Protocol | Published:

Quantitative analysis of chromosome conformation capture assays (3C-qPCR)

Nature Protocols volume 2, pages 17221733 (2007) | Download Citation

Subjects

Abstract

Chromosome conformation capture (3C) technology is a pioneering methodology that allows in vivo genomic organization to be explored at a scale encompassing a few tens to a few hundred kilobase-pairs. Understanding the folding of the genome at this scale is particularly important in mammals where dispersed regulatory elements, like enhancers or insulators, are involved in gene regulation. 3C technology involves formaldehyde fixation of cells, followed by a polymerase chain reaction (PCR)-based analysis of the frequency with which pairs of selected DNA fragments are crosslinked in the population of cells. Accurate measurements of crosslinking frequencies require the best quantification techniques. We recently adapted the real-time TaqMan PCR technology to the analysis of 3C assays, resulting in a method that more accurately determines crosslinking frequencies than current semiquantitative 3C strategies that rely on measuring the intensity of ethidium bromide-stained PCR products separated by gel electrophoresis. Here, we provide a detailed protocol for this method, which we have named 3C-qPCR. Once preliminary controls and optimizations have been performed, the whole procedure (3C assays and quantitative analyses) can be completed in 7–9 days.

Introduction

Insight into genomic organization is key to understanding gene regulation in mammals. However, owing to technical limitations, we still have little idea about how the mammalian genome is structured in vivo at the scale at which long-range physical interactions between genes and dispersed regulatory elements most often take place (1–103 kbp). The recent development of the “Tagging and recovery of associated proteins”1 and 3C (see ref. 2) assays allowed the very first glimpse into this crucial level of organization of the genome3,4,5. However, the RNA-TRAP technique, which is based on the targeting of peroxidase activity to nascent transcripts, is restricted to physical interactions occurring with actively transcribed genes, while 3C assays potentially allow identification of physical interactions between any chromatin segments. 3C technology is particularly suited to identify chromatin loops formed in genomic regions of up to several hundreds of kilobases in size. 5C technology6,7 offers a robust high-throughput alternative for this analysis, based on large-scale sequencing or microarray analysis. 5C is however more laborious to set up. To identify DNA segments that interact over distances larger than several hundreds of kilobases, we recommend using 4C technology8,9,10,11, which allows for an unbiased genome-wide screen for DNA elements that interact with a genomic site of choice.

The principle of 3C technology2 (Fig. 1) is based on formaldehyde crosslinking of interacting chromatin segments, followed by restriction digestion and intramolecular ligation of crosslinked fragments. Ligation products are subsequently analyzed by PCR using primers specific for the restriction fragments of interest. The mere detection of a ligation product between two segments does not reveal much. To identify a chromatin loop between two segments, it needs to be demonstrated that they interact more frequently with each other than with neighboring DNA fragments. A meaningful 3C analysis therefore critically relies on the accurate quantification of the different ligation products, which in turn requires that measurements are taken when the DNA amplification reaction is in the linear range.

Figure 1: Schematic diagram showing the principles of 3C and qPCR.
Figure 1

(a) Principle of the 3C assays (adapted, with permission, from Dekker et al.2). 3C assays involve three essential steps: Step 1: interacting chromatin segments are crosslinked by a formaldehyde treatment; Step 2: DNA is digested by an appropriate restriction enzyme; and Step 3: crosslinked fragments undergo intramolecular ligation and give ligation products that will be quantified by qPCR. (b) TaqMan probe and qPCR primers design. The locations of the constant fragment (black segment) and candidate interacting fragments (red, blue, green and pink segments) are shown in the upper part of the figure. Restriction sites that will be used in the 3C assay are depicted as small vertical bars in blue. The relative positions of the TaqMan probe (Q--F) and test primers (black arrows T1–T4) that will be used for qPCR quantifications are also depicted. The TaqMan probe should be designed on the strand antisense to the “Constant” primer (black arrow C), but in the same restriction fragment, as depicted in the figure. During the elongation step of the PCR, the 5′–3′ exonuclease activity of the Taq DNA polymerase degrades the probe and releases the fluorophore from the quencher, thus generating a fluorescent signal that is strictly specific to the amplification of the relevant ligation products. T, test primer; F, fluorophore; Q, quencher.

In traditional 3C experiments, a standard PCR protocol, which uses a standard number of PCR cycles and a standard amount of DNA template, is applied to the analysis of all different ligation products. Amounts are then estimated by measuring the intensity of ethidium bromide-stained PCR products separated by gel electrophoresis. This semiquantitative method is prone to provide inaccurate data, as measurements will not always be taken when the DNA amplification reaction is in the linear range.

The TaqMan quantitative PCR technology12,13,14 provides more accurate measurements and was recently used to analyze higher order chromatin structure at the mouse HoxB1 (see ref. 10), α-globin15 and β-globin loci16 using the protocol described here. TaqMan technology makes use of a locus-specific oligodeoxynucleotide probe modified at its 5′ and 3′ ends by a fluorophore and a quencher, respectively. During the elongation step of the PCR, the 5′–3′ exonuclease activity of the Taq DNA polymerase17 degrades the probe and releases the fluorophore from the quencher (Fig. 1), thus generating a fluorescent signal that is quantified at each PCR cycle12,13. The latter allows measurements to be always taken during the linear range of amplification. Quantifications correspond to a “threshold cycle” (Ct) that is determined as close as possible to the base of the exponential phase by defining a baseline to eliminate the background found in early cycles of amplification. Here we explain how TaqMan technology can be applied to the quantitative analysis of the ligation products generated by the 3C procedure. We refer to this strategy as 3C-qPCR. In addition, we provide a detailed step-by-step protocol for the preparation of 3C DNA templates prepared from mouse tissues that can subsequently be used for quantitative (or semiquantitative) PCR analysis. Modifications to the method presented below may be required to prepare 3C templates from other sources.

Experimental design

Preparations before 3C assays, as described below, are the most laborious and time-consuming part of the procedure. It is also a critical part of the procedure, as both the resolution and sensitivity of the technique largely depend on its experimental design.

Restriction enzyme choice

The restriction enzyme should dissect the locus such that it allows for the separate analysis of the relevant regulatory elements (gene bodies, promoters, enhancers, insulators, etc.). Analyzing the topology of small loci (<10–20 kb) requires the use of frequently cutting restriction enzymes such as DpnII or NlaIII (4-base cutters). When analyzing larger loci, 6-base cutters such as EcoRI, BglII or HindIII can also be used. Enzymes that generate cohesive ends are recommended, as blunted ends do not allow efficient ligation (Step 13 of PROCEDURE). Restriction enzymes that are sensitive to DNA methylation should not be used, as they can introduce a bias into the assay owing to differences in the digestion efficiency of the restriction sites analyzed. An additional digestion of the ligation products with a different restriction enzyme may optionally be performed for reliable quantification; this should help to minimize potential PCR biases owing to differences in template accessibility15 (see Box 1). Obviously, the corresponding restriction site (EcoRI in the example in Box 1) should be absent from the PCR amplicons that will be used for ligation product quantification. Therefore, this additional restriction enzyme should be chosen when designing PCR primers used for quantifications. A prerequisite is thus to possess a reliable genomic sequence of the locus pertaining to the biological samples that will be studied.

Box 1: Optional complementary digestion • TIMING 3–4 h

This additional digestion of religated DNA can help to minimize potential PCR biases that may result from limited template accessibility20. We have used EcoRI as an example below, but any suitable restriction enzyme and buffer can be used (see the Experimental design section).

  1. To the sample from Step 34, add 28 μl of dH2O and 20 μl of 1 × EcoRI buffer (commercial 10 × buffer for EcoRI (or any other selected restriction enzyme) diluted in water).

  2. Place the sample in a 1.5 ml tube and add 2 μl of 50 U μl−1 EcoRI enzyme (100 U final).

  3. Incubate for 2 h at 37 °C.

  4. Add 200 μl of phenol–chloroform, mix vigorously.

  5. Centrifuge for 5 min at 16,100g at room temperature (18–22 °C).

  6. Transfer the supernatant into a new tube and add 20 μl of 2 M sodium acetate pH 5.6, then 0.5 ml of ethanol.

  7. Mix and place at −80 °C until frozen (about 45 min).

  8. Centrifuge for 20 min at 16,100g at 4 °C.

  9. Remove the supernatant and add 200 μl of 70% ethanol.

  10. Centrifuge for 4 min at 16,100g at room temperature (18–22 °C).

  11. Remove the supernatant and dry the pellet at room temperature (18–22 °C).

  12. Resuspend the pellet into 150 μl of 10 mM Tris-HCl pH 7.5.

  13. Proceed from Step 35 of the main procedure.

TaqMan probe design

From the quantity of genomic DNA required to detect ligation products, we estimate that some locus-specific ligation events may occur in only 1/2,000 to 1/20,000 mammalian cells. Therefore, reliable quantification of ligation products is possible only if large amounts of DNA (i.e., many genome equivalents) are added to each PCR; we routinely use 50–200 ng of 3C template (8 × 104 to 3 × 105 genome equivalents) per PCR. As such large amounts of DNA preclude the use of SybRGreen incorporation as a means to quantify PCR products4, the highly specific TaqMan chemistry is required for quantifications by real-time PCR. The 3C analysis quantifies ligation products generated by ligation of a “constant fragment” with candidate interacting fragments of the locus. Quantification of these ligation products requires a TaqMan probe and a “constant primer” (C) both located in the “constant fragment”, combined with a battery of “test primers” (T) designed in the candidate interacting fragments. For efficient amplification of ligation products, the amplicon needs to be small, and test primers therefore should be designed to hybridize as close as possible to, and preferably within 50 bp of, the restriction sites of interest. The TaqMan probe and the “constant primer” should also be designed as close as possible to the relevant restriction site and, importantly, they should hybridize to opposite strands of the “constant fragment”16. This configuration results in a fluorescent signal that is strictly specific to the amplification of the ligation product selected for analysis (see Fig. 1). As the TaqMan probe is designed to hybridize within the constant fragment, this probe can be used for quantification of all products ligated to the “constant fragment”. Further details on primer and probe design can be found in the REAGENT SETUP section.

Control template/primer efficiency

A control template that contains all ligation products in equal amounts is used to optimize real-time PCRs and to establish the minimal amount of ligation product that can still be quantified in a reliable manner. For the PCR control template, we recommend the use of a single Bacterial Artificial Chromosome (BAC) clone covering the genome segment under study. Alternatively, a set of minimally overlapping BAC clones mixed in equimolar amounts could be used18,19. This BAC is then cut with the restriction enzyme of choice (e.g., HindIII) and religated by T4 DNA ligase. A secondary restriction enzyme (e.g., EcoRI) can be used to linearize DNA circles, which may otherwise affect primer hybridization efficiency20. It is then necessary to make serial dilutions of this reaction to obtain standard curves that cover the same range of ligation product concentrations as those that will be obtained in the 3C samples. To mimic 3C sample conditions, the final DNA concentration in these dilutions is adjusted to the amount of DNA used in the 3C samples (see Step 38 of PROCEDURE). Thus, depending on the amount of 3C sample used, these dilutions are performed on a 50–200 ng genomic DNA solution.

Experimental controls

Several controls need to be performed on each 3C sample to obtain meaningful data. Sample purity (Steps 35–37) and digestion efficiency (Box 2) should be carefully assessed, as they may impair accurate quantifications. Finally, the level of random collisions needs to be assessed carefully to distinguish between relatively frequent but nonfunctional interactions from physiologically relevant interactions (see the ANTICIPATED RESULTS section). To correct for differences in crosslinking and digestion efficiencies between samples, 3C-qPCR data are normalized toward a set of “control interaction frequencies”. For this purpose, we routinely use the locus of the ERCC3 gene (excision repair cross-complementing rodent repair deficiency, complementation group 3), a ubiquitously expressed gene. As this gene is expressed at very similar levels in all mouse tissues, higher order chromatin architecture at this locus is assumed to be identical and can be used for normalization of 3C-qPCR data (for primer sequences used in combination with HindIII, see Splinter et al.16).

Box 2: Determination of digestion efficiency • TIMING 7–8 h

Digestion efficiencies should be carefully assessed for each restriction site involved in the analysis. Indeed, a twofold drop in digestion efficiency of a given site causes a twofold reduction in the available amount of a given restriction end, which would affect the number of ligation products it can form. Therefore, digestion efficiencies should be in the same range for each of the sites of interest.

  1. Add 500 μl of 1 × PK buffer and 1 μl of 20 mg ml−1 PK (20 μg final) to the control aliquots saved in Steps 12 (UND) and 14 (D).

  2. Incubate for 30 min at 65 °C (or overnight at 65 °C if performed in parallel to Steps 22–34).

  3. Equilibrate for a few minutes at 37 °C, then add 1 μl of 1 mg ml−1 RNase A (1 μg final) and incubate for 2 h at 37 °C.

  4. Add 500 μl of phenol–chloroform, mix vigorously.

  5. Centrifuge for 5 min at 16,100g at room temperature (18–22 °C).

  6. Transfer the supernatant into a new tube and add 50 μl of 2 M sodium acetate pH 5.6, then 1.5 ml of ethanol.

  7. Mix and place at −80 °C until frozen (about 45 min)

  8. Centrifuge for 20 min at 16,100g at 4 °C.

  9. Remove the supernatant and add 500 μl of 70% ethanol.

  10. Centrifuge for 4 min at 16,100g at room temperature (18–22 °C).

  11. Remove the supernatant and dry the pellet at room temperature (18–22 °C).

  12. Resuspend the pellet in 60 μl of water.

  13. Perform real-time PCR quantifications (SybRGreen) on both samples (UND and D) using the qPCR conditions detailed in Steps 38A(ii) and (iii) of the main procedure. To check digestion efficiencies, use primer sets that amplify across each restriction site of interest (R); these should be designed as outlined in the REAGENT SETUP section. To correct for differences in the amount of template added to the PCR, also PCR-amplify control regions not containing the restriction sites of interest (use the “internal” primers used for loading adjustments, Step 38).

  14. Use the cycle thresholds (Ct values, i.e., the number of PCR cycles required to generate a threshold amount of PCR product) to calculate the restriction efficiency according to the following formula:

    % restriction = 100−100/2^((CtR − CtC)D − (CtR − CtC)UND)

    CRITICAL STEP The efficiency of the restriction enzyme digestion should be above 60–70%, but ideally >80% should be digested. Samples with lower digestion efficiencies should be discarded.

Materials

Reagents

  • Mice or appropriate cultured cells

  • 10% (v/v) FCS/DMEM

  • 10% (v/v) FCS/PBS

  • 0.25% (w/v) trypsin

  • 37% formaldehyde (vol/vol; 47629 Fluka)

  • Glycine, 1 M

  • Lysis buffer (10 mM Tris-HCl, pH 7.5; 10 mM NaCl; 5 mM MgCl2; 0.1 mM EGTA; 1 × complete protease inhibitor; 11836145001 Roche)

    Critical

    • For some cell types, a more stringent lysis buffer is required. See REAGENT SETUP for additional information.

  • 2.5% (w/v) collagenase type 1 (Sigma C0130)

  • 20% (w/v) sodium dodecyl sulfate (SDS) (EU0460B Euromedex)

  • Triton X-100 (2327140 BDH Chemicals)

  • High-concentration restriction enzyme (40 U μl−1)

  • T4 DNA ligase, high concentration (20 U μl−1; Promega)

  • 10 × ligation buffer (660 mM Tris-HCl, pH 7.5; 50 mM DTT; 50 mM MgCl2; 10 mM ATP)

  • Ribonuclease A (RNase A), 1 mg ml−1

  • Proteinase K (PK), 10 mg ml−1

  • PK buffer (5 mM EDTA, pH 8.0; 10 mM Tris-HCl, pH 8.0; 0.5% SDS)

  • Control PCR primers for assessing DNA digestion efficiency, user-specific (see REAGENT SETUP)

  • Forward and reverse “internal” primers for loading adjustments, user-specific (see REAGENT SETUP)

  • Constant and test PCR primers for quantifying 3C reaction, user-specific (see REAGENT SETUP)

  • PCR primers for determining digestion efficiency, user-specific (see REAGENT SETUP)

  • PCR primers for determining control interaction frequencies, user-specific (see REAGENT SETUP)

  • TaqMan probe, user-specific (see REAGENT SETUP and Experimental design)

  • QuantiTect Probe PCR Master Mix (204343 Qiagen)

  • QuantiTect SYBR Green PCR Master Mix (Qiagen) or as described in ref.14

  • TE pH 8.0 (10 mM Tris-HCl, pH 8.0; 1 mM EDTA)

  • Agarose (A9539 Sigma)


Equipment

  • Microcentrifuge, for example Eppendorf 5415 D, rotor F45-24-11 (for 1.5 ml tubes)

  • Eppendorf centrifuge 5810R (for 15 and 50 ml Falcon tubes).

  • LightCycler apparatus (Roche) and LightCycler Software, version 3.5 (TaqMan qPCR analyses were performed in “F1/F2” mode)

  • 40 μm cell strainer (352340 Falcon).


Reagent setup

  • Lysis of cells The choice of lysis buffer depends on the tissue or cell type used. For most tissues, the lysis buffer described above will be sufficient. If cells are not properly lysed, a different lysis buffer containing NP-40 can be used (10 mM Tris-HCl, pH 7.5; 10 mM NaCl; 0.2% NP-40; 1 × complete protease inhibitor (11836145001 Roche)). If this is not sufficient, an additional douncing step can be included as described in troubleshooting Step 5. Staining cells with Methylgrün-Pyronin and viewing them under a light microscope can confirm cell lysis.

  • PCR primer design All PCR primers used in this protocol should be typically 20–22-mers with a Tm in the range 55–65 °C with a 2 °C maximum difference between all primers used in a single experiment. Primers could be designed using software available on the internet (e.g., http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi). PCR primers used for analyzing digestion efficiency or the 3C assays should be designed close (within 50 bp) to the restriction site used for the 3C assays.

  • TaqMan probe design The annealing temperature of the probe should be 8–10 °C higher than that of the PCR amplification primers. The probe should be 20–30 nucleotides long and contain more C than G residues. Avoid G in the most 5′ position (which could quench the fluorescent reporter), and stretches of four consecutive identical nucleotides (particularly G). We have used 5′FAM–3′BHQR and 5′FAM–3′TAMRA probes12,13. TaqMan probes are dissolved in TE pH 8.0 and stored at −20 °C as 1.5 μM aliquots.

Procedure

Single-cell preparation

Timing: 30 min–1 h

  1. Obtain single-cell preparations from mouse tissue (option A), adherent cultured cells (option B) or from cultured cells in suspension (option C).

    1. Single-cell preparations from mouse tissues

      1. Dissect mouse tissues according to approved methods and transfer to a 1.5 ml Eppendorf tube.

      2. For more solid tissues (e.g., fetal brain and most adult tissues), incubate in 10% (v/v) PBS/FCS supplemented with 0.125% (w/v) collagenase at 37 °C with gentle shaking at 300 r.p.m.; the exact duration of the collagenase treatment will need to be optimized for each tissue. For other tissues (e.g., fetal liver), omit the collagenase treatment and add 250 μl of 10% (v/v) FCS/DMEM per 1 × 107 cells; resuspend the tissue with a blue tip.

        Critical step

        • Disrupting more solid tissues by force (e.g., repeated pipetting) will often cause cell death, which may obscure the 3C results.

      3. Centrifuge the cell suspension for 1 min at 400g at room temperature (18–22 °C).

      4. Discard the supernatant and resuspend the pellet in 500 μl of 10% (v/v) FCS/PBS.

      5. Filter through the 40 μm cell strainer to make a single-cell suspension. The cells are now ready for crosslinking (Step 2).

    2. Single-cell preparations from adherent cultured cells

      1. Remove medium and wash cells twice with PBS.

      2. Add 1 ml of 0.25% (w/v) trypsin and incubate for 5 min at 37 °C.

      3. Resuspend cells in 9 ml of 10% (v/v) FCS/DMEM.

      4. Centrifuge the cell suspension for 1 min at 400g at room temperature (18–22 °C).

      5. Discard the supernatant and resuspend the pellet in 500 μl of 10% (v/v) FCS/PBS per 1 × 107 cells.

      6. Filter through the 40 μm cell strainer to make a single-cell suspension. The cells are now ready for crosslinking (Step 2).

    3. Single-cell preparations from suspended cultured cells

      1. Centrifuge the cell suspension for 1 min at 400g at room temperature (18–22 °C).

      2. Discard the supernatant and resuspend the pellet in 500 μl of 10% (v/v) FCS/PBS per 1 × 107 cells.

      3. Filter through the 40 μm cell strainer to make a single-cell suspension. The cells are now ready for crosslinking (Step 2).


Formaldehyde crosslinking

Timing: 20–25 min

  1. Per 1 × 107 cells, add 9.5 ml of 2% formaldehyde/10% FCS/PBS and incubate, while tumbling, for 10 min at room temperature (18–22 °C). This is the crosslinking step.

    Critical step

    • 1% formaldehyde is also routinely used. Formaldehyde concentration may affect digestion efficiencies; if digestion efficiency is likely to be a problem, use 1% instead of 2% formaldehyde.

  2. Transfer the reaction tubes to ice and add 1.425 ml of 1 M glycine (ice cold) to quench the crosslinking reaction.

    Critical step

    • Glycine will inactivate the excess formaldehyde that did not react with the biological sample. The glycine concentration described here is sufficient for most biological samples. However, if the sample is low in free amino groups, quenching may not be complete; in this case, the glycine concentration should be increased before proceeding immediately to Step 4.

  3. Spin for 8 min at 225g at 4 °C and carefully remove all the supernatant.


Cell lysis

Timing: 20–25 min

  1. Take up the pellet in 5 ml cold lysis buffer and incubate for 10 min on ice.

    Critical step

    • It is important to obtain a homogeneous preparation of nuclei, which may be facilitated by gently pipetting the mixture up and down. Depending on the cell type, more stringent lysis buffers may be needed to prepare nuclei (see REAGENT SETUP).

    Troubleshooting

  2. Centrifuge for 5 min at 400g at 4 °C and remove the supernatant.

    Troubleshooting

    Pause point

    • The pelleted nuclei can be frozen in liquid nitrogen and stored at −80 °C for several months.


Digestion

Timing: 18–20 h

  1. Take up the nuclei in 0.5 ml of 1.2 × restriction enzyme buffer and transfer to a safe-lock tube.

  2. Place the tube at 37 °C and add 7.5 μl of 20% (w/v) SDS (final: 0.3% SDS).

  3. Incubate for 1 h at 37 °C while shaking at 900 r.p.m.

    Troubleshooting

  4. Add 50 μl of 20% (v/v) Triton X-100 (final: 2% Triton X-100).

  5. Incubate for 1 h at 37 °C while shaking at 900 r.p.m.

  6. Take a 5 μl aliquot of the sample and label as an undigested genomic DNA control (UND). This sample may be stored at −20 °C until it is needed to determine the digestion efficiency (see Step 14 and Box 2).

  7. Add 400 U of the selected restriction enzyme to the remaining sample and incubate overnight at 37 °C while shaking at 900 r.p.m.

  8. Take a 5 μl aliquot of the sample and label as a digested genomic DNA control (D). To process the remaining sample, proceed to Step 15. To determine the digestion efficiency, analyze the control aliquots from Steps 12 and 14 as described in Box 2. This analysis can be carried out in parallel with Steps 22–34.

    Troubleshooting


Ligation

Timing: 8–9 h

  1. Add 40 μl of 20% (w/v) SDS (final 1.6%) to the remaining sample from Step 14.

  2. Incubate for 20–25 min at 65 °C; shake at 900 r.p.m.

  3. Transfer the digested nuclei to a 50 ml falcon tube.

  4. Add 6.125 ml of 1.15 × ligation buffer.

  5. Add 375 μl of 20% (v/v) Triton X-100 (final 1% Triton X-100).

  6. Incubate for 1 h at 37 °C while shaking gently.

  7. Add 5 μl ligase (100 U total) and incubate for 4 h at 16 °C followed by 30 min at room temperature (18–22 °C).

  8. Add 30 μl of 10 mg ml−1 PK (300 μg final).

  9. Incubate at 65 °C overnight to de-crosslink the sample.


DNA purification

Timing: 5–6 h

  1. Add 30 μl of 10 mg ml−1 RNase (300 μg final).

  2. Incubate for 30–45 min at 37 °C.

  3. Add 7 ml of phenol–chloroform and mix vigorously.

  4. Centrifuge for 15 min at 2,200g at room temperature (18–22 °C).

    Troubleshooting

  5. Transfer the supernatant into a new 50 ml tube and add 7 ml of distilled water, 1.5 ml of 2 M sodium acetate pH 5.6 followed by 35 ml of ethanol.

    Critical step

    • Increasing the volume before precipitation will dilute the DTT present in the ligation buffer and prevent it from precipitating (white pellet).

  6. Mix and place at −80° C for approximately 1 h.

  7. Centrifuge for 45 min at 2,200g at 4 °C.

  8. Remove the supernatant and add 10 ml of 70% (v/v) ethanol.

  9. Centrifuge for 15 min at 2,200g at 4 °C.

  10. Remove the supernatant and briefly dry the pellet at room temperature (18–22 °C).

  11. Dissolve the DNA pellet in 150 μl of 10 mM Tris pH 7.5. The 3C template is now ready for qPCR analysis. If an additional digestion step is required to maximize template accessibility, follow the procedure in Box 1 before continuing with Step 35 and qPCR analysis.

    Troubleshooting

    Pause point

    • 3C template may be kept at −20 °C for several months.


Assessment of sample purity

Timing: 2–4 h

  1. Dilute an aliquot of the 3C sample (from Step 34) two- to tenfold. Add genomic DNA to the diluted reaction samples such that the total amount of DNA in each reaction is constant and between 50 and 200 ng.

  2. Set up 10 μl qPCRs as detailed in the table below. Any 3C test primer can be used.

    Perform a standard curve in the same run using serial dilutions of the control template (see Experimental design) according to the instructions of the manufacturer of the qPCR apparatus.

    Stock componentAmountFinal
    H2O2 μl 
    Test primer (10 μM)0.5 μl (5 pmol)0.5 μM
    Constant primer (10 μM)0.5 μl (5 pmol)0.5 μM
    QuantiTech Probe PCR Master5 μl 
    Mix—Qiagen  
    TaqMan probe (1.5 μM)1 μl (1.5 pmol)0.15 μM
    3C sample (from Step 35)1 μl 
  3. Run the qPCR using the conditions tabulated below. Check that real-time PCR quantifications are reduced according to the dilution factors (see Example in the ANTICIPATED RESULTS section). If this is not the case, sample purity is not appropriate and the sample should be discarded.

    Cycle numberDenatureAnneal and extend
    115 min at 95 °C 
    2–4610 s at 95 °C1 min at 60 °C

Performing loading adjustments

Timing: 2 h

  1. Assess the concentration of samples that pass the purity assessment (Steps 35–37). To accurately determine the DNA concentration of a 3C sample, it needs to be compared to a reference sample of genomic DNA of known concentration. This can be done by SybGreen quantitative PCR on 50 × dilutions of 3C samples and the reference sample, using “internal” primer sets that do not amplify across sites recognized by any of the restriction enzymes used (option A) (also see Example in the ANTICIPATED RESULTS section). Alternatively, dilutions of 3C samples and reference samples can be run side by side on an agarose gel (option B).

    Critical step

    • Optical density (OD260) measurements fail to provide an accurate estimate of DNA concentration in 3C samples, probably because of their limited purity.

    1. SybGreen quantitative PCR

      1. Prepare a 50 × dilution of an aliquot of the 3C sample (Step 34) and serial dilutions of a reference DNA (10–0.01 ng μl−1 of DNA).

      2. Set up 20 μl qPCRs as follows (LightCycler apparatus, Roche)

        Stock componentAmountFinal
        H2O14 μl 
        Forward internal primer (10 μM)1 μl (10 pmol)0.5 μM
        Reverse internal primer (10 μM)1 μl (10 pmol)0.5 μM
        qPCR mix142 μl 
        50 × dilution of 3C sample (Step 38A(i))2 μl 
      3. Carry out qPCR using the conditions tabulated below:

        Cycle numberDenatureAnneal and extend
        13 min at 95 °C 
        2–461 s at 95 °C5 s at 70 °C and 15 s at 72 °C
        47Increase temperature to 95 °C at a rate of 0.2 °C s−1 

        Critical step

        • If using another apparatus than the LightCycler or a different commercial qPCR mix, please follow the instructions of the manufacturers.

      4. Determine the DNA concentration by comparing the Ct value of the sample to the standard curve.

        Troubleshooting

    2. Agarose gel analysis

      1. Prepare 1/5, 1/10, 1/20 and 1/40 dilutions of sample and reference sample of known DNA concentration.

      2. Run corresponding dilutions of sample and reference DNA side by side on an agarose gel (0.8–2%); the ligated 3C samples (and the undigested genomic reference sample) will run as a single high-molecular-weight band.

      3. Quantify and compare the amount of DNA between the ligated 3C samples and the reference sample by measuring ethidium bromide incorporation using a Typhoon 9200 imager (GE Healthcare). Processed images are analyzed using ImageQuant software (GE Healthcare) to determine the DNA concentration.

  2. Adjust the original 3C assays (from Step 34) with H2O to 50 ng μl−1 ±10% and optionally verify this concentration by repeating the measurements as described in Step 38.


Real-time PCR quantifications of ligation products

Timing: 2–4 days (2–3 h per run; number of runs depends on the number of primer pairs and 3C samples to be analyzed)

  1. Perform TaqMan real-time PCR quantifications of ligation products on 1 μl (containing 50 ng of DNA) of the “adjusted” 3C samples (from Step 39). Use the reaction conditions described in Step 36.

  2. For each ligation product, run the qPCR using the PCR parameters tabulated in Step 37. Perform standard curves in each run using serial dilutions of the control template (see Experimental design) according to the instructions of the manufacturer of the qPCR apparatus. Analyze data as described in the ANTICIPATED RESULTS section.

    Critical step

    • 3C-qPCR data have to be normalized toward a “loading control” (“internal” primers located in the gapdh gene in the example experiment below) and a set of “control interaction frequencies”. This latter control should be performed on each 3C assay using a pair of test/constant primer localized in a locus that contains a ubiquitously expressed gene. For this purpose, we routinely use primer pairs within the locus of the ERCC3 (see ref. 16) or PDHB genes (see Example in the ANTICIPATED RESULTS section).

    Troubleshooting

Troubleshooting

Troubleshooting advice can be found in Table 1.

Table 1: Troubleshooting table.

Timing

Step 1, preparation of the cells: 30 min–1 h

Steps 2–4, formaldehyde crosslinking: 20–25 min

Steps 5 and 6, cell lysis: 20–25 min

Steps 7–14, digestion: 18–20 h

Steps 15–23, ligation: 8–9 h

Steps 24–34, DNA purification: 5–6 h

Steps 35–37, sample purity: 2–4 h

Steps 38 and 39, loading control: 2 h

Steps 40 and 41, real-time PCR quantification of ligation products: 2–4 days

Anticipated results

Data analysis

As recently stressed elsewhere18, for each 3C sample, the level of random collisions needs to be assessed carefully to distinguish relatively frequent but nonfunctional interactions from physiologically relevant interactions. The background of random collisions can be estimated by determining the interaction frequencies between sites separated by increasing genomic distances (see Example of experiment). As mentioned previously, “a specific interaction is detected when a local peak in interaction frequency is observed”18. Using the 3C protocol to simply look for an interaction between two sites (e.g., promoter to enhancer) with just a few primer sets can easily result in misinterpretation of the data.

However, some parameters—like the size of the restriction fragments—may influence the quantification of the ligation products4 and create “artifactual local peaks”. Therefore, the comparison of relative crosslinking frequencies between two different samples is the best indication of physiologically relevant interactions.

Example of experiment

3C-qPCR was first validated by Splinter et al.16 at the mouse β-globin locus. To further validate the 3C-qPCR protocol, we analyzed the enhancers/promoter interaction at the imprinted H19 locus on mouse chromosome 7, which is a mammalian locus recently analyzed by 3C assays using the standard procedure21,22. We prepared nuclei from liver samples from 7-day-old mice (when Igf2 and H19 are highly expressed)23 and proceeded until Step 34 of the 3C protocol as described above using BamHI (Step 13) and EcoRI (Box 1) restriction enzymes. Sample purity was then assessed on aliquots of three independent 3C assays (A1, A2 and X) as described above (Steps 35–37) using TaqMan PCR with a constant primer located into the fragment containing the H19 enhancers and test primer 20 (see Fig. 2 and Table 2 for primer sequences). The results of this purity control are given in Table 3. The values correspond to the quantifications of the ligation product formed between fragment 20 and the constant fragment (C). These values are calculated using the parameters of the standard curve (b: intercept; a: slope) as follows: value=10(Ct−b)/a.

Figure 2: 3C-qPCR analysis of long-distance interactions at the mouse imprinted Igf2/H19 locus.
Figure 2

The relative level of each ligation product (fragments −1 to 8 and 17 to 20) has been plotted according to its distance (in kb) from the Igf2 promoter P0 (see map below graphs). The “constant” primer and the TaqMan probe were designed as described in Figure 1 (also see PCR primer design in the REAGENT SETUP section) in a BamHI fragment (fragment 21) containing the H19 endodermic enhancers (see Tables 2 and 5). Two independent 3C-qPCR experiments (A1, blue curve; A2, pink curve) were performed on the same biological samples. The data were normalized either to a Gapdh loading control (a) or to Pdhb “control interaction frequencies” (b). Below the graphs, the BamHI restriction fragments are indicated. BamHI fragments are numbered from fragment −1 to 21. Arrows P0–P3 indicate the four Igf2 promoters. The locations of the imprinting-control region (ICR, white rectangle in fragment 18), Igf2 and H19 genes (black rectangle in fragment 19), the endodermic enhancers (black circles in fragment 21), the matrix attachment region 3 (MAR3, white rectangle in fragment 8) and the differentially methylated region 1 (DMR1, white rectangle in fragment 2) are also indicated.

Table 2: Sequence of qPCR primers used for 3C-qPCR analysis of the mouse imprinted Igf2/H19 locus (see Fig. 2).
Table 3: Sample purity assessment.

One 3C assay (X) was discarded from the analysis because impurities impaired faithful quantifications (see Table 3). The purity of two other 3C samples (A1 and A2) was appropriate and we proceeded with loading adjustments (Step 38), as shown in Table 4. SYBR Green quantifications were performed on an aliquot of a 50 × dilution of these 3C samples as described in Step 38A. Once the original 3C samples had been appropriately diluted to 50 ng μl−1 (i.e., dilution factor of 5 for sample A1 or 9 for sample A2), we verified their concentration on an aliquot of a 50 × dilution of these adjusted samples (Step 39). Standard curves were obtained using serial dilutions of a reference sample (see Step 38A), giving a value of 1 for the 1 ng μl−1 point. The values were obtained using the “internal” primer set (here we used primers located in the Gapdh gene; see Table 2 for primer sequence). These values are calculated using the parameters of the standard curve (b: intercept; a: slope) as follows: value=10(Ct−b)/a.

Table 4: Loading adjustments.

Digestion efficiency was controlled as described in Box 2 (data not shown; for primer sequences see Table 5). Quantifications of ligation products were then performed using real-time PCR with a TaqMan probe designed in the fragment that contains the H19 enhancers (fragment 21; Fig. 2; for primer sequences, see Table 2). These enhancers are endoderm specific and are necessary for both Igf2 and H19 gene expression24. As shown in Figure 2a (loading control normalization) as well as in Figure 2b (normalization to control interaction frequencies, right graphs), for each 3C sample analyzed (A1 and A2), a local interaction peak is found with the restriction fragment that contains the H19 promoter (fragment 19). Raw data from this experiment are given in Table 6 as an example of data analysis.

Table 5: Sequence of qPCR primers used to determine the digestion efficiency at each BamHI site used for 3C-qPCR analysis of the mouse imprinted Igf2/H19 locus; all amplifications were performed as described in PROCEDURE (Step 38A).
Table 6: Data analysis for interaction between H19 enhancers and the H19 gene promoter.

In the upstream part of the locus, we found that the H19 enhancers contact the sequences containing the differentially methylated region 1 (DMR1, fragment 2) (Fig. 2a,b, left graphs). These results confirm the recently described data21 and thus further validate the 3C-qPCR methodology.

References

  1. 1.

    , , , & Long-range chromatin regulatory interactions in vivo. Nat. Genet. 32, 623–626 (2002).

  2. 2.

    , , & Capturing chromosome conformation. Science 295, 1306–1311 (2002).

  3. 3.

    A closer look at long-range chromosomal interactions. Trends Biochem. Sci. 28, 277–280 (2003).

  4. 4.

    , & 3C technology: analyzing the spatial organization of genomic loci in vivo. Methods Enzymol. 375, 493–507 (2004).

  5. 5.

    , , , & Looping and interaction between hypersensitive sites in the active beta-globin locus. Mol. Cell 10, 1453–1465 (2002).

  6. 6.

    & Mapping networks of physical interactions between genomic elements using 5C technology. Nat. Protoc. 2, 988–1002 (2007).

  7. 7.

    et al. Chromosome Conformation Capture Carbon Copy (5C): a massively parallel solution for mapping interactions between genomic elements. Genome Res. 16, 1299–1309 (2006).

  8. 8.

    et al. Interchromosomal interactions and olfactory receptor choice. Cell 126, 403–413 (2006).

  9. 9.

    et al. Nuclear organization of active and inactive chromatin domains uncovered by chromosome conformation capture-on-chip (4C). Nat. Genet. 38, 1348–1354 (2006).

  10. 10.

    & Genome-wide scanning of HoxB1-associated loci in mouse ES cells using an open-ended Chromosome Conformation Capture methodology. Chromosome Res. 14, 477–495 (2006).

  11. 11.

    et al. Circular chromosome conformation capture (4C) uncovers extensive networks of epigenetically regulated intra- and interchromosomal interactions. Nat. Genet. 38, 1341–1347 (2006).

  12. 12.

    , , & Real time quantitative PCR. Genome Res. 6, 986–994 (1996).

  13. 13.

    , , , & Oligonucleotides with fluorescent dyes at opposite ends provide a quenched probe system useful for detecting PCR product and nucleic acid hybridization. PCR Methods Appl. 4, 357–362 (1995).

  14. 14.

    & Performing quantitative reverse-transcribed polymerase chain reaction experiments. Methods Enzymol. 410, 386–400 (2006).

  15. 15.

    , , , & Long-range chromosomal interactions regulate the timing of the transition between poised and active gene expression. EMBO J. 26, 2041–2051 (2007).

  16. 16.

    et al. CTCF mediates long-range chromatin looping and local histone modification in the beta-globin locus. Genes Dev. 20, 2349–2354 (2006).

  17. 17.

    , , & Detection of specific polymerase chain reaction product by utilizing the 5′— —3′ exonuclease activity of Thermus aquaticus DNA polymerase. Proc. Natl. Acad. Sci. USA 88, 7276–7280 (1991).

  18. 18.

    The three “C”s of chromosome conformation capture: controls, controls, controls. Nat. Methods 3, 17–21 (2006).

  19. 19.

    et al. The beta-globin nuclear compartment in development and erythroid differentiation. Nat. Genet. 35, 190–194 (2003).

  20. 20.

    et al. A real-time polymerase chain reaction assay for quantification of allele ratios and correction of amplification bias. Anal. Biochem. 320, 252–258 (2003).

  21. 21.

    et al. CTCF binding at the H19 imprinting control region mediates maternally inherited higher-order chromatin conformation to restrict enhancer access to Igf2. Proc. Natl. Acad. Sci. USA 103, 10684–10689 (2006).

  22. 22.

    , & Interaction between differentially methylated regions partitions the imprinted genes Igf2 and H19 into parent-specific chromatin loops. Nat. Genet. 36, 889–893 (2004).

  23. 23.

    et al. H19 gene expression is up-regulated exclusively by stabilization of the RNA during muscle cell differentiation. Oncogene 19, 5810–5816 (2000).

  24. 24.

    , , , & An enhancer deletion affects both H19 and Igf2 expression. Genes Dev. 9, 2079–2089 (1995).

Download references

Acknowledgements

We thank Robert Feil and Franck Court for discussion and comments on the manuscript. This work was supported by grants from the Association pour la Recherche contre le Cancer (ARC contract no. 3279), the “GIS Longévité” (contract no. GISLO401), the “Fond National de la Science” (ACI jeune chercheur) given to T. Forné and by funds from the “Centre National de la Recherche Scientifique” (CNRS). C.B. was supported by an ARC fellowship (JR/MLD/MDV—P05/2 and P06/2). W.L. was supported by grants from the Dutch Scientific Organization (NWO) (016-006-026) and (912-04-082). J.D. was supported by a grant from the NIH (HG003143).

Author information

Affiliations

  1. UMR5535 CNRS-UMII; IFR122; Institut de Génétique Moléculaire de Montpellier; IGMM; 1919, Route de Mende, 34293 Montpellier Cedex 5, France.

    • Hélène Hagège
    • , Caroline Braem
    • , Guy Cathala
    •  & Thierry Forné
  2. Department of Cell Biology and Genetics, Erasmus MC, PO Box 2040, 3000 CA Rotterdam, The Netherlands.

    • Petra Klous
    • , Erik Splinter
    •  & Wouter de Laat
  3. Program in Gene Function and Expression and Department of Biochemistry and Molecular Pharmacology, University of Massachusetts Medical School, Worcester, Massachusetts 01605-0103, USA.

    • Job Dekker

Authors

  1. Search for Hélène Hagège in:

  2. Search for Petra Klous in:

  3. Search for Caroline Braem in:

  4. Search for Erik Splinter in:

  5. Search for Job Dekker in:

  6. Search for Guy Cathala in:

  7. Search for Wouter de Laat in:

  8. Search for Thierry Forné in:

Competing interests

The authors declare no competing financial interests.

Corresponding authors

Correspondence to Wouter de Laat or Thierry Forné.

About this article

Publication history

Published

DOI

https://doi.org/10.1038/nprot.2007.243

Further reading

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.