direct cloning of PCR products

Bruce Elder elder at CGL.UCSF.EDU
Tue Jan 19 12:38:54 EST 1993

Yin-Chai writes:

>... I have been trying to clone a 2200 bp fragment from a RAPD PCR
>of mouse DNA (using a single 12 bp primer). I ran out the PCR products
>on a 1% agarose/TBE gel. (Type of agarose don't seem to matter. I have
>tried SeaPlague, SeaKem, and BRL.) I then cut out the band, ligate it to
>Promega's pGEM-T vector, and transform BRL's DH5alpha (no white colonies)
>or Promega's JM109 (24 white colonies max). A miniprep of these whites
>revealed 18 false positives (no insert) and 6 inserts that are smaller than
>the band I cut out (700 bp to 1300 bp). I have tried many variations of the
>above protocol (in gel ligation, NACS purification of the band,
>electroelution, plain phenol choloroform purification of the band etc.) and
>24 white colonies is the best so far, and I have never got any insert
>bigger than 1500 bp.

>Any ideas on what's going on? Someone suggeted homologous recombination.
>I am trying Stratagene's SURE cell (recA, recB, recC and recJ minus). Any
>other ideas? Thanks.  Yin-Chai

We have been using a method that has given clones from 200bp to just over
3kb with great success. It is basically just like yours except for these

1) We isolate small 200bp-2kb) bands from acrylamide gels by cutting out
   the band and eluting over-night at 37 oC in 0.45M AmOAc/1mM EDTA (about
   400ul). This is then phenol:chloroformed, precipitated with glycogen (see
   previous nets on this one!) and resuspended in about 10ul of water. I hurl
  all of this into my ligations (some people in the lab us a fraction).

2) We also use agarose gels to purify PCR products and then freeze/squeeze
  (also discussed in earlier net-news) or geneclean. Once again, I use all
  of the resulting fragment in my ligations.

3) The biggest difference is the vector. We use a vector called pDK101 that
  to my knowledge is not commercially available. You can refer to Kovalic,
  et al. (1991) NAR 19:4560-4563. I think it is ok if I send you some, but
  will have to check on this.

Good luck.

Bruce Elder
elder at
Univ. of Calif. San Francisco

More information about the Methods mailing list