Inconsistent DNA methylation results from bisulfite-PCR!

[Ryohichi]

Dear all,

I am currently trying to analyze DNA methylation pattern of certain imprinted gene with bisufite-PCR method follwed by sequencing. The PCR has been always working in my hand. However, I got inconsistent results of methylation pattern in one particular sample. The sample is DNA from 10,000 of somatic cells from mouse (sorted by flow cytometry). When I triplicated the PCR individually from the same sample, one result is that all of PCR products were almost free from methylation. Second result was that almost of PCR products were fully methylated. Third result was that ratio of methylated and unmethylated PCR products were the same.
Since other samples prepared in the same way (same number of somatic cells PCRed for the same gene) have been giving us consistent results, I am not sure what is happening to bisulfite-PCR of that my particular sample.

Does anyone have the same experience?

Thank you.
Rio Sugimura

[Mathijs]

When you are using a kit to perform the bisulfite conversion itself, it is IMO unlikely that this could be the cause of the inconsistent results. Just make sure you don't convert more DNA than is advised in the kit manual and use freshly made reagent. Primers and PCR programs to amplify from bisulfite template can be tricky. If your primers are not specific enough for the bisulfite treated DNA you may amplify reisdual or contaminating unconverted genomic DNA. Same goes for the program used. I would run a negative control by amplifying from small amount of unconverted genomic DNA to check how well my primers/program amplify fragments from unconverted template.

[Ryohichi]

Thanks Mathijs so much for the advice!
I agree that DNA content was the key in my previous experiments.

Rio