Learning the language of immunology: an interview with Professor Carmen Molina-París
23 July 2012
Professor Carmen Molina-París is coming to the end of a BBSRC Research Development Fellowship which she commenced in 2009. Here she talks about her experiences as a physicist amongst immunologists.
What inspired you to apply for a Research Development Fellowship?
The story behind this is that before I had the Fellowship, I coordinated a BBSRC-funded Immunology Imaging and Modelling Network, and so I got to know about what was happening in the UK in immunology. I talked to lots of immunologists and Professor Robin Callard at UCL, in particular, helped to encourage me and gave the confidence to apply for the fellowship.
When I found out about the Research Development Fellowship, I thought it would be really good for me as it would allow me to take some time out of teaching. The fellowship would allow me to visit different people and really spend some time to immerse myself in immunology.
It has been exceptional in opening up lines of research. Now we are collaborating with experimental immunologists. We're discussing, making models and they are helping me know that what we're doing is immunologically relevant.
Was the fellowship especially important because you were coming from a different background? Did you feel a need to develop your knowledge, to actually learn more immunology?
Yes, I had to learn to speak the language of immunology. Now when I give a talk to an audience of immunologists they go 'wow you know a lot'. When I started meeting with immunologists, things used to go back and forth for hours because we didn't share a common language. Now we are all on the same wavelength and they are looking at my plots and equations with no fear!
It has taken time to develop this shared language and that is what the Research Fellowship has allowed me to do.
Have you got some good results out of the research you've been able to dedicate yourself to?
It's quite difficult because when you're given a fellowship you are expected to start publishing and having an impact quite quickly but up until last year we had only produced the one paper. Now though we're really producing some interesting results and we have recently published three papers and have a number of publications in the pipeline.
How did you end up in immunology from a background in physics?
In Spain you have to choose to specialise in a science subject in your last year at school. I didn't know what to do – medicine, biology, physics – but at the last minute wanting to understand the universe won out – I suppose I've always been more theoretical than experimental. After my post doc at LANL, I got a research fellowship in Madrid to do physics but the research institute was quite interdisciplinary. I met some virologists who explained to me how maths was being used to understand the evolution of viruses and they asked if I could help out. We used to have weekly meetings, just like a journal club, where I would help out and a few papers came out of that. I ended up moving to England for personal reasons and a friend of mine pointed me at a research position on modelling T-cell activation. I knew nothing about modelling T-cell activation!
The position was with a group at The University of Warwick who had money from EPSRC and they wanted someone with my background. My friend told me that I had nothing to lose and that I should apply. So I did and I got the position! My plan was to stay for a year because I had a 5 year fellowship to go back to in Spain but I'm still here.
And you developed an interest in immunology?
The amazing complexity of immunology was really fascinating and I enjoyed it a lot but I felt that there was no way that I would be able to produce anything significant in the area.
Being supported by Robin Callard and Hugo van den Berg helped a lot. Hugo had made the opposite transition from biology to maths, so he gave me the confidence that other people have made these transitions and that I could too. The lucky thing for me was that the work at Warwick allowed me to secure a lectureship – everything was very timely. Once I had an academic position I could follow my research interests a bit more.
Do you think you look at immunological problems differently to your colleagues with a pure biology background?
Yes, I think so. Every time I meet immunologists I explain some of my maths and they go "that's a great question – I'd never thought about it that way!" The way as physicists that we look at the same facts is very different because of how we have been trained. We are looking through different eyes.
BBSRC is keen to encourage interdisciplinary research and the use of mathematics to understand biology. Do you think it's important that you combine people with different skills and backgrounds?
Yes, yes. I believe that maths and physics can help biology a lot, but first we need a common language. Mathematicians and physicists like me need to know their biology and how experiments are designed. That was an essential bit of my fellowship. I knew that, without going to a bench, my models wouldn't be grounded and wouldn't have the same quality and immunological relevance. It's the same the other way, slowly, as we talk about mathematics; the immunologists are starting to think differently about their experiments.
What do you think that a mathematical approach can bring to our understanding of the immune system?
Sometimes biologists are happy to concentrate on small questions – Is a molecule where we think or not. But mathematics and a quantitative approach can help you to test mechanisms in a very defined way which, I think, sometimes biologists don't see as a priority.
So are you saying that maths allows biologists to address slightly bigger questions .To take a step back from the minutiae of their experiments?
Yes. In fact our most recent paper is an example of exactly that. We started with the known outcomes but wanted to decipher the bigger picture of what was going on behind the experimental data.
Could you tell us more?
We used data about how T-cells bind and unbind to proteins, called ligands, which are on the surface of dendritic cells, for example, to understand how a T-cell response is triggered in our immune system. Basically we have compared two different hypotheses that make use of the experimental data. You can frame these hypotheses mathematically so we developed models and found that one hypothesis was a better fit.
What does this tell us about how our bodies detect pathogens?
We found that the time it takes for your immune system to respond is governed by the length of time that a certain number of T-cell receptors are bound to the ligands. This hypothesis also helps to confirm an immunological paradigm. It explains why, when you have an infection, you might only need a few proteins from the pathogen to cause an immune response. This was a consensus amongst the immunological community but it was hard to pin down in a quantitative way. Using our model we were able to show that as long as the T-cells were bound for a long time an immune response would still be triggered even with just a few ligands.
The model also shed light on an important concept called self-nonself discrimination. This is extremely important as it means that the immune system doesn't target your own healthy tissues causing an autoimmune disease. We've been able to show that our model naturally yields the self-nonself discrimination that we know the immune system has. So the model really ties everything together nicely.
How has the immunological community responded to this research?
Really well. We were at a meeting of immunologists in March and one of the world's leading immunologists from NIH in the US came to our poster – it was too much! I was the odd one out, the only mathematician or physicist there. I was explaining the research and he asked – 'please send me the paper as soon as it's out.' I was really proud!
You said you were the only mathematician there. Do you see that changing?
I think so. At that meeting our collaborator, Ed Palmer, gave a talk including slides from our paper and did a very good job of explaining the equations. Because it was coming from an immunologist there was a sense from the other researchers that it had been validated and it was worth making the effort to understand.
That is one big difference I've seen during the course of my fellowship. When I first came to the UK I approached lots of immunologists about my models but people are busy, applying for grants, giving lectures, working with students, and the maths takes time and effort to understand. Unless there is a clear benefit then why make the effort to understand it. But at that meeting I sensed that a number of immunologists were seriously considering the maths and whether they needed to be working with or even recruiting mathematicians and physicists to help with their research.
Do you still try to maintain an interest in physics?
It's difficult. At the moment my physics is pushed to the side. I have the occasional discussion every few months but it's more of a hobby.
Immunology has been my lucky charm. Ever since I was awarded the Maths and Systems Biology Network award from BBSRC I have felt that someone must believe in me. But sometimes I still think that I don't know what I'm doing – like I've jumped in at the deep end without knowing how to swim. I don't think I know that much! The big defining step of the fellowship has been able to publish in immunology journals or interface journals. It suggests that I do have something to add. I may not know a lot of immunology, but at least now I know enough immunology to make sense to immunologists.
Could you tell me about the other paper that you have recently published?
It uses modelling to investigate an important hypothesis in immunology about how our immune system maintains a certain number of T-cells. Our bodies don't have enough energy to maintain an unlimited number of T-cells and too many can be dangerous, so we have to balance the ability to mount an immune response without immune cells getting out of control. The hypothesis is that the T-cells need to be able to count themselves. A similar concept, called quorum sensing, is well known amongst bacteria and they have been modelled in great detail. The fellowship allowed me to visit the Pasteur institute to work with people who knew about CD4 T-cells. It took us a year to build a model to try to explain the possible mechanism which cells could use to"count'' one another. The model allows us to simulate an infection and then see what happens in our model immune system. We have found that a subset of cells in the immune system called regulatory T-cells are maintained at a minimum level, that allows mounting an immune response, yet not causing an autoimmune disease.
I've discussed this with a number of immunologists. I talked to Robin Callard who himself doesn't believe in regulatory T-cells. My student presented the paper to him and after it had been explained we convinced him that this might be happening and he went away to go and look at the paper in more detail.
I guess that's a great thing about mathematical models. As long as your assumptions are right then your answer is quite hard to argue with.
Exactly! One of the great things about my fellowship is that I've learned how to write my equations in a way that makes sense to immunologists. So we took Robin through the hypotheses one-by one and it is easy to identify where you might disagree. It's much harder to do this with experiments. There are so many variables, things are so much more fussy. But Robin is great he's really part of a new wave of immunologists because he has done an open university degree in mathematics. That's really brave, but it's been really great for me to have someone to argue with!