Hello,
I am improving my X-ChIP protocol (x stands for DNA crosslinked to proteins with formaldehyde). In my protocol I do not isolate nuclei. I just use the whole cell lysate prepared with a buffer containing 1% SDS. My protein could be in nuclear and cytoplasmic fractions. I now think I might need to isolate nuclei first. What would be the best approach and do I really need this?
When I inspect my whole cell lysate under a microscope, it looks like nuclei stay intact. It is only after a couple of rounds of sonication nuclei star to disappear. I already tested spinning down my whole cell lysate for 13000rpm (max speed, benchtop centrifuge), 20 min and I got a pellet that contained 90% of all DNA in the sample. The question is, I did not see that anybody isolated nuclei in this way (lysing cells with 1% sds containing buffer and then isolating nuclei 13000 rpm 20 min). Does this sound right?
When you do ChIP or X-ChIP you want to identify the binding sites of the protein of interest in the genome. Therefore, the fraction of the protein in the cytosol is not really relevant in most cases. Nuclei isolation is a logical step and it improves the recovery of the protein-DNA complexes quite substantially. In most cases, sonication is not sufficeint to lyse the nuclei, you, typically, want to use a lysis buffer to break the nuclei prior sonication. So what I would suggest is that:
1. Lyse the cells
2. Isolate the nuclei
3. Lyse the nuclei
4. Then shear the chromatin with sonication.
Best of luck
Dear Waleed,
Thank you very much for your help. Do you happen to have a protocol to do listed 4 steps?
You're very welcome. You can find a detailed protocol I used in the supplementary figures and tables document in this paper
https://academic.oup.com/jxb/article/69/11/2847/4985136
which is a modified version of the following protocols
https://www.nature.com/articles/nprot.2008.66
https://www.nature.com/articles/nprot.2009.244
As you may have noticed already that those are two it is a protocol to perform ChIP on Arabidopsis plants so you might need to tweak the protocol a little bit if you work on non-plant tissue/cells. I have used the same composition of most buffers to perform ChIP, RIP and CLIP-seq on yeast cells and it worked well with no problems. For instance, you do not need to to use vaccuum infiltration to crosslink cells and you do not need to include sucrose in the buffers if you are working with cells or mammalian tissue. You can also omit the liquid nitrogen grinding step if you're working on cells. Depending on the sonicator you use you may have to optimize the sonication step to get the right fragment size as sonication is a very variable step. You may also have to optimize the IP step, if you have a good antibody that works well with protein A/B magentic beads then I recommend using that instead of Agarose/Sepharose beads.
Best of luck
- Badges
- Science method