This is the fourth in a series of Forum interviews with PRF’s eight new science advisors for 2014-2015.
David Bennett, MB, PhD, is professor of neurology at the University of Oxford and a consultant neurologist at Oxford University Hospitals in the UK. He is also a Wellcome Trust senior clinical scientist fellow. He has a longstanding interest in the role of trophic factors in mediating plasticity of the sensory nervous system. His current focus is to understand the pathogenesis of neuropathic pain and to develop novel therapeutics. Bennett takes a translational approach ranging from cell-based models, to human pain models, to patients with chronic pain. He administers a neuropathic pain clinic based at Oxford University Hospitals and receives national referrals for painful channelopathies. He is a member of the London Pain Consortium and is vice director of Europain, a collaboration of scientists working in academia and industry that is funded by the European Union’s Innovative Medicines Initiative. Bennett spoke with Neil Andrews, PRF executive editor, to discuss his current research, the importance of translational studies, and the role of systems biology in pain research. Below is an edited transcript of their conversation.
What was your path to pain research?
I always enjoyed lectures on sensory biology, and one summer, as an undergraduate, I did a project with Steve McMahon, who at that time was a senior lecturer at Guy’s and St. Thomas’ Medical School in London. That summer, Steve was supposed to go to San Francisco to work with a biotech company, Genentech, but he couldn’t go, so I took his place. I had a great time there working on neurotrophic factors and sensory neurons, and it really piqued my interest. Shortly after that, Guy’s and St. Thomas’ Medical School decided to start an MB/PhD program [MB is the equivalent of an MD degree]. They approached me and asked if I wanted to enter the program, and said they would fund me, so of course I said yes. Then I started a PhD with Steve. So there was an element of interest in the general topic of sensory biology, and then a certain level of serendipity—I happened to run into the right person at the right time and had the chance to go to the US in Steve’s place. That’s how it all started.
Who are the mentors who helped shape you as a researcher?
Steve certainly has been a great mentor to me over many years, and he is still a very good friend and colleague. I also worked with Pat Wall when I was an undergrad, and I got to know him quite well when I was doing my PhD. I always found him quite an inspirational figure who was very imaginative in the way he did his science. I’m also dual trained as a clinician, and I was very fortunate to do my neurology training in London, where I worked with some great clinical neurologists who were important mentors as well.
What guiding principles direct your research?
Anything that I work on has to be something that I find biologically interesting. Of course, many of those topics are of clinical relevance, and I try to work on the interface between the preclinical and the clinical, but clinical relevance is not an exclusive factor. As long as something is biologically interesting, even if there may not be a simple translation to the clinical sphere, I am still very happy to work on it. So to some extent, like many scientists, I follow my nose rather than set ground rules as to what we’re going to work on. I’m fairly flexible in that respect—it’s something I learned from my mentors like Steve and Pat, who also liked to follow interesting biology.
How have your research interests evolved over time?
Interestingly, many of the same themes remain since I started out. For instance, from the very outset I was interested in plasticity within the sensory nervous system, and the role of growth factors and how they may shape sensory neuronal function. I’m still very interested in that topic with regard to persistent pain states and how they may be modulated on a molecular level.
One area that has evolved over the years is in trying to understand more about intercellular signaling as well as pain. I also have a more general interest in how the nervous system repairs itself and how the interaction between neurons and glia may be important in the repair process.
Because I am a practicing clinician, I have also increasingly embraced translational research, and I have come to apply some of what I know from the preclinical sphere to the clinical realm. Overall, I work on a very broad base, and that is something that has evolved over the years, partly because I’ve been referred interesting patients, which naturally piques your interest, and then I try to back-translate some of that clinical research.
What are some of the specific projects your group is working on now?
At the clinical level, we are trying to understand the sensory phenotype of patients suffering from neuropathic pain. The question is, If you take a group of people and expose them to the same insult, why do only a proportion of those go on to experience persistent pain? Some of this work involves phenotyping quite large cohorts, which we are doing with diabetic neuropathy patients and those with carpal tunnel syndrome. We study these patients in a very multifactorial manner. We look at their symptoms and do quantitative sensory testing. We use neurophysiology and sensory testing to determine the somatosensory phenotype in combination with genetics, and more recently with functional imaging, to tease out what is special about that group that goes on to have persistent neuropathic pain. We also ask whether we can cross-correlate features such as histology with a sensory phenotype and genetics. We have some fairly large cohorts to do all of this work, and it involves collaboration with many people. For instance, I’m lucky to be at the University of Oxford where Irene Tracey runs an extremely effective functional imaging group, so we are linking up with her and applying some of her newer imaging protocols to our very well-phenotyped patients.
We are also looking at much rarer pain conditions such as small-fiber neuropathy and erythromelalgia. Here, too, we are looking at the genetic basis of these conditions and trying to relate that to the exact sensory phenotype that these individuals have, and attempting to study the pathophysiology.
We are also working quite hard to see if we can obtain human cellular models for some of these pathophysiological pain states. We are part of a big consortium called StemBANCC that is generating stem cell lines from clinically phenotyped patients. StemBANCC itself is a very large project, but I’m working within a subsection looking at sensory disorders. We are now improving protocols to differentiate human induced pluripotent stem [iPS] cells, which have been generated from a skin biopsy, into sensory neurons. Then we can start trying to understand pathophysiology in a dish from well-phenotyped patients—to really understand the basis of their disease—and then, possibly, somewhat in the future, we can use the cells to screen for novel analgesic agents. So quite a large part of my lab now is trying to improve these cellular models of sensory neurons, and of sensory neurons interacting with other cell types.
Finally, we also have a program that is more preclinical, which often involves generating or using transgenic mice. For instance, as part of another collaborative consortium called Europain, we have generated a transgenic line that has a knock-in mutation of SCN9A [which encodes the sodium channel Nav1.7]. We know that this causes a pain phenotype in humans, and we are asking if we can use that knowledge to obtain better outcome measures of pain behavior in the rodent. This is another example of taking something from the clinical realm and then seeing how that may facilitate our preclinical studies.
How do you find the state of translational research in the pain field in general?
The pain field is becoming more joined up in that respect, and it’s important to keep that link. It certainly keeps my research honest to see that we still have a long way to go in improving the lives of patients with chronic pain. Treatment options are widening; however, these are often reformulations of existing drug classes, and there is still a massive need for novel analgesics. When you run clinics looking after patients with long-term neuropathic pain, the degree of suffering that these patients still experience is striking.
At the recent IASP meeting, there was a mixture of clinicians and preclinical scientists, which is great because it’s very important to keep a dialogue between and connect the two—there is really an effort to do so. One example of this is research that examines sensory profiles in patients and then asks if a certain pattern of sensory dysfunction can be related to particular pathophysiological mechanisms. I don’t think there is a simple translation between the clinical and the preclinical, but people are trying to make the links, and it’s tremendously important. Clinicians have a lot to learn from preclinical scientists, and vice versa.
It’s not trivial combining a career in clinical work and preclinical research wherever you are in the world, and in the UK, the National Health Service is undergoing a lot of pressure at the moment, which actually makes research, at times, even more difficult. But it’s very important to try, and to support people who are doing both types of research, because in the long run it’s how we are going to make progress.
You also have an interest in systems biology. Why will work in that area be important for pain research?
As in many areas of science, as we are generating very large data sets, we really need to have effective tools to mine those data sets in order to extract the maximum impact from them that we can. This is driven partly by technology. For instance, in some areas, whole exome genotyping is now almost standard practice where I am in Oxford. We can look transcriptionally across the whole genome, we can do proteomics, and we can obtain massive data sets from functional imaging. Joining up all of these data and getting the most from them is an important problem, and people in many areas of science are struggling with it.
Toward this effort, one of the good things that happened in the London Pain Consortium, which I am a member of, was that we realized that we needed to work with bioinformaticians more often. I developed a very fruitful collaboration with Christine Orengo, who runs a bioinformatics research group at University College London. One example of this is the Pain Networks database [see PRF related news story], which is a systems biology tool I generated with Christine. In addition, I still have a student with Christine who is working on experiments investigating long non-coding RNAs, which we are learning more about now that RNA sequencing technology has come along.
I’m a biologist, not a computational scientist. But having those links and really making the effort to try and speak each other’s language is a large part of this, and it takes a bit of give and take on both sides—of me understanding more about bioinformatics and computing, and the bioinformaticians trying to apply their tools in the context of biology, and working together as we do so.
You mentioned Europain earlier—what has your experience been like with that collaboration?
This, too, has been a really enjoyable and very productive collaboration. Europain came out of an idea from the EU [European Union] to try to improve public-private partnerships, and so it was jointly funded by the EU and the pharmaceutical industry. The idea was that we would work together to improve early-stage development of analgesics based on the understanding of pain. The rationale for these collaborations was that we could do as a group what we probably wouldn’t be doing individually. I think Europain has been a success, and we are certainly seeing some of the fruits of that. In fact, the Pain Networks database was partly developed with Christine within Europain. And we have been doing some quite large, newer transcriptional profiling experiments as part of that collaboration.
Being a part of Europain certainly improved my links with industry and offered the opportunity to become more familiar with some of the problems in drug development, from their perspective, and how we can remove some of the barriers to progress and smooth the path toward new analgesics. Europain has moved forward on a very broad scale, from the level of molecules and cells, to the level of generating better animal models of pain, and right up to better phenotyping patients and using this to better target treatment. So the idea of public-private partnerships can be and has been very productive.
Looking beyond your own work, what research appears particularly promising to you?
There is a lot of research that is catching my eye. One of the nice things about being in the pain field is that there is an embrace of advances coming from other areas of neuroscience. For instance, we are seeing work using optogenetics to selectively activate or deactivate particular neuronal populations. People in the pain field are using these new tools to unravel neuronal circuitry, a concept that has been so influential in neuroscience in general, so that we can dissect out circuits that may have particular roles in pathological pain states. And we are seeing this on quite a broad scale, with use of transgenic animals, with new technology like functional magnetic resonance imaging, which is allowing a view of somatotopy that we wouldn’t have had five or 10 years ago, along with huge advances in human genetics—for instance, whole genome sequencing, which we were just dreaming of only about five years ago, is now beginning to be rolled out in patients within the National Health Service in the UK. Again, this creates problems in terms of how to manage all the data, and that’s why I think bioinformatics is important.
Finally, on a much more pragmatic, clinical level, I like the general theme that we need to think very carefully about how we phenotype patients and try to categorize different groups of patients. Based on this work, we may find that particular groups of patients might be especially responsive to certain analgesic drugs.
What are some of the challenges you see facing pain research?
We still need to understand how we can better model persistent pain states. There have been improvements in this area. For example, work from Frank Porreca using conditioned place preference has given us some really interesting insight into ongoing pain states.
What I would guard against is what I sometimes hear from people in industry, who say that we should completely forget about animal models because they are never predictive and are not helpful. I think that’s the wrong course of action—it would be throwing the baby out with the bath water. You have to be cautious with certain animal models, but they can be very helpful in understanding basic biology. You just need to interpret findings from those models in context. So I would warn against this trend of saying that no animal model is predictive and that we should just move into cellular models. Certainly, human cellular models can be very exciting, but they are also going to have their own limitations. For example, a sensory neuron derived from a human iPS cell is still a surrogate model of a human sensory neuron. Nothing is going to be perfect.
Is there any dogma about our understanding of pain that you think will be overturned?
Actually, I think the pain field is probably a bit less dogmatic than it was. For example, there used to be a lot of controversy between the labeled line versus pattern theory of pain, but we are coming to realize that both arguments are valid—peripheral sensory neurons are highly specialized, but there is a large amount of integration within the spinal cord and higher central nervous system centers. We are coming to more of a consensus that, like so many controversies in science, there is a middle ground that more accurately reflects the reality.
What is the funding environment like for biomedical research in the UK?
It’s challenging, just as it is in the US. Biomedical research in the UK has, to an extent, been protected from what has been a severe recession, but it’s certainly more challenging than it was 10 years ago if you consider funding rates within our Medical Research Council or the number of pharmaceutical firms doing biomedical research; it is true that a number of companies have left the pain field.
Having said that, one theme of biological research in the UK is that there is, more and more, a move toward research that has direct clinical relevance, and it is undoubtedly true that pain research has huge clinical relevance. We still have a situation where about 20 percent of the population lives in persistent pain, and almost 50 percent of those are not adequately controlled on an analgesic. This is a big clinical problem that’s not going away, so despite the challenging funding environment, pain research is a good place to be because pain is such an important clinical problem.
Do you have any suggestions for young investigators just starting out in pain research?
You need a bit of luck and a lot of enthusiasm, but with some application there’s no reason why people can’t make a success out of pain research. Sometimes you can see negativity building and people thinking that they are wasting their time because of the difficult funding environment, and because they think there aren’t going to be any tenured jobs, but that’s not the case. There are jobs that come up, and I’ve had very good postdocs who have gone on to get excellent tenured positions. So my advice is, don’t be too overwhelmed by what is a challenging environment. I think very good people still rise to the top, and pain research is and will continue to be a really exciting area, partly because it is so broad, ranging from studying molecules and cells right up to studying patient populations, and because it embraces so many different techniques. Young investigators should stick with it and enjoy it. With some enthusiasm, great things can be achieved.
Thanks for taking the time to speak with PRF.
PRF Related Content:
See "Related Content" box in the right column of this page.
Schmid AB, Bland JDP, Bhat MA, Bennett DLH.
Brain. 2014 Dec; 137(Pt 12):3186-99.
Bennett DLH, Woods GC.
Lancet Neurol. 2014 Jun; 13(6):587-99.
Perkins JR, Lees J, Antunes-Martins A, Diboun I, McMahon SB, Bennett DLH, Orengo C.
Pain. 2013 Dec; 154(12):2586.e1-12.
Antunes-Martins A, Perkins JR, Lees J, Hildebrandt T, Orengo C, Bennett DLH
Wiley Interdiscip Rev Syst Biol Med. 2013 Jan-Feb; 5(1):11-35.
Segerdahl AR, Xie J, Paterson K, Ramirez JD, Tracey I, Bennett DLH
Pain. 2012 May; 153(5):1122-7.
Calvo M, Zhu N, Tsantoulas C, Ma Z, Grist J, Loeb JA, Bennett DL.
J Neurosci. 2010 Apr 14;30(15):5437-50.
Bennett DL, Michael GJ, Ramachandran N, Munson JB, Averill S, Yan Q, McMahon SB, Priestley JV.
J Neurosci. 1998 Apr 15;18(8):3059-72.
Other Forum Interviews with PRF's 2014-2015 Science Advisors: