IUBio Biosequences .. Software .. Molbio soft .. Network News .. FTP

Recombination Question

Graham Dellaire popa0206 at PO-Box.McGill.CA
Wed May 24 17:06:21 EST 1995

Gary Williiams wrote:

>Are there regions in the genome which have large numbers of
>recombinations, aided by recognition of sites by proteins?

Well first it helps to define a system... ie bacterial or mammalian or yeast etc.
But generally recombination is most similar between bacteria and yeast and 
a little more complicated in mammalian cells.....

In yeast and mammalian cells you have fragile sites that usually consist of 
repeats like ctg, cag or expansions of the dinucleotide CpG etc (ex. look
 for articles on fragile X syndrome), that are quite prone to breakage and 
this double stranded break can initiate recombination.

 As well sometimes in bacteria you have high  recombination between inverted 
or direct repeats. 

 Essentially in bacteria  and yeast, sequences highly homologous to each other 
can recombine very efficiently.  In bacteria the protein involved in the recognition 
of homology and thought to mediate recombination between these sequences
is known as REC A.... it is not sequence specific though.  

In yeast you have FLP recombinase that recognizes specifically two identicle
target site that can be used to flank two regions you wish to recombine.  The P1
phage of bacteria has a similar sytem known as the Cre recombinase... which
recognizes two loxp sites.

In mammalian cells.... you have plenty of extra protein to wade through (ie histones)
before you get to the DNA to recombine.  So basically lack of protein in a region
(removal of histones for example) greatly increase the chance of recombination.

Also DNA conformation can have an effect.  It has been shown by my laboratory
for instance that bent DNA regions flank retroviral integration sites (Milot et al 1994)
SO in some way the integrase of the virus is recognising bent regions.....

A good example of site directed recombination (ie perhaps similar to Cre or FLP)
is the rearragement of the immuniglobin locus in maturing B cells and the TCR
region in T cells.  This seems to involve alot of proteins generally involved in 
DNA repair (ie. Ku80/70  and DNA protien Kinase (recently determined as 
the gene involved in the SCID mouse defect), as well as two gene products
RAG-1 and RAG-2 which seem to make up a recombinase that is specific
for the signal sequences found in the native immuniglobin loci.  

In the human genome you also can get recombination between short and
intermediate repeats but probably by a different process than in bacteria
and yeast (ie. gene conversion rather than reciprocal exchange).  For example
between ALU sequences and between Line-1 elements.   

>Are the signals for such sites known?

If you mean site directed.... yes as described above.  As for
fragile sites... I think that using cytogenetic staining procedures
you can visualize these fragile sites. 

>Is recombination within exons less frequent than outside them? (This may
>be difficult to measure, assuming there may be lethal disruption of the
>cell's machinery when this occurs). 

Recombination can occur at a higher rate in regions of DNA that transcribed,
most likely due to increase access to this region for recombination to occur, ex.
by simply removing the histones and/or changing DNA's conformation so it
is more available for recombination.

Of course if this is homologous recombination with the allel on the homologous
chromosome there is no harm done.

Actually there is a good evolutionary reason for recombination occuring
at higher rates with in transcribed regions.  By continually recombining during
mitosis and/or meiosis you prevent lethal mutations from building up in a gene
and also ensure proper disjunction at meiosis (chromosomes that do not 
recombine with their homolog will not segregate properly).

So agian it depends on what type of recombination we are talking about
as well <grin>

I hope this will help you get you started.  After a bit of reading you may
see how all this fits you application (or not <grin> )

Graham Dellaire			Snail Mail:
                                  Red Cross, Research		
McGill Univeristy                 Montreal Blood Services	  	
Faculty of Medicine               3131 Sherbrooke St. East         
Div. of Experimental Medicine     Montreal, QC, Canada           
E-mail: popa0206 at po-box.mcgill.ca H1W 1B2			   
B2XE at musicb.mcgill.ca						
Fax: (514) 525 0881							  
Voice: (514) 527 1501 ext 175 					

More information about the Biochrom mailing list

Send comments to us at biosci-help [At] net.bio.net