CpG methylation - (Sep/14/2005 )
Dear Nick and PCRman,
My bisulfite treatment PCR cloning sequencing results show: In a monoallelic patient (see attachment), I have done 20 clones , and 2 of these 20 clones show some CpG methylation (3 out of 13 CpG sites per clone). in 6 controls, there is not any CpG methylation. I am wondering if this will be significant to show it is hypermethylated in this patient? Interestingly, one methylated CpG site is the SP1 binding site.
There is no polymorphism in the CpG island, so I can not show the allelic specificity. But from genomic DNA and cDNA combining result I know there is only paternal allele expression.
Thank you very much in advance for your help and your time.
2 out of 20 showed methylation, while no methylation in all 6 controls. The difference is too little to be of any statisitic significance.
Have you seen any non-cpg methylation in your clones? If yes, those cpg methylation may be just artifact.
I would suggest that you screen more patients with direct BSP sequencing. If there is any methylation, then do a cloning to get a acurate quantitation of the methylation.
Thank you very much for your reply PCRman.
Yes, I understand it is too poor to be statisticly significant
In the sequence, all other C have been changed to T which convinces me that the treatment is complete...
If the maternal allelic expression is partly depressed instead of completely imprinted, do you think the cloning result will be reasonable?
I would go with Pcrman with regard to direct sequencing and also that is not enough to show that it is methylated, statistically speaking.
I see that of the two clones you have methylation on, they show identical methylation patterns suggesting you are getting clonal amplicons, meaning you happened to have sequenced maybe, the same amplicon. This can be due to PCR bias and is a technical problem when you come to cloning and sequencing which is why direct sequencing of you PCR amplicon would give you a more representative answer, however it's more challenging.
To circumvent PCR bias, I normally perform my PCR in triplicate and then pool the amplicons for cloning and sequencing, clonality is an issue and someone in the methylation field would pick up on that. the other way of going about this is to sequence more clones.
hope this helps, apologies for the late reply, i did see your message haiyan, but I am swamped with grant writing, one of the joys of science
Sincerely appreciate your help and time, although the result is frustrating...
I have only 3 of this kind monoallelic patients, it is not common of monoallele with the gene I am studying with. No reports said it is imprinted. In all these 3 patients, the maternal alleles are missing which make me to think about imprinting...I do not know, maybe I should think about histone acetylation as well...but have no experience in it (even with DNA methylation, I have strugelled for a long time...)
Direct sequencing does not give me a nice picture (remember last time I asked why there was one T missing after bis treatment? hoho...so with those different amout of T, the sequencing did look nightmare, but there were some C with T at the CpG site in patients but not in control)
I send you another 2 patients CpG map. all 6 biallelic controls are identically negative. Sorry to disturb you for so many times and do appreciate your kind help
I didn't clarify this earlier, does a red dot signify a methylated CpG site?? if so, then the regions are not methylated at all.
The chromatograms are of direct seqeuncing of PCR amplicons? it's something I get occasionally, messy sequence and it's quite difficult to determine methylation status.
By looking at the direct sequencing chromatograms, I have impression that there is clearly a difference between patients and controls. Although there is backgroud noise, at the position where two CpG sites are located, there are cytosine peaks in patient samples, while in controls, there is not. I would suggest that you go ahead with cloning these samples and sequencing ten clones. I would predict that you can get 2-3 methylated clones out of ten.
Dear Nick and PCRman,
Million thanks for the very helpful reply.
Yes, the red dot signifies the methylated site. The chromatograms are from direct sequencing.
Now I am changing the sequencing primer avoiding the troubling T groups to see if I can get some nice sequence showing the difference. Also, I am doing the MSP with specific primers to see if I can get some significant results between patients and controls. Cross my fingers.
By the way, the gene is only expressed in muscle, unfortunately I do not have any muscle cell lines from any of these 3 patients. Do you think it is available to do CHIP if DNA methylation is excluded? (absolute layman in this field)
Thanks again! Have a nice day!