What is the "p" in CpG?

s060965 at AIX1.UOTTAWA.CA s060965 at AIX1.UOTTAWA.CA
Tue Jun 17 07:43:14 EST 1997


A correction is in order here:

CpG islands are clusters of CpG dinucleotides usually present at the 5'
region of housekeeping genes.

They are typically NOT methylated even though they are a great substrate
for DNA methyltransferase.  While Nicolas's statement regarding the low
frequency of CpG and deamination are correct (this only applies to normal
CpG) and not too so called "CpG islands" which are refractory to DNA
methylation.  Remember that methylated DNA is typically transcriptionally
inactive ... so if CpG islands are methylated this would cause a problem
for housekeeping genes.

For references see:  Bird, A.P. (1986).  Nature 321:  209-213. or
Antequera, F. and Bird, A.P. (1988) EMBO J 7:  2295-2299.

Cristofre Martin
Loeb Institute
Ottawa Civic Hosp
Ottawa, ON
Canada
K1Y 4E9

On 17 Jun 1997, Nicolas  Le Novere wrote:

> In article <bregeon-1706971048590001 at ravel.ijm.jussieu.fr>,
> Bregeon Damien <bregeon at ijm.jussieu.fr> wrote:
> >In article <5o3fqs$8vd$1 at vixen.cso.uiuc.edu>, dneece at ux1.cso.uiuc.edu
> >(David Neece) wrote:
> 
> 
> An interesting point about CpG islands. The "school" explanation of their 
> rarity is the methylation. The methylases act specifically on these sites.
> The methylated C is easely transformed by deamination into T. The increased
> mutations frequency would be counterselected. 
> 
> However, another fact is the thermodynamic behviour of CpG. Indeed this 
> "Crick's pair" possesses the highest deltaH of all, for DNA homoduplexes 
> (Breslauer et al (1986) PNAS 83 : 3746-3750) but mostly for DNA/RNA
> heteroduplexes (Sugimoto et al (1995) Biochemistry 34 : 11211-11216).  
> The Tm being increased (Wetmur (1991) Crit Rev Biochem Mol Biol 26 : 227-259),
> some problems could occur during transcription. There would be a 
> counterselection of the CpG but the action of methylases would be the 
> consequence of the CpG rarity (for specificity reasons) and not the cause.
> 
> Some idea about that?
>  
> 
> -- 
> Nicolas Le Novere
> INSTITUT PASTEUR,Neurobiologie Moleculaire,25, rue du DR ROUX, 75O15 PARIS, FRANCE
> Ph: 33-1-45-68-88-44,Fax: 33-1-45-68-88-36,lenov at pasteur.fr
> http://www.pasteur.fr/units/neubiomol
> 
> 




More information about the Mol-evol mailing list