I am performing chromatin immpunoprecipitation assays (ChIP) in which I transfect HT-1080 cells with vector control or p53. When I use an anti-p53, I consistently pull down less DNA in the p53 overexpressing cells when compared to the vector control cells. This is even observed at the p53 binding site in the p21 promoter, which should function as the positive control for this experiment. We know the transfection efficiency is high (at least 75%) and the cells are not dying. Are there any other reasons this could occur?
Hi Jillian,
When you say you're using anti-p53, are you using it to precipitate the DNA in the ChIP kit?
In addition, How are you measuring your DNA concentration? p53 is a DNA binding protein and it may mask some measurement of the DNA (espeically if you are using native gel electrophoresis)
Interesting initial experiment, I recommend you should go to the next molecular level to find more results and conclusions.
In the first thought it does seem to act counterintuitively. However, more protein does not always mean that it is more active/binds more DNA. I do not know much about p53 but it is one of the most studied proteins. Does it need a post translational modification in order to bind to DNA? Perhaps a cofactor? Large excess of p53 can sequester some of the needed factors. The cellular localization can also change - high protein concentration can cause proteins to aggregate and take them out of their normal function. Another option is that since there is large amount of p53 after over expression that is not bound to the DNA, you are preferentially pulling down p53 that is not chromatin bound. I hope you can figure it out!
Thank you all so much for your answers! Hareth, yes I am precipitating chromatin using the Zymo-Spin ChIP kit. I extract genomic DNA from 5 million cells for this experiment, verify shearing via agarose gel, and then perform RT PCR for analysis.
Silvia, I have done almost the exact same experiment but instead of overexpressing p53, we use the HCT-116 cell line and compare the results to the HCT-116 p53 -/- cell line and get beautiful results with all of our controls working perfectly. Further, we have confirmed our findings by western blot, measuring changes in mRNA, and evaluating functional changes in cells so we know that the strange ChIP results are due to a technical issue.
Tonis, these are all excellent points. p53 does undergo various types of post-translational modifications. My main problem though is determining which one might be causing these issues! While by IF I see that overexpressed p53 does enter the nucleus, you are correct in that we don't know that it necessarily is all binding to the chromatin and functioning or if it's aggregating and thus non-functional. I hope I can figure it out too!
Andrei - thank you this paper has given me some great ideas on what I can try! We have not yet attempted to induce DNA damage or change the cell's redox state to see if it facilitates binding. We are using the same p53 antibody as was used in this paper, so at least that alleviates that concern. Thank you so much for sharing this paper with me, hopefully the cellular manipulation methods that are described within will help resolve our issue.
I understand, according your comments, that it was not binding at all in the p53 overexpressing cells when compared with your negative control vector - that is in your first experiment. Which was the binding problem? Thank you Jillian
Hi Silvia, there is no measurable increase in the amount of p53 bound under any condition that we have tried yet when we overexpress p53. So we are not exactly sure what the issue is. I'm also precipitating with a second antibody to measure changes, and in that case, the readings are always as expected so I don't believe it's a general technical problem but rather specific to how p53 is or is not interacting with the DNA.
Hi Jillian, I'm assuming that you're overexpressing WT human p53. Have you checked the sequence of the intrinsic p53? I hope it doesn't have a mutation(s) that has preferential binding to promoters. I'm actually surprised that p53 is binding to promoters without any stress if there are no mutations associated with it. Literature point toward PTMs being critical in DNA binding.
Have you looked at the mRNA of p21 or MDM2 when p53 is overexpressed to confirm your observation on the p21 promoter?
Is there a possibility to regulate the amount of p53 expression, as in the case of a Tet-ON system? As Tonis pointed, overexpression of p53 may predominantly pull down unbound p53.
Hi Ramesh,
These are all excellent points you bring up. We have sequenced our construct and there are no mutations in it. Using luciferase assays we have seen clear and specific changes, although perhaps due to overexpressing a plasmid along with p53 we are increasing the amount of p53 that will bind? I have not looked at p21 or MDM2 mRNA, but by western we see an increase in the protein levels of at least p21. We are only now purchasing an MDM2 antibody. Do you have a good MDM2 antibody you can recommend? I have also ordered primers for MDM2 as you suggest.
Regarding the amount of p53, I think the overexpression must be causing the pulldown of the unbound p53 as Tonis and you suggest. We have not considered a Tet-On system though; thank you for this suggestion I will certainly look into that!
- Badges
- Science method