terminal transferase & PCR, Brian Foley are you listening?

Brian Foley brianf at med.uvm.edu
Sat Aug 13 12:32:49 EST 1994

Zophonias O. Jonsson (zjons at vetbio.unizh.ch) wrote:

: Some months ago you posted a message on the methds-reagnts newsgroup where
: you suggested using terminal transferase to change PCR bands into "smears"
: before cutting them with restriction enzymes. 

	I wrote back to Zophonias personally by e-mail, but
I'd like to add a bit more here.

	This was one method suggested to improve the efficiency of
cutting a PCR product in which the restriction site(s) are at the
very 3' end of the primers.
	If you are designing PCR primers with restriction sites, 
first check the table of restriction enzymes to find out which enzymes
will cut when they are at the very end of linear DNA.  One
such table can be found on page 274 of the 1994 Stratagene cloning
systems catalog.  This should help you avoid needing tricky
protocols for cutting your PCR product.

	On the realted subject of cloning PCR products.  Many
people recommend using Klenow fragment to "polish" the ends of
the PCR product.  Current Protocols In Molecular Biology 
(J Wiley and Sons publishers) mentions that Klenow will
remove the 3' A residue left by TAQ and allow blunt-end
ligation to occur.  
	A recent publication (not peer-reviewed I assume?)
in Strategies volume7 number 2, pp 47-48 shows that Klenow
does not produce blunt ends from PCR products.  They
show that T4 and Pfu polymerases do a good job, but that using
Klenow was no better than leaving the PCR product as-is.

	This paper also claims that Taq polymerase does not always
add a 3' A , but that it will preferentially ad a 3' G if the 
primer for the other strand ends in a C.

	Can we trust publications in catologs?  Is there
anyone watching these people for truth in advertising?

	If is search the archives of this newsgroup for "cloning
PCR products" I can find hundreds of postings on this topic.
There are many protocols for generating your own "TA" cloning
vector, for "polishing" the PCR product ends with Klenow, and
so on.  Thus I assume that I am not alone in finding it
sometimes difficult to ligate PCR products.  A "quick and
dirty" ligation is often good enough, if we only need one
or two clones.  But some applications call for very high 
efficiency of ligation in order to generate a library of
cloned products.

	Can we trust advice given out in this newgroup?
What can we do to increase the level of trust we can
have in this nbewsgroup?

	My own solution has been to add EcoRI or BamHI 
restriction sites to the 5' end of my primers, plus 3 or
4 bases so that the recognition sequence is not at the very
end of the linear product.  This works well in my hands.
I still do not have an efficient method for ligating 
PCR products "blunt ended".  I can get a few clones,
but always with low efficiency.
*  Brian Foley               *     If we knew what we were doing   *
*  Molecular Genetics Dept.  *     it wouldn't be called research  *
*  University of Vermont     *                                     *

More information about the Methods mailing list