C57 BL/6J mice were purchased from Jackson Laboratory (Bar Harbor, ME) and maintained in Oriental Yeast Co. Ltd (Tokyo, Japan). Brain specimens from mice at three different developmental stages: E15, 15-day-old embryo; NB, new born; and AD, 12-week adult; were disected and stored as described previously (11).

Twenty primary neuroblastoma tumors were obtained in Nihon University Hospital (Tokyo, Japan) at the time of diagnosis, from 1999 to 2007. All the analyses of those specimens were performed under the approval of Nihon University Institutional Review Boards (IRB no. 51). Neither neoadjuvant chemotherapy nor irradiation therapy was given preoperatively to any patient. Four adrenal samples were collected from a nephroblastoma patient undergoing nephrectomy and from 3 neuroblastoma patients (cases 3, 8 and 20) undergoing tumor resection. All of the samples were immediately snap-frozen in liquid nitrogen and stored at −80°C until use. Summary of these patients is shown in Table I.

Human neuroblastoma cell lines NB9 and NB69 were obtained from Riken Cell Bank (Tsukuba, Japan). Both human neuroblastoma cell lines were maintained in RPMI-1640 (Nalarai Tesque, Kyoto, Japan) supplemented with 15% fetal bovine serum (Nichirei Biosciences, Tokyo, Japan), 100 IU/ml penicillin (Gibco™, Carlsbad, CA) and 100 μl/ml streptomycin (Gibco). The cells were cultured in a 37°C humidified atmosphere containing 5% CO₂ maintained in appropriate conditions recommended by the manufacturers.

Total genomic DNA was extracted from mouse brains, primary tumors, neuroblastoma cell lines and normal adrenal medullas with DNeasy Tissue Kit (Qiagen, Valencia, CA) and modified by sodium bisulfite using the EZ DNA Methylation Kit (Zymo Research, Orange, CA), by following the manufacturer’s instructions.

Sequenom MassARRAY quantitative methylation analysis (12) using the MassARRAY Compact System (www.sequenom.com) was employed for the quantitative DNA methylation analysis at CpG dinucleotides. This system utilizes mass spectrometry (MS) for the detection and quantitative analysis of DNA methylation using homogeneous MassCLEAVE (hMC) base-specific cleavage and matrix-assisted laser desor ption/ionization time-of-f light (MALDI-TOF) MS (13). The MethPrimer program (http://www.urogene.org/methprimer/index1.html) (12) was used to design bisulfite PCR primers (Table II). Each reverse primer has a T7-promotor tag for in vitro transcription (5′-cagtaatacgactcactatagggagaaggct-3′), and the forward primer is tagged with a 10 mer to balance melting temperature (TM) (5′-aggaagagag-3′). All primers were purchased from Operon (Tokyo, Japan). Polymerase chain reaction (PCR) amplification was performed using HotStarTaq Polymerase (Qiagen) in a 5 μl reaction volume using PCR primers at a 200 nM final concentration, and bisulfate treated DNA (∼20 ng/ml). After the treatment of shrimp alkaline phosphatase, 2 μl of the PCR products was used as a template for in vitro transcription and RNase A Cleavage for the T-reverse reaction (3′ to either rUTP or rCTP), as described in the manufacturer’s instructions (Sequenom hMC, Sequenom, San Diego, CA). The samples were desalted and spotted on a 384-pad SpectroCHIP (Sequenom) using a MassARRAY nanodispenser (Samsung Seoul, Korea), followed by spectral acquisition on a MassARRAY Analyzer Compact MALDITOF MS (Sequenom). The resultant methylation calls were analyzed by EpiTYPER software v1.0 (Sequenom) to generate quantitative measurements for each CpG site or an aggregate of multiple CpG sites. Since maldi-TOF mass methylated peaks do not denote a particular CpG site, but rather corresponds to the number of CpG sites methylated within the cleavage fragment, we decided to present average percent methylation of all CpG sites in the bisulfite PCR fragment with the standard curve.

Standard curve of DNA methylation level was made by using 0, 25, 50, 75 and 100% methylated samples. BAC DNA (RPMI-11 341L6) obtained from Roswell Park Cancer Institute (Buffalo, NY) was used as 0% methylation and M.Sss-1 double treated BAC DNA was used as 100% methylation. The PCR was carried out with a final volume of 50 μl, containing 1.0 μl of each 10.0 μM primer (final concentration 0.2 μM), 8.0 μl of 2.5 mM dNTP, 25 μl of 2 times GC Buffer (Takara Bio, Shiga, Japan), 0.5 U of LA taq (Takara Bio) and 1 μl of genomic DNA as a template. Amplification was carried out with an initial denaturing at 94°C for 1 min followed by 45 cycles of denaturing at 94°C for 30 sec, annealing for 1 min at the annealing temperature of each primer (60°C), extension for 3 min at 72°C, and then a final extension for 5 min at 72°C. The methylation reactions were carried out in 1X M.SssI buffer with 160 μM SAM (New England Biolabs, Ipswich, MA). In total reaction volume of 50 μl, 500 ng PCR product was treated with 4U M.SssI for 1 h at 37°C. Reactions were stopped for 20 min at 65°C and PCR product was purified Qiagen PCR purification kit (Qiagen). This CpG methyltransferase reaction was performed twice. Then M.SssI treated PCR product was produced (14). Curve was fitted and methylation levels were modified.

All of samples were collected and total cell lysates were prepared in M-PER mammalian protein extraction reagent (Thermo, Rockford, IL) containing a protease-inhibitor cocktail (Nalarai Tesque). Proteins (20 μg) were loaded on NuPAGE + 10% Bis-Tris gels (Invitrogen Life Technologies, Carlsbad, CA) for electrophoresis. The proteins were separated at 100 mA for 1 h, then transferred to polyvinylidene difluoride membranes by using iblot transfer for 7 min (Invitrogen). The membranes were incubated with Tris-buffered saline (TBS), containing 5% non-fat milk, 0.2% Tween-20 and a rabbit anti-NR4A3 polyclonal antibody (1:100) overnight (SC-30154, Santa Cruz Biotechnology, Santa Cruz, CA). The membranes were washed three times with a TBS containing 0.2% Tween-20. The immunocomplexed proteins were identified by reaction with a peroxidase-linked goat antibody to rabbit IgG (GE Healthcare, Little Chalfont, UK). Then these immunocomplexed proteins were detected by enhanced chemiluminescent reaction (Amersham Bioscience Inc., Piscataway, NJ). Immunoblotting with antibody to actin (Abcam, Cambridge, MA) provided an internal control for equal protein loading. Chemiluminescent detection was performed by LAS4000 (Fujifilm, Tokyo, Japan).

Formalin-fixed, paraffin-embedded serial sections (4 μm) were deparaffinized in xylene, rehydrated through graded alcohols, and immersed for 15 min in phosphate-buffered saline (PBS). The sections were soaked in 10 mmol/l of sodium citrate buffer (pH 6.9) and treated in a microwave for 15 min for antigen retrieval. After antigen retrieval the endogenous peroxidase activity was blocked with 3% hydrogen peroxidase in methanol for 30 min, and non-specific staining was then blocked by 1-h incubation with normal goat serum (Nichirei Biosciences). The sections were then incubated overnight at 4°C with 2 μg/ml of a rabbit anti-NR4A3 polyclonal antibody (SC-30154: Santa Cruz Biotechnology). The sections were treated for 30 min at room temperature with goat secondary antibody against rabbit immunoglobulins (Nichirei Biosciences). The sections were stained at room temperature for 25 min with AEC substrate kit (Vector Laboratories, Burlingame, CA). After staining with AEC substrate kit the sections were counterstained with hematoxylin.

Chromatin immunoprecipitation assays were performed essentially as previously described (15–17) with the following minor modifications. Neuroblastoma cell lines (NB9 and NB69) were fixed in 0.33 M (1%) formaldehyde for 10 min, before adding 4 volumes of ice-cold PBS containing 0.125 M glycine, to give an approximately 2-fold molar excess of glycine over formaldehyde. Then cells were washed with cold PBS containing Protease K (Nalarai Tesque). Crude cell lysates were sonicated to generate 200–1,000 bp DNA fragments. Chromatin was immunoprecipitated with 10 μl of rabbit antiserum raised against human CTCF (07–729; Millipore, Billerica, MA) per 1×10⁷ cells or with 2 μg normal rabbit immunoglobulin G (IgG; Millipore) used as a control, according to manufacturer’s protocols (Millipore). PCR amplification was performed using AccuPrime (Invitrogen) in a 25 μl reaction volume using PCR primers at 200 nM final concentration and 5 μl immunoprecipitated DNA as a template. Amplification was carried out with an initial denaturing at 94°C for 1 min followed by 38 cycles of denaturing at 94°C for 30 sec, annealing for 30 sec at the annealing temperature of each primer, extension for 1 min at 68°C, and then a final extension for 5 min at 72°C, using specific primers as follows: for NR4A3 exon 3, 5′-CTTCCCGCTCTTCCACTTC-3′; and 5′-TCACCTTGAAAAAGCCCTTG-3′, Tm 58°C; for cMYC, 5′-GTTTTAAGGAACCGCCTGTCCTTC-3′ and 5′-GGA TTGCAAATTACTCCTGCCTCC-3′, Tm 62°C (18). All primers were purchased from Operon.

The Mann-Whitney U test was used to evaluate the statistical significance of the difference in the methylation level of NR4A3 among the samples. The methylation levels were categorized by Youden index using 17 patients passed the observation period (19). The cutoff point between high and low levels of DNA methylation at each DMR was calculated by ROC curve analysis. Survival curves were calculated according to Kaplan-Meier analysis and compared with a log-rank test. Event-free survival was calculated as the time from diagnosis to event or last examination if the patient had no event. Recurrence, progression of disease and death from disease were counted as events. Death resulting from therapy complications or from second malignancy was not counted as an event but censored for event-free survival. The data were analyzed by the SPSS (Chicago, IL) for Windows. Differences were considered significant at p<0.05.

Methylation levels of each CpG site at the Nr4a3 exon 3 CpG island in mouse brains were analyzed at three different developmental stages. This CpGi (mouse chr4:48072571-48072905 in the USCS database, February, 2006 assembly) showed higher methylation level in the brain specimens of 12-weeks-old mice (AD brain), compared with brain specimens from 15-days-old embryos (E15) or new-born mice (NB) in the analysis using Mass ARRAY EpiTYPER (Fig. 1A). The average methylation level of all CpG sites in this region was significantly higher in AD brain than in E15 and NB brain specimens (Fig. 1B). Western blot analysis revealed higher expression level of NR4A3 protein in AD brain specimens, compared with the other developmental stages (Fig. 1C).

We analyzed methylation level of human homologous region (chr4:48072371-48072905, in USCS genome database, March, 2006, NCBI36/hg18) in the surgical resected NB specimens and found that all 3 NB specimens showed significantly lower methylation level at NR4A3 exon 3 CpGi compare to those from the matched adrenal tissues (Fig. 2). The CpGi located between NR4A3 promoter and intron 1 was not methylated at all in either neuroblastoma or adrenal samples.

Additional 17 neuroblastoma samples and 1 adrenal sample were used to confirm aberrant methylation at NR4A3 exon 3 CpGi. Adding a new adrenal sample, all the 4 adrenal samples were hypermethylated at NR4A3 exon 3 CpGi. The average methylation level in 4 adrenal samples was 64.6±2.1%. The methylation level of neuroblastoma specimens varied from *83.3±.. to *−0.5±0.9%, and the average value of the all samples was 27.0±25.8%, which was significantly lower than the average methylation level of 4 normal samples (p=0.005, Mann-Whitney U test) (Table I, Fig. 3A). In 17 out of 20 neuroblastoma specimens, methylation levels at NR4A3 exon 3 CpGi were low, compared with the average methylation level in 4 adrenal samples. Methylation level in 9 MYCN amplified neuroblastoma specimens was significantly lower, compared with that in 11 specimens without MYCN amplification (Fig. 3B) (p=0.005, Mann-Whitney U test).

For Kaplan-Meier analysis, cut off value of the methylation level was calculated as 11.93% by using youden index and the methylation levels of 20 patients passed the observation period. Twenty neuroblastoma patients were divided into two groups depending on their methylation levels at the NR4A3 exon 3 regions. The hypermethylation group has methylation level higher than the cut off value, and the hypomethylation group showed lower than that. Eight out of 20 neuroblastoma specimens were classified to hypomethylation group. There was significant association between the methylation level at NR4A3 exon 3 CpGi and patient outcome (p=0.034, log-rank test) (Fig. 3C).

Immunohistochemical analysis using NR4A3 antibody showed a strong signal in adrenal tissue sections and pronounced cytoplasmic staining. On the other hand, faint staining was seen in neuroblastoma specimens (Fig. 4).

The average methylation level at NR4A3 exon 3 CpGi in NB9 cells was 44±23.2%, on the other hand, it was 97.1±5.8% in NB69 cell (Fig. 5A and B). Western blot analysis revealed higher expression level of NR4A3 protein in NB69 compared with NB9 (Fig. 5C). This result also indicates that NR4A3 protein expression was correlated with NR4A3 exon 3 CpGi methylation in human neuroblastoma cell lines.

We examined chromatin immunoprecipitation assays to elucidate the mechanism in which methylation level of NR4A3 exon 3 CpGi regulates the expression level. DNA purified from the immunoprecipitated chromatin using anti-CTCF antibody was amplified by PCR for a candidate CTCF binding site of NR4A3 exon 3. In this analysis, the amplified PCR fragment was detected clearly in NB9 cells, but not in NB69 (Fig. 5D).