conservation of intron-exon boundaries

Nicolas Le Novere lenov at pasteur.fr
Mon Apr 29 03:05:21 EST 1996


In article <4lm00b$5ke at swen.emba.uvm.edu>, brianf at med.uvm.edu (Brian Foley) writes:
|> J. David Spafford (jspaffor at gpu2.srv.ualberta.ca) wrote:
|> 
|> : What does conservation of intron-exon boundaries mean?  Does it mean that
|> : the genes are regulated similarly?
|> 
|> 	I don't think we know yet what the implications of intron/exon
|> structure are.  For another example of a gene which has been sequenced
|> in a variety of organisms and found to have some conserved intron sites
|> and some movable intron sites, see the Pax-6 gene and its homologs
|> (eyeless in Drosophila, Sey in mouse etc...). 

|> 	There has been a lot of talk about whether introns are more
|> often inserted in genes than deleted, but no conslusions have yet
|> been drawn, to my knowledge.
|> 	As for regulation, look for conserved elements within
|> introns and other non-translated regions, rather than just
|> intron/exon structure overall.  
|> 

In the ligand-gated ion channel superfamily (GABAa receptors, glycine
receptors, 5-HT3 receptors and nicotinic receptors), a cladistic 
analysis of the int/ex boundaries (as well as a compatibility analysis)
match perfectly the phylogeny established from sequence analyses.
Moreover, inside a family the conservation of a boundary varies from
100% to almost 0%, the most conserved boundaries being located in the
most conserved parts of the coding sequence. 
That means:
(i)- The behaviour of ex/int boundaries looklikes other sequence features
(their evolution is poissonian).
(ii)- The conservation of the boundaries reflects in fact the conservation
of sequence. Indeed, a movement of an ex/int boundary is likely to 
change the coding sequence (because of the plicing tags).

I don't know if these conclusions are specific of this group of molecules
or if they are more general laws.



More information about the Mol-evol mailing list