Hi,
This is a brief description of my latest head scratcher:
I am digesting a vector with XhoI and HindIII-HF (both from NEB), then I dephosphorylate the digested vector with the Antarctic dephosphatase. Finally, I purify the digested and dephosphorylated vector by gel extraction.
I ran a negative control in which I did a ligation for the vector only (using the Quick Ligation kit by NEB) followed by transformation of DH5a and the following day I was shocked by finding more than 40 colonies in my vector only plate
I am trying to find an explanation for this and a solution too. I did notice that the restriction enzymes don't have a 100% efficiency, even doubling their concentration and running the digestion for twice the time: indeed, if I do not gel purify the digestion, but I run, instead, a PCR clean-up, transformation of this reaction (after ligation WITHOUT LIGASE) leads to a full carpet of transformed bacteria on the Petri dish.
Since I remove the uncut vector by gel purification, I suspect there my purified vector contains some molecules of single cut vector. The only explanation I can come up with is that the dephosphorylation failed at removing the P from all 5', leading therefore to a scenario in which the the few molecules of vector digested by a single RE and still possessing a phosphate group can anneal and self ligate.
What do you think about this possibility? I haven't tried to transform bacteria with the gel purified digested vector without ligase to ensure that no uncut vector is left, but that seems an unlikely scenario.
I haven't tried a ligation with dephosphorylated vector and phosphorylated insert yet.
Should I prolong the incubation with the Antarctic dephosphatase to increase the chances that molecules have been deprived of their 5' P? Am I missing something?
Hello Matteo,
Why don't you try to perform the actual cloning, i.e. vector plus insert ? After transforming, patch 10-20 colonies and perform colony PCR on them to check for the presence of the insert. This is really quick and all you need is just one clone carrying the desired insert. If this fails, you can also digest the ligation mixture prior to transformation with a restriction enzyme that lies between XhoI and HindII in the multiple clolning site of your vector, but is of course absent in your insert (if available)
Hi Pierre,
Thanks for your advice.
Indeed I am going to try the actual cloning (vector and insert): the abundance of insert should help reduce the likelihood of transformation by self-ligated vectors.
Your suggestion of digesting the ligation with a third restriction enzyme is brilliant, I'll keep that in mind if my attempts fail!
Have a great weekend
I agree with Pierre Béguin but would add that it is really difficult to reduce the background to zero. If you take a plasmid prep and digest it with an enzyme, even if your digestion was 99.999% efficient the remaining 0.0001 would still give you lots of colonies. There also might be a really tiny fraction of plasmid in your prep that underwent some deletion event or mutation event such that the restriction site is no longer present. So there are all sorts of reasons why the background is almost never nil. But if you have a good ligation then the number of colonies should be well above background.
The 3rd enzyme trick is really really useful, my lab does it all the time when it is possible.
- Badges
- Science topic
- Similar topics
- Vectorization