Inefficient PCR Cloning (T-myth?)

Jim Graham graham at biodec.wustl.edu_nospammmmmmm
Wed Aug 26 07:20:32 EST 1998

Inefficient PCR Cloning (T-myth?)

A couple things. Only some 30% of Taq extension products have
non-templated residues in my own determination and as stated in the 98
NEB catalougue. Also, a single base overhang is MORE difficult to ligate
than a blunt end, according to the latter. Therefore, why would the
T-vector strategy have been necessary? (See also orignal papers
describing at best <20% white colonies in TA cloning.)

In trying to create a "library" of random primed cDNA, I'm finding that
no matter whether kinasing and blunt end ligating, or digesting linker
sites (either 5 or 15 bp from the end) I get either poor restriction
and/or terrible concatamerization or vector ligation (self ligation is
fairly good after simple T4 pol and kinase treatment -Biotechniques
17(2) 1994 L. Hood.)

A recent paper (Biotechniques 24(4), 578 1998) clearly demostrates a
contaminant or residual activity which interferes with ligation of PCR
products. The authors show that a phenol extraction/ proteinase K
treatment helps, and succeeds only in combination with two choatropic
salts binding column treatments. This extensive regime is still not
working in my hands, as similar “restricted” PCR fragments will still
not ligate together efficiently. (Quiagen disavowed any knowledge of
residual Taq in Quiaquick eluates and seems unaware of the data in the
paper above when I last spoke to them.)

It seems high time someone describe the nature of the historically
recognized reduced clonability of PCR generated fragments. It cannot be
a "non-templated residue" bogeyman, as this effects only a minority of
of the product. Likewise, I see the effect quite readily with
restriction sites >10 bp from the ends. Even when all ends are converted
to blunt with T4 pol and kinased, I see a drastically reduced blunt
cloning efficiency relative to blunt-end fragments derived from

(Please don't bother to describe how one can obtain a single clone of
interest by screening a few colonies, the issue here is reduced
efficiency rather than a routine 10-20% success rate.)

Anyone with insight here please respond. I would like to hear from
anyone who has achieved library efficiency cloning PCR generated
material by any strategy. 

-I'm left thinking i) Taq is not eliminated, and interferes with as many
as 50 bp at DNA ends in a manner different than adding residues to some
30% of molecules ii) or residual Taq reacts directly with other enzymes
like ligase and restriction enzymes. My best clue so far comes from the
fact that only a 70C incubation with EDTA and subsequent
NaCl/isopropanol precipitation (Wang et al. Biotechniques 17(2) 1994)
does appear to allow subsequent good concateramerization ligation in my
hands (although still apparetnly reduced vector ligation efficiency).

(The evolution of a seemingly "irrational" T-vector strategy suggests
that residual Taq may not react with the minority of molecules with
non-templated extensions, but be preventing ligation of the majority

E mail me and remove the _nospammmmm

J. Graham PhD
Biology Department

More information about the Methods mailing list