Cloning sticky PCR products

Vladimir Svetlov svetlov at
Fri Apr 17 11:24:06 EST 1998

In article <353624BF.117C at biodec.wustl.edu_nospammmmmmm>,
graham at biodec.wustl.edu_nospammmmmmmm wrote:

> Hello Brian and Vladimir,
> The recent Biotechniques article demonstrates a visibly striking
> increase in ligate-ability of digested PCR products with proteinase
> K/phenol and 2X silca column chromatography over proteinaseK and phenol
> alone (2.6X by cfu).

Hey, if my salary would increase 2.6 fold I'd call it striking... 
> The confusion generated by the "distance from the end" has been a big
> setback in understanding this decreased clonability of PCR generated
> fragments. Clearly when restrictions sites are 10 or more base pairs
> from the end this should not be relevant.

Maybe I did not make it clear in the first post, but as a rule I use
"overhangs" minimally sufficient for the overnight digestion (90+%
efficiency) as they are listed in the NEB catalog chart. So, for NdeI it is
extra 7 nt's appended to the cloning site in the primer sequence, 2 nt's
for BamHI, 1 - for EcoRI and AscI etc. On rare occasion that I've used
something not listed there I've used 5 nt extensions with no problem yet. 

> Attention must also be paid to
> the use of polymerases other than Taq, or possiblily different Taq
> preparations. (No definitive evidence of the contaminant responsible has
> been presented, rather its effect is reduced by additional protease
> treatment, organic extraction, precipitations, and column binding.)

I've used Taq-amplified products for cloning only on occasion of making a
randomly mutagenized insert, whereas cloning requiring high fidelity of
amplification was and is being done using Pwo (I've also used Pfu for a
short period of time). I guess the difference in views on this subject that
we have stems from some differences in our methods that are not apparent in
such superficial comparison. I've heard for example that isolation of DNA
from the agarose gels using DE81 paper "does not work". Having made
literally hundreds of clonings using DNA's purified this way, ranging from
10 kb plasmid backbones to mutagenic oligos, I find such a slunder hard to
believe. However, when I tell (in unusual for me mild form) that to the
accuser, s/he usually replies that I must be satisfied with very low
efficiency of the method that s/he finds unacceptable. 
To pay back to all those folks I can only say that if one somehow reduces
the cloning efficiency greatly enough than s/he might expect all sorts of
perturbations like residual Taq activity to start affecting the overall
result to a detectable degree. Scary-scary, aren't we mean...
> The Wybranietz article aslo cites two articles (Gene 112:29-35 1992 and
> NAR 19 p.184 1991) in terms of failure of gel-purification to elimnate
> residuual Taq activity. 

I guess I must have missed those papers when I was making my Taq-amplified
libraries. It seems to work like voodoo - if you don't believe in it, it
can't get ya. 

> Actually seldom are products first gel purified,
> then restricted and then gel purified again, which would be necessary to
> prevent residual Taq from either filling in sticky ends or adding
> non-templated As to blunt ends during the restriction digest.

I never gel-purify PCR products twice. Once they've been through the gel
after amplification I either heat-kill the enzyme(s) after the digest or at
worst PCI them (as I said I started using Promega's PCR clean-up kit lately
to get rid of BamHI and such, works fine).
> (I'm trying a set of ligations with an additional protK, CIA, and 2nd
> Quiaquick cleaup now and will report back. These are the same PCR
> fragments, identicle vector fragment, and same frozen competent cells I
> had twice only inefficiently ligated using Quaiquick and
> gel-purification alone (~200 colonies/10 plates).

So if I'm getting about 100 colonies per plate (having plated 1/8 of the
ligation that in turn used up 1/10 of the gel-purified and
NdeI/BamHI-digested PCR products to clone'em into pET33) I should expect a
"lawn" with those additional steps? BTW having picked 9 colonies from each
of two plates with different inserts, I can report 100% efficiency of
cloning. NB - Pwo was used, not Taq. 

Vladimir Svetlov, Ph. D.
McArdle Lab for Cancer Research
Dept. Oncology
1400 University Ave.
Madison, WI 53706

More information about the Methods mailing list