Primers for unmethylated DNA work too well! - (Nov/10/2004 )

Hi!

I have designed primers with MethPrimer for five genes using the suggested parameters. The methylated pairs work well, they only make products from my bisulfide modified methylated control DNA, but not the bisulfide treated sperm DNA.

The unmethylated pairs, instead, are producing bands from every bisulfide modified DNA I have tried, including commercial fully methylated DNA and DNA methylated with SssI by myself. (They do not amplify DNA without bisulfide conversion.)

I have now tried higher and higher annealing temperatures, so that I finally loose the products.
The bands seen in agarose gel in methylated controls are little weaker than for example in non-methylated sperm DNA, but still are visible. There is 35 to 40 cycles in my PCR. I would not like to lower the cycles because my samples don´t have so much DNA and their bands might get weak, too.

Is there any solution to my problem? And has anyone else these same problems?

I have seen this so often since many of my colleagues use MSP and almost all samples they have studied show USP band while only a few show MSP band. That is why I never trust MSP result. It is understandable why USP primers tend to work better than MSP primers because T is easier than C to form mismatch.

Has it really been so with your controls, too, pcrman? I could understand that the genes people are working with are normally "in use" so that samples usually amplify only with USP, but it would be so ....... nice NOT to have the methylated controls amplifying!

So, at least I have the same problem with 5 different USPs for 5 different genes, is it really this common problem, or do I just have bad luck? I think that there are USPs that work, isn´t there? Anybody?

OR could it be my methylated DNA, that the bisulfide conversion goes over, and converts some of the methylated C´s also? I do the conversion with CpGenome Kit, as the protocol says.