UNIT 21.3 Chromatin Immunoprecipitation for Determining the Association of Proteins with Specific Genomic Sequences In Vivo



37% formaldehyde: store up to 1 year at room temperature
2.5 M glycine, heat sterilized
APPENDIX 2), ice cold
UNIT 3.13)
UNIT 2.5A)
Primary antibody against protein or epitope of interest
50% (v/v) Ultralink Protein A/G-Silica beads (Pierce) or equivalent in 13.1 & 13.2), phenol/chloroform extraction and ethanol precipitation (UNIT 2.1A), PCR (UNIT 15.1 & 15.7), agarose gel electrophoresis (UNIT 2.5A), and nondenaturing acrylamide gel electrophoresis (UNIT 2.7).

Cross-link protein-DNA complexes in vivo

1. For each sample, grow 200 ml Saccharomyces cerevisiae to OD600 = 0.6 to 0.8 (UNITS 13.1 & 13.2).

CAUTION: Keep cultures covered or work in a fume hood to avoid noxious formaldehyde fumes.

The volumes of culture can be reduced (20 ml is a reasonable minimum) or increased depending on need. Typically, 20 to 40 ml yeast is used for an individual immunoprecipitation, so the 200-ml volume permits multiple immunoprecipitations from the same cells. This is particularly useful for experiments involving the analysis of multiple factors or for carrying out independent immunoprecipitations involving the same factor for data reproducibility.

2. Add 5.5 ml of 37% formaldehyde (1% final). Cross-link 15 to 20 min at room temperature by occasionally swirling flask or shaking slowly on a platform.

3. Add 30 ml heat-sterilized 2.5 M glycine and incubate an additional 5 min at room temperature.

Glycine stops the cross-linking by reacting with formaldehyde.

Harvest cells

4. Centrifuge cells 5 min at 2500 ´ g, 4°C. Discard supernatant into a chemical waste container and resuspend pellet in 50 to 200 ml ice-cold TBS. Repeat once.

5. Centrifuge cells for a third time 5 min at 2500 ´ g, 4°C. Discard supernatant and resuspend cells in 10 ml ice-cold FA lysis buffer.

6. Pellet cells by centrifuging in a benchtop centrifuge 5 min at 3000 rpm, 4°C. Discard supernatant.

The cells can remain on ice for a few hours while other samples are being collected so that all samples may be processed as a group from this point onward. Alternatively, the cells may be frozen in liquid nitrogen or a dry ice/ethanol bath and stored up to several months at -80°C. This is particularly helpful if multiple samples are being generated during a time-course experiment. If cells are frozen, they must be thawed on ice before continuing with the procedure.

Lyse cells and isolate chromatin

For lysis using a mini bead beater (preferred)

7a. Resuspend the cell pellet in 1 ml ice-cold FA lysis buffer/2 mM PMSF. Fill three-quarters of a 2-ml flat-bottomed screw-cap microcentrifuge tube with ~0.5-mm-diameter silica-zirconia or glass beads. Add cells, taking care to avoid introduction of bubbles, and screw the cap on tightly. Make sure there are no leaks.

The mini bead beater is recommended, because it is more efficient at breaking cells (multiple samples can be broken simultaneously). Silica-zirconia beads are more efficient at breaking cells than glass beads and are also recommended. To facilitate cell breakage with the mini bead beater, it is important that the final suspension nearly fill the tube. Do not break >160 OD600 units of cells (i.e., <5 ´ 109 cells) in a single 2-ml tube; for larger cultures, split the cells into multiple tubes.

8a. Lyse cells 3 min with a mini bead beater at maximum speed. Remove sample and incubate 1 min in an ice-water bath. Repeat five times for a total breakage time of 18 min.

This step assumes breakage with silica-zirconia beads. The cell breakage time with glass beads may be longer.

For lysis using an individual or multivortexer

7b. Resuspend in 250 ul FA lysis buffer/2 mM PMSF. Add 350 ul silica-zirconia or glass beads to a 2-ml microcentrifuge tube with relatively flat bottom. Add cells.

When using a multivortexer (or standard vortexer), it is important to keep the volume small as this improves cell breakage. Do not break >160 OD600 units of cells (i.e., <5 ´ 109 cells) in a single 2-ml tube; for larger cultures, split the cells into multiple tubes.

8b. Vortex continuously on an individual or multivortexer 30 min at full speed, 4°C.

Success and reproducibility of the ChIP procedure is aided by complete (or near-complete) breakage of cells. In this regard, formaldehyde-cross-linked cells are considerably harder to break than untreated cells. The use of 1.5-ml microcentrifuge tubes with conical bottoms should be avoided because the narrow shape constricts bead movement, resulting in unequal lysis among samples. The 2-ml microcentrifuge tubes have a nearly flat bottom that allows the beads to vortex vigorously. The indicated vortexing or bead-beating conditions should be tested if a different device is used.

Isolate lysate

9. Cut a 5-ml syringe ~1 cm below the flared opening (i.e., where the plunger is inserted) with a razor. Insert the smaller portion into a 15-ml disposable conical tube so that the flared portion of the truncated syringe rests on top of the conical tube opening, forming a microcentrifuge-tube holder.

10. Invert the sample tube and punch a hole in the bottom with a 25-G needle. Place the sample tube into the syringe/conical tube and punch a hole in the top cover with the same needle.

11. Spin the assembly in a benchtop centrifuge 1 min at 1000 rpm, 4°C. Place the conical tube on ice. Discard the 2-ml centrifuge tube containing the dry beads after confirming the sample has been transferred to the 15-ml tube.

Occasionally, beads will clog the pierced hole and prevent complete transfer of the sample. If this occurs, pierce the tube one or two more times and repeat the step in the same 15-ml tube. No additional buffer should be added.

Shear DNA

12. Transfer the sample to a standard 1.5-ml microcentrifuge tube. Microcentrifuge 15 min at maximum speed, 4°C. Discard the supernatant and add 1 ml ice-cold FA lysis buffer to the pellet.

The pellet contains the cross-linked chromatin, cell debris, and unbroken cells. The purpose of this centrifugation step is to remove soluble protein, most of which is not cross-linked to DNA, as it might contribute to nonspecific background in the subsequent immunoprecipitations step. There is no need to resuspend the pellet at this point.

13. Holding the microtip probe near the bottom of the tube to prevent foaming, sonicate the sample 30 sec at 4°C using a continuous pulse at a power output of 20%. Cool in an ice-water bath >1 min. Repeat two more times.

Take great care that the sample does not get too hot.

If a different sonication device is used, empirically determine the conditions necessary to achieve the desired level of DNA shearing. The shear size is determined as described below (see UNIT 2.1A).

While it is convenient to perform the reaction in a PCR machine overnight, it could just as easily be done in heat blocks or water baths. The same is true of the incubation described in step 26.

16. Resuspend in 30 ul TE, pH 7.5, add 1 ul of 20 mg/ml DNase-free RNase A, and incubate 15 min at 37°C. Add 3 ul of 10´ loading buffer and electrophoretically separate the material on a 1.5% agarose gel (UNIT 2.5A).

Fragments should be between 100 to 1000 bp, with an average length of 400 to 500 bp.

It is important to shear DNA fragments down to an average length of 400 to 500 bp. Longer fragments will increase the background and will decrease the resolution of the region to which the protein associates (see Commentary).


17. Incubate 800 ul chromatin solution with 10 ul primary antibody against the protein or epitope of interest and 20 ul of 50% (v/v) Ultralink Protein A/G-Silica beads in TBS on an end-over-end rotator 90 min at room temperature.

For ~5 ´ 108 cells, use 800 ul chromatin solution to cross-link.

The actual amount of antibody needed has to be empirically determined and can vary considerably. The idea is to have an excess of antibody to efficiently precipitate at least 50% of the antigen in question. One way to assess the efficiency of antigen immunoprecipitation is to determine the amount of antigen present in the sample before and after the immunoprecipitation. An aliquot of 30 ul chromatin solution, taken before and after immunoprecipitation, is usually sufficient to visualize the protein of interest via immunoblotting and standard chemiluminescent detection (UNIT 10.8); however, the samples have to be boiled in SDS/PAGE sample buffer for 30 min prior to loading in order to reverse the formaldehyde cross-links. The immunoprecipitation conditions can be varied (e.g., time, temperature, salt concentration, presence of detergents) if necessary.

Protein A/G-Silica beads are used here because they bind to a wider range of antibodies from different organisms and of different isotypes. In most cases, however, Protein A-Sepharose or Protein A-agarose may be substituted without a decrease in antibody-binding ability.

18. Microcentrifuge beads 1 min at 3000 rpm, room temperature. Transfer 300 ul supernatant into a 0.5-ml PCR tube labeled "INPUT." Discard the rest of the liquid.

Wash beads

19. Resuspend beads in 700 ul FA lysis buffer, room temperature, and transfer mixture into a Spin-X centrifuge-tube filter.

The use of Spin-X filters aids in the recovery of the beads after washes and results in better uniformity between different samples. The procedure is also substantially faster with the filters, particularly when multiple samples are processed simultaneously. Alternatively, one could use conventional microcentrifuge tubes for the washes and aspirate the supernatant with a narrow-bore pipet tip after each spin.

20. Place the filter into a 1.5-ml microcentrifuge tube and mix sample 3 min on an end-over end rotator. Microcentrifuge 2 min at 3000 rpm, room temperature. Discard the flow-through liquid at the bottom of the tube.

21. Add 700 ul FA lysis buffer, room temperature, to the beads and repeat step 20.

Elute protein from beads

22. Wash beads for 3 min each with 700 ul FA lysis buffer/0.5 M NaCl, 700 ul ChIP wash buffer, and finally 700 ul TE.

For many polyclonal antibodies, the more stringent washes in this step result in a cleaner signal, while gentle washes frequently lead to an unacceptably high background. For some antibodies (e.g., monoclonal against peptide epitopes; see Alternate Protocol), repeated washes with FA lysis buffer, which are gentler, might be more appropriate.

23. Place filter unit containing the beads into a new 1.5-ml microcentrifuge tube and add 100 ul of ChIP elution buffer. Gently pipet up and down two or three times in order to dislodge beads from the filter. Incubate 10 min in a 65°C water bath.

A water bath is used instead of other heating apparatuses in order to improve heat transfer.

24. Microcentrifuge beads 2 min at 3000 rpm, room temperature. Discard filter with beads. Transfer the eluate into a 0.5-ml PCR tube labeled "IP."

Reverse cross-links and purify DNA

25. Add 80 ul TE and 20 ul Pronase in TBS to the IP tube. Combine 20 ul INPUT material (step 18), 100 ul ChIP elution buffer, 60 ul TE, and 20 ul TBS into a new 0.5-ml PCR tube.

26. To reverse cross-links, place tubes into a PCR machine. Incubate 2 hr at 42°C, followed by 6 hr at 65°C. Store samples at 4°C until use.

The incubation at 42°C allows for Pronase digestion of cross-linked polypeptides, while the 65°C incubation results in a reversal of the formaldehyde cross-links.

27. Purify DNA using a Qiagen PCR-purification spin column as per manufacturer's instructions.

This will require double loading of the spin column (i.e., 600 ul spin through and then repeat).

Alternatively, add 20 ul of 4 M LiCl and purify by extracting with 25:24:1 phenol/chloroform/isoamyl alcohol, followed by extraction with chloroform and ethanol precipitation (UNIT 2.1). It is useful to add 2 ul of Pellet Paint (Novagen) prior to the addition of ethanol, as this aids both the ethanol precipitation and visualization of the very small pellet.

28. Resuspend in 300 ul TE and store at -20°C.

DNA pellets stored in this fashion should be stable for years.

Perform quantitative PCR

29. Design primer pairs for the desired genomic regions to be examined.

Success in obtaining high-quality data is critically dependent on good primer design (see UNIT 15.1 for more information.

30. Dilute INPUT DNA (obtained from step 18) in three separate tubes by a factor of 5, 10, and 20. Set up standard PCR reactions (UNITS 15.1 & 15.7) with 2 ul DNA sample, primers at 1 pmol/ul, and total reaction volumes of 10 to 50 ul. If PCR products will be detected by radioactivity, add 1 mCi of 3000 Ci/mmol [32P]dATP.

For a typical measurement, the three dilutions of input DNA are tested along with duplicate immunoprecipitated samples (or undiluted and 5-fold diluted immunoprecipitated samples). This permits an assessment of whether the assay is being performed in the linear range and reproducibility of the PCR reaction. The immunoprecipitated DNA is typically used without dilution, although it is useful to analyze different amounts to ensure that it is also in the linear range.

There are several key parameters for achieving an optimum PCR reaction. For example, it is very important to have a quality repeat pipettor that can reproducibly dispense 2 ul DNA. Pipetting inaccuracies at this stage will lead to greater well-to-well variability and poorer reproducibility among identical samples. Additionally, multiple primer pairs (up to 4 to 5) can be included in the same reaction, provided that the PCR products can be unambiguously resolved from each other by gel electrophoresis. This permits simultaneous and internally controlled analysis of multiple genomic regions in a single reaction. However, it is critical to ensure that there is no competition between the different primer pairs and PCR products. Also, comparable results are obtained when PCR reactions are performed in volumes between 10 to 50 ul; using smaller volumes reduces the cost and facilitates loading of the reaction products on gels. See UNIT 2.7) or agarose gels (UNIT 2.5A).

The gels should be stained either with ethidium bromide or SYBR green dyes, or analyzed by autoradiography or PhosphorImager.

33. Quantitate the relative amount of PCR products using appropriate software for the accompanying instrument.

34. Calculate the apparent immunoprecipitation efficiency for a specific fragment by dividing the amount of PCR product obtained with the immunoprecipitated DNA by the amount obtained with the input DNA.

A volume of 2 ul immunoprecipitated DNA sample (1/150 total immunoprecipitated material) contains ~200 times the number of cell equivalents as 2 ul INPUT sample that has been diluted 5-fold (1/30,000 of the original aliquot that was immunoprecipitated). Thus, if the amount of PCR product in the immunoprecipitated sample is equal to the amount of PCR product in the 5-fold diluted INPUT sample, the apparent immunoprecipitation efficiency is 0.5%. The apparent immunoprecipitation efficiency for the background signal is typically ~0.025% to 0.05%, and it should not be higher than 0.1%.

From Current Protocols in Molecular Biology Online
Copyright © 2003 John Wiley & Sons, Inc. All rights reserved.