38 replies to this topic

[jaydfox]

Okay, I'm putting a brainstorming section here for myself more than anything, but feel free to jump in or, better yet, provide references that address my concerns.

The concerns I'm about to raise are not ones that are show-stoppers. I am merely looking for more information. I will be going through as much of the info on de Grey's website as possible in the coming weeks to get caught up, but I want to get something written down while it's still in my head.

Okay, Mitochondrial DNA (mtDNA) damage. Under some theories of aging, this is actually one of the most critical aspects of aging, and in one theory, it is the critical aspect. Regardless of which theory we go with, it's an important issue.

The basic problem is already well understood, so I won't waste a bunch of time trying to summarize it. In short, mitochondria produce a lot of reactive oxygen species (ROS), and these can damage mtDNA. Once mtDNA is damaged, dysfunctional proteins are produced, increasing the rate of ROS production. An obvious feedback cycle presents itself. What's worse, the ROS can escape the mitochondria and damage proteins, RNA, and most importantly, nuclear DNA.

Okay, now how do we fix this. Well, similar to the chromosomal DNA problem, one obvious answer is to increase mtDNA integrity. This can be accomplished through either or both of the following:
1) Reduce production of ROS (including reducing the lifetime of ROS after they are produced, e.g. antioxidants).
2) Increase mtDNA repair/maintenance.

Okay, fairly obvious. But how effective can these be? The literature so far is discouraging, but antioxidants, long held as the holy grail of anti-aging medicine, simply are not targetting effectively. In theory, if we accept that ROS production is the primary cause of mtDNA damage, and if we can reduce ROS by 90% with antioxidants, then this should have a profound impact on mtDNA damage. In experiment, it does not. One conclusion is that the theory that ROS causes mtDNA damage is wrong. I don't buy it. Another conclusion is that antioxidants are not mopping up ROS before it can cause the damage, but rather after. This would be especially true if antioxidants cannot get to or very near the source of ROS production in high doses.

So, we have room for improvement with targeted antioxidants, and some are in the research pipeline.

As for mtDNA repair/maintenance? Well, one solution is upregulation of key DNA Repair/Maintenance (kDRM) factors. Another solution is allotropic expression of currently mtDNA expressed genes.

The second option is clearly superior in one sense: ROS production in the mitochondria is not nearly as big an issue. To affect nuclear genes, the ROS would have to leave the mitochondria, traverse the cell to the nucleus, penetrate the nucleus, and then attack the DNA. It is for this reason that nuclear DNA is orders of magnitude more stable than mtDNA.

The first option, increasing kDRM factors, is inferior in that mtDNA remains in a very hostile environment.

[jaydfox]

Were these the only concerns, allotropic expression of mitochondrial DNA would indeed be preferable.

However, there are other considerations. I should state up front that none of these is a showstopper for me, I'm just trying to make sure that everybody (and not just the SENS experts) is aware of the issues.

First, expressing the genes in the correct amounts in the first place. Dr. de Grey has covered this:

Regulation: we're lucky there, because every one of the 13 mt-coded proteins is a subunit of a complex that also has nuclear-coded ones in 1:1 stoichiometry. So even if it turns out to be necessary to mimic existing expression patterns really well (and in fact there is some evidence that this is not actually very crucial, e.g. heterozygous deletion of nuclear-coded OXPHOS complex subunits never has a phenotype in flies), we can just slap our coding region into the regulatory DNA of such a nuclear gene.

There was some disagreement over this point at first, but eventually all sides agreed, so I'm satisfied.

Second, getting nuclear-coded proteins into the mitochondria. This involves two parts. First, there must be an existing mechanism for such imports that we can hijack, so to speak. de Grey covers this:

Your new question about which mitochondria in a given cell receive the protein... Answer: Most nuclear-coded mitochondrial proteins (including all those that are subunits of respiratory chain complexes) are directed to mitochondria by short amino-acid sequences attached to the start of the protein, which are removed after import. It was established a decade or two ago that these targeting sequences can be attached to any old protein and cause it to be imported into mitochondria (so long as it's not too hydrophobic). Exactly how an individual mitochondrion activates and deactivates its import machinery to turn import on and off is not well understood, but it is clear that when import is occurring it is not specific for a subset of mitochondrial proteins; rather, everything nearby that has the signal sequence is imported.

(my italicization)

Hydrophobicity is the next issue, and de Grey has covered it (though not to everyone's satisfaction, but the issue was dropped as far as I can remember).

Hydrophobicity: the main problem, but now seems very likely to be solvable just by judicious amno-acid substitutions that marginally reduce hydrophobicity without affecting function. Candidates can be found by looking in other species. The post-import modification idea is also plausible, and indeed a version of it, using inteins, was the main novel component of my TibTech paper mentioned above. But your concern is invalid, because (a) some such post-import modifications, inteins among them, do not require a peptidase but rearrange purely autocatalytically, and (b) even if a peptidase is needed, it will be a soluble protein so not hydrophobic, so its import will be easy.

Okay, so we've addressed production and import. But there's still the issue of transport. It's an open question for now, but I assume will be addressed shortly, if I don't find the answer on my own.
http://www.imminst.o...t=20#entry49665

However, the machinery that transports the nuclear-coded proteins to the correct mitochondria is in the cytoplasm, or at least partially in the cytoplasm, right? ...can this machinery be as easily coaxed into operating on the mtDNA-coded genes that we just moved to the nuclear genome? ...

Can we ensure that each mitochondrion will get the proper amounts of the proteins that we are moving out of the mtDNA?

I had a lot of questions on the mitochondria thing, and de Grey answered all of them except this one. However, John Schloendorn offered an answer:

If this question means something like if 50 subunits are needed, how do we assure one mito does not get 75 subunits, while another one gets only 25, then the answer is simple. We don't. For statistical reasons, distributions close to the average are just much more likely than distributions that are far away. This is ultimately due to the same forces that drive the universe towards ever increasing entropy, but I would recommend to google a little bit for introductory thermodynamics, which can be quite revealing about this kind of phenomena.

to which I replied:

I hope it is that simple, but prudence tells me that I cannot assume that, I must see it verified in experiment.

There are a great many reasons I can think of why mitochondria will not require relatively similar amounts of various proteins at various times.

First, ...
Second, ...
...
Fifth, ...

The list could go on and on. Some of the items on the list may not be concerns. Others definitely will be.

So I can't rely on statistical distribution to fix the problem. We need to know that there is a system that not only ensures proper transport into a mitochondrion upon arriving at one (which de Grey covered), but that we get the proteins to the mitochondria in the first place.

Now, if nature doesn't already have machinery to do this, but relies instead on statistical uniformity to get the job done, then there's no work on our part. BUT, if there IS an existing machinery to cause a non-uniform distribution of proteins, then we must ensure that the 13 new allotropic genes take advantage of that same machinery.

Okay, this is where that question stands, and once answered, I think it will help solidify the case for the technical hurdles. Unless anyone else knows of any good technical hurdles?

[jaydfox]

I can't find the reference to who brought the issue up, so I apologize. Somebody brought up the issue of what would happen to the 13 mtDNA genes once they are expressed allotopically.

Since I don't have the reference, I'm going to wing it.

What does happen? Do we allow the mtDNA to continue producing proteins? This has two big problems, and another small one:

1) This upsets the 1:1 stoichiometry that de Grey used to justify the relative ease of moving mtDNA to the nucleus in the first place.
2) If damaged versions of proteins are built by damaged mtDNA, then what is to prevent the damaged proteins from being used instead of the supply of pristine proteins being transported/imported from the nucleus?
3) Extra work: raw materials, energy, and production capacity are used up on unnecessarily redundant proteins, possibly taxing the cell further and adding (minutely, probably) to the very damaging processes we're trying to avoid.

Issue 3 is one of effeciency, and is probably not critical. The first two, however, are.

Unless someone can address both of these issues, the only tenable solution is to stop the production of those proteins within the mitochodria.

So how do we stop the production? One option is to simply prevent the mitochondria from expressing these genes, or if that fails, interfering with the RNA, thus preventing protein synthesis.

OR, we could delete the 13 genes from the mitochondrial DNA, along with their regulatory systems. Or, if there's no harm in expressing a loop of DNA that doesn't encode any proteins, then we can just delete the genes, and ignore the regulatory system.

Option 1 seems unlikely to be successful, given the A) long lifespan we're facing, with chances for screwups, B) the difficulty of affecting all the mitochondria all the time.

Option 2 seems better, especially given A above, and affecting B can be simply a matter of multiple procedures or one very comprehensive procedure that reaches as many mitochondria as possible.

Both options seem to remain major technical hurdles. However, science continues to advance, so I'm optimistic for now.

However, if we do delete genes from the mtDNA, then we have another problem: Sexual reproduction.

If a woman has her 13 mtDNA genes deleted and moved to the nucleus, what happens to her kids? They'll get her modified mitochondria, and only one set of the mtDNA genes in the nucleus. de Grey said that at least one heterozygous genotype in flies did not have a phenotype, but that's a lot to bet one's children on.

Of course, it would be simpler if only immortalists who'd undergone the treatment mated, but this is obviously obsurd. Unless a socially-driven class division is to be instituted (such as don't mate with a midget or someone with Down's syndrome or who hasn't had a SENS procedure), it will be completely impractical.

Okay, that means that mtDNA deletion is probably out of the question, at least for people still interested in having kids, especially if they don't already have a life-partner who could also undergo the treatment.

So we're back to gene silencing. Are there other options?

[John Schloendorn]

Another conclusion is that antioxidants are not mopping up ROS before it can cause the damage, but rather after.

When a ROS causes damage, it's gone. They react with their substrates, becoming part of them, so to speak, and that constitutes the damage. So every antioxidant that scavenges a ROS potentially saves another molecule.

This would be especially true if antioxidants cannot get to or very near the source of ROS production in high doses.

The idea that the problem is one of localization probably comes from the evidence that endogenous free radical defenses are concentrated to the mitochondria. [1] I don't know if there is anything known about their submitochondrial localization, e.g to the respiratory chain complexes.

John: If this question means something like if 50 subunits are needed, how do we assure one mito does not get 75 subunits, while another one gets only 25, then the answer is simple. We don't.
Jay: Now, if nature doesn't already have machinery to do this, but relies instead on statistical uniformity to get the job done, then there's no work on our part. BUT, if there IS an existing machinery to cause a non-uniform distribution of proteins, then we must ensure that the 13 new allotropic genes take advantage of that same machinery.

There is indeed no such system known for transport to any subcellular organelle. Cellular organelles of each type are, to the best of my knowlege, indistinguishable.

Do we allow the mtDNA to continue producing proteins? This has two big problems, and another small one:
1) This upsets the 1:1 stoichiometry that de Grey used to justify the relative ease of moving mtDNA to the nucleus in the first place.

Good point. It is possible that mtdna in mitos that do not fulfill the 1:1 stochiometry becomes relatively rapidly degraded due to increased ROS production. But this most probably be undesirable due to ROS leakage.
One could instead target a soluble unspecific nuclease to the mitochondria to clean up. The nuclease could be expressed from a somatic cell, or postmitotic cell specific promoter to restict its effect to the desired tissues.
If immortals insist to have children with non-immortals, I would recommend IVF followed by some genetic engineering, such as deletion of the nuclease, if the child is destined to be mortal, and implementation of all SENS, if he is destined to be immortal...

[John Schloendorn]

However, the structural integrity of the mtDNA molecules could be required for proper mitochondrial division and its regulation. If that turns out to be so, one could

- use an RNAse to scavenge the mRNAs instead (easiest, but eventually metabolic strain)
- try to delete the mitochondrial ribosomal subunits in the nuclear DNA (I can't think of a way to do this in the adult. But if this sci-fi technology is being developed in the course of WILT anyways, in order to achieve site-specific telomerase deletion, then we may as well use it for this one [thumb] ) or else knock them down with RNAi
- attempt to engineer a wobbly recombinase, or a DNA polymerase, such as this one, just much more error-prone, as to erase the mtDNA sequence information content, but leave the molecular structure intact. (perhaps a good compromise)
- pharmacologically inhibit mitochondrial protein synthesis (e.g. chloramphenicol) to bridge the time while some of the above rearrangements take place

Edited by John Schloendorn, 16 February 2005 - 02:26 AM.

[Lazarus Long]

Have you considered that the mitochondrion is the only organelle that is passed down, in its entirety with no recombination from another nucleus (as the nucleus is) from female to female down each generation?

The reason for this is probably associated with the fact that Mitos are reproducing asexually in a mitotic fashion consistent with their genomic character. We might be better off treating mtDNA with techniques developed for bacteria and prokaryotes IMHO.

However one caveat, if I remember correctly there is an anomaly associated with paternal mtDNA that appears in the offspring from time to time and I also wonder if when this occurs, a kind of *episomal recombination* that can also occur, with the possibility contributing to germline mutagenesis for mitochondria. There do appear to be examples of very limited but also very relevant mtDNA mutations that may be the result of cross parental recombination, and some genetic related disease. Could this be the result of what are essentially mtDNA *plasmids* from the father?

(I will go back for cites on this later)

The problem here, at least in part is that mitos are not just like other organelles, they carry their own genes. They are quasi independent, in some senses analogous to intestinal micro-flora.

I also have wondered if we should focus on what has evolved during the *oocyte production* phase in terms of discovering how the female appears to both mass produce mitos and also set their somatic mutation clock back to *zero* during the production of oocytes.

I have came to the opinion that this may also have something to do with embryonic stem cells as well because the oocytes are actually manufactured during the gestation of the embryo. One key is to find at what specific embryonic phase the oocytes are produced and then decipher all the associated mechanisms.

What I am interested in is that there may be an associated instruction on the *host's* nuclear DNA that can organize mtDNA reproduction during this phase. If it does, I suspect the process mimics bacterial genetics and that is one reason we may only see it function once *normally*.

The second concern is that mtDNA concentration and amplification in the oocyte is at least in part a result of the meiotic cell division that discards the nuclear DNA of the polar bodies and moves the cytoplasm from those bodies to the oocyte. If the increase in proportional quantity of mitos is the result of simply this process then the ratio of mitos will be predictable as roughly 4:1 and the result is not from a complex instruction to reproduce in quantity but the almost mechanical theft of the cytoplasm from the discarded polar bodies.

However, aside from the concentration of mitos in the oocyte there is a second period after fertilization when there appears to be a *bloom* of mitos associated with the same period we observe with undifferentiated Stem Cells in the zygote as the mitos are being made available for later differentiation.

Lastly another anomaly associated with oocytes that influences Mitos and their DNA that could influence what we are looking for is something that puts them in relative suspension for aging while in the oocyte. I am inferring this from the obvious but please examine the fact that if all the oocytes are produced while the mother is still an embryo then aging for these oocytes is significantly slowed in relation to most other cells in the body.

Yes oocytes age for a variety of reasons but not for the same exact reasons of oxidative stress as most other cells. They appear to have their aging stopped until released in menstruation and then their clock starts and they either die and are discarded after a designated period or are fertilized. They do not grow or consume much nutrient/oxygen at all during their stasis in the ovaries.

But we must ask ourselves how it is that the female mitochondrion manages to survive from fertilized female egg to developed female to fertilized female egg comparatively intact - a process that continues along the female germline indefinitely.

So while I agree that in order to prevent long term *somatic* mutation of the mtDNA germline as it is transferred from generation to generation that a *renewal or reset* instruction may (and probably does) exist, we must also see that a large part of the reason we can see very little aging of the maternal mtDNA is that it is simply not involved in the normal aging process of the mother at all but is being reproduced during her zygotic phase from mtDNA that was last used during her mother's embryonic phase (and so on) and is being kept in a kind of *stasis* in the ovaries like a *vault*.

[Michael]

All:

Okay, Mitochondrial DNA (mtDNA) damage. Under some theories of aging, this is actually one of the most critical aspects of aging, and in one theory, it is the critical aspect. Regardless of which theory we go with, it's an important issue.

The basic problem is already well understood, so I won't waste a bunch of time trying to summarize it. In short, mitochondria produce a lot of reactive oxygen species (ROS), and these can damage mtDNA. Once mtDNA is damaged, dysfunctional proteins are produced, increasing the rate of ROS production. An obvious feedback cycle presents itself. What's worse, the ROS can escape the mitochondria and damage proteins, RNA, and most importantly, nuclear DNA.

Ah, time must be wasted . The above is the popular understanding of the mtROS theory/theories of aging, still trotted out by supplement companies and by many scientists who think they know something about the role of mt in aging. However (as de Grey pointed out ages ago (1,2,2a)), this "vicious cycle" theory just doesn't fit with the empirical evidence -- a fact which has become even clearer with time. As a result, there are still people who, having not read de Grey, are still announcing that the mt theory/theories of aging are incompatible with the data (3) -- which, again, the standard ("vicious cycle") theories are not. This understandably makes Aubrey a bit frustrated ((4) -- and note the testy title ), since he explained all of this years ago, and provided a viable (if horrendously more complicated) alternative mitochondrial free radical theory of aging (MiFRA).

On top of that, "most importantly, nuclear DNA" definitely does not apply. The rate of aging in mammals is, on good interspecies evidence (5,7), related to the rate of accumulation of mt, but not nuclear, DNA damage and esp deletions.

Okay, now how do we fix this. Well, similar to the chromosomal DNA problem, one obvious answer is to increase mtDNA integrity. This can be accomplished through either or both of the following:
1) Reduce production of ROS (including reducing the lifetime of ROS after they are produced, e.g. antioxidants).
2) Increase mtDNA repair/maintenance.

The problem with the latter is that one cannot repair or maintain mtDNA against deletions: the genome cannot look outside of itself for a template to tell it with what to replace the missing chunk. Since it is deletions which actually accumulate with age, and which seem to be related to aging (see again de Grey's papers), the latter won't get us far.

The third option, as you note later, is allotopic nuclear expression of the mt-coded genes.

if we accept that ROS production is the primary cause of mtDNA damage, and if we can reduce ROS by 90% with antioxidants, then this should have a profound impact on mtDNA damage. In experiment, it does not. One conclusion is that the theory that ROS causes mtDNA damage is wrong. I don't buy it.

Good call. CR lowers mtROS production (5) -- which we can be confident on interspecies grounds and from the CR example are indeed involved in aging (5,7) -- and also reduces accumulation of deletions (6). Importantly, CR reduces the generation, but not the progression, of "ragged red" (COX negative and SDH hyperreactive (COX-/SDH++)) phenotype in muscle, which is a functional result of the "common mtDNA deletion."

Another conclusion is that antioxidants are not mopping up ROS before it can cause the damage, but rather after. This would be especially true if antioxidants cannot get to or very near the source of ROS production in high doses.

And we know that this is essentially true (5,7,8). This explains why mtROS generation and endogenous antioxidant levels are both inversely proportional to species-specific LS & why CR lowers the former but has no consistent effects on the latter: mtROS drive aging, and when you lower production there is less need to try to mop up the mess post hoc.

The second option is clearly superior in one sense: ROS production in the mitochondria is not nearly as big an issue. To affect nuclear genes, the ROS would have to leave the mitochondria, traverse the cell to the nucleus, penetrate the nucleus, and then attack the DNA. It is for this reason that nuclear DNA is orders of magnitude more stable than mtDNA.

You're right about the reasons for the greater stability of mt vs nuDNA, but I don't see how this supports the superiority of increasing mtDNA maintenance or that ROS production in mt is not as big an issue.


-Michael

1. de Grey AD. The reductive hotspot hypothesis of mammalian aging: membrane metabolism magnifies mutant mitochondrial mischief. Eur J Biochem. 2002 Apr;269(8):2003-9. Review. PMID: 11985576 [PubMed - indexed for MEDLINE]

2. de Grey AD. The mitochondrial free radical theory of aging. 1999; Austin, TX: Landes Bioscience. (ISBN 1-57059-564-X).

2a. de Grey AD. A proposed refinement of the mitochondrial free radical theory of aging. Bioessays. 1997 Feb;19(2):161-6. Review. PMID: 9046246 [PubMed - indexed for MEDLINE]

3. Jacobs HT.
The mitochondrial theory of aging: dead or alive?
Aging Cell. 2003 Feb;2(1):11-7.
PMID: 12882330 [PubMed - indexed for MEDLINE]

4. de Grey AD.
Mitochondria in homeotherm aging: will detailed mechanisms consistent with the evidence now receive attention?
Aging Cell. 2004 Apr;3(2):77. No abstract available.
PMID: 15038822 [PubMed - indexed for MEDLINE]

5. Barja G.
Endogenous oxidative stress: relationship to aging, longevity and caloric restriction.
Ageing Res Rev. 2002 Jun;1(3):397-411. Review.
PMID: 12067594 [PubMed - indexed for MEDLINE]

6. Lee CM, Aspnes LE, Chung SS, Weindruch R, Aiken JM.
Influences of caloric restriction on age-associated skeletal muscle fiber
characteristics and mitochondrial changes in rats and mice.
Ann N Y Acad Sci. 1998 Nov 20;854:182-91. Review.
PMID: 9928429 [PubMed - indexed for MEDLINE]

7. Barja G.
Rate of generation of oxidative stress-related damage and animal longevity.
Free Radic Biol Med. 2002 Nov 1;33(9):1167-72. Review.
PMID: 12398924 [PubMed - indexed for MEDLINE]

8. de Grey AD. The non-correlation between maximum lifespan and antioxidant enzyme levels among homeotherms: implications for retarding human aging. J. Anti-Aging Med. 3(1):25-36.

9. Bua E, McKiernan SH, Aiken JM.
Calorie restriction limits the generation but not the progression of
mitochondrial abnormalities in aging skeletal muscle.
FASEB J. 2004 Mar;18(3):582-4. Epub 2004 Jan 20.
PMID: 14734641 [PubMed - indexed for MEDLINE]

[jaydfox]

I will read your references before responding, to avoid wasting any more time.

[jaydfox]

Except (2), which is $115 web-price, list-price $139. I apologize, as that ref might address the crux of the matter.

[Lazarus Long]

Except (2), which is $115 web-price, list-price $139. I apologize, as that ref might address the crux of the matter.

This sounds like a good start for the reference sharing effort.

[Michael]

All:

First: Prometheus, I got a mess of pseudo-html when I tried to respond to your post using the "Quote" function. I run into this sometimes on Imminst forums. Did you originally respond using the "Quote" function, or some alternative posting mechanism?

[quote]
[quote]
The rate of aging in mammals is, on good interspecies evidence (5,7), related to the rate of accumulation of mt, but not nuclear, DNA damage and esp deletions.[/quote]
Good God! How on earth do you make such an authoritative sounding statement like that?[/quote]

Because that's what the evidence shows .

[quote]
Lets look at your references. [In (5),] Barja's main contention in this paper is that the rate of mtDNA damage is higher that nDNA damage ... No one disagrees with the premise that mtDNA has more oxidative damage than nDNA.

However, where is the connection made that mtDNA damage is the limiting factor in maximum lifespan, in contrast to nDNA?[/quote]

From the fact that "long-lived animals show lower levels of oxidative damage in their mitochondrial DNA (mtDNA) than short-lived ones, whereas this does not occur in nuclear DNA (nDNA)." (5) This means longevous species, not individual organisms, and is measured in terms of max LS. Likewise, "the rates of mitochondrial oxygen radical generation [and] oxidative damage to mitochondrial DNA" "are negatively correlated with maximum longevity." (7) This originally goes back to a series of comparative studies, most notably (10):

[quote]
If oxidative damage to DNA is involved in aging, long-lived animals (which age slowly) should show lower levels of markers of this kind of damage than short-lived ones. ... In this study, steady-state levels of ... (8-oxodG) referred to deoxyguanosine (dG) were measured ... in the mitochondrial (mtDNA) and nuclear (nDNA) DNA from the heart of eight and the brain of six mammalian species ranging in maximum life span (MLSP) from 3.5 to 46 years. ... 8-oxodG/dG in nDNA did not correlate with MLSP across species either in the heart (r=-0.68; P<0.06) or brain (r = 0.53; P<0.27). However, 8-oxodG/dG in mtDNA was inversely correlated with MLSP both in heart (r=-0.92; P<0.001) and brain (r=-0.88; P<0.016) tissues following the power function y = a(.)x(b), where y is 8-oxodG/dG and x is the MLSP.[/quote]

... and (11) also finds a more consistent relationship of mtDNA than nuDNA damage to maximum LS, somewhat more weakly because based on comparisons across 2 different orders of vertebrates who already show unexpected differences in rate of mtROS generation based on body size and metabolic rate, rather than within one order as in (10):

[quote]
Lower steady-state 8-oxodG values were observed in all cases in the heart mtDNA in birds than in mammals. 8-oxodG levels were also lower in brain mtDNA in pigeons than in rats, in brain nDNA in canaries than in mice, and in heart nDNA in parakeets compared with mice. The rest of the comparisons did not show significant differences between species. These results taken together indicate that oxidative damage to DNA tends to be lower in birds (highly long-lived species) than in short-lived mammals, especially in the case of mtDNA.[/quote]

The fact that CR does not lower mtDNA deletions brain (12,13), unlike in mucle, heart, and liver, may also be important to interpreting this finding.

[quote]
Barja does not dare make this assertion directly since there is no evidence to support it, instead he hints that maximum lifespan (MSLP) is negatively correlated in heart and brains with 8-oxodG formation in mtDNA but not in nDNA.[/quote]

But that is exactly evidence that "The rate of aging in mammals is, on good interspecies evidence (5,7), related to the rate of accumulation of mt, but not nuclear, DNA damage", as I'd said.

[quote]
On the other hand Barja contradicts his claim by reporting on the findings that increased 8-oxodG levels are equally present in aging organs:
[quote]
the majority of the investigations have shown that tissue 8-oxodG levels in brain, heart or liver nDNA or mtDNA are moderately higher in old than in mature adult rodents or humans[/quote][/quote]

Note that this isn't actually a contradiction. To observe the fact that some kind of damage increases, and even accumulates, with age is distinct from showing that the rate at which this happens it is related to the rate of aging. To show the latter, we need evidence either from intervention (interventions that reduce a particular kind of damage with age increase max LS, and interventions that do not do the former do not do the latter) or from interspecies comparisons (more longevous spp suffer less of this damage with age than more short-lived ones). Barja and others have shown that this is true of oxidative mtDNA and not of nuDNA.

[quote]
And one more thing, you may be interested to know that your mentor is not fond of 8-oxodG as an indicator of DNA damage:

[quote]
One of the most pervasive errors in DNA analysis is to presume that rises in the amount of a pre-mutagenic lesion translate to proportional rises in that of bona fide mutations: in fact the relationship is nowhere near that, because the chance of a given 8oxodG becoming a mutation depends on its halflife, i.e. how long on average before it is repaired.[/quote][/quote]

Aubrey is not saying that 8-OHdG is not an indicator of any damage to DNA, but hat it is not a reliable proxy of actual mutations. 8-OHdG is not an accumulating lesion but a steady-state damage snap shot, and in fact it directly measures the rate of repair of oxidative DNA lesions rather than their rate of formation. But you have to swallow the whole pill. It's fortunate that the evidence from the CR model avaialable, as it has drawn the link to actual damage accumulation in the form of mtDNA deletions (5-9, 12, 13). Likewise, the fact that 8-OHdG is more readily repaired in nuDNA than mtDNA further strengthens the non-relevance of the former in aging (although, of course, it causes cancer).

On this front, it should be noted that while individual cells and their progeny certainly can be expected to become dysfunctional when their nuDNA aquire mutations, the low rate of cell division in vivo in most tissues -- and the virtual nonexistance of same in postmitotic tissues like heart and brain -- means that individual cell's mutations get little chance to "take over" the tissue and render the whole dysfunctional. As I suggested in reply to Estep, age-related changes in nuclear gene expression appear to be most clearly secondary to other, primary lesions:

[quote][quote]
Many recent experiments, especially large-scale microarray transcript profiles of aging, support this general concept of time-dependent decay of orderliness (for examples, see references [21, 22, and 23]).[/quote]

However, such changes are in fact almost certainly addressed by the SENS platform because there is a very strong case to be made that such shifts in gene expression with age are secondary to other changes whcih are themselves subject to the SENS panel of interventions.

That is: such shifts are either the result of an "aging program," or they are secondary to some primary, and fundamentally entropic, aging process. Since the former is rejected on theoretical grounds by the consensus of researchers into the role of evolutionary pressure into aging (8), the latter must hold. And indeed, the most useful investigations into such shifts -- those in which shifts in gene expression associated with normal aging are compared with those undergone in animals subjected to calorie restriction (CR) (29), which (as Estep well knows) is the sole intervention known to retard biological aging in mammals. These studies have found that the most prominent classes of genes undergoing shifts which both occur with aging are retarded by CR (and which are thus most likely related, as cause or as effect, to primary aging processes) are those involved in inflammation and antioxidant defense -- gene classes, that is, whose natures imply precisely that their expression has been altered in response to underlying, [primary] molecular lesion(s). [Compare the parallel findings in humans (30)]. But all such entropic processes appear to be embraced by the SENS platform (24,25); therefore, the redressing of such processes via the SENS panel of interventions is predicted to obviate the secondary shifts in gene expression associated with aging.[/quote]

By contrast, of course, cells with cancerous lesions by their very nature quickly multiply to pathological levels.

It seems obvious to me (and I expect that Aubrey agrees) that it is likely that nuDNA mutations would eventually become pathological if not repaired over the course of a greatly extended LS, as eventually all cells would have accumulated a great many true mutations; however, evidence to hand indicates that they are not accumulating at high enough levels over the course of a "normal" lifespan to significantly contribute to aging per se. Of course, WILT will itself deal with much of this by replacing and/or supplementing such cells periodically wth pristine stem cells.

[quote]
Perhaps Michael, you can enlighten us as to why you think oocyte mitochondria appear to be so comparatively robust - potentially biologically immortal, in fact.[/quote]

Careful. Individual oocyte mt are not necessarily terribly robust, nor are oocytes themselves. The immortality and agelessness of the germ line has been rather misrepresented. One of the main reasons that the germ line is retained intact is that the body is so much more rigorous in apoptosing (neologism!) defective cells in the line -- not that the cells themselves are individually retained pristine. The body selects for healthy ova using atresia, keeping ova quiescent until ovulation, and even more rigorous selection of teh fittest during oogenesis. Also, defective mitochondria in the germ line (I seem to recall -- but perhaps others can provide either documentation or correction) are more likely to lead to flat-out cell death than the same phenomenon in somatic cells.

Result: a woman is born with 1-2 million ovarian follicles; by puberty she has only 300,000 & despite the fact that ovulation per se only leads to the "wastage" of 1 (or a few) eggs per month, only a few hundred remain at menopause.

Really, then, the germline is only immortal/ageless in the sense that the species is: individuals die, but the line passes forward, from generation unto generation.

I now see that Aubrey has elsewhere given reasons why mt in oocytes also have less damage -- but again, not because the mt are actually, individually more robust:

[quote]
It's complicated. (a) When the germ line is rapidly dividing (in early embryogenesis) there may be selection against mutant mitochondria (see below). (b) When it's not dividing (in the oocyte during the mother's life until fertilisation) the host cell has very low energy requirements so is respiring very little, so is producing few mutagenic free radicals. © The mitochondria that get into the oocyte are apparently put through a population bottleneck, which means that if any mutant mitochondria get into a given oocyte then it is virtually guaranteed that lots will; this is good because it will cause that oocyte to fail to ovulate (or to abort very early in embryogenesis) whereas a small number of mutants may not kill the offspring until much later. Kearns-Sayre syndrome is likely to be a case of this last trick not working.[/quote]

Whatever you think about the value of keeping up your germ line, let's fix the problem in individual humans, people! Give till it hurts to the Methuselah Mouse Prize, and take political action for anti-aging biomedicine.

[quote]
I will read your references before responding, to avoid wasting any more time.

Except (2), which is $115 web-price, list-price $139. I apologize, as that ref might address the crux of the matter.[/quote]

If you actually want to own a hard copy, you can get it used for ~US$86:

http://www.bookfinde...64X&mode=direct

Alternatively, it appears that the entire volume is available online from Eurekah:

http://www.eurekah.c...kid=44&catid=42

For US$39, you can get access to their entire online library, including this, for a year.

As well, I'm sure you can get it for free or for a few dollars via interlibrary loans.

de Grey's book is unbelievably useful: in addition to explaining his MiFRA in greater detail than is available elsewhere, it provides an extremely readable introduction to mitochondria generally. I highly reccomend it.

-Michael

4. de Grey AD.
Mitochondria in homeotherm aging: will detailed mechanisms consistent with the evidence now receive attention?
Aging Cell. 2004 Apr;3(2):77. No abstract available.
PMID: 15038822 [PubMed - indexed for MEDLINE]

5. Barja G.
Endogenous oxidative stress: relationship to aging, longevity and caloric restriction.
Ageing Res Rev. 2002 Jun;1(3):397-411. Review.
PMID: 12067594 [PubMed - indexed for MEDLINE]

6. Lee CM, Aspnes LE, Chung SS, Weindruch R, Aiken JM.
Influences of caloric restriction on age-associated skeletal muscle fiber characteristics and mitochondrial changes in rats and mice.
Ann N Y Acad Sci. 1998 Nov 20;854:182-91. Review.
PMID: 9928429 [PubMed - indexed for MEDLINE]

7. Barja G.
Rate of generation of oxidative stress-related damage and animal longevity.
Free Radic Biol Med. 2002 Nov 1;33(9):1167-72. Review.
PMID: 12398924 [PubMed - indexed for MEDLINE]

9. Bua E, McKiernan SH, Aiken JM.
Calorie restriction limits the generation but not the progression of
mitochondrial abnormalities in aging skeletal muscle.
FASEB J. 2004 Mar;18(3):582-4. Epub 2004 Jan 20.
PMID: 14734641 [PubMed - indexed for MEDLINE]

10. Barja G, Herrero A.
Oxidative damage to mitochondrial DNA is inversely related to maximum life span
in the heart and brain of mammals.
FASEB J. 2000 Feb;14(2):312-8.
PMID: 10657987 [PubMed - indexed for MEDLINE]

11. Herrero A, Barja G.
8-oxo-deoxyguanosine levels in heart and brain mitochondrial and nuclear DNA of
two mammals and three birds in relation to their different rates of aging.
Aging (Milano). 1999 Oct;11(5):294-300.
PMID: 10631878 [PubMed - indexed for MEDLINE]

12: Kang CM, Kristal BS, Yu BP.
Age-related mitochondrial DNA deletions: effect of dietary restriction.
Free Radic Biol Med. 1998 Jan 1;24(1):148-54.
PMID: 9436624 [PubMed - indexed for MEDLINE]

13: Cassano P, Lezza AM, Leeuwenburgh C, Cantatore P, Gadaleta MN.
Measurement of the 4,834-bp mitochondrial DNA deletion level in aging rat liver and brain subjected or not to caloric restriction diet.
Ann N Y Acad Sci. 2004 Jun;1019:269-73. Review.
PMID: 15247027 [PubMed - indexed for MEDLINE]

21. Ly DH, Lockhart DJ, Lerner RA, Schultz PG. Mitotic misregulation and human aging. Science. 2000 Mar 31;287(5462):2486-92.

22. Whitney AR, Diehn M, Popper SJ, Alizadeh AA, Boldrick JC, Relman DA, Brown PO. Individuality and variation in gene expression patterns in human blood. Proc Natl Acad Sci U S A. 2003 Feb 18;100(4):1896-901. Epub 2003 Feb 10.

23. Welle S, Brooks AI, Delehanty JM, Needler N, Thornton CA. Gene expression profile of aging in human muscle. Physiol Genomics. 2003 Jul 07;14(2):149-59.
>
24: De Grey AD.
Challenging but essential targets for genuine anti-ageing drugs.
Expert Opin Ther Targets. 2003 Feb;7(1):1-5.
PMID: 12556198 [PubMed - in process]

15: de Grey AD, Ames BN, Andersen JK, Bartke A, Campisi J, Heward CB, McCarter RJ, Stock G.
Time to talk SENS: critiquing the immutability of human aging.
Ann N Y Acad Sci. 2002 Apr;959:452-62; discussion 463-5.
PMID: 11976218 [PubMed - indexed for MEDLINE]

29. Weindruch R, Kayo T, Lee CK, Prolla TA.
Gene expression profiling of aging using DNA microarrays.
Mech Ageing Dev. 2002 Jan;123(2-3):177-93.
PMID: 11718811 [PubMed - indexed for MEDLINE]

30. Lu T, Pan Y, Kao SY, Li C, Kohane I, Chan J, Yankner BA.
Gene regulation and DNA damage in the ageing human brain.
Nature. 2004 Jun 24;429(6994):883-91. Epub 2004 Jun 09.
PMID: 15190254 [PubMed - indexed for MEDLINE]v

[jaydfox]

I asked:

Somebody brought up the issue of what would happen to the 13 mtDNA genes once they are expressed allotopically.

Since I don't have the reference, I'm going to wing it.

What does happen? Do we allow the mtDNA to continue producing proteins? ...

1) This upsets the 1:1 stoichiometry that de Grey used to justify the relative ease of moving mtDNA to the nucleus in the first place.
2) If damaged versions of proteins are built by damaged mtDNA, then what is to prevent the damaged proteins from being used instead of the supply of pristine proteins being transported/imported from the nucleus?
3) ...

BioEssays Vol. 19 no. 2
A proposed refinement of the mitochondrial free radical theory of aging
Aubrey D.N.J. de Grey
Accepted 1 November 1996
Pages 161-166

Page 165, left column, just before Conclusions

Problems of stoichiometry may arise, due to simultaneous expression of the mt-coded and transgenic copies of genes, but this can if necessary be tackled by, for example, disruption of nuclear-coded components of the mitochondrial transcription and translation machinery.

From the original paper! Figures, shame on me...

Not a detailed solution, but at least he brought it up. And this was still early theoretical framework, leaving the details to the lab scientists.

[jaydfox]

Another conclusion is that antioxidants are not mopping up ROS before it can cause the damage, but rather after. This would be especially true if antioxidants cannot get to or very near the source of ROS production in high doses.

And we know that this is essentially true (5,7,8). This explains why mtROS generation and endogenous antioxidant levels are both inversely proportional to species-specific LS & why CR lowers the former but has no consistent effects on the latter: mtROS drive aging, and when you lower production there is less need to try to mop up the mess post hoc.

(my emphasis added)
It explains why those levels are inversely proportional, but only to the extent that scavenging mtROS yielded diminishing, but not necessarily zero, returns, which were not sufficient to offset the extra cost of producing those enzymes. de Grey's point was that lifespan-affecting systems were affected proportional to their relative cost/benefits, so that increasing antioxidant enzyme levels became less important than decreasing leakage and oxidizability (8). Less important, not unimportant. The same could be said of any lifespan-affecting system: tweaking it will only be affective to the degree that other systems don't still lead to mortality or induce age-related damage. For example, if you completely solve the leakage problem of the inner membrane (thus preventing any superoxide from being in the MIMS), antioxidants would not be completely obviated. There's still the issue of oxidizability in the inner surface of the inner membrane (where the mtDNA resides), so if oxidizability is not also addressed, antioxidants will still be needed. Thus no one system gets overoptimized by evolution. But eliminating all leakage form the inner membrane wouldn't be an undesirable goal, would it?

Just because evolution failed to optimize a system for longevity, that doesn't mean that we can't do better. Increasing antioxidants, if properly targeted, and especially if in conjunction with other systems (remember, we want to affect all lifespan-relevant systems, hence the comprehensiveness of SENS), is not an undesirable goal.

Just a point of clarification. I'm not arguing against allotopic mtDNA expression. Just pointing out that I still believe that mopping up antioxidants can work, if targetted correctly.

Speaking of which, in (8) de Grey addresses the very issue of targetting anti-oxidants to the MIMS, the very region that he hypothesized was causing the seemingly counter-intuitive negative correlation of antioxidants and MLSP. Which is what I was saying sort of: the antioxidants weren't in the right place. I was assuming they needed to be closer to the inner surface of the inner membrane, but de Grey concluded the need to be in the MIMS.

Prometheus, remember that peptide anti-oxidant you pointed out? It targets the inner membrane (j1), but the abstract you posted didn't specify which side of the membrane. Even if it's on the inner surface (i.e. not in the MIMS), that would presumably have some impact on the leakage factor and oxidizability of that inner surface, both of which is positively correlated with MLSP.



(8) de Grey AD. The non-correlation between maximum lifespan and antioxidant enzyme levels among homeotherms: implications for retarding human aging. J. Anti-Aging Med. 3(1):25-36.

(j1) J. Biol. Chem., Vol. 279, Issue 33, 34682-34690, August 13, 2004
Cell-permeable Peptide Antioxidants Targeted to Inner Mitochondrial Membrane inhibit Mitochondrial Swelling, Oxidative Cell Death, and Reperfusion Injury

[jaydfox]

So I wonder that in addition of increasing DNA repair systems in mitochondria (as well as the nucleus), should we be not investigating how we may increase or the turnover or fission of mitochodrial organelles within a post-mitotic cell as a strategy of decreasing mtDNA damage? Once again, a strategy far simpler than allotopic expression.

I'll spare Michael the effort of commenting on that one.

In his "ground-breaking" paper (2a), de Grey points out that it is precisely "the turnover or fission of mitochodrial organelles within a post-mitotic cell" that is the problem. You and I have pointed out the abundant benefits of ablation and replacement by division for somatic cells. The cells with the highest integrity are selected for.

According to de Grey's hypothesis, the reverse logic actually comes into play. Because of the dynamics of the ablation process of mitochondria (via consumption by lysosomes), the mitochondria with lower integrity are sometimes selected for. They eventually are selected for until all mitochondria in a cell are mutant. If the mutations causes slight dysfunction, the effect is reduced mitochondrial output, and thus increased stress on the cell.

In the worst-case scenario, a completely dysfunctional mitochondrion is selected for, and the cell becomes anaerobic. Which leads us to his LDL-oxidation theory in (j2).

Interesting stuff, wish I'd read it earlier. I can understand Michael's frustration that the "vicious" cycle theory still has its hold, assuming de Grey's work hasn't been refuted.

It's worth noting that (2a) doesn't say that fission has to be bad. It simply says that the current dynamics of selection for lysosomal destruction can choose the non-mutant organelles occassionally. Potential solutions?

1) Engineer a better selection method.
2) Prevent the mutations in the first place.

If you can suggest a way to engineer a better selection method, then increasing turnover or fission might be an option. Otherwise, we're stuck with 2. Unless there's a hole in de Grey's paper (2a).


2a. de Grey AD. A proposed refinement of the mitochondrial free radical theory of aging. Bioessays. 1997 Feb;19(2):161-6. Review. PMID: 9046246 [PubMed - indexed for MEDLINE]

j2. de Grey ADNJ. A mechanism proposed to explain the rise in oxidative stress during aging. J Anti-Aging Med 1998; 1(1):53-66.

[jaydfox]

Potential solutions?

1) Engineer a better selection method.
2) Prevent the mutations in the first place.

If you can suggest a way to engineer a better selection method, then increasing turnover or fission might be an option. Otherwise, we're stuck with 2.

Options for solution 2:

2.1) allotopic expression of mtDNA
2.2) increased mtDNA integrity through increasing antioxidant availability in the correct targets, and reducing oxidizability through DNA maintenance and repair factors. I'm not sure how to address the leakage issue, but addressing oxidizability and ROS potency are both possible.

Gentlemen, I want a clean fight. [":)] [pirate]

[jaydfox]

My proposal is to increase the rate of fission before mitochondria have had a chance to mutate into more survivable but less functional versions - if that is indeed what happens.

As I understand it, the point mutations thought responsible need time to drift into all 5-10 copies of the genome (that is what is meant by homozygous, yes?), so if you could catch the problem before then (e.g. via mitochondrial "tumor suppressors" ), you could in theory eliminate the problem.

Or if you could fission the mitochondria faster than the mutation could drift. I doubt that's possible, but IANAMB.

[jaydfox]

Thank's Jay, I was aware of Aubrey's hypothesis that mitochondria with genomic damage and impaired function can be selected for and become more proliferative than healthy mitochondria

I didn't doubt you were aware. I was just saving Michael Rae the trouble of pointing it out to you, complete with references.