From: Robert J. Bradbury (bradbury@aeiveos.com)
Date: Thu Dec 12 2002 - 21:27:58 MST
On Thu, 12 Dec 2002, Joao Magalhaes wrote:
> At 11:02 10-12-2002 -0500, Rafal wrote:
Cool. "The Devil went down to Georgia" and dueling banjos.
(you may not catch those references, but they are worth learning [IMO]).
> As far as I know, ultraviolet radiation, except perhaps now due to
> the Ozone problem, has remained constant for millions of years.
Actually, a nearby supernova could cause a significant spike in
UV radiation (by causing a depletion of atmospheric ozone). It
is currently thought these would have to be in the vicinity of 25-100
light years so they don't happen very often. Its more than millions but
perhaps less than billions of years.
> Curiously, I remember reading
> about teeth evolution and how the teeth and jaws of modern animals have
> evolved so much since, say, the dinosaurs. So unlike the assumption in your
> argument, toothy predators have evolved a lot in the past hundreds of
> millions of years.
Considering entities such as saber toothed tigers and wooly mammoths
I'd say that the evolutionary time scale for the evolution in the
predator-prey relationship is of the order of thousands to millions
of years (not hundreds of millions). However considering how close
the human and mouse genomes seem to be, there is a lot less selection
pressure on genes that are somewhat less critical to survival.
> Yet a bacterial
> culture, even when some cells suffer DNA mutations, survives. The question
> I ask is: why do we assume a multicellular organism can't cope with DNA
> damage? Why do we assume that DNA mutations accumulate in human cells when
> they don't in a bacterial culture? You can argue that non-dividing human
> cells can accumulate damage, but most cells in the body can be replenished
> by stem cells--even neurons.
It could (cope with the damage) if the genomic program were designed for
it (as say the Deinococcus genome seems to be). Bacterial cultures are
always subject to selection effects -- mutated genes and you are asta
la vista baby (sp?) [unless of course the environment has changed to
prefer those genes].
I would disagree with the statement that "most cells in the body can
be replenished by stem cells". It is only those cells normally exposed
to wear and tear that have reasonable replication capabilities (skin, gut, to
a lesser extent endothelial cells, fibroblasts replacing damaged tissue, etc.)
While there is some evidence that some parts of the human body can
be re-invigorated by stem cells (bone marrow transplants being the
key example) the evidence for other tissues is much less clear.
I'd cite any major "mature" organ system (brain, kidney, heart, etc.)
These systems may not be structured such that stem cells can effectively
repair them.
> No. Although when a system is "damaged", there's "something wrong with it",
> there could be "something wrong" with a system without it being "damaged".
> If a car runs out of fuel, there's something wrong with it, but it doesn't
> mean it's damaged. In the same way, if a cell, for example, overexpresses
> p21waf, the cell cycle is blocked but it doesn't mean the cell is damaged.
This is a key point. We have to get out of the framework of saying aging
is "this" or its "that". Cells (and organisms) have a set of hazards
they have to deal with. Genetic programs are designed to optimize the
survival chances of a cellular "collection". (Or more importantly they are
designed to optimize the survival chances of copies of cellular collections).
A program that causes more cells to undergo apoptosis when detecting
multiple double strand breaks (the condition that seems to be true
in humans vs. mice) is helpful in preventing a single cell (and humans
have many more than mice) from killing the organism. *But* unless you
complement it with a program that allows the continual renewal of tissues
with stem cells (that must have *very* good DNA repair) it is only
effective at preventing cancer and not at extending longevity (or the
production of copies of cellular collections). So I think it is
critical to view all genomes as "under construction".
> True, but--as I pointed to Robert, though probably off-list--you also need
> to go through hundreds of fetal cells to clone animals.
I may have missed this. It would be nice to see some replicated work
in this area (there is a lot of rumor floating around in this area
and it would be nice to see us move into an environment where we can
at least all agree on the data).
> Yet according to
> the theory, fetal cells should not have DNA mutations. My conclusion when
> both controls (i.e. fetal cells) and adult cells have a low chance of
> yielding viable cloned animals is that something is wrong in the process of
> cloning and not with the cells themselves. See the papers by Michael West
> and there's a few papers by a Korean team too.
I did post a very concrete reference a few months ago about how
high the mutation rate is in embryos. All one can say about fetal cells
is that they work up until one harvests them. That doesn't tell you
anything about whether "is this a 'perfect' genome" or not. It depends
entirely on what fraction of the genome is essential prior to the "harvesting"
vs. what fraction of the genome is essential after the "harvesting".
Actually, it is unclear to me that "fetal cells should not have DNA mutations".
I would agree that it seems less likely that they should have mutations than
an average adult cell. But to answer this with certainty we would need to
know the DNA repair gene expression profiles, DNA copying error rates, etc.
for the "average" fetal cell and the "adult" cells from which cloning
statistics might be derived.
I'll cut short this message here. I may review and comment on some of the
additional comments at a later date.
It is, as always, a stimulating discussion given my interests.
Robert
This archive was generated by hypermail 2.1.5 : Wed Jan 15 2003 - 17:58:42 MST