Name: David Oxley
Position: Head of Protein Mass Spectrometry Laboratory, Babraham Institute
Background: Researcher, Proteome Systems; PhD, Chemistry, University of Hull
This month, the Babraham Institute in Cambridge, UK, a biomedical research center focusing on the molecular mechanisms of cellular processes and disease, hosted a symposium celebrating the installation of new equipment, including four mass spectrometers, purchased with £2.1 million ($3.4 million) from the UK's Biotechnology and Biological Sciences Research Council.
David Oxley is head of the institute's protein mass spectrometry lab, which provides mass-spec analysis for the institute's various research groups.
He spoke this week to ProteoMonitor about the new machines, how they are facilitating work at the institute, and the use of mass spec in proteomics and metabolomics research generally.
The following is an edited version of the interview.
What types of mass spectrometers did the institute purchase and what sort of work will they be used for primarily?
We got money from the BBSRC to expand the mass spec facility. It was partly for protein [research] and partly for lipid [research]. For the protein work we got the Thermo Orbitrap Velos. For the lipid work we got two Qtrap 4000s from AB Sciex and a [Thermo Fisher] GC-MS, a TSQ Quantum.
What will you be doing in terms of protein and lipid research?
The lipid [work] basically falls into two main areas. [The first is] lipidomics, where in a particular sample we analyze 200 or 300 different phospholipids in a single analysis, comparing the total lipidome from various samples. The second area we're interested in is signaling lipids, particularly phosphoinositides. That's more specialized, looking for a small group of low-abundance lipids in amongst the total lipid extract. The main features we need for that are sensitivity and robustness and being able to quantitate over a broad range of concentrations. So we ended up getting the 4000 Qtrap for that.
For the protein work there's basically going to be one machine, so it had to be as versatile as possible. We obviously wanted something that was the latest technology and had the highest specifications but also that could do a wide range of things. The Orbitrap has very high mass accuracy and resolution, which is useful in just about every aspect of mass spectrometry. On top of that it has very high sensitivity, high scan speed.
[Electron-transfer disassociation] is another feature [of the Orbitrap] that is particularly interesting to us. We do a lot of work on post-translational modifications and ETD is something that we've never been able to do before. [The Orbitrap] can do [Collision-induced disassociate] and ETD. ETD has some advantages. It's complementary to CID in many ways. It gives you some extra information that you can't get just by CID alone. It fragments in a different way, so labile modifications like phosphorylation, which often drop off in CID and aren't detectable or can't be located, are more resistant in ETD and you can actually detect them and localize them to a particular amino acid. It's a whole new level of analysis that we can do with PTMs.
What type of machines were you using before getting the new ones?
The lipid work was done mainly on a Shimadzu ion-trap TOF machine, and the protein work was done on an ABI QSTAR. We still have those machines and they're still used. We also have a MALDI-TOF mass spec that we use for sort of general protein analysis
Given that you already had an ABI machine, did you consider the AB Sciex TripleTOF 5600 machine?
The TripleTOF wasn't available then. It's only just recently come onto the market. This whole process started last year. The tender offer [to vendors] was put out last year, and the TripleTOF wasn't on the market at that stage. But that's an interesting looking machine.
Did you need to upgrade your bioinformatics software when you purchased the new instruments?
Some software comes with the machines, so with Thermo obviously we got the Proteome Discoverer software. We also got the Scaffold software [from Proteome Software] as well.
We're looking at additional software for quantitation so potentially [software for] SILAC, iTRAQ and label-free [quantitation], but we're still assessing our needs, so that's still ongoing. There's MaxQuant [developed by Matthias Mann's group at Munich's Max Planck Institute – ed.], which a lot of people use for doing the SILAC analysis. We're evaluating that. For label-free we're evaluating the Nonlinear Dynamics software Progenesis LC-MS.
What are the primary research interests of the groups you work with at the institute?
Two of the main themes of research at the institute are [protein] signaling and chromatin biology. One of the most common things that we do is identify proteins in signaling complexes and chromatin remodeling complexes. Then, of course, for signaling, identifying new phosphorylation sites is important and being able to measure how phosphorylation changes during some biological process or activation or inhibition of some signaling pathway. Those are the sort of things that we do.
[ pagebreak ]
Identifying and quantifying PTMs like phosphorylation has long been a challenge in mass spec. How are the new machines improving your capabilities on that front?
We've already seen that the quality that we get from the Orbitrap in terms of the mass accuracy and mass resolution helps quite a lot, because a lot of phospho MS-MS spectra are generally quite poor, so having the extra mass accuracy helps to identify the phosphopeptides when the data quality isn't that good, when you're lacking key fragments.
The other thing – which we haven't really fully explored yet – is the use of ETD for [identifying] phosphorylation. We're still testing that out and seeing its benefits in terms of identifying phosphorylation sites, but I'm pretty confident that it's going to enhance things quite a bit. We're doing individual tests of CID alone and ETD alone and then combinations of the two and using a sort of decision-tree approach where you decide whether to do CID or ETD based on the charge state and the m/z value and then comparing which gives the best result.
We've done a couple of comparisons of complex samples – like an enriched phosphopeptide fraction from some particular cells – and we [identified] at least double the number of phosphopeptides with the Orbitrap as we got with the QSTAR.
Do you have any plans to buy additional mass spectrometers in the near future?
Unfortunately, I think that in the current financial environment it would be highly unlikely that we could get another mass spec in the near future. Obviously we'd love to get more. We'd always like more, but I really don't expect to get anything new for several years to come.
If you were able to buy another machine, what would it be?
I've just seen the [AB Sciex TripleTOF 5600]. On paper that looks like a very nice machine and would be well worth looking at, I think. I suppose if I were to buy one I would try to find something that would complement what we've got now. Probably something like the TripleTOF – if it's as good as it sounds. With the extremely high scanning speed and mass accuracy it could be rather nice.
Do you see the industry moving towards developing more qualitative-quantitive machines like the TripleTOF 5600?
I think obviously the Orbitrap has been at the forefront for a few years now. It was so far ahead in terms of mass accuracy and resolution of what was available a few years ago that it sort of took off and has been really successful. All the other manufacturers have, I guess, been trying to catch up with that. Several manufacturers have come out recently with quadrupole-TOF [machines]. The Bruker Maxis machine looks pretty good, and the TripleTOF. High mass accuracy and high scanning speed are a great benefit for looking at complex mixtures, which a lot of people are [doing], and, of course, the faster you can scan, the better, and sensitivity is always a big issue.
With these new products, have the other vendors managed to catch up to the Orbitrap?
I think the Orbitrap is still leading. Again, I haven't tested these machines side by side. You've got to take the manufacturers' claims with a pinch of salt. It seems to me the TripleTOF may have similar but not quite as good mass accuracy and resolution, but it seems to have a big advantage in scan speed. They were claiming 100 Hz. If it really could be used at that sort of speed for real samples, that's a significant benefit over the Orbitrap.