Library prep issue

[nisha barak]

Hi,
Any help on this would be greatly appreciated!

I use the NEB mastermix library prep kit to make libraries from my ChIP samples. For the past month or so I have had poor yield for my ChIP libraries (but not for the input libraries). One thing I've noticed is that during the ligation step when I add the ligase buffer to ChIP samples I get a white precipitate which turns the sample cloudy. On the other hand when I add the ligase buffer to the input samples the sample stays clear. Any ideas why the precipitate forms and if it could affect the efficiency of adaptor ligation?
I tried asking technical support at NEB and they suggest the following:

'It sounds like maybe there is still protein left in the sample, maybe the reverse crosslinking did not get completed? It sounds like the protein in the sample is most likely interfering with the Blunt/TA ligase formulation. I would suggest to check your sample as this may interfere with the ligation and PCR step.'

I reverse crosslink for 5 hours with NaCl at 65C and treat with Proteinase K overnight followed by phenol-chloroform extraction and ethanol precipitation. I treat my input and ChIP samples the same so not sure why protein contamination would not also happen in my input sample.

[Genohub]

First, what do you mean by "input" libraries ? Two things regarding the precipitate come to mind. Could there be residual phenol-chloroform in the sample causing the protein to precipitate? Any trace amount of chloroform will certainly inhibit the ligation reaction. Or could it be excess DTT, which is precipitating out of the ligation buffer?

- Genohub

[nisha barak]

Hi,
Thanks for your response.
Input libraries refers to an aliquot taken from the sonicated chromatin which was not ChIPed with an antibody but was RNAse treated, reverse cross-linked, proteinase K treated, etc alongside the ChIP sample.
I use phase-lock gels for phenol-chloroform extraction so I think this helps reduce contamination. I think its definitely possible that DTT is coming out of solution.
One big difference between ChIP and Input is that I pool around five ChIPs to increase yield and for each of these ChIPs I am adding Proteinase K in excess (5microlites of 20mg/ml each sample). So possibly there is contamination. Also I ethanol precipitate each sample separately and add 2 microlitres glycogen (10mg/ml) to each. So potentially by the time I have pooled all samples into a small 30 microlitre value I have a high enough concentration of glycogen to inhibit downstream reactions. I'm now lowering the amount of proteinase K and glycogen I use and will see if this helps.