• Log in with Facebook Log in with Twitter Log In with Google      Sign In    
  • Create Account
  LongeCity
              Advocacy & Research for Unlimited Lifespans

Photo
- - - - -

Can SENS produce an immortal yeast ?


  • Please log in to reply
49 replies to this topic

#1 noam

  • Guest
  • 34 posts
  • 0

Posted 02 January 2006 - 05:29 PM


I was wondering, SENS seems to be based upon the assumption that there is no active genetic mechanism for aging, meaing there is no simple clock-gene that governs how much an organism will live. In order for SENS to work, aging must be regarded as accumulation of random damage over time.

If this is correct, is there any reason to believe things won't be the same for our most basic organism that we use for a model of eukaryotic aging - the yeast ?.

If de Grey's model for combating aging is correct, shouldn't we expect for it to work for yeast ?.

Let's see, from the seven categories:

1. Cell loss.
2. Nuclear Mutations.
3. Mitochondria Mutations.
4. Death-resistant cells.
5. Extracellular crosslinks.
6. Extracellular junk.
7. Intracellular junk.

For yeast aging, categories #1,4,5,6 are not relevant (yeast is a single cell organism).

So, we are left with:

1. Nuclear Mutations
2. Mitochondria Mutations
3. Intracellular junk

But on a second thought, according to de Grey, Nuclear mutations aren't relevant as well, because our main concern from this regard is cancer, which isn't relevant for yeast.

So, we are left with:

1. Mitochondria Mutations.
2. Intracellular junk.

Could this be it ?.

So, according to my interpretation of de Grey's vision, all we have to do is move the mitochondrial genes into the nuclear DNA of the yeast (assuming we know who they are, and assuming that like in mammals, each protein whose gene is in the mitochondria's genom, is part of a complex which has other participants which are encoded in the nuclear DNA) and find the right enzymes to brake the specific intracellular yeast junk ?.

Is it really all it takes ?. You do realise that people in the lab are trying countless amounts of mutations (for several years now) in order to prolong yeast life, and our best success as of yet, was an increase of 60% in yeast life span (replicative life span, regarding chronological I think the maximum was 30%).

I think that if SENS was to prove itself in yeast, which should be much, much easier than in mouse, or even drosophila, it might not win the hearts of the general public, but it will certainly win the hearts of many scientists. (I am in no way saying it should replace proof in concept in mouse, but if we can make proof of concept in yeast, within 2-3 years, it might be a much smaller pill to swallow that we can ever make it for mouse, it might give a much needed in-between step).

Also, I don't think that using yeast as first proof of concept is such a deviation from the mouse. Afterall, we will have develop protocols for moving those mitochondrial genes in the mouse too, and first doing it for the single celled yeast might prove like a smarter strategy. Regarding the inclusion of intracellular junk braking enzymes, again, first dealing with yeast might make it easier to see what are the basic problems of doing such a thing, before we jump for the extremely complex mouse, with its trillion cells and immune system...

Not to mention that even if someone had the money, and wanted to start proving SENS in mice (or drosophila, or any other multicellular eukaryotic animal) already today, he really can't. We still don't have the needed quality vectors for gene therapy, nor enough knowledge about how to work with stem cells. We still have to wait several years. But, we do have all the technology necessary to try it for yeast already today. In 5 years when the necessary technology will make it possible to start with mice, we won't start from zero, but we'll be able to use much of the information from the yeast (not to mention it will be much easier to raise capital after you already done proof of concept for some eukaryotic organism).

EDIT:
I do realize that the above scheme is only "a first generation therapy" for the yeast. We haven't dealt with Nuclear mutations, which will probably be the next problem the yeast will meet.

Saying that, I'm not sure DNA repair will be such a bottleneck afterall. Today the longer lived yeast strains devide for ~30 times before they die. I don't think anyone is claiming that after 30 divisions they die from DNA mutations. Nuclear DNA mutations might still allow the SENS treated yeast to live for 120 divisions (if not much more), which will be great proof of concept that SENS practically works for a eukaryotic organism.

I think that if you'll ask S.Jay Olshansky, or other SENS opposers (most of the scientific community), they will tell you that SENS will definitely not dramatically prolong yeast life in the near future.

Edited by noam, 02 January 2006 - 07:41 PM.


#2

  • Lurker
  • 1

Posted 03 January 2006 - 10:04 AM

Welcome Noam. Hopefully Aubrey himself will address your question.

Your point on engaging the scientific community before seeking to court the general public is well taken.

Are you aware of the 2004 report by Maringele et al ?

#3 John Schloendorn

  • Guest, Advisor, Guardian
  • 2,542 posts
  • 156
  • Location:Mountain View, CA

Posted 03 January 2006 - 10:55 AM

In order for SENS to work, aging must be regarded as accumulation of random damage over time

As far as I can tell there is no requirement for the damage to be random, or why would there?

#4 noam

  • Topic Starter
  • Guest
  • 34 posts
  • 0

Posted 03 January 2006 - 03:38 PM

Hello Prometheus and John,

As far as I can tell there is no requirement for the damage to be random, or why would there?


Is this better ?:

In order for SENS to work, aging must be regarded as random accumulation of damage over time.

The damage indeed can be made by many specific mechanisms, but it has to accumulate randomly over time, and not mainly according to some specific longevity-genes, which act upon some independent timer.

I say "independent timer" because while it is true that according to the theory of "Shadow Selection", we might possess a numerous number of pleiotropic genes, which become deadly at old age, still, the assumption for SENS must be that the timer they all use in order to know when to exhibit their negative phenotype, is dictated by some, or all, of the 7 categories of damage, and without input from these categories, their clock will remain "off" (meaning they have no timer which is independant from all of the 7 categories, and in the case of the yeast, from the 2-3 categories mentioned, so if you treat these categories, those pleiotropic genes will never show their negative phenotype).

Edited by noam, 03 January 2006 - 03:54 PM.


#5 John Schloendorn

  • Guest, Advisor, Guardian
  • 2,542 posts
  • 156
  • Location:Mountain View, CA

Posted 03 January 2006 - 03:50 PM

If the damage were exclusively caused by "some specific longevity-genes" (which might then better be called "aging" genes, but never mind), then targeting the damage à la SENS (not bothering with the genes) might still seem like a valid approach to ensure freedom from suffering. SENS seeks to repair the damage of aging regularly, before it reaches pathogenic levels. This idea remains consistent, no matter how the damage is caused.
Although, if the damage-causing genes were few, dispensable, and easy to remove from vivo, then it seems that targeting them would be easier than it really is, and the SENS argument might have less force. However, even if you deleted all such damage-causing genes, it seems inconceivable to me how the body should be able to avoid all types of irreversible damage unmediated by such genes.

Btw, the above is not meant to affect your yeast argument in any way. I don't know anything about yeast aging.

#6 noam

  • Topic Starter
  • Guest
  • 34 posts
  • 0

Posted 03 January 2006 - 04:06 PM

Although, if the damage-causing genes were dispensable, and easy to remove from vivo, then it seems that targeting them would be easier than it really is, and the SENS argument might have less force.


Exactly.

However, even if you deleted all such damage-causing genes, it seems inconceivable to me how the body should be able to avoid all types of irreversible damage unmediated by such genes.


That's my opinion too. You should know that some of the researchers I taked with don't rule out the option that the so called ageless animals (ala the giant Galapagos turtle) indeed do not age. If this was to be correct, then according to our understanding of things (you and I, that is), they also shouldn't show meaningful deterioration in any of the 7 categories.

Saying this, thinking that the long lived phenotype exhibited by such animals is governed by just a few "aging genes", seems to be pretty far fetched to me. I think that over evolution, the high ability of these animals to survive extrinsic forces, pushed the maintenance genes toward exhibiting their negative phenotype at much older age.

Edited by noam, 03 January 2006 - 04:34 PM.


#7 John Schloendorn

  • Guest, Advisor, Guardian
  • 2,542 posts
  • 156
  • Location:Mountain View, CA

Posted 03 January 2006 - 04:12 PM

I see, we basically agree. One problem with ageless animals (or in particular, translating their agelessness into animals that do age) is that the intervention would be excessively long-term to test. Targeting damage has the advantage here that you can measure it instantly, but that comes at the cost of not being able to correlate it precisely with health in many cases...

As for antagonistic pleiotropy, I know no example of a pleiotropic gene that would be "off" during youth and suddelny revert to an "on", damage-causing state in old-age. The mechanism of pleiotropic genes I know of is to cause damage very slowly all the time, so that the damage-level does not matter in youth, but becomes pathogenic in old-age (e.g. build-up of an atherosclerotic lesion over the lifespan, accumulation of senescent cells - see recent work by campisi).

#8 noam

  • Topic Starter
  • Guest
  • 34 posts
  • 0

Posted 03 January 2006 - 04:21 PM

As for antagonistic pleiotropy, I know no example of a pleiotropic gene that would be "off" during youth and suddelny revert to an "on", damage-causing state in old-age


I agree that your definition of pleotropy is the improtant one. But you can look at it from another angle too:

You can think of any protein, that while the cell is "healthy", it does its job well, but when things starts to deteriorate in the cell, and so the micro envronment in the cell changes for that protein, then it suddenly start to work in wrong directions and boost the rate of damge accumulation to much higher levels than before. So, in effect it seems that this protein started to show its negative phenotype at old age, but like I said above, the timer is damage and not some independent clock, and this notion is of course compatible with SENS.

So, if you take this into the "ageless animals", then they simply evolved better maintance genes (to match their high ability to survive extrinsic factors), and this also delyed the onset of negative contribution from these other (pseudo) pleotropic genes.

Edited by noam, 03 January 2006 - 04:52 PM.


#9 John Schloendorn

  • Guest, Advisor, Guardian
  • 2,542 posts
  • 156
  • Location:Mountain View, CA

Posted 03 January 2006 - 05:02 PM

Sounds good. Now I can't wait for a yeast aging person to drop by. Is yeast more like the timer-controlled pleiotropic system (extrachromosomal DNA circles??) and thus unfit for SENS?

#10 ag24

  • Honorary Member, Advisor
  • 320 posts
  • 29
  • Location:Cambridge, UK

Posted 03 January 2006 - 05:44 PM

The closest one could get in yeast to a proof of concept of SENS would be to extend the time it can live without the cells dividing. Note that this is not the most common definition of yeast lifespan - normally what is measured is how often a cell can bud off new cells. (This latter definition, termed replicative lifespan, is possible because the yeast we're talking about, S. cerevisiae, divides by budding off a smaller cell rather than just cleaving into two morphologically similar cells.) The reason we wouldn't use that measure is because the things that go wrong are so very different than in mammals -- extrachromosomal circles are the best example.

So, the better measure of lifespan - what yeast people call chronological lifespan in order to distinguish it from replicative lifespan - is worth looking at. The main gap in your logic is that the fact that yeast don't get cancer means that they probably do accumulate non-cancer nuclear DNA damage fast enough for that sort of damage to contribute to their aging. This is a bit of a showstopper really, because it's something that's explicitly excluded from SENS for reasons that don't apply to yeast. The same problem exists in using flies or worms to test SENS, because they also don't die of cancer.

Nonetheless, yeast is still a useful organism to work with, because proof of concept of any of the SENS components in one organism is a step towards getting it working in mammals too by showing that it's possible in something. The demonstration that it increases lifespan is a second thing, which would be confirmation of a different tenet of SENS (namely that the seven strands are adequately exhaustive). I would not be surprised if yeast were the first organism in which comprehensive allotopic expression is made to work, not least because only six genes need to be moved instead of 13. (This is for a boring reason unfortunately - yeast have no Complex I at all, so we can't copy the nuclear versions of the seven subunits because they aren't there.)

The Maringele work isn't directly relevant because yeast normally expresses telomerase. (Whether it's relevant to ALT in mammalian cancers is a very important question though - I'm following that group's work with care.)

It's indeed correct that SENS doesn't care whether the damage is random -- except that if it were not, the "aging program" might "try harder" to age us if SENS caused it not to work as normal. However, and despite the number of attempts still being made to argue otherwise, I still agree with the mainstream view that aging is programmed only in a few unusual species like salmon.

#11 noam

  • Topic Starter
  • Guest
  • 34 posts
  • 0

Posted 03 January 2006 - 05:47 PM

Is yeast more like the timer-controlled pleiotropic system (extrachromosomal DNA circles??) and thus unfit for SENS?


First of all, you can regard ERCs (extra ribosomal circles) just like intracellular junk. It is just that.

Secondly, the model today for ERCs formation, involves a gene named fob1 that encodes a protein which blocks the replication fork from moving in one of the directions (this suppose to somehow improve some synchronization for the yeast). The blocked fork sometimes create DNA breakage, which is fixed by homologoues recombination, that can create an ERC.

Thus, a mutation in the fob1 gene, eliminate the replication fork block, and is assumed to eliminate ERCs. It prolonges yeast life by ~30% (depend on the strand). The amount of ERCs found in the post mortem for fob1 mutant yeast, is very low (in comparison to very high, for w.t).

Edited by noam, 03 January 2006 - 06:02 PM.


#12 noam

  • Topic Starter
  • Guest
  • 34 posts
  • 0

Posted 03 January 2006 - 06:00 PM

The closest one could get in yeast to a proof of concept of SENS would be to extend the time it can live without the cells dividing. Note that this is not the most common definition of yeast lifespan - normally what is measured is how often a cell can bud off new cells. (This latter definition, termed replicative lifespan, is possible because the yeast we're talking about, S. cerevisiae, divides by budding off a smaller cell rather than just cleaving into two morphologically similar cells.) The reason we wouldn't use that measure is because the things that go wrong are so very different than in mammals -- extrachromosomal circles are the best example.


See my explanation of ERCs above.

So, the better measure of lifespan - what yeast people call chronological lifespan in order to distinguish it from replicative lifespan - is worth looking at.


It might not be wise to base conclusions on chronological aging in yeast. In order to make yeast reach a postmitotic state, it is being given a stress (like food deprivation). There is no knowing what this stress does to the cell. It is a very non objective system for measuring postmitotic life span imo.

In order to make a better model of the yeast for chronological aging, I think we must create a strain with a conditional mutation that will arrest the cell cycle. This way we will be able to raise the post mitotic yeast in normal conditions, without any stress. But then if Nuclear DNA mutations are the bottleneck, we might be at point zero again.


The main gap in your logic is that the fact that yeast don't get cancer means that they probably do accumulate non-cancer nuclear DNA damage fast enough for that sort of damage to contribute to their aging.

This is a problem of course. But you do realize it is possible to elongate their replicative life span by playing with genes that are not related to DNA repair. Also, if we are talking about DNA mutation as the yeast's longevity bottleneck, wouldn't you expect mitochondrial DNA mutations to be their first longevity bottleneck, and Nuclear DNA mutations a far second ?. Unless you are suggesting that in yeast, Nuclear DNA repair is not better than mitochondrial ?.


If the DNA repair mechanism is not such a limiting factor, I don't see why a replicative life span is not a good measure. A young yeast cell does everything well, and no matter what will happen to it in the future which will cause it to die/age, it must have its roots at molecular damage. If we can prevent this damage, it will only die of nuclear DNA mutations.

Edited by noam, 03 January 2006 - 06:51 PM.


#13 jaydfox

  • Guest
  • 6,214 posts
  • 1
  • Location:Atlanta, Georgia

Posted 03 January 2006 - 06:50 PM

Let's see, from the seven categories:

1. Cell loss.
2. Nuclear Mutations.
3. Mitochondria Mutations.
4. Death-resistant cells.
5. Extracellular crosslinks.
6. Extracellular junk.
7. Intracellular junk.

For yeast aging, categories #1,4,5,6 are not relevant (yeast is a single cell organism).

So, we are left with:

[2]. Nuclear Mutations
[3]. Mitochondria Mutations
[7]. Intracellular junk

But on a second thought, according to de Grey, Nuclear mutations aren't relevant as well, because our main concern from this regard is cancer, which isn't relevant for yeast.

So, we are left with:

[3]. Mitochondria Mutations.
[7]. Intracellular junk.

(I renumbered the list for consistency)

Fascinating ideas regarding implementing points 3 and 7 only. Point 2 has been covered by others, and I think we can agree that this point would be a limiting factor in yeast more so than in mammals. Here, nuclear DNA repair and maintenance would become critical, and indeed, several of the known longevity-modulating genes involve nuclear DNA repair/maintenance. See the next paragraph, though, on potential benefits of this fact for validating points 3 and 7.

As for replicative versus chronological lifespan, I can see that this will skew the interpretation of the results. One thing I might suggest, should points 3 and/or 7 be implemented in yeast, is that multiple strains be produced with other known longevity-modulating genes (e.g. sirtuins, anti-oxidants, DNA repair enzymes, etc.), and help pin down the source of any life extension for both replicative (perhaps including versions with telomerase inhibition, via one or both subunits?) and chronological lifespans. This should validate the gains by allowing statistical analysis of confounding variables. My suspicion is that there will be low overlap with other forms of longevity-modulatiing genes, and hence a SENS 3/7-enhanced yeast should see cumulative benefits from other genetic mods.

As for the massive parallel testing? Hey, it's yeast: short lifespan, low cost. All the major/popular genetic mods could be tested in conjunction with a SENS-like strain of yeast, in a so-called "shotgun" analysis. Let the statistics gurus sort out the details.

#14 noam

  • Topic Starter
  • Guest
  • 34 posts
  • 0

Posted 03 January 2006 - 07:17 PM

The main gap in your logic is that the fact that yeast don't get cancer means that they probably do accumulate non-cancer nuclear DNA damage fast enough for that sort of damage to contribute to their aging.


On a second thought, I'm not sure I'm following this logic.

What do you mean by "yeast don't get cancer" ?. A culture of yeast cells will divide indefinitely, what more can we ask ?.

Are you implying that in "yeast cancer", the the yeast mother cell should "live forever" ?, that's not a fair request.

So by "yeast cancer" I take it you probably mean that the yeast mother cell will lose the cell cycle control, and in its limited life span, will create a much larger amount of daughter cells. But most chances this will decrease the life span of the mother cell, because losing cell cycle control will incorporate mutations which will hurt the ability of it to survive.

Ok I give up, moment of weakness, can someone explain to me what is "cancer" in yeast, and why if it happened, it would have implied they have better DNA repair mechnisms than if not ?.

#15 noam

  • Topic Starter
  • Guest
  • 34 posts
  • 0

Posted 03 January 2006 - 07:36 PM

What about this test:
If you sequence the genome of some haploid yeast today (just 13,000kb total, but I guess just one chromosome would be enough), and then sequence the genome of one of its great great great.... granddaughters in 2 months from now (which grew in the same conditions all the time). How much nucleotide difference will there be ?. If the DNA repair and replication mechanisms are not such a bottleneck, then there won't be much of a difference between the two genomes.

#16 ag24

  • Honorary Member, Advisor
  • 320 posts
  • 29
  • Location:Cambridge, UK

Posted 03 January 2006 - 07:42 PM

Hi again,

Yes, I know that ERC-mediated aging can be modulated as you describe - but at a presumed cost in terms of the accunmulation of nDNA mutations. Replicative lifespan will certainly not be limited by mt mutations unless it's extended enormously by other means first, because replication in vitro is so fast - far faster than one would expect in typical natural environments. Ditto for most intracellular junk I think - ERCs seem to be the only junk that matters. (Bud scars might be next, but not for a while.) I agree that chronological lifespan is best examined by a method that does not involve stress, but deleting a cell cycle gene may not be appropriate - might cause other stresses if the cell gets into a state that would normally trigger division and then finds it can't. Best would be to mimic natural conditions. nDNA repair is surely better than mtDNA repair in yeast, but perhaps only on a per-kb basis.

#17 noam

  • Topic Starter
  • Guest
  • 34 posts
  • 0

Posted 03 January 2006 - 08:08 PM

but deleting a cell cycle gene may not be appropriate - might cause other stresses if the cell gets into a state that would normally trigger division and then finds it can't.

What about a mutation in one of the appropriate components of the cell cycle control system. You can create a situation where such a component always senses an inhibitory signal, and this way the arrest is more natural.

Let's assume for a moment we do succeed to do the following things:

1. Creating a post mitotic yeast cell, that experience no stress.
2. Transferring the mitochondrial genes to the nuclear genome.

Let's assume these three implications/hopes:
A. Because there is no DNA replication, nDNA mutations are not a close bottleneck.
B. Because there is no DNA replication, ERCs won't accumulate.
C. The post mitotic yeast (without the mtDNA transfer) does not live way too much (it probably won't, but if it will, people will be reluctant to run an experiment with such a cell).

So, now we'll have to compare the life span of this yeast cell, to the life span of a yeast without the mtDNA genes at the nDNA. If we're extremly lucky, there will be some difference in life span (beginners luck).

If there won't be much of a difference (probably), we'll have to figure out what accumulated in the yeast and was the bottleneck (again, assuming implication A holds). After we'll isolate this accumulated damaging agent, find the specific enzymes that digest it, and find how to insert them without them digesting entire parts of the yeast, we'll then run another life span test. (this last paragraph probably isn't going to be easy..., but in comparisson for doing the same with a mouse, piece of cake).

Assuming the SENS'ed yeast will now live significantly more (how much is singnificant, I'm not sure), do you think this will be worth the hard work ?. What will the immidiate reaction of the naysayers will be to such a test, using such a model ?, do you think it is convincing ?.

I think that a really "convincing" result will come only if we'll get to a point that we'll prove that we also juiced the advantage of the mtDNA transfer.

Edited by noam, 03 January 2006 - 09:15 PM.


#18 Mind

  • Life Member, Moderator, Secretary
  • 16,307 posts
  • 2,003
  • Location:Wausau, WI

Posted 03 January 2006 - 09:54 PM

Do yeast age "chronologically"?

#19 JonesGuy

  • Guest
  • 1,183 posts
  • 8

Posted 03 January 2006 - 10:25 PM

We most often measure the age of a yeast by how many times it has budded. The time aspect can be changed based on nutrient availability.

#20 noam

  • Topic Starter
  • Guest
  • 34 posts
  • 0

Posted 03 January 2006 - 10:56 PM

We most often measure the age of a yeast by how many times it has budded.

Right, this is called "replicative life span", and it has nothing to do with chronological life span, meaning, if yeast A gave 40 buds during the 2 days in which it lived, and yeast B gave 20 buds during the 4 days in which it lived, then the replicative life span of yeast A, is higher than that of yeast B, even though yeast B lived twice more time than yeast A.

The 4 days that yeast B lived (or the 2 days that yeast A lived) are by definition, *not* called their chronological life span.

Chronological life span in yeast, refers to a situation where you measure the time that a yeast cell stays alive, without replicating, before it dies. In order to induce such post-mitotic state in the yeast, so that we could check this definition of chronologial life span, we subject the yeast to a stress, like food deprivation. Food deprivation indeed stops the yeast from dividing (it induces the cell cycle control system sensors that sense prohibitive outside signals).

The problem with this technique of inducing a post mitotic state, is that it is hard to differentiate between the stress induced aging, and the normal aging. You basically kill the cell with the stress!. let's say you found one yeast strain that lives twice longer than another, the question remains, does indeed it has twice longer chronological life span, or does it have twice the ability to make up for lack of nutrients in the medium ?.

#21 Mind

  • Life Member, Moderator, Secretary
  • 16,307 posts
  • 2,003
  • Location:Wausau, WI

Posted 03 January 2006 - 11:03 PM

What about just isolating the parent yeast cell after each division, and see how long it lives? Or is the distinction parent/offspring meaningless in yeast. (as you can tell, I am not a yeast expert)

#22

  • Lurker
  • 1

Posted 04 January 2006 - 05:24 AM

The same problem exists in using flies or worms to test SENS, because they also don't die of cancer.


Drosophila (flies) do get and die of cancer due to mutations in genes that result in cellular overproliferation in certain tissues. The following link provides a well referenced summary (see Table 2b) of research to date.

In order for SENS to work, aging must be regarded as random accumulation of damage over time.


SENS would work - if it works - irrespective of whether damage over time is random or not. However as Aubrey states,

It's indeed correct that SENS doesn't care whether the damage is random -- except that if it were not, the "aging program" might "try harder" to age us if SENS caused it not to work as normal. However, and despite the number of attempts still being made to argue otherwise, I still agree with the mainstream view that aging is programmed only in a few unusual species like salmon.


I still think the brutally cathartic approach of WILT - if it works - would not be affected by an underlying aging program. Nevertheless, Aubrey's concern reveals why he would prefer that the added complication of an aging program not exist.

There are subtle hints of the existence of such a program, however. In yeast as in many other model organisms we observe that under certain environmental conditions (such as starvation) extended lifespan will be conferred - generally by upregulating somatic maintenance. Mutations in genes associated with somatic maintenance mechanisms or nutrient sensing mechanisms allow extended lifespan to be conferred irrespective of environmental conditions (i.e. in yeast - Gpr1, Ras2, cAMP, Sch9, Man2, SOD, etc. - [Longo & Finch 2003]). The modulation of such genetic pathways suggests that a program may exist that regulates the level of protection against damage. Another interesting observation in yeast is that long lived mutants have a lower incidence of DNA mutations compared to shorter lived wild type cells [Fabrizio 2003].

If the rate of DNA mutation is proportional to the speed of adaptation in changing environments it would not be unreasonable to consider that under certain conditions long lived cells would be at a disadvantage over shorter lived cells. The implications for higher organisms are self-evident.

#23 noam

  • Topic Starter
  • Guest
  • 34 posts
  • 0

Posted 04 January 2006 - 06:12 AM

There are subtle hints of the existence of such a program, however. In yeast as in many other model organisms we observe that under certain environmental conditions (such as starvation) extended lifespan will be conferred - generally by upregulating somatic maintenance.


It is possible to look at things from another angle:

First of all, as you state, the prolonged yeast lifespan phenotype observed under a condition of partial nutrient deprivation (also called "Calorie Restriction"), can also be triggered by a large amount of other minimal stress conditions (ala heatshock, osmotic stress, etc.). Experiments revealed, that if you subject the cell to just one of the minimal stressors, it becomes more immune to all the stressors. The phenomena has a name: Hormesis (and thanks to prometheus I know how to spell it).

Now, think about that: In the wild, practically all animals do suffer from a constant condition of different stresses, for example they are usally always hungry. Now, we bring those organisms to our lab, and let them grow without any stress. The result is that they live less than when they are subjected to a small stress.

In conclusion, what we are doing might be just proving that under an unnatural condition of no-stress, some mechanism in the organism will perform *less than normal*.

Under this view, calorie restriction in lab organisms does not prolong their normal life span, it is the lack of calorie restriction that shorten their potential normal life span.

A mechanistical explanation to what is going on, may be along the lines that maintenance proteins (ala DNA repair) are not by default turned on to an optimum level for the organism. Just a small amount of stress is needed in order for those maintenance proteins to work much more. If you remove all types of stress, you might put the maintenance proteins, "to sleep", sort of say. They do need this constant stimulation, in order to work more optimally, which is the natural state in the wild.

Edited by noam, 04 January 2006 - 01:56 PM.


#24

  • Lurker
  • 1

Posted 04 January 2006 - 06:33 AM

(hOrmesis). It's a valid point but it does not diminish the support that such studies lend to the prospect of an aging program.

#25 noam

  • Topic Starter
  • Guest
  • 34 posts
  • 0

Posted 04 January 2006 - 06:47 AM

I would like to rephrase (I also corrected the post above): It's not that the life expectancy of animals in nature is longer than in the lab (it is shorter, due to higher extrinsic forces), but it is possible the maintanace mechanisms of animals in the wild, in comparison to the lab, work to a higher level.

When you bring an organism to the lab, and grow it without: A. high extrinsic lifespan reducing factors it encounters in nature, and B. without the constant minimal stress it is used to from nature, the organism will live more than in nature, but it's possible this phenotype is only thanks to A, and not B.

Its maintenance mechanisms, as a consequence of the lack of sufficient stimulation, are now downregulated in comparison to their state in nature, and this leads to an unnatural bottleneck for life span, "unnatural" because that organism does have the ability to operate its maintance mechanisms to a better level in nature, but without any stress, it can't do it.

By inducing some minimal stress in the lab, it is possible that all we are doing is upregulating those maintenance mechanisms back to their natural level (not more than that!). The phenotype will of course be a prolonged life span. (in this way the organism enjoy both worlds at once : 1. low extrinsic life threatening force, 2. normal, and not reduced, level of maintenance mechanism).

If you choose to view things this way, in my opinion it does diminish the support that such studies lend to the prospect of an aging program.

Edited by noam, 04 January 2006 - 02:10 PM.


#26 John Schloendorn

  • Guest, Advisor, Guardian
  • 2,542 posts
  • 156
  • Location:Mountain View, CA

Posted 04 January 2006 - 01:56 PM

can someone explain to me what is "cancer" in yeast, and why if it happened, it would have implied they have better DNA repair mechnisms than if not ?

In humans, the argument is that genomic DNA mutations in non-cancer related genes do not matter as much as mutations in cancer-related genes. One mutation in one cell's GAPDH gene costs us a cell, which is no cost at all. One mutation in p19/ARF can kill us. So the evolution of genomic DNA repair was driven by cancer, rather than by mutations in non-cancer related genes and non-cancer related genes are protected much better than they would be if there were no cancer. Thus Aubrey's argument was that genomic DNA mutations do not play a significant role in human aging other than cancer. If there is no cancer in yeast, or it has no deleterious effects, then this argument would obviously not hold in yeast.

I'd like to qualify that: If there is no class of genes in yeast, the deletion of which has no greater negative effect than the loss of the cell in which it occurs, then the argument does not hold. However, I would suggest that there are such genes, namely those that create cells which continue to expand and consume resources, but are not optimally competitive (like cancer cells in us). For example, I vaguely remember one experiment where they let a wild-type strain compete with an apoptosis-deficient strain, and the apoptosis-deficient one lost only after a while. Anyway, this cannot be decided by rambling, but by experiment.

Drosophila (flies) do get and die of cancer due to mutations in genes that result in cellular overproliferation in certain tissues

Interesting, has this been shown to occur spontaneously in wild-type flies?

Under this view, calorie restriction in lab organisms does not prolong their normal life span, it is the lack of calorie restriction that shorten their normal life span

Although as for vertebrates, only small irregularities in the CR protocol can ruin the effect (basically to avoid malnutrition or other stresses), which is why life-extension by CR was until relatively recently regarded as a fringe science that suffered from lack of reproducibility. It is hard to conceive that nature would have stuck to the CR protocols to the degree that is required for CR laboratory experiments. Wild vertebrates rather seem to live a "feast and famine" lifestyle.
(It seems to me that this idea, too does not directly support an aging program. It could be a thousand other things, that I would try to summarize as "more complicated stress response and longevity determination in vertebrates than in yeast")

And most important question:

Assuming the SENS'ed yeast will now live significantly more (how much is singnificant, I'm not sure), do you think this will be worth the hard work ?

I think work in yeast is not going to strongly impress vertebrate scientists, who will at some point back-pedal thinking "not my field"... However, if we are aiming at the masses via pop science, the largest claimed live-extension should get us the greatest benefit. Sadly, this seems to work even if the reasoning leading from the experiment to the claim is rather dodgy. Remember the recent "yeast lives six times" sirtuin hype. If SENS could extend yeast replicative life span by six times in comparison to wild-type under standard condidtions (which I think is rather optimistic) people will say, bah old story, we've had that before.

However, I do think the experiment you envision will yield important implications for mammalian allotopic mtDNA expression and should generate credibility for SENS, which directly translates into donations for the MF. Especially, if the alternative is not doing anything relevant to SENS, I think it would be great if you had some way of going ahead with this.

#27 noam

  • Topic Starter
  • Guest
  • 34 posts
  • 0

Posted 04 January 2006 - 02:32 PM

Thanks for your explanation, John.

However, I do think the experiment you envision will yield important implications for mammalian allotopic mtDNA expression and should generate credibility for SENS, which directly translates into donations for the MF. Especially, if the alternative is not doing anything relevant to SENS, I think it would be great if you had some way of going ahead with this.

Well, "generating credibility for SENS" is a worthy goal, for me. As for my head of the lab, I don't think he'll do it unless he'll be convinced we can really do something impressive for the yeast lifespan.

#28

  • Lurker
  • 1

Posted 04 January 2006 - 02:47 PM

has this been shown to occur spontaneously in wild-type flies?


There are studies going as far back as 1975 on caffeine-induced tumors in drosophila larvae as well as other substances that are carcinogenic to this species. Naturally - and just as in humans - some strains are more sensitive than others to carcinogens.

In humans, the argument is that genomic DNA mutations in non-cancer related genes do not matter as much as mutations in cancer-related genes.


It should be noted that this is a SENS-centric hypothesis. Aubrey has a forthcoming paper where he explains his rationale on this.

So the evolution of genomic DNA repair was driven by cancer, rather than by mutations in non-cancer related genes and non-cancer related genes are protected much better than they would be if there were no cancer.


Unless damage to non-cancer related DNA - and perhaps RNA - is also harmful to the cell. More importantly, however, is that DNA repair rate is inversely proportional to mutation-driven adaptation. The pressure to adapt to changing circumstances beyond a cell's ability to respond from its genomic library presents a powerful selective force.

#29 ag24

  • Honorary Member, Advisor
  • 320 posts
  • 29
  • Location:Cambridge, UK

Posted 04 January 2006 - 04:13 PM

Yes, I had a feeling prometheus wouldn't let me get away with that simplification! Flies (like nematodes) are normally described as "wholly postmitotic", but this really refers only to the somatic cells of the adult: germ cells in the adult and many cells in the larva are mitotically competent. The question thus arises: if some tissues in a given organism are mitotically competent and others are not, does my logic break down in those that are postmitotic -- can DNA maintenance and repair have a fidelity that is tissue-specific? My opinion is that in principle it certainly can, and that this is quite likely to be the situation in flies and nematodes, but that in mammals there is sufficient potential for a postmitotic cell to transform a nearby mitotically competent cell by secreting growth-promoting/permitting things (this should really be officially named the "Campisi effect" by now) that it's important to be careful in the postmitotic cells too. The forthcoming paper of mine to which prometheus refers has a section is which I discuss this -- basically it's an argument that I think is a priori quite fragile but well supported by the available data.

#30 olaf.larsson

  • Guest
  • 583 posts
  • 21
  • Location:Sweden

Posted 04 January 2006 - 04:26 PM

I still agree with the mainstream view that aging is programmed only in a few unusual species like salmon.


My idea about the salomon phenoptosis is following:

The salomons die after mating becouse by dying in the place of mating they produce an increase in microorgansims in the water that can serve as food supply
for their offspring. If they would die in another place before mating a second time, and they most likely would, there would not be this beneficial effect for the offspring.




1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users