Sensitive detection of chromatin co-‐associations using enhanced chromosome conformation capture on chip Tom Sexton1, Sreenivasulu Kurukuti2, Jennifer A. Mitchell3, David Umlauf4, Takashi Nagano5 and Peter Fraser5,6 1
Laboratory of Chromatin and Cell Biology, Institute of Human Genetics, CNRS UPR 1142, 141 rue de
la Cardonille, 34396 Montpellier Cedex 5, France. 2
Department of Animal Sciences, School of Life Sciences, University of Hyderabad, Hyderabad-‐
500046. A.P. India. 3
Department of Cell and Systems Biology, University of Toronto, 25 Harbord Street, Toronto, Canada
M5S 3G5 4
Department of Functional Genomics, Institut de Genetique et de Biologie Moleculaire et Cellulaire, 1
rue Laurent Fries, BP 10142, 67404 Illkirch, France. 5
Nuclear Dynamics Programme, Babraham Institute, Babraham Research Campus, Cambridge, UK.
CB22 3AT. 6
Corresponding author. E-‐mail:
[email protected]. Telephone: +44 (0)1223 496644. Fax:
+44 (0)1223 496002. First published in [Nature Protocols. 2012 Jun 21;7(7):1335-‐50. doi: 10.1038/nprot.2012.071. PMID: 22722369] Nature Publishing Group, a division of Macmillan Publishers Limited] KEYWORDS Chromosome conformation capture; nuclear organization; microarrays; enhanced 4C; e4C
1
ABSTRACT Chromosome Conformation Capture (3C) is a powerful technique for analyzing spatial chromatin organization in vivo. Technical variants of the assay (‘4C’) allow the systematic detection of genome-‐ wide co-‐associations with bait sequences of interest, enabling the nuclear environments of specific genes to be probed. We describe enhanced 4C (e4C), a technique incorporating additional enrichment steps for bait-‐specific sequences, and thus improving sensitivity in the detection of distal chromatin co-‐associations. In brief, e4C entails the fixation, restriction digestion and ligation steps of conventional 3C, with an optional chromatin immunoprecipitation (ChIP) step to select for subsets of chromatin co-‐associations, followed by bait enrichment by biotinylated primer extension and pull-‐ down, adapter ligation and PCR amplification. Chromatin co-‐associations with the bait sequence can then be assessed by hybridizing e4C product to microarrays or sequencing. The e4C procedure takes approximately one week to go from tissue to DNA ready for microarray hybridization. INTRODUCTION Chromosome Conformation Capture (3C) has revolutionized research into spatial genome organization, allowing chromatin co-‐associations to be assessed at resolutions unattainable by light microscopic methods. More generalized variants of the technique have emerged to probe chromatin co-‐associations and hence nuclear organization in a more global manner. One such technique, enhanced chromosome conformation capture on chip (e4C), is described in this protocol and has been used to identify the spatial transcriptional partners of the mouse globin genes1. Briefly, the 3C technique and its variants entail formaldehyde fixation and restriction digestion of nuclei in their native state, followed by ligation under dilute conditions which favor intramolecular reactions between covalently cross-‐linked restriction fragments. Hybrid ligation products are generated between restriction fragments which are separated on the linear chromosome (or reside on different chromosomes), but which are physically proximal in vivo, and their relative abundances can be assessed by PCR to draw conclusions about chromatin topology2,3. Over the last decade, 3C has been used to identify chromatin loops between distal regulatory elements and target genes and to correlate these loops with gene regulation4. Several alternative techniques to PCR assessment of limited numbers of ligation products have been developed in an effort to map chromatin contacts genome-‐wide; the suitability of each technique depends on the specific aspect of nuclear organization that is being investigated. A relatively inexpensive and informative way to map all chromatin interactions with one bait sequence of interest, such as a specific gene or regulatory element, in a genome-‐wide manner is the 2
‘4C’ (defined in this article as circular chromosome conformation capture) method, developed in parallel by multiple groups5-‐8. The technical differences are described elsewhere9, but all methods essentially entail circularization of the 3C template and an inverse PCR approach, using primer pairs within the bait restriction fragment to amplify all bait-‐linked 3C ligation products. Such 4C, coupled with hybridization of the material to microarrays, has identified networks of genes which co-‐ associate in the nucleus6,8,10,11. However in at least one case, the sensitivity of 4C has been questioned: RNA FISH (fluorescent in situ hybridization) studies identified multiple significant interchromosomal co-‐associations between the globin genes and other expressed genes in erythroid cells1, which were not detected by 4C in the same tissue6. Conversely, the majority of these transcriptional co-‐associations were uncovered by e4C, the protocol described here and designed to optimize detection of rarer but significant chromatin co-‐associations1. Increased sensitivity is conferred by replacing the circularization step with a bait enrichment step, employing primer extension with a bait-‐specific primer and pull-‐down on streptavidin-‐coated beads (Fig. 1a). This step confers at least a hundred-‐fold enrichment of bait sequences (Box 1 and Anticipated Results), reducing the amount of non-‐specific genomic DNA in the (e)4C product which also undergoes microarray hybridization, thus improving signal-‐to-‐noise ratios. e4C products are then amplified by incorporating an adapter and PCR using a nested, bait-‐specific primer and an adapter-‐specific primer (Fig. 1a). The major limitation of e4C, like all 3C-‐based methods to date, is that chromatin interactions are pooled from a large population of nuclei. Alternative techniques, such as FISH, are required to assess specific genomic co-‐associations within single cells. Another limitation of e4C is that the results of each experiment are limited to contacts with a specific bait region. The 3C-‐carbon copy (5C) technique allows the repertoires of contacts between multiple specific targets to be assessed simultaneously12, but requires the expensive and technically challenging use of hundreds of primers and cannot assess chromatin interactions outside of the tested regions. The ‘Hi-‐C’ technique couples 3C with high-‐throughput sequencing to offer a truly genome-‐wide view of chromatin contacts13, but the numbers of sequences required to obtain high-‐resolution profiles in metazoan genomes is prohibitively expensive when assessment of only the interactions with specific genes is desired. This is caused by the complexity of 3C material generated in large genomes: a genome of n restriction fragments has a number of possible 3C ligation products in the order of n2. The ChIA-‐PET (chromatin interaction assay with paired end tags) method incorporates a chromatin immunoprecipitation (ChIP) step to reduce the complexity of sampled 3C ligation products14, but the numbers of sequences needed for comprehensive maps remain very large. e4C coupled to microarray hybridization is thus an inexpensive and efficient method for the assessment of genomic interactions with one or a few 3
specific genomic regions of interest. A ChIP step can also be introduced in the e4C protocol to reduce the complexity of the sampled 3C ligation products, and was previously done with an antibody recognizing the active form of RNA polymerase II to focus more fully on transcriptional co-‐ associations with the globin genes1. The detailed e4C protocol described below is that used for assessing chromatin co-‐ associations with the beta-‐globin gene, Hbb-‐b1, in mouse fetal liver, with an optional incorporation of a ChIP step with an antibody recognizing active RNA polymerase II. For clarity, we call the method e4C to denote the whole 3C, primer extension, streptavidin capture, adapter incorporation and PCR amplification steps in this protocol (Fig. 1a). ChIP-‐e4C denotes the same method with an incorporation of an immunoprecipitation step between the 3C digestion and ligation steps, analogous to the ChIP-‐loop technique15 and previously used to assess transcriptional co-‐associations of the globin genes1. When adapting e4C to different tissues and biological questions, various experimental considerations will need to be made and are highlighted below. Experimental design Chromosome Conformation Capture (3C): The quality of e4C results is critically dependent on the 3C material used in their generation. Technical considerations which are important for 3C experiments, such as restriction enzyme choice and obtaining a good efficiency of restriction digestion3 are equally important for e4C (see also Troubleshooting and Box 2). If possible, it is also desirable to have some known 3C results for the cell or tissue type used in the e4C study; reproduction of these results makes an ideal quality control before proceeding with subsequent steps of the e4C protocol. As described elsewhere3, several steps of the 3C may need to be modified to optimize restriction digestion and ligation in more technically challenging tissue types, such as muscle or plant tissue16. Chromatin immunoprecipitation (ChIP): The inclusion or omission of the ChIP step in the e4C protocol depends on the biological question being addressed. The ChIP step selects for a subset of genomic interactions involving chromatin fragments that are bound by a specific protein, thus increasing sensitivity of detecting rarer co-‐associations of this type. However, ChIP enrichments will be non-‐ uniform, skewing e4C detection of chromatin co-‐associations to those between the most strongly enriched binding sites and making quantitative conclusions of e4C results more difficult. It is thus important to verify such seemingly strong detected co-‐associations by performing e4C without the ChIP step (where the interaction may be expected to be detected, albeit more weakly) or by completely independent methods such as FISH. Furthermore, the protocol has been optimized for an 4
antibody recognizing the active form of RNA polymerase II1; other antibodies will need to be tested for compatibility with the e4C protocol. Analogous to known 3C results, it is desirable to have positive and negative ChIP controls for the tissue type used, which can be tested by qPCR once the 3C DNA is purified. In this manner, any required application-‐specific modifications to the protocol can be assessed for both 3C and ChIP efficiency before proceeding with e4C. Primer design: Compared to 5C, e4C requires the design of only two bait-‐specific primers. However, the fidelity and specificity of these primers is crucial to e4C and the primers need to be thoroughly tested before use (Box 3). Previous applications of e4C1 used the six-‐cutter restriction enzyme BglII for the 3C step (primary restriction enzyme X) and the four-‐cutter restriction enzyme NlaIII (secondary restriction enzyme Y) for incorporation of the adapter. In principle, e4C can be performed with any combination of restriction enzymes fulfilling these criteria (see also Fig. 1b): •
the enzyme X digests fixed chromatin efficiently in 3C experiments and cuts less frequently than the enzyme Y;
•
the bait fragment is of a sufficient length that two nested bait-‐specific primers can be designed within a single X-‐Y restriction fragment, with the primers facing site X, at a distance of ~50-‐200 bp;
•
to test e4C primer fidelity (Box 3), a site for enzyme Y is present ~20-‐200 bp downstream of the bait site X, and closer than the next site for enzyme X.
Note that the e4C adapter sequence will need to be altered to give compatible cohesive ends if a different restriction enzyme to NlaIII is used as enzyme Y. Sequencing or microarray hybridization: e4C was originally designed for hybridization to custom microarrays1. While these identify clusters of significantly-‐interacting sequences over large genomic distances (see Anticipated Results), microarrays tend to be saturated for close-‐range (few to tens of kilobases) interactions, similarly to other 4C studies6. Microarrays are thus not suitable for assessing specific chromatin loops within gene loci. However, conventional cloning and sequencing of a few thousand e4C products reproduced known looped interactions within the mouse beta-‐globin locus (see Anticipated Results), implying that the microarray technology is limiting the short-‐range analysis of chromatin topology and not the e4C procedure per se. By the incorporation of Illumina adapter sequences into the PCR primers, the e4C technique has been made compatible with high-‐throughput sequencing17. This specific application obtained a high-‐resolution profile of the local environment of a human tRNA gene, demonstrating that tRNA genes preferentially coalesce together and exclude RNA polymerase II-‐transcribed genes. However, insufficient sequences were obtained to provide a genome-‐wide view of the bait’s nuclear environment. The decision between using microarrays or 5
sequencing for e4C thus depends on the specific biological question being addressed and the budget of the experiment. Note that all microarray data deposited in public databases should adhere to the MIAME (minimal information for microarray experiments) guidelines18. Analysis: Bioinformatically, analysis of 4C and e4C datasets are identical. Original processing methods involved classical statistics, such as running means and medians6 or t-‐tests1 on sliding windows of fixed size. However, it is often difficult to predict a sensible window size for optimal detection of genomic interactions. ‘Domainogram’ approaches19 allow statistical analysis of biological profiles at multiple scales, and have recently been optimized for analysis of 4C data by other groups20-‐22. All of the analytical approaches have successfully identified gene co-‐association partners; we recommend that multiple strategies are tried to explore new datasets. However, validation of co-‐associations by independent methods such as FISH remains an indispensable part of projects involving e4C. MATERIALS Reagents •
Pregnant (E14.5) mice CAUTION: Approved national and institutional regulations for animal work must be adhered to, and experiments reported according to the ARRIVE guidelines23
•
Fetal bovine serum (FBS; Sigma F9665)
•
Dulbecco’s modified Eagle medium (DMEM; Gibco 41966)
•
10% (v/v) FBS/DMEM CRITICAL: Solution must be made under aseptic conditions. 50 ml aliquots can be stored at 4°C for a few weeks.
•
37% (v/v) formaldehyde (histology-‐grade, free from acid; Merck 103999) CAUTION: Formaldehyde is toxic. Perform fixation in a fume hood.
•
Fix solution (see Reagent Setup)
•
2 M glycine (Sigma-‐Aldrich G8898) CRITICAL: Ensure glycine is chilled on ice before use in quenching fixation. Stocks can precipitate with prolonged storage at 4°C, so should be stored at room temperature.
•
Sodium chloride (NaCl; Sigma-‐Aldrich S3014). Stocks made at 5 M concentration can be stored for months at room temperature.
•
Disodium hydrogen phosphate (Na2HPO4; Sigma-‐Aldrich S3264)
•
Potassium dihydrogen phosphate (KH2PO4; Sigma-‐Aldrich P9791)
•
Phosphate buffered saline (PBS; see Reagent Setup)
•
1 M Tris-‐HCl, pH 8.0 (see Reagent Setup) 6
•
NP-‐40 (also known as Igepal CA-‐630; Sigma-‐Aldrich I8896)
•
Complete protease inhibitor, EDTA-‐free (Roche 04693132001; see Reagent Setup)
•
Cell permeabilization buffer (see Reagent Setup)
•
NEB3 buffer (10 x stock; New England Biolabs B7003S) CRITICAL: When thawing stock to make working solution, vortex thoroughly to ensure dithiothreitol (DTT) is fully dissolved.
•
20% (w/v) sodium dodecyl sulfate (SDS; Sigma-‐Aldrich 05030)
•
20% (v/v) Triton-‐X100 (Sigma-‐Aldrich; T8787)
•
BglII (50,000 U ml-‐1; New England Biolabs R0144M)
•
Dynabeads protein A-‐coated magnetic beads (Invitrogen 100-‐02D)
•
Bovine serum albumen (BSA; Sigma 05482); 5 mg ml-‐1 in PBS. Store at 4°C for a few weeks maximum.
•
Anti-‐RNA polymerase II antibody (serine-‐5-‐phosphate; ChIP-‐grade rabbit IgG Abcam Ab5131)
•
0.5 M EDTA, pH 8 (Sigma-‐Aldrich E6758)
•
Dilution buffer (see Reagent Setup)
•
ChIP buffer (see Reagent Setup)
•
ChIP wash buffer I (see Reagent Setup)
•
ChIP wash buffer II (see Reagent Setup)
•
TE buffer (see Reagent Setup)
•
T4 DNA ligase buffer (see Reagent Setup)
•
T4 DNA ligase (400,000 U ml-‐1; New England Biolabs M0202L)
•
Proteinase K (10 mg ml-‐1; Sigma-‐Aldrich P2038)
•
Ribonuclease A (RNase A; 20 mg ml-‐1; Sigma-‐Aldrich R6513)
•
Phenol-‐chloroform-‐isoamyl alcohol (25:24:1; pH 8; Sigma-‐Aldrich 77617) CAUTION: Phenol and chloroform are toxic chemicals; handle in a fume hood.
•
Glycogen (UltraPure 20 µg ml-‐1; Invitrogen 10814010)
•
Isopropanol (Sigma-‐Aldrich I9516)
•
Ethanol (Sigma-‐Aldrich 02860)
•
Molecular biology-‐grade water (Qiagen 129114)
•
Phenol, pH 8 (Sigma-‐Aldrich P4557) CAUTION: Phenol is toxic; handle in a fume hood.
•
Chloroform (Sigma-‐Aldrich 25666) CAUTION: Chloroform is toxic; handle in a fume hood.
•
3 M sodium acetate, pH 5.2 (Sigma-‐Aldrich S5636)
•
Quant-‐iT PicoGreen dsDNA assay kit (Invitrogen P7589) or Qubit dsDNA BR assay kit (Invitrogen Q32853)
•
VentR (exo-‐) DNA polymerase (2000 U ml-‐1; New England Biolabs M0257L), provided with 10 x ThermoPol buffer 7
•
dNTPs (2 mM solution, made from 100 mM PCR-‐grade stocks; Roche 11969064001)
•
Biotinylated Hbb-‐b1 primer (Sigma-‐Aldrich): 5’-‐biotin CTCAGAGCAGTATCTTTTGTTTGC 3’
•
NlaIII (10,000 U ml-‐1; New England Biolabs R0125L), provided with 10 x NEB4 buffer and 100 x (10 mg ml-‐1) BSA solution CRITICAL: Store at -‐80°C. When thawing NEB4 buffer to make working solutions, vortex thoroughly to ensure DTT is fully dissolved.
•
QiaQuick PCR purification kit (Qiagen 28106)
•
Dynabeads M-‐280 Streptavidin (Invitrogen 112-‐06D)
•
Dynabeads kilobaseBINDER kit (Invitrogen 601-‐01)
•
Streptavidin beads wash buffer 1 (SBWB1; see Reagent Setup)
•
Streptavidin beads wash buffer 2 (SBWB2; see Reagent Setup)
•
NlaIII reaction solution (see Reagent Setup)
•
e4C
adapter
forward
strand
(Sigma-‐Aldrich):
5’
TTTGGATTTGCTGGTGCAGTACAACTAGGCTTAATAGGGACATG 3’ NlaIII cohesive end is indicated in bold •
e4C
adapter
reverse
strand
(Sigma-‐Aldrich):
5’-‐phosphorylated
TCCCTATTAAGCCTAGTTGTACTGCACCAGCAAATCC 3’-‐amine C7 •
High-‐concentration T4 DNA ligase (2,000,000 U ml-‐1; New England Biolabs M0202M)
•
Adapter ligation reaction solution (see Reagent Setup)
•
HotStar Taq DNA polymerase (5 U µl-‐1; Qiagen 203205), provided with 10 x HotStar buffer
•
e4C adapter-‐specific primer (Sigma-‐Aldrich): 5’ GGATTTGCTGGTGCAGTACA 3’
•
e4C nested Hbb-‐b1 primer (Sigma-‐Aldrich): 5’ AGGATGAGCAATTCTTTTTGC 3’
Equipment •
70 µm cell strainer (Falcon 352350)
•
Centrifuge (Heraeus Megafuge 3.0R)
•
Orbital shaker
•
Haemocytometer
•
Phase contrast microscope
•
Thermal mixer (Eppendorf Thermomixer shaker EF4283)
•
Magnet (Invitrogen DynaMag-‐2 magnet 123-‐21D)
•
Rotating wheel
•
Microcentrifuge (Eppendorf 5417R)
•
PCR thermal cycler (MJ Research PTC-‐200) 8
•
NanoDrop spectrophotometer (Thermo Scientific ND2000)
Reagent Setup Fix solution: 2% formaldehyde in 10% (v/v) FBS/DMEM. Warm FBS/DMEM to room temperature (23°C), then add formaldehyde just before use. CRITICAL: Make solution fresh for each fixation. Phosphate buffered saline: 155 mM NaCl; 3 mM Na2HPO4; 1 mM KH2PO4; pH 7.4 1 M Tris-‐HCl, pH 8: Tris base (Sigma-‐Aldrich T5941), adjusted to pH 8.0 with concentrated HCl. Stocks can be stored at room temperature. Complete protease inhibitor, EDTA-‐free: Dissolve one tablet in 1 ml molecular biology-‐grade water to make 50 x working stock, which can be stored at 4°C for a maximum of three days. Cell permeabilization buffer: 10 mM Tris-‐HCl, pH 8.0; 10 mM NaCl; 0.2% NP-‐40; 1 x complete protease inhibitor, EDTA-‐free. CRITICAL: Make buffer fresh on the day of the experiment. Dilution buffer: 83 mM Tris-‐HCl, pH 8; 42 mM EDTA; 2.7 x complete protease inhibitor, EDTA-‐free. CRITICAL: Make buffer fresh on the day of the experiment. ChIP buffer: 16.7 mM Tris-‐HCl, pH 8; 167 mM NaCl; 1.2 mM EDTA; 1.1% Triton-‐X100; 0.01% SDS; 1 x complete protease inhibitor, EDTA-‐free. CRITICAL: Make buffer fresh on the day of the experiment. ChIP wash buffer I: 20 mM Tris-‐HCl, pH 8; 150 mM NaCl; 2 mM EDTA; 1% Triton-‐X100; 0.1% SDS; 1 x complete protease inhibitor, EDTA-‐free. CRITICAL: Make buffer fresh on the day of the experiment. ChIP wash buffer II: 20 mM Tris-‐HCl, pH 8; 500 mM NaCl; 2 mM EDTA; 1% Triton-‐X100; 0.1% SDS; 1 x complete protease inhibitor, EDTA-‐free. CRITICAL: Make buffer fresh on the day of the experiment. TE buffer: 10 mM Tris-‐HCl, pH 8; 1 mM EDTA. Stocks can be stored at room temperature. T4 DNA ligase buffer: 10 x stock; New England Biolabs B0202S. CRITICAL: When thawing stock to make working solution, vortex thoroughly to ensure DTT is fully dissolved. Streptavidin beads wash buffer 1: 10 mM Tris-‐HCl, pH 8; 2 M NaCl; 1 mM EDTA. Stocks can be stored at room temperature. Streptavidin beads wash buffer 2: 10 mM Tris-‐HCl, pH 8. Stocks can be stored at room temperature.
9
NlaIII reaction solution (per reaction): 20 U NlaIII in 50 µl total volume 1 x NEB4 plus 100 µg ml-‐1 BSA. CRITICAL: Make buffer fresh on the day of the experiment. Adapter ligation reaction solution (per reaction): 2000 U high-‐concentration T4 DNA ligase plus 200 pmol e4C adapter in 40 µl total volume 1 x T4 DNA ligase buffer. CRITICAL: Make buffer fresh on the day of the experiment. e4C primer design: The same principles are used in designing the biotinylated extension primer as for all other PCR primers in the protocol. Primers are 20-‐24mers with a Tm in the range of 60-‐65°C (62°C optimally), designed with the primer3 website (http://frodo.wi.mit.edu/primer3). Before use, primers are also checked for specificity in the mouse genome by BLASTN alignment to the mouse genome with the “near-‐exact matches (oligo)” setting on the Ensembl website (http://www.ensembl.org/Multi/blastview). e4C adapter setup: The adapter strand sequences given generate an NlaIII cohesive end when annealed; the terminal sequences will need to be changed if a different restriction enzyme is used. The reverse strand is 3’-‐aminoacylated and the forward strand is non-‐phosphorylated and contains a three nucleotide overhang to prevent adapter concatemerization during ligation reactions. To make 100 µM stock of adapter, mix equal volumes of 200 µM solutions of each adapter strand in a 1.5 ml eppendorf and incubate at 65°C in a water bath for 15 min. Switch off the water bath and allow the mixture to cool slowly overnight for strand annealing. The stock can be stored at 4°C for several weeks. Microarray design: All BglII-‐NlaIII fragments within the repeat-‐masked mouse genome are retrieved and fragments containing fewer than 45 bp of non-‐repetitive sequence are discarded. The remainder is used for in-‐house 45-‐65mer isothermal oligonucleotide probe design and custom microarray printing by NimbleGen Systems. PROCEDURE Single-‐cell preparation and fixation TIMING: 30 min-‐1 h 1 Dissect 5 fetal livers (containing ~2 x 108 cells; see step 8 for required cell amounts in e4C or ChIP-‐ e4C experiments) according to approved methods and transfer to a 1.5 ml eppendorf tube with cold 10% (v/v) FBS/DMEM. Obtain a single-‐cell suspension by pipetting vigorously. CAUTION: Approved national and institutional regulations for animal work must be adhered to. 10
CRITICAL STEP: Dissection and fixation steps must be as fast as possible to conserve nuclear organization. Other tissues may need harsher treatments to obtain a single-‐cell suspension3,16. For use of new tissue types, it is recommended to perform haemocytometry in pilot experiments to confirm single-‐cell suspensions and estimate cell yields per amount of tissue dissected. If working with cell lines, use appropriate scraping or trypsin treatment methods to obtain single-‐cell suspensions. 2 Filter cells through a 70 µm strainer into a 50 ml falcon tube and increase volume to 40 ml with cold FBS/DMEM. 3 Centrifuge cells at 400 g, 4°C, 8 min. Remove supernatant and resuspend cells in 40 ml fix solution. Incubate with gentle shaking on an orbital shaker for 5 min at 23°C. CAUTION: Formaldehyde is toxic. Perform fixation reaction in a fume hood. CRITICAL STEP: Add formaldehyde to FBS/DMEM just before use in fixation. Fixation medium should be at room temperature (23°C). Depending on the tissue used and the specific downstream applications, percentage of formaldehyde and fixation times may need to be varied3. 4 Quench fixation reaction by adding 2.67 ml 2 M cold glycine (0.125 M final concentration) and mix by inverting tube 4-‐5 times. CRITICAL STEP: Keep samples on ice between subsequent centrifugation steps (steps 4-‐7). 5 Centrifuge cells at 400 g, 4°C, 8 min. Remove supernatant and wash cells by resuspending in 50 ml cold PBS. Centrifuge again at 400 g, 4°C, 8 min. CRITICAL STEP: It is important to get a homogeneous suspension for the wash step. One way to achieve this is to initially resuspend the cells by vigorous pipetting with 1 ml PBS, then to add the remaining 49 ml of PBS buffer after resuspension. If cells stick to the pipette tips and numbers are limiting, consider gentle vortexing of the suspension. Cell permeabilization Timing: 30-‐40 min 6 Remove supernatant and resuspend cells in 50 ml cold cell permeabilization buffer. Incubate for 30 min on ice, mixing by inverting the tube once every 10 mins. CRITICAL STEP: Cell permeabilization buffer needs to be made fresh on the day of the experiment and chilled on ice before use. A homogeneous solution is required for efficient permeabilization. One 11
way to achieve this is to initially resuspend the cells by vigorous pipetting with 1 ml permeabilization buffer, then to add the remaining 49 ml of buffer after resuspension. If cells stick to the pipette tips and numbers are limiting, consider gentle vortexing of the suspension. Different tissue types may require different buffers, such as those used for 3C experiments in plant16 and Drosophila24 tissues. TROUBLESHOOTING 7 Collect nuclei by centrifuging at 764 g, 4°C, 5 min. TROUBLESHOOTING PAUSE POINT: The pelleted nuclei can be frozen in liquid nitrogen and stored at -‐80°C for several months. Restriction digestion Timing: 18-‐20 h 8 Resuspend nuclei in 1 ml 1.2 x NEB3 buffer and count with a haemocytometer. Make up 500 µl aliquots of 1 x 107 nuclei aliquots each in 1.2 x NEB3. CRITICAL STEP: One or two aliquots of 1 x 107 nuclei is usually sufficient for standard 3C experiments or one e4C experiment which does not include a ChIP step. If the ChIP step is incorporated, two to three aliquots of 1 x 107 nuclei are often needed per immunoprecipitation. In turn, depending on the immunoprecipitation efficiency, three to six immunoprecipitation experiments may be needed per e4C experiment. CRITICAL STEP: When thawing 10 x NEB3 buffer to make working solution, vortex thoroughly to ensure all DTT is dissolved. If using a restriction enzyme different to BglII, use the compatible restriction buffer at this step. 9 Add 7.5 µl 20% (w/v) SDS (0.3% final concentration) and incubate for 1 h at 37°C, 950 rpm on a thermal mixer. 10 Add 50 µl 20% (v/v) Triton-‐X100 (1.8% final concentration) and incubate for 1 h at 37°, 950 rpm on a thermal mixer. 11 Add 30 µl (1500 U) BglII and incubate overnight at 37°C, 950 rpm on a thermal mixer. If the e4C experiment contains a ChIP step, after 2 h of digestion transfer a 50 µl aliquot to a new 1.5 ml eppendorf tube for subsequent quantification of the input DNA (Box 4) before proceeding directly to step 12. If the ChIP step is omitted, skip to step 23. 12
CRITICAL STEP: Efficient restriction digestion is important, especially around the bait fragment. A method for testing digestion efficiency can be found in Box 2. ChIP preparation steps Timing: 18-‐20 h 12 Prepare Dynabeads protein A-‐coated magnetic beads (100 µl per immunoprecipitation) in 1.5 ml eppendorf tubes. Use magnet to separate beads from storage solution and resuspend in 1 ml 5 mg ml-‐1 BSA/PBS by pipetting gently. Wash the beads a total of three times in 1 ml BSA/PBS in this manner, and leave the beads in the suspension after the final wash. CRITICAL STEP: Time between removing supernatant and adding the new solution should be minimized to prevent the beads from drying out. This is true for every step involving the protein A-‐ coated magnetic beads. 13 Add 3 µg of anti-‐RNA polymerase II antibody (exact volume depends on the concentration of the antibody batch provided) to bead suspension and incubate overnight at 4°C on a rotating wheel. Chromatin immunoprecipitation Timing: 14-‐18 h 14 Wash protein A-‐coated beads (from step 13) three times with 1 ml PBS, using the magnet as per step 12. Resuspend in 1 ml BSA/PBS and store beads at 4°C until step 18. 15 Pool BglII-‐digested chromatin samples (from step 11; aliquot quantified in Box 4) to make the equivalent of 100 µg DNA aliquots per immunoprecipitation in separate 1.5 ml eppendorf tubes. Centrifuge at 14,000 g, 23°C, 1 min and remove supernatant. CRITICAL STEP: For fetal liver nuclei, ~90% of the chromatin remains within the pellet as intact nuclei at this stage. For different tissues, this should be confirmed by DNA purification and quantitation of the supernatant and pellet fractions. If a significant amount of chromatin is in the supernatant, pool greater quantities of digestions to obtain the same amount of input chromatin for the immunoprecipitation step. 16 Resuspend nuclei in 230 µl 1x NEB3 buffer and add 20 µl 20% SDS (1.6% final concentration). Incubate for 30 min at 65°C, 950 rpm on a thermal mixer. CRITICAL STEP: When thawing 10 x NEB3 buffer to make working solution, vortex thoroughly to ensure DTT is fully dissolved. 13
17 Add 150 µl dilution buffer and 600 µl ChIP buffer to the chromatin sample, then remove the nuclear debris by centrifuging at 14,000 g, 4°C for 10 min. Aliquots taken at this stage can be used as input controls for assessment of ChIP efficiency by qPCR (for example, as described in ref 25). CRITICAL STEP: Make dilution buffer and ChIP buffer fresh on the day of the experiment. CRITICAL STEP: The pellet of insoluble chromatin can be quite loose. Take care not to disturb the pellet when taking the soluble chromatin, as contamination with insoluble chromatin reduces the efficiency of the immunoprecipitation. 18 Use a magnet to remove the liquid from the protein A-‐coated beads (from step 14). Resuspend the beads with the supernatant from the BglII-‐digested chromatin (from step 17) and incubate overnight at 4°C on a rotating wheel. ChIP washes and elution Timing: 90 min-‐2 hr 19 Wash the beads three times for 5 min each with 1.5 ml ChIP wash buffer I, incubating on a rotating wheel at 23°C and using a magnet to remove the supernatants at each wash. CRITICAL STEP: Make ChIP wash buffer I fresh on the day of the experiment. 20 Wash the beads three times for 5 min each with 1.5 ml ChIP wash buffer II, incubating on a rotating wheel at 23°C and using a magnet to remove the supernatants at each wash. CRITICAL STEP: Make ChIP wash buffer II fresh on the day of the experiment. 21 Wash the beads three times for 5 min each with 1.5 ml TE buffer, incubating on a rotating wheel at 23°C and using a magnet to remove the supernatants at each wash. 22 Resuspend beads in 50 µl 1 x NEB3 buffer and add 4 µl 20% SDS (1.6% final concentration). Incubate for 30 min at 65°C, 950 rpm on a thermal mixer. Separate beads on magnet and transfer supernatant to a new 2 ml eppendorf tube. CRITICAL STEP: Take care not to contaminate the sample with residual protein A-‐coated beads. Any chromatin still adsorbed may subsequently undergo intermolecular ligation reactions and thus increase the noise of the e4C results. CRITICAL STEP: When thawing 10 x NEB3 buffer to make working solution, vortex thoroughly to ensure DTT is fully dissolved. 14
Ligation and DNA purification Timing: 20-‐24 h 23 Dilute the chromatin and neutralize the SDS, then purify the DNA. Due to the different volumes involved, the procedure is different if a ChIP step is incorporated (option A) or omitted (option B). (A) ChIP-‐e4C (i.) To the eluted chromatin (step 22), add 1 ml 1.1 x T4 DNA ligase buffer and 55 µl 20 % Triton-‐ X100 (1% final concentration). Incubate for 30 min at 37°C. CRITICAL STEP: Final DNA concentration will depend on the amount of chromatin eluted from the immunoprecipitation step. A yield greater than 10% of the input is very unlikely, but would create DNA concentrations greater than the 10 ng µl-‐1 used in e4C experiments without ChIP. If this is a concern, consider increasing the ligation volume. Percentages of non-‐specific ligation can be estimated as shown in Box 5. CRITICAL STEP: When thawing 10 x T4 DNA ligase buffer to make working solution, vortex thoroughly to ensure all DTT is dissolved. (ii.) Add 2 µl (800 U) T4 DNA ligase (400,000 U ml-‐1) and incubate for 4 h at 16°C, then 30 min at 23°C. If testing for restriction digestion efficiency (Box 2), omit T4 DNA ligase from one aliquot (‘no ligase control’) and process with the rest of the ChIP-‐e4C samples to step 24. (iii.) Add 40 µl 5 M NaCl (200 mM final concentration), 25 µl 0.5 M EDTA (12.5 mM final concentration) and 10 µl 10 mg ml-‐1 proteinase K (100 µg ml-‐1 final concentration) and incubate overnight at 65°C. (iv.) Add 2 µl 20 mg ml-‐1 RNase A (40 µg ml-‐1 final concentration) and incubate for 1 h at 37°C. (v.) Add 1 ml phenol/chloroform/isoamyl alcohol and mix by vortexing. Centrifuge at 14,000 g, 23°C, 4 min and transfer aqueous (top) layer to new 2 ml eppendorf tube. CAUTION: Phenol and chloroform are toxic. Handle these chemicals in a fume hood. (vi.) Add 2 µl (40 µg) glycogen and 700 µl isopropanol and mix by vortexing. Centrifuge at 14,000 g, 4°C, 30 min. (vii.)
Remove supernatant carefully, wash pellet with 500 µl 70% ethanol and centrifuge at
14,000 g, 4°C, 5 min. (viii.)
Remove supernatant, air-‐dry pellet and dissolve DNA in 100 µl molecular biology-‐
grade water. (B) e4C without ChIP
15
(i.) Add 50 µl 20% SDS (1.6% final concentration) to the digested chromatin (from step 11) and incubate for 30 min at 65°C, 950 rpm on a thermal mixer. (ii.) Transfer chromatin to a 15 ml falcon tube and add 7 ml 1.1 x T4 DNA ligase buffer and 400 µl 20% Triton-‐X100 (1% final concentration), making a final DNA concentration of ~10 ng µl-‐1. Incubate for 1 h at 37°C. CRITICAL STEP: When thawing 10 x T4 DNA ligase buffer to make working solution, vortex thoroughly to ensure all DTT is dissolved. (iii.) Add 2 µl (800 U) T4 DNA ligase (400,000 U ml-‐1) and incubate for 4 h at 16°C, then 30 min at 23°C. If testing for restriction digestion efficiency (Box 2), omit T4 DNA ligase from one aliquot (‘no ligase control’) and process with the rest of the e4C samples to step 24. (iv.) Add 70 µl 10 mg ml-‐1 proteinase K (100 µg ml-‐1 final concentration) and incubate overnight at 65°C. (v.) Add 14 µl 20 mg ml-‐1 RNase A(40 µg ml-‐1 final concentration) and incubate for 1 h at 37°C. (vi.) Transfer solution to a new 50 ml falcon tube and add 10 ml phenol, pH 8. Mix thoroughly by vortexing and centrifuge at 2900 g, 23°C, 15 min. Transfer aqueous (top) layer to a new 50 ml falcon tube. CAUTION: Phenol is toxic; handle in a fume hood. (vii.)
Add 10 ml chloroform and mix thoroughly by vortexing. Centrifuge at 2900 g, 23°C,
15 min. Transfer aqueous (top) layer to a new 50 ml falcon tube. CAUTION: Chloroform is toxic; handle in a fume hood. (viii.)
Add 700 µl 3 M sodium acetate, pH 5.2, and 17.5 ml 100% ethanol. Mix vigorously by
vortexing and incubate at -‐20°C for 1 h. (ix.) Centrifuge at 2900 g, 4°C, 1 h. Remove supernatant and vortex pellet with 10 ml 70% ethanol. (x.) Centrifuge at 2900 g, 4°C, 30 min. Remove supernatant, then air-‐dry pellet at 37°C for 5 min. Dissolve pellet by pipetting vigorously with 100 µl molecular biology-‐grade water and incubate at 37°C for 1 h. Transfer DNA to a new 1.5 ml eppendorf tube. TROUBLESHOOTING 24 Quantify 3C DNA with Quant-‐iT PicoGreen dsDNA or Qubit dsDNA BR assay. Approximately 10 µg DNA is obtained from one 3C experiment (taken from one digestion reaction of 107 nuclei); the yields from ChIP-‐e4C experiments will vary according to antibody efficiency. DNA at this stage can be used for quality controls, such as qPCR assessment of ChIP efficiency and detection of known 3C products (see Anticipated Results). ‘No ligase control’ aliquots can be used to assess restriction digestion efficiency (Box 2; see Supplementary Table 1 for primer sequences). 16
CRITICAL STEP: Standard A260 measurements are not reliable for quantification of 3C material. These fluorimetric assays are preferred. TROUBLESHOOTING PAUSE POINT: 3C DNA can be stored at -‐20°C for several weeks. Primer extension Timing: 14-‐18 h 25 Mix primer extension reaction components on ice in PCR tubes, in 50 µl reaction volumes: Component
Amount per reaction (µl)
Final
10 x ThermoPol buffer
5
1 x
2 mM dNTPs
5
200 nM
10 µM biotinylated Hbb-‐b1
0.5
5 pmol
38.5 (total)
100 ng DNA if ChIP step is
primer Molecular biology-‐grade water plus 3C DNA from step 24
included; 500 ng DNA if ChIP step is omitted
Vent (exo-‐) DNA polymerase
1
2 U
CRITICAL STEP: 12-‐15 primer extension reactions are routinely required to generate enough e4C material for one microarray hybridization reaction. 26 Run the primer extension reaction in a thermal cycler: Cycle number
Denature
Anneal
Extend
1
95°C, 4 min
60°C, 2 min
72°C, 10 min
Place the reactions on ice immediately and leave on ice for 5 min. CRITICAL STEP: The fidelity of the primer extension reaction is crucial for e4C. The melting temperature needs to be optimized for each primer (Box 3). Furthermore, to prevent primer mis-‐ annealing, the reactions should be kept on ice until the thermal cycler has reached the denaturation temperature, and the reactions should be put on ice immediately once the program has completed, without allowing it to cool slowly. 27 Add 2 µl (20 U) NlaIII and incubate overnight at 37°C. 17
CRITICAL STEP: NlaIII is an unstable enzyme and should be stored at -‐80°C. If using a different secondary enzyme to NlaIII, check that it is active in the ThermoPol buffer of the primer extension reaction. If not, the material will need to first be diluted in the compatible buffer. Purification on streptavidin beads Timing: 3-‐4 h 28 Remove excess biotinylated primer by purifying the DNA with a QiaQuick PCR purification kit, following the manufacturer’s instructions. Elute the DNA with 50 µl EB buffer (elution buffer; 10 mM Tris-‐HCl; provided with the kit). 29 Mix Dynabeads M-‐280 streptavidin-‐coated magnetic beads stock thoroughly and transfer 20 µl beads (200 µg) to a new 1.5 ml eppendorf tube (one tube per primer extension reaction). Remove storage solution with a magnet and wash twice with 50 µl binding buffer provided in the Dynabeads kilobaseBINDER kit. Resuspend beads in 50 µl kilobaseBINDER binding buffer. CRITICAL STEP: Time between removing supernatant and adding the new solution should be minimized to prevent the beads from drying out. This is true for every step involving the streptavidin-‐ coated beads. The kilobaseBINDER binding buffer contains detergent. To optimize binding, pipette solutions very carefully to keep air bubbles to a minimum. 30 Add 50 µl purified primer extension products from step 28 to the 50 µl of streptavidin-‐coated beads suspension and incubate for 3 h at 23°C, 1200 rpm on a thermal shaker. 31 Wash beads twice with 100 µl SBWB1 (streptavidin beads wash buffer 1), then once with 100 µl SBWB2 (streptavidin beads wash buffer 2), using the magnet to remove the solutions at each wash. Aliquots of material at this stage can be used to assess bait enrichment (Box 1). Re-‐digestion Timing: 14-‐18 hr 32 Use magnet to remove SBWB2 and resuspend beads in 50 µl freshly prepared NlaIII reaction solution. Incubate overnight at 37°C, 1200 rpm on thermal shaker. CRITICAL STEP: Make NlaIII reaction solution fresh on the day of the experiment. When thawing 10 x NEB4 buffer to make working solution, vortex thoroughly to ensure DTT is fully dissolved.
18
Adapter ligation Timing: 3-‐4 h 33 Wash beads twice with 100 µl SBWB1 then once with 100 µl SBWB2, using the magnet to remove the solutions at each wash. 34 Use magnet to remove SBWB2 and resuspend beads in 40 µl 1 x T4 DNA ligase buffer. Incubate at 50°C for 5 min, then immediately place on ice for 5 min. In parallel, do the same heating and chilling treatment to the 100 µM e4C adapter stock. CRITICAL STEP: This treatment denatures the NlaIII cohesive ends for more efficient ligation of adapter. Too high temperatures or incubation times may completely denature the e4C adapter. CRITICAL STEP: When thawing 10 x T4 DNA ligase buffer to make working solution, vortex thoroughly to ensure DTT is fully dissolved. 35 Use magnet to remove DNA ligase buffer and resuspend beads in 40 µl freshly prepared adapter ligation reaction solution. Incubate for 4 hr at 37°C, 750 rpm on a thermal shaker. CRITICAL STEP: Make adapter ligation reaction solution fresh on the day of the experiment. When thawing 10 x T4 DNA ligase buffer to make working solution, vortex thoroughly to ensure DTT is fully dissolved. 36 Wash beads twice with 100 µl SBWB1 then once with 100 µl SBWB2, using the magnet to remove the solutions at each wash. PCR amplification Timing: 2-‐4 h 37 Use magnet to remove SBWB2 and resuspend beads in 50 µl freshly prepared PCR reaction solution, as tabulated below. Transfer each 50 µl reaction solution to individual PCR tubes. Component
Amount per reaction (µl)
Final
Molecular biology-‐grade water
35
10 µM e4C adapter-‐specific
2
400 nM
2
400 nM
2 mM dNTPs
5
200 µM
10 x HotStar buffer
5
1x
primer 10 µM bait-‐specific primer (e.g. nested Hbb-‐b1 primer)
19
5 U µl-‐1 HotStar Taq DNA
1
5U
polymerase CRITICAL STEP: Make PCR reaction solution fresh on the day of the experiment. 38 Run the reaction in a thermal cycler, using the program tabulated below: Cycle number
Denature
Anneal
Extend
1
95°C, 15 min
2-‐36
94°C, 30 s
55°C, 30 s
72°C, 1 min
37
72°C, 10 min
CRITICAL STEP: The melting temperatures need to be optimized for each bait-‐specific primer (Box 3). 39 Transfer reaction solutions to new 1.5 ml eppendorf tubes and use a magnet to separate the beads from the supernatant. Pool the supernatants from four PCR reactions to new 1.5 ml eppendorf tubes. 40 Purify the DNA with a QiaQuick PCR purification kit, following the manufacturer’s instructions. Elute the DNA with 50 µl EB buffer (elution buffer; 10 mM Tris-‐HCl; provided with the kit). 41 Assess e4C product yield and quality by measurement of A260, A280 and A230 with a NanoDrop spectrometer. Expected yield from 12-‐15 pooled primer extension reactions is ~5 µg DNA. TROUBLESHOOTING PAUSE POINT: e4C products can be stored at -‐20°C for several weeks. BglII/NlaIII digestion Timing: 6-‐7 h 42 Pool all purified e4C products to one 1.5 ml eppendorf tube (~150-‐250 µl total volume) and make up to 300 µl in 1 x NEB4 buffer, 100 µg ml-‐1 BSA and 50 U NlaIII. Incubate at 37°C for 2 h. To make control material for competitive hybridization with the e4C products, perform the same reaction in parallel using ~20 µg mouse genomic DNA and perform the same steps (steps 42-‐47) on this control material.
20
CRITICAL STEP: This step is only required if the e4C products are going to be analyzed by microarray hybridization. Efficient BglII and NlaIII digestion is necessary to cleave unknown 3C-‐ligated sequences from constant bait and adapter sequences before microarray hybridization. As these constant bait and adapter sequences are present on every e4C product, cross-‐hybridization between e4C products which maintain the constant sequences may reduce efficiency of their hybridization to the microarray probes. Sequential NlaIII then BglII digestion is required to provide compatible buffers for digestion. If using different enzymes to BglII and NlaIII, use the correct buffer for efficient double digestion in one step, or perform sequential digestion in appropriate buffers. 43 Add 235 µl molecular biology-‐grade water, 60 µl 10x NEB3 buffer and 5 µl (250 U) BglII, and incubate at 37°C for 2 h. CRITICAL STEP: When thawing 10 x NEB3 buffer to make working solution, vortex thoroughly to ensure DTT is fully dissolved. 44 Add 600 µl phenol/chloroform and mix by vortexing. Centrifuge at 14,000 g, 23°C, 4 min and transfer aqueous (top) layer to new 2 ml eppendorf tube. CAUTION: Phenol and chloroform are toxic; handle in a fume hood. 45 Add 60 µl 3 M sodium acetate, pH 5.2, and 1300 µl 100% ethanol and mix by vortexing. Incubate at -‐20°C for 1 h, then centrifuge at 14,000 g, 4°C, 30 min. 46 Carefully remove supernatant and wash DNA pellet by vortexing with 300 µl 70% ethanol. Centrifuge at 14,000 g, 4°C, 5 min. Remove supernatant and air-‐dry pellet, before dissolving in 30 µl molecular biology-‐grade water. 47 Assess e4C product and control-‐digested genomic DNA yields and qualities by measurement of A260, A280 and A230 with a NanoDrop spectrophotometer. Material is now ready for in-‐house labeling and microarray hybridization by NimbleGen Systems. CRITICAL STEP: All microarray data should be compliant with the MIAME (minimal information for microarray experiments) guidelines when deposited in public databases18. Ensure that all background information on probe and experimental design is recorded and organized accordingly. TIMING
21
The exact timings of the e4C steps depend on whether a ChIP step is incorporated or omitted, and which controls are included. Overall, 3C takes three days and ChIP-‐3C takes four days to perform, respectively. Once the 3C or ChIP-‐3C DNA has been validated, the e4C procedure takes an additional three days to perform, with an extra day required for digestion of the material, prior to processing for microarray hybridization. Downstream hybridization and analysis step timings will vary. See Fig. 2 for more details. Steps 1-‐5: Single-‐cell preparation and fixation: 30 min-‐1 h Steps 6-‐7: Cell permeabilisation: 30-‐40 min Steps 8-‐11: Restriction digestion: 18-‐20 h Steps 12-‐13: ChIP preparation steps: 18-‐20 h Steps 14-‐18: Chromatin immunoprecipitation: 14-‐18 h Steps 19-‐22: ChIP washes and elution: 90 min-‐2 h Steps 23-‐24: Ligation and DNA purification: 20-‐24 h Steps 25-‐27: Primer extension: 14-‐18 h Steps 28-‐31: Purification on streptavidin beads: 3-‐4 h Step 32: Re-‐digestion: 14-‐18 h Steps 33-‐36: Adapter ligation: 3-‐4 h Steps 37-‐41: PCR amplification: 2-‐4 h Steps 42-‐47: BglII/NlaIII digestion: 6-‐7 h Box 1: Assessing bait enrichment by e4C: 4-‐6 h Box 2: Assessment of 3C restriction digestion efficiency: 2-‐3 h Box 3: e4C primer testing and optimization: 3 days Box 4: Quantification of ChIP input DNA: 18-‐20 h Box 5: Quantifying ChIP-‐e4C non-‐specific ligation events: 7 days 22
TROUBLESHOOTING The efficiency of the 3C and ChIP steps are best assessed by PCR for known positive and negative controls (see Anticipated Results). See Table 1 for troubleshooting details. Troubleshooting of 3C (ref 3) and ChIP (ref 25) experiments are also outlined in other published protocols. ANTICIPATED RESULTS Some expected e4C results can be demonstrated by following the example of the mouse beta-‐globin gene, Hbb-‐b1, in fetal liver nuclei1. 3C DNA (at step 24), with or without incorporation of a ChIP step using an antibody recognizing the active form of RNA polymerase II, is validated by two means. First, the known distal intrachromosomal co-‐associations between Hbb-‐b1 and the erythroid-‐expressed genes Ahsp (alpha-‐globin stabilizing protein, formerly called Eraf) and Uros (uroporphyrinogen III synthase), are qualitatively confirmed by PCR26 (Fig. 3a; primers given in Supplementary Table 2). These products are observed in fetal liver tissue but not kidney, and an interaction with the non-‐ expressed gene, P2ry6 (pyrimidinergic receptor P2Y6), which is closer in chromosomal distance to Hbb-‐b1 than Eraf or Uros, is not detected in either tissue. Ligation products between two adjacent BglII fragments within the Calr (calreticulin) gene serve as a positive control for the digestion and ligation steps of 3C, and are detected in both fetal liver and kidney tissues (Fig. 3b; primers given in Supplementary Table 2).As the second quality control, the enrichment of actively expressed genes by the ChIP procedure is assessed by qPCR1 (Fig. 3c; primers given in Supplementary Table 3). The promoter/first exon regions of the expressed genes Hba-‐a1 (alpha-‐globin), Slc4a1 (Band III anion exchange protein) and Ahsp are highly enriched compared to the non-‐expressed region within the immunoglobulin heavy-‐chain locus (VH16).
Having established the quality of the 3C or ChIP-‐3C material, two controls are then
performed to assess the efficiency and fidelity of the e4C primers and reagents. Firstly, the amount of Hbb-‐b1 bait sequence is quantified by qPCR on 3C material before and after e4C primer extension, streptavidin pull-‐down and elution by BfaI digestion (Fig. 3d; see Box 1 for details and Supplementary Table 4 for primer sequences). Three different picogram quantities of 3C and e4C-‐ processed material were used as templates for qPCR, with quantification compared to a standard curve of genomic DNA of known concentrations, and at least 100-‐fold bait enrichment was detected each time for the e4C-‐processed material. The final control of the e4C primers is to perform e4C on a genomic DNA template, instead of 3C or ChIP-‐3C material (Fig. 3e; see Box 3 for details). e4C normally produces a smear of products corresponding to multiple different 3C ligation partners with 23
the Hbb-‐b1 bait fragment, but e4C with genomic DNA template produces a single band, confirmed by cloning and sequencing to correspond to the expected 239 bp product (72 bp nested e4C primer to BglII site + 130 bp BglII-‐NlaIII fragment directly downstream of the bait in the contiguous sequence + 37 bp adapter sequence; Fig. 1b and Fig. 3e). Importantly, no products are obtained from the e4C PCR step using only the adapter-‐specific primer and omitting the nested bait-‐specific primer, indicating that the e4C material is not contaminated with spurious genomic NlaIII fragments that have obtained adapter sequences on both ends.
The e4C material generated (at step 41) can then be processed by cloning and sequencing
(Fig. 4a) or BglII/NlaIII digestion (steps 42-‐47) and microarray hybridization (Fig. 4b-‐e), depending on the desired information. 4847 out of 8377 (58%) sequenced and uniquely mapped e4C clones located within 80 kb of the Hbb locus; the remainder were sparsely distributed throughout the genome and gave little information. However, known looped interactions between Hbb-‐b1 and upstream hypersensitive sites within the locus control region (LCR)27,28 are apparent when looking at e4C sequences within the Hbb locus (Fig. 4a). Conversely, signals are saturated within a few hundred kilobases of the Hbb locus when e4C material is hybridized to a microarray, and such local chromatin loops are not observed (Fig. 4b). We note that sequencing thousands of clones is not a cost-‐effective way to analyze chromatin topology, and modifications of the e4C protocol allowing compatibility with high-‐throughput Illumina sequencing have recently been reported17. However, insufficient sequencing depth was obtained in this study to assess chromatin interactions genome-‐wide. For long-‐range cis (Fig. 4c) and interchromosomal (Fig. 4d-‐e) chromatin interactions, microarrays typically reveal clusters of interacting sequences, spanning a few tens to a few hundred kilobases. These clusters are easier to visualize when running means of the e4C hybridization signal are plotted (Fig. 4c-‐e), but are also apparent when looking at the raw data (Supplementary Fig. 1). The genes highlighted in Fig. 4 have been shown to significantly co-‐associate with the Hbb-‐b1 gene by RNA FISH1,26, and can be systematically identified by a variety of statistical analyses. e4C and ChIP-‐e4C microarray profiles appear quite similar in this particular study (Fig. 4); by inspection, incorporation of the ChIP step for active RNA polymerase II appears to slightly boost hybridization signals within the interacting clusters of microarray probes. However, the globin genes are very highly transcribed in erythroid tissues and may be predicted to constantly be in an RNA polymerase II-‐associated nuclear environment. e4C and ChIP-‐e4C profiles may be considerably more different for other bait sequences or antibodies used in the ChIP step. Overall, e4C affords sensitive bait-‐specific detection of chromatin interaction partners. The use of sequencing versus microarrays, or the incorporation of a ChIP step, allows different aspects of these interactions to be further explored. 24
ACKNOWLEDGEMENTS This work was supported by the Medical Research Council, the Biotechnology and Biological Sciences Research Council, and by a long-‐term EMBO fellowship to D.U. AUTHOR CONTRIBUTIONS T.S., S.K. and P.F. designed the experiments, T.S. and S.K. developed the method, J.A.M. and D.U. optimized the chromatin immunoprecipitation steps, T.N. developed primers for restriction digestion efficiency tests, and T.S. and P.F. wrote the manuscript. COMPETING FINANCIAL INTERESTS The authors declare no competing financial interests. BOXES Box 1. Assessing bait enrichment by e4C Timing: 4-‐6 h The amount of bait sequence is quantified by qPCR, before and after enrichment by biotinylated primer extension and streptavidin pull-‐down. A qPCR primer pair needs to be designed within the bait fragment, between the biotinylated primer sequence and the downstream 3C restriction site. A restriction enzyme also needs to be found which cuts just once within this region, between the biotinylated primer sequence and the qPCR primers (see Fig 1b). For the Hbb-‐b1 bait described in this protocol, the primers are given in Supplementary Table 4 and the eluting enzyme is BfaI. (i.) Take one eppendorf tube at step 31 (one aliquot of 3C or ChIP-‐3C material that has undergone the primer extension and streptavidin purification steps of e4C), remove supernatant with magnet and resuspend in 50 µl elution digestion mixture (20 U BfaI in 1 x NEB4 buffer; New England Biolabs). Incubate for 2 h at 37°C, 1200 rpm on a thermal mixer. (ii.) Remove the eluted DNA from the beads with magnet and add 50 µl molecular biology-‐grade water and 100 µl phenol/chloroform. Mix by vortexing. Centrifuge at 14,000 g, 23°C, 4 min and transfer aqueous (top) layer to a new 1.5 ml eppendorf tube. CAUTION: Phenol and chloroform are toxic. Handle chemicals in a fume hood.
25
(iii.) Add 10 µl 3 M sodium acetate, pH 5.2, and 250 µl 100% ethanol and mix by vortexing. Incubate at -‐20°C for 1 h, then centrifuge at 14,000 g, 4°C, 30 min. (iv.) Carefully remove supernatant and wash DNA pellet by vortexing with 300 µl 70% ethanol. Centrifuge at 14,000 g, 4°C, 5 min. Remove supernatant and air-‐dry pellet, before dissolving in 20 µl molecular biology-‐grade water. (v.) Quantify DNA with the Quant-‐iT PicoGreen dsDNA or Qubit dsDNA BR assay, following the manufacturer’s instructions. (vi.) With compatible SYBR Green master mixes to the qPCR machine being used, make up triplicate qPCR reaction volumes containing 200 nM each primer. As DNA templates, use a serial dilution of 100 pg to 4 pg eluted e4C material, and the same picogram quantities of the 3C input material (from step 24). Serial dilutions of known amounts of genomic material can also be used to generate standard curves. (vii.)
Run the qPCR reactions, according to the manufacturer’s instructions. The fold
enrichment from e4C purification can be estimated by comparing the amount of bait sequence in equivalent picogram quantities of DNA before and after the enrichment (see Anticipated Results). Box 2. Assessment of 3C restriction digestion efficiency Timing: 2-‐3 h Restriction digestion efficiency is assessed by qPCR quantitation of genomic regions spanning sites for the restriction enzyme used in the 3C reaction. One genomic region that does not span a restriction site is also used to normalize the reaction inputs. Example primers for assessing BglII digestion in mouse tissues are given in Supplementary Table 1. (i.) With compatible SYBR Green master mixes to the qPCR machine being used, make up triplicate qPCR reaction volumes containing 2 ng ‘no ligase control’ 3C DNA aliquots (see steps 23 and 24) and 200 nM each primer. Serial dilutions of known amounts of genomic material (typically 10 to 0.01 ng per reaction) can also be used to generate standard curves. (ii.) Run the qPCR reactions, according to the manufacturer’s instructions. The amount of input DNA, I, is derived from the quantitation of the genomic region not spanning a restriction site. (iii.) For each restriction site tested, the quantitation of the amplicon, U, derives from undigested DNA. The percentage of digestion from this site, D, is thus given as: D = 100 x (1 -‐ (U/I)) Digestion efficiencies should typically be more than 70%. 26
Box 3. e4C primer testing and optimization Timing: 3 days If the e4C primers work specifically, then performing steps 25-‐41 of the main Procedure on genomic DNA template instead of 3C material should generate only one kind of product: the BglII-‐NlaIII fragment directly downstream of the bait fragment in the contiguous genomic sequence, flanked by the bait fragment and the adapter sequence (see Anticipated Results). Before using new e4C primers, they should thus be tested in trial reactions on genomic DNA templates, trialing different annealing temperatures for the primer extension (step 26) and PCR (step 38) steps. The products should then yield single bands in agarose gel electrophoresis, as opposed to the smears of product produced in real e4C experiments (see Anticipated Results). These single bands should be excised, cloned and sequenced to verify that they correspond to expected sequences. Box 4. Quantification of ChIP input DNA Timing: 18-‐20 h Total DNA from an aliquot of chromatin after 2 h of digestion (step 11) is purified for quantitation of input material. This is necessary to ensure that controlled amounts of input chromatin are used for the immunoprecipitations (step 15). (i.) Add 1 µl 2 mg ml-‐1 RNase A (diluted from 20 mg ml-‐1 stock; 40 µg ml-‐1 final concentration) to the chromatin aliquot (from step 11) and incubate for 1 h at 37°C. (ii.) Add 48.5 µl molecular biology-‐grade water and 100 µl phenol/chloroform/isoamyl alcohol and mix by vortexing. Centrifuge at 14,000 g, 23°C, 4 min and transfer aqueous (top) layer to a new 1.5 ml eppendorf tube. CAUTION: Phenol and chloroform are toxic. Handle these chemicals in a fume hood. (iii.) Add 10 µl 3 M sodium acetate, pH 5.2, and 250 µl 100% ethanol and mix by vortexing. Incubate at -‐20°C for 1 h, then centrifuge at 14,000 g, 4°C, 30 min. (iv.) Carefully remove supernatant and wash DNA pellet by vortexing with 300 µl 70% ethanol. Centrifuge at 14,000 g, 4°C, 5 min. Remove supernatant and air-‐dry pellet, before dissolving in 50 µl molecular biology-‐grade water. (v.) Quantify DNA with the Quant-‐iT PicoGreen dsDNA or Qubit dsDNA BR assay, following the manufacturer’s instructions. Assume that the total amount of BglII-‐digested chromatin (from step 11) contains nine times the amount of DNA calculated to be present in the aliquot. CRITICAL STEP: Standard A260 measurements are not reliable for quantification of 3C material. These fluorimetric assays are preferred. 27
Box 5. Quantifying ChIP-‐e4C non-‐specific ligation events TIMING: 7 days Non-‐specific ligation events between chromatin fragments that are not covalently linked during formaldehyde fixation will increase noise in e4C experiments and should be avoided. To assess such spurious ligation, perform the 3C digestion procedure (steps 1-‐17) separately with equal numbers of nuclei of the tissue used in the experiment (e.g. mouse fetal liver), and with an unrelated tissue in a different species, whose genome sequence is available and for which the antibody has comparable efficiency (e.g. human U2OS cells). At step 18 of the main Procedure, mix equal volumes (500 µl each) of the two digested samples and combine in the same immunoprecipitation. Proceed with e4C (steps 18-‐41) and clone and sequence ~100 e4C products. Non-‐specific ligation frequency can be gauged by the number of ligation products detected between the e4C bait and sequences derived from the heterologous species, and was found to be ~1% in a previous study1. FIGURE LEGENDS Figure 1. Overview of the e4C procedure. (a) The e4C protocol. (i) 3C or ChIP-‐3C material containing a pool of hybrid DNA sequences ligated at BglII (B) sites undergoes primer extension with a biotinylated primer (red arrow) complementary to sequence on a specific bait fragment (red). (ii) Bait sequences tagged by biotin (red circle) are specifically captured on streptavidin beads (grey sphere). In the illustrated example, the black bar denotes an uncharacterized BglII fragment which is ligated to the bait fragment. (iii) Bead-‐captured bait sequences are digested with NlaIII and an adapter with a complementary cohesive end (cyan bar) is ligated to the NlaIII (N) site, before PCR using a nested bait-‐specific primer (red arrow) and an adapter-‐specific primer (cyan arrow). (iv) The PCR products are digested with BglII and NlaIII, generating a pool of bait fragments (red bars), adapter fragments (cyan bars) and uncharacterized fragments which formed 3C ligation products with the bait sequence (black bars). This e4C library is then hybridized to a custom-‐made microarray. (b) Schematic of the Hbb-‐b1 e4C bait fragment (red) and surrounding sequences (black), as used in a previous study1, shown to scale. The biotinylated and nested primers (black arrows) are contained on the same bait BglII-‐NlaIII fragment. The red arrows denote the positions of the qPCR primers when measuring bait enrichment after elution of e4C material by digesting with BfaI, which cuts uniquely within the bait fragment at the denoted position (see Box 1). The adjacent BglII-‐NlaIII fragment in the contiguous genomic sequence is sufficiently large to allow unambiguous mapping when sequenced during trials of e4C primer fidelity (see Box 3). 28
Figure 2. Flowchart of e4C timings. Rectangular boxes denote the steps in the e4C procedure which take one day to perform. Red text denotes extra steps if ChIP is incorporated into the procedure. Hexagons show the time-‐points for control and validation steps; further details are given in the denoted Boxes. Figure 3. Examples of quality controls for e4C. a) Gel showing specific 3C ligation products between the Hbb-‐b1 gene and the genes Ahsp or Uros when PCR is performed on 3C (E) or ChIP-‐3C (Ch; using an antibody for active RNA polymerase II) material generated from erythroid tissue. These erythroid-‐ specific interactions are not observed when 3C is performed on kidney (K); an interaction between the genes Hbb-‐b1 and P2ry6 is not observed in either tissue. The no-‐template control for the PCR is denoted (H2O). b) The orientations of the 3C control Calr primers (mapping to adjacent BglII fragments) are shown above a gel, which shows the corresponding ligation products when PCR is performed on 3C material generated from erythroid (E) or kidney (K) tissue. The no-‐template control for the PCR is denoted (H2O). c) qPCR results for ChIP enrichments of specific erythroid-‐expressed genes (Hba-‐a1, Slc4a1, Ahsp) and a negative control (VH16) when ChIP-‐3C is performed on erythroid tissue, using an antibody for active RNA polymerase II. RNA polymerase II binding is scored as fold-‐ enrichment over VH16. d) qPCR quantification, in arbitrary units, of bait sequence before (black bars) and after (white bars) 3C material has undergone the primer extension and streptavidin pull-‐down steps of e4C. Three different qPCR template quantities are shown, and in each case, ~100-‐fold bait enrichment is observed. Error bars denote the standard error of the mean from triplicate qPCR reactions. e) Gel of e4C products when the procedure is applied on 3C material (3C) or genomic DNA (G) as template. The lane marked A shows the lack of product when the nested bait-‐specific primer is omitted from the final e4C PCR step and only the adapter primer is used. The no-‐template control for the PCR is denoted (H2O). Figure 4. Typical e4C and ChIP-‐e4C results, using Hbb-‐b1 sequence as bait. a) Profile of the Hbb locus revealed by cloning and sequencing of e4C material. Bars indicate numbers of e4C clones mapping to each BglII fragment, drawn to scale; black bar represents the BglII fragment containing the Hbb-‐b1 bait primer (position indicated by arrow under the profile). Schematic of the Hbb locus is shown underneath the profile, with DNase-‐hypersensitive sites (vertical arrows), expressed globin genes (red rectangles), silent embryonic globin genes (grey rectangles) and olfactory receptor genes 29
(black rectangles). b) e4C microarray profile showing e4C enrichments over genomic control for probes within a ~365 kb window spanning the Hbb locus. The arrow denotes the position of the e4C bait primer; the genomic orientation has been flipped for easier comparison with the results in (a). Black bars underneath denote the positions of genes, and the positions of the LCR and some of the Hbb genes are marked. c-‐e) e4C (red) and ChIP-‐e4C (blue; using an antibody for the active form of RNA polymerase II) microarray profiles, showing the running means of e4C hybridization signal enrichment over genomic control for 100 kb windows, centered on c) the intrachromosomal co-‐ association with Uros (~650 kb window); d) the interchromosomal co-‐association with Epb4.9 and Xpo7 (~510 kb window); e) the interchromosomal co-‐association with Epb4.1 (~470 kb window). Black bars underneath denote the positions of genes; marked are the genes whose co-‐associations with Hbb-‐b1 have been characterized by RNA FISH1. TABLES Table 1: Troubleshooting Step
Problem
Possible cause
6
Nuclei are aggregated
7
No visible pellet
23
DNA precipitate does not resuspend
24
Inefficient digestion from test (Box 2)
24
Low ChIP-‐3C DNA yield
Sub-‐optimal resuspension Pipette up and down more vigorously conditions while trying to avoid air bubbles. If the cells are very difficult to handle, consider using a different lysis buffer or including a douncing step Insufficient numbers of Start from step 1 with more tissue nuclei Pellets over-‐dried Dissolve DNA by incubating at 37°C, 950 rpm for 30 min on a thermal shaker Sub-‐optimal conditions Ensure restriction buffer is compatible for the restriction enzyme with the enzyme being used. Try varying the SDS/Triton X-‐100 concentrations before digestion and/or reducing the formaldehyde percentage at fixation. Many enzymes are not suitable for 3C, so the choice of enzyme may also need to be reconsidered Antibody is of low affinity Increase antibody concentration in to antigen immunoprecipitation, or consider changing antibody supplier or batch. Reduce the stringency of washing steps by reducing salt concentrations of ChIP wash buffers I and II
30
Solution
Antibody is of low affinity to protein A Washes not stringent enough
24
High qPCR signal from ChIP negative control
41
Non-‐specific e4C products from test (Box 3)
Sub-‐optimal PCR conditions
41
Low e4C or ChIP-‐e4C DNA yields
Poor binding of primer-‐ extended material to streptavidin beads
Inefficient adapter ligation
‘Negative’ control actually bound by immunoprecipitated protein Sub-‐optimal primer extension conditions
Check that antibody should not be used with protein G beads instead Increase stringency of washes by increasing salt concentrations of ChIP wash buffers I and II Design new qPCR primers for different negative control regions
Alter primer extension annealing temperature. Ensure reaction mixtures are kept on ice until thermal cycler has reached denaturation temperature, and are placed on ice immediately once the reaction is completed. Consider designing new biotinylated primers Alter PCR annealing temperature. Consider designing new bait-‐specific PCR primers Ensure there are no air bubbles in bead binding solutions. Consider changing batch of Dynabeads M-‐280 Streptavidin beads Ensure that adapter is not denatured by excessive temperatures or by repeated freeze-‐thawing
SUPPLEMENTARY INFORMATION Supplementary Figure 1. Typical raw e4C and ChIP-‐e4C results, using Hbb-‐b1 sequence as bait. Supplementary Table 1. List of primers used for assessing BglII restriction digestion efficiency in mouse tissues. Supplementary Table 2. List of 3C primers used in Anticipated Results section. Supplementary Table 3. List of ChIP primers used in Anticipated Results section. Supplementary Table 4. List of primers used for assessing e4C bait enrichment in Anticipated Results section. REFERENCES 31
1. 2. 3. 4. 5. 6. 7. 8.
9. 10. 11. 12.
13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23.
Schoenfelder, S. et al. Preferential associations between co-‐regulated genes reveal a transcriptional interactome in erythroid cells. Nat Genet 42, 53-‐61 (2010). Dekker, J., Rippe, K., Dekker, M. & Kleckner, N. Capturing chromosome conformation. Science 295, 1306-‐11 (2002). Hagege, H. et al. Quantitative analysis of chromosome conformation capture assays (3C-‐ qPCR). Nat Protoc 2, 1722-‐33 (2007). Sexton, T., Bantignies, F. & Cavalli, G. Genomic interactions: chromatin loops and gene meeting points in transcriptional regulation. Semin Cell Dev Biol 20, 849-‐55 (2009). Lomvardas, S. et al. Interchromosomal interactions and olfactory receptor choice. Cell 126, 403-‐13 (2006). Simonis, M. et al. Nuclear organization of active and inactive chromatin domains uncovered by chromosome conformation capture-‐on-‐chip (4C). Nat Genet 38, 1348-‐54 (2006). Wurtele, H. & Chartrand, P. Genome-‐wide scanning of HoxB1-‐associated loci in mouse ES cells using an open-‐ended Chromosome Conformation Capture methodology. Chromosome Res 14, 477-‐95 (2006). Zhao, Z. et al. Circular chromosome conformation capture (4C) uncovers extensive networks of epigenetically regulated intra-‐ and interchromosomal interactions. Nat Genet 38, 1341-‐7 (2006). Ohlsson, R. & Gondor, A. The 4C technique: the 'Rosetta stone' for genome biology in 3D? Curr Opin Cell Biol 19, 321-‐5 (2007). Hakim, O. et al. Diverse gene reprogramming events occur in the same spatial clusters of distal regulatory elements. Genome Res 21, 697-‐706 (2011). Noordermeer, D. et al. Variegated gene expression caused by cell-‐specific long-‐range DNA interactions. Nat Cell Biol 13, 944-‐51 (2011). Dostie, J. et al. Chromosome Conformation Capture Carbon Copy (5C): a massively parallel solution for mapping interactions between genomic elements. Genome Res 16, 1299-‐309 (2006). Lieberman-‐Aiden, E. et al. Comprehensive mapping of long-‐range interactions reveals folding principles of the human genome. Science 326, 289-‐93 (2009). Fullwood, M.J. et al. An oestrogen-‐receptor-‐alpha-‐bound human chromatin interactome. Nature 462, 58-‐64 (2009). Horike, S., Cai, S., Miyano, M., Cheng, J.F. & Kohwi-‐Shigematsu, T. Loss of silent-‐chromatin looping and impaired imprinting of DLX5 in Rett syndrome. Nat Genet 37, 31-‐40 (2005). Louwers, M., Splinter, E., van Driel, R., de Laat, W. & Stam, M. Studying physical chromatin interactions in plants using Chromosome Conformation Capture (3C). Nat Protoc 4, 1216-‐29 (2009). Raab, J.R. et al. Human tRNA genes function as chromatin insulators. Embo J (2011). Brazma, A. et al. Minimum information about a microarray experiment (MIAME)-‐toward standards for microarray data. Nat Genet 29, 365-‐71 (2001). de Wit, E., Braunschweig, U., Greil, F., Bussemaker, H.J. & van Steensel, B. Global chromatin domain organization of the Drosophila genome. PLoS Genet 4, e1000045 (2008). Bantignies, F. et al. Polycomb-‐Dependent Regulatory Contacts between Distant Hox Loci in Drosophila. Cell 144, 214-‐26 (2011). Noordermeer, D. et al. Variegated gene expression caused by cell-‐specific long-‐range DNA interactions. Nat Cell Biol (2011). Tolhuis, B. et al. Interactions among Polycomb Domains Are Guided by Chromosome Architecture. PLoS Genet 7, e1001343 (2011). Kilkenny, C., Browne, W.J., Cuthill, I.C., Emerson, M. & Altman, D.G. Improving bioscience research reporting: the ARRIVE guidelines for reporting animal research. PLoS Biol 8, e1000412 (2010).
32
24.
25.
26. 27. 28.
Comet, I., Schuettengruber, B., Sexton, T. & Cavalli, G. A chromatin insulator driving three-‐ dimensional Polycomb response element (PRE) contacts and Polycomb association with the chromatin fiber. Proc Natl Acad Sci U S A 108, 2294-‐9 (2011). Sandmann, T., Jakobsen, J.S. & Furlong, E.E. ChIP-‐on-‐chip protocol for genome-‐wide analysis of transcription factor binding in Drosophila melanogaster embryos. Nat Protoc 1, 2839-‐55 (2006). Osborne, C.S. et al. Active genes dynamically colocalize to shared sites of ongoing transcription. Nat Genet 36, 1065-‐71 (2004). Carter, D., Chakalova, L., Osborne, C.S., Dai, Y.F. & Fraser, P. Long-‐range chromatin regulatory interactions in vivo. Nat Genet 32, 623-‐6 (2002). Tolhuis, B., Palstra, R.J., Splinter, E., Grosveld, F. & de Laat, W. Looping and interaction between hypersensitive sites in the active beta-‐globin locus. Mol Cell 10, 1453-‐65 (2002).
33
Sexton et al., Figure 1 a (i) B
B B
(ii) B B
N B
(iii) B
N
(iv)
b
Hbb-b1 Biotin primer
NlaIII
Nested primer
BfaI 100 bp
BglII
NlaIII
Sexton et al., Figure 2 Steps 1-11 Single-cell preparation and fixation Cell permeabilization Restriction digestion Steps 12-13 ChIP preparation steps
Tissue from different species for assessment of non-specific ligation (Box 5)
Steps 14-18 Quantitation of ChIP input (Box 4) Chromatin immunoprecipitation
Steps 19-22 ChIP washes and elution Step 23 Ligation Cross-link reversal
Steps 23-24 DNA purification
Genomic DNA for e4C primer test (Box 3)
Steps 25-27 Primer extension
Steps 28-32 Streptavidin pull-down Re-digestion Gel to assess primer fidelity (Box 3) Clone and sequence to assess non-specific ligation (Box 5)
3C validation ChIP validation Assessment of digestion efficiency (Box 2)
Steps 33-41 Adapter ligation PCR amplification
Steps 42-47 BglII/NlaIII digestion
Assessment of bait enrichment (Box 1)
Sexton et al., Figure 3
a
b H2O K
E
Calr
Ch
Hbb-b1/P2ry6 BglII
Hbb-b1/Ahsp
BglII H2O K
E
Hbb-b1/Uros
d Hbb-b1 sequence
30
20
10
100
146-fold 115-fold
10
102-fold
1 0.1 0.01 0.001
4 pg
V
H
16
sp Ah
1 4a Slc
a-
a1
0
Hb
Fold enrichment over VH16
c
20 pg
100 pg
Amount of PCR template
e 1 kb 500 bp 300 bp 200 bp
H2O
G
A
3C
Sexton et al., Figure 4 2800
a
b log2(e4C enrichment)
No. e4C clones
1600 60
40
20
8
Chr. 7
0 103.9
0 -80
-60
-40 LCR
-20 y
0 kb
bh1
b1
20 b2
40
Chr. 7
log2(e4C enrichment)
ChIP-e4C
0 133.5
133.7
y
b1
3
b2
e4C
Chr. 14
0 3
ChIP-e4C
0
133.9 Mb
69.1
69.3
Uros
69.5 Mb
Epb4.9
e log2(e4C enrichment)
133.3
log2(e4C enrichment)
e4C
log2(e4C enrichment)
log2(e4C enrichment) log2(e4C enrichment)
LCR
103.6 Mb
d
0 4
103.7
3’HS1
c 4
103.8
3
e4C
Chr. 4
0 3
ChIP-e4C
0 131.1
131.25
Epb4.1
131.4 Mb
Xpo7
Sexton et al., Supplementary Figure 1
e4C
Chr. 7
7
log2(e4C enrichment)
0
ChIP-e4C
0 133.5
133.7
133.9 Mb
7
e4C
Chr. 14
0 7
ChIP-e4C
0 69.1
69.3
Uros
69.5 Mb
Epb4.9
c log2(e4C enrichment)
133.3
log2(e4C enrichment)
7
b
log2(e4C enrichment)
log2(e4C enrichment)
log2(e4C enrichment)
a
6
e4C
Xpo7
Chr. 4
0 6
ChIP-e4C
0 131.1
131.25
131.4 Mb
Epb4.1
Supplementary Figure 1. Typical raw e4C and ChIP-e4C results, using Hbb-b1 sequence as bait. a-c) e4C (red) and ChIP-e4C (blue; using an antibody for the active form of RNA polymerase II) microarray profiles, showing the e4C hybridization signal enrichment over genomic control, centered on a) the intrachromosomal co-association with Uros (~650 kb window); b) the interchromosomal coassociation with Epb4.9 and Xpo7 (~510 kb window); c) the interchromosomal co-association with Epb4.1 (~470 kb window). Black bars underneath denote the positions of genes; marked are the genes whose co-associations with Hbb-b1 have been characterized by RNA FISH.
Supplementary Table 1. List of primers used for assessing BglII restriction digestion efficiency in mouse tissues (see Box 2). These primers assess the digestion efficiency at two BglII sites on mouse chromosome 17; the region amplified by the ‘no site’ primers is also nearby on chromosome 17 and is used to normalize DNA inputs in the qPCR. Name
Sequence
No BglII site forward
5’ GTCACCATCCTCATCAATGCTATC 3’
No BglII site reverse
5’ ACCAGTCCCTGTAGAAATCGAAAC 3’
BglII site I forward
5’ CTCATCCAACTTTACGTGAACAGC 3’
BglII site I reverse
5’ GAAGAGGAGGCAGTGTCCATTAC 3’
BglII site II forward
5’ AGACAGTAACGAGGGCTTTCTCT 3’
BglII site II reverse
5’ CTATGGAAACTAACCCAGGAGGTA 3’
Supplementary Table 2. List of 3C primers used in Anticipated Results section. Name
Sequence
Hbb-‐b1 first
5’ CTCAGAGCAGTATCTTTTGTTTGC 3’
Hbb-‐b1 nested
5’ AGGATGAGCAATTCTTTTTGC 3’
Ahsp first
5’ TGTATCACTTGCCAAATCTGACT 3’
Ahsp nested
5’ TGCCAAATCTGACTTAGACTGC 3’
Uros first
5’ TCCAGGCCTTATAGGACTTCAA 3’
Uros nested
5’ CCCAGGCCTTATAGGACTTCA 3’
P2ry6 first
5’ CAGACTCTCCGAGCATAGGAA 3’
P2ry6 nested
5’ CGTCTACCGTGAGGATTTCAA 3’
Calr1
5’ CCCTTGTCTTTCCTATGTCTCACCTG 3’
Calr2
5’ GATGAGGGCTGAAGGAGAATTAAAG 3’
Supplementary Table 3. List of ChIP primers used in Anticipated Results section. Name
Sequence
Hba-‐a1 forward
5’ TTCTGACAGACTCAGGAAGAAACCA 3’
Hba-‐a1 reverse
5’ AGCACCATGGCCACCAATCT 3’
Slc4a1 forward
5’ TGGGAGCTCAGCCAGTCACA 3’
Slc4a1 reverse
5’ CGGGACAGATGCCAAAGGAC 3’
Ahsp forward
5’ GTGAAAATGTAACTTCAGAGCAGAGCGG 3’
Ahsp reverse
5’ CCACCACCCCTGTTAAACATCCTTC 3’
VH16 forward
5’ GGAGGGTCCACTAAACTCTCTTG 3’
VH16 reverse
5’ GCATAGCCTTTTCCACTCTCATC 3’
Supplementary Table 4. List of primers used for assessing e4C bait enrichment in Anticipated Results section. Name
Sequence
Hbb-‐b1 bait forward
5’ CCATAAAGATAGGATGAGCAA 3’
Hbb-‐b1 bait reverse
5’ ATTACTGATCTTCATTAAGTCAAG 3’
