From: Lorant Lakatos, lakatos at hubi.abc.hu
Date: Thu, 21 Jul 1994 08:29:34 GMT
In article <lakatos.774779374 at hubi.abc.hu> Lorant Lakatos,
lakatos at hubi.abc.hu writes:
>I am looking for a good detailed protocol for
>inverse-PCR. I have tried some but I did not get
>Thanks so much
I don't think I can give a *very* detailed protocol, because of
variations in the target size, starting material etc, but here are a few
points that I needed to work on before I got any bands. You need to
think about the concentration of DNA in the recircularisation step. For
most targets, monomeric circles will be favoured at about 5ng/yl (a la
Maniatis). Measure DNA concentration after the digest but before
dilution, to get an accurate figure. So, digest, kill the enzyme (I used
heat), recircularise at 5ng/yl DNA, kill the ligase and then digest with
the second enzyme. Kill the second enzyme, and then in to the PCR. Here
is the second major snag - Mg++ concentration in the PCR. Taq likes
about 1.5 mM Mg++, but your DNA will be in about 15mM, from the digest.
In order to get enough DNA into the PCR (about 100ng of mammalian genomic
worked for me) you will probably need to desalt it, and I also made my
own magnesium-free PCR buffer and titrated the Mg++. For large targets,
I used Vent (NEB) and this has the advantage of liking higher Mg levels.
I never precipitated my DNA as I didn't have that much to start with, but
it's obviously an option.
After that, just play with the PCR as with any other PCR. In my hands,
even after a lot of optimisation, the thing was still bloody unreliable,
but it worked well enough to clone a transgene junction or two.