Yeast ChIP-seq; Sonication/Bioanalyzer question

[AaronS]

Hi,
I've just started doing ChIP-seq, and am having difficulty getting my INPUT and IP DNA fragment sizes in the proper size range for Illumina sequencing. My ChIP protocol seems to work fine, as I can show enrichment of a 360bp region surrounding a known binding site in my IP sample compared to the INPUT sample by pcr.

However, our Bioanalyzer reports that the size fragmentation is too small for the library prep. This seems to indicate that I'm over-sonicating the samples. I've been following a protocol published earlier this year by Mike Snyder's lab http://dx.doi.org/10.1016/S0076-6879(10)70004-5. In this protocol the sonication procedure uses a Branson Digital Sonifier @ 5x30sec @ power 50.

In our lab, we have a Branson Analog Sonifier and I've tried sonicating 6x10sec @ power 20 and 3x10sec @ power 20. In both cases, the bioanalyzer indicates that my DNA size fragmentation is mostly <100bp. However, when I run the INPUT DNA on a 2% Agarose gel, I see a smear with the greatest DNA concentration ~1,000bp and no DNA below 100bp.

[jorgemon]

Hi AaronS.
I think that the not real result is the agarose gel. The bioanalyzer is more specific than gel, where you have less resolution to see small DNA quantity and migration is not the same (the upper marker of a DNA HS is 10380bp, but you see all the largest fragmented sizes grouped around 1500bp). If you try with DNA 1000 or 7500, you will have the same result (you will see all large sizes grouped around 1500bp too).
I think that the smear of fragmented sizes that you see in bioanalyzer, you can't see in agarose...you only can see sizes with more concentration (usually largest sizes but they are in correct size compared with marker, not as occurs in bioanalyzer).
The sample that you see in bioanalyzer at 100bp, may be DNA degradation of Chip and Input. If you purify the samples by column (Qiaquick for example), and run again the samples at bioanalyzer, you will not see anything less than 150bp.
You can try to fragment more Inputs with different times. Then you can run it in a bioanalyzer and know if you are fragment too much your samples.
It can be a protein contamination from previous step too (it could explain this results).
In order to can see better the DNA purified by column, you can put 2-3ul evaporated (quantitative range of DNA HS is up 0.5ng/ul).
I hope you can solve this problem with my comments.
Regards.