“… we try to build molecules to solve problems”
Transcript of the telephone interview with Roger Y. Tsien following the announcement of the 2008 Nobel Prize in Chemistry, 8 October 2008. The interviewer is Adam Smith, Editor-in-Chief of Nobelprize.org.
[Roger Tsien] Hello.
[Adam Smith] Hello, this is Adam Smith from Stockholm. Is that Roger Tsien?
[AS] Thank you very much for making time for this. How have the last 24 hours been?
[RT] A little bit surreal, but all I can do is put one foot in front of the other. Just try to keep going.
[AS] Yes, it will be the weekend soon, although I don’t know if things will abate.
[AS] I’ve read that, like many Laureates, you started your science career as a child doing chemistry experiments at home. Did you always know you were going to be a scientist?
[RT] No, I didn’t know. Obviously, it was one of the well-known possible options. I think I did go through the usual small-child phase of imagining being a fireman and so on, but yeah, I think being a scientist was more of a possibility for me than probably for most kids.
[AS] When did you actually realise you were a scientist?
[RT] I guess I realised I was most interested in being a scientist in the … Oh I guess it would have been junior high school, say, probably the age of 9 or 10.
[AS] That’s pretty young, that’s pretty young.
[RT] Well I, yeah, I guess I think … I’m trying to remember when I started playing around with chemistry experiments at home. It was probably around then. Much to the horror of … Well, a mixture of fear in my parents and a little bit of encouragement.
[AS] Right. Yes. The Prize has been awarded for your work on GFP of course, and for your revelation of how GFP fluoresced, and then using that knowledge to create mutants which extended the palette of colours of tags available. What does this extended range of tags allow you to do?
[RT] Well, several things. It means you can, basically you can monitor many signals at once inside the cell and you can monitor interactions of proteins by the phenomenon of fluorescence resonance energy transfer. If that’s what you’re referring to – the use of the different colours.
[RT] Of course, in the process we had to improve all of the colours, and the original wild-type GFP was very difficult to use, for many reasons. It was made by the jellyfish for its own reasons, some of which, as I said, have been very obscure. One remarkable property of the original jellyfish protein is that it actually isn’t just green. It excites, well it emits green, but it excites mostly in the UV, and only a small amount in the blue.
[RT] And that’s a strange paradoxical phenomenon because we discovered that almost any mutation of one amino acid right next to the chromophore will shift it to being all of one or all of the other – either all UV or all blue. And most people for biological purposes would rather have all blue. Occasionally it’s useful to have nothing but UV, and there’s I think only one amino acid that will make it either UV or blue. And when it is UV it is unable to do the energy transfer from aequorin. And therefore the jellyfish would presumably glow blue. And one of the great mysteries that I’ve never figured out is why the jellyfish chose the only amino acid that would compromise, and be schizophrenic, and be partly UV and partly blue in its absorbance. And it could have so easily shifted it to all of one or all of the other. And instead it’s preserved this split character. We don’t know why. But that is a nuisance for anyone else using it pretty much. Almost anyone else.
[AS] Apparently it doesn’t bother the jellyfish in evolutionary terms – it gets on fine. So …
[RT] Well, we have even wondered if there are times the jellyfish wants to turn on GFP’s ability to switch blue into green, and there are times it wants to turn it off. But I’m not enough of an ecologist to … Since we don’t even know why it wants to glow, and if it does want to glow why does it sometimes want to glow green, then I cannot answer the question. I’m certainly not going to answer the question of why doesn’t it always want to glow green?
[AS] But from your point-of-view, modulating these proteins, it’s beautiful that the selectivity resides in just single amino acids.
[RT] Well, it’s more complicated than that. At least for the wild-type, there’s one amino acid that controls really primary colour of the emission, and the next one to it controls, if you are going to be green, what is your absorbance spectrum. I think that’s a fair summary.
[RT] And then many other surrounding amino acids have influences as well, which we’ve gradually pieced together.
[AS] And learnt to tweak, yes.
[RT] But simply changing the colour of the fluorescent protein’s a sort of very visible thing to do, but I don’t know that it’s that terribly significant.
[AS] Are there are applications of the use of the probes you’ve produced that you’re particularly fond of?
[RT] I have to say that of course the fluorescent proteins have actually only been less than, it’s a minority part of my career, which was mostly spent building probes based on other things. But the fluorescent GFP work is obviously now the most famous. So I started out my, the successful part of my scientific career started out building calcium dyes …
[RT] … By synthetic organic chemistry. But, and more recently we’ve begun to get out of fluorescent proteins more and more and back towards synthesis, because we cannot treat people with gene therapy. And so we need non-genetic means of imaging processes in the human body. But, as far as the GFP, we soon switched on to the homologous proteins from corals, and those were discovered by the Russians not by us.
[AS] But you modified them to make them work.
[RT] Again, they needed extensive modification to make them user-friendly for the biological community. Most notably killing the intense tendency to form tetramers. It was actually great good luck that the green fluorescent protein from jellyfish was not nearly as obligately multimeric as the corals were. Because if they had been, a lot of the early applications of GFP would have been choked away, and the whole revolution would take a lot more time to get started. But because GFP at least didn’t have a tetramerization or severe dimerization problem – it only has a very mild one – that sort of got people hooked on what a wonderful tool this whole system was. So we were more able to cope with the tetramerization when it came up a few years later.
[AS] Does it ever have a detrimental effect on the things you’re studying? I mean, obviously if it’s starting to form …
[RT] Yeah. There … It is … It can do. One of the maybe more-forgotten facts about GFP, for example, is that in its maturation process it inherently generates a molecule of hydrogen peroxide. Hydrogen peroxide can be bad news inside cells. There are other times when if you fuse GFP to a particular protein, that particular protein just doesn’t like it. It says “No, I will collapse into a heap if you stick anything onto me, or at least onto my N-terminus or onto my C-terminus”. So you hope it’s not both of them, you know. But there are times you want to follow very small proteins or even peptides, and then to put GFP on them is putting a dog on the end of a tail. The probe is much, much bigger than what you want to study. We have strategies to overcome that too, which are mixed organic synthesis and genetics, but they’re lesser well-known. Most of the time, surprisingly, GFP is not – surprisingly often it’s not injurious, seemingly.
[AS] Right, right.
[RT] And I might actually point out that the first fusions to GFP ever were made by Marty Chalfie’s wife, Tulle Hazelrigg.
[AS] Right, right. Yes, yes. Yes, and indeed it was his marriage to her that accounted for the missing two years in his development of the technique.
[RT] Well, she made up for it after, once the GFP got going.
[AS] Sorry, we were on the question of favourite applications. I don’t know if you …
[RT] Sorry if I wandered off to something …
[AS] No, no, you wandered off very interestingly. It was fascinating. But did you want to mention anything in particular that you like that has been done with all these probes?
[RT] It’s hard to single anything out. There’s been thousands of, I think tens of thousands of papers, I have to say most of them using GFP or the red varieties or the yellow varieties in a fairly routine way. There are obviously some very cute things that have been done with it, a few I think we might have actually contributed to. And, one of the showiest applications is the one say The New York Times picked for its lead page today, which is from Jeff Lichtman’s group. This trick of combinatorially painting neurons a whole kaleidoscope of colours.
[AS] This is the thing they call the brainbow, is that right?
[RT] Brainbow, yes. And, you know, is it fair to single that one out? Well, it’s easy for me to remember it because there was this beautiful picture this morning, but there are actually lots and lots of cute things GFPs and RFPs have done. The one thing they have never done is get us to really long wavelengths, like excitation beyond 600 nanometres. That is truly red, deep red excitation and infrared emission. And that would be very important for going into whole animals that have a lot of blood in them, like you and me, and more particularly mice. Haemoglobin absorbs quite severely below 600 nanometres, so you want to be above 600 nanometres to avoid the blood absorbance as much as possible.
[AS] So …
[RT] And finally we think we’re getting there.
[AS] So that gives you whole-body imaging, right?
[RT] Well, it will help. I mean that alone isn’t enough. We’re still pretty, mammals are still pretty opaque, but this at least gets rid of one of the big barriers.
[AS] And the dream is to have multiple sensors of body processes there at the same time?
[RT] Biochemical processes. Not always necessarily at the same time, but at least to be able to look at protein kinase activity, redox state, gene transcription, protein degradation, localization of DNAs and RNAs by slightly more indirect means, which if you use GFP and RFP to label a protein that in turn binds specifically to your favourite nucleic acid, and the list goes on and on.
[AS] I wanted just to ask about what we call all this, because you work in a Department of Pharmacology, and you’ve been awarded the Nobel Prize for Chemistry, for work that is basically involved in cell biology. Does it matter how it’s all labelled?
[RT] Yes and no. It shouldn’t, but we have to respect the fact that we work in universities with department names, and attempts to make an amoeboid department that covers all of science probably don’t tend to work very well. Some people call this area chemical biology. Sometimes it’s been hard to wonder so what’s the difference between chemical biology and biochemistry. And sometimes I like to call it molecular engineering – at least our approach to it, because we try to build molecules to solve problems. And it’s the part of chemistry that is maybe slightly more applied, at least to some people’s way of thinking. Do I really worry about it? I try not to, but I recognise that the granting agencies, university departments, students … That’s when it’s dangerous, is when students let themselves get pigeon-holed, or let their thought processes get pigeon-holed, and say “Oh, I could never do that, that’s chemistry. I don’t know any chemistry”. And it’s surprising how often biologists have that attitude. And actually chemists sometimes have that attitude about biology too. This instinctual fear that, “Oh, that’s a subject I can’t do, and nobody should expect me to know how to do, and so I will just not pay any attention to questions that lead me in that direction.”
[AS] So as a final thought, what advice would you give to students thinking of moving into the field of building sensors?
[RT] Well, I might echo the advice that my then Department Chairman gave me when I was an Assistant Professor, when I was worried about certain, choosing between some safe projects and some risky ones, and he said, “Trust your heart and your gut and go for the one that you think is really interesting and don’t worry too much about trying to think to game the system, and think about what’s going to be the most appealing to outside people. Pick the one that you think is the most interesting”. And maybe to that I would myself add in my case since I love pretty colours, that helped me often decide that I would do things that involved pretty colours, so that at least without having to worry about whether the work would be successful in 5 or 10 or 20 years, I could get some direct aesthetic pleasure out of the experiments as I went along. I have to say I myself do not find pipetting colourless droplets of liquid from one plastic tube to another awfully inspiring, and that’s what much of molecular biology often seems to be at the bench. And so it helps to have things where you can see, for me, when you can see cells do things in real time, and if they can do so in pretty colours, so much the better. And that’s part of why we wound up trying to emphasize that.
[AS] It’s nice. Keeps the left brain and the right brain happy at the same time, yes.
[RT] One tries, yeah.
[AS] Splendid. Well, that’s a nice note to stop on. Thank you very much indeed for speaking to us.
[AS] And congratulations.
[RT] Sure, bye.
Did you find any typos in this text? We would appreciate your assistance in identifying any errors and to let us know. Thank you for taking the time to report the errors by sending us an e-mail.