May 1, 2016 (Vol. 36, No. 9)
Brian Egan Ph.D. Director of Genomic Applications and Services Active Motif
Chromatin immunoprecipitation (ChIP) is conceptually a simple method. Briefly, formaldehyde is used to crosslink DNA binding proteins to DNA. Sonication is then used to shear the DNA into small fragments, followed by immunoprecipitation to pull down a protein of interest along with its associated genomic DNA. Crosslinks are then reversed, proteins removed and DNA isolated and analyzed by qPCR or sequenced (ChIP-Seq) to reveal the genome-wide binding pattern of any protein of interest. Today, Chip is undoubtedly the most widely used method for the identification of loci specific protein/DNA interactions.
Described originally in 1984, the ChIP technique has evolved from a method to confirm protein interactions at specific promoters to a way to detect all genomic binding sites of a transcription factor in a single experiment (ChIP-chip). Today ChIP-Seq has enabled an even more comprehensive readout that may even include chromatin bar-coding to enable the analysis of multiple protein/DNA interaction profiles in a single reaction.
As a technique, ChIP has had a tremendous impact in the scientific community. One could argue that without it the field of epigenetics might still be in its infancy today. ChIP has made it possible for the ENCODE consortium to generate binding maps of hundreds of transcription factors and dozens of histone modifications across multiple cell types. This comprehensive dataset is used as a reference for hundreds of researchers and has helped turn chromatin biology and epigenetics into more mainstream disciplines that are now routinely considered in other fields such as neurobiology, development, and oncology.
Current Limitations of ChIP-Seq in Disease Research
Chromatin immunoprecipitation is not only used as a tool to understand the basic concepts of chromatin biology, epigenetics, and gene regulation, but now ChIP is also used to explore protein/DNA interactions in relation to disease. Consequently, it is critically important to be able to directly compare ChIP-Seq samples (disease vs normal or treated vs untreated). Herein lies the problem. ChIP-Seq has never been considered a quantitative assay. This is mainly a consequence of picogram to low nanogram amounts of DNA that are isolated from ChIP reactions resulting in the need to PCR amplify the material prior to sequencing. Presumably, this step has the most significant impact on quantitation but other technical steps such as sonication, immunoprecipitation, and DNA purification are also subject to technical variation. These challenges make it difficult to reliably identify meaningful differences in ChIP-Seq datasets.
A Novel ChIP-Seq Spike-in Strategy
Many labs have tried bioinformatics approaches to appropriately normalize ChIP-Seq data, however none of these programs or algorithms have made a significant impact on ChIP-Seq differential analysis and none of them have been widely adopted by the scientific community. Active Motif has recognized this issue and has taken a different approach by not just tackling the problem from a bioinformatics angle, but from a combined technical and bioinformatics approach.
Taking a cue from the RNA-Seq world, where synthetic RNA spike-in is used as a way to monitor technical variation, Active Motif has developed a ChIP-Seq spike-in technique and accompanying bioinformatic normalization. In short, prior to immunoprecipitation, small amounts of Drosophila melanogaster chromatin are mixed with chromatin from human, mouse, or other sample of interest. This chromatin mixture is immunoprecipitated using an antibody that recognizes the target protein and a second antibody that specifically recognizes a Drosophila DNA binding protein. An antibody against the Drosophila specific histone variant H2A was chosen for this purpose. The combination of the Drosophila chromatin and Drosophila-specific antibody provides a mechanism to consistently pull out Drosophila chromatin as a minor fraction of the ChIP DNA.
Drosophila chromatin is ideal since 1) it is structurally similar to mammalian chromatin with conservation of most major chromatin associated proteins thus minimizing bias in the immunoprecipitation 2) the genome is small so minimal amounts can be added relative to the test chromatin and 3) the genome is sufficiently different from mammalian genomes to allow for unique genomic mapping of DNA sequences.
Following immunoprecipitation the ChIP-DNA is sequenced and mapped to both the test genome and the Drosophila genome. The sequence tag counts that align to the Drosophila genome are used as a reference to normalize the sequence tag counts from the test genome. The spike-in approach normalizes for any variation that was introduced during sample processing and makes it possible to accurately compare samples. This has been demonstrated in experiments at Active Motif using cells treated with a compound that blocks the activity of the histone methyltransferase EZH2, which trimethylates lysine 27 of histone H3. Treatment of cells with EZH2 inhibitors result in global reductions of H3 lysine 27 trimethylation (H3K27me3). However, traditional H3K27me3 ChIP-Seq approaches have not been able to detect these differences. Introduction of Active Motif’s ChIP-Seq spike-in reagents into this model reveals the expected decrease in H3K27me3 ChIP-Seq signal.
The need to perform accurate sample-to-sample comparisons of ChIP-Seq data sets is apparent since multiple approaches have now been described in the literature. The ChIP-Seq spike in approach was first described in 2014 researchers spiked mouse chromatin into human chromatin samples and relied on antibody cross reactivity to pull out the mouse spike-in chromatin. This approach was not ideal given the issues related to cross-mapping of mouse sequence to the human reference genome and the method was improved when another group spiked Drosophila chromatin into human ChIP-Seq reactions. This approach does alleviate mapping issues but, it is limited to only experiments exploring histone modifications since it still relies on antibody cross reactivity to pull out the spiked in chromatin.
A Universal Solution
As chromatin immunoprecipitation methodologies have continued to evolve new challenges have arisen and with new challenges, new solutions. The key to the solution developed at Active Motif is the use of the second Drosophila specific antibody. This technique is proven to work with histone modifications as described above and also applicable to all other DNA binding proteins since it does not rely on antibody cross reactivity. The addition of spike-in to the ChIP-Seq repertoire is another step forward in enabling researchers from diverse fields to perform sample-to-sample comparisons of ChIP-Seq data.