PCR: Why is my TAQ unfaithful

Robert Preston rapr at med.pitt.edu
Tue Nov 29 13:50:14 EST 1994

In article <3bd5st$2tp at sun4.bham.ac.uk>, K.M.Toellner at bham.ac.uk (Kai-M.
Toellner) wrote:

> The Taq buffer supplied with Promega Taq Polymerase has pH 9.0, which is 
> rather high compared to the usually used Taq buffer, which has pH 8.3.
> In
>   JN: NUCLEIC ACIDS RESEARCH 1990 Vol.18 No.13 pp.3739-3744
> was shown, that the error rate of Taq-polymerase rises 60-fold, when the 
> pH is increased from pH 5-6 to pH 8.2. I assume, the error rate will 
> rise further, when the pH is increased to pH 9.0. 
> So try another source for your Taq or carry on with the cheap Promega 
> stuff and ask PE for some tubes of buffer;-)

Citations of pH values supposedly used during PCR are not very informative
unless the identity of the buffering components are known.  Especially
with Tris, which has a pKa temp. coefficient of 0.03, huge changes in
pH occur during temperature cycling.  Different buffers (it can't be
assumed that everyone always uses Tris) that have different temp.
coefficients will give very different pH values under cycling
conditions that are usually far from 25 C.

Suzanne Cheng and Wayne Barnes used pH 9 (pH measured and adjusted at room
temp) Tris or Tricine buffers for some of their "Long and Accurate PCR"
protocols. (but they also added small amounts of proofreading Pfu or Vent
polymerases and/or cosolvents, and sometimes 32 mM NH4+, which also
buffers, with pKa=9.1...the situation can be complicated...with NH4+
present, you can also lose volatile NH3 in uncapped containers, esp.
at high temps, and that drops the pH substantially).

A small change in efficiency per cycle (e.g., due to pH problems) can be
critical for the success of an exponential process like PCR.  You need
to "think exponential" to deal with PCR problems. (this must have SOME-
thing to do with Taq error rates?  ;-)  Hmmmm, wonder what.

Robert Preston                                    rapr at med.pitt.edu
U. Pittsburgh Pathology Dept.                   Pittsburgh PA 15261
vox 412-648-9573                                   fax 412-648-1916

More information about the Methods mailing list