to BioTechniques free email alert service to receive content updates.
Colony PCR
Kristie Nybo, Ph.D.
BioTechniques, Vol. 53, No. 6, December 2012, pp. 345–347
Full Text (PDF)

This month's question from the Molecular Biology Forums (online at 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 do not represent endorsements by, or the opinions of, BioTechniques.

Molecular Biology Techniques Q&A

What can cause colony PCR to fail? (Thread 32239)

Q I am attempting colony PCR using a reaction mix and PCR program that have worked well in the past, but this time I saw no bands in the gel for either the wild type or mutant strain. Repeating the experiment also produced no results. What could cause this?

I used 150 μl water, 8 μl of foward and reverse primers (1:10 dilution), 1 μl of platinum Taq polymerase, 20 μl buffer, 6 μl MgCl2, and 2 μl dNTPs (20 mmol/l) in my reaction mix. I picked the colonies and added them to 10 μl Milli-Q purified water and used 1 μl of this in 9 μl of the reaction mix. The PCR program included a 12 min cell lysis phase at 94°C, as well as a 3 min, 30s extension period at 68°C. I was trying to amplify about 3.5 kb.

A Your mix adds up to 187 μl. If the buffer stock is 10×, you should add that to a volume of 180 μl. Currently, your buffer is a little too concentrated. If the annealing temperature you use is near the Tm for one of the primers, the change in salt concentration could alter the Tm such that the primer won't be able to anneal. I looked up the Platinum Taq protocol and it looks like your concentration of enzyme is correct. For the primers, you will need to determine the best concentration empirically. The Mg concentration looks okay if you are using a 50 mM stock as listed in the protocol. The dNTP concentration also looks fine.

For colony PCR, a colony control is always a good idea. If you have your desired PCR product cloned in a different vector, you might try using colonies carrying that construct. 3.5 kb is a big product to amplify using standard Taq, especially when your mix is contaminated with bacteria. I'm surprised that this worked for you previously. Ideally, you should change your primers to amplify a smaller product.

A I agree, a PCR product of 3.5 kb is too big for Taq. If your primers are based on the vector sequence, you can redesign the primers based on the insert sequence, starting around 500 bp from either end. Then you can use one primer based on the vector sequence and the other based on the insert sequence to screen for clones carrying the insert in the correct orientation. Another reason you might not see bands using colony PCR is if none of the colonies you picked contain the insert. This is usually caused by incomplete digestion or dephosphorylation of your vector.

Q2 I am also doing colony PCR and not seeing the correct bands in my gel. Even the positive control lane doesn't look right.

I am using 66 μl HPLC water, 10 μl 10× buffer (15 mM MgCl2), 5 μl MgCl2 (25 mM), 2 μl dNTPs, 5 μl forward primer (100 pmole/μl), 5 μl reverse primer, and 2 μl Taq polymerase. I inoculate one loop of bacterial cells into 100 μl of TE and boil it. Then, I spin down the tube and remove the supernatant.

After PCR and electrophoresis, I see multiple bands of nonspecific sizes. What could be causing this?

A I add my bacteria directly into the PCR reaction, so in my hands, the colony PCR is often messy. Sometimes the bacteria inhibit the PCR or cause a smear of nonspecific products to show up in the gel. Amplifying a small PCR product works best for me; this usually requires a primer that anneals within the insert.

A Have you done PCR with these primers before? What is the expected amplicon size? I usually pick up some bacteria from a colony using a filtered pipet tip and resuspend directly in the PCR reaction mix by pipetting up and down. I also occasionally use 1 μl of an overnight culture. Since you resuspend a loopful of bacteria in 10 μl TE but only use 4 μl of that suspension, there may not be enough bacteria for the reaction. When you add the cells, do you see the PCR mix become cloudy?

Q My insert is 1.8 kb. I used 4 μl of the removed supernatant, so checking the cloudiness of the PCR mixture won't give me an indication of the template quantity. I just don't understand why I see multiple bands on the gel that are not running at the correct size.

A If you put a loopful of bacteria in 100 μl of TE, centrifuge, and use 4 μl of the supernatant for PCR, then you have no template. That explains why you don't see a band for PCR. The plasmid is in the bacteria, which you spun down and discarded. You should suspend the bacteria in 100 μl of TE and take 4 μl of the suspension for PCR.

Q After adding the bacteria to the TE buffer, I placed the tube in a boiling water bath for 10 min, and then the centrifuge for 3 min at 8000 rpm.

A After boiling, the DNA should be released into the TE. Still, it is possible that you don't have enough DNA template in the 4 μl of TE that you use. Try using a filtered pipet tip to pick up half of a colony and suspend the bacteria directly in the PCR reagent mix. Pipet up and down and stir the solution to dislodge hanging bacteria. The solution should be cloudy, indicating that the bacteria are in the reagent mix. At the initial denaturation step of the PCR program, the bacteria will burst open and release DNA.

A You might try this alternate protocol based on SDS-NaOH lysis of the colony bacteria. Choose a colony less than or equal to 1 mm in diameter and use a 0.8% agarose gel. Prepare a 2× lysis solution of 0.1 mM NaOH, 10 mM EDTA, 1% SDS, 1× loading buffer, and 10% glycerol. Collect colonies on a p10 tip, placing as much bacteria as possible at the bottom of a 1.5 ml Eppendorf tube. (The tip could be added to 5 ml selective medium if you would like to use it for miniprep after an overnight culture at 37°C.) Add 10 μl of water to the Eppendorf tubes, vortex, centrifuge briefly, and add 10 μl of 2× lysis buffer. Let this sit for 2-5 min at room temperature. Then, load into agarose gel wells not covered by TAE buffer. (The TAE buffer must be at half the height of the agarose gel to ensure electrical contacts without letting the wells leak.) Run the gel for 10 min and then cover the gel with TAE and continue the electrophoresis. Good colonies containing recombinant plasmids will show a relatively higher molecular weight band than non recombinant plasmids. Just remember that since your DNA was denatured, plasmid size might differ from the expected size. Here, it is the shift in size that you are looking for.

A It looks like you might be using too much primer. You reported using 5 μl of a 100 μM stock in a 100 μl reaction, so that comes to 5 μM of each primer. The normal range of primer concentrations for PCR is 0.2-1.0 μM, with 0.5 μM being used most commonly. Too much primer can cause nonspecific bands by recruiting reagents away from the correct product. DNA primers also bind magnesium, lowering the effective Mg concentration, which could also be a problem since you are using a minimal amount of Mg.

Q Can I use m13_puc primers to check my inserts ligated in pk18mobsacB vectors? I cloned the insert into the PstI site of the vector.

A Be aware that there are several primers for M13, so you really need to check whether the primers are able to bind to your insert or not. Previously, I found multiple bands using M13 primers, but was then able to get a single band when I changed one of the primers to the T7 promoter.