Primer synthesis problems

Paul Shinn pshinn at
Fri Dec 3 20:46:14 EST 2004

I synthesize our  oligos on a 96-well Polyplex.  Most
of the oligos are 43 and 56mers used for PCR/cloning.  There
is a SfiI restriction site at the end of each oligo that
allows us to directionally clone our products into our home
prepared vector using T4 DNA ligase.  The vector/PCR junction
sequence must be perfect and right now we are having some major
issues with that.

I monitor the trityls after the deblocking step and they are
visually good.  I don't have a spec but after 5 years of doing
this, they look very good--nice bright orange even after 40
couplings.  There are no mechanical indications from the synthesizer
that any valves are blocked/broken.  The support/amidites/activator/deblock
must be good for me to see good trityls.  I can't really troubleshoot
the CapA/CapB/oxidizer other than I know the valves aren't blocked
and the QC from the manufacturer is good.

The PCRs are good (first 21 bases anneal to our target) but even
oligos with bad ends will give good PCR.  We are getting decent
transformants but the junctions are bad.  Some oligos will have
deletions or insertions (bad Caps?).  Some are blunt end ligated
to the vector and the restriction site is missing.  Sometimes a
few bases on the 5' overhang from the vector side are missing as
well as several bases from the oligo side and it's blunt-end ligated.

Now with long oligos like these, only a percentage of them will be
full length under ideal conditions.  (56% of the 56mers and 64% of
the 43mers).  Indeed more problems are from the 56mer side but I
don't see why we seem to be getting so many more bad clones now.
Under ideal conditions, only PCR products with good Sfi digested
ends should preferentially clone into our vector.  If one good end is
ligated on and the other end is bad and flapping in the wind, it
does seem possible that overnight the bad end will be ligated.
What I need to do is maximize the number of good ends.

Can anyone address these issues?  The synthesizer is doing the best it
can do.  I don't see any problems with it but am I missing something?
Could our cloning problems be from bad vector?

We sodium acetate/ethanol precip our PCR rxns and then Sfi digest
overnight.  The digest is sodium acetate/ethanol precipitated
and ligated overnight at 16C.  We are doing 96 rxns at a time so
gel extractions of our PCRs or PAGE purification of our oligos
really isn't feasible.

Thanks, Paul

More information about the Methods mailing list