Using Westerns to check crosslinking in ChIP - (Apr/05/2007 )

I have been trying to perform ChIP on a particular transcription factor and have had a lot of difficulty with it. My shearing on an argarose gel looks fine, but when it comes time to run PCR I have no appreciable signal (but the input DNA works just fine, so there's nothing wrong with the PCR.)

Thinking it might have been my antibody, I then used an antibody against RNA PolII (antibody 8WG16) at a concentration known to work in my lab and got negative results again. So I then thought maybe I was doing something wrong in the actual process of IP and to test against this I did ChIP with chromatin prepared by a colleague of mine which we knew should work. Sure enough, when running his chromatin and my chromatin (from the same cells, cell number, etc.) side by side with the same antibody (RNA PII again), his chromatin gave good clear results and my chromatin did not. So I decided it wasn't anything wrong in my ChIP protocol.

So my next thought is that there must be something wrong with my crosslinking. After looking through this forum I know that one possibility is that if you crosslink too much, the epitope can be masked and that could explain my results. Another possibility is that if you don't crosslink enough you lose your DNA when you precipitate. I was initially crosslinking at a final concentration of 1% formaldahyde for 30 minutes at RT, and I later tried 10 minutes at RT. Up till now I have not been quenching the reaction, instead just washing 2X with PBS.

A colleague suggested one method of checking crosslinking without having to go through the entire process of ChIP. If after crosslinking, you sonicated and then just ran it on an SDS-PAGE gel and performed a Western Blot analysis, we think one could assay crosslinking. If there is not enough crosslinking, the protein of interest should show up at its regular molecular weight because it will be alone. If there is too much crosslinking, the protein of interest will not appear at all because the epitope will be masked. But if there is an appropriate amount of crosslinking, the protein of interest should show up but not at its regular molecular weight, but instead heavier because it crosslinked with other proteins. It could be so heavy that it will even in the loading section.

I am planning on trying this out sometime in the next couple of days, but I was wondering if anyone on here had any input. Perhaps you've tried it before, or have some ideas that might improve the method or see something which we've overlooked that will doom it to certain failure.

I apologize if this topic has already been brought up before, I've looked through what I thought were all the ChIP topics and didn't see anything, but if I missed something please just point me in the right direction.

Thanks for any help you can give us!
mdgalen