Announcement

Collapse

Welcome to the New Seqanswers!

Welcome to the new Seqanswers! We'd love your feedback, please post any you have to this topic: New Seqanswers Feedback.
See more
See less

ChIP-Seq: Single-tube linear DNA amplification (LinDA) for robust ChIP-seq.

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • ChIP-Seq: Single-tube linear DNA amplification (LinDA) for robust ChIP-seq.

    Syndicated from PubMed RSS Feeds

    Single-tube linear DNA amplification (LinDA) for robust ChIP-seq.

    Nat Methods. 2011 Jun 5;

    Authors: Shankaranarayanan P, Mendoza-Parra MA, Walia M, Wang L, Li N, Trindade LM, Gronemeyer H

    Genome-wide profiling of transcription factors based on massive parallel sequencing of immunoprecipitated chromatin (ChIP-seq) requires nanogram amounts of DNA. Here we describe a high-fidelity, single-tube linear DNA amplification method (LinDA) for ChIP-seq and reChIP-seq with picogram DNA amounts obtained from a few thousand cells. This amplification technology will facilitate global analyses of transcription-factor binding and chromatin with very small cell populations, such as stem or cancer-initiating cells.

    PMID: 21642965 [PubMed - as supplied by publisher]



    More...

  • #2
    Anyone have a complete protocol for this method ?

    Comment


    • #3
      ChIP and RIP protocol

      Can anyone give me a protocol for ChIP and RIP assays...

      Comment


      • #4
        linear amplification

        the detailed protocol is in Nature protocols
        Nat Protoc. 2012 Jan 26;7(2):328-38. doi: 10.1038/nprot.2011.447.
        Single-tube linear DNA amplification for genome-wide studies using a few thousand cells.
        Shankaranarayanan P, Mendoza-Parra MA, van Gool W, Trindade LM, Gronemeyer H.

        Comment


        • #5
          Has anyone not in Hinrich Gronemeyer's lab gotten this process to work? I've been trying for well over a month to get it working. After a tremendous amount of troubleshooting I've made a few discoveries.

          First, an aside: you can't buy the SAP they recommend (not sold anymore). The only one I can find still for sale is from USB. I assume a recombinant SAP cannot be used because they seem far pickier about the buffers used.

          I have great difficulty getting the T-tailing and T7-BpmI-oligoA(15) addition to work. I discovered this week I get far better efficiency when I skip the SAP reaction.

          Here's an example:


          Lane 2 and Lane 3 are identical, except that Lane 2 had the SAP reaction done. And yes, I did heat inactivate at 70 C for 10 min.

          Still, as you can see from the doublet in Lane 3, I'm not getting full linker addition. It looks about 50% efficient. The authors show a near 100% efficiency in a supplemental figure, so I obviously have some improving to do.

          I took two of the reactions (set up like Lane 3), one with 50 PICOgrams and the other with 50 NANOgrams of the PCR product and proceeded forward with the in vitro transcription.

          From the 50 ng input, I got a pretty good RNA yield (assayed by NanoDrop) - 580 ng/ul in 30 ul elution. For the 50 pg input I got a terrible to non-existant yield - 3 ng/ul in 30 ul elution. I haven't had the opportunity to confirm these yields on a BioAnalyzer. I should be able to do that next week. From experience, I am quite confident that 3ng/ul is really nothing.

          I'm assuming my yield from the 50 pg input is so poor because of the inefficient T7 oligo addition.

          Sorry for all of the rambling. Does anyone have any tips for me?

          I'm using aliquoted T-mix, dNTPs, and T7-BpmI-oligoA(15), to reduce freeze/thaw degradation. The T7-BpmI-oligoA(15) was synthesized by IDT DNA and PAGE purified. All of the reagents (except the SAP, are the same brand/type as recommended in the protocol). I'm following the Nature Methods papers exactly, except that I'm adding 2 ul of 2.5 mM CoCl2 (the protocol calls for adding 1 ul of 5 mM CoCl2).

          Comment


          • #6
            the detailed protocol is in nature protocols where the exact reaction volumes are given.
            1. Unfortunately Promega seems to have discontinued SAP, however, the thermostable phosphatase seems to be a good alternative.
            2. In your gel you mention that NO dNTP was added for the reactions in
            lanes 1 2, 3. Is that right? If there is no dNTP added with the T7 oligo klenow would start chewing back the strand.
            3. Perhaps there is something wrong with your SAP. The SAP step is expendable, ie. it is there only to improve the efficiency.
            4. Try your reaction with 1 ng of DNA, 50 ng maybe too much substrate and that is why you see only 50% efficiency.
            5. Do a Qubit HS for RNA quantitation, Qubit seems to be the most accurate and sensitive test for DNA/RNA. For 50 pg of substrate you should get around 10 to 20 ng of RNA (in 30 µl), ie. 0.3 to 0.6 ng / µl (nanodrop volume). this is far below the detection limit for nanodrop or for that matter bioanalyser.
            6. so try Qubit, and elute in 20 µl (imp for the next step)

            Comment


            • #7
              1. Thanks, I'll give it a try.
              2. Sorry, I wasn't clear - if the Klenow was added, then 10 mM dNTPs were added. What I meant was that 100 uM dNTPs were added instead of 100 uM T-mix. As a control, of course, since they wouldn't be very useful for adding the T7-BpmI-OligoA(15)
              3. Perhaps.
              4. That's true - I was using 50ng because it is difficult to visualize less on a gel.
              5. We don't have a Qubit, but we do have BioAnalyzer Pico chips, which can measure in the hundreds of picograms of RNA.
              6. Yes, I was using the SS III kit with the 10x buffer (hence I could use 26ul of RNA instead of 22ul), since that is what I had on hand. I'll switch to the kit with the 5x buffer in the future.

              If you only expect 20ng of RNA total from 50pg of DNA, then I should expect to get only around 5-10ng of DNA in the end? Which would not be a challenging amount to build a library for next-gen sequencing. So I would need at around 200 pg of input into the system to comfortably generate enough material for next-gen sequencing?

              Originally posted by linampli View Post
              the detailed protocol is in nature protocols where the exact reaction volumes are given.
              1. Unfortunately Promega seems to have discontinued SAP, however, the thermostable phosphatase seems to be a good alternative.
              2. In your gel you mention that NO dNTP was added for the reactions in
              lanes 1 2, 3. Is that right? If there is no dNTP added with the T7 oligo klenow would start chewing back the strand.
              3. Perhaps there is something wrong with your SAP. The SAP step is expendable, ie. it is there only to improve the efficiency.
              4. Try your reaction with 1 ng of DNA, 50 ng maybe too much substrate and that is why you see only 50% efficiency.
              5. Do a Qubit HS for RNA quantitation, Qubit seems to be the most accurate and sensitive test for DNA/RNA. For 50 pg of substrate you should get around 10 to 20 ng of RNA (in 30 µl), ie. 0.3 to 0.6 ng / µl (nanodrop volume). this is far below the detection limit for nanodrop or for that matter bioanalyser.
              6. so try Qubit, and elute in 20 µl (imp for the next step)

              Comment


              • #8
                someone pointed out a mistake in the amplification protocol, in the reagent list it is given as 1mM of rATP, rGTP, rCTP, rUTP. It should be 100 mM each as is supplied by the kit, please do not adjust the concentration.

                Comment


                • #9
                  I am a PHD student from China
                  Similar to thedavid, I think I have the same problem in LinDA: very low efficiency of the‘T’ tailing or low efficiency of attaching T7 primer.
                  I choose another Shrimp alkaline phosphatase produced by Fermentas. I have succeeded in LinDA for only one time, when I add dTTP without ddCTP, and inactivated Klenow at 75° for 20min (start with 50ng DNA [input DNA, fragmented by Ultrasound] produces 17000ng RNA and finally 5000ng DNA). But when repeat the experiment, I failed again and again, with no RNA yield from 20-50ng DNA starting material and no ampification of ChIP-DNA confirmed by realtime PCR.
                  Working hard on it for over two months, I'm now can confirm that this protocol can work, but I don't understand why it is so hard to stabilize the system!
                  Has anyone not in Hinrich Gronemeyer's lab gotten this process to work?
                  Has anyone know why the efficiency of ‘T’ tailing/attaching T7 primer is low?
                  Thanks a lot!

                  Comment


                  • #10
                    LinDA

                    Hi, so I think I'm managing to get it to work. After amplifying 50pg I got back 48ng which is more than I expected. Unfortunately when I try to use this DNA for qPCR validation some primer sets give my much higher ct values than others. It looks as if there is some kind of a bias introduced but it doesn't seem to be dependent on GC. Any idea what this might be?
                    I'm also thinking about the kit I want to use to construct the libraries. Illumina's ChIP-seq seems to be laborious. Any thoughts about the new kits like NextFLEX ChIP seq or the NuGene low input library preparation kit (ready for 1ng of DNA)?

                    Comment


                    • #11
                      LinDA

                      Hi, I have seem the same phenomenon in the qPCR. But whether it's a bias or not, is under question. If the amount of gene B is higher in the starting material, it may amplify more after LinDA than A. I think you can try to compare the amplification ratio of one gene between specific antibody group and IgG control group.
                      for instance, gene A in specific antibody group is originally "10", then after LinDA is "100"; gene A in IgG control group is "1", then after LinDA is "9", the ratio of amplification ratio is 10/9. gene B in specific antibody group is originally "20", then after LinDA is "400"; gene B in IgG control group is "2", then after LinDA is "38", the amplification ratio is 20/19. If that so, I think it may not influence result of the ChIP-seq.
                      I'm glad to hear that another people have got the protocol to work. However, I am still have trouble with ‘T’ tailing and attaching T7 primer. I follows the protocol except that using another ThermoSensitive Alkaline Phosphatase(Fermentas) and inactivating at 75° 5 min according to the user manual. Can you share some experience of your LinDA?
                      thanks in advance!

                      Originally posted by epignome View Post
                      Hi, so I think I'm managing to get it to work. After amplifying 50pg I got back 48ng which is more than I expected. Unfortunately when I try to use this DNA for qPCR validation some primer sets give my much higher ct values than others. It looks as if there is some kind of a bias introduced but it doesn't seem to be dependent on GC. Any idea what this might be?
                      I'm also thinking about the kit I want to use to construct the libraries. Illumina's ChIP-seq seems to be laborious. Any thoughts about the new kits like NextFLEX ChIP seq or the NuGene low input library preparation kit (ready for 1ng of DNA)?

                      Comment


                      • #12
                        hi,
                        Did you purify the DNA after LinDA with Qiagen columns or SPRI beads? I found that the T7 poly A primer doesnt get completely removed with column purification, this leads to interference in qPCR. SPRI beads purification (eg. Ampure) manages to remove the primers efficiently and qPCRs improve significantly.

                        Originally posted by epignome View Post
                        Hi, so I think I'm managing to get it to work. After amplifying 50pg I got back 48ng which is more than I expected. Unfortunately when I try to use this DNA for qPCR validation some primer sets give my much higher ct values than others. It looks as if there is some kind of a bias introduced but it doesn't seem to be dependent on GC. Any idea what this might be?
                        I'm also thinking about the kit I want to use to construct the libraries. Illumina's ChIP-seq seems to be laborious. Any thoughts about the new kits like NextFLEX ChIP seq or the NuGene low input library preparation kit (ready for 1ng of DNA)?

                        Comment


                        • #13
                          hi,
                          to get back to the library prep, I am now finalising a procedure utilising LinDA for Illumina library prep. With the modified protocol one should get an Illumina library directly with LinDA, without using PCR cycles as is the case with NuGene. Basically I incorporate the Illumina primers in the RT and 2nd strand synthesis steps.

                          Originally posted by epignome View Post
                          Hi, so I think I'm managing to get it to work. After amplifying 50pg I got back 48ng which is more than I expected. Unfortunately when I try to use this DNA for qPCR validation some primer sets give my much higher ct values than others. It looks as if there is some kind of a bias introduced but it doesn't seem to be dependent on GC. Any idea what this might be?
                          I'm also thinking about the kit I want to use to construct the libraries. Illumina's ChIP-seq seems to be laborious. Any thoughts about the new kits like NextFLEX ChIP seq or the NuGene low input library preparation kit (ready for 1ng of DNA)?

                          Comment


                          • #14
                            Hi,

                            I am struggling with LinDA now, and thanks for all your helpful suggestions and ideas on the subject, especially from Linampli.

                            I used my ChIP pulled down chromatin 2ng(maybe nanodrop not acurate) to do LinDA and got specific amplification, total amount 180ng; and when I use 5opg of pulled down DNA for LinDA, I got low yield non-specific amplification.I think I should start with accurate 50pg of Input DNA to set up the experiment first.

                            Here are my questions:
                            1. 50pg of sonicated chromatin(100bp to 1kb) and 50pg of 250bp DNA fragment is different. Did someone get ideal amplification from 50pg of sonicated chromatin?

                            2. Several post above said they got lower efficiency of T-tailing and oligo T7-annealing. how to improve the efficiency, other than usage of TSAP, eg, increase reaction time, increase the enzyme de-activation temparature..or..?

                            Hope to get some help here. Thanks a lot.

                            Comment


                            • #15
                              Has anyone made any progress with this protocol? We can't yield higher than ~10 ng of RNA.

                              Comment

                              Working...
                              X