Subcloning Problem - (Jun/09/2011 )

Hey guys, I've been unsuccessful for the past couple of weeks at inserting a 700kb insert into a 5.4kbp vector. As far as I know, this should be a relatively simple process, but I just can't get it to work. Here's how I've been doing it:

The insert was PCR amplified with primers containing EcoRI and XhoI restriction sites. The correct PCR product was detected and gel purified. Then I took that and digested at 37C for 1 hour with 0.5uL of EcoRI and XhoI.
The vector already had an insert with it and digestion with EcoRI and XhoI (37C, 1 hour) yielded 2 bands corresponding to the vector and insert. This was then gel purified to isolate the digested and linearized vector.

The digested insert and vector are then ligated at room temperature for 30 minutes. Vector amount is about 50ng. I've tried vector/insert ratios of 1:2, 1:3 and 1:6. 1uL of T4 ligase is used in a total reaction volume of 20uL.

5uL of the ligation mix is then mixed with 40-50uL of competent cells on ice. The cells are kept on ice for 30 min. Cells are heat shocked at 42C for 1min and then put back on ice for 2-3min. 400uL of LB is then added and the mix is incubated for 1hour at 37C. This is then centrifuged to remove ~350uL of LB, resuspended, and finally plated onto amp plates. I have yet to get a single positive colony.

I have transformed an intact vector to verify competency. I have ligated a different vector and insert (predigested and given to me by supervisor) to verify the ligase works. The fact that the original vector with different insert was successfully cut means the EcoRI and XhoI are working. Gels of the insert and vector prior to ligation show only the band of DNA with correct size.

I'm beginning to suspect that the insert was not properly digested. I have 2 extra bases flanking the restriction sites, is this too little? Any suggestions would be helpful!

You can verify correct digestion of your insert by cutting it with only one enzyme at a time, then ligating it to itself. This should result in double length insert size fragments. You need to heat kill the ligase (and avoid quick ligase) to make this work well.

You need to test your competent cells with 10 pg/ul plasmid, transforming only 1 ul of this. Unless you get many colonies, your competent cells are not sufficiently competent.

See this table for overhang lengths required.
http://www.neb.com/nebecomm/tech_reference/restriction_enzymes/cleavage_linearized_vector.asp

In my hands, and in the hands of almost everybody I have worked with, this approach very rarely works. For some reason, restriction enzymes cut the ends of PCR products very poorly. Not sure why, maybe something to do with 'breathing' ( partial denaturation) of the ends of the PCR product at the required temp of 37. We do this sort of ligation all the time and never even bother to try to ligated the PCR product in directly. For us, the most reliable way is to first clone the PCR product into a PCR cloning vector (we use PCR blunt, but there are plenty of others) and then digest it out. This gives you the added reassurance that the RE sites that you have designed into your oligos, do, indeed work. Donto forget to gel purify your linearised vector; absolutely essential!!

Good luck

I agree totally with the above, though i have in the past digested PCR products, the reduction in fragment length is so small that it is impossible to know if the digest has been succesful... my weapon of chice is pGEM -T easy vector.. fast simple super reliable and there is no ambiguity over the restriction products ends..

The other option you have is to blunt end ligate in then sequence your recombinants for directionality