High G/C PCR

Jerry Kropp jkropp at itsa.ucsf.edu
Thu Feb 13 15:25:23 EST 1997

In article <josh-1202971251450001 at>, josh at panda.uchc.edu (Josh
Goldstein) wrote:

> In article <32F8498F.1765 at ic.ac.uk>, k.desmet at ic.ac.uk wrote:
> > X Jia wrote:
> > > 
> > > Hello:
> > > can anyone give me some advises for my PCR on a piece of DNA that contains
> > > about 70% G/C? I have a lot of troubles to make the PCR working.
> > > thank you very much
> > > X. Jia,
> > > Xjia at uci.edu
>    I've found that 10% Formamide makes all the difference in high GC PCR. 
> Some people also talk about DMSO but I haven't tried that.
>    Also I use Vent polymerase, which is more thermostable than Taq.
> Josh Goldstein
> University of Connecticut Health Center
> josh at panda.uchc.edu

Weird. Some people swear by formamide and I had no success when I tried it
on two human genomic templates. Adds weight to my suspicion that PCR is
more art than science.
What has worked for me is 7.5 to 10.5% DMSO, usually with a hot start:
cook the mix sans enz. at 98° ~5min., lower temp to 90°, add enz., begin
cycling with the melting step. That worked for at least one other poster
to the group.
One Leishmania template (75% C/G) also required 50mM tetramethylammonium
chloride (TMAC), whereas another region of the same organism that had a
more normal base composition bombed with the stuff.
Like I said, weird...

"If it ain't broke, it needs more features"-plagerized

More information about the Methods mailing list