This study was approved by the ethics committee of Nagoya City University (Nagoya City, Japan). Before participation, informed consent was obtained from all subjects in accordance with the Declaration of Helsinki. FLSs were cultured from the synovial tissues of 10 patients who had RA, who were undergoing total knee arthroplasty, and who met the American Rheumatism Association 1987 revised criteria for the classification of RA 23, as previously described 242526. The clinical characteristics of these patients are shown in Table 1. FLSs were maintained in Dulbecco's modified Eagle's medium supplemented with penicillin (100 units per mL), streptomycin (100 μg/mL), and 10% fetal bovine serum at 37°C in a 5% CO₂ atmosphere. The cultures were found to be completely free of lymphoid and monocytic cells. Cells were allowed to adhere overnight, and then non-adherent cells were removed and adherent FLSs were split at a 1:3 ratio when they reached 70% to 80% confluency. FLSs were used between passages 3 and 9, during which time they were a homogeneous population of cells.

The 1,243-base pair (bp) GLS/TP promoter fragment was a kind gift from Mayumi Ono (Department of Pharmaceutical Oncology, Kyushu University, Fukuoka, Japan) 19. The fragment was cloned into luciferase gene Basic Vector 2 (NIPPON GENE CO., LTD, Tokyo, Japan) and digested with XhoI and HindIII. This reporter construct was designated pTP-Luc1. To prepare other deletion constructs, pTP-Luc1 was digested with different restriction enzymes, including XhoI, StuI, SacI, and KpnI. Both 5' overhangs and 3' overhangs of the digested product were blunted by T4 DNA polymerase (Takara Bio Inc., Otsu, Japan) and self-ligated by T4 DNA ligase (Takara Bio Inc.) to produce pGLS/TP I, pGLS/TP II, and pGLS/TP III (Figure 1).

FLSs were seeded into a 48-well culture plate (5 × 10³ cells per well) and incubated at 37°C in a humidified atmosphere of 5% CO₂/air for 1 week. They were then co-transfected with luciferase reporter vector (400 ng/well) and pRL-SV40 internal control construct (10 ng/well) (Promega Corporation, Madison, WI, USA) by using Lipofectamine™ LTX (0.5 μL/well) and Plus™ Reagent (0.5 μL/well) (Invitrogen Corporation, Life Technologies Corporation, Carlsbad, CA, USA). The medium was replaced after 12 hours, and FLSs were treated with 100 or 300 nM mithramycin (Sigma-Aldrich, St. Louis, MO, USA) for 24 hours. Cells were harvested by using passive lysis buffer (65 μL/well), and the luciferase activity was measured by using a Dual-Luciferase Assay System (Promega Corporation) and normalized to the internal control.

Chromatin immunoprecipitation (ChIP) assays were performed by using ChIP-IT™ Express kits (Active Motif, Carlsbad, CA, USA) in accordance with a reported protocol 27 with minor modifications. In brief, 1.0 × 10⁶ FLSs cells from 60-mm dishes were treated with or without 300 nM mithramycin for 24 hours and then harvested. Cells were cross-linked by formaldehyde (1% final concentration) and incubated at room temperature for 10 minutes. Cells were then washed with ice-cold phosphate-buffered saline (PBS) containing protease inhibitors, and the fixation reaction was stopped by adding 10 mL of Glycine Stop-Fix Solution (Active Motif). Samples were lysed for 30 minutes in lysis buffer on ice, and the chromatin was sheared by sonicating 20 times for 40 seconds each at maximum power with an ultrasonic processor, including 30 seconds of cooling on ice between each pulse (Misonix, Inc., Farmingdale, NY, USA). Cross-linked and released chromatin fractions were immunoprecipitated with magnetic beads, Sp1 antibodies, and non-specific IgG (rabbit) on a rolling shaker overnight at 4°C. Cross-linking of the immunoprecipitates containing fragmented DNA was chemically reversed, and the polymerase chain reaction (PCR) was performed with MESA GREEN qPCR MasterMix Plus for SYBR Assay I Low ROX (Eurogentec, San Diego, CA, USA). The PCR primers used for amplifying promoters containing the Sp1-binding site furthest upstream were 5'-AACTGTGGGCCTTCCCACTC-3' and 5'-TGCTGAGGTCCTCGAAGAAAC-3', which produced a 227-bp fragment (Figure 2a).

FLSs were incubated in 60-mm dishes with 1 ng/mL TNF-α (R&D Systems, Minneapolis, MN, USA) in the presence or absence of mithramycin. To prepare nuclear extracts, cells were rinsed in ice-cold PBS and lysed in 10 mM Tris-HCl (pH 7.5), 1 mM ethylenediaminetetraacetic acid (EDTA), 0.5% Nonidet P-40, and a protease inhibitor cocktail (Sigma-Aldrich) for 10 minutes at 4°C. Cell lysates were centrifuged at 20,000 g for 10 minutes at 4°C to separate the cytoplasmic fraction (supernatant). Insoluble materials were dissolved in sodium dodecyl sulphate sample buffer to collect the nuclear extracts. After measuring the protein concentration, samples (20 μg of total protein/lane) were loaded and separated by electrophoresis on a 7.5% polyacrylamide gel (Bio-Rad Laboratories, Hercules, CA, USA) and then transferred to a polyvinylidene difluoride membrane (Immobilin-P; EMD Millipore, Billerica, MA, USA). The blots were blocked for 60 minutes with 5% non-fat dry milk in Tris-buffered saline containing 0.1% Tween 20 (TBS-T) and then incubated with an anti-Sp1 antibody (1:1,000; Cell Signaling Technology, Inc., Danvers, MA, USA), anti-lamin C (1:1,000; ImmuQuest Ltd, Seamer, North Yorkshire, UK), and anti-α-tubulin antibody (1:1,000; Cell Signaling Technology, Inc.) as loading controls in TBS-T for 2 hours at room temperature. After three washes with TBS-T, membranes were incubated with appropriate horseradish peroxidase-conjugated secondary antibodies in TBS-T (1:3,000; GE Healthcare, Little Chalfont, Buckinghamshire, UK) for 1 hour at room temperature and washed three times with TBS-T. Protein bands were detected by using the ECL Plus Western Blotting Detection System (GE Healthcare) and then quantified with a densitometric scanner by using ImageJ software 28.

Expression of the GLS gene was assessed by using reverse transcription-PCR (RT-PCR). Total RNA was isolated by using TRIzol reagent (Invitrogen Corporation), and RT was carried out by using random primers and Ready-To-Go You-Prime First-Strand Beads (GE Healthcare). The resultant cDNA was subjected to real-time PCR by using a 7500 Fast Real-time PCR System (Life Technologies Corporation) with MESA GREEN qPCR MasterMix Plus for SYBR Assay I Low ROX (Eurogentec) and primers. The PCR conditions were an initial denaturation at 95°C for 10 minutes, followed by 40 cycles at 95°C for 5 seconds and 60°C for 1 minute. The relative quantification value of GLS was normalized to an endogenous control, β-actin, after confirming that the GLS and β-actin cDNAs were amplified with the same efficiency. The primers were as follows: GLS, 5'-ACAGGAGGCACCTTGGATAA-3' and 5'-CCGAACTTAACGTCCACCAC-3' (272 bp), and β-actin, 5'-GACCTGACTGACTACCTCAT-3' and 5'-TCGTCATACTCCTGCTTGCT-3' (542 bp).

Sp1 Stealth siRNAs and negative control siRNA were purchased from Invitrogen Corporation. The following Sp1 siRNA oligos were used: sense sequence 5'-GACAGGUCAGUUGGCAGACUCUACA-3' and anti-sense sequence 5'-UGUAGAGUCUGCCAACUGACCUGUC-3'. FLSs were transfected with the indicated combinations of siRNA against Sp1 or negative control siRNA at a final concentration of 20 nM by using Lipofectamine RNAiMAX transfection reagent (Invitrogen Corporation) in accordance with the recommendations of the manufacturer. FLSs were harvested for Western blotting 48 hours after transfection. FLSs were incubated for 24 hours after transfection and further incubated with 1 ng/mL TNF-α for 24 hours and harvested for RT-PCR.

GLS was measured by using an enzyme immunoassay system, as described by Hirano and colleagues 29. Polyclonal antibodies on the solid phase were obtained by immunizing New Zealand albino rabbits with 40 μg of purified natural human GLS, and the monoclonal antibody was used with the β-galactosidase-labeled secondary antibody. The detection limit of this assay was 150 pg/mL, and no significant cross-reactivity or interference was observed. The GLS concentration in cultured FLSs was normalized to the protein content, as measured by using a Pierce BCA protein assay kit (Thermo Fisher Scientific, Waltham, MA, USA).

Confluent FLSs in chambered slides coated with BD Matrigel Matrix (BD Biosciences, Franklin Lakes, NJ, USA) were fixed for 30 minutes in 3% paraformaldehyde, permeabilized with 0.2% Triton X-100 for 5 minutes, washed with PBS, and blocked for 60 minutes at room temperature with blocking solution comprising 3% bovine serum albumin and 0.1% glycine in PBS. After washing, cells were labeled overnight at 4°C with the primary antibody (1:1,000 anti-TP/GLS antibody; a gift from Chugai Pharmaceutical Co., Ltd., Tokyo, Japan). Cells were washed and labeled with an Alexa Flour 594-labeled (red) goat anti-mouse IgG (1:1,000; Molecular Probes Inc., now part of Invitrogen Corporation) secondary antibody. After washing, sections were mounted on glass slides containing ProLong Gold Antifade with 4',6-diamidino-2-phenylindole (DAPI) (Invitrogen Corporation). Stained cells were visualized by using an AX70 fluorescence microscope (Olympus, Tokyo, Japan). Total red intensity of immunostaining in nine random fields was quantified by ImageJ. The numbers of cells were counted in the fields. Data were presented as mean of intensity per cell.

All data were entered into an electronic database and analyzed by using GraphPad Prism 4 (GraphPad Software, Inc., San Diego, CA, USA). Data are expressed as the mean ± standard error of the mean unless otherwise stated. The statistical significance of the differences was examined by using the Mann-Whitney U test. In all cases, P values of less than 0.05 were considered to be statistically significant.

We prepared three deletion constructs of the GLS/TP promoter and fused them to the reporter luciferase plasmid (Figure 1) to investigate whether mithramycin inhibited GLS/TP promoter activity. The promoter activity was expressed relative to the internal control. Treatment with mithramycin inhibited the promoter activity in RA FLSs transfected with pTP-Luc1 (55.5%, 100 nM mithramycin; 59.2%, 300 nM mithramycin), whereas promoter activity was measured at 41.8% in RA FLSs transfected with pGLS/TP I, relative to those transfected with pTP-Luc1. The observed inhibition was accelerated following treatment with mithramycin (38.2%, 100 nM mithramycin; 35.1%, 300 nM mithramycin) in RA FLSs transfected with pGLS/TP I. The promoter activity of pGLS/TP II and pGLS/TP III was largely inhibited with or without mithramycin.

We performed a ChIP assay to confirm Sp1 binding to the promoter and examined the inhibitory effect of mithramycin on GLS gene expression. Sp1 was shown to bind to the furthest upstream Sp1-binding site of the GLS/TP promoter fragment (1,243 bp). Treatment with mithramycin suppressed binding at this site (Figure 2b). No amplification band was observed by using the non-specific rabbit IgG, indicating the specificity of the DNA immunoprecipitation and ChIP assays.

To examine whether the nuclear localization of Sp1 changed following treatment with TNF-α or mithramycin, we prepared nuclear extracts from FLSs treated with or without 300 nM mithramycin for 30 minutes, further incubated with 1 ng/mL TNF-α for 24 hours, and then subjected to immunoblotting with anti-Sp1, anti-lamin C, and anti-α-tubulin antibodies. TNF-α treatment led to a remarkable accumulation of Sp1 in nuclear extracts, which was inhibited by treatment with mithramycin (Figure 3). The nuclear extracts were stained by the anti-lamin C antibody as a nuclear lamin marker but not by the anti-α-tubulin antibody as a cytoplasmic marker.

To investigate the direct effect of Sp1 on GLS expression, we employed an siRNA against Sp1 for as an approach to inhibit GLS expression. The efficiency and specificity of siRNA gene knockdown of Sp1 were determined by Western blotting for Sp1 expression. Sp1 siRNA suppressed the Sp1 protein expression (Figure 4a). Sp1 siRNA reduced TNF-α-induced GLS gene expression by 56.7% at 48 hours post-transfection (Figure 4b).

FLSs were treated in the presence or absence of 10 to 300 nM mithramycin for 30 minutes and then further incubated with 1 ng/mL TNF-α for 24 hours. GLS mRNA (Figure 5a) and protein (Figure 5b) levels were significantly induced by treatment with TNF-α alone (11.3-fold and 2.1-fold compared with the control, respectively), and these inductions were suppressed by mithramycin treatment in a dose-dependent manner. GLS was not induced by treatment with mithramycin alone. We confirmed the non-toxic concentration at least 48 hours after incubation with 10 to 300 nM mithramycin and 1 ng/mL TNF-α by using WST-8 assays (Cell Counting Kit-8; Dojindo Laboratories, Kumamoto, Japan) (data not shown). GLS mRNA expression increased in response to TNF-α (1 ng/mL) and reached a maximum at 24 hours (21.9-fold compared with the level at 0 hours) (Figure 5c). GLS protein expression increased in response to TNF-α and continued to rise until at least 48 hours after treatment (5.6-fold compared with the level at 0 hours) (Figure 5d). These increases in GLS mRNA and protein were significantly suppressed by 300 nM mithramycin treatment. To confirm whether TNF-α directly regulates GLS expression, FLSs were pre-treated with the protein synthesis inhibitor, cycloheximide. TNF-α-induced GLS mRNA production significantly decreased in a dose-dependent manner with treatment of cycloheximide (Additional file 1).

FLSs were cultured to confluence in the presence or absence of 300 nM mithramycin for 30 minutes, followed by further incubation with or without 1 ng/mL TNF-α for 24 hours. We observed no morphological change in FLSs during immunocytochemical staining. FLSs were immunostained with GLS antibody (red). Staining was weakly diffuse in the cytoplasm of FLSs that had not undergone treatment (Figure 6a). Treatment with mithramycin alone had no effect on GLS staining. The expression of GLS protein was significantly induced by treatment with TNF-α alone (1.5-fold compared with the control). This induction was significantly suppressed by mithramycin treatment (0.6-fold compared with the sample induced by TNF-α alone) (Figure 6b).