MSP/USP in human tissue - USP not working! (Oct/16/2006 )

Hey All,

I have a problem with my MSP, or better: the unmethylation part of it. Back in the day I tested several cell lines with my MSP and USP primers. As expected, several cancer cell lines were positive for methylation and negative for unmethylation, whereas some primary cells were negative for methylation and positive for unmethylation.
So far, so good. But here's my problem: I also wanted to test in human cancer and matching non-cancer samples. I started with the MSP and it worked beatifully. But when I tested the same samples with the USP primers, I got a band in every sample. I know, that I probably have both templates (for MSP and USP) in every sample. Therefore I decreased the cycle number for my USP from 18 to 16 to 12. But still I get bands every where. I also diluted the template (which is the universal PCR, 1µl in 25µl RXN) 1:100 and 1:1000. Still bands every where! I also increased the annealing temp but all I got was loss of signal for every sample (originally at 54°C, went up to 59°C: bands every where, then up to 62°C: no signal). I also used DMSO to increase the specificity (no effect).
After nothing had worked, I went ahead and ordered some new different primers (designed with Methprimer), but still get the same result.

Does anyone here have any idea how it is possible that the MSP works beatifully and that my USP doesn't (especially since it worked for several cell lines)? Also, does anyone have any suggestions what else I could change? I guess I could go even lower in the cycle number, but would results obtained with - let's say - 8 cycles be valid?

Please, if you have any suggestions, let me know.

Thanks a lot!

Hazel

hazel,

you could have contamination. have your no template, water controls come up with a band?

With the MSP assay, you should be performing the PCR for both M and U primers together, in the same run and machine for a vaild result. It is not correct to lower the number of cycles in your U set and not in your M set.

If you have normal cells in your tumor sample, this would also give rise to bands in both M and U primer sets.

Nick

Hey Nick,

thanks for your reply. No, I don't think that I have a contamination, because I never saw a band coming up in the water control of any of my USP runs.

As far as normal cells in my tumor sample are concerned: I agree with you - I think that I also have normal cells within that specimen, which should give me a band in the USP run in cancer.
However, I thought that this band would come up later. Because, when I performed the MSP I didn't get any bands in the healthy tissue but very strong bands in the cancer samples. Therefore I thought (since the MSP result was so striking) that I should at least see differences between my cancer and healthy tissue samples in the USP run.

Do you - or anyone else reading this - have any other idea what I could change (since the MSP was no problem, I think I could change conditions there too)? Are there maybe any other sites on the internet like MethPrimer? Or do you have any suggestions on what to consider when I design primers? Should the 3' end possibly lie on a CpG site?

Thanks,

Hazel

Hazel,

try methylprimer express from ABI for primer design (see the pinned thread

Nick