Perplexing Bisulfite Sequencing Problem - (Aug/18/2009 )
I am trying to diagnose Prader-Willi/Angelmans Syndrome by using bisulfite conversion, followed by cycle sequencing. Normal individuals (with one methylated copy and one unmethylated copy) should appear heterozygous (A&G or C&T) at certain base pairs. I used BiSearch to make the primers (I can put the sequences up if you want) and ABI Amplitaq Gold to do the PCR's. I used Dye Terminator sequencing, with the forward and reverse primers used for the PCR and came up with homozygosity in all the amplicons I tried. I used Taqa1 to digest the PCR product to see if there were two different products, and sure enough, there were (the enzyme cuts the non-bisulfite modified strand and leaves the bisulfite-modified one alone). It would appear as if somehow, I'm not sequencing both strands of DNA, even though they are getting PCR'd.
Any idea what might be causing this??? It's really annoying, and I've never had a problem like it in the past.
Thanks!
Hi there,
it may help to see your primers, as you are amplifying strand specifically, your TaqI digest will tell you if you have good bisulfite conversion of methylated HpaII sites.
Are your primers targeted to the methylated template alone or the unmethylated template alone? sounds like to be case.
methylnick on Aug 19 2009, 03:03 AM said:
it may help to see your primers, as you are amplifying strand specifically, your TaqI digest will tell you if you have good bisulfite conversion of methylated HpaII sites.
Are your primers targeted to the methylated template alone or the unmethylated template alone? sounds like to be case.
Here are my primers sequences (for Human SNRPN_IVS4):
1F-GTTTGTAAATTTTTTTTTTTGAA
2F-GTTTAAGTTGTGTTTATTTTAATT
3F-TTTTTTTTGTTATAAATGTTTGT
1R-CATTACTATAATTTTCTCTAACCC
2R-CATTACTATAATTTTCTCTAACCC
3R-ATATATTCCTCAAACCAATCCTA
BiSearch made the primers, and from what I see in the literature, the program is designed to make primers away from any CpG regions. Since non-CpG C's aren't methylated, then they'll all turn into T's and the primer will be OK for both strands, but it's very possible that I have a rather serious logic flaw!
I've tried all 9 combinations of primers and none of them give me heterozygous sequencing (even though some gave me the expected band after Taq1 digestion of the PCR products).


potentially you have incomplete conversion or the primers you are using are not specifically amplifying fully converted DNA, this is why you are seeing the mixing from TaqI but not on the direct sequencing results,
are you seeing C/T double peaks (or G/A depending on the strand) in your chromatograms at all?
Nick