branson S250 shearing crosslinked DNA for ChIP

[ChipBrian]

I've been trying to shear formaldehyde-crosslinked chromatin using a Branson S250 for ChIP-Seq, and I've been having some problems with attaining the correct size distribution.

I tried reading all of the posts related to shearing, so if someone could point me in the right direction, that would be great.

Basically, I keep getting a bimodal size distribution with a cluster at ~2.5 kb that is really recalcitrant to being sheared. I can break it up with longer shearing times, but at this point my fragments are too small (mostly <300 bp). I heard that one should have a tight size distribution (200-800 bp) for going on to the ChIP and library prep.

I've attached a shearing time course using Bioanalyzer.

Here are the details:

I've been following a procedure from a 2006 Nature Methods paper from Richard Young's group for the crosslinking (10 minutes in media with 1% methanol-free formaldehyde at room temp.) and lysing conditions (nuclear isolation).

The shearing conditions are:

1. 15 million cells in ~1 mL of sonication buffer (10 mM Tris-HCl, 1 mM EDTA, 0.5 mM EGTA, 100 mM NaCl, 0.5% N-laurylsarkosyl, 0.1% Deoxycholate).

2. 45% amplitude, 0.7 s pulse on, 1.3 pulse off, then timecourse.

I tried using SDS (0.1% or 0.2%) instead of N-laurylsarkosyl with not much success. I was thinking about going to 1%.

Any help would be greatly appreciated!!!

[Hamid]

Hi Brian,

What you are noticing is due to one of the following reasons:
1. over-fixed chromatin. when chromatin is over-fixed areas with high protein content tend to over-fix and is quite resistant to shearing. On a bioanalyzer trace it often appears at a peak of 1-7kb. You can resolve this by reducing the fixation time. Since you are using a high power probe sonicator, your choice of reducing the fixation might be limited since the epitopes will be destroyed otherwise. Do you have access to a Covaris instrument? If you let me know where you are located, I can see if there is an instrument at your institution you can use.
2. Cell culturing. damaged and dead cells tend to fix more easily, and will contribute to the peak that you are noticing.
3. use of High sensitivity bioanalyzer chip. use regular sensitivity 12k chip only. The high sensitivity chip always seems to show secondary peaks which do not appear on regular sensitivity chips, or agarose gels.

I have attached a bioanalyzer traces that illustrate effects of over-fixation, and one that show the gradual reduction in size and it the fragment distribution of Covaris fragmented chromatin.

Thank you

hamid

[kwaraska]

Brian-
I don't know where in Boston you are but we at the BioPolymers Facility at Harvard Med School have a Covaris instrument that Hamid talks about.

Please go to our website if you want more information

https://www.genome.med.harvard.edu

[ChipBrian]

Thanks for the suggestions and comments; it is much appreciated.

Is the peak from the high sensitivity DNA chip an artifact? The peak seems to be broken down after awhile with more shearing.

I saw the earlier posts on over-crosslinking and the effects of methanol and media on the rate of crosslinking.

I did a little cross linking timecourse study for our protein of interest (it doesn't directly contact DNA) using ChIP-qrtPCR. I did 5 min crosslink in PBS (after 1 PBS wash), 5 min crosslink in media w/serum, 10 min crosslink in media w/serum, and 15 min crosslink in media w/serum, all at RT with 1% formaldehyde. The 10 min crosslink in media was the only condition that gave us good ChIP-qPCR (overall % input and difference between positive and negative promoters).

I switched over to shearing in serum free media, and it seems to help. As far as I understand, serum helps permeabilize the membrane, so all I'm doing is reducing the crosslinking by getting rid of it.

I heard great things about the Covaris; the only thing is that our lab has a Branson so it's very convenient and the qrt-PCR studies have already been worked out.

One last question I have is, how important is it for the size distribution to be 200-700 bp for going onto the library prep? I've seen other people suggest that 100-600 bp is sufficient. I'm going to gel purify and cut out what corresponds to an insert size that is 150-200 bp anyways.

Thanks again for the help!

[qsonica]

Probe sonication works well for most people but you must be sure the tip is submerged properly each time. If its too deep or too shallow it will affect the results.

You also need to ensure you have the correct size tip for your specific sample volume. What is your sample volume? What size tip are you using?