Jeffrey Perkel has been a scientific writer and editor since 2000. He holds a PhD in Cell and Molecular Biology from the University of Pennsylvania, and did postdoctoral work at the University of Pennsylvania and at Harvard Medical School.
They say a rising tide lifts all ships, and that holds true for ‘omics, too. The rapid advances in mass spectrometry (MS) technology that have propelled proteomics over the past few years have given other ‘omics disciplines a leg up, as well.
One such discipline is lipidomics. A subset of metabolomics, lipidomics concerns itself with cataloging and quantifying lipids, fatty acids, sterols and the like, whether in cells, tissues or entire organisms.
Given that most lipids are implicated in building cellular and subcellular membranes, that may sound dull, but it’s not, says Andreas Huhmer, director of marketing for proteomics tools at Thermo Fisher Scientific. “Lipids are structural molecules, but they’re also signaling molecules.” One classic example is the conversion of the membrane phospholipid phosphatidylinositol-4,5-bisphosphate into diacylglycerol and inositol triphosphate—second messengers that subsequently can activate downstream kinases and induce release of intracellular calcium stores.
More than one way to slice a lipidome
One key tool for studying lipids, of course, is mass spectrometry. But, as with proteomics and metabolomics, there’s more than one way to slice a lipidome and different mass spectrometry strategies to consider. It all comes down to what a researcher is trying to accomplish.
According to Giuseppe Astarita, principal scientist for metabolomics and lipidomics at Waters Corp., there are three broad classes of lipidomics research: untargeted (i.e., discovery-mode lipidomics), targeted (i.e., lipid quantification and validation) and imaging.
Targeted and untargeted approaches typically require upstream chromatographic lipid separation of lipid extracts, Astarita says, for instance using a C18 with ultrahigh-performance liquid chromatography, supercritical fluid chromatography or two-dimensional liquid chromatography. Imaging uses a laser or other surface-based, ambient desorption/ionization and ablation strategy to introduce molecules into the mass spectrometer while retaining their spatial context. The result is a map of an ion’s molecular abundance, the equivalent of viewing, say, one color channel in a red-green-blue (RGB) digital image.
According to Astarita, most imaging-mass spectrometry studies actually profile lipids, whether the researchers want to or not. “Turns out, when you put a biological sample underneath these sources, because lipids are so abundant in cells, what happens is the signal you get from the ionization process is mostly lipids, and they ionize very well.” Indeed, researchers have discovered they can use these lipid signatures to distinguish healthy from cancerous tissue in vivo, a fact that Waters is exploring through its recent acquisition of MediMass, a company that was pursuing a surgical application of this approach called the iKnife.
Traditionally, targeted lipidomics has been accomplished with a simple triple quadrupole, which offers only nominal mass accuracy but a wide dynamic range and exceptionally good quantification. Alternatively, researchers can use higher-end tandem instrumentation, such as a quadrupole-time-of-flight (Q-TOF), using the quadrupole as an ion filter and the TOF for fragmentation analysis.
Discovery-mode studies require high mass-accuracy instruments, for instance, using Orbitrap or Q-TOF mass analyzers. That’s because chemically similar lipids—those differing by just a single hydrogen, in fact—can exhibit dramatically different biology, says Huhmer. The addition or subtraction of just one hydrogen can introduce a kink in a fatty-acid chain and increase or decrease molecular fluidity and aggregation, for instance, Huhmer notes. “Rather small structural changes can have a large biological impact.”
Imaging studies also require high mass accuracy, but coupled to a source capable of extracting ions from a mounted tissue sample, such as desorption electrospray ionization (DESI) or matrix-assisted laser desorption/ionization (MALDI), Astarita says.
Ion separation is key
A good lipidomics-focused mass spectrometer requires more than a good mass analyzer, however. Multiple ion-fragmentation strategies, for instance, provide greater latitude for probing molecular structure. Thermo’s new Orbitrap Fusion™ Lumos™ Tribrid™ mass spectrometer supports CID, HCD and ETD, Huhmer says, providing a suite of options for fragmenting (and thus dissecting) novel interesting molecular species.
Another popular feature for lipidomics mass spectrometry is ion-mobility separation. Available on such instruments as Waters’ Synapt™ HMDS and new Vion™ IMS QTof and Agilent Technologies’ 6560 IMS Q-TOF, ion-mobility spectrometry (IMS) is a gas-phase strategy that separates molecules by their collisional cross section, or shape. Among other things, IMS fractionation provides an orthogonal separation strategy following chromatography, thereby enabling researchers to differentiate isobaric molecular species that may be otherwise impossible to resolve. IMS “gives a fourth dimension of separation for resolving lipid subclasses and isomers,” says Steve Madden, software product manager at Agilent.
IMS is also featured in SCIEX’s new Lipidyzer™ lipidomics platform. According to Fadi Abdi, senior global market manager for lipidomics, metabolomics and imaging at SCIEX, Lipidyzer combines SCIEX’s QTRAP® mass spectrometer with sample preparation kits and data-analysis tools from metabolomics services provider Metabolon® in one integrated, easy-to-use package; included is a differential ion-mobility separation (DMS) technology called SelexION®. According to Abdi, the DMS cell enables researchers to distinguish even structural isomers. “Because the dipole moment associated with each lipid class is different and forms a linear relationship with the compensation voltage, the SelexION device is able to separate one from the other and remove isobaric interferences.”
Researchers simply prepare their samples using SCIEX’s proprietary sample-preparation kits, which include more than 50 internal standards, inject them onto the Lipidyzer platform, and the software takes care of the rest, configuring and running the mass spectrometer and outputting the biological findings. The system can automatically identify and quantify more than 1,000 lipid species at once, Abdi says. “There is no method development, no experimental design on behalf of the user.” He adds, “The platform is so cool that the user doesn’t even see a single spectrum or a single peak. It is literally taking samples directly to the biology.”
SCIEX also supports an alternative discovery-mode lipidomics strategy on its TripleTOF® mass spectrometers. Called MS/MSALL, the strategy is a data-independent acquisition protocol that (like SWATH® acquisition) enables researchers to collect spectral fragmentation data on all molecules in a complex mixture for subsequent analysis, even if they weren’t aware of a given molecule’s significance at the time of data collection. Thus, researchers can retrospectively re-mine a dataset for any species that may be of value in a later study without having to recollect the data. (Waters offers a data-independent acquisition mode of its own, called MSE.)
GC and LC are OK
Many researchers use liquid chromatography (LC)-coupled mass spectrometers for lipidomics work (or, alternatively, direct sample infusion), but some prefer gas chromatography (GC) separation instead of, or in conjunction with, LC-MS. Thermo launched its Q Exactive GC Orbitrap at the American Society for Mass Spectrometry conference earlier this year; using it, researchers can analyze those lipids that do not ionize or fractionate well by LC, Huhmer says—although chemical derivitization may be required to do so.
“Traditionally people do [lipid analysis by] GC, but the number of lipid classes you can cover is limited. But the depth of what you can cover is probably better with LC,” he says. For instance, long-chain fatty acids may resolve better on GC columns, whereas sterols generally work well with both technologies.
What do the data mean?
However you collect your lipidomics data, the goal is to make sense of the information. Lipidomics-focused analysis packages do exist, including Thermo’s LipidSearch and PREMIER Biosoft’s SimLipid software.
A bigger challenge is integrating data across data types, says Wiktor Jurkowski, integrative genomics group leader at The Genome Analysis Centre (TGAC) in the United Kingdom. Jurkowski has developed a software package called ONION, which uses pre-existing knowledge of biochemical pathways and combined lipidomic and transcriptomic data to improve the identification of genes that may be implicated in metabolite changes.
Whether the resulting hits truly represent the underlying biology, of course, remains to be determined. But thanks to new advances in mass spectrometry technology, figuring that out will be easier than ever.