Phospho NF-κB p65 (Ser536; cat. 3031), 65 kDa, NF-κB p65 (cat. 3034) 65 kDa and Fn14 polyclonal antibody were purchased from Cell Signaling (Beverly, MA, USA). NF-κB p65 (sc-109), TWEAK (FL-249), and TWEAK (S-20) antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Secondary anti-goat Alexa 594 and anti-rabbit Alexa 488 antibodies were purchased from Invitrogen (Carlsbad, CA, USA). Human recombinant IL-1β, TNF-α and TWEAK were purchased from R&D Systems (Abingdon, UK), and human TWEAK ELISA was purchased from PeproTech (London, UK).

Primary neuroblastoma samples from tumors and non-malignant adrenals were obtained during surgery, snap-frozen in liquid nitrogen, and transferred to −80°C for future analysis. Twenty six neuroblastoma samples derived from children of different ages and all clinical stages, including different biological subsets (MYCN amplification, 7 of 27; 1p deletion, 9 of 27; Table I) were analyzed. Three childhood ganglioneuromas and three samples of non-malignant adrenals from children aged 12–25 months were also included. Ethical approval was obtained from the Karolinska University Hospital Research Ethics Committee.

Human neuroblastoma cell lines [SK-N-BE(2), SK-N-DZ, SH-SY5Y, SK-N-SH, SK-N-FI, IMR-32, SK-N-AS and SHEP-1] were grown in RPMI-1640 medium supplemented with 10% fetal calf serum (FCS), 2 mM L-glutamine, and antibiotics. The cultures were kept at 37°C in a humified 5% CO₂ atmosphere.

Total RNA was extracted from cultured neuroblastoma cells using RNeasy Mini kit (cat. no. 74104, Qiagen, Valencia, CA, USA) according to the manufacturer’s instructions. cDNA was synthesized using 2.0 μg total RNA that was reverse transcribed in a final volume of 50 μl using the SuperScript preamplification kit (Life Technologies, Inc., Gaithersburg, MD, USA).

Gene specific PCR was performed in 50 μl of reaction mixture containing 2–10 μl cDNA (from isolated RNA), 2.5 U of Taq DNA polymerase (Promega, Madison, WI, USA), 10 mM Tris-HCl (pH 8.0), 100 mM KCl, 0.1% Triton^® X-100, 2.0 mM MgCl₂, 1 mM deoxynucleotide triphosphate mix, and 1 μM of each primer.

PCR for TWEAK and Fn14 was performed as follows: 94°C for 5 min (first denaturation/hot start) and then at 94°C for 1 (denaturation), 52°C for 1.5 min (annealing), and 72°C for 1 min (extension) for 35 cycles with a 10 min final extension at 72°C. PCR conditions for β-actin were identical except for a 1 min annealing at 55°C, and a total of 27 cycles, with a 10 min final extension.

PCR amplifications were performed in a PTC-200 Peltier Thermal Cycler (MJ Research Inc. Waltham, MA, USA). PCR products were analyzed by agarose gel (1.5%) electrophoresis and photographed under UV light. Nucleotide sequences of PCR primers used were as follows: Fn14, 5′-GAC CTG GAC AAG TGC ATG GAC-3′ (sense) and 5′-AGC TGT TTT GTG TGA GCC AGC- 3′ (antisense); TWEAK 5′-ATC GCA GCC CAT TAT GAA GTT C-3′ (sense) and 5′-GAT GGA AAA CAC GTG AAC AGG C-3′ (antisense); β-actin, 5′-TGA CGG GGT CAC CCA CAC TGT GCC CAT CTA-3′ (sense) and 5′-ACT CGT CAT ACT CCT GCT TGC TGA TCC A-3′ (antisense); PCR fragments of 500 (Fn14), 607 (TWEAK), and 625 bp (β-actin) were expected.

Raw data files from four European expression microarray studies generated from two Affymetrix platforms (HU133A and HU133plus2) were obtained from ArrayExpress (www.ebi.ac.uk/microarray-as/ae/) and the r2 data base (http://hgserver1.amc.nl/cgi-bin/r2/main.cgi). The three studies run on the HU133A platform (25–27), were reanalyzed using gcRMA by Bioconducter for R 2.9.2 (library BioC 2.4) in two separate groups: i) De Preter data set comprising preamplified primary neuroblastoma samples (n=17, stages 1–4), ii) McArdle (n=16) and Wilzén (n=8) data sets comprising non-preamplified primary neuroblastoma samples. Also, expression values (log2) from 76 MAS5.0 normalized neuroblastoma samples (stages 1–4) run on the Affymetrix HU133plus2 platform were obtained from the r2 database (28) and referred to as the Versteeg data set.

Neuroblastoma samples from all three data sets (25–27) were divided into two groups based on their clinical stage (INSS stage) (29) and investigated for differential expression of TNFSF12 (TWEAK) and TNFRSF12A (Fn14) between groups. The significance was tested by Welch’s t-test (2-tailed, 2 sample comparison, unequal variance).

To prevent influence of endogenously produced TWEAK, neuroblastoma cells were serum starved in an RPMI-1640 medium containing 0.1% FCS for 24 h prior to the incubation of recombinant human TWEAK for the indicated concentrations and time-points.

Formalin-fixed and paraffin-embedded tissue sections were deparaffinized in xylene and graded alcohols, hydrated and washed in PBS. After antigen retrieval in sodium citrate buffer (pH 6.0) in a microwave oven, the endogenous peroxidase was blocked by 0.3% H₂O₂ for 15 min. Sections were incubated overnight at 4°C with primary antibody (TWEAK; FL-249). As a secondary antibody, the anti-rabbit horseradish peroxidase (HRP) SuperPicTure Polymer detection kit was used (Zymed-Invitrogen, San Francisco, CA, USA). A matched isotype control was used as a control for non-specific background staining. Routine standard staining showing a normal histology of neuroblastoma was performed with hematoxylin and eosin.

For immunofluorescence studies, cells were grown on fibronectin-coated chamber slides (Nunc, Roskilde, Denmark) for 24 h. Cultures were then washed and fixed with 2% paraformaldehyde for 15 min and 70% cold methanol for 5 min. After washing with PBS buffer, goat-anti TWEAK (S-20) and rabbit-anti Fn14 antibodies were incubated with cultures overnight at 4°C. After rinsing in PBS, cultures were incubated with secondary antibodies conjugated with Alexa 488 and Alexa 599, respectively. A matched isotype control was used as a control for non-specific background staining. The cells were examined in a Zeiss axiophot photomicroscope (Carl Zeiss, Oberkochen, Germany). Nuclear translocation of NF-κB upon stimulation with TWEAK was performed by immunofluorescence studies using anti-NF-κB p65 antibody (sc-109) that recognizes both non-phosphorylated and phosphorylated forms of NF-κB p65.

TWEAK ELISA was performed to measure the endogenous production of TWEAK in neuroblastoma cells upon stimulation by cytokines IL-1β and TNF-α. Neuroblastoma cells [SK-N-AS, SK-N-BE(2) and SH-SY5Y] were seeded in a regular growth medium in 96-well plates and allowed to attach. Cells were then treated with two concentrations (10 and 50 ng/ml) of IL-1β and TNF-α, respectively, for 12 h. Supernatants from treated cells were collected and 100 μl medium per well were analyzed by using the Human TWEAK ELISA Development kit (PeproTech, Rocky Hill, NJ, USA), following the manufacturer’s instructions.

Proteins from TWEAK-treated cells and control cells were extracted in RIPA lysis buffer (cat. no. 20–188, Upstate Biotechnology, USA) containing complete, mini, EDTA-free protease inhibitor (cat. no. 11 836 170 001, Roche). The protein content was measured using Bradford reagents (Bio-Rad Laboratories, CA, USA). Equal amounts of protein were separated by NuPAGE, Novex and Tris-Acetat Mini Gels (Invitrogen) 4–12% in reduced conditions, and proteins were transferred to a PVDF (Pierce, Rockford, IL, USA) membrane and incubated with primary antibodies at 4°C overnight. Alkaline-phosphatase conjugated antibodies were used as secondary antibodies. Detection and visualization were performed using Pierce Super Signaling solutions (Pierce), the Fujifilm Luminescent Image Analyzer LAS-3000 and the Fujifilm MultiGauge (Ver. 3.0) analysis software.

The NF-κB-responsive reporter plasmid κBcon A-LUC was provided by E. Sontag (30) and also described in Johannesen et al(31). SK-N-AS and SK-N-BE cells were seeded in 6-well plates in triplicates the day before transfection. Cells were transfected using Lipofectamine 2000 (Invitrogen), 4 μg/μl DNA per well and calf thymus DNA (Amersham Pharmacia, Sweden), and incubated for 6 h at 37°C in a humidified 5% CO₂ atmosphere. Cells were then serum starved for 14 h and subsequently incubated for 6 h with TWEAK (100 ng/ml) or TNF-α (30 ng/ml). Cells were washed in 1X PBS and lysed in TROPIX lysis buffer containing 0.5 mM DTT. Luciferase activity was determined using the Dual-Light Luciferase Gene Assay System (Applied Biosystems Inc., Foster City, CA, USA) in a Luminoscan RT (Labsystems, Helsinki, Finland). Luciferase measurements were corrected to protein concentrations in each cell lysate, and protein measurements were performed as described above. The difference between the groups was analyzed on log-transformed values after normalization to total protein content in the cell lysates using a repeated measures one-way ANOVA test, followed by a Bonferroni multiple comparison test.

SK-N-AS and SK-N-BE(2) cells were seeded in 6-well culture plates in RPMI medium at a 30-50% confluence. Cells were transfected with target-specific Fn14 (sc-43764), TWEAK (sc-37522), control (Fluorescein conjugate-A) (sc-36869) or scrambled control (sc-37007) siRNA (Santa Cruz Biotechnology), respectively, at a concentration of 33 nM using Lipofectamine 2000 in OptiMEM. To evaluate cell viability, western blot analysis of protein extracts and trypan blue exclusion assay were performed 72 h after the initial transfection. Transfection efficiency was assessed in SK-N-AS cells transfected with control (Fluorescein conjugate-A) using flow cytometry.

The presence of MMP-2 and MMP-9 in serum-free media from cells treated with or without recombinant TWEAK was determined by SDS-gelatin zymography. Approximately 30,000 cells were seeded in 96-well plates and left to attach in 10% RPMI-1640 overnight. Then, SK-N-AS and SK-N-SH cells were starved in 0.1% RPMI-1640 for 24 h prior to incubation with TWEAK (0–1000 ng/ml) for 48 h. SDS-substrate PAGE was done as previously described (32,33) with gels containing 0.1% (w/v) gelatin. The gelatin zymograms were calibrated with a mixture of conditioned serum-free medium from THP-1 and humans skin fibroblast cells (32). Conditioned medium (10 μl) was mixed with 2.5 μl of loading buffer (250 mM Tris-HCl, pH 6.8, 10% SDS, 0.03% bromophenol blue and 50% glycerol). Eight μl of this non-heated mixture was applied to the gel, which was run at 20 mA/gel at 4°C. Thereafter, the gel was washed twice in 100 ml of washing buffer [2.5% (v/v) Triton X-100 in water], and then incubated in 100 ml of assay buffer (50 mM Tris-HCl, pH 7.5, 5 mM CaCl₂, 0.2 M NaCl and 0.02% Brij-35) for ∼20 h at 37°C. Gels were stained with 0.2% Coomassie brilliant blue R-250 (30% methanol) and destained in a solution containing 30% methanol and 10% acetic acid. Gelatinase activity was evident as cleared regions.

All statistical analyses were performed with GraphPad Prism Software (GraphPad Software, San Diego, CA, USA). The t-test was used to determine whether the mean of a single sample differed significantly from control. To compare several treatment groups, one-way ANOVA with Tukey multiple-comparisons tests were used. P<0.05 was considered statistically significant.

We investigated eight different neuroblastoma cell lines for the expression of TWEAK and Fn14. All human neuroblastoma cell lines investigated showed varying degrees of both TWEAK and Fn14 mRNA and protein expression as detected by RT-PCR and western blot analysis, respectively (Fig. 1A and B). The 30–35-kDa band for TWEAK represents the transmembrane form of TWEAK as the soluble form could not be detected by western blot analysis. We selected 3 cell lines SK-N-AS, SK-N-BE(2) and SH-SY5Y for the in vitro experiments. The rationale behind choosing these cell lines was based on the genetical and phenotypical differences between the cell lines. SK-N-AS and SH-SY5Y are typical non-MYCN-amplified cell lines, whereas the SK-N-BE(2) cell line is MYCN-amplified and P53-mutated. Furthermore, while the SK-N-BE(2), SK-N-SH and SK-N-AS show multi-drug resistant (MDR) phenotype, the SH-SY5Y cell line does not. In order to investigate whether neuroblastoma cells were able to produce and secrete TWEAK in vitro, we assessed the TWEAK production by ELISA. The TWEAK levels in cell supernatants from SK-N-AS cells showed an elevated secretion upon stimulation with pro-inflammatory cytokines IL-1β and TNF-α (Fig. 1C). Similar results were obtained for SH-SY5Y and SK-N-BE(2) cell lines (data not shown).

Immunofluorescence staining of neuroblastoma SK-N-AS cells stained with antibodies towards TWEAK (red) and Fn14 (green) revealed the cellular distribution of the ligand and receptor, demonstrating a distribution of TWEAK and Fn14 in the cytoplasm but also to a certain degree to the nuclear compartment (Fig. 1D).

The staining of primary neuroblastoma tumor tissue with antibodies against TWEAK and Fn14 revealed significant cytoplasmic and nuclear expression of both TWEAK and Fn14 in all primary tumors that were analyzed (Fig. 1E and Table I). No significant difference in staining intensity between favorable and non-favorable tumors (MYCN-amplified vs non-MYCN-amplified) could be detected by immunohistochemistry and no staining was observed in sections incubated with isotype control antibody. Moreover, mRNA levels of both TWEAK and its receptor Fn14 were found to be generally higher in primary high stage tumors (stage 3–4) compared to low stage tumors (stage 1–2), when investigating three European micro-array data sets (Fig. 2). The up-regulation was significant for TWEAK and Fn14 in one out of three data sets respectively, i.e. the De Preter and Versteeg data sets (p<0.05, Welch’s t-test; Fig. 2). Also, the expression variance of both TWEAK and Fn14 was considerably higher in the high-stage group, and the up-regulation seemed to involve a sub-set of tumor cases. Ten out of 117 cases from all three data sets showed up-regulation of TWEAK (fold change >2), 13 showed up-regulation of Fn14 (fold change >2), and 9 showed up-regulation of both genes (fold change >2) compared to the mean expression levels of the low-stage group (Fig. 2).

TWEAK treatment has previously been shown to stimulate NF-κB activation in different cell types (15,34–36). To determine if TWEAK could induce activation of NF-κB in neuroblastoma cells, we examined the intracellular localization of NF-κB in TWEAK stimulated SK-N-AS cells. No nuclear staining for NF-κB was observed in untreated cells, but after 20 min of stimulation with recombinant TWEAK, nuclear trans-location of NF-κB was detected (Fig. 3A). To validate the TWEAK induction of NF-κB in SK-N-AS cells grown in 0.1% FCS, we isolated total protein fractions of cells treated with TWEAK (100 ng/ml) and immunoblotted for both the NF-κB and the pNF-κB subunit. Blots revealed an increased level of pNF-κB (p65) subunit shortly after stimulation by TWEAK at the same time that NF-κB showed a decreased signal (Fig. 3B). This is consistent with the immunostaining results showing activation of pNF-κB (Fig. 3A), and the results of SK-N-AS cells upon stimulation with TWEAK showing increase of NF-κB transcriptional activity upon transient transfection using the NF-κB-responsive reporter plasmid κBcon A-LUC (Fig. 3C).

Furthermore, immunohistochemical analysis of neuroblastoma primary tumors using phospho-specific NF-κB antibody (p65) revealed a significant nuclear staining (Fig. 1E).

To investigate the influence of TWEAK on neuroblastoma cell survival, SK-N-AS cells were transfected with siRNA targeting TWEAK or Fn14. As shown in Fig. 4A, the silencing of both TWEAK and Fn14 resulted in significant decrease in neuroblastoma cell survival compared to cells transfected with a scrambled siRNA construct (p<0.05). Addition of recombinant TWEAK partly restored TWEAK expression and cell survival in TWEAK siRNA-treated SK-N-AS cells, underscoring a role for TWEAK in cell survival (Fig. 4A and B). Increased expression of Fn14 in TWEAK siRNA-treated cells compared to scramble- and non-treated cells (Figs. 1B and 4B) is possibly due to the absence of the ligand resulting in lowered receptor internalization and degradation of receptor-ligand complex. Additional western blotting was performed to also confirm the specific down-regulation of TWEAK or Fn14 expression in SK-N-AS cells following siRNA transfection (Fig. 4C).

Since MMP-9 is a NF-κB responsive gene and TWEAK has been shown to induce MMP-9 protein level in other cell systems (21,22), we investigated the effect of TWEAK on the secretion of MMP-2 and MMP-9 in four different neuroblastoma cell lines. Zymography performed on conditioned medium from neuroblastoma cell lines revealed that MMP-2 was constitutively expressed, whereas TWEAK induced the release of MMP-9 in a dose-dependent manner in SK-N-AS and SK-N-SH cells (Fig. 5). Similar results were obtained for SK-N-BE(2) cells whereas SK-N-SY cells did not release MMP-9 upon stimulation with TWEAK (data not shown), indicating a heterogeneity with respect to TWEAK induction of MMP-9 in neuroblastoma cells.