Toxicology Letters 170 (2007) 238–247
Curcumin activates human glutathione S-transferase P1 expression through antioxidant response element Toru Nishinaka a,∗ , Yusuke Ichijo b , Maki Ito b , Masayoshi Kimura b , Masato Katsuyama b , Kazumi Iwata b , Takeshi Miura a , Tomoyuki Terada a , Chihiro Yabe-Nishimura b a
Laboratory of Biochemistry, Faculty of Pharmacy, Osaka Ohtani University, 3-11-1 Nishikiori-kita, Tondabayashi, Osaka 584-8540, Japan b Department of Pharmacology, Kyoto Prefectural University of Medicine, Kawaramachi-Hirokoji, Kamigyo-ku, Kyoto, Kyoto 602-8566, Japan Received 30 October 2006; received in revised form 15 March 2007; accepted 15 March 2007 Available online 24 March 2007
Abstract Curcumin is a plant-derived diferuloylmethane compound extracted from Curcuma longa, possessing antioxidative and anticarcinogenic properties. Antioxidants and oxidative stress are known to induce the expression of certain classes of detoxification enzymes. Since the upregulation of detoxifying enzymes affects the drug metabolism and cell defense system, it is important to understand the gene regulation by such agents. In this study, we demonstrated that curcumin could induce the expression of human glutathione S-transferase P1 (GSTP1). In HepG2 cells treated with 20 M curcumin, the level of GSTP1 mRNA was significantly increased. In luciferase reporter assays, curcumin augmented the promoter activity of a reporter construct carrying 336 bp upstream of the 5 -flanking region of the GSTP1 gene. Mutation analyses revealed that the region including antioxidant response element (ARE), which overlaps AP1 in sequence, was essential to the response to curcumin. While the introduction of a wild-type Nrf2 expression construct augmented the promoter activity of the GSTP1 gene, co-expression of a dominant-negative Nrf2 abolished the responsiveness to curcumin. In addition, curcumin activated the expression of the luciferase gene from a reporter construct carrying multiple ARE consensus sequences but not one with multiple AP1 sites. In a gel mobility shift assay with an oligonucleotide with GSTP1 ARE, an increase in the amount of the binding complex was observed in the nuclear extracts of curcumin-treated HepG2 cells. These results suggested that ARE is the primary sequence for the curcumin-induced transactivation of the GSTP1 gene. The induction of GSTP1 may be one of the mechanisms underlying the multiple actions of curcumin. © 2007 Elsevier Ireland Ltd. All rights reserved. Keywords: Curcumin; Glutathione S-transferase P1; Antioxidant response element; Nrf2; AP1; Promoter
1. Introduction
∗ Corresponding author. Tel.: +81 721 24 9953; fax: +81 721 24 9953. E-mail address:
[email protected] (T. Nishinaka).
The expression of drug-metabolizing enzymes such as cytochrome P450 and glutathione S-transferases is known to be induced by various xenobiotic drugs and chemicals (Hayes and Pulford, 1995). One of the physiological roles of these enzymes is to protect cells against carcinogens, toxins, etc. On the other hand, resistance
0378-4274/$ – see front matter © 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.toxlet.2007.03.011
T. Nishinaka et al. / Toxicology Letters 170 (2007) 238–247
to anticancer drugs in tumor cells or tolerance to pharmaceutical agents is partly due to elevated levels of detoxification enzymes (Gottesman, 2002). Glutathione S-transferases (EC 2.5.1.18: GSTs) represent a group of multifunctional proteins which not only catalyze the conjugation of reduced glutathione to electrophilic xenobiotics but also have the capacity to bind various exogenous hydrophobic compounds as well as endogenous compounds (Hayes et al., 2005). GSTs are involved in cellular protection against xenobiotics and oxidative stress as well as resistance to chemotherapeutic compounds such as doxorubicin (Duvoix et al., 2004). Human cytosolic GSTs are highly polymorphic and can be divided into seven classes: Alpha, Mu, Pi, Theta, Sigma, Omega, and Zeta (Hayes et al., 2005; Mannervik et al., 2005; Nebert and Vasiliou, 2004). Pi class GSTs are known to be expressed in hepatocellular carcinomas (Yusof et al., 2003) and considered to be a tumor marker (Sato, 1989). The transcriptional regulation of a human Pi class GST, GSTP1, is reported to be mediated via the GC box, AP1 site, and NF-B site within the 5 -flanking region of its gene (Morceau et al., 2004; Xia et al., 1996). Antioxidant response element (ARE) is one of the cis-elements found in the promoter region of various detoxifying enzymes and antioxidative stress response proteins such as NADPH:quinone oxidoreductase (Favreau and Pickett, 1995), ␥-glutamylcysteine synthetase (Moinova and Mulcahy, 1999) and heme oxygenase-1 (Prestera et al., 1995). ARE is responsible for several groups of xenobiotics including phenolic antioxidants, flavonoids, and isothiocyanate, such as -naphthoflavone (-NF) and 2(3)-tert-butyl4-hydroxyanisole (t-BHA) (Jiang et al., 2003). The transcription factor Nrf2 (NF-E2-related factor 2), a member of the Cap ‘n’ Collar subfamily of basic regionleucine zipper (bZip) transcription factors, is known to form a heterodimer with small Maf proteins and bind to ARE (Itoh et al., 1995; Rushmore et al., 1991). Under the basal condition, Nrf2 binds to Kelch-like ECH-associated protein 1 (Keap1) in the cytoplasm. Furthermore, Keap1 facilitates the ubiquitination and degradation of Nrf2. Once electrophiles and oxidative stress disrupt the Nrf2-Keap1 complex, Nrf2 is allowed to translocate to the nucleus (Kobayashi et al., 2004). The core sequence of ARE, 5 -TGACNNNGC-3 , resembles the AP1 consensus sequence. In fact, the AP1 site of the human GSTP1 promoter (5 -TGACTCAGC-3 ) overlaps an ARE consensus sequence, suggesting that this site possibly acts as an ARE (Xia et al., 1996). Curcumin, an active yellow pigment of turmeric and curry, is a plant-derived diferuloylmethane compound
239
extracted from Curcuma longa, possessing antiinflammatory, antioxidative, and anticarcinogenic properties (Joe et al., 2004). Several mechanisms for the physiological actions of curcumin have been proposed (Leu and Maa, 2002; Leu et al., 2003; Lin et al., 2000). Curcumin is a potent inhibitor for reactive oxygengenerating enzymes such as lipoxygenase, and for signal transduction molecules such as epidermal growth factor (EGF) receptor kinase and IB kinase. Subsequently, curcumin has been shown to inhibit the expression of cjun, c-fos, and the inducible form of nitric oxide synthase (iNOS). As mentioned above, antioxidants such as t-BHA are known to induce the expression of certain classes of detoxification enzymes. Changes in the expression of detoxifying enzymes affect the drug metabolism and cell defense system. We have been studying the redox regulation of detoxification enzymes, and have shown that the gene expression of aldose reductase, a member of aldo-keto reductase superfamily, is upregulated by reactive oxygen species such as hydrogen peroxide (Nishinaka and Yabe-Nishimura, 2001). Thus, we are interested in antioxidant curcumin and initiated a search for curcumin-responsive gene in human liver. In this study, we demonstrated that curcumin augmented the expression of human-type GSTP1 and its promoter activity in a human hepatocellular carcinoma cell line, HepG2. Our results suggested that ARE-mediated transcription was involved in the curcumin-induced GSTP1 expression. 2. Materials and methods 2.1. Materials Curcumin was purchased from Nacalai Tesque, Inc. (Kyoto, Japan). Luciferase reporter plasmids, pTAL-luc and pTALAP1-luc, were obtained from BD Biosciences Clontech (Palo Alto, CA, USA). Anti-Nrf2 antibody (C-20X) was a product of Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA). Other reagents were of the highest grade available. 2.2. Plasmids The 5 -flanking region of human GSTP1 (−336 to +74) was amplified by PCR from HepG2 genomic DNA with primers, 5 -ggggtaccgcagcggtcttagggaatttc-3 (sense) and 5 -ctgcgggttggccccatgctgggagctc-3 (antisense). The DNA obtained was digested with KpnI and SacI, and inserted into the vector pGL3-basic (Promega Corp., Madison, WI, USA) to obtain −336-GSTP1-luc. The wild-type and mutant DNA fragments for the human GSTP1 promoter region (−78 to +74) were obtained by PCR using −336-GSTP1-luc as a
240
T. Nishinaka et al. / Toxicology Letters 170 (2007) 238–247
template with GL primer 2 (Promega) and the following primers: 5 -ggggtaccggcgccgtgactcagcactgg-3 (−78-GSTP1luc); 5 -ggggtaccggcgccgggactcagcactgg-3 (−78mt1-GSTP1luc); and 5 -ggggtaccggcgccgtgactccgcactgg-3 (−78mt2GSTP1-luc). The DNA fragments were digested with KpnI and SacI, and inserted into pGL3-basic. The nucleotide sequences of the PCR products were confirmed with a Prism 310 Sequencer (Applied Biosystems Japan, Ltd., Tokyo, Japan). The Nrf2 expression construct (pNrf2) was a gift from Dr. Cecil B. Picket (Schering-Plough Research Institute, NJ, USA) (Nguyen et al., 2000). The luciferase construct carrying 2xARE and the dominant-negative Nrf2 expression plasmid (DN-Nrf2) were prepared as described previously (Nishinaka and YabeNishimura, 2005). 2.3. Cell culture Human hepatoma HepG2 cells and human colorectal adenocarcinoma Caco2 cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) at 37 ◦ C under an atmosphere of 95% air and 5% CO2 . The cells were seeded in six-well plates 1 day before the transfection. The transfection was carried out using Fugene 6 reagent (Roche Diagnostics K.K., Tokyo, Japan), according to the manufacturer’s instructions. To normalize the transfection efficiency, the -galactosidase expression plasmid, pSV--GAL, was co-introduced into the cells.
with a luciferase assay system (Promega Corp.) with a Micro Lumat LB96P Luminometer (Berthold Japan, Co. Ltd., Tokyo). -Galactosidase activity was measured spectrophotometrically with a -galactosidase enzyme assay system (Promega Corp.). Luciferase activity was normalized to the -galactosidase activity for each sample. 2.6. Electrophoretic mobility shift assays (EMSAs) Fluorescein-labeled oligonucleotides containing human GSTP1 ARE/AP1 (5 -ccggcgccgtgactcagcactggggcg-3 and 5 -cgccccagtgctgagtcacggcgccgg-3 ) were prepared by Invitrogen K.K. Japan (Tokyo, Japan). The HepG2 nuclear extracts were prepared as described previously (Nishinaka and YabeNishimura, 2005). The nuclear extracts and the labeled probe were incubated at room temperature for 30 min, resolved in a polyacrylamide gel, and analyzed with a FLA-5100 fluoroimage analyzer (Fuji Film, Tokyo, Japan). Unlabeled probe or anti-Nrf2 antibody was incubated with nuclear extracts for 15 min prior to the addition of labeled probe. 2.7. Statistical analysis Statistical significance was assessed by the Student’s ttest. Values were expressed as means ± S.E.M. and differences between groups were considered to be significant at P < 0.05.
3. Results 2.4. RT-PCR
3.1. Curcumin increases the mRNA level of GSTP1 HepG2 cells were treated with 20 M curcumin for 8 h and then total RNA was prepared with Sepasol RNA extraction reagent (Nacalai Tesque, Inc.). Two hundred nanograms of total RNA was subjected to RT-PCR. The RT-PCR was carried out with a Qiagen one-step RT-PCR system (Qiagen K.K., Tokyo, Japan). The primers used for the detection of human GSTP1 were 5 -tccgctgcaaatacatctcc-3 (sense) and 5 tgtttcccgttgccattgat-3 (antisense) (Crawford et al., 2000). The reverse transcription was performed at 50 ◦ C for 30 min. PCR conditions were denaturing at 95 ◦ C for 15 min, then 35 cycles of denaturing at 94 ◦ C for 30 s, annealing at 55 ◦ C for 30 s, and elongation at 72 ◦ C for 1 min. The expression of GSTP1 was normalized to that of -actin amplified by 15 cycles of PCR using primers 5 -tgaaccccaaggccaaccgc-3 (sense) and 5 ttgtgctgggtgccagggca-3 (antisense). The 35th cycle for GSTP1 and the 15th cycle for -actin were confirmed to be in the exponential phase of PCR amplification by real-time PCR analyses. Real-time PCR was performed using QuantiTect SYBR Green RT-PCR system (Qiagen) with an Opticon 2 real-time PCR apparatus (Bio-Rad Laboratories, Inc., Tokyo, Japan). 2.5. Reporter assay The cells were treated with curcumin for 24 h after the transfection and then harvested. Luciferase activity was measured
Antioxidants such as t-BHA are known to induce the expression of certain classes of detoxification enzymes (Jiang et al., 2003). Since changes in the expression of detoxifying enzymes affect the drug metabolism and cell defense system, it is important to understand the gene regulation by such agents. In this study, we initiated a search for curcumin-responsive genes, and found that curcumin could induce human-type GSTP1 expression. In HepG2 cells treated with 20 M curcumin for 8 h, an approximately 2.5-fold increase in the GSTP1 mRNA level was detected using the RT-PCR method (Fig. 1A and B). A real-time PCR analysis also revealed a significant increase of 1.5-fold (Fig. 1C). 3.2. Curcumin activates GST-P1 promoter activity To determine the region of the human GSTP1 promoter responsive to curcumin, luciferase reporter analyses with respect to the 5 -flaking region of the GSTP1 gene were performed. Fig. 2 summarizes the luciferase reporter constructs used in this study. When the reporter construct carrying the region from −336 to +74 (−336-GSTP1-luc) was examined, the promoter
T. Nishinaka et al. / Toxicology Letters 170 (2007) 238–247
241
Fig. 1. (A) Expression of GSTP1 induced by curcumin. HepG2 cells were treated with 20 M curcumin for 8 h. Total RNA was extracted and subjected to RT-PCR. The data represent the results obtained from three independent treatments with curcumin or solvent alone. (B) Optical density measurements of GSTP1 and -actin bands using NIH Image software. Values relative to the intensity obtained in control cells (mean ± S.E.M.) are presented. (C) Real-time PCR analysis of GST-P1 induction by curcumin. The level of GSTP1 expression was normalized by that of -actin. * P < 0.05 (n = 3–5).
Fig. 2. Luciferase reporter constructs used in this study.
activity was found to increase in cells treated with 20 M curcumin by 2.8-fold (Fig. 3A). This region contains a consensus NF-B sequence and an AP1 sequence. The AP1 site also overlaps ARE consensus sequence. When the reporter construct with a shorter region, from −78
to +74 (−78-GSTP1-luc), lacking the NF-B sequence was analyzed, curcumin-induced promoter activation was still observed (Fig. 3A). Thus, we speculated that the AP1/ARE site might be involved in the response to curcumin.
242
T. Nishinaka et al. / Toxicology Letters 170 (2007) 238–247
3.3. Antioxidant response element-mediated transcriptional activation is involved in the induction of GSTP1 expression by curcumin The overlapping nucleotide sequence of AP1/ARE site of the human GSTP1 promoter, 5 -CGTGACTCAGCAC-3 (−71 to −59), includes consensus sequences for both palindromic AP1 (5 -TGACTCA-3 ) and the non-palindromic antioxidant response element, ARE (5 -TGACNNNGC-3 ). We also found that GSTP1 expression was augmented by an antioxidant, ethoxyquin (data not shown), which is known to activate ARE-dependent gene expression, indicating that GSTP1 expression could be regulated through ARE. In order to clarify whether this region functions as an AP1 site or ARE for curcumin, we prepared the AP1/ARE mutant constructs −78mt1-GSTP1-luc and −78mt2GSTP1-luc. The −78mt1 contains a critical mutation for both AP1 and ARE, and the −78mt2 contains a mutation more critical to AP1 (Favreau and Pickett, 1995). Both constructs significantly lost the responsiveness to curcumin, indicating that the AP1/ARE site is important. On the other hand, 78mt2-GSTP1-luc retained better responsiveness than −78mt1-GSTP1luc (Fig. 3B). These results suggested that ARE is the primary sequence for the curcumin-induced GSTP1
Fig. 3. AP1/ARE-mediated transcriptional activation of the human GSTP1 gene. (A) HepG2 cells were transfected with 0.5 g of −336GSTP1-luc or −78-GSTP1-luc. (B) HepG2 cells were transfected with 0.5 g of wild-type (−78-GSTP1-luc) or mutant (−78mt1-GSTP1-luc and −78mt2-GSTP1-luc) luciferase reporter construct. The transfected cells were treated with 20 M curcumin for 24 h, harvested, and then subjected to measurements of luciferase activity. Values indicate the fold-increase relative to the activity in the absence of curcumin with −336-GSTP1-luc (A) or −78-GSTP1-luc (B). Bars represent the mean ± S.E.M. obtained from three experiments. * P < 0.05, ** P < 0.01 vs. each control (n = 3–4).
Fig. 4. Dominant-negative Nrf2 suppressed the curcumin-induced activation of the GSTP1 promoter. HepG2 cells were co-transfected with 0.5 g of −336-GSTP1-luc, and 0.2 g of either dominantnegative Nrf2-expression construct (DN-Nrf2) or control plasmid (pcDNA3). The transfected cells were treated with 20 M curcumin for 24 h, harvested, and then subjected to measurements of luciferase activity. Values indicate the fold-increase relative to the activity with pcDNA3 in the absence of curcumin. Bars represent the mean ± S.E.M. obtained from three experiments. ** P < 0.01 vs. each control (n = 3–4).
T. Nishinaka et al. / Toxicology Letters 170 (2007) 238–247
243
Fig. 5. Curcumin augmented ARE-mediated transcriptional activation. (A) HepG2 cells were transfected with 0.5 g of 2xARE-luc. (B) HepG2 cells were transfected with 0.5 g of pTAL-AP1-luc. As a positive control, the pTAL-AP1-transfected Caco2 cells were treated with 0.1 g/ml phorbol 12-myristate 13-acetate (PMA) instead of curcumin. (C) HepG2 cells were co-transfected with 0.5 g of −78-GSTP1-luc, and 0.2 g of either pNrf2 or pcDNA3. The transfected cells were treated with 20 M curcumin for 24 h, harvested, and then subjected to measurements of luciferase activity. Values indicate the fold-increase relative to the activity with pcDNA3 in the absence of curcumin (A), the activity of each control (B), or the activity of control with pcDNA3 (C). Bars represent the mean ± S.E.M. obtained from three experiments. ** P < 0.01 (n = 3).
expression. In order to confirm the importance of ARE sequence, the effect of a dominant negative Nrf2 on the reporter assay was examined. The introduction of a dominant negative Nrf2 expression construct into the cell almost completely abolished the responsiveness to curcumin (Fig. 4). Next, we examined whether curcumin can activate ARE-mediated gene expression using
the reporter construct 2xARE-luc.This construct carries a tandem repeat of two ARE consensus sequences (5 -TGACCCAGC-3 ) derived from the mouse aldose reductase promoter, which does not perfectly match the typical AP1 sequence. Curcumin significantly augmented the reporter activity of 2xARE-luc, and this augmentation was suppressed by the introduction of
244
T. Nishinaka et al. / Toxicology Letters 170 (2007) 238–247
dominant negative Nrf2 (Fig. 5A). On the other hand, pTAL-AP1-luc, which contains six copies of the AP1 consensus sequence, did not respond as well as 2xAREluc to curcumin (Fig. 5B) while phorbol 12-myristate 13-acetate (PMA), an AP1 activator, could activate the reporter, indicating that pTAL-AP1 was functional (Fig. 5B). These results indicated that curcumin activated the human GSTP1 promoter through ARE but not AP1. In order to further clarify whether the transcription factor Nrf2 can activate the GSTP1 promoter, an Nrf2 expression plasmid was co-introduced in the luciferase reporter assay. As shown in Fig. 5C, overexpression of Nrf2 could activate the GSTP1 promoter. Curcumin also further augmented the luciferase activity under Nrf2 overexpression. 3.4. Curcumin augments the DNA-binding activity for ARE Next, to examine whether curcumin increases the binding activity for the GSTP1 ARE in HepG2, electrophoretic mobility shift assay was performed. Augmented levels of the ARE-binding complex were observed in the nuclear extracts prepared from curcumintreated cells (lane 2) compared with those from vehicle-treated cells (lane 1), suggesting that curcumin activated ARE-mediated transcription (Fig. 6). When anti-Nrf2 antibody was added to the reaction mixture, the binding of nuclear factors to the GSTP1 ARE was markedly attenuated, suggesting that the antibody prevented Nrf2 protein from binding to the probe (lane 3). We previously demonstrated a similar result that the same antibody inhibited the formation of binding complex with aldose reductase ARE probe using HepG2 nuclear extracts (Nishinaka and Yabe-Nishimura, 2005). These results indicate that curcumin activates Nrf2 to bind to GSTP1 ARE. 4. Discussion In this study, we demonstrated that curcumin could induce the gene expression of human-type GSTP1 in a human hepatocellular carcinoma cell line, HepG2, by activating transcription via ARE within its 5 -flanking region. The ARE consensus sequence spans nucleotide −69 to −61 from the transcription start site of the human GSTP1 gene. Since this sequence contains a complete AP1 consensus sequence, it is questioned whether this region works as an ARE or AP1 for the curcumininduced activation of the promoter. We concluded that this region functions as ARE. The major lines of evidence provided in this study are as follows: (1) The curcumin-
Fig. 6. Electrophoretic mobility shift assay. HepG2 nuclear extracts (5 g) were incubated with a fluorescein-labeled probe containing the GSTP1 ARE sequence and the DNA-protein complex was resolved in a 6% polyacrylamide gel. Anti-Nrf2 antibody (6 g) was added to the reaction mixture and subjected to the assay (Anti-Nrf2). The rabbit IgG was used for negative control (rb. IgG). An autoradiograph representative of two experiments is shown.
induced activation was suppressed by the introduction of dominant negative Nrf2. (2) Introduction of Nrf2 protein could activate the promoter. (3) Curcumin activated the gene expression from the reporter construct carrying two ARE consensus sequences, 2xARE-luc, while it only slightly activated the expression from the one with multiple AP1 consensus sequences, pTAL-AP1-luc. The presence of a functional ARE has been reported for the genes of class Pi GSTs of mouse and rat (Ikeda et al., 2002, 2004). Curcumin has been shown to induce GSTP1 expression in a human breast cancer cell line, MCF-7, by a microarray analysis (Ramachandran et al., 2005) and in the liver of curcumin-fed mice (Singh et al., 1998). Our present study also revealed that the expression of GSTP1 is induced by curcumin in human hepatic cells and demonstrated the involvement of transcription factor Nrf2 in the curcumin-induced GSTP1 expression. Curcumin is known to possess many physiological effects such as anti-inflammatory and anti-tumor activities in addition to its role as an antioxidant (Joe et al., 2004). However, the mechanisms of its actions have not been well defined. In this study, we demonstrated that
T. Nishinaka et al. / Toxicology Letters 170 (2007) 238–247
curcumin activated the ARE-mediated transcription of a phase II detoxification enzyme, GSTP1, in HepG2 cells. The result suggested that induction of detoxification enzymes via ARE-mediated transcriptional activation is one of the mechanisms underlying the multiple functions of curcumin. Balogum, et al. reported that curcumin activated heme oxygenase-1 gene expression in an AREdependent manner in renal epithelial cells (Balogun et al., 2003). Thus, the activation of ARE-mediated transcription appears to be a common function of curcumin. They also showed that the p38 MAP kinase pathway was involved in the curcumin-mediated activation of Nrf2 (Balogun et al., 2003). In our study, we examined the effects of dominant-negative MKK3 and MKK6, upstream kinases for p38, on the curcumininduced transcriptional activation through ARE. These dominant-negative proteins, however, failed to suppress the action of curcumin. In addition, neither dominantnegative MEK1, an upstream kinase of ERK, nor dominant-negative SEK1, an upstream kinase of SAPK, could suppress the effect of curcumin (data not shown). These results suggested that MAP kinase pathways were not involved in the curcumin-induced activation of Nrf2 in HepG2 cells. Nrf2 is known to be retained in the cytoplasm under the basal condition by interacting with Keap1 that also facilitates ubiquitination and degradation of Nrf2. Electrophiles such as t-BHA and oxidative stress disrupt the Nrf2-Keap1 complex, allowing Nrf2 to translocate to the nucleus (Kobayashi et al., 2004). Since curcumin could still augment luciferase reporter activity in the Nrf2 overexpressing cells (Fig. 5C), curcumin may directly interact with Keap1 protein to inhibit the degradation of Nrf2 by proteasomes and facilitate the nuclear translocation of Nrf2. Further investigation is needed to clarify the activation mechanisms of Nrf2 by curcumin in these cells. Several mechanisms for the physiological actions of curcumin have been proposed. Curcumin itself is thought to act as an antioxidant (Leu and Maa, 2002). Curcumin has been shown to directly inhibit protein kinases such as EGF receptor kinase (Leu et al., 2003). It was also shown to have an inhibitory effect on NF-B- or AP1mediated transcription (Lin et al., 2000). The expression of human GSTP1 is known to be regulated by NF-B and AP1 (Morceau et al., 2004; Xia et al., 1996). Duvoix et al. reported that curcumin inhibited the TNF-␣-induced expression of GSTP1 by inhibiting these transcription factors in leukemia cells (Duvoix et al., 2003). Their observation appears to contradict our results. However, there was a marked difference in the experimental conditions used. In our study, we examined the direct effect of curcumin on the cells without any additional treatment
245
while they tested the effect of curcumin on TNF-␣ treatment by which the activities of AP1 and NF-B were augmented and GSTP1 expression was increased. Thus, our results simply demonstrate that curcumin alone was able to activate ARE-mediated transcription. Since Nrf2 and AP1 may compete with each other to bind DNA because of the similarity of their recognition sequences, curcumin may inhibit AP1-mediated transcription by activating Nrf2 in cells where AP1 is more dominantly involved in GSTP1 expression than Nrf2. There are other aspects of the interaction between curcumin and GSTP1. Iesrel et al. reported that curcumin inactivates human GSTP1 via covalent modification (Iersel et al., 1996). In addition, GSTP1 is known to catalyze the conjugation of curcumin with glutathione (Awasthi et al., 2000). These findings indicate a highly complex relationship between curcumin and GSTP1. That is, curcumin can either activate the GSTP1 expression or inactivate GSTP1 enzyme activity while curcumin is metabolized by the enzyme. Apparent effect of curcumin on GSTP1 activity may depend on the amount of curcumin incorporated into the cells. Further investigation will be needed to clarify the relationship between GSTP1 activity and cellular concentration of curcumin. Nrf2 is a major protein that binds to ARE and activates ARE-dependent transcription. Microarray analyses of Nrf2-knockout mice revealed that Nrf2 is involved in the regulation of glutathione-related proteins, antioxidative proteins such as ferritin, NADPH-producing enzymes such as glucose 6-phosphate dehydrogenase, and antiinflammatory proteins (Leu et al., 2003; Thimmulappa et al., 2002). These proteins play important roles in cell protection. Accordingly, the actions of curcumin appear to show a good correlation with the function of Nrf2. The physiological role of the expression of drug-metabolizing enzymes including GST is to protect cells against the adverse effects of compounds such as carcinogens and toxins. The ARE-mediated promoter activation of the detoxifying enzymes can account for the chemopreventive and anti-tumorigenic actions of curcumin. The expression of cell-protective proteins via ARE-mediated transcriptional activation may be one of the mechanisms underlying the multiple actions of curcumin. Acknowledgments This work was supported in part by grant-in-aid for young scientists (B) 14771339 from the Ministry of Education, Culture, Sports, Science, and Technology, Japan (T.N.), and by research grant from the Japan food chemical research foundation (T.T.) and Osaka
246
T. Nishinaka et al. / Toxicology Letters 170 (2007) 238–247
Ohtani University research fund (Pharmaceutical Science) (T.T.). The authors thank Dr Cecil B. Pickett (Schering-Plough Research Institute, NJ, USA) for providing the Nrf2 expression construct. References Awasthi, S., Pandya, U., Singhal, S.S., Lin, J.T., Thiviyanathan, V., Seifert Jr., W.E., Awasthi, Y.C., Ansari, G.A., 2000. Curcuminglutathione interactions and the role of human glutathione S-transferase P1-1. Chem. Biol. Interact. 128, 19–38. Balogun, E., Hoque, M., Gong, P., Killeen, E., Green, C.J., Foresti, R., Alam, J., Motterlini, R., 2003. Curcumin activates the haem oxygenase-1 gene via regulation of Nrf2 and the antioxidantresponsive element. Biochem. J. 371, 887–895. Crawford, E.L., Khuder, S.A., Durham, S.J., Frampton, M., Utell, M., Thilly, W.G., Weaver, D.A., Ferencak, W.J., Jennings, C.A., Hammersley, J.R., Olson, D.A., Willey, J.C., 2000. Normal bronchial epithelial cell expression of glutathione transferase P1, glutathione transferase M3, and glutathione peroxidase is low in subjects with bronchogenic carcinoma. Cancer Res. 60, 1609–1618. Duvoix, A., Delhalle, S., Blasius, R., Schnekenburger, M., Morceau, F., Fougere, M., Henry, E., Galteau, M.M., Dicato, M., Diederich, M., 2004. Effect of chemopreventive agents on glutathione Stransferase P1-1 gene expression mechanisms via activating protein 1 and nuclear factor kappaB inhibition. Biochem. Pharmacol. 68, 1101–1111. Duvoix, A., Morceau, F., Delhalle, S., Schmitz, M., Schnekenburger, M., Galteau, M.M., Dicato, M., Diederich, M., 2003. Induction of apoptosis by curcumin: mediation by glutathione S-transferase P1-1 inhibition. Biochem. Pharmacol. 66, 1475–1483. Favreau, L.V., Pickett, C.B., 1995. The rat quinone reductase antioxidant response element. Identification of the nucleotide sequence required for basal and inducible activity and detection of antioxidant response element-binding proteins in hepatoma and non-hepatoma cell lines. J. Biol. Chem. 270, 24468–24474. Gottesman, M.M., 2002. Mechanisms of cancer drug resistance. Annu. Rev. Med. 53, 615–627. Hayes, J.D., Flanagan, J.U., Jowsey, I.R., 2005. Glutathione transferases. Annu. Rev. Pharmacol. Toxicol. 45, 51–88. Hayes, J.D., Pulford, D.J., 1995. The glutathione S-transferase supergene family: regulation of GST and the contribution of the isoenzymes to cancer chemoprotection and drug resistance. Crit. Rev. Biochem. Mol. Biol. 30, 445–600. Iersel, M.L., Ploemen, J.P., Struik, I., van Amersfoort, C., Keyzer, A.E., Schefferlie, J.G., van Bladeren, P.J., 1996. Inhibition of glutathione S-transferase activity in human melanoma cells by alpha,beta-unsaturated carbonyl derivatives. Effects of acrolein, cinnamaldehyde, citral, crotonaldehyde, curcumin, ethacrynic acid, and trans-2-hexenal. Chem. Biol. Interact. 102, 117–132. Ikeda, H., Nishi, S., Sakai, M., 2004. Transcription factor Nrf2/MafK regulates rat placental glutathione S-transferase gene during hepatocarcinogenesis. Biochem. J. 380, 515–521. Ikeda, H., Serria, M.S., Kakizaki, I., Hatayama, I., Satoh, K., Tsuchida, S., Muramatsu, M., Nishi, S., Sakai, M., 2002. Activation of mouse Pi-class glutathione S-transferase gene by Nrf2(NF-E2-related factor 2) and androgen. Biochem. J. 364, 563–570. Itoh, K., Igarashi, K., Hayashi, N., Nishizawa, M., Yamamoto, M., 1995. Cloning and characterization of a novel erythroid cellderived CNC family transcription factor heterodimerizing with the small Maf family proteins. Mol. Cell. Biol. 15, 4184–4193.
Jiang, Z.Q., Chen, C., Yang, B., Hebbar, V., Kong, A.N., 2003. Differential responses from seven mammalian cell lines to the treatments of detoxifying enzyme inducers. Life Sci. 72, 2243–2253. Joe, B., Vijaykumar, M., Lokesh, B.R., 2004. Biological properties of curcumin-cellular and molecular mechanisms of action. Crit. Rev. Food Sci. Nutr. 44, 97–111. Kobayashi, A., Kang, M.I., Okawa, H., Ohtsuji, M., Zenke, Y., Chiba, T., Igarashi, K., Yamamoto, M., 2004. Oxidative stress sensor Keap1 functions as an adaptor for Cul3-based E3 ligase to regulate proteasomal degradation of Nrf2. Mol. Cell. Biol. 24, 7130–7139. Leu, T.H., Maa, M.C., 2002. The molecular mechanisms for the antitumorigenic effect of curcumin. Curr. Med. Chem. Anti-Canc. Agents 2, 357–370. Leu, T.H., Su, S.L., Chuang, Y.C., Maa, M.C., 2003. Direct inhibitory effect of curcumin on Src and focal adhesion kinase activity. Biochem. Pharmacol. 66, 2323–2331. Lin, J.K., Pan, M.H., Lin-Shiau, S.Y., 2000. Recent studies on the biofunctions and biotransformations of curcumin. Biofactors 13, 153–158. Mannervik, B., Board, P.G., Hayes, J.D., Listowsky, I., Pearson, W.R., 2005. Nomenclature for mammalian soluble glutathione transferases. Meth. Enzymol. 401, 1–8. Moinova, H.R., Mulcahy, R.T., 1999. Up-regulation of the human gamma-glutamylcysteine synthetase regulatory subunit gene involves binding of Nrf-2 to an electrophile responsive element. Biochem. Biophys. Res. Commun. 261, 661–668. Morceau, F., Duvoix, A., Delhalle, S., Schnekenburger, M., Dicato, M., Diederich, M., 2004. Regulation of glutathione S-transferase P1-1 gene expression by NF-kappaB in tumor necrosis factor alphatreated K562 leukemia cells. Biochem. Pharmacol. 67, 1227– 1238. Nebert, D.W., Vasiliou, V., 2004. Analysis of the glutathione Stransferase (GST) gene family. Hum. Genom. 1, 460–464. Nguyen, T., Huang, H.C., Pickett, C.B., 2000. Transcriptional regulation of the antioxidant response element. Activation by Nrf2 and repression by MafK. J. Biol. Chem. 275, 15466–15473. Nishinaka, T., Yabe-Nishimura, C., 2001. EGF receptor-ERK pathway is the major signaling pathway that mediates upregulation of aldose reductase expression under oxidative stress. Free Radic. Biol. Med. 31, 205–216. Nishinaka, T., Yabe-Nishimura, C., 2005. Transcription factor Nrf2 regulates promoter activity of mouse aldose reductase (AKR1B3) gene. J. Pharmacol. Sci. 97, 43–51. Prestera, T., Talalay, P., Alam, J., Ahn, Y.I., Lee, P.J., Choi, A.M., 1995. Parallel induction of heme oxygenase-1 and chemoprotective phase 2 enzymes by electrophiles and antioxidants: regulation by upstream antioxidant-responsive elements (ARE). Mol. Med. 1, 827–837. Ramachandran, C., Rodriguez, S., Ramachandran, R., Raveendran Nair, P.K., Fonseca, H., Khatib, Z., Escalon, E., Melnick, S.J., 2005. Expression profiles of apoptotic genes induced by curcumin in human breast cancer and mammary epithelial cell lines. Anticanc. Res. 25, 3293–3302. Rushmore, T.H., Morton, M.R., Pickett, C.B., 1991. The antioxidant responsive element. Activation by oxidative stress and identification of the DNA consensus sequence required for functional activity. J. Biol. Chem. 266, 11632–11639. Sato, K., 1989. Glutathione transferases as markers of preneoplasia and neoplasia. Adv. Cancer Res. 52, 205–255. Singh, S.V., Hu, X., Srivastava, S.K., Singh, M., Xia, H., Orchard, J.L., Zaren, H.A., 1998. Mechanism of inhibition of benzo[a]pyreneinduced forestomach cancer in mice by dietary curcumin. Carcinogenesis 19, 1357–1360.
T. Nishinaka et al. / Toxicology Letters 170 (2007) 238–247 Thimmulappa, R.K., Mai, K.H., Srisuma, S., Kensler, T.W., Yamamoto, M., Biswal, S., 2002. Identification of Nrf2-regulated genes induced by the chemopreventive agent sulforaphane by oligonucleotide microarray. Cancer Res. 62, 5196–5203. Xia, C., Hu, J., Ketterer, B., Taylor, J.B., 1996. The organization of the human GSTP1-1 gene promoter and its response to retinoic
247
acid and cellular redox status. Biochem. J. 313 (Pt. 1), 155– 161. Yusof, Y.A., Yan, K.L., Hussain, S.N., 2003. Immunohistochemical expression of pi class glutathione S-transferase and alphafetoprotein in hepatocellular carcinoma and chronic liver disease. Anal. Quant. Cytol. Histol. 25, 332–338.
