Dr. Gordon Lithgow of the Buck Institute for Research on Aging, who spoke to us recently about the Caenorhabditis Intervention Testing Program, talks today about how the field of aging research has evolved over his career and about how engagement with the wider biomedical community will need to change for geroscience research to impact society.
You worked on hormesis and stress response in the early days of aging research, what was that like?
So back in the early 90s when I was a postdoc, I was working on worms in Tom Johnson’s lab, and Tom’s lab was the first to discover a mutation that extended lifespan, the gene age-1. And I had gone to Tom’s lab because of this, and it was very, very early days, and I think most people didn’t believe that you could find mutations that extended lifespan up until then. So we were working on the very first mutation, and the only mutation in the world at that time. And I found that these mutant worms were resistant to heat, and there was a tumbling out of possibilities from there.
As soon as I saw it, I was very, very excited about it, and there was no one in the building that I could tell, so I was running around the building trying to find someone, but it was late at night and everyone had gone home. But I realized right then and there that if the animals were resistant to heat, chances were they were making more molecular chaperones, and if they were making more, then maybe that was why they were long-lived as well, because they were able to maintain protein shape. And all of that happened very fast in my head. Of course, I wasn’t the first person to think about proteins in aging. People have been looking at protein damage, especially oxidative damage, for years. But this idea of an involvement with molecular chaperones was pretty new.
But I also realized that if you gave the animals a mild stress they would elevate the molecular chaperones, and I wondered if that was sufficient for them to be longer lived. I should say there were people working in oxidative stress and lifespan on this mutant before me, so I wasn’t the first person to think about stress, but certainly the first person to try heat shock. So I ran to Tom going, “You won’t believe this, Tom, but you can give the animals a stress so bad that it knocks out their fertility, but it makes them live longer!” And he opened his filing cabinet and pulled out a paper, and he said, “Yeah, you’ve rediscovered hormesis. It was published in the 1950s.”
He showed me a paper by John Maynard Smith in the ‘50s, who was interested in fertility and lifespan, and so they were doing heat shocks on drosophila to knock out fertility. What they didn’t know at the time was that they were also turning on molecular chaperone genes. And then he showed me all these papers on radiation and Hiroshima, and the idea that there was a band of distance from the epicenter where there was evidence of beneficial effects of the radiation exposure. So catastrophic all the way through, and then a band of people who were exposed at a low level with some evidence of benefit. And this was all from many years previous.
So we published in PNAS and said mild stress increases lifespan, and suddenly everybody was doing it. You would go to all these conferences and everybody was talking about hormesis and lifespan and chaperones, and for a while it was really, really exciting. Not many people are really working on it now. There was a series of papers from Michael Ristow’s lab where, and this is what muddies the water with oxidative stress, they were inducing oxidative stresses of various sorts and showing a lifespan increase, and they could suppress the lifespan increase with an antioxidant. But the model is actually supportive of the oxygen radical theory of aging, because they said we’re causing a small oxidative stress, that’s raising defenses, and that keeps the animal living longer. And because it was mitochondrial stress they called it mitohormesis. Then just this year Malena Hansen at Scripps published a lovely paper where she showed that autophagy was required for the heat shock hormesis that we’d been doing all those years ago, which was nice to see.
So many people have been working on caloric restriction, and very few people are working on hormesis now, but essentially they’re both environmental manipulations that increase lifespan, so I suspect that lots of us will go back to looking at hormesis. It’s just that the effects are not enormous, and when you’re working in a worm and one intervention gives you a 70% increase in lifespan, and this other intervention gives you 20%, you work on the bigger one, because it’s much easier.
Are different hormetic effects additive?
I would expect them to be additive, and we published one example of this, but it was only with heat stress. So we heat stressed them, left them for a few days, heat stressed them, left them for a few days. We got diminishing returns, so it wasn’t completely additive, but we also saw a little bit of benefit: if you heat shocked them three times, that was still better than heat shocking them twice. That was like 10 years ago. So yeah, I would expect additivity, and I think if we were doing mitochondrial stress and heat shock together we might see additivity as well. I don’t know if anyone is doing those experiments. We actually are looking at additivity of compound treatments as well, because there could be similar scenarios where you’ve got a caloric restriction mimetic and something else that’s working in a different way.
Have you undergone any significant shifts in your approach to studying the aging process or your conception of the aging process since beginning your career?
Yeah, aging is a bit more complicated than I hoped it would be. I hoped it would be really simple. The biggest conceptual shift of course was the discovery of the first aging mutation, so Tom Johnson’s work and then Cynthia Kenyon’s work. For the people who physically saw those worms down the microscope and they were alive when they shouldn’t be, it usually has a massive effect on you, and you want to then devote your career to working out why this is happening. And I think we had the same feeling with the antioxidant drugs in 2000, where again it was, “Oh my goodness, we can do this with a compound, that’s incredible.” And that was definitely a shift for my lab, even through all the difficulties of reproducing things, to realize that you could pull off the same thing with small molecules that you could do with genetics. And that meant you could probably translate that much more rapidly into potential therapeutics, which was massive.
So a year later the Buck opens, and at the time I’ve got no interest in humans, no interest in human disease, no time for the anti-aging crowd. I just was studying a fundamental and interesting biological process. The people who arrived here were people like myself, and then there were people who were studying Alzheimer’s or breast cancer or stroke or different disease states, and I didn’t think I would have much to talk to them about. But over a four or five year period we realized just how intimately linked aging was with these disease mechanisms. Of course I’ve been thinking about protein conformation for many years, so parts of this were somewhat obvious. But the very fact that this lab over here that’s studying lifespan in flies was studying the TOR pathway, and then this lab that’s studying breast cancer was also studying the TOR pathway. It was like, oh wow, this is all really intimately connected in ways we can find out, we can actually find these connections and explain them.
We were increasingly focusing on protein homeostasis and started making worms that overexpressed molecular chaperones, and then we started talking to people who work on disease and realized that we should be modeling disease in our aging experiments, and so we started using Alzheimer’s and Parkinson’s models. There’s a personal story here as well, because my wife Julie Andersen came here to work on Parkinson’s mouse models, and our research didn’t overlap at all then. But over the years we realized that we were working on the same mechanisms, and Julie showed that the compounds we published in the Science paper also helped prevent Parkinson’s in mouse models. Over the years we’ve published a lot of studies going back and forth between longevity and neurodegenerative disease. So that was a slow conceptual shift, but it’s a massive one in the way it influences what experiments we do and how we talk to our colleagues, and then how I talk to the public. I’ve talked to press my entire career, but I was always talking about longevity, and I wasn’t talking about the fact that we have an opportunity to prevent some of the most expensive, awful diseases that we know about. We have a chance to do something about this, and we need to change the way that we think about aging as a society in order to bring this about.
Is it controversial to approach age-related diseases from an aging standpoint, rather than attempting to treat each disease individually?
It hasn’t even reached a level of being controversial. For it to be controversial, people have to be talking about it and fighting about it in journals, and I would say that the wider biomedical community has very little appreciation at all for what’s happened in the biology of aging. It’s not a surprise at all – we haven’t communicated it. We talk to ourselves a lot. We go to worm meetings, we go to fly meetings. We go to aging meetings and we talk to each other. We really don’t do a good job of getting it out to the wider biomedical community. I think there are some individuals out there who really know and appreciate what’s going on, and they’re advocates for the field, but I would imagine that if I was doing a residency in the local hospital here, even if I’m a geriatrician, I’m not necessarily aware of what’s going on. We’re very fortunate, we’ve got a chap who’s just joined us, John Newman, who is a geriatrician and has been following the biology of aging, and I know of only about two people in the country like that. I think there are only two grants in the entire NIH portfolio that are aimed at geriatricians doing biology of aging research. So yeah, it’s not controversial because nobody knows about it.
Why isn’t the general biomedical community more proactive about interfacing with aging research?
Well, there are two things. I think that in order to have a proper debate about this, we have to have a success story in humans. We have to show that an intervention that arose from the biology of aging and all this research that has been going on for 25 years is actually helping someone or a population of people. The current debate about the TAME trial with metformin, that might provide it. But we kind of need like 100 TAME studies, you know? We need success in humans. We’ve ample success in invertebrates and even mice now to say that the course of disease can be changed, but that’s not impacting clinicians.
We don’t necessarily make the kind of measures that impress clinicians. I mean, you go to a clinician and say, we’ve made mouse lifespan longer, what are they gonna do with that? You know, on a daily basis, their experience is trying to cure people of disease and maintain functions. They measure blood pressure, they measure heart rate, they measure glucose intolerance, they measure things that give them certain information, and we don’t measure the same things. Simon Melov, who’s also here, his research is trying to do things that clinicians would deem important. Do in mice what you do in humans, give them the kind of data that they would have coming out of humans, and maybe they’ll pay attention.
So if it’s anyone’s fault it’s our fault for not doing the most relevant research over the years, or for not following up on our interventions. But it’s also resources, because we have 100 compounds downstairs in freezers that have really interesting effects on aging in worms, and we’re only working on one or two of them. If we had the resources, they would all be in mouse studies, and we’d be looking at healthspan measures in mouse and function and cardiac function, but each of those experiments is a half a million dollar experiment.
Would having valid biomarkers help in terms of getting clinicians more on board?
Definitely. Especially if they’re the biomarkers they already understand, that they already measure, inflammatory measures and so on, the sorts of things that you learn about in medical school. If those were biomarkers of aging as well, then that would be terrific. But it’s a major issue. Partly it’s an issue because if you’re thinking about human clinical trials for longevity interventions, you’re thinking about longer periods of time than normal, and any clinical trial that’s over three months long is an expensive clinical trial. So we’re talking about effects that we might not be able to detect for years, because we need that time to see something play out. You have to have dysfunction in your control population in order to see some sort of rescue. So any sort of biomarkers that take into account those long timescales and that have a high dynamic range would be fantastic. It’s been a very difficult issue for decades.
Money was spent decades ago trying to find really good biomarkers in mice, and it failed for the most part. I think we’ll end up looking at function, and this is something that we do now, where you have a large panel of measures, you know, tail stiffness, fur quality, some of which are actually quite subjective. But as a panel with equal weighting, they can work effectively at tracking aging. I think that’s what we would do in humans as well.
What the TAME people are tracking, or proposing to track, is events. So they’re not tracking a biomarker, but the frequency with which somebody turns up in the hospital with a particular condition. So the prediction is that events in total will be lowered by something like metformin over time. And that’s good, because that’s not expensive to measure, you just wait for people to turn up in the hospital. You’re not actually measuring something directly.