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.
These days it’s a given that researchers can sift through thousands upon thousands of molecular species to find those that differ with disease, developmental state, drug treatment and more. Whether it’s peptides or metabolites, lipids or sugars, the scientific community takes it on faith that, if there’s a molecular biomarker to be found, they can find it.
To an ever-larger extent, the tool researchers use to find, identify and quantify those molecules is mass spectrometry. But not just any mass spec: To be confident that a peak represents what the researcher thinks it does, he or she must be confident that each spectral peak represents a distinct molecular species, and that its measured mass is as close to the “true” mass as possible.
Enter high-resolution mass spectrometry
As its name implies, a high-res mass spectrometer is any instrument capable of resolving closely spaced spectral peaks, such that what appears to be, say, a broad peak at low resolution, comes into focus as multiple distinct, sharp peaks as resolution increases. At high resolution, compounds with identical “nominal masses” but different actual masses can be distinguished—such as the trimethyl and acetyl modifications of histone proteins, which differ in size by just 0.036 Da.
That resolution also enables researchers, especially in metabolomics, to compute not just the elemental composition but even the isotopic ratios of unknown species, data from which they can constrain the set of metabolites that could possibly have generated them. In other words, given a sufficiently well resolved and mass-accurate set of peaks, they can identify the molecules they represent, even if those molecules have never before been observed.
Resolution typically is expressed as a unitless number that is the ratio of peak mass-to-charge ratio (m/z) divided by the peak width at half its height (i.e., full-width/half-maximum or FWHM). Thus, a peak at m/z 400 that is 1 m/z unit wide at FWHM yields a resolution of 400; if that same peak is 0.025 m/z wide, the resolution is 16,000. (Clearly, resolution is a function of mass.)
There is no hard and fast cutoff definition for high-resolution mass specs, but in general, says Matthew Willetts, group leader for electrospray ionization applications at Bruker Daltonics, low-resolution instruments such as ion traps and quadrupoles tend to offer “unit resolution”—that is, the ability to resolve peaks separated by about 1 m/z unit—with FWHM resolution values of about 1,000. High-res instruments have resolution specs in the range of about 10,000 and up, into the millions.
These days, researchers can choose between three basic flavors of high-res mass spectrometers: Orbitraps, quadrupole-time-of-flights (qTOF) and Fourier transform-ion cyclotron resonance (FT-ICR) instruments.
Orbitrap mass spectrometers
Available exclusively from Thermo Fisher Scientific, Orbitrap mass spectrometers calculate the mass of an ion based on the axial frequency measured as it rotates around the central electrode inside the trap itself. (This is the same basic concept underlying FT-ICR instruments, which differ from Orbitraps in the manner in which they induce that orbital rotation.)
The company offers five basic flavors of Orbitrap mass spectrometers, including the Thermo Scientific Exactive (standalone Orbitrap); Thermo Scientific Q Exactive (hybrid quadrupole-Orbitrap); and three linear ion trap-Orbitrap configurations, the Thermo Scientific LTQ Orbitrap XL, Thermo Scientific Orbitrap Velos Pro and Thermo Scientific Orbitrap Elite. Of these, the Q Exactive™ and the Orbitrap Elite™ are the two newest, says Iain Mylchreest, vice president of research & development for chromatography and mass spectrometry applications at Thermo Fisher Scientific.
The benchtop Q Exactive, essentially a triple-quad MS/MS instrument with an Orbitrap in the Q3 position, has a resolution of 140,000 and is capable of screening, quantifying and confirming molecular identity in one run. The Orbitrap Elite, with an LTQ Velos ion trap coupled to a “high-field” Orbitrap, offers a resolution of 240,000 at m/z 400 at one scan/second (1 Hz). “That’s the highest mass resolution in the industry other than some high-field FT-ICR systems,” says Mylchreest.
At that resolution, says Mylchreest, the Orbitrap Elite can provide mass accuracy of 1 to 2 parts per million (ppm) using external calibration—that is, without adding internal standards to the sample. “You can differentiate [masses] to the fourth decimal place,” he says.
To determine mass accuracy, calculate the difference between the observed mass of an ion and its actual mass, divide by the actual mass and multiple by one million. For instance, as illustrated in this review, "Debating Resolution and Mass Accuracy in Mass Spectrometry", from Spectroscopy Online, if an ion of calculated m/z 400 is observed at m/z 400.002, the mass accuracy is (0.002/400) x 1,000,000 = 5 ppm. High-resolution instruments usually offer high mass accuracy, but the one does not necessarily follow the other, says Willetts. An instrument can be high-resolution and yet have poor mass accuracy. But a machine with poor resolution is unlikely to offer high mass accuracy unless the sample it is measuring is very pure (that is, no other peaks overlap it.)
Orbitraps are popular choices for labs looking for high-resolution mass spectrometers. Ruedi Aebersold’s lab at the Institute of Molecular Systems Biology at ETH Zurich has three of them, including an Orbitrap Elite that the lab installed just this spring. According to Alexander Leitner, a senior assistant in the lab who oversees Aebersold’s MS collection, the lab uses these high-res instruments for proteomics discovery work.
The lab, Leitner explains, has pioneered the use of “targeted proteomics,” a technique in which researchers search for and quantify specific, pre-determined molecules using multiple reaction monitoring (MRM). In that case, they typically use triple-quadrupoles or AB SCIEX QTRAP® (hybrid triple-quad/linear ion trap) mass specs, which sacrifice resolution for sensitivity and quantitative acumen, detecting and quantifying molecules by their characteristic fragmentation products.
The lab uses the high-res instruments at an earlier stage, when trying to identify those molecules in the first place. “The main reason we use high-resolution mass spectrometers is for the higher mass accuracy that comes to some extent with high-resolution, and that is important for better [molecular] identification,” Leitner says.
Quadrupole-time-of-flight mass spectrometers
In addition to its Orbitraps, Aebersold’s lab has a pair of high-resolution qTOF instruments. These are essentially triple-quadrupoles in which a time-of-flight mass analyzer stands in for Q3. Available from Agilent Technologies, Bruker Daltonics, AB SCIEX and Waters, qTOFs offer resolutions in the tens of thousands, ranging from about 17,500 for the Bruker Daltonics MicroTOF-Q2 to 60,000 for the Bruker Daltonics maXis 4G.
But these instruments are not merely high-resolution; many also are exquisitely sensitive, to the point, says Willetts, that some are beginning to encroach on the quantitative work normally performed by triple-quads, with an approach called “Qual/Quant.”
In traditional biopharmaceuticals development, for instance, researchers typically identify specific metabolites of interest and program triple quads to search for them using MRM experiments. The problem with this approach is that only those preselected molecules are monitored and recorded; everything else is ignored.
But qTOFs don’t work like that, Willetts explains. Instead, they collect full-scan spectra. That means researchers can collect everything and then monitor for particular drugs or metabolites during data processing. “If you later discover a new metabolite, you can look again at the data to see if it’s there,” he says.
Aebersold’s lab, for instance, recently developed a workflow called SWATH-MS for use on the AB SCIEX 5600, of which the lab has two, says Leitner. (The company’s newest qTOF, the 4600, which offers resolution of 30,000 and was launched in early May, is not compatible with SWATH-MS, Leitner says. In SWATH-MS, as reported in Molecular and Cellular Proteomics, the instrument fragments and quantifies all ions across a given mass range—say m/z 400 to 1,200—in 25-m/z-wide windows (or “swaths”) as they emerge from the liquid chromatography column. The resulting dataset can then be probed post hoc to identify molecules that were not even on a researcher’s radar when the data were originally collected.
“What you get is a multiplexed tandem mass spectrum, because [each window contains] not just one precursor but multiple precursors,” Leitner explains. That’s kind of like a shotgun approach, Leitner continues, except that the ions are identified not using a conventional database search but rather by the ions they expect to see based on precalculated m/z ratios. “So we are not doing database searches, but looking at fragment ions that we predict are there.”
“This is a very powerful workflow,” says Dominic Gostick, senior director for the academic business at AB SCIEX, about SWATH-MS. “This could be the technology of the future and could change the way people do their experiments.”
FT-ICR mass spectrometers
For those who want the top-of-the-line in resolution, there are the FT-ICR mass spectrometers. Like the Orbitrap, FT-ICRs measure mass based on the orbital track of ions cycling inside the instrument, but their resolution is a function of the field strength of the superconducting magnets that drive them. The Bruker Daltonics solariX, for instance, comes with magnets of 7 Tesla (T), 9.4T, 12T and 15T. An 18T magnet was recently announced but is not yet widely available.
According to product manager Michael Easterling, the 7T solariX has a mass resolution of one million, four times that of the Orbitrap Elite. For the 15T version, he says, “We spec that out at double that amount.” But that resolution comes at a significant cost. Though he declines to disclose pricing, Easterling estimates that the cost of a 7T is about one-and-a-half times more than that of a high-end qTOF—between $800,000 and $1 million. The 15T instrument costs about four times what the qTOF would cost, he says.
Still, Easterling says, for certain applications, an FT-ICR MS makes perfect sense. For instance, given sufficient resolution, researchers can distinguish distinct post-translational forms of intact proteins (Tran JC, et al. Nature 2011)—a workflow called “top-down” proteomics that differs from the more common “bottom-up” proteomics in that the proteins are not first digested with proteinases. Another application is drug metabolism and pharmacokinetics, or DMPK. In one example, researchers at GlaxoSmithKline used a MALDI-coupled FT-ICR to do “MALDI imaging” of drug metabolites in an animal model (Castellino S, et al., Bioanalysis 2011). In one figure in that review, the researchers distinguish four metabolites within 0.25 m/z units of one another, all without the chromatographic separation that normally helps resolve closely spaced molecules.
“You need an extremely high-performance detector that’s capable of separating these fragments in m/z space,” Easterling says. Otherwise, “you’re going to miss that information.”
Buying advice
When making a purchasing decision, don’t just rely on printed specifications, says Lester Taylor, LC/MS marketing director at Agilent Technologies, whose 6550 iFunnel qTOF system couples resolution of up to 40,000 with enhanced sensitivity thanks to redesigned ion injection optics. The best option may not necessarily be the one with the highest resolution or tightest mass accuracy. Just as when buying a car, he says, “there are lots of subjective factors.”
How easy is the system to use? How much will the hardware set you back, and what are its anticipated running costs? How user-friendly is the analytical software? Most importantly, can the instrument detect and quantify your molecule(s) of interest in your specific samples? And then there are the more quantifiable variables, like sensitivity, dynamic range and data quality.
Take the opportunity to run your own samples on a test instrument, either by visiting a company demonstration lab, or virtually. Agilent, for instance, has several “Centers of Excellence” around the world that customers can visit to take an instrument for a test drive, says George Stafford, director of R&D for LC/MS hardware.
“There are two components to that,” Stafford says. “Can you get the right answer, and how easy is it to do that?”
The image at the top of the page is Agilent Technologies' 6550 iFunnel Q-TOF LC/MS System.