Site directed mutagenesis

[Dr. Nico Stuurman]

Dear everyone,

I have been performing site directed mutagenesis using the two primer 
method (original reference Deng and Nickoloff (1992), Ananl. Bioch. 200:
81).  In short, one primer introduces the mutation of interest, the second 
primer mutates a unique restriction site in the vector into another site.  
After a first round of transformation pooled plasmid is digested with the 
enzyme specific for the parent vector and the digest used for a second 
transformation.  The method works reasonably well for me. However, two 
problems occurred.  When I used double stranded plasmid as template (as 
the original procedure calls for) I got only strange, much shorter 
plasmids.  This problem was obliviated by starting with ssDNA as a 
template.  My latest problem is, however, that I am getting additional 
mutations.  About 35 bp upstream of the aimed mutation is an extremely GC 
rich region (actually, it never resolves well on sequencing gels).  The 
mutants that I got all have extra mutations in that region, probably 
insertions of G or C residues (the gel is hard to read over there).  Two 
other mutants, made with different primers (although in about the same area) 
did not have these additional mutations.  I assume that the T4 DNA 
polymerase that I am using has some difficulties.  Does any one know how I 
could improve the fidelity of the DNA synthesis reaction? Is it possible to 
include DMSO to reduce problems arising from secundary structure?  
Thanks for your help.


Nico Stuurman
Dept. of Pharmacological Sciences
SUNY at Stony Brook, 
Stony Brook, NY11794
Nico at pharm.som.sunsyb.edu