You may also try using 0.5% Tris-glycine for running. It heat much more less than TBE and allow a better resolution. In my hand some complex that dissociate in tbe are pretty stable in tris glycine. and the picture is really nicer, with sharper bands. In some of your lanes, you also have a signal just above your probes, it may reflect complex that disrupted during the migration.  I'm also agree with several people before: A bigger gel is better, you should prerun the gel (I was not putting loading buffer). Wash the wells before loading. Load at the bottom of the well in a way that your sample is not diluted in running buffer and concentrated in a very sharp band in the bottom of the well. Put your running buffer at 4°C the day before the migration....

I think your biggest problem is that there is a lot stuck in your wells and I think if you tackle that a lot of your other problems might be solved. I have had this before myself and the main problem then was the gel and the time of running. I use the following recipe for a 4% native PAGE gel to look at p53 super shifts: 5 ml 40% acrylamide, 3.3 ml 5XTBE, 50 ul Triton, 41.45 ml H2O, 500 ul APS and 50 ul TEMED. I use a gel that is about 25 cm long and let it run for about 3 hrs at 200 volt at 4 degrees. I use either 421 or DO-1 for super shift. I hope this helps!!

Hi Maoyu!

The oligos are labelled but not p53, and these are crude nuclear extracts. Could one simple yet depressing explanation be that you are not looking at p53 binding, but that some other protein is binding your oligo, causing the shift? The observation that the size doesn´t shift with p53 antibody supports this, doesn´t it? The only weird thing is that the band intensity changes. There also seems to be less probe stuck in the wells in the antibody treated lanes. Dont know why that would happen, but this could explain the increased intensity on the gel.

Best,

Vegard

Dear Maoyu,

1. Are you sure about the proper Nutlin treatment of your cells?Because what seems from the western blot picture, the ratio of p53 (nuclear: cytoplasmic:: lane 2:lane 1) is not quantitatively much distinct.This may cause less difference in comparative band intensities in your EMSA in treatment vs. control sample. 

2. My experience tells that supershifts DO change the binding band intensities i.e the your supershift may be hidden in the binding band itself due to less concentration of gel.But before increasing gel concentration, you have to make sure that your protein is not going to stick in the well.I use to reduce the amount of protein and play with the binding reaction to maximize the band intensity along with flushing the well intensely.

Hope this helps a bit....

Thank you all for your kind help!

@Patricia: Thank you so much! Would you please tell me a little bit more about how you get the P53 DO-1 antibody to work with the supershift? Like the ratio of antibody to protein, the sequence you add the ingredients (before or after the oligos), incubation time and temperature, shake/no shake and anything else. Currently I'm incubating antibody with the protein for 15 min at room temperature with gently shake before adding poly(dl.dc) and oligos. 

I thought about going with a lower percent gel, however we use precast native gels from Thermo and they don't have any thing lower than 6%, if you have any recommendations please let me know. Thanks!

@Vegard: I have exactly the same concern, I can't say it is P53 binding unless the supershift works. For now I'll try again with p21 oligos to see if I can get better luck with the supershift. Thank you.

@Debomita: I appreciate you reply, you made a great point. I may try again with less proteins to see if it helps. Do you have any tips on flushing the wells? Currently I'm using a P1000 pipettor to flush the wells, is there a better way of doing it?  

Dear Maoyu, we generally use the syringes (5ml or 10 ml) to flush the wells with buffer and just pre-run the gel with loading dye.....they seem perfect....you may try that.

Hi Maoyu!

I my hands, the binding activity of a protein towards DNA is not always reflected in an EMSA. Using ChIP is more specific for quantitating DNA binding. I discovered that I would sometimes hardly see a difference in EMSA, but that the sensitivity is much higher with ChIP. It might be worth trying that approach if you can't get the EMSA to detect a difference.

Hi Maoyu,

It seems that you don't have a specific binding. what you saw on the gel is a non-specific binding of other protein recognizing the same RE. I would suggest the following

1. optimize your salt concentration. I usually use 0.1-0.2 M NaCl. 

2. for supershift I usually use 1ug of Ab , incubate on ice for 30 min before adding the binding reaction.

3. I incubate the binding reaction at room temp for 20min before loading into the gel.

Hi Maoyu,

I am in agreement with others in that you need to verify specificity.  You may want to try doing the shift with a lysate lacking p53 (p53 null cell line, immuno-depleted, etc.) to see if you form complex.

In my experience, the order-of-addition of components (extract, oligos, antibodies) is critical, but largely empirical for the system you are studying.  For example, if your protein binds cooperatively to your sequence, the antibody may interfere with cooperative interactions, so you may want to add the antibody after nucleoprotein complexes are formed.  p53 is known to bind cooperatively to, what many refer to as, low-affinity sequences.

You may also want to consider that p53 may not bind this sequence independently, and may require another protein for interaction with DNA.  If this is the case, the amount of complex formed may be limited by this auxiliary protein.  

A few general notes, If you suspect the involvement of a Zn-finger protein, avoid EDTA in all of your buffers.  This is often overlooked.  As for the problem of well retention, you can try to use a different formulation of polyacrylamide.  Most pre-cast gels are approximately 37.5:1 or 40:1 Acrylamide:bis-acrylamide.  Several sources have recommended a lower amount of crosslinker to help resolve large complexes and to reduce well retention (see Hellman and Fried (2008) Nature Protocols - attached).  Also, include a lane on your gel that contains only probe in binding buffer.  This lets you know if the well retention is due to your buffer or something else.  Centrifuge your lysates really well before setting up your reactions, and check your reactions for sign of precipitation before loading.  

Good luck!

thank you very much everyone, I'm testing out some new oligos and I'll keep you posted if I get a better outcome.

You may also try using 0.5% Tris-glycine for running. It heat much more less than TBE and allow a better resolution. In my hand some complex that dissociate in tbe are pretty stable in tris glycine. and the picture is really nicer, with sharper bands. In some of your lanes, you also have a signal just above your probes, it may reflect complex that disrupted during the migration.  I'm also agree with several people before: A bigger gel is better, you should prerun the gel (I was not putting loading buffer). Wash the wells before loading. Load at the bottom of the well in a way that your sample is not diluted in running buffer and concentrated in a very sharp band in the bottom of the well. Put your running buffer at 4°C the day before the migration....

Hi guys,

I am a having the same problem. Some complex is getting stucked into the well repeatedly..

In case of running gel in Tris Glycine buffer...Should gel also be prepared in Tris buffer like for normal protein gel? I am attaching picture of my gel.S1S4 are my samples and free probe in between.

Please help.

We have always used a very old procedure with 4% native acrylamide gels in TBE buffer containing 2% glycerol. The glycerol, as well as careful washing of the wells immediately prior to loading the gels, appears to aid in generating tight bands with little residue remaining at the top. Our nuclear extracts also contain glycerol.