Chromatin Interaction Analysis by Paired-End Tag Sequencing (ChIA-PET) is a method for de novo detection of chromatin interactions, for better understanding of transcriptional control.
Genomes are organized into three-dimensional structures, adopting higher-order conformations inside the micron-sized nuclear spaces 7, 2, 12. Such architectures are not random and involve interactions between gene promoters and regulatory elements 13. The binding of transcription factors to specific regulatory sequences brings about a network of transcription regulation and coordination 1, 14.
Chromatin Interaction Analysis by Paired-End Tag Sequencing (ChIA-PET) was developed to identify these higher-order chromatin structures 5,6. Cells are fixed and interacting loci are captured by covalent DNA-protein cross-links. To minimize non-specific noise and reduce complexity, as well as to increase the specificity of the chromatin interaction analysis, chromatin immunoprecipitation (ChIP) is used against specific protein factors to enrich chromatin fragments of interest before proximity ligation. Ligation involving half-linkers subsequently forms covalent links between pairs of DNA fragments tethered together within individual chromatin complexes. The flanking MmeI restriction enzyme sites in the half-linkers allow extraction of paired end tag-linker-tag constructs (PETs) upon MmeI digestion. As the half-linkers are biotinylated, these PET constructs are purified using streptavidin-magnetic beads. The purified PETs are ligated with next-generation sequencing adaptors and a catalog of interacting fragments is generated via next-generation sequencers such as the Illumina Genome Analyzer. Mapping and bioinformatics analysis is then performed to identify ChIP-enriched binding sites and ChIP-enriched chromatin interactions 8.
We have produced a video to demonstrate critical aspects of the ChIA-PET protocol, especially the preparation of ChIP as the quality of ChIP plays a major role in the outcome of a ChIA-PET library. As the protocols are very long, only the critical steps are shown in the video.
A. Chromatin Immunoprecipitation (ChIP) (see Figure 1)
1. Dual Crosslinking of Chromatin-Bound Proteins and Cell Harvesting
(at 02:10 of the video)
Chromatin immunoprecipitation (ChIP) is the first critical step involved in constructing a ChIA-PET library. This step is important to reduce the level of complexity, background noise and add specificity. The preparation of ChIP will need to be optimized for cell type and factor of interest. This protocol is based on the RNA Polymerase II ChIA-PET library prepared from MCF-7 cells 9 constructed in our laboratory. To ensure that the resulting library is of sufficient complexity, we recommend using 1 x 108 cells to prepare the ChIP material. Depending on the target and cell line, the yield obtained can be between 100 ng to 300 ng. We recommend that a minimum of 100 ng ChIP should be used to construct a ChIA-PET library with fewer than 20 PCR cycles to minimize redundancy during sequencing. Higher amounts of ChIP material will allow for further reductions in redundancy, improving the quality of the libraries.
2. Cell Lysis
(at 04:15 of the video)
3. Nuclear Lysis
(at 05:00 of the video)
Nuclear lysis is performed to release crosslinked chromatin before chromatin fragmentation. In the absence of nuclear membrane, the crosslinked chromatin can be sonicated using gentler conditions. Sometimes, sonication strength may not be sufficient to break up the nuclear membrane in which case, less chromatin will be obtained because the intact nuclei will be discarded after centrifugation. However, different cell types may require different conditions.
4. Fragmentation of Chromatin
(at 07:03 of the video)
5. Washing, Preclearing and Coating of Antibody to Beads
(at 08:17 of the video)
6. Chromatin Immunoprecipitation
(at 10:03 of the video)
To reduce the level of complexity and background noise, antibodies against specific protein factors are used to enrich specific chromatin fragments of interest before proximity ligation 6.
Here, we used a mouse RNA polymerase II monoclonal antibody (8WG16) that recognized the initiation form of the protein. By enriching DNA fragments that are associated with RNA polymerase II, specificity of the library can be increased, allowing identification of long-range chromatin interactions between active promoters and their corresponding regulatory regions 9.
7. Washing and Elution of Immunoprecipitated DNA-Protein Complexes
(at 11:13 of the video)
B. Chromatin Interaction Analysis using Paired-End Tag Sequencing (ChIA-PET)
The second half of the video will highlight the key steps in the construction of a ChIA-PET library.
1. End-Blunting of ChIP DNA Fragments
This chapter of the video highlights two main points, a step-by-step procedure of washing magnetic beads to remove enzymes and buffering salt of the previous reaction (Step 1.1, 12:22 to 12:54 of the video) and the procedure of setting up enzymatic reactions involving magnetic beads in the reaction mixture (Step 1.2, 12:54 to 13:25 of video).
2. Ligation of Biotinylated Half-Linkers to ChIP DNA
This chapter of the video highlights the characteristics of the half-linker oligonucleotides used in the construction of ChIA-PET and the use of nucleotide barcode composition to distinguish between non-specific and specific ligation products (13:28 to 14:24) of the video).The chapter also shows the step-by-step procedure of setting up a half-linker ligation reaction (Step 2.2, 14:24 to 15:54 of the video).
Two types of biotinylated half-linkers are introduced in this protocol and are designed with an internal barcode of four nucleotides (TAAG or ATGT) and a recognition site for the type IIS restriction enzyme MmeI (TCCAAC). After ChIP enrichment, sonicated chromatin fragments are divided equally into two aliquots and are first ligated with an excess of either half-linker A or half-linker B 5,9.
3. Elution and Proximity Ligation of ChIP DNA Fragments
This chapter of the video explains the role of Buffer EB, SDS and Triton X-100 during the elution of chromatin complexes from beads and also shows the step-by-step procedure of setting up a circularization reaction (Step 3.9, 15:54 to 17:10 of the video).
After ligating the half-linkers to the chromatin fragments, both fractions are combined and eluted off the beads. Interacting DNA fragments will then be connected by a complete linker sequence during proximity ligation.
Using the nucleotide barcode composition, the sequences can be classified into three categories, namely sequences with heterodimer AB linkers (barcode ATGT / TAAG) and sequences with homodimer AA or BB linkers (barcode TAAG / TAAG or ATGT / ATGT) to distinguish non-specific and specific ligation products respectively 9.
Hence, the use of two half-linkers with different nucleotide barcodes allows specification of different experiments or replicates, as well as to monitoring of non-specific chimeric ligation rate between different ChIP complexes 5.
4. Reverse-Crosslinking and DNA Purification
This chapter of the video shows the step-by-step procedure of phenol:chloroform extraction (Step 4.3, 17:11 to 17:59) and the close-up of the pelleted DNA after centrifugation in Step 4.4.
5. Immobilization of ChIA-PET DNA to Streptavidin Beads
The introduction of this chapter talks about the presence of the MmeI recognition site and the biotinylated T present in the half-linker oligonucleotides to facilitate the extraction of tag-linker-tag constructs (“PETs”, 18:02 to 19:10 of the video). In addition, this chapter of the video also gives a quick overall view of Step 5.2, the step-by-step procedure of setting up a PCR reaction (Step 6.1, 19:10 to 20:00 of the video) and excising a successful ChIA-PET DNA (Step 6.9, 20:00 to 20:30).
Half-linkers A and B contain flanking MmeI recognition sites, allowing this type IIS restriction enzyme to cut 18/20 base pairs downstream of their target binding sites to generate short “tags” of the chromatin fragment, producing paired tag-linker-tag constructs (“PETs”).
To enable the purification of PET constructs by streptavidin-coated magnetic beads, both half-linker A and B are modified with biotin, allowing capture and purification of the ChIA-PET constructs.
6. Amplification of ChIA-PET DNA
Initial Step | 30 seconds | 98 °C | (denaturation) |
18 to 25 cycles | 10 seconds | 98 °C | (denaturation) |
30 seconds | 65 °C | (annealing) | |
30 seconds | 72 °C | (extension) | |
Final Step | 5 minutes | 72 °C | (final extension) |
7. Quality Check and Amplification of ChIA-PET DNA
C. Representative ChIA-PET Results
We successfully constructed ChIA-PET libraries using RNA Polymerase II antibody (8WG16) in MCF-7 cells (CHM160 and CHM163) 9 using 672 ng of ChIP material as described above. The initial diagnostic gel run of this PCR-amplified library, as mentioned in Step 6.2 of the ChIA-PET protocol, displayed a bright and well-defined band at the expected size of 223 base pairs for all PCR cycling used (see Figure 4).
16 PCR cycles was used to amplify the ChIA-PET library and a total yield of 17.1 ng was obtained. A single, intense electropherogram peak was observed at the expected size of 223 base pairs via Agilent DNA 1000 analysis as mentioned in Step 7.1 of the ChIA-PET protocol (see Figure 5).
Figure 1. ChIP overview. MCF-7 cells are dual cross-linked with EthylGlycol bis(SuccinimidylSuccinate (EGS) and formaldehyde sequentially, resulting in covalent links between spatially adjacent chromatin. The cross-linked chromatin was obtained from the fixed MCF-7 cells by cell lysis and nuclear lysis. The chromatin was then subjected to fragmentation to a size range of 200-600 base pairs. After pre-clearing the sonicated chromatin with Protein G magnetic beads to remove non-specific DNA, the pre-cleared chromatin was immunoprecipitated overnight with antibody-coated beads to capture chromatin of interest.
Figure 2. Gel analysis of sonicated chromatin fragments. A 100 bp DNA ladder is shown in the first and last lane for size reference. The sonicated chromatin displayed strong intensity between 200 to 600 bp which is ideal to capture long-range chromatin interactions.
Figure 3. ChIA-PET overview. Fragmented chromatin fragments are end-blunted and ligated to biotinylated half-linkers containing flanking MmeI restriction sites. Intact chromatin complexes are then eluted off the beads and subjected to proximity ligation under extremely dilute conditions, such that interacting DNA fragments are preferentially ligated to one another. After reverse cross-linking to remove DNA-associated proteins, MmeI digestion is carried out to release tag-linker-tag (PET) constructs, which are then purified by selective binding to streptavidin beads. The PET constructs are ligated with adapters for high-throughput sequencing.
Figure 4. Gel analysis of ChIA-PETs after PCR amplification. A 25 bp DNA ladder is shown in lane 1 and 5 for size reference. Lanes 2 to 4 are PCR products generated after 16, 18 and 20 cycles of PCR amplification from 2 μl of bead-immobilized template, respectively. This is a successful library, as indicated by the bright, well-defined bands at the expected size of 223 bp. The non-specific smear is generated when the number of PCR cycles is increased while the lowest band of each PCR reaction is made up of primer dimers.
Figure 5. Agilent 2100 Bioanalyzer analysis of purified Illumina-454 adapter-ligated ChIA-PETs. Screen capture of Agilent 2100 Bioanalyzer electropherograms profiling a successful library, with a single intense peak at the expected size of 223 bp. Note that the Agilent Bioanalyzer assay usually reports a slightly higher-than-expected size; in this case, the desired peak is displayed at 237 bp instead of 223 bp. This is within the 10% error range of the Agilent assay.
ChIA-PET is a method developed to identify long-range interactions in transcription regulation. One of the critical factors that determine the quality of a ChIA-PET library is the quality of ChIP material.
The protocol shown in the video incorporates the use of EGS and formaldehyde to cross-link the cells. The use of formaldehyde combined with a second cross-linking reagent bearing a longer spacer arm may help in the binding of proteins which could not be bound by formaldehyde alone 3,11,15. We have constructed libraries with this method which have demonstrated robust binding sites and long-range interactions 9. However, the cross-linking and ChIP conditions should be optimized for each factor of interest, and it is important not to over-crosslink as too much cross-linking will result in difficulties in fragmentation by sonication, and may possibly result in spurious chromatin interactions. Chromatin interactions identified by ChIA-PET should be validated by a different method, such as fluorescence in-situ hybridization 4.
We recommend a minimum of 100 ng of chromatin material. While we have constructed good quality libraries from 50 ng of chromatin material, we have observed that large amounts of starting material allowed construction of ChIA-PET libraries with less than 16 PCR cycles, thereby minimizing amplicons and redundancy of each library. This lower redundancy correlated with higher unique mapped tags and also a high percentage of usable data, thereby enabling a more comprehensive chromatin interaction map with fewer lanes of sequencing. The final packed volume of beads in each tube should be 50 μl and 100 μl for magnetic and Sepharose beads respectively. If the packed bead volume is less than stated, bring to the minimum packed volume with similarly pre-cleared blank magnetic or Sepharose beads to minimize loss of DNA-bearing beads in subsequent steps. Sawed-off tips or large-core tips should be used for pipetting Sepharose beads.
The following modifications were incorporated following the previously published ChIA-PET protocol 5. Firstly, magnetic G beads were used to minimize sample loss during washes. In addition, we identified non-specific bands with approximate sizes of 100 bp and 138 bp to be amplicons of self-ligated half-linkers or/and adaptors. Hence, we reduced the concentration of biotinylated half-linkers and 454 GS20 adaptors to minimize non-specific bands during PCR amplification. The proximity ligation volume was reduced from 50 ml to 10 ml to minimize sample loss during subsequent purification steps and also save on reagent costs. We also increased the incubation time to immobilize ChIA-PET DNA to beads to ensure maximal capture of ChIA-PET DNA on the streptavidin beads.
During the proximity ligation step, chimeric ligations that do not represent true in vivo chromatin interactions are inevitably generated in a non-specific and random manner. Hence, to evaluate the quality of data from any ChIA-PET experiment, the rate of chimerism is estimated from the use of two different half-linkers with specific nucleotide barcodes TAAG and ATGT 5. After high-throughput sequencing, the ChIA-PET sequences are first analyzed for linker barcode composition and sequences derived from specific ligation products and non-specific ligation products can be distinguished 8. The percentage of known chimeras (i.e. heterodimers AB linkers) present in our in-house MCF-7 RNA Polymerase II ChIA-PET libraries is less than 15%.
ChIA-PET sequences are subsequently classified into two categories, namely self-ligation PETs and inter-ligation PETs. Self-ligation PETs are obtained from self-circularization ligation of the chromatin fragments while inter-ligation PETs are derived from inter-ligation between two different DNA fragments. The latter is then sub-divided into three different categories based on the genomic distance of each tag on the same chromosome (intrachromosomal inter-ligation PETs) or that both tags are mapped to two different chromosomes (interchromosomal inter-ligation PETs). We have developed a ChIA-PET tool software package to sort out the different categories 8. This will be based on the DNA fragments that are in the library. Generally, smaller ChIP fragments will give a higher resolution and the cut off for these RNA Polymerase II ChIA-PET libraries is about 4 kb.
In addition, real chromatin interactions can be distinguished from random noise by counting the number of inter-ligation PETs in an interaction cluster; in other words, a cluster of high PET count is said to have a higher probability of being a real chromatin interaction 8.
To filter our false positives arise from highly enriched anchors which can form inter-ligation PETs by random chance, a statistical analysis framework has also been formulated to account for random formation of any inter-ligation PETs between two anchors 8.
In conclusion, the ChIA-PET technique allows mapping chromatin interaction networks on a global scale. The implementation of ChIP in ChIA-PET allows the reduction of library complexity and background noise. In addition, ChIP adds specificity to chromatin interactions, enabling the examination of specific chromatin interactions associated with particular transcription factors 5.
The authors have nothing to disclose.
The authors are supported by A*STAR of Singapore. In addition, M.J.F. is supported by an A*STAR National Science Scholarship, a L’Oreal For Women in Science National Fellowship and a Lee Kuan Yew Post-Doctoral Fellowship. Y.R. is supported by NIH ENCODE grants (R01 HG004456-01 and R01 HG003521-01). The authors also acknowledge the videography team of 8 Pixels Productions, Singapore, in particular Mr. Kelvin Issey, Mr. Chang Kai Xiang and Mr. Sherwin Gan for filming the scenes, Ms. Siti Rahim for video editing and Ms. Michelle Teo for the voice-over.
Name of the reagent | Company | Catalogue Number | Comments |
4-20% gradient TBE gel | Invitrogen | EC6225BOX | Step B, 6.3 |
5 x T4 DNA Ligase Buffer with PEG | Invitrogen | 46300018 | Step B, 2.2 |
6% TBE gel | Invitrogen | EC6263BOX | Step B, 6.8 |
Agilent DNA 100 Assay | Agilent Technologies | 5067-1504 | Step B, 7.1 |
Agilent High Sensitivity DNA Assay | Agilent Technologies | 5067-4626 | |
Agilent Bioanalyzer 2100 | Agilent Technologies | G2940CA | |
Centrifuge Tube Filters (Spin-X) | Corning | CLS8160 | Step B, 3.7 and 6.9 |
Dark Reader Transilluminator | Clare Chemical Research | DR46B | Step B, 6.8 |
Digital Sonifier Cell Disrupter | Branson | 450D-0101063591 | Step A, 4.3 |
DynaMag-2 magnet (Magnetic Particle Concentrator) | Invitrogen | 123-21D | Step A, 7.5.1 Step B, for all magnetic beads washing steps |
DynaMag-15 Magnet (Magnetic Particle Concentrator) | Invitrogen | 123.01D | Step A, 5 and 6 |
DynaMag-PCR (Magnetic Particle Concentrator) | Invitrogen | 49-2025 | Step B, 6.5 |
Escherichia coli DNA Polymerase I | NEB | M0209 | Step B, 5.6 |
GlycoBlue | Ambion | AM9516 | Step A, 7.5.1 Step B, 4.3 and 6.5 |
Illumina cBot Cluster Generation System | Illumina | SY-301-2002 | Step B, 7.4 |
Illumina Genome Analyzer IIx | Illumina | SY-301-1301 | Step B, 7.5 |
Illumina PE primers | Illumina | PE-102-1004 | Step B, 5.4 (if necessary) |
Intelli-Mixer | Palico Biotech | RM-2L | Step B, any incubations with rotation |
LightCycler 480 Real-Time PCR System | Roche | 04 640 268 001 | Step A, 7.5.3 Step B, 7.3 |
LightCycler480 DNA SYBR Green I MasterMix | Roche | 03 752 186 001 | Step B, 7.5.3 |
M-280 Streptavidin Dynabeads | Invitrogen | 11206D | Step B, 5.2 |
Magnetic (Dynabeads) Protein G | Invitrogen | 100.03D | Step A, 5.1.1 and 5.2.1 |
MaXtract High Density (2ml) | Qiagen | 129056 | Step A, 7.5.1 |
MaXtract High Density (50ml) | Qiagen | 129073 | Step B, 4.2 |
MmeI | NEB | R0637 | R0637 |
RNase ONE Ribonuclease | Promega | M426C | Step B, 4.8 |
RNA Polymerase II (8WG16) monoclonal antibody | Covance | MMS-126R | Step A, 5.2.4 |
Oak Ridge Centrifuge Tubes (Polypropylene) | Nalgene | 3119-0050 | Step A, 3.1 |
Oak Ridge Centrifuge Tubes (Teflon FEP) | Nalgene | 3114-0050 | Step B, 4.3 |
Phusion High Fidelity Master Mix | Finnzymes | F-531 | Step B, 6 |
Picogreen (Quant-iT) dsDNA Reagent | Invitrogen | P11495 | Step A, 7.5.2 |
Polystyrene Round Bottom Test Tube | BD Biosciences | 352057 | Step A, 4.1 |
Protease Inhibitor Cocktail Tablets (cOmplete, EDTA-free) | Roche | 11873580001 | Step A, 1.6 onwards |
Proteinase K Solution (20mg/ml) | Fermentas | E00491 | Step A, 4.4, 7.5.1 Step B, 4.1 |
T4 DNA Ligase | Fermentas | EL0013 | Step B, 2.2, 3.9 and 5.4 |
T4 DNA Ligase Buffer (NEB) | NEB | B0202S | Step B, 3.3 and 3.9 |
T4 DNA Polymerase | Promega | M4215 | Step B, 1.2 |
T4 DNA Polynucleotide Kinase | NEB | M0201 | Step B, 3.3 |
TruSeq SBS Kit v5–GA | Illumina | FC-104-5001 | Regents for Illumina Genome Analyzer IIx system |
TruSeq PE Cluster Kit v2–cBot–GA | Illumina | PE-300-2001 | to be use with Illumina cBot Cluster Generation System |
SYBR Green I | Invitrogen | S-7585 | Step B, 6.3 and 6.8 |
XCell SureLock Mini-Cell Electrophoresis System | Invitrogen | EI0001 | Step B, 6.3 and 6.8 |
Name | Sequence | Comments | |
Biotinylated half-linkers A (200ng/μl) | Top | 5′ GG CCG CGA/iBiodT/ ATC TTA TCC AAC 3′ | 250 nmole scale HPLC Purified Internal Biotin dT(9) |
Bot | 5′ GTT GGA TAA GAT ATC GC 3′ | 250 nmole scale HPLC Purified | |
Biotinylated half-linkers B (200ng/μl) | Top | 5′ GGC CGC GA/iBiodT/ ATA CAT TCC AAC 3′ | 250 nmole scale HPLC Purified Internal Biotin dT(9) |
Bot | 5′ GTT GGA ATG TAT ATC GC 3′ | 250 nmole scale HPLC Purified | |
Non-biotinylated half-linkers (200ng/μl) | Top | 5′ GGC CGC GAT ATC GGA TCC AAC 3′ | 250 nmole scale PCR Grade |
Bot | 5′ GTT GGA TCC GAT ATC GC 3′ | 250 nmole scale PCR Grade | |
GS20 Adaptor A (200ng/μl) | Top | 5′ CCA TCT CAT CCC TGC GTG TCC CAT CTG TTC CCT CCC TGT CTC AGN N 3′ | 250 nmole scale PCR Grade |
Bot | 5’CTG AGA CAG GGA GGG AAC AGA TGG GAC ACG CAG GGA TGA GAT GG 3′ | 250 nmole scale PCR Grade | |
GS20 Adaptor B (200ng/μl) | Top | 5′ CTG AGA CAC GCA ACA GGG GAT AGG CAA GGC ACA CAG GGG ATA GG 3′ | 250 nmole scale PCR Grade |
Bot | 5′ CCT ATC CCC TGT GTG CCT TGC CTA TCC CCT GTT GCG TGT CTC AGN N 3′ | 250 nmole scale PCR Grade | |
Illumina-NN adaptors | Top | 5′ ACA CTC TTT CCC TAC ACG ACG CTC TTC CGA TC 3′ | 250 nmole scale HPLC purified See Step B, 5.4 |
Bot | 5′ phos – GAT CGG AAG AGC GGT TCA GCA GGA ATG CCG AG 3′ | 250 nmole scale HPLC purified Phosphorylated 5′ end See Step B, 5.4 | |
Illumina 1-454 (forward) primer (10 μM) | 5′ AAT GAT ACG GCG ACC ACC GAG ATC TAC ACC CTA TCC CCT GTG TGC CTT G 3′ | 250 nmole scale PCR Grade | |
Illumina 2-454 (reverse) primer (10 μM) | 5′ CAA GCA GAA GAC GGC ATA CGA GAT CGG TCC ATC TCA TCC CTG CGT GTC 3′ | 250 nmole scale PCR Grade | |
qPCR Primer 1.1 (10 μM) | 5′ AAT GAT ACG GCG ACC ACC GAG AT 3′ | 10nmole scale PCR Grade | |
qPCR Primer 2.1 (10 μM) | 5′ CAA GCA GAA GAC GGC ATA CGA 3′ | 10nmole scale PCR Grade | |
Illumina 3-454 sequencing primer (100 μM) | 5’TGC GTG TCC CAT CTG TTC CCT CCC TGT CTC AG 3′ | 100nmole scale HPLC Purified | |
Illumina 4-454 sequencing primer (100 μM) | 5’GTG CCT TGC CTA TCC CCT GTT GCG TGT CTC AG 3′ | 100nmole scale HPLC Purified |
Table of oligos and adaptors. Order oligos from Integrated DNA Technologies (IDT) and prepare linkers and adaptors in the same manner as described previously10. Half-linkers and adaptors may be prepared beforehand and stored for several months at -20 °C.