PCR GC-rich genomic DNA

Fri Jul 21 15:41:38 EST 1995

Nancy J. Philips writes:

>	I am having trouble obtaining a PCR product of a 70%GC region from
genomic DNA but not from a control plasmid with the target region. I have tried
1. Annealing 0 - 6 degrees above calculated (Rychlik-Rhoades) Tm 2. various
conc'ns formamide and DMSO 3. template, Mg++, primer conc'ns as for usual PCRs. 
Result: excellent 200 bp product from positive control plasmid template, no
200 bp band from human genomic DNA but a 500-1000 bp faint smear instead. These
primers are themselves 60-65% GC, 20-24 bp, Tms 71-73 degrees, no obvious
periodicities. ...[affects multiple primer pairs]...

I think your problem is that the primers are overly stable, and getting
tied up in nonproductive complexes.  For example, if you add the
effect of 2.5 mM Mg, (no DMSO) on top of the Tm's you quote, these
primers might be annealing to places with as much as 30% mismatching.

1) If it's primer/primer or hairpin interaction, then the primers will behave
as if their conc. is low.  Ampl. will be less than 2x per cycle. This can still
give a nice band from ng amounts of template but the starting conc. of that
seq. in human DNA is much lower so it may not make it to a visible band.
Test by diluting the control template down to the actual conc. present in
a human DNA sample.  You'll need some noncomplex DNA as carrier, or maybe some
tRNA.  If this is the problem, just go more cycles.

2) If it's template to nonspecific template interactions, then you
test by diluting the control template into human DNA to reconstruct a
single copy conc. and see if the human DNA really quenches the control
amplification.  If not, then your PCR results are accurate after all
and you clone isn't representative of the DNA in this human DNA
sample.  If the human DNA quenches the control signal, then you may
want to try reducing the amount of human DNA and just going more
cycles.  Also, raising annealing up to the polymerization temp.,
lowering Mg to barely over the total dNTP conc., and adding DMSO or
formamide may help.  It sounds like you may have already tried some or
all of these; but maybe you should try more than one of these measures

3) I assume that you have some other primers that work in this human DNA.
Otherwise, it could just be general impurities in the human DNA that are
inconsistent with any PCR.

4) If all else fails, make shorter primers; shoot for a computed Tm around 52
calculated without Mg correction, then anneal at 55 using 1.3 mM Mg plus DMSO
or formamide.

Good luck.

Steve Hardies, Assoc. Prof. of Biochem., Univ. of Texas HSC at San Antonio
Hardies at uthscsa.edu

For information on our graduate program in Biochemistry:

More information about the Methods mailing list