Tempol differently affects cellular redox changes and antioxidant enzymes in various lung-related cells

Tempol (4-hydroxy-2,2,6,6-tetramethylpiperidine-1-oxyl) is a potential redox agent in cells. The present study investigated changes in cellular reactive oxygen species (ROS) and glutathione (GSH) levels and in antioxidant enzymes, in Tempol-treated Calu-6 and A549 lung cancer cells, normal lung WI-38 VA-13 cells, and primary pulmonary fibroblasts. Results demonstrated that Tempol (0.5–4 mM) either increased or decreased general ROS levels in lung cancer and normal cells at 48 h and specifically increased O2•− levels in these cells. In addition, Tempol differentially altered the expression and activity of antioxidant enzymes such as superoxide dismutase, catalase, and thioredoxin reductase1 (TrxR1) in A549, Calu-6, and WI-38 VA-13 cells. In particular, Tempol treatment increased TrxR1 protein levels in these cells. Tempol at 1 mM inhibited the growth of lung cancer and normal cells by about 50% at 48 h but also significantly induced cell death, as evidenced by annexin V-positive cells. Furthermore, down-regulation of TrxR1 by siRNA had some effect on ROS levels as well as cell growth inhibition and death in Tempol-treated or -untreated lung cells. In addition, some doses of Tempol significantly increased the numbers of GSH-depleted cells in both cancer cells and normal cells at 48 h. In conclusion, Tempol differentially increased or decreased levels of ROS and various antioxidant enzymes in lung cancer and normal cells, and induced growth inhibition and death in all lung cells along with an increase in O2•− levels and GSH depletion.

. H 2 O 2 is then processed to O 2 and/or H 2 O by catalase or glutathione (GSH) peroxidase 12 . GSH is an important antioxidant peptide which can protect cells from toxic insults 13 . In addition, thioredoxin (Trx) is a small antioxidant protein (~ 12 kDa) that has redox-active cysteine residues at its active site 14 . The oxidized form of Trx is reduced by NADPH-dependent Trx reductase (TrxR) 14 . While Trx1 and TrxR1 are usually localized in the cytoplasm, Trx2 and TrxR2 are found in mitochondria 14 . The Trx system is involved in cell survival, tumor development, and inflammatory diseases, particularly lung cancer [15][16][17][18] . Oxidative stress due to overproduction of ROS, a lack of antioxidants, or both, can lead to permanent modifications of proteins, lipids, and DNA, leading to cell death and tissue inflammation, consequently resulting in the chronic progression of many diseases including cancer 19,20 . More importantly, oxidative stress and chronic inflammation are associated with each other.
Tempol (4-hydroxy-2,2,6,6-tetramethylpiperidine-1-oxyl) is a synthetic cyclic nitroxide compound that has been commonly utilized as a contrast material in magnetic resonance spectroscopy 21,22 . It undergoes rapid reversible transfer between three forms: nitroxide, hydroxylamine and the oxoammonium cation form 21 . Therefore, Tempol is a potential redox agent that may function as a reductive or oxidative mediator depending on its concentration in the cell 21 . Treatment with Tempol removes a variety of ROS and diminishes oxidation, subsequently protecting cells and tissues from oxidative injuries 21,[23][24][25][26] . Our previous reports have shown that Tempol prevents lung cancer cells and juxtaglomerular cells from undergoing cell death in response to cytotoxic insult 27,28 . Potential antioxidant mechanisms of Tempol have been proposed to explain these beneficial effects. One possibility is that it may act as a mimic of SOD and reduce the formation of • OH either by scavenging O 2 •− or by decreasing the intracellular concentration of Fe (II) 21,23,29 . However, increasing attention on Tempol in the clinical area has prompted more comprehensive studies on its potential toxicity. Abundant evidence suggests that Tempol interrupts ferritin synthesis and Fe metabolism, which could stimulate cell death 30 . The toxicity of Tempol is dose-dependent and preferentially targets cancer cells 31 . It has been reported that Tempol at millimole concentrations inhibits cell growth and induces apoptosis, depending on cell types and origins [31][32][33] . Moreover, high concentrations of, or extended exposure to Tempol increases ROS levels in juxtaglomerular cells 32 , breast cancer cells 34 , and ovarian cancer cells 33,35 . Thus, in order to clarify the different effects of Tempol as pro-oxidant or antioxidant in cells and tissues, further in depth studies need to be undertaken to elucidate its biological mechanisms.
In this study, alterations to cellular redox molecules such as ROS, GSH, and antioxidant enzymes in response to Tempol were examined in A549 and Calu-6 lung cancer cells, as well as in primary normal human PF (HPF) cells and an SV40-transformed normal HPF line (WI-38 VA-13 subclone 2RA). Moreover, it was examined whether down-regulation of Trx system, especially TrxR1 protein affects cell growth and death as well as ROS levels in Tempol-treated or -untreated lung cancer and normal cells.

Materials and methods
Cell culture. Human NSCLC A549 adenocarcinoma cells, SCLC Calu-6 cells, and normal lung WI-38 VA-13 subclone 2RA cells were obtained from the American Type Culture Collection (Manassas, VA, USA). Primary normal HPF cells were purchased from PromoCell GmbH (C-12360, Heidelberg, Germany) and used between passages five and ten. Lung cells were kept in a standard humidified incubator at 37 °C with 5% CO 2 and cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS; Sigma-Aldrich Co., St. Louis, MO, USA) and 1% penicillin-streptomycin (GIBCO BRL, Grand Island, NY, USA). Cells were grown in 100 mm plastic cell culture dishes (BD Falcon, Franklin Lakes, NJ, USA) and harvested with trypsin-EDTA (Gibco BRL).

Reagents.
Tempol was obtained from Sigma-Aldrich Co. and dissolved in methanol (Sigma-Aldrich Co.) at 1 M as a stock solution. 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) was also obtained from Sigma-Aldrich Co. and dissolved in distilled water. SOD and catalase proteins were acquired from Sigma-Aldrich Co. and dissolved in 50 mM potassium phosphate buffer at 4733 U/ml. Activity units of SOD and catalase were defined as the amount of enzyme needed to exhibit 50% dismutation of O 2 •− and to exhibit 50% decomposition of H 2 O 2 to H 2 O and/or O 2 at 25 °C for 10 min, respectively, as previously described 36 . TrxR1 protein was also acquired from Sigma-Aldrich Co. All stock solutions were wrapped in foil and kept at 4 °C or -20 °C. Methanol (0.02%) was used as a vehicle control.

Determination of intracellular ROS and O 2
•− levels. Overall  Western blot analysis. Expression levels of various antioxidant proteins were evaluated via western blotting. Briefly, 5 × 10 6 cells in 100 mm culture dishes (BD Falcon) were incubated with the indicated concentrations of Tempol (0.5-4 mM) for 48 h. Cells were washed with PBS and four volumes of lysis buffer (Intron Biotechnology, Seongnam, Gyeonggi-do, Korea) were added. Total proteins (30 μg) were resolved in 12.5% SDS-PAGE gels and then transferred to Immobilon-P PVDF membranes (Millipore, Billerica, MA, USA) by electroblotting. Membranes were probed with anti-SOD1, anti-catalase, anti-Trx1, anti-TrxR1, and anti-β-actin (Santa Cruz Biotechnology, Santa Cruz, CA, USA). Membranes were incubated with horseradish peroxidase-conjugated secondary antibodies (Santa Cruz Biotechnology). Western blots were developed using an EZ-Western Lumi Pico ECL solution kit (DoGen, Seoul, Korea). All band intensities were analyzed using the ImageJ software program (FujiFilm, Tokyo, Japan).

Measurement of cellular SOD activity.
The activity of cellular SOD enzyme was measured using an SOD assay kit (Sigma-Aldrich Co.), as previously described 37 . In brief, 1 × 10 6 cells in 60 mm culture dishes (BD Falcon) were incubated with 1 mM Tempol for 48 h. Supernatant samples that contained 30 μg protein were used for the measurement of SOD activity. These were added to wells of 96-well microtiter plates (Nunc, Roskilde, Denmark) with enzyme working solution containing a color reagent, and then the plates were incubated at 25 °C for 10 min to produce a water-soluble formazan dye. Formazan crystal formation was measured at 450 nm using a microplate reader (BioTekR Instruments Inc. Winooski, VT). SOD activity is expressed in arbitrary units.

Measurement of cellular catalase activity.
The activity of cellular catalase was measured using a catalase assay kit from Sigma-Aldrich Co., as previously described 37 . In brief, 1 × 10 6 cells in 60 mm culture dishes (BD Falcon) were incubated with 1 mM Tempol for 48 h. Supernatant samples containing 30 μg protein were used for the determination of catalase activity. Samples were added to a microcentrifuge tube with assay buffer (50 mM KH 2 PO 4 /50 mM Na 2 HPO 4 , pH 7.0) and the reaction was commenced by the addition of 200 mM H 2 O 2 solution at 25 °C for 5 min and ended by addition of stop solution. One hundred μl of the above reaction mixture was removed and added to another microcentrifuge tube with the color reagent (150 mM potassium phosphate buffer, pH7.0, containing 0.25 mM 4-aminoantipyrine, 2 mM 3,5-dichloro-2-hydroxybenzenesulfonic acid, and the peroxidase solution), and was incubated at 25 °C for 10 min. The dye was measured at 520 nm using a microplate reader (BioTekR Instruments Inc). Catalase activity is expressed in arbitrary units.

Measurement of cellular TrxR1 activity.
The activity of TrxR1 was assessed using a thioredoxin reductase assay kit according to the manufacturer's instructions (Sigma-Aldrich Co.). In brief, 1 × 10 6 cells in 60 mm culture dishes (BD Falcon) were incubated with 1 mM Tempol for 48 h. Supernatant samples containing 30 μg protein were used for the determination of TrxR1 activity. These were added to wells of 96-well microtiter plates (Nunc) with 5,5'-dithiobis-(2-nitrobenzoic acid) from the kit at 25˚C for 1 h. The optical density of each well was measured at 412 nm using a microplate reader (BioTekR Instruments Inc). TrxR1 activity is expressed in arbitrary units.

Transfection of cells with TrxR1 siRNAs.
Gene silencing of TrxR1 was performed as previously described 39  Cell growth inhibition assay. The effects of Tempol on the growth of lung cells were determined using MTT assays. In brief, 1.5 × 10 4 cells per well were seeded in 96-well microtiter plates (Nunc) for MTT assays. After incubation with Tempol at 1 mM for 48 h, 20 μL of MTT solution (2 mg/mL in PBS) was added to each well. Plates were then incubated at 37 °C for 3-4 h. Medium in each well was removed by pipetting. Then 100-200 μL of DMSO was added to each well to solubilize formazan crystals. Optical density was measured at 570 nm with a microplate reader (BioTekR Instruments Inc).

Detection of apoptosis.
Apoptosis was identified via annexin V-FITC (Life Technologies) as previously described 40  Detection of intracellular GSH levels. Cellular GSH levels were evaluated using fluorescent probe 5-chloromethylfluorescein diacetate (CMFDA, Ex/Em = 522 nm/595 nm; Invitrogen Molecular Probes), as previously described 37 . In brief, 1 × 10 6 cells in 60 mm culture dishes (BD Falcon) were treated with Tempol at indicated concentrations (0.5-4 mM) for 48 h. Then, the cells were washed with PBS and incubated with 5 µM CMFDA at 37 ºC for 30 min. The mean CMF fluorescence intensity was determined using a FAC Star flow cytometer (BD Sciences). CMF negative (−) staining indicated the depletion of GSH content in cells.

Statistical analysis.
The results show the mean of more than two independent experiments (mean ± SD).
The data were analyzed using Instat software (GraphPad Prism 5.0, San Diego, CA, USA). A Student's t-test or one-way analysis of variance with post-hoc analysis using Tukey's multiple comparison test was applied to determine statistical significance, which was defined as p-values of < 0.05.  (Fig. 1C). While Tempol at 0.5 and 1 mM significantly decreased ROS (DCF) levels in primary HPF cells at 48 h by about 50%, 2 mM Tempol had no effect (Fig. 1D). Intracellular O 2 •− (DHE) levels were reduced in A549 cells treated with 0.5 and 1 mM Tempol at 48 h but they were significantly increased at 2 and 4 mM, with 4 mM Tempol increasing by about 500% (Fig. 2A). In Calu-6 cells, all Tempol doses increased O 2

Effects of Tempol on the expression and activity of various antioxidant enzymes in lung cancer and normal cell cells.
Changing ROS levels result from alterations to antioxidant enzyme expression and activity. Thus, it was examined whether Tempol influenced the expression and activity of SOD, catalase, Trx1, and TrxR1 in lung cancer and normal cells at 48 h. As shown in Fig. 3A,B, and Supplementary Fig. 1A, Tempol at 0.5-2 mM did not affect SOD1 protein levels in A549 cells but increased the levels in Calu-6 cells. While 0.5 mM and 1 mM doses of Tempol diminished SOD1 protein levels in WI-38 VA-13 cells, 2 mM slightly increased the protein level ( Fig. 3C and Supplementary Fig. 1A). For catalase, apparent differences in expression levels were  Supplementary Fig. 1B). In addition, Tempol increased Trx1 protein expression in A549 cells and WI-38 VA-13 cells (Fig. 3A,C, and Supplementary  Fig. 1C) but did not affect Trx1 expression in Calu-6 cells ( Fig. 3B and Supplementary Fig. 1C). Meanwhile increases in TrxR1 protein levels in response to Tempol were observed in all of these lung cells ( Fig. 3 and Supplementary Fig. 1D).
Regarding the activity of these enzymes, it was found that the activity of SOD was down-regulated in A549 cells treated with 1 mM Tempol at 48 h while it was up-reregulated in 1 mM Tempol-treated Calu-6 cells (Fig. 4A). At 1 mM it did not affect the activity of SOD in WI-38 VA-13 cells (Fig. 4A). In addition, the activity of catalase was enhanced in Tempol-treated A549 and Calu-6 cells whereas it was significantly decreased in WI-38 VA-13 cells (Fig. 4B). Moreover, 1 mM Tempol significantly decreased the activity of TrxR1 at 48 h in all of these lung cells as compared to control cells (Fig. 4).

Effects of TrxR1 silencing on cell growth, cell death, and ROS levels in Tempol-treated lung cancer and normal cells.
Because treatment with Tempol dose-dependently increased TrxR1 protein levels in A549, Calu-6, and Wi-38 VA-13 cells at 48 h (Fig. 3), it was assumed that changes in TrxR1 were closely related to cell growth and death as well as ROS levels. Hence whether gene silencing of TrxR1 using siRNA knock-down alters these cellular effects was examined. TrxR1 siRNA knock-down efficiently decreased protein levels more than 90% in all cell types at 48 h, as compared to scramble siRNA-transfected cells (Fig. 5A and Supplementary Fig. 2). Transfection of TrxR1 siRNA decreased the growth of A549 cells at 48 h by about 20% compared with control transfected A549 cells whereas cell growth was unaffected in Calu-6 and Wi-38 VA-13 cells (Fig. 5B). Treatment with 1 mM Tempol decreased the growth of scramble siRNA-transfected A549, Calu-6, and Wi-38 VA-13 cells by about 55%, 50%, and 55%, respectively (Fig. 5B). Tempol at 1 mM also reduced the growth of TrxR1 siRNA-transfected A549, Calu-6, and Wi-38 VA-13 cells by about 50%, 40%, and 50%, respectively (Fig. 5B).

Effect of Tempol on GSH levels in lung cancer and normal cells. Changes in GSH levels were
assessed in lung-related cells using CMF fluorescent dye. As shown in Fig. 7A, Tempol at 0.5 and 1 mM did not augment the number of GSH-depleted cells in Calu-6 cells at 48 h, whereas 2 and 4 mM concentrations significantly increased the number of GSH-depleted cells by about 44% and 74%, respectively. Tempol at 0.5-4 mM significantly increased the number of GSH-depleted cells in A549 cells at 48 h, with 1 and 4 mM Tempol increasing numbers by about 20% and 60%, respectively (Fig. 7B). In WI-38 VA-13 cells, 0.5 and 1 mM Tempol did not affect the number of GSH-depleted cells but 2 and 4 mM concentrations significantly augmented the number of GSH-depleted cells by about 28% and 90%, respectively (Fig. 7C). In addition, Tempol significantly induced GSH depletion in primary HPF cells at 48 h, with 0.5, 1, and 3 mM Tempol increasing the depletion by about 19%, 38%, and 80%, respectively (Fig. 7D).

Discussion
Tempol has been shown to protect cells and tissues from oxidative damage 21,[23][24][25][26] . In contrast, it has also been reported that Tempol enhances inflammation and oxidative damage in numerous cell and tissue models 21,23 . In addition, millimole concentrations of Tempol or extended exposure to it increases ROS levels in juxtaglomerular cells 32 , breast cancer cells 34 , and ovarian cancer cells 33,35 . The current study has focused on investigating changes in cellular redox changes and antioxidant enzymes in Tempol-treated lung cancer and normal cells.
Results from the present study showed that ROS (DCF) levels were significantly increased in A549 cells treated with 0.5-4 mM Tempol and in WI-38 VA-13 cells treated with 2 and 4 mM Tempol. However, ROS (DCF) were significantly reduced in Calu-6 cells treated with 0.5-4 mM Tempol and in WI-38 VA-13 cells treated with 0.5 mM Tempol. In addition, Tempol at 0.5 and 1 mM significantly decreased ROS (DCF) in primary HPF cells. These results imply that Tempol can raise or lower ROS (DCF) levels depending on the cell type without significant differences between lung cancer and normal lung cells. Tempol at 1 mM significantly increased the www.nature.com/scientificreports/ cancer and normal cells, death from Tempol was more correlated with changes in O 2 •− levels than ROS (DCF) levels. The apparent diverse effects of Tempol in different conditions and in different cells could be explained by assuming that each cell has a different basal activity or metabolism of mitochondria and a variety of antioxidant enzymes present in the cytoplasm or other organelles.
The Trx system consists of Trx, TrxR, and NADPH and is a fundamental enzymatic complex that preserves cellular redox homeostasis 14 . Trx1 and/or TrxR1 are overexpressed in various pulmonary diseases including cancer 16,[42][43][44] . Hence, it is plausible that the Trx system is a promising target for treatment of lung cancer. Tempol significantly decreased the activity of TrxR1 in A549, Calu-6, and Wi-38 VA-13 cells. However, TrxR1 protein levels were increased in all these lung cells. Thus, the increased activity of TrxR1 in Tempol-treated lung cells is more likely to be attributed to other factors such as Trx and Trx-interacting proteins rather than simple expression levels. In addition, Tempol treatment increased Trx1 protein expression in A549 cells and WI-38 VA-13 cells but not in Calu-6 cells. Further research is needed on the mechanism underlying alterations in the Tempol-mediated Trx system.
Treatment of Tempol at 1-5 mM inhibits cell growth and induces apoptosis in various cancer cells [32][33][34][35]41,45 . Likewise, 1 mM Tempol reduced the growth of scramble siRNA-transfected A549 and Calu-6 lung cancer cells, and Wi-38 VA-13 normal cells by about 55%, 50%, and 55% at 48 h, respectively. Moreover, Tempol at 1 mM also significantly increased the amounts of annexin V-positive cells in scramble siRNA-transfected A549, Calu-6, and Wi-38 VA-13 cells by about 30%, 24%, and 22% at 48 h, respectively. Therefore, Tempol is likely to induce lung cell death via apoptosis and/or necrosis. There was no significant difference in the susceptibility to Tempol among these lung cancer and normal cells. TrxR1 silencing itself did not influence the growth of Calu-6 and Wi-38 VA-13 cells compared with scramble siRNA-transfected control cells but it did inhibit the growth of A549 control cells by about 20%. Transfection of TrxR1 siRNA did not significantly alter the growth inhibition in both Tempoltreated lung cancer and normal cells. Thus, reduced TrxR1 protein level was involved in growth inhibition only in A549 cells among these lung cells and did not affect growth inhibition in these lung cells treated with Tempol. In addition, TrxR1 siRNA alone did not increase the amounts of annexin V-positive cells in A549, Calu-6, and Wi-38 VA-13 control cells. Furthermore, knock-down of TrxR1 slightly decreased cell death in Tempol-treated A549 cells and significantly attenuated cell death in Tempol-treated Wi-38 VA-13 cells. These results suggest that TrxR1 silencing does not have a direct effect on cell survival in Tempol-untreated control lung cells, but has a differential effect on cell survival in various lung cells treated with Tempol.
A549 cells transfected with TrxR1 siRNA showed slightly decreased ROS (DCF) levels at 48 h while transfected Calu-6 cells showed slight increases. TrxR1 siRNA reduced the increased levels of ROS (DCF) in Tempoltreated A549 cells whereas silencing partially recovered the decreased levels of ROS (DCF) in Tempol-treated Calu-6 cells. In Wi-38 VA-13 cells, TrxR1 siRNA did not alter ROS (DCF) levels regardless of 1 mM Tempol treatment. It has been reported that Trx or TrxR1 could regulate the basal cellular metabolism of glycolysis, tricarboxylic acid cycle, and mitochondria to produce adenosine triphosphate (ATP) or NAD(P)H 46 . In addition, there are potentially harmful interactions between Tempol and enzyme complexes, which are inolved in glucose transport, glutamine metabolism and mitochondiral ATP production 21,23,33 . Therefore, further studies are needed on the difference in cellular ROS levels in lung cancer and normal cells following downregulation of TrxR1 and treatment with Tempol.
Depletion of GSH promotes cell death [47][48][49] and this has been reported to occur in Tempol-treated leukemia and cervical cancer cells 41,50 . Similarly, the present study demonstrated an increased proportion of GSH-depleted cells in A549, Calu-6, WI-38 VA-13, and primary HPF cells treated with 2 or 4 mM Tempol. In addition, lower doses of Tempol (0.5 and 1 mM) significantly induced GSH-depleted cells in Calu-6 and primary HPF cells. These results support the notion that the induction of cell death was inversely proportional to GSH content [47][48][49] . However, 1 mM Tempol, which showed cell growth inhibition and cell death in A549 and WI-38 VA-13 cells, did not significantly augment the number of GSH-depleted cells in these cells. As there was little difference in susceptibility to cell death to Tempol between lung cancer and normal cells, the effect of GSH depletion by Tempol was not significantly different between lung cancer and normal cells. According to the present study, GSH content is a crucial factor or indicator of cell death induced by Tempol, but its content alone is not sufficient to accurately predict cell death.
In summary, the present results have shown that Tempol at 0.5-4 mM could either increase or decrease ROS (DCF) levels in lung cancer and normal cells and this drug specifically increased O 2 •− levels and GSH depletion. Tempol at 1 mM inhibited the growth of lung cancer and normal cells and also induced their cell death. There was no significant difference in Tempol sensitivity between lung cancer and normal cells. In addition, Tempol had other effects on the expression and activity of antioxidant enzymes, especially TrxR1 in lung cancer and normal cells. Furthermore, down-regulation of TrxR1 partially affected cell growth and death, as well as ROS (DCF) levels in both Tempol-untreated or -treated cells, depending on the cell type. The current results provide valuable information for understanding the cellular effects of Tempol on lung cancer and normal cells with regard to cell growth inhibition and cell death, as well as redox states (ROS and GSH levels) and various antioxidant pathways.
Ethics declarations. The material in this paper has not been published or is not under active consideration by another journal. The research was conducted in accordance with the declaration of Helsinki.

Data availability
Data collected during the present study are available from the corresponding author upon reasonable request.