Protocol Online logo
Top : Forum Archives: : Protein and Proteomics

Failure to show Super Shift in EMSA Experiments - (Jul/06/2006 )

Hiiii Everybody,

I am currently doing mobility shift assays to determine if my protein of interest binds to a sequence on the 5'-end of my gene. I have not been successful till now to show the super shift in my experiments although I do get shifts. I have used a poly clonal anti-sera raised against the DNA binding protein. The experiment has been shown to work invitro with purified proteins but the challenge is to show that this is the case invivo as well.

I am working for my master thesis so I havent got much time. I would really appreciate if someone can help me in getting over this problem. And is it acceptable to show shifts using EMSA and then confirm the identity of the protein using a western blot by transferring the complex to a pvdf/nitro cellulose membrane for a publication?

Also I have some very silly questions: I use Igepal (detergent ) in my gel running buffer and also in my acrylamide gel. Why is this if EMSA is a native page technique?

How do I know when should I stop using my 32-p labelled Oligos and label them afresh and How do I know what is the specific activity of the labelled Oligos? I could only calculate % incorporation which ranged from 76% to 47%.

Thanking you in adavance,

Regards smile.gif


If you antibody works on purified protein then theoretically in should would on the in vitro samples you have from cells. If it doesn't supershift your complex in these samples it implies that the binding complex might not be the protein you think it is or is at least more complex than the purified protein. Is the protein you're interested in one member of a family of DNA-binding proteins ? In which case you could test more antibodies against other members of the family or test more antibodies with different epitopes against the protein you orginally thought it was.

Other things you could try would be to mutate the consensus binding site or try competition experiments with 10-fold excess of specific/non-specific cold competitor probes to look for specificity of binding.

All the best,


what sort of different parameters have you tried for the binding reaction of the supershift? like, adding Ab first or last, pre-incubating different components, incubations at different times/temperatures...


oh, hey, and my understanding is that igepal doesn't denature like SDS, it's very mild and just helps keep things soluble...? perhaps that's not right?



Thanks for your replies. Indeed the DNA binding protein of interest is one of a family of proteins. However in Retrovirus Transformed Cells it is expressed as a fusion protein with Gag and I couldnot observe any super shift by incubating with the Gag Ab as well. I am also running the competition reaction with 100 times more of cold probe and the interaction seems to be quite specific!!!!! As you have suggested I am going to try out with a different antibody against its binding partner.

As far as trying different parameters are concerned I have tried out incubating the reaction containing the hot oligo with antibody for 30 mins @ RT. This however didnot work ( shift but no super shift). Then I tried incubating the reaction without antibody @ 4 degree C for 20 mins and then added the antibody with a further incubation of 20 mins @ 4 degree C. Then I ran my gel (6%) @ 200 V for 1 hr 20 mins @ 4 degree C. However this experiment failed utterly!!!! I couldn't see anything on the gel. Can anyone suggest changes to my DNA Binding Buffer Composition???

The Recipe I use is as follows:

1 M HEPES ph 8.0 (Final Con. 50 mM)
1 M MgCl2 (25 mM)
0.5 M EDTA (0.5 mM)
5 M NaCl (100 mM)
Glycerol (50%)
1 M DTT (10 mM)
0.5 M Spermidine (10 mM)
H20 to 1 ml.

Also some advice if I could still use my oligos which have been labelled with 32-p for almost 2 weeks. I understand that by 14 days the radio activity would have declined by almost 50% but my dilemma is whether I could still use the labelled oligos that I have with me for my further experiments or should I first label fresh oligos and then start my expts?????


Hi Mik,
I don't know if this will make a difference but I've always added the antibody for 15-20mins then added the labelled probe for 20mins. If the epitope is masked when the protein binds to DNA I think this could make a difference and adding the ab first might block the binding so you will see loss of the band. The labelled probe I use up to 4 weeks after labelling although your sensitivity will drop off and you'll have to expose the films for longer. For best results use freshly-labelled probe from fresh radiolabel.

All the best,


I have shifted a couple different transcription factor proteins. one shifts if you add Ab, incubate at 25 for 20 min, then add labeled oligo and incubate an additional 20 min at 25. with the other, I add Ab, leave on ice 20 min, then add oligo and incubate 10 min at 25.

I would recommend trying it several different ways and times. for example, Ab first, then oligo...vice versa too...for different times and at different temperatures. I know it's enough to make you crazy, all the combinations, very time-consuming...but supershift is the coup d'etat, you know? I struggled mightily with one (the one with the 20min/20min incubation), and found that adding BSA into the binding buffer helped stabilize the supershifted wouldn't see a supershift band without it at 50ug/mL in the binding buffer