New Mass Spec Fragmentation Technique Could Boost Shotgun Mass Spec Coverage

By May 24, 2017 No Comments

by Adam Bonislawski

NEW YORK (GenomeWeb) – Researchers at the University of Wisconsin-Madison have implemented a mass spec fragmentation technique that could significantly increase the depth of coverage of shotgun proteomics experiments.

Described in a pair of studies published this month in Analytical Chemistry, the method is an improved version of electron transfer dissociation (ETD) fragmentation called activated ion ETD (AI-ETD) that uses photoactivation of ions during ETD to more thoroughly fragment peptides for mass spec analysis.

In the Analytical Chemistry studies, the UW-Madison researchers showed that the method nearly doubled the peptide identifications made using standard ETD while more than tripling the number of peptide phosphosites identified compared to traditional ETD methods.

As the name implies, ETD fragments peptides by transferring an electron to them, which causes the peptide backbone to break apart. Developed more than a decade ago, the method has grown in popularity in recent years due to its ability to retain many post-translational modifications and to provide even fragmentation along the entire length of large peptides.

ETD struggles, however, with peptides with low charge states or high mass-to-charge ratios, said Josh Coon, professor of chemistry at UW-Madison and senior author of both studies. Such molecules are able to fold in on themselves, which allows hydrogen bonding to keep them together even after the peptide backbone has been fragmented by ETD.

“So the hydrogen bonding stays intact and it looks like we didn’t get any dissociation,” Coon noted, adding that this phenomenon is known as “ET no D” for electron transfer with no dissociation.

AI-ETD tackles this problem by exciting the peptide with a laser before and during the ETD process, which keeps the molecule unfolded so that the hydrogen bonds that keep it together don’t form.

“It gives it enough vibrational energy to unfold, so that when the electron transfer happens, the cleavage will occur and the [peptide] will readily separate,” Coon said. “You can take this very simple step of having the [peptide] ions in this environment, where they receive low-energy photons and ETD efficiency goes way up.”

In the past, proteomics researchers have combined ETD with other fragmentation techniques to improve overall fragmentation. For instance, electron-transfer/higher-energy collision dissociation fragmentation (EThcD), has seen uptake in a number of labs. That method was introduced by a 2012 paper in Analytical Chemistry, published by Utrecht University researcher Albert Heck.

Coon said he and his colleagued found that AI-ETD also outperformed EThcD and other combined fragmentation methods. One challenge of EThcD, he said, was the free-radical-driven chemistry generated by following ETD with HCD, which “can throw your search engines for a loop.”

Nicholas Riley, a graduate student in Coon’s lab and the first author on the two new Analytical Chemistry papers, added that the fact that the laser excitation occurs during the ETD process also boosts its performance compared to EThcD, as it allows for faster mass spec cycle times.

“Having the laser on only during the ETD reaction and not having to do any activation after the reaction is completed makes the entire scan sequence faster,” he said. He added that the manipulation of ions involved in combining fragmentation methods like ETD and HCD can also increase ion loss, which can lead to a reduction in sensitivity.

Riley said the researchers achieved a 10 to 15 percent boost in peptide IDs using AI-ETD compared to EThcD and a roughly 25 percent boost in phosphopeptide identifications.

Coon noted that he and his colleagues actually developed the AI-ETD approach five or six years ago and implemented it on a Thermo Fisher Scientific Orbitrap instrument at the time, but that the architecture of that generation of Orbitraps made it too technically challenging to pursue as a method for general use.

“We always had our eye on using infrared photoactivation [combined with ETD],” he said. “It took us a few years to get around to being able to build a device to test it on, because getting the infrared photon beam into the ion trap was technically pretty challenging. So we had to shelf that for a while, knowing that we had this really great solution, but that it really wasn’t terribly simple for the current hardware configuration.”

The architecture of Thermo Fisher’s current top-of-the-line instrument, the Orbitrap Fusion Lumos, is much more amenable to the approach, Coon said. “The backside of the ion trap is completely open, and it’s very simple now to introduce the photon beam. So we, through our partnership with Thermo, modified the system to introduce the laser beam into that space.”

Coon said he didn’t know if Thermo Fisher had any plans to develop a commercial version of the AI-ETD-capable instrument used in the Analytical Chemistry papers, but he said they had shown interested in the idea.

“I can’t speak for the company, but I would think it’s likely to happen, because I think it will make a big difference for people,” he said. “The addition of this laser to the system to complement and improve the quality of the dissociation is a pretty big deal in terms of the data quality.”

Coon said that his lab now uses the AI-ETD method for any experiments in which ETD would have been used. “There’s really no penalty for doing it,” he said.

The researchers are also testing the approach in other types of molecules where ETD appears to have potential but has struggled due to the issue of hydrogen bonding. Glycoproteomics is one such area, Coon noted.

“Glycosylated peptides and proteins have traditionally been almost impossible to do with collisional activation methods,” he said. “ETD had a lot of promise, but, because [glycopeptides] were often in a low-charge state, they would not perform very well under ETD. But now, with this tool, we think we’re going to really be able to get them.”

This article was not written by Coon Labs. Please see the original, featured on GenomeWeb.