Before founding Impetus Grants, a non-profit that aims to speed up longevity research by handing out money to scientists, minus the bureaucracy, Martin Borch Jensen did his doctorate in the Bohr lab at the NIA and his postdoc at the Buck Institute, and he received an NIH Pathway to Independence award that would have enabled him to start his own lab, although he turned it down for his own ambitiously named start-up Gordian. He also wrote a book while serving in the Danish military.
We spoke with Martin about the bold ideas behind Impetus Grants and Gordian, his view of the longevity field, and why lifespan versus healthspan is a false dichotomy.
You are a European who moved to the US, which is not uncommon in the longevity field. How does Europe compare to the US as an environment for aging research?
I think there’s quite a lot happening in Europe. There are some differences in what the most popular topics are. For example, DNA damage seems to be more studied in Europe than in the US. Part of the reason I moved here was to go to the Buck Institute, which was a choice of professor rather than of institution. The US and the Bay Area specifically is certainly the top spot for translation of science and for start-up companies, a much better ecosystem. I think that’s the main difference.
Americans are also better at putting themselves out there than Europeans are on average. As a Dane, to come here and say that you did something cool feels almost shameful, embarrassing.
Also, maybe some labs and institutes in Europe are just more isolated. They’re not putting effort into being visible. A small anecdote on that point: a lab in the US usually has a standalone website, not just on the university page, and you can see the people there and you can often find contact information for them. When I need to do that for European labs for various reasons, half the time, I can only find some info on the university website. I can’t find people’s personal emails, and then I don’t write them.
On the other hand, Europe has way better biobanks and annotated electronic health records than the US. The US is a total mess for electronic health records. They all are siloed in individual hospitals, it’s disorganized and hard to access. So, for the essential work of making biomarkers of aging, I think Europe is the place to go, at least if you need lots of samples.
I understand that you wrote a book while serving in the Danish military. What is it about and is it going to be translated?
The book is about intermittent fasting – the concept that, in various model organisms, if you restrict their eating to certain times of day or week, like alternate days, they can live longer, and this has ties to various aging mechanisms. This was something that I had been doing in my PhD for two and a half years.
While I was still in my PhD, in 2012, a friend of a friend who was a journalist thought that this sounded really interesting. I fasted every other day, and the guy probably thought, “Oh, this person is doing something really weird, but he also has a PhD in aging, so, maybe, it’s more than just weird.”
There were some newspaper articles, and they led a PR rollercoaster, including appearances on radio and TV. At one point, another journalist who later became a coauthor of the book, talked to a publisher and said to me: “Hey, we should write a book about this thing”.
The timing just happened to overlap with my mandatory military service in the Danish air force, so I was writing on nights and weekends in Denmark. The book sold well, not quite a bestseller, but almost, but there are currently no plans to translate it, to my knowledge.
Anyway, it would have to be rewritten. In the last eight years, we’ve learned more about how intermittent fasting works, although we still, for the most part, have no idea what the appropriate regimen of fasting would be. Even if you think that the mouse data that we have should work in humans, how much and how often you’re supposed to fast is totally unknown. That was true then, and it’s still true, but if you’re using intermittent fasting to not eat too much, then it’ll probably be beneficial.
Do you still do this day-in, day-out fasting?
There’s some mental overhead of doing that; you have to schedule around it. At some point, I became so busy that I just stopped. It’s an interesting optimization problem when your job is to work on drugs for aging. If you impair your ability to do your job by any amount, you’re reducing the probability that you will live longer, even if the thing you’re doing might also let you live longer.
After your postdoc at Buck, you received a grant to study Alzheimer’s, but you gave it away to start your own company, Gordian. That’s unusual. What was the rationale behind that?
I thought, okay, I can go and start a lab and focus on this thing that the grant was about. On the other hand, what I want to do is using aging biology to make the maximum impact on human health; that’s my life mission. If I start my lab around this grant, is this the most important thing to do for addressing aging in humans? The answer was no. When will I get to work directly towards solutions that would impact humans? The answer was maybe in 10 years, after I get tenure and multiple grants. That felt too long.
It’s not that I had no doubts. I was definitely thinking about whether that made sense. I’ve been working towards a career in academia for this whole time. What are the odds that you start doing something else, and you just flub it, and then you’ve got nothing? I was encouraged by certain investors who, even before there was any notion of a company, wanted to fund me to do stuff. Then it’s less scary of a jump, right? That’s different from having money in the bank, but it was certainly helpful.
But still, no regrets, right?
For me, this was the right decision. First, because it does feel much more energizing to be working directly towards what I think has the best chance of addressing aging. Second, I personally like making sure things work, at least as much as I like going into the unknown and making new discoveries.
I have kind of this engineering mentality, plus I love working with people and having a diverse team who are focused on the same goal. That feels really good. For me, that was exactly the right time and the right choice.
Everything you do seems a bit unorthodox. As I understand, Impetus Grants was set up to fill the void between regular grants and corporations. Why is it important, and what do you expect to achieve?
With Impetus Grants, we had a couple of goals. It was obviously inspired by the COVID-19 Fast Grants, which were organized by Patrick Carlson and Tyler Cohen. The NIH put out an RFA for grants on COVID, and people would get the money in March 2020. They were like, well, that doesn’t make any sense, can we fund research in a faster way? Which they did. That’s also our idea: to do things faster with way less bureaucracy for everyone involved.
When I submitted the application for my grant, the science part was about 12 pages, and the total thing was a hundred pages of stuff: all kinds of justification, descriptions of animal procedures etc. Obviously, that took a while. It took me like a month to write that thing. I don’t know how long it took the reviewers to read it, but probably also a while.
Lots of scientists’ time is put into paperwork, in reviewing each other, etc. I’m not saying we must get rid of this all, but the more you make people write, the more you make people read, and time is spent on that. From the polls I’ve seen, most professors say that they spend somewhere between 25% and 75% of their time on writing grants and stuff.
Many people in the longevity field have this feeling of urgency, because, you know, time flies.
Right? That’s the idea behind the name Impetus. I’ve been in the aging field for 10 years now. We know a lot of things, though there are even more things that we don’t know, but let’s at least say that we want to move forward. You have to ask yourself, what is our specific goal? Is this thing the most impactful if we want to reach that goal? If it isn’t, you should be doing something else. We had identified some focus areas such as biomarkers, which are especially important because it doesn’t matter how much we study if we must wait an entire human lifetime to get answers on things.
We wanted something impactful, like this new thing, TIME-seq, where they basically took epigenetic clocks and made them much cheaper. If it’s a hundred times cheaper, everyone can use it in their studies, and you can imagine how this pulls the whole field forward. That’s the sort of thing we are trying to do with Impetus Grants.
Applications were closed November 1st. How did it go? Are you excited about the proposals you have received?
Very much so. We’ve received more than 600 applications. We have funded 43 so far, but we haven’t reviewed all the applications yet. The acceptance rate will be around 15%, which is, I think, higher than with the NIH.
There are tons of ideas out there. We looked for the good ideas that are getting ignored, and people proposed things that I hadn’t thought of, so, yes, I’m excited. I think there will be several things funded by us that will make people say, “Oh, wow, this really changed the way we see things.”
There will also certainly be grants that, whatever hypothesis they were testing – totally not the case. In that regard, we are organizing a special journal issue so that negative findings can be published.
How do you recruit sponsors for Impetus? Are those people close to the longevity field or know little about it?
The list of donors is already public. They are Juan Benet of Filecoin, Vitalik Buterin of Ethereum, Jed McCaleb, who’s also done multiple things in crypto, Karl Pfleger, a philanthropist, and Fred Ehrsam, who also comes from the crypto field.
As you can tell, we received a lot of money from people who are in crypto and who are interested in longevity. I guess that’s the type of person whom this message resonates with: let’s cut down on bureaucracy, let’s do things in a new way, let’s do usable research.
Gordian is an unusual company as well. You are building a high-throughput in vivo screening system, which, frankly, sounded like an oxymoron to me till I learned how this works. Now, I think it’s a fascinating concept. Please walk us through it.
Sure. We run multiple separate experiments in cells where we put some sort of biological perturbation in, which could be giving a drug or making a knockout or something like that, and then we ask what happens.
This is the same thing that all of biology does, but for aging and diseases of aging, there is no good in vitro system: since we don’t know how aging works, we can’t model it perfectly. We don’t even know everything we’re supposed to model.
Even for what we do know, there is no model that captures all of it except an actual aged animal. So, we wanted to run all of those experiments but in the context where aging and all the aspects of aging is present. Can we just take the entire lab and all the experiments that are being run in it and move them inside an aging organism?
The answer is yes. We do this by delivering those perturbations to individual cells rather than to all cells at once, and then we find a way to have readouts that also come from individual cells. To achieve that, we use gene therapy for the delivery side, where we can package perturbations increasing or decreasing the expression of a gene into viruses. Each virus will have exactly one perturbation, and each perturbation will have a unique DNA barcode in it. We then put them into the animal, at a very low dose, so that in a liver or a different organ, 99% of the cells will not receive a perturbation, and so, you’ll have these ‘islands’ of biological perturbation within an unperturbed, deceased system.
Then we leave it there, and we pull out those cells that were perturbed, and use single cell sequencing as what I call a pheno-target screening. We measure the phenotype of that cell, the expression of every gene, everything the cell is trying to do at the time. We can then do bioinformatics, to interpret that in the context of different cellular pathways like metabolism, mitochondrial function, inflammation and so forth.
We can use the DNA barcodes to measure which perturbation was in there. Because it’s gene therapy, we know exactly what it did. With a small molecule, you might not know what happened. You just know something happened and then you have to find the target, while we know already what the target is.
So, you’re basically doing many perturbations in one animal?
Yes, that’s right.
How much potential do you think pooled screening like this can have in the longevity field?
First, right now there is no practical way to target “aging” as your indication, because it is a poorly defined concept with no good metrics. So, like many other companies, at the current stage at Gordian, we are targeting specific diseases of aging, including NASH [nonalcoholic steatohepatitis, an aggressive form of fatty liver disease] and osteoarthritis. For those diseases, which are complex disorders where in vivo environment matters a lot, we will be tremendously more efficient than standard drug development at identifying drugs that really work. Hopefully, that will lead to more drugs on the market that will treat these individual diseases.
Then there’s the other thing that we are hoping to do. The way the aging field works right now, people are putting individual interventions into an animal. You do this one thing to an animal, and then you read out what happens to this whole animal. We are finding some things that can extend the lifespan of animals.
The majority of those things relate to metabolism and nutrient sensing, and I suspect that anti-aging will not work the same in all types of cells of the body. Different cells will need different things. There are plenty of published examples of this already, like IGF-1 signaling – do you want it high or low? You get opposite answers: in the brain, it’s different than in spinal cord hematopoietic stem cells. Or TGF-ß signaling: do you want more or less of it? It depends: in your cartilage, you want more, in your fibrotic lungs, you want less.
If this is the case, then the interventions that we keep finding, like mTOR and FOXO (I’m oversimplifying a bit here, but we do keep finding them), interventions that have such a beneficial effect averaged across all cells that we actually see lifespan extension, may actually improve the resilience of certain cells and organs, but have negative effects in others. Then, to achieve really big results in terms of lifespan, you’re going to need targeted perturbations in specific contexts, you need to optimize the system. Currently, there are not many ways of doing that.
The platform that we’ve developed at Gordian enables us to see different perturbations in specific contexts and cell types. In a sense, aside from being a successful company (hopefully), it is also, a sort of backup plan for the entire field.
That means, if we fail to find silver bullets, if mTOR or partial reprogramming or whatever is not a silver bullet that will rejuvenate the entire body efficiently, and we’ll need to do many small things in combination, then we’ll have a platform for that.
This is really exciting. And you want to make the technology available, right?
On one hand, Gordian is a company that has a unique technology that others don’t have. In order for us to exist as a company, if someone wants to use that technology, they have to give us a bunch of money.
On the other hand, we are not the only people in the world who are trying to do this. It’s a small group, but there are other academics who are doing these kinds of methods more for biological understanding. Once you show that it’s possible, it becomes crazy to not be looking for drugs in this way.
I think, 10 years from now, this is going to be a standard part of drug development workflows, and big companies will be doing that. In that sense, it probably should and will become a commonly accessible thing.
I do appreciate the name choice for Gordian, but could you explain it in your own words?
I had spent close to 10 years in academia trying to understand all of aging. There’s one approach to fixing something, which is to understand exactly how it works and then go in and do exactly what is needed, but we’re not going to fully understand the complexity of aging in the next 10 years.
So, what’s the other approach? Instead of trying to untie this incredibly complex knot of biology, what is the cheat code? How can you cut this Gordian knot? And of course, the answer we came up with is: find out what answer you want and in what context you want that answer in, and then find a way to ask a whole lot of questions. It’s about tackling the problem in a different way.
It’s inspiring, but let’s hope it doesn’t turn out to be wishful thinking. Would you say you’re out of the woods with that?
Yes, the platform works. We’re still pretty stealthy, but we have done some conference presentations et cetera, so it’s no secret that we have actually done this. We really are doing in vivo screening.
We’ll be closely watching you, of course, and I wish you a lot of luck. Now, you seem to have a holistic view of the situation in longevity research. Can you share it with us?
As I mentioned for Impetus Grants, I don’t pretend to know everything, I don’t have a master plan to do all the right things, but I do have thoughts and ideas. We’ve already talked about biomarkers. It’s just so obvious that we need biomarkers that actually work. That’s fortunately an area where many people are working, so that’s not what I’m focusing on.
I call myself a systems biologist, because I think what happens in aging is unlikely to be any single process. Take my PhD thesis that was in DNA damage repair. There is this DNA damage theory of aging – that it sorts of leads to all of aging. Already during my PhD, it was obvious to me that this was not the case. We were doing DNA damage repair, but those guys over there were doing metabolism, and they too could make mice live longer, and other people were doing some third thing.
You could argue that each of these things cause something, and they are linearly additive, but I don’t think it’s true, for two reasons. One is that if you look at biology in general, how many things are linear? It’s the minority, right? Much more often you have threshold effects, nonlinear synergies and so forth. So, biology is generally non-linear. And the only reason that we choose to sort of ignore this is because that makes it really hard to do anything.
Say, there are different hallmarks of aging, and if only we could address them all individually, then we’ll just add all those things up in a system. But people have tried. It’s not like nobody has tried to simultaneously target this pathway and that pathway. Sometimes it works, but more often it doesn’t. We don’t see that everything that works just adds up linearly.
Personally, I think that aging is better understood as a sort of an information system’s decline where you have different parts and different signals. It’s signal-to-noise ratio. You get this desynchronization of parts that are supposed to work closely together, and then you get a decline. Why does everyone have multiple aspects of aging going wrong? Why is it that if you fall and break your hip, or if you have one disease, you are now at much greater risk for different features of aging? It’s because the system is interconnected.
So, that’s my personal take on this thing we call aging that nobody has a real definition for. I see it as systems-level information decline. And if this hypothesis that I have and some other share is correct, then we need different tools. Biology has lived in the reductionist study era for a long time because that’s what’s worked for the most part.
I’ve done all that work as well, like genetic screens and epistasis experiments, but it’s like something you see in a textbook: gene, arrow, gene, arrow. And we know that’s not true. We know there are all those other arrows. But we still work in this linear world because we don’t have the tools to do it differently. This is what we are trying to change.
You strike me as a bold and imaginative person. What is the limit for the longevity field? Do you think we’ll ever be able to drastically increase human lifespan?
I don’t know if we’re going to wipe ourselves out in a nuclear Holocaust, but if not, I’m pretty sure that sometime between 50 to 5,000 years we will attain sufficient mastery of biology so that our bodies will be sustainable, existing in a homeostatic state.
If we will even have bodies.
Right, who knows? In the very long run, there is no doubt that we will have enough control over biology to make people not immortal, but ageless, with mortality risks not increasing with age. Will we get there in the next 20 years? Nobody knows, and it doesn’t even seem a useful discussion to me.
Do people want to have effective treatments for multiple diseases at once? Everyone would probably say yes. Do you want treatments that you take before you get the disease instead of treating the symptoms after you get the disease? Yes. Do you want that done in the most efficient way possible? Again, yes. Well, then we’re looking for geroprotectors or whatever you want to call them, right? There are all those false dichotomies, like lifespan vs healthspan. Look at the studies: are these things correlated? Yes. The vast majority of interventions that extend lifespan also extend healthspan.
I don’t think most of these questions have any practical implications. It’s mostly metaphysics. Let’s just go solve problems and make health better in the most efficient way. We know we’ll have to do this by targeting the fundamental mechanisms of aging. If we solve everything, we can ask people: okay, you’ve lived a hundred of extra years in good health, do you want some more of this, or do you want to die tomorrow? I think they’re going to want a few more years.