Вы находитесь на странице: 1из 4

BBA - Molecular and Cell Biology of Lipids 1862 (2017) 782–785

Contents lists available at ScienceDirect

BBA - Molecular and Cell Biology of Lipids


journal homepage: www.elsevier.com/locate/bbalip

Opinion article

Green light for lipid fingerprinting MARK


a b,c,⁎
Richard P. Haslam , Ivo Feussner
a
Rothamsted Research, Department of Plant Science, Harpenden AL5 2JQ, UK
b
University of Goettingen, Albrecht-von-Haller-Institute for Plant Sciences, Department of Plant Biochemistry, Justus-von-Liebig-Weg 11, 37077 Goettingen, Germany
c
University of Goettingen, Goettingen Center for Molecular Biosciences (GZMB), Department of Plant Biochemistry, Justus-von-Liebig-Weg 11, 37077 Goettingen,
Germany

A R T I C L E I N F O A B S T R A C T

Keywords: The use of targeted lipidomic approaches for the analysis of plant lipids has steadily increased during recent
Lipid profiling years. We review recent developments of these methods and suggest the introduction of discovery lipidomics as
Lipid fingerprinting additional approach through a new workflow, lipid fingerprinting, that integrates the advantages of shotgun
lipidomics (quantitative data) with LC-MS-based strategies (higher resolution and/or coverage). This article is
part of a Special Issue entitled:BBALIP_Lipidomics Opinion Articles edited by Sepp Kohlwein.

1. Introduction resolution and/or coverage).

From the early days of chromatography, when Tswett first separated 2. The challenges of lipidomics
leaf pigments [1], lipid analysis has been always connected with
analytical developments. The continued development of analytical 2.1. Lipid extraction
tools and techniques has defined the discipline of lipidomics, focusing
on improvements mainly in mass spectrometry and in separation Prior to any discussion of lipid analytical measurement, sample
techniques [2,3]. Specifically, these improvements have addressed processing and extraction must first be addressed. Effective improve-
two key analytical challenges: naturally occurring lipid signals are ments in lipidomic analysis begin with sample extraction; the choice of
typically found in relatively narrow mass ranges and can often suffer methodology defines the range of metabolites available for analysis and
from isobaric interferences; and lipids present in high concentration the likely throughput. Traditionally, isolation of lipids has relied on
partially or fully suppress the detection of numerous important low general extraction procedures established by Folch et al. [4] and Bligh
abundance lipids. The resolution of these two issues has been addressed and Dyer [5], with frequent modifications for specific tissues, scaling of
in recent years by a number of technological analytical advances. volumes and matrices. However, these methods typically encounter
However, further improvements are now often accompanied by com- problems when adapting them to automated liquid handling systems
promise, namely that growing the number of detected lipid molecules and processing large numbers of samples. The development of protocols
often reduces the number of analyzed samples per time. As investiga- using the less toxic methyl tert-butyl ether (MTBE) has now provided an
tors attempt to capture the huge array of lipids typically found in alternative in which the low density lipid-containing organic phase
biological samples with ever greater accuracy, improvements are made forms the upper layer during phase separation [6]. The successful
in sample preparation, chromatographic separation and mass spectro- extraction of the main lipid classes using MTBE has enabled the
metric detection, but this is always at the expense of measurement time development of well-plate based high-throughput analysis. Accurate
and costs. It is possible that developments may allow the capture of and rapid lipid extraction using well plates has been the ambition of
many hundreds of different plant lipid molecular species across multi- many in the lipidomic community. For example, an automated chloro-
ple classes in a single analytical run. Here, we present our thoughts on form-free method was developed by Lofgren et al. [7,8] for the 96-well
the opportunities for discovery lipidomics provides and a new work- extraction of plasma lipids. With compatible recovery of the main lipid
flow, lipid fingerprinting, that integrates the advantages of shotgun classes to the Folch method, this BUME (butanol/methanol) method
lipidomics (quantitative data) with LC-MS-based strategies (higher enables the effective processing of large sample numbers. Following the


This article is part of a Special Issue entitled:BBALIP_Lipidomics Opinion Articles edited by Sepp Kohlwein.

Corresponding author at: University of Goettingen, Albrecht-von-Haller-Institute for Plant Sciences, Department of Plant Biochemistry, Justus-von-Liebig-Weg 11, 37077 Goettingen,
Germany.
E-mail address: ifeussn@uni-goettingen.de (I. Feussner).

http://dx.doi.org/10.1016/j.bbalip.2017.04.005
Received 22 November 2016; Received in revised form 15 April 2017; Accepted 17 April 2017
Available online 20 April 2017
1388-1981/ © 2017 Elsevier B.V. All rights reserved.
R.P. Haslam, I. Feussner BBA - Molecular and Cell Biology of Lipids 1862 (2017) 782–785

same line, a single phase method based on isopropanol, hexane and ions. In particular, ion mobility-mass spectrometry (IM-MS) in which
water was developed for plant tissues [9]. However, even with these ions are separated by their size, shape, charge and mass depending on
developments extracts contain complex mixtures of metabolites which different mobility in low or high electric fields, has provided a new
have the capacity to interfere with accurate mass spectrometry (MS) dimension in the analytical workflow. The use of IM-MS specifically
based analysis. Therefore, when trying to undertake the in-depth allows for the separation of isobaric species, reduction of background
characterization of lipid classes, further clean-up is often required, noise and increased selectivity via the separation of interfering
involving either liquid/liquid partitioning or solid-phase extraction. metabolites from target lipids. Moreover, IM-MS provides a unique
The issue of lipid stability must also be considered within any descriptor of the physiological properties of a compound (collision
extraction approach; specific lipid species are more likely than others cross-section value; CCS) based on the time required for any given ion
to undergo degradation during the extraction process. For example, to cross the chamber. CCS values provide another means of compound
oxylipins require specific extraction methodologies to retain their annotation and the benefits of IM-MS in lipidomics have been demon-
original metabolic form [10]. Overall, when selecting any lipidomic strated, for example, differences in drift time have been used to
analytical strategy, the number and type of operations necessary for separate phospholipids with different degrees of unsaturation, linkage
lipid extraction must be considered. type and head-group [17,18]. Research has now established ion
mobility as a method that provides a separate and complimentary
2.2. Analysis using mass spectrometry approach to increasing selectivity, enhancing the detection of low
abundance species [19]. Indeed, it also has the potential to reduce the
In terms of mass spectrometry detection, a major milestone in the number of steps required for successful sample extraction and clean-up.
field of lipid analysis was the development of electrospray ionization However, any issues involving ion suppression at the ion source will
(ESI) and its connection to tandem mass spectrometry (MS/MS) in the still remain, limiting the utility of IM-MS for the quantification of low
field of mammalian lipids [11,12]. Cell extracts are analyzed by direct abundant ions in complex biological samples.
infusion MS using precursor ion scans (PIS) and neutral loss scans (NL) Often not represented in any total extraction of lipids is the spatial
to identify key lipid fragments. The relative simplicity of direct infusion distribution or compartmentation of lipid species in tissues or samples.
ESI-MS/MS, fast analysis, and flexibility to detect various lipid classes The development of MS-imaging has, particularly in plant lipidomics,
within a single run has resulted in the widespread adoption of this begun to address this lack of spatial information. Matrix-assisted laser
approach. In its current form, ESI-MS/MS methods now widely use desorption ionization (MALDI) coupled to high-resolution mass spectro-
automated nanospray techniques. The use of NanoESI-MS/MS has metry (MS) has emerged as a new tool for the localization of lipid
extended the lipid analysis of low volume samples that would otherwise metabolites directly in plant tissues with 5-to-50 μm spatial resolution
not be practical with other approaches. The transition of ESI-MS/MS [20,21]. Imaging the location of phosphatidylcholines (PCs) and
methods to plants, which have a much more complex matrix took a triacylglycerols (TAGs) in seeds by MALDI-MS not only demonstrated
further five more years [13]. Since these methods relied exclusively on the compartmentalisation of lipid metabolism but also the metabolic
a simple extraction procedure and direct mass analysis for lipid precursor-product relationships of seed oil biosynthesis [22].
identification, they increased dramatically the speed and coverage of Therefore, even with the technological advances described above,
analyzed lipid samples at relatively moderate costs. Complimenting the analysis of (mainly) minor lipids such as lipid mediators has
traditional lipid analytical approaches, the establishment of direct required attention and has led to the development of numerous specific
infusion ESI-MS/MS for the characterization of plant lipids enabled liquid chromatographic separations prior to mass analysis [10,23–26].
the comparison of the lipid profiles from plants subjected to forward- or Primarily these methods have included hydrophilic interaction chro-
reverse-genetic manipulation, developmental and physiological pheno- matography (HILIC) and reversed phase (RP) chromatography. The
typing, and unraveled metabolic bottlenecks associated with molecular former is very well suited for lipid class discovery [27], but is limited in
engineering. However, it is apparent that methods using tandem MS in resolving lipids within the same class [28]. However, this drawback can
a class-specific or targeted way are restricted in their potential to detect be overcome by combining it with RP methods that allow the
and identify unknowns [11]. Furthermore, during recent years it separation of molecular species including isomeric pairs [29] and will
became clear that direct infusion methods suffer from ion suppression. most likely lead to the development of two dimensional LC-MS methods
Not all molecular species are identified with equal efficiency, in in the near future. In any case, when relying on a single chromato-
particular acyl chain length, degree of unsaturation and lipid concen- graphic run both HILIC and RP methods usually focus on certain lipid
tration all impact on the instrument response. Indeed in plant lipido- classes for optimizing separation and thereby increasing the number of
mics, which covers a huge diversity of starting material including crops, detected species, but this may compromise the analysis of other lipid
microalgae, roots, shoots and seeds, matrix interference and ion classes. Only recently, a method for plant lipidomics has been devel-
suppression can make the analysis of low abundant species extremely oped that increased the coverage of measured species at a reasonable
difficult [14]. In addition, even if with increasing mass resolution, the time by covering a great number of lipid classes [9]. This approach
high number of possible isobaric ions, which may occur in lipid (Fig. 1) used an organism-specific database, created from a system-wide
extracts, limits the selectivity of the direct infusion approach [15]. In analysis, to generate a targeted list of candidate multiple reaction
the search to characterize the lipidome at a greater depth, it has been monitoring (MRM) transitions. Such an approach provided a high
observed that isobaric interference can be often neglected for major selective and sensitive method for the analysis of 393 molecular lipid
species, whereas for minor species, fragment ion spectra often remain species in leaves from the model plant Arabidopsis thaliana. However, it
unclear and false positives may occur [16]. However, it should be should be noted, that almost all lipidomic methods that rely on the
pointed out that a major strength of all direct infusion or so-called combination of liquid chromatography with mass spectrometry (LC-
shotgun approaches, besides their high speed and comparatively low MS) suffer as quantitative methods. Since the ionizing conditions of
costs, is that at least semi-quantitative data can be obtained by adding lipid molecules at different retention times within a gradient profile
just a reasonable number of isotopically labelled standards. may differ remarkably in LC-MS-derived methods; typically, they would
require a much larger number of internal standards that are often not
2.3. The development of analytical approaches for lipidomics available. Therefore, the main challenge for lipidomics still remains the
development of a single-method with a broad-coverage of almost all
As discussed, the enormous diversity of lipids in biological extracts lipid classes and abundances, whilst retaining the selectivity, sensitivity
has created an ongoing analytical challenge that often requires the and sample throughput of methods that focus traditionally on much
adoption of new strategies for the deconvolution of complex isobaric smaller groups of lipids.

783
R.P. Haslam, I. Feussner BBA - Molecular and Cell Biology of Lipids 1862 (2017) 782–785

Fig. 1. Lipid profiling workflow for a comprehensive organism-specific lipidome analysis.

Fig. 2. Comprehensive and comparative lipidomic strategies for lipid marker identification.

3. A new lipidomic approach [33]. Interestingly the software can be used for systems-wide ap-
proaches, combining multiple data sets from different omics-ap-
More broadly, metabolite analysis with its numerous developments proaches, to identify pathways or groups of lipid markers affected by
in metabolomic workflows is facing the same challenges [30]. In environmental or genetic perturbations. In analogy to this strategy, we
addition to even higher differences in metabolite concentrations, the propose now the term lipid fingerprinting as a non-targeted LC-MS-
major challenge is the great chemical diversity which makes quantifica- based strategy for lipidomics (Fig. 2), because we feel that identifying
tion almost impossible and substance identification difficult [31]. To additional lipid species or even classes will be the next challenge in the
overcome this dilemma a workflow called metabolite fingerprinting has field [34].
been developed which is based on LC-MS approaches [32]. Here, in the Moreover, the same authors suggested the term metabolite profiling
first experimental step, samples of at least two different conditions are for a targeted strategy to measure as many as possible known
analyzed. Data processing steps (peak picking and peak alignment) metabolites in a quantitative way [32]. In order to get semi-quantitative
result in features, which consist of a retention time, an exact mass and data with LC-MS-based strategies, because of their higher resolution
the corresponding intensity profile. Assuming that the analytical and/or coverage in comparison to shotgun approaches, we feel that
conditions for every analytical run are the same, absolute quantification lipidomic profiling is nowadays the most successful strategy. Such a
may not be necessary and the result of the experiment is either an workflow would be formed around LC-MS-based analysis methods,
intensity profile in case of only a single sample that is being analyzed thereby having the advantage of covering more lipid classes and
against an empty control or differential profiles that show the differ- detecting different lipid molecules over a larger range of concentra-
ences or changes between two analyzed samples. Recently the software tions. Furthermore, any lack of quantitative data as discussed above
suite MarVis was developed to support the analysis of these workflows would be compensated by gaining relative lipid profiles through the

784
R.P. Haslam, I. Feussner BBA - Molecular and Cell Biology of Lipids 1862 (2017) 782–785

comparison of different samples, i.e. unstressed vs. stressed or before on multiplexed liquid chromatography–mass spectrometry reveals additional in-
sights into cold- and drought-induced membrane remodeling, Plant J. 84 (2015)
and after a stimulus. The before mentioned workflow (Fig. 1) follows 621–633.
this strategy by using only four different gradients to cover 23 different [10] C. Göbel, I. Feussner, Methods for the analysis of oxylipins in plants,
lipid classes from a plant leaf extract [9]. Lipid profiling is a valuable Phytochemistry 70 (2009) 1485–1503.
[11] X. Han, R.W. Gross, Electrospray ionization mass spectroscopic analysis of human
compromise because it enables the comprehensive analysis of cellular erythrocyte plasma membrane phospholipids, Proc. Natl. Acad. Sci. U. S. A. 91
lipidomes with high lipid class and species selectivity at a reasonable (1994) 10635–10639.
sample throughput. The combination of lipidome profiling and finger- [12] B. Brügger, G. Erben, R. Sandhoff, F.T. Wieland, W.D. Lehmann, Quantitative
analysis of biological membrane lipids at the low picomole level by nano-
printing provides now researchers with complementary strategies for electrospray ionization tandem mass spectrometry, Proc. Natl. Acad. Sci. U. S. A. 94
capturing the diversity of the cellular lipidome (Fig. 2). (1997) 2339–2344.
As our understanding of the role lipids play in cellular metabolism [13] R. Welti, W. Li, M. Li, Y. Sang, H. Biesiada, H.E. Zhou, C.B. Rajashekar,
T.D. Williams, X. Wang, Profiling membrane lipids in plant stress responses. Role of
continues to expand, there is an ever increasing requirement for
phospholipase D alpha in freezing-induced lipid changes in Arabidopsis, J. Biol.
methods that accurately capture lipid turnover and acyl-exchange. Chem. 277 (2002) 31994–32002.
The discipline of lipidomics has for the most part developed targeted [14] R. Harkewicz, E.A. Dennis, Applications of mass spectrometry to lipids and
analytical workflows so far. At first they have been mostly used for membranes, Annu. Rev. Biochem. 80 (2011) 301–325.
[15] H.C. Köfeler, A. Fauland, G.N. Rechberger, M. Trotzmüller, Mass spectrometry
studying individual lipid classes e.g. signaling molecules, while over the based lipidomics: an overview of technological platforms, Meta 2 (2012) 19–38.
last decade methods have been developed for lipid profiling and [16] T. Iven, C. Herrfurth, E. Hornung, M. Heilmann, P. Hofvander, S. Stymne, L.-H. Zhu,
biomarker discovery. However, to better understand metabolic net- I. Feussner, Wax ester profiling of seed oil by nano-electrospray ionization tandem
mass spectrometry, Plant Methods 9 (2013) 24.
works and their response to perturbation, new high-throughput [17] K.A.Z. Berry, R.M. Barkley, J.J. Berry, J.A. Hankin, E. Hoyes, J.M. Brown,
approaches are required that allow increased coverage of lipid classes R.C. Murphy, Tandem mass spectrometry in combination with product ion mobility
and individual species within a system. We propose lipid fingerprinting for the identification of phospholipids, Anal. Chem. 89 (2017) 916–921.
[18] S. Jackson, M. Ugarov, J. Post, T. Egan, D. Langlais, J.A. Schultz, A. Woods, A study
as such a method, capable of expanding the breadth of coverage whilst of phospholipids by ion mobility TOFMS, J. Am. Soc. Mass Spectrom. 19 (2008)
retaining selectivity and sensitivity. 1655–1662.
[19] A. Basit, S. Pontis, D. Piomelli, A. Armirotti, Ion mobility mass spectrometry
enhances low-abundance species detection in untargeted lipidomics, Metabolomics
Transparency Document
12 (2016) 50.
[20] D. Sturtevant, Y.-J. Lee, K.D. Chapman, Matrix assisted laser desorption/ionization-
The http://dx.doi.org/10.1016/j.bbalip.2017.04.005 associated mass spectrometry imaging (MALDI-MSI) for direct visualization of plant metabo-
lites in situ, Curr. Opin. Biotechnol. 37 (2016) 53–60.
with this article can be found, in online version.
[21] P.J. Horn, K.D. Chapman, Lipidomics in situ: insights into plant lipid metabolism
from high resolution spatial maps of metabolites, Prog. Lipid Res. 54 (2014) 32–52.
Acknowledgements [22] S. Marmon, D. Sturtevant, C. Herrfurth, K. Chapman, S. Stymne, I. Feussner, Two
acyltransferases contribute differently to linolenic acid levels in seed oil, Plant
Physiol. 173 (2017) 2081–2095.
The authors are grateful to Kirstin Feussner, Pablo Tarazona and [23] G. Astarita, A.C. Kendall, E.A. Dennis, A. Nicolaou, Targeted lipidomics strategies
Cornelia Herrfurth for preparing the figures and for critical discussions. for oxygenated metabolites of polyunsaturated fatty acids, Biochim. Biophys. Acta
Financial support by the Deutsche Forschungs Gemeinschaft (DFG, 1851 (2015) 456–468.
[24] Richard W. Gross, X. Han, Lipidomics at the interface of structure and function in
INST 186/822-1, ZUK 45/2010, INST 186/1167-1), the systems biology, Chem. Biol. 18 (2011) 284–291.
Bundesministerium fuer Bildung und Forschung (BMBF) to BioFung, [25] A. Fauland, H. Köfeler, M. Trötzmüller, A. Knopf, J. Hartler, A. Eberl, C. Chitraju,
the Biotechnology and Biological Sciences Research Council (BBSRC) to E. Lankmayr, F. Spener, A comprehensive method for lipid profiling by liquid
chromatography-ion cyclotron resonance mass spectrometry, J. Lipid Res. 52
Alpha and Omega and the European Union to ICON and GIAVAP is (2011) 2314–2322.
gratefully acknowledged. We apologize to those colleagues whose work [26] A. Ibrahim, A.-L. Schütz, J.-M. Galano, C. Herrfurth, K. Feussner, T. Durand,
we overlooked or were not able to include because of space constraints. F. Brodhun, I. Feussner, The alphabet of galactolipids in Arabidopsis thaliana, Front.
Plant Sci. 2 (2011) 95.
[27] Y. Okazaki, H. Otsuki, T. Narisawa, M. Kobayashi, S. Sawai, Y. Kamide, M. Kusano,
References T. Aoki, M.Y. Hirai, K. Saito, A new class of plant lipid is essential for protection
against phosphorus depletion, Nat. Commun. 4 (2013) 1510.
[28] E. Cifkova, M. Holcapek, M. Lisa, M. Ovcacikova, A. Lycka, F. Lynen, P. Sandra,
[1] M. Tswett, Adsorptionsanalyse und chromatographische Methode. Anwendung auf
Nontargeted quantitation of lipid classes using hydrophilic interaction liquid
die Chemie des Chlorophylls, Ber. Dtsch. Bot. Ges. 24 (1906) 384–393.
chromatography-electrospray ionization mass spectrometry with single internal
[2] K. Yang, X. Han, Lipidomics: techniques, applications, and outcomes related to
standard and response factor approach, Anal. Chem. 84 (2012) 10064–10070.
biomedical sciences, Trends Biochem. Sci. 41 (2016) 954–969.
[29] W.W. Christie, X. Han, Lipid Analysis - Isolation, Separation, Identification and
[3] M. Wang, C. Wang, R.H. Han, X. Han, Novel advances in shotgun lipidomics for
Lipidomic Analysis, (4th Edition) Ed, Oily Press, Bridgwater, U.K., 2010.
biology and medicine, Prog. Lipid Res. 61 (2016) 83–108.
[30] I. Feussner, A. Polle, What the transcriptome does not tell — proteomics and
[4] J. Folch, M. Lees, G.H.S. Stanley, A simple method for the isolation and purification
metabolomics are closer to the plants' patho-phenotype, Curr. Opin. Plant Biol. 26
of total lipides from animal tissues, J. Biol. Chem. 266 (1957) 497–509.
(2015) 26–31.
[5] E.G. Bligh, W.J. Dyer, A rapid method of total lipid extraction and purification, Can.
[31] R.J. Bino, R.D. Hall, O. Fiehn, J. Kopka, K. Saito, J. Draper, B.J. Nikolau, P. Mendes,
J. Biochem. Physiol. 37 (1959) 911–917.
U. Roessner-Tunali, M.H. Beale, Potential of metabolomics as a functional genomics
[6] V. Matyash, G. Liebisch, T.V. Kurzchalia, A. Shevchenko, D. Schwudke, Lipid
tool, Trends Plant Sci. 9 (2004) 418–425.
extraction by methyl-tert-butyl ether for high-throughput lipidomics, J. Lipid Res.
[32] O. Fiehn, Combining genomics, metabolome analysis, and biochemical modelling to
49 (2008) 1137–1146.
understand metabolic networks, Comp. Funct. Genomics 2 (2001) 155–168.
[7] L. Löfgren, M. Stahlman, G.-B. Forsberg, S. Saarinen, R. Nilsson, G.I. Hansson, The
[33] A. Kaever, M. Landesfeind, K. Feussner, A. Mosblech, I. Heilmann, B. Morgenstern,
BUME method: a novel automated chloroform-free 96-well total lipid extraction
I. Feussner, P. Meinicke, MarVis-Pathway: integrative and exploratory pathway
method for blood plasma, J. Lipid Res. 53 (2012) 1690–1700.
analysis of non-targeted metabolomics data, Metabolomics 11 (2015) 764–777.
[8] L. Löfgren, G.-B. Forsberg, M. Ståhlman, The BUME method: a new rapid and simple
[34] V. Bruckhoff, S. Haroth, K. Feussner, S. König, F. Brodhun, I. Feussner, Functional
chloroform-free method for total lipid extraction of animal tissue, Sci. Rep. 6 (2016)
characterization of CYP94-genes and identification of a novel jasmonate catabolite
27688.
in flowers, PLoS One 11 (2016) e0159875.
[9] P. Tarazona, K. Feussner, I. Feussner, An enhanced plant lipidomics method based

785