On Sat, 6 May 2000, phil osborn wrote:
> I had previously written:
> >John, has pointed out the points in my argument that I'm glossing over.
> >The question is: What are the sources of mutation that lead to malignancy
> >(and perhaps aspects of aging)?
> >(a) Radiation
> >(b) Oxidative stress
> >(c) Toxins in the cellular "environment" (derived from food).
> >(d) Mutations by DNA Polymerase in copying the DNA.
> >
> Thus, the engineering tradeoffs dictate that in a system as complex as a
> cell, there WILL be bugs and weaknesses, but that they can't be prevalent or
> serious enough that most cells will succumb to them much of the time.
I would argue many of the weaknesses are inherent in the system or perhaps
local optima in the system. The fact that you have a system that rapidly
recycles any proteins that do not fold properly means that upstream
mutations in DNA or mistakes in the manufacturing of the protein don't
get propagated downstream. Now of course, when you get mutations in the
recycling machinery things start to come unglued...
> Statistically, there is the tiny chance that one of the lethal
> variants of the signals will get through.
At some point that must happen.
>
> However, as I'm vaguely recalling from the Sci. Am. article, cells also tend
> to go crazy when pushed somehow outside the normal operating parameters.
>
Most programs have this property. Windows 2K goes bonkers when you push it to
the edge of available swap space.
> Koestler wrote about the DNA code being taped over - as with a piano with
> certain keys taped over - such that only certain potentialities are
> expressed. As cells age, however, they tend to drift back toward a more
> general expression - to become less differentiated, as the "tape" comes
> loose at random points, until they conclude that somehow they are not
> fulfilling their mission - or the demented version of it that now is the
> controlling goal.
This is Cutler's dysdifferentiation theory (which seems to have been developed
circa 1985). It sounds like Koestler is recycling it (do you have a ref. for that?)
> When cells conclude that they are not fulfilling their mission - or, more
> concretely, when signals from the cell surround indicate that it is not
> capable of meeting its programmed challenges, what is the response going to
> be? ... To divide, obviously. If one cell is not enough, then maybe a
> million will do the trick.
Not clear. Cells that divide are those that are always posed on the edge
of division because division is what that tissue normaly does (or at least
does in response to specific hormones, e.g. mammary glands). There are
many tissues in the body (consider ~300 different cell types), that
are terminally differentiated. You almost *never* get cancer in those
cells because you require a lot more mutation to restore the program
to a state where it is "primed" for division.
>
> So, what kinds of code correction systems keep these cells in line well
> beyond the normal limits, as we have been discussing?
>
The major ones I can think of are
(a) the DNA repair process;
(b) blocks on replication when damage is present (e.g. p53 gene & friends);
(c) apoptosis when things get really out of whack;
(d) a patrolling immune system looking for cells that seem to have
"wierd" proteins inside.
>
> Is it possible that one of the reasons for the apparent extended
> youthfulness, etc., exhibited in the recent calve clones or some of the cell
> culture experiments is due to the fact that the stable environment, with
> only cells of the same kind present, made for a much stronger reinforcement
> signal pattern to the cells of each generation? Instead of the chance of
> stray signals from surrounding tissue or debris or viruses, etc., confusing
> the cell as to its identity and slipping through one of the programming
> holes, each newly divided heart muscle cell received the unambiguous
> message, "you're a heart muscle cell... you're a heart muscle cell...
> you're a heart muscle cell... "
You never hear of "cancer of the heart", precisely because muscle cells
are terminally differentiated and multi-nucleated (meaning they have
multiple copies of the program in each cell). But I agree with the
basic premise, that in tissue culture you have a much cleaner environment
and are unlikely to receive any "wrong" signals from the guy down the block
whose program counter is executing random code segments.
> Thus, the normal correction procedures that keep a cell on track might use
> that unambiguous signal as a referent, and actually clean up a lot of
> accumulated error.
I believe the current thinking is that most cancers requires the presence
of a promoting signal and the absence of a restricting signal. Then there
are failsafes like telomeres and the immune system to try and "catch"
the cells that elude those constrants. The system "works" for ~75 years
for about 2/3 of the population.
Robert
This archive was generated by hypermail 2b29 : Thu Jul 27 2000 - 14:10:38 MDT