insert:vector ratio for ligation - religation of the vector? (Jun/06/2007 )

Hi,
I've tried the same thing about 7 times now, with variations, and I seem to just do it wrong.

I've got a 700 bp insert, PCR-product with Xho sites on both ends, that is supposed to replace an insert in a vector which also has Xho sites. I digest over night, dephosphorylate the vector for 1 hour and purify both, the new insert and the digested vector on a gel.

For ligation I used ratios of 1:5 - 1:50 (vector:insert) of the DNA solution from the gel extraction (I haven't determined the concentration, though), and ligate over night at 16 °C with T4 ligase from usb.
I transform into DH5alpha bacteria and plate on Ampicillin plates.

The next day, there are usually 10-20 colonies of religated vector (where I didnt put any insert into the ligation mix), and the same number on the other plates.

Sometimes, there are more insert-colonies than religation colonies, and I do a colony-PCR, only to find that they contain the insert that was originally supposed to be cut OUT of the vector by XhoI, not the insert I wanted them to have.

If anyone understood this explanation :-D, I'd be happy for suggestions for modifications.

More specifically:
What exactly is the optimal ratio of insert:vector:T4 ligase?
Why can the vector religate when it's been dephosphorylated? (I did see improvement, when I bought a new batch of SAP)
How can the cut-out insert get back into the vector, when I excised the vector band from a gel? (The gel has run for 2 hours and it was REALLY well separated from the undigested vector, and I saw the insert band below the vector)
Why do I lose something like 50% of DNA through gel extraction and XhoI digest over night?

Many thanks,

Moly

For single digested ligations, I would use 1:5 or even 1:10 ratio for ligations. I would verify the conc for both the insert and vector in a gel.

you will lose the some DNA thru gel purification and also you can to account for the insert which is being digested out. For eg. if the vector is 3000 bps and your insert is 600 bps. If you digest 1ug of vector, you lose 200ng as insert and you are purifying only 800ng and you will lose some thru the columns.

Alkaline phosphatase is atleast reducing your negative control colonies. I would try antarctic AP, I find it better. If you do have more colonies in the insert-vector plate, then I would check all of them for the insert. I would try to do miniprep and digest them with which might reveal the actual orientation and insert.

Digest the insert-PCR for 1 hr and it should be sufficient for complete digestion. Overnight digestion could damage the ends and could reduce the number of ligated colonies.

Good Luck !!!

my money is on the dephosphorylation step. 1hr is very long by my book, over dephosphorylation will damage your DNA ends. How many unit of CIP are you using, and how much vector (in pmol) are you dephosphorylating?

The rule of the tumb is
1pmol DNA with 0.1 Unit CIP, in 50ul volume for 1 hr at 37 Celsius.

If my suspecion is correct, treating your dephos vector with PNK (polynucleotide kinase) and ligating it, will not increase the number of colonies as all the vector's ends have been damaged.

Try recalibrating the dephosphorylation times.