NAD, Sirtuins and Aging - Leonard P. Guarente

2016 Roy Walford lecture: NAD, Sirtuins and Aging - Leonard P. Guarente, PhD David Geffen School of Medicine at UCLA From an accredited US medical school Learn how experts define health sources in a journal of the National Academy of Medicine 36,063 views Oct 24, 2016 Leonard P. Guarente, PhD Novartis, Professor of Biology Massachusetts Institutes of Technology Harvard University Chapters View all Show transcript David Geffen School of Medicine at UCLA 7.87K subscribers Videos About Created using YouTube Video Editor 20 Comments rongmaw lin Add a comment... @psiclops521 @psiclops521 4 years ago What I get from this is that, at this point in human history, mice must be healthier than ever in their existence due to the massive number of medical breakthroughs. 3 years ago Amazing what has evolved in biochemistry in the 47 years since my first biology class. Modern scientific miracles' of hope in these otherwise dark days. 3 years ago For very different reasons, NAD+ has also attracted a wave of attention from cancer researchers. Recent studies suggest that cancer cells of many types depend on NAD+ to sustain their rapid growth and that cutting off the NAD+ supply could be an effective strategy for killing certain cancers. The data from these studies paint a more complicated picture of NAD+ and raise new questions about the diverse ways taking an NAD+-boosting supplement might influence health. “It might still slow down the aging part, but it might fuel the cancer part,” says Versha Banerji, a clinician-scientist at the University of Manitoba. “We just need to figure out more about the biology of both of those processes, to figure out how we can make people age well and also not get cancer.” This is just a fellow up to my previous post. Interesting that most sites are ignoring the possibilities of its effects on cancer. 6 Reply 2 replies @abdullayt588 @abdullayt588 2 years ago wow 1 Reply @rui-9-cs315 @rui-9-cs315 3 years ago 💕 1 Reply @rezarabbi7419 @rezarabbi7419 2 years ago Nice 1 Reply @antMight @antMight 1 year ago Super, thank you for a video. I’m passionate about aging and supplements to do it possible 🙂 Reply @fondrees @fondrees 3 years ago What does this have to do with roy walford? 1 Reply @biogerontology7646 @biogerontology7646 2 years ago 26:35 organoid colonies Reply 1 reply @ganje3869 @ganje3869 4 years ago Looks like it works hand is hand with the endocannabanoid system. 1 Reply @sheafisher3578 @sheafisher3578 5 years ago So now the whole pro-SIRT/anti-mTOR view is complicated by the intervening deacetylation of S6K1. 2 Reply 2 replies @petercoderch589 @petercoderch589 2 years ago Audio so low couldn't understand anything. Transcript Follow along using the transcript. Introduction 0:14 thank you very much for that kind introduction and it's a real pleasure to 0:20 be here at UCLA so what I want to do today is give a pretty broad 0:27 introduction into the topic for those of you who might not have followed it closely over the years and then talk 0:35 about some work in the lab that has been focused on adult stem cells in mice and 0:42 if there's time at the end tell you a little bit about a new project were 0:48 undertaking to try to study Aging in humans so this work began about 25 years Early work 0:56 ago in the lab and it was driven by two new graduate students at the time Brian 1:02 Kennedy and Nick Austria Co Brian is now the president and CEO of the buck 1:08 Institute Nevada California and Nick has gone on to do other things in life but 1:17 both were a very creative and driven 1:23 students and they really made this early work lead to something interesting and 1:28 what we were studying in those days was aging in budding yeast which is what my 1:34 lab worked on at the time and we chose the system because we thought it was tractable and we thought it was 1:41 interesting we certainly didn't think we were gonna find anything generalizable about it but you know whatever you 1:50 choose to work on it should be something that is doable so that's why we chose it 1:55 and the way this works in budding yeast is that the daughter cell is new the 2:00 mother cell is old and what you do is follow the this mother cell in the 2:06 microscope and push away the daughter every cell division and simply count how many times the mother 2:12 so in the file and this was a system that was well established and it was 2:18 known that mother cells could divide roughly twenty generations or so and then they semester and they semester Model of aging 2:25 with a characteristic phenotype they were enlarged sterile alter wrinkled so 2:32 that was the model of aging that we were working with and we spent about five 2:38 years trying to identify genes that control this process such that if we 2:46 mutated the gene it would change the lifespan as defined by this metric in a 2:53 significant way and the most interesting findings that came out of these studies was the identification of a gene called 3:01 surtout as an anti aging gene in yeast so this is a survival assay of mother 3:07 cells and you can see for the wild type strain the number of generations is on 3:14 the x axis here and there's this precipitous decline in viability at 3:19 about twenty generations or so in this case actually about 30 generations for this strain and the importance of 3:26 surtout is illustrated by the fact that when the gene is ablated that lifespan 3:32 is shortened and when the gene is overexpressed here two copies instead of 3:38 one the lifespan is longer than the wild-type control strain so that sort of 3:45 focused our attention on this gene and we continued working on it in yeast and Sirtuins 3:53 paper only four years ago from another lab jazz Winsky lab did a study that was 3:59 surveying all the genes in the genome for importance and dictating lifespan 4:05 and the top of their list was Sir - so we've sort of stumbled into an important 4:11 gene at least as far as east-asian is concerned now in the years since a lot 4:17 of studies have been carried out in other systems including the Ron worm flies and mice 4:23 and in these cases as well it turns out again I would not have predicted this or 4:29 expected it but it turns out that sir to homologues and they've been called 4:34 sirtuins can extend lifespan when their 4:40 activity is increased in these systems so that means sirtuins have a broad 4:46 effect to slow down aging and to extend lifespan so we were interested very 4:52 early on then in what sir - did okay and 4:57 its history comes from something called silencing in yeast and actually Mike 5:06 grunts team here was one of the players who identified the importance of histones in silencing and that 5:15 importance is that the amino-terminal tails of histones h3 and h4 have lysines 5:23 that tend to be Isetta lated on the epsilon amino group for active chromatin 5:29 and are deacetylated when the chromatin is silenced and surtout is somehow was 5:37 known to be involved in this silencing process so that's adjusted right away 5:43 that sir - could be a protein or histone deacetylase but many attempts to show 5:48 that in multiple laboratories failed so we started working on that the 5:56 biochemistry of sir tube and we purified the yeast sir - protein and the 6:02 mammalian ortholog of sir - and we're studying it for several years and came 6:09 upon this activity which is an nad dependent deacetylation of targets and 6:15 so what this means is that without the nad cofactor sir - and sirtuins in 6:22 general are inactive and nad is actually a co substrate and is split every 6:28 reaction cycle concomitant with the acceleration of the substrates for 6:33 sirtuins now those substrates and be histones and in yeast the 6:39 histones are certainly critical substrates of sir - but in mammalian cells in addition to histones substrates 6:46 include other proteins in other certain multiple cellular compartments that can 6:53 be deacetylated and that are important physiologically so Shin Amon my postdoc 6:59 and I were excited about this finding because it provided a link between protein acetylation metabolism and 7:07 epigenetics it also had implications for aging because it gave a possible 7:13 explanation for how a low calorie diet or calorie restriction might extend 7:19 lifespan and slow aging and we proposed that it worked by activating sirtuins 7:24 and I think that hypothesis has proven to be robust as I'll get to in a few 7:30 minutes so with respect to a settle Asian then this was the third and final I think 7:38 class of enzymes shown to influence the protein of settle Asian level in cells 7:43 the first was the histone acetyltransferases identified by David Allis second was the H stack proteins 7:50 which are deacetylases identified by Schreiber and colleagues and these have no cofactor requirements and the third 7:57 are the nad dependent deacetylases or the search joints so in mammals and my Mammalian sirtuins 8:07 talk today will focus on mammals there are seven homologues of the yeasts or 8:12 two which is shown here on this phylogenetic tree the closest is called certain one shown here and I'll talk 8:21 mostly about 31 today and I in many senses this is the most important of the 8:26 seven mammalian sirtuins and 31 along with 36 and seven are nuclear proteins 8:34 31 tends to be in the nucleus deacetylated many targets off of the 8:41 chromatin and thirty six and seven tend to be more closely associated with the 8:46 chromatin suggesting that histones might be particularly important substrates for them now what's 8:52 interesting is that another three of these proteins 34 5 & 3 are encoded in 9:01 the nucleus but target to mitochondria so three of the seven sirtuins are reside in the mitochondria and 32 is 9:09 cytosol ik so is there a rhyme or reason to why there are seven of these proteins 9:17 are they redundant no they're not redundant for one thing they reside in different cellular compartments one way Sirtuins and mitochondria 9:26 to view this is to look at sirtuins with this color key here and proteins that 9:34 they're known to do settle a and many tissues have substrates for sirtuin 9:39 proteins shown here I'm not gonna go through this all but let me just point out P G c1 alpha as an example because 9:46 this is a protein that's deacetylated by sir T 1 and which when deacetylated can 9:51 drive mitochondrial biogenesis and mitochondrial function and in fact if 9:57 you go through this list one of the themes that emerges is that what 10:02 sirtuins do in a coordinated fashion is they drive oxidative metabolism in 10:09 mitochondria okay and so they force the 10:15 cell to produce ATP via electron transport in mitochondria now 10:20 concomitant with that they drive stress resistance to reactive oxygen species so 10:27 enzymes like superoxide dismutase are up regulated by particularly 33 in 10:34 mitochondria they also drive pathways that would be 10:39 consistent with oxidative metabolism such as oxidation of fatty acids inside of mitochondria so I think that's a kind 10:46 of an overarching principle and note that oxidative metabolism would be 10:52 particularly important and the conditions when the fuel source is limiting because of the efficiency with 10:58 which ATP is produced by mitochondria tambul ISM so what about this idea that 11:05 sirtuins are necessary for the effects of calorie restriction well there's a 11:11 pretty vast literature that is in the hundreds of papers I will choose one example here because it's particularly Hearing loss in mice 11:17 clear-cut and the subject here is hearing loss in mice and calorie 11:25 restriction so what these authors did in this study is to examine hearing loss in 11:31 12 month old mice on the control diet shown here so what you can see is that 11:37 it takes a much higher decibel level of sound to elicit a response in the old 11:45 mice over a broad range of frequencies shown here they're not even that old 11:51 these mice will live past 2 years so this is kind of midlife and already their hearing is severely damaged and 11:58 it's damaged due to oxidative destruction of hair cells in the cochlea 12:05 of the inner ear of these animals now what the study shows is that if the mice 12:11 are aged matched to the 12 month old but on calorie restriction they're completely protected against hearing 12:18 loss so that's good that's kind of in line with what calorie restriction does it slows down the ravages of aging in 12:27 many ways and the point of the slide is 12:32 to show what happens in a serie knockout mouse now note that this mouse has 12:38 almost no phenotype - sort of casual analysis it's just like the wild-type 12:45 but if you look at it in this assay you can see that the knockout mice is completely refractory the calorie 12:52 restriction calorie restriction doesn't work in this mice in this mouse and hearing loss is observed so that's one 13:02 point about the sirtuins oxidative metabolism especially under conditions of limiting fuel sources and 13:08 one reason that might slow aging again is because of beefing up resistance 13:15 to oxidative stress and oxidative damage in cells now an unexpected feature NAD and Circadian Clock 13:23 emerged more recently and that is that nad and soar to ins are intimately 13:30 connected to the circadian clock and this was a field that started about six 13:37 seven years ago and the way this works is the following so this is the oscillator that is present in the brain 13:45 but also in peripheral tissues that oscillates with a diurnal cycle and the 13:51 nature of the oscillation is that this is a transcription factor called bmail 13:57 and clock a dimer and this transcription factor drives the expression of these 14:03 feedback regulators here which build up and inhibit the transcription factor and 14:10 this occurs with a site and when the transcription factor is inhibited then 14:16 the cycle is reset and this oscillates with a period of about 24 hours now this 14:23 wouldn't do anything to metabolism at all if it couldn't reach out and affect 14:29 proteins that are not a part of the oscillator and what these papers show is 14:34 that one of the critical output genes that's not a part of the oscillator but that's regulated by B Mel and clock is a 14:42 gene encoding this enzyme in a.m. PT and this is the rate-limiting enzyme for nad 14:48 synthesis and so what that means is B Mel and clock oscillate and therefore 14:54 expression of an amputee will oscillate and therefore any D synthesis itself is 15:00 circadian and it peaks at the right time and that means sirtuins our circadian in 15:07 their activity and I won't go through this but there are also more connections 15:13 between 31 and 36 and the components of the clock itself so there's this really 15:18 close connection and I think that part of the reason for this is that metabolism needs to be staged in a 15:25 temporal way and if it's not things go awry and there are many instances of in mice if you 15:33 damage genes and coding components of the oscillator the mice have bad health 15:39 in humans there's a considerable data that people who live life styles for example shift workers that mess up their 15:45 circadian clocks are not as healthy much less healthy so that's another dimension 15:51 to this is that there's a considerable input into this system that makes 15:57 metabolism circadian now what about diseases and again I'm still in this NAD and Diseases 16:04 introduction part in rodent models multiple sirtuins have been shown to be 16:10 protective against a series of diseases cancer diabetes neurodegenerative cardiovascular osteoporosis kidney 16:17 disease information and the list is longer in fact so let me just say a few words about cancer where three of the 16:25 sirtuins t3 t4 which are mitochondrial and t6 which is nuclear have been found 16:32 to be lost in roughly 20 to 40 percent of human tumors so that means their 16:37 tumor suppressors and how did they suppress how do these protein suppress 16:43 cancer and one of the major mechanisms I think's I think relates to cancer 16:49 metabolism and particularly to the propensity of cancer cells to one a 16:55 favor glycolysis which is the Warburg effect but also glutamine utilization 17:01 and so the connection then to the sirtuins is shown on the left part of 17:07 this slide so certain six represses glycolysis and also ribosomal biogenesis 17:14 and those are two pathways that cancer cells need in order to grow in the 17:21 mitochondria 34 suppresses this critical enzyme for glutamate utilization 17:27 glutamate dehydrogenase and thus represses glutamine metabolism and 33 17:34 shown here represses several enzymes again this is by deacetylation that are 17:42 involved in our OS production so cert III suppresses production of reactive oxygen species so when any one of these 17:49 three proteins is lost then the cells metabolism is more geared towards tumor 17:54 metabolism favoring cancer now on top of 18:00 that 31 and 36 are intimately involved in DNA repair that includes both single 18:06 strand break repair and double strand break repair and I'm not going to go through this in detail but it includes 18:12 repair at telomeres it includes non-homologous end joining and 18:19 homologous recombination so again the removal of one of these two sirtuins 18:25 would create higher DNA damage which again could favor a tumor tumorigenic 18:32 phenotype now the last thing I want to NAD and Aging 18:38 tell you in this introduction is something that was found still more recently and that is nad is depleted 18:45 with normal aging so this is a study that was done in mice so basically the data is very clear-cut in Ron worms in 18:52 mice and emerging data exists in humans and what's shown here is young versus 18:59 old mice and any D levels first simply in young versus old liver and over here 19:07 is muscle and there's about a two-fold reduction in nad levels do you say that's not very much is that enough to 19:14 do anything and this is a block of PG suin alpha a set elation in muscle of 19:20 young versus old mice and what this shows is in the old mice we just want 19:26 alpha is hyper satellited and the deacetylase that works on 19:31 peaches when alpha is certain so this implies that this reduction in nad of 19:38 twofold is enough to deactivate thirteen one in these old mice now I'm gonna move 19:44 on we could talk about why nad declines and someone wants to ask a question later I'll be happy to tell you what's 19:51 known about that but it does decline and the real exciting part is that this is 19:57 actionable and the reason it's actionable is you can replenish the lost and a bee by feeding the animal nad 20:04 precursors and again there's a whole series of papers that show this and again I want to credit Qin mi who first 20:12 noticed this and what he noticed is that in thirty-one transgenic mouse the mouse 20:19 had a phenotype okay which was in improved glucose tolerance but when the 20:24 mouse got old it lost a phenotype okay but the protein was still overexpressed so he wondered why that could be and 20:31 what he hit upon is nad levels were plummeting in the old mice and if he 20:38 supplied nad in the form of this nad precursor called nicotinamide mononucleotide he could restore any D 20:45 levels and restore the phenotype to that Mouse now it turns out that just wild-type mice when supplied with 20:52 nicotine I'm on a nucleotide or nicotinamide rabbit side and our have an 20:58 improvement in their health for example they're more resistant to becoming diabetic as 21:04 old mice there have been other health benefits described so this then leads to 21:10 a notion that therapeutically one can think in terms of activating certain 21:15 ones for health in two ways one would be nad replenishment as I just mentioned 21:21 and the other way is something that was found still earlier by David Sinclair 21:27 and colleagues that there are certain small molecules that can activate thirty-one such as resveratrol and more 21:33 recently compounds that are being developed at GSK that will be in the clinic in about a year by binding to an 21:41 allosteric domain in the protein and these compounds the stacks will target 21:47 31 but nad replenishment will replenish the activity of all the surgeons 21:55 okay so that concludes the introduction went a little bit long so I'm gonna have to speed up so I'm going to talk about Adult Stem Cells 22:02 adult stem cells and the stem cells are these intestinal stem cells so this is 22:08 how the gut is organized its these finger-like structures called villi and the base of the villi switch villi which 22:16 are called crypts and the stem cells are at specific positions within these crypts and they can divide to give more 22:23 of themselves or to differentiate to give all of the differentiated cell types of the gut which move up the villi 22:30 and Arce left off very rapidly it's a very rapidly dividing tissue now the 22:36 point of departure in this analysis for us was a paper published by illness and Sabbatini a few years ago and they were 22:44 studying the interaction between the intestinal stem cell and a niche cell 22:49 called the Paneth cell and what they found is that calorie restriction 22:55 repressed the activity of tor and in particular a complex of Tor called M 23:00 torque one and that led to the Paneth cells synthesizing this small molecule cyclic adp-ribose which was then sensed 23:09 by the stem cell to promote self renewal in the stem cell so that means calorie 23:15 restriction gave rise to an increase in stem cells so the study carried out by 23:22 masaki Igarashi in my lab then utilizes mice that are either wild-type 4:31 in 23:29 the gut shown here or overexpress 31 shown here are knocked out in the gut 23:36 431 shown there and what we did is we looked at the effect of calorie 23:42 restriction on all of these mice and tried to piece together what was 23:48 happening particularly in the stem cells so here's an introductory slide that 23:54 shows a cross-section of the gut in wild-type mice ad libitum versus calorie restriction and as young 24:02 man saw we saw an increase in the size of the crypt 24:07 due to an increase in the number of stem cells shown here with when this was done 24:13 in a villain cream knockout mouse that specifically knocked out in the gut you don't see this increase the date is 24:19 shown here for wild-type crypt size increases and the mecha doesn't increase 24:27 now the key experiment is this one here where we're staining for this marker of 24:33 stem cells to actually get a count on the number of stem cells and what you 24:38 can see is that there's an increase in the number of stem cells in calorie restriction just shown here and that 24:44 does not occur in the knockout mouse we wanted to have a backup way to assess 24:52 this so we use the mouse now where the stem cells are marked with a GFP driven by a stem cell spell at the cell 24:58 specific promoter LGR five shown here and again calorie restriction gives rise 25:04 to an increase in the number of stem cells and if you knock out thirty-one in 25:10 the stem cells this time with tamoxifen you don't get that increase so that backs up the histology next we looked at 25:19 the overexpressing mice that have higher levels of certainty one in the gut and these mice now are fed ad libitum 25:25 they're not calorie restricted and what you can see is the changes that normally are induced by calorie restriction 25:31 they're induced simply by over expressing 31 so this mimics the calorie restriction phenotype you get the 25:38 increase in the crypt shown here and you get the increase in the number of stem cells per crypt shown there so this is 25:46 very nice it says both lost and gain-of-function mutations in 31 show that this protein 25:54 is necessary and its activation is sufficient for the response of is cs2 calorie restriction in vivo so the in 26:01 vivo is I think a pretty good foundation so we want to know what are the relevant Organoids 26:08 cells for this is it the stem cells or is at the Paneth cells and one way to 26:13 approach this is to purify the stem cells in the panas cells which can be done 26:19 by marking them appropriately and then sorting them in the cell sorter so you 26:25 have now a pure population of stem cells and a pure population of Panna cells and 26:30 then you can cultivate them in this matrix and they give rise to these 26:35 colonies called organoid colonies I'll show you a picture in a second this is developing cleavers lab and what 26:41 you can do then is add a define member of cells of one type or both to a 26:49 microtiter well and count the number of these organs that are formed so these 26:56 are the organoids shown here and they are seeded by a stem cell so a stem cell 27:03 initiates the formation of one of these colonies and then they differentiate to give all the cell types of the gut and 27:09 they're thought to be an ex-vivo assay for the function of is CS okay so let's 27:16 look at an experiment if you co-culture stem cells and Paneth cells and the 27:23 painted cells come from an ad libitum fed mouse you get this number of organize but if the Paneth cells are 27:30 from a calorie restricted Mouse you see an increase this is what was observed in the young mass papper okay 27:36 this was the bedrock assay the calorie restriction is activating the panas cells now you can ask the experiment 27:43 well what if the stem cells but not the Parana cells stem cells come from a knockout Mouse no response okay so that 27:51 says you need thirteen one in the stem cells in order to get the response to the calorie-restricted Paneth cells however if the Paneth cells come from a 28:00 knockout Mouse they work just fine so thirteen one is required in the IOC's 28:05 but not the Paneth cells to get the effect of calorie restriction so we Cyclic ADPribose 28:12 began then to focus on the stem cell and to piece together the pathway of signaling and again I'll go through this 28:18 briefly since it was published a month or two ago now cyclic adp-ribose the 28:23 signal from the pana cell is known to stimulate calcium release in the target cell so we looked for many 28:30 pathways that were calcium-dependent and we found that cam kinase kinase was 28:36 important in this signaling pathway here's an example of that this is mixing stem cells and Paneth cells this is the 28:43 effect of calorie restriction here and if you use a specific inhibitor of cam 28:48 kinase kinase you block that effect and a nice control to this experiment is 28:54 instead of using calorie restriction to drive the increase in organoid colonies 29:01 you could use the sir to overexpressing mouse right remember that mimics calorie restriction and when you use that mouse 29:07 then the inhibitor doesn't do anything and so what that means is that the way 29:13 we interpret this is cam kinase kinase then it's downstream of calcium signaling and see one is downstream of 29:20 of that and if you have a gain of function mutation in 31 it's now independent of upstream signaling and 29:27 you get the increase in organoids the other nice thing about cam kinase kinase 29:33 is one of its known targets it's a MP kinase which is known to respond to 29:38 calorie restriction in many settings so we took a look at a MP kinase in the MP kinase 29:45 Crips and you can see that phosphorylation which equates with activation of the kinase is induced by 29:52 calorie restriction shown here and it's induced in the Cir 231 overexpressing 29:58 Mouse shown here and furthermore if you use an inhibitor of MP kinase then this 30:04 increase by the Paneth cells in calorie restriction is blocked and if you use a 30:10 car which is an activator of AMP you kind of it's sufficient to give the increase so this I think firmly places 30:17 EMP kinase in this pathway and lastly AM 30:22 P kinase is known to regulate expression of our friend here in a.m. Pt the 30:28 rate-limiting enzyme for nad synthesis and in fact calorie restriction shows a two-fold induction of n a.m. PT 30:36 so we can now trace a pathway that looks like this from in the stem cell from cam 30:42 kinase kinase through a and P kinase dam PT nad and thirty-one activation that's 30:50 what we believe and the question then is what's downstream of this so remember the output of calorie restriction is to 30:57 expand the number of stem cells so we looked at known pro-growth pathways such 31:04 as pi3 kinase Akt or beta catenin and both of these in other contexts for 31:10 example colon cancer play an important role in driving growth but they're not 31:16 in play in the case of calorie restriction we also looked at M torque 31:21 one okay and in particular the signaling from M torque 1 to phosphorylate s6 31:27 kinase and to phosphorylate s 6 to drive protein synthesis and cell growth and we 31:33 found that bingo this was the guy and it's very surprising result because I 31:38 told you earlier and I'll come back to this that M torque 1 is canonically repressed by calorie restriction and 31:45 here it's up regulated so is this true let me show you some of the evidence for that Evidence for NAD 31:51 ok this is looking at the relevant phosphorylation event and s6 kinase ad 31:57 libitum or calorie restricted mice looking at crypts this is the same experiment within test 6 again this is 32:04 with crypts we did the same thing with purified stem cells now ok this is now 32:12 immuno staining experiment looking at phosphorylation of s6 wild-type calorie 32:19 restriction let's see the increase is very evident shown here and from a knockout Mouse you don't see this 32:25 increase you can also see it and this is a little bit heroic in a Western blot on 32:31 isolated stem cells calorie restriction gives an increase in Essex kinase 32:36 phosphorylation at the relevant phosphate but not at an on tour regulated serine s6 same thing now 32:46 what's really interesting is another substrate of em torque one which is called for EBP it's not affected by calorie restriction 32:54 so somehow the activity of M torque one is parsed so the calorie restriction is 32:59 giving an increase in phosphorylation of one substrate s6k but not another 33:04 substrate for ATP how is that possible and we really were befuddled by this for 33:12 some time then a paper came along shown here that showed that there was 33:18 crosstalk between 31 and s6k and tour and with the shown in the paper 33:26 is that 31 actually dear satellites s6k and that makes s6k a better substrate 33:33 for M torque 1 so we wanted to see if that mechanism was in play and it is if 33:40 we look at the effect of calorie restriction shown here in Crips we can 33:46 see it leads to the deacetylation of s6k and that requires 31 not shown here also 33:54 if we look at the overexpressing mouse for 31 it leads to the deacetylation of s6k so what we believe is that calorie 34:02 restriction induces 31 deacetylation of s6k this residue and that makes it a 34:09 better substrate for M torque 1 and that's why the signaling through that part of the pathway increases so here's 34:16 what it looks like then shown here 31 is up regulated by these upstream steps I 34:22 mentioned deacetylase s6k and this leads to protein synthesis and 34:29 expansion of the is CS now we also believe and I won't show data for this but we could talk about it is that the 34:35 actual nutrient sensing here is done by the Paneth cell which is what sense is that calories are low and this stem cell 34:45 then is entrained by the pan itself via cyclic adp-ribose and M torque 1 which 34:51 normally would be suppressed in calorie restriction is not suppressed it stays the same but it's output is increased 34:58 because of Sir T 1 activation and the acetylation of s6k to make it a 35:04 better substrate for tour now what about functional debt is this true or is this just fantasy so if this were true than 35:13 blocking these downstream steps in the tour pathway should have real consequences okay so here's our 35:20 experiment with organoids again the control the calorie restricted pan is 35:26 cell give the increase organoids and that's blocked by an inhibitor of s6k 35:31 and it's blocked by rapamycin the famous inhibitor of M torque 1 what about in 35:39 vivo so we've done multiple runs of this experiment in which we calorie restrict 35:47 the mice and either treat them mark or with rapamycin and ask what is the 35:52 effect of rapamycin now note that prior to this study one would have assumed 35:58 that rapamycin itself would be a mimic of calorie restriction because that's 36:03 how it was thought to act and does act in many cases but let's see what happens 36:08 in this case ok so this is the data here here's the tabulation ok this is the 36:16 sized calorie restriction there's an increase blocked by rapamycin shown there and 36:22 here's the bottom line you count the stem cells and color restriction 36:28 increase and it's blocked by dosing the mice with rapamycin so this is 36:35 interesting because 31 and M torque 1 are usually working in opposite 36:41 directions so the genetic says that 3 T 1 up regulation slows aging and Fort or 36:47 down regulation slows aging during aging shirt C 1 activity is thought to 36:52 decrease which is bad and torque activity to increase which is bad and 36:59 calorie restriction does the opposite straight C one activity increases which is good and torque 1 decreases which is 37:05 good but I just showed you that's not true in the is CS and a therapeutic then that 37:12 would mimic calorie restriction would be something that would activate 31 the nad precursor or the 37:17 stacks or would inhibit tor rapamycin but one would predict that that would 37:23 not be a good thing for these stem cells and for the gut okay now what about Aging 37:29 aging so I really had told you so far relates to diet okay there's studies were all done with young mice what's the 37:36 effect of normal aging on the stem cells in the gut so we did that study and 37:42 here's the data so I'm just showing you the data on counting the stem cells shown here and young mice versus old 37:49 mice and the number of stem cells goes down from about five per crypt to about 37:55 three per crap so that's consistent with what we think about adult stem cells 38:02 that they decrease in number with aging now interestingly if we look at by 38:09 immunostaining or by Western that's signaling through the pathway I just described you so in the pathway I 38:15 described to you relates to calorie restriction that may or may not play a role in normal aging we don't know we 38:20 wanted a query whether it did and in fact if you look at a six staining with 38:26 aging it goes down and the old versus the young gut and by Western you can see 38:32 this clearly phosphorylation of s6k it's down phosphorylation of s6 is down 38:39 and importantly the levels of Nan PT go down with aging so that's the stem cells 38:49 what about the Paneth cells did the same thing young versus old mice and 38:54 surprisingly we did not see any change in the number of Paneth cells and the old mice stayed the same 39:01 no decrease we could also look at the function of 39:06 the stem cells in the pen itself by the organoid assay ex vivo and I just look 39:12 here if we mix together young stem cells or stem cells from young mice with 39:18 Paneth cells from young mice you get this if you mix old old you get that 39:24 there's clear deficit okay but here's the revealing part if you mix old stem 39:29 cells from old with Paneth cells from young deficit but if the stem cells are 39:37 young and the Paneth cells are old no problem so by this functional test as well the 39:42 problem with aging is in the stem cells so the stem cells seem to be taking the hit and not the pan of cells and is this Is this reversible 39:55 reversible so we looked by the organoid assay again young versus old there's a 40:03 deficit and the organoid formation this is just the stem cells now and if we 40:08 supplement with the nad precursor NR that's rescue of all okay so that's consistent with the idea that an amputee 40:14 is down and the everything downstream of that and the pathway is gonna be down so 40:20 NR can rescue that furthermore that rescue by NR is blocked by rapamycin the 40:29 inhibitor of M torque 1 and it's blocked by ex5 2 7 which is an inhibitor of Surti one do the same experiment with 40:37 now this is with Crips with purified stem cells same result and our rescues 40:43 the old purified stem cells and that's blocked by rapamycin what about in vivo In vivo study 40:51 study here was then to take young versus old mice and let's just look at this which is the stem cell number just sort 40:58 of the bottom line and old you see a decrease in the number of stem cells I 41:04 showed you that earlier and that's rescue able by NR in vivo which is 41:09 really surprising to me that you could actually rescue that but I think nice 41:16 so the model then for aging now is that the same pathway is in play here and 41:23 that aging leads to a deficit that we can trace at least as far up as nan PT 41:28 here which means that NR can come in at this level and bypass the deficit and 41:34 the deficit leads to a decrease in mTOR Qin signalling and a decrease in ISC 41:40 number and function with aging okay 41:46 so again if we return to the canonical 41:52 view the genetics what happens during aging in calorie restriction therapeutic we can see that in every case what 41:59 happens in the is sees is the opposite of what one would predict by the canonical view of M torque one so 42:08 possible that the ISEs are the only exception to the canonical view or it's 42:13 possible there are more exceptions that are yet to be uncovered 42:18 so to summarize then this part the aging part of the talk is that the wiring of Aging is complex 42:25 pathway is important in agent can be cell type-specific and I just showed you that the stem 42:30 cells seem to be wired differently and the implication is a drug to treat aging globally may affect different tissues 42:38 differently and even oppositely and mammalian aging is complex okay so 42:49 there's a little time so I'm just gonna take a few minutes to shift gears and talk about a new project what I'm 42:57 interested in these days not yeast so much anymore or worms or even mice but 43:04 I'm interested in aging in humans and part of that interest is of everything 43:10 that's been found in the model systems how much of that would carry over to humans if any and the other part is is 43:18 there a way to directly interrogate humans to learn about aging so this is a study Study 43:26 to study Aging in the human brain that is in its early stages being led by a 43:33 postdoc in the lab Kristin Glorioso and as a collaboration and it's a 43:38 collaboration with computer programmers okay which include andreas pfennig and 43:44 Manolis Kellis MIT and other 43:49 collaborators that supply us with tissue and data and these are centers that have 43:57 collected banks of human brains okay which is what you need to do a study of 44:04 this sort and I'll note that these first two banks are disease-free 44:11 so quote-unquote brains and this cohort of David Bennett is weighed towards 44:19 Alzheimer's disease some Parkinson's disease but also some control disease free so here's an example of the cohorts 44:31 and the age distribution the Pittsburgh cohort has a nice distribution many 44:39 different ages here NIH pretty nice Ross map the disease cohort is very very 44:48 narrow age distribution skewed towards old and gtex I'm not going to say anything about today so the region we're 44:55 looking at then in this study is the prefrontal cortex and it's a particular 45:01 region of PFC called BA 47 and the 45:07 reason for that is that the brain is so heterogeneous with so many compartments 45:12 that you have to focus in on one specific region otherwise it's almost 45:18 impossible so what we do then is it's very simple we look at genes that change 45:25 as a function of age and of course there are genes that go down like this be TNF 45:32 being a pretty important protein for brain health and that go up 45:39 such as chief I'll show them there and identify these aging regulated genes of 45:46 which there's roughly a thousand or so in the Pittsburgh cohort and most of 45:52 those overlap with similar changes in the other cohort okay so that's good and 45:59 we use this data then to develop a metric of what we call the molecular age of the brain by and I'm not I don't have 46:08 time to go through that but basically we try to break the data down into principal components and throw out the 46:14 principal components that seem not relevant to aging and way the principal 46:21 components that seem relevant to aging and develop this metric so what does it 46:27 look like then if we let's say compare the molecular age of the brain as 46:33 defined in this way with the chronological age of the subject at the time of death okay so here's that data Delta Age 46:41 for the Pittsburgh cohort molecular versus chronological age molecular here 46:46 chronological there and as you'd expect there's a pretty good correlation okay 46:52 but what we wanted is what you see a correlation that's good but not perfect 46:59 and the reason is we want to take for each data point each data point here is 47:05 a brain the difference from the regression line shown here and the 47:12 actual data point to define as a variable called Delta age okay which is 47:18 a molecular age minus the chronological age and that tells you how different that brain is from the average so 47:24 something above the line here is a brain that's older than the average and below 47:30 the line is younger than the average and some of them are older or younger by a fair amount it's a continuous variable 47:37 that applies to every single brain and we take that as a proxy for the aging 47:43 rate in these brains fully aware we're not measuring rate because this is not longitudinal this is 47:50 all cross-sectional nonetheless we take this measure of Delta EIJ as a proxy for 47:56 the aging rate and our goal long-term is to find identify genes and pathways that 48:03 determine this aging rate and determine why this brain is aging fast this brain 48:09 is aging slow okay but what I'll just give you a little tiny vignette today is 48:15 how this data this proxy for aging rate 48:21 what it would tell us if we compared a cohort that had Alzheimer's or 48:26 Parkinson's brains versus disease-free control brains what were we see and let 48:36 me call your attention here so this is from the Ross map cohort and these are 48:42 the control brains here so on the y-axis is Delta age - means molecular ly 48:49 younger than average and positive is older than average and this is the 48:57 Alzheimer's set there and what you can see is there's a significant difference between the mean and the Alzheimer's 49:03 brains are significantly older molecularly than the controls okay you can extend 49:09 that to looking at amyloid levels in the brain shown here or tangles 49:16 so again younger brains now on this side older brains on that side versus amyloid 49:23 levels and significant correlation the 49:28 younger brains more protected against amyloid same for parkinson's okay so 49:35 this would lead to a hypothesis than one hypothesis there are others but one hypothesis would be that people whose 49:44 brains have aged more slowly or more protected against Alzheimer's disease and conversely people with a faster rate 49:52 of brain aging are more susceptible to Alzheimer's disease now many of you would say well that's 49:57 obvious we knew that already but I think this is a more rigorous 50:03 demonstration of that now it becomes interesting also when you compare this 50:08 to the known major risk factor for Alzheimer's disease which is the April 50:16 efore allele okay which as far as I know stands way above any other risk factor 50:24 for non familial Alzheimer's disease there's nothing else even close and I'll 50:32 show you a little data here then that shows the relationship between Delta EIJ 50:38 and April re4 okay so we took the cohort 50:46 and divided it into groups that significantly deviated from the average 50:52 Delta age so this is - five years is the youngest I think decile and the cohort in terms 51:00 of aging rate and plus five years it's the oldest decile highest aging rate and 51:09 the number of apoe4 alleles is shown here what you can see is it's quite remarkable that it this big risk factor 51:17 for Alzheimer's April what's plotted here is the relative risk of a B in the 51:22 cohort having one allele of apoe4 normally a very significant risk factor 51:28 is blunted almost to insignificance in the young brains so what we interpret 51:36 this to mean is that slow brain aging can suppress having an April re4 allele 51:42 and that these two variables of brain aging and April re4 functioning 51:48 independently conversely in the older brains you can see that it's much higher 51:55 with no April re4 alleles there's a higher risk of Alzheimer's and the defective April re4 is very evident so 52:01 one way of thinking about this then is April we four it's a risk to Alzheimer's and rapid aging 52:09 for Alzheimer's and we also I didn't show you the data but we asked whether 52:15 the April we for brains compared to control brains showed more rapid aging 52:21 and they didn't they were the same so what we think then is that these two 52:27 risk factors are independent roughly similar magnitude and together is the 52:34 worst combination leading to Alzheimer's I list other phenotypes here 52:42 I think our data is strongest for Alzheimer's disease so with and again 52:50 what we want to use this tool for is to probe more deeply into the mechanism of 52:56 brain aging and what determines it so with that I'll stop and take questions Funding Sources 53:02 and I just will acknowledge my funding sources disclaimers here and misaki 53:09 gurashi who did all the gut work and kristin gloria so the work on the brain Sirtuin regulation of metabolism and stem cells - Danica Chen SENS Research Foundation 10,624 views Feb 25, 2014 The metabolic network is coordinately regulated in response to nutritional status to maintain homeostasis. Perturbed metabolic homeostasis is integral to the aging process and underlies many aging-associated diseases. Recent studies strongly suggest that metabolic enzymes are concertedly regulated via acetylation to allow coordination of the directionality and the rate of the metabolic flux upon changes in nutritional status. This mode of metabolic regulation is conserved evolutionarily and is regulated by the sirtuin family of deacetylase. SIRT3, a mammalian mitochondrial sirtuin, regulates the global acetylation landscape of mitochondrial proteins and triggers a metabolic reprogramming to reduce oxidative stress. SIRT3 regulation of oxidative stress has profound physiological relevance, such as stem cell maintenance and tissue homeostasis at an old age, and prevents many aging-associated diseases, including cancer, heart failure, and hearing loss. The SIRT3 regulatory program is suppressed with aging and, intriguingly, SIRT3 reactivation is an effective means of rejuvenation. SIRT7, a mammalian nuclear sirtuin, senses nutritional status and functions at the chromatin to epigenetically regulate metabolic homeostasis. 10K views 9 years ago … 8 Comments rongmaw lin 6 years ago Of the seven human sirtuins, SIRT6 and 7 are located in the nucleus, and SIRT3, 4, and 5 in the mitochondrial matrix. SIRT1, while in the nucleus, interacts with core histone tails, but it can also be present in the cytoplasm and interact with many non-histone proteins. 3 years ago I’m confused: Numerous animal studies, some quite recent, appear to show a reduction in metabolic rate in calorie restriction. Here, the opposite claim is made. Which is true? Reply 3 replies @abvmoose87 @abvmoose87 4 years ago Was this talk next to the hotell wardrobe or whats going on? Why am I hearing adjusting clothes, zippers, moving chairs constantly through this talk? Such a shame for otherwise great talk. Prop great info too but it's all lost in the background noise.