Heme-Oxygenase-1 Promotes Polychlorinated Biphenyl Mixture Aroclor 125
http://www.100md.com
《毒物学科学杂志》
Department of Environmental Medicine, University of Rochester of School of Medicine and Dentistry, Rochester, New York 14642
ABSTRACT
Dopaminergic (DAergic) systems have been identified as putative targets for polycholorinated biphenyl (PCB) actions. However, the precise mechanisms leading to neurotoxicity are unresolved. Reactive oxygen species (ROS) were recently shown to mediate injury in DAergic MN9D cells following exposure to Aroclor 1254 (A1254), a commercial PCB mixture. The oxidative stress response in DAergic cells included a persistent expression of heme oxygenase-1 (HO-1). This study tested the hypothesis that a sustained PCB-induced HO-1 response leads to abnormally high Fe levels, which generates ROS production and mediates death in the MN9D DAergic cell model. Accordingly, results indicated that A1254 augmented intracellular Fe levels in MN9D cells after 24 h. Fe chelation by desferoxamine or pharmacologic inhibition of HO activity with tin-protoporphyrin reduced Fe accumulation, ROS production, and cytotoxicity following A1254 exposure. HO-1 over-expression predisposed MN9D DAergic cells to enhanced ROS production and cell death in response to PCBs. Conversely, antisense inhibition of HO-1 expression prevented PCB-induced ROS production and cell death. These observations suggest that enhanced HO-1 catalytic activity and subsequent liberation of Fe participate in neurotoxic DAergic cell injury caused by A1254 exposure in vitro.
Key Words: PCB; neurotoxicity; reactive oxygen species (ROS); iron; neurodegeneration.
INTRODUCTION
Although polychlorinated biphenyls (PCBs) are no longer produced in the United States, exposure remains a profound health hazard because these toxicants are resistant to biodegradation and remain widely distributed in the environment (Giesy and Kannan, 1998; Safe, 1993). PCB exposure has been associated with cognitive and motor neurobehavioral abnormalities in both humans (Jacobson and Jacobson, 2003; Schantz et al., 2003) and experimental animals (Branchi et al., 2005; Eriksson et al., 1991; Widholm et al., 2001). Dopaminergic (DAergic) systems appear to be cellular and neurochemical targets for PCBs (Bemis and Seegal, 2004; Mariussen et al., 2001; Mariussen and Fonnum, 2001; Seegal, 2003; Seegal et al., 2002), but the mechanisms of neurotoxic action are not well defined. In vitro studies have indicated that PCBs disrupt intracellular calcium homeostasis (Bemis and Seegal, 2000; Inglefield et al., 2001, 2002) and interfere with ryanadine receptor function (Howard et al., 2003; Wong et al., 1997a,b). Transient increases in reactive oxygen species (ROS) have also been previously observed following PCB exposure in isolated rat brain synaptosomes (Voie and Fonnum, 2000) and in cerebellar granule neurons in vitro (Mariussen et al., 2002). However, a recent study from our laboratory demonstrated that the commercial PCB mixture Aroclor 1254 (A1254) generated a sustained elevation in ROS production, which preceded DAergic cell death (Lee and Opanashuk, 2004).
Whereas oxidative stress might be involved in A1254 induced DAergic cell injury, the precise intracellular sources and mechanisms of ROS generation in mediating PCB neurotoxicity have not been identified. DAergic neurons are particularly susceptible to oxidative damage due to the high levels of inherent ROS that are produced during dopamine (DA) breakdown by monoamine oxidases or via auto-oxidation to quinones (Cohen et al., 1997; Graham, 1978; Maker et al., 1981). Additionally, H2O2 is generated during DA synthesis by tyrosine hydroxylase (Haavik et al., 1997). Perhaps more importantly, iron bound to neuromelanin within DAergic neurons can subsequently react with metabolically liberated H2O2 through the Fenton reaction to produce extremely toxic hydroxyl radicals (OH). If not properly buffered, excess OH can stimulate lipid peroxidation, which will eventually lead to macromolecular injury and neuronal death. In a previous study, we demonstrated that the PCB-induced generation of ROS in MN9D DAergic cells was accompanied by a depletion of intracellular glutathione (GSH) (Lee and Opanashuk, 2004). This characteristic index of oxidative stress was accompanied by an antioxidant defense response that included a robust increase in heme-oxygenase–1 (HO-1) protein expression.
The heme oxygenase (HO) pathway is considered to be a fundamental defense system for neurons undergoing oxidative challenges (Chen et al., 2000; Dore, 2002). However, recent evidence suggests that HO-1 activity could reach levels that are detrimental to cells (Dennery, 2000; Ryter and Tyrrell, 2000). The microsomal HO-1 enzyme isoform is a 32kDa stress response protein that is readily inducible by a variety of stimuli, including DA, hydrogen peroxide, ultraviolet A radiation, and tumor necrosis factor- (Keyse and Tyrrell, 1989; Mehindate et al., 2001; Schipper et al., 1999). HO-2 is the constitutively expressed isoform and represents the main source of HO activity in the brain (Maines, 2000). HO enzyme activity plays a principal cellular role in catalyzing the breakdown of free heme groups into several byproducts that include carbon monoxide (CO), biliverdin (BV), and ferrous iron (Fe2+) (Ryter and Tyrrell, 2000). Much attention has been focused on HO-1 induction as an adaptive response to oxidative stress since biliverdin, which is readily metabolized into bilirubin, has antioxidant properties (Chen et al., 2000; Fukuda et al., 1996; Maines et al., 1998). Paradoxically, recent studies have reported that HO-1 induction did not protect against hyperoxia in fibroblasts (Suttner and Dennery, 1999) or hemin mediated death in catecholaminergic PC12 cells (Leon et al., 2003). In primary cortical neuronal cultures, HO-2 gene deletion was shown to reduce oxidative stress and decrease vulnerability in response to hemin (Regan et al., 2004) or hemoglobin (Rogers et al., 2003). Therefore, in contrast to its traditional role as a cellular defense mechanism, the induction of HO-1 could be toxic under certain conditions.
As indicated above, Fe is liberated following heme breakdown by HO enzymes. If the excess labile Fe resulting from elevated HO catalytic activity following PCB exposure is not buffered, it may produce neuronal injury by participating in oxidative processes to form highly reactive OH (Ryter and Tyrrell, 2000) or neurotoxic DA-derived semiquinone intermediates (Hermida-Ameijeiras et al., 2004; Pezzella et al., 1997). Indeed, PCB exposure was recently associated with elevated Fe levels in hepatocytes in adult rats (Whysner and Wang, 2001). Considering that HO-1 was profoundly elevated in DAergic cells following A1254 treatment (Lee and Opanashuk, 2004), a potential intracellular source of Fe could arise from the HO-mediated breakdown of heme moieties (Dennery, 2000). Therefore, this study tested the hypothesis that a sustained PCB-induced HO-1 response leads to abnormally high Fe levels, which generates ROS production and mediates death in the MN9D DAergic cell model. The results of this study are consistent with the notion that both enhanced HO-1 catalytic activity during heme breakdown and the subsequent liberation of Fe participate in neurotoxic DAergic cell injury caused by in vitro A1254 exposure.
MATERIALS AND METHODS
Reagents.
Chemicals were purchased from Sigma-Aldrich (St. Louis, MO), unless otherwise stated. 1-Methyl-4-phenylpyridinium (MPP+) was purchased from RBI (Natick, MA), 2',7'-dichlorofluorescein-diacetate (DCF-DA) was obtained from Calbiochem (San Diego, CA). Aroclor 1254 (Lot 124-191; 99% purity) was purchased from Accustandard Inc (New Haven, CT). Cell culture media was acquired from Gibco/BRL (Grand Island, NY).
Dopaminergic MN9D cell line and Aroclor 1254 treatments.
MN9D cells (a generous gift from Drs. Alfred Heller and Lisa Won, Univ. Chicago) were maintained in Dulbecco's modified Eagle's medium containing 15% fetal bovine serum (FBS), and 10,000 U/ml penicillin/streptomycin, as previously described (Heller et al., 1996). Cells were grown to 90% confluency on 100 mm2 dishes before plating on poly-D-lysine (100 μg/ml) coated 12-well plates for experiments (1.5 x 105 cells/well).
Arolor 1254 (A1254) was resuspended in 100% DMSO. The experimental dilutions were prepared as 1000x stock solutions in 100% DMSO. The experimental working concentrations were prepared from the 1000x stocks in culture media and placed in a sonifying water bath the night before the exposures were initiated. MN9D cells were exposed to 0–20 ppm A1254 for 24–48 h. This experimental paradigm was previously shown to cause concentration and time dependent increases in DAergic cell death after 48 h (Lee and Opanashuk, 2004). Cell death was shown to be preceded by and dependent upon oxidative stress. In some experiments, cultures were co-treated with 100 μM desferoxamine (DES), an Fe chelator, or pretreated for 3 h with 5 μM tin- protoporphyrin (SnPPIX), a pharmacologic HO-1 inhibitor. Reactive oxygen species (ROS) production was analyzed after 24 h. Cytotoxicity was assessed 48 h following exposure.
Recombinant adenovirus generation and treatment.
Adenovirus carrying the green fluorescent protein, GFP (Ad.GFP), HO-1 cDNA (Ad.HO-1), and the HO-1 cDNA oriented in the anti-sense direction (Ad.antiHO-1) were kindly provided by Dr. Heiling Lu (CRC, Mass. Gen. Hospital, Harvard Medical School, MA). Dr. Lu and coworkers have previously described the generation of these replication-deficient adenoviruses (He et al., 1998). Multiplicity of infection (MOI) of 10–200 (number of virus particles per cell) was based on the concentrations of the stock solutions that were provided by Dr. Liu. MN9D cells in DMEM containing 15% FBS were transfected with 0–200 MOI of adenovirus carrying the Ad.GFP, Ad.HO-1, and Ad.antiHO-1 HO-1 cDNAs for 24 h prior to treatment with A1254 (He et al., 1998; Liu et al., 2003). The Ad.GFP vector served as a control for the transfection procedure. LDH release was measured to evaluate the toxicity of adenoviral infections. The GFP adenovirus control construct (that lacked either HO-1 or anti-sense HO-1) was used to rule out adverse effects associated with the transfection procedure. GFP fluorescence was monitored by microscopy for all transfections. Transfection efficiency as measured by GFP fluorescence intensity was observed to be similar following infection with all three adenoviral constructs (i.e., 60% for 200 MOI). The respective increases and decreases in HO-1 protein following transfection with either HO-1 or antisense HO-1 were quantified by immunoblot analyses. For MOI 100–200, heme oxygenase levels were altered by 50–80% (data not shown).
Immunoblot analysis.
MN9D cells were harvested in phosphate buffered saline (PBS) containing 0.3% Triton, protease inhibitors (100 mM AEBSF, 0.08 mM aprotinin, 2 mM leupeptin, 1.5 mM pepstatin A, 4 mM bestatin, 1.4 mM E-64; Sigma), 5 mM EDTA, and 2 mM phenylmethyl sulfonyl fluoride. Protein concentrations were determined using a MicroBCA assay from Pierce (Rockford, IL). Twenty-five μg of protein were separated on 10% acrylamide gels before transfer of proteins onto a polyvinylidene difluoride (PVDF) membrane. Membranes were blocked with 5% powdered milk solution in 0.3% Triton-PBS solution before incubation with 1:2000 HO-1 (StressGen, Victoria, B.C.). Actin (Sigma, St. Louis, MO; 1:1500) was used as a loading control. Protein bands were detected by a chemiluminescent substrate for horseradish peroxidase (LumiGLO, KPL, Gaithersburg, MD) and quantified by Scion Image analysis software (v.4.0.2).
Atomic absorption spectroscopy.
Total intracellular Fe was quantified by atomic absorption spectroscopy following exposure to 20 ppm Aroclor 1254 for 24 h. Parallel cultures were co-treated with 100 μM DES to chelate Fe or pretreated for 3 h with 5 μM SnPPIX to inhibit HO activity. Cellular extracts (1 mg protein) were digested in concentrated nitric acid (HNO3) at 90°C in ultra-clean Teflon vials. Samples were resuspended in 2% HNO3 and analyzed on a Perkin-Elmer AA600 Graphite Furnace atomic absorption spectrometer with a Zeeman background correction. Fe content was measured using a single element (Fe) lamp after a pyrolysis (1300°C) and atomization (2150°C) step. A matrix modifier solution of Mg (NO3)2 was used before readings were taken at a wavelength of 248.3 nm. The estimated limit of detection was 1.542 ng/ml.
Dichlorofluorescein-diacetate (DCF-DA) assay.
ROS levels were analyzed 24 h following exposure to 5–20 ppm A1254 with the DCF-DA assay as previously described (Mariussen et al., 2002; Wang and Joseph, 1999). DCF-DA readily permeates cellular membranes whereby its acetate moieties are cleaved by intracellular esterases. Upon reaction with ROS, such as lipid and hydrogen peroxide radicals, the probe fluoresces (Ex: 485nm and Em: 530nm). After PCB treatment, MN9D cells were incubated with 10 μM DCF-DA dissolved in DMSO for 30 min at 37°C, rinsed with Dulbecco's phosphate buffered saline (DPBS), and transferred into 96-well plates for fluorescence measurements.
Lactate dehydrogenase measurement.
Lactate dehydrogenase (LDH) release was measured to assess cell death (Decker and Lohmann-Matthes, 1988) 48 h following exposure to 5–20 ppm A1254. This measure of cell viability is based on the rate of change in NADH concentration as pyruvate is converted to lactate. Media was collected from individual wells at the end of each experiment and assayed using a spectrophotometer. The extinction coefficient of NADH at 340 nm and pH 7.5 is 0.00622 and was used in the calculation of the total amount of LDH released from the slopes collected during spectrophotometric analysis.
Statistical analysis.
Statistical differences were analyzed by comparing means among different treatment groups using StatView version 5.0 (SAS Institute, Cary, NC). Analysis of variance (ANOVA) was applied with treatment and cotreatment variables (where applicable) to detect differences between these parameters and toxicant exposure. Post-hoc comparisons of mean values were by Fisher's PLSD and statistical differences were considered significant with a p-value of <0.05.
RESULTS
Fe Chelation and HO-1 Inhibition Reduces PCB-Induced Intracellular Fe Accumulation in DAergic MN9D Cells
Treatment with A1254 elicited a significant increase in total intracellular iron (Fe) compared to DMSO controls (Fig. 1.; CON; black bars). In the presence of DES, an Fe chelator, Fe levels declined in both DMSO and A1254 compared to untreated controls (treatment effect p < 0.001). PCB-induced Fe levels diminished by approximately 56% in the presence of DES (Fig. 1; +DES; white bars; p < 0.05, compared to A1254). Furthermore, inhibition of HO-1 following a pretreatment with SnPPIX, significantly reduced Fe accumulation by 50% in response to A1254 (Fig. 1; +SnPPIX; vertically-striped bars p < 0.05). These observations suggest that Fe accumulation in MN9D DAergic cells following PCB treatment could emanate from elevated HO-1 activity.
Fe Chelation and HO-1 Inhibition Prevent PCB-Induced ROS Production and DAergic Cell Death
A1254 produced a concentration dependent elevation in ROS levels in MN9D cells following exposure (Fig. 2A; effect of treatment p < 0.0001). Iron chelation with DES and HO inhibition with SnPPIX significantly diminished the ROS production in response to PCBs (Fig. 2A; p < 0.05). To determine whether excessive Fe accumulation or elevated HO-1 activity could mediate neurotoxicity, LDH release was measured to monitor DAergic cell viability. Both Fe chelation and pharmacologic HO-1 inhibition reduced the PCB-induced cell death observed following A1254 treatment (Fig. 2B). Together these findings suggest that HO-1 activity and Fe accumulation produce ROS and cytotoxicity in MN9D cells following A1254 treatment.
HO-1 Over-expression Enhances PCB-Induced ROS Production and DAergic Cell Death
As an additional strategy to test the hypothesis that HO-1 upregulation contributes to PCB-induced oxidative stress and cell death, HO-1 levels were over-expressed in DAergic cells. In the absence of PCBs, MN9D cells expressed concentration-dependent increases in HO-1 protein levels incrementally with multiplicity of infection (MOI) (Fig. 3A). To determine if oxidative stress contributes to PCB-induced cell death following HO-1-transfection, MN9D cells were treated with A1254 prior to measuring ROS production with the DCF-DA assay. At the higher MOI (100 and 200) levels, ROS production was significantly increased in both the control and DMSO groups (Fig. 3B; effect of MOI, p < 0.0001) in the absence of PCBs. PCB treatment further enhanced free radical generation in MN9D cells that over-expressed HO-1 with a significant effect of treatment (p < 0.0001), MOI (p < 0.0001), and an interaction between treatment and MOI (p < 0.0001). Interestingly, HO-1 over-expression alone at the highest MOI (100–200) increased MN9D cell death by approximately 60–120% (Fig. 3C; p < 0.0001 ). Cell death was more pronounced following treatment with A1254 in MN9D cells that over-expressed HO-1 with a significant effect of treatment (p < 0.0001), MOI concentration (p < 0.0001), and an interaction between treatment and MOI concentration (p < 0.0001) (Fig. 3C). These observations support the notion that the sustained elevation of HO-1 following PCB exposure contributes to oxidative stress and subsequently, DAergic cell death.
Anti-sense Inhibition of HO-1 Expression Diminishes PCB-Induced ROS Production and DAergic Cell Death
To determine whether the inhibition of HO-1 expression would protect against PCB-induced ROS generation and toxicity, MN9D cells were transfected with adenovirus carrying an HO-1 cDNA construct in the anti-sense orientation. The reduction in HO-1 protein expression following transfection with the anti-sense HO-1 cDNA construct (Ad.HO-1) was confirmed by immunoblot analyses (Fig. 4A). Uninfected, adenovirus-infected control (Ad.GFP), and HO-1 over-expressing MN9D cells treated with PCBs expressed greater levels of HO-1 protein, compared to DMSO controls. Furthermore, HO-1 expression was not observed in DMSO controls or PCB treated MN9D cells that were infected with the antisense Ad.HO-1 construct. PCB treatment in nave cells resulted in elevated ROS production (Fig. 4B). The PCB-induced ROS levels diminished with increasing MOI in cultures that contained the Ad.HO-1 construct (Fig. 4B). Whereas treatment with 20 ppm A1254 significantly increased cell death, 5 ppm A1254 had a minimal effect on viability (Fig. 4C). However, DAergic cells infected with the Ad.HO-1 construct remained viable following A1254 treatment (Fig. 4C), suggesting that inhibition of HO-1 expression and activity is neuroprotective against PCB toxicity. Together these studies verify that ROS production precedes PCB-induced DAergic cell death. Furthermore, the observations from these experiments support the contention that a robust HO-1 induction following PCB treatment can lead to excessive free radical production, which compromises cell survival.
DISCUSSION
Although DAergic systems have previously been identified as targets for PCB actions (Bemis and Seegal, 2004; Mariussen et al., 2001; Mariussen and Fonnum, 2001; Seegal, 2003; Seegal et al., 2002), the precise mechanism leading to neurotoxicity at the cellular level is unresolved. Our recent study demonstrated that ROS are involved in the biochemical events that contribute to DAergic cell injury following in vitro exposure to A1254, a PCB mixture (Lee and Opanashuk, 2004). The most notable index of oxidative stress was the persistent expression of HO-1 protein following PCB treatment. Therefore, the present study tested the hypothesis that elevated HO-1 expression and catalytic activity produces excess Fe, which promotes PCB-induced oxidative stress and neurotoxicity in DAergic cells. Our results indicate that A1254 exposure increased intracellular Fe levels in MN9D cells. Moreover, Fe chelation by DES or pharmacologic inhibition of HO activity with SnPPIX precluded the Fe accumulation, ROS production, and cytotoxicity induced by A1254 exposure. These findings suggest that A1254 activates an oxidative cascade that includes both increased HO-1 catalytic activity and Fe accumulation. Presumably, this response leads to PCB-mediated DAergic cell injury.
The release of redox active Fe (Fe2+) following heme degradation has been postulated to mediate the pro-oxidant consequences of HO activity (Ryter and Tyrrell, 2000; Schipper, 2004). In our study, atomic absorption spectroscopy (AAS) analyses indicated that Fe levels were elevated in DAergic cells following PCB treatment. However, the disadvantage of AAS is that this method is unable to distinguish free from protein-bound Fe. This makes it difficult to determine the origin and oxidation states of the excess intracellular Fe observed following PCB exposure (Fig. 1). The reduction in Fe levels observed in cells treated with DES suggests that a potential source of Fe arises from the extracellular media. Serum-bound Fe could be imported into the cell via the transferrin receptor or the divalent metal transporter (Moos and Morgan, 2004). It should be acknowledged, however, that DES is not a specific Fe chelator, i.e., DES has been shown to sequester zinc and cadmium (Kaur and Andersen, 2002). Therefore, the possibility cannot be excluded that other metals might also be involved in PCB neurotoxicity. Regardless of origin, the reduced Fe levels following pretreatment with SnPPIX to inhibit HO activity suggest that the Fe is liberated from the HO-mediated heme degradation reaction in MN9D cells in response to A1254 treatment (Fig. 1). HO inhibitors and Fe chelators have previously been shown to diminish lipid peroxidation and Fe accumulation in pulmonary artery smooth muscle cells treated with hemoglobin (Lamb et al., 1999), suggesting that the release of redox active Fe from HO-mediated heme breakdown can lead to cellular injury via lipid peroxidation. Indeed, our study demonstrated that both HO-1 inhibition and Fe chelation protected DAergic cells against PCB-mediated ROS production and cell death.
Inhibition of HO by SnPPIX pretreatment suggested that elevated HO-1 enzyme activity following PCB treatment liberates excess Fe that becomes available to participate in oxidative reactions. However, a disadvantage of using SnPPIX and other metalloporphyrins is that they are associated with nonselective effects apart from inhibiting HO activity (Grundemar and Ny, 1997; Maines and Trakshel, 1992). Metalloporphyrins have been shown to interfere with guanylate cyclase (Zakhary et al., 1996), nitric oxide synthase (Meffert et al., 1994), and calcium signaling (Chen et al., 2000). Therefore, an additional strategy used in our study for evaluating the potential consequences of elevated HO-1 protein levels in response to PCBs involved genetically manipulating the cellular expression of the HO-1 enzyme isoform. Over-expression of HO-1 predisposed MN9D DAergic cells to enhanced ROS production and cell death in response to A1254 exposure. Conversely, antisense inhibition of HO-1 expression prevented PCB-induced ROS production and cell death. These observations further support our hypothesis that HO-1 protein induction and concurrent increases in enzyme activity are sources of Fe and ROS that lead to DAergic cell injury following exposure to A1254 in the MN9D cell model.
It currently remains unclear whether excessive HO-1 activation is the primary intracellular signaling event leading to the mechanism of PCB neurotoxicity in DAergic cells. PCBs could also interfere with Fe efflux systems to augment intracellular accumulation of labile Fe, which can interact with other oxygen radicals to result in cytotoxicity. Unfortunately, the precise mechanisms of Fe efflux from neurons are not well characterized (Moos and Morgan, 2004). Previous studies have suggested that PCBs mediate neurotoxicity by disrupting intracellular Ca+ homeostasis in non-DAergic neurons (Bemis and Seegal, 2000; Inglefield et al., 2001, 2002). Additionally, ortho-substituted PCBs have been shown to interfere with ryanodine receptors (RyR), which regulate the release of Ca+ from intracellular stores (Howard et al., 2003; Wong et al., 1997a,b; Wong and Pessah, 1996). It is possible that abnormal intracellular Ca+ signaling, perhaps through disruption of RyR function (Howard et al., 2003; Wong et al., 1997a,b), could similarly exert PCB neurotoxicity in DAergic cells. As an alternative possibility, DAergic cell injury could arise from ROS produced as a consequence of increased DA turnover following A1254 exposure (Lee and Opanashuk, 2004).
The limitations of the MN9D DAergic cell model are recognized. However, this system is advantageous because the direct effects of PCB exposure on the biochemical pathways that mediate neuronal damage can be studied in a homogeneous population of DAergic cells. It is acknowledged that the A1254 concentrations tested in our study appear to be relatively high compared to expected tissue levels following in vivo exposure (Meacham et al., 2005). The problems with trying to make valid dosimetric comparisons between in vitro and in vitro systems are not restricted to PCB neurotoxicity but represent a common issue encountered with other toxicants, for example, acrylamide (Barber and LoPachin, 2004). This does not necessarily invalidate our in vitro findings but highlights the toxicokinetic differences and cumulative physiochemical events that mediate neurotoxicity in vivo. Additional studies are necessary to determine whether HO-1 and Fe are involved in the mechanism underlying DAergic neurotoxicity in vivo. It will also be important to determine whether HO-1 expression is associated with PCB-induced injury in non-DAergic neurons and other neural cell types. Nevertheless, our data suggest that PCB neurotoxicity involves a complex mechanism that includes ROS production, HO activation, and dysregulation of Fe management processes in DAergic cells following A1254 exposure.
The conclusion that HO-1 mediates PCB neurotoxicity in DAergic cells is contrary to the cytoprotective role typically ascribed. Rapid elevation in HO-1 expression as a response to oxidative stress is thought to be a cellular defense mechanism against macromolecular damage in various experimental models (Chen et al., 2000; Choi et al., 2004; Ferris et al., 1999; Zhang et al., 2004). However, increasing evidence suggests that HO-1 induction (Lamb et al., 1999; Leon et al., 2003; Schipper et al., 1999; Suttner and Dennery, 1999), as well as constitutive HO-2 activity (Regan et al., 2004; Rogers et al., 2003), can also produce cell injury, presumably through excessive ROS generation. DAergic cells are particularly vulnerable to oxidative damage given their inherent redox active microenvironment. Perhaps the most intriguing observation in our study was that HO-1 over-expression in MN9D control cells (in the absence of PCBs) was associated with elevated ROS levels and cytotoxicity (Fig. 3). These findings further strengthen the notion that excessive HO-1 activity, even in the absence of exogenous stimulation, can generate ROS that interfere with DAergic cell survival and function. Our study indicates that A1254, a commercial PCB mixture, further increases this deleterious oxidative cascade in MN9D cells. It is proposed that PCB neurotoxicity is dependent upon the duration and intensity of HO protein induction and enzyme activity. Taken together, our data suggest that there is a critical threshold of HO-1 activation following PCB exposure that is detrimental to DAergic cells.
The increased expression of HO-1 and elevated Fe levels resulting from A1254 exposure represents a potential mechanism by which PCBs exert their neurotoxicity in DAergic cells. It has been hypothesized that environmental factors, such as pesticide or metal exposure, contribute to DAergic neurodegeneration in Parkinson's disease (PD) (Di Monte et al., 2002; Tanner et al., 1999). A potential link between PCB exposure and neurodegeneration was suggested in a previous study that reported elevated PCBs and other organochlorine compounds in the brains of PD patients (Corrigan et al., 1998). Interestingly, both HO-1 and Fe have been connected to neurodegenerative disorders such as PD. HO-1 protein and Fe levels were shown to be elevated in the substantia nigra of PD brains (Schipper et al., 1998; Sofic et al., 1988). It is conceivable that both excess HO activation and Fe accumulation generate ROS that produce DAergic cell loss in PD. However, a causal relationship between HO-1, Fe, and DAergic neurodegeneration remains to be established. Understanding the role of HO and Fe in PCB induced DAergic neuronal injury could have global implications for the pathophysiology of Parkinsonism. It is reasonable to speculate that PCB accumulation results in the persistent elevation of HO-1 and Fe to produce abnormally high ROS levels, which could lead to DAergic cell loss. Therefore, exposure to PCBs and other environmental contaminants that produce oxidative stress, HO-1 induction, and disrupt Fe homeostasis, should be considered as potential risk factors for development of neurodegenerative disorders, including PD.
ACKNOWLEDGMENTS
The authors thank Drs. Alfred Heller and Lisa Won for providing MN9D cells and advice related to these studies. We express our gratitude to Dr. Heiling Lu for generously providing the HO-1 adenoviral constructs. We also appreciate the editorial assistance provided by Dr. Randy LoPachin and Sarah Notter. This research was supported by the National Institute of Health (NIH) grants ES00375, ES01247, and T32 ES07026.
REFERENCES
Barber, D. S., and LoPachin, R. M. (2004). Proteomic analysis of acrylamide-protein adduct formation in rat brain synaptosomes. Toxicol. Appl. Pharmacol. 201, 120–136.
Bemis, J. C., and Seegal, R. F. (2000). Polychlorinated biphenyls and methylmercury alter intracellular calcium concentrations in rat cerebellar granule cells. Neurotoxicology 21, 1123–1134.
Bemis, J. C., and Seegal, R. F. (2004). PCB-induced inhibition of the vesicular monoamine transporter predicts reductions in synaptosomal dopamine content. Toxicol. Sci. 80, 288–295.
Branchi, I., Capone, F., Vitalone, A., Madia, F., Santucci, D., Alleva, E., and Costa, L. G. (2005). Early developmental exposure to BDE 99 or Aroclor 1254 affects neurobehavioural profile: Interference from the administration route. Neurotoxicology 26, 183–192.
Chen, K., Gunter, K., and Maines, M. D. (2000). Neurons overexpressing heme oxygenase-1 resist oxidative stress-mediated cell death. J. Neurochem. 75, 304–313.
Choi, B. M., Pae, H. O., Jeong, Y. R., Oh, G. S., Jun, C. D., Kim, B. R., Kim, Y. M., and Chung, H. T. (2004). Overexpression of heme oxygenase (HO)-1 renders Jurkat T cells resistant to fas-mediated apoptosis: Involvement of iron released by HO-1. Free Radic Biol. Med. 36, 858–871.
Cohen, G., Farooqui, R., and Kesler, N. (1997). Parkinson disease: A new link between monoamine oxidase and mitochondrial electron flow. Proc. Natl. Acad. Sci. U.S.A. 94, 4890–4894.
Corrigan, F. M., Murray, L., Wyatt, C. L., and Shore, R. F. (1998). Diorthosubstituted polychlorinated biphenyls in caudate nucleus in Parkinson's disease. Exp. Neurol. 150, 339–342.
Decker, T., and Lohmann-Matthes, M. L. (1988). A quick and simple method for the quantitation of lactate dehydrogenase release in measurements of cellular cytotoxicity and tumor necrosis factor (TNF) activity. J. Immunol. Methods 115, 61–69.
Dennery, P. A. (2000). Regulation and role of heme oxygenase in oxidative injury. Curr. Topics Cell. Regul. 36, 181–199.
Di Monte, D. A., Lavasani, M., and Manning-Bog, A. B. (2002). Environmental factors in Parkinson's disease. Neurotoxicology 23, 487–502.
Dore, S. (2002). Decreased activity of the antioxidant heme oxygenase enzyme: Implications in ischemia and in Alzheimer's disease. Free Radical Biol. Med. 32, 1276–1282.
Eriksson, P., Lundkvist, U., and Fredriksson, A. (1991). Neonatal exposure to 3,3',4,4'-tetrachlorobiphenyl: Changes in spontaneous behaviour and cholinergic muscarinic receptors in the adult mouse. Toxicology 69, 27–34.
Ferris, C. D., Jaffrey, S. R., Sawa, A., Takahashi, M., Brady, S. D., Barrow, R. K., Tysoe, S. A., Wolosker, H., Baranano, D. E., Dore, S., Poss, K. D., and Snyder, S. H. (1999). Haem oxygenase-1 prevents cell death by regulating cellular iron. Nature Cell Biology 1, 152–157.
Fukuda, K., Richmon, J. D., Sato, M., Sharp, F. R., Panter, S. S., and Noble, L. J. (1996). Induction of heme oxygenase-1 (HO-1) in glia after traumatic brain injury. Brain Res. 736, 68–75.
Giesy, J. P., and Kannan, K. (1998). Dioxin-like and non-dioxin-like toxic effects of polychlorinated biphenyls (PCBs): Implications for risk assessment. Crit. Rev. Toxicol. 28, 511–569.
Graham, D. G. (1978). Oxidative pathways for catecholamines in the genesis of neuromelanin and cytotoxic quinones. Mol. Pharmacol. 14, 633–643.
Grundemar, L., and Ny, L. (1997). Pitfalls using metalloporphyrins in carbon monoxide research. Trends Pharmacol. Sci. 18, 193–195.
Haavik, J., Almas, B., and Flatmark, T. (1997). Generation of reactive oxygen species by tyrosine hydroxylase: A possible contribution to the degeneration of dopaminergic neurons J. Neurochem. 68, 328–332.
He, T. C., Zhou, S., da Costa, L. T., Yu, J., Kinzler, K. W., and Vogelstein, B. (1998). A simplified system for generating recombinant adenoviruses. Proc. Natl. Acad. Sci. U.S.A. 95, 2509–2514.
Heller, A., Price, S., and Won, L. (1996). Glial-derived neurotrophic factor (GDNF) induced morphological differentiation of an immortalized monoclonal hybrid dopaminergic cell line of mesencephalic neuronal origin. Brain Res. 725, 132–136.
Hermida-Ameijeiras, A., Mendez-Alvarez, E., Sanchez-Iglesias, S., Sanmartin-Suarez, C., and Soto-Otero, R. (2004). Autoxidation and MAO-mediated metabolism of dopamine as a potential cause of oxidative stress: Role of ferrous and ferric ions. Neurochem. Int. 45, 103–116.
Howard, A. S., Fitzpatrick, R., Pessah, I., Kostyniak, P., and Lein, P. J. (2003). Polychlorinated biphenyls induce caspase-dependent cell death in cultured embryonic rat hippocampal but not cortical neurons via activation of the ryanodine receptor. Toxicol. Appl. Pharmacol. 190, 72–86.
Inglefield, J. R., Mundy, W. R., Meacham, C. A., and Shafer, T. J. (2002). Identification of calcium-dependent and -independent signaling pathways involved in polychlorinated biphenyl-induced cyclic AMP-responsive element-binding protein phosphorylation in developing cortical neurons. Neuroscience 115, 559–573.
Inglefield, J. R., Mundy, W. R., and Shafer, T. J. (2001). Inositol 1,4,5-triphosphate receptor-sensitive Ca(2+) release, store-operated Ca(2+) entry, and cAMP responsive element binding protein phosphorylation in developing cortical cells following exposure to polychlorinated biphenyls. J. Pharmacol. Exp. Ther. 297, 762–773.
Jacobson, J. L., and Jacobson, S. W. (2003). Prenatal exposure to polychlorinated biphenyls and attention at school age. J. Pediatr. 143, 780–788.
Kaur, D., and Andersen, J. K. (2002). Ironing out Parkinson's disease: Is therapeutic treatment with iron chelators a real possibility Aging Cell 1, 17–21.
Keyse, S. M., and Tyrrell, R. M. (1989). Heme oxygenase is the major 32-kDa stress protein induced in human skin fibroblasts by UVA radiation, hydrogen peroxide, and sodium arsenite. Proc. Natl. Acad. Sci. U.S.A. 86, 99–103.
Lamb, N. J., Quinlan, G. J., Mumby, S., Evans, T. W., and Gutteridge, J. M. (1999). Haem oxygenase shows pro-oxidant activity in microsomal and cellular systems: Implications for the release of low-molecular-mass iron. Biochem. J. 344, 153–158.
Lee, D. W., and Opanashuk, L. A. (2004). Polychlorinated biphenyl mixture aroclor 1254-induced oxidative stress plays a role in dopaminergic cell injury. Neurotoxicology 25, 925–939.
Leon, A., Le Foll, I., Charriault-Marlangue, C., Leprince, J., Vaudry, H., Gabriel, C., and Duval, D. (2003). Level of haem oxygenase does not obligatorily reflect the sensitivity of PC12 cells to an oxidative shock induced by glutathione depletion. J. Neurochem. 84, 459–470.
Liu, H., Nowak, R., Chao, W., and Bloch, K. D. (2003). Nerve growth factor induces anti-apoptotic heme oxygenase-1 in rat pheochromocytoma PC12 cells. J. Neurochem. 86, 1553–1563.
Maines, M. D. (2000). The heme oxygenase system and its functions in the brain. Cell. Mol. Biol. 46, 573–585.
Maines, M. D., Polevoda, B., Coban, T., Johnson, K., Stoliar, S., Huang, T. J., Panahian, N., Cory-Slechta, D. A., and McCoubrey, W. K., Jr. (1998). Neuronal overexpression of heme oxygenase-1 correlates with an attenuated exploratory behavior and causes an increase in neuronal NADPH diaphorase staining. J. Neurochem. 70, 2057–2069.
Maines, M. D., and Trakshel, G. M. (1992). Differential regulation of heme oxygenase isozymes by Sn- and Zn-protoporphyrins: Possible relevance to suppression of hyperbilirubinemia. Biochimica et Biophysica Acta 1131, 166–174.
Maker, H. S., Weiss, C., Silides, D. J., and Cohen, G. (1981). Coupling of dopamine oxidation (monoamine oxidase activity) to glutathione oxidation via the generation of hydrogen peroxide in rat brain homogenates. J. Neurochem. 36, 589–593.
Mariussen, E., Andersson, P. L., Tysklind, M., and Fonnum, F. (2001). Effect of polychlorinated biphenyls on the uptake of dopamine into rat brain synaptic vesicles: A structure-activity study. Toxicol. Appl. Pharmacol. 175, 176–183.
Mariussen, E., and Fonnum, F. (2001). The effect of polychlorinated biphenyls on the high affinity uptake of the neurotransmitters, dopamine, serotonin, glutamate and GABA, into rat brain synaptosomes. Environ. Res. 86, 2–11.
Mariussen, E., Myhre, O., Reistad, T., and Fonnum, F. (2002). The polychlorinated biphenyl mixture aroclor 1254 induces death of rat cerebellar granule cells: The involvement of the N-methyl-D-aspartate receptor and reactive oxygen species. Toxicol. Appl. Pharmacol. 179, 137–144.
Meacham, C. A., Freudenrich, T. M., Anderson, W. L., Sui, L., Lyons-Darden, T., Barone, S., Jr., Gilbert, M. E., Mundy, W. R., and Shafer, T. J. (2005). Accumulation of methylmercury or polychlorinated biphenyls in in vitro models of rat neuronal tissue. Toxicol. Appl. Pharmacol. 205, 177–187.
Meffert, M. K., Haley, J. E., Schuman, E. M., Schulman, H., and Madison, D. V. (1994). Inhibition of hippocampal heme oxygenase, nitric oxide synthase, and long-term potentiation by metalloporphyrins. Neuron 13, 1225–1233.
Mehindate, K., Sahlas, D. J., Frankel, D., Mawal, Y., Liberman, A., Corcos, J., Dion, S., and Schipper, H. M. (2001). Proinflammatory cytokines promote glial heme oxygenase-1 expression and mitochondrial iron deposition: Implications for multiple sclerosis. J. Neurochem. 77, 1386–1395.
Moos, T., and Morgan, E. H. (2004). The metabolism of neuronal iron and its pathogenic role in neurological disease: Review. Ann. N.Y. Acad. Sci. 1012, 14–26.
Pezzella, A., d'Ischia, M., Napolitano, A., Misuraca, G., and Prota, G. (1997). Iron-mediated generation of the neurotoxin 6-hydroxydopamine quinone by reaction of fatty acid hydroperoxides with dopamine: A possible contributory mechanism for neuronal degeneration in Parkinson's disease. J. Med. Chem. 40, 2211–2216.
Regan, R. F., Chen, J., and Benvenisti-Zarom, L. (2004). Heme oxygenase-2 gene deletion attenuates oxidative stress in neurons exposed to extracellular hemin. BMC Neurosci. 5, 34.
Rogers, B., Yakopson, V., Teng, Z. P., Guo, Y., and Regan, R. F. (2003). Heme oxygenase-2 knockout neurons are less vulnerable to hemoglobin toxicity. Free Radic. Biol. Med. 35, 872–881.
Ryter, S. W., and Tyrrell, R. M. (2000). The heme synthesis and degradation pathways: Role in oxidant sensitivity. Heme oxygenase has both pro- and antioxidant properties. Free Radical Biol. Med. 28, 289–309.
Safe, S. (1993). Toxicology, structure-function relationship, and human and environmental health impacts of polychlorinated biphenyls: Progress and problems. Environ. Health Perspect. 100, 259–268.
Schantz, S. L., Widholm, J. J., and Rice, D. C. (2003). Effects of PCB exposure on neuropsychological function in children. Environ. Health Perspect. 111, 357–576.
Schipper, H. M. (2004). Heme oxygenase-1: Transducer of pathological brain iron sequestration under oxidative stress. Ann. N.Y. Acad. Sci. 1012, 84–93.
Schipper, H. M., Bernier, L., Mehindate, K., and Frankel, D. (1999). Mitochondrial iron sequestration in dopamine-challenged astroglia: Role of heme oxygenase-1 and the permeability transition pore. J. Neurochem. 72, 1802–1811.
Schipper, H. M., Liberman, A., and Stopa, E. G. (1998). Neural heme oxygenase-1 expression in idiopathic Parkinson's disease. Exp. Neurol. 150, 60–68.
Seegal, R. F. (2003). PCBs and dopamine function: Neurological effects of polychlorinated biphenyls: does occupational exposure alter dopamine-mediated function Ann. N.Y. Acad. Sci. 991, 322–325.
Seegal, R. F., Okoniewski, R. J., Brosch, K. O., and Bemis, J. C. (2002). Polychlorinated biphenyls alter extraneuronal but not tissue dopamine concentrations in adult rat striatum: An in vivo microdialysis study. Environ. Health Perspect. 110, 1113–1117.
Sofic, E., Riederer, P., Heinsen, H., Beckmann, H., Reynolds, G. P., Hebenstreit, G., and Youdim, M. B. (1988). Increased iron (III) and total iron content in post mortem substantia nigra of Parkinsonian brain. J. Neural Transmission - General Section 74, 199–205.
Suttner, D. M., and Dennery, P. A. (1999). Reversal of HO-1 related cytoprotection with increased expression is due to reactive iron. FASEB J. 13, 1800–1809.
Tanner, C. M., Ottman, R., Goldman, S. M., Ellenberg, J., Chan, P., Mayeux, R., and Langston, J. W. (1999). Parkinson disease in twins: An etiologic study. JAMA 281, 341–346.
Voie, O. A., and Fonnum, F. (2000). Effect of polychlorinated biphenyls on production of reactive oxygen species (ROS) in rat synaptosomes. Arch. Toxicol. 73, 588–593.
Wang, H., and Joseph, J. A. (1999). Quantifying cellular oxidative stress by dichlorofluorescein assay using microplate reader. Free Radic. Biol. Med. 27, 612–616.
Whysner, J., and Wang, C. X. (2001). Hepatocellular iron accumulation and increased cell proliferation in polychlorinated biphenyl-exposed Sprague-Dawley rats and the development of hepatocarcinogenesis. Toxicol. Sci. 62, 36–45.
Widholm, J. J., Clarkson, G. B., Strupp, B. J., Crofton, K. M., Seegal, R. F., and Schantz, S. L. (2001). Spatial reversal learning in Aroclor 1254-exposed rats: Sex-specific deficits in associative ability and inhibitory control. Toxicol. Appl. Pharmacol. 174, 188–198.
Wong, P. W., Brackney, W. R., and Pessah, I. N. (1997a). Ortho-substituted polychlorinated biphenyls alter microsomal calcium transport by direct interaction with ryanodine receptors of mammalian brain. J. Biol. Chem. 272, 15145–15153.
Wong, P. W., Joy, R. M., Albertson, T. E., Schantz, S. L., and Pessah, I. N. (1997b). Ortho-substituted 2,2',3,5',6-pentachlorobiphenyl (PCB 95) alters rat hippocampal ryanodine receptors and neuroplasticity in vitro: Evidence for altered hippocampal function. Neurotoxicology 18, 443–456.
Wong, P. W., and Pessah, I. N. (1996). Ortho-substituted polychlorinated biphenyls alter calcium regulation by a ryanodine receptor-mediated mechanism: Structural specificity toward skeletal- and cardiac-type microsomal calcium release channels. Mol. Pharmacol. 49, 740–751.
Zakhary, R., Gaine, S. P., Dinerman, J. L., Ruat, M., Flavahan, N. A., and Snyder, S. H. (1996). Heme oxygenase 2: Endothelial and neuronal localization and role in endothelium-dependent relaxation. Proc. Natl. Acad. Sci. U.S.A. 93, 795–798.
Zhang, X., Shan, P., Jiang, D., Noble, P. W., Abraham, N. G., Kappas, A., and Lee, P. J. (2004). Small interfering RNA targeting heme oxygenase-1 enhances ischemia-reperfusion-induced lung apoptosis. J. Biol. Chem. 279, 10677–10684.(Donna W. Lee, Robert M. Gelein and Lisa )
ABSTRACT
Dopaminergic (DAergic) systems have been identified as putative targets for polycholorinated biphenyl (PCB) actions. However, the precise mechanisms leading to neurotoxicity are unresolved. Reactive oxygen species (ROS) were recently shown to mediate injury in DAergic MN9D cells following exposure to Aroclor 1254 (A1254), a commercial PCB mixture. The oxidative stress response in DAergic cells included a persistent expression of heme oxygenase-1 (HO-1). This study tested the hypothesis that a sustained PCB-induced HO-1 response leads to abnormally high Fe levels, which generates ROS production and mediates death in the MN9D DAergic cell model. Accordingly, results indicated that A1254 augmented intracellular Fe levels in MN9D cells after 24 h. Fe chelation by desferoxamine or pharmacologic inhibition of HO activity with tin-protoporphyrin reduced Fe accumulation, ROS production, and cytotoxicity following A1254 exposure. HO-1 over-expression predisposed MN9D DAergic cells to enhanced ROS production and cell death in response to PCBs. Conversely, antisense inhibition of HO-1 expression prevented PCB-induced ROS production and cell death. These observations suggest that enhanced HO-1 catalytic activity and subsequent liberation of Fe participate in neurotoxic DAergic cell injury caused by A1254 exposure in vitro.
Key Words: PCB; neurotoxicity; reactive oxygen species (ROS); iron; neurodegeneration.
INTRODUCTION
Although polychlorinated biphenyls (PCBs) are no longer produced in the United States, exposure remains a profound health hazard because these toxicants are resistant to biodegradation and remain widely distributed in the environment (Giesy and Kannan, 1998; Safe, 1993). PCB exposure has been associated with cognitive and motor neurobehavioral abnormalities in both humans (Jacobson and Jacobson, 2003; Schantz et al., 2003) and experimental animals (Branchi et al., 2005; Eriksson et al., 1991; Widholm et al., 2001). Dopaminergic (DAergic) systems appear to be cellular and neurochemical targets for PCBs (Bemis and Seegal, 2004; Mariussen et al., 2001; Mariussen and Fonnum, 2001; Seegal, 2003; Seegal et al., 2002), but the mechanisms of neurotoxic action are not well defined. In vitro studies have indicated that PCBs disrupt intracellular calcium homeostasis (Bemis and Seegal, 2000; Inglefield et al., 2001, 2002) and interfere with ryanadine receptor function (Howard et al., 2003; Wong et al., 1997a,b). Transient increases in reactive oxygen species (ROS) have also been previously observed following PCB exposure in isolated rat brain synaptosomes (Voie and Fonnum, 2000) and in cerebellar granule neurons in vitro (Mariussen et al., 2002). However, a recent study from our laboratory demonstrated that the commercial PCB mixture Aroclor 1254 (A1254) generated a sustained elevation in ROS production, which preceded DAergic cell death (Lee and Opanashuk, 2004).
Whereas oxidative stress might be involved in A1254 induced DAergic cell injury, the precise intracellular sources and mechanisms of ROS generation in mediating PCB neurotoxicity have not been identified. DAergic neurons are particularly susceptible to oxidative damage due to the high levels of inherent ROS that are produced during dopamine (DA) breakdown by monoamine oxidases or via auto-oxidation to quinones (Cohen et al., 1997; Graham, 1978; Maker et al., 1981). Additionally, H2O2 is generated during DA synthesis by tyrosine hydroxylase (Haavik et al., 1997). Perhaps more importantly, iron bound to neuromelanin within DAergic neurons can subsequently react with metabolically liberated H2O2 through the Fenton reaction to produce extremely toxic hydroxyl radicals (OH). If not properly buffered, excess OH can stimulate lipid peroxidation, which will eventually lead to macromolecular injury and neuronal death. In a previous study, we demonstrated that the PCB-induced generation of ROS in MN9D DAergic cells was accompanied by a depletion of intracellular glutathione (GSH) (Lee and Opanashuk, 2004). This characteristic index of oxidative stress was accompanied by an antioxidant defense response that included a robust increase in heme-oxygenase–1 (HO-1) protein expression.
The heme oxygenase (HO) pathway is considered to be a fundamental defense system for neurons undergoing oxidative challenges (Chen et al., 2000; Dore, 2002). However, recent evidence suggests that HO-1 activity could reach levels that are detrimental to cells (Dennery, 2000; Ryter and Tyrrell, 2000). The microsomal HO-1 enzyme isoform is a 32kDa stress response protein that is readily inducible by a variety of stimuli, including DA, hydrogen peroxide, ultraviolet A radiation, and tumor necrosis factor- (Keyse and Tyrrell, 1989; Mehindate et al., 2001; Schipper et al., 1999). HO-2 is the constitutively expressed isoform and represents the main source of HO activity in the brain (Maines, 2000). HO enzyme activity plays a principal cellular role in catalyzing the breakdown of free heme groups into several byproducts that include carbon monoxide (CO), biliverdin (BV), and ferrous iron (Fe2+) (Ryter and Tyrrell, 2000). Much attention has been focused on HO-1 induction as an adaptive response to oxidative stress since biliverdin, which is readily metabolized into bilirubin, has antioxidant properties (Chen et al., 2000; Fukuda et al., 1996; Maines et al., 1998). Paradoxically, recent studies have reported that HO-1 induction did not protect against hyperoxia in fibroblasts (Suttner and Dennery, 1999) or hemin mediated death in catecholaminergic PC12 cells (Leon et al., 2003). In primary cortical neuronal cultures, HO-2 gene deletion was shown to reduce oxidative stress and decrease vulnerability in response to hemin (Regan et al., 2004) or hemoglobin (Rogers et al., 2003). Therefore, in contrast to its traditional role as a cellular defense mechanism, the induction of HO-1 could be toxic under certain conditions.
As indicated above, Fe is liberated following heme breakdown by HO enzymes. If the excess labile Fe resulting from elevated HO catalytic activity following PCB exposure is not buffered, it may produce neuronal injury by participating in oxidative processes to form highly reactive OH (Ryter and Tyrrell, 2000) or neurotoxic DA-derived semiquinone intermediates (Hermida-Ameijeiras et al., 2004; Pezzella et al., 1997). Indeed, PCB exposure was recently associated with elevated Fe levels in hepatocytes in adult rats (Whysner and Wang, 2001). Considering that HO-1 was profoundly elevated in DAergic cells following A1254 treatment (Lee and Opanashuk, 2004), a potential intracellular source of Fe could arise from the HO-mediated breakdown of heme moieties (Dennery, 2000). Therefore, this study tested the hypothesis that a sustained PCB-induced HO-1 response leads to abnormally high Fe levels, which generates ROS production and mediates death in the MN9D DAergic cell model. The results of this study are consistent with the notion that both enhanced HO-1 catalytic activity during heme breakdown and the subsequent liberation of Fe participate in neurotoxic DAergic cell injury caused by in vitro A1254 exposure.
MATERIALS AND METHODS
Reagents.
Chemicals were purchased from Sigma-Aldrich (St. Louis, MO), unless otherwise stated. 1-Methyl-4-phenylpyridinium (MPP+) was purchased from RBI (Natick, MA), 2',7'-dichlorofluorescein-diacetate (DCF-DA) was obtained from Calbiochem (San Diego, CA). Aroclor 1254 (Lot 124-191; 99% purity) was purchased from Accustandard Inc (New Haven, CT). Cell culture media was acquired from Gibco/BRL (Grand Island, NY).
Dopaminergic MN9D cell line and Aroclor 1254 treatments.
MN9D cells (a generous gift from Drs. Alfred Heller and Lisa Won, Univ. Chicago) were maintained in Dulbecco's modified Eagle's medium containing 15% fetal bovine serum (FBS), and 10,000 U/ml penicillin/streptomycin, as previously described (Heller et al., 1996). Cells were grown to 90% confluency on 100 mm2 dishes before plating on poly-D-lysine (100 μg/ml) coated 12-well plates for experiments (1.5 x 105 cells/well).
Arolor 1254 (A1254) was resuspended in 100% DMSO. The experimental dilutions were prepared as 1000x stock solutions in 100% DMSO. The experimental working concentrations were prepared from the 1000x stocks in culture media and placed in a sonifying water bath the night before the exposures were initiated. MN9D cells were exposed to 0–20 ppm A1254 for 24–48 h. This experimental paradigm was previously shown to cause concentration and time dependent increases in DAergic cell death after 48 h (Lee and Opanashuk, 2004). Cell death was shown to be preceded by and dependent upon oxidative stress. In some experiments, cultures were co-treated with 100 μM desferoxamine (DES), an Fe chelator, or pretreated for 3 h with 5 μM tin- protoporphyrin (SnPPIX), a pharmacologic HO-1 inhibitor. Reactive oxygen species (ROS) production was analyzed after 24 h. Cytotoxicity was assessed 48 h following exposure.
Recombinant adenovirus generation and treatment.
Adenovirus carrying the green fluorescent protein, GFP (Ad.GFP), HO-1 cDNA (Ad.HO-1), and the HO-1 cDNA oriented in the anti-sense direction (Ad.antiHO-1) were kindly provided by Dr. Heiling Lu (CRC, Mass. Gen. Hospital, Harvard Medical School, MA). Dr. Lu and coworkers have previously described the generation of these replication-deficient adenoviruses (He et al., 1998). Multiplicity of infection (MOI) of 10–200 (number of virus particles per cell) was based on the concentrations of the stock solutions that were provided by Dr. Liu. MN9D cells in DMEM containing 15% FBS were transfected with 0–200 MOI of adenovirus carrying the Ad.GFP, Ad.HO-1, and Ad.antiHO-1 HO-1 cDNAs for 24 h prior to treatment with A1254 (He et al., 1998; Liu et al., 2003). The Ad.GFP vector served as a control for the transfection procedure. LDH release was measured to evaluate the toxicity of adenoviral infections. The GFP adenovirus control construct (that lacked either HO-1 or anti-sense HO-1) was used to rule out adverse effects associated with the transfection procedure. GFP fluorescence was monitored by microscopy for all transfections. Transfection efficiency as measured by GFP fluorescence intensity was observed to be similar following infection with all three adenoviral constructs (i.e., 60% for 200 MOI). The respective increases and decreases in HO-1 protein following transfection with either HO-1 or antisense HO-1 were quantified by immunoblot analyses. For MOI 100–200, heme oxygenase levels were altered by 50–80% (data not shown).
Immunoblot analysis.
MN9D cells were harvested in phosphate buffered saline (PBS) containing 0.3% Triton, protease inhibitors (100 mM AEBSF, 0.08 mM aprotinin, 2 mM leupeptin, 1.5 mM pepstatin A, 4 mM bestatin, 1.4 mM E-64; Sigma), 5 mM EDTA, and 2 mM phenylmethyl sulfonyl fluoride. Protein concentrations were determined using a MicroBCA assay from Pierce (Rockford, IL). Twenty-five μg of protein were separated on 10% acrylamide gels before transfer of proteins onto a polyvinylidene difluoride (PVDF) membrane. Membranes were blocked with 5% powdered milk solution in 0.3% Triton-PBS solution before incubation with 1:2000 HO-1 (StressGen, Victoria, B.C.). Actin (Sigma, St. Louis, MO; 1:1500) was used as a loading control. Protein bands were detected by a chemiluminescent substrate for horseradish peroxidase (LumiGLO, KPL, Gaithersburg, MD) and quantified by Scion Image analysis software (v.4.0.2).
Atomic absorption spectroscopy.
Total intracellular Fe was quantified by atomic absorption spectroscopy following exposure to 20 ppm Aroclor 1254 for 24 h. Parallel cultures were co-treated with 100 μM DES to chelate Fe or pretreated for 3 h with 5 μM SnPPIX to inhibit HO activity. Cellular extracts (1 mg protein) were digested in concentrated nitric acid (HNO3) at 90°C in ultra-clean Teflon vials. Samples were resuspended in 2% HNO3 and analyzed on a Perkin-Elmer AA600 Graphite Furnace atomic absorption spectrometer with a Zeeman background correction. Fe content was measured using a single element (Fe) lamp after a pyrolysis (1300°C) and atomization (2150°C) step. A matrix modifier solution of Mg (NO3)2 was used before readings were taken at a wavelength of 248.3 nm. The estimated limit of detection was 1.542 ng/ml.
Dichlorofluorescein-diacetate (DCF-DA) assay.
ROS levels were analyzed 24 h following exposure to 5–20 ppm A1254 with the DCF-DA assay as previously described (Mariussen et al., 2002; Wang and Joseph, 1999). DCF-DA readily permeates cellular membranes whereby its acetate moieties are cleaved by intracellular esterases. Upon reaction with ROS, such as lipid and hydrogen peroxide radicals, the probe fluoresces (Ex: 485nm and Em: 530nm). After PCB treatment, MN9D cells were incubated with 10 μM DCF-DA dissolved in DMSO for 30 min at 37°C, rinsed with Dulbecco's phosphate buffered saline (DPBS), and transferred into 96-well plates for fluorescence measurements.
Lactate dehydrogenase measurement.
Lactate dehydrogenase (LDH) release was measured to assess cell death (Decker and Lohmann-Matthes, 1988) 48 h following exposure to 5–20 ppm A1254. This measure of cell viability is based on the rate of change in NADH concentration as pyruvate is converted to lactate. Media was collected from individual wells at the end of each experiment and assayed using a spectrophotometer. The extinction coefficient of NADH at 340 nm and pH 7.5 is 0.00622 and was used in the calculation of the total amount of LDH released from the slopes collected during spectrophotometric analysis.
Statistical analysis.
Statistical differences were analyzed by comparing means among different treatment groups using StatView version 5.0 (SAS Institute, Cary, NC). Analysis of variance (ANOVA) was applied with treatment and cotreatment variables (where applicable) to detect differences between these parameters and toxicant exposure. Post-hoc comparisons of mean values were by Fisher's PLSD and statistical differences were considered significant with a p-value of <0.05.
RESULTS
Fe Chelation and HO-1 Inhibition Reduces PCB-Induced Intracellular Fe Accumulation in DAergic MN9D Cells
Treatment with A1254 elicited a significant increase in total intracellular iron (Fe) compared to DMSO controls (Fig. 1.; CON; black bars). In the presence of DES, an Fe chelator, Fe levels declined in both DMSO and A1254 compared to untreated controls (treatment effect p < 0.001). PCB-induced Fe levels diminished by approximately 56% in the presence of DES (Fig. 1; +DES; white bars; p < 0.05, compared to A1254). Furthermore, inhibition of HO-1 following a pretreatment with SnPPIX, significantly reduced Fe accumulation by 50% in response to A1254 (Fig. 1; +SnPPIX; vertically-striped bars p < 0.05). These observations suggest that Fe accumulation in MN9D DAergic cells following PCB treatment could emanate from elevated HO-1 activity.
Fe Chelation and HO-1 Inhibition Prevent PCB-Induced ROS Production and DAergic Cell Death
A1254 produced a concentration dependent elevation in ROS levels in MN9D cells following exposure (Fig. 2A; effect of treatment p < 0.0001). Iron chelation with DES and HO inhibition with SnPPIX significantly diminished the ROS production in response to PCBs (Fig. 2A; p < 0.05). To determine whether excessive Fe accumulation or elevated HO-1 activity could mediate neurotoxicity, LDH release was measured to monitor DAergic cell viability. Both Fe chelation and pharmacologic HO-1 inhibition reduced the PCB-induced cell death observed following A1254 treatment (Fig. 2B). Together these findings suggest that HO-1 activity and Fe accumulation produce ROS and cytotoxicity in MN9D cells following A1254 treatment.
HO-1 Over-expression Enhances PCB-Induced ROS Production and DAergic Cell Death
As an additional strategy to test the hypothesis that HO-1 upregulation contributes to PCB-induced oxidative stress and cell death, HO-1 levels were over-expressed in DAergic cells. In the absence of PCBs, MN9D cells expressed concentration-dependent increases in HO-1 protein levels incrementally with multiplicity of infection (MOI) (Fig. 3A). To determine if oxidative stress contributes to PCB-induced cell death following HO-1-transfection, MN9D cells were treated with A1254 prior to measuring ROS production with the DCF-DA assay. At the higher MOI (100 and 200) levels, ROS production was significantly increased in both the control and DMSO groups (Fig. 3B; effect of MOI, p < 0.0001) in the absence of PCBs. PCB treatment further enhanced free radical generation in MN9D cells that over-expressed HO-1 with a significant effect of treatment (p < 0.0001), MOI (p < 0.0001), and an interaction between treatment and MOI (p < 0.0001). Interestingly, HO-1 over-expression alone at the highest MOI (100–200) increased MN9D cell death by approximately 60–120% (Fig. 3C; p < 0.0001 ). Cell death was more pronounced following treatment with A1254 in MN9D cells that over-expressed HO-1 with a significant effect of treatment (p < 0.0001), MOI concentration (p < 0.0001), and an interaction between treatment and MOI concentration (p < 0.0001) (Fig. 3C). These observations support the notion that the sustained elevation of HO-1 following PCB exposure contributes to oxidative stress and subsequently, DAergic cell death.
Anti-sense Inhibition of HO-1 Expression Diminishes PCB-Induced ROS Production and DAergic Cell Death
To determine whether the inhibition of HO-1 expression would protect against PCB-induced ROS generation and toxicity, MN9D cells were transfected with adenovirus carrying an HO-1 cDNA construct in the anti-sense orientation. The reduction in HO-1 protein expression following transfection with the anti-sense HO-1 cDNA construct (Ad.HO-1) was confirmed by immunoblot analyses (Fig. 4A). Uninfected, adenovirus-infected control (Ad.GFP), and HO-1 over-expressing MN9D cells treated with PCBs expressed greater levels of HO-1 protein, compared to DMSO controls. Furthermore, HO-1 expression was not observed in DMSO controls or PCB treated MN9D cells that were infected with the antisense Ad.HO-1 construct. PCB treatment in nave cells resulted in elevated ROS production (Fig. 4B). The PCB-induced ROS levels diminished with increasing MOI in cultures that contained the Ad.HO-1 construct (Fig. 4B). Whereas treatment with 20 ppm A1254 significantly increased cell death, 5 ppm A1254 had a minimal effect on viability (Fig. 4C). However, DAergic cells infected with the Ad.HO-1 construct remained viable following A1254 treatment (Fig. 4C), suggesting that inhibition of HO-1 expression and activity is neuroprotective against PCB toxicity. Together these studies verify that ROS production precedes PCB-induced DAergic cell death. Furthermore, the observations from these experiments support the contention that a robust HO-1 induction following PCB treatment can lead to excessive free radical production, which compromises cell survival.
DISCUSSION
Although DAergic systems have previously been identified as targets for PCB actions (Bemis and Seegal, 2004; Mariussen et al., 2001; Mariussen and Fonnum, 2001; Seegal, 2003; Seegal et al., 2002), the precise mechanism leading to neurotoxicity at the cellular level is unresolved. Our recent study demonstrated that ROS are involved in the biochemical events that contribute to DAergic cell injury following in vitro exposure to A1254, a PCB mixture (Lee and Opanashuk, 2004). The most notable index of oxidative stress was the persistent expression of HO-1 protein following PCB treatment. Therefore, the present study tested the hypothesis that elevated HO-1 expression and catalytic activity produces excess Fe, which promotes PCB-induced oxidative stress and neurotoxicity in DAergic cells. Our results indicate that A1254 exposure increased intracellular Fe levels in MN9D cells. Moreover, Fe chelation by DES or pharmacologic inhibition of HO activity with SnPPIX precluded the Fe accumulation, ROS production, and cytotoxicity induced by A1254 exposure. These findings suggest that A1254 activates an oxidative cascade that includes both increased HO-1 catalytic activity and Fe accumulation. Presumably, this response leads to PCB-mediated DAergic cell injury.
The release of redox active Fe (Fe2+) following heme degradation has been postulated to mediate the pro-oxidant consequences of HO activity (Ryter and Tyrrell, 2000; Schipper, 2004). In our study, atomic absorption spectroscopy (AAS) analyses indicated that Fe levels were elevated in DAergic cells following PCB treatment. However, the disadvantage of AAS is that this method is unable to distinguish free from protein-bound Fe. This makes it difficult to determine the origin and oxidation states of the excess intracellular Fe observed following PCB exposure (Fig. 1). The reduction in Fe levels observed in cells treated with DES suggests that a potential source of Fe arises from the extracellular media. Serum-bound Fe could be imported into the cell via the transferrin receptor or the divalent metal transporter (Moos and Morgan, 2004). It should be acknowledged, however, that DES is not a specific Fe chelator, i.e., DES has been shown to sequester zinc and cadmium (Kaur and Andersen, 2002). Therefore, the possibility cannot be excluded that other metals might also be involved in PCB neurotoxicity. Regardless of origin, the reduced Fe levels following pretreatment with SnPPIX to inhibit HO activity suggest that the Fe is liberated from the HO-mediated heme degradation reaction in MN9D cells in response to A1254 treatment (Fig. 1). HO inhibitors and Fe chelators have previously been shown to diminish lipid peroxidation and Fe accumulation in pulmonary artery smooth muscle cells treated with hemoglobin (Lamb et al., 1999), suggesting that the release of redox active Fe from HO-mediated heme breakdown can lead to cellular injury via lipid peroxidation. Indeed, our study demonstrated that both HO-1 inhibition and Fe chelation protected DAergic cells against PCB-mediated ROS production and cell death.
Inhibition of HO by SnPPIX pretreatment suggested that elevated HO-1 enzyme activity following PCB treatment liberates excess Fe that becomes available to participate in oxidative reactions. However, a disadvantage of using SnPPIX and other metalloporphyrins is that they are associated with nonselective effects apart from inhibiting HO activity (Grundemar and Ny, 1997; Maines and Trakshel, 1992). Metalloporphyrins have been shown to interfere with guanylate cyclase (Zakhary et al., 1996), nitric oxide synthase (Meffert et al., 1994), and calcium signaling (Chen et al., 2000). Therefore, an additional strategy used in our study for evaluating the potential consequences of elevated HO-1 protein levels in response to PCBs involved genetically manipulating the cellular expression of the HO-1 enzyme isoform. Over-expression of HO-1 predisposed MN9D DAergic cells to enhanced ROS production and cell death in response to A1254 exposure. Conversely, antisense inhibition of HO-1 expression prevented PCB-induced ROS production and cell death. These observations further support our hypothesis that HO-1 protein induction and concurrent increases in enzyme activity are sources of Fe and ROS that lead to DAergic cell injury following exposure to A1254 in the MN9D cell model.
It currently remains unclear whether excessive HO-1 activation is the primary intracellular signaling event leading to the mechanism of PCB neurotoxicity in DAergic cells. PCBs could also interfere with Fe efflux systems to augment intracellular accumulation of labile Fe, which can interact with other oxygen radicals to result in cytotoxicity. Unfortunately, the precise mechanisms of Fe efflux from neurons are not well characterized (Moos and Morgan, 2004). Previous studies have suggested that PCBs mediate neurotoxicity by disrupting intracellular Ca+ homeostasis in non-DAergic neurons (Bemis and Seegal, 2000; Inglefield et al., 2001, 2002). Additionally, ortho-substituted PCBs have been shown to interfere with ryanodine receptors (RyR), which regulate the release of Ca+ from intracellular stores (Howard et al., 2003; Wong et al., 1997a,b; Wong and Pessah, 1996). It is possible that abnormal intracellular Ca+ signaling, perhaps through disruption of RyR function (Howard et al., 2003; Wong et al., 1997a,b), could similarly exert PCB neurotoxicity in DAergic cells. As an alternative possibility, DAergic cell injury could arise from ROS produced as a consequence of increased DA turnover following A1254 exposure (Lee and Opanashuk, 2004).
The limitations of the MN9D DAergic cell model are recognized. However, this system is advantageous because the direct effects of PCB exposure on the biochemical pathways that mediate neuronal damage can be studied in a homogeneous population of DAergic cells. It is acknowledged that the A1254 concentrations tested in our study appear to be relatively high compared to expected tissue levels following in vivo exposure (Meacham et al., 2005). The problems with trying to make valid dosimetric comparisons between in vitro and in vitro systems are not restricted to PCB neurotoxicity but represent a common issue encountered with other toxicants, for example, acrylamide (Barber and LoPachin, 2004). This does not necessarily invalidate our in vitro findings but highlights the toxicokinetic differences and cumulative physiochemical events that mediate neurotoxicity in vivo. Additional studies are necessary to determine whether HO-1 and Fe are involved in the mechanism underlying DAergic neurotoxicity in vivo. It will also be important to determine whether HO-1 expression is associated with PCB-induced injury in non-DAergic neurons and other neural cell types. Nevertheless, our data suggest that PCB neurotoxicity involves a complex mechanism that includes ROS production, HO activation, and dysregulation of Fe management processes in DAergic cells following A1254 exposure.
The conclusion that HO-1 mediates PCB neurotoxicity in DAergic cells is contrary to the cytoprotective role typically ascribed. Rapid elevation in HO-1 expression as a response to oxidative stress is thought to be a cellular defense mechanism against macromolecular damage in various experimental models (Chen et al., 2000; Choi et al., 2004; Ferris et al., 1999; Zhang et al., 2004). However, increasing evidence suggests that HO-1 induction (Lamb et al., 1999; Leon et al., 2003; Schipper et al., 1999; Suttner and Dennery, 1999), as well as constitutive HO-2 activity (Regan et al., 2004; Rogers et al., 2003), can also produce cell injury, presumably through excessive ROS generation. DAergic cells are particularly vulnerable to oxidative damage given their inherent redox active microenvironment. Perhaps the most intriguing observation in our study was that HO-1 over-expression in MN9D control cells (in the absence of PCBs) was associated with elevated ROS levels and cytotoxicity (Fig. 3). These findings further strengthen the notion that excessive HO-1 activity, even in the absence of exogenous stimulation, can generate ROS that interfere with DAergic cell survival and function. Our study indicates that A1254, a commercial PCB mixture, further increases this deleterious oxidative cascade in MN9D cells. It is proposed that PCB neurotoxicity is dependent upon the duration and intensity of HO protein induction and enzyme activity. Taken together, our data suggest that there is a critical threshold of HO-1 activation following PCB exposure that is detrimental to DAergic cells.
The increased expression of HO-1 and elevated Fe levels resulting from A1254 exposure represents a potential mechanism by which PCBs exert their neurotoxicity in DAergic cells. It has been hypothesized that environmental factors, such as pesticide or metal exposure, contribute to DAergic neurodegeneration in Parkinson's disease (PD) (Di Monte et al., 2002; Tanner et al., 1999). A potential link between PCB exposure and neurodegeneration was suggested in a previous study that reported elevated PCBs and other organochlorine compounds in the brains of PD patients (Corrigan et al., 1998). Interestingly, both HO-1 and Fe have been connected to neurodegenerative disorders such as PD. HO-1 protein and Fe levels were shown to be elevated in the substantia nigra of PD brains (Schipper et al., 1998; Sofic et al., 1988). It is conceivable that both excess HO activation and Fe accumulation generate ROS that produce DAergic cell loss in PD. However, a causal relationship between HO-1, Fe, and DAergic neurodegeneration remains to be established. Understanding the role of HO and Fe in PCB induced DAergic neuronal injury could have global implications for the pathophysiology of Parkinsonism. It is reasonable to speculate that PCB accumulation results in the persistent elevation of HO-1 and Fe to produce abnormally high ROS levels, which could lead to DAergic cell loss. Therefore, exposure to PCBs and other environmental contaminants that produce oxidative stress, HO-1 induction, and disrupt Fe homeostasis, should be considered as potential risk factors for development of neurodegenerative disorders, including PD.
ACKNOWLEDGMENTS
The authors thank Drs. Alfred Heller and Lisa Won for providing MN9D cells and advice related to these studies. We express our gratitude to Dr. Heiling Lu for generously providing the HO-1 adenoviral constructs. We also appreciate the editorial assistance provided by Dr. Randy LoPachin and Sarah Notter. This research was supported by the National Institute of Health (NIH) grants ES00375, ES01247, and T32 ES07026.
REFERENCES
Barber, D. S., and LoPachin, R. M. (2004). Proteomic analysis of acrylamide-protein adduct formation in rat brain synaptosomes. Toxicol. Appl. Pharmacol. 201, 120–136.
Bemis, J. C., and Seegal, R. F. (2000). Polychlorinated biphenyls and methylmercury alter intracellular calcium concentrations in rat cerebellar granule cells. Neurotoxicology 21, 1123–1134.
Bemis, J. C., and Seegal, R. F. (2004). PCB-induced inhibition of the vesicular monoamine transporter predicts reductions in synaptosomal dopamine content. Toxicol. Sci. 80, 288–295.
Branchi, I., Capone, F., Vitalone, A., Madia, F., Santucci, D., Alleva, E., and Costa, L. G. (2005). Early developmental exposure to BDE 99 or Aroclor 1254 affects neurobehavioural profile: Interference from the administration route. Neurotoxicology 26, 183–192.
Chen, K., Gunter, K., and Maines, M. D. (2000). Neurons overexpressing heme oxygenase-1 resist oxidative stress-mediated cell death. J. Neurochem. 75, 304–313.
Choi, B. M., Pae, H. O., Jeong, Y. R., Oh, G. S., Jun, C. D., Kim, B. R., Kim, Y. M., and Chung, H. T. (2004). Overexpression of heme oxygenase (HO)-1 renders Jurkat T cells resistant to fas-mediated apoptosis: Involvement of iron released by HO-1. Free Radic Biol. Med. 36, 858–871.
Cohen, G., Farooqui, R., and Kesler, N. (1997). Parkinson disease: A new link between monoamine oxidase and mitochondrial electron flow. Proc. Natl. Acad. Sci. U.S.A. 94, 4890–4894.
Corrigan, F. M., Murray, L., Wyatt, C. L., and Shore, R. F. (1998). Diorthosubstituted polychlorinated biphenyls in caudate nucleus in Parkinson's disease. Exp. Neurol. 150, 339–342.
Decker, T., and Lohmann-Matthes, M. L. (1988). A quick and simple method for the quantitation of lactate dehydrogenase release in measurements of cellular cytotoxicity and tumor necrosis factor (TNF) activity. J. Immunol. Methods 115, 61–69.
Dennery, P. A. (2000). Regulation and role of heme oxygenase in oxidative injury. Curr. Topics Cell. Regul. 36, 181–199.
Di Monte, D. A., Lavasani, M., and Manning-Bog, A. B. (2002). Environmental factors in Parkinson's disease. Neurotoxicology 23, 487–502.
Dore, S. (2002). Decreased activity of the antioxidant heme oxygenase enzyme: Implications in ischemia and in Alzheimer's disease. Free Radical Biol. Med. 32, 1276–1282.
Eriksson, P., Lundkvist, U., and Fredriksson, A. (1991). Neonatal exposure to 3,3',4,4'-tetrachlorobiphenyl: Changes in spontaneous behaviour and cholinergic muscarinic receptors in the adult mouse. Toxicology 69, 27–34.
Ferris, C. D., Jaffrey, S. R., Sawa, A., Takahashi, M., Brady, S. D., Barrow, R. K., Tysoe, S. A., Wolosker, H., Baranano, D. E., Dore, S., Poss, K. D., and Snyder, S. H. (1999). Haem oxygenase-1 prevents cell death by regulating cellular iron. Nature Cell Biology 1, 152–157.
Fukuda, K., Richmon, J. D., Sato, M., Sharp, F. R., Panter, S. S., and Noble, L. J. (1996). Induction of heme oxygenase-1 (HO-1) in glia after traumatic brain injury. Brain Res. 736, 68–75.
Giesy, J. P., and Kannan, K. (1998). Dioxin-like and non-dioxin-like toxic effects of polychlorinated biphenyls (PCBs): Implications for risk assessment. Crit. Rev. Toxicol. 28, 511–569.
Graham, D. G. (1978). Oxidative pathways for catecholamines in the genesis of neuromelanin and cytotoxic quinones. Mol. Pharmacol. 14, 633–643.
Grundemar, L., and Ny, L. (1997). Pitfalls using metalloporphyrins in carbon monoxide research. Trends Pharmacol. Sci. 18, 193–195.
Haavik, J., Almas, B., and Flatmark, T. (1997). Generation of reactive oxygen species by tyrosine hydroxylase: A possible contribution to the degeneration of dopaminergic neurons J. Neurochem. 68, 328–332.
He, T. C., Zhou, S., da Costa, L. T., Yu, J., Kinzler, K. W., and Vogelstein, B. (1998). A simplified system for generating recombinant adenoviruses. Proc. Natl. Acad. Sci. U.S.A. 95, 2509–2514.
Heller, A., Price, S., and Won, L. (1996). Glial-derived neurotrophic factor (GDNF) induced morphological differentiation of an immortalized monoclonal hybrid dopaminergic cell line of mesencephalic neuronal origin. Brain Res. 725, 132–136.
Hermida-Ameijeiras, A., Mendez-Alvarez, E., Sanchez-Iglesias, S., Sanmartin-Suarez, C., and Soto-Otero, R. (2004). Autoxidation and MAO-mediated metabolism of dopamine as a potential cause of oxidative stress: Role of ferrous and ferric ions. Neurochem. Int. 45, 103–116.
Howard, A. S., Fitzpatrick, R., Pessah, I., Kostyniak, P., and Lein, P. J. (2003). Polychlorinated biphenyls induce caspase-dependent cell death in cultured embryonic rat hippocampal but not cortical neurons via activation of the ryanodine receptor. Toxicol. Appl. Pharmacol. 190, 72–86.
Inglefield, J. R., Mundy, W. R., Meacham, C. A., and Shafer, T. J. (2002). Identification of calcium-dependent and -independent signaling pathways involved in polychlorinated biphenyl-induced cyclic AMP-responsive element-binding protein phosphorylation in developing cortical neurons. Neuroscience 115, 559–573.
Inglefield, J. R., Mundy, W. R., and Shafer, T. J. (2001). Inositol 1,4,5-triphosphate receptor-sensitive Ca(2+) release, store-operated Ca(2+) entry, and cAMP responsive element binding protein phosphorylation in developing cortical cells following exposure to polychlorinated biphenyls. J. Pharmacol. Exp. Ther. 297, 762–773.
Jacobson, J. L., and Jacobson, S. W. (2003). Prenatal exposure to polychlorinated biphenyls and attention at school age. J. Pediatr. 143, 780–788.
Kaur, D., and Andersen, J. K. (2002). Ironing out Parkinson's disease: Is therapeutic treatment with iron chelators a real possibility Aging Cell 1, 17–21.
Keyse, S. M., and Tyrrell, R. M. (1989). Heme oxygenase is the major 32-kDa stress protein induced in human skin fibroblasts by UVA radiation, hydrogen peroxide, and sodium arsenite. Proc. Natl. Acad. Sci. U.S.A. 86, 99–103.
Lamb, N. J., Quinlan, G. J., Mumby, S., Evans, T. W., and Gutteridge, J. M. (1999). Haem oxygenase shows pro-oxidant activity in microsomal and cellular systems: Implications for the release of low-molecular-mass iron. Biochem. J. 344, 153–158.
Lee, D. W., and Opanashuk, L. A. (2004). Polychlorinated biphenyl mixture aroclor 1254-induced oxidative stress plays a role in dopaminergic cell injury. Neurotoxicology 25, 925–939.
Leon, A., Le Foll, I., Charriault-Marlangue, C., Leprince, J., Vaudry, H., Gabriel, C., and Duval, D. (2003). Level of haem oxygenase does not obligatorily reflect the sensitivity of PC12 cells to an oxidative shock induced by glutathione depletion. J. Neurochem. 84, 459–470.
Liu, H., Nowak, R., Chao, W., and Bloch, K. D. (2003). Nerve growth factor induces anti-apoptotic heme oxygenase-1 in rat pheochromocytoma PC12 cells. J. Neurochem. 86, 1553–1563.
Maines, M. D. (2000). The heme oxygenase system and its functions in the brain. Cell. Mol. Biol. 46, 573–585.
Maines, M. D., Polevoda, B., Coban, T., Johnson, K., Stoliar, S., Huang, T. J., Panahian, N., Cory-Slechta, D. A., and McCoubrey, W. K., Jr. (1998). Neuronal overexpression of heme oxygenase-1 correlates with an attenuated exploratory behavior and causes an increase in neuronal NADPH diaphorase staining. J. Neurochem. 70, 2057–2069.
Maines, M. D., and Trakshel, G. M. (1992). Differential regulation of heme oxygenase isozymes by Sn- and Zn-protoporphyrins: Possible relevance to suppression of hyperbilirubinemia. Biochimica et Biophysica Acta 1131, 166–174.
Maker, H. S., Weiss, C., Silides, D. J., and Cohen, G. (1981). Coupling of dopamine oxidation (monoamine oxidase activity) to glutathione oxidation via the generation of hydrogen peroxide in rat brain homogenates. J. Neurochem. 36, 589–593.
Mariussen, E., Andersson, P. L., Tysklind, M., and Fonnum, F. (2001). Effect of polychlorinated biphenyls on the uptake of dopamine into rat brain synaptic vesicles: A structure-activity study. Toxicol. Appl. Pharmacol. 175, 176–183.
Mariussen, E., and Fonnum, F. (2001). The effect of polychlorinated biphenyls on the high affinity uptake of the neurotransmitters, dopamine, serotonin, glutamate and GABA, into rat brain synaptosomes. Environ. Res. 86, 2–11.
Mariussen, E., Myhre, O., Reistad, T., and Fonnum, F. (2002). The polychlorinated biphenyl mixture aroclor 1254 induces death of rat cerebellar granule cells: The involvement of the N-methyl-D-aspartate receptor and reactive oxygen species. Toxicol. Appl. Pharmacol. 179, 137–144.
Meacham, C. A., Freudenrich, T. M., Anderson, W. L., Sui, L., Lyons-Darden, T., Barone, S., Jr., Gilbert, M. E., Mundy, W. R., and Shafer, T. J. (2005). Accumulation of methylmercury or polychlorinated biphenyls in in vitro models of rat neuronal tissue. Toxicol. Appl. Pharmacol. 205, 177–187.
Meffert, M. K., Haley, J. E., Schuman, E. M., Schulman, H., and Madison, D. V. (1994). Inhibition of hippocampal heme oxygenase, nitric oxide synthase, and long-term potentiation by metalloporphyrins. Neuron 13, 1225–1233.
Mehindate, K., Sahlas, D. J., Frankel, D., Mawal, Y., Liberman, A., Corcos, J., Dion, S., and Schipper, H. M. (2001). Proinflammatory cytokines promote glial heme oxygenase-1 expression and mitochondrial iron deposition: Implications for multiple sclerosis. J. Neurochem. 77, 1386–1395.
Moos, T., and Morgan, E. H. (2004). The metabolism of neuronal iron and its pathogenic role in neurological disease: Review. Ann. N.Y. Acad. Sci. 1012, 14–26.
Pezzella, A., d'Ischia, M., Napolitano, A., Misuraca, G., and Prota, G. (1997). Iron-mediated generation of the neurotoxin 6-hydroxydopamine quinone by reaction of fatty acid hydroperoxides with dopamine: A possible contributory mechanism for neuronal degeneration in Parkinson's disease. J. Med. Chem. 40, 2211–2216.
Regan, R. F., Chen, J., and Benvenisti-Zarom, L. (2004). Heme oxygenase-2 gene deletion attenuates oxidative stress in neurons exposed to extracellular hemin. BMC Neurosci. 5, 34.
Rogers, B., Yakopson, V., Teng, Z. P., Guo, Y., and Regan, R. F. (2003). Heme oxygenase-2 knockout neurons are less vulnerable to hemoglobin toxicity. Free Radic. Biol. Med. 35, 872–881.
Ryter, S. W., and Tyrrell, R. M. (2000). The heme synthesis and degradation pathways: Role in oxidant sensitivity. Heme oxygenase has both pro- and antioxidant properties. Free Radical Biol. Med. 28, 289–309.
Safe, S. (1993). Toxicology, structure-function relationship, and human and environmental health impacts of polychlorinated biphenyls: Progress and problems. Environ. Health Perspect. 100, 259–268.
Schantz, S. L., Widholm, J. J., and Rice, D. C. (2003). Effects of PCB exposure on neuropsychological function in children. Environ. Health Perspect. 111, 357–576.
Schipper, H. M. (2004). Heme oxygenase-1: Transducer of pathological brain iron sequestration under oxidative stress. Ann. N.Y. Acad. Sci. 1012, 84–93.
Schipper, H. M., Bernier, L., Mehindate, K., and Frankel, D. (1999). Mitochondrial iron sequestration in dopamine-challenged astroglia: Role of heme oxygenase-1 and the permeability transition pore. J. Neurochem. 72, 1802–1811.
Schipper, H. M., Liberman, A., and Stopa, E. G. (1998). Neural heme oxygenase-1 expression in idiopathic Parkinson's disease. Exp. Neurol. 150, 60–68.
Seegal, R. F. (2003). PCBs and dopamine function: Neurological effects of polychlorinated biphenyls: does occupational exposure alter dopamine-mediated function Ann. N.Y. Acad. Sci. 991, 322–325.
Seegal, R. F., Okoniewski, R. J., Brosch, K. O., and Bemis, J. C. (2002). Polychlorinated biphenyls alter extraneuronal but not tissue dopamine concentrations in adult rat striatum: An in vivo microdialysis study. Environ. Health Perspect. 110, 1113–1117.
Sofic, E., Riederer, P., Heinsen, H., Beckmann, H., Reynolds, G. P., Hebenstreit, G., and Youdim, M. B. (1988). Increased iron (III) and total iron content in post mortem substantia nigra of Parkinsonian brain. J. Neural Transmission - General Section 74, 199–205.
Suttner, D. M., and Dennery, P. A. (1999). Reversal of HO-1 related cytoprotection with increased expression is due to reactive iron. FASEB J. 13, 1800–1809.
Tanner, C. M., Ottman, R., Goldman, S. M., Ellenberg, J., Chan, P., Mayeux, R., and Langston, J. W. (1999). Parkinson disease in twins: An etiologic study. JAMA 281, 341–346.
Voie, O. A., and Fonnum, F. (2000). Effect of polychlorinated biphenyls on production of reactive oxygen species (ROS) in rat synaptosomes. Arch. Toxicol. 73, 588–593.
Wang, H., and Joseph, J. A. (1999). Quantifying cellular oxidative stress by dichlorofluorescein assay using microplate reader. Free Radic. Biol. Med. 27, 612–616.
Whysner, J., and Wang, C. X. (2001). Hepatocellular iron accumulation and increased cell proliferation in polychlorinated biphenyl-exposed Sprague-Dawley rats and the development of hepatocarcinogenesis. Toxicol. Sci. 62, 36–45.
Widholm, J. J., Clarkson, G. B., Strupp, B. J., Crofton, K. M., Seegal, R. F., and Schantz, S. L. (2001). Spatial reversal learning in Aroclor 1254-exposed rats: Sex-specific deficits in associative ability and inhibitory control. Toxicol. Appl. Pharmacol. 174, 188–198.
Wong, P. W., Brackney, W. R., and Pessah, I. N. (1997a). Ortho-substituted polychlorinated biphenyls alter microsomal calcium transport by direct interaction with ryanodine receptors of mammalian brain. J. Biol. Chem. 272, 15145–15153.
Wong, P. W., Joy, R. M., Albertson, T. E., Schantz, S. L., and Pessah, I. N. (1997b). Ortho-substituted 2,2',3,5',6-pentachlorobiphenyl (PCB 95) alters rat hippocampal ryanodine receptors and neuroplasticity in vitro: Evidence for altered hippocampal function. Neurotoxicology 18, 443–456.
Wong, P. W., and Pessah, I. N. (1996). Ortho-substituted polychlorinated biphenyls alter calcium regulation by a ryanodine receptor-mediated mechanism: Structural specificity toward skeletal- and cardiac-type microsomal calcium release channels. Mol. Pharmacol. 49, 740–751.
Zakhary, R., Gaine, S. P., Dinerman, J. L., Ruat, M., Flavahan, N. A., and Snyder, S. H. (1996). Heme oxygenase 2: Endothelial and neuronal localization and role in endothelium-dependent relaxation. Proc. Natl. Acad. Sci. U.S.A. 93, 795–798.
Zhang, X., Shan, P., Jiang, D., Noble, P. W., Abraham, N. G., Kappas, A., and Lee, P. J. (2004). Small interfering RNA targeting heme oxygenase-1 enhances ischemia-reperfusion-induced lung apoptosis. J. Biol. Chem. 279, 10677–10684.(Donna W. Lee, Robert M. Gelein and Lisa )