Hi Juliette, 

What are you trying to immunoprecipitate, a histone or a transcription factor? 

Where you indicate frozen retinas, you mean they were freeze following fixation? 

Hi Raquel,

Once the fixation/sonication conditions set-up I will try to immunoprecipitate with an antibody against a transcription factor. I did not tell in my first message that I'm using Diagenode shearing optimization kit. The retinas were either fixed freshly after dissection, or frozen for 3 days at -80°C after dissection. In fact I want to set-up the conditions for the two cases frozen and fresh tissues...

Do you have any suggestion?? I'm really worried about this number of cycles and also about the duration time of fixation. Do you think that fresh retinas need less fixation time duration?

Thanks for your help

Juliette

I can't help you with the frozen tissue, I have never freeze and then fix the material. I normally fix and then freeze.

In general terms I agree that lane 5 of frozen tissue is properly sonicated. However, if you are trying to immunoprecipitate a transcription factor, that sonication could be too much.

Transcription factors are normally bigger proteins than histones and the more you sonicate the material, the more difficult it is to ChIP them. 

Maybe you could try to ChIP with chromatin in lane 5, if ChIP doesn't work, I recommend you to decrease sonication time. 500 - 800 bps sheared chromatin is not so optimal but sometimes mandatory to get a proper ChIP signal with transcription factors.

As long as you have proper positive and negative controls there shouldn't be a problem.

Good luck!

Hola Raquel,

Thank you very much for your answer. The trasncription factor we will immunoprecipitate is approximately twice the size of an histone (arounf 37kDa). I will follow your advices! When you say decrease the sonication time, do you mean I keep 3x10 cycles but I change the 30"ON-30"OFF into 25"ON-25"OFF, or I decrease the number of cycles? I guess nobody can answer this question before doing the experiment, but I'm just trying! ;-)

Juliette

We normally decrease the number of cycles, but I suppose it doesn't matter.

All the best,

Raquel

Ok! i guess there is no ideal way to do it and so many parameters! Let's follow our intuitions!

Thanks Raquel for your help!

Best regards

Juliette

Hi Juliette,

0.65 µl tubes (which are not TPX) can shear only one unique sample volume 100 µl. Any deviation from this leads to the lack of shearing!

Please contact me at [email protected] if you need more help

Hi Irina,

Thank you for your advice! I'miss in contact with Céline from Diagenode who advised me the same! We will do so in the next experiment. 

Thanks again

Juliette

Hi Juliette,

have you done any progress in relation with this issue?

I am struggling in the same point of ChIP. My goal is also to ChIP a transcription factor for subsequent exonuclease digestion (ChIP-exo) and sequencing (I do not have any known target to test performance of the ChIP by PCR so the best I can do to control before going to sequencing is a parallel independent positive control). I am using cell lines and I always fix fresh samples and freeze the fixed cell pellet, so I cannot help you with the fresh and frozen samples either. I am using an older version of Bioruptor and 1.5ml tubes with 250µl sample.

The first thing I am struggling with is the lysis of the cells. I have understood that it is better to do a 2-step lysis, meaning that cell membranes are first lysed to obtain a nuclear pellet (with a "gentle" lysis buffer which should only disrupt the plasma membrane and leave the nuclei intact) and then the nuclei are resuspended in a more harsh lysis buffer (containing SDS or other ionic detergents) and sonicated for chromatin fragmentation. Another option is the one-step lysis: using a lysis buffer that will disrupt both plasma and nuclear membranes at once, but in this case also cytoplasmic proteins will be present in the IP reaction leading to higher unspecific binding and poorer enrichment of target sites (??? I am not 100% sure is this so?). However, I know other people has used one-step lysis and obtained good IP enrichment for ChIP-PCR but I do not know is it ok to use it for ChIP-seq? or is it necessary to get rid of cytoplasmic contents and extract nuclei in order to do ChIP-seq? What type of lysis protocol do you use?

I have monitored the lysis under the microscope and it seems cells are still intact after incubation with lysis buffer (often grouped as large sheets probably due to have being collected by scrapping from the culture plate). Also the use of a dounze homogenizer did not improve this situation. I therefore tried the NEXSON protocol in which cells are gently sonicated (3 cycles of 15 sec ON; 30 sec OFF at low power) in the gentle lysis buffer. This treatment results in lysis of the cells as observed under the microscope (smaller structures which could be nuclei and lots of cell debris under the microscope). I was concerned about the risk of lysing nuclei with this method and therefore loosing part of the sample in the supernatant when pelleting nuclei but this doesn't seem to be a problem as I obtain good amounts of DNA to load in the agarose gel. Do you monitor cell lysis?

Also I do not know how much does the lysis affect the sonication. I mean that does sonication efficiency depend on whether cells have been effectively lysed or not ... because it seems many do not even monitor lysis.

I have been reading a lot about what other people has done to achieve a good sonication and at least one thing that I will use in my next test is to include a non-sonicated sample in the agarose gel. In this way you can see what sonication is actually doing to the samples and whether the poor fragmentation you see is due to degradation that takes place independently of the sonication (although in your case it seems clear that you are able to actually fragment the chromatin to some extent after 3 cycles of sonication when cells are fixed for 10 min (lane 5))

I have also reduced cross-linking time from 15 min to 10 min (15 min was indicated in the protocol from the Active Motif kit I am using) and I am also thinking to either reduce to 8 min or keep the time but reduce FA concentration from 1% to 0.75%. At least in your case the fixation time has a clear effect on sonication efficiency after 3 cycles (lanes 5 and 6)

Also I have learned that what you see in the gel depends a lot on how you process the samples. How do you de-crosslink, do you use proteinase K and RNase A, what amounts and for how long and of course how much DNA you load.

Good luck!

Hi Alejandra!

I'm sorry I didn't see and answer to your comments earlier! Thank you very much. In fact this experiment I was talking about I did it with a student of the lab. We did not have time to try again. I saw the same as you about isolation of the nuclei first. I think I will try this next time. And also we will try to fix 5 and 8 minutes too because I know now that too long fixation times in some way impairs sonication. The problem is that we are lysing retinas and we are trying to sonicate photoreceptors DNA which is known to have a very compact chromatin. So... But we will try again and if we got good results. I'll tell you!

Thank you again for all sharing with me your experience.

Juliette