Insert alone gives colonies?

Alistair Forrest zzalforr at
Thu Aug 24 02:24:44 EST 2000

You need to do a simple experiment.

bugs alone - check they are not contaminated
bugs plus mock ligation - check your water, ligase etc is not contaminated
insert alone no vec - check the insert is not contaminated
insert plus ligase - contaminating linear
vector alone - check that it's fully cut
vector plus ligase - check your CIP treatment and/or double digest is ok
vector plus insert plus ligase - the actual cloning

Do colony counts on all of these. Use the same batch of ligase, bugs etc
for the whole set, if necessary pool batches of bugs.

If you get colonies for the mock ligation you know one of your reagents is
contaminated. Make sure you do a Mock purification of insert. eg. do a
purification on some water. pretend you have DNA there.

I have seen plasmid contamination of:
-glass milk used in geneclean/bresaclean kits. Aliquot your glass milk,
esp. if shared use.
-BUGS, double check
-electroporation cuvettes, especially if you recycle them (I still use
chemically competent)
-insert with uncut vector
-vector with uncut vector - make sure you gel purify, and CIP just in case
it's amazing how much single cut vector can be there in your double digest
and if they're the same size, you have no idea till it's too late.

Regarding dh5alpha I have seen them spit an 11kb plasmid in a maxiculture
overnight. Growing too quick or something uses up the amp. Tested this by
doing a mini on the starter culture - plasmid ok, and an aliquot of the
maxiculture - no plasmid. When I switched to XL1-blues they maintained the
11kb plasmid fine. They grow slower. I've also switched to STBL2 when I
started getting probs with the XL1 blues.

I found I get higher competency with the XL1 blues, but higher stability
with the STBL2's.( In a caesium the XL1 blues would give more variable
yields 4mg -1mg or nothing, whereas the STBL2's gave lower yields 1-2mg
but just about always gave something).

Anyway, cheers and good luck.

More information about the Methods mailing list