The quality of the gels is below the limit. However, it looks for me that in both case you get similar size of sonicated DNA (mostly longer than 3kb). Old sample looks a little bit better just because there is a lot of RNA in it. Did you try to remove RNA from your samples?

Hi Fedor

Thanks for your response.

I am planning to do this again with RNase and protease treatment before running the sample on gel. I hope that helps.

Hi Fedor,

I have attached the results after treating the samples with RNase for 1hr at 370C followed by reverse crosslinking at 650C overnight. The gels show three distinct bands. What is your take on this result? should I sonicate longer? 

Hi Saptha,

Something very wrong with your gel. It looks like your samples have some "contamination" - high salt, high SDS... Did you purify your DNA or just directly load it on the gel after reverse crosslinking? You should purify them (for example by phenol-chloroform extraction followed by ethanol precipitation)

I didn't purify in the gel shown but I did perform a phenol-chloroform-isoamyl alcohol extraction of same samples and ran them on gel. I have attached the picture of the same, I find a smear still not very conclusive.

I would really appreciate if you could tell me the steps to be performed before running the sample on gel

Hi Saptha,

I think your problem is  that you didn't treat your samples with proteinase K after the RNAse treatment. So the steps before loading the DNA into the gel are: 1) RNAse 2) PTNase K 3) Revese crosslinking 4) DNA purification

Hope that helps, good luck!

Antonio.

Hi Antonio,

Thanks for the response. I did in the order you mentioned and did another sonication to optimize by using bioruptor. I used high setting and 30s On/30s Off for different times.

I have attached the result of the same . Ignore the not purified samples. Considering the purified one's which do you think is the best condition?

Hi Saptha,

I think the best would be somewhere between 20'-30', but as there is little correlation between sonication patterns and enrichment, you will only know the best conditions after doing the enrichment qPCR. So I suggest you now take the 20', 30' and 40' sonicated samples and proceed until the end, test enrichment by qPCR and choose the one with the best results.

Good luck!

Antonio.

 As an additional note, formaldehyde crosslinking is very sensitive to temperature changes around 25C. If your lab is 5 degrees warmer one day than another, you may over-xlink your DNA and never get it to shear down as small as a sample that was formaldehyde treated on a cooler day.

Helllo Saptha,

I am senior scientist at Diagenode in charge of fragmentation projects. The image showing "old" shearing looks like your chromatin was not de-cross-linked ( a typical distinct band). I prefer the new gel, where you can see some shearing but it is not optimal yet.

Could you please tell me how many cells do you sonicate per tube, what is the sample volume per sonication, what kind of tubes do you for sonication. Finally, what Br model do you use?

You can some tips regarding chromatin shearing at https://www.diagenode.com/files/protocols/The_Ultimate_Guide_for_Chromatin_Shearing_Optimization_with_Bioruptor_protocol.pdf

f

You can also concat me directly at [email protected]

Best regards,

Hi, Saptha, I think based on the standard steps which mentioned by Antonio, maybe your sample were over cross-link.Did you work with transcription factor or epigenetic modification protein? For me,if you work with a epigenetic modified proteion, it is better to decrease the fixation time to 5 minutes. Actually, I think if you want to get a desirable fragment size, the only way is longer the sonicate process. Any way, cool down the sonicate sample is very essential to get DNA at the end. For different specises,we chose different time for sonication. So you had better try to do fixation and sonication steps with a ladder and find the best condition before purify the DNA.

Hi Saptha,

Based on the results you show the main problem seems to be an analysis of shearing bur not the shearing itself. In the first image showing old and new results, "old" results have a typical pattern of cross-linked chromatin, like your reversal of cross-linking did not work al all. I would trust more to the smear you see for "new" results. The gel showing purified chromatin might correspond to small fraction of free DNA which was not efficiently cross-linked. I would recommend to carefully preformed decrosslinking  for at eats 4h at 65°C or longer in SDS-contaning buffer.  RNAse treatment before decross-linking is recommended. Samples have to be purified before loading on agarose gel. You could check the page 10-12 of the following guide to some more tips about chromatin shearing assessment https://www.diagenode.com/files/protocols/bioruptor-pico-chromatin-preparation-guide.pdf.

Please be aware also that efficient sonication using Bioruptor required using specific tubes. The ultrasound has to go through tube wall to reach samples, so the plastics itself, the shape of tubes etc have a significant impact on the sonication efficiency. More info on recommended tubes at https://www.diagenode.com/files/organigram/bioruptor-organigram-tubes.pdf.  Pay attention to a sample volume, you could find recommended range in Br manuals.

Please fell free to contact me at [email protected] if you need more assistance about Bioruptor.

Hi Irina, 

I use Bioruptor standard USD200 and this doesn't include the cooling system so I add ice manually to keep the sample cool. I replace with fresh ice every 10 min. I use 300ul sample in 1.5 ml tubes (as recommended in the manual)

I get pretty reproducible results using this Bioruptor.

Hi Antonio, 

As per your suggestion, I used 20', 30', 40' and also used 50', 60'. I found 20', 40' and 60' gives good enrichment for ChIP-qPCR whereas the other one's give enrichment but not so good.

Then I went ahead and the submitted the ChIP sample for QC before sequencing but Unofrtunately, the ChIP performed sample was not enriched in the desired size region.

We get totally different size distribution for the I/p by gel electrophoresis (mostly in the range of 400-100bp)  and the ChIP sample by fragment analyzer (mostly in the 2Kb range). Obviously we are again stuck and not able to proceed with sequencing.

I tried loading a 2 fold serially diluted Input DNA (after sonication,purification and crosslinking). The result of the same is attached here. I find them to be a big smear and low conc. and suddenly it shows 300-100bp size. Is my interpretation on the size of the fragment right? 

I have also attached the QC results for 40, 50 and 60 min sonication. (these are samples after IP)

Any idea why do we see this differential fragment size and how do we fix this issue? 

Hi Li,

I am working with transcription factors and my protein doesn't directly interact with DNA. It requires other proteins help to bind to DNA. So I can't reduce the fixation time any further and this is the optimum minimal time.  

Hi, Saptha, 

For me, the differential fragment size of DNA maybe comes from the tube during shearing. Some kinds of protocol suggests that you should spin down the tube for every 5 mins in case of big size. And the DNA low binding tube may gives us a good results. Perhaps, if your fixation time were determined, maybe you could try to increase the shearing cycles or derease the initiate tissues. 

Good luck!

ym

Hello everyone ! This post was very helpful !

We are struggling with JLat cells sonication (jurkat cells that harbor viral minigenome), I'm very new at this and after a talk with another team that  used to sonicate JLat cells only 7 cycles using bioruptor I tried it. But when I analyzed my samples on a 1,2% gel I found this: (image)

1, 2 and 1' comes from the same cell sample.

1' : reverse crosslink without NaCl

1 and 2: this time I added NaCl I extracted the DNA using two almost similar methods and 2 is better : "pure DNA"

Now some people do not quantify the DNA quantity they load on the gel, but what I see here is that quantity completely alters the migration result ... I find myself completely unable to say if my sonication did work or not and the real size of fragments?! 

Thank you and good luck :)

Hi Ghina,

I am still struggling with this problem. I tried using different concentrations of SDS in my sonication buffer and that seemed to help for the first time but unfortunately I am not successful in reproducing that result yet. It is pretty frustrating so far!!

Hi Saptha,

I finally solved my issues it was the reversed cross-link, the DNA extraction quality and also the amount of EtBr in my gel. I now  treat 1h with Rnase A at 37°C, reverse the cross-link ON 65°C with 0,2M NaCl (NaCl really changed the profile! ) then treat 1h with PK at 45°C, I am using ActivMotif kit to extract DNA for qPCR (with the Nanodrop I can see i have no phenol/Chlo contamination my DNA is clean)  and now i can see my samples even with only 500ng on my gel also I now add more EtBr on my gel (10µL for like 150mL) . My fragments are about 200-400bp I use JLat cells (jurkat) I sonicate 10M cells (Diagenode kit medium SDS)  , 7 cycles  and I only fix 10'.  Analysing the fragments size seems more complicated than the ChIp itself !! Hope this can help you out  ! https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2077865/pdf/1746-4811-3-11.pdf      I think more than 10 -15 cycles is way too much for  jurkat cells probably like me ,you also have an issue after sonication  : reverse cross link dna isolation , gel migration?! try to run sonicated and non sonicated samples on your gel +/- NaCl +/- PK and try a different DNA extraction technique 1µg-500ng should be ok try not to load too much on your gel ! Good luck!!

Hi Ghina,

I am happy that you got things working for you. Thanks for your inputs those are variations which I haven't tried so far. I will definitely give them a try and update with what happens.

I had few questions: my goal is pull down transcription factors which are not directly bound to DNA and so I have to fix them for 15' though 10' give good sonication result but I can't see the desired enrichment through qPCR. Hence I use 15' but I tried 10' long time back and I have changed a lot ever since so probably I should give that another try. What are your target proteins? 

what do you mean by 7cycles? 30'On/30'off for 7' or 7 repetitions of  10' cycles?

I fixed the issue with my DNA isolation, quality and gel, they are good now. My biggest problem is getting the right fragment size after sonication. I do get a nice fragments around 300-100 bp but unfortunately there is always a little bit of 3-2 kb fragment, which is just too stubborn to be broken down. After performing ChIP, even this 3-2 kb fragment gets enriched.

May be the cross linking with NaCl might help I will give that also a try.

Thanks again for your inputs

Hi Saptha !

As I said in my first comment I am very new at that ChIp thing, all I know I get it from literature commercial protocols home made peotocols and through research gate questions haha!

My target proteins are histones that is why I can fix my cells only 10min from what I read when trying to solve my issues when you work on transcription factors or factors that do not directly bind on your dna you have to fix longer .. just as you do.  I really don't know if this will impact sonication drasticallyI think you can find an answer for that in diagenode protocol at the very end of the booklet.

I sonnicate mine 7cycles 30sOn/30sOff I tried 9 don't see big difference and 10 is too much for me 100-300bp

NaCl really helps with reversing cross-link and if you fixed you cells 15min you will defently need NaCl .. I read some people needed to decross link for almost 24h when they fixed too long (the internet is full of ChIp-desperate students like us haha).

About the 3kb fragment have you checked for contamination in your cells? Do you use salmon sperm DNA somewhere in your protocol for instance as a carrier? Or to avoid non specific binding during you chIP?

Also ising the extraction kit helps gain time and procede to several tests. If I can give you an advice if you want to solve your issue : fresh start put one standard protocol (you can find a paper that focus on transcription factors and check their protocol) than start changing one parameter at a time! Also compare the different old protocols you used to gain time. Good Luck be patient and stubborn like your 3kb fragment!! ;)

Hi Ghina, 

Thanks for the NaCl tip. That does wonder to the reverse crosslinking process. Now I am able to get the desired fragment length but now my ChIP-qPCR is not working. Would you mind sharing your entire ChIP protocol with me? 

Hi Saptha,

I'm very happy this could help you... since our last discussion i found out I had to use the proteinase K  before the ON decrosslink to get rid of it ....in fact residues of proteinase K inhibit your qPCR....may be this is what you have ? The ON decrosslink must be the last step of the "decrosslink process".

I also use the extraction kit of activmotif with column to have clean samples.. any residue could inhibit your qpcr...  hope this will help you I'm almost sure this is the issue... do you have like very very high AE?

Ghina

May be proteinase treatment is the issue. Earlier, I used to do RNase and proteinase treatment before ON reverse crosslink and recently I changed it to ON crosslink followed by RNase and protease treatment (I might have seen this in some literature). 

What is AE? 

AE is for Amplification efficiency ( that you have to evaluate for each primer couple) when you have residues of proteinase K or any kind of proteins, phenols etc you will have high AE which is a sign of a pcr inhibition.  I had this issue because when you treat with proteinase K after the ON decrosslink this doesn't affect the gel migration (fragment analysis you have the good size etc) but it does affect the qpcr.. that is why in the diagenode protocole they keep the ON decrosslink for last so that you get rid of all the enzymes you used before. Hope this little change will fixe the issue :)

Ghina

Ok the efficiency of the qPCR reaction for test genes have a range of 115-160% whereas the house keeping gene falls in the range of 90-100% 

well 115% is not perfect but ok I guess.. but if it's higher for your target genes you can play with the temperature (even though it has to do with the Tm of each couple you can use higher temperature like 60°C to reduce non specific binding if it doesn't work i think you have to change your primers) sometimes changing the primers quantity can work too .. I had like way more than 160%  amplification when my qPCR was inhibited by the pk residues ... Once I left the ON decrosslink for last  and thn amplified at 60°C instead of 55° C my  efficiency dropped to 100-85%  and I had only one amplification and a good curve.