Coleen Murphy is a professor in the Princeton Department of Molecular Biology and Director of LSI Genomics who studies the molecular mechanisms of aging. She’s also the author of the 2023 popular science book How We Age: The Science of Longevity. We sat down together to talk about the book, why she wrote it, and lessons learned in the field of research into aging. The interview below has been lightly edited.

The Birth of an Idea   

I guess my first question for you is, why did you decide to write a book?

It came out of some discussions with an editor at Princeton University Press. It became quite obvious a few years ago that there are lots of books about aging and longevity that are aimed at a lay audience. I think a lot of lay readers at this point know a lot about aging because it's such a popular topic. And at the time, I had taught for many years a class on aging to undergraduates and graduate students about cutting edge science in the field. That kind of class actually sets itself up really well for this kind of book where you can really delve deeply into the current papers and what's happening in the field.

Of course, the problem with this book is that the minute I sent off the proofs, there was still cool science coming out. I have to get over this feeling, “Oh, no, I didn't include that in my book,” because it is just always such a fast moving field, more information is always coming out about how all these pathways that extend longevity work at a molecular and cellular level. And although this is not what my lab does—or is not what we've done so far—I'm excited to see that it's not just a lot of talk. There are people actually trying to take some of these pathways and find targets and molecules that can turn into drugs that could help people.

Who is your target audience for this book?

My audience would be any of those people who say, I want to know more. I don't think it's a beginner's book; I go right into all the research. But I want to give people credit: there are a lot of smart people out there who are not directly in my field but who would read some of this stuff and at least want to know how some of these things work. 

Also, I also wanted to mention that in all of these other books, women are hardly ever mentioned. That’s crazy. Reproductive aging is the first real sign of aging and menopause accelerates aging. So I did want to make sure that people know this is not all about 80-year-old men.

A lot of the work in the lifespan field has been done with female animals, and in the book you discuss that longevity is not a trait that's selected for by evolution. Why is that?

Intuitively, we know that if you have really old organisms around, they're probably not contributing much one way or the other to the population. There used to be these ideas that even for organisms like the worm C. elegans, which I study in my lab, you'd want to get rid of old worms because they're eating food the young could use. But actually, most worms don't even eat much food after about day four of adulthood, which correlates with their decline in reproduction. So really, all these pathways that we think of as longevity pathways are actually reproduction optimization pathways. 

Animals need nutrients in order to reproduce. If they don't have enough nutrients, one smart thing to do is to slow down your reproductive system until times are good again, right? And so, the pathways that tell the animal to slow down reproduction and keep their body healthy at the same time, those are the classically studied longevity pathways.

So the idea is that females need to be robust so that they can carry the reproductive process through to completion, and anything we get after reproduction, lifespan-wise, is just gravy?

We noticed that C. elegans has a really long post-reproductive lifespan, which is weird to think about because they don't really care for their young. All these social theories that work for mammals, I don't think we can apply to non-social invertebrates. So I had an undergraduate in my lab ask, what does it correlate with?

And we found this really weird thing, which was that if you build a ratio of the size of the mother to the size of the progeny, then the bigger the progeny are relative to the mother, the longer that post-reproductive lifespan is. Let me give you an example. If you have a sea urchin, that's a fairly large animal, but its progeny are basically microscopic. The sea urchin can go on to reproduce basically until the day before it dies. It has almost no post-reproductive lifespan because it's so easy for them to reproduce that they don't need to be somatically very healthy, right? 

If you look at C. elegans—and actually this is true for humans as well—they have really large progeny relative to the body size of the mother. The worms have to be pretty healthy to lay eggs. It’s gross to say, but they literally explode if they're in poor health when they're trying to lay eggs. That's what made us think of this idea in the first place, but it scales really well across a bunch of different animals.

In humans, male reproduction is easy, yet they get to live almost as long as females. 

But they can also reproduce much longer, so actually, their post-reproductive lifespan is not that long. So it fits.

The Battle of the Sexes

You talk in the book about how male reproductive strategies are often counter to a female’s desire to live a little longer after having children.

For males, the goal is to get as many of their genomes out there in the world as possible. Mariana Wolfner and  Linda Partridge have shown that male flies deliver accessory proteins in the seminal fluid that make females less receptive to the next male. That’s very gentle compared to C. elegans, where the male delivers seminal fluid that basically poisons the mother and kills her just after she's produced all that male’s progeny. 

In C. elegans the mothers are all hermaphrodites, right? Your book explains that the hermaphrodites only mate with themselves, and avoid males until they’re down to their last few eggs. 

Yes, when they've got left over oocytes they switch their pheromones so suddenly they are not so repelled by the males. Then they'll mate and produce those progeny, but then they die. 

So that sounds like the males have won, right? They've gotten what they want. But the species has programmed in a self-destruct button for the males: the males produce a pheromone that is very toxic to other males, even themselves. Eventually because of this male infighting the population goes back down to almost all hermaphrodites again within a few generations.

Males are an accident in C. elegans, right?

Exactly. They’re made through a loss of the X chromosome, and in a normal well-fed population they're one in a thousand. But in times of stress they'll start to make more males and this increases crossing, which you could imagine helps with fitness. But they're very quickly driven back down by this male pheromone issue. 

That sounds like a good strategy for the hermaphrodites.

It's so successful that C. elegans have evolved hermaphroditism multiple times independently. They have some males around when they need them, then get rid of them again. There's a real feminist bent to all my research but it's not on purpose, that's just what C. elegans does.   

The book gets into the nitty gritty on the molecular pathways underlying all these reproductive-optimization-slash-lifespan-extension pathways. For example, you spend a lot of time in the book talking about daf-2 mutant worms, which are deficient in the insulin/IGF-1 receptor.

Daf-2 mutants seem to have it good. They live twice as long, they’re healthier all that time, and the cost is just that they don't have quite so many offspring?

That’s a big cost, though. If daf-2 mutants are out in the wild, they're going to lose out to wild type almost instantly because they only have two thirds as many progeny. Kind of a big deal. But certainly in the lab they've been a great model for us to pull things apart and really understand why do these animals have better preserved oocytes? Why do they have better memory? That's been the most eye-opening thing: they really do have better memory, and molecularly, that’s for completely separate reasons than why their body stays healthy longer with age.

This touches on one of the concepts you discuss a lot in the book, which is the idea of “healthspan”. What is that and why focus on it? 

It's one thing to want to live to be really old, but nobody wants to live forever unless they're in good health, right? People don't assume that living long equals living in good health. In some cases, that's true. In some cases, it's not. And so the focus on healthspan is reflecting what's happening in the broader field as well: this idea that there are age-related diseases and if we could slow those down, not just by slowing down aging, this would improve quality of life. 

Biotech companies are using these age-related diseases as proxies for lifespan. That's pretty smart because there is no FDA-approved assay for longevity, and you can’t get FDA approval for a drug to treat aging. So really they have to do that to test their drugs. I think that is very smart, but also it's just what people are interested in, right? We want to have our brains working as long as possible and be able to walk around and, you know, hold our grandkids and stuff like that. So I think that focus on healthspan is really intuitive, and there's a lot of scientific grounding now that we can base this on.

Live Long and Prosper

You explain that one of the genes that contributes to healthspan in worms is called daf-16, also known as FOXO in mice. This gene encodes a transcription factor that acts downstream of Daf-2 to activate a whole host of pathways in the cells…  

I think most people have heard of many of these pathways before. Things like proteostasis that keep proteins functional longer, or repair damage. It’s really the collection of all of them being boosted together that makes the worms live so much longer. And those are things that are in us as well. At the cellular level that's all extremely well evolutionarily conserved. 

DAF-16 is also required for neuronal health but it picks different targets in neurons. The super scientist in my lab, Rachel Kaletsky, discovered a while back that you could take apart C. elegans in adulthood. She got neurons out of them and looked at what was going on in daf-2 and daf-16-mutant worms and found that they had a very small set of overlapping targets from those in the body. They still have turned on those same pathways as in the rest of the body but they also regulate things like the CREB transcription factor, which is responsible for memory and keeping the synapses functioning longer. 

The reason we're excited about that is because when we looked at one of the proteins that is involved in this this CREB transcription factor pathway, we figured out ways to hyperactivate it in both young and old worms, and that boosted memory. Then we got our collaborators to make the same mutation in the homologous protein in mice, and that was able to boost memory in those really old mice. So, you know, I have fun talking about C. elegans but if we can find something where it really is a good model for something we care about, like cognitive aging, that's super exciting. It justifies all the time we spent trying to work on this because we don't really care about worm memory. We want to make things better in humans. That's the goal.  

The counterweight to DAF-16 is a protein called PQM-1. What drives PQM-1 activity?

In developing animals, PQM-1 is in the nucleus turning on a lot of the growth and developmental genes. And then under stress, DAF-16 comes in and seems to kick PQM-1 out. They're not entirely mutually exclusive; there are some conditions that will have both of them in the nucleus together, but most of the time it’s one or the other. So there’s fighting stress versus development. This is a pretty constant theme that cells can either be dividing and reproducing, or the cells can be fighting to maintain themselves.

But this raises a good question, which is that if you take some sort of drug to turn on your FOXO, is there going to be a downside at all? The longevity field is having to ask this question a lot, for example, with senescence. Judith Campisi talked about the reason that there are senescent cells is in order to avoid developing cancer.  We as a field need to be careful that we don't do something in our efforts to prolong life that's going to turn off surveillance systems for things like cancer.

I think you can definitely get rid of senescent cells with no cost, but that's probably an area that people are trying to understand better. Under what conditions can you get rid of them using senolytics or senomorphics and really be all healthy, versus the conditions where you'd cause extra problems by turning them off? The senescent cell field is actually very visibly addressing those questions. There are actually some of these things getting to clinical trials so I think we'll know some of this stuff soon.

We want to get rid of these pathological cells, but also renew functional cells, right? There’s a lot of excitement about the four proteins, the Yamanaka factors, that can induce pluripotent stem cells. You can see it all over the internet, all these ads saying let us give you stem cells to make you look younger, get a new liver, or whatever. I'm not sure some people are properly skeptical on that front.

Yeah. I think those are the right words. Properly skeptical. That field is moving particularly fast because there's so much interest in it. I think we'll know pretty quickly whether they're useful.  But I’ll tell you a story. It’s the kind of thing I hate, an “n of one” kind of thing. But I'm the biggest skeptic of all these things, and I went home a couple years ago to my parents‘ house and my mom showed me this packet. She’d been having a lot of problems with her knees and hips and was going to go off to this clinic to get some stem cells injected into her joints.  

I was like, that sounds like a terrible idea. I was really worried about her getting cancer or something.  And also I didn't think it would work, right? Then she showed me all these papers. I guess knowing how skeptical I would be, she'd actually already done her research. The papers looked at people who’d had this procedure done and didn't ask them whether they felt good or how had better mobility, they just asked how much ibuprofen they were taking every day. And it was pretty amazing, it went down. After a year it's hard to imagine a psychosomatic effect or some short term thing like a placebo effect.  So I said okay, sure, go on.  Not that she was asking for my permission, she just did it.  And I tell you, it's two years later and she's walking around, no problem. My parents are in their 80s but they’re super active. It’s just remarkable me that this actually worked. I don't know if it's actual stem cells or something else in the fluid. I think everyone should be cautious, but maybe also cautiously optimistic that maybe some of these things are moving forward and not everyone's trying to pull a fast one. I’m usually skeptical but…

Be skeptical until you've read the papers and decided they’re good ones and the statistics are done well?

Right. For my mom it was kind of miraculous, so I can't be too much of a naysayer when there's some actual evidence of things working. Again, that's an n of one, but still. Pretty impressive.  

The Future of Aging

So it seems there are lots of angles from which we could tackle, let’s say, the symptoms of aging. But for us humans, it doesn’t seem we can just flip one switch and suddenly you're living twice as long. We can only address one pathway at a time.

Yes. To some degree, that's the approach for now. People will get a disease and they treat that disease and move on to the next one. But there are a lot of age-related diseases, and it turns out a lot of things that are helpful for one particular age-related disorder might actually be beneficial for a lot of them simultaneously.

But of course, the only way that you can get something FDA-approved is to have a particular disease or disorder that you're going to test on. You can't just say, I'm going to go and fix everything—although there are the metformin and rapamycin trials that are going right now that are kind of asking that question.

But in humans, there doesn't appear to be a switch like DAF-2 in C. elegans, where if you knock down this one protein then you get super worms?

Yeah, I haven't seen any evidence of a similar thing in humans. Caloric restriction advocates will say that it works like that. And maybe that does have a lot of benefits for your body, but it doesn't seem great for your brain and it’s not good for anybody's mood. There are people working on the mimetics of caloric restrictions and perhaps those might be systemically switch-like.

So no magic gene that works like the human version of Daf-2 in C. elegans?  

I never want to say never. The reason I say that is because before 2020, you would never have said that there would be one molecule that people could inject and they would lose weight magically without any obvious side effects.

Yet now we have Wegovy, also known as semaglutide, which also seems to have benefits for the cardiovascular system, compulsive behaviors, and other things... 

Exactly, right? Today was the first time I read something bad about that: a very rare condition called gastroparesis where the stomach becomes paralyzed. But there seem to be a lot of other beneficial effects.

 Which leads me to a question: what happens if you drop Wegovy on worms?

I don't know. I probably can't afford the prescription to do that experiment. [Laughs]

[Laughs] Okay, but I think one point the book makes quite clear is that we would never have learned how any of these pathways work had we not had the worms and other model systems to show us the way.

Right. And there's stuff that we didn't think C. elegans or flies would be a good model for, but they turn out to be much better models once you figure out what the analogy is. Once you get down to the molecular level, evolutionarily things can be very well conserved, depending on what you're looking at. You can do experiments really quickly and really understand mechanisms in a way that get quite hairy and expensive even when you go up to just mice. So yeah, I love model systems. 

At the end of the book, you talk about the many interventions that are the products of work in model systems, and that someday soon might be coming out of all these new biotech companies. It was really long list.

Yeah, it's a lot, right? But that's good!

Of course, there are some things we can do that you mention a few times that we can do already that I don’t involve drugs.

I mean, those are things that people probably knew before they picked up the book.

[Laughs] Fair. But what are they?

Well, everyone can eat a little bit less, probably, although that's not my favorite thing to do.  And then the most important thing I think is exercise because it turns out that the more we study exercise, the more we learn all the things that are going on when we exercise that really help the body. And then there is sleep, but that's easier said than done, I think.

And the last one is “be rich”, which, as you point out, is not accessible to everyone.

Exactly. And many people have limited control over their own diet, depending on their situation, right?  So we have to recognize that as well.

  When we opened the interview, you mentioned that as soon as you submitted the book, you regretted that new things kept coming out that you wanted to include. Like what? 

I should keep a running list! But here's a nice example: there’s a whole chapter where I talked about epigenetic clocks that things that Steve Horvath and Morgan Levine have found with DNA methylation, right? Well, there was a very nice paper soon after the book was done by Tony Wyss-Coray and Anne Brunet where they actually looked at transcriptional clocks. They were interested in a part of the brain, the sub-ventricular zone, where new cells are made to make new neurons, and if there’s a transcriptional clock that they could use. Once they found this clock they could then show that exercising mice, for example, slowed the clock down.

And in fact along that theme, I saw paper just the other day from Bjorn Schumacher saying that these DNA methylation clocks are not real clocks, they’re just marks of increasing stochasticity in cells. That's why it's been so hard to tie the clocks back to a molecular understanding of what's happening. All these things are interesting and from a clinical standpoint they're important because people are starting to sell these clocks as ways to monitor biological age, to help inform about whether to start taking an intervention or see if it worked or not. 

Well, maybe you can put that in the sequel?

[Laughs] Yeah, maybe the sequel. Like, ten years from now.