Recombinant PCR

Robert Horton horton at molbio
Wed Jun 3 22:19:57 EST 1992

timmer at CCWF.CC.UTEXAS.EDU (, Richard T. Timmer) writes:
: Dear Bionetters:
: I was wondering if anyone out there in "netland" has experience with 
: recombinant PCR....
: My problem is with the second round of PCR (I have no problem with the first
: round of PCR and I get good yields of a single product).  In the second round
: of PCR my problems include the following:  1) no amplification unless a large
: amount of DNA is used for first few cycles (approx. 200 ng); and 2) a
: plurality of products.  
In my experience, the second round of PCR (the overlap extension reaction) has
been quite robust; once I lost my DNA pellet from the first round and just
rinsed whatever traces of DNA might have been left in the tube and went on.
I know that every template, etc., behaves differently, but after using
overlap extension on several unrelated projects, we found only two real snags;
First, if the intermediate products aren't gel-purified you can (not always)
get extra junk in the final reaction. Second, if there is a mistake in the
primer design so that the "overlaps" don't really match, it just won't work
(experience speaking:)).

: Any input on procedure, conditions, and type of thermophilic polymerase to
Some suggestions:
        1) check your oligo sequences.
        2) make sure your intermediate products are clean (we get rid of
           the agarose, too)
        3) make the overlaps a little longer? 10 bp may be too short; our
           rule has been to shoot for a "Wallace temperature" of ~50oC,
           which is usually 15-17 bp.
        4) check the ends of the sequences for potential hairpins. These
           might make it (I guess) difficult to make an overlap. If this
	   the case, longer overlaps might help if they extend beyond
	   the hairpin.

: (e.g. would Vent work better because it doesn't leave a 3' terminal
: non-complementary monophosphate)

Lately I've tried to design primers so that if an "A" gets added by the
terminal transferase activity, it will still match the other strand.
However, in the "old days" before we worried about such things, it 
didn't cause any problems. I did hear someone at a meeting describe a
mutant recombinant PCR-generated molecule that had an extra "A" added
at the recombination site, however. The last time I tried Vent I was
not impressed, though they say the consistency of their preps has gone up.
I do know that Taq CAN work just fine.
	If you figure/d out what is/was wrong, please let me know!
Bob Horton            /\ "Crash programs fail because of the theory that
U. of Minnesota, CBS  || with nine women pregnant you get a baby a month" 
1479 Gortner Ave.    /||\                    -Werner von Braun. 
St. Paul, MN 55108    ^^   horton at 624-3790

More information about the Methods mailing list