Are my bisulfite treated sequences methylated or not?? - (Jan/27/2011 )
Hi all,
I have just started working in epigenetics, specifically I am looking at methylation in fruit flies. The advantage is that I have a genome from which I can design reliable primers. I have designed BS primers in a candidate gene using Methprimer (I know this is has been advised not to be the best but options for primer design seem limited with macs...). I extracted and treated whole flies with sodium bisulfite using the Qiagen EpiTect extraction and bisulfite treatment kit. I then performed PCR using hot taq start.
My first results were promising if messy. I got nice sequences but each peak had a minor peak of roughly half the size of the major peak. However, I did see cytosine major peaks and these only occurred at some CpA positions as predicted for flies, so I was hopeful. I thought the minor peaks might be due to differences in cell types across the whole fly, but also thought it might be inefficient BS conversion and amplification on non-converted DNA. Testing my primers with normal DNA showed they did amplify it, but minimally. So I thought to optimise my PCR reaction by increasing Ta and reducing cycle number from 45 to 40, as I have been getting lots of DNA with clean tight bands.
When I did this, I got very clean sequences but with no indication of methylation! So was I previously seeing unconverted sequence amplification? If so why were major peaks cytosines?? It is worth noting that I raised the Ta quite a bit above the predicted melting temp of the primer....this may mean something... Could the BS converted DNA degrade over time or with freeze thawing? Having a positive control would be useful, does anyone have a recommendation?
Here are my primers and the unconverted sequence (sorry for the reverse being backwards - was easier to paste the seq from the methprimer file..):
Forward:
TAGTATTTCTGAACTCAGCAAATAGGTAAG
||||||||:||||:|:||:|||||||||||
TAGTATTTTTGAATTTAGTAAATAGGTAAG
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
Reverse:
ATAATCACTAAATTACCTTGTCTAGACTTG
|||||:|:|||||||::||||:||||:|||
ATAATTATTAAATTATTTTGTTTAGATTTG
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
If anyone can help, I would be grateful, it is frustrating to get nice sequences but not have enough knowledge to know what I am seeing!
Thanks,
Epiconfused.
Hi there epiconfused,
how did you obtain the sequencing data? was it direct sequencing or cloning and then sequencing?
I suspect if you are using direct sequencing then your primers seem to be mispriming but it's hard to comment without further information.
Hi there epiconfused,
how did you obtain the sequencing data? was it direct sequencing or cloning and then sequencing?
I suspect if you are using direct sequencing then your primers seem to be mispriming but it's hard to comment without further information.
Hi there,
yes, sorry it was direct sequencing as they are known homozygous lines meaning I don't need to worry about heterozygotes. I know my primers will anneal to un-converted DNA as I did a little test of his, so mispriming does occur.
So would this mean I should simply re-design primers with more converted C's so as to ensure amplification of the converted strands? Previous posts have indicated Methprimer to not be so good for primer design, any suggestions of other software for a Mac user?
Cheers,
EpiC