3MC, 3-(4,5-dimethylthiazol)-2,5-di-phenyltetrazolium bromide (MTT), S-adenosylmethionine (SAM), S-adenosylhomocysteine (SAH), L-amino acids analytical standard, hypotaurine, taurine, o-phthaldiadehyde, 7-benzo-2-oxa-1,3-diazole-4-sulfonic acid (SBD-F), tris(2-carboxyethyl)phosphine hydrochloride (TCEP), N-(2-mercaptopropionyl)glycine (MPG), GSH, tert-butyl hydroperoxide (t-BHP) and crystal violet were purchased from Sigma Aldrich; Merck KGaA (St. Louis, MO, USA). Anti-cysteine dioxygenase (CDO) antibody was obtained from Abcam (Cambridge, MA, USA). Anti-γ-glutamylcysteine ligase catalytic subunit (GCL) antibody and anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH) antibody were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). Peroxidase-conjugated antibodies against mouse or rabbit IgG were obtained from Thermo Fisher Scientific (Waltham, MA, USA).

The non-tumorigenic mouse liver cell line BNL CL.2 (designated as CL2) and 3MC-transformed BNL 1MEA. 7R.1 cells (designated as 1MEA), which were derived from CL2 cells chronically treated with 3MC, were obtained from the Korea Research Institute of Bioscience and Biotechnology (Daejeon, Korea). Both cell lines were established and characterized by Gold et al (17). The cells were grown in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum, 100 U/ml penicillin and 100 µg/ml streptomycin (GenDEPOT, Barker, TX, USA) at 37°C in a humidified incubator with 5% CO₂.

A cell pellet of 25 mg was transferred to a 4-mm NMR nanotube (Agilent Technologies, Santa Clara, CA, USA). D₂O (25 µl) containing 2 mM 3-(trimethylsilyl) propionic-2,2,3,3-d₄ acid sodium salt (TSP-d₄) was then transferred to a separate NMR nanotube. Samples were measured using high-resolution magic angle spinning (HR-MAS) NMR spectroscopy. All spectra were acquired with a 600.167-MHz Agilent NMR spectrometer equipped with a 4-mm gHX NanoProbe (Agilent Technologies) and a spinning rate of 2,050 Hz. The Carr-Purcell-Meiboom-Gill (CPMG) pulse sequence was used due to the suppression of water and the presence of high molecular mass compounds. The acquisition time was 1.703 msec, the relaxation delay was 1 sec, and 128 transients were collected. All spectra were processed and assigned using Chenomx NMR Suite 7.1 professional with the Chenomx 600 MHz library database (Chenomx, Edmonton, AB, Canada). Spectra were binned from 0.5 to 10 ppm for multivariate statistical analysis. The binning size was 0.001 ppm and water peak (4.5–4.9 ppm) and spinning side band peak were excluded (1.15–1.2, 3.61–2.69, 6.56–7.16 and 8.0–8.4 ppm). The binning data were normalized to total area. Principal component analysis (PCA) was conducted with the SIMCA-P⁺ software package (version 12.0; Umetrics, Umeå, Sweden). R² parameter indicates the explained variation in the data and goodness of fit and Q² parameter represents the predictive power of the model (18).

Washed cells in 100-mm culture plates were harvested 24 h after a medium change by scraping in either 5% perchloric acid to determine SAM, SAH, cysteine and GSH, or in ice-cold methanol to determine the levels of methionine, glutamate, glycine, hypotaurine and taurine. The protein pellet was dissolved in 0.1 M NaOH solution, and the protein concentration was determined using a bicinchoninic acid (BCA) protein assay kit (Thermo Fisher Scientific).

SAM and SAH were analyzed by high performance liquid chromatography (HPLC) (Ultimate 3000™; Thermo Fisher Scientific) equipped with a UV detector (UV-3000; 254 nm; Thermo Fisher Scientific), using the modified method of She et al (19). Separation was accomplished using a Hector T-C18 column (5 µm × 4.6 mm × 100 mm) (Chiral Technology Korea, Daejeon, Korea).

Methione, glutamate, glycine, hypotaurine and taurine were derivatized with o-phthalaldehyde/2-mercaptoethanol and quantified using HPLC with a fluorescence detector (FLD-3100; excitation wavelength, 338 nm; and emission wavelength, 425 nm; Thermo Fisher Scientific) (20,21). They were separated using a Hector T-C18 column (3 µm × 4.6 mm × 100 mm).

Total cysteine and GSH were analyzed using the SBD-F sample derivatization method (22). MPG (50 µl), as an internal standard, was added to the samples (50 µl) and briefly mixed by vortex. Following the addition of 10 µl of a 10% (w/v) TCEP solution, the tubes were incubated at room temperature for 30 min. Subsequently, 90 µl of a 10% (w/v) trichloroacetic acid solution with 1 mM ethylenediaminetetraacetic acid (EDTA) was added to each sample, briefly mixed by vortex and centrifuged (13,000 × g for 10 min). The supernatant (50 µl) was then added to another tube containing 10 µl of 1.55 M NaOH, 125 µl of 0.125 M borate buffer (pH 9.5) with 4 mM EDTA, and 50 µl of 0.1% (w/v) SBD-F in borate buffer (0.125 M with 4 mM EDTA). The samples were incubated at 60°C for 1 h, and a 20 µl aliquot was then injected into the HPLC equipped with a fluorescence detector (excitation wavelength, 385 nm; and emission wavelength, 515 nm). The chromatographic separation was achieved using a Hector M-C18 column (3 µm × 4.6 mm × 150 mm).

Cells were lysed with ice-cold PRO-PREP™ protein extract solution (Intron, Seongnam, Korea), and the protein concentration was quantified using the BCA procedure (Thermo Fisher Scientific). Equal amounts of protein samples were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis using a 10% polyacrylamide gel, and then transferred onto a polyvinylidene difluoride membrane (Millipore, Billerica, MA, USA). The membrane was blocked with 5% skimmed milk in 100 mM Tris-HCl (pH 7.5), 150 mM NaCl and 0.2% Tween-20 (TBST) for 1 h at room temperature. The blots were incubated overnight with primary antibodies diluted in TBST containing 5% milk at 4°C. After three washes with TBST, the blot was incubated with the appropriate horseradish peroxidase-conjugated secondary antibodies. The antigen was detected using a Western Bright ECL HRP substrate kit (Advansta, Menlo Park, CA, USA).

Cells were plated in a 96-well plate and cell viability was determined 24 h after t-BHP treatment by the MTT assay, according to the manufacturer's instructions. Briefly, after incubation with MTT (0.5 mg/ml) for 4 h at 37°C, formazan precipitates formed by mitochondrial dehydrogenases in viable cells were extracted with dimethyl sulfoxide. The absorbance of the converted dye was measured at 540 nm using the Multiskan GO reader (Thermo Fisher Scientific), and the results were expressed as a percentage (%) compared with the vehicle treatment.

Apoptotic and live cells were determined by FACS analysis using Annexin V-FITC staining. The cells were treated with t-BHP for 24 h, and were subsequently harvested, trypsinized, washed once in cold phosphate-buffered saline and suspended in 1X binding buffer. The counted cells were stained in prop-idium iodide and Annexin V-FITC solution (Annexin V-FITC apoptosis detection kit; BD Biosciences, Bedford, MA, USA) at room temperature for 15 min in the dark. The stained cells were analyzed by flow cytometry within 1 h. Apoptotic and live cells were analyzed by Becton Dickinson FACSscan flow cytometer and BD FACSDiva software (BD Biosciences).

All the results are expressed as mean ± standard deviation (n=5) and were analyzed by a two-tailed Student's t-test. The acceptable level of statistical significance was set at P<0.05.

Upon ¹H NMR analysis, pattern recognition using PCA of the NMR spectra in metabolite profiling revealed clear clustering between CL2 and 1MEA cells (Fig. 1A). The PCA loading plots revealed that several metabolites, such as acetate, alanine, creatine, glutamate, glutamine, glycine, sn-glycero-3-phosphocholine, GSH, myoinositol, lactate and o-phosphocholine, differed significantly between the two cell types (Fig. 1B). In detailed spectra analysis, it was identified that acetate (×0.21), choline (×0.72), formate (×0.51), GSH (×0.59), glycerol (×0.38), isoleucine (×0.61), leucine (×0.62), niacinamide (×0.44), o-phosphocholine (×0.06), phenylalanine (×0.63), serine (×0.71), succinate (×0.56), tyrosine (×0.68), uridine (×0.42), valine (×0.70) and sn-glycero-3-phospho-choline (×0.31) were significantly lower in the 1MEA cells compared with the CL2 cells (P<0.05). Alternatively, glutamate (×1.61), glutamine (×6.17), glycine (×1.45), proline (×2.02) and myo-inositol (×1.32) were significantly higher in 1MEA cells compared with CL2 cells (P<0.05; Fig. 1C).

GSH plays a critical role in removing reactive metabolites from the cell, and a disturbance in its synthesis reduces a cell's resistance to oxidative stress. Since there are limited data available showing that 3MC causes oxidative stress, attention was paid to the significantly lower level of intracellular GSH in 1MEA cells compared with that in CL2 cells. Specifically, we focused on the transsulfuration pathway that produces GSH and taurine from the essential amino acid methionine (Fig. 2). To investigate and validate all the major metabolites in the transsulfuation pathway, HPLC with a UV or fluorescence detector was used. The cellular levels of SAM, cysteine, glutamate and glycine were signifi-cantly increased 2.0-, 2.7-, 2.1- and 1.4-fold, respectively, in 1MEA cells compared with CL2 cells (Fig. 3B, D–F). On the contrary, the concentration of SAH was slightly decreased in 1MEA cells (Fig. 3C). The levels of methionine, hypotaurine and taurine did not change (Fig. 3A, H and I), but the GSH levels in 1MEA cells were significantly reduced compared with those in CL2 cells (Fig. 3G). Cysteine is the precursor metabolite used to synthesize both GSH and taurine, with GSH being a tripeptide composed of glutamate, cysteine and glycine. Considering the increased levels of cysteine, glutamate and glycine in 1MEA cells, the decreased levels of GSH cannot be fully explained. To investigate why the GSH levels decreased in 1MEA cells, the expression of CDO and GCL, which are the rate-limiting enzymes in the production of taurine and GSH, respectively, was subsequently investigated. In 1MEA cells, CDO expression decreased to 60% of that in CL2 cells, whereas GCL expression decreased to 11% of that in CL2 cells (Fig. 4). Thus, it is suggested that reduced GCL protein expression contributes to decreased GSH synthesis.

To investigate whether low concentrations of GSH in 1MEA cells affected the rate of cell death by exogenous oxidative stress, the cells were exposed to t-BHP, a producer of reactive oxygen species, with cell viability being examined 24 h post-treatment (Fig. 5). As expected, 1MEA cells were more sensitive to t-BHP treatment, with a 50 µM t-BHP treatment leading to 50% 1MEA cell death rate, while the survival of CL2 cells was not affected. To determine oxidative stress-induced apoptotic cell death, flow cytometry analysis was performed. As shown in Fig. 6, increase of early apoptosis (lower right quadrant) and late apoptosis (upper right quadrant) were clearly observed in t-BHP-treated 1MEA cells compared with CL2 cells.