Michael Lones <mal111 at ohm.york.ac.uk> wrote in article
<82gseh$hd1$1 at pump1.york.ac.uk>...
> Could anybody tell me, in fairly simple terms, why (according to the
> Holliday model) branch migration occurs during genetic recombination? Is
> this considered a phase which developed to increase evolvability of an
> organism (through gene conversion, which I assume has some evolutionary
> significance) - or is it just a consequence of chemistry or a side effect
of
> some other cellular mechanisms?
> Thanks for any help. I'm but a computer scientist trying to understand
> biology!
>> --
>> Michael Adam Lones
> Biologically-Inspired Systems
> Department of Electronics
> University of York
> England
>>Four strand exchange reactions generally start in regions where one
of the substrates is single stranded, and that is an important point.
In general, though:
A Holliday junction between two dsDNA molecules, in the absence of
other molecules, will diffuse randomly along the substrates. You
can model this using Kramer's rate theory, which is not very
complicated.
In the context of biology, this is of very little value. The motion of
the junction must have a net directionality if it is to consistently
accomplish a completed exchange reaction, and if it is to do so
in a reasonable amount of time. In prokaryotes, this is facillitated
by any of a variety of molecular mechanisms. Two of them are
RecA and the RuvAB heterodimer.
RecA (which I work on) accomplishes that homology search
between two substrates and can accomplish
extensive directed strand exchange as well. I have a paper
in review which proposes a model of how this occurs
(I will email you a copy if you are interested). IMO, there
is only one model currently in the literature which is
viable in light of what is know experimentally: the
"facillitated rotation" model of M. M. Cox et al.. Look
in the Journal of Biological Chemistry if you want
details of the work done on RecA, and Cox's model.
If you want a good pedestrian overview, go to the following site:
http://robles.callutheran.edu/BioDev/omm/gallery.htm
Download the necessary free software to view the
molecules, and then go to the "Hall of DNA Recombination",
where there is an excellent short tutorial on RecA.
RuvAB is perhaps even neater than RecA. It does not accomplish
the homology search, as RecA does, but only the directed
motion of the Holliday junction. Generally, it takes over once
RecA has set things up. A *great* little animation of this
baby in action is at the following site:
http://www.sdsc.edu/journals/mbb/ruva.html
which is also available closer to your home and my heart at:
http://www.shef.ac.uk/~mbb/ruva/ruva.html
I will leave those sites to provide more information about RuvAB, since
the authors of those sites certainly know more about it than I do.
I hope this is of use to you.
Cheers,
Kevin
