Skip to content

Multiplexed MeDIP-seq protocol using Illumina’s TruSeq Adapters

December 31, 2011

After looking at some MeDIP-seq protocols there were two issues that concerned me. The first is that all the protocols I have been able to find use the old Illumina adapters. I find that the Y-shaped TruSeq adapters migrate much more slowly through agarose gels than dsDNA, so I think it is better to convert them to dsDNA before size selection on an agarose gel. Thus, I modified existing protocols and moved the size selection to after the immunoprecipitation step so I could do a couple of cycles of PCR before the size selection.

The second issue that occurred to me is that the ligation of the adapters to the DNA molecules could result in base pairing between the adapters, which would result in DNA molecules that are being immunoprecipitated via their methyl-cytosine pulling down other DNA molecules via base pairing between the adapters. To alleviate this theoretical issue, I add a short oligo in excess to the immunoprecipitation reaction that is complementary to where the adapters anneal to each other. This should prevent any base pairing between the adapters during the immunoprecipitaiton. As a bonus, I figure it couldn’t hurt to have some non-specific DNA competitor in the immunoprecipitation reaction, which is easily removed afterwards by purifying the immunoprecipitation with Ampure XP beads.

I would also like to note that the Covaris S2 machine is fantastic for shearing. The sample to sample reproducibility and size distribution is perfect. Also because the excellent temperature control of the system, GC-biases should be minimized.

I’ve performed the protocol with HeLa cells and did a qPCR after the immunoprecipitation step and the results are below. I continued with the library prep and got a nice looking library with 8 cycles of PCR, i.e. an appropriately sized smear and no adapter dimer. Now if Beckman Coulter Genomics would be kind enough to send me the Ampure XP beads I ordered some time ago, I’ll prep up some real samples and send them out.

The protocol is here:

One other thing I forgot to mention is that LifeTech sells a kit called the ‘MethylMiner Methylated DNA Enrichment Kit’ which they claim works better than antibody based enrichment. It uses a biotinylated MBD2 protein and streptavidin Dynabeads to pull down methylated DNA. Millipore and Diagenode sell something similar but as a GST-fusion. This scheme, called MBD-seq, has the additional benefit of working on dsDNA so you do not have to denature the DNA for the enrichment step and avoid issues associated with having single stranded adapters on every DNA molecule. I would be interested to see how these two methodologies perform in a side by side test from a less biased source. The kits are pretty reasonable priced and comes out to $20/rxn in US pricing. If you are on a budget, I bet it would be really simple to express GST-MBD in E. coli and have enough protein for thousands of experiments.

Update: A couple of pubs comparing methlyation sequencing methods:

Comparison of sequencing-based methods to profile DNA methylation and identification of monoallelic epigenetic modifications

Genome-wide mapping of DNA methylation: a quantitative technology comparison


From → Posts

  1. Sushma permalink

    Hello Ethan,

    Your blog is Fantastic! Thank you for sharing your experience with MeDIP protocol optimization.
    I have a question regarding the competitor oligo that you add to prevent base-pairing between the adapters. The IP with 5-meC antibody is done with denatured single-stranded DNA (denatured at 98 degrees, and then immediately cooled). Would that not be enough to prevent re-annealing between complementary adapter sequences too? Did you see a lot of base-pairing between adapters after the initial denaturation? I’m only asking because I want to know if I NEED to add the oligo as well in my meDIP.

    Thank you.


    • ethanomics permalink

      Hi Sushma,
      Thanks for the compliments, I haven’t been doing much lab work the last months so not much new material on the blog. But as far as MeDIP-seq, I don’t know if the ‘blocking oligo’ actually helps. I’ve been meaning to try a MeDIP-seq without it but never really got around to it. Seems like it could be a quick Biotechniques paper to do it side-by-side and see if it actually does help (anybody out there want to do this???). The oligo is short and thus very cheap so I see no reason not to use it. I am getting great enrichment with my protocol. There is clearly, in theory, a way the fragments can anneal to each other by the adapters, so I’d say it’s a good idea. My guess is that it reduces background but as I wrote I have not checked. I have not seen any other MeDIP-seq protocols where they use a blocking oligo so it clearly works without it, but maybe not as well.

      I think I read somewhere that one of the exome capture protocols use a blocking oligo for the same reasons but I can’t remember where I read that. So I’m not the only person coming to the same conclusion.

      Hope that helps.


      • Sushma permalink

        Great, thanks Ethan. I will try the protocol without the oligo and let you know how it works. Keep up with your blog posts whenever possible and update us on your latest research and other exploits!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: