Chromatin Immunoprecipitation Approaches to Determine Co-transcriptional Nature of Splicing

Transcription

1 Chapter 23 Chromatin Immunoprecipitation Approaches to Determine Co-transcriptional Nature of Splicing Nicole I. Bieberstein, Korinna Straube, and Karla M. Neugebauer Abstract Chromatin immunoprecipitation (ChIP) is a common method used to determine the position along DNA where an antigen is found. The method was initially devised for protein antigens that come in direct contact with genomic DNA, such as components of the transcriptional machinery and histones. However, ChIP can also be extended to antigens that bind RNA, as demonstrated by the specific localization of spliceosomal components to particular gene regions that correlate with when and where introns and exons are transcribed. The activities of any RNA binding protein can in principle be monitored using ChIP, and RNA dependency of binding can also be assessed through RNase treatment. Combined with qpcr or high-throughput sequencing, this method allows the detection of RNA bound proteins at individual genes or genome-wide. Here, we present a detailed protocol for splicing factor ChIP in tissue culture cells. Key words Chromatin immunoprecipitation (ChIP), Splicing regulatory proteins, Spliceosomal small nuclear ribonucleoprotein particles (snrnps), Spliceosome assembly, Quantitative PCR (qpcr), ChIP-Seq 1 Introduction Splicing the removal of introns and ligation of exons by the spliceosome can take place co-transcriptionally, while the premrna is still attached to chromatin via RNA polymerase II [ 1 ]. Thereby, the nascent RNP lies close to the DNA axis, allowing for interactions between the splicing machinery and chromatin (reviewed in ref. 2 ). The co-transcriptional binding of splicing regulatory proteins and spliceosome assembly on nascent RNA can thus be monitored by chromatin immunoprecipitation (ChIP) [ 3 6 ]. The basic principle of this technique is in vivo crosslinking followed by the immunoprecipitation of an RNA binding protein of interest and finally the isolation of the corresponding DNA fragment (Fig. 1 ). First, unperturbed cells are usually crosslinked together by formaldehyde, because formaldehyde is cell-permeable and efficiently forms CH 2 linkages between amino acid side chains and nearby nitrogen atoms in nucleic acids. In the resulting complex, Klemens J. Hertel (ed.), Spliceosomal Pre-mRNA Splicing: Methods and Protocols, Methods in Molecular Biology, vol. 1126, DOI / _23, Springer Science+Business Media, LLC

2 316 Nicole I. Bieberstein et al. Fig. 1 Schematic of the splicing factor ChIP approach. ( a ) In vivo, splicing factors ( diamond ) can bind cotranscriptionally to nascent RNA. ( b ) Formaldehyde crosslinking covalently binds the RNA binding protein to RNA, which is attached to DNA via the polymerase ( open circle ). (c ) After cell lysis, the DNA is sheared by sonication. ( d ) Immunoprecipitation using an antibody directed against the splicing factor of interest will isolate the DNA RNA protein complex to which the splicing factor was bound. ( e ) After washing and uncrosslinking, the corresponding DNA fragment is purifi ed. ( f ) Analysis of the recovered DNA fragments by qpcr or next generation sequencing identifi es the genomic region where the splicing factor was bound. The 5 end cap on the nascent RNA is indicated by the small open circle the splicing factor is covalently bound to RNA, which is in turn linked to the polymerase and to DNA. Next, the cells are lysed and the chromatin is sheared to fragments of ~200 bp by sonication. The protein nucleic acid complex is immunoprecipitated and purified, using magnetic beads. After washing and elution, the complex is uncrosslinked by heating, and proteins are digested by Proteinase K. The DNA fragments are isolated by phenol chloroform extraction and residual RNA is removed by the addition of RNase A. Finally, the recovered DNA is analyzed by qpcr or highthroughput sequencing (ChIP-Seq). 2 Materials Prepare all solutions using deionized water. All reagents are stored at room temperature unless indicated otherwise % Formaldehyde. 2. PBS, store at 4 C. 3. Protease Inhibitor Cocktail 25 in PBS, store at 4 C.

3 Mammalian Splicing Factor ChIP SDS lysis buffer: 1 % (w/v) SDS, 10 mm EDTA, 50 mm Tris HCl ph 8.1, add 1 protease inhibitor before use. 5. Bradford reagent. 6. ChIP dilution buffer: 0.01 % (w/v) SDS, 1.1 % (v/v) Triton X-100, 1.2 mm EDTA, 16.7 mm Tris HCl ph 8.1, 167 mm NaCl, add 1 protease inhibitor before use. 7. Dynabeads coated with protein G or A. 8. Low Salt Immune Complex wash buffer: 0.1 % (w/v) SDS, 1 % (v/v) Triton X-100, 2 mm EDTA, 20 mm Tris HCl ph 8.1, 150 mm NaCl, store at 4 C. 9. High Salt Immune Complex wash buffer: 0.1 % (w/v) SDS, 1 % (v/v) Triton X-100, 2 mm EDTA, 20 mm Tris HCl ph 8.1, 500 nm NaCl, store at 4 C. 10. LiCl Immune Complex wash buffer: 0.25 M LiCl, 1 % (v/v) NP-40, 1 % (w/v) deoxycholic acid, 1 mm EDTA, 10 mm Tris HCl ph 8.1, store at 4 C TE: 10 mm Tris HCl ph 8.1, 1 mm EDTA, store at 4 C. 12. Elution buffer: 1 % (w/v) SDS, 0.1 M NaHCO 3, prepare freshly before use M NaCl M EDTA M Tris HCl ph mg/ml Proteinase K. 17. Phenol chloroform/isoamyl alcohol (25:24:1). 18. Chloroform/isoamyl alcohol (24:1) M NaOAc ph mg/ml Glycogen % Ethanol % Ethanol. 23. Deionized water. 24. RNase A. 3 Methods 3.1 Cell Culture and Crosslinking 1. Grow cells to confluency on four 14 cm dishes. This should give cell material for four immunoprecipitations with approximately ~10 8 cells per immunoprecipitation ( see Note 1 ). 2. Crosslinking: Add 540 μl formaldehyde (37 % solution) directly to 20 ml culture medium to a final concentration of 1 %, mix and incubate for 10 min at RT (work and incubate under the fume hood!) ( see Note 2 ).

4 318 Nicole I. Bieberstein et al. 3.2 Harvesting the Cells 3.3 Preparing the Lysate 3.4 Bradford Assay to Determine Protein Concentration (See Note 8 ) 3.5 Immunoprecipitation 3.6 Washing the Beads All following steps are performed on ice, if not stated otherwise. 1. Aspirate medium thoroughly. Wash cells twice using 5 ml cold PBS + protease inhibitor (1:100) ( see Note 3 ). 2. Add 5 ml cold PBS + protease inhibitor (1:100), scrape cells using a plastic cell scraper and transfer the cells to a 50 ml tube. Repeat the scraping with 5 ml cold PBS + protease inhibitor. Pool the cells from four plates (final volume 40 ml). 3. Pellet cells for 5 min at 1,500 g at 4 C. 4. Cell pellets can be frozen in liquid nitrogen and stored at 80 C or processed directly ( see Note 4 ). 1. Resuspend the cell pellet in 1 ml of SDS lysis buffer + 1 protease inhibitor, pipette up and down to homogenize the lysate, transfer the lysate to a 15 ml tube and incubate for 10 min on ice ( see Note 5 ). 2. Sonicate the lysate to shear the DNA to lengths between 200 and 500 bp. Keep the samples on ice/ethanol bath. Recommended sonication conditions: 30 % amplitude, s pulses, 20 s pauses ( see Note 6 ). 3. Centrifuge the lysate for 10 min at 20,000 g at 4 C and transfer the supernatant to a 1.5 ml tube. Keep the cleared lysate on ice ( see Note 7 ). 1. Prepare a standard curve with 10, 5, 2.5, and 1.25 μg/ml BSA in 800 μl ddh 2 O. Use 800 μl ddh 2 O without BSA as a blank. 2. Dilute the ChIP lysate 1:1,000, 1:5,000, and 1:10,000 in 800 μl ddh 2 O. 3. Add 200 μl Bradford reagent, mix and incubate 5 min at RT. 4. Measure absorbance at 595 nm and calculate the amount of total protein in the lysate. 1. Dilute 3 mg total protein ChIP lysate in 2 ml ChIP dilution buffer + 1 protease inhibitor. Prepare one 2.0 ml tube for each immunoprecipitation ( see Note 9 ). 2. Freeze one aliquot as input (1/4 of IP volume, i.e., 0.75 mg total protein) to be used in Subheading Add 5 μg of the immunoprecipitating antibody and precipitate overnight at 4 C with rotation. Include a mock IP as control ( see Notes 10 and 11 ). 4. Add 18 μl of Dynabeads for 1 h at 4 C with rotation to collect the antibody protein complex ( see Notes ). 1. Capture the beads with a magnetic rack. Carefully remove the supernatant that contains unbound, nonspecific DNA.

5 Mammalian Splicing Factor ChIP Resuspend the beads in 1 ml Low Salt Immune Complex wash buffer and transfer the beads to a new 1.5 ml tube ( see Note 16 ). 3. Wash the bead/antibody/protein complex for 4 min on a rotary shaker. 4. Capture the beads with a magnetic rack and remove the supernatant. 5. Repeat the washing with 1 ml of each of the buffers in order as listed below ( see Note 17 ): 1 High Salt Immune Complex wash buffer. 1 LiCl Immune Complex wash buffer. 1 TE. 3.7 Elution 3.8 Uncrosslinking and Proteinase K Treatment 3.9 Recover DNA by Phenol Chloroform Extraction All following steps are performed at room temperature, if not stated otherwise. 1. Freshly prepare elution buffer. 2. Capture the beads in a magnetic rack and remove the supernatant from the last washing step. 3. Elute the protein complex from the antibody by adding 250 μl elution buffer. Vortex briefly to mix and incubate at room temperature for 15 min with rotation. 4. Capture the beads, and carefully transfer the supernatant fraction (eluate) to a 1.5 ml tube and repeat elution with 250 μl fresh elution buffer. 5. Combine eluates (total volume 500 μl). 1. Take the frozen input (from Subheading 3.5 ) and add ChIP dilution buffer to a total volume of 500 μl. Uncrosslink the input together with the immunoprecipitation samples. This sample is considered to be the input/starting material for all the immunoprecipitations done with this extract and serves as a background control ( see Note 11 ). 2. To all samples add 20 μl of 5 M NaCl, 10 μl of 0.5 M EDTA, 20 μl of 1 M Tris HCl, ph 6.5 and 10 μl Proteinase K (10 mg/ml) incubate for 6 h at 65 C. 1. Add 560 μl (=1 volume) phenol chloroform/isoamyl alcohol (25:24:1) ph 8, vortex and incubate 2 3 min at RT. 2. Centrifuge for 15 min at 20,000 g and 4 C. 3. Transfer the upper aqueous phase to a new 1.5 ml tube. 4. Add 560 μl (=1 volume) chloroform/isoamyl alcohol, (24:1), vortex. 5. Centrifuge for 15 min at 20,000 g and 4 C.

6 320 Nicole I. Bieberstein et al. 6. Transfer the upper aqueous phase to a new 1.5 ml tube. 7. Precipitate the DNA by adding 1 ml 100 % EtOH, 50 μl 3 M NaAC ph 5.4, and 1 μl Glycogen (20 mg/ml). Vortex and incubate overnight at 80 C. 8. Centrifuge for 30 min at 20,000 g and 4 C. 9. Discard the supernatant and wash the pellet with 750 μl 70 % EtOH. 10. Centrifuge for 20 min at 20,000 g and 4 C. 11. Discard the supernatant, dry the pellet, and resuspend the DNA in ddh 2 O or TE + 50 μg/ml RNase A ( see Notes 18 and 19 ). 4 Notes 1. A large number of cells is required as starting material for ChIP of spliceosomal proteins and splicing regulators. We recommend using at least 10 8 cells for each immunoprecipitation and each control. In the case of HeLa cells, one 14 cm cell culture dish provides enough material for one IP. However, due to differences in cell density and also in chromatin content between different cell lines (HeLa cells are polyploid), the optimal number of cells has to be determined experimentally for each cell line ( see also Notes 5 9 for optimization). 2. Formaldehyde crosslinking will covalently link proteins and nucleic acids. Thus, it is thought that splicing factors bound to RNA are crosslinked to DNA and chromatin via RNA polymerase II. However, the exact size of these protein nucleic acid complexes is not known. The RNA binding protein of interest might be able to crosslink to a window of several nucleotides of DNA, therefore limiting the nucleotide resolution of this method. Keep that in mind, when interpreting your results. Moreover, the dependency of such a ChIP signal on bridging RNA molecules can be tested by treating the initial lysate with RNase A [ 6 ]. The crosslinking reaction can be quenched with glycine (125 mm final concentration). However, we obtained good results by directly removing the formaldehyde containing medium and washing the cells with cold PBS + protease inhibitor. 3. Ice-cold PBS is required for washing the cells after crosslinking to preserve the crosslinked protein nucleic acid complexes. We therefore recommend to store PBS for ChIP at 4 C. 4. Pellets of crosslinked cells can either be lysed directly or shock frozen in liquid nitrogen and stored at 80 C for future use. However, we recommend being consistent and treating the samples in exactly the same way for each experiment, i.e., always

7 Mammalian Splicing Factor ChIP 321 Fig. 2 DNA fragmentation by Sonication. Cells were harvested, lysed in SDS lysis buffer and sonicated for 5 10 s, s, or s pulse with 20 s pause at an amplitude of 30 %. After centrifugation, the supernatant was uncrosslinked and treated with proteinase K before the DNA was purifi ed by phenol chloroform extraction. Fragmentation was assessed on a 1 % agarose gel. With increasing number of pulses, the fragment size is reduced to bp in the last lane, which represents the desired fragment distribution around 200 bp proceed directly OR always freeze the pellets. In our experience, freezing the cell pellets is safer than freezing and reusing lysates. 5. Pipette slowly up and down to resuspend the pellet and lyse the cells. Avoid foam! You might have to optimize the ratio of cells to SDS lysis buffer ( see also Notes 1, 8, and 9 ) to obtain a homogenous and concentrated lysate. 6. The sonication conditions vary between cell types and sonicators and have to be optimized experimentally to yield fragments of approximately 200 bp. Sonication quality and fragment size can be assessed by agarose gel electrophoresis (Fig. 2 ). The size of DNA fragments determines the positional resolution of the ChIP assay. 7. Centrifugation should yield a small, white pellet of cell debris and a clear or milky supernatant containing the fragmented chromatin. A rather big and dense pellet indicates incomplete lysis and sonication. 8. Any assay to determine protein concentration could be used. However, due to the high SDS concentration in the lysate (SDS lysis buffer), the samples have to be diluted accordingly. We recommend Bradford or the Amido Black assay, which is insensitive to SDS, and a sample dilution of 1:1,000 to 1:10,000 in ddh 2 O. 9. The amount of total protein used as staring material for each IP depends on the abundance of the protein of interest.

8 322 Nicole I. Bieberstein et al. For splicing factor ChIP with HeLa cells, we recommend 3 4 mg of total protein per IP. For ChIP of RNA polymerase II or histones, 2 mg of total protein per IP are sufficient. The exact amount might vary between cell lines ( see Note 1 ). At maximum 200 μl of lysate can be diluted in a total volume of 2 ml ChIP dilution buffer, otherwise the high SDS concentration will affect the IP. You might have to optimize the lysis buffer to cells ratio, the lysis itself and sonicating conditions in order to obtain a concentrated lysate yielding ~3 mg total protein in 200 μl ( see Notes 1, 5 8 ). 10. The success of a ChIP experiment largely depends on the quality of the antibody. In general, polyclonal antibodies are preferred over monoclonal, as individual epitopes might not be accessible in the crosslinked state. Our lab previously showed that using a GFP-tag in combination with an anti- GFP antibody, can enhance the ChIP enrichment and signal to noise ratio for splicing factor ChIPs [ 6 ]. The GFP-tag should not be engaged in protein protein or protein nucleic acid interactions and thus protrude from the crosslinked complex, providing highest accessibility. Furthermore, the same anti-gfp antibody can be used for ChIPs of multiple tagged proteins of interest, thus increasing comparability. If tagging is not an option, we recommend using specific ChIP-grade antibodies. The amount of antibody per IP has to be determined experimentally; we recommend 5 μg per IP as starting point. 11. The required controls depend on the downstream analysis, i.e., qpcr or next generation sequencing. In any case, an input control is required. The input represents the total fragmented genomic DNA, which did not undergo an IP. For qpcr, the ChIP enrichment is calculated relative to Input. For ChIP-Seq, the input is required to assess the background distribution of reads due to sonication bias. Nucleosome occupancy and GC content influence the fragmentation by sonication. Thus, input samples will not yield a homogenous distribution of reads, but rather indicate hot spots of DNA shearing. It is therefore recommended to compare the ChIP- Seq signals of IP and input to test, if the enrichment is significant or simply reflects sonication bias. In addition, a mock IP using IgG is recommended for qpcr. This sample passes through all IP steps except for the precipitation of any chromatin complexes. The IgG control can therefore be used to determine the background of nonspecific material that was carried through the procedure by sticking to tubes, beads, pipettes, etc. Comparing the ChIP enrichment of the specific IP to the mock IP will determine, if the enrichment is significant. 12. We recommend using magnetic beads, however, we also have good experiences with sepharose beads (e.g., GammaBind

9 Mammalian Splicing Factor ChIP 323 beads). The advantage of magnetic beads is that the washing is more stringent. The supernatant can be removed more completely when the magnetic beads are held back by a magnetic rack, compared to sepharose beads that were pelleted by centrifugation. In our hands, using magnetic beads greatly improved the signal to noise ratio by reducing background. 13. Whether to use protein A or protein G coupled beads depends on the Ig origin of the antibodies ( see supplier information for more details). 14. Beads have to be washed twice in ChIP dilution buffer before use. 15. An alternative approach is to pre-couple the antibody to beads before adding the ChIP lysate. Aliquot 18 μl beads into a 2 ml tube, add 500 μl ChIP dilution buffer, and 5 μg antibody. Incubate for 2 h at 4 C with rotation. Wash the beads twice in ChIP dilution buffer to remove excess antibody. Add the diluted ChIP lysate to the pre-couples beads, incubate overnight at 4 C with rotation and proceed with Subheading Chromatin can also stick to plastic tubes and thus be carried through the whole procedure increasing the nonspecific background. One option is using non-sticky tubes, or alternatively, transferring the beads to a new tube during the washing step. 17. We recommend using washing buffers with increasing stringency. 18. How the DNA is finally resuspended depends on the downstream analysis. 50 μg/ml RNase A should be added to remove residual RNA. 19. The relative amount of DNA recovered can be determined by qpcr. The ChIP signal is calculated as enrichment over input using the following equation: ΔCt = 2 ( C tinput C tip), where Ct Input is the threshold cycle of the input sample and Ct IP that of the specific IP. The ChIP enrichment is further normalized to a control primer pair such as an intergenic gene desert region as ΔΔCt = ΔCt experiment /ΔCt control. References 1. Brugiolo M, Herzel L, Neugebauer KM (2013) Counting on Co-transcriptional Splicing. F1000Prime Reports 5: Carrillo Oesterreich F, Bieberstein N, Neugebauer KM (2011) Pause locally, splice globally. Trends Cell Biol 21(6): Görnemann J, Kotovic KM, Hujer K et al (2005) Cotranscriptional spliceosome assembly occurs in a stepwise fashion and requires the cap binding complex. Mol Cell 19:53 4. Listerman I, Sapra A, Neugebauer K (2006) Cotranscriptional coupling of splicing factor recruitment and precursor messenger RNA splicing in mammalian cells. Nat Struct Mol Biol 13: Lacadie S, Rosbash M (2005) Cotranscriptional spliceosome assembly dynamics and the role of U1 snrna:5 ss base pairing in yeast. Mol Cell 19:65 6. Sapra A, Ankö ML, Grishina L et al (2009) SR protein family members display diverse activities in the formation of nascent and mature mrnps in vivo. Mol Cell 34:179