Puzzling Cloning Problem With a Specific Vector (View forum version)


Posted 12 July 2013 - 11:52 AM

Hello, over the last couple of months I have been attempting to generate artificial aggrobacterial vectors containing DNA inserts that correspond to the tripartite genome of the Cowpea Chlorotic Mosaic Virus. These vectors will be used to generate mutant virions in-Plantae. I am using a vector provided to our lab by a collborator. I must use this vector for all cloning experiments. It's a large vector ( about 9000 bp) and the insert is large as well (about 3000bp). So far I have had no luck in doing so.

Here is an overview of my protocol.

I begin by PCR amplifying the insert. The primers used have been engineered to incorporate a STU1 restriction site on one end, and a BAMH1 site on the other. In addition, the primers contain a 4 nucleotide overhang on both sides of the insert to allow for restriction enzyme binding. The PCR reaction is gel purified using a Qiagen kit, and then double digested with both restriction enzymes for 2 hours at 37c. The resulting digested insert is then gel purified, using EtBr as the stain with high wavelength UV. Afterwords I have tried both treating the insert with kinase, and not, and neither subsequent ligation has worked. As a control to ensure that the RE’s were cutting both ends of my insert, I performed a ligation with simply insert and T4 ligase. This resulted in a ladder of insert oligomers, which indicates that the inserts have been cut on both ends.
The vector is double digested as well, using the same RE’s for 2 hours at 37c. I was concerned about the proximity of the two restriction sites (they’re 18 nucleotides apart) so I have also tried two single digestions with a Qiagen column purification step between the two digests. The resulting double-cut vector is then gel-purified, treated with CIP for 1hr at 37c, and column purified to remove the CIP.
The ligation reactions are set up as follows. Using a 3:1 ( I’ve also tried 2:1 and 5:1) insert:vector ratio reactions are set up in a 20uL reaction with a total DNA concentration of about 100ng ( 50ng of insert and 50ng of vector, which, based on the length of both species, is a 3:1 mole ratio). Also included is a negative control, which consists of just digested, CIP treated vector. The ligation reactions are done at 16c for 12 hours. After liagtion, Top-Ten cells are transformed with 2uL of ligation, and transformed with heat-shock for 30 seconds at 42c. They recover for 1 hr at 37c in SOC media, and are plated on pre-warmed plates containing Kanamycin.
After 12-16 hours, this is what I see. The negative control plate contains no colonies, while all plates containing insert have many colonies ( 50-300). Upon screening using a double digest, I find that ALL colonies screen contain ONLY vector, with no insert whatsoever. This is very perplexing, because this seems to indicate an issue CIP treating or digesting the vector, yet the negative control contains no colonies. I am currently at my wits end with this, as I’ve chased down every issue I can think of with the appropriate controls. I know my RE’s work, as I’ve done a control digestion on lambda DNA, and I know they cut my vector as well, as I have done the appropriate control there as well. If you have any thoughts on the matter, I’d love to hear them.

Thank you.


Posted 12 July 2013 - 12:27 PM

The first thing I would try is to do this reaction without CIP -- understanding that there would be high vector only background. You might try gel purifying the cut vector to try to eliminate the short fragment. Second, I would switch from CIP to a heat killable dephosphorylating enzyme, such as SAP or Antarctic Phosphatase.

Colony PCR is probably a higher throughput lower overhead way of detecting transformants. Use one primer on the insert, another on the vector.

The way I would really try to do this would be to PCR the agrobacterium vector with whatever 5' RE sites I wanted, purify, DpnI digest to remove template, and simultaneously cut with the REs. I would avoid BamHI which can't be heat killed. I would definitely rule out StuI which is a blunt cutter, and will make your ligations unselective and difficult. Heat kill enzymes, mix, ligate, go.