Re: AGING: Accumulation of DNA damage

From: Robert J. Bradbury (
Date: Mon Jun 26 2000 - 15:05:50 MDT

On Mon, 26 Jun 2000 wrote:

> By itself this observation doesn't seem to establish your conclusion.

I would agree. I've suggested before, I think, that it will take
something like extensive SNP analysis or a lot of DNA sequencing
on a cell by cell basis to *really* settle this question. It is
however a smoking gun...

> The fact that DNA is damaged, and damage accumulates, doesn't imply
> that it is a cause of the phenomenon that we call aging. It could
> be that aging is caused by something else, and that while DNA damage
> would eventually lead to death, the symptoms and the time scale would
> be completely different from what we think of as aging. Maybe you'd
> live 1000 years and then your cells would all turn to cancers.

Actually, probably not. It seems that after a period of time, (70-90 years),
cancer rates start to decline. When the unravel the genetic background,
I suspect they will discover a "gene combination" that just "happens"
to have fantastic DNA repair and a lack of genes that turn environmental
substances into carcinogens.

> There may be other reasons to believe that DNA damage is a cause of aging,
> but the mere fact that damage accumulates over time isn't enough.
Its the observations in aging (decline in protein synthesis, changes
in hormone levels, decreased cellular replication ability, etc.)
that lead to the plausibility that the original program is being
damaged over time. This fits into the dysdifferentiation hypothesis
of aging if as mutations accumulate they gradually distrupt the
ability of the cell to maintain its differentiated state.

> As we've discussed before, intentionally producing brainless humans is
> not going to be considered by most people as a way to resolve ethical
> issues. Rather, it raises a host of questions of its own. Start off
> with Huxley's Brave New World and discuss whether that system, which has
> some similarities to this idea, is moral.

Well, individuals such as myself who haven't read BNW aren't going
to be interested in what it says if we can make a pre-emptive argument.
a) Here is a live person with "lots" of information/experience/etc.
b) Here is a "body" we can grow that will be constructed in such
   a way that it has nothing other than pre-programmed information and
   no experience (as well as no consciousness, self-awareness, free-will,
c) (a) is more valuable than (b) [from an information content standpoint]
d) Making (b) and using it to preserve (a) is the correct thing to do
   (if one assumes organized information is valuable as presumably
    most extropians do).

> In particular, the epsilons are produced by poisoning their brains
> during development so they are nearly mindless, drooling automatons.
> It's not a very attractive image.

Yep, and "attractiveness" isn't something I personally care about
when it comes to whether we will kill individual. I'll *rationally*
adopt the position of a physician in a triage situation who has to
save the greatest number of lives with the resources and techniques
at his disposal.

If a person doesn't want to be "saved" using a "clone", then that
is their choice, but if the technology could exist, then I strongly
object to society making that choice for me.

> Is it reasonable that as many as one cell in 30 has escaped harmful
> DNA mutations, so that it is in good enough shape to create a whole new
> organism from scratch? These are ordinary cells that have been dividing
> and growing and metabolizing for a substantial fraction of the lifespan of
> the animal. If DNA damage is the limiting factor, I would have expected
> ordinary tissue cells to be more than 30 times worse than the egg cells
> (which presumably have extra protection against damage).

The devil is in the details here of course. I would have to go through
the literature to refresh my memory, but I'm fairly certain that
transcribed genes and replicating tissues have higher levels of DNA repair
than untranscribed/replicated genes (RNA or DNA polymerase get "blocked"
while running down the strand and that triggers a repair response). So that
allows damage to preferentially accumulate in the non/rarely transcribed
or copied genes (particularly in the developmental program). There is a low
level of generic repair that occurs but it is not clear how complete or
effective it is.

This makes sense -- why would you want to expend resources repairing
genes that the average cell will never use again? Now the *sticky*
part comes into play with the *rarely* executed subroutines. You
only execute the apoptosis (cell-death) program once, but presumably
when the immune system tells a cancer cell to execute that program
you *would* like it to work (at least most of the time)...

In looking at this it seems to me that nature will have tuned
repair mechanisms (you don't think running ECC checks on 3x10^9 BP/cell
that are receiving thousands+ mutations/day is cheap do you???)
to match the requirements -- "frequently used programs" ==
"highly active repair", "less used programs" == "less active repair",
"never used programs" == "very rare repair".

Since it isn't a well designed system you would expect some
overlap between the groups.

So I stand by my position that the program is likely to be highly
defective in those processes that clones will require (leading to
many clonal failures). [This agrees with normal human reproductive data.]
The corrolary of that situation would appear to me to be part of
the aging process, in that lesser used parts of the program
(e.g. those subroutines that stimulate slow-dividing stem cells
to divide) are likely to have problems. Lose your stem cells
(in replicative tissues) and you eventually lose the organism.


This archive was generated by hypermail 2b29 : Thu Jul 27 2000 - 14:14:35 MDT