tricky digestion for cloning gurus - (Jan/24/2006 )
hey folks. I have been struggling with a cloning project lately, and I am quite certain that the problem is because the two RE's I'm using in the MCS are very close together, in fact there is one base that overlaps the two sites.
I am using BamHI / XhoI, following a published protocol so I know it can work. here is the vector map for pET-14b: http://www.emdbiosciences.com/docs/NDIS/TB044-000.pdf
I have tried this 4 times now and I am not getting anything that is not re-ligated vector.
well, I have run gels at all conceivable steps and the last time I ran through the protocol I called NEB to get their idea of the best reaction conditions to favor cutting by both enzymes, and I did everything they said, and I still got a bunch of vector on my plates. I know both enzymes are good and I know that my vector is linearizing well from the bands that I see on the gel; this is why I suspect that the problem is due to the proximity of the sites.
Does anyone have any ideas to help me figure out a good way to get both enzymes to digest properly? has anyone experienced such a thing?
Thanks to anyone who can help!
Sorry to hear about your troubles. You say it has been documented that this has worked before? I didn't think overlapping sites would work together since many enzymes need the site to be:
1) double stranded, and
2) have at least 2 bases present after the site in order to cleave.
If the sites are lined up as is stated in the manual it would look like this:
when one enzyme cuts (we'll say BamHI) it would yield:
----G + GATCCTCGAG----
----CCTAG + GAGCTC----
OK I would suggest one of two things, 1 use NdeI.
Or two, try using BstKT I instead of BamH I. It uses the 4 base sequence internal to the BamH I sequence (GATC) and would leave a single base prior to the Xho I site. Unfortunately you will probably find other BstKT I sites since it is only a 4 base cutter .
If you do find more, try Taqα I or TthHB8 I instead of Xho I. Also internal 4 base cutters, so check if there are other TCGA sites in your vector.
Dang, I looked through the sequence and those 4 base cutters are indeed everywhere. Looks like Nde will be your only choice .
I have followed the published strategy to get an in-frame gene for expression. I went through the isoschizomer chart with both vector and insert sequence and was unable to find one that didn't whack my DNA into 60 pieces
I have tried to email the author of the paper, but their address bounces back; I have a message into their institution and I'm hoping I can get some new contact info.
I think I will have to go back to the drawing board and design a separate primer for one end, so I can get the insert in-frame with another site. dammit
I have tried a couple different sequential digestions; some of the reason this has been so tricky is due to the BamHI...a wonderful cheap enzyme with star activity that can't be heat-inactivated. aarrgghhhh
OH well. I have about given up on the published method.
Oh, hey, I have nothing against Nde, cept it cuts my insert twice.
I'm wondering how the hell the authors got it to work????
btw, thanks for your help, Captain DNA!
Can you send a link to their paper? Maybe we can find a clue there?
I am cloning the LukF-PV into pET-14b, same S aureus strain for the insert, same vector, primers as published. The only thing I am doing differently, I am using JM109 instead of DH5a for the initial cloning, but that shouldn't matter in the slightest for what I am doing
The details are very sketchy in the paper. I have done all the steps in the routine fashion...gels at various points to make sure the DNA is there and that sort of thing...qiaquick clean-up columns (although the first time I used gel extraction through glass wool; I went to the columns trying to beef up my insert:vector ratio and get a little less sample-loss)...ligation o/n at 16C, the sorts of routine cloning steps that always work. I also cloned S and got that one to work on the first try, but I used a different vector
I am wondering if someone screwed up in the methods section and it was a different pET vector after all?
That is a tough one... I've used Nde I and Bam HI to open pET vectors, but not Xho I and Bam HI.
I suppose you could cut the vector with Bam HI, and clone some irrelevant piece of Bam HI digested DNA into it that has no internal Bam HI or Xho I sites. Then cut those plasmids with Xho I, then with Bam HI, and recover the appropriately sized vector fragment that way.
after much sober reflection, and banging my head repeatedly against the wall, I have decided to go with the BlpI site a little farther down. I will order a new 3' oligo and start over.
I suspect it will be faster in the long run
thank you guys for your help!