DFO (deferoxamine), mimosine (MIM), ferric citrate (FC), alkaline-phosphatase-conjugated monoclonal mouse anti-rabbit IgG, and p-nitrophenyl phosphate tablets were purchased from Sigma-Aldrich (St. Louis, MO). All reagents and media for tissue culture experiments were tested for their LPS contents with a colorimetric Limulus amebocyte lysate assay (detection limit 10 pg/Mℓ Sigma-Aldrich). Human IL-1β was obtained from Invitrogen (Carlsbad, CA). TNF-α and polyclonal goat anti-human IL-8 were obtained from R & D Systems (Minneapolis, MN). SB203580 and PD98059 were purchased from Calbiochem (La Jolla, CA). Polyclonal rabbit anti-human IL-8 was from Endogen (Woburn, MA). Abs against HRP-conjugated anti-rabbit IgG was from Amersham Biosciences (Little Chalfont, U.K.). Anti-human I-κBα or antiphospho-I-κBα was from Santa Cruz Biotechnology (Santa Cruz, CA). Dulbecco's modified Eagle's medium (DMEM), fetal bovine serum (FBS), and other tissue culture reagents were purchased from Gibco BRL (Grand Island, NY).

Informed consent was obtained from all subjects. The study protocol was approved by ethical committee and institutional review boards of Wonkwang University, dental college, Iksan, Republic of Korea (WKDIRB 20048-13). This study complies with the Helsinki Declaration in its recent Korean version. The trial will also be carried out in keeping with local legal and regulatory requirements.

HPV-immortalized human oral keratinocytes (IHOK) were in vitro-established cell lines, which immortalized by transfection of normal human oral keratinocytes with PLXSN vector containing the E6/E7 open reading frames of HPV type 16 as previously described 272829. The IHOK cells were obtained from Dr. Myung-Hee Park (NIDCR, NIH, MD), and cultured in the keratinocyte growth medium (KGM, Gibco, Grand Islands, NY) supplemented with 2 mL of bovine pituitary extract (13 mg/ml), 0.5 ml each of hydrocortisone (0.5 mg/ml), human epidermal growth factor (0.5 μg/ml), insulin (5 mg/ml), epinephrine (0.5 mg/ml), transferrin (10 mg/ml), triiodothyronine (6.5 μg/ml), and GA-1000 and 0.05 mM CaCl₂.

HN12 (= HNSCC12) cells were cultured in Dulbecco's modified Eagle's medium (DMEM, Biofluid, Rockville, MD) containing 10% fetal bovine serum (FBS, Gibco, USA) with 100 U/ml penicillin and 100 U/ml streptomycin (Life Technologies, Gaithersburg, MD). HN12 from metastatic carcinoma of the oral cavity 30 were obtained from the laboratory of Dr. John F. Ensley (Wayne State University). All the cell lines were grown at 37°C in a humidified atmosphere of 5% CO₂ and 95% air. Cells were dissociated with 0.25% trypsin just before transfer for experiments and were counted using a hemocytometer.

IHOK and HN12 cells were seeded at 2 × 10⁴ into 12-well plates (Nalge Nunc International, Rochester, NY) and cultured in serum-containing medium for 24 h prior to treatment. Cells were treated with fresh medium containing stimuli as indicated. The supernatants were collected, cleared by centrifugation, and kept at -20°C until evaluation by ELISA. For measurement of IL-8 concentrations in cell culture supernatants, 96-well microtiter plates (MaxiSorp; Nunc) were coated with 0.2 μg/well goat anti-human IL-8 Abs (R & D Systems) in 50 μℓ of PBS at 4°C overnight. All further steps were conducted at room temperature. After washing three times with PBS, nonspecific binding sites were blocked by incubation with 150 μℓ PBS + 1% BSA/0.05% Tween-20/well for 2 h. After three washes with PBS, 50 μℓ of samples or IL-8 standards were added and incubated for 2 h. As a second antibody, 0.05 μg/well polyclonal rabbit anti-human IL-8 (Endogen) was added and incubated for 2 h. As a third antibody, alkaline phosphatase-labeled monoclonal mouse anti-rabbit IgG (Sigma-Aldrich) was diluted in 50 μℓ of PBS + 0.1% BSA/0.05% Tween-20 to 1:50,000 and incubated for 2 h. Finally, alkaline phosphatase substrate p-nitrophenyl phosphate (Sigma-Aldrich) was added at a concentration of 1 mg/ml in 0.1 M glycine buffer (pH 10.4) containing 1 mM MgCl₂ and 1 mM ZnCl₂. After overnight incubation, plates were read at 405 nm on a microplate reader (Molecular Devices, Sunnyvale, CA). The detection limit of the ELISA was 30 pg/ml.

Cells (5 × 10⁵) were grown in 60-mm culture dishes for 24 h prior to treatment, and incubated for 4–24 h in a fresh medium containing stimuli as indicated. After discarding growth medium, total RNA was isolated from cells using easy-Blue (iNtRON Biotechnology, Daejon, Korea), following the manufacturer's instructions. Reverse transcription of the RNA was performed using AccuPower RT PreMix (Bioneer, Daejon, Korea). One microgram of RNA and 20 pmol primers were preincubated at 70 degrees for 5 min and transferred to a mixture tube. The reaction volume was 20 μℓ. cDNA synthesis was performed at 42 degrees for 60 min, followed by RT inactivation at 94 degrees for 5 min. Thereafter, the RT-generated DNA (2–5μℓ) was amplified using AccuPower PCR PreMix (Bioneer). The primers used for cDNA amplification and PCR conditions were as follows: IL-8 31, 5-ATGACTTCCAAGCTGGCCGTGGCT-3 (sense) and 5-TCTCAGCCCTCTTCAAAAACTTCTC-3 (antisense); GAPDH, 5-CG GAGTCAACGGATTTGGTCGTAT-3 (sense) and 5-AGCCTTCTCCA TGGTGGTGAAGAC-3 (antisense); an initial denaturation at 94 degrees for 5 min; 25 cycles being conducted for the time course measurements of IL-8 and GAPDH and 30 cycles for the detection of mRNA decay, each cycle with 30 s of denaturation at 94 degrees, 30 s of annealing at 62 degrees and 30 s of extension at 72 degrees; and a final dwell at 72 degrees for 7 min. The expected PCR products were 289 bp (for IL-8) and 306 bp (for GAPDH). PCR products were resolved on a 1.5% agarose gel and stained with ethidium bromide.

Cells (1 × 10⁶) were grown in 100-mm culture dishes and incubated for various times with DFO. Cells were mechanically scraped in PBS and washed, then 1 × 10⁷ cells were resuspended in 500 μl of lysis buffer A (15 mM KCl, 10 mM HEPES, 2 mM MgCl₂, 0.1 mM EDTA, 1 mM PMSF, 1 mM dithiothreitol (DTT), 10 μg/ml aprotinin, 2 μg/ml leupeptin, 0.1% NP-40, pH 7.6). Cell suspensions were then incubated for 10–15 min on ice with occasional vortexing, and centrifuged for 30 s to pellet nuclei, which were rinsed with wash buffer B (2 mM KCl, 25 mM HEPES, 0.1 mM EDTA, 1 mM PMSF, 1 mM DTT, 10 μg/ml aprotinin, 2 μg/ml leupeptin, pH 7.6) and incubated at 4°C for 20 min. Nuclear extracts were then prepared by centrifugation at 20 000 × g for 15 min in buffer C (25 mM HEPES, 0.1 mM EDTA, 20% glycerol, pH 7.6) and stored at -80° until used for EMSA. The probes contained the NF-kB or the AP-1 oligonucleotide consensus sequence and were labeled with γ-³²P (Amersham Life Science) (3000 Ci/mmol, 250 μCi) using T4 polynucleotide kinase (Boehringer Mannheim). In competition assays, 100 × cold NF-kB competitor was added. The DNA-protein complex was separated on a non-denaturating 4% polyacrylamide gel in TBE buffer (Tris-HCl, boric acid, EDTA 2 mM, pH 8.0). After electrophoresis, the gel was dried and autoradiographed by overnight exposure to X-ray film.

IHOK and HN 12 cells (1 × 10⁶) were grown in 100 mm-dishes and incubated for 0.5–16 h in fresh medium containg stimuli as indicated. For the analysis of phosphorylation and degradation of the I-κB, stimulated cells were rinsed twice with ice-cold PBS and then lysed in ice-cold lysis buffer (50 mM Tris-HCl (pH 7.4), containing 150 mM NaCl, 1 % Nonidet P-40, 0.1 % SDS, 0.1 % deoxycholate, 5 mM sodium fluoride, 1 mM sodium orthovanadate, 1 mM 4-nitrophenyl phosphate, 10 μg/Mℓ leupeptin, 10 μg/Mℓ pepstatin A, and 1 mM 4-(2-aminoethyl)benzenesulfonyl fluoride). Cell lysates were centrifuged at 15,000 rpm for 20 min at 4°C, and the supernatant was mixed with a one-fourth volume of 4 × SDS sample buffer, boiled for 5 min, and then separated through a 12 % SDS-PAGE gel. After electrophoresis, proteins were transferred to a nylon membrane by means of Trans-Blot SD semidry transfer cell (Bio-Rad, Hercules, CA). The membrane was blocked in 5 % skim milk (1 h), rinsed, and incubated with primary antibody (for phosphorylated MAPKs or I-κB) in TBST and 3 % skim milk overnight at 4°C. Excess primary antibody was then removed by washing the membrane four times in TBST, and the membrane was incubated with 0.1 μg/Mℓ peroxidase-labeled secondary antibody (against rabbit) for 1 h. Following three washes in TBST, bands were visualized by ECL Western blotting detection reagents and exposed to x-ray film.

Differences among groups were analyzed using one-way analysis of variance combined with the Bonferroni test. All values were expressed as means ± standard deviations, and differences were considered significant at p < 0.05.

We examined whether chelation of iron from immortalized human oral keratinocyte (IHOK) and oral squamous cell carcinoma cells (HN12) was sufficient to induce a signal that would increase the production IL-8. We found that IL-8 secretion was elevated upon exposure to DFO in immortalized IHOK cells, but no significant change in IL-8 concentration was seen in HN12 cells (Fig. 1A). The effect of DFO was concentration-dependent in the range of 0 to 2 mM; however, higher concentrations of DFO did not increase the production of IL-8. Maximal IL-8 production was achieved using 0.75 mM DFO in IHOK cells (Fig. 1A).

MIM is an iron chelator that is structurally distinct from DFO. MIM also induced IL-8 secretion in IHOK cells, but again, no significant change was observed in HN12 cells (Fig. 1B). Conversely, the addition of FC (Fe³⁺, 0.5 mM) significantly prevented DFO-induced IL-8 production (Fig. 1B), indicating that the target of DFO is specific for intracellular iron in IHOK cells. As shown in Fig. 1C, the IL-8 concentrations induced by DFO were comparable to those induced by IL-1β (10 ng/ml) and TNF-α (10 ng/ml), and combining DFO with IL-1β had an additive effect on IL-8 secretion in IHOK cells. Conversely, IL-8 secretion did not differ significantly between the DFO and the DFO plus cytokine treatments in HN12 cells (Fig. 1D).

We found that DFO induced IL-8 secretion in IHOK and HN12 cells in a time-dependent manner (Fig. 2), and that maximal induction occurred after 16 h of incubation (Fig. 2A and 2C). This is different from that observed in IL-1β elicited IL-8 secretion, where a plateau of IL-8 reached within 24 h after IL-1β treatment in IHOK cells. We also found that the increase of IL-8 protein secretion appears to correspond to increased IL-8 mRNA levels in IHOK cells (Fig. 2B).

The activation of NF-κB is usually associated with the phosphorylation of IκB-α. NF-κB is then released and translocated to the nucleus to activate targeted gene expression and the phosphorylated IκB-α is degraded by a proteasome. We examined the effect of DFO on NF-κB activation by measuring I-κBα and pI-κBα degradation (Fig. 3), and DNA binding of NF-κB (Fig. 4). To determine the levels of I-κBα and pI-κBα, IHOK and HN12 cells were incubated with 1.0 mM DFO for 0.5 to 16 h. Cell lysates were prepared to analyze the degradation and phosphorylation of I-κBα by Western blot analysis using anti-I-κBα and pI-κBα antibodies. As shown in Fig. 3, IL-1α treatment induced significant I-κBα degradation in 1 h whereas significant DFO-induced I-κBα degradation was only seen after 16 h in IHOK and HN12 cells.

Given that NF-κB activity is dependent on a steady concentration of I-κB, we investigated the effect of DFO on I-κB phosphorylation. Phosphorylated I-κB was detected 16 h after DFO stimulation in IHOK and HN12 cells (Fig. 3A and 3B). This demonstrated that the dissociation of NF-κB from I-κB plays a crucial role in NF-κB signal transduction, and that NF-κB was translocated to the nucleus when DFO was added to IHOK and HN12 cells.

It has been demonstrated that NF-κB is the central regulator of IL-8 gene expression (32). We therefore examined the effect of DFO on the activation of NF-κB using the electrophoretic mobility shift assay (EMSA) to analyze the NF-κB binding activity of extracted proteins. The nuclear extracts were isolated from DFO-treated IHOK and HN12 cells, and NF-κB oligomers were used as probes. We observed that DFO-treated IHOK and HN12 cells showed increased DNA binding of NF-κB (Fig. 4A), and that maximum binding activity was detected after 16 h of incubation (Fig. 4B). These results indicate that an NF-κB dependent pathway regulates iron chelator-mediated IL-8 production in IHOK and HN12 cells.

Previous studies have shown that three MAP kinase subfamilies contribute to cell survival and the induction of apoptosis in some cell systems 33343536. Furthermore, the activation of p38 and ERK MAP kinase was observed during iron deprivation-induced apoptosis in immortalized and malignant oral keratinocytes 28. To examine whether the activation of p38 and ERK1/2 by DFO treatment contributes to the stimulatory effect of DFO on IL-8 production, we investigated the effects of pharmacological agents that modulate MAP kinase activities. IHOK and HN12 cells were treated for 48 h with DFO (1.0 mM), PD98059 (20 μM), SB203580 (20 μM), or combinations thereof (Fig. 5). Treatment of IHOK cells with SB203580, the selective inhibitor for p38, blocked DFO-induced IL-8 production whereas the ERK pathway inhibitor PD98059 weakly blocked IL-8 production (Fig. 5A). SB203580 and PD98059 treatment in HN12 cells, however, did not significantly block DFO-induced IL-8 production (Fig. 5C). Furthermore, IL-8 secretion appeared to correspond with the level of IL-8 mRNA accumulation in IHOK and HN12 cells (Fig. 5B and 5D).

The accumulation of mRNA is the result of a balance between mRNA synthesis and degradation 37. Given that our results demonstrated that NF-κB is involved in iron chelator-mediated IL-8 production in IHOK and HN12 cells (Fig. 3), we sought to determine whether MAP kinase activation by DFO is involved in the stabilization of IL-8 mRNA transcripts to augment IL-8 protein secretion. IHOK and HN12 cells were treated with DFO (1.0 mM) for 16 h, followed by treatment with actinomycin D (5 μg/ml), an inhibitor of mRNA synthesis, in the presence or absence of MAP kinase inhibitors (Fig. 6). Total RNA was isolated at various time points, and the remaining IL-8 mRNA was measured by semiquantitative RT-PCR. The inhibitory effect of SB203580 was stronger than that of PD98059 during actinomycin D treatment for 4 h (Fig. 6); however, the effect of MAP kinase inhibitors was weaker in HN12 cells than in IHOK cells. The inhibition of p38 and ERK resulted in the rapid disappearance of IL-8 mRNA compared to actinomycin D-treated control cells, suggesting that the posttranscriptional regulation of the IL-8 gene transcript is dependent on the iron chelator in IHOK and HN12 cells.

We examined whether DFO induces IL-8 production through a variety of different signal transduction mechanisms (Fig. 7). Recent evidence indicates that nitric oxide (NO) influences Fe metabolism 38. We therefore examined whether DFO induces IL-8 production via a NO-dependent mechanism. We initially observed that cells treated with 0.3 mM S-nitrosoglutathione (GSNO) expressed increased levels of IL-8 mRNA. We then observed that IHOK and HN12 cells incubated with 0.3 mM S-nitrosoglutathione (GSNO) in the presence of DFO showed decreased IL-8 mRNA production. These results indicated that the NO implicated in transcriptional activation of the DFO-induced IL-8 gene.

We also examined the involvement of the redox-sensitive NF-κB transcription factor in DFO-mediated up-regulation of IL-8 gene expression by co-incubation with DFO and pyrrolidine dithiocarbamate (PDTC), an antioxidant and potent inhibitor of NF-κB in IHOK cells. We observed that significant inhibition of the DFO-dependent IL-8 induction by PDTC occurred in IHOK cells (Fig. 7A and 7B). Similar results were obtained with wortmannin, an inhibitor of phosphatidylinositol 3-kinase-dependent activation of NAD(P)H oxidase. These findings suggest that NAD(P)H oxidase is the main source of ROS required for DFO-dependent transcriptional activation of the IL-8 gene. Finally, DFO had a negligible effect on IL-8 gene induction by cycloheximide (CHX) de novo protein synthesis in IHOK and HN12 cells.