for all you emsa buffs - (Oct/03/2005 )

all right...I have been doing emsa for quite some time now and usually have pretty good luck. I have looked at both NF-kB and AP-1 binding in the promoter regions of 3 different human genes and had all sorts of success. I have supershifted the NF-kB stuff and also had all sorts of success.

here's the problem: I am trying to supershift the AP-1 stuff for one of the 3 promoter regions. I am trying a Jun panel and a Fos panel. I have done this about 60 ways from Sunday with all sorts of variables in the binding reaction involving adding the antibody, the extract, and the probes all at different times; I have also monkeyed with the binding buffer and made about 6 different versions with differing glycerol, BSA, salts, all the other stuff.

When I get to the finished product, I see a vast reduction in the shifted band, but no supershifted band or any other band appears. I have been able to find a few random websites with an explanation - the antibody has tied up all the TF and so it is unable to bind the probe, which only leaves a tiny amount of TF/probe complex to appear on the blot. perhaps the complex (AP-1 can be a very large multimer) does not enter the gel.

So, basically the supershift is working or the shifted band would not be lessened. But, I do not know if this is publishable because I have not seen it mentioned in a journal reference.

What are your opinions? does anyone know of a reference for this phenomenon?

any help is really appreciated!!

Aimee

Hi, I don't know about the publishability, and i have only considered EMSA theoretically, and have yet to attempt practical application, but here are my ideas....

maybe you should change the antibody you are supershifting with, if it recognizes the DNA binding domain and prevents interaction with the oligo then an antibody with another recognition site should supershift efficiently??

Also, maybe it is possible to incubate with the oligo and TF first then later add the antibody so that it is not able to prevent the interaction with the oligo?

anyway I hope these thoughts help...Good Luck!

thank you for the help, becca

I have tried adding the oligo first, but that results in no change from the original shift. I do agree that probably the epitope recognized by the antibody is important for binding the oligo.

I have not tried another antibody. AP-1 can consist of at least two members of a very large family...the two antibodies I am trying each recognize a panel of 4 proteins involved in two different AP-1 subfamilies

I could spend a hell of a lot more money and order about 20 antibodies, but I would like to be able to screen before I go that route...and based on the reduction of the original shift, I know that it's working but I'm having a difficult time getting visual proof so it can be added to the paper

arrggghhhh!

Anyways, thanks for the input

thank you for the help, becca

I have tried adding the oligo first, but that results in no change from the original shift. I do agree that probably the epitope recognized by the antibody is important for binding the oligo.

I have not tried another antibody. AP-1 can consist of at least two members of a very large family...the two antibodies I am trying each recognize a panel of 4 proteins involved in two different AP-1 subfamilies

I could spend a hell of a lot more money and order about 20 antibodies, but I would like to be able to screen before I go that route...and based on the reduction of the original shift, I know that it's working but I'm having a difficult time getting visual proof so it can be added to the paper

arrggghhhh!

Anyways, thanks for the input

Actually, yes I did find it and get the band to appear. YAY! the paper went out Monday...what a relief, it was the last piece of the puzzle that we needed.

I upped the BSA a bit in the binding buffer, decreased the amount of antibody a little, and increased the incubation time with the oligo probe a little. This allowed the antibody to bind up the TF (the BSA helped, and the decrease in amount was just because I didn't think the excess was getting me anywhere), then gave more time for the oligo to interact with the complex.

Anyways, I found the magic combination of incubation and buffer components and got lucky.

I think the deal with supershifts is that 10 minutes here or there, 50 uM here or there in the buffer, all these tiny little changes that can influence the binding reaction in big ways are difficult to iron out, and it is a little different for every TF and every promoter.

Good luck with yours; that is an interesting observation. I have not yet used CHIP but we were going to go there next

Actually, yes I did find it and get the band to appear. YAY! the paper went out Monday...what a relief, it was the last piece of the puzzle that we needed.

I upped the BSA a bit in the binding buffer, decreased the amount of antibody a little, and increased the incubation time with the oligo probe a little. This allowed the antibody to bind up the TF (the BSA helped, and the decrease in amount was just because I didn't think the excess was getting me anywhere), then gave more time for the oligo to interact with the complex.

Anyways, I found the magic combination of incubation and buffer components and got lucky.

I think the deal with supershifts is that 10 minutes here or there, 50 uM here or there in the buffer, all these tiny little changes that can influence the binding reaction in big ways are difficult to iron out, and it is a little different for every TF and every promoter.

Good luck with yours; that is an interesting observation. I have not yet used CHIP but we were going to go there next