HI Vijay,
One thing I would like to mention first is that often when a DNA fragment is
smaller than or close to 1Kb, it could enter the low EtBr region on an agarose
gel if allowed to run to long. Its bound EtBr could be stripped off so
significantly that the corresponding band become very faint. Sometimes these
could give you false negatives. I would run the gel with accordingly shorter
time when expecting small fragments. Or I will chose another pair of enzyme,
one of which cuting inside the insert and the other in the vector part. And I
will try to choose a pair that give relatively easy to read gel patterns. Or I
will try screening the constructs with PCR, which has some other disadvantages
as well as some advantages...
Another suggestion is to check your insert and vector fragments' concentration
by running them on an agarose gel. For fragments longer than 100bp, the
brightness of the band should reflect the DNA's total mass - not molarity. The
relative molarity is inversely related to the fragment's length. For example,
if your insert preparation looks almost as bright as your vector, and the
insert is 1.2Kb, the vector is 3.6Kb long, then if you mix them 1:1, you get a
3:1 molarity ratio for the insert:vector.
UV quatitation is not very useful for restrictive fragments, because the
gel-purified DNA samples are often too diluted. If you do a 300nm~200nm UV
spectrum scan for a gel-purified restriction fragment sample, Even without
further dilution, you might only see at 260nm there is a very tiny bump sitting
on the shoulder of the huge salt peak extends to as far as 290nm - or nothing
but the shoulder of the salt peak. But you could still be able to see the DNA
on the gel.
By the way, I would like to take this chance give a little criticism to the
traditional 260~280 two-point reading technique for quatitation of DNA.
Although it is kind of a criterion for how complete the protein contaminants
have been removed from the DNA preparation (but bear in mind that most proteins
have much lower extinction coefficients at 280nm (A=1~2 at 1ug/uL) than DNA at
260nm (A=20 for 1ug/uL), which means this ratio would not be very sensitive to
protein contaminants), I regard it as an extremely poor technique for
quantitating diluted DNA samples. Since we are now only taking two points, we
have no idea if the two points are really representing a DNA peak, or just two
numbers taken from the huge salt shoulder that often appear in gel-extraction
samples. You may still get a good 260:280=1.8 reading, but what you are
actually measuring might be the salt shoulder or 80% coming from the salt. In
the early days, UV specs used to have no computer connection, but an electric
gauge to give the single point reading, that is why people had to only take a
few points - otherwise to assay a plasmid could take a day. There is really no
reason to continue doing that these days with computer-controlled UV specs
capable of producing UV spectrums at 1nm wavelenth resolution within 5 seconds.
To convince myself, I have experimentally proved that the "scan" and the
"DNA/RNA" program(which does the 2-point) on our lab's Cary50 instrument
produce same readings at 260 and 280nm, when properly zeroed (or "baseline"d
for the scan program). But with scan you can see the peak and you can substract
the salt shoulder. With DNA/RNA, you can fool yourself to be happy with a very
poor DNA preparation and use an inflated concentration of that sample.
OK, where am I... As to the cases of getting self-ligated vectors, our lab do
the following to deal with it:
1. As Raji had already mentioned, use a phosphatase to dephosphorylate the
vector. I prefer the Antarctic phosphatase than CIP, for it has lower activity
and can be heat-inactivated (and also require Zn2+ cofactor, which is normally
not present in ligation mixture - safer). Residual phosphatase activity in you
ligation mixture could cause failure of ligation, and CIP is a really strong
and robust enzyme. (You might think gel-purification should have removed all
the CIP, but, who knows where the CIP runs on an agarose gel? It is just like a
native protein eletrophresis. The CIP band can be overlapping your vector band
- just a bad possibility.)
Whenever I can be sure about the completeness of the double-digestion of the
vector fragments, I would avoid using phosphatases, because they could also
cause you to have zero colonies on the plate, which is even more distressing.
But it is a good method when it works properly, and definitely is among the
first things to try when you have self-ligation problems..
2. If we already have a construct with the same vector and has the same
restriction sites, and the insert is relatively long(~1kb or more), I would use
that as the "vector" plasmid for making the vector fragments. When the double
digest is working, you can easily see the released insert on the gel when you
are doing gel-purification. And you should also be albe to see the vector
fragment running at its correct position instead of being 1kb higher. Now the
vector must be completely double-digested. If you are not expecting exonuclease
contamination or star-activity, then you can use it with confidence.
Also, if you gel-purify the popped-out insert of the previous construct, it
would be a very good positive control for your preparation of inserts, since
you know this fragment must have good ends. If you can find some early
preparations of insert or vector fragments with compatiible ends with your
current cloning, you can use them as your positive controls to figure our
whether it is your vector or your insert that has got problem, too.
3. We always setup a no-insert control ligation reaction, same as the cloning
one, but replacing the volume of insert with water. When you see this negative
control plate grow like a star night, well, cut the vectors longer time, add
more enzyme, or strictly follow the enzyme manufacturer's manual.
In theory, ratio of colony numbers on the negative control plate and the
cloning plate should reflect how many colonies on the cloning plate is false
positive. So people are advised to try to achieve a good ratio on the two
plates before going to the screening. But in my experience, even sometimes the
ratio does not look so great, or even the cloning plate has less colonies than
the negative control plate, as long as some difference can be observed, that
suggests that you insert has done something to the vector: so while repeating a
new round of cloning, go ahead to screen some colonies at the same time, you
might be lucky!
4. Yes, as you did, varying the ligation ration. However, varying the ligation
ration within such narrow range as 1:3 and 1:2, is probably not a great idea.
Such small variation would not change the situation too much. And the covered
range is so small, that potential (actually ussally quite significant) DNA
quatitation errors would not be corrected with such small coverage. You might
have better to give it a greater variation like 1:3, 3:1, 1:1, or even 1:10,
1:1, 10:1. Especially, for short DNA inserts (like 100bp or 200bp or smaller),
inserts in excess often is not a big problem, and most of the colonies should
be positive. The tolerance of ligation ratio is actually pretty broad. For
small fragments (tens of bps), I used to try 1000:1 (I was managing to insert
multiple copies), and it still worked: with only one copy of the insert
inserted.
Zhijie
----- Original Message -----
From: vijay srivastava
To: CCP4BB@JISCMAIL.AC.UK
Sent: Monday, September 01, 2008 3:05 AM
Subject: [ccp4bb] regarding cloning
Hi,
I am trying to clone a 1.2kb insert into a expression vector pET 23a
through T/A cloning. The restriction enzyme used is Nhe1(NEB) and BamH1 (NEB)
in the forward and reverse primer recpectively. I was succesful in subcloning
(T/A vector) and getting my insert at 1.2kb after double digestion and also
the vector at 3.7kb ,for the ligation i am using the ratio of vector to insert
is 1:3,1:2,getting the colony after the transformation but some how when i
used to confirm my clone through double digestion i am not getting my insert at
the correct position.Some time in the gel only the size of the vector was
there.
------------------------------------------------------------------------------
Be the first one to try the new Messenger 9 Beta! Click here.
