Extra Long PCR reamplification

upyers mcoon at u.washington.edu
Fri Mar 22 12:52:31 EST 1996

In article <chumakov.1177826046B at newssrv.dcrt.nih.gov>,
chumakov at helix.nih.gov (Konstantin M. Chumakov) wrote:

> Does anybody know why Extra Long PCR reaction (XL PCR kit, Perkin/Elmer)
> works great the first time when it is seeded with cDNA, but when I try to
> re-amplify the product with the same primers , I can see only smear all over
> the molecular weigth range?  When homogenous (even gel-purified) XL PCR-made
> DNA about 7500 bp is being amplified, it is not before dilution factor
> reaches E-8 I can see a weak specific band in agarose, which disappears very
> soon upon further dilutions, and never is as strong as in the first round
> primed with cDNA?
> I would appreciate any explanations and suggestions on how to effectively
> re-amplify XL PCR products.

Dr. Chumakov

In my experience I've found that using nested (or hemi-nested) sets of
primers for second round PCR usually minimizes the smearing background.
However I've found two other causes (perhaps this is where your trouble

   1. Too much target DNA in either the first or second rounds (or both).
Diluting the second round sometimes clears up the problem, but sometimes
it requires dilutions of the first round. The rationale for this is that
the presence of significant amounts of misprimed products at early steps
in either round will result in a significant loss of one (or both) primers
in subsequent cycles. The smearing results from linear (not 2nth)
amplification of misprimed targets.

   2. Sometimes I've been able to reduce background smearing by modifying
the primer concentration. I've found that concentrations ranging from 5
pmol/100 ul rxn to 40 pmol/100 ul rxn work well, but the range can be
narrow for some primer pairs and/or targets and needs to be determined

It is unclear to me why, but cDNA often gives me smeary bands, although
they tend to be smaller MW smears than my intended product. I think it has
to do with the lack of processivity of RT, but I haven't rigorously tested

When I've run into these problems I was using nested sets of primers. If
you don't use nested sets, I cannot be sure any of these suggestions will

Good luck


"The two most common things in the universe are hydrogen and stupidity" Harlan Ellison

More information about the Methods mailing list