deleting one base
Dennis J. Templeton
djt2 at po.CWRU.Edu
Wed Jan 20 11:04:08 EST 1993
In a previous article, wsun at jeeves.ucsd.edu (Fiberman) says:
>This is a question for molbio experts:
>I have a linear piece of double stranded DNA with the sequence
>something like this:
>I wish to delete the first base (C) so I would end up with:
>what would be the best way to do this? I need to end up with a
>uniform population of DNA because I will be using it for a
I can think of two ways, one elegant, and the other involving cloning :=)
A related PCR approach is also included.
1) Many folks ignore the fact that Klenow fragment of Pol I has a 3' to 5'
exonuclease, that is limited only by the fact that the rate of synthesis
exceeds the rate of depolymerization.
Since you want to remove one terminal G from the 3' end of your molecule
(looking at the sequence upside down from the way you wrote it above), I
would incubate the DNA with Klenow in the presence of dA, dC, and dTTP (no
dGTP) the Klenow will remove the G but not go further because any further
deletion will be rapidly replaced from the pools of nucleotide. I would
run this reaction about 10 minutes at 30 degrees, then stop with EDTA and
phenol extract and precipitate. BTW, T4 pol should work for this too, but
we always have Klenow around.
You will now have a one-base 5' extension, and want to remove the single
unpaired nucleotide. For this I would use mung bean nuclease. Use the
regular MBN buffer, 300 mM NaAc pH 4.5, 1 mM ZnCl, and only a few units of
nuclease (maybe 100 to 1000, note that it comes around 10,000 units per ml;
don't just shoot in a ul!) Do the reaction at 4C, for maybe 10 minutes.
Stop the reaction with EDTA, add carrier, phenol extract and precipitate.
Note that the other end of your molecule has been modified too, you may
have to restriction digest or whatever to generate a clean end for cloning
if a blunt end is unsuitable.
2) Cloning the thing.
Find a vector with a unique HphI site (this is not easy, but Invitrogen
sells one that they use for cloning PCR fragments (PCR 1000 I think) Cut
with HphI to leave an end with the sequence GGTGAXXXXXXXX
i.e., a one base 3' extention. Trim off the 3' end with mung bean nuclease
above, and blunt clone in your end. When you re-cleave your piece with
HphI, it will cut off one NT of your blunt cloned piece of DNA, leaving you
with a single NT 3' extention. Remove this with MBN as above, and clean up
A related method would be to include an Hph1 site in a PCR primer. Then
you could cut the PCR fragment with HphI and MBN without having to clone
Which would I do? If you have lots of DNA, I'd use the first method, but
note that the final yield might be only 25% of what you want. The second
way allows you to make as much of the intermediate as you want, and could
have an advantage if you have little DNA to start with.
More information about the Methods