Protocol Online logo
Top : Forum Archives: : DNA Methylation, Histone and Chromatin Study

COBRA - trouble with quantification series - (Nov/16/2005 )


first I would like to say hello to all.
I got a problem concerning my COBRA methylation studies.
To quantify our COBRA we clone both fully methylated (e.g. SssI treated DNA) and
fully unmethylated bisulfite PCR-products into the pcr2.1 vector of Invitrogen.
The plasmids we obtain are mixed in a calibration standard from 0% to 100% cloned
methylated pcr-product.
The methylation status of cloned pcr-products is evaluated by sequencing.

Now the problem:
Despite the fact the pcr-product in my vector of bisulfite treated DNA is completely methylated,
the digestion (in this case with MluI) remains incomplete in 100% sample.
We do the digestion with about 60 ng of pcr-product obtained after gel extraction. The amount of enzyme is about 10 units per reaction.

I don't know, where to look first.
1. It could be a matter of reaction conditions in the digest. That's what I think the problem is.
But how should I optimize the digest?
2. The clone is not a clone but a mix of different clones. But I should see this in sequencing, I think.

What other reasons could be thougt of?
Which further informations should I give?

I'd be thankful for any statement or idea given!



what is the status of your bacterial cells? do they have methylase activity and thus maintain the methylation of your vector as you transform it into cells.

the mixed methylation status could be just that.

I can't recall off the top of my head which cells will maintain methylation.

It would be better to clone your amplicon into a plasmid.....midiprep it and then SssI methylate the plasmid. Assuming 100% efficency of SssI methylase you can then titrate methylated to unmethylated vector for your bisulfite.



Thanks for your answer.
I think you misunderstood my posting. We put the SssI-treated DNA into
our bisulfite treatment. The pcr-product of this DNA we clone into the vector.
So there is no methylation still left at all, it is completely transferred into our
sequence. I don't have to maintain methylation, because its the sequence that
differs methylated from unmethylated.
Or did I get you wrong?
Did you mean that my vector gets de-novo methylated in my DH5alpha and
my COBRA enzyme is blocked by this methylation?


Your proposal lacks some features that we wanted to be included:
We need a standard that can easily be put into our bisulfite pcr and doesn't
need to be treated by SssI and bisulfite all the time. So we want to clone a sequence
where all CpG-C's are maintained and all nonCpG-C's are converted to T. That's what I
called "fully methylated", but it in fact there is no methylation left after the pcr, so maybe
that is the point why you misunderstood me.


QUOTE (methyl-tom @ Nov 17 2005, 06:42 AM)
Did you mean that my vector gets de-novo methylated in my DH5alpha and
my COBRA enzyme is blocked by this methylation?

yeah that was what I was getting at!


I checked for the possibility of de-novo methylation, but MluI is not
sensitive for Dam or Dcm methylation.
There must be another reason.



hi methyl-tom,

I have had a look at rebase and you are indeed right about Dam and Dcm methylation however, MluI is sensitive to CpG methylation, thus if your samples were in fact 100% methylated you will not get complete digestion!!!