Why can the MSP represent the methylation of CpG island? - (Apr/14/2005 )

Hello:
I have a question,maybe so simple.
There are so many CpG sites in one CpG island ,and the CpG sites investigated by a MSP are so limited. Only less than 10 CpG sites included in the two pairs of primers for MSP that can be investgated by a MSP. Why so little CpG sites mehtylation informatiom can repsent the whole CpG island ???

---

it is safe to assume that within an island, if the residues you pick for MSP represent the methylation status of the whole island. Having said that you can not be absolutely sure unless you perform bisuflite PCR and sequencing.

Nick

---

BUT I am still wonder why "it is safe to assume that within an island, if the residues you pick for MSP represent the methylation status of the whole island. "Can you explain to me?
Thank you very much!

---

I think this question is still a unsolved puzzle.

>>Only less than 10 CpG sites included in the two pairs of primers for MSP that can be investgated by a MSP.

Even you have included 10 CpG sites in your msp primers, the PCR result doesn't tell you all 10 sites have the same status of methylation. Possibly it can only represents the true methylation status of the 3' most 'C' in your primers.

Even you got the true methylation status of all 10 sites by sequencing, you are still not sure whether the methylation of them has any impact on transcription.

You may use demethylating agent such as 5azaC to treat cells, if the methylation disappears and expression of your gene comes back, you can say oh the methylation of those 10 cpg sites kills my gene. No way, 5azaC demethylates everything everywhere.

We have studied methylation for so many years, so many papers have been published, so many genes been typed, so many samples been analyzed, what have we gained? found the cause of cancer? Known how methylation is originated, controlled? Used for clinical diagnosis? Nothing!

So that is a real puzzle.

---