Questions about primer design for BSP in plant - shouldn't we take the mC in CHG CHH context into consideration? (Nov/18/2008 )

Hi, all!
I am new to this forum and I am happy to find so many colleagues doing BSP. I am now working on DNA methylation in plant, hope to get some helpful discussion from here.

For a long time, I have been puzzled by a question in designing BSP primers. For a given sequence, if we want to design BSP primer toward its sense strand, we converted all C to T except C in CG context, cause CG is most likely to be methylated and CHG and CHH context are both considered to be non-methylated. But that may not be the case in plant. CHG context may also be highly methylated in some highly methylated region. If we design primers in that standard, what we amplified is just 90% (or even higher percent) of bisulfite converted DNA, but the DNA with mCHG or mCHH will not be amplified, that part of sequence information will be lost and will not brought to the final result ! That might be very serious lost in plant.

For example, the FWA locus in Arabidopsis, methylation status for CG, CNG and assymetric sites will be ~90%, ~20%(sometimes up to 30%), ~10% (sometimes up to 30%) in some cases. (One-Way Control of FWA Imprinting in Arabidopsis Endosperm by DNA Methylation, SCIENCE VOL 303 23 JANUARY 2004). In this paper, they used degenerated primers.

Is it necessary to use degenerate primers all the time? or is there any mistake for my understanding of BSP primer design?

THX!

arabidopsis is a tricky one,

degenerate primers must be used because it is not known the methylation status of the C's that your primer targets, and if you do not design degenerate primers then you are selectively biasing your PCR amplification to the unmethylated or methylated templates (or in the usual case, converted vs unconverted DNA). Degeneracy helps overcome this but also opens up the possibility of false priming to similar sequences that are not your target!

Hope this helps

Nick

Hi, Nick

Thanks a lot! I totally agree with you. But I am wondering why in most papers, they just use primer simply convert all C to T. I have ever tried the primers for SUPERMAN gene in arabidopsis but failed to get any PCR product, until I synthesized additional flanking primer for semi-nested PCR and finally got the product. I don't know what is the problem, why can't I get the product from 1 round PCR? Is the low concerntration of converted DNA the main problem? By the way, I haven't the experience of getting the product in one round for BSP. That make me doubt about the original paper.

Regards

Atos

with standard bisulfite conversions I usually perform two rounds hemi-nested.

you would need to start with a lot of gDNA for bisulfite conversion to only do one round.

There is only one kit I know that does not degrade your DNA (ie: what you start with is what you get) from Genetic Signatures.

Nick