Promising aging approaches from an NIA perspective | Ronald Kohanski, National Institute on Aging

7:36AM Apr 9, 2021

Speakers:

Nir

Keith

Ron

Keywords:

aging

intervention

mouse

question

people

disease

nih

health

phenotypes

rejuvenated

sorts

set

understand

population

methylation

function

genes

lab

older

score

Thank you everyone for joining again, I'm super excited for another keynote meeting of our health extension group sponsored by 100 plus capital. Thank you all for joining. It's really nice to see so many of you. And I think, just as a brief announcement, you will be getting a lot more emails for scheduling a lot more meetings, because a lot more meetings have been proposed. And if you can't make individual bonus meetings, that's totally fine. Those keynote meetings here really are the ones that that we love everyone here for. So for this one, I think in particular, I'm super excited to welcome Monica henskee, from the NIH for a meeting that many of you have been looking forward to for a really long time. And that has also been in the planning for a really long time. And Ronald has been promoted to the Acting Director of the Division of aging biology at the National Institute on Aging from the NIH. And we're incredibly honored, I think, to have him today and present to us a little bit. I know he's loving, you have to take it. I will introduce this. And I were really excited to have him first present on the naa and then answer your questions. And I think he wants to make an open session starting, starting after his presentation. So we're really looking forward to a really broad discussion. And Neil unmuted himself already Do you have something to say?

I love I love seeing Ron when he's awake. That's all, you know. Always a pleasure.

All right, Ron. Well, no question. I knew that. And I think, yeah, without further ado, and we're super, super happy for your presentation, I will share the more of your bio and background in the chat to not take for the time away from you. And I'll be admitting more people as they join. Really, really, really excited to have you here. And I'm looking forward to this meeting very much. All right.

Thank you, Allison. And good morning near? Yes. So I'm going to treat this actually as as something of a therapy session, I'm going to tell you what's bothering me. And then you can sit there and say, That's interesting. How do you feel. But what the way I feel about this is biology of aging is has literally come of age over the last 30 years. The Division of aging biology. Put this topic on the map really through a series of requests for applications and funding of applications. That was done by Dr. Anna McCormack when she was here. And it was on longevity assurance genes. And it helped to people to understand that there's a molecular basis for what was then a relatively obscure topic, which is what is aging? And how does it happen. So I hope you can see the slides. And I'll just jump right in. Also, thank you for the invitation and for your forbearance and listening to this presentation. So I mean, it's just begin with a little perspective, which is that living is what we all do, as long as we can. And it's pretty much what you do day to day. And at least when you're a young person, you expect to increase your capacity and your activities and your abilities. And as an adult, you'd like to maintain a certain level of activity. And this is all related to some indices of health. And in fact, what you can do day to day, would indicate good health as long as you maintain it. But when you have aging, your activities are harder to do certainly year to year. And you also do expect them to diminish, even though these are indices of health. So one of the questions becomes, in the face of diminishing capacity, are you still healthy? And the answer is actually Yes. But the question is, how do we understand that? How do we better maintain it? And for those whose health is declining a little bit more rapidly than we'd like? How do we ameliorate that? So one of the things that we talked about frequently is the rate of aging. And it's something that we think we know what we mean, but it's not necessarily so obvious what the metrics are. And I'll go into why I think that although there are multiple metrics, and I'm proud to say that the NIH has supported quite a few of those studies. So aging encompasses, as mentioned, the decline of function and also an increase in risk for disease. It's basically the geroscience hypothesis, that are the foundation for general sciences that aging is this major risk factor. If we could slow the rate of aging, we could decrease the severity of disease and delay its onset. But it also has to do with functions. So walking speed is a very good indicator of expectation of mortality actually. And walking speed diminishes with age, it doesn't really decrease in a linear fashion. Actually, it accelerates a bit as one goes from one decade to the next. It also there's a rise of course, as function is acquired, counterbalanced by the loss of underlying capability. But a lot goes into walking, it's a complex activity, even if not chewing gum at the same time, because it involves muscle strength, it involves bone quality, joint quality, speed of neuronal conduction, perception, balance, a great many things go into walking. And all of those parameters are and can be analyzed pretty well. And a major disease that is associated with age, and certainly the increase with age is cancer, all sorts of cancers. And these are data taken from the NCI website where the age diagnosis is shown here. And the disease prevalence per 100,000 is shown here. And this is an exponential increase.

So there are a lot of clocks, of course, for aging the early on, in 2013, hanham, in collaboration with Ida kurunjang, came up with an epigenetic clock, along with Steve Horvath, and Morgan Levine entirely separate work. But basically looking for an elastic net linear fit to a set of methylations on DNA. And using elastic net, of course, you're going to get a linear outcome. And the question is, what is your methylome age as it were related to your chronological age or time since birth. And it is interpreted as saying that those who have methylation ages above the average set are aging faster, and those who are below the average set are aging slower. And it looks like the evidence is supporting that in a lot of cases, and methylation age can be determined from tissues. And all the constraints of those are becoming better known and better understood. And there are a lot of methylation Cox at this point. And they cross a lot of species. But one of the problems is though that the methylation age is linear and the risk, the appearance of the risk anyway is not. So it's exponential in terms of disease, and it has some type of loss of loss when it comes shape when it comes to loss of function. So here's the methylation clock, it's linear. If you look at risk, our prevalence from a single factor, it would be linear. But if you had two factors interacting, as we know from cancer, and many other diseases, where there's environmental influence interacting with a genetic factor, you'll get a nonlinear behavior. And if you have three factors that are interacting, it begins to look like an exponential change. So it's not quite clear how one relates this to that to the actual appearance of the disease, and therefore the understand the risk. But it's likely that thing that whatever the metrics are for aging. The methylation clock, if it's causal is interacting with at least one other factor. So the other thing about the lack of linearity and some features of aging comes from the work of Tony was Korean your bars lie in a paper where he was the lead author, and they looked at roughly 3000 proteins in plasma. And they found that there are changes and there are basically three peaks of change in these aging proteins are occurring early, late mid life and in old age. So what kind of clock represents this pattern of aging is not yet clear. Then we get on to the question of healthspan, which is the period of life that's free of disease, that's the accepted definition. So here would be prevalence of disease in human populations are from some data compiled to 2014 and earlier, and this is the period of life that's where most of the popular You can say would be free of disease. So healthspan would seem to come to an end, by the time you reach midlife, which is depressing. But there may be other ways to look at this, the term health span, at least in the point of view of NIH, a big became prominent because the metric we had been using for 15 years was lifespan. But we were pivoting toward health. And so health span became a natural, not necessarily neologism, but the term that we went to. But when you get out to these older ages, you have multi morbidities. And I'm not going to steal any Thunder from from Neil by any means. But by pointing out that this is an area where the team trial becomes extremely important.

And we're hoping to see that results from that in the not too distant future. But of course, it's take five years from the time it starts. So here's a question, starting to put some things together. So you have the How do you relate all of these changes that you can detect in the plasma proteome or the methylome, or metabolome, or almost any other way of looking at molecular changes across age? How could you relate this to the prevalence of disease, which of these 3000 proteins, if any, are risk factors for any of these conditions? That's a challenge. And that's the challenge for geroscience to unravel. So there's another way of looking at it, which isn't a term already in use in, in insurance companies, that's health expectancy, right? They come in, and if you get met by an insurance agent at home, they'll come in, draw some blood, take some information, go back with that information, try to determine what's your next likely disease and how healthy you're going to be? And how much should they charge you. So they don't just been by your chronological age, they also been by some of your health metrics. So there is an expectation of what your health is going to look like in the not too distant future. And I think something like that might actually be useful when it comes to geroscience. So it's biology or physiology over time, that's what we're asking for. When it comes to health expectancy. Either you look at your recent trajectory, or you take a single time point. And then relative to other known trajectories or collections of time points, you ask, how will this person compare relative to the rest of the population. So there are lots of ways of looking at biology over time. There's fetal development, which is that the Ballard score actually applies to people. But you get the same concept when it looks at mouse development in embryos, or embryonic development in Drosophila, or in zebrafish, or almost anything out there that goes through an embryonic phase. So you have set of scores where you can look at parameters, and you can tell where you are relative in the course of development. And you can also tell where things are not going well, you can find the outliers in any of these organs, if they do not fit a specific pattern. So things generally speaking, go in concert, but not always. And in fact, for mammals, approximately 40% of this process fails. It's not really 100% efficient, not even close. So the failures can occur in all sorts of places. Of course, a birth can happen, where some failures along the way have taken place, but they aren't severe enough to abort the process. Then there's health at birth, which is the app car score, these five parameters named Actually, it's a reverse acronym, Virginia app car was the person who came up with this, an anesthesiologist at Columbia University. And this is a way of assessing a healthy baby with a score from zero, which means inactive, no respiration, etc. but can be treated or all the way up to 10, where things look pretty good. So there's a score at birth. There's also the transition from being a baby to being a teenager. And there are sets of scales for what is going on during maturation and puberty, the tenor scales, and these all have scores as well. And you can tell because The multiple components, how well a person is doing in terms of their maturation, the development of those capacities that will allow them to do their daily activities, including reproduction, and their multiple components. So you can find out if things are in synchrony, or have they become this synchronous. And you can also see in shifts in populations effects of environmental influences. As the the, I'd say the onset of monarch, for example, has shifted to earlier ages, probably due to environmental influences.

And we also know that what constitutes health during child development, the American Academy of Pediatrics has these charts that anybody can follow, you can fill them in at home with your own child. And basically, what happens here is that you have deciles of condition, height and weight at birth. And pretty much if you stay in your lane, you're healthy. So a person who's light and short will be healthy, staying there. And a person who is taller and heavier will also be healthy, if they continue along that pattern. But if you break across lanes, there is then some indication of an unhealthy condition. It's pretty simple. And as to what it is there. There's a long history, of course, pediatric analysis, so it's possible to identify what the problems are. And in many cases, there are reasonable solutions that can be used. So what do you have for aging? Assuming from my point of view, aging is an adult process on it could begin earlier, it's subject to debate, but I would draw your attention to work that's emerging from Vadim gladyshev slab on that topic, which is quite interesting. But come to that another time. So you could have an aging score. And if you can figure out what it is you get your name there. And you have various components, you have physiology, which could be viewed as resilience or frailty, there are a lot of frailty indices out there, which some of which are based on function, some of which are based on molecular characteristics, or clinical characteristics in some which combine these. And then there are extrinsic risks, which affect age, and will, of course, affect health at age. And then I'm not going to develop it in today's talk. But there's a viewpoint that the minimum unit of a biological system is two interacting components in a feedback loop, or some form of feedback loop in a network. Because in those situations, if you change one, you change both. And those results spread through a network. So it wouldn't be just one node, it would be interacting nodes, but could debate that on another time. But one, one of the two things that that happens here, when you look at these health expectancy space would be what kinds of outcomes are you going to predict? So are you going to predict time to death? That's very important and very useful. It can be disconcerting to be told. But that's an ethical question beyond my scope. But there's also the question of what do you expect for your health? What not necessarily, what's your next disease? But what is your expectation for continued function, and to the best that could be maintained. And that best is again, softer. So the next thing is intervention. So you have some score, or some way of assessing health. And you want to develop interventions as you do and which is one of the reasons that I listened to this group is to try to get some idea of your ideas about what interventions might be there, and how one goes about screening for interventions and how you score interventions. But aging is an inherently heterogeneous process. So it's a little bit difficult. So here's one way of looking at it once again, from the cloud, the shift lab. This is taking interventions that affect lifespan, and in almost all cases, I actually can't think of an exception when in the intervention testing program that NIH supports and also in other research that's done outside of that purview. Taking laboratory mice, whether inbred strain or the heterogeneous mice used in the intervention testing program. If you extend lifespan, the older mice tend to look healthier than the control mice and What cloud is Chevron and Miller, I was thinking, I might have been on this. But in any case, what the gladyshev group did was they, they took the liver from these, and they looked at gene expression signatures in the liver. They also looked at other tissues, but shown here is for liver. These are a range of interventions, which can buy clustering, you can see that some of them clearly have more in common than others. Some just don't seem to have much in common with any of them. So you can map these through computer tools to a network. And you can see that there are a range of

here. These are the growth hormone receptor knockout, little mice, Ames, no FGF 21. These are all genetically modified mice that have extended lifespans. There are also some chemical interventions that were tested. And there's also caloric restriction, which is pretty well established that it works in all, if not most, most, if not all, laboratory mice. Some question about that. But rapamycin is an outlier. So you can find signatures from the livers. This is these livers I believe were taken before the or somewhere near the breakover point in the curve. So they indicate in a sense of future performance following the intervention, so this is the proposal here is intriguing, because the question is, if you can use this as a screen for an intervention, which would extend based on extension of lifespan, might you then have an intervention that would improve health. So then we get to the other thing, which of course is well, it's not really about mice, it's about people. So if you look at relative age, often referred to as biological age, or physiological age, but let's say relative age, that is within a group of people at one age, as in the Dinesen study, and this is a quick representation of the kind of study they they do by belski and Terry Moffitt and Caspi and others. So it's a study of about 1000, people who keep coming back year over year they've been followed since birth, I think they're 45. Now, they come back from all over the world to spend time in this program, although they were all born in New Zealand. So what you get here is what you expect that for any population is going to be some average set of characteristics, whatever they are, they're 19 biomarkers used, and you get an aggregate score. And you can say there are people who seem to be aging slowly, and people who seem to be aging quickly, relative to the average. And in fact, Terry Moffitt and others in the, in this program have shown pictures of people who are, let's say, by these metrics, with the standard doctor's visit data, standard dentist visit data and pulmonary metrics, which is not exactly standard, but it's not a big deal to go through. This takes about an hour, I think you can you can take the faces of these people and make an aggregate using computer assistance. And the people who by these metrics are aging faster, look older, and the people who are aging slower, actually look younger, males and females. So it's a remarkable piece of work. So this is the why is this a starting concept for intervention is that if you look at a population, so let's say you're going to do a study for 20 years of an intervention, you have some biomarkers score, you have some people who look younger, and some people who look older, if you knew who they are using this image, facial image, you might be able to stratify the population. And you might suspect that people who look older would benefit more from the intervention. So if you use the whole population and took averages, how much will that average shift, perhaps not that much relative to your reference population, which is the 70 year olds, you have at the same time you start the intervention. So what you're trying to do is take an older looking or relatively older, 50 year old, and by the time they're 70, you'd like them to look like a relatively younger 70 year old. Another way of saying that is an old 50 year old, relatively old, 50 year old looks like a young 70 year old and is there an intervention that you can do that would move the person From this trajectory to this trajectory, not everybody wants to wait 20 years to know if their intervention will work,

or at least is promising enough to bring it to market. So here's a set of here's a study from Mike Snyder's lab at UCSF, is a geneticist, but has taken an interest in aging, and actually taking an interest in a lot of things. But here's the here's an interesting observation. Each one of these lines here is one person brought into the clinic and the lab to have set of metabolites essayed, some other functions essayed, and some clinical metrics taken, and then an aggregate score developed. So people have, of course, aging, their aging is a typic, in that, if you have 20 metabolites used, they won't all show the same pattern, if they are aging at the same apparent rate, they'll show different patterns. That's why you use the aggregate score, as you know. But what's intriguing here is that these are doctor's visits over a period of two and a half to three years, from which you can deduce, or derive a rate of aging, some appear to be aging more quickly, some appear to be aging more slowly, some might actually be for whatever they're doing, improving their relative age, over this three year time period. But what's encouraging about this is that in principle, you could be testing an intervention over two to three year time period. Now, that hasn't been demonstrated here. It's just taking what Snyder calls he types and analyzing them within a relatively short period of time. So that's encouraging, because that's actually the same timeframe as to do a study with laboratory mice. So here, you go with these interventions, and there's some what might be organizing principles. One of them is you have to accept that aging is heterogeneous. And that means both at the organ, tissue, Cell and Molecular levels, so you have a challenge they have, unfortunately, the computers are up to dealing with those challenges. And the mathematicians, and we're all actually capable of understanding these things through some of the dimensional reduction plots. But the other thing is, you could been by phenotype instead of chronological age. And that might actually allow you to amplify your signal, it might allow you to find out if you're doing an intervention that has a strong impact, the moderate impact or no impact. Or you could also look at the other end of the spectrum and ask, are you these people, you don't expect to benefit, but they might, they might skew this curve even farther toward the younger age, or for whatever reason, you might get a deleterious effect. And then again, of course, the question is, you really get the average, shifted very much from the reference to the intervention. So the other thing is you need the appropriate phenotypic assay. So there are lots of phenotypic assays out there. lifespan being the one that NIH has supported for the longest period of time. But there are all these tissues, those that you can sample directly or indirectly would be good to access. And you could also ask for what is common or what is unique. So you might be looking at liver muscle or white adipose tissue aging, for whatever reason you want to focus on that. And you could find that there are genes upregulated and downregulated, and your expression profiles and find that maybe there aren't that many that gives you a universal answer for what is aging, what have you done in terms of an intervention using this particular set. But there are other ways of going about it. And you can find things that are a common set as it were. That's useful. So the other thing about subpopulations is there is actually evidence for the idea that dividing a population might be beneficial. And this is from Mendenhall redrawing some work from Mendenhall.

The paper is here. And this is a situation where there's a marker for one of the hallmarks of aging in this case, it's pretty restasis and they have a Heat Shock Protein reporter and in the center have died is that's used here. Some have a brighter display and some have a lower display of this Heat Shock Protein reporter at baseline so it's not induced. It's just what did they bring you endogenously and then if you separate this population From this, you'll find that those that have the higher level of this seat Shock Protein reporter will age better. That is they have longer lifespans, and those that have a lower level have shorter lifespan. So in principle, segregating populations can be quite informative. Now, the question, of course, is when you're testing interventions, the granularity. So this is loss of proteostasis is one of the hallmarks of aging, and the loss of the proteostasis. Underlying that occurs in organelles. autophagy is involved, ATP, mediated protein degradation, all sorts of things that go into it, and various organelle levels. And you can actually ask, say, organelle, or function microscopically, and you can see differences in the quality of produce basis mechanism at the level of organelles. Or you could look at the very complex network that's underlying proteostasis. And you could ask, it's not gene by Gene, node by node in this network, using those as an assay if you had a reporter for a specific part of the network. So this whole spectrum is available in terms of what would be a useful phenotypic assay. So then we get to rejuvenation and this is, this is great stuff. I guess. So. You heard a talk by Irina convoy, who was really Irina and Mike when they were in Tom randos. Lab, as she pointed out, along with other collaborators really got the ball rolling on this, although the methodology has been around for more than 100 years. It's it's a, it's a method where they actually used an injury repair model. But in other situations, where it's not an injury repair model, people have also shown that when you can join a young mouse and an old mouse, that some of the features of the young mouse can appear to rejuvenate the old mouse. And contrary, some of the something in the old mouse pass through the circulation to the young mouse seems to produce accelerated aging. And of course, the question immediately is what is doing that. And there were some various suggestions and various ways of going about finding out what it is that's moved through the circulation. And you can also do as arena has developed heterochronic blood exchange, that's what this letter is letter stands for heterochronic, parabiosis, and heterochronic blood exchange, where here you can do all sorts of things to one of the mice and then take the blood and transfer it thinking that well, exercise improves all sorts of things, cognitive function, and so on. And in fact, Salvi ate his group at UCSF did an experiment, where they found that an enzyme with a long name involved in clinical lipid metabolism is elevated in exercise in the young mouse. And when that is given to the old mouse actually helps with cognitive function in the older mouse. It's very nice study. But we also know that caloric restriction has all sorts of effects. You could even take blood at different times of day where the circadian rhythms seem to have an impact and where the circadian rhythm is dysregulated. You may see accelerated aging, you can also do deleterious exposures and find out how if that is something that's transferred through the blood, chemical interventions, genetic modifications, all sorts of things are available. Of course, this has more options open because there's a limit to how much an old mouse can survive surgery.

So there are lots of ways of going about this. But one of the things that you can ask in this and other experimental paradigms about rejuvenation, what is necessary and sufficient to turn back the hands of time? That's a tough question. So there's more fibrosis. And sometimes you see that fibrosis is resolved in one of these experiments, or if it's not necessarily resolved, maybe there's less of it depends on the timeframe is the epi genome, reprogram something to Tom randos working on his DNA repaired so DNA damage is a driver of aging. So if you've rejuvenated is the DNA now repaired? damage is a driver of aging in multiple compartments of molecular compartments, has that damage now been removed or repaired in as part of the rejuvenation is complexity restored? Now I This question comes actually following the work of Louis lipsitz. In Boston, and he looks at cardia, he has looked at cardiac function along with gait, and range of things in older humans. And if you look at an electron cardiogram young, healthy adults, you'll see a very complex set of signals that come out of it. And if you look at an older person, much older person, a lot of that complexity has gone away for a range of physiological reasons, probably partly from increased stiffness in the heart, or problems with electrical conduction. But be that as it may, his observation is that there's less complexity with age, maybe more noise, but less complexity in his view. So his complexity restored, has the heterogeneity be really been reconfigured, we know from looking at single nucleotide and single cell RNA profiles, that the cells are quite heterogeneous, and you can find subsets of cells that are more or less in a particular class with age that may be linked to some of these aging phenotypes. Or maybe that's all still there, but there's an increased tolerance for it. And so the system is buffered against the inefficiency of every enzyme catalyzed reaction, DNA repair is not 100% efficient, but maybe for whatever reason, there's more tolerance for it, a barent messages might be cleared better alternative splicing, or splices ohms might suddenly be more efficient in producing the protein intended from the transcription of the DNA. So let's go back to this problem of undulating waves of protein with age. And let's actually do a little thought experiment. Let's say that we try to superimpose the waves of plasma proteins. Now, again, this is humans, but you can expect, I suppose the same thing in mice. And let's look at what was done in terms of the heterochronic parabiosis. The standard model is a mouse that's 24. That's the young and 20 months, that's the old this will tolerate the searcher sutures and the conjoining course the most for different sizes and different capacities. So you have to have some reserved capacity. But if you map this age relative to let's say, human age, so mouse might be 38 months and a person 90 years. So here's here's where you are. If If you make this analogy to the prevalence of disease now, it's not an exact analogy, but bear with me, please. So the question is, did we just we, the people who do the research, just get lucky. And so they found that these aging phenotypes were moved, even though you might be in an area where there aren't that many aged phenotypes to deal with? Or could you in some fashion, perhaps in heterochronic, blood exchange? Ask, what would you take what would happen if you took this kind of mouse that is the expression profile here is different from the four month old mouse, it's still before the advent of the disease states. But it's earlier and with a different

blood profile from the 20 month old mouse, or even more, so a 13 month old mouse were probably a lot of this stuff has happened. So we know that the diseases themselves affect the aging profile or the rate of aging. Maybe this has some information in it as well, that's different from the standard format, and might also be different from this format. So of course, you know, it's easy for me to say I'm not running a lab, I can ask anybody to do anything, and they don't have to pay any attention. But it's an interesting question. Have we just by gotten lucky with this experimental paradigm? And is there more to it that we have not yet leveraged? And of course, you could take another yet another intermediate state where there's some more disease, but it's not probably quite as bad as later aged mouse. But this for technical reasons, of course, might be better done through the blood exchange than the parabiosis. So let me just go briefly through a few more things. I asked about fibrosis before. Here's a study from Henry Jasper's lab. So he, he's looking as a candidate factor man, which is mesencephalic astrocyte derived neurotrophic factor And it seems to have an effect on cell death by as an inhibitor. So he's testing this. And this actually was supported by money that we gave to the lab is an award on request for support for heterochronic parabiosis. And he worked with solvated to do the actual experiments. And so if you join together the course there's the controls without the conjoining. So there's the the old isochronic pair bionic. And then there's the young and the old, and you see that there's some rejuvenation and less fibrosis. And if you do the math, heterozygous, so the level is much lower, or at least half lower. It's more like the old mouse in the heterochronic, in the parabiosis paradigm. So what's laid out here is a way of analyzing this, because you have the genetic, the genotype has been altered, at least in one gene. So if you look at all liver genes affected by aging, that's a young versus an old, without the conjoining. That's this set, then you have those that are rejuvenated by the heterochronic parabiosis. And that's this liver gene expression profile, sorry, these are gene expression data, relative to this pair. And so that's this subset of those affected in the liver, by old versus young. So here's your pool, from which you would expect the rejuvenation to be significant. And then the genes that still change in the manner heterozygote, those can be considered relatively independent of math, because you've reduced the signal from math. So then you have a set of other smaller set of genes to work with. And those genes provide clues as to the mechanism by which the fibrosis is reduced, or whatever other phenotypes you're looking for in the liver. So that's all about proteins. And as Tony was, Corey says, they look at proteins because it's convenient, it's it's something for which they have the technology. But here's something from the gentleman lab and an injury repair model, the previous one was just aging, not injury repair. So here's a fracture it we know that fractures in older people and older mammals in general. They, they heal not as well. And so Allman showed that there are macrophages that move into the fracture area. And they affect progenitor differentiation. So they affect a cell fate decision to become either fiber blastic or osteogenic. And, interestingly enough, using gene expression profiles, and some clues as to the molecular nature of these macrophages, which come and go, they appear in the wound site, have their effect and depart.

And they come from the young mouse and are moved into the old mouse through the circulation. So he's been able to identify but not publish, which are those macrophages, what is their level, relative and age, and what are their actual characteristics and their origin of origin and also their developmental origin. So I'm just showing the T's the plots as as of give you the flavor of the kind of thing he did, although this is differences between normal and tumor, and comes from an entirely unrelated study, it's just to illustrate this is a way of looking at complexity, with which you're all familiar and which Allman has used in what I hope will be published soon. But Allman also did this other thing. So when you're dealing with rejuvinating, in an acute situation, you might expect the timing is important. So we know if you're going to do a chemical intervention, or pharmacological intervention after heart attack, the sooner you get to the patient and give them the appropriate pharmaceuticals, the better off the patient will be, on the other hand work done in Germany by Alex cyren. And colleagues, where they made the same assumption about trying to do a cell therapy uncovered the interesting observation that if you actually wait a bit longer before you try to extract cells to use cell therapy to treat the heart attack patient, you're going to get a better result because what's happened is the injury has signaled to the bone marrow compartment, and the cells there are responding to that signal. So that takes a few days. So they just came by being alert, which is good science, they found that delaying actually was beneficial. In that case, you could do a two regimen, intervention with the pharmaceuticals and then try to get a longer term repair by using cell therapy. I haven't followed that recently. So I don't know how well it's worked out. But here's a situation where the timing for repairs understood in a molecular fashion. This is when signaling pathway as so many things are injury repair. And what happens is that there's a tendency towards the fibroblasts state and those progenitors in the older mouse. But if you activate the windset signaling pathway, which with login produced by the macrophage, then you can suppress the fibroblasts state and increase the osteoblasts state. But you have to get there at the right time. If you wait too long, you will not have the effect. So, here's another example of a cellular effector moving through the blood, this was bone marrow taken or bone marrow cell cells taken from a young animal injected then into the bone of an old animal. And what was rejuvenated was cognitive function, along with some activity, so the cognitive function was measured, as do these laboratory mice explore as a young mouse would a lot of the cage or as an old mouse would either just the old mouse, you're getting old bone marrow as a control, they tend to stay around the periphery. But if you give them the young bone marrow cells, then you rejuvenate the young phenotypes, there's a phenotypic screen that could be used, as well. So we just did another request for applications for heterochronic blood exchange. And we're going to make nine awards. This, I'm very pleased with this, it's our second round of such a request for applications. Unfortunately, the awards haven't gone out at this point. We are bureaucracy, and we are slow, unfortunately. But the money will get out there were those awards made, I could tell you who got them in for what. But if you go to reporter and another three months, you'll be able to see for yourself, selves, who got it and for what processes but just in general, what's intriguing about this set of applications and awards is that they are uncovering the origins of factors that will be anti Germanic, that is they will be rejuvenating. And they come from all over the body and affect all sorts of tissues, they affect cognitive function, they affect the blood brain barrier, they affect the brain vasculature, they affect the body, the peripheral vasculature, they come from bone, they have targets in bone, they come from muscle and have targets and muscle is actually remarkable what this kind of experimental paradigm is able to do. And I hope you're all suitably impressed with Irina's presentation, because her impact on the field is tremendous. So in summary,

some of the things that kind of bother me are, we have yet to really identify rate of aging, we have lots of tools out there, we have lots of things that are useful and being applied. And as long as we understand that they are being refined as we go along. That's very good. There's no reason to pause the pause button in any of this, all of this work on rate of aging and linear clocks, nonlinear clocks, all sorts of clocks, that really needs to go forward. Because as in all things in science, it's it's testing whether the hypothesis is supported by the data. But it's also useful to stratify research populations. And I think that's really it's happening. It's time has come if you pardon the phrase. And it's it's not that new, but it could be emphasized a bit more. And it's also I think, useful to think in terms of health expectancy. Based on what is health span, health expectancy might be a little more hopeful, if nothing else, and we actually do know that how you feel about your condition does have an impact on your treatment. And it's not just the placebo effect. It's actually epidemiological data and social researchers, I think have some pretty compelling evidence that that is the case. And also, you can leverage subpopulations to screen and validate interventions, that's probably not the worst thing to do. Also, in terms of these various displays, we know how to visualize aging better now than we did five years ago. But we also have available a lot of the actual visualization techniques through microscopy is of various sorts. And it is possible to train visualization techniques to look at that to identify differences in aging and I heard about this some years ago to QB Three presentation. And it's being done. And ni is supporting some studies in this direction, of course, you have to embrace heterogeneity, you don't have a choice because it's there. So I would just like to put in a plug, if I may, for the Nathan shock centers, we renewed this program, it's been going on, I don't know, 30 years, perhaps, maybe 25. We have some new Nathan shock centers. Now we managed to add to to the six that were there. There are multiple labs, they cover multiple topics in the biology of aging. They have all sorts of things now, imaging bulk and single cell height, all sorts of great technology that's out there. They work with laboratory animals, human populations, human materials, they have a lot of technology. And I'm sure that collaborations could be arranged with them, under the appropriate circumstances, either as the shock centers themselves, or the investigators who are there. And I can say that one of the leading shock centers is that Albert Einstein College of Medicine, Mumbai near Basel II. So if you wanted to find out what they are, here are his website. Allison has the slide set, you're welcome to it, you can do whatever you want with it, edit it, print it, use it for obscure purposes, where it's all yours. So thank you very much for sticking with me. And I'm happy to answer any questions you have.

Fantastic. Thank you so so much for joining. see many real clapping hands here. Thank you. Thank you. I think in the interest of time, I'll go immediately to add to a q&a. This was a fantastic array, I think of information. And I think the first one up, we have Keith.

Yes. Firstly, great talk, Ron, really impressive stuff. I have a question regarding the topic of scoring the results of interventions. You were mentioning earlier, how it's easier to you know, physically do things when you're healthier. And we know that exercise itself has a general protective effect. So it stands to reason that the eventual beneficial effects of a given intervention, say if the end results of team is making your biological age five years younger, by certain measures, for example, can actually be split into first order and second order causes right the biological effect of whatever the intervention was, and then maybe if that makes you feel more energized, and you exercise more, etc, etc, that itself will have an effect. So when it comes to trying to rank, which interventions are having the most significant first order effects? Is there currently any system of how we maybe normalize or tease out these two potentially compounding component of AI normalizing based on mobility changes or anything like that? Is the question Make sense?

Yes, it does. That's the sort of question that triallist would be better at answering. But I can speculate a little bit and perhaps your other trial is here would correct me if I can call near a trial list. And that is you can you you track what's going on. And you can you can look for the secondary effects, if they feed back, they could be treated as confounders of the primary outcome. And there are actually graphical ways of displaying that that will tell you if something is significant in terms of relating the intervention to the outcome, or if it's a modifier of the outcome.

Near for example, is that kind of analysis happening as part of the plan team trial to kind of normalize for the mobility increases and things like that? Well,

I absolutely look a third Oh, yeah, 25% of the budget is a grant from an NIH to take all those biomarkers and make sense to them, because what we're missing mainly is not having biomarkers but which one are changing with aging. So we don't have to do phase three trial, right, we can look for few weeks and few months and see if you were on the right track. And, and so yeah, and you're right, it's very, a you have to plan it very carefully, and do all the regression and linear analysis and everything in order to really understand what's going on. But But Keith, you raise a really important point, because what you would like to get from an intervention, if it's, let's say, a pharmacological as opposed to behavioral intervention, is that the intervention Well, we're from the economic point of view, you might want to keep giving the drug all the time or from the financial point of view. But from the point of view of a person's health or from his social impact of it, you'd like people to be able to do those things better, whether they need less of the intervention, because they're doing the other things that have the beneficial outcome. But that's, that's more of a personal point of view that you'd like to like that character and To Kill a Mockingbird, that scout read to all the time at the end trying to get off morphine. That's the same concept, we want to have some way of getting the person's independence increased, because that also has an effect on on.

Yeah, just just to clarify, for societal, it almost doesn't matter, you just care about the end result, right. But if, if most of the effect of an intervention happens to be it makes you feel much more energized, and you're much more mobile, that will have a limit at some point, you know, like, you know, that won't stop aging, eventually that benefit will run out, right. So we would at some point, as scientists want to understand what's going to have the first order effect as much as possible, and that it might be important to be able to guess, system to rank what's happening in which therapy should be prioritized based on that.

And I say this only slightly in a Cavalier way is, but it would be good if you could die happy.

Yeah, I think yeah, there's a Don't get me wrong, do everyday. results will can be confounded. I think we as users out if we want to get more information,

but but the tools that but the mathematical tools are there to do that teasing out.

Thank you. I want to ask you do you have to hop off right at the when the hour strikes? Nope. Okay, just checking.

Excuse me. Yeah, I don't want to go to my next meeting anyway.

Well, and do you have like five minutes more or?

No, no, my time is yours.

Oh, my God. Be careful what you wishing for.

Okay, that works. Okay.

And then after Aaron, I'm going to turn off the recording in case people have questions that don't shouldn't go on the record. Okay, Aaron, you go.

Sure. Glad to be the last final recorded question. Thank you for your talk. Dr. Galinsky. My question is within the NIH, who are the other people that you're working with who are championing this direction of research? And as a bonus question, are there any opponents and people who are trying to stymie this direction of research?

We're all in this together. We collaborate quite a lot. There are four divisions, geriatrics and clinical, gerontology, behavioral and social research, and division of neuroscience. We all work together. We also work with other NIH institutes. And pretty much every Institute now has is paying attention to aging as a parameter. And then there's also an NIH wide approach to including older people in clinical trials and clinical research. So as far as opposition, I wouldn't say there's opposition, but there is do caution. Because when you ask older people to participate, then you're asking a group of not everybody but some with some cognitive impairment. And then you have serious concerns about ethics because of what is informed consent, so I wouldn't say there's any opposition, I'd say In fact, there's a great deal of enthusiasm across the entire NIH enterprise, but it's it's moderated by the necessary caution.