Protocol Online logo
Top : New Forum Archives (2009-): : PCR, RT-PCR and Real-Time PCR

Problem with 3 step PCR - (Jun/26/2013 )


I've been trying to do a 3 step PCR to make an insert to be used for homologous recombination. I am trying to do an allele exchange between my gene of interest and an antibiotic resistance cassette, flanked by FRT sites. First time around, I was fooled into believing that the plasmid I'm taking the resistance cassette from had a promoter/terminator region. I designed my primers to introduce FRT sites on both sides of the cassette. The 3step PCR worked GREAT - first time, every time. Of course my bacteria were never antibiotic resistant.

Anyhow, once I picked up a new plasmid w/ the resistance cassette with a promoted/terminator regions, I tried to do the exact same thing. The overhang sequences on my primers are IDENTICAL to last time (they make up the FRT site, and are the only homologous part between the fragments), and the Tm's of my primers are almost identical (both the partial Tm and the full-length Tm). Fragments 1, 2 and 3 all got amplified just fine by themselves. I then managed to fuse fragments 2 and 3 just fine (first time, every time) but I could not get fragment 1 to fuse with fragment 2. I tried this a few times, played around with annealing temps but nada (while the fragment 2-3 fusion worked every time). *The homologous sequence between 1 and 2, is identical to the one between 2 and 3 (that would be the FRT site, in the same orientation. Theoretically if I tried to fuse 1 and 3, I should be able to)
*the fragments are rather short - <600bp (fragment 1 and 3) and ~1500bp (fragment 2), so length shouldn't be an issue.

I furthermore used a few different softwares to check for primer homo- and heterodimers, and for secondary structure. The only significant secondary structure had a Tm of 41 - well below my annealing temp.

I really have no idea what's going on.

Any suggestions?

Thanks in advance.


You could sequence your template to make sure it has the sequence you think it has. You could make some other primers which do not have 5' overhangs, and which perhaps bind to other regions of your template. You could try one of these primers and one of the new primers, to determine if one of your primers is causing problems. You could reorder the primer -- sometimes they are made incorrectly. You could try additives to your PCR reaction. Is the template GC rich or GC poor?


The GC content is about 50%. I guess one of my primers could be made incorrectly. The 5' overhangs are only there during the amplification of the individual fragments and had no negative consequences. When trying to fuse the 2 fragments, there should be no overhang anymore (since I'm using the same F primer as i did for Fragment 1 and the same R primer as I did for Fragment 2).


i tried your method before and ended up getting a lot of unspecific bands.
i ended up using the old method, which is multiple cloning.