Protocol Online logo
Top : Forum Archives: : Molecular Biology

which method is more efficient for oligo end-labeling? - (Mar/01/2006 )

As so far i search, there are two methods available to end label my DNA probe (24nts) for gel-shift assay.

1. using sequenase to label 5' protruding ds oligos
2. using T4 polynucleotide kinase to exchange the 5'-phosphate of ds oligos

Since i don't have much experience on this, does anyone tell me which method is more efficient and popularly used now. Thx.



I have never used sequenase, but I use T4 polynucleotide kinase on a daily basis. It's a doddle and very quick.

Simple oligo labelling protocol:

Take 5 pmol of ssDNA oligo (1 µL at 5 µM)
Add 1 µL of 10x T4 Kinase Buffer
Add 5 µL water
Add 1 µL T4 kinase
Add 2 µL (32P-gamma-dATP) at 10 µCi/µL

Incubate for 30-60 minutes at 37 degrees Celsius

Following incubation, heat the mixture to 70 degrees Celsius for 10 minutes to denature the enzyme.

Subsequently, add 5 pmol of complementary oligo and 89 µL of annealing buffer (varies in lab but is generally about 10 mM Tris-HCl pH 7.4; 50 mM NaCl; 0.1 mM EDTA pH 8.0)

Heat the solution to 94 degrees Celsius in a heat block or preferably a PCR machine if you are using small tubes. Now either turn of the heat block and let the solution cool naturally or set the PCR machine to cool to 4 deg C over 45-90 minutes at a rate of 1 or 2 deg per min.

You will get nearly 100% annealing this way.
Unfortunately you still need to separate the dsDNA oligo from the unincorporated nucleotide. To do this, run a 6% non-denaturing polyacrylamide gel (ratio 37.5:1) until the bromophenol blue dye is two-thirds down the gel. The 24-mer should be just below it and the dATP will either run out or more likely form a horrible dark smear at the bottom of the gel when you visualise it.

From there, just visualise, excise and crush the right band. Elute into Tris or TN and clean by EtOH precip'n.

Child's Play


for purification, the Qiaquick nucleotide removal kit from qiagen before annealing will prefectly do the job


Unless tings have got a lot better recently I've found that nucleotide removal kits are totally inadequate for separating radiolabelled nucleotide from short oligos like the 24-mer in question. Most of these kits only efficiently retain molecules of at least 40 bp.

Whilst the QIAquick Nucleotide Removal Kit (I'm reading it right now) states that it will recover 17-40 bp oligos, I have found that not only is the % recovery v.low below 25 bp but that the resultant oligo is often 'dirty' and does not produce nice bands when run in EMSA.

Whilst this method is v.quick, in my opinion PAGE produces a more superior probe in the end. Depends how nice you want your pictures to look.


I still have three questions regarding the radio-labeling of dsDNA using T4 PNK method.

First, is it okay for me to use 32P-alpha ATP with similar radio-activity but not 32-gamma ATP in labeling? Is the only difference the exposure time when developing films between these two??

Second, can i anneal two ssDNA to dsDNA first followed by radio-labeling instead of the reverse order? Is there any difference in the labeling efficiency between these two??

Third, should i dilute T4 polynucleotide kinase prior to the reaction and just use 2-5 units in each reaction since as the instruction mentioned that concentrated enzyme can result in lower levels of incorporation?? Thank you.



You must use gamma-labelled nucleotide triphosphate otherwise you will not get any incorporation. Only the gamma phosphate is transferred to the nucleic acid using T4 kinase. For alpha-labelled ATP you need to use reverse transcriptase to fill in the sticky ends of a restriction fragment, then the whole ATP molecule is incorporated. Hence, alpha-labelled ATP is not appropriate for your purposes.

To answer your second question, yes and no, you could anneal and then label, but it would be difficult to get the labelling conditions correct as the annealing procedure uses a larger volume and different buffer conditions to the labelling. You would need to pellet and resuspend your dsDNA in a smaller volume to acheive this. Otherwise you would end up using a lot of enzyme and it might have no activity anyway.

As for dilution of T4 kinase. I never bothered. I simply bought a vial from Promega and added 1 µL (I think that's 5 units) to a 10 µL reaction.


Thanks, but if i want to do the fill-in reaction, which enzyme usually does the best job? Is the commonly-used Taq working?