Chronic lymphocytic leukemia (CLL) is the most common leukemia in the western world. NFAT transcription factors are important regulators of development and activation in numerous cell types. Here, we present a protocol for the use of chromatin immunoprecipitation (ChIP) in human CLL cells to identify novel target genes of NFAT2.
Chronic lymphocytic leukemia (CLL) is characterized by the expansion of malignant B cell clones and represents the most common leukemia in western countries. The majority of CLL patients show an indolent course of the disease as well as an anergic phenotype of their leukemia cells, referring to a B cell receptor unresponsive to external stimulation. We have recently shown that the transcription factor NFAT2 is a crucial regulator of anergy in CLL. A major challenge in the analysis of the role of a transcription factor in different diseases is the identification of its specific target genes. This is of great significance for the elucidation of pathogenetic mechanisms and potential therapeutic interventions. Chromatin immunoprecipitation (ChIP) is a classic technique to demonstrate protein-DNA interactions and can, therefore, be used to identify direct target genes of transcription factors in mammalian cells. Here, ChIP was used to identify LCK as a direct target gene of NFAT2 in human CLL cells. DNA and associated proteins are crosslinked using formaldehyde and subsequently sheared by sonication into DNA fragments of approximately 200-500 base pairs (bp). Cross-linked DNA fragments associated with NFAT2 are then selectively immunoprecipitated from cell debris using an αNFAT2 antibody. After purification, associated DNA fragments are detected via quantitative real-time PCR (qRT-PCR). DNA sequences with evident enrichment represent regions of the genome which are targeted by NFAT2 in vivo. Appropriate shearing of the DNA and the selection of the required antibody are particularly crucial for the successful application of this method. This protocol is ideal for the demonstration of direct interactions of NFAT2 with target genes. Its major limitation is the difficulty to employ ChIP in large-scale assays analyzing the target genes of multiple transcription factors in intact organisms.
Chronic lymphocytic leukemia (CLL) represents the most common leukemia in adults in western countries, exhibiting distinct accumulation of CD19, CD23, and CD5 expressing mature B cells1. Most patients exhibit an indolent disease course, which does not necessitate specific treatment for many years. In contrast, some patients show rapid progression requiring immediate therapeutic interventions with immune-chemotherapy or other targeted therapies2,3. Nuclear factor of activated T cells (NFAT) is a family of transcription factors controlling various developmental and activation processes in numerous cell types4,5,6. We recently demonstrated overexpression and constitutional activation of NFAT2 in CLL cells from patients with indolent disease7. Here, it regulates an unresponsive state to B cell receptor stimulation called anergy7. To demonstrate that NFAT2 binds to the lymphocyte-specific protein tyrosine kinase (LCK) promoter and regulates LCK expression in human CLL cells, a specific chromatin immunoprecipitation assay (ChIP) was developed and employed.
ChIP is one of the several techniques to investigate the role of transcription factors in gene expression8. Gene expression is tightly orchestrated in a very complex manner by several regulators with transcription factors taking an irreplaceable part in this process9,10,11,12. Transcription factors regulating the gene expression in a spatial and temporal context have been identified in numerous species (e.g., for development and differentiation)13,14,15,16,17,18. Errors in the intricate control mechanisms involving transcription factors can lead to a variety of pathologic processes including cancer19,20. Hence, identification of transcription factors and their respective targets might offer novel therapeutic avenues21,22. To investigate this intriguing field several methods are available like ChIP, electrophoretic mobility shift assay (EMSA), various DNA pull-down assays and reporter-assays8,11,12,23,24.
To demonstrate that a certain transcription factor interacts with specific regions of the genome in vivo, ChIP is an ideal technique25. For this purpose, DNA and associated proteins in living cells are cross-linked using UV irradiation or formaldehyde (cross-linked ChIP, XChIP). This step is omitted to obtain better DNA and protein recovery in the so-called native ChIP (NChIP)26. The DNA-protein complexes are subsequently sheared by sonication into fragments of approximately 200-500 base pairs (bp) and immunoprecipitated from the cell debris using a specific antibody against the transcription factor of interest. The associated DNA fragments are then purified and characterized by PCR, molecular cloning, and sequencing. Alternative techniques use microarrays (ChIP-on-Chip) or the next-generation sequencing (ChIP-Seq) to analyze the immunoprecipitated DNA.
ChIP was first introduced by Gilmour and Lis in 1984 when they used UV light to covalently cross-link DNA and bound proteins in living bacteria27. Upon cell lysis and immunoprecipitation of bacterial RNA polymerase, specific probes of known genes were used to map the in vivo distribution and density of RNA polymerase. The method was subsequently used by the same investigators to analyze the distribution of eukaryotic RNA polymerase II on heat shock protein genes in Drosophila28. The XChIP assay was further refined by Varshavsky and coworkers who first used formaldehyde cross-linking to study the association of histone H4 with heat shock protein genes29,30. The NChIP approach, which carries the advantage of a better DNA and protein recovery due to naturally intact epitopes and, therefore, greater antibody specificity, was first described by Hebbes and colleagues in 198831.
The advantage of ChIP in comparison to other techniques to analyze DNA-protein interactions is in fact, that the actual interaction of a transcription factor can be investigated in vivo and no probes or artificial conditions created by buffers or gels are employed8,11,12. By combining ChIP with next-generation sequencing, multiple targets can be identified simultaneously.
Major limitations of this technique are its limited applicability to large-scale assays in intact organisms25. The analysis of differential gene expression patterns can also be challenging using ChIP techniques if the respective proteins are expressed only at low levels or during narrow time windows. Another potentially limiting factor is the availability of an appropriate antibody suited for ChIP11.
The ChIP protocol presented here can be employed for the in vivo identification of target genes of a transcription factor by quantitative real-time PCR (qRT-PCR). Specifically, the goal was to identify novel target genes of NFAT2 in CLL. ChIP was chosen because of its potential to directly demonstrate the binding of NFAT2 to the promoter regions of different target genes under natural conditions in human CLL patient cells.
All experiments conducted with human material were approved by the Ethics Committee of the University of Tübingen and written informed consent was obtained from all patients who contributed samples to this study.
1. Isolation and Stimulation of Jurkat cells
NOTE: To optimize the protocol, use the Jurkat cell line which is known to express the high levels of NFAT2. All steps are performed under a laminar flow hood.
2. Isolation and Stimulation of Primary CLL Cells from Human Patients
NOTE: Patient samples were acquired and stimulated as previously described7. All steps are performed under a laminar flow hood.
3. Fixation, Cell Lysis, and Chromatin Shearing
NOTE: Patient samples and Jurkat cells are fixed and lysed with a commercially available ChIP kit according to the manufacturer's instructions with modifications as described previously7. The fixation is performed under a laminar flow hood.
4. Chromatin Immunoprecipitation
NOTE: The chromatin immunoprecipitation was performed with a commercially available ChIP kit according to the manufacturer's instructions with modifications7.
5. Detection of NFAT target genes in CLL cells.
6. Normalization and Data Analysis
NOTE: Relative enrichment for the promoter regions of interest (IL-2, CD40L or LCK) is calculated using the IgG-control for normalization.
Figure 1 shows an exemplary flow cytometry analysis of a CLL patient performed after staining with CD19-FITC and CD5-PE antibodies. Figure 1a shows the gating of the lymphocytes, representing the majority of cells in the blood of CLL patients. Figure 1b shows the proportion of CD19+/CD5+ CLL cells, which represent 89.03% of lymphocytes in this example. The proportion of CD19+/CD5– B cells is 3.72% in the respective patient.
Figure 2 shows examples of antibody performance in Jurkat cells fixed for different time periods as assessed by SDS-PAGE and Western Blot. Different antibodies from different manufacturers were analyzed. Figure 2a shows the performance of the αNFAT2-antibody (clone 7A6) from different manufacturers detected with a green fluorescent anti-mouse antibody. The antibody of manufacturer 1 showed binding to its epitope without fixation (band A) and even when Jurkat cells were fixed for 2.5 min or 5 min (band B or C, respectively). The αNFAT2-antibody (clone 7A6) from manufacturer 2, on the other hand, showed a weaker performance even in Jurkat cells not fixed (band D). Upon fixation, the antibody of this manufacturer achieved even weaker results (bands E (2.5 min fixation) and F (5 min fixation)). Figure 2b shows the performance of the αNFAT2-antibody (clone D15F1) from a different manufacturer detected with a red fluorescent anti-rabbit antibody. This antibody also performed weakly in unfixed samples (band A) and the absence of a band indicates the masking of its epitope by the fixation (bands B (2.5 min fixation) and C (5 min fixation)).
Figure 3 shows examples of sheared chromatin from Jurkat cells obtained after different fixation and shearing times of good and poor quality as assessed by gel electrophoresis. Bands A and B (0 min fixation or 2.5 min fixation, respectively) show good quality sheared chromatin, which is characterized by a DNA fragment size of 200-500 bp which can be detected on an agarose gel as a typical smear. Sheared chromatin of poor quality, on the other hand, can be recognized by almost complete or complete absence of the expected DNA smear (band C) as well as a smear in a larger or smaller than expected fragment size range (bands D-F). The complete absence of the smear indicates the use of insufficient amounts of starting material. The detection of smaller DNA fragments hints to excessive DNA shearing (band E) while larger DNA fragments hint to insufficient shearing.
Figure 4 shows the results of a typical experiment with Jurkat cells. LCK was analyzed as a potential NFAT2 target. IL-2 which is a well-defined target gene of NFAT2 in T cells was used as a positive control. An IgG-control antibody was utilized for normalization. Figure 4a shows an experiment with optimal results documented by significant enrichment of the IL-2 positive control. From this experiment, it can be concluded that there is a significant enrichment of LCK sequences in the precipitated DNA demonstrating that LCK is a direct NFAT2 target in Jurkat cells. Figure 4b shows an experiment performed with poor quality DNA due to missing fixation or a suboptimal shearing procedure causing a low degree of enrichment in the IL-2 positive control.
Figure 5 shows the results of an experiment performed with primary human CLL cells. CD40L which is a known NFAT2 target in B cells served as the positive control and LCK was analyzed as the experimental target. Figure 5a shows an experiment with a considerable level of enrichment of CD40L DNA in the positive control. From the even stronger enrichment of LCK DNA, it could be concluded that LCK is also a direct NFAT2 target in primary human CLL cells. Figure 5b shows an experiment performed with poor quality DNA most likely due to inadequate shearing. No substantial enrichment of CD40L DNA could be detected in the positive control rendering the experimental data meaningless.
Figure 1: Exemplary Flow cytometry analysis of CLL patients. (a) Samples of CLL patients were analyzed via flow cytometry after staining with CD19-FITC and CD5-PE antibodies and lymphocytes were gated. (b) Subsequently, the proportion of CD19+/CD5+ CLL cells and CD19+/CD5– B cells were determined. Here, CD19+/CD5+ CLL cells account for 89.03% of lymphocytes, whereas CD19+/CD5– B cells represent 3.72% of lymphocytes. One exemplary patient is shown. Please click here to view a larger version of this figure.
Figure 2: Examples of antibody performance in fixed Jurkat cells assessed by SDS-PAGE and Western-Blot. (a) Jurkat cells were fixed for 0 min (bands A and D), 2.5 min (bands B and E), or 5 min (C and F) and the αNFAT2-antibody (clone 7A6) from different manufacturers (bands A-C = manufacturer 1, bands D-F = manufacturer 2) was compared. The antibody of manufacturer 1 showed a better overall performance, binding with the higher affinity even to fixed samples (compare bands A-C and bands D-F). (b) The αNFAT2-antibody (clone D15F1) from a different manufacturer was used. This antibody showed only a poor performance in unfixed samples (band A) and the epitope was masked upon fixation. Therefore, no binding of the antibody could be detected after fixation (bands B and C). Please click here to view a larger version of this figure.
Figure 3: Examples of chromatin of good and poor shearing quality from Jurkat cells assessed by gel electrophoresis. The Jurkat cells were fixed and sheared for different time periods. Fixation was done for 0 min (bands A and D), 2.5 min (bands B and E), or 5 min (C and F). Shearing was performed either for 10 min (bands A-C) or 20 min (bands D-F). Chromatin of good shearing quality is characterized by a DNA fragment size of 200-500 bp which can be detected as a smear in the respective region (bands A and B). Chromatin of poor quality can be recognized either by an almost complete or complete absence of the DNA smear due to an insufficient amount of starting material used (band C) or by a smear in a smaller or larger size region because of inappropriate shearing conditions (bands D-F). Please click here to view a larger version of this figure.
Figure 4: ChIP of Jurkat cells assessed by qRT-PCR. (a) NFAT2 binds to LCK and the IL-2 promoter in Jurkat cells. The relative enrichment of LCK and IL-2 promoter regions precipitated with the NFAT2 antibody is shown as analyzed by qRT-PCR. Three independent ChIP-experiments are shown as mean ± SEM (Student's t-test, *P < 0.05; **P < 0.01 ***P < 0.001). (b) DNA from samples with poor quality sheared chromatin was used. No relevant binding of NFAT2 to the target sequences could be detected. A similar picture can be detected if there was no fixation or the incubation time with the NFAT2 antibody was insufficient. Three independent ChIP-experiments are shown as mean ± SEM (Student's t-test, *P < 0.05; **P < 0.01 ***P < 0.001). Please click here to view a larger version of this figure.
Figure 5: ChIP of primary human CLL cells assessed by qRT-PCR. (a) NFAT2 specifically binds to the LCK and CD40L promoters in primary human CLL cells from patients with an indolent course of the disease. The diagram shows the relative enrichment of LCK and CD40L promoter regions precipitated with the NFAT2 antibody as analyzed by qRT-PCR. Three independent ChIP-experiments are shown as mean ± SEM (Student's t-test, *P < 0.05; **P < 0.01 ***P < 0.001). (b) DNA from patient samples with poor quality sheared chromatin was used. No binding of NFAT2 to the respective target sequences could be observed. Three independent ChIP-experiments are shown as mean ± SEM (Student's t-test, *P < 0.05; **P < 0.01 ***P < 0.001) Please click here to view a larger version of this figure.
item | volume (µL) per IP |
5% BSA | 6 |
100 x protease inhibitor | 3 |
5 x ChIP buffer 1 | 56 |
sheared chromatin | x (15 µL or 70 µL) |
Protein A coated magnetic beads | 20 |
ChIP seq grade water | 185-x-y |
antibody (αNFAT2 or IgG-control) | y (1.09 µL or 1 µL) |
total | 270 |
Table 1: Schedule for pipetting the chromatin immunoprecipitation. The chromatin immunoprecipitation was performed by preparing the reactions as indicated in the table.
item | volume (µL) per qRT-PCR |
immunoprecipitated DNA | 9 |
qRT-PCR Mix | 10 |
primer forward (LCK/CD40L/IL-2) | 0.5 |
primer reverse (LCK/CD40L/IL-2) | 0.5 |
total | 20 |
Table 2: Schedule for pipetting the qRT-PCR. The qRT-PCR was performed by preparing the reactions as indicated in the table.
The critical steps of performing a successful ChIP assay are the selection of an appropriate antibody and the optimization of the chromatin shearing process25. The selection of the αNFAT2 antibody proved to be particularly challenging during the development of this protocol. While there are several αNFAT2 antibodies commercially available and the majority of these works fine for western blotting and other applications, clone 7A6 was the only antibody which could be successfully used for ChIP7. Even supposedly identical 7A6 antibodies from different manufacturers exhibited significant differences with respect to their performance in ChIP. A major challenge is the potential disruption of target protein epitopes by the fixation process used in XChIP protocols25.
The chromatin shearing procedure is also critical for acquiring appropriate results in XChIP. A fragment size of 200-500 bp as assessed by agarose gel electrophoresis was found to be optimal in this NFAT2 ChIP protocol7. Inappropriate shearing conditions resulting in a shortage of DNA starting material or DNA fragments of smaller or larger size typically lead to suboptimal results when performing this assay. Suboptimal shearing was also observed when using frozen and rethawed PBMCs. Hence, thawing of PBMCs resulted in a lower yield of DNA from cells as well as an increase in excessive shearing of DNA-fragments.
The availability of a broad variety of ChIP kits and ChIP-grade antibodies is challenging, but also offers the possibility to use the technique in a wide context. Thus, a diversity of transcription factors can be investigated in different cell types. For example, other members of the NFAT family (NFAT1 and NFAT4) have been investigated recently using ChIP35,36,37.
The ChIP kit used here also had to be modified to suit our needs. First, the time of fixation was adjusted to avoid masking of the epitope recognized by the used antibody and to enable optimal shearing. The volume of buffers used for lysis and shearing was also changed to reduce the loss of cells during the different washing steps. Additionally, the shearing conditions were optimized to obtain DNA fragments in the range of 200-500 bp. The protease inhibitor and the IgG-control antibody were exchanged with other commercially available reagents as they proved to be comparable and more cost-effective. The step involving the carrier provided by the manufacturer was omitted as it interfered with the precipitation of the DNA. In the end, the amount of buffer used to elute the DNA was adapted to yield a sufficient volume to be used in the qRT-PCR.
There are several other kits available from various companies and there is also the possibility to perform the ChIP without a kit. However, testing and establishment are very challenging, as there are many factors to be considered, like the compatibility of analyzed cells and proteins with different fixation and shearing methods. The mentioned kit was used as it was well-established in our laboratory.
A major restriction of this method is its limited applicability to the large-scale investigations in intact model organisms because appropriate antibodies have to be identified or generated for each individual transcription factor. The analysis of genes which are expressed only at low levels, in a small number of cells or during a restricted time window can also be challenging using ChIP.
While ChIP is the gold standard to demonstrate a direct interaction of a given transcription factor with its respective target genes in intact eukaryotic cells25, there is a number of other techniques to investigate an interaction between proteins and DNA. EMSA is a useful method to detect low-abundance DNA binding proteins in cell lystates24. It can also be used to characterize the binding affinity of a particular protein through a systematic DNA probe mutational analysis. A major advantage of EMSA in comparison to ChIP is the fact that it is generally substantially less time-consuming to establish the respective assay. DNA pull-down assays, microplate capture and detection assays and reporter assays are other techniques to analyze protein-DNA interactions23.
More recent developments have made it possible to apply the ChIP assay to genome-wide approaches by its combination with microarray technology (ChIP-on-chip)38,39,40 or next-generation sequencing (ChIP-Seq)41,42,43.
The authors have nothing to disclose.
This work was supported by the DFG grant MU 3340/1-1 and the Deutsche Krebshilfe grant 111134 (both awarded to M.R.M.). We thank Elke Malenke for excellent technical assistance).
1 X PBS | Sigma Aldrich | D8537 | |
1.5 mL tube shaker Themomixer comfort | Eppendorf | 5355 000.011 | Can be substituted with similar instruments |
10X Bolt Sample Reducing Agent | Thermo Scientific | B0009 | |
20X Bolt MES SDS Running Buffer | Thermo Scientific | B0002 | |
37 % Formaldehyde p.a., ACS | Roth | 4979.1 | |
4X Bolt LDS Sample Buffer | Thermo Scientific | B0007 | |
Anti-NFAT2 antibody | Alexis | 1008505 | Clone 7A6 |
Anti-NFAT2 antibody | Cell Signaling | 8032S | Clone D15F1 |
Anti-NFAT2 antibody ChIP Grade | Abcam | ab2796 | Clone 7A6 |
big Centrifuge | Eppendorf | 5804R | Can be substituted with similar instruments |
CD19-FITC mouse Anti-human | BD Biosciences | 555412 | Clone HIB19 |
CD5-PE mouse Anti-human CD5 | BD Biosciences | 555353 | Clone UCHT2 |
Density gradient medium Biocoll (Density 1,077 g/ml) | Merck | L 6115 | |
DNA LoBind Tube 1.5 mL | eppendorf | 22431021 | |
FBS superior | Merck | S0615 | |
Flow Cytometer | BD Biosciences | FACSCalibur | Can be substituted with similar instruments |
Halt Protease and Phosphatase Inhibitor Cocktail (100X) | Thermo Scientific | 78440 | |
iBlot 2 Gel Transfer Device | Thermo Scientific | IB21001 | |
iBlot 2 Transfer Stacks, nitrocellulose, regular size | Thermo Scientific | IB23001 | |
iDeal ChIp-seq kit for Histones | Diagenode | C01010059 | |
Ionomycin calcium salt | Sigma Aldrich | I3909 | |
IRDye 680LT Donkey anti-Rabbit IgG (H + L), 0.5 mg | LI-COR Biosciences | 926-68023 | |
IRDye 800CW Goat anti-Mouse IgG (H + L), 0.1 mg | LI-COR Biosciences | 925-32210 | |
LI-COR Odyssey Infrared Imaging System | LI-COR Biosciences | B446 | |
LightCycler 480 Multiwell Plate 96, white | Roche | 4729692001 | Can be substituted with other plates in different real-time PCR instruments |
Lysing Solution OptiLyse B | Beckman Coulter | IM1400 | |
M220 AFA-grade water | Covaris | 520101 | |
M220 Focused-ultrasonicator | Covaris | 500295 | |
Magnetic rack, DynaMag-15 Magnet | Thermo Scientific | 12301D | Can be substituted with similar instruments |
MEM Non-Essential Amino Acids Solution 100X | Thermo Scientific | 11140050 | |
Microscope Axiovert 25 | Zeiss | 451200 | Can be substituted with similar instruments |
microTUBE AFA Fiber Pre-Slit Snap-Cap 6x16mm | Covaris | 520045 | |
Neubauer improved counting chamber | Karl Hecht GmbH & Co KG | 40442012 | Can be substituted with similar instruments |
NH4 Heparin Monovette | Sarstedt | 02.1064 | |
Nuclease-free water | Promega | P1193 | |
NuPAGE 4-12% Bis-Tris Protein Gels, 1.0 mm, 15-well | Thermo Scientific | NP0323BOX | |
Odyssey® Blocking Buffer (TBS) 500 mL | LI-COR Biosciences | 927-50000 | |
Penicillin/Streptomycin 100X | Merck | A2213 | |
PerfeCTa SYBR Green FastMix | Quanta Bio | 95072-012 | |
PMA | Sigma Aldrich | P1585 | |
Primer CD40L promotor region forward | Sigma Aldrich | 5’-ACTCGGTGTTAGCCAGG-3’ | |
Primer CD40L promotor region reverse | Sigma Aldrich | 5’-GGGCTCTTGGGTGCTATTGT -3’ | |
Primer IL-2 promotor region forward | Sigma Aldrich | 5’-TCCAAAGAGTCATCAGAAGAG-3’ | |
Primer IL-2 promotor region reverse | Sigma Aldrich | 5’-GGCAGGAGTTGAGGTTACTGT-3’ | |
Primer LCK promotor region forward | Sigma Aldrich | 5’-CAGGCAAAACAGGCACACAT-3’ | |
Primer LCK promotor region reverse | Sigma Aldrich | 5’-CCTCCAGTGACTCTGTTGGC-3’ | |
Rabbit mAb IgG XP Isotype Control | Cell Signaling | # 3900S | Clone DA1E |
Real-time PCR instrument | Roche | LightCycler 480 | Can be substituted with similar instruments |
Roller mixers | Phoenix Instrument | RS-TR 5 | |
RPMI 1640 Medium, GlutaMAX Supplement | Thermo Scientific | 61870010 | |
Safety-Multifly-needle 21G | Sarstedt | 851638235 | |
SeeBlue Plus2 Pre-stained Protein Standard | Thermo Scientific | LC5925 | |
Shaker Duomax 1030 | Heidolph Instruments | 543-32205-00 | Can be substituted with similar instruments |
small Centrifuge | Thermo Scientific | Heraeus Fresco 17 | Can be substituted with similar instruments |
Sodium Pyruvate | Thermo Scientific | 11360070 | |
ß-Mercaptoethanol | Thermo Scientific | 21985023 | |
Tris Buffered Saline (TBS-10X) | Cell Signaling | #12498 | |
Trypan Blue solution | Sigma Aldrich | 93595-50ML |