by Jeffrey M. Perkel
Suppose you identify a transcription factor that you believe regulates some gene of interest. It’s easy enough to demonstrate this protein is capable of binding the gene’s promoter or operator sequences in vitro—a simple gel-shift assay will do the trick. But how do you prove they interact in vivo?
For that, researchers often turn to ChIP, or chromatin immunoprecipitation. In ChIP, cellular chromatin is crosslinked, freezing protein-DNA interactions in vivo. The chromatin is then isolated, sheared into fragments a few hundred bases in length and subjected to immunoprecipitation using antibodies against the specific transcription factor or histone modification of interest. Finally, the crosslink is reversed and the enriched DNA analyzed, using either endpoint PCR, quantitative PCR (qPCR), microarrays (called “ChIP-on-chip”) or DNA sequencing (“ChIP-Seq”).
Chromatin immunoprecipitation has become a staple of the epigenetics community, which uses it (or variants thereof) to study interactions between DNA and such proteins as sequence-specific transcription factors (e.g., NF-kappaB and Oct4), general transcription factors (e.g., RNA polymerase II) and modified histones (e.g., histone H3-dimethyl-Lys4). Fortunately, a wide variety of ChIP kits and associated reagents (such as validated antibodies and controls) are now commercially available, freeing researchers from much of the troubleshooting the assay originally required and compressing assay time considerably.
Anyone can do it, says Chris Wasden, marketing manager at Active Motif. The key is to do your homework. “It’s not difficult, but [as] with any experiment there are many details you need to get correct,” he says. So study the protocols, talk to colleagues and consult with technical support teams before diving in. “Plan ahead, tap knowledge that exists and you’ll get much, much better results,” he says.
Chromatin preparation
By all accounts, the ChIP protocol contains two key steps: chromatin preparation and immunoprecipitation. The antibodies used for the IP are key, says Michael Sturges, senior product manager for epigenetics at EMD Millipore—after all, they provide the assay with its sensitivity and specificity—but chromatin is the substrate on which they work. If you don’t have good chromatin, he says, “It doesn’t matter how good your antibody is.”
There are three components to preparing high-quality chromatin: crosslinking, purification and fragmentation. Crosslinking is typically accomplished using formaldehyde; the protocol for EMD Millipore’s Magna ChIP™ G kit, for instance, crosslinks with formaldehyde mixed in culture media (final concentration, 1%) for 10 minutes. However, the troubleshooting section of the kit’s manual recommends optimizing this step, either by using a fixed amount of formaldehyde for varying lengths of time or by varying its concentration. (Not all proteins need to be crosslinked; according to the Magna ChIP manual, histones may not need to be crosslinked because they are tightly associated with DNA.)
Following crosslinking, the chromatin is isolated. “There is a little technique involved there, because you want the chromatin without having other cellular components,” says Rizwan Farooqui, market segment manager for Thermo Scientific protein function products. Yet at the same time, he says, you cannot use conditions that are so harsh they disrupt protein-DNA interactions.
In the Thermo Scientific Pierce® Agarose ChIP kit, chromatin isolation is accomplished by first gently lysing the cell membrane and isolating the nuclei by centrifugation and then resuspending the nuclei in a buffer containing micrococcal nuclease to fragment the DNA.
Micrococcal nuclease is one of two basic approaches that can be used to fragment the chromatin; the other method is sonication. Nucleases may exhibit some sequence bias in their digestion patterns. But, says Sturges, sonication can be difficult to control because of, for instance, heating and frothing of the sample. Plus, not everyone has access to a sonicator. By contrast, says Farooqui, nuclease digestion reduces time and increases reproducibility. “We can control what size chromatin fragments we get after digestion. It also makes it easier to optimize, because you can set up titrations to get to the ‘sweet spot’ [of digestion].”
The generally desired range for ChIP fragmentation is between 200 and 1,000 bp, which should be monitored by agarose gel electrophoresis. Generally, any kit will accommodate both sonication and nuclease, but users may need to purchase the digestion enzymes separately. Active Motif’s ChIP-IT® Express Enzymatic Kit, for instance, is preconfigured for nuclease digestion, whereas EMD Millipore’s Magna ChIP kits are not; instead, EMD Millipore offers a separate EZ-Zyme™ Chromatin Prep Kit containing a “proprietary enzymatic cocktail” for chromatin digestion.
For those who don’t wish to go through all that optimization, prepared chromatin is commercially available for a handful of widely used cell lines. Active Motif, for instance, sells four flavors of pre-sheared, “Ready-to-ChIP Chromatin” (for HeLa, HepG2, K-562 and NIH/3T3 cells).
Antibodies and immunoprecipitation
The next step after chromatin purification is immunoprecipitation. That, of course, requires an antibody —but not all antibodies are equal. Some antibodies, for instance, might recognize a particular histone modification, but only in the context of a peptide and not in the intact molecule. Alternatively, the antibody may not be specific enough—recognizing not only a trimethyl-modified histone, for instance, but also the dimethylated and monomethylated forms. Or the antibody’s epitope may be hidden in the protein’s DNA-bound form. In any case, just because an antibody works well in Westerns or even in immunoprecipitation experiments doesn’t mean it will work in ChIP.
“Sourcing good antibodies is definitely one of the hardest steps,” says Farooqui.
Look for antibodies that are classified as “ChIP-validated.” Thermo Scientific Pierce has more than 30 antibodies in its collection that are classified as such, and Cell Signaling Technology (which also offers ChIP kits) offers more than 100. Active Motif’s portfolio includes 137 ChIP-validated antibodies, and EMD Millipore has nearly 130—including 61 “ChIPAb+” products, each of which is validated for ChIP not just once, but for every lot, Sturges says. (ChIPAb+ antibody kits also include PCR control primers against a known positive locus as well as a negative control IgG antibody.)
For researchers who may be using an antibody that has not yet been ChIP-validated, Sturges recommends first calling the manufacturer to see if the antibody has ever been tested for that application. “It’s possible that maybe the antibody was tested in ChIP and it doesn’t work, and that’s why it’s not [indicated] on the [product documentation] sheet,” he says.
If it hasn’t been tested, Sturges recommends testing it first in Western blots (does the antibody produce a single band?) and immunoprecipitation experiments. Next, test the antibody in dot blots against a range of related peptides (for instance, differently modified forms of the same peptide) to test specificity. If everything looks good, you can move on to ChIP. But be sure you have a good assay, he adds—you need to know a sequence that it absolutely does or does not bind to, a positive and negative control for your ChIP.
If you happen to be working with a novel protein for which no antibody is available, you can either make your own—a process that can take months—or try Promega’s HaloCHIP™ approach. Based on the company’s HaloTag® platform, HaloCHIP removes the need for antibodies by tagging the protein of interest with a protein tag (called HaloTag) instead.
The name HaloCHIP, explains Chad Brueck, global product manager for proteomics at Promega, is “a misnomer.” He explains, “We chose to include the word ChIP in the product name to better connect customers with the application of this product.” ChIP implies the use of an antibody; instead, this assay is more like a classical pull-down. In the presence of a chloroalkane substrate, HaloTag covalently couples to the substrate, which may be linked to a fluorescent ligand, or in the case of HaloCHIP, a Sepharose® bead. Researchers simply make a fusion-clone of their protein with HaloTag. (Or they can buy one; according to Brueck, the company has more than 1,000 transcription factor and “epigenetics-related genes” already cloned in its HaloTag Human ORF collection.) They then express that clone in cells, crosslink protein to DNA, shear it and capture with HaloLink resin—it’s the same basic process as ChIP, but without an antibody.
“You don’t need to search or screen for antibodies, you can use exactly the same process regardless of the gene you’re studying,” Brueck says.
Actually, all ChIP variants use some form of solid resin to pull-down the ChIP’d DNA. Some kits use magnetic beads, others do not. HaloCHIP uses non-magnetic beads (meaning it requires a spin-column or other filtration step to collect the beads), but Active Motif’s ChIP-IT Express and Millipore’s Magna ChIP kits use magnetic beads and a magnet instead. If you are actually doing ChIP (as opposed to HaloCHIP), you’ll need a bead capable of capturing the antibody; that means one coupled to either protein A, protein G or both. (Thermo offers beads containing both epitopes, as will Millipore, soon).
The final steps of the IP procedure clean up the sample. Most kits include sample clean-up reagents (such as buffers and a spin column), but Zymo Research sells a dedicated kit that can be used throughout the process. The company’s ChIP DNA Clean & Concentrator™ kit, says senior research associate Onyi Chima, was designed to handle the high levels of detergent, enzymes and antibodies present at various stages of the ChIP procedure.
Purification, says Chima, is “a really important step.” “If you don’t have a kit, ChIP can take days,” she says. “I don’t really think it’s worth it to lose your sample at the end of it.”
ChIP variants
Standard ChIP has been around for more than a decade, but variants have arisen. Active Motif, for instance, offers an RNA ChIP-IT kit for capturing chromatin-associated RNA-protein interactions. Another Active Motif offering is the Re-ChIP-IT kit, which enables researchers to carry out multiple rounds of ChIP to determine, for instance, if two separate proteins are present at a specific region.
“That allows people to look at complexes, cofactors, transcription factors and histone modifications,” says Wasden. “You can do a lot of things by performing subsequent ChIPs.”
Other ChIP variants allow the isolation of methylated DNA—a process sometimes called MeDIP (methyl-DNA immunoprecipitation). EMD Millipore’s CpG MethylQuest DNA Isolation Kit, for instance, uses a methyl-DNA binding protein to pull-down methylated DNA sequences.
Downstream analysis
The vast majority of users, says Farooqui, use qPCR to probe their ChIP’d DNA. The approach is fast, readily available and quantitative. But it is only useful for a relatively small number of loci. To get a genome-wide perspective, researchers can use either microarrays or next-generation sequencing.
Any ChIP’d material can technically be applied to either of these approaches, though extra steps are required. Sequencing requires the construction of a library, for instance. Some companies offer sequencing-ready versions of their ChIP kits, such as EMD Millipore’s Magna ChIP-Seq™ Chromatin Immunoprecipitation and Next Generation Sequencing Library Preparation Kit.
According to Farooqui, interest in ChIP-Seq is rising as the cost of sequencing falls. But arrays still retain certain advantages, including ease of use, ease of analysis, reproducibility and cost, according to Stan Trask, scientific program manager at Affymetrix.
To probe ChIP’d material using a microarray, the DNA must be labeled prior to hybridization. To hybridize to an Affymetrix GeneChip microarray, for instance, the DNA is end-labeled with biotin using terminal transferase; following washing, hybridizing fragments are detected with a streptavidin-coupled fluorescent dye.
Any tiling array may be used for this purpose, Trask says. Affymetrix, for instance, offers a completely tiled human genome on seven GeneChip microarrays for full, unbiased coverage. Alternatively, researchers can choose arrays that focus on promoters, CpG islands, specific chromosomes or custom designs. Human promoter-focused arrays, for instance, are available from Affymetrix, RocheNimbleGen and Agilent Technologies (the latter being included in EMD Millipore’s Magna ChIP2 ™ Chromatin Immunoprecipitation Human Promoter 244K Microarray Kit).
But just because any array can be used doesn’t mean it necessarily should be, says Trask. He advises researchers to ask whether the array content overlaps the areas you want to study, and if its resolution (that is, probe-to-probe spacing) is sufficient to capture the detail you need. In other words: “What do you want to get from your study? What are your research goals?”
The image at the top of this page is Thermo Scientific's Pierce Agarose ChIP Kit.