HaCaT cells were obtained from the American Type Culture Collection (Manassas, VA, USA) and maintained in Dulbecco's modified Eagle's medium (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 10% (v/v) fetal bovine serum (FBS; Gibco; Thermo Fisher Scientific, Inc.) and antibiotics (100 μg/ml gentamycin and 100 μg/ml penicillin-streptomycin; Gibco; Thermo Fisher Scientific, Inc.). A hypoxia chamber was used to create the low-oxygen environment composed of 1% O₂, 5% CO₂ and 94% N₂.

Total protein was isolated from harvested cells (using lysis buffer comprising phenylmethanesulfonyl fluoride, Na₃VO₄ and complement C), suspended in phosphate-buffered saline (PBS). The pellets were re-suspended and sonicated in buffer (Sigma-Aldrich, St. Louis, MO, USA) containing 20 mM Tris, pH 7.5, 1% Triton X-100, 1 mM EDTA, 1 mM ethylene glycol tetraacetic acid (EGTA) as well as protease and phosphatase inhibitors (Sigma-Aldrich). The lysates were subjected to western blot analysis. Total protein from HaCaT cells was isolated by homogenization in cold radioimmunoprecipitation assay buffer (Sigma-Aldrich) containing 50 mM Tris, pH 7.5, 150 mM sodium chloride, 1% NP-40, 0.5% sodium deoxycho-late, 0.1% SDS, 0.1 mM EDTA and 0.1 mM EGTA, as well as the appropriate protease and phosphatase inhibitors. The lysates were centrifuged and the supernatants were subjected to western blot analysis.

Protein concentrations were estimated using a protein assay (Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. Nitrocellulose (NC, Merck Millipore, Milford, MA, USA) membranes were incubated with 5% nonfat milk to block non-specific binding. The membranes were subsequently exposed to antibodies that recognized polyclonal LC3 (cat. no. 4108; Cell Signaling Technology, Inc., Danvers, MA, USA); monoclonal P62 (cat. no. MABC32; Merck Millipore) and monoclonal HIF-1α (cat. no. sc-53546; Santa Cruz Biotechnology, Inc., Dallas, TX, USA). The primary antibodies were incubated at a dilution of 1:1,000 in 5% bovine serum albumin or 5% non-fat milk for 18 h at 4°C. The membranes were exposed to polyclonal goat anti-rabbit immunoglobulin (Ig)G (cat. no. ADI-SAB-300-j; Enzo Life Sciences, Inc., Farmingdale, NY, USA), polyclonal goat anti-mouse IgG (cat. no. ADI-SAB-100-j; EnzoLife Sciences, Inc.) or anti-goat secondary antibodies conjugated with horseradish peroxidase (dilution, 1:10,000 for all) in Tris-buffered saline containing Tween-20 (TBS-T) and 5% nonfat milk for 1 h at room temperature. Electrophoresis was performed using an electrophoresis chamber (Bio-Rad Laboratories, Inc., Hercules, CA, USA). A Westsave Gold enhanced chemiluminescence kit (AbFrontier; Young In Frontier Co., Ltd., Seoul, South Korea) was used, and the signals were detected using a chemiluminescence detection system (Fusion Fx7 version 15.18; Vilber Lourmat, Eberhardzell, Germany) and exposed to X-ray film.

Cell viability was determined by crystal violet staining (Sigma-Aldrich), as described previously (18). Briefly, HaCaT cells were pre-incubated under hypoxic conditions (1% O₂ for 24 h) and exposed to 100–400 ng/ml TRAIL (AbFrontier; Young In Frontier Co., Ltd.) for 6 h. The cells were pre-treated with autophagy inhibitors [200 μM 3-MA (Sigma-Aldrich) or 50 μM CQ (Sigma-Aldrich)] for 3 h and exposed to 200 ng/ml TRAIL for 6 h under hypoxic (1% O₂ for 24 h) or normoxic conditions. Cell viability was calculated based on the relative dye intensity compared with that of the controls.

Cytotoxicity was assessed by the LDH assay using the supernatant and a LDH Cytotoxicity Detection kit (Takara Bio Inc., Tokyo, Japan) according to the manufacturer's instructions. LDH activity was determined by measuring the absorbance at a wavelength of 490 nm using a SpectraMax M Series spectrophotometer.

Values are expressed as the mean ± standard error. Results of different treatments and time courses were analyzed using one-way analysis of variance. Comparisons between two groups were analyzed by two-tailed Student's t-test, analysis of variance and Duncan's multiple range test using the SAS statistical package 9.1 (SAS Institute, Cary, NC, USA)

HaCaT cells were cultured under hypoxic or normoxic conditions and treated with TRAIL (100–400 ng/ml). Morphological examination of the cell population by microscopy (Eclipse TS100; Nikon Corporation, Tokyo, Japan) indicated decreased TRAIL-induced apoptosis under hypoxia as compared with normoxia (Fig. 1A). The influence of hypoxia on TRAIL-induced apoptosis in HaCaT cells was then quantified by crystal violet staining (Fig. 1B). HaCaT cells were responsive to TRAIL treatment (10–60% reduction in cell viability) under normoxic conditions, whereas under hypoxic conditions, TRAIL only had a minor effect (5% reduction in cell viability) on cell viability (Fig. 1C), indicating that hypoxia prevented TRAIL-induced apop-tosis. Consistent with these results, the LDH assay also showed that hypoxia prevented TRAIL-induced apoptosis (Fig. 1D). These results confirmed that hypoxia prevented TRAIL-mediated apoptosis.

HIF-1α is a transcription factor and a major regulator of cell adaptation to low oxygen levels. The present study evaluated HIF-1α protein levels in HaCaT cells under hypoxic and normoxic conditions by western blot analysis. Cells were incubated under hypoxic conditions (1% O₂; 0, 6, 12 or 24 h), and a western blot analysis was performed to determine HIF-1α protein levels (Fig. 2A). Cells displayed increased HIF-1α protein levels under the hypoxic conditions but not under normoxia. It is known that mild hypoxia activates autophagy in keratinocytes and reduces p62 protein levels (12). Autophagy controls cell survival, growth and cellular homeostasis as well as cellular defense. Thus, the present study assessed the autophagy marker, LC3 and the protein, P62 that is cleared by autophagic flux, by western blot analysis (Fig. 2B–D). Decreased p62 protein levels were observed under hypoxic conditions as compared to those under normoxic conditions. By contrast, the autophagic flux marker, LC3-II was increased after 24 h of hypoxia. These results indicated that hypoxia increased autophagic flux.

Autophagy inhibitors were utilized to inhibit autophagic flux in order to examine the protective role of autophagy in TRAIL-induced cell death. Hypoxia prevented TRAIL-induced apoptotic death in cells not treated with autophagy inhibitors. By contrast, treatment with the autophagy inhibitors 3-MA and CQ blocked hypoxic inhibition of TRAIL-induced apoptosis (Fig. 3A). Cell viability and LDH assays confirmed that hypoxia-mediated induction of autophagy protected hypoxic HaCaT cells from TRAIL-induced apoptosis (Fig. 3B and C). These results demonstrate that TRAIL-induced apoptosis was blocked by autophagy inhibitors.

To confirm the inhibition of autophagic flux by the autophagy inhibitors, expression of the autophagy marker, LC3 and the levels of p62 protein were assessed by western blot analysis. Treatment with 3-MA decreased the levels of LC3 in hypoxic HaCaT cells, confirming the inhibition of autophagy; however, treatment with CQ increased the levels of LC3-II (Fig. 4A and B). Autophagy inhibitor-treated cells demonstrated increased p62 protein levels (Fig. 4C and D). As 3-MA inhibits conversion of LC3-I to LC3-II and CQ blocks fusion of autophagolysosomes, treatment with 3-MA decreased the LC3-II levels whereas CQ increased the LC3-II levels. These results confirmed that autophagic flux was inhibited by the autophagy inhibitors, 3-MA and CQ.