Weird bioanalyser profile

[nisha barak]

Dear all,
I am trying to set up ChIP-seq in our lab. My problem is that, following ChIP, the bioanalyser profile of my input and samples is very different. The input shows plenty of material in the 2-300bp range. However, all my ChIP samples show a peak at around 90bp and go to baseline after that (see attached).
Just to give you some info on the procedure I am using:
Starting with 7 x 10^7 cells (Drosophila Kc cells)
Fix with 1% formaldehyde for 10 minutes
Sonicate with the Bioruptor for 7 pulses (30secs ON 30secs OFF)
(I've optimised the number of pulses to make sure I get a smear of between 100bp and 1Kb when I look at an aliquot on a gel following reversal of crosslinking)
I block samples with Rabbit IgG for 3-5 hours and treat with the antibody for the TF I am studying O/N.
The beads are agarose G. I do the RNase, reversal of crosslinks and Proteinase K steps, and do Phenol-Chloroform extraction to purify using Glycoden and carrier tRNA.
I have also done this exact procedure with the positive control H3K4me3. qPCR indicated an enrichment of a positive site by 40-50% so I assumed the ChIP had worked. However, the bioanalyser profile is the same as for my samples with a peak at 90bp and nothing bigger.
If anyone has seen this kind of thing before and has any ideas I'd be really grateful.
Thank you!

[Chipper]

The spike and dip is present also in input, and is likely instrument related. I would not expect to see a mear on the dna1000 chip, we usually can not detect our sample on hs chips before library amplification.

[nisha barak]

Thanks for your reply. I forgot to add: I also measured my samples on the Qubit and am getting very high (unusually for ChIP) concentrations, around 80 ng per microlitre (in a total of 30 microlitres). Attempts at library preparation (done by the sequencing centre) of these samples has failed as there is not enough material at around 200-300bp.

[lterhune]

Thanks for your reply. I forgot to add: I also measured my samples on the Qubit and am getting very high (unusually for ChIP) concentrations, around 80 ng per microlitre (in a total of 30 microlitres). Attempts at library preparation (done by the sequencing centre) of these samples has failed as there is not enough material at around 200-300bp.

Hi Nisha, I also recently did a ChIP experiment that we have done previously, and got 52ng/ul in a total of 50ul on the Qubit. We haven't tried the library prep yet but are wondering if it is a waste of time and reagents. Let me know if you find anything else out!

[nisha barak]

Hi Iterhune,
I have discovered that the high amount of DNA indicated by the Qubit was actually the tRNA carrier I was using during phenol-chloroform extraction. When I ran my samples on a gel I just got a concentrated 90 bp band. On the other hand if I used spin columns (and therefore did not use a tRNA carrier) instead of phenol-chloroform extraction, this band disappeared. Now I know that my ChIP had too low a DNA amount to successfully make a library so I am instead trying to do several ChIPs which I will pool together.

[pmiguel]

Hi Iterhune,
I have discovered that the high amount of DNA indicated by the Qubit was actually the tRNA carrier I was using during phenol-chloroform extraction. When I ran my samples on a gel I just got a concentrated 90 bp band. On the other hand if I used spin columns (and therefore did not use a tRNA carrier) instead of phenol-chloroform extraction, this band disappeared. Now I know that my ChIP had too low a DNA amount to successfully make a library so I am instead trying to do several ChIPs which I will pool together.

Can I just pause at this point to offer a blanket denunciation on the use of "tRNA carrier" on/with any sample that is to be sequenced? I have sequenced a fair amount of yeast in my day deriving from this source. As a result I am carrying a lot of unsublimated rage over this practice.

--
Phillip