- Email: [email protected]
Redox cycling of the herbicide paraquat in microglial cultures
Molecular Brain Research 134 (2005) 52 – 56 www.elsevier.com/locate/molbrainres
Review
Redox cycling of the herbicide paraquat in microglial cultures Dafna Bonneh-Barkay, Stephen H. Reaney, William J. Langston, Donato A. Di Monte* The Parkinson’s Institute, 1170 Morse Avenue, Sunnyvale, CA 94089-1605, USA Accepted 17 November 2004 Available online 6 January 2005
Abstract Mechanisms involved in paraquat neurotoxicity that selectively target nigrostriatal dopaminergic neurons remain relatively unknown. In this study, we tested the hypotheses that paraquat exposure leads to the production of reactive oxygen species (ROS) through a process of redox cycling and that microglia represent an important site for the initiation of redox cycling reactions. Addition of paraquat to N9 microglial cultures resulted in a dose- and time-dependent release of superoxide radicals. Other agents that share with paraquat the property of redox cycling, i.e., benzyl viologen and diquat, also induced a marked production of superoxide radicals by microglia. The ability of paraquat, benzyl viologen, and diquat to induce superoxide release was correlated to their one-electron reduction potentials and thus their tendency to redox cycle. Nitric oxide synthase and NADPH oxidase were identified as enzymatic sources of electrons that triggered paraquat redox cycling by microglia. Taken together, these data provide evidence in favor of a new mechanism by which microglia could play a role in oxidative injury during neurodegenerative processes. Microglial NOS and NADPH oxidase could promote the generation of ROS via the redox cycling of paraquat-like toxicants. D 2004 Elsevier B.V. All rights reserved. Theme: Disorders of the nervous system Topic: Degenerative disease: Parkinson’s Keywords: Parkinson; Paraquat; Microglia; Redox cycling; NOS; NADPH oxidase; Superoxide
Contents 1. 2.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Materials and methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Cell cultures and treatments . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Redox cycling assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4. Statistical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Paraquat-induced superoxide release . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Comparison of the effects of paraquat, diquat, and benzyl viologen on superoxide 3.3. Role of NOS and NADPH oxidase in paraquat-induced superoxide release . . . . 4. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
* Corresponding author. Fax: +1 408 734 8522. E-mail address: [email protected] (D.A. Di Monte). 0169-328X/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.molbrainres.2004.11.005
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . release . . . . . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
53 53 53 53 53 54 54 54 54 55 55 56 56
D. Bonneh-Barkay et al. / Molecular Brain Research 134 (2005) 52–56
1. Introduction
2. Materials and methods
Exposure of mice to the herbicide paraquat yields a valuable model of Parkinson’s disease (PD)-like neurodegeneration induced by environmental risk factors [9]. Features of paraquat toxicity that mirror PD pathology include a loss of dopaminergic cells in the substantia nigra pars compacta accompanied by microglial and astrocytic activation [15]. Other neuronal cell populations, such as GABAergic neurons in the substantia nigra pars reticulata, are spared by paraquat toxicity, and no glial response occurs outside the nigrostriatal dopaminergic pathway [15]. The mechanism of paraquat-induced dopaminergic cell damage, although presently unknown, is likely to involve a process of redox cycling that begins with the enzymatic one-electron reduction of the herbicide, proceeds with the electron transfer to molecular oxygen and ultimately leads to the formation of superoxide anion and other reactive oxygen species (ROS) [3,5]. Dopaminergic neurons may be preferentially targeted by paraquat neurotoxicity because of their greater vulnerability to ROS-mediated oxidative injury [6,10]. A variety of cellular diaphorases are capable of transferring electrons to paraquat, thus initiating its redox cycling. These enzymes are typically flavin-containing oxidoreductases that use NADH or NADPH as electron donors [4,19]. Of particular relevance to paraquat neurotoxicity is the reported ability of nitric oxide synthase (NOS) to catalyze paraquat reduction and NADPH oxidation [8]. This finding suggests that paraquat bioactivation in the CNS may be promoted within cells that express constitutive (neurons) or inducible (predominantly microglia) NOS isoforms [7,14]. Another enzyme that could potentially play an important role in catalyzing paraquat redox cycling is NADPH oxidase. Although NADPH oxidase is present in various cell types including neurons [21], its activity has mostly been associated with the function of immunocompetent cells such as macrophages and microglia. After a toxic insult to the CNS and upon microglial activation, the subunits that compose NADPH oxidase are assembled at the plasma membrane level where the enzyme is then capable of reducing oxygen to the superoxide radical [22]. It is conceivable that, in the presence of paraquat, NADPH oxidase could transfer electrons to the herbicide and enhance superoxide formation via a redox cycling mechanism. The current study was undertaken to test the hypothesis that microglia play a role in paraquat neurotoxicity by promoting its redox cycling. More specifically, experiments were designed to determine the ability of cultured microglia to generate superoxide radicals after paraquat exposure, to relate this ROS formation to a process of redox cycling and, finally, to identify the roles of NOS and NADPH oxidase in microglia-mediated paraquat redox cycling.
2.1. Materials
53
Paraquat, diquat, benzyl viologen, nitroblue tetrazolium (NBT), superoxide dismutase (SOD), apocynin and NN-nitro-l-arginine methyl ester (l-NAME) were purchased from Sigma (St. Louis, MO). RPMI 1640 medium and fetal calf serum were from GIBCO (Gaithersburg, MD). 2.2. Cell cultures and treatments N9 murine microglial cells (kindly provided by Dr. Paola Ricciardi-Castagnoli) were plated onto 96-well plates and maintained in RPMI 1640 medium supplemented with 5% heat-inactivated fetal calf serum and 25 Am h-mercaptoethanol. When cells reached maximum confluence, experiments were conducted in serum- and phenol red-free RPMI 1640 medium. In experiments in which the effects of NOS and NADPH oxidase inhibitors were tested, cells were incubated in the presence of apocynin (0.5 mM) and/or l-NAME (5 mM) for 30 min prior to paraquat or vehicle addition. 2.3. Redox cycling assay Redox cycling was measured as the production of superoxide anion after treatment with paraquat, diquat or benzyl viologen. Superoxide radical release was quantified as the SOD-sensitive reduction of NBT [13]. NBT is a water-soluble reagent (yellow solution) that, in the presence of superoxide radicals, is converted to nitroblue formazan (insoluble purple precipitate). N9 cells were incubated with vehicle (medium) or different concentrations of the redox cycling agents for 30 min prior to the addition of NBT (1 mg/ml, final concentration). After this addition, incuba-
Fig. 1. Paraquat-induced dose-dependent increase in superoxide release. N9 cells were treated for 90 min with various concentrations (mM) of paraquat. Superoxide radical production was quantified by measuring NBT reduction. Data (n z 6) are mean F SEM and are expressed as % of control values measured in vehicle-treated cultures. *P b 0.001 compared to the other treatment groups.
54
D. Bonneh-Barkay et al. / Molecular Brain Research 134 (2005) 52–56
when differences were observed in the ANOVA testing ( P b 0.05).
3. Results 3.1. Paraquat-induced superoxide release
Fig. 2. Paraquat-induced time-dependent increase in superoxide release. N9 cells were incubated in the presence of 0.5 mM paraquat for 0, 60, 90 or 120 min. Superoxide production was quantified by the reduction of NBT. Data (n z 6) are mean F SEM and are expressed as % of control values measured in vehicle-treated cultures. *P b 0.05 compared to the other treatment groups.
tions proceeded at 37 8C for various periods of time (30 to 90 min post NBT). The medium was then removed, and the cells were resuspended in water and sonicated to solubilize the formazan precipitate. For quantification, optical density was immediately measured at 560 nm in a microtiter plate reader.
In the first set of experiments, N9 cells were exposed for 90 min to concentrations of paraquat (0.1 to 2 mM) similar to those used to trigger redox cycling reactions in other in vitro systems over a relatively short incubation period [8,18]. These incubations resulted in a dose-dependent enhancement of NBT reduction (Fig. 1), with substantial superoxide release caused by concentrations of 0.5 mM or greater. Paraquat-induced changes were also time-dependent. After addition of 0.5 mM paraquat, a significant increase in NBT reduction was already evident at 60 min and became progressively more pronounced at the 90- and 120-min time points (Fig. 2). NBT reduction caused by paraquat was counteracted by inclusion of SOD (600 U/ml) in the incubation medium (data not shown), demonstrating its relation to the formation and release of superoxide radicals. 3.2. Comparison of the effects of paraquat, diquat, and benzyl viologen on superoxide release
Differences among means were analyzed using a one-way ANOVA. Fisher’s post hoc analysis was employed
The rapid NBT reduction caused by paraquat in microglial cultures suggests the involvement of redox cycling reactions. Other bipyridyl pesticides, such as diquat and benzyl viologen, share the ability to generate superoxide radicals through a redox cycling mechanism [18]. We therefore tested the effects of these agents in our in vitro system. As compared
Fig. 3. Effects of bipyridyl compounds on superoxide release. N9 cells were incubated in the presence of paraquat (black bars), benzyl viologen (dark gray bars) or diquat (light gray bars) at 0.1 or 0.5 mM concentrations. Superoxide production was quantified by measuring NBT reduction 90 min after the administration of bipyridyls. Data (n z 6) are mean F SEM and are expressed as % of control values measured in vehicle-treated cultures. *P b 0.001 compared to the other treatment groups at the same concentration.
Fig. 4. Effects of NOS and NADPH oxidase inhibitors on paraquat-induced superoxide release. N9 cells were treated with 5 mM l-NAME, an inhibitor of NOS, and/or 0.5 mM apocynin, an inhibitor of NADPH oxidase, for 30 min prior to the addition of 0.5 mM paraquat. Superoxide production was quantified by measuring NBT reduction 90 min after paraquat administration. Data are (n = 5) mean F SEM and are expressed as % of control values measured in vehicle-treated cultures. *P b 0.05 compared to the other treatment groups.
2.4. Statistical analysis
D. Bonneh-Barkay et al. / Molecular Brain Research 134 (2005) 52–56
to paraquat, both benzyl viologen and diquat induced a more pronounced NBT reduction (Fig. 3). When the three compounds were added for 90 min at a concentration of 0.1 mM, only diquat caused a significant stimulation of superoxide release. At a concentration of 0.5 mM, NBT reduction was increased by approximately three-, four- and six-folds in the presence of paraquat, benzyl viologen and diquat, respectively (Fig. 3). 3.3. Role of NOS and NADPH oxidase in paraquat-induced superoxide release As cellular diaphorases, both NOS and NADPH oxidase could catalyze the redox cycling of paraquat in microglia. Incubation of N9 cells for 30 min in the presence of the NOS inhibitor l-NAME [17] prior to the addition of paraquat resulted in a significant decrease in paraquat-induced NBT reduction (Fig. 4). Similarly, apocynin, an inhibitor of NADPH oxidase [16], significantly prevented the release of superoxide radicals triggered by paraquat exposure. The effect of 0.5 mM paraquat (90 min exposure time) on superoxide release was attenuated by approximately 50% after pre-incubations with either l-NAME or apocynin. However, when the two inhibitors were used together, an additive action was indicated by the finding that NBT reduction returned to the basal levels measured in paraquatfree control cultures (Fig. 4).
4. Discussion Oxidative stress has long been suggested to play an important role in the neurodegenerative process underlying PD [6,10]. Various mechanisms of free radical generation pertinent to PD have been proposed, including the NOSmediated production of NO and peroxynitrite that results from microglial activation and the enhanced formation of ROS after partial inhibition of mitochondrial complex I
Fig. 5. Redox cycling reactions triggered by paraquat (PQ) in microglia.
55
[2,14]. Experimental evidence showing selective dopaminergic cell loss after treatment of mice with paraquat also suggests that neurotoxicants could trigger ROS generation and induce oxidative damage through a process of redox cycling [15]. As a redox cycling agent, paraquat must undergo a one-electron reduction catalyzed by cellular reductases [3,5]. The results of this study provide first evidence that this bioactivation can occur within microglia. Addition of paraquat to microglial cultures caused a time- and dose-dependent production of superoxide radicals. This finding is consistent with the following sequence of redox cycling reactions: the one-electron reduction of paraquat produces a cation radical which, by transferring its electron onto molecular oxygen, readily forms superoxide anion and re-generates the parent compound (Fig. 5). Further evidence in favor of a role of microglia in triggering redox cycling derives from experiments with diquat and benzyl viologen. The rate of redox cycling is dependent upon the electrochemical properties and, in particular, the one-electron reduction potentials of specific compounds [5,12]. Agents that accept electrons less readily are characterized by a more negative potential and are therefore poorer redox cyclers. A comparison of the reduction potentials of paraquat ( 0.44 V vs. a normal hydrogen electrode), benzyl viologen ( 0.35 V) and diquat ( 0.33 V) would predict the ranking of redox cycling to be paraquat b benzyl viologen b diquat [11,12]. Indeed, when the three compounds were added to microglial cultures, superoxide formation was most pronounced in the presence of diquat and least stimulated after paraquat exposure. Taken together, these data suggest that microglia are important sites for the bioactivation of redox cycling toxicants and that, through this novel mechanism, they could play a role in oxidative injury during neurodegenerative processes. Another important component of the present work was to identify specific enzymes responsible for the reduction of paraquat and the initiation of its redox cycling. Previous work has suggested that paraquat redox cycling in cultures of endothelial cells and cytokine-activated macrophages is attenuated by NOS inhibitors [8]. Here, we show that NOS inhibitors are also capable of preventing paraquat-induced superoxide formation by microglia. Furthermore, we identified an additional pathway involved in the reduction of paraquat that was catalyzed by NADPH oxidase (Fig. 5). Both microglial NADPH oxidase and inducible NOS have already been implicated in neurodegenerative processes targeting dopaminergic cells [14,22]. The toxic potential of NADPH oxidase has been related to the direct production of superoxide radicals, whereas NOS activity could promote oxidative injury via the initial formation of NO, the rapid reaction of NO with superoxide radical, and ultimately, the generation of the strong oxidant peroxynitrite. Our results indicate a complementary mechanism by which these two enzymes could contribute to oxidative reactions in the presence of specific toxicants. They could act as electron sources for the redox cycling of paraquat-like compounds,
56
D. Bonneh-Barkay et al. / Molecular Brain Research 134 (2005) 52–56
and through this process, the formation of ROS would likely be significantly enhanced. Other important considerations arise from the participation of microglial NOS and NADPH oxidase in redox cycling reactions. Our data show that NOS inhibitors reduced extracellular levels of superoxide in paraquat-exposed cultures. This finding is consistent with the interpretation that membrane-soluble superoxide radicals, once generated within microglial cells via NOS activity, are released into the extracellular space. In relation to NADPH oxidase, the transmembrane location of this enzyme would likely facilitate the bioactivation of hydrophilic toxicants, including paraquat, characterized by a relatively poor capability of crossing cell membranes. Moreover, NADPH-mediated redox cycling at the plasma membrane level is likely to produce superoxide radicals outward directly into the extracellular space [1]. The release of these radicals from microglia into the extracellular compartment raises the possibility that, in the in vivo setting of human and animal brain, they could ultimately lead to the oxidative damage of nearby neuronal cells. In summary, the present data provide insight into the neurotoxicity of redox cycling agents, pointing to microglia as cell mediators of their metabolic activation and to NOS and NADPH oxidase as enzymatic triggers of redox cycling reactions. These findings are likely to be relevant to our understanding of the sequence of events that begins with exposure to paraquat and paraquat-like compounds and may ultimately lead to the selective degeneration of dopaminergic neurons [15]. Besides bipyridyl pesticides such as paraquat, diquat, and benzyl viologen, a variety of naturally occurring and synthetic molecules (e.g., quinones) share the ability to redox cycle [12,20]. This ability could therefore characterize an entire class of environmental risk factors of particular interest for PD.
Acknowledgments This work was supported by grants from the National Institute of Environmental Health Sciences (ES10442, ES10806 and ES12077, DAD) and the Michael J. Fox Foundation (DB-B). We thank Alison L. McCormack for her valuable comments on this manuscript.
References [1] B.M. Babior, NADPH oxidase: an update, Blood 93 (1999) 1464 – 1476. [2] R. Betarbet, T.B. Sherer, G. MacKenzie, M. Garcia-Osuna, A.V. Panov, J.T. Greenamyre, Chronic systemic pesticide exposure reproduces features of Parkinson’s disease, Nat. Neurosci. 3 (2000) 1301 – 1306. [3] J.S. Bus, J.E. Gibson, Paraquat: model for oxidant-initiated toxicity, Environ. Health Perspect. 55 (1984) 37 – 46.
[4] L. Clejan, A.I. Cederbaum, Synergistic interactions between NADPH-cytochrome P-450 reductase, paraquat, and iron in the generation of active oxygen radicals, Biochem. Pharmacol. 38 (1989) 1779 – 1786. [5] G.M. Cohen, M. d’Arcy Doherty, Free radical mediated cell toxicity by redox cycling chemicals, Br. J. Cancer, Suppl. 8 (1987) 46 – 52. [6] W. Dauer, S. Przedborski, Parkinson’s disease: mechanisms and models, Neuron 39 (2003) 889 – 909. [7] V.L. Dawson, T.M. Dawson, Nitric oxide in neurodegeneration, Prog. Brain Res. 118 (1998) 215 – 229. [8] B.J. Day, M. Patel, L. Calavetta, L.Y. Chang, J.S. Stamler, A mechanism of paraquat toxicity involving nitric oxide synthase, Proc. Natl. Acad. Sci. U. S. A. 96 (1999) 12760 – 12765. [9] D.A. Di Monte, The environment and Parkinson’s disease: is the nigrostriatal system preferentially targeted by neurotoxins? Lancet Neurol. 2 (2003) 531 – 538. [10] S. Fahn, G. Cohen, The oxidant stress hypothesis in Parkinson’s disease: evidence supporting it, Ann. Neurol. 32 (1992) 804 – 812. [11] M. Faro, C. Gomez-Moreno, M. Stankovich, M. Medina, Role of critical charged residues in reduction potential modulation of ferredoxin-NADP+ reductase, Eur. J. Biochem. 269 (2002) 2656 – 2661. [12] D.M. Frank, P.K. Arora, J.L. Blumer, L.M. Sayre, Model study on the bioreduction of paraquat, MPP+, and analogs. Evidence against a bredox cyclingQ mechanism in MPTP neurotoxicity, Biochem. Biophys. Res. Commun. 147 (1987) 1095 – 1104. [13] T.W. Kirby, I. Fridovich, A picomolar spectrophotometric assay for superoxide dismutase, Anal. Biochem. 127 (1982) 435 – 440. [14] G.T. Liberatore, V. Jackson-Lewis, S. Vukosavic, A.S. Mandir, M. Vila, W.G. McAuliffe, V.L. Dawson, T.M. Dawson, S. Przedborski, Inducible nitric oxide synthase stimulates dopaminergic neurodegeneration in the MPTP model of Parkinson disease, Nat. Med. 5 (1999) 1403 – 1409. [15] A.L. McCormack, M. Thiruchelvam, A.B. Manning-Bog, C. Thiffault, J.W. Langston, D.A. Cory-Slechta, D.A. Di Monte, Environmental risk factors and Parkinson’s disease: selective degeneration of nigral dopaminergic neurons caused by the herbicide paraquat, Neurobiol. Dis. 10 (2002) 119 – 127. [16] R.B. Muijsers, E. van Den Worm, G. Folkerts, C.J. Beukelman, A.S. Koster, D.S. Postma, F.P. Nijkamp, Apocynin inhibits peroxynitrite formation by murine macrophages, Br. J. Pharmacol. 130 (2000) 932 – 936. [17] H. Nakanishi, J. Zhang, M. Koike, T. Nishioku, Y. Okamoto, E. Kominami, K. von Figura, C. Peters, K. Yamamoto, P. Saftig, Y. Uchiyama, Involvement of nitric oxide released from microgliamacrophages in pathological changes of cathepsin D-deficient mice, J. Neurosci. 21 (2001) 7526 – 7533. [18] M.S. Sandy, P. Moldeus, D. Ross, M.T. Smith, Role of redox cycling and lipid peroxidation in bipyridyl herbicide cytotoxicity. Studies with a compromised isolated hepatocyte model system, Biochem. Pharmacol. 35 (1986) 3095 – 3101. [19] J. Sarlauskas, A. Nemeikaite-Ceniene, Z. Anusevicius, L. Miseviciene, M.M. Julvez, M. Medina, C. Gomez-Moreno, N. Cenas, Flavoenzymecatalyzed redox cycling of hydroxylamino- and amino metabolites of 2,4,6-trinitrotoluene: implications for their cytotoxicity, Arch. Biochem. Biophys. 425 (2004) 184 – 192. [20] M.T. Smith, Quinones as mutagens, carcinogens, and anticancer agents: introduction and overview, J. Toxicol. Environ. Health 16 (1985) 665 – 672. [21] S.P. Tammariello, M.T. Quinn, S. Estus, NADPH oxidase contributes directly to oxidative stress and apoptosis in nerve growth factordeprived sympathetic neurons, J. Neurosci. 20 (2000) RC53. [22] D.C. Wu, P. Teismann, K. Tieu, M. Vila, V. Jackson-Lewis, H. Ischiropoulos, S. Przedborski, NADPH oxidase mediates oxidative stress in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine model of Parkinson’s disease, Proc. Natl. Acad. Sci. U. S. A. 100 (2003) 6145 – 6150.
Review
Redox cycling of the herbicide paraquat in microglial cultures Dafna Bonneh-Barkay, Stephen H. Reaney, William J. Langston, Donato A. Di Monte* The Parkinson’s Institute, 1170 Morse Avenue, Sunnyvale, CA 94089-1605, USA Accepted 17 November 2004 Available online 6 January 2005
Abstract Mechanisms involved in paraquat neurotoxicity that selectively target nigrostriatal dopaminergic neurons remain relatively unknown. In this study, we tested the hypotheses that paraquat exposure leads to the production of reactive oxygen species (ROS) through a process of redox cycling and that microglia represent an important site for the initiation of redox cycling reactions. Addition of paraquat to N9 microglial cultures resulted in a dose- and time-dependent release of superoxide radicals. Other agents that share with paraquat the property of redox cycling, i.e., benzyl viologen and diquat, also induced a marked production of superoxide radicals by microglia. The ability of paraquat, benzyl viologen, and diquat to induce superoxide release was correlated to their one-electron reduction potentials and thus their tendency to redox cycle. Nitric oxide synthase and NADPH oxidase were identified as enzymatic sources of electrons that triggered paraquat redox cycling by microglia. Taken together, these data provide evidence in favor of a new mechanism by which microglia could play a role in oxidative injury during neurodegenerative processes. Microglial NOS and NADPH oxidase could promote the generation of ROS via the redox cycling of paraquat-like toxicants. D 2004 Elsevier B.V. All rights reserved. Theme: Disorders of the nervous system Topic: Degenerative disease: Parkinson’s Keywords: Parkinson; Paraquat; Microglia; Redox cycling; NOS; NADPH oxidase; Superoxide
Contents 1. 2.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Materials and methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Cell cultures and treatments . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Redox cycling assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4. Statistical analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Paraquat-induced superoxide release . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Comparison of the effects of paraquat, diquat, and benzyl viologen on superoxide 3.3. Role of NOS and NADPH oxidase in paraquat-induced superoxide release . . . . 4. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
* Corresponding author. Fax: +1 408 734 8522. E-mail address: [email protected] (D.A. Di Monte). 0169-328X/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.molbrainres.2004.11.005
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . release . . . . . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
. . . . . . . . . . . . .
53 53 53 53 53 54 54 54 54 55 55 56 56
D. Bonneh-Barkay et al. / Molecular Brain Research 134 (2005) 52–56
1. Introduction
2. Materials and methods
Exposure of mice to the herbicide paraquat yields a valuable model of Parkinson’s disease (PD)-like neurodegeneration induced by environmental risk factors [9]. Features of paraquat toxicity that mirror PD pathology include a loss of dopaminergic cells in the substantia nigra pars compacta accompanied by microglial and astrocytic activation [15]. Other neuronal cell populations, such as GABAergic neurons in the substantia nigra pars reticulata, are spared by paraquat toxicity, and no glial response occurs outside the nigrostriatal dopaminergic pathway [15]. The mechanism of paraquat-induced dopaminergic cell damage, although presently unknown, is likely to involve a process of redox cycling that begins with the enzymatic one-electron reduction of the herbicide, proceeds with the electron transfer to molecular oxygen and ultimately leads to the formation of superoxide anion and other reactive oxygen species (ROS) [3,5]. Dopaminergic neurons may be preferentially targeted by paraquat neurotoxicity because of their greater vulnerability to ROS-mediated oxidative injury [6,10]. A variety of cellular diaphorases are capable of transferring electrons to paraquat, thus initiating its redox cycling. These enzymes are typically flavin-containing oxidoreductases that use NADH or NADPH as electron donors [4,19]. Of particular relevance to paraquat neurotoxicity is the reported ability of nitric oxide synthase (NOS) to catalyze paraquat reduction and NADPH oxidation [8]. This finding suggests that paraquat bioactivation in the CNS may be promoted within cells that express constitutive (neurons) or inducible (predominantly microglia) NOS isoforms [7,14]. Another enzyme that could potentially play an important role in catalyzing paraquat redox cycling is NADPH oxidase. Although NADPH oxidase is present in various cell types including neurons [21], its activity has mostly been associated with the function of immunocompetent cells such as macrophages and microglia. After a toxic insult to the CNS and upon microglial activation, the subunits that compose NADPH oxidase are assembled at the plasma membrane level where the enzyme is then capable of reducing oxygen to the superoxide radical [22]. It is conceivable that, in the presence of paraquat, NADPH oxidase could transfer electrons to the herbicide and enhance superoxide formation via a redox cycling mechanism. The current study was undertaken to test the hypothesis that microglia play a role in paraquat neurotoxicity by promoting its redox cycling. More specifically, experiments were designed to determine the ability of cultured microglia to generate superoxide radicals after paraquat exposure, to relate this ROS formation to a process of redox cycling and, finally, to identify the roles of NOS and NADPH oxidase in microglia-mediated paraquat redox cycling.
2.1. Materials
53
Paraquat, diquat, benzyl viologen, nitroblue tetrazolium (NBT), superoxide dismutase (SOD), apocynin and NN-nitro-l-arginine methyl ester (l-NAME) were purchased from Sigma (St. Louis, MO). RPMI 1640 medium and fetal calf serum were from GIBCO (Gaithersburg, MD). 2.2. Cell cultures and treatments N9 murine microglial cells (kindly provided by Dr. Paola Ricciardi-Castagnoli) were plated onto 96-well plates and maintained in RPMI 1640 medium supplemented with 5% heat-inactivated fetal calf serum and 25 Am h-mercaptoethanol. When cells reached maximum confluence, experiments were conducted in serum- and phenol red-free RPMI 1640 medium. In experiments in which the effects of NOS and NADPH oxidase inhibitors were tested, cells were incubated in the presence of apocynin (0.5 mM) and/or l-NAME (5 mM) for 30 min prior to paraquat or vehicle addition. 2.3. Redox cycling assay Redox cycling was measured as the production of superoxide anion after treatment with paraquat, diquat or benzyl viologen. Superoxide radical release was quantified as the SOD-sensitive reduction of NBT [13]. NBT is a water-soluble reagent (yellow solution) that, in the presence of superoxide radicals, is converted to nitroblue formazan (insoluble purple precipitate). N9 cells were incubated with vehicle (medium) or different concentrations of the redox cycling agents for 30 min prior to the addition of NBT (1 mg/ml, final concentration). After this addition, incuba-
Fig. 1. Paraquat-induced dose-dependent increase in superoxide release. N9 cells were treated for 90 min with various concentrations (mM) of paraquat. Superoxide radical production was quantified by measuring NBT reduction. Data (n z 6) are mean F SEM and are expressed as % of control values measured in vehicle-treated cultures. *P b 0.001 compared to the other treatment groups.
54
D. Bonneh-Barkay et al. / Molecular Brain Research 134 (2005) 52–56
when differences were observed in the ANOVA testing ( P b 0.05).
3. Results 3.1. Paraquat-induced superoxide release
Fig. 2. Paraquat-induced time-dependent increase in superoxide release. N9 cells were incubated in the presence of 0.5 mM paraquat for 0, 60, 90 or 120 min. Superoxide production was quantified by the reduction of NBT. Data (n z 6) are mean F SEM and are expressed as % of control values measured in vehicle-treated cultures. *P b 0.05 compared to the other treatment groups.
tions proceeded at 37 8C for various periods of time (30 to 90 min post NBT). The medium was then removed, and the cells were resuspended in water and sonicated to solubilize the formazan precipitate. For quantification, optical density was immediately measured at 560 nm in a microtiter plate reader.
In the first set of experiments, N9 cells were exposed for 90 min to concentrations of paraquat (0.1 to 2 mM) similar to those used to trigger redox cycling reactions in other in vitro systems over a relatively short incubation period [8,18]. These incubations resulted in a dose-dependent enhancement of NBT reduction (Fig. 1), with substantial superoxide release caused by concentrations of 0.5 mM or greater. Paraquat-induced changes were also time-dependent. After addition of 0.5 mM paraquat, a significant increase in NBT reduction was already evident at 60 min and became progressively more pronounced at the 90- and 120-min time points (Fig. 2). NBT reduction caused by paraquat was counteracted by inclusion of SOD (600 U/ml) in the incubation medium (data not shown), demonstrating its relation to the formation and release of superoxide radicals. 3.2. Comparison of the effects of paraquat, diquat, and benzyl viologen on superoxide release
Differences among means were analyzed using a one-way ANOVA. Fisher’s post hoc analysis was employed
The rapid NBT reduction caused by paraquat in microglial cultures suggests the involvement of redox cycling reactions. Other bipyridyl pesticides, such as diquat and benzyl viologen, share the ability to generate superoxide radicals through a redox cycling mechanism [18]. We therefore tested the effects of these agents in our in vitro system. As compared
Fig. 3. Effects of bipyridyl compounds on superoxide release. N9 cells were incubated in the presence of paraquat (black bars), benzyl viologen (dark gray bars) or diquat (light gray bars) at 0.1 or 0.5 mM concentrations. Superoxide production was quantified by measuring NBT reduction 90 min after the administration of bipyridyls. Data (n z 6) are mean F SEM and are expressed as % of control values measured in vehicle-treated cultures. *P b 0.001 compared to the other treatment groups at the same concentration.
Fig. 4. Effects of NOS and NADPH oxidase inhibitors on paraquat-induced superoxide release. N9 cells were treated with 5 mM l-NAME, an inhibitor of NOS, and/or 0.5 mM apocynin, an inhibitor of NADPH oxidase, for 30 min prior to the addition of 0.5 mM paraquat. Superoxide production was quantified by measuring NBT reduction 90 min after paraquat administration. Data are (n = 5) mean F SEM and are expressed as % of control values measured in vehicle-treated cultures. *P b 0.05 compared to the other treatment groups.
2.4. Statistical analysis
D. Bonneh-Barkay et al. / Molecular Brain Research 134 (2005) 52–56
to paraquat, both benzyl viologen and diquat induced a more pronounced NBT reduction (Fig. 3). When the three compounds were added for 90 min at a concentration of 0.1 mM, only diquat caused a significant stimulation of superoxide release. At a concentration of 0.5 mM, NBT reduction was increased by approximately three-, four- and six-folds in the presence of paraquat, benzyl viologen and diquat, respectively (Fig. 3). 3.3. Role of NOS and NADPH oxidase in paraquat-induced superoxide release As cellular diaphorases, both NOS and NADPH oxidase could catalyze the redox cycling of paraquat in microglia. Incubation of N9 cells for 30 min in the presence of the NOS inhibitor l-NAME [17] prior to the addition of paraquat resulted in a significant decrease in paraquat-induced NBT reduction (Fig. 4). Similarly, apocynin, an inhibitor of NADPH oxidase [16], significantly prevented the release of superoxide radicals triggered by paraquat exposure. The effect of 0.5 mM paraquat (90 min exposure time) on superoxide release was attenuated by approximately 50% after pre-incubations with either l-NAME or apocynin. However, when the two inhibitors were used together, an additive action was indicated by the finding that NBT reduction returned to the basal levels measured in paraquatfree control cultures (Fig. 4).
4. Discussion Oxidative stress has long been suggested to play an important role in the neurodegenerative process underlying PD [6,10]. Various mechanisms of free radical generation pertinent to PD have been proposed, including the NOSmediated production of NO and peroxynitrite that results from microglial activation and the enhanced formation of ROS after partial inhibition of mitochondrial complex I
Fig. 5. Redox cycling reactions triggered by paraquat (PQ) in microglia.
55
[2,14]. Experimental evidence showing selective dopaminergic cell loss after treatment of mice with paraquat also suggests that neurotoxicants could trigger ROS generation and induce oxidative damage through a process of redox cycling [15]. As a redox cycling agent, paraquat must undergo a one-electron reduction catalyzed by cellular reductases [3,5]. The results of this study provide first evidence that this bioactivation can occur within microglia. Addition of paraquat to microglial cultures caused a time- and dose-dependent production of superoxide radicals. This finding is consistent with the following sequence of redox cycling reactions: the one-electron reduction of paraquat produces a cation radical which, by transferring its electron onto molecular oxygen, readily forms superoxide anion and re-generates the parent compound (Fig. 5). Further evidence in favor of a role of microglia in triggering redox cycling derives from experiments with diquat and benzyl viologen. The rate of redox cycling is dependent upon the electrochemical properties and, in particular, the one-electron reduction potentials of specific compounds [5,12]. Agents that accept electrons less readily are characterized by a more negative potential and are therefore poorer redox cyclers. A comparison of the reduction potentials of paraquat ( 0.44 V vs. a normal hydrogen electrode), benzyl viologen ( 0.35 V) and diquat ( 0.33 V) would predict the ranking of redox cycling to be paraquat b benzyl viologen b diquat [11,12]. Indeed, when the three compounds were added to microglial cultures, superoxide formation was most pronounced in the presence of diquat and least stimulated after paraquat exposure. Taken together, these data suggest that microglia are important sites for the bioactivation of redox cycling toxicants and that, through this novel mechanism, they could play a role in oxidative injury during neurodegenerative processes. Another important component of the present work was to identify specific enzymes responsible for the reduction of paraquat and the initiation of its redox cycling. Previous work has suggested that paraquat redox cycling in cultures of endothelial cells and cytokine-activated macrophages is attenuated by NOS inhibitors [8]. Here, we show that NOS inhibitors are also capable of preventing paraquat-induced superoxide formation by microglia. Furthermore, we identified an additional pathway involved in the reduction of paraquat that was catalyzed by NADPH oxidase (Fig. 5). Both microglial NADPH oxidase and inducible NOS have already been implicated in neurodegenerative processes targeting dopaminergic cells [14,22]. The toxic potential of NADPH oxidase has been related to the direct production of superoxide radicals, whereas NOS activity could promote oxidative injury via the initial formation of NO, the rapid reaction of NO with superoxide radical, and ultimately, the generation of the strong oxidant peroxynitrite. Our results indicate a complementary mechanism by which these two enzymes could contribute to oxidative reactions in the presence of specific toxicants. They could act as electron sources for the redox cycling of paraquat-like compounds,
56
D. Bonneh-Barkay et al. / Molecular Brain Research 134 (2005) 52–56
and through this process, the formation of ROS would likely be significantly enhanced. Other important considerations arise from the participation of microglial NOS and NADPH oxidase in redox cycling reactions. Our data show that NOS inhibitors reduced extracellular levels of superoxide in paraquat-exposed cultures. This finding is consistent with the interpretation that membrane-soluble superoxide radicals, once generated within microglial cells via NOS activity, are released into the extracellular space. In relation to NADPH oxidase, the transmembrane location of this enzyme would likely facilitate the bioactivation of hydrophilic toxicants, including paraquat, characterized by a relatively poor capability of crossing cell membranes. Moreover, NADPH-mediated redox cycling at the plasma membrane level is likely to produce superoxide radicals outward directly into the extracellular space [1]. The release of these radicals from microglia into the extracellular compartment raises the possibility that, in the in vivo setting of human and animal brain, they could ultimately lead to the oxidative damage of nearby neuronal cells. In summary, the present data provide insight into the neurotoxicity of redox cycling agents, pointing to microglia as cell mediators of their metabolic activation and to NOS and NADPH oxidase as enzymatic triggers of redox cycling reactions. These findings are likely to be relevant to our understanding of the sequence of events that begins with exposure to paraquat and paraquat-like compounds and may ultimately lead to the selective degeneration of dopaminergic neurons [15]. Besides bipyridyl pesticides such as paraquat, diquat, and benzyl viologen, a variety of naturally occurring and synthetic molecules (e.g., quinones) share the ability to redox cycle [12,20]. This ability could therefore characterize an entire class of environmental risk factors of particular interest for PD.
Acknowledgments This work was supported by grants from the National Institute of Environmental Health Sciences (ES10442, ES10806 and ES12077, DAD) and the Michael J. Fox Foundation (DB-B). We thank Alison L. McCormack for her valuable comments on this manuscript.
References [1] B.M. Babior, NADPH oxidase: an update, Blood 93 (1999) 1464 – 1476. [2] R. Betarbet, T.B. Sherer, G. MacKenzie, M. Garcia-Osuna, A.V. Panov, J.T. Greenamyre, Chronic systemic pesticide exposure reproduces features of Parkinson’s disease, Nat. Neurosci. 3 (2000) 1301 – 1306. [3] J.S. Bus, J.E. Gibson, Paraquat: model for oxidant-initiated toxicity, Environ. Health Perspect. 55 (1984) 37 – 46.
[4] L. Clejan, A.I. Cederbaum, Synergistic interactions between NADPH-cytochrome P-450 reductase, paraquat, and iron in the generation of active oxygen radicals, Biochem. Pharmacol. 38 (1989) 1779 – 1786. [5] G.M. Cohen, M. d’Arcy Doherty, Free radical mediated cell toxicity by redox cycling chemicals, Br. J. Cancer, Suppl. 8 (1987) 46 – 52. [6] W. Dauer, S. Przedborski, Parkinson’s disease: mechanisms and models, Neuron 39 (2003) 889 – 909. [7] V.L. Dawson, T.M. Dawson, Nitric oxide in neurodegeneration, Prog. Brain Res. 118 (1998) 215 – 229. [8] B.J. Day, M. Patel, L. Calavetta, L.Y. Chang, J.S. Stamler, A mechanism of paraquat toxicity involving nitric oxide synthase, Proc. Natl. Acad. Sci. U. S. A. 96 (1999) 12760 – 12765. [9] D.A. Di Monte, The environment and Parkinson’s disease: is the nigrostriatal system preferentially targeted by neurotoxins? Lancet Neurol. 2 (2003) 531 – 538. [10] S. Fahn, G. Cohen, The oxidant stress hypothesis in Parkinson’s disease: evidence supporting it, Ann. Neurol. 32 (1992) 804 – 812. [11] M. Faro, C. Gomez-Moreno, M. Stankovich, M. Medina, Role of critical charged residues in reduction potential modulation of ferredoxin-NADP+ reductase, Eur. J. Biochem. 269 (2002) 2656 – 2661. [12] D.M. Frank, P.K. Arora, J.L. Blumer, L.M. Sayre, Model study on the bioreduction of paraquat, MPP+, and analogs. Evidence against a bredox cyclingQ mechanism in MPTP neurotoxicity, Biochem. Biophys. Res. Commun. 147 (1987) 1095 – 1104. [13] T.W. Kirby, I. Fridovich, A picomolar spectrophotometric assay for superoxide dismutase, Anal. Biochem. 127 (1982) 435 – 440. [14] G.T. Liberatore, V. Jackson-Lewis, S. Vukosavic, A.S. Mandir, M. Vila, W.G. McAuliffe, V.L. Dawson, T.M. Dawson, S. Przedborski, Inducible nitric oxide synthase stimulates dopaminergic neurodegeneration in the MPTP model of Parkinson disease, Nat. Med. 5 (1999) 1403 – 1409. [15] A.L. McCormack, M. Thiruchelvam, A.B. Manning-Bog, C. Thiffault, J.W. Langston, D.A. Cory-Slechta, D.A. Di Monte, Environmental risk factors and Parkinson’s disease: selective degeneration of nigral dopaminergic neurons caused by the herbicide paraquat, Neurobiol. Dis. 10 (2002) 119 – 127. [16] R.B. Muijsers, E. van Den Worm, G. Folkerts, C.J. Beukelman, A.S. Koster, D.S. Postma, F.P. Nijkamp, Apocynin inhibits peroxynitrite formation by murine macrophages, Br. J. Pharmacol. 130 (2000) 932 – 936. [17] H. Nakanishi, J. Zhang, M. Koike, T. Nishioku, Y. Okamoto, E. Kominami, K. von Figura, C. Peters, K. Yamamoto, P. Saftig, Y. Uchiyama, Involvement of nitric oxide released from microgliamacrophages in pathological changes of cathepsin D-deficient mice, J. Neurosci. 21 (2001) 7526 – 7533. [18] M.S. Sandy, P. Moldeus, D. Ross, M.T. Smith, Role of redox cycling and lipid peroxidation in bipyridyl herbicide cytotoxicity. Studies with a compromised isolated hepatocyte model system, Biochem. Pharmacol. 35 (1986) 3095 – 3101. [19] J. Sarlauskas, A. Nemeikaite-Ceniene, Z. Anusevicius, L. Miseviciene, M.M. Julvez, M. Medina, C. Gomez-Moreno, N. Cenas, Flavoenzymecatalyzed redox cycling of hydroxylamino- and amino metabolites of 2,4,6-trinitrotoluene: implications for their cytotoxicity, Arch. Biochem. Biophys. 425 (2004) 184 – 192. [20] M.T. Smith, Quinones as mutagens, carcinogens, and anticancer agents: introduction and overview, J. Toxicol. Environ. Health 16 (1985) 665 – 672. [21] S.P. Tammariello, M.T. Quinn, S. Estus, NADPH oxidase contributes directly to oxidative stress and apoptosis in nerve growth factordeprived sympathetic neurons, J. Neurosci. 20 (2000) RC53. [22] D.C. Wu, P. Teismann, K. Tieu, M. Vila, V. Jackson-Lewis, H. Ischiropoulos, S. Przedborski, NADPH oxidase mediates oxidative stress in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine model of Parkinson’s disease, Proc. Natl. Acad. Sci. U. S. A. 100 (2003) 6145 – 6150.