For the best experience on the Abcam website please upgrade to a modern browser such as Google Chrome
ChIP-chip and ChIP-seq technologies have shed light on the complex mechanisms and interactions involved in gene regulation. ChIP techniques have proven particularly useful for generating detailed genomic maps of epigenetic marks and transcription factor binding sites.
Though useful for mapping protein-DNA interactions, ChIP data does not provide direct insights into chromatin structure. While DNA itself is linear, chromatin is a complex three-dimensional (3D) structure capable of dynamic conformational changes.
Such changes are believed to facilitate regulation of gene expression. 3D structure on the cellular scale is difficult to resolve, but it can be inferred from the frequency of interaction of relevant genomic loci.
In 2002, Job Dekker et al. introduced chromosome conformation capture (3C), a method that allows the inference of 3D chromatin structure in vivo (Dekker et al, 2002).
In 3C, formaldehyde cross-linked chromatin is digested with a restriction enzyme that excises a known or predicted interaction point where two DNA regions meet via a protein complex. After ligating these fragments, the crosslinks are reversed, and quantitative PCR (qPCR) is used to assess recombinant fragment abundance and thus interaction frequency of the two regions.
After its initial introduction, 3C was used to analyze the chromosome organization and dynamics in higher eukaryotes.
Small (30-40 kb) regions, such as the mouse immunoglobulin kappa (Ig) gene domain (Liu and Garrard, 2005) and chicken alpha globin gene domain (Gavrilov and Razin, 2008) as well as large (~150-200 kb) regions such as mammalian alpha and beta globin gene domains were successfully analyzed (Palstra et al., 2003; Vakoc et al., 2005; Vernimmen et al., 2007; Zhou et al., 2006).
At the human beta globin locus, the upstream locus control region (LCR) was shown to physically interact with the active genes, looping the intervening sequence. In all cases, it was clear that the activation or repression of gene expression dramatically changed the 3D structure of the domain.
Further, that these changes oscillated through development and coincided with specific developmental checkpoints.
While the original 3C method was and still is a very appropriate technology for many research settings, it required some prior knowledge of interactions to be assayed by qPCR.
Moreover, some of the interactions that 3C surveys are infrequent, which can lead to substantial noise in 3C data. The method did provide a great foundation for additional innovation, however, and has led to the introduction of several derivate techniques that provide increased utility in broader experimental scenarios.
4C: Circularized Chromosome Conformation Capture and Chromosome Conformation Capture on Chip
There are two variations of 4C (Zhao et al, 2006 and Simonis et al, 2006). Both involve the amplification of circular, double ligation products containing the known DNA region and an unknown DNA interacter.
The unknown DNA is amplified by PCR primers, which bind in the known region and face outward. The PCR products are then detected using microarrays.
The other 4C variant differs in that a second, frequently cutting restriction enzyme creates smaller fragments allowing easier circularization. The use of microarrays and the ability to amplify unknown interacters dramatically increases the scale of 4C over 3C, and allows novel interactions to be detected.
5C: Chromosome Conformation Capture Carbon Copy
5C (Dostie et al, 2007) is a further refinement of 3C which uses special 5C primers that anneal across the ligated junction of the DNA fragments. The primer tails contain a universal sequence, allowing the amplification of all ligation products.
This “carbon copy” library is analysed using either microarrays or quantitative DNA sequencing. Advantages of 5C over 4C and 3C include large scale multiplexing, compatibility with DNA sequencing technology, and a larger scale interaction matrix generating more accurate 3D reconstructions.
Drawbacks, however, are that not all sites permit the design of 5C primers, meaning that 4C can have higher resolution.
Hi-C (Belton et al, 2012) increases the scale of investigated interactions further. Before the initial ligation step, biotin-labeled nucleotides are annealed to the fragment ends.
A blunt end ligation followed by a biotin pull down isolates ligated fragments only. Quantitative DNA sequencing is then performed and reads are mapped to the genome, allowing the identification of the frequency of interaction across the genome.
Hi-C offers a 10-fold increase in resolution, but in turn requires a 100-fold increase in sequencing depth.
ChIA-PET (Chromatin Interaction Analysis by Paired-End Tag Sequencing)
ChIA-PET (Fullwood et al, 2009) combines ChIP-based methods and 3C technology to direct analysis to DNA interactions involving a specific protein. The cross-linked chromatin is digested and then pulled down with an antibody against a protein of interest.
Following slightly modified ligation and purification procedures, the resulting library is sequenced yielding a set of DNA regions that interact with each other and the protein of interest. The ChIP based nature of ChIA-PET is both its greatest advantage and limitation: ChIA-PET only works if the researcher has a protein with which to screen, so it is not appropriate for all experiments.
It should also be noted that ChIA-PET cannot determine if the protein of interest is actually binding either side of the DNA loop interaction, only that it is present.
3D data from 3C-based technologies are offering new insights that weren’t possible just a few years ago. A great example can be found in a recent study that investigated the lncRNA Xist and its localization pattern in X inactivation.
The research team found that Xist interacts with distal regions of the X chromosome during silencing, but did so independent of any sequence motifs (Engreitz et al., 2013). Early Xist binding was found to correlate with H3K27 methylation, gene density, and lack of LINE repeats; however, these findings did not explain to which locations Xist first localizes.
Using a previous Hi-C map of mouse X-chromosome conformation (Lieberman-Aiden et al., 2009), the team found that the Xist RNA spread was tied more closely to the three dimensional structure of the X chromosome.
3C has also been used to assess the role of chromosome conformation in cancers. Rickman et al. (2012) assessed the role of chromosome conformation in expression of the oncogene ERG in prostate cancer cell lines using Hi-C.
They found not only does ERG overexpression lead to broad changes in chromosome conformation, but that binding of the ERG transcription factor is associated with changes in chromatin conformation and expression of prostate cancer-relevant genes.
These are just a few examples of how 3C-based methods have led to new insights into the structural basis of disease and development that may otherwise go undiscovered.