Protocol Online logo
Top : Forum Archives: : DNA Methylation, Histone and Chromatin Study

ChIP IgG has the same signal than the antibody tested - (May/15/2007 )

hello

I'm trying to make ChIP with antibody against transcription factors ( I tried with Active Motiv and Upstate) but the control (when I use anti-IgG) has the same signal than all the other even with primer designed for GAPDH which is transcribed and should recruit transcritpion factors.
I tried more wash but it didn't change the problem so if you have some suggestions...
thanks a lot

-sweet-

QUOTE (sweet @ May 15 2007, 07:13 AM)
hello

I'm trying to make ChIP with antibody against transcription factors ( I tried with Active Motiv and Upstate) but the control (when I use anti-IgG) has the same signal than all the other even with primer designed for GAPDH which is transcribed and should recruit transcritpion factors.
I tried more wash but it didn't change the problem so if you have some suggestions...
thanks a lot


To make sure that your antibody works in ChIP you might try running a western blot for your protein after reversal of crosslinking.

To make sure that your ChIP is actually working you might try ChIP with a different antibody. Since TFs can be a little tricky you mighy try an antibody to histone H3 or to RNA pol II since these are both abundant proteins on the chromatin (at least on transcribed genes for Pol II). I've used two Pol II antibodies (Santa Cruz sc-899 for the N-terminus of Rpb1 and Abcam ab5408 for the CTD of Rpb1) and one H3 antibody (Abcam ab1791) in ChIP and all have very good signal to noise ratios (mainly because of the abundance of the proteins).

-KPDE-

thanks
in fact I already have done a western blot ( just after the IP) and the antibody works well.
I also tried with histone H3 and I've a signal, which is more important than the signal of the control IgG but the histone is much more abundant than the TFs I'm looking and the band of IgG is thus more thin compared to this with H3 but comparable to the band of the IP with antibodies against TFs which are less abondant in the cell
Thanks for your answer and if you have another idea.. Wich kit do you use for the ChIP?

-sweet-

QUOTE (sweet @ May 15 2007, 02:43 PM)
thanks
in fact I already have done a western blot ( just after the IP) and the antibody works well.
I also tried with histone H3 and I've a signal, which is more important than the signal of the control IgG but the histone is much more abundant than the TFs I'm looking and the band of IgG is thus more thin compared to this with H3 but comparable to the band of the IP with antibodies against TFs which are less abondant in the cell
Thanks for your answer and if you have another idea.. Wich kit do you use for the ChIP?


Is it possible that the primer you are using is missing the site where the TF binds?

We don't use a kit for ChIP. We developed a ChIP method (Nucleic Acids Research 34(1) e2; Nature Protocols 1(1) 179-185) which is faster and cheaper than any of the kits. If you are interested I can send you a reprint of the Nature Protocols paper.

-KPDE-

Thanks, yes I'm really interested . If you can send me the protocole (I will send you my email adress by PM. In fact I see a signal for the input so I think the primer is right..I don't know, in fact I'm in last year of master in chemistry and I've choosen to do biochemistry but I don't have many experience . I try to learn a lot but I don't have many time before the end of the year.
Thank you for your answers

-sweet-

QUOTE (sweet @ May 15 2007, 09:06 PM)
Thanks, yes I'm really interested . If you can send me the protocole (I will send you my email adress by PM. In fact I see a signal for the input so I think the primer is right..I don't know, in fact I'm in last year of master in chemistry and I've choosen to do biochemistry but I don't have many experience . I try to learn a lot but I don't have many time before the end of the year.
Thank you for your answers


Though the primer may work in that it amplifies the region you expect it to, is it possible that the TF is binding to a region outside the amplicon.

-KPDE-

the primer works in the region I'm interested in because it works with a plasmid containing this region and the band is located at the same place with the IP, the problem is really that the IP with the antibodies against TFs don't seem to give an enriched quantity of DNA in PCR compared to antibody against IgG, which was my negative control ( or my mesure of unspecific association with the beads)
Thanks a lot for all your suggestions..I'm really glad to discuss (with my PC) with scientists from all over the world . I learn a lot. Thanks again

-sweet-

Hello,

I also found that these things will also help in bringing down IgG background (not perfect but it works well so far):

1. use magnetic beads (I use Protein G Dynabeads from In Vitrogen for both rabbit and mouse Abs) instead of agarose beads from Upstate or Santa Cruz
2. for each wash at the end of IP, I do 10-15 min per wash
3. use normal IgG from Vector Laboratories ( I used Rabbit IgG from Upstate before and found that it gives me more background than this one....go figure)

Anyway, these are my 2 cents worth of info. Hope it helps. Good luck smile.gif

-sonixchip-

QUOTE (sweet @ May 17 2007, 01:03 PM)
the primer works in the region I'm interested in because it works with a plasmid containing this region and the band is located at the same place with the IP, the problem is really that the IP with the antibodies against TFs don't seem to give an enriched quantity of DNA in PCR compared to antibody against IgG, which was my negative control ( or my mesure of unspecific association with the beads)
Thanks a lot for all your suggestions..I'm really glad to discuss (with my PC) with scientists from all over the world . I learn a lot. Thanks again


If your primers aren't a problem (although I would suggest tiling a few primers on either side of the primer pair you're using now, just to make sure you're not missing the place where your TF binds) then another possibility is that your crosslinking is not efficient enough to allow you to bring down your piece of DNA of interest with the TF. You might try optimizing your crosslinking (try longer incubations or higher concentrations of formaldehyde) or using other crosslinking reagents. Some have tried using paraformaldehyde which has a longer crosslink than the methylene bridge formed by formaldehyde crosslinking. Also a two step crosslinking protocol has been used for pulling down DNA associated with NF-kB (Biotechniques 39(5) p715-725). The first step stablizes protein-protein interactions that can be important in maintaining TF complexes on the DNA, while the second step is the normal formaldehyde crosslinking.

If it's possible that you are using a very large number of cells per ChIP assay you may be overwhelming the antibody. In this case the high background would overwhelm what you are pulling down with the antibody. You could try titrating the amount of chromatin you use and see if you get a signal above background with the lower concentrations. In a few cases this has worked for me for factors which have a small enrichment above background.

Anyways, hope you figure it out.

Joel

-KPDE-

hi, thanks for your advices. I saw the protocol you send me . We don't have Chelex 100 in the lab so I asked if I could buy some. I'll try your protocol when these beads arrive.ยต
To answer to sonixchip,
we tried magnetic beads with active motif but it didn't work, even if I agree it's easier for manipulation. Perhaps should we have made more washes. Thanks for your intervention too
I've already sonicated..but I'm waiting for the beads so I'll tell you more later.
Thanks again

-sweet-