I apologize if this appears twice, but it did not appear on the message
board. I waited for 24 hours before reposting.
I'm trying to dephosphorylate cosmid DNA as part of Stratagene's SuperCos I
genomic library protocol. For the past couple of months, I have been
consistently unable to make this reaction proceed to anything that even
remotely resembles completion. I'll describe what I've done.
First, I digested the circular cosmid with Xba I, which cleaves between the
Cos sites, leaving a 4-base overhang on the 3' ends and burried 5'
phosphates. As instructed, I then phenol-chloroform extracted,
chloroform-rextracted, ethanol precipitated, and washed the resulting linear
DNA. A gel confirmed that this procedure gave almost entirely the linear
product with high yield.
Now I must dephosphorylate the 5' phosphates.
First tried Stratage's CIP
[Only registered users see links. ]
I used the number of units recommended in the procedure, ten-fold more, and
hundred-fold more (in separate reactions). I have tried the standard 60
minute incubation at 37 C, and extending the time all the way to 2.5 hours!
After each reaction, I phenol-chloroform extracted, chloroform-rextracted,
ethanol precipitated, and washed the resulting DNA. I then took out an
amount that can be visualized on a gel (~500 ng). I diluted it in water and
ligase buffer. I then split it between two tubes. To one I added ligase and
to the other I added an equal amount of water. I incubated the reaction at
16 C overnight. CONSISTENTLY, ligase is able to catalyze the ligation of the
vector despite these amounts of enzyme and extending the incubation time.
I have also tried NEB's CIP, as well as NEB's Antarctic Phosphatase, which
is thermolabile and supposedly more docile,
[Only registered users see links. ] [Only registered users see links. ]
with the same result. Additionally, I chemically transformed the circular
cosmid vector into E. coli and harvested it using a Qiagen plasmid prep,
rather than simply using the sample that Stratagene provides with the kit.
What could I be doing wrong? What sorts of variables might be affecting my
CIP reaction? Has anyone used this plasmid vector (SuperCos I
[Only registered users see links. ]
). I have already been advised to try melting the strands and chilling
quickly so that CIP can get at the burried phosphoryl groups. Any other
I thank you for taking the time to read through this.
Surf and talk on the phone at the same time with broadband Internet access.
Get high-speed for as low as $29.95/month (depending on the local service
providers in your area). [Only registered users see links. ]
having a real positive control for CIP / SAP means P3(2/3)radiolabelling
some dsDNA - rather nasty.
Anyway, you could switch the template instead: what about some plasmid
sitting in your fridge or on your bench? Most have an XbaI site, so you
would not need to change anything else in your setup.
Also run some tests for exonuclease if you can't re-ligate the plasmid when
you don't CIP/SAP it before. (e.g. incubate linearized or cut plasmid
overnight with solution to test and look for smear or disapperings on gels
afterwards). Be sure to hav a negative control without ligase because small
fractions of uncut plasmid will be overrepresented in the ligation. If you
like, omit the gel purification here.
Make sure ATP in your ligase is present and fresh (some companies include
it in the ligation buffer, some don't). If you're not sure, just add 1ul of
a fresh aprox. 10 mM NaATP stock, a slight excess doesn't harm.
If possible, use SAP instead of CIP, it's easier to get rid of (20 min 60
degC kill it for sure). And make sure your sample doesn't smell like phenol
anymore, most enzymes don't like it at all, so even traces might kill
Instead or in addition, you could use a PCR cleanup kit (like that from
Qiagen - no affil) for the cleanup - they're also suitable for cleaning
other enzymatic reactions on DNA. Just make sure, your cosmid comes off again.
My feeling is that generally the vector alone, even dephosphorylated,
will be forced to self-ligate by the (generally plenty) amount of
ligase and overnight incubation at RT. With insert available however,
there's clear preference for the insert over self ligation.
I can tell you my example where I got stuck: 10 kb vector, blunt ended
(!) and dp. Insert cut, then blunt ended. Ligation with NEB T4 ligase,
with and without insert. Always plenty of transformants
(electroporation), generally more or less similar to expectation. Did
hundreds of colony screening pcr's for insert, generally none. Where's
Finally I added 15% PEG8000 to the rxn mixture; almost all
transformants were with insert, however the control without insert in
the ligation mix also had almost similar amount of transformants,
however (of course) empty and therefore self-ligated.
This isn't similar to your case, as I only got in problems when
ligation didn't work without PEG, and for some reason the plasmid did
it with himself rather than with the insert.
Never needed PEG with sticky ends, therefore not had problems before.
But I guess, when I try vector alone I will get similar results.
In a following up thought, from my last posting you might conclude
there's something wrong with the insert. The XbaI site might be too
close to the end, although my NEB catalog claims it's not very
sensitive to that. Leave the rxn o/n at 37oC anyway!