ChIP - some basic FAQs - (Mar/24/2005 )
Hi,
I want to study histone modifications affecting a particular gene that I'm working with. I guess ChIP is the method to use, but I (or my lab) have no experience in this method. How hard is it to do, how much time should I expect for this kind of experiment? Any advices? Are there any companies that do it for you?
Thanks!
I must say ChIP is quite challenging. It usually takes 2-3 days besides the time needed for cell treatment. I don't know if there is any company offering such service but Upstate does sell good ChIP kit. Also check their website for more info on ChIP such as protocol, faqs.
Hi Molbio,
may the ChIP gods look upon you favourably.
Pcrman is right in saying it is a very challenging technique to master, but once mastered it's very rewarding to see the results.
Upstate have a very good kit, I would also like to add that Abcam are very good with their antibodies, some if not most histone modification antibodies are guaranteed for ChIP.
I know abcam have sourced their H3K9 trimethylated from the Jenuwien Lab, a very good antibody for heterochromatin.
Be aware with the antibodies though, even though they may well be guarnteed for ChIP I have experienced that this is not the case, so don't go out and buy a big batch of a particular antibody until you are sure the darn thing works! Note the bacth number of that antibody that you test and insist on getting that batch as the activity can vary between batches.
good luck with it and may the ChIP-force be with you always!
Nick 
Thanks for our answeres. I'm going to try it at least...
I´ve been looking into some antibodies. I suspect that my gene of interest is deacetylated (due to TSA studies) do I have to try H2A, H2B, H3 and H4, or is one of these sites more likely to be acetylated, or is all of them acetylated if one is...
Could you recommend which one to start with?
Thanks again!
H3 and H4 are most studied for acetylation. Try them first.
Ok,
I'm already stuck... I've been going through the protocol for the ChIP kit from Upstate. Since I have noone to ask at my lab, I turn to you experts!
So, I want to see if a gene that is up-regulated after treatment with TSA really is deacetylated.
Upstate protocol:
"Part A - optimization of DNA shearing
1. Cells should be treated under conditions for which transcriptional activation has been demonstrated"
- should I treat the cells by TSA? But my interest lies in if it is normally deacetylated... (The gene is expressed but at lower levels also without TSA)
I'm also a bit confused about the controls.
"Part B - experimental control
For a negative control, prepare a sample to use as a no-antibody IP control"
- ok, makes sens
"additionally, transcriptionally unactivated DNA should be prepared as controls in PCR."
- Cell not treated with TSA? Which is what I'm interested in, but as control?
I'm really confused, could anyone please clarify this for me?!
And in the PCR, I've understood that real-time PCR is best. Can I design normal TaqMan probes as for expression analyzes and use them? Designed around the promoter region then, I guess...
By the way, does anyone know of some kind of course in epigenetics and epigenetic methods, think I really would need it...
Thanks again!
Hi Molbio! If this is your first ChIP experiment, I won't do any treatment with TSA or anything. I would advise you to take the cells you're going to use and do a simple antibody/ no-antibody experiment just to get your hands on and see where the problems lie and familiarize yourself with the different steps of the technique.
I mean, you should harvest and sonicate enough cells to do two precipitations and then set up the IP with the antibody you're using (for what you're saying, I guess you're using an anti-acetyl histone), and in parallel an IP with no antibody or with an unrealted antibody (some non-immune serum, or an ab against a cytoplasmic protein. I generally use anti-insulin).
As for the transcriptional activation and inactivation controls in the Upstate protocol, your cells will have constitutively active genes and silenced genes that you can use as positive and negative controls, respectively. I use beta-actin as a positive control for acetylation, as it's expressed in all cell types and is highly acetylated in all cell lines I've used 'til now. As a negative control I use tissue-specific genes, such as insulin (when I'm not working with beta-cells) or albumin (when I'm not working with hepatocytes). In fibroblasts, both these genes are silenced and nonacetylated. In hepatocytes, insulin is silenced and nonacetylated. So what you need to do is to get primers for different genes that you can use as positive or negative controls. Usually the primers will be designed in the promoter region, as you say.
Then, in your IP'd antibody/ no-antibody samples, you can check the amplification of your positive controls (for which you should have a band in the anti-acetylhistone and input samples, but not in the no-ab sample) and your negative controls (for which you should have a band only on the input sample). If you get these results, you can assume that the technique is working and you can proceed to check your gene, and do TSA treatments, etc.
Also, if you haven't done it yet, I would advise you to check the quality of your sonicated samples before proceeding to immunoprecipitation.
Well, hope that it helps. Good luck!!
Oh, and for molecular biology courses, if you're in Europe, you can check the EMBO schedule of courses and seminars. Most of the practical courses are free. I've been to several of them and they're really fun 
Thank you for your answeres, the "control issue" is no much clearer to me. Just two more questions...
"Check your quality of your sonicated samples"
- In what regard? Check the sizes of the sonicated fragments or what do you mean?
The second thing is primer design. I want to use TaqMan real-time PCR... For my gene of interest, is it enough to design one primer pair? I´ve seen in publications that some people design a couple of pairs for different regions, some not even in the promoter region. And they seem to get different results, but if I'm just interested in finding things that can affect the expression of the gene is it enough to look at the promoter? And should I design several primers so I cover the entire promoter?
Thanks again!
- In what regard? Check the sizes of the sonicated fragments or what do you mean?
hi molbio,
that is exactly what is meant about the quality of sonicated samples. you need to have small fragments, (sub 2kb) so they are of a size that can be IP'ed by your antibody.
As for the primer design, it is safer to pick more that one set of primer for the same region as a way to be sure your region of interest is in fact binding to the protien.
good luck matey!
Nick