This month's question from the Molecular Biology Forums (online at molecularbiology.forums.biotechniques.com) comes from the “DNA and General PCR Methods” section. Entries have been edited for concision and clarity. Mentions of specific products and manufacturers have been retained from the original posts, but they do not represent endorsements by, or the opinions of, BioTechniques.
Molecular Biology Techniques Q & A Will electroporating fewer cells lead to better antibiotic selection? (Thread 21958)Q I am trying to clone a GFP-kanamycin cassette into the stop codon of a gene by recombineering with the Gene Bridges kit. I made the targeting vector with 50-bp homology arms by PCR and then electroporated the cassette into DH10B bacteria containing the BAC for the gene. I selected with 15 µg/mL kanamycin for the insertion and 15 µg/mL chloramphenicol for presence of the BAC and got colonies on my kanamycin/chloramphenicol plates.
As recommended by Gene Bridges, I also ran control experiments to check if the antibiotic selection was working. For these controls, I just plated DH10B cells without antibiotic selection vectors on kanamycin plates. I got colonies on the kanamycin control plates, but not on the chloramphenicol plates. I am sure that my plating technique is correct and that my antibiotics are fresh.
Prior to electroporating the targeting vector, I grew the cells for 22–24 hours and then diluted using 30 µg preculture added into 1.4 mL LB media. Then I induced the rec proteins with arabinose and electroporated. The size of the pellets I used for electropration varied. Does the pellet size influence selection? Will changing the quantity of starting cells result in better antibiotic selection?
A Try plating cells without any electroporation at all. If this results in colonies, it indicates that your DH10B cells have some intrinsic resistance to kanamycin.
If this happens and you ordered the cells directly from a company, request that they send you some new cells free of charge. If your lab made the cells electrocompetent, redo the preparation from a fresh stock.
DH10B cells from Invitrogen are resistant to steptomycin, but should be susceptible to ampicillin, kanamycin, and tetracycline. I'm not sure about chloramphenicol.
A Not all kanamycin genes are created equal. Some will give resistance up to 25 µg/mL or even more when they are present as a single copy. Try your selection again with a higher concentration of kanamycin.
A The size of the pellet you electroporate won't affect antibiotic resistance. Instead, the competency of your cells will be affected by the growth phase they are in, which will affect the size of the pellet.
Whenever I do recombineering, I grow the cells overnight, dilute them first thing in the morning, and then grow them to an OD600 of 0.6, which takes about 3 hours. Then I induce the red genes and electroporate.
Q By overnight growth, how many hours do you mean? I am guessing it has to be more than 16 hours. Is that correct?
A I probably grow them for about 17 hours overnight, but the length of the overnight step isn't that important. That growth is just intended to make a very good culture for inoculating the 3-hour culture the next day. I haven't encountered a protocol for making competent cells by just growing them overnight.
Either way, you need to start with kanamycin-sensitive cells.
A I found that in my situation, the kanamycin gene was expressed at different levels depending on the target gene's chromosomal location. So now when I target a new gene, I plate the transformation solution with different concentrations of kanamycin.
For cell strains, I use TOP10 from Invitrogen and MG1655. These are not inherently resistant to kanamycin.
Is there a chance that you added the kanamycin while the LB agar was too hot?
Another possibility is that your 50-bp homology arms are specific to more than one region on the E. coli chromosome. That wouldn't resolve the problem of kanamycin-resistant cells, but might explain any other background and it's something else you could check.
Q I cooled my agar down to 50 degrees in a water bath for an hour and then added the antibiotics. I stirred it well, ordered new antibiotics, and did titration experiments from 15 µg/mL to 40 µg/mL with bacterial strains DH5α and DH10B.
In the titration experiments, my bacteria did not include a resistance vector and the plates contained only kanamycin. The bacteria grew extensively on 15- and 25-µg/mL kanamycin plates. The growth started to taper off at higher concentrations so that at 35 µg/mL, there was only growth of a colony or two. So I decided to use 35 µg/mL kanamycin since my recombineering plates have more than 50 colonies at this concentration.
I also found in a paper from 1999 that Francis Stewart uses a concentration of 35 µg/mL with the DH10B strain.
A What are you using as a template for the PCR? If it is plasmid DNA, then it is likely that you have plasmid contamination in your PCR fragment. A simple way to test for this is to take a few microliters of the PCR mix and transform some competent cells. Spread those on kanamycin-containing plates. If you get colonies, you have plasmid contamination. (I am assuming, of course, that the kanamycin cassette in your plasmid has a prokaryotic promoter.) This is a much more likely cause for colonies on the negative control plates than background resistance in your cells.
A If this is the case, you can treat your samples with DpnI to destroy any plasmid or template. This would also be an easy way to determine if plasmid contamination was causing the problem.
A For most common strains of E. coli, DpnI will work. But if the cells don't methylate with Dam, then a DpnI digest won't do anything.
XL1, XL2, and DH5 would be good strains to use for this test. I think that TOP10 won't work for this purpose, but you could check the genome.
