Questions related on Fast ChIP protocol - (Oct/20/2010 )
Hi,
I'm relative new to ChIP. I'm employing the Fast ChIP method by Nelson et al (2006). So far, it has worked pretty well. However, I'm a bit concerned about my data for a couple of reasons.
First of all, I am getting very low "percent input" values, even for the transcription factors (including Sp1, p50, and p65) that I expect to bind well to the sequence I am amplifying. For example, in a recent assay I obtained the following mean Ct values:
Input (1:100 dilution): 27.1 - equivalent to 20.5 after calculating dilution factor
Rb IgG: 34.9
Sp1: 29.8
After performing calculations with my primer amplification efficiency I get:
Rb IgG % input: 0.0016
Sp1 % input: 0.08
Thus, from these data, Sp1 is present at 50-fold enrichment at my site of interest. However, I'm concerned at the very low percent input values. Also, as several other of the factors I've pulled down are giving mean Ct values in the 30-32 range, variability between replicates is high, and I'm not sure how much confidence to have that these are true pull-downs due to the extremely low percent input values, even if they are significantly lower than the IgG control antibody Ct reading.
In conclusion, are Ct readings in the 29-33 region legitimate for calculating ChIP results? How about percent input values in the ranges I'm showing? I'm using 500,000 cells per pulldown, so I'm testing different cell numbers to see if that shifts the curve, but since the assay is working well otherwise I don't want to change too many parameters.
Any help or advice would be greatly appreciated!
VirusGuy on Wed Oct 20 20:35:48 2010 said:
Hi,
I'm relative new to ChIP. I'm employing the Fast ChIP method by Nelson et al (2006). So far, it has worked pretty well. However, I'm a bit concerned about my data for a couple of reasons.
First of all, I am getting very low "percent input" values, even for the transcription factors (including Sp1, p50, and p65) that I expect to bind well to the sequence I am amplifying. For example, in a recent assay I obtained the following mean Ct values:
Input (1:100 dilution): 27.1 - equivalent to 20.5 after calculating dilution factor
Rb IgG: 34.9
Sp1: 29.8
After performing calculations with my primer amplification efficiency I get:
Rb IgG % input: 0.0016
Sp1 % input: 0.08
Thus, from these data, Sp1 is present at 50-fold enrichment at my site of interest. However, I'm concerned at the very low percent input values. Also, as several other of the factors I've pulled down are giving mean Ct values in the 30-32 range, variability between replicates is high, and I'm not sure how much confidence to have that these are true pull-downs due to the extremely low percent input values, even if they are significantly lower than the IgG control antibody Ct reading.
In conclusion, are Ct readings in the 29-33 region legitimate for calculating ChIP results? How about percent input values in the ranges I'm showing? I'm using 500,000 cells per pulldown, so I'm testing different cell numbers to see if that shifts the curve, but since the assay is working well otherwise I don't want to change too many parameters.
Any help or advice would be greatly appreciated!
Those are pretty high Cts. I wouldn't put a lot of confidence in those values, especially if, as you say, your signal to noise ratio is pretty bad. Have you tried using an antibody to a factor which will give you a high signal, like for Pol II (4H8 monoclonal to Pol II CTD) or H3K4m3, looking just downstream of the transcription start site. When starting up with ChIP it's always a good idea to use a "positive control" antibody like this, to make sure the assay is working in your hands. Not long after developing Fast ChIP we were getting % inputs for Pol II around 7-15%, just to give you a target.
Having said that, you can try increasing the chromatin amount you are using while as the same time increasing the amount of antibody (no need to increase the amount of beads as you should be pretty far from saturating them). This should increase your yield without decreasing your signal to background (IP/IgG) ratio.
Hope that helps but let me know if you have any more questions.
Joel
Hi Joel,
Thanks very much for your response. I've located some pol II antibody and will give that a try to check the pull-down efficiency.
One question. I notice in your original fast ChIP protocol you use a lysis buffer that lacks SDS to do the cell lysis/sonication steps. I've been using a pretty standard ChIP lysis/sonication buffer that contains 1% SDS. I'm using THP-1 cells in my assay in which nuclei can be somewhat difficult to lyse, so that's why I modified your protocol. I dilute my sonicated chromatin 1:10 with your lysis buffer (+5% BSA and herring sperm DNA) to get a final SDS concentration of 0.1% and then proceed with the pull-downs. However, I've read that sonicating in 1% SDS and even the 0.1% SDS concentration in the pull-down buffer can had adverse effects on pull-down efficiency, especially for some antibodies. What is your experience? Did you consciously choose to do your lysis/sonication sans SDS?
I appreciate your help. I really like the protocol and have found it easy to employ as a ChIP first-timer.
Luke
VirusGuy on Sat Oct 23 23:50:34 2010 said:
Hi Joel,
Thanks very much for your response. I've located some pol II antibody and will give that a try to check the pull-down efficiency.
One question. I notice in your original fast ChIP protocol you use a lysis buffer that lacks SDS to do the cell lysis/sonication steps. I've been using a pretty standard ChIP lysis/sonication buffer that contains 1% SDS. I'm using THP-1 cells in my assay in which nuclei can be somewhat difficult to lyse, so that's why I modified your protocol. I dilute my sonicated chromatin 1:10 with your lysis buffer (+5% BSA and herring sperm DNA) to get a final SDS concentration of 0.1% and then proceed with the pull-downs. However, I've read that sonicating in 1% SDS and even the 0.1% SDS concentration in the pull-down buffer can had adverse effects on pull-down efficiency, especially for some antibodies. What is your experience? Did you consciously choose to do your lysis/sonication sans SDS?
I appreciate your help. I really like the protocol and have found it easy to employ as a ChIP first-timer.
Luke
I don't have any experience using SDS in ChIP as I haven't needed it for any of my cell types. I suppose you could always try diluting your chromatin 2-fold more and then use twice as much during the IP.
I do ChiP on virus infected cells (if that's what you doing, if not disregard) and have noticed that, and I think KPDE would agree, it is best to perform PCR on regions of cellular genome where you know your factor of interest binds and one where it doesn't. There is lots of ChiP/CHIP and ChiP/Seq data available nowadays to find these regions for your specific cell type. I have found that IP efficiencies are lower for viral sequence for my ChIP'ed stuff but my positive and negative controls for cellular sequence works out fine. One of my guesses is that I don't get rid of membrane associated virus after the infection period. Also DNA viral genomes (at least those who have a nuclear step in their life cycle sometimes are treated heterogeneously during infection (ie- some get the factor bound but others don't. FISH for certain viral genomes has shown that when multiple particles infect the same cell that some act as template for transcription and for whatever reason others are transcriptionally stagnant. Lastly, if viral DNA replication is occurring during the time point at which you fix the cells there will be a set of genomes which are not associated with chromaitn. Hope this rather late reply helps.