I'm using transformation-associated recombination (TAR) in yeast to capture a biosynthetic gene cluster (32 kb) by transforming gDNA and a capture vector (11 kb) with homology arms. I identified a region-positive clone and moved the plasmid into E. coli. I did a midiprep and ran the plasmid extract on a DNA gel with the empty capture vector (no insert, 11 kb).

By size comparison, the midiprep extract appears to contain the empty capture vector, and there are some more faint bands near the top of the ladder at 15 and 20 kb. However, when I run a PCR on the plasmid DNA to screen for a 500 bp region in the middle of the biosynthetic operon, I get a clear band, which suggests I also have my plasmid of interest, since that DNA does not exist in E coli. I made sure the primers were orthogonal to the E. coli genome (a common source of contamination in plasmid extracts).

I guess where I'm stuck is how would I separate these seemingly cotransformed plasmids out from the E coli? I was thinking of doing a gel extraction of the higher MW band on the gel, but our kit caps the size limit at 23 kb, and my plasmid of interest is 44 kb... Any suggestions or advice would be appreciated.

I'm also thinking of transforming a very small amount of my midiprep extract and hoping that single colonies will have only picked up on plasmid. The issue is that they both have the exact same backbone, and I presume the smaller plasmid will replicate much faster than the plasmid of interest, but I'm not certain.

  • $\begingroup$ Do you have any selection marker available for your engineered plasmid? $\endgroup$
    – Chris
    Commented Apr 25 at 18:30
  • $\begingroup$ The empty capture vector has both SpecR and AmpR, which also means the engineered plasmid has the exact same selection markers. The only difference between the capture vector and the recombinant plasmid is the gDNA insertion, which is just an operon, not other things are inserted. $\endgroup$ Commented Apr 25 at 18:32
  • $\begingroup$ I'd recommend trying for the single-colony screening. It should be pretty fast: transform into cells and plate today, colony PCR tomorrow. I think the fitness cost of the big plasmid shouldn't be too high to show up on a plate. Give it a shot and let us know how it goes! $\endgroup$ Commented Apr 25 at 21:04

1 Answer 1


There's a couple of options here - it could be that your plasmid extract isn't very pure, so some genomic DNA has come over and the ROI is in that. The PCR would detect this even if you don't have any captured DNA in the plasmid giving you a false positive.

You are also using two very large bits of DNA there. 32 Kb is a massive fragment to try to clone into a plasmid, most people would try to take it out in chunks of a few Kb at a time and then reconstruct. In my experience the general rule is that the insert should be about the same size or smaller than the plasmid. It's not a hard and fast rule, you can definitely do it the other way around, but it tends to be more challenging for some reason. A YAC (yeast artificial chromosome) or similar might be a better option to capture the full length.

A classic gel extraction should work. This is not a kit extraction, but rather done by eluting the DNA in the gel into solution and then performing precipitation on the DNA. However, for this to work you need quite a lot of DNA as the efficiency of the extraction is very low; usually much less than 10%, more in the 1-2% range.

Basically the protocol is to get the liquid containing DNA out of the gel chunk. I used to do this (>20 years ago) with a 0.5 ml tube - poke a small hole through the bottom with a heated paperclip and add your gel chunk. Place 0.5 ml tube with gel inside a 1.5 ml eppendorf tube and spin the hell out of it in a higher speed microcentrifuge - you need one that goes above about 20K RCF. You may need to place a small piece of filter paper in the hole to prevent the gel getting squeezed through if it is a low percentage gel. Collect any liquid that ends up in the bottom of the 1.5 ml tube and precipitate. There are also less DIY protocols at Current Protocols in Molecular Biology, I can vouch for the NA-45 paper one as working very well, but these still require some refinement of technique and things you are unlikely to have.

You might be better off just amplifying your whole fragment of interest from your extraction and then re-inserting it into a plasmid of your choice. This allows you to choose which plasmid to put it into and gives you some control over the method of insertion. This is easily achievable with a high fidelity polymerase such as Phusion, Q5, or LA polymerases.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .