Bonehead ligation driving me crazy

John Ladasky ladasky at
Wed Feb 27 13:45:52 EST 2002

Hi there.

This should be really simple, but somehow it isn't.

I have a plasmid which contains a gene I'm studying, fused to a yellow
fluorescent protein (YFP) gene.  I have decided that I want to do some
fluorescent energy transfer (FRET) studies with this gene, and
therefore I need the cyan fluorescent protein (CFP) chimera.  CFP and
YFP differ by only five codons.  If one sequence isn't toxic, the
other shouldn't be, either.  We express both CFP- and YFP-containing
plasmids on a regular basis, in prokaryotes and eukaryotes.

I have another plasmid which contains the CFP sequence.  I had two
ways to make the swap.  At the 5' end of the fluorescent protein
genes, I have a BamHI site in both plasmids.  At the 3' end, I have
BsrGI or AflII in both plasmids.  Initially I chose AflII for the 3'

I cut both plasmids, removed restriction enzymes with
phenol/chloroform, ethanol precipitated, resuspended in water,
dephosphorylated the vector fragment, mixed, and ligated for an hour
at room temperature.  I had about 250 ng each of vector fragment (4
kb) and insert (1 kB) in each 20 microliter ligation reaction.  All
very standard.  I've done it a hundred times before.  I transformed my
E. coli with some of my ligation and, as a control, a few nanograms of
uncut plasmid.  I was surprised to see that I had hundreds of colonies
on my control plate, but none from my ligation.

I had a closer look at the ligations.  I performed controls with
plasmid DNA digested with only AflII or only BamHI, which suggested
that AflII sticky ends were not ligating well.  I tried a vial of
AflII from another lab and got the same result.  O.K., then, I
thought, I'll try BsrGI.  Perhaps the weak TTAA overhang of AflII
needs a lower ligation temperature?  EcoRI, with an AATT overhang,
does not require low temperatures in my hands.  I've also done some
BsrGI cloning recently with no troubles (overhang is CTAG), so this
seemed the easiest way to proceed.

I performed both cloning procedures in parallel so that I could
observe the differences.  When things start geting weird on me, I
always perform control ligations.  These are a self-ligation with cut
vector and with cut, dephosphorylated vector.  This way I can see that
the phosphatase is working, and also that there are fragments present
in the ligation with insert which are not present in the
self-ligation.  I could clearly see that the BamHI/AflII ligations
with insert produced no fragments which were not also present in the
self-ligation.  The BamHI/BsrGI ligations with insert looked much more
like I would expect when things have gone well, with several novel

So, once again I transformed E. coli with the BamHI/BsrGI ligation,
along with uncut plasmid as a control.  And guess what?  Once again I
got hundreds of colonies on the control plate and NOTHING from my

Shoot me now!  :^P

Actually, any advice is greatly appreciated.

John J. Ladasky Jr., Ph.D.
Department of Biology
Johns Hopkins University
Baltimore MD 21218

More information about the Methods mailing list