priming pcr using wrong species sequences

Pearse Ward misra at
Thu Jul 11 02:26:03 EST 1991

In article <175132 at>, hoopes at (Laura L. M. Hoopes) writes:
> I am trying to use Gnebank sequences for human DNA repair genes to prime
> RT-PCR reactions in mice, and the only protection from evolution I've
> incorporated into the experiment so far is to choose regions that are
> homolous betwe the proteins I want to study and other known proteins,
> hoping they will be the most conserved parts.  The primers are not very
> effective and I now want to use primers that are multiple in the third
> positions of the codons.  Does anyone have good/bad experience with this
> type of experiment?  Any suggestions for length, maximum or minimum amount
> of multiple sequences, temperatures?{  Thanx, Laura Hoopes, Hoopes at

I have used this approach to develop a PCR diagnostic test for an RNA virus
whose genomic sequence is pretty variable. Like you, I chose regions of amino
acid homology (minimum six amino acids), which also had a minimal amount of
codon degeneracy. Then I stuck a ten base restriction enzyme recognition site
to the five prime end of each primer. The resulting primers have 32-128 fold
degeneracy, and are 27 or 28 bases long. I lower the annealling temperature for
the first five cycles to allow the 17-18 homologous bases to bind and the extra
ten bases to be incorporated into product, then raise the annealling
temperature to allow only binding of the full 28 'mer. There does not seem to
be a problem with non-specific binding in the first five cycles if the
annealling temperature is calculated correctly.

More information about the Methods mailing list