problems with cloning into a pET15b vector - (Aug/09/2006 )

Hello-
Well, I have recently been attempting to clone an 800 bp insert into a pET-15b vector. I incorporated BamHI and NdeI sites onto the gene (as well as the "tails" suggested by NEB to allow for endonuclease binding). I did a double digest of BamHI and NdeI on the pET vector, and then dephosphorylated with human placental alkaline phosphatase (the lab didn't have and CIAP). I purified the digested vector on a low melt agarose gel, excised the band, and purified with glass milk beads. I also did a double digest with the same enzymes on the PCR product, purified by phenol:chloroform, ethanol precipitation, and then ligated with the QuickLigase kit from NEB. I also ran a control, in which I ligated the digested, dephosphorylated vector. I transformed the ligation products into OneShot TOP10 cells, the ligation reaction (with the insert) produced about 21 colonies on the LB/Amp agar plate, the ligated digested, dephosphorylated vector produced zero colonies, the unligated digested, dephosphorylated vector produced 4 colonies, the negative control (transformed with dH2o) produced zero colonies.
I picked ten colonies, grew them up in LB/Amp broth, and miniprepped the plasmids using the Wizard kit from promega. (I have done over 30 minipreps total, but only ten using a dephosphorylated vector). I then did a restriction digest analysis on all of the plasmids, with BamHI and NdeI. None of the miniprepped plasmids showed the 800 bp insert, and the only bands that appeared on the gel were around 3 kBp...which is strange considering the pET-15b vector is ~5000 bp. I have also performed PCR off of the miniprep plasmids using the primers used in the initial amplification of the gene. The PCR produced bands at ~800 bp on multiple plasmids that had not shown fragments in the restriction digest. In short, I am confused, and would appreciate any help I can get. Thanks very much.

sometimes you don't get good cutting after religating the ends together. perhaps the ~3kb band on your miniprep digestions was supercoiled, uncut plasmid that may be correct?

if I were in your shoes, I would try to cut with an enzyme that's not Nde or Bam. options would be, one that cuts inside the insert or vector that will linearize so you can check size; another option is to find one that cuts once in the insert and once in the vector.

it may be that you have it, you just have to find the right way to show it

Did you engineer your NdeI site onto one of the primers used to generate your insert?

NdeI does not cut well at all when it's at the end of a sequence (see here. When we use NdeI in this way (which we do somewhat frequently), we will first TA clone the PCR product, and then digest the TA plasmid (say with NdeI and BamHI) and recover our digested insert from a gel of that digest. This is the only way we've found to be assured of having a properly digested NdeI end on a PCR product.

The second suggestion I'd make is to skip the pET15b dephosphorylation step entirely -- it's not needed because your plasmid is linearized with two enzymes that generate incompatible ends. Thus, it is of no help to you, and can actually hurt you -- dephosphorylation can damage the ends of the vector if done incorrectly (too much enzyme, too long a time, etc.).