Fredrik Elinder is a Professor of Molecular Neurobiology at Linköping University, Sweden. He is also the Vice Dean of the Faculty of Medicine and Health Sciences at Linköping University and the head of a research group studying how different small-molecule compounds affect voltage-gated ion channel activity.
Fredrik actively collaborates with computational biophysicists and believes that we must be better using high-throughput electrophysiology in academia.
Aside from that, Fredrik is a world-class ultramarathon runner, having done 35 ultramarathon competitions, including the famous Spartathlon – a 245.3km road footrace.
I spoke with Fredrik Elinder about his career, drug development in academia, automated electrophysiology, his collaboration with computational biophysicists, ion channels in Sweden, new Physiologica journal and… ultramarathons. I greatly enjoyed talking with Fredrik and happy to share this interview with you.
Have a good read.
If you don’t have time to read the whole interview, you can jump right to the section you’re interested in the most.
Fredrik on:
OK, Fredrik, let’s start by talking a little bit about your career background and your current position.
Sure. I have a medical background and I did my PhD in Neurophysiology at the Karolinska Institute in ion channel biophysics lab led by Peter Århem. Then I did a postdoc at the University of Cambridge with Richard Keynes, who did a great work with Alan Hodgkin in the early 1950s and is a co-discoverer of the gating currents. It was a very unusual postdoc as Richard was already retired at that time. We’ve worked on the analysis of gating currents and met many times at his place to discuss our results. That was fun. And then, I got an assistant professor position supported by the Swedish Research Council and started my own group at the Karolinska Institutet. And I think I was lucky that at the same time, Peter Larsson went back to Sweden from the US and set up his group at the same institute. So, we worked together and that was a very productive and truly inspiring period for both of us. We studied the HCN channel gating as well as slow inactivation of the Shaker channel and published some nice papers together. But then Peter moved back to the US, whereas I got the full professorship at Linköping University. I set up my own group at Linköping and I’ve been there since then. And I’m very happy with the research environment at Linköping as we have built up a very strong ion channel community there. So, a former PhD student of mine, Sara Liin, after her postdoc, got the ERC starting grant last year and started her own research group at Linköping University. She’s very successful and continuing her research on ion channels. And two years ago, Antonios Pantazis from UCLA joined Linköping University as a senior lecturer and a group leader. So, we do have a very strong ion channel community now at Linköping University.
Could you tell me about your current scientific projects?
Yeah. What we’re working on now is to discover and also develop new compounds that can regulate the activity of voltage-gated ion channels. The overall goal is to find any type of compounds that can regulate ion channels, but our current focus is on the Kv7.2 and Kv7.3 channels, or M-channel, and we are mostly interested in how to activate the voltage sensor domain to open the channel. You know that many people with epilepsy don’t have good treatment, so there is a need for new drugs. M-currents control neuronal excitability and have been implicated in the mechanisms of epilepsy, so that’s one of the reasons why we are focusing on it. Another reason is that from the pharmaceutical point of view, from industrial and commercial perspectives, it’s much more attractive to work with the mechanisms that we know are working. So, with regard to M-current, there has already been an antiepileptic drug on the market, Retigabine, acting via opening the M-channels. It was rather effective, but due to strong side effects, it was discontinued. So, we know that this is a working mechanism, but we need to get rid of the side effects.
Therefore, we are looking for new molecules that can be used to treat epilepsy via the M-channel opening. And, in my opinion, compounds from plants are very, very interesting in this regard. Actually, I became interested in naturally occurring compounds about 20 years ago, when we were first contacted by clinicians from Karolinska University Hospital. They tried to explore the ketogenic diet for epilepsy patients and found it to be quite efficient for some of them. So, they approached us and asked if we had any idea why this diet is working. We’ve tested a bunch of compounds from this diet and found out that polyunsaturated fatty acids (PUFAs) activated voltage-gated potassium channels, leading to reduced neuronal excitability. And we’ve also revealed that PUFAs exert their effect on potassium channels by acting on the voltage sensor domain and affecting the voltage-sensing mechanism. That was really interesting work, but at that time, I was rather skeptical about the future of PUFAs as pharmaceutical drugs. My understanding was that in order to develop a drug, we need more specific and potent compounds. We had some ideas that these compounds had to be fairly hydrophobic and they also had to be charged, and so, we started to explore available literature on this subject and stumbled upon resin acids, which are found naturally in resin from conifers. These resin acids possessed the properties we were looking for, I mean, they are lipid-soluble charged molecules, and they were also shown to be effective on some ion channels. So, we started to explore these resin acids and found that the dehydroabietic acid potently opened voltage-gated potassium channels by electrostatically activating the voltage-sensor domain. Since then, we’ve created more than 200 synthetic molecules based on the dehydroabietic acid and identified some potent compounds which we hope could be developed into drugs.
Well, as I talk about molecules and drugs, you might have the impression that we are the drug development lab, but we are not. Our main goal is to uncover the mechanisms of activation and so we are really doing basic science here. Having said that, I’d like to mention that we have an early drug development project in collaboration with the Drug Discovery and Development Platform at the SciLifeLab, the largest life science research infrastructure facility in Sweden.
That’s interesting. I’m curious to know more about your drug development project with SciLifeLab.
Well, this drug development project started in 2018 and it’s based on our findings that resin acids can regulate potassium channels. We’ve got patents protecting our findings, and together with the Drug Discovery and Development (DDD) Platform, we are trying to explore and understand if resin acids could be developed into drugs and progress towards a preclinical proof-of-concept. We also got some funding from the Novo Nordisk Foundation to support our drug development activities. And it’s very important for us because it’s not easy for a basic scientist to find the money for kind of drug development activities. I mean, if I get grants from the Swedish Research Council – this is for basic research. So, we were really happy when our project has been accepted for the DDD Platform. And now, we are working with toxicologists and chemists on a proof-of-concept. I must also say that the DDD platform helps us a lot with our strategic decisions regarding this drug development project, and this is not an obvious thing for basic scientists.
So, as I understand, you will collaborate with the DDD platform until you get a preclinical proof-of-concept. What would be your next steps? What’s your plan?
If we are successful, and if we are really making progress, at a certain point we’ll need to exit this platform, obviously. And then you have several choices. You could start your own company to develop your project further, or you could sell the project to another company. I don’t know which choice is the best for us. At the moment, I would probably love to create a dedicated company for this project, but I’m not sure I’m the most suitable person to do it.
You know, here in Sweden, we have a very unusual system – you as a researcher, you own your ideas, discoveries and inventions. Not the university, but you. Let’s call it an old Swedish tradition. And although it might seem very attractive to many scientists to own their inventions, it has been questioned if this is a good practice, as, in the end, this may prevent technological development. In most cases, as a researcher, you’re a good scientist but not a good businessman. And so when it comes to strategic business decisions, many scientists have a hard time and this could negatively influence the future of their project. Sometimes, in order to make your idea into reality, it’s better to get more “muscles” from the university. And it’s great that the state, the government, has realized that this is a problem and so nowadays, every university has its own innovation structure. So, for example, at Linköping University, we have LiU Innovation – the university’s innovation office that supports students, researchers and employees in their start-up journey. And so, if you want to build a company based on your inventions, they can help you in different ways. They can guide you during your patent applications, or they can advise on business strategy and licensing, or they can even support you financially. And later on, you can move to the business incubator to really put your company on its feet.
In many cases, company creation is the only possible way for scientists to develop their projects into a development stage.
FREDRIK ELINDER
Interestingly, in many cases, company creation is the only possible way for scientists to develop their projects into a development stage. As I told you before, it’s very difficult to do drug development in the lab. Scientists are not usually funded for this. And at the same time, in order to convince other companies to in-license your project, you need to show a great deal of work done. So, collaboration with drug development platforms and start-up creation could really be just a necessity for some projects to be developed.
It’s great to hear that your findings progress into a drug development stage and I wish you good luck with this. Could you tell me about the technologies you use in your research and drug development?
Our main working horse is the oocyte rig. This is what we are mostly using for a number of reasons. One is that we have a long experience with oocytes. It’s working very well and we do get very good recordings. It’s also easy to study mutations of ion channels and, in fact, this is what we are doing a lot. We mutate different sites of action and see how this modifies compound binding to ion channels. In order to speed up the process, we have even got an automated oocyte workstation OpusXpress (Axon Instruments), allowing for automated TEVC recording from 8 Xenopus oocytes at a time in parallel.
Maybe I’m just a bit conservative, but I really prefer oocytes over mammalian cell lines like HEK or CHO. Of course, we always have to defend our choice and answer reviewers’ question: “Why are you using oocytes when there are mammalian cell lines?” But in our hands, the results are very very similar. And we do get a very good signal-to-noise ratio, and sometimes we are recording gating currents, and it’s very easy to record gating currents in oocytes. You know, with the advent of new technologies, many people tend to implement them in their research, but very often, they go back to the old techniques because they just work quite well.
However, for some of our academic projects, we need to have access to state-of-the-art high-throughput technologies, and for this reason, we initiated a collaboration with Sophion to perform high-throughput electrophysiological experiments at their site. Sophion is located in Copenhagen, and by train from Linköping, it’s just four hours or so. It’s very convenient. So, you can go there in the morning, run your experiments and go back in the evening with a dose-response curve for many compounds. It’s amazing. And it’s great that Sophion is willing to collaborate with academic groups in order to understand our demands and to improve their technologies. So, it’s a kind of a win-win situation for both scientists and the company.
I believe that within academia, we must have better access to high-throughput electrophysiology technologies.
FREDRIK ELINDER
You know, I believe that within academia, we must have better access to high-throughput electrophysiology technologies. In other fields, we’re already using high-throughput technologies, but in the ion channel field, it’s still very rare. At the moment, we don’t have an automated patch clamp system at the SciLifeLab Drug Discovery and Development Platform, and so, I’m trying to convince people here in Sweden that this is something we should spend money on. At SciLifeLab, they understand that this is important and at the DDD platform they would love to have this. But, you know, it all comes down to money, and so you always have to prioritize and choose what’s more important for now.
And what about computational techniques? From your publication record, I see that you are interested in molecular dynamics simulations and you collaborate with computational biophysicists Erik Lindahl and Lucie Delemotte on some of your projects. What’s the secret of effective collaboration between experimental and computational biophysicists?
Well, I think the secret is in CLOSE collaboration. People should try to solve problems together, and this is what we’ve been doing with Erik Lindahl for a while now. Our collaboration is not that I just send him my data and he is sending back some structures to me. Absolutely not. Usually, when I think that computational techniques could be useful for my project, I visit Erik in his lab and we discuss my hypothesis in detail. “Could these two residues be close to each other? Is it possible that the binding site for a molecule is here? Will this structure be stable at certain conditions?” – these are some of the questions we discuss. Then, Erik makes some first simulations and approaches me with his predictions: “This structure is not stable, but that one is very stable instead. These residues seem to be very close and here is the possible drug-binding pocket of a channel.” And then we test these predictions in my lab and we come back to him, and so it’s a very much back and forth work. The same process is also true for the projects initiated by Erik. So, in any case, it’s very close collaboration.
And now, I’ve had it with Lucie Delemotte (see Lucie’s recent interview to the IonChannelLibrary) for two years, and we just submitted a paper describing multiple binding sites for resin-acid derivatives on the voltage-sensor domain of the Shaker channel. Our collaboration with Lucie is also very close. We are working together, continuously sharing and testing each other’s data. And I think this is the way it should be done.
Effective collaboration implies that people try to learn a little bit about each other’s techniques, try to understand how others work.
FREDRIK ELINDER
In my opinion, effective collaboration implies that people try to learn a little bit about each other’s techniques, try to understand how others work. So, for example, I have been doing structural work for a very long time. I mean, even before we had the X-ray structures, I did my own electrostatic calculations. And although reviewers heavily criticized this approach, in the end, it turned out that these calculations were extremely accurate. What I want to say is that I have some background and interest in structural work and that makes me a person kind of predisposed to collaborations with computational biophysicists, I suppose.
OK, as we started to talk about your collaborations in Sweden, maybe we can talk a little bit about ion channels in Sweden. Could you describe the centers of ion channel research in Sweden?
Sure, I’ve been in this community for many years and I wouldn’t say that we have a very large ion channel community in Sweden. Let me give you a short overview. At Linköping University, as I told you before, except for my own group, we have Sara Liin (studying voltage-gated potassium channels) and Antonios Pantazis (dealing with calcium channels), and they’re both very strong and focused ion channel researchers, ion channel biophysicists. In Stockholm, we have three big research centers: Stockholm University, Royal Institute of Technology and Karolinska Institutet. But with respect to ion channels, there are not so many people there. For example, Eric Lindahl (from Stockholm University) and Lucie Delemotte (from Royal Institute of Technology), they both work at SciLifeLab and they are computational biophysicists performing molecular dynamics simulations of ion channels and other membrane proteins. Also, I cannot fail to mention the group of my PhD supervisor Peter Århem at Karolinska Institutet, studying voltage-gated potassium channels. Then, at Uppsala University, the only person I know doing ion channel work is Bryndis Birnir, working on GABA receptors. At Lund University, Peter Zygmunt is working on TRP channels and also some other people use patch-clamp in their studies like, for example, Lena Eliasson and Erik Renström, studying insulin secretion.
Then at the University of Gothenburg, there is a very well-known scientist, Patrik Rorsman, who learned the patch-clamp technique from Bert Sakmann and has been working in Oxford for many years in close collaboration with Frances Ashcroft. He is mostly focusing on calcium-dependent mechanisms behind the insulin release. Then in Gothenburg, I also know a newly recruited Alesia Tietze. She is focused on the chemical synthesis of complex molecules (including membrane proteins) and has also been working on toxins and voltage-gated sodium channels. Finally, at Umea University, Staffan Johansson and Kristoffer Sahlholm, former students of my supervisor, are dealing with GABA receptors and GIRK channels, respectively.
So, these are the groups I can recall at the moment. But, I need to say that here I was mentioning people having ion channels as one of the main research topics in their labs. Of course, we have a very strong neuroscience community in Sweden, but I would say that most of these people are interested in neurophysiology, and not in ion channels per se. And also, from time to time, we can see some ion channel papers from the labs doing structural biology or biomarker research. I didn’t’ mention those people here as they do not have ion channels as their primary research focus.
Do you have some local ion channel meetings in Sweden?
Actually, we had. For more than 10 years, we have had an ion channel subgroup at the Scandinavian Physiological Society. It was called something like “Ion channels, transporters and pumps”. And I was in charge of this subgroup for a number of years. We had local meetings every second year, and these were very nice and fun meetings, I remember. So, on average, we had about 40-50 participants or so, and these were mainly people from Sweden, Denmark, Norway and Finland. We also had some international invitations, some really good people and very interesting talks.
But now, we don’t have this subgroup and the meetings are not organized anymore. Since I’ve become the Vice Dean, I had no possibility to continue managing all this and we didn’t find people willing to take on these duties. So, naturally, all this petered out with time.
However, it’s possible to start the subgroup again, and I believe that the Scandinavian Physiological Society would support it.
And as we’ve mentioned Scandinavian Physiological Society, I’d like to say that I have just been appointed editor-in-chief of a new open-access journal Physiologica. The journal is owned by the Scandinavian Physiological Society and is a sister journal to the well-known Acta Physiologica. The new Physiologica journal is aimed for physiology in a broad sense, not only ion channels, and the first issue is projected to be published somewhere in a year. We think there is a need for society-owned open-access journals in the field of physiology, as a counterbalance to all new commercial journals. This is, of course, a true challenge for me to get this up and running and I have to work hard to make people know about this new journal. So, I thought that your ionchannellibrary.com is a good place to inform the ion channel community about our new Physiologica journal.
Sure, that’s a great idea. I’m looking forward to the new Physiologica journal coming out. And so, my last question will be about the thing that really amazed me. I’m talking about your passion for running, and more specifically, running ultramarathons. I found the video on YouTube with you finishing 100 km race. I’m curious to know more about this.
Well, I’ve been running all my life. I started running ultramarathons in 2005, and I was very active in this field for many years. Now, I’m much less active due to some injuries, which are not related to running, by the way. So, an ultramarathon is any footrace that is longer than the traditional marathon (42.195 km), and many different kinds of ultramarathons exist. There could be fixed-length races (50 km, 75 km, 100 km, and so on), fixed-time races (6h, 12h, 24h, 48h), or staged races. For example, once I was running in Wales and it was a five-day staged race from the North to the South of Wales. We were running during the days and were sleeping at nights, so you didn’t have to run for five days non-stop. We ran about 60-70 km per day in the mountains and I counted that we’ve climbed more than 35 hills (they’re not like in Himalaya, but still, some of them were quite heavy). We had everything that we could expect from the weather: it was raining, and it was hailing, and it was foggy, and it was hot. That was an absolutely beautiful and interesting experience.
I also had been competing in the national team for Sweden at the world championships for 100 km and 24h. Overall, I’ve done 35 competitions above the marathon.
The YouTube video you are referring to was an official 100 km race in the south of Sweden, and I won that race with a result of 8 hours 10 minutes and 4 seconds.
When you are running 100 km, are you running non-stop?
Yes. If you’re running 100 km in 8 hours 10 minutes, this is roughly 4 minutes 50 seconds per km. So, there is no time to stop.
What about traditional marathons? What’s your record time?
My best time in the traditional marathon is only 2 hours 53 minutes. At that time, I was about 42 years old and I always thought that I could go down to 2 hours 40 minutes, but I’ve never got the chance to test it. (ed. The official men’s marathon world record is held by Kenyan long-distance runner Eliud Kipchoge and is 2:01:39 hours (Berlin Marathon 2018)).
And what was the longest distance you’ve ever run non-stop?
It’s 245.3 km. It was in 2008 in Greece – the Spartathlon. Extremely nice competition with really good runners. And it was probably the best running experience I’ve ever had.
And how long did it take for you to finish this race?
It took me roughly 33 hours 2 minutes, so it was quite heavy.
Non-stop, no-sleep?
Yes. Non-stop, no-sleep.
I’m thankful to Prof. Fredrik Elinder for taking the time to talk with me and sharing his interesting insights.
If you have questions to Prof. Elinder, you can contact him via LinkedIn.
Visit Fredrik Elinder lab website here.
Read more about Linköping University here.
Read more about SciLifeLab here.
Watch Fredrik Elinder talking about resin-acid derivatives and potassium channels at the Ion Channel Modulation Symposium 2019 here.
Watch Fredrik Elinder talking about his research and ultra-running here.
Watch Fredrik Elinder finishing 100km ultramarathon here.
Photos provided by Fredrik Elinder
