Summary

An Integrated Workflow to Study the Promoter-Centric Spatio-Temporal Genome Architecture in Scarce Cell Populations

Published: April 21, 2023
doi:

Summary

Gene expression is regulated by interactions of gene promoters with distal regulatory elements. Here, we descirbe how low input Capture Hi-C (liCHi-C) allows the identification of these interactions in rare cell types, which were previously unmeasurable.

Abstract

Spatiotemporal gene transcription is tightly regulated by distal regulatory elements, such as enhancers and silencers, which rely on physical proximity with their target gene promoters to control transcription. Although these regulatory elements are easy to identify, their target genes are difficult to predict, since most of them are cell-type specific and may be separated by hundreds of kilobases in the linear genome sequence, skipping over other non-target genes. For several years, Promoter Capture Hi-C (PCHi-C) has been the gold standard for the association of distal regulatory elements to their target genes. However, PCHi-C relies on the availability of millions of cells, prohibiting the study of rare cell populations such as those commonly obtained from primary tissues. To overcome this limitation, low input Capture Hi-C (liCHi-C), a cost-effective and customizable method to identify the repertoire of distal regulatory elements controlling each gene of the genome, has been developed. liCHi-C relies on a similar experimental and computational framework as PCHi-C, but by employing minimal tube changes, modifying the reagent concentration and volumes, and swapping or eliminating steps, it accounts for minimal material loss during library construction. Collectively, liCHi-C enables the study of gene regulation and spatiotemporal genome organization in the context of developmental biology and cellular function.

Introduction

Temporal gene expression drives cell differentiation and, ultimately, organism development, and its alteration is closely related to a wide plethora of diseases1,2,3,4,5. Gene transcription is finely regulated by the action of regulatory elements, which can be classified as proximal (i.e., gene promoters) and distal (e.g., enhancers or silencers), the latter of which are frequently located afar from their target genes and physically interact with them through chromatin looping to modulate gene expression6,7,8.

The identification of distal regulatory regions in the genome is a matter which is widely agreed upon, since these regions harbor specific histone modifications9,10,11 and contain specific transcription factor recognition motifs, acting as recruiting platforms for them12,13,14. Besides, in the case of enhancers and super-enhancers15,16, they also have low-nucleosome occupancy17,18 and are transcribed into non-coding eRNAs19,20.

Nonetheless, each distal regulatory element's target genes are more difficult to predict. More often than not, interactions between distal regulatory elements and their targets are cell-type and stimulus specific21,22, span hundreds of kilobases, bridging over other genes in any direction23,24,25, and can even be located inside intronic regions of their target gene or other non-intervening genes26,27. Furthermore, distal regulatory elements can also control more than one gene at the same time, and vice versa28,29. This positional complexity hinders pinpointing regulatory associations between them, and therefore, most of each regulatory element's targets in every cell type remain unknown.

During recent years, there has been a significant boom in the development of chromosome conformation capture (3C) techniques for studying chromatin interactions. The most widely used of them, Hi-C, allows to generate a map of all the interactions between every fragment of a cell's genome30. However, to detect significant interactions at the restriction fragment level, Hi-C relies on ultra-deep sequencing, prohibiting its use to routinely study the regulatory landscape of individual genes. To overcome this economic limitation, several enrichment-based 3C techniques have emerged, such as ChIA-PET31, HiChIP32, and its low-input counterpart HiCuT33. These techniques depend on the use of antibodies to enrich for genome-wide interactions mediated by a specific protein. Nonetheless, the unique feature of these 3C techniques is also the bane of their application; users count on the availability of high-quality antibodies for the protein of interest and cannot compare conditions in which the binding of the protein is dynamic.

Promoter Capture Hi-C (PCHi-C) is another enrichment-based 3C technique that circumvents these limitations34,35. By employing a biotinylated RNA probe enrichment system, PCHi-C is able to generate genome-wide high-resolution libraries of genomic regions interacting with 28,650 human- or 27,595 mouse-annotated gene promoters, also known as the promoter interactome. This approach allows one to detect significant long-range interactions at the restriction fragment level resolution of both active and inactive promoters, and robustly compare promoter interactomes between any condition independently of the dynamics of histone modifications or protein binding. PCHi-C has been widely used over recent years to identify promoter interactome reorganizations during cell differentiation36,37, identify the mechanism of action of transcription factors38,39, and discover new potential genes and pathways deregulated in disease by non-coding variants40,41,42,43,44,45,46,47,48, alongside new driver non-coding mutations49,50. Besides, by just modifying the capture system, this technique can be customized according to the biological question to interrogate any interactome (e.g., the enhancer interactome51 or the interactome of a collection of non-coding alterations41,52).

However, PCHi-C relies on a minimum of 20 million cells to perform the technique, which prevents the study of scarce cell populations such as the ones often used in developmental biology and clinical applications. For this reason, we have developed low input Capture Hi-C (liCHi-C), a new cost-effective and customizable method based on the experimental framework of PCHi-C to generate high-resolution promoter interactomes with low-cell input. By performing the experiment with minimal tube changes, swapping or eliminating steps from the original PCHi-C protocol, drastically reducing reaction volumes, and modifying reagent concentrations, library complexity is maximized and it is possible to generate high-quality libraries with as little as 50,000 cells53.

Low input Capture Hi-C (liCHi-C) has been benchmarked against PCHi-C and used to elucidate promoter interactome rewiring during human hematopoietic cell differentiation, discover potential new disease-associated genes and pathways deregulated by non-coding alterations, and detect chromosomal abnormalities53. The step-by-step protocol and the different quality controls through the technique are detailed here until the final generation of the libraries and their computational analysis.

Protocol

To ensure minimal material loss, (1) work with DNA low-binding tubes and tips (see Table of Materials), (2) place reagents on the tube wall instead of introducing the tip inside the sample and, (3) if possible, mix the sample by inversion instead of pipetting the sample up and down, and spin down afterward to recover the sample.

1. Cell fixation

  1. Cells growing in suspension
    1. Harvest 50,000 to one million cells and place them in a DNA low-binding 1.5 mL tube.
      NOTE: The cell types used for the present study are detailed in the representative results section.
    2. Centrifuge the cells for 5 min at 600 x g (at 4 °C) using a fixed-angle rotor centrifuge and remove the supernatant by pipetting.
    3. Resuspend the cells in 1 mL of RPMI 1640 supplemented with 10% fetal bovine serum (FBS) at room temperature.
    4. Add 143 µL of methanol-free 16% formaldehyde to reach a concentration of 2% and mix.
    5. Incubate the cells for 10 min while rotating at room temperature to fix the cells.
      NOTE: Try to be as accurate as possible with the 10 min incubation. Over- or under-fixation of the cells can lead to a decrease in the quality of the library.
    6. Quench the reaction by adding 164 µL of ice-cold 1 M glycine and mix. Incubate the cells for 5 min, rotating at room temperature.
    7. Further incubate the cells for 15 min on ice, mixing by inversion every ~5 min.
    8. Centrifuge the cells for 10 min at 1,000 x g (at 4 °C) using a fixed-angle rotor centrifuge and remove the supernatant.
    9. Wash the cells by resuspending the pellet in 1 mL of ice-cold 1x phosphate-buffered saline (PBS).
    10. Centrifuge the cells for 10 min at 1,000 x g (at 4 °C) and remove the supernatant.
      ​NOTE: The pelleted cells can be flash-frozen in liquid nitrogen or dry ice and stored at -80 °C.
  2. Adherent cells
    1. Wash the cells in the culture dish with 1x PBS.
    2. Prepare enough RPMI 1640 supplemented with 10% FBS and methanol-free 2% formaldehyde at room temperature to cover the culture dish.
    3. Add the supplemented media to the culture dish with the cells and incubate for 10 min, rocking at room temperature to fix the cells.
      ​NOTE: Try to be as accurate as possible with the 10 min incubation. Over- or under-fixation of the cells can lead to a decrease in the quality of the library.
    4. Quench the reaction by adding 1 M glycine until 0.125 M and mix by rocking.
    5. Incubate the cells for 5 min, rocking at room temperature.
    6. Further incubate the cells for 15 min at 4 °C, mixing by rocking every 3-4 min.
    7. Remove the media and wash the cells with cold 1x PBS.
    8. Scrape the cells and transfer them into a 1.5 mL DNA low-binding tube. Wipe the culture dish clean with 0.5-1 mL of cold 1x PBS.
    9. Centrifuge the cells for 10 min at 1,000 x g (at 4 °C) using a fixed-angle rotor centrifuge and remove the supernatant. The pelleted cells can be flash-frozen in liquid nitrogen or dry ice and stored at -80 °C.

2. Lysis and digestion

  1. Resuspend the cells in 1 mL of cold lysis buffer (Table 1) to disrupt the cell membrane. The addition of the buffer alone should be enough to resuspend the cells, but they can be further resuspended if necessary by light vortexing.
  2. Incubate on ice for 30 min, mixing by inversion every ~5 min. Centrifuge the nuclei for 10 min at 1,000 x g (at 4 °C) and remove the supernatant. Resuspend the nuclei with 500 µL of cold 1.25x restriction buffer 2 (see Table of Materials).
  3. Centrifuge the nuclei for 10 min at 1,000 x g (at 4 °C) and remove the supernatant. Resuspend the nuclei in 179 µL of 1.25x restriction buffer 2.
  4. Add 5.5 µL of 10% sodium dodecyl sulfate (SDS; see Table of Materials) and mix. The cells may form clumps; this is normal and needs to be disaggregated as much as possible by vortexing. Incubate the sample for 1 h at 37 °C in a thermoblock, with shaking at 950 rpm.
  5. Quench the SDS by adding 37.5 µL of 10% Triton X-100 and mix. Incubate the sample for 1 h at 37 °C in a thermoblock, with shaking at 950 rpm.
  6. Digest the chromatin by adding 7.5 µL of HindIII (100 U/µL; see Table of Materials) and mix. Incubate the sample overnight at 37 °C in a thermoblock, with shaking at 950 rpm.
  7. The next morning, add an extra 2.5 µL of HindIII (100 U/µL) and further incubate the sample for 1 h at 37 °C in a thermoblock, with shaking at 950 rpm to ensure proper chromatin digestion.
    NOTE: As part of the digestion efficiency controls (see step 5.1), transfer the equivalent of 20,000 to 40,000 nuclei to another tube to represent the undigested control. After digestion, transfer the same number of cells to another tube again to represent the digested control. It is recommended to test the digestion efficiency as a separate experiment if the availability of the starting material is scarce.

3. Ligation and decrosslinking

  1. Chill the sample on ice. Prepare a mastermix and add the following reagents to fill in and biotinylate the restriction fragment overhangs: 3 µL of 10x restriction buffer 2, 1 µL of nuclease-free water, 0.75 µL of 10 mM dCTP, 0.75 µL of 10 mM dTTP, 0.75 µL of 10 mM dGTP, 18.75 µL of 0.4 mM biotin-14-dATP, and 5 µL of 5 U/µL Klenow (see Table of Materials). Incubate the sample for 75 min at 37 °C, mixing by inversion every ~15 min.
  2. Chill the sample on ice. Prepare a mastermix and add the following reagents to ligate the filled-in DNA ends: 50 µL of 10x ligation buffer, 2.5 µL of 20 mg/mL bovine serum albumin (BSA), 12.5 µL of 1 U/µL T4 DNA ligase, and 173 µL of nuclease-free water (see Table of Materials).
  3. Incubate the sample for 4-6 h at 16 °C, mixing by inversion every ~1 h. Further, incubate the sample for 30 min at room temperature. Decrosslink the chromatin by adding 30 µL of 10 mg/mL Proteinase K and mix. Incubate the sample overnight at 65 °C.
  4. The following morning, add an extra 15 µL of 10 mg/mL Proteinase K and further incubate the sample for 2 h at 65 °C to ensure proper chromatin decrosslinking.

4. DNA purification

  1. Cool the sample to room temperature and transfer it to a suitable tube for phenol-chloroform purification.
  2. Add 1 volume (545 µL) of phenol:chloroform:isoamyl alcohol (25:24:1) to purify the DNA and mix by vigorously shaking.
  3. Centrifuge the sample for 5 min at 12,000 x g at room temperature and transfer the upper aqueous phase (545 µL) to a 2 mL DNA low-binding tube.
  4. Add the following reagents to precipitate the DNA: 1,362.5 µL of 100% ethanol chilled to -20 °C, 54.5 µL of 3 M sodium acetate (pH 5.2), and 2 µL of 15 mg/mL glycogen as a coprecipitant.
  5. Incubate for 1 h at -80 °C or overnight at -20 °C.
  6. Centrifuge the sample for 30 min at 16,000-21,000 x g at 4 °C and remove the supernatant. The DNA pellet must be visible.
  7. Wash the pellet by adding 1 mL of 70% ethanol, vortexing, and centrifuging for 10 min at 16,000-21,000 x g at room temperature.
  8. Remove the supernatant and let the pellet air-dry. Resuspend the DNA pellet in 130 µL of nuclease-free water.
  9. Assess the concentration by fluorometric quantification (see Table of Materials). Store the purified 3C material at -20 °C for several months before proceeding with the protocol.

5. Optional quality controls

  1. Assess the digestion efficiency. Perform decrosslinking and phenol:chloroform DNA purification, as described previously, to the undigested and digested controls obtained in step 2.13. Resuspend the DNA pellet in 10 µL of nuclease-free water. Quantify the concentration and dilute the obtained DNA, if necessary, to 4 ng/µL.
  2. Perform quantitative PCR with 4 ng of DNA of both the undigested and digested controls, with primers spanning an open chromatin locus with and without a HindIII target (see Table 2 for primer design). Calculate the efficiency of the digestion following a previously published report35.
    NOTE: The efficiency of ligation is calculated as a percentage using the formula: digestion (%) = 100 -100/(2^[(Ct digested with HindIII – Ctdigested without HindIII) – (Ct undigested with HindIII – Ct undigested without HindIII)]) (see Table 3), which takes into account the differential of the different Cts obtained for each primer pair in the undigested and digested controls.
  3. Assess the sensitivity of the interaction detection by performing conventional polymerase chain reaction (PCR) (0.2 mM dNTP, 0.4 µm both F + R primers, 0.1 and U/µL hot start polymerase), with primers spanning both long- and short-range cell-invariant interactions (see Table 2 for primer design). Use 50-100 ng of 3C material (for short- and long-range interactions, respectively) and amplify using the following conditions: 98 °C for 15 min, followed by 37 cycles of 98 °C for 30 s, 60 °C for 1 min, 72 °C for 1 min, and finish by 72 °C for 10 min. Hold at 4 °C.
    NOTE: If the amount of DNA obtained is less than 2 µg, only check a long- and a short-range interaction instead of the whole interaction panel.
  4. Run the product on 1x tris-borate-EDTA (TBE) using 1.6% agarose gel and look for the presence of the corresponding PCR amplicon54.
    NOTE: Due to the unpredictable "cutting-and-pasting" of the genome, unspecific bands may appear. As long as the correct band size is observed, it is counted as correct.
  5. Assess the efficiency of biotin fill-in and ligation by differentially digesting a short-range PCR product with HindIII, NheI (see Table of Materials), both enzymes or none (water), and running the product on a 1x TBE 1.6% agarose gel. A correct fill-in and ligation eliminate the previous HindIII target and create a new NheI target, so the amplicon should be cut only in the presence of NheI.
    ​NOTE: To minimize material loss, retrieve 2.5 µL of a short-range PCR product from the interaction controls and reamplify it five times to perform the fill-in and ligation controls.

6. Sonication

  1. Transfer the 130 µL of the sample (top up with nuclease-free water if some was used for the controls) to a cuvette suitable for sonication.
  2. Set up a water bath sonicator and sonicate using the following parameters (optimized for the model and cuvettes described in Table of Materials): duty factor: 20%; peak incidence power: 50; cycles per burst: 200; time: 65 s; and temperature range: 6-10 °C (8 °C optimal).
  3. Transfer the sample into a new 1.5 mL DNA low-binding tube.

7. End-repair

  1. Prepare a mastermix and add the following reagents to repair the DNA fragments' uneven ends created during the sonication: 18 µL of 10x ligation buffer, 18 µL of 2.5 mM dNTP mix each, 6.5 µL of 3 U/µL T4 DNA polymerase, 6.5 µL of 10 U/µL T4 PNK, and 1.3 µL of 5 U/µL Klenow (see Table of Materials).
  2. Incubate the sample for 30 min at 20 °C and top up with Tris-low EDTA (TLE) buffer (see Table 1) to 300 µL.

8. Biotin pull-down

  1. Transfer 150 µL of C1 streptavidin beads (see Table of Materials) per sample to a 1.5 mL tube, place them on a 1.5 mL tube magnet, and wait 2-3 min or until all the beads are stuck to the wall. Remove the supernatant, leaving the beads behind.
  2. Wash the beads with 400 µL of 1x Tween buffer (TB; see Table 1). To wash the beads, add the buffer and resuspend them by soft vortexing. Place the tube back in the magnet and wait 2-3 min or until all the beads are stuck to the wall. Remove the supernatant, leaving the beads behind.
  3. Wash the beads with 300 µL of 1x no Tween buffer (NTB; see Table 1). Resuspend the beads in 300 µL of 2x NTB (see Table 1).
    NOTE: The beads may form a dusty layer around the tube wall when washing with buffers without detergent. This is normal and doesn't affect the outcome of the protocol.
  4. Combine this 300 µL of beads in 2x NTB with the 300 µL of the sample. Incubate for 15 min, rotating at room temperature to pull down the informative DNA fragments with biotin. The library is now stuck to the C1 streptavidin beads.
  5. Wash the beads with 400 µL of 1x NTB. Wash the beads with 100 µL of TLE buffer and afterward resuspend them in 35.7 µL of TLE buffer.

9. dATP-tailing, adapter ligation, and PCR amplification

  1. Prepare a mastermix and add the following reagents to the sample to dATP-tail the ends of the repaired DNA fragments: 5 µL of 10x restriction buffer 2, 2.3 µL of 10 mM dATP, and 7 µL of 5 U/µL Klenow exo-. Incubate the sample for 30 min at 37 °C.
  2. Inactivate Klenow exo- by further incubating the sample for 10 min at 65 °C. Cool the sample on ice. Wash the beads with 300 µL of 1x TB. Wash the beads with 300 µL of 1x NTB.
  3. Wash the beads with 100 µL of 1x ligation buffer and afterward resuspend them in 50 µL of 1x ligation buffer. Add 4 µL of 15 µM pre-annealed adapter mix (see Table 2) and 1 µL of 2,000 U/µL T4 DNA ligase (see Table of Materials) to the sample.
  4. Incubate for 2 h at room temperature. Wash the beads with 400 µL of 1x TB. Wash the beads with 200 µL of 1x NTB. Wash the beads with 100 µl of 1x restriction buffer 2.
  5. Wash the beads with 50 µL of 1x restriction buffer 2 and afterward resuspend them in 50 µL of 1x restriction buffer 2.
  6. Mix the following reagents to prepare the PCR reaction to amplify the library: 50 µL of beads with the library, 250 µL of 2x PCR mastermix with enzyme, 12 µL of F + R primers (25 µM each; see Table 2), and 188 µL of nuclease-free water.
  7. Perform the PCR with the following conditions (split the PCR reagent mix into 50 µL reactions): 98 °C for 40 s, followed by X cycles of 98 °C for 10 s, 65 °C for 30 s, 72 °C for 30 s, and finish by 72 °C for 10 min. Hold at 4 °C.
    NOTE: Use the following number of cycles as a starting point for optimizing the protocol in diploid cells aiming for a 500-1,000 ng output before library capture: one million cells for eight cycles; 250,000 cells for 10 cycles; 50,000 cells for 12 cycles.
  8. Pool all the 50 µL reactions from the same sample into a 1.5 mL DNA low-binding tube, place it on a 1.5 mL tube magnet, and wait 2-3 min or until all the beads are stuck to the wall.
  9. Transfer the supernatant containing the library (500 µL) to a new 1.5 mL DNA low-binding tube. Top up with TLE buffer to 500 µL if some of the supernatant was lost. The C1 streptavidin beads are no longer needed.
  10. Perform a double-sided selection55 using paramagnetic bead purification (0.4-1 volume). This allows for the selective elimination of too-large (>1,000 bp) and too-small fragments or PCR primers (<200 bp), depending on the concentration of polyethylene glycol and salt of the paramagnetic beads added.
  11. Add 200 µL (0.4 volumes) of stock beads to the library and mix by vortexing. Incubate for 10 min, rotating at room temperature.
  12. Place on a magnet, wait 2-3 min or until all the beads are stuck to the wall, and transfer the supernatant containing the library (without the larger fragments) to a new 1.5 mL DNA low-binding tube.
  13. Concentrate the beads by taking 750 µL of stock beads, place them in a 1.5 mL tube on a magnet, wait for 2-3 min or until all the beads are stuck to the wall, remove the supernatant, and resuspend the beads by vortexing in 300 µL of new stock beads.
  14. Add this 300 µL of concentrated beads to the sample (1 volume) and mix by vortexing. Incubate for 10 min, rotating at room temperature. Place on a magnet, wait 2-3 min or until all the beads are stuck to the wall, and remove the supernatant (containing the smaller fragments and PCR primers).
  15. Wash the beads three times with 1 mL of 70% ethanol. To do this, add the ethanol while the tube with the beads is still on the magnet, trying not to disturb the beads, and wait for 30-60 s. Afterward, remove the supernatant without disturbing the beads.
  16. Allow the beads to air-dry and resuspend them in 21 µL of TLE buffer by vortexing.
    NOTE: Excessive drying of the beads can reduce the yield when eluting the DNA. Aim to resuspend them in TLE buffer immediately after they are no longer "shiny" from the ethanol.
  17. Incubate the sample for 10 min at 37 °C in a thermoblock to elute the library from the beads. Place the tube in a magnet and transfer the supernatant containing the library to a new 1.5 mL DNA low-binding tube.
  18. Quantify the size and concentration by automated electrophoresis (see Table of Materials). Purified Hi-C material can be stored at -20 °C for several months before proceeding with the protocol.

10. Library capture

  1. Work with 500-1,000 ng of the library. Concentrate the library by drying the DNA using a vacuum concentrator and resuspending the material in 3.4 µL of nuclease-free water.
  2. Add the following blockers from the target enrichment kit (see Table of Materials) to the sample: 2.5 µL of Blocker 1, 2.5 µL of Blocker 2, and 0.6 µL of custom oligo blocker for the adapters.
  3. Resuspend thoroughly, transfer the solution to a 0.2 mL PCR strip, and incubate in a thermocycler for 5 min at 95 °C, followed by 5 min at 65 °C with a heated lid. Leave the tube incubating at 65 °C.
  4. Prepare the hybridization solution by combining the following reagents from the target enrichment kit (see Table of Materials) per sample (13 µL). Keep on the bench at room temperature: 6.63 µL of Hyb 1, 0.27 µL of Hyb 2, 2.65 µL of Hyb 3, and 3.45 µL of Hyb 4.
  5. Dilute 0.5 µL of RNase block from the target enrichment kit with 1.5 µL of nuclease-free water per sample. Thaw 5 µL of the biotinylated RNA per sample on ice and add to it the 2 µL of the diluted RNase block. Keep on the bench at room temperature.
  6. Add the 13 µL of the hybridization solution to the 7 µL of the biotinylated RNA with RNase and mix well.
  7. While on the thermocycler at 65 °C, transfer the hybridization solution with the biotinylated RNA (20 µL) to the blocked library. Close the tube lid firmly and incubate in the thermocycler overnight at 65 °C.
    ​NOTE: To minimize sample evaporation (which may lead to suboptimal RNA-DNA hybridization) when carrying out multiple samples at the same time, use a multichannel pipette to transfer the biotinylated RNA to each library at the same time.

11. Biotin pull-down and PCR amplification

  1. Transfer 50 µL of T1 streptavidin beads per sample to a 1.5 mL DNA low-binding tube, place them on a 1.5 mL tube magnet, and wait 2-3 min or until all the beads are stuck to the wall. Remove the supernatant, leaving the beads behind. Wash the beads three times with 200 µL of the binding buffer from the target enrichment kit.
  2. Resuspend the beads in 200 µL of binding buffer. While on the thermocycler at 65 °C, transfer the sample to the resuspended T1 streptavidin beads and incubate for 30 min, rotating at room temperature.
  3. Wash the beads with 200 µL of wash buffer 1 from the target enrichment kit. Incubate for 15 min, rotating at room temperature. Wash the beads three times with 200 µL of wash buffer 2 from the target enrichment kit heated to 65 °C. Incubate for 10 min at 65 °C in a thermoblock, shaking at 300 rpm between washes.
  4. Wash the beads with 200 µL of 1x restriction buffer 2 and afterward resuspend them in 30 µL of 1x restriction buffer 2.
  5. Mix the following reagents to prepare the PCR reaction to amplify the library: 30 µL beads with the library, 150 µL of 2x PCR mastermix with enzyme, 7.2 µL of F + R primers (25 µM each; see Table 2), and 112.8 µL of nuclease-free water.
  6. Perform the PCR with the following conditions (split the PCR reagent mix in 50 µL reactions): 98 °C for 40 s, followed by four cycles of 98 °C for 10 s, 65 °C for 30 s, 72 °C for 30 s, and finish with 72 °C for 10 min. Hold at 4 °C.
  7. Pool all the 50 µL reactions from the same sample into a 1.5 mL DNA low-binding tube, place it on a 1.5 mL tube magnet, and wait 2-3 min or until all the beads are stuck to the wall.
  8. Transfer the supernatant containing the library (300 µL) to a new 1.5 mL DNA low-binding tube. Top up with TLE buffer to 300 µL if some of the supernatant was lost. The T1 streptavidin beads are no longer needed.
  9. Perform a DNA purification using paramagnetic beads (0.9 volumes; see Table of Materials). Add 270 µL of stock beads to the sample and mix by vortexing.
  10. Incubate for 10 min, rotating at room temperature.
  11. Place on a magnet, wait 2-3 min or until all the beads are stuck to the wall, and remove the supernatant.
  12. Wash the beads three times with 1 mL of 70% ethanol. To do this, add the ethanol while the tube with the beads is still on the magnet, trying not to disturb the beads, and wait 30-60 s. Afterward, remove the supernatant without disturbing the beads.
  13. Allow the beads to air-dry and resuspend them in 21 µL of TLE buffer by vortexing.
    NOTE: Excessive drying of the beads can reduce the yield when eluting the DNA. Aim to resuspend them in TLE buffer immediately after they are no longer "shiny" from the ethanol.
  14. Incubate the sample for 10 min at 37 °C in a thermoblock to elute the library from the beads.
  15. Place the tube in a magnet and transfer the supernatant containing the library to a new 1.5 mL DNA low-binding tube.
  16. Quantify the size and concentration by automated electrophoresis.

Representative Results

liCHi-C offers the possibility of generating high-quality and resolution genome-wide promoter interactome libraries with as little as 50,000 cells53. This is accomplished by – besides the drastic reduction of reaction volumes and the use of DNA low-binding plasticware throughout the protocol – removing unnecessary steps from the original protocol, in which significant material losses occur. These include the phenol purification after decrosslinking, the biotin removal, and subsequent phenol-chloroform purification and ethanol precipitation. Besides that, reorganizing the steps of the Hi-C library preparation (biotin pulldown, A-tailing, adapter ligation, PCR amplification, and double-sided paramagnetic bead selection-also as the PCR product purification) allows us to remove yet another unnecessary DNA purification step. An overview of the experimental workflow can be found in Figure 1A.

To assess library quality, several controls throughout the protocol are performed, the first of which is the calculation of genome digestion efficiency; values over 80% are considered acceptable (Table 3). Checking for the digestion efficiency of the cell type in a separate experiment is suggested in order to not lose a significant amount of material from a single liCHi-C experiment. Second, before sonication and end-repair, it is recommended to check for the sensitivity of interaction detection by amplifying cell-type invariant chromatin interactions by conventional PCR (Figure 1B). If the specific product is detected, a third control should be performed, focusing on the efficiency of biotinylation and ligation by differentially digesting one of the previously obtained PCR products with HindIII and NheI (Figure 1C,D). When filling a digested HindIII restriction site and blunt-end ligating it with another one, a new NheI restriction site is generated instead of regenerating the original HindIII one. Therefore, the digestion of the PCR amplicon should only be observed when NheI is present. Finally, just before and after the entire capture, the concentration and size distribution should be checked using automated electrophoresis. Pre-capture library amplification must aim at obtaining 500-1,000 ng of Hi-C material, the exact amount needed to perform the RNA probe capture, since excessive amplification of both the pre- and post-captured library leads to a high percentage of PCR duplicates and the consequent loss of sequencing reads during analysis. Libraries can be reamplified again under the same conditions if not enough material is obtained during the first conservative PCR amplification. The amount of post-captured library material can vary, but as a rule of thumb, it should be approximately tenfold to 20-fold less than before the capture. The size distribution of the library should fall around 450-550 bp (Figure 2A,B), invariable between pre- and post-captured libraries. Collectively, correct results of these controls ensure the generation of excellent liCHi-C libraries.

Finished liCHi-C libraries are then (at least) 100 bp paired-end sequenced and analyzed. Raw sequencing data53 is processed using the HiCUP pipeline56 for mapping and filtering out artifacts. The ideal HiCUP report shows a fivefold to tenfold increase in the distribution of cis (inside the same chromosome) compared to trans (between different chromosomes) paired-end reads, as described previously according to in-nucleus ligation Hi-C57 (Figure 3B). The obtention of more than 100 million unique, valid reads after the removal of PCR duplicates is enough to proceed to the following step in the analysis (Figure 3B), which is to assess the capture efficiency. Paired-end reads in which none of their ends map into a captured restriction fragment by the RNA probe enrichment system, are discarded, keeping only the ones representing the promoter interactome of the cell (i.e., those reads in which at least one of their ends maps into restriction fragments containing one or more gene promoters [Figure 3C], ideally more than 60%).

Finally, significant interactions are called with the CHiCAGO pipeline, as described in58,59. Two or more biological replicates are needed for the final set of significant promoter interactions. The data quality can also be validated using principal component analysis (PCA), since biological replicates must cluster together and cell types must be separated. For instance, by analyzing liCHi-C datasets from four different primary cell types from the human hematopoietic tree (common myeloid progenitors, monocytes, megakaryocytes, and erythroblasts), we can observe in a PCA that liCHi-C libraries cluster in a developmental trajectory-reflecting fashion (Figure 3D). A closer examination of the significant interactions detected for the four cell types reveals that promoter interactomes are cell-type specific and dynamic during cell development. For example, the AHSP gene, a key chaperone synthesized in erythroid precursors which oversees the correct folding of hemoglobin60,61,62, shows a gain of interactions with potentially active regulatory elements (i.e., H3K27ac and H3K4me1 enriched regions) in erythroblasts, but not in other cell types (Figure 4). This demonstrates that the liCHi-C method can uncover potential regulatory interactions in rare cell types.

Figure 1
Figure 1: Protocol overview and quality controls of the sample before sonication. (A) Schematic overview of the liCHi-C protocol divided by days. B and blue hands represent, respectively, biotin molecules and steps in which one can safely stop the protocol for a large period of time. (B) Representative results of the 3C interaction controls. Both interaction sets for human (left) and mouse (right) are shown. The bands to expect are marked in dark blue, while an unspecific band is marked in light blue. (C) Representative fill-in and ligation controls using the "Dekker" human interaction primer pair. The band is cut only in lanes 2 and 3, where NheI is added. (D) Schematic representation of the generation of a new NheI restriction site during fill-in and ligation of a HindIII restriction site. Please click here to view a larger version of this figure.

Figure 2
Figure 2: Representative automated electrophoresis profiles from libraries pre- and post-capture. (A) Automated electrophoresis profile from a library just before capture. The total amount of DNA obtained is 994 ng (49.7 ng/µL in 20 µL). (B) High-sensitivity automated electrophoresis profile from a finished liCHi-C library. The sample is loaded half-diluted to preserve as much material as possible. The total amount of DNA obtained is 61.2 ng (1.53 ng/µL in 20 µL x2 to account for the dilution). Please click here to view a larger version of this figure.

Figure 3
Figure 3: Representative HiCUP pipeline output and sample replicate clustering by PCA. (A) Classification of the validity of read pairs by percentages and total counts. Invalid read pairs are subclassified by the experimental artifact type. (B) Deduplication percentages and classification of the interaction types. Cis interactions are further subclassified as cis-close (less than 10 kb) and cis-far (more than 10 kb). (C) Percentage of capture efficiency. Captured read-pairs are further subclassified whether one end, the other, or both contain one or more promoters. (D) Principal component analysis of CHiCAGO significant interaction scores from both replicates of liCHi-C libraries from common myeloid progenitors (CMP), erythroblasts, megakaryocytes, and monocytes. Please click here to view a larger version of this figure.

Figure 4
Figure 4: AHSP interaction landscape in human primary hematopoietic cells. Representative example of the AHSP promoter-centered interaction landscape in common myeloid progenitors (CMP), erythroblasts (Ery), megakaryocytes (MK), and monocytes (Mon) as seen in the WashU Epigenome Browser63. Arcs represent significant interactions. The dark blue shade shows the AHSP gene promoter, while the light blue shades overlap potential active regulatory elements that interact specifically with the AHSP promoter in erythroblasts. Please click here to view a larger version of this figure.

Table 1: Buffer composition and preparation. Please click here to download this Table.

Table 2: PCR primers and adapter sequences. Please click here to download this Table.

Table 3: Example of the calculation for the digestion efficiency. Please click here to download this Table.

Discussion

liCHi-C offers the capability of generating high-resolution promoter interactome libraries using a similar experimental framework from PCHi-C's but with a vastly reduced cell number. This is greatly achieved by eliminating unnecessary steps, such as phenol purification and biotin removal. In the classical in-nucleus ligation Hi-C protocol57 and its subsequent derivative technique PCHi-C, biotin is removed from non-ligated restriction fragments to avoid pulling down DNA fragments that are afterward uninformative. Skipping this part and its subsequent DNA purification does not translate in a significant reduction in the percentage of valid reads (Figure 3A) while cutting out potential DNA-wasting steps, which are DNA purifications. The reorganization of the Hi-C library preparation after the sonication allows to skip yet another unnecessary purification by using the double-sided selection as the purification step itself. All of this enhances the performance of the whole protocol by employing minimal tube changes, and together with the reduction in reaction volume, changes in reagent concentration, and the use of DNA low-binding plasticware, is what allows the generation of high-quality libraries using as little as 50,000 cells. It is important to keep in mind that the starting cell number determines, in part, the number of significant interactions due to the library's complexity. Although libraries generated with 50,000 cells retain the cell-type specific and invariant topological features of more complex libraries53, our recommendation is to use, if possible, over 100,000 cells per biological replicate in order to capture a higher number of significant promoter interactions.

The resolution of the interactions detected in 3C techniques is essentially given by the restriction enzyme used. Here, the application of liCHi-C is described using HindIII, a 6 bp-cutting enzyme that gives a theoretical mean resolution of 4,096 bp. liCHi-C allows for the restriction enzyme to be replaced, for example, toward a 4 bp-cutting enzyme or even a micrococcal nuclease, thus increasing the resolution of the significant interactions detected. The generation of liCHi-C libraries, switching the HindIII restriction enzyme for the 4 bp-cutting MboI enzyme, has been reported to deliver excellent results detecting nearly double the total interactions, albeit with the shifting of the mean linear distance of the interactions detected down to half the distance53.

Regarding the actual RNA probe enrichment system, one of the main advantages of using this type of capture over an antibody-based one, such as HiChIP32 or HiCuT33, among others, is the ability to compare conditions independently of the binding of the protein, as well as not having to rely on the availability of a working antibody for the protein of interest. Furthermore, the RNA enrichment system can be tailored to capture any specific regions genome-wide to suit each investigator's need (the design is discussed in35,64).

In addition to antibody-based capture methods, several outstanding single-cell (such as scHi-C65, Dip-C66, or Sci-Hi-C67 among others) or low-input methods (such as Low-C68 or Easy Hi-C69) to investigate the 3D genome architecture have been developed in recent years. However, these generate sparse contact maps with low-resolution that do not allow the identification of contacts between distal regulatory elements and target genes. liCHi-C is a method that is able to overcome this limitation, opening the possibility of studying the promoter centric genome architecture in scarce cell types and providing the opportunity to advance our understanding of cellular and developmental biology and disease development.

Despite all of its features, liCHi-C is not exempt of limitations. First, processing the raw sequencing data is not trivial, and fair computational skills are needed to analyze the data and interpret its results. Moreover, liCHi-C does not discriminate between functional and structural interactions; it is required for the liCHi-C data to be integrated with epigenetic data and/or functional analysis to validate the potential functional interactions of gene promoters with their target regulatory elements. Lastly, library complexity is sacrificed when working on the lower end of cell numbers. This is reflected on the amount of unique interactions detected compared to higher cell number liCHi-C libraries and their deduplication rate, which can reach up to 80%. However, low-cell number liCHi-C libraries retain the topological features of higher cell number libraries in a more focal manner53, demonstrating that it is feasible to perform liCHi-C libraries that recapitulate the cell's promoter interactome with as little as 50,000 cells.

Overall, liCHi-C is a cost-effective and customizable method to generate high-quality and high-resolution promoter interactome libraries in scarce cell types. It is the first low-input method to map the promoter interactome and call for significant loops at the restriction fragment resolution. We foresee that this new tool, as its predecessor PCHi-C, will provide new insights in cell differentiation and organism development, both in health and disease.

Disclosures

The authors have nothing to disclose.

Acknowledgements

We thank the rest of the members from the Javierre lab for their feedback on the manuscript. We thank CERCA Program, Generalitat de Catalunya, and the Josep Carreras Foundation for institutional support. This work was financed by FEDER/Spanish Ministry of Science and Innovation (RTI2018-094788-A-I00), the European Hematology Association (4823998), and the Spanish Association against Cancer (AECC) LABAE21981JAVI. BMJ is funded by La Caixa Banking Foundation Junior Leader project (LCF/BQ/PI19/11690001), LR is funded by an AGAUR FI fellowship (2019FI-B00017), and LT-D is funded by an FPI Fellowship (PRE2019-088005). We thank the biochemistry and molecular biology PhD program from the Universitat Autònoma de Barcelona for its support. None of the funders were involved at any point in the experimental design or manuscript writing.

Materials

0.4 mM Biotin-14-dATP Invitrogen 19524-016
0.5 M EDTA pH 8.0 Invitrogen AM9260G
1 M Tris pH 8.0 Invitrogen AM9855G
10x NEBuffer 2 New England Biolabs B7002S Referenced as restriction buffer 2 in the manuscript
10x PBS Fisher Scientific BP3994
10x T4 DNA ligase reaction buffer New England Biolabs B0202S
16% formaldehyde solution (w/v), methanol-free Thermo Scientific 28908
20 mg/mL Bovine Serum Albumin New England Biolabs B9000S
5 M NaCl Invitrogen AM9760G
5PRIME Phase Lock Gel Light tubes Qiuantabio 2302820 For phenol-chloroform purification in section 4 (DNA purification). Phase Lock Gel tubes are a commercial type of tubes specially designed to maximize DNA recovery after phenol-chloroform purifications while avoiding carryover of contaminants in the organic phase by containing a resin of intermediate density which settles between the organic and aqueous phase and isolates them. PLG tubes should be spun at 12,000 x g for 30 s before use to ensure that the resin is well-placed at the bottom of the tube
Adapters and PCR primers for library amplification Integrated DNA Technologies Bought as individual primers with PAGE purification for NGS
Cell scrappers Nunc 179693 Or any other brand
Centrifuge (fixed-angle rotor for 1.5 mL tubes) Any brand
CHiCAGO R package 1.14.0
CleanNGS beads CleanNA CNGS-0050
dATP, dCTP, dGTP, dTTP Promega U120A, U121A, U122A, U123A Or any other brand
DNA LoBind tube, 1.5 mL Eppendorf 30108051
DNA LoBind tube, 2 mL Eppendorf 30108078
DNA polymerase I large (Klenow) fragment 5000 units/mL New England Biolabs M0210L
Dynabeads MyOne Streptavidin C1 beads Invitrogen 65002 For biotin pull-down of the pre-captured library in section 8 (biotin pull-down)
Dynabeads MyOne Streptavidin T1 beads Invitrogen 65602 For biotin pull-down of the post-captured library in section 11 (biotin pull-down and PCR amplification)
DynaMag-2 Invitrogen 12321D Or any other magnet suitable for 1.5 ml tubeL
Ethanol absolute VWR 20821.321
FBS, qualified Gibco 10270-106 Or any other brand
Glycine Fisher BioReagents BP381-1
GlycoBlue Coprecipitant Invitrogen AM9515 Used for DNA coprecipitation in section 4 (DNA purification)
HiCUP 0.8.2
HindIII, 100 U/µL New England Biolabs R0104T
IGEPAL CA-630 Sigma-Aldrich I8896-50ML
Klenow EXO- 5000 units/mL New England Biolabs M0212L
Low-retention filter tips (10 µL, 20 µL, 200 µL and 1000 µL) ZeroTip PMT233010, PMT252020, PMT231200, PMT252000
M220 Focused-ultrasonicator Covaris 500295
Micro TUBE AFA Fiber Pre-slit snap cap 6 x 16 mm vials Covaris 520045 For sonication in section 6 (sonication)
NheI-HF, 100 U/µL New England Biolabs R3131M
Nuclease-free molecular biology grade water Sigma-Aldrich W4502
PCR primers for quality controls Integrated DNA Technologies
PCR strips and caps Agilent Technologies 410022, 401425
Phenol: Chloroform: Isoamyl Alcohol 25:24:1, Saturated with 10 mM Tris, pH 8.0, 1 mM EDTA Sigma-Aldrich P3803
Phusion High-Fidelity PCR Master Mix with HF Buffer New England Biolabs M0531L For amplification of the library in sections 9 (dATP-tailing, adapter ligation and PCR amplification)
and 11 (biotin pull-down and PCR amplification)
Protease inhibitor cocktail (EDTA-free) Roche 11873580001
Proteinase K, recombinant, PCR grade Roche 3115836001
Qubit 1x dsDNA High Sensitivity kit Invitrogen Q33230 For DNA quantification after precipitation in section 4 (DNA purification)
Qubit assay tubes Invitrogen Q32856
rCutsmart buffer New England Biolabs B6004S
RPMI Medium 1640 1x + GlutaMAX Gibco 61870-010 Or any other brand
SDS – Solution 10% for molecular biology PanReac AppliChem A0676
Sodium acetate pH 5.2 Sigma-Aldrich S7899-100ML
SureSelect custom 3-5.9 Mb library Agilent Technologies 5190-4831 Custom designed mouse or human capture system, used for the capture
SureSelect Target Enrichment Box 1 Agilent Technologies 5190-8645 Used for the capture
SureSelect Target Enrichment Kit ILM PE Full Adapter Agilent Technologies 931107 Used for the capture
T4 DNA ligase 1 U/µL Invitrogen 15224025 For ligation in section 3 (ligation and decrosslink)
T4 DNA ligase 2000000/mL New England Biolabs M0202T For ligation in section 9 (dATP-tailing, adapter ligation and PCR amplification)
T4 DNA polymerase 3000 units/mL New England Biolabs M0203L
T4 PNK 10000 units/mL New England Biolabs M0201L
Tapestation 4200 instrument Agilent Technologies For automated electrophoresis in section 9 (dATP-tailing, adapter ligation, and PCR amplification) and
section 11
(Biotin pull-down and PCR amplification). Any other automated electrophoresis system is valid
Tapestation reagents Agilent Technologies 5067-5582, 5067-5583, 5067-5584, 5067-5585, For automated electrophoresis in section 9 (dATP-tailing, adapter ligation, and PCR amplification) and
section 11
(Biotin pull-down and PCR amplification). Any other automated electrophoresis system is valid
Triton X-100 for molecular biology PanReac AppliChem A4975
Tween 20 Sigma-Aldrich P9416-50ML

References

  1. Hatton, C. S., et al. α-thalassemia caused by a large (62 kb) deletion upstream of the human α globin gene cluster. Blood. 76 (1), 221-227 (1990).
  2. Toikkanen, S., Helin, H., Isola, J., Joensuu, H. Prognostic significance of HER-2 oncoprotein expression in breast cancer: A 30-year follow-up. Journal of Clinical Oncology. 10 (7), 1044-1048 (1992).
  3. Church, C., et al. Overexpression of Fto leads to increased food intake and results in obesity. Nature Genetics. 42 (12), 1086-1092 (2010).
  4. Bhatia, S., et al. Disruption of autoregulatory feedback by a mutation in a remote, ultraconserved PAX6 enhancer causes aniridia. American Journal of Human Genetics. 93 (6), 1126-1134 (2013).
  5. Herranz, D., et al. A NOTCH1-driven MYC enhancer promotes T cell development, transformation and acute lymphoblastic leukemia. Nature Medicine. 20 (10), 1130-1137 (2014).
  6. Carter, D., Chakalova, L., Osborne, C. S., Dai, Y. F., Fraser, P. Long-range chromatin regulatory interactions in vivo. Nature Genetics. 32 (4), 623-626 (2002).
  7. Rao, S. S. P., et al. A 3D map of the human genome at kilobase resolution reveals principles of chromatin looping. Cell. 159 (7), 1665-1680 (2014).
  8. Schoenfelder, S., Fraser, P. Long-range enhancer-promoter contacts in gene expression control. Nature Reviews Genetics. 20 (8), 437-455 (2019).
  9. Heintzman, N. D., et al. Distinct and predictive chromatin signatures of transcriptional promoters and enhancers in the human genome. Nature Genetics. 39 (3), 311-318 (2007).
  10. Zentner, G. E., Tesar, P. J., Scacheri, P. C. Epigenetic signatures distinguish multiple classes of enhancers with distinct cellular functions. Genome Research. 21 (8), 1273-1283 (2011).
  11. Creyghton, M. P., et al. Histone H3K27ac separates active from poised enhancers and predicts developmental state. Proceedings of the National Academy of Sciences. 107 (50), 21931-21936 (2010).
  12. McPherson, C. E., Shim, E. Y., Friedman, D. S., Zaret, K. S. An active tissue-specific enhancer and bound transcription factors existing in a precisely positioned nucleosomal array. Cell. 75 (2), 387-398 (1993).
  13. He, A., Kong, S. W., Ma, Q., Pu, W. T. Co-occupancy by multiple cardiac transcription factors identifies transcriptional enhancers active in heart. Proceedings of the National Academy of Sciences. 108 (14), 5632-5637 (2011).
  14. Dogan, N., et al. Occupancy by key transcription factors is a more accurate predictor of enhancer activity than histone modifications or chromatin accessibility. Epigenetics and Chromatin. 8, 16 (2015).
  15. Whyte, W. A., et al. transcription factors and mediator establish super-enhancers at key cell identity genes. Cell. 153 (2), 307-319 (2013).
  16. Hnisz, D., et al. Super-enhancers in the control of cell identity and disease. Cell. 155 (4), 934-947 (2013).
  17. He, H. H., et al. Nucleosome dynamics define transcriptional enhancers. Nature Genetics. 42 (4), 343-347 (2010).
  18. Song, L., et al. Open chromatin defined by DNaseI and FAIRE identifies regulatory elements that shape cell-type identity. Genome Research. 21 (10), 1757-1767 (2011).
  19. De Santa, F., et al. A large fraction of extragenic RNA Pol II transcription sites overlap enhancers. PLoS Biology. 8 (5), e1000384 (2010).
  20. Kim, T. K., et al. Widespread transcription at neuronal activity-regulated enhancers. Nature. 465 (7295), 182-187 (2010).
  21. ENCODE Project Consortium. An integrated encyclopedia of DNA elements in the human genome. Nature. 489 (7414), 57-74 (2012).
  22. Wu, H., et al. Tissue-specific RNA expression marks distant-acting developmental enhancers. PLoS Genetics. 10 (9), e1004610 (2014).
  23. Banerji, J., Rusconi, S., Schaffner, W. Expression of a β-globin gene is enhanced by remote SV40 DNA sequences. Cell. 27 (2), 299-308 (1981).
  24. Amano, T., et al. Chromosomal dynamics at the Shh locus: limb bud-specific differential regulation of competence and active transcription. Developmental Cell. 16 (1), 47-57 (2009).
  25. Shi, J., et al. Role of SWI/SNF in acute leukemia maintenance and enhancer-mediated Myc regulation. Genes and Development. 27 (24), 2648-2662 (2013).
  26. Lettice, L. A., et al. A long-range Shh enhancer regulates expression in the developing limb and fin and is associated with preaxial polydactyly. Human Molecular Genetics. 12 (14), 1725-1735 (2003).
  27. Tuvikene, J., et al. Intronic enhancer region governs transcript-specific Bdnf expression in rodent neurons. eLife. 10, e65161 (2021).
  28. Tasic, B., et al. Promoter choice determines splice site selection in protocadherin α and γ pre-mRNA splicing. Molecular Cell. 10 (1), 21-33 (2002).
  29. Perry, M. W., Boettiger, A. N., Levine, M. Multiple enhancers ensure precision of gap gene-expression patterns in the Drosophila embryo. Proceedings of the National Academy of Sciences. 108 (33), 13570-13575 (2011).
  30. Lieberman-Aiden, E., et al. Comprehensive mapping of long-range interactions reveals folding principles of the human genome. Science. 326 (5950), 289-293 (2009).
  31. Fullwood, M. J., et al. An oestrogen-receptor-α-bound human chromatin interactome. Nature. 462 (7269), 58-64 (2009).
  32. Mumbach, M. R., et al. HiChIP: Efficient and sensitive analysis of protein-directed genome architecture. Nature Methods. 13 (11), 919-922 (2016).
  33. Sati, S., et al. HiCuT: An efficient and low input method to identify protein-directed chromatin interactions. PLoS Genetics. 18 (3), e1010121 (2022).
  34. Schoenfelder, S., et al. The pluripotent regulatory circuitry connecting promoters to their long-range interacting elements. Genome Research. 25 (4), 582-597 (2015).
  35. Schoenfelder, S., Javierre, B. M., Furlan-Magaril, M., Wingett, S. W., Fraser, P. Promoter capture Hi-C: High-resolution, genome-wide profiling of promoter interactions. Journal of Visualized Experiments. (136), e57320 (2018).
  36. Rubin, A. J., et al. Lineage-specific dynamic and pre-established enhancer-promoter contacts cooperate in terminal differentiation. Nature Genetics. 49 (10), 1522-1528 (2017).
  37. Siersbæk, R., et al. Dynamic rewiring of promoter-anchored chromatin loops during adipocyte differentiation. Molecular Cell. 66 (3), 420-435 (2017).
  38. Schoenfelder, S., et al. Polycomb repressive complex PRC1 spatially constrains the mouse embryonic stem cell genome. Nature Genetics. 47 (10), 1179-1186 (2015).
  39. Zhang, N., et al. Muscle progenitor specification and myogenic differentiation are associated with changes in chromatin topology. Nature Communications. 11 (1), 6222 (2020).
  40. Javierre, B. M., et al. Lineage-specific genome architecture links enhancers and non-coding disease variants to target gene promoters. Cell. 167 (5), 1369-1384 (2016).
  41. Jäger, R., et al. Capture Hi-C identifies the chromatin interactome of colorectal cancer risk loci. Nature Communications. 6, 6178 (2015).
  42. Martin, P., et al. Identifying causal genes at the multiple sclerosis associated region 6q23 using capture Hi-C. PLoS One. 11 (11), e0166923 (2016).
  43. Burren, O. S., et al. Chromosome contacts in activated T cells identify autoimmune disease candidate genes. Genome Biology. 18 (1), 165 (2017).
  44. Choy, M. K., et al. Promoter interactome of human embryonic stem cell-derived cardiomyocytes connects GWAS regions to cardiac gene networks. Nature Communications. 9 (1), 2526 (2018).
  45. Miguel-Escalada, I., et al. Human pancreatic islet three-dimensional chromatin architecture provides insights into the genetics of type 2 diabetes. Nature Genetics. 51 (7), 1137-1148 (2019).
  46. Law, P. J., et al. Association analyses identify 31 new risk loci for colorectal cancer susceptibility. Nature Communications. 10 (1), 2154 (2019).
  47. Speedy, H. E., et al. Insight into genetic predisposition to chronic lymphocytic leukemia from integrative epigenomics. Nature Communications. 10 (1), 3615 (2019).
  48. Li, T., et al. Epigenomics and transcriptomics of systemic sclerosis CD4+ T cells reveal long-range dysregulation of key inflammatory pathways mediated by disease-associated susceptibility loci. Genome Medicine. 12 (1), 81 (2020).
  49. Orlando, G., et al. Promoter capture Hi-C-based identification of recurrent noncoding mutations in colorectal cancer. Nature Genetics. 50 (10), 1375-1380 (2018).
  50. Cornish, A. J., et al. Identification of recurrent noncoding mutations in B-cell lymphoma using capture Hi-C. Blood Advances. 3 (1), 21-32 (2019).
  51. Madsen, J. G. S., et al. Highly interconnected enhancer communities control lineage-determining genes in human mesenchymal stem cells. Nature Genetics. 52 (11), 1227-1238 (2020).
  52. Dryden, N. H., et al. Unbiased analysis of potential targets of breast cancer susceptibility loci by Capture Hi-C. Genome Research. 24 (11), 1854-1868 (2014).
  53. Tomás-Daza, L., et al. Low input capture Hi-C (liCHi-C) identifies promoter-enhancer interactions at high-resolution. Nature Communications. 14 (1), 268 (2023).
  54. Lee, P. Y., Costumbrado, J., Hsu, C. Y., Kim, Y. H. Agarose gel electrophoresis for the separation of DNA fragments. Journal of Visualized Experiments. 62 (62), e3923 (2012).
  55. Bronner, I. F., Quail, M. A. Best practices for Illumina library preparation. Current Protocols in Human Genetics. 102 (1), 86 (2019).
  56. Wingett, S., et al. HiCUP: pipeline for mapping and processing Hi-C data. F1000Research. 4, 1310 (2015).
  57. Nagano, T., et al. Comparison of Hi-C results using in-solution versus in-nucleus ligation. Genome Biology. 16 (1), 175 (2015).
  58. Cairns, J., et al. CHiCAGO: Robust detection of DNA looping interactions in Capture Hi-C data. Genome Biology. 17 (1), 127 (2016).
  59. Freire-Pritchett, P., et al. Detecting chromosomal interactions in Capture Hi-C data with CHiCAGO and companion tools. Nature Protocols. 16 (9), 4144-4176 (2021).
  60. Kihm, A. J., et al. An abundant erythroid protein that stabilizes free α-haemoglobin. Nature. 417 (6890), 758-763 (2002).
  61. Feng, L., et al. Molecular mechanism of AHSP-mediated stabilization of α-hemoglobin. Cell. 119 (5), 629-640 (2004).
  62. Favero, M. E., Costa, F. F. Alpha-hemoglobin-stabilizing protein: An erythroid molecular chaperone. Biochemistry Research International. 2011, 373859 (2011).
  63. Li, D., et al. WashU Epigenome Browser update 2022. Nucleic Acids Research. 50 (W1), W774-W781 (2022).
  64. Mifsud, B., et al. Mapping long-range promoter contacts in human cells with high-resolution capture Hi-C. Nature Genetics. 47 (6), 598-606 (2015).
  65. Nagano, T., et al. Single-cell Hi-C reveals cell-to-cell variability in chromosome structure. Nature. 502 (7469), 59-64 (2013).
  66. Tan, L., Xing, D., Chang, C. H., Li, H., Xie, X. S. 3D genome structures of single diploid human cells. Science. 361 (6405), 924 (2018).
  67. Ramani, V., et al. Massively multiplex single-cell Hi-C. Nature Methods. 14 (3), 263-266 (2017).
  68. Díaz, N., et al. Chromatin conformation analysis of primary patient tissue using a low input Hi-C method. Nature Communications. 9 (1), 4938 (2018).
  69. Lu, L., Jin, F. Easy Hi-C: A low-input method for capturing genome organization. Methods in Molecular Biology. 2599, 113-125 (2023).

Play Video

Cite This Article
Rovirosa, L., Tomás-Daza, L., Urmeneta, B., Valencia, A., Javierre, B. M. An Integrated Workflow to Study the Promoter-Centric Spatio-Temporal Genome Architecture in Scarce Cell Populations. J. Vis. Exp. (194), e65316, doi:10.3791/65316 (2023).

View Video