Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
BBRC Biochemical and Biophysical Research Communications 305 (2003) 662–670 www.elsevier.com/locate/ybbrc Induction of phase 2 enzymes by serum oxidized polyamines through activation of Nrf2: effect of the polyamine metabolite acroleinq Mi-Kyoung Kwak,a,* Thomas W. Kensler,a,b and Robert A. Casero Jr.a,b a b Department of Environmental Health Sciences, Johns Hopkins University, Bloomberg School of Public Health, 615 N. Wolfe St., Baltimore, MD 21205, USA Department of Oncology, Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins School of Medicine, Baltimore, MD, USA Received 22 April 2003 Abstract The naturally occurring polycationic polyamines including putrescine, spermidine, and spermine play an important role in cell growth, differentiation, and gene expression. However, circulating polyamines are potential substrates for several oxidizing enzymes including copper-containing serum amine oxidase. These enzymes are capable of oxidizing serum polyamines to several toxic metabolites including aldehydes and H2 O2 . In this study, we investigated the effects of polyamines as inducers of phase 2 enzymes and other genes that promote cell survival in a cell culture system in the presence of bovine serum. Spermidine and spermine (50 lM) increased NAD(P)H quinone oxidoreductase (NQO1) activity up to 3-fold in murine keratinocyte PE cells. Transcript levels for glutathione S-transferase (GST) A1, GST M1, NQO1, c-glutamylcysteine ligase regulatory subunit, and UDP-glucuronyltransferase 1A6 were significantly increased by spermidine and this effect was mediated through the antioxidant response element (ARE). The ARE from the mouse GST A1 promoter was activated about 9-fold by spermine and 5-fold by spermidine treatment, but could be inhibited by the amine oxidase inhibitor, aminoguanidine, suggesting that acrolein or hydrogen peroxide generated from polyamines by serum amine oxidase may be mediators for phase 2 enzyme induction. Elevations of ARE-luciferase expression and NQO1 enzyme activity by spermidine were not affected by catalase, while both were completely repressed by aldehyde dehydrogenase treatment. Direct addition of acrolein to PE cells induced multiple phase 2 genes and elevated nuclear levels of Nrf2, a transcription factor that binds to the ARE. Expression of mutant Nrf2 repressed the activation of the ARE-luciferase reporter by polyamines and acrolein. These results indicate that spermidine and spermine increase the expression of phase 2 genes in cells grown in culture through activation of the Nrf2–ARE pathway by generating the sulfhydryl reactive aldehyde, acrolein. Ó 2003 Elsevier Science (USA). All rights reserved. Keywords: Polyamines; Acrolein; Antioxidant response element; Nrf2; Phase 2 detoxifying enzymes The polyamines putrescine, spermidine, and spermine are naturally occurring polycations that play an important role in cell growth and differentiation [1,2]. Many studies using inhibitors of polyamine synthesis and polyamine deficient mutants have established the fact that polyamines are essential for cell growth [1–3]. Polyamines can alter the expression of genes; for exq Abbreviations: ARE, antioxidant response element; GST, glutathione S-transferase; NQO1, NAD(P)H:quinone oxidoreductase; UGT1A6, UDP-glucuronyltransferase; cGCL, c-glutamylcysteine ligase; MnSOD, Mn-superoxide dismutase; NF-jB, nuclear factor jB; FBS, fetal bovine serum; PUT, putrescine; SPD, spermidine; SPM, spermine; RT-PCR, reverse transcriptase polymerase chain reaction. * Corresponding author. Fax: 1-410-955-0116. E-mail address: [email protected] (M.-K. Kwak). ample, transcriptional expression of c-myc and c-fos was facilitated by polyamines [4,5]. Other reports have shown that polyamines are involved in the regulation of c-jun and the heat shock protein hsp 70 in animal cells, and the recA gene in Escherichia coli [6,7]. Spermidine and spermine can be oxidized by several intracellular and extracellular oxidases. The classical intracellular polyamine oxidase (PAO) is a FAD-dependent oxidase that preferentially oxidizes N1 -acetylated spermine and spermidine [8,9] and its activity is rate limited by the availability of its acetylated substrate that is produced by spermidine/spermine N1 -acetyltransferase [10]. A newly discovered intracellular oxidase, PAOh1/SMO, is also a FAD-dependent oxidase; however, it preferentially oxidizes spermine to 0006-291X/03/$ - see front matter Ó 2003 Elsevier Science (USA). All rights reserved. doi:10.1016/S0006-291X(03)00834-9 M.-K. Kwak et al. / Biochemical and Biophysical Research Communications 305 (2003) 662–670 spermidine, 3-aminopropanal, and H2 O2 and does not oxidize spermidine [11,12]. The primary extracellular polyamine-oxidizing enzyme is the copper-containing serum amine oxidase found in bovine serum [13–15]. Amine oxidase in bovine serum can oxidize spermidine and spermine to aminoaldehydes and hydrogen peroxide [1,15]. Recently Sharmin et al. [16] showed that aminoaldehydes formed from polyamines by this oxidase spontaneously undergo b-elimination to form acrolein and putrescine. Similar products are formed from polyamines by another family of copper-containing oxidases, the diamine oxidases (DAO) that are found in serum and many tissues [14,17–19]. It is known that hydrogen peroxide and aminoaldehyde produced from polyamines oxidized intracellularly or extracellularly can cause cytotoxicity [20–22]. Therefore, it is not surprising that cells would possess mechanisms to protect themselves from oxidative damage induced by polyamine metabolites. Nrf2 is a basic-leucine zipper transcription factor that binds to the antioxidant response element (ARE), which is found in the promoter region of various phase 2 detoxifying genes [23,24]. The ARE is a central regulatory element in the expression of basal and inducible phase 2 genes, which can protect cells from environmental stresses [25]. Studies with knockout mice indicate that Nrf2 is an essential component in the transcription complex on the ARE for the regulation of phase 2 and antioxidant genes containing this element [14,26–28]. The actin-bound protein Keap1 acts as a repressor of Nrf2 by binding to its N-terminal domain and retaining it in the cytosol [29]. Electrophiles and oxidants interact with cysteine sulfhydryls in Keap1 allowing the release of Nrf2, its translocation to the nucleus, and transactivation of the ARE of phase 2 genes [30]. In this study, exposure of cells to the polyamines spermidine and spermine increased multiple phase 2 enzymes in murine keratinocytes grown in the presence of serum. The mediator of this enzyme induction was acrolein, a catabolic metabolite of polyamines, produced by serum amine oxidase. Acrolein, a highly reactive a; b-unsaturated aldehyde, induced phase 2 genes through the ARE. The rapid accumulation of Nrf2 following treatment with acrolein implicates the involvement of Nrf2 in the action of acrolein. Also, overexpression of mutant Nrf2 repressed the activation of the ARE by acrolein. These results indicate that the oxidation of spermidine and spermine by amine oxidases can induce phase 2 enzymes by generating acrolein and triggering Nrf2 activation. Materials and methods Cell culture and treatment. Murine keratinocyte PE cells [31] were maintained in EagleÕs minimum essential media containing 10% of 663 heat-inactivated and chelex-treated fetal bovine serum (FBS, Life Technologies, Gaithersburg, MD), 2 mM CaCl2 , and antimycotics/ antibiotics (Life Technologies). Cells were treated with putrescine, spermidine, or spermine for 18 h. When indicated aminoguanidine, catalase (Calbiochem, San Diego, CA), or aldehyde dehydrogenase (Sigma, St. Louis, MO) was coincubated with polyamine-treated cells for 18 h. Measurement of NQO1 activity and intracellular polyamine concentrations. PE cells were plated on 96-well plates at a density of 3000 cells/well and grown at 37 °C for up to 48 h. When cells became 80% confluent, the polyamines (50 lM) were added alone or together with aminoguanidine for 18 h. The cells were lysed by repeated freeze– thawing and the NQO1 activity was measured using menadione as substrate [32]. Protein concentration was determined with the bicinchoninic acid protein assay (Pierce, Rockford, IL). Intracellular polyamine concentrations were measured by the methods of Kabra et al. [33]. Total RNA isolation and RT-PCR analysis. Cells were treated with 50 lM spermidine, spermidine with aminoguanidine (100 lM), or acrolein (10 or 25 lM) for 18 h. Total RNA was isolated by the procedure of Chomcynski and Sacchi [34]. RNA samples were electrophoresed on 1% agarose gels containing 2.2 M formaldehyde and transferred to nylon membranes (Schleicher and Schuell, Keene, NH). cDNAs for rat GST A1 were labeled with [a-32 P]dCTP using a random primer labeling kit (Amersham–Pharmacia Biotech., Piscataway, NJ), hybridized, and washed. After washing, the membranes were exposed to X-ray film (Eastman Kodak, Rochester, NY) and developed using a Konica film processor (Shinjuku-ku, Tokyo, Japan). Analyses of mRNA levels of NQO1, UGT1A6, c-glutamylcysteine ligase, Mn-superoxide dismutase, catalase, and b-actin were performed using RT-PCR. For the synthesis of cDNA, 100 ng total RNA was incubated with 10 mM Tris, 5 mM KCl, 5 mM MgCl2 , 4 mM dNTPs, 0.125 lg oligo(dT)12–18 , and 30 U M-MRV (Moloney Murine Leukemia Virus) reverse transcriptase (Life Technologies) for 15 min. PCR amplification was performed by incubating cDNA with gene specific PCR primers as described previously [26]. PCR products were electrophoresed on 1.5% agarose gels and the gel image was quantified using a Un-scan-it gel image analysis program (Silk Scientific, Orem, Utah). Preparation of nuclear extracts and Western blot analysis. Nuclear extracts from PE cells were prepared as described previously [26], loaded on a 6% of SDS–polyacrylamide gel, and separated by electrophoresis. Immunoblotting was carried out using Nrf2 antibodies recognizing the N-terminal of murine Nrf2. Immunoblotted membranes were developed using the ECL Western blotting system (Amersham–Pharmacia Biotech.) according to the manufacturerÕs instructions. Transient transfection of plasmids and measurement of luciferase activity. A DNA sequence containing the GST A1 ARE ()833 to )533 from the start codon) was prepared by PCR from the mouse GST A1 promoter ()1094 to )10) [35], which was isolated from mouse brain cDNA and inserted into a luciferase reporter vector (ARE-TATALucþ ). Cells were plated on 12-well plates at a density of 2–3 104 cells/well. The cells were grown overnight and the transfection complex containing 0.7 lg plasmid DNA, 0.05 lg pRLtk, and lipofectamine reagent (Life Technologies) was added to each well. Twenty-four hours after transfection, spermidine or spermine was added to cells, which were then lysed 18 h later. Luciferase activity was measured using the Dual Luciferase Assay kit (Promega, Madison, WI) with a luminometer (Turner Design, Sunnyvale, CA). ARE-luciferase activities were normalized using Renilla luciferase activity as an internal control. For the overexpression of mutant Nrf2 in cells, a mutant Nrf2 construct was generated by excluding the transactivation domain as described previously [36]. 664 M.-K. Kwak et al. / Biochemical and Biophysical Research Communications 305 (2003) 662–670 Results Increase of NQO1 enzymatic activity by spermidine and spermine in cell culture system—inhibition by amine oxidase inhibitor, aminoguanidine Murine keratinocyte PE cells were incubated with 50 lM putrescine, spermidine, or spermine for 18 h and the specific activity of NQO1 was measured. Spermidine and spermine increased NQO1 activity up to 3-fold; however, putrescine did not change the enzymatic activity (Fig. 1A). It is important to note that the exposure of PE cells to the various polyamines resulted in no significant changes in intracellular polyamine pools. Putrescine was not detected in PE cells and the extracellular addition of spermidine and spermine did not make a remarkable difference in the intracellular polyamine levels in the presence or absence of fetal bovine serum (Table 1). GST activity was also increased by 70% by spermidine (data not shown). These effects were serum-dependent as NQO1 activity was not increased by any of the polyamines in the absence of FBS. To clarify the enhancing effect of serum, the serum amine oxidase inhibitor, aminoguanidine, was incubated with polyamines. Coincubation of aminoguanidine (100 lM) with spermidine or spermine completely blocked the induction of NQO1 activity by these polyamines (Fig. 1A). This result indicates that metabolites of polyamines produced by amine oxidase might mediate the induction of NQO1 and GST. Polyamine exposure induces the expression of multiple detoxifying enzymes through the Nrf2–ARE pathway Spermidine, either with or without aminoguanidine, was incubated with PE cells in the presence of bovine serum, and mRNAs for detoxifying genes, catalase, and superoxide dismutase were measured using RT-PCR. Spermidine (50 lM) increased the transcript levels for GST A1, NQO1, and the light chain of glutamylcysteine ligase by 7.8-, 9.7-, and 5.9-fold, respectively (Table 2). Levels of GST M1 and UDP-glucuronyltransferase 1A6 were also significantly elevated. These inducible effects were inhibited substantially by treatment with aminoguanidine. Transcripts of antioxidant enzymes, heme oxygenase-1, and ferritin light chain were also increased Fig. 1. Effects of polyamines on NQO1 specific activity and GST A1 ARE-luciferase reporter activity. (A) Effect of polyamines on NQO1 enzyme activity. PE cells were incubated with 50 lM putrescine (PUT), spermidine (SPD), or spermine (SPM) with or without the amine oxidase inhibitor aminoguanidine (100 lM) in the presence or absence of fetal bovine serum (10% FBS) for 18 h. Values are means SE from three experiments and are expressed as ratios of treated to vehicle control. a P < 0:05 compared to vehicle-treated control. (B) Effects of polyamines on murine GST A1 ARE-luciferase reporter activity. Spermidine (SPD) and spermine (SPM) were incubated at a concentration of 50 lM without or with aminoguanidine (AG, 100 lM). Luciferase activities were normalized by cotransfecting Renilla luciferase control vectors. Values are means SE (n ¼ 3) and expressed as ratios of treated over vehicle control. a P < 0:05 compared to vehicle-treated control; b P < 0:05 compared to spermine alone treated group; and c P < 0:05 compared to spermidine alone treated group. Levels of mRNA for GST A1 were also measured by Northern blot. Total RNA was isolated from PE cells 18 h after administration of 50 lM SPD or SPM. (C) Nrf2 levels were determined in nuclear extracts by immunoblot analysis following treatment with vehicle (control), SPM (50 lM), and SPD (50 lM) for 6 h. M.-K. Kwak et al. / Biochemical and Biophysical Research Communications 305 (2003) 662–670 665 Table 1 Concentration of polyamines in cells following treatment with 50 lM putrescine (PUT), spermidine (SPD), and spermine (SPM) with or without amine oxidase inhibitor aminoguanidine (AG, 100 lM) in the presence or absence of fetal bovine serum (10%) for 18 h Serum Treatment No serum FBS (10%) FBS (10%) Polyamines (nmol/mg protein) PUT SPD SPM Vehicle PUT SPD SPM — 10.81 0.51 9.88 0.47 10.04 0.14 11.17 0.67 12.05 0.99 12.68 0.95 11.62 1.44 12.18 1.46 Vehicle PUT SPD SPM — 10.09 0.45 11.85 0.14 10.85 0.78 11.95 1.22 10.47 0.14 9.26 0.34 11.56 0.71 9.28 0.66 Vehicle + AG PUT + AG SPD + AG SPM + AG — 8.74 0.13 12.50 0.75 11.46 0.67 12.25 0.24 10.01 0.61 8.34 0.18 8.56 0.47 8.93 0.39 — — — — — — — — — Values are means SD from four measurements. Table 2 Induction of mRNAs for phase 2 and antioxidant enzymes following treatment with spermidine (SPD, 50 lM) or aminoguanidine (AG, 100 lM) + SPD for 18 h in PE cells Genes SPD SPD + AG GST A1 GST M1 NQO1 UGT1A6 cGCSr MnSOD Catalase 7.87a 2.16a 9.66a 2.66a 5.89a 1.12 1.29 2.06b 0.88b 2.82b 1.63b 2.29b 1.01 1.08 Values are the ratios of treated over vehicle control levels and are expressed as mean of three samples per group. Levels of RNA for each gene were normalized to b-actin mRNA levels. a P < 0:05, compared to vehicle-treated control. b P < 0:05, compared to SPD-treated group. by spermidine (data not shown). However, levels of mRNA for Mn-superoxide dismutase and catalase were not increased by treatment with these reagents in this cell system. To clarify whether the inductive effect of polyamine metabolites on phase 2 and antioxidative genes is mediated through the ARE, murine GST A1 ARE-luciferase reporter activity was measured following treatment of PE cells with polyamines. Spermidine and spermine exposure increased ARE-luciferase activity by 6- and 10fold, respectively, and these effects were largely repressed by the coincubation with aminoguanidine (Fig. 1B). In accord with the activation of the ARE, GST A1 mRNA levels were increased 8–10-fold by treatment with spermidine and spermine, and this accumulation of transcripts was inhibited by aminoguanidine (Fig. 1B). As expected, putrescine showed no effect on ARE activity (data not shown). Increase of nuclear Nrf2 is needed for the activation of the ARE. The nuclear level of Nrf2 was increased 6 h after treatment with spermine and spermidine in these cells (Fig. 1C). These results demonstrate that metabolites of polyamines produced by amine oxidase can induce multiple phase 2 detoxifying genes through the Nrf2–ARE pathway. Polyamine exposure activates the ARE through the generation of the reactive aldehyde, acrolein The inhibition by aminoguanidine on the induction of phase 2 genes by spermidine indicated that hydrogen peroxide or acrolein may be possible mediators for the activation of the ARE. To clarify which metabolite of the polyamines was involved in the induction of phase 2 genes, inactivating enzymes of hydrogen peroxide (catalase) and acrolein (aldehyde dehydrogenase) were used and NQO1 and ARE-luciferase activities were measured. Activation of GST A1 ARE-luciferase reporter by spermidine exposure was not affected by catalase treatment, while the effect of extracellular spermidine exposure was completely prevented by aldehyde dehydrogenase treatment (Fig. 2A). That hydrogen peroxide exposure can induce the ARE regulated genes in this system is readily demonstrated (Fig. 2A). However, the effect of hydrogen peroxide on ARE activity was completely inhibited by catalase, while aldehyde dehydrogenase treatment did not change the effect of hydrogen peroxide. By contrast, aldehyde dehydrogenase completely repressed activation of the ARE by acrolein and catalase did not inhibit the activation of the ARE by acrolein (Fig. 2A). Similar results were obtained in the measurement of NQO1 activity in these cells (Fig. 2B). Collectively, these results indicate that the inducer generated from spermidine is greatly affected by aldehyde dehydrogenase, suggesting that it is not a hydrogen peroxide, but an aldehyde. 666 M.-K. Kwak et al. / Biochemical and Biophysical Research Communications 305 (2003) 662–670 Fig. 2. Effects of catalase and aldehyde dehydrogenase (ALDH) on the induction of NQO1 activity and the activation of GST A1 ARE-luciferase reporter by spermidine (SPD). (A) PE cells were treated with SPD (50 lM), hydrogen peroxide (25 lM), or acrolein (25 lM) with catalase (5 U/well) or aldehyde dehydrogenase (3 U/well) following transient transfection of ARE-TATALucþ and pTATALucþ tk plasimds. (B) NQO1 activity was measured following treatment with SPD, hydrogen peroxide, or acrolein with catalase or aldehyde dehydrogenase. Values are means SE (n ¼ 3) and are expressed as ratios of treated over vehicle control. a P < 0:05 compared to spermidine treated control; b P < 0:05 compared to hydrogen peroxide treated control; and c P < 0:05 compared to acrolein treated control. Acrolein increases the expression of phase 2 genes through Nrf2–ARE pathway like polyamines To determine the effect of the aldehyde on phase 2 enzyme induction, acrolein was incubated with PE cells and levels of transcripts for phase 2 genes were monitored. No effects on cell viability were observed at the concentrations (up to 25 lM) used. Fig. 3A shows that levels of mRNAs for GST A1, NQO1, c-glutamylcysteine ligase regulatory subunit, and UDP-glucuronyltransferase 1A6 were elevated by treatment with acrolein for 18 h. As it was seen with the polyamines, the transcript levels of Mn-superoxide dismutase and catalase were not changed. In accord with the induction of multiple phase 2 genes, the nuclear accumulation of Nrf2 level was increased rapidly by treatment with acrolein as measured by immunoblot analysis (Fig. 3B). Fig. 3. Effects of acrolein on the transcriptional expression of phase 2 and antioxidative enzymes, and nuclear Nrf2 accumulation. (A) Levels of mRNAs for GST A1, M1, NQO1, regulatory domain of c-glutamylcysteine ligase (cGCLr), UDP-glucuronosyltransferase 1A6 (UGT1A6), Mn-superoxide dismutase (MnSOD), catalase, and b-actin were measured by RT-PCR analysis following treatment with acrolein (10 and 25 lM) for 18 h. (B) Nrf2 levels were determined in nuclear extracts by immunoblot analysis following treatment with acrolein (10 and 25 lM) for 6 h. Each lane contains three pooled samples. Overexpression of mutant Nrf2, which has no transactivation domain, inhibited the activation of GST A1 ARE-luciferase reporter by acrolein by 45%, while stimulation of the ARE by spermidine was inhibited by 67% (Fig. 4C). Cotreatment with sulfur-containing antioxidants such as glutathione and N-acetylcysteine completely blocked the activation of the ARE by acrolein (Fig. 4A). However, when the reactive unsaturated carbon of acrolein was reduced to propionaldehyde, this chemical did not activate the ARE (Fig. 4B). These results suggest that acrolein, an a; b-unsaturated aldehyde, activates the Nrf2–ARE pathway through its sulfhydryl reactive chemical property. Discussion Little is known as to how extracellular polyamines and their metabolites can effect gene expression. Changes in endogenous polyamine concentrations have been demonstrated to alter the expression of several important growth-related genes. The transcription of proto-oncogenes c-myc, c-jun, and c-fos is specifically downregulated when intracellular polyamines are M.-K. Kwak et al. / Biochemical and Biophysical Research Communications 305 (2003) 662–670 667 Fig. 4. (A) Effect of mutant Nrf2 overexpression on the activation of GST A1 ARE-luciferase reporter by spermidine (SPD) and acrolein. Mutant Nrf2 (mNrf2-pcDNA3, 0.6 lg/well) or blank plasmid (pcDNA3) with ARE-TATALucþ was transfected into PE cells and ARE-luciferase activity was measured following treatment with SPD (50 lM) or acrolein (25 lM). Luciferase activity was normalized using Renilla luciferase activity. Values are means SE (n ¼ 4) and expressed as ratios of treated over vehicle control. a P < 0:05 compared to vehicle blank plasmid control; b P < 0:05 compared to SPD-treated blank plasmid transfected group; and c P < 0:05 compared to acrolein-treated blank plasmid transfected group. (B) Glutathione (GSH, 500 lM) and N-acetylcysteine (NAC, 500 lM) were incubated with acrolein for 18 h and ARE-luciferase activity was measured. Propionaldehyde (Pro-Ald, 25 lM) was incubated and luciferase activity was also measured 18 h after treatment. Values are means SE (n ¼ 3). a P < 0:05 compared to SPD-treated control. depleted by inhibition of ornithine decarboxylase (ODC) [4,37–39]. Similarly, polyamine depletion can inhibit the heat shock response increase in hsp 70, c-jun, and c-fos expression [40]. By contrast increases in intracellular polyamines have been demonstrated to specifically induce the expression of growth related genes in cells overexpressing ODC [5,41]. Although molecular mechanisms by which polyamines induce or inhibit the expression of growth related genes are still to be determined a more direct role for the polyamine-induced transcription of the polyamine catabolic enzyme spermidine/spermine N1 -acetyltransferase has been determined. In the case of SSAT, increases in intracellular polyamines result in an upregulation of the transcription factor polyamine modulated factor-1 (PMF-1) that activates the transcription of SSAT through a polyamine response element (PRE) [42,43]. Interestingly, PMF-1 activates transcription, by first binding to Nrf2, which has been found to be constitutively bound to the PRE [43,44]. Consequently, there may be an interplay between the activation of genes modulated by intracellular polyamines and those modulated by extracellular polyamines or their metabolites. In the present study, we show that metabolites of polyamines can induce the expression of multiple phase 2 genes through activation of the transcription factor Nrf2. The results presented here support the fact that the effect of polyamines on phase 2 genes is through the formation of acrolein. Activation of Nrf2, which is a central element in the regulation of many detoxifying genes by polyamines, may bring broad alterations of gene expression patterns in cells. Spermidine and spermine are oxidized by serum amine oxidase and cleaved to aminoaldehydes and hy- drogen peroxide in cell culture [1,45]. Cytotoxicities evoked by polyamines have been explained by the hydrogen peroxide formation. Gaugas and Dewey [21] proposed that excessive production of hydrogen peroxide by serum amine oxidase contributes to polyamineinduced toxicity. Meanwhile, other reports have proposed that cytotoxicity by polyamines was not altered by catalase treatment [16,45], suggesting that aminoaldehydes were the major cause of cytotoxicity by exogenous polyamines [2,21]. Recently, it has been shown that acrolein can be formed from aminoaldehydes by nonenzymatic processes [16]. In our study, effects of polyamines on NQO1 and GST A1 ARE-luciferase activity were completely abolished by coincubation with aldehyde dehydrogenase; however, catalase treatment had no effect. Acrolein directly increased mRNA levels for multiple phase 2 detoxifying genes with activation of the ARE-luciferase reporter and accumulation of Nrf2 within nuclei. Hydrogen peroxide is also known to be a phase 2 enzyme inducer; concentrations below the cytotoxic levels of hydrogen peroxide (25 lM) activated the ARE in PE cells and transcripts of multiple phase 2 genes including GST A1, NQO1, c-glutamylcysteine ligase, and UDP-glucuronosyltransferase 1A6 (data not shown). However, inducible effects of polyamine exposure on NQO1 enzyme activity and ARE activity were not affected by catalase (Fig. 2). These results suggest that inducers derived from spermidine and spermine are aldehydes, and that reactive oxygen species derived from hydrogen peroxide have minimal effects. Tissue polyamine oxidases oxidize spermidine and spermine producing putrescine, 3-acetaminopropanal [1,9], or 3-aminopropanal [11,12], and hydrogen peroxide [1]. Accumulated 3-aminopropanal from polyamines 668 M.-K. Kwak et al. / Biochemical and Biophysical Research Communications 305 (2003) 662–670 has been associated with cytotoxicity in various tissues [46–50] and it was shown that acrolein is spontaneously generated from 3-aminopropanal to a greater extent than from spermidine itself in vitro [16,21]. Increased spermidine and spermine within cells in certain physiological conditions, particularly cancers [1,49], could cause alteration in Nrf2-dependent gene expression by generating acrolein. It is tempting to speculate that the induction of phase 2 enzymes by polyamine metabolites through Nrf-2 activation is a protective mechanism to guard cells from the disregulation of polyamine metabolism that frequently accompanies tumorigenesis [1]. A failure of this protective mechanism could result in an increase in oxidative damage, thus leading to the multiple genetic defects that are required for cancer to develop [50,51]. Nrf2 is a critical transcription factor in the regulation of ARE-dependent phase 2 enzymes. This transcription factor heterodimerizes with other b-ZIP transcription factors and transactivates the ARE in the promoter of phase 2 genes [23,24]. Induction of these genes by the antioxidant t-butylhydroxyanisole, sulforaphanes as well as dithiolethiones was largely lost by deleting this gene in mice [24,26,27]. Defects in the expression of protective phase 2 genes made nrf2-knockout mice more sensitive to the acute toxicities of different chemicals and carcinogens [52–57]. In the regulation of Nrf2, cytosolic protein Keap1 was proposed as an inhibitor of Nrf2 by anchoring this protein in the cytoplasm [29,58]. Binding between these two proteins is sensitive to sulfhydryl modifying reagents and captured oxidative signals by these proteins can dissociate this binding, leading to translocation of Nrf2 into nuclei [29]. Acrolein is an a; bunsaturated carbonyl-containing aldehyde that rapidly binds to cellular nucleophiles and induces cytotoxicity and genotoxicity [59–61]. This unsaturated carbonyl can act as a Michael center and has a high reactivity to sulfhydryl groups within the cells [60,62]. Involvement of Nrf2 in the induction of phase 2 genes by acrolein can be supported by the accumulation of Nrf2 within nuclei following treatment with acrolein. Furthermore, overexpression of mutant Nrf2 significantly repressed the activation of ARE by acrolein. Sulfhydryl reactive acrolein might act on the binding of Keap1 and Nrf2 to activate Nrf2–ARE pathway. Even though hydrogen peroxide and acrolein are theoretically produced in equal amounts in the catabolism polyamines, the leading effect of acrolein implicates the higher reactivity of acrolein to sulfhydryl groups of Keap1 than hydrogen peroxide. Sustained release of acrolein from polyamines by amine oxidase may stimulate the signal pathway continuously, accounting for the higher efficacy of spermidine than acrolein itself as an inducer. The effect of acrolein on ARE activation was completely inhibited by adding other sulfhydryl-reacting reagents GSH or Nacetylcysteine. Propionaldehyde does not have an un- saturated carbonyl group similar to acrolein and this chemical did not activate the ARE, suggesting that the a; b-unsaturated carbon in acrolein is essential in the activation of Nrf2 (Fig. 4D). Other reports have proposed that the a; b-unsaturated aldehyde, 4-hydroxy-2nonenal induces GST class-pi and c-glutamylcysteine ligase heavy chain in animal cells [63,64]. Recently, Tirumalai et al. [65] have also shown that acrolein increased NQO1 mRNA levels by activation of Nrf2 in human lung epithelial cells. In summary, the oxidation of polyamines by serum amine oxidase increases multiple phase 2 genes by activating Nrf2 pathway and the likely metabolite responsible for this induction is the sulfhydryl reactive aldehyde acrolein. Therefore, these results demonstrate a novel effect of exogenous polyamines on gene expressions in cell culture and suggest additional pathways that must be considered when examining the effects of polyamine catabolism. Acknowledgments We are grateful to Ryan Dick and Elin Simms for helpful discussions. This work was supported by Grants CA39416, CA94076, and CA51085. References [1] A.E. Pegg, Polyamine metabolism and its importance in neoplastic growth and a target for chemotherapy, Cancer Res. 48 (1988) 759–774. [2] C.W. Tabor, H. Tabor, Polyamines, Annu. Rev. Biochem. 53 (1984) 749–790. [3] L.J. Marton, A.E. Pegg, Polyamines as targets for therapeutic intervention, Annu. Rev. Pharmacol. Toxicol. 35 (1995) 55–91. [4] P. Celano, S.B. Baylin, R.A. Casero Jr., Polyamines differentially modulate the transcription of growth-associated genes in human colon carcinoma cells, J. Biol. Chem. 264 (1989) 8922–78927. [5] A. Tabib, U. Bachrach, Activation of the proto-oncogene c-myc and c-fos by c-ras: involvement of polyamines, Biochem. Biophys. Res. Commun. 202 (1999) 720–727. [6] M.A. Desiderio, L. Tacchini, E. Anzon, G. Pogliaghi, L. Radice, A. Bernelli-Zazzera, Effects of polyamine imbalance on the induction of stress genes in hepatocarcinoma cells exposed to heat shock, Hepatology 24 (1996) 150–156. [7] I.G. Kim, T.J. Oh, SOS induction of the recA gene by UV-, cirradiation and mitomycin C is mediated by polyamines in Escherichia coli K-12, Toxicol. Lett. 116 (2000) 143–149. [8] N. Seiler, Polyamine oxidase, properties and functions, Prog. Brain Res. 106 (1995) 333–344. [9] S. Vujcic, P. Liang, P. Diegelman, D. Kramer, C.W. Porter, Genomic identification and biochemical characterization of the mammalian polyamine oxidase involved in polyamine backconversion, Biochem. J. Pt (2002). [10] R.A. Casero Jr., A.E. Pegg, Spermidine/spermine N1 -acetyltransferase—the turning point in polyamine metabolism, FASEB J. 7 (1993) 653–661. [11] Y. Wang, W. Devereux, P.M. Woster, T.M. Stewart, A. Hacker, R.A. Casero Jr., Cloning and characterization of a human M.-K. Kwak et al. / Biochemical and Biophysical Research Communications 305 (2003) 662–670 [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] polyamine oxidase that is inducible by polyamine analogue exposure, Cancer Res. 61 (2001) 5370–5373. S. Vujcic, P. Diegelman, C.J. Bacchi, D.L. Kramer, C.W. Porter, Identification and characterization of a novel flavin-containing spermine oxidase of mammalian cell origin, Biochem. J. 367 (2002) 665–675. C.W. Tabor, H. Tabor, U. Bachrach, Identification of the aminoaldehyde produced by the oxidation of spermine with purified plasma amine oxidase, J. Biol. Chem. 239 (1964) 2194–2203. S.B. Baylin, B.J. Rosenstein, L.J. Marton, D.H. Lockwood, Agerelated abnormalities of circulating polyamines and diamine oxidase activity in cystic fibrosis heterozygotes and homozygotes, Pediatr. Res. 14 (1980) 921–925. W.A. Gahl, H.C. Pitot, Reversal by aminoguanidine of the inhibition of proliferation of human fibroblasts by spermidine and spermine, Chem. Biol. Interact. 22 (1978) 91–98. S. Sharmin, K. Sakata, K. Kashiwagi, S. Ueda, S. Iwasaki, A. Shirahata, K. Igarashi, Polyamine cytotoxicity in the presence of bovine serum amine oxidase, Biochem. Biophys. Res. Commun. 282 (2001) 228–235. M.C. Wolvekamp, R.W. de Bruin, Diamine oxidase: an overview of historical, biochemical and functional aspects, Digest. Dis. 12 (1994) 2–14. M. Le Calve, J. Segalen, D. Quernee, M.T. Lavault, D. Lescoat, Diamine oxidase activity and biochemical markers in human seminal plasma, Hum. Reprod. 10 (1995) 1141–1144. A. Keskinege, S. Elgun, E. Yilmaz, Possible implications of arginase and diamine oxidase in prostatic carcinoma, Cancer Detect. Prev. 25 (2001) 76–79. D.A. Averill-Bates, E. Agostinelli, E. Przybytkowski, B. Mondovi, Aldehyde dehydrogenase and cytotoxicity of purified bovine serum amine oxidase and spermine in Chinese hamster ovary cells, Biochem. Cell. Biol. 72 (1994) 36–42. J.M. Gaugas, D.L. Dewey, Oxygen-dependent free radicals in spermine oxidation cytostatis and chemiluminescence and the role of superoxide dismutase, Br. J. Cancer 41 (1980) 946–955. Y.C. Ha, M. Patrick, M. Woster, J.D. Yager, R.A. Casero Jr., The role of polyamines catabolism in polyamine analogue-induced programmed cell death, Proc. Natl. Acad. Sci. USA 94 (1997) 11557–11562. R. Venugopal, A.K. Jaiswal, Nrf1 and Nrf2 positively and c-Fos and Fra1 negatively regulate the human antioxidant response element-mediated expression of NAD(P)H:quinone oxidoreductase 1 gene, Proc. Natl. Acad. Sci. USA 93 (1996) 14960–14965. K. Itoh, T. Chiba, S. Takahashi, T. Ishii, K. Igarashi, Y. Katoh, T. Oyake, N. Hayashi, K. Satoh, I. Hatayama, M. Yamamoto, Y. Nabeshima, An Nrf2/small Maf heterodimer mediates the induction of phase II detoxifying enzyme gene through antioxidant response elements, Biochem. Biophys. Res. Commun. 236 (1997) 313–322. W.W. Wasserman, W.E. Fahl, Functional antioxidant response element, Proc. Natl. Acad. Sci. USA 94 (1997) 5361–5366. M-K. Kwak, K. Itoh, M. Yamamoto, T.R. Sutter, T.W. Kensler, Role of transcription factor Nrf2 in the induction of hepatic phase 2 and antioxidative enzymes in vivo by the cancer chemoprotective agent, 3H-1,2-dithiole-3-thione, Mol. Med. 7 (2001) 135–145. M. MacMahon, K. Itoh, M. Yamamoto, S.A. Chanas, C.J. Henderson, L.I. McLellan, C.R. Wolf, C. Cavin, J.D. Hayes, The CapÕnÕCollar basic leucine zipper transcription factor Nrf2 (NFE2 p45-related factor2) controls both constitutive and inducible expression of intestinal detoxification and glutathione biosynthetic enzymes, Cancer Res. 61 (2001) 3299–3307. T. Ishii, K. Itoh, S. Takahashi, H. Sato, T. Yanagawa, Y. Katoh, S. Bannai, M. Yamamoto, Transcription factor Nrf2 coordinately [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] 669 regulates a group of oxidative stress-induced genes in macrophages, J. Biol. Chem. 275 (2000) 16023–16029. K. Itoh, N. Wakabayashi, Y. Katoh, T. Ishii, K. Igarashi, J.D. Engel, M. Yamamoto, Keap1 represses nuclear activation of antioxidant responsive elements by Nrf2 through binding to the amino-terminal Neh2 domain, Genes Dev. 13 (1999) 76–86. A.T. Dinkova-Kostova, W.D. Holtzclaw, R.N. Cole, K. Itoh, N. Wakabayashi, Y. Katoh, M. Yamamoto, P. Talalay, Direct evidence that sulfhydryl groups of Keap1 are the sensors regulating induction of phase 2 enzymes that protect against carcinogens and oxidants, Proc. Natl. Acad. Sci. USA 99 (2002) 11908–11913. S.H. Yuspa, D. Morgan, U. Lichti, E.F. Spangler, D. Michael, A. Kilkenny, H. Hennings, Cultivation and characterization of cells derived from mouse skin papillomas induced by an initiation– promotion protocol, Carcinogenesis 7 (1986) 949–958. H.J. Prochaska, A.B. Santamaria, P. Talalay, Rapid detection of inducers of enzymes that protect against carcinogens, Proc. Natl. Acad. Sci. USA 89 (1992) 2394–2398. P.M. Kabra, H.K. Lee, W.P. Lubich, L.J. Marton, Solid-phase extraction and determination of dansyl derivatives of unconjugated and acetylated polyamines by reversed-phase liquid chromatography: improved separation systems for polyamines in cerebrospinal fluid, urine and tissue, J. Chromatogr. 380 (1986) 19–32. P. Chomcynski, N. Sacchi, Single-step method of RNA isolation by acid guanidinium thiocyanate–phenol–chloroform extraction, Anal. Biochem. 162 (1987) 156–159. V. Daniel, R. Sharon, Y. Tichauer, S. Sarid, Mouse glutathione S-transferase Ya subunit: gene structure and sequence, DNA 6 (1987) 317–324. J. Alam, D. Stewart, C. Touchard, S. Boinapally, A.M.K. Choi, J.L. Cook, Nrf2, a CapÕn Collar transcription factor, regulates induction of the heme oxygenase-1 gene, J. Biol. Chem. 274 (1999) 26071–26078. P. Celano, S.B. Baylin, F.M. Giardiello, B.D. Nelkin, R.A. Casero Jr., Effect of polyamine depletion on c-myc expression in human colon carcinoma cells, J. Biol. Chem. 263 (1988) 5491–5494. J.Y. Wang, L.R. Johnson, Expression of protooncogenes c-fos and c-myc in healing of gastric mucosal stress ulcers, Am. J. Physiol. 266 (1994) G878–886. J.Y. Wang, S.A. McCormack, M.J. Viar, H. Wang, C.Y. Tzen, R.E. Scott, L.R. Johnson, Decreased expression of protooncogenes c-fos, c-myc, and c-jun following polyamine depletion in IEC-6 cells, Am. J. Physiol. 265 (1993) G331–338. M.A. Desiderio, L. Tacchini, E. Anzon, G. Pogliaghi, L. Radice, A. Bernelli-Zazzera, Effects of polyamine imbalance on the induction of stress genes in hepatocarcinoma cells exposed to heat shock, Hepatology 24 (1996) 150–156. A. Tabib, U. Bachrach, Role of polyamines in mediating malignant transformation and oncogene expression, Int. J. Biochem. Cell Biol. 31 (1999) 1289–1295. Y. Wang, L. Xiao, A. Thiagalingam, B.D. Nelkin, R.A. Casero Jr., The identification of a cis-element and a trans-acting factor involved in the response to polyamines and polyamine analogues in the regulation of the human spermidine/spermine N1 -acetyltransferase gene transcription, J. Biol. Chem. 273 (1998) 34623–34630. Y. Wang, W. Devereux, T.M. Stewart, R.A. Casero Jr., Cloning and characterization of human polyamine-modulated factor-1, a transcriptional cofactor that regulates the transcription of the spermidine/spermine N(1)-acetyltransferase gene, J. Biol. Chem. 274 (1999) 22095–22101. Y. Wang, W. Devereux, T.M. Stewart, R.A. Casero Jr., Characterization of the interaction between the transcription factors human polyamine modulated factor (pmf-1) and nf-e2-related factor 2 (nrf-2) in the transcriptional regulation of the spermidine/ 670 [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] M.-K. Kwak et al. / Biochemical and Biophysical Research Communications 305 (2003) 662–670 spermine N1 -acetyltransferase (ssat) gene, Biochem. J. 355 (2001) 45–49. V.G. Brunton, M.H. Grant, H.M. Wallace, Mechanism of spermine toxicity in baby-hamster kidney (BHK) cells, Biochem. J. 280 (1991) 193–198. J.R. Foster, L.L. Smith, P.M. Hext, A. Brammer, A.R. Soames, I. Wyatt, Target cell toxicity of inhaled spermidine in rat lung, Int. J. Exp. Pathol. 71 (1990) 617–630. A. Dogan, R.A. Muralikrishna Rao, J. Hatcher, V.L. Muralikrishna Rao, M.K. Baskaya, R.J. Cempsey, Effects of MDL 72527, a specific inhibitor of polyamine oxidase, on brain edema, ischemic injury volume, and tissue polyamine levels in rats after temporary middle cerebral artery occlusion, J. Neurochem. 72 (1999) 765–770. S. Ivanova, G. Botchkina, Y. Al-Abed, M. Meistrell III, F. Batliwalla, J.M. Dubinsky, C. Iadecola, H. Wang, P.K. Gregersen, J.W. Eaton, K.J. Tracey, Cerebral ischemia enhanced polyamine oxidation: identification of enzymatically formed 3aminopropanal as an endogenous mediator of neuronal and glial cell death, J. Exp. Med. 188 (1998) 327–340. F.M. Giardiello, S.R. Hamilton, L.M. Hylind, V.W. Yang, P. Tamez, R.A. Casero Jr., Ornithine decarboxylase and polyamines in familial adenomatous polyposis, Cancer Res. 57 (1997) 199–201. B. Vogelstein, E.R. Fearon, S.R. Hamilton, S.E. Kern, A.C. Preisinger, M. Leppert, Y. Nakamura, R. White, A.M. Smits, J.L. Bos, Genetic alterations during colorectal-tumor development, N. Engl. J. Med. 319 (1988) 525–532. E.R. Fearon, B. Vogelstein, A genetic model for colorectal tumorigenesis, Cell 61 (1990) 759–767. K. Chan, Y.W. Kan, Nrf2 is essential for protection against acute pulmonary injury in mice, Proc. Natl. Acad. Sci. USA 96 (1999) 12731–12736. K. Chan, X.D. Han, Y.W. Kan, An important function of Nrf2 in combating oxidative stress: detoxification of acetaminophen, Proc. Natl. Acad. Sci. USA 98 (2001) 4611–4616. H.-Y. Cho, A.E. Jedlicka, M.S. Sekhar, P.M. Reddy, L.Y. Zhang, T.W. Kensler, M. Yamamoto, S.R. Kleeberger, Linkage analysis of susceptibility to hyperoxia: Nrf2 is a candidate gene, Am. J. Respir. Cell Mol. Biol. 26 (2001) 42–51. [55] M.-K. Kwak, P.A. Egner, P.M. Dolan, M. Ramos-Gomez, J.D. Groopman, K. Itoh, M. Yamamoto, T.W. Kensler, Role of phase 2 enzyme induction in chemoprotection by dithiolethiones, Mutat. Res. 480–481 (2001) 305–315. [56] Y. Aoki, H. Sato, N. Nishimura, S. Takahashi, K. Itoh, M. Yamamoto, Accelerated DNA adduct formation in the lung of the Nrf2 knockout mouse exposed to diesel exhaust, Toxicol. Appl. Pharmacol. 173 (2001) 154–160. [57] M. Ramos-Gomez, M.-K. Kwak, P.M. Dolan, K. Itoh, M. Yamamoto, P. Talalay, T.W. Kensler, Sensitivity to carcinogenesis is increased and chemoprotective efficacy of enzymes inducers is lost in nrf2 transcription factor-deficient mice, Proc. Natl. Acad. Sci. USA 98 (2001) 3410–3415. [58] S. Dhakshinamoorthy, A.K. Jaiswal, Functional characterization and the role of INrf2 in antioxidant response element-mediated expression and antioxidant induction of NAD(P)H:quinone oxidoreductase, Oncogene 20 (2001) 3906–3917. [59] J.P. Kehrer, S.S. Biswal, The molecular effects of acrolein, Toxicol. Sci. 57 (2000) 6–15. [60] K.D. Bunting, A.J. Townsend, Dependence of aldehyde dehydrogenase-mediated oxazaphosphorine resistance on soluble thiols: importance of thiol interactions with the secondary metabolite acrolein, Biochem. Pharmacol. 56 (1998) 31–39. [61] H. Jaeschke, C. Kleinwaechter, A. Wendel, The role of acrolein in allyl alcohol-induced lipid peroxidation and liver cell damage in mice, Biochem. Pharmacol. 36 (1987) 51–57. [62] T. Prestera, W.D. Holtzclaw, Y. Zhang, P. Talalay, Chemical and molecular regulation of enzymes that detoxify carcinogens, Proc. Natl. Acad. Sci. USA 90 (1993) 2965–2969. [63] R.-M. Liu, Z. Borok, H.J. Forman, 4-Hydroxy-2-nonenal increases c-glutamylcysteine synthetase gene expression in alveolar epithelial cells, Am. J. Respir. Cell Mol. Biol. 24 (2001) 499–505. [64] A. Fukuda, Y. Nakamura, H. Ohigashi, T. Osawa, K. Uchida, Cellular response to the redox active lipid peroxidation products: induction of glutathione S-transferase P by 4-hydroxy-2-nonenal, Biochem. Biophys. Res. Commun. 236 (1997) 505–509. [65] R. Tirumalai, K.T. Rajesh, K.H. Mai, S. Biswal, Acrolein causes transcriptional induction of phase II genes by activation of Nrf2 in human lung type II epithelial (A549) cells, Toxicol. Lett. 132 (2002) 27–36.