Chromatin length after fixation--before sonication - chromatin is sheared before sonicating? (Jul/22/2005 )

I am doing sonication optimization for ChIP assays. I ran a gel using different intensities and number of pulses for sonication (always 10s on and 15s off--on ice) As a control, I also ran fixed, but unsonicated DNA on the gel for comparison, and all of the lanes are basically the same. I get a smear of DNA (for unsonicated and for all the levels of sonication) that starts at 500bp and goes to 100bp. It seems like my chromatin was already fragmented at the time of fixation.

I am using NCI H292 cells. I fix the cells with 1%FA for 10min at 37C according to the Upstate protocol which I am using.

I am running a 1% agarose gel and staining afterward with EtBr.

I add RNase to the samples during crosslink reversal, and I have already confirmed the activity of RNAse in the lysis buffer using e.coli rRNA.

DNase should not be a problem as I am using 1%SDS lysis buffer.

I do proteinase K digestion and then directly run a 10ul sample on the gel. I was under the impression that it is not necessary to do phenol chloroform extraction in order to check sonication on the gel, but I have not confirmed this.

Does anyone have any suggestions? Has anyone else had a similar problem? Any ideas as to why this may be happening?

Could you run your gel after phenol chloroform extraction? I still feel the smear is RNA, although you said you have added RNase

Thanks bullfrog!!! My problem was due to RNA contamination, as well as DNA concentration. I am now working to design sonication conditions based upon the DNA instead of RNA


Here are some suggestions so that others hopefully won't have the same problems as I had.

You cannot proteinase K treat the samples in lysis buffer because of the presence of AEBSF so phenol-chloroform extraction/precipitation is necessary. You should add RNase to the resuspension buffer to ensure only DNA shows up on the gel (this is another good reason to precipitate because although the RNase does have activity in the lysis buffer, it is not sufficient to completely degrade the RNA) And finally make sure that you check sonication on a sufficient amount of DNA, i can see about 850ng. This is equivalent to 25ul of lysate when 10^6 cells are lysed in 200ul (Upstate protocol)

The 850ng estimate is based upon the celera prediction of 6.6pg of genomic DNA/cell and assumes 100% lysis, extraction/precipitation, etc... so if you are quantitating directly, 500ng is probably enough.