The rat model of 4NQO-induced oral carcinogenesis was established as described previously (16). A total of 20 male Sprague-Dawley (SD) rats (4 weeks of age and weighing 75–100 g) were divided into 2 groups. In the experimental group (n=14), rats were fed daily with 20 ppm 4NQO (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) solution in their drinking water. At weeks 16–24, the experimental rats were sacrificed, as visible lesions of tongue dysplasia or squamous cell carcinoma had developed. The normal control group (n=6) was fed with distilled water only. The rat tongues were harvested and sagittally cut into 2 halves. One half was fixed in 10% buffered formalin, embedded in paraffin and cut into 4-mm-thick sections for hematoxylin and eosin staining to confirm the pathological diagnosis. The other half was collected and stored in liquid nitrogen. Finally, 6 dysplasia samples (1 mild, 3 moderate and 2 severe) and 8 well-differentiated carcinoma samples were collected. All the animal procedures were conducted in accordance with the Guidelines for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee at Sun Yat-sen University (Guangzhou, China).

Another 22 male SD rats (4 weeks of age and 75–100 g in weight) were divided into 2 groups. In the experimental group (n=16), the rats were fed daily with 20 ppm 4NQO (Sigma-Aldrich; Merck KGaA) solution in their drinking water for 24 weeks. The rats in the normal control group (n=6) were fed with distilled water only. At week 18, the experimental rats were divided into 2 groups as visible lesions of tongue dysplasia had appeared. One group was injected with PX-12 (12 mg/kg) weekly via the tail vein. The other group and the normal control group were injected with an equal amount of saline. The rats were sacrificed at week 24. The tongues were dissected, and a longitudinal mid-lingual incision was made. The specimens were collected and analyzed as described below.

The normal control, dysplasia and squamous cell carcinoma specimens (n=6, 6 and 8, respectively) were digested in 1.8 U/ml dispase II (Roche Applied Science, Penzberg, Germany) at 4°C overnight. The epithelial tissues were enzymatically and mechanically separated from the connective tissues. The samples were detected using iTRAQ-based proteomic analysis at the Beijing Genomics Institute (BGI; Shenzhen, China). The protocol for proteomics analysis was conducted according to the BGI guidelines, as previously described (7). In this experiment, the total proteins of the epithelial tissues from the 3 groups were extracted and digested into peptides, respectively. The desalted peptides were labeled with iTRAQ reagents (SCIEX, Framingham, MA, USA). The normal control, dysplasia and squamous cell carcinoma were labeled with 114, 116 and 119 iTRAQ tags, respectively [normal (N)-114, dysplasia (D)-116 and carcinoma (C)-119]. The LTQ-Orbitrap-Velos hybrid mass spectrometer (Thermo Fisher Scientific, Inc., Waltham, MA, USA) coupled with strong cation exchange chromatography (SCX) and liquid chromatography (Thermo Fisher Scientific, Inc.) was used to analyze the mixed peptides.

All raw data files were searched using Mascot 2.3.02 (Matrix Science Ltd., London, UK) against the Rattus RefSeq database (www.ncbi.nlm.nih.gov/protein) and the Uniprot Human database (www.uniprot.org), containing 29,964 sequences. The following identification parameters were selected: MS/MS ion search; enzyme, trypsin; fragment mass tolerance, ±0.02 Da; mass values, monoisotopic mass; max missed cleavages, 1; peptide mass tolerance, 10 ppm; variable modifications, Gln-> pyro-Glu (N-termQ), oxidation (M) and iTRAQ8plex (Y); fixed modifications, carbamidomethyl (C), iTRAQ8plex (N-term) and iTRAQ8plex (K). The identified proteins and peptides were filtered with confidence corresponding to the 1% FDR. The ratios of 116:114 and 119:114 were calculated. A fold change of ≥1.5 or ≤0.67 and a P-value <0.05 were set as the thresholds for screening differentially expressed proteins.

Specimens were obtained from the oral mucosa of 60 subjects, including patients with OSCC (n=35), patients with oral leukoplakia (n=15) and healthy control individuals (n=10). The human healthy mucosa tissues were obtained from the cheeks and gingiva of patients who underwent orthognathic surgery. The diagnosis was based on clinical appearance and histological analysis. Each subject provided written consent to participate after being informed about the aims and protocol of the research. This study was approved by the Ethics Committee of Guanghua School of Somatology, Sun Yat-sen University.

The CAL33 cell line was originally purchased from the Leibniz Institute DSMZ-German Collection of Microorganisms and Cell Cultures GmbH (Braunschweig, Germany). The HSC6 cell line was obtained from the National Cancer Center Research Institute (Tokyo, Japan). Both cell lines were preserved at the Guangdong Provincial Key Laboratory of Stomatology. The cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) (both from Gibco; Thermo Fisher Scientific, Inc.) and 1% penicillin-streptomycin (Gibco, Grand Island, NY, USA) at 37°C in a 20% O₂ and 5% CO₂ humidified incubator (normoxic conditions). Hypoxic conditions were induced in a tri-gas incubator at 37°C with 1% O₂, 5% CO₂ and 94% N₂. The cells were transferred into the specific incubator where necessary. PX-12 was dissolved in dimethyl sulfoxide (DMSO) at 100 mM as a stock solution. N-acetyl cysteine (NAC) (all from Sigma-Aldrich; Merck KGaA) was diluted to 500 mM by filtering. In this study, the cells were treated with 5 mM NAC for 1 h prior to treatment with PX-12, and 0.03% DMSO was used as a vehicle control.

Cell viability was detected by Cell Counting Kit-8 (CCK-8) assay (Dojindo, Kumamoto, Japan). The HSC6 (5×10³ cells/well) and CAL33 cells (6×10³ cells/well) were seeded into 96-well plates. After 24 h, the cells were treated with various concentrations of PX-12 (0–50 µM) for 24 h. Prior to the detection of the absorbance at 450 nm using a microplate reader (Thermo Fisher Scientific, Inc.) the cells were incubated with CCK-8 reagent (10 µl/well) for 2 h at 37°C. All experiments were performed in triplicate.

The invasion assays were performed using 24-well Transwell units (BD Biosciences, Franklin Lakes, NJ, USA). Matrigel (50 µl) diluted with serum-free DMEM (1:5) was coated on the upper chamber. Subsequently, 8×10⁴ cells in 200 µl serum-free medium with or without PX-12 (0, 10, 20 or 40 µM) were seeded in the upper chamber of the system. In the experiments with NAC, the cells were treated with NAC for 1 h prior to the addition of PX-12. A total of 600 µl DMEM with 10% FBS was added to the bottom wells. The cells were incubated for 24 h. The invading cells on the bottom membrane of the upper chamber were fixed with 4% paraformaldehyde and stained with 0.4% crystal violet. Invading cell numbers were counted in 5 random fields (original magnification, ×100).

For apoptosis assay, the Annexin V-FLUOS/propidium iodide (PI) double-staining apoptosis detection kit (Roche Diagnostics GmbH, Mannheim, Germany) was used. The OSCC cells were treated with various concentrations of PX-12 for 24 h, following pretreatment with or without 5 mM NAC for 1 h. The cells were collected and stained with 5 µl Annexin V-fluorescein isothiocyanate and 5 µl PI, according to the manufacturer’s instructions. The acquisition and analysis of the apoptosis data were performed using a flow cytometer (FACSCalibur; BD Biosciences). Basal apoptosis was determined using the same method in control cells.

TUNEL assays were used to identify the apoptotic cells using the FragEL™ DNA Fragmentation Detection kit (Calbiochem; EMD Chemicals Inc., Gibbstown, NJ, USA) according to the manufacturer’s instructions. Briefly, the tissue sections were deparaffinized, rehydrated and incubated with proteinase K for 15 min. Following treatment with 3% H₂O₂ for 5 min, the sections were incubated with terminal deoxynucleotidyl transferase (TdT) enzyme with TdT buffer and biotin-tagged nucleotides in a humidified chamber at 37°C. Tagged nucleotides were detected using HRP solution. After washing, the sections were stained with diaminobenzidine (DAB) solution and counterstained with hematoxylin. For the evaluation of the slides, 100 tumor or epithelial cells were counted per high-power field (original magnification, ×400).

The intracellular ROS levels were measured using a dichlorofluorescein assay (Beyotime Institute of Biotechnology, Haimen, China). 2,7-Dichlorodihydrofluorescein diacetate (DCFH-DA) was used to evaluate the generation of ROS during oxidative damage. The cells were incubated with 100 µM DCFH-DA for 20 min after being washed 3 times in serum-free medium. Finally, the cells were harvested and DCFH-DA fluorescence was assessed using a flow cytometer (FACSCalibur; BD Biosciences) at 494/525 nm. All experiments were performed in triplicate.

IHC staining was performed according to the manufacturer’s instructions. Tissue sections were incubated with primary antibodies against Trx-1 (1:600, #2429; Cell Signaling Technology, Inc., Danvers, MA, USA), hypoxia-inducible factor (HIF)-1α (1:500, ab1) and Bax (1:100, ab32503) (both from Abcam, Cambridge, MA, USA) overnight at 4°C, after being blocked in normal goat serum for 20 min. Subsequently, the sections were incubated with peroxidase-conjugated goat anti-rabbit secondary antibody (GK600510; Gene Tech, Shanghai, China) for 30 min at room temperature. Finally, the slides were visualized with DAB (R&D Systems, Inc., Minneapolis, MN, USA) for antigen detection and counterstained with hematoxylin. All steps were separated by phosphate-buffered saline (PBS) or PBST washes. The expression of proteins was quantified using a visual grading system based on the extent of staining (percentage of positive cells graded on a scale from 0–3 as follows: 0, <5%; 1, 5–30%; 2, 30–70%; and 3, >70%) and the intensity of staining (graded on a scale from 0–3 as follows: 0, none; 1, weak; 2, moderate; and 3, strong) (17). A total of 5 representative fields at ×400 magnification were evaluated.

The cells were lysed with RIPA buffer supplemented with protease (both from Sigma-Aldrich; Merck KGaA) and phosphatase inhibitors (Roche Applied Science). A BCA protein assay kit (ComWin Biotech Co., Ltd., Beijing, China) was used to measured the concentrations of the lysates. The lysates were then incubated at 99°C for 5 min and mixed with loading buffer (4:1; ComWin Biotech Co., Ltd.). The samples (30 µg/lane) were separated on 10 or 12% SDS-PAGE gels and electrophoretically transferred onto a PVDF membrane (EMD Millipore, Billerica, MA, USA). The membrane was blocked in 5% non-fat milk for 1 h at room temperature and then incubated with primary antibodies against Trx-1 (1:1,000, #2429; Cell Signaling Technology, Inc.), HIF-1α (1:1,000, ab1), Bax (1:1,000, ab32503) (both from Abcam), Bcl-2 (1:2,000, #2870s), β-actin (1:1,000, #4970s), poly(ADP-ribose) polymerase (PARP) and cleaved PARP (1:1,000; Cle-PARP; #9532s) (all from Cell Signaling Technology, Inc.) overnight at 4°C, respectively. Subsequently, the membrane was washed in TBST 3 times and incubated with HRP-conjugated secondary antibody (1:3,000, #7074s; Cell Signaling Technology, Inc.) for 1 h at room temperature. The immunoreactive bands were visualized with an enhanced chemiluminescence detection system (EMD Millipore). Immunoreactive bands were quantified by densitometry with ImageJ 1.48 (National Institutes of Health, Bethesda, MD, USA). Similar results were obtained from 3 independent experiments.

All results shown represent the means ± standard deviation from triplicate experiments performed in a parallel manner, unless otherwise indicated. Statistical analyses were performed using a two-tailed Student’s t-test, Mann-Whitney U test, Fisher’s exact test and one-way ANOVA, where appropriate. A value of P<0.05 was considered to indicate a statistically significant difference.

To explore protein profiles during oral carcinogenesis, a rat model of 4NQO-induced oral carcinogenesis was established (Fig. 1A and B). The epithelia from the normal, dysplasia and squamous cell carcinoma stages were collected for proteomics analysis. In this experiment, proteins were found to be differentially expressed in the specimens at the dysplasia stage (74 upregulated and 35 downregulated) and squamous cell carcinoma stage (119 upregulated and 170 downregulated) compared with the normal group specimens (data not shown). In total, 15 proteins were upregulated in both the dysplasia and carcinoma stages (Table I). Notably, the Trx, Grx and Prx proteins, which have been characterized as ‘guardians’ of the intracellular redox state (18), were indicated to be upregulated during oral carcinogenesis, including Trx-1, Grx-1 and Prx-2. Among these three oxidation-associated proteins, Trx-1 was identified to be the most upregulated protein among the 3 oxidation-associated proteins in the dysplasia stage (Table I).

To confirm the expression pattern of Trx-1 during oral carcinogenesis, IHC was performed. The results demonstrated that Trx-1 expression gradually increased during oral malignant transformation, both in the human samples and the rat specimens. As shown in Fig. 1C, none or weak Trx-1 staining was detected in the human normal oral mucosa (the score of Trx-1 positive staining was 3.08±0.72), while the staining of Trx-1 was particularly prominent in the abundant epithelial cytoplasm and nuclei in the oral leukoplakia (4.68±0.47) and OSCC samples (5.32±0.55). These scores were significantly higher compared with the normal samples (P<0.05). The clinicopathological characteristics of the patients are shown in Table II. Moreover, similar characteristics were observed in the animal model. As shown in Fig. 1D, Trx-1 expression was increased in the dysplasia tissues (5.4±0.33) compared with the control tissues (3.93±0.39; P<0.05), as well as the tongue carcinogenesis tissues (5.91±0.16), depending on the extent and intensity of staining.

In solid tumors, hypoxia is a common feature that leads to oxidative stress and results in redox imbalance (19). It has been demonstrated that hypoxia is involved in the growth and aggressiveness of many types of cancer (20–22). In this study, two OSCC cell lines, CAL33 and HSC6, were cultured under hypoxic conditions for 12, 24 and 48 h. Cell proliferation was significantly increased compared with normoxic conditions, as demonstrated by CCK-8 assays (P<0.05) (Fig. 2A and B). The cell invasive ability was also promoted at 24 h, as demonstrated by Transwell assay (P<0.05) (Fig. 2C). Moreover, Trx-1 expression was detected by western blot analysis. The cells were cultured under hypoxic conditions for 3, 6, 12 and 24 h. Trx-1 expression gradually increased in a time-dependent manner upon hypoxic stimulation, as was the expression of HIF-1α (P<0.05) (Fig. 2D and E).

Subsequently, experiments were carried out to examine the effect of Trx-1 on the phenotypes of OSCC cells under hypoxic conditions. When Trx-1 was specifically inhibited by PX-12 at various concentrations (data not shown), cell viability and the cell invasive capacity gradually decreased in a dose-dependent manner (P<0.05) (Fig. 3A and B). Moreover, the apoptosis of the OSCC cells was measured by flow cytometry. As a result, the percentages of OSCC cells undergoing early apoptosis were increased by treatment with PX-12 in a concentration-dependent manner (P<0.05) (Fig. 3C).

Since the inhibition of the activation of Trx-1 has been hypothesized to affect the redox state of cells (23), the intracellular ROS levels of OSCC cells were detected in the presence of PX-12 (10, 20 or 4 µM) for 24 h, and were markedly increased compared with those of controls under hypoxic conditions (P<0.05) (Fig. 4A). Treatment of the OSCC cells with the antioxidant agent, NAC (5 mM), 1 h prior to the addition of PX-12 caused a significant decrease in ROS levels (P<0.05) (Fig. 4B and C) and markedly prevented cellular apoptosis (P<0.05) (Fig. 4D and E). Furthermore, the addition of NAC also attenuated the inhibitory effects of PX-12 on cell proliferation and invasion (data not shown). These results demonstrated that the effects of PX-12 were ROS-dependent.

In addition, the mechanisms underlying the apoptosis-promoting effects of PX-12 in OSCC cells were assessed by western blot analysis. The levels of apoptosis-related proteins, such as the anti-apoptotic protein, Bcl-2, the pro-apoptotic protein, Bax, and PARP cleavage (Cle-PARP) were examined in the total protein from CAL33 and HSC6 cells. As shown in Fig. 5, the expression of Bax and Cle-PARP was upregulated and the expression of Bcl-2 was downregulated by PX-12 under hypoxic conditions. Additionally, the ratio of Bcl-2/Bax was decreased. However, pretreatment with NAC reversed the changes in Bax, Bcl-2 and Cle-PARP expression, and in the ratio of Bcl-2/Bax. Thus, these results suggested that the inhibition of Trx-1 induced apoptosis via ROS accumulation.

The in vitro study results indicated that Trx-1 was a promising therapeutic target for oral carcinogenesis. Thus, in our in vivo experiments, we randomly subdivided the rats exposed to 4NQO into the PX-12 treatment group (n=8) and the disease-control group (n=8) at week 18, when dysplasia was clearly observed. The rats were administered PX-12 (12 mg/kg) via tail vein injection weekly, from week 18 to the end of the experiment (Fig. 6A). The gross weight and emergence rate of the tumor was measured periodically. Notably, the PX-12-treated rats had a lower cancerization rate (3/8) than the rats in the disease-control group (7/8) (Fig. 6B and C). In addition, the weights of the rats in the PX-12 treatment group (417.63±13.22 g) were higher than those of the rats in the disease-control group (356.38±19.56 g) (P<0.05) (Fig. 6D). Additionally, apoptosis was determined by TUNEL assay and the expression of Bax was detected by IHC. As shown in Fig. 6E, the apoptotic rate of the PX-12 treatment group appeared higher than that of the disease-control group (9.93±1.86 vs. 2.09±0.61%, P<0.05). Similarly, the expression of Bax was increased in the PX-12 treatment group (5.6±0.4, P<0.05). These results demonstrated that Trx-1 may be a preventative and therapeutic target during oral epithelial malignant transformation.