PCR with 3' mismatch

Curt Ashendel ashendel at aclcb.purdue.edu
Thu Sep 1 11:07:46 EST 1994


On 31 Aug 1994 14:30:24 -0700, 
HARDIES at THORIN.UTHSCSA.EDU  <HARDIES at THORIN.UTHSCSA.EDU> wrote:
>m-brazil at nimr.mrc.ac.uk (Melanie) wrote:
>> Regarding primers with a mismatch at the 3' end. Has anybody used this
>> method to distinguish between the presence (or absence) of point mutations?
>> Any tips greatly appreciated.
>We messed around with this for a long time and it's a bear.  The primers tend
>to prime in spite of the mismatches.   We could get preferential priming by
>cranking the annealing temp. up to its maximum workable value; but we never
>completely suppressed the false positive signal.  I don't recommend this
>approach.   My understanding is that LCR has much higher specificity, although
>we haven't tried it.
>
I have never done this, but I heard of a way around the lack of complete 
suppression of the mismatch priming:  To make the penultimate nucleotide a 
deliberate mismatch.  The double mismatch never primes, but if the 3' 
nucleotide matches priming still occurs, albeit somewhat less efficiently 
than normal.  This method has a name/acronym, but I don't remember what it 
is (sorry).  I am under the impression that this method is fairly widely 
used for detecting mutations, but I could be mistaken as this is not my 
field.  Obviously, to get a positive PCR signal with mutations, the 3' 
nucleotide would have to be three-fold degenerate (all three possible 
mutations), while for a positive PCR signal only without a mutation, use 
the wild-type 3' nucleotide.

My apologies if this point has been made already in this thread, as I 
did not notice this thread until I saw this reply.


Curt Ashendel
Purdue University
West Lafayette, IN
ashendel at aclcb.purdue.edu



More information about the Methods mailing list