Univeristy of Cambridge
Name: Carol Robinson
Position: Professor of mass spectrometry, University of Cambridge, 2001 to present; senior research fellow, Churchill College, Cambridge, 2003 to present; Royal Society Research professorship, 2006 to current
Background: Titular professor, University of Oxford, 1999 to 2001; Royal Society University Research Fellow, Oxford University, 1995 to 2001
Mass spectrometry has played a crucial role in helping scientists decipher protein structure, but in one class of protein, membrane proteins, the technology has yielded little structural information.
In a paper published online in the June 12 online edition of Science Express, researchers from the UK say that membrane proteins complexes have been hampered by a number of issues, including “an inherent insolubility in buffers compatible with electrospray,” and the ready dissociation of subunit interactions.
The researchers have developed a method to protect interactions between cytoplasmic and transmembrane subunits and release intact membrane complexes by maintaining detergent micelles in solutions “well above the critical micelle concentration.”
This week, ProteoMonitor spoke with Carol Robinson, the corresponding author of the article. Below is an edited version of the conversation.
Was the key to this strategy maintaining detergent micelles in solution?
Yes. … The complexes are held together by interactions that are a bit fragile once you take away the solutions. It wasn’t really thought that it was possible to maintain these interactions because it was always assumed you needed the water there.
And of course in the mass spectrometer, you take away the water as you go from solution to [the] vacuum. We thought that they would fall apart, including me. But surprisingly we were able to find conditions whereby if we kept it protected until the last moment and then just exposed it, we could trap it into a sort of meta-stable state whereby we could then observe it.
[This was discovered by Nelson Barerra, a post-doctoral researcher in Chile] and interestingly, he hadn’t been working in mass spectrometry for very long, so he didn’t realize a lot of the difficulties and the history and just tried something a bit different, and it worked. It was quite exciting.
Is this technique complementary to imaging techniques such as X-ray crystallography or electron microscopy?
Yes, very complementary to that. Obviously, they have atomic detail. We’re just going to get stoichiometry, the numbers of the interacting subunits. … But we’re doing it via mass whereas they have atomic positions. They would argue that they’re much more detailed, but we’re much quicker.
If someone is using your strategy, will they need to go back and use those imaging technologies to get a better view of the structure of the proteins?
It depends on what the question is … if you wanted to know how many protein molecules [are] interacting, then this would be the way to do it. Or if there was a drug binding, then you could look at that [and] you could do it relatively quickly, so you could get an answer within a week, whereas if you wanted to crystallize it, this can be extremely difficult and very few membrane protein complexes have been successfully crystallized.
It’s getting better but there still not many.
To get a full picture of the membrane protein structure, would you need to do both?
In the Science Express paper, you say that there have been a few examples of intact membrane protein complexes that have been reported. How were those researchers able to do that, and why haven’t those techniques been applicable across all membrane protein complexes?
The ones that I talked about, they weren’t with both the membrane and the soluble parts intact, so there was a part that was in the membrane previously and there [were?] two examples where they had managed to trap that. But they had never seen the interactions with the other proteins, which always used to always fall apart.
There weren’t many membrane [protein complexes], two, and people have been working on it for years. … With this new method we’ve managed to get five in a relatively short space of time.
Is that one of your key findings? That you were able to look at the proteins with both the membrane and soluble parts intact?
The ones that interact with the membrane proteins, yes. Obviously, the ones that are embedded in the membrane form a complex but then they quite often interact with other proteins in the cytoplasm of the cell. The key thing was that we could see those as well as membrane ones, the interactions between them.
Does your technique address the low abundance of protein membranes?
Not really. That’s a tricky part. This one was expressed outside the cell, but we have now taken some other ones which have not been overexpressed and we’ve managed to get those to work as well.
At the moment, we haven’t done that … that’s something we’d like to do in the future.
Can this technique be used to address this?
It could be. I wouldn’t like to say, definitely, because we haven’t done it yet, but it’s possible.
Does this approach work better for prokaryotic or eukaryotic species?
It could work for [both]. It doesn’t matter.
In real-world applications for drug development, what does your strategy mean?
We’d like to be able target some ligands. Obviously, a lot of these membrane protein complexes are drug targets so it gives us a way of looking at how it stabilizes and interacts with the subunits within the membrane complex.
And what can researchers do now that the interactions between these complexes can be maintained intact in the mass spectrometer?
Now we can measure what’s binding to them, how many subunits make up this, whether subunits affect the stability of other subunits, whether we stabilize all the interactions, or we weaken some interactions, so we can actually get quite an interesting topological map of the interactions between subunits that we couldn’t get before.
And those are the fundamental questions about membrane proteins?
Yes, they are. Also it’s been a goal for a number of years to get these complexes to work. We’ve been working on it, off and on, for a number of years, and never had any success, so it’s exciting for us.
Are there certain types of membrane proteins where this method would be suitable?
We haven’t found them as yet, so I don’t know. … We’ve only done five, and they’ve all worked, but maybe there’ll be some that don’t. Maybe it’s too early days to say.
Does this work for other types of membrane complexes aside from membrane protein complexes?
We haven’t tried. We’re just looking at the membrane protein complexes at the moment.
How are you using this new ability that you’ve developed?
We’ve established lots of collaborations with people across the world and we’re looking at lots of different complexes and trying to answer questions that have been longstanding in the field.
Can you give me some details about these collaborations?
We’re looking at ATPases, which are involved in energy production with people in Australia, in the UK and the US and across Europe, particularly because it’s very difficult to see both parts of that assembly using traditional methods.