site directed mutagenesis

Tracy Aquilla aquilla at
Tue Sep 6 17:37:08 EST 1994

In Article <CvIE3A.ILt at>,
jgraham at (the End) wrote:
>Don't waste you time with any kit, delve into the literature as that's 
>where you will be headed as soon as you complete the steps on the 
>"card" and don't find your desired mutants.

I disagree. While you should certainly read as much of the literature as
possible, I strongly suggest looking into the Altered Sites protocol, and
NOT wasting your time with a PCR-based mutagenesis protocol. As stated
previously, PCR is very likely to introduce more mutations than you want.
The Altered Sites kit has the highest success rate of any mutagenesis
protocol used in our lab (it has worked every time on the first attempt,
with no additional mutations, EVER), and has the added advantage of
employing a selection procedure that makes recovery of the mutants highly
efficient, obviating the need for screening many mutants, and thus making it
faster, as well as more reliable than PCR. However, if you do try PCR, more
suggestions follow.

>After trying many of the basic approaches, I find that you can almost 
>always obtain your mutant in a few days by incorporating a mismatch 
>primer by PCR. Limit your cycle number to less than 15 and increase the 
>template plasmid level to 200 ng / 100 ul rxn. You will not create 
>additional unwanted changes.

While keeping the number of cycles down is extremely important, the
incorporation of unwanted mutations also depends on the dNTP and Mg2+
concentrations (and particularly the dNTP conc. relative to Mg2+), reaction
pH and temperature, length of the PCR product and the polymerase used. To
minimize unwanted mutations, I suggest the following: keep both the Mg2+ and
dNTP conc. as low as possible and do not let the [Mg2+] exceed the [dNTPs];
use an enzyme with proof-reading activity, such as Pfu or Vent pol; and keep
the pH at about 6 (@ 70C). You should also design the experiment so that you
are amplifying as small a fragment as possible. Once you get some mutants,
it is also easier to confirm the sequence if it is shorter. This is
important, particularly if you plan on expressing a mutant protein;
sequencing is essential to confirm that only the desired mutation(s) is present.

>Should your site lie within a region devoid of flanking restriction sites,
>you can either add sites by further mismatch in the primers if you are 
>altering a protein target, or otherwise try the two PCR "megaprimer"
>method where the entire PCR product is used as a "primer" in a second
>PCR with a third primer which encompasses a restriciton site. Keep the 
>first target as small as possible (<~500 bp) for good results with this 
>Good luck,
>J. E. Graham 
>Institute for Unrestricted Information Exchange
>Ameriscam Inc. Liberal, KS

P.S. Ameriscam Inc.? Is this a scam company, or what? ;-)
Tracy Aquilla, Ph.D.
Molecular Physiology and Biophysics
University of Vermont
aquilla at

More information about the Methods mailing list