Is The 'Longevity' Gene Sirtuin One Big Research Error?
    By Catarina Amorim | September 21st 2011 11:59 AM | 14 comments | Print | E-mail | Track Comments
    About Catarina

    After many years as a scientist (immunology) at Oxford University I moved into scientific journalism and public understanding of science. I am...

    View Catarina's Profile

    A study out tomorrow in Nature by researchers from the Institute of Healthy Ageing at the University College of London and colleagues is questioning the anti-aging effects of sirtuin – which is “just”the most important anti-aging gene of the decade - claiming that its capacity to increase longevity was nothing more than an experimental error, and showing that,once the flaws are corrected, sirtuin has no effect on lifespan.

    The work by Filipe Cabreiro, Camilla Burnett, Sara Valentini, David Gems and colleagues in Hungary and the US will shake the anti-aging research world, as well as the multi-million dollar industry linked to it, and it will not be easily accepted. But even if this is not the first time that some experiments are questioned, it is the first time that researchers are able to identify problems in the original experiments and show that when they are corrected the outcome is very different (and not in one, but in six of them). In other words, these new results will not be easily fought off.

    Like Oscar Wilde’s "Dorian Gray" character, modern society obsesses about losing youth so much that we have turned the anti-aging industry into a multi-million dollar giant with one of the fastest growing markets in the world.

    And in the last decade, sirtuin has probably been one of the industry’s biggest bets, ever since high levels of this protein were linked to longer, healthier lives in a variety of animals and it was suggested that they could be behind the increased longevity seen with calorie restriction (drastic restriction of calories, without malnutrition is known to increase longevity and retard age-related diseases).

     So how did we get here, 10 years on, concluding that it is all a mistake?

    The whole story starts in 1999 when it was discovered that an extra copy of the sirtuin gene increased yeasts’ longevity up to 30 %. Amazingly,a similar effect was found in a variety of different animals: fruit fly, the roundworm Caenorhabditis elegans and even mice. In roundworms,hyper-activation of the sirtuin gene was described to increase longevity are markable 50% more. But it was when sirtuins were linked to the high longevity obtained with calorie restriction that things got really exciting.

     Calorie restriction increases lifespan and delay age-related diseases in a variety of organisms, including mammals and, apparently(since the experiments have not finished yet) even non-human primates,suggesting that it might also be relevant in humans. If sirtuins were behind calorie restriction, then their study could be the key to understanding our own aging, and fighting it.

    To the media the idea of being able to extend life was nothing less than irresistible, but to researchers the possibility of reducing diseases associated with aging – such as diabetes, cardiovascular diseases and neuro-degeneration - was no less tempting and in 2003 the first molecule capable of activating sirtuin was found. Resveratrol is a natural ingredient of red wine and its discovery was followed by a frenetic search for synthetic equivalents (preferentially more potent) to patent and develop. By then sirtuin's potential was considered so high that the company launched by those involved in its discovery was bought by the pharmaceutical giant GlaxoSmithKlein for $720 million. Since then several synthetic equivalents of sirtuin have been found and at least one tested as far as human trials, with no conclusive results. Meanwhile anti-aging creams with red grapes extract have already been launched in the market riding on the wave of attention that grabbed the media, with even the New York Times giving sirtuins and resveratrol several main pages.

    Inside the academic world another mood was stirring, with several groups starting to challenge revesratrol’s capacity to activate sirtuins among claims of artifacts and improper controls, only to have the “pro-sirtuin field” producing more supportive experiments and new explanations further rustling the field. And then, sirtuins’ effects on longevity started to be challenged too.

    So what was happening, and why were scientists having such a hard time agreeing on the results?

    The problem is that aging is a complex process and one very hard to study.  Longevity – contrary to things like number of limbs or eye color - is not an exact number or a “yes”or “no” result, but instead a subjective measurement that can be affected by many factors with some being particularly difficult to detect like laboratory environment or inbreeding. This also means that to be able to take conclusions when it refers to longevity, the experiments, and particularly their controls, have to be flawless. 

    Normal 0 false false false EN-GB X-NONE X-NONE

    So the contradictory results regularly popping up in the sirtuin research field led Camilla Burnett, Sara Valentini, Filipe Cabreiro,David Gems and colleagues to wonder if there could be some kind of experimental design problems going on, and, finally, go back for a close and through look at the original experiments.
    One of the biggest issues in this type of assays - where a gene is studied by introducing it into a species without it and then have this new population compared with control animals (animals from the same species but without the new gene) to know the function of the gene- is the confounding effect of the genetic background (the remaining genome). In fact, not only these new genes (called transgenes - genes brought from another species) tend to be carried with “activators” or “carriers” that can themselves have an effect on longevity, but also these new animals need to be bred in order have numbers enough for a statistically significant comparison, what can create problems. This comes with the fact that during the time needed to achieve enough animals, both transgenic and control populations change, losing and gaining genes as well as mutations.So when the two populations are finally compared they have many more differences than just the transgene.  

    The way to sort this problem is by crossing the transgenic population back with controls (this is called backcross) a few times assuring that both groups have the same genetic background (only differ in the gene to test). And this was not done in any of the original sirtuin experiments.

    So to investigate if this lack could have led to errors in the experiments that launched sirtuin’s effects on longevity, Cabreiro and colleagues looked back at two of the initial transgenic roundworms - both with multiple copies of the sirtuin gene (so high quantities of sirtuin) and higher longevity – and backcrossed them several times with control animals. What they discovered is that the resulting roundworms now showed normal lifespan despite conserving high levels of sirtuin.

    This basically meant that whatever was increasing longevity it was not sirtuin. In one of the transgenic worms the researchers were  even able to identify the gene behind the increased longevity as a known anti-aging gene called Dyf. On one of the roundworms they also blocked the sirtuin gene to see that longevity was not affected, confirming that sirtuin had no effect on lifespan. 

    Next Cabreiro and colleagues testeda transgenic fruit fly – also from the original experiments - with a “switch” that hyper-activated the sirtuin gene and increased lifespan. This time, after the transgenic population was backcrossed, the resulting flies still lived longer than control animals but not than flies with only the “switch (and no sirtuin). In other words, longevity was not linked to sirtuin but, instead, to the “switch”. Confirming these results, when the researchers constructed a new transgenic fly with much higher levels of sirtuin and no “switch”, all obtained flies had a normal lifespan.

    The researchers also tested resveratrol, the component of wine that supposedly activates sirtuins and is at the basis of several anti-aging products already in the market. But, despite the multiple tests done by the different laboratories using different methods, no effect of resveratrol on sirtuins could be found.

    Finally they looked at one of the most interesting sirtuin experiments – the one that launched the possibility that calorie restriction worked through sirtuins – where fruit flies without sirtuin were put under calorie restriction and showed no increase in life span. When Cabreiro and colleagues backcrossed these transgenic flies with normal animals so all had the same genetic background and tested the resulting animals this time they responded to calorie restriction living longer showing that, yet again, previous results resulted from the effects of genetic background. 

    The conclusion, proved in this study over and over, is that sirtuins have no effect on longevity and that several of the essential experiments of the” anti-aging sirtuin theory” were wrong due to design flaws what raises the question; why did it take 11 years to detect these?

     The new results will not be accepted without a fight, there is just too much at stake, and in this same edition of Nature researchers linked to some of the original experiments in worm  reply accepting that they did mistakes but also claiming to have produced new transgenic roundworms that do not lose their high longevity when backcrossed. Interestingly they also refer , as proof that sirtuins work, to a published article where again no backcross was done.

    Whatever happens now, and even if Glaxo and the original researchers keep pursuing the “anti-aging sirtuin theory”, its credibility has been strongly shaken. Will the anti-aging creams based on resveratrol disappear from the market? Will investors notice this article? The answer to both is “probably not”. But it will be difficult for Glaxo or any other company, to justify human trials until they can prove without doubts that sirtuin has an effect on lifespan and aging-linked diseases. This is an exciting story and one that people inside and outside of the field will no doubt be following close...

     Citation: Burnett, C et al. Absence of effects of Sir2 over-expression on lifespan in C. elegans and Drosophila. doi:10.1038/nature10296Nature; e-pub in advance, 22/09/2011.


    The effects of resveratrol have already been debunked by a follow up study showing that the effects of the resveratrol were related to insufficient controls in the study ( no fluorescence done with background cleaved product Borra, 2005). Unfortunately, Sirtris could not make money on their new combinatorial library if it did not have this effect on Sirtuins, so they followed it with a followup experiment to cover themselves. They made their money, and most people never knew enough about the in vitro assay to realize the mistake, so they pulled off a 3/4 of a billion dollar sale.

    Sirtis actually claimed that the fact that revesratrol only worked in the presence of the fluorochrome was because the fluorochrome mimicked a natural substrate... they do have an answer to everything
    haha. If someone offered me $720 million I wonder if I would say eating rocks cures all cancers. They are after all patented proprietary rocks.

    This basically meant that whatever was increasing longevity it was not sirtuin. ... Dyf ... longevity was not linked to sirtuin but, instead, to the “switch”
    Is this promising or are we likely confronted with publication bias (no or negative results never published) or maybe that the trangenic population always gets treated especially well during their multiplication phase? I mean, if I were naive and did not know about the publishing game, the fact that something apparently prolonged lifespan in many of these flawed studies seems not to be questioned in the new article, or are the authors just politically correct not calling the whole lot a pile of bunk?

    How can one tell people to look at recent scientific research for their decisions? Certainly in the medical field, one cannot trust anything before not at least 20 years have passed.

    I think that to be interesting it needs to be relevant for humans or at least mamals and so far nothing, beside sirtuins, seem that interesting. Dyf is related to neurons although sensorial neurons, the problem is that too many things affect longevity because they make life a little more difficult/easy , not because they really affect longevity. When it concerns to believe or not data if you know the filed - I read a lot in the last 3 weeks- you could sense something was not right. If people keep coming with objections over and over to the same guys, if they are involved in something that involves a lot of money and have an answer for evrything and more if they are personal in their attack you start saying "hum... should i really trust these guys?". The NY Times also made an article , have alook they make the whole thing a war between continents (they have been supporting the sirtuin field from the beggining so coldn't really take another position but the personal attacks and the fact that none of the authos from this paper was able to commment on their article does not look good)
    It is important to note that the drive for anti-aging drugs is not to prolong lifespan per se, because that would take too long to prove for a primate drug trial, but to treat diseases of aging. One of the founders of Sirtis has stated in seminars that a key result for the purchase of the company was that their compounds extended the life of obese mice. Lab rodents fed to look like most American adults have shorter lifespans similar to humans, and the large increase in the weight of the average American, and consequent diabetes and other related diseases, is a huge potential market. These studies will need more scrutiny as actual drugs are prepared and tested for market.
    Most of us reading these conversations will not live to see if an effective anti-aging drug is ever produced because the validation of such a result will take too long. As a society and as individuals, we would be far better off focusing on how to convince the general population to support a healthier lifestyle and appropriate preventative medical care in the face of a barrage of unhealthy foods, lax attitudes towards the dangers of being overweight and a lack of education for unhealthy bad habits (and, perhaps, more focus on personal responsibility).

    The problem with these mice is that they receive such a toxic diet that is difficult to know what is going on. For example they never manage to live as long or be as healthy as normal animals and also they use activators of sirtuin when most people believe that resveratrol and equivalent molecules do not activate sirtuins and apparently sirtuin are not affecting longevity. It's very good to say that longevity is not the most important - ageing is not considered a disease anyway - but when everything else is not fitting togethr how can we trust anything derived from it?
    With regard to the mouse toxic diet, obese Americans feed themselves a highly toxic diet with shortened lifespans and a larger burden on the healthcare system. Consequently, it is whether the original observation that the compounds do something to alleviate these health effects is the central question and are experiments that must be repeated independently. Since it is a drug for human consumption, it needs the highest scrutiny prior to approval in any case.
    In terms of deriving conclusions from the controversial experiments, the issue again is whether the promising results that occurred under a flawed paradigm are independently repeatable. While we do want to know the mechanism, making sure that the phenomenon is true (when its potential impact is so high) should be the main focus. Multiple examples of useful discoveries made using models which were mechanistically incorrect are present in today's society. My favorite is the capacitor, which was designed in Ben Franklin's time as a Leyden jar to store electricity when it was thought to be a fluid. Electricity is not a fluid, but capacitors are everywhere today. Consequently, the observations claimed in the last ten years in the sirtuin field that can be of most potential medical use are worth repeating in a careful manner even if the mechanism is unclear (the same way aspirin used to be a couple of decades ago).

    That I know there are 2 studies in mice, in the first the mice are still living when the experiment finishes and in the second they don't have increased longevity. There are also a few strange deaths in the middle. To me two things here count - 1st I no longer feel confortable with the people bringing out the data as they do have a problem accepting different results and 2nd if they improve ageing related diseases shouldn't the animals theoretically last long - ok not necessarily all of them but not at all?
    A few points that almost everyone is missing in their analysis of this study, even those who actually read it and are qualified to interpret the findings and data.

    1. This is one study, which only addressed the Sirtuins, and only from the standpoint of longevity. It is an extremely narrowly focused and short term animal study only. The non-scientific press is again jumping to a conclusion that is not supported by the data and findings of the study.
    2. There are over 5,000 published peer-reviewed studies on resveratrol, as well as the principal analogs of resveratrol, which elucidate the stilbenes’chemo protective properties with respect to the so-called diseases of aging and obesity, including type 2 diabetes, coronary artery disease, cancer, neurological conditions and dysfunction, inflammatory conditions such as arthritis and other autoimmune conditions, physical fitness, endurance and mitochondrial function and density, and many others.
    3. Resveratrol operates via multiple clearly identified pathways apart from sirtuin modulation. These include Nf Kappa B, Cox 2, P-53, pro apoptiic and autophagy activation and regulation, mTOR, FOX0, and over 100 others.
    4. The benefits of resveratrol are in its ability to extend the period of healthy lifespan, not necessarily to directly extend lifespan per se. However, if resveratrol prevents one from developing diabetes, heart disease or a cancerous tumor it certainly has extended the life and healt span of that individual.
    5. As the other sturcturally related molecules to resverarol, such as pterostilbene and polydatin are studied and analyzed additional beneficial effects are being confirmed.
    6. In the limited number of human clinical trials either underway or recently completed the health span extension and disease prevention properties are confirmed.
    7. The issue of longevity is extremely complicated. There is probably no gene or set of genes that directly promote or inhibit longevity. Natural selection for longevity is not in the interest of the survival of any species, with the possible exception of humans. Nature is ambivalent when it comes to longevity past the individual’s age of reproduction.

    James I am not questioning what resveratrol does,in fact resveratrol seemed so right when it was linked to sirtuin activation because people already thought it improved aging-related diseases. This article is about sirtuins and even when resveratrol appears is in its context. The big issue here is a story about these proteins , what is believed , and the mess in the field. Sirtuins were always particularly interesting because of their potential link to calorie restriction which is just not there. This article shows that the original experiments were messed up and we have to re-start again and be clear about what is happening. It might be alimited message but is one that is at the core of all the theory. And resveratrol activation of sirtuins has been questioned too, in one of the most famous experiments it only worked with the fluorochrome for example. There are quite a few good review articles on that. Human trials I only know one , the one that tried to treat leukemia I think and finished because of side effects. So I know none with positive results.
    1) 5000 peer reviewed studies on resveratrol? Where did you come up with this number. I searched elsevier and only had 250 and most quoted previous studies or had nothing to do with pharmaceutical efficacy.
    2) invitro assays are easy to do, but when dealing with compounds like natural products, people tend not to do sufficient controls to insure that the results are valid. Many natural products interfere with chromophores and fluorophores either by quenching, changing the ph just enough to change the flourescence etc. This is one of the reasons why most in vitro research, even peer reviewed research, related to turmeric is so unreliable. Keep in mind that many of the fluorophores used in in vitro assays are derivatives of natural products. You should be skeptical when you see that a compound works on several targets.
    3) "The benefits of resveratrol are in its ability to extend the period of healthy lifespan" - proof that it works in humans. . zero.
    4)what clinical trials? If it is not an FDA or equivalent approved clinical trial it is NOT a clinical trial.
    I assume whomever wrote this has a financial stake in these products. Sigh

    Gerhard Adam
    So what was happening, and why were scientists having such a hard time agreeing on the results?
    I guess I'm a bit confused since it should have been obvious that the science was certainly not a closed book.  Confirming results and more experimentation seemed to be ignored in favor of marketing.  What was ANY business doing trying to capitalize on what was, at its absolute best, preliminary biological knowledge?

    It's little wonder that the public is increasingly distrustful of scientists and moneyed interests.
    Mundus vult decipi
    The most important question is: What is the real role of the sirtuin enzymes ?
    There are seven sirtuins in humans, and their cellular level change in response to different signal, in addition
    all of us know that these enzimes are conserved in several organism, why in your opinion? because are unnecessary?
    I don't believe in the sirtuins activator, for me resvelatrol is only a pleiotropic molecule with antioxidant activity!!! but for understand the sirtuins function we need of new inhibitors!!!