Dear Somnath Paul,

I hope I got you right, that you want to use the msp primers for the methylated version and see if the ct of your sample is different (i.e. higher ct) from the fully methylated one. Yes, you can do that, but the information that you get from that is very poor. The only thing that you can tell about your sample, is that it is less methylated compared to the fully methylated standard, not even by how much. You would need to do a standard curve for the meth and unmeth MSP primers to be able to quantify methylation levels. I never did a qMSP, so please check other refs on how a qMSP is done. Methylation of LINE1 is very tricky, I´m also struggling with that at the moment (but want to do that via HRM). The only method that I came across, that really works great is streptavidin coupled Pyroseq (qiagen). 

all the best

Dear Olivier,

You got me correct. I want to take the opportunity to quantify this. Actually for every sample, two sets of DNA will be done. Set A will be MSssI treated and the Set B will be natural, i.e untreated. Let me elaborate it for a better understanding of what I think out as possible technique. But again, with no reference at hand, let me tell you, this is just a thought over the coffee table. Do let me know whether you think it's feasible. Others are also welcome:

As we all know repeat elements are in quite a large number within the human genome. But methylation pattern of all might not be equal. It's all upon chance, in this respect. I want to detect the methylation status of L1 (LINE-1) in arsenic exposed human populations with various grades of patho-physiological outcomes.Now, since all the copies of variably methylated for L1 in a human genome, I would like to follow the following method:

1) Use two sets of DNA from each human sample. Set A will be treated with M.SssI to methylate all the C's maximally. This would incorporate the L1 C's which are not methylated. So we can consider all the L1 C's are methylated. Set B would be the same DNA from same sample, but not treated with M.SssI. So we have differentially methylated C's depending on the natural conformation of the C's in the human sample.

2) Both Set A and Set B are bisulfite converted. We use Qiagen kit.

3) Then both the sets shall be subjected to qRT PCR for MSP primer. The MSP primer will only be for the methylated L1 target region. This primer is reported and designed from already reported studies.

4) The Ct value of Set A and Set B shall be recorded in triplicates. Now, to detect the variation in methylation profile (more methylated L1/ less methylated L1) of each human sample, difference of mean Ct values for individual samples Set A and Set B will be calculated.

The concept which I am trying to figure out is that since MSssI will methylate all the C's compared to the natural genome; then if the natural untreated genome posses less methylated L1 C's then the Ct value difference will be high, as the MSP primer will form amplicons from methylated C's (as per their design). Based on the difference in the values we can seggregate the groups with higher methylated (hypermethylated) L1 compared to lower methylated (hypomethylated) L1 and then associate the target patho-physiological outcomes, if any.

Thanks,

Somnath

Every new method has to be validated to a currently used one, and you should also show, that your method can distinguish between dna standards (0%, 25%, 50%, 75%, 100% methylation level).

You cannot quantify the methylation level, you only get the information, that it is not fully methylated. If you want to do statistics you cannot use raw ct values, as they reflect the template amount. You could include a reference gene, from which you know it is there twice per cell. But at the end you will still have normalized ct values. Ct from the fully methylated dna, ct from the untreated dna, and the reference ct, The next think is to check, that their ct values are more or less the same (-+5), but L1 elements are 10,000 times more abundant. You could normalize that by measuring the DNA content via nanodrop, but the accuracy might not be sufficient for downstream statistics.

Your untreated sample will have less methylation grade compared to your M.SSSl. treated one, as no one has completely methylated L1 elements.

Zymo has two and cheap methods to quantify the dna methylation level, have a look at them

OneStep qMethyl™

5-mC DNA ELISA Kit

All the best

Yes. Its quite a huge job, specially standardization. But thanks for your inputs.