year: 201036920946818
9. Roberts L.D.,McCombie G.,Titman C.M.,Griffin J.L.. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci.871Year: 2008174
10. Köfeler H.C.,Fauland A.,Rechberger G.N.,Trötzmüller M.. Metabolites2Year: 201219
11. Han X.,Yang J.,Cheng H.,Ye H.,Gross R.W.. Anal. Biochem.330Year: 200431715203339
12. Schwudke D.,Liebisch G.,Herzog R.,Schmitz G.,Shevchenko A.. Methods Enzymol.433Year: 200717517954235
13. Han X.,Yang K.,Yang J.,Cheng H.,Gross. J. Lipid Res.47Year: 200686416449763
14. Sewell G.W.,a Hannun Y.,Han X.,Koster G.,Bielawski J.,Goss V.,Smith P.J.,Rahman F.Z.,Vega R.,Bloom S.L.,Walker A.P.,Postle A.D.,Segal A.W.. Int. J. Biochem. Cell Biol.44Year: 2012183922728312
15. Ejsing C.S.,Sampaio J.L.,Surendranath V.,Duchoslav E.,Ekroos K.,Klemm R.W.,Simons K.,Shevchenko A.. Proc. Natl. Acad. Sci. U. S. A.106Year: 2009213619174513
16. Carvalho M.,Sampaio J.L.,Palm W.,Brankatschk M.,Eaton S.,Shevchenko A.. Mol. Syst. Biol.8Year: 20121
17. Yang K.,Han X.. Metabolites1Year: 20112122905337
18. Kalo P.,Kemppinen A.,Ollilainen V.,Kuksis A.. Lipids39Year: 200491515669768
19. Fauconnot L.,Hau J.,Aeschlimann J.-M.,Fay L.-B.,Dionisi F.. Rapid Commun. Mass Spectrom.18Year: 200421814745773
20. MacDougall K.M.,McNichol J.,McGinn P.J.,O’Leary S.J.B.,Melanson J.E.. Anal. Bioanal. Chem.401Year: 2011260921915640
21. Ikeda K.,Oike Y.,Shimizu T.,Guchi R.T.. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci.877Year: 20092639
22. Cai S.,Short L.,Syage J.,Potvin M.,Curtis J.. J. Chromatogr. A1173Year: 20078817964589
23. Buchgraber M.,Ulberth F.,Emons H.,Anklam E.. Eur. J. Lipid Sci. Technol.106Year: 2004621
24. Lin J.-T.,Woodruff C.L.,McKeon T.A.. J. Chromatogr. A782Year: 199741
25. Pueyo E.,a Martínez-Rodríguez M.C.,Polo G.,Santa-María B.,Bartolomé. J. Agric. Food Chem.48Year: 200011610637062
26. Hutchins P.M.,Barkley R.M.,Murphy R.C.. J. Lipid Res.49Year: 200880418223242
27. Kim H.,Min H.K.,Kong G.,Moon M.H.. Anal. Bioanal. Chem.393Year: 2009164919194696
28. Schwarz E.,Prabakaran S.,Whitfield P.,Major H.,Leweke F.M.,Koethe D.,Mckenna P.,Bahn S.. J. Proteome Res.7Year: 2008426618778095
29. Henderson C.M.,Lozada-Contreras M.,Naravane Y.,Longo M.L.,Block D.E.. J. Agric. Food Chem.Year: 20111276121995817
30. Pacetti D.,Malavolta M.,Bocci F.,Boselli E.,Frega N.G.. Rapid Commun. Mass Spectrom.18Year: 2004239515386630
31. Hermansson M.,Uphoff A.,Käkelä R.,Somerharju P.. Anal. Chem.77Year: 2005216615801751
32. Baumbach I.,Schmid A.,Hein E.,Blank L.M.,Heyland J.,Hayen H.. Rapid Commun. Mass Spectrom.23Year: 2009163619408252
33. Rezanka T.,Siristova L.,Matoulková D.,Sigler K.. Lipids46Year: 201176521479912
34. Moir R.D.,Gross D.A.,Silver D.L.,Willis I.M.. PLoS Genet.8Year: 2012e100289022927826
35. Buré C.,Ayciriex S.,Testet E.,Schmitter J.-M.. Anal. Bioanal. Chem.405Year: 201320323064709
36. Shui G.,Guan X.L.,Low C.P.,Chua G.H.,Goh J.S.Y.,Yang H.,Wenk M.R.. Mol. Biosyst.6Year: 2010100820485745
37. Sommer U.,Herscovitz H.,Welty F.K.,Costello C.E.. J. Lipid Res.47Year: 200680416443931
38. Nie H.,Liu R.,Yang Y.,Bai Y.,Guan Y.,Qian D.,Wang T.,Liu H.. J. Lipid Res.51Year: 2010283320526000
39. Graeve M.,Janssen D.. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci.877Year: 20091815
40. Holcapek M.,Volná K.,Jandera P.,Kolárová L.,Lemr K.,Exner M.,Církva A.. J. Mass Spectrom.39Year: 2004435014760612
41. McLaren D.G.,Miller P.L.,Lassman M.E.,Castro-Perez J.M.,Hubbard B.K.,Roddy T.P.. Anal. Biochem.414Year: 201126621402047
42. Ikeda K.,Mutoh M.,Teraoka N.,Nakanishi H.,Wakabayashi K.,Taguchi R.. Cancer Sci.102Year: 20117920946475
43. Fauland A.,Köfeler H.,Trötzmüller M.,Knopf A.,Hartler J.,Eberl A.,Chitraju C.,Lankmayr E.,Spener F.. J. Lipid Res.52Year: 2011231421960706
44. Brachmann C.B.,Davies A.,Cost G.J.,Caputo E.,Li J.,Hieter P.,Boeke J.D.. Yeast14Year: 19981159483801
45. Hanscho M.,Ruckerbauer D.E.,Chauhan N.,Hofbauer H.F.,Krahulec S.,Nidetzky B.,Kohlwein S.D.,Zanghellini J.,Natter K.. FEMS Yeast Res.12Year: 201279622780918
46. Villa-García M.J.,Choi M.S.,Hinz F.I.,Gaspar M.L.,Jesch S.,Henry S.A.. Mol. Genet. Genomics285Year: 201112521136082
47. Eichmann T.O.,Kumari M.,Haas J.T.,Farese R.V.,Zimmermann R.,Lass A.,Zechner R.. J. Biol. Chem.287Year: 20124144623066022
48. Folch J.,Lees M.,Sloane G.H.. J. Biol. Chem.226Year: 195749713428781
49. Plumb R.S.,Johnson K.A.,Rainville P.,Smith B.W.,Wilson I.D.,Castro-Perez J.M.,Nicholson J.K.. Rapid Commun. Mass Spectrom.20Year: 2006198916755610
50. Hartler J.,Trötzmüller M.,Chitraju C.,Spener F.,Köfeler H.C.,Thallinger G.G.. Bioinformatics27Year: 201157221169379
51. Laaksonen R.,Katajamaa M.,Päivä H.,Sysi-Aho M.,Saarinen L.,Junni P.,Lütjohann D.,Smet J.,Van Coster R.,Seppänen-Laakso T.,Lehtimäki T.,Soini J.,Oresic M.. PloS One1Year: 2006e9717183729
52. Ogiso H.,Suzuki T.,Taguchi R.. Anal. Biochem.375Year: 200812413118206977
53. Bateman R.H.,Carruthers R.,Hoyes J.B.,Jones C.,Langridge J.I.,Millar A.,Vissers J.P.C.. J. Am. Soc. Mass Spectrom.13Year: 200279212148804
54. Guan X.L.,Wenk M.R.. Yeast23Year: 200646516652392
55. Angus W.W.,Lester R.L.. Arch. Biochem. Biophys.151Year: 19724834339932
56. Ejsing C.S.,Moehring T.,Bahr U.,Duchoslav E.,Karas M.,Simons K.,Shevchenko A.. J. Mass Spectrom.41Year: 200637216498600
57. Ruiz-Gutiérrez V.,Pérez-Camino M.C.. J. Chromatogr. A885Year: 200032110941680
58. Nusshold C.,Kollroser M.,Koefeler H.,Rechberger G.,Reicher H.,Ullen A.,Bernhart E.,Waltl S.,Kratzer I.,Hermetter A.,Hackl H.,Trajanoski Z.,Hrzenjak A.,Malle E.,Sattler W.. Free Radic. Biol. Med.48Year: 2010158820226853
Appendix A  Supplementary data


Acknowledgements

We thank Myriam Visram for careful and critical reading of the manuscript. Also, we thank Robert Zimmermann and Irina Mrak for providing us with BMP and PG standard. This work was supported by the Austrian Ministry for Science and Research (GOLD project in the framework of GEN-AU/Genome Research in Austria), by the PhD program ‘Molecular Enzymology’ to O.L.K. and SFB LIPOTOX F3005, funded by the Austrian Science Fund (FWF).


Article Categories:
  • Article

Keywords: Keywords Lipidomics, Ultra-performance liquid chromatography, Mass spectrometry, Neutral lipids, Glycerophospholipids.
Keywords: Abbreviations BMP, bis(monoacylglycero)phosphate, Cer, ceramide, CL, cardiolipin, C/M, chloroform/methanol (2/1, v/v), DG, diacylglycerol, ESI, electrospray ionization, MM, minimal media, NL, neutral lipid, PA, phosphatidic acid, PC, phosphatidylcholine, PE, phosphatidylethanolamine, PG, phosphatidylglycerol, PI, phosphatidylinositol, GP, glycerophospholipid, PS, phosphatidylserine, SE, steryl ester, SM, sphingomyelin, TG, triacylglycerol, TIC, total ion current, WT, wild type, XIC, extracted ion chromatogram.

Previous Document:  A simple quantitative method analysing amikacin, gentamicin, and vancomycin levels in human newborn ...
Next Document:  Phosphorylation of Carma1, but not Bcl10, by Akt regulates TCR/CD28-mediated NF-?B induction and cyt...