I'm attempting to make several different constructs and they all seem to be recombining. The strategy I'm using is straightforward: I digest my vector and insert, CIP treat the vector, gel extract both, ligate overnight, and then transform into DH5 alpha. My negative control transformation (ligation done with the vector only) is always negative, and I usually get plenty of colonies on the plates with cells transformed with the vector/insert ligation. However, when I miniprep the colonies the resulting plasmids are very small. I expect the resulting plasmids to be over 8 kb (depends on insert), but for these minipreps the most abundant band is at about 1 kb. When I miniprep the original vector I get bands of the expected size.
When I attempt to digest the minipreps, the only enzyme that cuts is one which should cut in the ampicillin resistance gene. When I sequence using vector specific primers, most of the primers don't prime and those that do give very short reads (100 bp or so).
I think that this is probably due to the plasmids recombining to delete not just my insert, but also most of the rest of the plasmid. Does anyone have any thoughts on ways to get around this? I have tried using different competent
cells (XL10-Gold and Stellar), incubating the plates at room temperature rather than 37, and incubating liquid cultures for minipreps at room temperature. No success.
One thing that surprises me is that I'm having this problem with ALL of the constructs I'm making, even though they contain different combinations of vectors and inserts. I know that some sequences are unstable and E. coli doesn't like to make them, but it seems particularly unlucky that this is the case for all the constructs I'm working on.
I hope someone has some advice for me!