ChIP on chip amplification problems - (Jun/16/2009 )
Hi Everyone,
Thanks to all the people who post on this forum - it is always full of useful tips and troubleshooting. I now have a question to throw out there.
I am currently working doing ChIP on Chip - for both Histone modifications and Methyl binding proteins.
By means of background
- I am using primary cells
- 10x10^6 per IP with sonication optimised 100-600bp
- IgG as a negative control
- ChIP eluted with QiaQuick MinElute
- ChIP confirmed as successful with published PCR primers for enriched/not enriched regions, IgG OK.
- SIGMA WGA kits used to amplify ChIP DNA - all of IP DNA and 100ng (not 10ng as recommended by kit) input used for WGA2. Am expecting to hybridise to a Nimblegen HT12 Human Promoter array (2 colour array requires 4ug DNA per chip)
- WGA2 kit yields insufficient DNA - I have had to reamplify with the WGA3 kit using 15ng DNA. (This 2 step amplification has been published several times and seems to result in satisfactory representation/minimal bias etc) I am not doing the fragmentation step in the WGA2 as sonication has already fragmented the DNA. SIGMA NA1020 PCR cleanup kit used for cleaup after both WGA steps.
My problem lies in the amplification step. Initially, to check the quality of the WGA2 rxn I re did the PCR positive and negative controls with appropriate results for both WGA2 and WGA3.
I have moved onto my second batch of samples and unfortuantely when I have done the PCRs to check for enrichment on the WGA2 sample I get a smear (200-600bp) or no product . (expected product 260 or 270bp for these particular primers)
PCR was done with 250ng of input WGA2 DNA and 200ng -ve control ChIP DNA or 20ng of +ve control ChIP WGA 2 DNA at optimised cycling conditions. 1/5 PCR loaded onto 1.5% gel.
I guess the smear just represents the template WGA2 DNA. Perhaps 250ng input DNA is excessive for PCR template? The smear means that there is obviously plenty of DNA there, it just won't PCR up for me - does the WGA2 now not represent my initial sample? While I'm repeating the PCRs, any ideas why the WGA2 or the PCR might not have worked would be greatly appreciated
Orla
Hi Orla,
Unfortunately I've been experiencing exactly the same problems as you. I don't understand why I don't get a PCR product - unless the amplified DNA is rubbish. I've been told before that you can get a false positive result from your amplifiction as the random primers can ligate together and amplify themselves producing a product which is completely useless!! Great!
I'm glad that you've used 100ng of starting DNA for your amplifications too as I've tried 10ng and barely get anything back from this, yet my input DNA seems to amplify ok. I don't understand why this is as I purify my input and ChIP samples in exactly the same way so if there was a contaminant then it would inhibit the reaction in all samples. It's all so confusing! Anyway, I hope someone can provide us both with some interesting answers! Sorry I can't be more help. Good luck!
Vicky
Hi,
We are also using the WGA2 kit from Sigma. At first it was a bit tricky, but now it's all sorted We use primary cells too (that have been frozen) and use approx. 6 million cells per IP. After ChIP we elute in 30ul using a QIAGEN PCR purification kit. For the majority of our samples amplifying either 5.5ul or 11ul gives us at least 3ug of DNA.
In terms of your experiment - perhaps using all of your ChIP DNA is too much, and you're actually inhibiting the reaction. Also, if your making a master mix for the amplification step, make sure you vortex it before adding to each sample!
Let me know if you need anymore advice
Clare
orlatron on Jun 17 2009, 05:48 AM said:
Thanks to all the people who post on this forum - it is always full of useful tips and troubleshooting. I now have a question to throw out there.
I am currently working doing ChIP on Chip - for both Histone modifications and Methyl binding proteins.
By means of background
- I am using primary cells
- 10x10^6 per IP with sonication optimised 100-600bp
- IgG as a negative control
- ChIP eluted with QiaQuick MinElute
- ChIP confirmed as successful with published PCR primers for enriched/not enriched regions, IgG OK.
- SIGMA WGA kits used to amplify ChIP DNA - all of IP DNA and 100ng (not 10ng as recommended by kit) input used for WGA2. Am expecting to hybridise to a Nimblegen HT12 Human Promoter array (2 colour array requires 4ug DNA per chip)
- WGA2 kit yields insufficient DNA - I have had to reamplify with the WGA3 kit using 15ng DNA. (This 2 step amplification has been published several times and seems to result in satisfactory representation/minimal bias etc) I am not doing the fragmentation step in the WGA2 as sonication has already fragmented the DNA. SIGMA NA1020 PCR cleanup kit used for cleaup after both WGA steps.
My problem lies in the amplification step. Initially, to check the quality of the WGA2 rxn I re did the PCR positive and negative controls with appropriate results for both WGA2 and WGA3.
I have moved onto my second batch of samples and unfortuantely when I have done the PCRs to check for enrichment on the WGA2 sample I get a smear (200-600bp) or no product .

PCR was done with 250ng of input WGA2 DNA and 200ng -ve control ChIP DNA or 20ng of +ve control ChIP WGA 2 DNA at optimised cycling conditions. 1/5 PCR loaded onto 1.5% gel.
I guess the smear just represents the template WGA2 DNA. Perhaps 250ng input DNA is excessive for PCR template? The smear means that there is obviously plenty of DNA there, it just won't PCR up for me - does the WGA2 now not represent my initial sample? While I'm repeating the PCRs, any ideas why the WGA2 or the PCR might not have worked would be greatly appreciated

Orla
Hi Clare,
How much ChIP DNA do you pull down after ChIP? I seem to get loads of DNA back but my PCR works perfectly. I also do P/C extraction of the DNA, as with columns i seemed to lose alot of the DNA.
Cheers,
Vicky
Hi Vicky,
We never quantitate how much DNA we have after ChIP as levels are too low for the nanodrop. We prefer columns for purification because we usually have loads of samples and it makes life much easier!
Clare
kitwai83 on Jul 3 2009, 11:29 AM said:
How much ChIP DNA do you pull down after ChIP? I seem to get loads of DNA back but my PCR works perfectly. I also do P/C extraction of the DNA, as with columns i seemed to lose alot of the DNA.
Cheers,
Vicky
kitwai83 on Jun 25 2009, 04:22 AM said:
Unfortunately I've been experiencing exactly the same problems as you. I don't understand why I don't get a PCR product - unless the amplified DNA is rubbish. I've been told before that you can get a false positive result from your amplifiction as the random primers can ligate together and amplify themselves producing a product which is completely useless!! Great!
I'm glad that you've used 100ng of starting DNA for your amplifications too as I've tried 10ng and barely get anything back from this, yet my input DNA seems to amplify ok. I don't understand why this is as I purify my input and ChIP samples in exactly the same way so if there was a contaminant then it would inhibit the reaction in all samples. It's all so confusing! Anyway, I hope someone can provide us both with some interesting answers! Sorry I can't be more help. Good luck!
Vicky
Thanks so much for the replies. I have been trying so hard to trouble shoot this issue!!! SO frustrating.
The saga continues
- I have repeated the ChIP and WGA2 and find that my yield is now terrible following purification!! (0.5ug only from all of the pulled down DNA which nanodrop quantified at around 2ng/uL - gave appropriate positive PCR products in controls etc). I'm thinking that perhaps this is for two reasons - either there was ACTUALLY too much DNA in my ChIP elution which is inhibiting the WGA2 reaction, or perhaps my DNA is sonicated too small? I haven't checked fragment size since I first optimised sonication...
- I can now get a positive product from the WGA2 product that I referred to in my previous post by increasing cycle number to 40 and increasing amount of template in PCR. However, if I then WGA3 this sample, although the yield is now fantastic, I can't get a product for positive control at 40 cycles.
This isn't making me too confident about hybridising any time soon!
I'm going to repeat the ChIP again today, and check sonication size etc and try what Clare suggested with doing two WGA 2 reactions with smaller amounts of IP'd DNA and see what happens...
As an aside, I had intended on using Nimblegen for the array. I got to the final stage of finalising our agreement and then found out that the samples need to be sent to Iceland...unfortunately being in Australia the quote I have been given to get the samples to Iceland on ice is hideously expensive (I could almost drop them off in person more cheaply)...So I'm not even sure if I can use WGA products on Affy chips anyway!
Good luck and hope to hear that you are having some more success than I am!
Orla

Hi again
Just a thought - are you using P/C to purify after ChIP? It is possible that some carry over may be affecting the WGA (it is a VERY fussy procedure). I would repeat the Chip, purify using columns (elute in water) and see how you go.
Also, do you check that your sonication has worked after ChIP? We always run a quick sample on a gel after sonication to ensure it worked properly. WGA needs fragmented DNA to work...
We use agilent arrays and the WGA material works fine. Why don't you hyb the samples yourself?
Where in Australia are you? I did my PhD in Melbourne
Clare
Clare on Jul 15 2009, 04:54 AM said:

Just a thought - are you using P/C to purify after ChIP? It is possible that some carry over may be affecting the WGA (it is a VERY fussy procedure). I would repeat the Chip, purify using columns (elute in water) and see how you go.
Also, do you check that your sonication has worked after ChIP? We always run a quick sample on a gel after sonication to ensure it worked properly. WGA needs fragmented DNA to work...
We use agilent arrays and the WGA material works fine. Why don't you hyb the samples yourself?
Where in Australia are you? I did my PhD in Melbourne

Clare
Hi Clare,
I am using Qiagen Min Elute columns (which I do think have a bit of ETOH carry over - when you load a few uL of elution + loading buffer into a gel it will sometimes evaporate out of the well) to purify the ChIP DNA...I sonicated some cells again and took aliqouts out every few cycles and I'm defintely sonicating the chromatin (average size about 400bp, range 200-1000)..I'm currently repeating the ChIP with new samples and dbl checking everything along the way and will hopefully know by late tonight if WGA2 worked. (fingers crossed!) I've had another thought about WGA2 failure - I don't think I realised how technique sensitive it is, we had a similar problem a few months ago with a cRNA prep for a microarray, there seemed to be a bit of voodoo involved (no more than 4 samples at a time, need to sit at a particular bench etc). I did do 8 samples at once for the batch of WGA2 which failed...
Do you use Dnase1 to fragment DNA prior to hybridisation to the chip? Affymetrix protocols want dUTP incorporated into amplification and use UDG/APE 1 to fragment (although some papers have used Dnase1)...We could hyb here.....but the AGRF (I'm in Brisbane

Thanks for your advice...I'll post once I know how things are going

Orla
Hi again
No, we don't fragment our DNA any further before hybridisation. We just label our WGA DNA with the Invitrogen total genomic labeling kit and away we go
Good luck with your latest ChIP - let us know how you get on with the WGA this time.
Clare
Hi Clare,
I am using Qiagen Min Elute columns (which I do think have a bit of ETOH carry over - when you load a few uL of elution + loading buffer into a gel it will sometimes evaporate out of the well) to purify the ChIP DNA...I sonicated some cells again and took aliqouts out every few cycles and I'm defintely sonicating the chromatin (average size about 400bp, range 200-1000)..I'm currently repeating the ChIP with new samples and dbl checking everything along the way and will hopefully know by late tonight if WGA2 worked. (fingers crossed!) I've had another thought about WGA2 failure - I don't think I realised how technique sensitive it is, we had a similar problem a few months ago with a cRNA prep for a microarray, there seemed to be a bit of voodoo involved (no more than 4 samples at a time, need to sit at a particular bench etc). I did do 8 samples at once for the batch of WGA2 which failed...
Do you use Dnase1 to fragment DNA prior to hybridisation to the chip? Affymetrix protocols want dUTP incorporated into amplification and use UDG/APE 1 to fragment (although some papers have used Dnase1)...We could hyb here.....but the AGRF (I'm in Brisbane ) will do it for us pretty easily and that sounds good to me!!!
Thanks for your advice...I'll post once I know how things are going
Orla