Shikonin targets cytosolic thioredoxin reductase to induce ROS-mediated apoptosis in human promyelocytic leukemia HL-60 cells

Free Radical Biology and Medicine 70 (2014) 182–193

Contents lists available at ScienceDirect

Free Radical Biology and Medicine journal homepage: www.elsevier.com/locate/freeradbiomed

Original Contribution

Shikonin targets cytosolic thioredoxin reductase to induce ROS-mediated apoptosis in human promyelocytic leukemia HL-60 cells Dongzhu Duan, Baoxin Zhang, Juan Yao, Yaping Liu, Jianguo Fang n State Key Laboratory of Applied Organic Chemistry and College of Chemistry and Chemical Engineering, Lanzhou University, Lanzhou, Gansu 730000, China

art ic l e i nf o

a b s t r a c t

Article history: Received 23 September 2013 Received in revised form 13 February 2014 Accepted 18 February 2014 Available online 26 February 2014

Shikonin, a major active component of the Chinese herbal plant Lithospermum erythrorhizon, has been applied for centuries in traditional Chinese medicine. Although shikonin demonstrates potent anticancer efficacy in numerous types of human cancer cells, the cellular targets of shikonin have not been fully defined. We report here that shikonin may interact with the cytosolic thioredoxin reductase (TrxR1), an important selenocysteine (Sec)-containing antioxidant enzyme with a C-terminal -Gly-Cys-Sec-Gly active site, to induce reactive oxygen species (ROS)-mediated apoptosis in human promyelocytic leukemia HL-60 cells. Shikonin primarily targets the Sec residue in TrxR1 to inhibit its physiological function, but further shifts the enzyme to an NADPH oxidase to generate superoxide anions, which leads to accumulation of ROS and collapse of the intracellular redox balance. Importantly, overexpression of functional TrxR1 attenuates the cytotoxicity of shikonin, whereas knockdown of TrxR1 sensitizes cells to shikonin treatment. Targeting TrxR1 with shikonin thus discloses a previously unrecognized mechanism underlying the biological activity of shikonin and provides an in-depth insight into the action of shikonin in the treatment of cancer. & 2014 Elsevier Inc. All rights reserved.

Keywords: Thioredoxin reductase Shikonin Reactive oxygen species Redox Apoptosis Free radicals

Natural products, the major sources of chemical diversity driving pharmaceutical discovery over the past century, have served humankind in the treatment of various diseases for centuries and continue to provide an indispensable source of bioactive lead compounds for drug discovery [1,2]. Natural products and their modified derivatives are an important fountain for a large fraction of the current pharmacopoeia. It is roughly estimated that half of modern marketed drugs originate from natural products. In the case of anticancer and anti-infective agents, this proportion is even higher [1,3]. Thus, the past decades have witnessed increasing interest in identifying the active chemical components and their cellular targets, which leads to discovering numerous new therapeutic agents, such as vinblastine, rapamycin, paclitaxel, roscovitin, camptothecin, and homoharringtonine. Quinones are a class of highly reactive organic compounds whose chemical structures allow them to interact with biological targets by forming covalent bonds and/or by acting as electron transfer agents in oxidation–reduction reactions [4,5]. These unique characteristics make the quinone moiety a core motif in a variety of therapeutic agents, such as the vitamin K family, mitomycin C, doxorubicin, and emodin. Shikonin, a naturally occurring naphthoquinone isolated from the Chinese herbal plant Lithospermum erythrorhizon, has been used for thousands of years in traditional Chinese medicine for the

n

Corresponding author. Fax: þ86 931 8915557. E-mail address: [email protected] (J. Fang).

http://dx.doi.org/10.1016/j.freeradbiomed.2014.02.016 0891-5849 & 2014 Elsevier Inc. All rights reserved.

treatment of diverse ailments [6,7]. Recent studies have demonstrated that shikonin has potent antitumor potential, inhibiting malignant cell growth and inducing cancer cell death. A number of molecular targets in various types of cells, such as NF-κB [8], proteasome [9], mitogen-activated protein kinase family [10,11], phosphatase and tensin homolog deleted on chromosome 10 (PTEN) and tyrosine phosphatases [12], β-catenin [13], p53 [14], Akt/ASK1/ p38 [15], tyrosine kinase [16], steroid sulfatase [17], and DNA topoisomerase [18], have been reported to be affected by shikonin. Despite its undoubted anticancer efficacy, the molecular mechanism underlying the action of shikonin is still elusive, and the primary cellular target and mode of action of this molecule remain unclear. Thioredoxin reductase (TrxR), thioredoxin (Trx), and NADPH comprise a highly conserved, ubiquitous system (the thioredoxin system), which plays a crucial role in maintaining intracellular redox homeostasis [19–22]. Two major isoforms of TrxR/Trx are present in various intracellular organelles: TrxR1/Trx1 are predominantly in the cytosol and nucleus, whereas TrxR2/Trx2 are mainly localized within mitochondria [23]. Despite the different localizations of the isoforms within cells, mammalian TrxR1 and TrxR2 have similar structures and share the same catalytic mechanism. TrxRs catalyze the NADPH-dependent reduction of the disulfide in Trxs, which serve a wide range of functions in cellular redox signaling. Mammalian TrxRs, compared to those from bacteria, are large selenium-containing proteins [24–26]. Owing to the high reactivity of selenide that is present within

D. Duan et al. / Free Radical Biology and Medicine 70 (2014) 182–193

the C-terminus of reduced TrxRs, the enzymes have broad substrate specificity and are easily inactivated by various alkylating agents [19,21]. TrxR1 is often overexpressed in many cancer cells and targeting its ablation leads to a reduction in tumor progression and metastasis [27], making this selenoenzyme a promising target for development of novel anticancer agents [19,28,29]. In this study, we discovered that shikonin may interact with TrxR1 both in vitro and in human promyelocytic leukemia HL-60 cells. Shikonin primarily targets the Sec residue in the antioxidant enzyme TrxR1 to inhibit its Trx-reduction activity, but further elicits a new function of generating reactive oxygen species (ROS). Accumulation of ROS disrupts the intracellular redox balance and eventually induces apoptosis in HL-60 cells. Overexpression of a functional TrxR1 in cells attenuates the cytotoxicity of shikonin, whereas knockdown of the enzyme in cells enhances the cytotoxicity of shikonin. Targeting TrxR1 by shikonin thus reveals an unprecedented mechanism underlying the biological action of shikonin and sheds deep light on the potential application of shikonin in the treatment of cancer.

Materials and methods Chemicals and enzymes The recombinant rat TrxR1 was essentially prepared as described [30] and was a gift from Professor Arne Holmgren at Karolinska Institute, Sweden. The recombinant U498C TrxR1 mutant (Sec-Cys) was produced as described [24]. Proteins were pure as judged by Coomassie-stained SDS–polyacrylamide gel electrophoresis (PAGE), and the recombinant TrxR1 had a specific activity of 50% of the wild-type TrxR1 with the 5,5'-dithiobis (2-nitrobenzoic acid) (DTNB) assay. Escherichia coli Trx was purchased from IMCO (Stockholm, Sweden, www.imcocorp.se). RPMI 1640 medium, Dulbecco's modified Eagle's medium (DMEM), G418 disulfate salt, puromycin, N-acetyl-L-cysteine (NAC), bovine insulin, L-buthionine-(S,R)-sulfoximine (BSO), N-acetyl-Asp-GluVal-Asp-p-nitroanilide (Ac-DEVD-pNA), reduced and oxidized glutathione (GSH and GSSG), dimethyl sulfoxide (DMSO), yeast glutathione reductase (GR), superoxide dismutase (SOD), trypan blue, 2-vinylpyridine, Hoechst 33342, and 2',7'-dichlorofluorescein diacetate (DCFH-DA) were obtained from Sigma–Aldrich (St. Louis, MO, USA). Cytochrome c was obtained from Sangon Biotech (Shanghai, China). NADPH was obtained from Roche (Mannheim, Germany). Sephadex G-25, fetal bovine serum (FBS), anti-TrxR1 antibody, and DTNB were purchased from GE Healthcare Life Sciences, Sijiqing (Hangzhou, China), Santa Cruz Biotechnology (Santa Cruz, CA, USA), and J&K Scientific (Beijing, China), respectively. 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), penicillin, and streptomycin were obtained from Amresco (Solon, OH, USA). Bovine serum albumin, phenylmethylsulfonyl fluoride (PMSF), sodium orthovanadate(V) (Na3VO4), and anti-actin antibody were obtained from Beyotime (Nantong, China). Shikonin was obtained from Pufei De Biotechnology (Chengdu, China). A 50 mM solution of shikonin was prepared in DMSO and stored at  20 1C. All other reagents were of analytical grade. Cell cultures A549, HeLa, HepG2, HL-60, L02, and HEK 293 T cells were obtained from the Shanghai Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences. HL-60 cells were cultured in RPMI 1640 supplemented with 10% FBS, 2 mM glutamine, and 100 units/ml penicillin/streptomycin and maintained in an atmosphere of 5% CO2 at 37 1C. A549, HeLa, HepG2, HL-60, L02, and HEK 293 T

183

cells were cultured in DMEM with 10% FBS under the same conditions. HEK-TrxR1 and HEK-IRES cells, kindly provided by Professor Constantinos Koumenis from the University of Pennsylvania School of Medicine [31,32], were cultured in DMEM with 10% FBS, 2 mM glutamine, 100 units/ml penicillin/streptomycin, 0.1 μM sodium selenite, and 0.4 mg/ml G418 and maintained at 37 1C in a humidified atmosphere of 5% CO2. Cytotoxicity assay MTT assay Unless otherwise noted, 2  104 cells were incubated with shikonin or other agents in triplicate in a 96-well plate for the indicated times at 37 1C in a final volume of 100 μl. Cells treated with DMSO alone were used as controls. At the end of the treatment, 10 μl MTT (5 mg/ml) was added to each well and incubated for an additional 4 h at 37 1C. An extraction buffer (100 μl, 10% SDS, 5% isobutanol, 0.1% HCl) was added, and the cells were incubated overnight at 37 1C. The absorbance was measured at 570 nm using a microplate reader (Thermo Scientific Multiskan GO, Finland). Trypan blue exclusion assay HL-60 cells were seeded at 2  104 cells per well in 24-well plates and treated with various concentrations of NAC and shikonin (1, 2, or 5 μM) for 24 h. Cells treated with DMSO alone were used as controls, and cell viability was determined by the trypan blue exclusion assay. After treatment, the cells were stained with trypan blue (0.4%, w/v), and the numbers of viable (unstained) and dead (stained) cells were determined under a microscope. Generation of stable TrxR1 knockdown cells The short hairpin RNA (shRNA) plasmid targeting coding regions of the TrxR1 gene (shTrxR1) and the control nontargeting shRNA (shNT) were kindly provided by Professor Constantinos Koumenis from the University of Pennsylvania School of Medicine [31]. HeLa cells were plated in a six-well plate at 3  105 cells/well in DMEM without antibiotics overnight and were transfected with either shTrxR1 or shNT plasmid using GeneTran III transfection reagent (Biomiga, CA, USA). After 48 h of transfection, the cells were maintained in DMEM, 10% FBS, 2 mM glutamine, 100 units/ ml penicillin/streptomycin at 37 1C in a humidified atmosphere of 5% CO2 and selected by supplementation with 1 μg/ml puromycin. In vitro TrxR activity assays [33,34] DTNB assay The TrxR activity was determined at room temperature using a microplate reader. The NADPH-reduced TrxR (170 nM) or U498C TrxR (700 nM) was incubated with various concentrations of shikonin for the indicated times at room temperature (the final volume of the reaction mixture was 50 μl) in a 96-well plate. A master mixture of TE buffer (50 mM Tris–HCl, pH 7.5, 1 mM EDTA, 50 μl) containing DTNB and NADPH was added (final concentrations: 2 mM and 200 μM, respectively), and the linear increase in absorbance at 412 nm during the initial 3 min was recorded. The same amounts of DMSO (0.1%, v/v) were added to the control experiments and the activity was expressed as the percentage of the control. Endpoint insulin reduction assay The NADPH-reduced TrxR (170 nM) was incubated with various concentrations of shikonin for 2 h at room temperature in a final

184

D. Duan et al. / Free Radical Biology and Medicine 70 (2014) 182–193

volume of 50 μl. A master mixture of TE buffer (50 μl) containing 4 μM E. coli Trx, 0.4 mM NADPH, and 0.32 mM insulin was added to the solution and the incubation continued at room temperature for 0.5 h. The reaction was terminated by the addition of 100 μl of 1 mM DTNB in 6 M guanidine hydrochloride (pH 8.0) and the absorbance at 412 nm was measured using a microplate reader. The same amounts of DMSO (0.05%, v/v) were added to the control experiments and the activity was expressed as a percentage of the control. GR assay The NADPH-reduced GR (0.25 unit/ml) in TE buffer was incubated with various concentrations of shikonin for 0.5 h in a 96-well plate at room temperature in a total volume of 100 μl. Reactions were initiated by the addition of GSSG and NADPH (50 μl, final concentrations: 1 mM and 400 μM, respectively). The GR activity was determined by measuring the decrease in absorbance at 340 nm during the initial 3 min. The same amounts of DMSO (0.1%, v/v) were added to the control experiments and the activity was expressed as a percentage of the control. Determination of TrxR activity in cells [33,34] After cells were treated with various concentrations of shikonin for 6 h or 12 h, the cells were harvested and washed twice with phosphate-buffered saline (PBS). Total cellular proteins were extracted with RIPA buffer (50 mM Tris–HCl, pH 7.5, 2 mM EDTA, 0.5% deoxycholate, 150 mM NaCl, 1% Triton X-100, 0.1% SDS, 1 mM Na3VO4 and 1 mM PMSF) for 30 min on ice with occasional vortexing. The total protein content was quantified using the Bradford procedure. TrxR activity in cell lysates was measured by the endpoint insulin reduction assay. Briefly, the cell extract containing 20 μg of total proteins was incubated in a final reaction volume of 50 μl containing 100 mM Tris–HCl (pH 7.6), 0.3 mM insulin, 660 μM NADPH, 3 mM EDTA, and 15 μM E. coli Trx for 30 min at 37 1C. The reaction was terminated by adding 200 μl of 1 mM DTNB in 6 M guanidine hydrochloride, pH 8.0. A blank sample, containing everything except Trx, was treated in the same manner. The absorbance at 412 nm was measured, and the blank value was subtracted from the corresponding absorbance value of the sample. The same amounts of DMSO (0.02%, v/v) were added to the control experiments and the activity was expressed as the rate of thiol production (nM/min/mg total protein). Measurement of total glutathione and oxidized glutathione Quantification of the total glutathione and GSSG was based on the enzymatic recycling method [35]. The cells (70–80% confluence) were treated with various concentrations of shikonin or NAC for 24 h. Then cells were collected and resuspended using ice-cold extraction buffer containing 0.1% Triton X-100 and 0.6% sulfosalicyclic acid in 0.1 M potassium phosphate buffer with 5 mM EDTA disodium salt, pH 7.5 (KPE buffer). After sonication of the suspension in ice water for 2–3 min with vortexing every 30 s, the solution was centrifuged at 3000g for 4 min at 4 1C, and the supernatant was immediately transferred. For assay of the total GSH, a 120-μl solution containing 0.33 mg/ml DTNB and 1.66 units/ml glutathione reductase was added to each sample (20 μl). Then NADPH (60 μl of 0.66 mg/ml) was added and the absorbance at 412 nm was immediately measured every 10 s for 2 min. The same amounts of DMSO were added to the control experiments. GSSG was measured after GSH derivation by 2vinylpyridine. Two microliters of 2-vinylpyridine was added to 100 μl of cell supernatant and mixed. The reaction was allowed to take place for 1 h at room temperature in a fume hood. Then 6 μl

of triethanolamine was added to the supernatant and the solution was mixed vigorously. Measurement of GSSG was performed as described above for total GSH. Assessment of the intracellular ROS [36] Cells were plated in 12-well plates and were treated with 2 or 5 μM shikonin in the presence or absence of 5 mM NAC for 1 h. After removal of the medium, the ROS indicator DCFH-DA (10 μM) in fresh FBS-free medium was added and incubation continued for 30 min at 37 1C in the dark. The cells were visualized and photographed under a Leica inverted fluorescence microscopy. Induction of NADPH oxidase activity and production of superoxide anion by shikonin-modified enzyme [34,36] The NADPH-reduced TrxR (1.3 μM) was incubated with 100 μM shikonin at room temperature for 2 h in TE buffer. The remaining enzyme activity was less than 10% of the control, as monitored by the DTNB reduction assay. After incubation, the unreacted shikonin was removed by a Sephadex G-25 desalting column. To determine the NADPH oxidase activity, 44 μl of modified enzyme was added to 256 μl of TE buffer containing 200 μM NADPH. Oxidation of NADPH was followed at 340 nm using a molar extinction coefficient of 6200 M  1 cm  1 in calculations. The production of superoxide anion was determined by the cytochrome c reduction assay. Briefly, to the above reaction mixture was added 34 μl of 0.82 mM cytochrome c, and the absorbance spectra from 500 to 650 nm were monitored. After the indicated times, SOD was added to reach a final amount of 300 U. The difference between the increment of the absorbance at 550 nm with and without SOD was calculated to quantify the superoxide anion production using a molar extinction of 21,000 M  1 cm  1. Assessment of the intracellular thiols HL-60 cells were treated with various concentrations of shikonin for 6 and 12 h in 100-mm dishes. The cells were collected, washed twice with PBS, and lysed with RIPA buffer for 30 min on ice with occasional vortexing. The protein content was quantified using the Bradford procedure. Total thiol levels were determined by DTNB titration [36,37]. Briefly, 10 μl of cell lysate was added to cuvettes containing 90 μl of 1 mM DTNB in 6 M guanidine hydrochloride, pH 8.0. After 5 min at room temperature, the absorbance was read at 412 nm. Thiol levels were calculated by a calibration curve using GSH as the standard. Western blot analysis For Western blot analysis, equal amounts of protein in each lysate sample were separated by SDS–PAGE and electroblotted onto polyvinylidene difluoride membrane (Millipore, Billerica, MA, USA). After being blocked with 5% nonfat milk at room temperature for 2 h, the membranes were incubated with primary antibodies at 4 1C overnight, followed by washing with TBST three times, and then incubated with the peroxidase-conjugated secondary antibodies at room temperature for 1 h. The signal was detected using an enhanced chemiluminescence kit (GE Healthcare Life Sciences). Real-time reverse transcription-PCR (RT-PCR) RNA was isolated from cells using RNAiso Plus (TaKaRa, Dalian, China) according to the manufacturer's protocol and quantified through 260/280 wavelength measurement. Reverse transcription was performed using the Primescript RT reagent kit according to the manufacturer's protocol (TaKaRa). RT-PCR was achieved on an

D. Duan et al. / Free Radical Biology and Medicine 70 (2014) 182–193

Agilent Technologies Stratagene Mx3005P RT-PCR system using Power SYBR Green PCR Master Mix. The primer sequences for detecting human TrxR1 cDNA were CCACTGGTGAAAGACCACGTT (forward) and AGGAGAAAAGATCATCACTGCTGAT (reverse). Primer sequences for GAPDH were AGAAGGCTGGGGCTCATTTG (forward) and AGGGGCCATCCACAGTCTTC (reverse). Annexin V/propidium iodide (PI) staining HL-60 cells were treated with 1 and 2 mM shikonin for 24 or 48 h in 12-well plates. The cells were harvested and washed with PBS. Apoptotic cells were identified by double staining with fluorescein 5-isothiocyanate (FITC)-conjugated annexin V and PI according to the manufacturer's instructions (Zoman Biotech, Beijing, China). Data were obtained and analyzed using a FACSCanto flow cytometer (BD Biosciences, San Jose, CA, USA) with CellQuest software (BD Biosciences).

185

The same amounts of DMSO were added to the control cells and the activity was expressed as a percentage of the control. Hoechst 33342 staining HL-60 cells were plated in 12-well plates and were treated with 2 or 5 μM shikonin for 12 h, followed by the addition of 5 μg/ml Hoechst 33342. After incubation for 10 min, the cells were visualized and photographed under a Leica inverted fluorescence microscope. Statistics Data are presented as the mean 7SE. Statistical differences between two groups were assessed by Student's t test. Comparisons among multiple groups were performed using one-way analysis of variance, followed by a post hoc Scheffe test. P o0.05 was used as the criterion for statistical significance.

Measurement of caspase-3 activity

Results and discussion

HL-60 cells were treated with various concentrations of shikonin for 14 h in 100-mm dishes. The cells were collected, washed twice with PBS, and then lysed with RIPA buffer for 30 min on ice. The protein content was quantified using the Bradford procedure. A cell extract containing 50 μg of total proteins was incubated with the assay mixture (50 mM Hepes, pH 7.5, containing 2 mM EDTA, 5% glycerol, 10 mM dithiothreitol, 0.1% Chaps, and 0.2 mM Ac-DEVD-pNA) for 2 h at 37 1C in a final volume of 100 μl. The absorbance at 405 nm was measured using a microplate reader.

Inhibition of TrxR by shikonin in vitro Analysis of the chemical structure of shikonin (Fig. 1A) reveals that this molecule belongs to naphthoquinones, a class of highly reactive organic compounds whose chemical structures allow them to interact with biological molecules by forming covalent bonds and/or by acting as electron transfer agents in redox reactions [4,5]. Thus we speculated that shikonin might be a novel inhibitor of TrxR. With preincubation of shikonin with the reduced recombinant

Fig. 1. Inhibition of TrxR by shikonin in vitro. (A) The chemical structure of shikonin. (B) Time-dependent inhibition of TrxR by shikonin. The NADPH-reduced recombinant rat TrxR1 was incubated with 5 μM shikonin for the indicated times at room temperature, and the activity of the enzyme was determined by the DTNB reduction assay by monitoring the increase in the absorbance at 412 nm. The activity was expressed as a percentage of the control. nnnP o0.001 vs the control groups. (C) Dose-dependent inhibition of TrxR, U498C TrxR, and GR by shikonin. The NADPH-reduced recombinant rat TrxR1 (WT TrxR), U498C TrxR1, and GR were incubated with the indicated concentrations of shikonin for 0.5 h at room temperature, and the enzymes’ activity was determined. For assaying TrxR and U498C TrxR1, the increase in A412 was monitored by the convenient DTNB reduction method; for assaying GR, A340 was followed after the addition of GSSG. All the activity was expressed as a percentage of the control. (D) Inhibition of TrxR by shikonin detected by the endpoint insulin reduction assay. The NADPH-reduced TrxR1 was incubated with the various concentrations of shikonin for 2 h at room temperature, and the Trx-mediated insulin reduction assay was adopted. The activity was expressed as a percentage of the control.

186

D. Duan et al. / Free Radical Biology and Medicine 70 (2014) 182–193

rat TrxR1, the enzyme activity was gradually lost as incubation time increased (Fig. 1B). Removal of shikonin from the incubation mixture by a Sephadex G-25 desalting column could not recover TrxR1 activity (data not shown), indicating that an irreversible inhibition occurs, which is consistent with the action of the known Michael acceptor-containing inhibitors [33,34,38,39]. Next, we compared the inhibition potency of shikonin toward TrxR1, glutathione reductase, and the U498C TrxR1, in which Sec498 was replaced by Cys. As shown in Fig. 1C, shikonin effectively inhibits TrxR1 with an IC50 value around 1.6 μM, whereas it exhibits weak inhibition of U498C TrxR1 (IC50 410 μM) and even less effect on GR. The selective inhibition of WT TrxR1 but not U498C TrxR1 or GR suggests the Sec residue is specifically targeted by shikonin. To further explore the inhibition of TrxR by shikonin, we also determined the potency of shikonin at inhibiting the reduction of Trx, a physiological substrate of TrxR. As shown in Fig. 1D, shikonin dosedependently inhibits TrxR1 to reduce Trx as determined by the Trxmediated insulin reduction assay. Collectively, shikonin selectively inhibits TrxR in vitro by primarily targeting the Sec residue of the enzyme, most likely through a previously proposed mechanism [33,34], i.e., the Sec residue attacks the unsubstituted position of the electron-deficient double bond (highlighted in Fig. 1A) to form a shikonin–TrxR1 covalent adduct. Cytotoxicity of shikonin toward various cells and inhibition of TrxR in cells We have firmly demonstrated that shikonin effectively inhibits the enzyme activity of TrxR1 in vitro. To extend this in vitro discovery, we determined the cytotoxicity of shikonin toward a

panel of cancer cell lines (A549, HeLa, HepG2, and HL-60). Among all the tested cell lines, shikonin displays the most potent efficacy toward human promyelocytic leukemia HL-60 cells with a submicromolar IC50 value (  0.8 μM) after 72 h treatment (Fig. 2A). To further understand if shikonin has selectivity toward cancer cells, we also examined the cytotoxicity of shikonin toward two normal cell lines, human hepatic L02 cells and HEK 293 T cells. Compared to little effect on L02 cells or HEK 293 T cells, shikonin exhibited greater cytotoxicity toward HL-60 cells (Fig. 2B) under the experimental conditions. In sum, shikonin potently kills various types of cells with high preference for HL-60 tumor cells. Further studies indicate that the selective cytotoxicity of shikonin toward HL-60 cells might involve its potent inhibition of the cellular TrxR activity, remarkable decrease in cellular GSH/ GSSG ratio, and exclusive induction of ROS in HL-60 cells (Supplementary Fig. S1). Thus, HL-60 cells were chosen to investigate the physiological significance of the interaction between TrxR1 and shikonin. As illustrated in Fig. 2C, treatment of HL-60 cells with shikonin causes a remarkable decrease in the cellular TrxR activity with an IC50 value around 5 μM. It should be noted that the IC50 value for inhibition of TrxR is obtained after the cells were treated with shikonin for only 6 h, whereas the IC50 value for cytotoxicity is for 72 h treatment. This might account for the higher IC50 value required for TrxR inhibition. Indeed, after extending the incubation time to 12 h, the IC50 value of cellular TrxR inhibition decreased to less than 2 μM (Supplementary Fig. S1B). As a small reactive quinone, shikonin is also detoxified by conjugation to GSH (Supplementary Fig. S4A). However, this conjugation only partially prevented inhibition of TrxR by shikonin (Supplementary Fig. S4B), which

Fig. 2. Cytotoxicity of shikonin and inhibition of cellular TrxR in cells. (A) Sensitivity of various types of cancer cells (A549, HeLa, HepG2, and HL-60) to shikonin. The cells were treated with varying concentrations of shikonin for 72 h, and the viability was determined by the MTT method. Data are expressed as the mean 7 SE of three experiments. nnnPo 0.001 vs the other cells. (B) Cytotoxicity of shikonin toward L02, HEK 293 T, and HL-60 cells. The cells were treated with the indicated concentrations of shikonin for 48 h, and the viability was determined by the MTT method. Data are expressed as the mean7 SE of three experiments. nnnP o 0.001 vs the other cells. (C) Inhibition of TrxR activity in HL-60 cells by shikonin. After the cells were treated with the indicated concentrations of shikonin for 6 h, the enzyme activity of TrxR in the cells was determined by the endpoint insulin reduction assay. Data are expressed as the mean 7 SE of three experiments. nnnPo 0.001 vs the control groups. (D) No alteration in TrxR1 protein levels in HL-60 cells treated with shikonin. Cells were treated with the indicated concentrations of shikonin for 12 and 24 h, and the cell extracts were prepared and analyzed by Western blotting with an antibody against TrxR1. Actin was used as a loading control.

D. Duan et al. / Free Radical Biology and Medicine 70 (2014) 182–193

is consistent with the observation that shikonin does inhibit TrxR in HL-60 cells. The decrease in TrxR activity in cells could be caused by either direct inhibition of the enzyme activity or indirect downregulation of the protein expression or promotion of the protein degradation after shikonin treatment. There is no apparent alteration in the TrxR1 protein level determined by Western blots (Fig. 2D). Thus, the decrease in cellular TrxR activity may be attributable to the direct inhibition of the enzyme activity by shikonin, consistent with the in vitro results. Induction of oxidative stress in HL-60 cells The major function of TrxRs is to maintain the intracellular redox homeostasis and defend against oxidative stress. Shikonin inhibits TrxR both in vitro and in cells, which would be expected to disturb the cellular redox balance. Stimulation of the cells with shikonin leads to intracellular burst of ROS, as evidenced by DCFH-DA staining, and the ROS accumulation is dependent on the shikonin concentration (Fig. 3A). Pretreatment of the cells with the antioxidant NAC remarkably alleviates the ROS level. Several TrxR inhibitors are known to modify the enzyme to shift the enzyme from an antioxidant to a source of ROS [34,36,40], and thus we determined if shikonin has a similar effect. As shown in Fig. 3B, shikonin-modified TrxR1 displayed steady cytochrome c

187

reduction activity in the presence of NADPH, and this cytochrome c reduction activity was inhibited by SOD, suggesting the production of superoxide anions in such a process. The intracellular redox states are predominantly controlled by the balance of thiols and disulfides. To map the overall redox states of the cells, the amount of total thiols in cells was determined by DTNB titration. When HL-60 cells were treated with shikonin, significant loss of thiols was observed (Fig. 3C), consistent with the previous observation [41]. The ROS accumulation would lead to thiol oxidation, and the TrxR inhibition would prevent the reduction of intracellular disulfides. Taken together, shikonin treatment incites the loss of total thiols. Hydrogen peroxide is a mild oxidant that is produced endogenously. Because shikonin drastically disturbs the redox balance within the cells, shifting the intracellular environment to an oxidative state, we further investigated if shikonin could enhance the cytotoxicity of hydrogen peroxide. As shown in Fig. 3D, pretreatment of the cells with 1 or 1.5 μM shikonin significantly sensitized the cells to hydrogen peroxide challenge. Under our experimental conditions, shikonin (1 μM) or hydrogen peroxide (80 μM) caused negligible cell death. However, combination of shikonin and hydrogen peroxide treatment incited more than 20% cell death, indicating that shikonin has synergistic cytotoxicity with hydrogen peroxide to induce cell death.

Fig. 3. Induction of oxidative stress by shikonin. (A) Accumulation of ROS in the cells. HL-60 cells were treated with various concentrations of shikonin in the absence or presence of 5 mM NAC for 1 h followed by incubation with the fluorescence probe DCFH-DA (10 μM) for 30 min. Phase-contrast (top) and fluorescence (bottom) images were acquired by fluorescence microscopy. (B) Induction of superoxide anion production by shikonin-modified TrxR. Superoxide anion generation was monitored by cytochrome c reduction assay. The inset shows the change in absorbance at 550 nm after addition of cytochrome c and SOD. SOD was added where the arrow indicates in the inset. (C) Consumption of intracellular thiols by shikonin. HL-60 cells were treated with various concentrations of shikonin (2, 5, and 10 μM) for 6 or 12 h. Total thiols in the cell extracts were quantified by DTNB titration. nP o0.05, nnP o0.01, nnnPo 0.001 vs the control groups. (D) Enhancement of hydrogen peroxide cytotoxicity by shikonin. HL-60 cells were pretreated with shikonin (1 or 1.5 μM) for 4 h. After removal of shikonin, the cells were further incubated with hydrogen peroxide (80 μM) for 24 h. Cell viability was determined by MTT assay. Data are expressed as the mean 7 SE of three experiments. nnPo 0.01 vs the control groups.

188

D. Duan et al. / Free Radical Biology and Medicine 70 (2014) 182–193

The role of NAC and GSH in shikonin-induced cell death As we had demonstrated that perturbation of cellular redox balance is involved in the biological action of shikonin, we next explored the functions of NAC and GSH in the process of shikonininduced cell death. NAC is a known antioxidant and a precursor for GSH, a major small-molecule antioxidant that neutralizes ROS in all mammalian cells. Owing to the interference caused by high concentrations of NAC in the MTT assay, we adopted the trypan blue exclusion assay to distinguish dead cells from live cells. Pretreatment of HL-60 cells with NAC can protect against shikonin-induced cell death, and higher concentrations of NAC completely abolish the cytotoxicity of shikonin (Fig. 4A). Shikonininduced cell death was blocked by NAC, but other general antioxidants, including resveratrol, vitamin C, and vitamin E, did not block this process (data not shown), suggesting that the induction of cell death by shikonin may be related to the GSH depletion. Thus, we further determined the effect of GSH on the cytotoxicity of shikonin. Consistent with the protective role of NAC, depletion of cellular GSH by pretreatment of the cells with BSO remarkably enhanced the cytotoxicity of shikonin (Fig. 4B). Under our experimental conditions, pretreatment of HL-60 cells with 50 μM BSO for 24 h downregulated the intracellular GSH level to less than 20% of the control (Supplementary Fig. S2). NAC had marginal effect on the cellular GSH levels (Fig. 4C), but could significantly restore the GSH/GSSG ratio after shikonin treatment (Fig. 4D). GSH is a pivotal component of the glutathione system, which is another redox regulation network in cells in addition to the thioredoxin system and also acts as a backup of the thioredoxin system [42]. Depletion of GSH sensitizing the cells to

shikonin supports the involvement of the thioredoxin system in the biological action of shikonin. Protection of cell death by TrxR To further disclose the involvement of TrxR in the cytotoxicity of shikonin, we turned to comparing the sensitivity of HEK cells stably overexpressing TrxR1 (HEK-TrxR1) and cells stably transfected with a vector (HEK-IRES) toward shikonin treatment. To confirm the transfection efficiency, we determined the protein expression as well as the enzyme activity in the cells. The TrxR1 expression level and the TrxR activity in HEK-TrxR1 cells are  3and  3.5-fold higher than those in HEK-IRES cells, respectively (Fig. 5A). As shown in Fig. 5B, shikonin displays significantly higher growth inhibition in HEK-IRES cells than in HEK-TrxR1 cells. To further address the physiological relevance of TrxR1mediated shikonin cytotoxicity, we generated a cell line with stably knocked down TrxR1 expression by transfection of shRNA specifically targeting the enzyme. Owing to the low transfection efficiency of HL-60 cells, we chose HeLa cells for the knockdown experiments. The knockdown of TrxR1 in HeLa cells was fully evaluated as illustrated in Figs. 5C and D. Compared with the control cells transfected with nontargeting shRNA (HeLa-shNT cells), both gene and protein expression of TrxR1 in HeLashTrxR1 cells (HeLa cells transfected with shRNA specifically targeting TrxR1) were drastically downregulated (Fig. 5C and the inset in Fig. 5D). The total TrxR activity in HeLa-shTrxR1 cells decreased to about half of that in HeLa-shNT cells (Fig. 5D). Importantly, shikonin shows elevating cytotoxicity toward HeLashTrxR1 cells (Fig. 5E). It should be noted that this shRNA

Fig. 4. Protection against cell death by NAC and GSH. (A) Protection of the cells by NAC. HL-60 cells were incubated with the indicated concentrations of NAC and shikonin (1, 2, or 5 μM) for 24 h. Cell viability was determined by the trypan blue exclusion assay. Data are expressed as the mean7 SE of three experiments. (B) Augmentation of the cytotoxicity by GSH depletion. HL-60 cells were treated with 50 μM BSO for 24 h to lower the intracellular GSH level, followed by shikonin treatment for an additional 48 h. Treatment of the cells with 50 μM BSO for 24 h decreased the intracellular GSH to less than 20% of the control cells (Supplementary Fig. S2). Data are expressed as the mean 7SE of three experiments. nnPo 0.01, nnnPo 0.001 vs the control groups. (C) Effect of NAC on the intracellular GSH levels in HL-60 cells. The cells were treated with NAC (2 mM) and shikonin (5 μM) for 24 h. Intracellular GSH levels were measured as described under Materials and methods. (D) Effect of NAC on the GSH/GSSG ratio in HL60 cells. The ratio was calculated according to the reduced GSH and GSSG levels. Data are expressed as the mean 7SE of three experiments. nnnP o0.001 vs the control groups.

D. Duan et al. / Free Radical Biology and Medicine 70 (2014) 182–193

189

Fig. 5. The role of TrxR in shikonin-induced cell death. (A) Quantification of TrxR activity in HEK-IRES and HEK-TrxR1 cells. The TrxR activity in different cells was determined by the endpoint insulin reduction assay, and the activity was expressed as the rate of thiol production. Data are expressed as the mean 7 SE of three experiments. Inset: cell extracts were prepared and analyzed by Western blotting with an antibody against TrxR1 and actin. (B) Cytotoxic effects of shikonin on HEK-IRES and HEK-TrxR1 cells. The cells (5000) were seeded in a 96-well plate and treated with the indicated concentrations of shikonin for 72 h, and cell viability was determined by the MTT assay. Data are expressed as the mean 7 SE of three experiments. (C) TrxR1 gene expression in HeLa-shNT and HeLa-shTrxR1 cells. The mRNA levels of TrxR1 were quantified by RT-PCR and normalized using GAPDH as an internal standard. (D) Quantification of TrxR activity in HeLa-shNT and HeLa-shTrxR1 cells. The TrxR activity in different cells was determined by the endpoint insulin reduction assay, and the activity was expressed as the rate of thiol production. Data are expressed as the mean 7 SE of three experiments. Inset: cell extracts were prepared and analyzed by Western blotting with an antibody against TrxR1 and actin. (E) Cytotoxic effects of shikonin on HeLa-shNT and HeLa-shTrxR1 cells. The cells (5000) were treated with the indicated concentrations of shikonin for 48 h, and cell viability was determined by the MTT assay. Data are expressed as the mean7 SE of three experiments. nnPo 0.01, nnnP o0.001 vs the control groups.

specifically targets TrxR1, with no effect on TrxR2 [31]. The mitochondrial isoform TrxR2 probably contributes to the majority of the remaining activity in the HeLa-shTrxR1 cells, as the insulin reduction assay is not specific for the TrxR1 isoform. Derivatization of TrxR2 by shikonin could cause production of intermitochondrial ROS leading to mitochondrial dysfunction and apoptosis, and the decrease/increase in TrxR1 levels might provide more/less chance to target TrxR2 by shikonin. This could be one possibility in accounting for the observations from the TrxR overexpression/ knockdown experiments. Collectively, our results strongly support that the biological action of shikonin in cells is related to its interaction with TrxR1. Induction of apoptosis in HL-60 cells It has been reported that shikonin has the ability to activate apoptotic signaling in numerous type of cells [8,10,14,15,18,41, 43–45]. Herein, we also demonstrate that shikonin kills HL-60 cells predominantly through the induction of apoptosis. The results from the annexin-V–FITC/PI double staining of the cells after treatment with shikonin are illustrated in Fig. 6A. Shikonin triggers apoptotic cell death in a dose- and time-dependent manner. After the cells were treated with shikonin (2 μM) for 48 h, FITC-positive cells accounted for  70% of the total cells, indicating that apoptosis is a major mechanism of the cytotoxicity of shikonin. When HL-60 cells were incubated with shikonin followed by Hoechst staining, the bulk of the cells displayed condensed nuclei, a characteristic morphology of cells undergoing apoptosis (Fig. 6B). Caspase-3 is a crucial component of the apoptotic machinery in various cell types, and the activation of

caspase-3 is a central event in the process of apoptosis [46]. Thus, we further determined the activity of caspase-3 after cells treated with shikonin. As shown in Fig. 6C, shikonin significantly increased the caspase-3 activity in HL-60 cells. However, a higher concentration of shikonin (5 μM) decreased the activation of caspase-3, indicating that high concentration of shikonin probably promotes nonapoptotic cell death. Taken together, our data reveal that shikonin mainly induces apoptotic cell death in HL-60 cells. The thioredoxin system has emerged as an important target in cancer chemotherapy, because both Trx and TrxR have been shown to be overexpressed in a variety of human cancer types and associated with increased tumor growth, drug resistance, and poor patient prognosis [47]. Thus, the past years have witnessed increasing attention to developing novel inhibitors of the system as potential antitumor agents [19,48,49]. It has long been known that shikonin could induce oxidative stress in various cell lines, and induction of ROS production is involved in the biological functions of shikonin [10,14,15,44,45,50]. However, to the best of our knowledge, there is no detailed mechanism accounting for how shikonin elevates oxidative stress. We have identified TrxR as a target of shikonin and demonstrated that shikonin induces apoptosis through a previously uncharacterized mechanism by targeting this enzyme. Binding of shikonin to TrxR inhibits the physiological functions of TrxR, but further turns the enzyme to an NADPH oxidase to directly generate superoxide anions, which leads to ROS accumulation within cells, consequently causing intracellular thiol depletion and finally eliciting oxidative stress. A number of anticancer drugs or physical treatments, such as doxorubicin, daunorubicin, mitomycin C, etoposide, cisplatin, and irradiation, act, at least in part, through induction of ROS.

190

D. Duan et al. / Free Radical Biology and Medicine 70 (2014) 182–193

Fig. 6. Induction of apoptosis in HL-60 cells. (A) Analysis of apoptosis by annexin V/PI double-staining assay. HL-60 cells were treated with various concentrations of shikonin for 24 or 48 h, and representative FACS analysis scattergrams of annexin V–FITC/PI staining are shown on the left. The cells show four different cell populations marked as follows: double-negative (unstained) cells showing live cell population (lower left, Q3), annexin V-positive and PI-negative stained cells showing early apoptosis (lower right, Q4), annexin V/PI double-stained cells showing late apoptosis (upper right, Q2), and finally PI-positive and annexin V-negative stained cells showing dead cells (upper left, Q1). The quantification of dead cells (Q1), apoptotic cells (Q2 and Q4), and normal cells (Q4) is illustrated on the right. The data are expressed as the mean 7SE of three independent samples. nnP o 0.01, nnnP o0.001 vs the control groups. (B) Analysis of apoptosis by nuclear condensation. The Hoechst 33342 staining showed typical apoptotic morphology changes after shikonin treatment. HL-60 cells were incubated with different concentrations of shikonin for 12 h followed by Hoechst 33342 staining. Phasecontrast (top) and fluorescence (bottom) images were acquired by fluorescence microscopy. (C) Activation of caspase-3 by shikonin in HL-60 cells. HL-60 cells were incubated with the indicated concentrations of shikonin for 14 h and the caspase-3 activity in the cell extracts was determined by a colorimetric assay. nnPo 0.01 vs the control group.

The observation that shikonin sensitizes HL-60 cells to hydrogen peroxide treatment opens a new window on treating tumors with shikonin in combination with existing oxidative stress-causing anticancer drugs or physical treatments, such as ionizing radiation and photodynamic therapy. Elevated oxidative stress is observed more frequently in malignant cells than in normal cells [51]. One strategy in cancer therapy is using ROS or oxidative stress-causing agents or physical therapy to kill cancer cells as it seems that normal cells are more resistant to ROS than cancer cells [31,52–55]. It is therefore expected that further exposure to a low level of ROS would promote tumor cells toward death, whereas normal cells might maintain redox homeostasis through adaptive antioxidant responses. Perhaps the ability to inhibit the antioxidant enzyme

TrxR and accumulate ROS is a key point in the molecular mechanism shikonin uses to exert its antitumor activity. Naphthoquinones are a class of highly reactive molecules whose chemical structures allow them to interact with cellular targets by forming covalent bonds and/or by acting as electron transfer agents in oxidation–reduction reactions [4,5]. It has been reported that TrxR interacts with a number of naturally occurring or synthetic quinones, such as indolequinones [56], juglone [57], pleurotin [58], pyrroloquinoline quinone [59], and doxorubicin [60]. Unlike the fully substituted quinones, such as ubiquinone [61] and tocopherol quinone [62], which usually act as electroncarrying molecules via a quinone–hydroquinone exchange reaction, shikonin has a potential electrophilic site (indicated with an

D. Duan et al. / Free Radical Biology and Medicine 70 (2014) 182–193

asterisk in Fig. 1A), which makes it particularly prone to attack by biological nucleophiles [7,9]. The specificity of shikonin–TrxR interaction was demonstrated by the following evidence. First, we compared the in vitro inhibition potency of shikonin on TrxR1, U498C TrxR1, and GR, a homolog of TrxR [26,63] (Fig. 1). A single mutation of Sec to Cys sharply decreased the enzyme sensitivity to shikonin, indicating that the Sec residue in TrxR is a primary target for this small molecule. The structure of GR is quite closely related to that of TrxR. However, even less inhibition of GR was observed under our experimental conditions. The selective inhibition of WT TrxR1 but not U498C TrxR1 or GR suggests a specific interaction of shikonin and TrxR1. Most biological functions of shikonin may be attributed to the accumulation of ROS in cells [10,15,44,45,50]. Nevertheless, how the ROS are produced is not known. Our result that shikonin-modified TrxR1 directly generates ROS thus provides a mechanistic interpretation in accounting for the previous observations. Second, we provided evidence in cellular context to support the unique role of TrxR in shikonin's biological action. HEK cells stably overexpressing TrxR1 (HEK-TrxR1 cells) show less growth inhibition compared to the cells transfected with only the vector (HEK-IRES) (Fig. 5B). More physiologically relevant evidence that genetic knockdown of TrxR1 elevates shikonin's cytotoxicity further supports that TrxR1 is critically involved in the biological function of shikonin (Fig. 5E). The glutathione and thioredoxin systems are two predominant networks that work independently but with some overlaps in maintaining the intracellular redox balance and defending against oxidative stress. Recent studies indicated that the glutathione system can act as a backup of the thioredoxin system [42]. Our results that downregulation of intracellular GSH by BSO enhances shikonin's cytotoxicity, whereas upregulation of GSH by NAC alleviates the cytotoxicity (Figs. 4A and B), also suggest that TrxR is involved in shikonin's biological action. Third, other cellular targets, such as PTEN [12], proteasome [9], p53 [14], tyrosine kinase [16], steroid sulfatase [17], DNA topoisomerase [18], and epidermal growth factor receptor [11],were also reported to be affected by shikonin in various cell types. However, higher concentrations were required for these targets compared to shikonin in triggering TrxR-mediated apoptosis in HL-60 cells. Taken together, our data suggest that shikonin targets TrxR in HL-60 cells with high specificity. As TrxR2 shares the same structure as well as catalytic mechanism of TrxR1, TrxR2 is also very like a target of shikonin. In principle, the biological significance of inhibition of TrxR is at least as follows. First, the inhibition of TrxR reduces the cellular available pool of reduced Trx, leading to a decrease in the activity of diverse antioxidant enzyme systems that require reduced Trx as an electron donor, which results in an accumulation of ROS and alteration in the cellular redox states [64]. Second, reduced Trx directly interacts with diverse apoptosis-related enzymes, such as ASK1 [65], procaspase-3 [66], and NF-κB [67], to suppress apoptosis, and thus TrxR inhibition promotes apoptosis. Third, TrxR inhibition could result in the formation of selenium-compromised thioredoxin reductase-derived apoptotic proteins (SecTRAPs) [68], in which the active-site Sec residue is modified by electrophilic molecules. These SecTRAPs lose the ability to reduce oxidized Trx, but maintain NADPH oxidase activity in constantly generating ROS, which eventually contributes to increased intracellular oxidative stress. We believe that all these factors contribute to the cellular function of shikonin. Apoptotic cell death is a consequence of a series of precisely controlled events that are frequently altered in cancer cells. Two known apoptosis signaling pathways, i.e., the extrinsic cell surface receptor pathway [69] and the intrinsic mitochondrial pathway [70], converge on caspase activation [71]. Abrogation of apoptotic pathways is frequently found in numerous malignant cells arising from a complex interplay of genetic aberrations and misregulated

191

pathways [72]. Shikonin manifests high cytotoxicity toward HL-60 cells predominantly through induction of apoptosis (Fig. 6), thus potentiating the clinical use of shikonin in the treatment of tumors. As we have unambiguously demonstrated that shikonin promotes accumulation of ROS, depletion of intracellular thiols, and sensitization of cells to oxidative stress in HL-60 cells (Fig. 3), it is most likely that shikonin prompts apoptosis through the oxidative stress-mediated intrinsic mitochondrial pathway, which is also supported by previously published results [41,43]. Summary In conclusion, we have discovered TrxR as a target of shikonin, both in vitro and in human promyelocytic leukemia HL-60 cells, and demonstrated that shikonin induces apoptotic cell death through a previously unrecognized mechanism. The observation that shikonin has synergistic cytotoxic effects with hydrogen peroxide sheds new light on the possibility of treating tumors with shikonin in combination with existing oxidative stresscausing anticancer drugs or physical treatments. The discovery of shikonin–TrxR interaction provides deep insight into the understanding of how this naphthoquinone acts in vivo, and this novel targeting mechanism may lead to the development of potent small-molecule TrxR inhibitors as potential cancer chemotherapeutic agents.

Acknowledgments Financial support from the National Natural Science Foundation of China (21002047), the Ministry of Education of China (20100211110027), Lanzhou University (Fundamental Research Funds for the Central Universities, lzujbky-2012-59), the National Natural Science Foundation of China for Fostering Talents in Basic Research of the (J1103307), and the 111 Project are gratefully acknowledged. The authors also express heartfelt appreciation to Professor Arne Holmgren for recombinant rat TrxR1 and to Professor Constantinos Koumenis from the University of Pennsylvania School of Medicine for shRNA plasmids and HEK-TrxR1 and HEK-IRES cells.

Appendix A.

Supplementary material

Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.freeradbiomed. 2014.02.016. References [1] Kingston, D. G. Modern natural products drug discovery and its relevance to biodiversity conservation. J. Nat. Prod. 74:496–511; 2011. [2] Koehn, F. E.; Carter, G. T. The evolving role of natural products in drug discovery. Nat. Rev. Drug Discovery 4:206–220; 2005. [3] Newman, D. J.; Cragg, G. M.; Snader, K. M. Natural products as sources of new drugs over the period 1981–2002. J. Nat. Prod. 66:1022–1037; 2003. [4] Kumagai, Y.; Shinkai, Y.; Miura, T.; Cho, A. K. The chemical biology of naphthoquinones and its environmental implications. Annu. Rev. Pharmacol. Toxicol. 52:221–247; 2012. [5] O'Brien, P. J. Molecular mechanisms of quinone cytotoxicity. Chem. Biol. Interact. 80:1–41; 1991. [6] Chen, X.; Yang, L.; Oppenheim, J. J.; Howard, M. Z. Cellular pharmacology studies of shikonin derivatives. Phytother. Res. 16:199–209; 2002. [7] Papageorgiou, V. P.; Assimopoulou, A. N.; Couladouros, E. A.; Hepworth, D.; Nicolaou, K. C. The chemistry and biology of alkannin, shikonin, and related naphthazarin natural products. Angew. Chem. Int. Ed. Engl. 38:270–300; 1999. [8] Min, R.; Tong, J.; Wenjun, Y.; Wenhu, D.; Xiaojian, Z.; Jiacai, H.; Jian, Z.; Wantao, C.; Chenping, Z. Growth inhibition and induction of apoptosis in human oral squamous cell carcinoma Tca-8113 cell lines by shikonin was partly through the inactivation of NF-kappaB pathway. Phytother. Res. 22:407–415; 2008.

192

D. Duan et al. / Free Radical Biology and Medicine 70 (2014) 182–193

[9] Yang, H.; Zhou, P.; Huang, H.; Chen, D.; Ma, N.; Cui, Q. C.; Shen, S.; Dong, W.; Zhang, X.; Lian, W.; Wang, X.; Dou, Q. P.; Liu, J. Shikonin exerts antitumor activity via proteasome inhibition and cell death induction in vitro and in vivo. Int. J. Cancer 124:2450–2459; 2009. [10] Mao, X.; Yu, C. R.; Li, W. H.; Li, W. X. Induction of apoptosis by shikonin through a ROS/JNK-mediated process in Bcr/Abl-positive chronic myelogenous leukemia (CML) cells. Cell Res 18:879–888; 2008. [11] Singh, F.; Gao, D.; Lebwohl, M. G.; Wei, H. Shikonin modulates cell proliferation by inhibiting epidermal growth factor receptor signaling in human epidermoid carcinoma cells. Cancer Lett 200:115–121; 2003. [12] Nigorikawa, K.; Yoshikawa, K.; Sasaki, T.; Iida, E.; Tsukamoto, M.; Murakami, H.; Maehama, T.; Hazeki, K.; Hazeki, O. A naphthoquinone derivative, shikonin, has insulin-like actions by inhibiting both phosphatase and tensin homolog deleted on chromosome 10 and tyrosine phosphatases. Mol. Pharmacol. 70:1143–1149; 2006. [13] Lee, H.; Bae, S.; Kim, K.; Kim, W.; Chung, S. I.; Yang, Y.; Yoon, Y. Shikonin inhibits adipogenesis by modulation of the WNT/beta-catenin pathway. Life Sci. 88:294–301; 2011. [14] Hsu, P. C.; Huang, Y. T.; Tsai, M. L.; Wang, Y. J.; Lin, J. K.; Pan, M. H. Induction of apoptosis by shikonin through coordinative modulation of the Bcl-2 family, p27, and p53, release of cytochrome c, and sequential activation of caspases in human colorectal carcinoma cells. J. Agric. Food Chem. 52:6330–6337; 2004. [15] Ahn, J.; Won, M.; Choi, J. H.; Kim, Y. S.; Jung, C. R.; Im, D. S.; Kyun, M. L.; Lee, K.; Song, K. B.; Chung, K. S. Reactive oxygen species-mediated activation of the Akt/ASK1/p38 signaling cascade and p21(Cip1) downregulation are required for shikonin-induced apoptosis. Apoptosis 18:870–881; 2013. [16] Kamei, R.; Kitagawa, Y.; Kadokura, M.; Hattori, F.; Hazeki, O.; Ebina, Y.; Nishihara, T.; Oikawa, S. Shikonin stimulates glucose uptake in 3T3-L1 adipocytes via an insulin-independent tyrosine kinase pathway. Biochem. Biophys. Res. Commun. 292:642–651; 2002. [17] Zhang, Y.; Qian, R. Q.; Li, P. P. Shikonin, an ingredient of Lithospermum erythrorhizon, down-regulates the expression of steroid sulfatase genes in breast cancer cells. Cancer Lett. 284:47–54; 2009. [18] Chen, C.; Shanmugasundaram, K.; Rigby, A. C.; Kung, A. L. Shikonin, a natural product from the root of Lithospermum erythrorhizon, is a cytotoxic DNAbinding agent. Eur. J. Pharm. Sci. 49:18–26; 2013. [19] Cai, W.; Zhang, L.; Song, Y.; Wang, B.; Zhang, B.; Cui, X.; Hu, G.; Liu, Y.; Wu, J.; Fang, J. Small molecule inhibitors of mammalian thioredoxin reductase. Free Radic. Biol. Med. 52:257–265; 2012. [20] Holmgren, A.; Lu, J. Thioredoxin and thioredoxin reductase: current research with special reference to human disease. Biochem. Biophys. Res. Commun. 396:120–124; 2010. [21] Arner, E. S. Focus on mammalian thioredoxin reductases—important selenoproteins with versatile functions. Biochim. Biophys. Acta 1790:495–526; 2009. [22] Go, Y. M.; Jones, D. P. Thiol/disulfide redox states in signaling and sensing. Crit. Rev. Biochem. Mol. Biol. 48:173–181; 2013. [23] Rigobello, M. P.; Bindoli, A. Mitochondrial thioredoxin reductase purification, inhibitor studies, and role in cell signaling. Methods Enzymol. 474:109–122; 2010. [24] Zhong, L.; Holmgren, A. Essential role of selenium in the catalytic activities of mammalian thioredoxin reductase revealed by characterization of recombinant enzymes with selenocysteine mutations. J. Biol. Chem 275:18121–18128; 2000. [25] Gladyshev, V. N.; Jeang, K. T.; Stadtman, T. C. Selenocysteine, identified as the penultimate C-terminal residue in human T-cell thioredoxin reductase, corresponds to TGA in the human placental gene. Proc. Natl. Acad. Sci. USA 93:6146–6151; 1996. [26] Zhong, L.; Arner, E. S.; Holmgren, A. Structure and mechanism of mammalian thioredoxin reductase: the active site is a redox-active selenolthiol/selenenylsulfide formed from the conserved cysteine–selenocysteine sequence. Proc. Natl. Acad. Sci. USA 97:5854–5859; 2000. [27] Yoo, M. H.; Xu, X. M.; Carlson, B. A.; Gladyshev, V. N.; Hatfield, D. L. Thioredoxin reductase 1 deficiency reverses tumor phenotype and tumorigenicity of lung carcinoma cells. J. Biol. Chem. 281:13005–13008; 2006. [28] Mahmood, D. F.; Abderrazak, A.; Khadija, E. H.; Simmet, T.; Rouis, M. The thioredoxin system as a therapeutic target in human health and disease. Antioxid. Redox Signaling 19:1266–1303; 2013. [29] Powis, G.; Kirkpatrick, D. L. Thioredoxin signaling as a target for cancer therapy. Curr. Opin. Pharmacol. 7:392–397; 2007. [30] Arner, E. S.; Sarioglu, H.; Lottspeich, F.; Holmgren, A.; Bock, A. High-level expression in Escherichia coli of selenocysteine-containing rat thioredoxin reductase utilizing gene fusions with engineered bacterial-type SECIS elements and co-expression with the selA, selB and selC genes. J. Mol. Biol 292:1003–1016; 1999. [31] Javvadi, P.; Hertan, L.; Kosoff, R.; Datta, T.; Kolev, J.; Mick, R.; Tuttle, S. W.; Koumenis, C. Thioredoxin reductase-1 mediates curcumin-induced radiosensitization of squamous carcinoma cells. Cancer Res. 70:1941–1950; 2010. [32] Nalvarte, I.; Damdimopoulos, A. E.; Nystom, C.; Nordman, T.; Miranda-Vizuete, A.; Olsson, J. M.; Eriksson, L.; Bjornstedt, M.; Arner, E. S.; Spyrou, G. Overexpression of enzymatically active human cytosolic and mitochondrial thioredoxin reductase in HEK-293 cells: effect on cell growth and differentiation. J. Biol. Chem. 279:54510–54517; 2004. [33] Fang, J.; Holmgren, A. Inhibition of thioredoxin and thioredoxin reductase by 4-hydroxy-2-nonenal in vitro and in vivo. J. Am. Chem. Soc 128:1879–1885; 2006. [34] Fang, J.; Lu, J.; Holmgren, A. Thioredoxin reductase is irreversibly modified by curcumin: a novel molecular mechanism for its anticancer activity. J. Biol. Chem. 280:25284–25290; 2005.

[35] Rahman, I.; Kode, A.; Biswas, S. K. Assay for quantitative determination of glutathione and glutathione disulfide levels using enzymatic recycling method. Nat. Protoc. 1:3159–3165; 2006. [36] Cai, W.; Zhang, B.; Duan, D.; Wu, J.; Fang, J. Curcumin targeting the thioredoxin system elevates oxidative stress in HeLa cells. Toxicol. Appl. Pharmacol. 262:341–348; 2012. [37] Ellman, G. L. Tissue sulfhydryl groups. Arch. Biochem. Biophys. 82:70–77; 1959. [38] Gan, F. F.; Kaminska, K. K.; Yang, H.; Liew, C. Y.; Leow, P. C.; So, C. L.; Tu, L. N.; Roy, A.; Yap, C. W.; Kang, T. S.; Chui, W. K.; Chew, E. H. Identification of Michael acceptor-centric pharmacophores with substituents that yield strong thioredoxin reductase inhibitory character correlated to antiproliferative activity. Antioxid. Redox Signaling 19:1149–1165; 2013. [39] Chew, E. H.; Nagle, A. A.; Zhang, Y.; Scarmagnani, S.; Palaniappan, P.; Bradshaw, T. D.; Holmgren, A.; Westwell, A. D. Cinnamaldehydes inhibit thioredoxin reductase and induce Nrf2: potential candidates for cancer therapy and chemoprevention. Free Radic. Biol. Med. 48:98–111; 2010. [40] Arner, E. S.; Bjornstedt, M.; Holmgren, A. 1-Chloro-2,4-dinitrobenzene is an irreversible inhibitor of human thioredoxin reductase: loss of thioredoxin disulfide reductase activity is accompanied by a large increase in NADPH oxidase activity. J. Biol. Chem 270:3479–3482; 1995. [41] Gao, D.; Hiromura, M.; Yasui, H.; Sakurai, H. Direct reaction between shikonin and thiols induces apoptosis in HL60 cells. Biol. Pharm. Bull. 25:827–832; 2002. [42] Du, Y.; Zhang, H.; Lu, J.; Holmgren, A. Glutathione and glutaredoxin act as a backup of human thioredoxin reductase 1 to reduce thioredoxin 1 preventing cell death by aurothioglucose. J. Biol. Chem 287:38210–38219; 2012. [43] Yoon, Y.; Kim, Y. O.; Lim, N. Y.; Jeon, W. K.; Sung, H. J. Shikonin, an ingredient of Lithospermum erythrorhizon induced apoptosis in HL60 human premyelocytic leukemia cell line. Planta Med. 65:532–535; 1999. [44] Gong, K.; Li, W. Shikonin, a Chinese plant-derived naphthoquinone, induces apoptosis in hepatocellular carcinoma cells through reactive oxygen species: a potential new treatment for hepatocellular carcinoma. Free Radic. Biol. Med 51:2259–2271; 2011. [45] Chang, I. C.; Huang, Y. J.; Chiang, T. I.; Yeh, C. W.; Hsu, L. S. Shikonin induces apoptosis through reactive oxygen species/extracellular signal-regulated kinase pathway in osteosarcoma cells. Biol. Pharm. Bull. 33:816–824; 2010. [46] Budihardjo, I.; Oliver, H.; Lutter, M.; Luo, X.; Wang, X. Biochemical pathways of caspase activation during apoptosis. Annu. Rev. Cell Dev. Biol. 15:269–290; 1999. [47] Urig, S.; Becker, K. On the potential of thioredoxin reductase inhibitors for cancer therapy. Semin. Cancer Biol. 16:452–465; 2006. [48] Wang, L.; Yang, Z.; Fu, J.; Yin, H.; Xiong, K.; Tan, Q.; Jin, H.; Li, J.; Wang, T.; Tang, W.; Yin, J.; Cai, G.; Liu, M.; Kehr, S.; Becker, K.; Zeng, H. Ethaselen: a potent mammalian thioredoxin reductase 1 inhibitor and novel organoselenium anticancer agent. Free Radic. Biol. Med. 52:898–908; 2012. [49] Schuh, E.; Pfluger, C.; Citta, A.; Folda, A.; Rigobello, M. P.; Bindoli, A.; Casini, A.; Mohr, F. Gold(I) carbene complexes causing thioredoxin 1 and thioredoxin 2 oxidation as potential anticancer agents. J. Med. Chem. 55:5518–5528; 2012. [50] Chen, C. H.; Chern, C. L.; Lin, C. C.; Lu, F. J.; Shih, M. K.; Hsieh, P. Y.; Liu, T. Z. Involvement of reactive oxygen species, but not mitochondrial permeability transition in the apoptotic induction of human SK-Hep-1 hepatoma cells by shikonin. Planta Med. 69:1119–1124; 2003. [51] Trachootham, D.; Alexandre, J.; Huang, P. Targeting cancer cells by ROSmediated mechanisms: a radical therapeutic approach? Nat. Rev. Drug Discovery 8:579–591; 2009. [52] Shaw, A. T.; Winslow, M. M.; Magendantz, M.; Ouyang, C.; Dowdle, J.; Subramanian, A.; Lewis, T. A.; Maglathin, R. L.; Tolliday, N.; Jacks, T. Selective killing of K-ras mutant cancer cells by small molecule inducers of oxidative stress. Proc. Natl. Acad. Sci. USA 108:8773–8778; 2011. [53] Raj, L.; Ide, T.; Gurkar, A. U.; Foley, M.; Schenone, M.; Li, X.; Tolliday, N. J.; Golub, T. R.; Carr, S. A.; Shamji, A. F.; Stern, A. M.; Mandinova, A.; Schreiber, S. L.; Lee, S. W. Selective killing of cancer cells by a small molecule targeting the stress response to ROS. Nature 475:231–234; 2011. [54] Wang, H. C.; Choudhary, S. Reactive oxygen species-mediated therapeutic control of bladder cancer. Nat. Rev. Urol 8:608–616; 2011. [55] Xu, Y.; Fang, F.; Miriyala, S.; Crooks, P. A.; Oberley, T. D.; Chaiswing, L.; Noel, T.; Holley, A. K.; Zhao, Y.; Kiningham, K. K.; Clair, D. K.; Clair, W. H. KEAP1 is a redox sensitive target that arbitrates the opposing radiosensitive effects of parthenolide in normal and cancer cells. Cancer Res. 73:4406–4417; 2013. [56] Yan, C.; Siegel, D.; Newsome, J.; Chilloux, A.; Moody, C. J.; Ross, D. Antitumor indolequinones induced apoptosis in human pancreatic cancer cells via inhibition of thioredoxin reductase and activation of redox signaling. Mol. Pharmacol. 81:401–410; 2012. [57] Cenas, N.; Nivinskas, H.; Anusevicius, Z.; Sarlauskas, J.; Lederer, F.; Arner, E. S. Interactions of quinones with thioredoxin reductase: a challenge to the antioxidant role of the mammalian selenoprotein. J. Biol. Chem. 279: 2583–2592; 2004. [58] Kunkel, M. W.; Kirkpatrick, D. L.; Johnson, J. I.; Powis, G. Cell line-directed screening assay for inhibitors of thioredoxin reductase signaling as potential anti-cancer drugs. Anticancer Drug Des. 12:659–670; 1997. [59] Xu, J.; Arner, E. S. Pyrroloquinoline quinone modulates the kinetic parameters of the mammalian selenoprotein thioredoxin reductase 1 and is an inhibitor of glutathione reductase. Biochem. Pharmacol 83:815–820; 2012. [60] Mau, B. L.; Powis, G. Inhibition of thioredoxin reductase (E.C. 1.6.4.5.) by antitumor quinones. Free Radic. Res. Commun. 8:365–372; 1990. [61] Xia, L.; Nordman, T.; Olsson, J. M.; Damdimopoulos, A.; Bjorkhem-Bergman, L.; Nalvarte, I.; Eriksson, L. C.; Arner, E. S.; Spyrou, G.; Bjornstedt, M. The

D. Duan et al. / Free Radical Biology and Medicine 70 (2014) 182–193

[62]

[63]

[64] [65]

mammalian cytosolic selenoenzyme thioredoxin reductase reduces ubiquinone: a novel mechanism for defense against oxidative stress. J. Biol. Chem. 278:2141–2146; 2003. Fang, J.; Zhong, L.; Zhao, R.; Holmgren, A. Ebselen: a thioredoxin reductasedependent catalyst for α-tocopherol quinone reduction. Toxicol. Appl. Pharmacol. 207:103–109; 2005. Arscott, L. D.; Gromer, S.; Schirmer, R. H.; Becker, K.; Williams Jr C. H. The mechanism of thioredoxin reductase from human placenta is similar to the mechanisms of lipoamide dehydrogenase and glutathione reductase and is distinct from the mechanism of thioredoxin reductase from Escherichia coli. Proc. Natl. Acad. Sci. USA 94:3621–3626; 1997. Lu, J.; Holmgren, A. The thioredoxin antioxidant system. Free Radic. Biol. Med. 66:75–87; 2014. Saitoh, M.; Nishitoh, H.; Fujii, M.; Takeda, K.; Tobiume, K.; Sawada, Y.; Kawabata, M.; Miyazono, K.; Ichijo, H. Mammalian thioredoxin is a direct inhibitor of apoptosis signal-regulating kinase (ASK) 1. EMBO J. 17:2596–2606; 1998.

193

[66] Mitchell, D. A.; Morton, S. U.; Fernhoff, N. B.; Marletta, M. A. Thioredoxin is required for S-nitrosation of procaspase-3 and the inhibition of apoptosis in Jurkat cells. Proc. Natl. Acad. Sci. USA 104:11609–11614; 2007. [67] Matthews, J. R.; Wakasugi, N.; Virelizier, J. L.; Yodoi, J.; Hay, R. T. Thioredoxin regulates the DNA binding activity of NF-kappa B by reduction of a disulphide bond involving cysteine 62. Nucleic Acids Res. 20:3821–3830; 1992. [68] Anestal, K.; Prast-Nielsen, S.; Cenas, N.; Arner, E. S. Cell death by SecTRAPs: thioredoxin reductase as a prooxidant killer of cells. PLoS One 3:e1846; 2008. [69] Locksley, R. M.; Killeen, N.; Lenardo, M. J. The TNF and TNF receptor superfamilies: integrating mammalian biology. Cell 104:487–501; 2001. [70] Desagher, S.; Martinou, J. C. Mitochondria as the central control point of apoptosis. Trends Cell Biol. 10:369–377; 2000. [71] Boatright, K. M.; Salvesen, G. S. Mechanisms of caspase activation. Curr. Opin. Cell Biol. 15:725–731; 2003. [72] Hanahan, D.; Weinberg, R. A. Hallmarks of cancer: the next generation. Cell 144:646–674; 2011.