NEW YORK, June 7 - George Stafford, director of research at Thermo Finnigan, is regarded as one of the great minds behind ion trap mass spectrometry (MS). Back in the early 1980’s he helped to redefine and commercialize the technology, which is today a staple in proteomics research.
Recently, the American Society for Mass Spectrometry recognized Stafford’s contributions, awarding him with their Distinguished Contribution in Mass Spectrometry Award.
GenomeWeb caught up with Stafford to discuss his contribution to mass spec and the future of the increasingly important technology.
GW: How does the ion trap work?
STAFFORD: MS always involves taking molecules and making them into ions and that means you put an electrical charge on them. Then they are introduced into a high vacuum, something like a hundred millionth of an atmospheric pressure. This is to prevent them from reacting, let’s say, with air or water. Then they are mass analyzed – this is what the ion trap does – and then they are detected. The scanning techniques we came up with involved the ion trap itself.
There are focusing fields, that basically are trap ions. You trap a range of masses of ions, and then after they’re trapped, you do analysis on those ions by selectively ejecting them out for detection. And by doing this technique, by storing a range of ions, and selectively ejecting them, what you have accomplished is the ability to do very fast analysis – you get lots of data quickly – at low levels. Initially for GC/MS that was a very nice improvement over other techniques, but then when you get to very complex samples, like those involved in proteomics, then we can do something that’s also very nice. Since you’re trapping ions, you can do multiple stages of mass analysis because it’s simple. You put a bunch of ions in the trap and then you go through and do one stage of mass analysis.
Since you’re trapped, you can repeat that several times. This has been referred to as tandem mass analysis. And for complex mixtures this is very powerful and it improves your specificity, the ability to say do you have this peptide versus another peptide which may be similar. The ability to do tandem mass analysis increases your specificity enormously.
GW: What was missing from MS that required the invention of the ion trap?
STAFFORD: In 1979, there was not even a hint or a clue that you could do things in mass spectrometry that you can do today. It was not used for biotech analysis or not even very much in pharmaceutical work. Mainly it was used at that time in universities and research labs for environmental analysis, such as looking at pesticides and smaller molecules, and also in petrochemical research. A lot of early work in MS was done with petrochemicals. So there was really no suggestion in my mind that we would use MS for proteomics or pharmaceutical work where you’re looking at very complex mixtures like human blood serum.
At the time I was thinking about [the ion trap device] as a relatively small and inexpensive analyzer that might be useful for something like GC/MS. It’s a combined technique where you couple together a gas chromatographer with a mass spectrometer. [The technique] was used extensively at the time for environmental analysis, and it was a large market for our company. So I saw [of the ion trap] as a relatively simple and hopefully inexpensive device to manufacture. There was a fair amount of research that’d been done on the device but it had never been commercialized. There had been a couple of attempts but it had never really been successful.
I had run across it in the literature and seen early work and early researchers. I had gone to a conference in Seattle in 1979 and heard a paper by Ray March and just got interested in it. I spoke to him after the meeting, came back to Finnigan and spoke to my boss about some alternative ways to operate the device that would make it commercially successful. And that was the start. And the whole area of proteomics hadn’t even been dreamed of.
The dream of being able to do that with MS only occurred within the last two to three years probably. So there were many other developments in MS that led to where we are today aside from ion trap MS.
GW: Were there milestones along the way that allowed the ion trap to go from being a tool for GC/MS to being a proteomics tool?
STAFFORD: There are a couple key things that I can think of. I mentioned earlier that you have to make ions out of the molecules, and there were several ionization techniques that came along which are very well suited for proteomics. One is called electrospray ionization and that basically enables you to take a liquid directly and form ions of all the components in it, and to be able to ionize very large molecules, like peptides. Somewhere in the mid-80s it was first disclosed and we started seeing it become more and more popular. It’s a very good technique for taking a liquid and making ions out of very large molecules.
Some ionization techniques are very violent and for something like a peptide you would wind up breaking it into a whole bunch of small pieces. When you do that it makes it a little more difficult to figure out what you started with. Before you break something into pieces you want to look at the whole thing first. And then maybe later break it up to look at the smaller pieces And electrospray ionization does that. It enables you to start off with a liquid and make ions of virtually everything that’s contained in it and make complete ions of the complete molecules. We call this universal ionization technique and that means it makes ions of everything.
There’s another ion source that has proven to be very useful for biochemistry called MALDI – matrix-assisted laser desorption ionization. Basically, you put your sample on a surface and you hit it with a laser and the laser will form ions. This is done usually under vacuum and there’s a matrix present that assists in the ability to make ions. That was a breakthrough — finding certain matrices that would help this technique to produce more ions. Biochemists can put a spot on a plate, put it in the instrument, mix it with the right matrix, and when you hit it with the laser you can produce a nice set of ions from your samples.
So I would put these two techniques as very important. Both of them provide a very useful way of taking liquid samples, making ions of large molecules, and getting them into the analyzer. Before those techniques were developed it just wasn’t possible to do large molecules.
We were working on these techniques in different areas. I don’t think that anyone had the vision or the dream that it could all be put together. But, as the technologies were starting to evolve, in the mid-80s, or early 90s, people started thinking this would actually be possible. And it evolved into what we think of as proteomics today.
GW: Are there competitors to the ion trap?
STAFFORD: Yes, there are competing technologies. A current product that actually was introduced before the ion trap by Thermo Finnigan is something called a triple quadrupole. It does tandem analysis, but it does it in space. What I mean is [that] you have hardware devices that are coupled together and each one of them does a stage of analysis. You can do tandem analysis by coupling together different analyzers and by using one analyzer to do a first stage of analysis and a second device [to] do a second stage. They can be different devices, they just have to be coupled together. This is called tandem analysis in space. What the ion trap does is tandem in time. That way you have one analyzer and by trapping the ions you can do successive analysis by simply a sequence of timed events.
GW: What makes the ion trap better?
STAFFORD: I think it’s a little more cost-effective, and it can also be faster. But the triple quadrupole does address some markets quite strongly and it does a good job for lots of areas.
GW: What kinds of things are being worked on in the future? What’s the next challenge?
STAFFORD: We’ve talked about some of the hardware advances that occurred in the 1980s. But basically [they only] allow mass spectrometrists to do these techniques. That is, very experienced people take these hardware devices and put them together and achieve results. In many cases you get an enormous amount of data, very complex data, that requires specialists to interpret and get the information out. The data would be a set of spectra, but getting the information out would require complex analysis by specialists.
The challenge for the future is to put these various techniques together and package it to provide a complete solution for the biochemists. [Such a system would allow a biochemist to] take his sample that’s undergone some sort of chemical work-up and have it automatically analyzed from start to finish. [It would be] introduced into the instrument, and the computer [would do] some sort of real-time control, and once the data comes out it’s analyzed to provide an answer.
There have been lots of companies, including Thermo Finnigan, that have made great strides towards this, but I think that’s going to be the future. Ultimately people want a computer to provide answers, and not for people to have to spend time analyzing the data.