Skip to main content
Premium Trial:

Request an Annual Quote

FAIMS Could Enable Deep Proteome Coverage Without Fractionation, UW Team Finds


NEW YORK (GenomeWeb) – Researchers at the University of Wisconsin-Madison and Thermo Fisher Scientific have completed a study demonstrating the ability of High Field Asymmetric Waveform Ion Mobility Spectrometry (FAIMS) to improve depth of coverage in shotgun proteomics experiments.

The study, published last week in Analytical Chemistry, indicates that FAIMS could be used in place of offline fractionation to streamline proteomics experiments without sacrificing depth of analysis. It also suggests renewed interested in the technology at Thermo Fisher, which launched a FAIMS device more than a decade ago, but saw little uptake among proteomics researchers, and has not promoted the technology for use with its instruments since.

The company declined to comment on any FAIMS work it is currently undertaking, but Alexander Hebert, first author on the Analytical Chemistry paper and an assistant scientist in the lab of UW-Madison research Joshua Coon, senior author on the study, said the FAIMS device used in the study was a newer-generation system that appeared to perform better than the initial instrument.

A form of ion mobility, FAIMS works by running mixtures of ions at atmospheric pressure between two electrodes and subjecting them to alternating high- and low-field conditions by applying an asymmetric RF waveform, which separates them based on differences in ion mobility under high and low electric fields. Depending on their mass and charge, some ions will drift into the electrodes while others will pass through the device, allowing, for instance, a researcher to select low-abundance peptides for analysis from a tryptic digest while weeding out higher-abundance ions.

The technology can be inserted between a traditional liquid chromatography system and a mass spec instrument, allowing for an additional level of separation.

Waters was the first large mass spec vendor to incorporate a FAIMS device, licensing the technology in 2003 from biotech firm Ionalytics. After Waters chose not to renew the license, Thermo Fisher Scientific — called Thermo Electron at the time — acquired Ionalytics and released a FAIMS device in 2006, but the system failed to catch on with proteomics researchers.

Sciex introduced a FAIMS-based system with the 2011 launch of its Selexion technology, which it has incorporated into various of its mass spectrometers.

Hebert said that a major problem with Thermo Fisher's initial implementation of FAIMS technology was poor transmission of ions from the electrospray ionization (ESI) source into the FAIMS device, but that the new device achieved significantly higher transmission levels. In 2014, Thermo Fisher researchers published a paper in the Journal of the American Society for Mass Spectrometry detailing a new ESI-FAIMS interface that offered a ninefold improvement in ion transmission and provided for an overall ion transmission rate of around 70 percent.

Coon said his lab began looking into FAIMS as part of its larger interest in simplifying mass spec sample prep and maximizing the number of proteins it can identify in single-shot experiments. Traditionally, when depth of coverage is a priority, researchers have used offline fractionation in which a sample is divided using chromatography into multiple fractions, each of which is then run separately via LC-MS.

The question, Coon said, was "could you skip the two-dimensional fractionation and get comparable results by relying on the FAIMS device?"

Hebert noted that the researchers were also interested in FAIMS due to their observations that mass spec performance had outstripped the capabilities of upfront separation techniques.

"When the [Thermo Fisher Orbitap] Lumos Fusion came out, we would run a sample and we would get hundreds of thousands of spectra, and they looked really nice, but we weren't really getting many more [peptide] identifications," he said.

That led the researchers to the realization that standard chromatography systems weren't sufficient to take full advantage of the speed of newer high-end mass spectrometers. They detailed this finding in a 2016 commentary in Cell Systems in which they demonstrated that with high-performance LC columns they built in-house, they were able to more than double the number of peptides they could identify in an analysis of whole yeast digests.

While fractionation addresses this problem, that solution is at odds with Coon's goal of developing streamlined, single-shot workflows capable of deep coverage. Another option, which Coon said the lab is exploring, is increasing the length of the LC columns used and the pressure used.

FAIMS provides an additional possibility, Hebert said. "When [Thermo Fisher] came and told us that they have this FAIMS [device] that was working pretty well, we kind of perked up, because we thought, 'OK, this is a way to essentially increase the peak capacity on our upfront separations.'"

In the Analytical Chemistry study, the researchers combined the FAIMS device with an Orbitrap Fusion Lumos mass spec to analyze the proteome of the human cell line K562.

Without using fractionation, they were able to identify more than 100,000 peptides and more than 8,000 proteins in a five-hour experiment. A six-hour, single-shot FAIMS experiment identified 8,007 proteins compared to 7,776 proteins for an experiment that ran a sample separated into four fractions using four 90 minute gradients.

Coon said that the results suggest that FAIMS could replace modest amounts of fractionation, adding that his lab has not had a chance yet to test how it compares against experiments using more extensive fractionation.

"If you wanted to make 20 fractions from a sample, this wouldn't necessarily replace that, but it might augment it," he said. "Or maybe you would only have to do 10 [fractions] instead of 20. We haven't gotten a chance to test that yet."

"If you really wanted to get 10,000 proteins or more, you'd still probably have to do fractionation," Coon said. "But if you wanted to get 8,000 and do it in a relatively short time with very simple sample prep, this is pretty convenient."