Generation of High Quality Chromatin Immunoprecipitation DNA Template for High-throughput Sequencing (ChIP-seq)

Summary

The combination of chromatin immunoprecipitation and ultra-high-throughput sequencing (ChIP-seq) can identify and map protein-DNA interactions in a given tissue or cell line. Outlined is how to generate a high quality ChIP template for subsequent sequencing, using experience with the transcription factor TCF7L2 as an example.

Abstract

ChIP-sequencing (ChIP-seq) methods directly offer whole-genome coverage, where combining chromatin immunoprecipitation (ChIP) and massively parallel sequencing can be utilized to identify the repertoire of mammalian DNA sequences bound by transcription factors in vivo. "Next-generation" genome sequencing technologies provide 1-2 orders of magnitude increase in the amount of sequence that can be cost-effectively generated over older technologies thus allowing for ChIP-seq methods to directly provide whole-genome coverage for effective profiling of mammalian protein-DNA interactions.

For successful ChIP-seq approaches, one must generate high quality ChIP DNA template to obtain the best sequencing outcomes. The description is based around experience with the protein product of the gene most strongly implicated in the pathogenesis of type 2 diabetes, namely the transcription factor transcription factor 7-like 2 (TCF7L2). This factor has also been implicated in various cancers.

Outlined is how to generate high quality ChIP DNA template derived from the colorectal carcinoma cell line, HCT116, in order to build a high-resolution map through sequencing to determine the genes bound by TCF7L2, giving further insight in to its key role in the pathogenesis of complex traits.

Introduction

For many years there has been an unmet need to identify the set of genes bound and regulated by a given protein genome wide, in particular those in the transcription factor class.

Odom et al.¹ used chromatin immunoprecipitation (ChIP) combined with promoter microarrays to systematically identify the genes occupied by pre-specified transcriptional regulators in human liver and pancreatic islets. Subsequently, Johnson et al.² developed a large-scale chromatin immunoprecipitation assay based on direct ultra high-throughput DNA sequencing (ChIP-seq) in order to comprehensively map protein-DNA interactions across entire mammalian genomes. As a test case, they mapped in vivo the binding of the neuron-restrictive silencer factor (NRSF) to 1946 locations in the human genome. The data displayed sharp resolution of binding position (+50 base pairs), which facilitated both the isolation of motifs and the identification of NRSF-binding motifs. These ChIP-seq data also had high sensitivity and specificity and statistical confidence (P < 10⁻⁴), properties that are important for inferring new candidate interactions.

Robertson et al.³ also used ChIP-seq in order to map STAT1 targets in interferon-γ (IFN-γ)-stimulated and unstimulated human HeLa S3 cells in vivo. By ChIP-seq, using 15.1 and 12.9 million uniquely mapped sequence reads, and an estimated false discovery rate of less than 0.001, they identified 41,582 and 11,004 putative STAT1-binding regions in stimulated and unstimulated cells, respectively. Of the 34 loci known to contain STAT1 interferon-responsive binding sites^4-8, ChIP-seq found 24 (71%). ChIP-seq targets were enriched in sequences similar to known STAT1 binding motifs. Comparisons with two existing ChIP-PCR data sets suggested that ChIP-seq sensitivity was between 70% and 92% and specificity was at least 95%. Additionally, it was clear that ChIP-seq offers both low analytical complexity and sensitivity that increases with sequencing depth.

As such, "next-generation" genome sequencing technologies provide 1-2 orders of magnitude increase in the amount of sequence that can be cost-effectively generated over older technologies⁹. ChIP-seq methods therefore directly provide whole-genome coverage for effective profiling of mammalian protein-DNA interactions³.

In 2006, a strong association of variants in the transcription factor 7-like 2 (TCF7L2) gene with type 2 diabetes was discovered¹⁰. Other investigators have already independently replicated this finding in different ethnicities and, interestingly, from the first genome wide association studies of type 2 diabetes published in Nature^11,12, Science^13-15 and elsewhere^16,17, the strongest association was indeed with TCF7L2; this is now considered the most significant genetic finding in type 2 diabetes to date^18-20. In addition, TCF7L2 has been linked to cancer risk ^21,22; indeed, this connection became more obvious when the 8q24 locus revealed by genome wide association studies of a number of cancers, including colorectal carcinomas, was shown to be due to an extreme upstream TCF7L2-binding element driving the transcription of MYC^23,24. As such, there is great interest in determining the downstream genes regulated by this key transcription factor.

Based on experience with TCF7L2 as an example of the methodology, this paper outlines how to generate high quality ChIP DNA template. ChIP was carried out in the colorectal carcinoma cell line, HCT116, for subsequent sequencing in order build a high-resolution map of the genes bound by TCF7L2²⁵ in an endeavor to yield further insight in to its key role in the pathogenesis of complex traits.

Protocol

1. Cross-link Chromatin

1. Grow cells in 100x20mm cell culture dishes. The amount of cells can range from 1 to 10 million cells per dish depending on cell type. Approximately 2 million cells is sufficient for one immunoprecipitation.
2. Cross-link cells in 1% Formaldehyde for 10 min at room temperature with occasional rocking.
3. Quench cross-linking by adding a final concentration of 125 mM Glycine and incubate for 5 min at room temperature.
4. Wash cells with 1X Phosphate Buffered Saline (PBS) twice, decant PBS, and then add 0.2 ml of PBS.
5. Harvest cells with a plastic cell scraper into a microcentrifuge tube.
6. Spin down cells at 2,000 rpm for 5 min at 4 °C.
7. Aspirate the supernatant. Resuspend cells in SDS Lysis buffer (1% SDS, 10mM EDTA, 50mM Tris-HCl pH 8.1) for whole cell lysate or keep them as a pellet for nuclear extraction.
8. Cells can be saved at -80 °C or one can proceed immediately with sonication.

2. Prepare Nuclei (Proceed to Step 3.5 for Whole Cell Lysate)

1. Supplement Cell Lysis Buffer (5 mM PIPES pH 8.0, 85 mM KCl, 0.5% NP-40) with 1X Proteinase Inhibitor each experiment.
2. Resuspend thawed cell pellet in approximately 10 times the pellet volume with Cell Lysis Buffer.
3. Dounce-homogenize 10 times with pestle then incubate on ice for 10 min.
4. Centrifuge sample at 4,000 rpm for 5 min at 4 °C, discard supernatant, and save nuclear pellet.

3. Sonication*

1. Warm up SDS Lysis Buffer and supplement amount of buffer to be used with Proteinase Inhibitor.
2. Resuspend nuclear pellet in SDS Lysis Buffer (approximately 0.5 ml of buffer per 1-10 million cells)
3. Incubate on ice for 10 min.
4. Add 0.5 ml aliquots of samples to microcentrifuge tubes.
5. Sonicate on wet ice using Misonix sonicator with 30 sec on and 45 sec off at an amplitude setting of 2. Number of cycles for ideal fragment size can be determined by first trying out various cycle numbers (ex. 2, 4, 8, 12, 16, and 20 or more cycles). A different brand of sonicator may be used, however, the conditions will vary. Experimentation with the number of cycles and amount of time on and off must be carried out in order to determine the ideal conditions.
6. Collect 20 μl of each sample to check sonication results and to do quantification,. The rest of the sample can be stored at -80 °C.
7. Dilute the 20 μl of sample by adding 30 μl of 0.1X TE Buffer.
8. Treat sample with 1 μl of RNase A at 37 °C for 1 hr then add 1 μl of Proteinase K and incubate at 62 °C for 2 hr.
9. Run 20 μl of the sample on a 2% agarose gel.
10. Purify the remaining amount of sample with QIAquick PCR Purification kit then quantify using NanoDrop spectrophotometer.

*For native ChIP, microccocal nuclease digestion may be used alternatively to shear the DNA.

4. Block Agarose Beads*

1. If beads are already blocked, proceed to step 5.1.
2. Use Protein A or Protein G agarose. For 5 immunoprecipitations (IPs), use 600 μl of 50% bead slurry (300 μl bead pellet)
3. To wash the beads, spin them down at 800 rpm for 1 min at 4 °C and discard supernatant. Add slightly more than 2 ml ChIP Dilution Buffer (0.01% SDS, 1.2 mM EDTA, 167 mM NaCl, 1.1% Triton X-100, 16.7 mM Tris-HCl pH 8.1) and mix by slowly inverting the tube 10X. Spin down again at 800 rpm for 1 min at 4 °C and discard supernatant. Repeat the wash 2 more times.
4. Block beads by rotating at 4 °C overnight in blocking solution. Refer to Table 1 for the recipe of the blocking solution.

5. Pre-clear Chromatin

1. Thaw sonicated chromatin on ice.
2. Spin down at 12,000 rpm for 10 min at 4 °C then put on ice right away to remove SDS (white pellet).
3. Collect supernatant, discard pellet, and combine samples if necessary.
4. Take out the amounts needed for the experiment based on calculations (1-10 ug of chromatin per IP).
5. Dilute chromatin 10X in ChIP Dilution Buffer supplemented with proteinase inhibitor.
6. Add 100 μl of blocked beads per IP.
7. Rotate at 4 °C for 1 hr.

6. Immunoprecipitation

1. Spin down samples at 800 rpm for 1 min and transfer the supernatant to a fresh tube.
2. Spin down supernatant at 800 rpm for 1 min and transfer to another clean tube.
3. Save 20 μl of the supernatant, to serve as the input control, at -20 °C.
4. Aliquot the chromatin to the number of IPs to be done in the experiment.
5. Add 2 ug of antibody per 1-10 ug of chromatin to each sample.
6. Incubate overnight 4 °C with rotation.
7. Add 100 μl of blocked beads to each IP sample.
8. Incubate for 1 hr at 4 °C with rotation.
9. Pellet the beads by spinning down at 800 rpm for 1 min and discard as much of the supernatant as possible.
10. Wash beads once with Low Salt Immune Complex wash buffer. Add 1 ml of buffer to each tube; rotate at room temperature for 5-8 min; spin down at 800 rpm for 1 min, then discard supernatant. Repeat wash once with High Salt Immune Complex wash buffer and LiCl Immune Complex wash buffer and twice with TE Buffer for a total of 5 washes (Table 2).

7. Elution

1. Thaw input samples from previous day to be processed with the eluants.
2. Make Elution Buffer fresh (Table 3).
3. Make a master mix of enough Elution Buffer needed for IPs and input control samples plus 1-2 extra samples.
4. Add 100 μl Elution Buffer to each IP sample and incubate at room temp for 15 min with rotation.
5. Spin down at 800 rpm for 1 min and add supernatant to a fresh tube.
6. Add another 100 μl of Elution buffer to each tube of beads and incubate at room temp for 15 min with rotation.
7. Vortex for 15 sec after incubation; spin down at 5,000 rpm for 1 min, then combine supernatant with the supernatant from the first elution. (Make sure there are no left over beads in the supernatants. If unsure, spin down supernatant again at 5,000 rpm for 1 min and collect the supernatant in a new tube.
8. Add 180 μl of Elution Buffer to the 20 μl of input control samples.

8. Reverse Cross-link

1. To the 200 μl of eluants and input controls, add 8 μl of 5 M NaCl.
2. Seal tubes with parafilm and incubate in water bath at 65 °C overnight.

9. DNA Purification

1. Treat each sample with 1 μl of RNase A for 1 hr at 37 °C.
2. Add 4 μl of 0.5 M EDTA, 8 μl 1M Tris-HCl, mix, then add 1 μl of Proteinase K to each sample and incubate at 45 °C for 2 hr.
3. Purify the samples using QIAquick PCR Purification kit. The samples can be saved at -20 °C and PCR check can be done on a later date.

*Alternatively, ChIP-grade magnetic beads can be used in place of agarose for the immunoprecipitation portion.

10. PCR check

1. For the PCR check, use primers for regions known to be bound by the protein of interest. Also, use primers for non-binding regions as negative controls.
2. Mix the reagents for the reaction. Dilute Input sample at 1:100 (Table 3).
3. Run reaction. PCR Program:

Step 1: 94 °C 3 min

Step 2: 94 °C 20 sec
59 °C 30 sec
72 °C 30 sec
(Repeat Step 2 for at least 30 cycles)

Step 3: 72 °C 2 min

4. Run samples on 1% agarose gel.
5. Enrichment can also be determined quantitatively with real-time PCR.

Results

Once the chromatin has been sonicated and have been treated with RNase and Proteinase, the samples run on the 2% agarose gel should present a smear with the bulk of the DNA at the desired size. If several different cycles are tested, a gradual decrease in size should be seen as the number of cycles increase (Figure 2).

After completing the immunoprecipitation portion of the protocol the enrichment can either be checked by PCR or real-time PCR. For PCR samples run on an a...

Discussion

It is now feasible to carry out a genome-wide profile of protein-DNA interactions association using ChIP-seq, as has been very recently demonstrated with other transcription factors^2,3. The key to a successful sequencing outcome is the generation of a high quality chromatin immunoprecipitation DNA template.

Once the DNA template has been generated and ascertained to be adequately enriched, one can then take it in library preparation for subsequent sequencing. For example, one...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
QIAquick PCR Purification Kit	Qiagen	28104
EZ-ChIP Kit	Millipore	17-371
GoTaq Hot Start Polymerase	Promega	M5001
Misonix Sonicator	Qsonica	XL-2000
NanoDrop 1000 Spectrophotometer	Thermo-Scientific
Positive control primer sequences (TCF7L2-1) Forward- 5'-TCGCCCTGTCAATAATCTCC-3' Reverse- 5'-GCTCACCTCCTGTATCTTCG-3' Negative control primer sequences (CTRL-1) Forward-5'-ATGTGGTGTGGCTGTGATGGGAAC-3' Reverse- 5'-CGAGCAATCGGTAAATAGGTCTGG-3'