feedback requested on ideas(modified)

Iuval clejan clejan at
Thu Feb 22 19:08:26 EST 2001

Thanks very much, Aubrey. I really appreciate your time and will try not to ask
for much more of it. I have only a few comments and a few more questions:
Aubrey de Grey wrote:.

> This is a fair start, but has several problems.  First, your category (2)
> should be extended to include any cumulative change that occurs as a side-
> effect of normal metabolism, not just the two that you mention (replicative
> senescence and differentiation).  One that may be very important is the
> buildup of undegradable junk in lysosmes.  In nerves and muscle this is
> lipofuscin and its causal role in aging is still controversial, but in
> some other cell types it is definitely very bad for us: a major example
> is the oxidised cholesterol that accumulates in arterial macrophages and
> turns them into foam cells initiating atherosclerosis.

Does  this buildup of undegradable junk in lysosomes correlate with the time
scale for decreased organ function? Also, why does it not happen in the
germline (or does it get selected out)?

> Second, mechanism (3) can occur due to something much more dramatic than
> mechanisms (1) and (2) happening to other cells, namely other cells
> dying and not being replaced.  It is certain that a lot of aging is of
> this type.  The decline of the immune system is due in large part to
> loss of cells in the thymus, for example.

Is it due to the remaining cells being negatively impacted (e.g. having to
"work harder" or missing signals from their missing neighbors) or is it just
not having enough cells in an organ decreases its function, in which case I
would rather classify it as a separate mechanism, or a special case of (2).

> Third, your mechanism (3) can occur due to changes in the extracellular
> environment, not driven by other cells at all.  This is why AGE-breakers
> are so exciting: AGEs probably accumulate intracellularly too, but in the
> extracellular matrix they seem to play a much more major role in making
> tissues less elastic.  This makes the heart work less well, for example,
> which clearly has rather widespread potential consequences.

I am ignorant of this, will look up AGEs. Are they degraded or cross linked
collagens and elastin? I thought extracellular matrix is made up of components
secreted by cells (mostly fibroblasts).

> Fourth, when you dismiss decreased replicative capacity as having much
> to do with aging you do so on the basis of replicative senescence, i.e.
> you essentially ignore the other cause (differentiation), which is not
> studied in typical in vitro cell senescence experiments.  Cell loss in
> muscle is a good example of something that should be (indeed, is) very
> largely reversible by stimulating cell division: in that case by the
> very low-tech method of appropriate exercise.

I think there may be a role for decreased replicative capacity, but in the end
it it damage repair that fails, except for your objection about cells dying and
not being replaced.

> > If hormones/nutrients/extracellular matrix change due to decreased
> > replicative capacity of some cells (3), then the change should be
> > artificially induced in vitro to see if other cell types (e.g. heart
> > or brain) react to that change by aging more quickly.
> You seem to imply that this is an easy experiment....

No, but I know someone who is making a liver bioreactor and I may be able to
her into this.

> > I think there are
> > very few mutations that can cause a continuous change in protein
> > expression per cell. These include (correct me if I'm wrong) mutations
> > to:
> Hang on.  This list covers only the simple case of mutations that cause
> modulation of expression of the same protein, in the same cell, that
> the mutation is in the gene of.

Not at all. All except possibly (2) are about mutations for genes other than
the ones for the
protein being modulated.

>   Also, mutations may have nothing to do with these changes --
> gene expression is modulated by cellular state in terms of degree of
> oxidative stress, for example.

I thought things like pH, temperature, oxidative stress, etc., are all
extremely tightly regulated in vivo on the average. Is there a drift
in any of these variables (which indeed can influence reaction kinetics in
continuous ways) with age? I don't think so.

> At any particular moment a particular
> gene may be being expressed or not being, but over a long period it
> will alternate between on and off and the proportion of time it spends
> in either state can vary continuously during aging due to arbitrarily
> complex interactions of other changes.

Excellent point. Does this actually happen? As far as I know nobody has looked
at time averaged mRNA expression. What is the time scale
for which genes (coding for proteins that are involved in decreased organ
function with age and associated proteins, not for genes associated with daily
metabolic changes) are transcriptionally active, all else being equal?

> > In a non replicating
> > cell, the only way to knock out a gene with random damage (due to toxic
> > environmental or metabolic oxidant species) is if there is only 1 (or
> > no) repair pathway for that type of damage (or one enzyme shared
> > between two or more pathways), and if a gene coding for an enzyme in
> > that pathway (or the shared enzyme)also gets the same type of damage.
> Wrong: you're forgetting that these repair processes are imperfect.
> Suppose that a methylcytosine is deaminated and so becomes thymine,
> paired to guanine.  There's an enzyme that spots these and replaces
> the thymine with cytosine.  But since that doesn't happen absolutely
> instantly, and there is also other damage going on at the same time,
> there's a chance that the other strand (the one with the guanine)
> will suffer damage that requires bulky lesion repair, for example,
> where a chunk of strand is just excised and rewritten using the
> other strand.  If that happens while the methylcytosine is still a
> thymine, you've got a fixed base pair substitution.

I didn't think base pair substitution happens in non replicating cells. The
scenario you propose seems possible, although I don't know if it really
happens.  I need to study more on DNA repair. But the concept of
non-instantaneous repair is the same in non-replicating cells and indeed it can
happen that damage to more than one repair pathway can occur "simultaneously".
Again I must ask for time scales. Is it possible with some finite probability
to damage only enzymes in two or more repair pathways without damaging other
DNA which would kill the cell? If the time to do this is too long, then there
will be time to repair the repair enzymes which have been damaged by the ones
that have not. If the time is too short, then it seems (but may not be)
improbable that the damage won't be of massive type that destroys alot more
than just the repair enzymes, and the cell will die.

> > Now the mtDNA may be the only DNA in mammals for which there is only
> > one repair mechanism for some some types of oxidative damage.
> Here your argument breaks down not only for the above reason but also
> because mitochondria continue to divide even in non-dividing cells.

This I did not know. I thought there must be some way to coordinate
mitochondrial replication ( in somatic cells, I know this is not the case for
eggs) with the cell cycle, otherwise there would be too many mitochondria. What
regulates their division? Has it been observed in a non dividing somatic cell
(or G0/G1/G2 part of the cell cycle in a dividing cell) that mitochondria

> Replication can make mutations out of thin air, with no intermediate
> oddity that can be recognised as damage by some repair enzyme.  Once
> the mutation has happened, it may take over the cell by virtue of
> having some sort of selective advantage -- see, for example, the model
> I proposed in BioEssays 19:161.
> Aubrey de Grey

More information about the Ageing mailing list