NEW YORK — A team led by researchers at the Francis Crick Institute has used a new data-independent acquisition mass spectrometry technique to enable extremely high-throughput proteomics work.
In a study published last month in Nature Biotechnology, the researchers used Sciex's Scanning Swath method to measure nearly 2,000 proteins in cell lysate in 30-second mass spec experiments. The findings indicate the technique could prove useful for large-scale proteomic experiments, bringing run times down to levels not previously possible.
The approach could also allow DIA mass spec to move into areas where it has thus far been little used — analysis of protein post-translational modifications, in particular, said Markus Ralser, a group leader at Francis Crick and senior author on the paper.
In a traditional DIA mass spec experiment, the instrument steps through wide m/z ranges, fragmenting all the ions in a given window. This avoids the stochastic sampling problems facing conventional shotgun proteomics and consequently improves reproducibility. However, because multiple precursor ions are fragmented in a given window, the technique produces complex, overlapping spectra that must be deconvoluted.
In Scanning Swath, the instrument does not step through discrete chunks of the m/z range but instead uses a sliding quadrupole window. As the quadrupole slides through the m/z range, precursor and product ions can be seen entering and leaving the window, and those entry and exit times provide another parameter linking precursors and products that can be used to resolve complex spectra.
This improved ability to resolve complex spectra means researchers are able to run experiments that use less LC separation time, Ralser said.
"If you squeeze a spectrum that would typically come in a classic one-hour proteomic gradient into a 30-second gradient, you get a massively higher amount of signal interference from overlaid spectra," he said. "But this additional data dimension helps us a lot to deconvolute it."
Sciex introduced the method at the 2019 American Society for Mass Spectrometry annual meeting, but has not yet made it commercially available.
Ralser and his colleagues published a bioRxiv preprint in May 2019 describing the use of Scanning Swath in combination with his lab's DIA-NN software, which combines signal correction strategies to reduce interference with neural networks to assign the confidence of peak identifications to boost performance even when using very short LC gradients. In that preprint, they quantified around 3,000 proteins in cell digest samples running five-minute high-flow chromatography gradients.
Since then, Ralser said his lab has been improving the software for the Scanning Swath approach. For instance, they have included Scanning Swath functionality into the DIA-NN software suite, which Raiser said is currently the only software package for deconvoluting Scanning Swath data.
Ralser said that to implement the approach users also need Sciex to enable scanning mode on their instruments. He and his colleagues developed the software on Sciex's TripleTOF 6600 mass spec, but he suggested that the company could be planning to launch a commercial version of Scanning Swath as part of a new instrument release. He noted that while he did not have any particular insight into Sciex's plans, the 6600 has now been on the market for a number of years, "and clearly Sciex is working on a successor instrument, and so this is of course a workflow that will boost the successor of the 6600."
Sciex launched the 6600 in 2014. In 2019 it launched its TripleTOF 6600+, which it said at the time would eventually be able to run Scanning Swath. In an email this week, Arianna Jones, senior manager at Sciex, called Scanning Swath the "next evolution" in the company's DIA workflow offerings, but declined to say when it might put out a commercial version.
In the Nature Biotechnology study, Ralser and his colleagues found that using LC gradients of between one and five minutes they quantified at least 70 percent more precursors than they did with conventional Swath experiments. Specifically, they found that in a five-minute Scanning Swath experiment they were able to identify 5,004 proteins compared to 3,594 proteins in a five-minute experiment using a Thermo Fisher Scientific Orbitrap Exploris 480 with FAIMS separation and an Evosep One LC. Scanning Swath also significantly outperformed conventional Swath in terms of precision, the authors noted.
While the researchers demonstrated the method's ability to measure large numbers of proteins in experiments as short as 30 seconds, Ralser said his lab has arrived at an LC gradient of around two minutes and 30 seconds as their standard for the approach.
He said his lab uses an LC system with two columns, one of which is washed and equilibrated while the other is actively separating, which allows the researchers to switch back and forth between them while only losing a few seconds of time between separations.
"It's a runtime of two minutes and 30 seconds per proteome," he said. "With the 30-second runs, it is nice to show just how fast it can be and there may be some applications for like low-complexity drug screen type of work, but we haven't had applications where it really made sense to go down to 30 seconds."
Beyond the increase in throughput, Ralser said an advantage of using high-flow chromatography is the increased stability compared to approaches like nanoflow LC, which he noted is a key consideration for the sort of large-scale experiments his lab often undertakes.
"Our main application at the moment is clinical proteomics with serum and plasma samples, so we are doing one study with 25,000 human proteomes and another with 15,000," he said. "There you need to have extremely good comparability between the first sample and sample number 25,000. So we use [Scanning Swath] for all of those projects where we need throughput and we need data that is consistent and of high quality."
"We don't use it for those projects where you have just, say, three samples and you want tease out every single peptide that you can possibly quantify," he said.
Ralser said he sees additional applications for Scanning Swath beyond improving throughput, highlighting in particular DIA analysis of post-translational modifications. This has been a challenge for DIA due to the complexity of the spectra generated.
"If you have an acquisition window of 10 or 20 m/z, you struggle to properly assign the PTMs [to peptides]," he said, adding that the ability of Scanning Swath to better link precursor and product ions should improve DIA PTM analysis.