Download Induction of phase 2 enzymes by serum oxidized polyamines

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.