DNA methylation vs gene expression - (May/15/2006 )

Hi there,

I'm having some weird results lately which i can't really explain.

I have been analysing the expression of a certain gene in several tissues/cell lines and correlating this to the binding of a transcriptional repressor (TR) to the promoter of this gene. At first, it looked quite ok. With ChIP i saw that the TR binds to the promoter in tissues where the gene is not expressed and vice versa.

Now, since this TR is a methyl-CpG binding protein, i decided to check the methylation (with HhaI/McrBC digestion followed by PCR) of the promoter in different tissues, and now it becomes weird: I checked two tissues (one with low gene expression and one with high gene expression) and one (cancer) cell line with high expression. It seems that the promoter is methylated in the cell line (which does express the gene, but does not IP the mCpG binding protein) but not in the tissues!!

So I would like to know:
- has anybody ever heard of a gene which shows higher expression when it is methylated?
- how is it possible that i ChIP the protein in my non-methylated promoter and not in my methylated promoter??

Thx.

---

I am interested to hear which protein you are looking at. Indeed MBD's can bind to non-methylated DNA and this has been shown in the past. Pubmed PA Wade and MeCP2.

There are instances of hypermethylation and gene expression (ie: dysregulated expression of genes independent of DNA methylation) again if you pubmed this, you will find something.

Good luck!

Nick

---

It is indeed MeCP2 I am ChIPing. Now I am thinking that maybe it is not directly bound to my promoter but via some transcriptional repression complex. Any idea how I could show this (preferably without too much effort)?

---

you could try coimmunoprecipitation, or ChIP-ChIP.

It is entirely possible that MeCP2 is bound to your promoter through another repressor/activator that is directly bound to the DNA sequence.

Nick

---