pET15b ligation probs.

[Harry Witchel]

In article <3vqqrr$f1b at lastactionhero.rs.itd.umich.edu>, jpsmith at umich.edu 
(James Smith) says:
>
>I have posted here before and have corresponded with some of you via
>email, but I'm still having ligation problems.  I have recently switched
>restriction sites in hopes that I may have better luck.  I have been 
>successfully amplifying a 1.5kb DNA fragment via RT-PCR with a Bam HI 
>restriction site included in the primers, 3 bases from the end of the
>fragment (according to NEB's catalog, Bam HI appears to be a good 
end-cutter).
>I've been digesting the plasmid with Bam for 3h (checked on gel for
>complete linearization) and then treating with CIAP.  I've been using
>Bio-Rad's Prep-a-Gene kit to purify the plasmid.  The DNA insert was
>digested for 3 h with Bam and also purified with Prep-a-Gene. 
>
>My ligation mix has included DNA & plasmid in a 3:1 (approx.) ratio.  
I've
>been using 2 U of T4 DNA ligase (BM) and the included 10x ligase buffer.
>My ligase appears to be good as it ligated a HindIII lambda digest ladder
>in an hour at RT.
>
>Following an hour ligation, I ran 10 uL of the 30 uL ligation mix on a 
gel
>and noted a fairly intense smear from my 1.5kb fragment up to a point 
>corresponding approximately to the location of circular plasmid with no
>insert (although no *distinct* bands were observed).  The smear continued
>up beyond the location of linear plasmid, but was less intense in this 
>region.  I attempted transformation with this ligation rxn, but no 
colonies
>were observed (transformation was successful on the pUC18 control 
plates).
>
>I'm at a loss.  I've been trying to get this to work for about a month 
now
>with no success.  I'm new at these procedures, but it doesn't seem like 
it
>should be this problematic!  The smear on the gel described above 
concerns
>me - it seems that if my insert was dimerizing, etc., I would see 
distinct
>bands corresponding to 1.5kb, 3kb, 4.5kb, etc. instead of a smear.  
However
>I've never seen a successful ligation rxn run on a gel.
>
>I'd appreciate any comments or advice anyone could offer.  Thank you very
>much!
>
>Jim Smith
>University of Michigan


Jim:
	Be brave!  You are not alone.  Getting a "trivial" PCR product 
ligation to work has wasted many lab-months.  It looks like you are 
receiving quite a few protocols for doing this, which is nice.  The thing 
to remember is that there are hundreds of ways of doing a ligation 
correctly, but there are millions of ways of doing it incorrectly; every 
single technique ever published works -- in at least one laboratory.  
Obviously getting a technique to work in your laboratory is a matter of 
remembering to do all the steps "between the lines" in whatever protocol 
you choose.
	One control experiment you did not mention is to take your BamHI 
CIAP treated vector, kinase it, and autoligate it side by side with your 
other ligations and then transform; this is a much more useful control 
than a pUC18 transformation because it tells you: your bugs are competent 
and you are using enough vector (and that the vector is not contaminated 
with a ligation inhibitor).  As usual with the potitive control, the 
negative control is to transform with the linear CIAP vector (which should 
still offer no colonies).
	All that being said, I think it likely that the kinase control 
will work for you (ie your vector is OK), and that the problem is that 
your PCR product is dirty, as PCR products typically have all sorts of 
junk with them that make them hard to clone.  At least you will be certain 
that the problem is cleaning up your PCR product (rather than bad vector, 
bugs, or ligase).  A confidence builder is to use your vector to clone any 
old insert from another vector you have lying around: it doesn't solve 
your problem, but at least you cloned something.
	In terms of cleaning up PCR products for cloning, the method of 
choice is running a NuSieve LMP agarose gel (no financial connection to 
FMC) and then cutting the band out of the gel and getting the DNA out of 
the agarose.  There is a huge thread in this newsgroup concerning the 1000 
ways of getting DNA out of agarose (and of the 1,000,000 ways of not 
getting it out of agarose).	
	Courage.  Remember that the longbowmen at Agincourt defeated the 
French knights who had armour and were on horseback, because the English 
had technology...
	Bon chance
	Harry Witchel
	University of Bristol
	UK