In article <1994Jan28.190047.6170 at iitmax.iit.edu>, garfinkl at iitmax.iit.edu (Mark D. Garfinkel) writes:
>>afc at gnv.ifas.ufl.edu (Andrew Cockburn) writes:
> *The readers of this thread would do well to review Britten &
> Kohne's 1968 Science paper for the definition & initial methodology on
> genomic DNA complexity measurements, as well as the long series of Cot
> curve experiments by Britten & Davidson & their colleagues.
Shameless plug: When I started working on mosquitoes, I wanted to know
what their interspersion patterns were like. Instead of doing Cot curves
(of which I have done too many), we simply hybridized total genomic DNA
to a few hundred random clones from a genomic library. In a week or
so we had determined interspersion patterns for ten species. We
published this in Arch. Insect Biochem.Physiol. 10:105-113 (1989).
>>What I would like to see is an explanation of how a small genome
>>can evolve from a large genome full of repeats. I believe that this
>>happens in insects fairly frequently. Allen Spradling suggested to me
>>that this might be tied up with the mechanism of polytene
>>chromosome formation, which involves deletion of repetitive DNA.
> *Not being privy to the exchange you had with him, I don't know
> what Allan had in mind. I wonder how the mechanisms at work in
> underreplicating or eliminating portions of the genome in *somatic*
> polytene tissues could bear on the behavior of *germline* chromosomes.
> Please elaborate for us.
Spradling's point was based on experiments he reported in _Cell_ a few
years ago. It is the convential wisdom that polyteny in Drosophila includes
underreplication of heterochromatin. His results indicate that
this is false; rather heterochromatin is specifically deleted *after*
being replicated. BTW, polytenes are known from many groups besides
insects, and in most of these other cases it is not assumed that under-
replication is responsible for elimination of heterochromatin. (This
is based on a conversation from several years ago and the subsequent
reading of his paper. I know nothing about the polytene literature.)
My argument is that there does exist a mechanism for specifically
eliminating heterochromatin and perhaps other repeats from the genome.
Although this is expressed in somatic tissue, it does not take too
great a leap of imagination to think that this might occur occasionally
in the germline, at least on an evolutionary time scale.
Some such mechanism must be acting, or genomes would have expanded to
infinity by now. We know of lots of mechanisms for expanding genomes,
and if there were no counteracting mechanism genome size would rachet up