Positive Control for Methylation Studies in BSP sequencing - A "Universally" methylated Gene ?? (Apr/24/2006 )

I thought it would be really convenient in performing methylation studies if there was a positive control. I know positive controls can be purchased and that they're quite costly. But isn't there a known gene that is consistently methylated throughout the human population? If so, anyone can easily design primers and use this as a positive control in their methylation study. Especially for BSP sequencing. I think everyone here may appreciate having access to a documented well established, standardized positive control with primers too.

I'm just throwing this idea out to the forum to see if anyone has any interesting insight to share about this. Having a positive control would be very helpful to know if the bisulfite reaction worked properly or not. Sometimes I feel that the Bisulfite reaction is maybe too strong and as a result converting not only unmethylated C's to T's but also methylated C's (cytosines) into T's(thymine or uracil). I'm not sure if this is actually possible but a positive control would certainly help determine that.

Thanks for any input

I thought it would be really convenient in performing methylation studies if there was a positive control. I know positive controls can be purchased and that they're quite costly. But isn't there a known gene that is consistently methylated throughout the human population? If so, anyone can easily design primers and use this as a positive control in their methylation study. Especially for BSP sequencing. I think everyone here may appreciate having access to a documented well established, standardized positive control with primers too.

I'm just throwing this idea out to the forum to see if anyone has any interesting insight to share about this. Having a positive control would be very helpful to know if the bisulfite reaction worked properly or not. Sometimes I feel that the Bisulfite reaction is maybe too strong and as a result converting not only unmethylated C's to T's but also methylated C's (cytosines) into T's(thymine or uracil). I'm not sure if this is actually possible but a positive control would certainly help determine that.

Thanks for any input

purple fetus,

by positive control I suspect you are meaning positively methylated within your samples. There are a number of ways of doing this, artificially spiking your samples with SssI methylase methylated DNA. SssI methylates all CpG resides with high efficency. another method depending on your cell lines is to use tumour suppressor genes (for cancer lines) or oncogenes (in normal lines) for positive methylation. However, it is not entirely necessary to control for this, as the major problem I usually see incomplete conversion and this can be seen in humans as non-CpG C's appearing on your chromatogram. Although it is possible to get over conversion, the reaction is in a way such that you need to react over many days for this to occur as it favours unmethylated cytosines a lot more.

good lcuk!

Nick

Hey Nick and Jim,

I'm currently working on the gene SMAC DIABLO. Nick, I was getting at using an oncogene for a positive control. A gene with numerous CpG's methylated. They don't have to be from my sample but it would make it easier since my sample is undergoing the bisulfite treatment.

However, the DNA I have extracted is from skin tissue. I don't use cell lines. I don't know if methylation statuses of oncogenes will be consistent from tissue to tissue. (ie. liver tissue DNA versus spleen tissue DNA) Especially since methylation seems to be an acquired phenomenon more than genetic. (Please correct me if I'm wrong) Is there a known oncogene in the skin which is always methylated?

Or even better, is there a known oncogene that is always methylated in the leukocyte DNA? The reason I ask this is because I think blood is easier to acquire for most people as opposed to liver tissue, spleen tissue, heart tissue, skin tissue,...etc. (Although personally for me, skin tissue is easy to get a hold of.) That way anyone can get their hands on some blood DNA, and use a designated pair of primers on a known methylated promotor area of a (most likely) oncogene; and this could be a convenient positive control.

-Thanks

Hi, Purplefetus:

I use pancreatic tissue for DNA extraction. But unfortunately, I can't get the DNA from human samples (The weight of the human samples is about 10 mg including the paraffin, The DNA from cell lines is so much.). I have no idea how many samples (weight) I should use. I tried several methods including Qiagen kit It seems that all of the methods I tried don't work. Could you help me ? Thanks a lot.

Linksky

Hi Linksky,

I don't have an answer for you. I guess this is off topic but I use the Qiagen kit myself. I usually let the lysis buffer dissolve my biopsy sample over 2 days. In my limited experience, the whole sample is usually gone after 2 days. I also mash my sample up pretty good in a 1.5 mL tube w/ the lysis buffer before I let it sit for 2 days. Hope it works out for you.

I have to say I am not aware of any universally methylated genes published, however an oncogene would be the best bet. For tissue specific methylation again there will be variations, it would be just a matter to picking a couple, and assaying them to find out what they are.

Nick

In my quest to find a universally methylated gene for a positive control... I found this paper by Bock C, etal (PMID 16520826 ) In this paper they correlated DNA sequence, and certain repeats with CpG island methylation.

They found this one gene PDE9 to be highly methylated in lymphocytic DNA. I THINK this can be used as a positive control in doing BSP sequencing to assure your bisulfite reaction has worked well. As long as you can spare a microliter or 2 of bisulfite treated DNA, doing PCR on one additional tube may be worth while. According to the previously mentioned predictability model, I don't see why the methylation status should change in any other tissue DNA. I will try this in skin and keep everyone updated with any new findings.