A total of 96 NPC patients were selected at The First Hospital of Sichuan Medical University, Luzhou, China. Written informed consent was obtained from the patients, and this program was approved by the Ethics Committee of the First Hospital of Sichuan Medical University.

The tissue specimens from the patients were immunostained with a rabbit polyclonal TIGAR antibody (1:200, Santa Cruz Biotechnology, Santa Cruz, CA, USA), according to the manufacturer's instructions. Two pathologists independently scored each slide. The percentage of positive tumor cells was evaluated (0, 0%; 1, 1–25%; 2, 26–50%; 3, 51–75%; 4, 76–100%), as well as the staining intensity (0, negative; 1, weak; 2, moderate; 3, strong; 4, very strong), as previously described (12). The intensity score × percentage score value was used to obtain the final overall score for TIGAR (0–16).

The human normal nasopharyngeal epithelial cell line NP69-SV40T (Sun Yat-sen University Cancer Center, Guangdong, China), was routinely maintained in keratinocyte serum-free medium supplemented with human recombinant epidermal growth factor (EGF 1–53) and bovine pituitary extract (BPE) (Invitrogen, USA). The human CNE-2 and 5–8F NPC cell lines were obtained from ATCC (American Type Culture Collection), and routinely maintained in RPMI-1640 medium supplemented with 10% fetal bovine serum (GE Healthcare Life Sciences, Logan, UT, USA), 100 U/ml penicillin, and 100 μg/ml streptomycin (Beyotime Biotecnology, China).

The Lenti-shRNA vector system against TIGAR was constructed, packed, and purified by GeneChem (Shanghai, China). The shRNA oligonucleotides were designed as TIGAR-shRNA (GCCAGCTTTACTGGAGAACTT). A scramble sequence was synthesized as control, and tagged as Scramble-shRNA (TTACCGAGACCGTACGTAT). Human NPC cells CNE-2 and 5–8F were infected, and colonies expressing a stable shRNA were selected using puromycin (Sigma-Aldrich, St. Louis, MO, USA), according to the manufacturer's protocol.

For cell growth assay, stable cells were seeded at a density of 5×10⁴ per well. The cells were stained by trypan blue and counted in the following 5 days. For colony formation assay, stable cell lines transfected with TIGAR-shRNA and Scramble-shRNA were seeded in 3.5-mm culture dishes at a density of 200. The colony formation was evaluated under the microscope after 10 days. Next, the cells were fixed with 4% PFA, and stained with 0.1% crystal violet. Data are expressed as the mean ± SD of five independent experiments.

Stable cells were seeded on coverslips in 24-well plate at a density of 2×10⁴ per well. Twenty-four hours later, cells were incubated with EdU (20 μM) for 1.5 h. Then, EdU assay were continued using EdU DNA Proliferation In Vitro Detection kit (RiboBio Co., Ltd., Guangzhou, China), according to the manufacturer's instructions. In total, three fields per slice were randomly selected and analyzed, and experiment was repeated three times. The EdU incorporation rate was calculated as the ratio of the EdU positive cell number to the total cell number in each feld.

Cells at 80–90% confluence were scraped using a sterile micro-pipette tip across the monolayer to obtain a longitudinal scratch without cells, to perform the wound healing assay. The empty area was monitored every 4 h. The Transwell assay was performed using Transwell chambers with polycarbonate filters (8-μm pore size, BD Biosciences, San Jose, CA, USA). The chambers were coated with 50 μl Matrigel prior to cell seeding, and immersed in a well containing 600 μl of complete medium, followed by the addition of 4×10⁴ cells in 100 μl serum-free medium to the upper chamber. After an incubation of 24–48 h at 37°C in a 5% CO₂ incubator, cells were fixed with 4% PFA, and stained with 0.1% crystal violet. The cells remained in the upper chamber were removed with a cotton swab. The migrated cells were counted under a microscope at ×400 magnification.

Stable cell lines were harvested at 48 h after adriamycin (0.5 μg/ml) stimulation, and stained for Annexin V-FITC and PI at room temperature for 15 min according to the manufacturer's instructions (Invitrogen). Cells were analyzed by flow cytometry (FACScan, Becton-Dickinson).

Eight-week-old male nude mice (BALB/c-nu) were bought from Beijing HFK Bioscience Co. Ltd., Beijing, China. The cells expressing TIGAR-shRNA or Scramble-shRNA were collected and resuspended in PBS at the density of 10⁷/ml, and 100 μl (10⁶ cells) were subcutaneously inoculated in the flank of each mouse. Tumor volumes were monitored every 3 days and the length, width and height were measured to evaluate the tumor volume through the following formula: Tumor volume (mm³) = 0.5 × length × width × height (13). After the designated days, mice were sacrificed by carbon dioxide asphyxiation, and the tumors were removed for analysis. The animal experiments were approved by the Ethics Committee of Luzhou Medical College, Luzhou, China.

Total cell proteins were extracted with RIPA lysis buffer with cocktail of protease inhibitors (Beyotime Biotecnology). The samples were resolved on a SDS-PAGE gel and then transferred to a PVDF membrane (Amersham Biosciences, Fairfield, CT, USA). The membranes were labeled with the following antibodies of the proteins of interest: TIGAR (sc-166290; Santa Cruz Biotechnology); Caspase-3 (19677-1-AP; Proteintech, Wuhan, China); p65 (AN365, Beyotime Biotecnology); IkB-α (AI096, Beyotime Biotecnology); Bcl-2 (12789-1-AP, Proteintech); MMP-2 (10373-2-AP, Proteintech); MMP-9 (10375-2-AP, Proteintech); Oct-1 (10387-1-AP, Proteintech); GAPDH (2118; Cell Signaling, Beverly, MA, USA). Secondary antibodies were purchased from Proteintech. ECL was used (Amersham Biosciences) for the detection of the target bands.

The experiments were performed at least in triplicate. The results were expressed as the mean ± SD. Statistical analysis was performed by SPSS software and GraphPad Prism. P<0.05 was considered statistically significant.

TIGAR expression was evaluated in NPC tissue specimens by immunohistochemistry (IHC). TIGAR staining intensity in NPC cells was significantly stronger compared to the staining in the adjacent normal epithelial cells (Fig. 1A). Next, the immunostaining of the 96 slides from the NPC patients was scored, showing a higher overall score in the tumor tissues compared with the normal nasopharyngeal epithelium (Fig. 1B). Among these slides, 31 specimens with both the NPC tissues and the adjacent normal epithelium were selected, showing a similar result as that shown in Fig. 1B (Fig. 1C). The remarkably elevated protein level in primary NPC tissues was confirmed using immunoblotting (Fig. 1D). A similar result was also observed in NPC cells (CNE-2, 5–8F) and normal nasopharyngeal epithelial cells (NP69) (Fig. 1E).

To explore the biological function of TIGAR in NPC cells, lentivirus was used to introduce shRNA (Lenti-shRNA) targeting TIGAR into CNE-2 and 5–8F cells (Fig. 2). The western blot analysis showed a remarkable reduction of TIGAR level in TIGAR-shRNA cells compared with TIGAR expression in Scramble-shRNA cells (Fig. 2A). ROS accumulation and GSH/GSSG reduction also confirmed the high depletion efficiency of TIGAR (data not shown). Cell growth assays revealed a reduced growth in TIGAR-shRNA cells compared to the growth of the Scramble-shRNA cells (Fig. 2B). EdU incorporation confirmed the remarkable reduction of the proliferation in the TIGAR-shRNA cells (Fig. 2C). The number of colonies in TIGAR-shRNA cells was apparently lower than the number in the Scramble-shRNA cells (Fig. 2D and E). In order to investigate whether TIGAR was involved in migration, wound healing assay and Transwell migration assay were performed, showing a significant decreased migration ability in TIGAR-shRNA cells (Fig. 2F and G) as well as a reduced invasion ability, as shown by Matrigel invasion assay (Fig. 2H).

To evaluate whether TIGAR depletion affected apoptosis in NPC cells, stable NPC cells were stained with Annexin V/PI and analyzed by flow cytometry (Fig. 3). The results indicated that the apoptotic rate was increased in TIGAR-shRNA cells compared to the rate in the Scramble-shRNA cells (Fig. 3A and B). To test whether TIGAR depletion can improve the adriamycin related apoptosis, cells were incubated with 0.5 μg/ml adriamycin for 48 h, and analyzed by flow cytometry. As shown in Fig. 3A and B, cell apoptosis was remarkably induced in TIGAR-shRNA cells compared with apoptosis in the Scramble-shRNA cells. We further examined the anti-apoptotic protein BCL2, which was downregulated as we expected, while cleaved caspase-3 was upregulated (Fig. 3C).

To evaluate the effect of TIGAR overexpression on NPC cells, we constructed stable cells overexpressing TIGAR (Fig. 4A). These cells exhibited an increased growth rate, colony formation, and enhanced migration and invasion (Fig. 4B–F). However, TIGAR overexpression did not reduce apoptosis (Fig. 4G).

To evaluate whether knockdown of TIGAR reduced tumor growth in vivo, NPC stable cells were subcutaneously inoculated into the flank of nude mice. As shown in Fig. 5, the tumor growth rate was decreased in TIGAR-shRNA group compared with the corresponding Scramble-shRNA group. In addition, the tumor weight in the TIGAR-shRNA group was one-half of the tumor weight in the Scramble-shRNA group (Fig. 5C and F). On the contrary, TIGAR overexpression accelerated xenograft tumor growth in nude mice (Fig. 5G and H).

The IκB-α and p65 were analyzed to evaluate whether NF-κB pathway was involved in TIGAR-regulated apoptosis in NPC cells. Our results showed increased expression of IκB-α, and inhibited translocation of p65 into the nucleus in both CEN-2 and 5–8F TIGAR-shRNA cells, indicating an inhibited NF-κB pathway (Fig. 6A and B). Immunofluorescence staining of p65 confirmed the reduced level of p65 in the nucleus (Fig. 6C). Next, NF-κB target genes matrix metalloproteinase- 2 (MMP-2) and MMP-9 were examed for a significant reduction in TIGAR-shRNA cells (Fig. 6B).