MATERIALS AND METHODS

Cell lines
K-562, Jurkat and U-937 were obtained from the ATCC (American Type Culture Collection, Rockville, USA) and EM-2, LAMA-84, CML-T1, BV-173, SD-1 and RPMI-8226 from the DSMZ (Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, Braunschweig, Germany). All cell lines, except BV-173, SD-1 and RPMI-8226, were IRF-4-negative.

Cell culture and stimulation
All cell lines were maintained at 5% CO2 in RPMI 1640 medium with 1% glutamine (Gibco/BRL Eggenstein, Germany) supplemented with 10% fetal calf serum (Gibco/BRL), 1% penicillin/streptomycin (Biochrom, Berlin, Germany). When indicated, cells were treated with 5-aza-2-deoxycytidine (AzadC) or 5-azacytidine (AzaC) (Sigma, Taufkirchen, Germany) for different time periods. Owing to their chemical instability fresh substances were re-added every 24 h.

Sequencing of the IRF-4 promoter
For analysis of the IRF-4 promoter region for permanent aberrations such as insertions/deletions or mutation, we PCR-amplified two fragments from genomic DNA, which was extracted from depicted cell lines with a commercial kit (Qiagen, Hilde, Germany) as recommended. The primers were 1-forward: 5′-TTGAGATGGAGTCTTGCTCTGT-3′, 1-reverse: 5′-CCAGGACCTCAGGAGGCCAGTCA-3′; 2-forward: 5′-AGCGGTGAAACTGAGAGTGCGAGGT-3′, 2-reverse: 5′-GCCACATCGCTGCAGTTTAG-3′. The products were cloned with the ‘TOPO TA cloning kit’ (Invitrogen, Groningen, The Netherlands). After bacterial amplification of the cloned PCR fragments by standard procedures, at least three clones from each sample were sequenced with an automated sequencer (ABI Prism 377, Applied Bio-systems, Foster City, USA) as recommended by the manufacturer.

Expression analysis
To analyze the IRF-4 transcriptional level, RNA was extracted from cells using the commercial RNAzol-kit (Paesel, Frankfurt, Germany). An aliquot of 1 μg total RNA was used for cDNA synthesis as described previously (27). RNA expression analysis for IRF-4 and the reference gene β-actin was carried out by semi-quantitative PCR as described previously (3,27). PCR products were verified by automated sequencing. PCR primers and conditions for expression analysis of DNMT or MBP (DNMT1 DNMT3A, DNMT3B, MeCP, MBD1, MBD2 and MBD4) were published elsewhere (28).
For analysis of IRF-4 protein expression, a standard immunoblotting assay was performed as described previously (29). Briefly, protein lysates were generated by incubating 1 × 106 cells in 100 µl RIPA buffer (1% NP-40, 0.5% sodiumdesoxycholate, 0.1% SDS, 100 µg/ml phenylmethylsulfonyl fluoride, 10 µl/ml protease-inhibitory-mix, 1 µmol/ml sodiumorthovanadate in phosphate-buffered saline) for 30 min on ice. After centrifugation, protein concentration of the supernatant was determined by BCA-method (Pierce, Rockford, IL) as recommended. Protein lysates (70–100 µg) were electrophoresed on polyacrylamide gels and transferred to a PVDF-membrane (Immobilon P, 0.45 µm; Millipore, Eschborn, Germany). Membranes were blocked with 2.5% blocking reagent (Boehringer Mannheim, Germany) in TBST buffer (4.44 g/l Tris–HCL, 2.65 g/l TrisOH, 8.07 g/l NaCl, 0.2 g/l KCl and 500 µl/l Tween-20 in H2O) and subsequently incubated with primary antibody as indicated and horseradish peroxidase-conjugated secondary antibody, anti-mouse or anti-goat IgG (DAKO, Hamburg, Germany), respectively. The membranes were then developed with an ECL detection kit (Amersham Pharmacia Biotech, Freiburg, Germany). The primary antibodies were goat anti-IRF-4/ICSAT (M-17) (Santa Cruz Biotechnology, Santa Cruz, CA) and mouse anti-β-actin (AC-74) (Sigma).

Methylation-specific restriction-PCR-assay
DNA was extracted with a commercial kit (Qiagen) as recommended. Since the restriction ability of several endonucleases is inhibited by methylation of their target sequence, we used methylation-sensitive enzymes HpaII and HaeII-isochizomer Bsp143II and Bsh1236I (MBI Fermentas, St Leon-Rot, Germany) (20,24). As control the methylation-resistant enzyme MspI and an enzyme with no recognition site in the target promoter, EcoRI, were used. DNA (0.8 µg) was digested by 40 U the respective enzyme for 6 h and, to ensure complete cleavage, additional 20 U for 16 h. Thereafter 100 ng of digested DNA was used to a PCR amplification of two fragments (F1 and F2) spanning part of the IRF-4 promoter (30) (GenBank U52683; see Figure 3A). The sequences of the primers were F1-forward: 5′-TTGAGATGGAGTCTTGCTCTGT-3′, F1-reverse: ATCACTTCCAGACTTCAGTTCACCT-3′ (341 bp); F2-forward: 5′-AAGGTGAACTGAAGTCTGGAAGTGA-3′, F2-reverse: 5′-CCAGGACCTCAGGAGGCCAGTCA-3′ (474 bp). The PCR conditions were described elsewhere (3). PCR was performed with an annealing temperature of 62°C and 35 cycles. When DNA was methylated at specific sites, the sensitive enzymes were not able to digest the DNA and amplification took place; in case of no methylation, DNA was digested and no product was generated. The PCR products were electrophoresed on a 3% agarose gel, were stained with ethidium bromide and photographed. PCR products were verified by automated sequencing.

Bisulfite treatment
DNA was extracted as described above. Bisulfite treatment of DNA, leading to conversion of unmethylated cytosine to uracil residues and no change of methylated cytosine residues, was performed as described as follows. Briefly, 1 µg of DNA and 2 µg of poly(dA–dT)(poly(dA–dT) copolymers (Amersham Pharmacia Biotech) were denaturated for 20 min at 42°C in 0.3 M NaOH in a volume of 50 µl. Fresh solutions of 30 µl of 10 mM hydrochinon (Sigma) and 530 µl of 3 M sodium bisulfite (pH 5.0; Sigma) were added, the solution was gently mixed, overlayed with mineral oil and incubated in the dark for 12–13 h at 50°C. The aqueous phase was recovered using the ‘Wizard DNA clean-up system’ (Promega, Mannheim, Germany). The purified DNA was subsequently mixed with 1 M NaOH to a final concentration of 0.3 M and incubated for 20 min at 37°C to ensure complete desulfonisation. DNA was ethanol precipitated in the presence of 1/10 vol of 3 M sodium acetate, washed with 70% ethanol and resuspended in 50 µl H2O. Subsequent PCR amplification of 4 µl bisulfite-treated DNA was used for cloning of two fragments of the IRF-4 promoter (BS-I and BS-II) into pCR2.1 vector with the ‘TOPO TA cloning kit’ (Invitrogen) (see Figure 3A). The primers used for PCR amplification of the BS-I and BS-II fragments contain the putative altered sequence of the sense strand due to bisulfite treatment (converted cytosine residues are written in bold letters): BS-I-forward 5′-TATTTGGATTTTTAGGGAGTTTTTTTT-3′, BS-I-reverse 5′-ACCCAACTCCCTTAAACTATTAAACT-3′ (187 bp); BS-II-forward 5′-AGTTTAATAGTTTAAGGGAGTTGGGT-3′, BS-II-reverse 5′-CTCACCCTAAACTCAAAACTAAAAAC-3′ (674 bp). After bacterial amplification of the cloned PCR fragments by standard procedures, eight clones from each sample were sequenced with an automated sequencer (ABI Prism 377, Applied Biosystems).

In vitro methylation and reporter gene assays
The IRF-4 promoter-reporter gene construct was generously provided by J. Hiscott (31). Constructs were methylated in vitro with CpG Methylase (M.Sss I) as recommended by the manufacturer (NE Biolabs) and complete methylation was checked via restriction analysis (Figure 5A). Reporter gene assays using the dual luciferase assay (Promega) were performed similar to previous reports (29). Briefly, 5 nM of the reporter construct and the transfection control construct expressing the renilla luciferase gene were transientlyco-expressed via electroporation. The control construct served as an internal reference for transfection efficiency. Forty-eight hours after transfection, luciferase activity was measured with a LB 96 P microlumat (EG&G Berthold, Bad Wildbad, Germany). IRF-4 promoter activation was quantified as a ratio of measured firefly light units (flu) relative to renilla (rlu). Each experiment was carried out at least three times.