[1] 13w Overexpression of NQO1 protects human SK-N-MC neuroblastoma cells against dopamine-induced cell death ଝ

[1] 248w NAD(P)H quinone oxidoreductase 1 (NQO1) can metabolize dopamine-derived quinones (DAQ) and absence of NQO1 due to the NQO1*2 polymorphism has been suggested to be a risk factor for Parkinson's disease. In order to define whether NQO1 plays a protective role in dopamine toxicity, we have examined the potential role of NQO1 in the SK-N-MC human neuroblastoma cell line. SK-N-MC cells were stably transfected with NQO1 to generate stable clones with NQO1 enzymatic activity of 245 nmol/mg min while vector control and parental cells had NQO1 activities of less than 12 nmol/mg min. Incubation of dopamine for 24 h in both parental and vector control SK-N-MC cells resulted in 85% and 72% cell death as assessed by annexin-V/propidium iodide analysis. In agreement, 88% and 84% of parental and vector control cells, respectively underwent loss of mitochondrial membrane potential (MMP) assessed by tetramethylrhodamine ethyl ester. In contrast, NQO1-transfected cells were resistant to dopamine toxicity and both cell death and loss of MMP were markedly abrogated in NQO1-transfected SK-N-MC cells. When dopamine was added to medium, oxygen uptake could be detected indicating autoxidation with concomitant formation of oxygen radicals and quinones. However, dopamine-induced cell death was not affected by the inclusion of either superoxide dismutase or catalase suggesting that superoxide and hydrogen peroxide were not involved in toxicity. Quinones formed in medium may exert toxicity extracellularly or intracellularly but the protective role of NQO1 argues for an intracellular mechanism. In summary, transfection of SK-N-MC cells with NQO1 protects against dopamine-induced toxicity.

[1] 123w Parkinson's disease (PD) is a progressive neuronal disease, which is associated with selective loss of dopaminergic neurons in the nigrostriatal pathway of the brain. Among the symptoms shown in PD include autonomic disturbances and motor abnormalities such as tremor and rigidity. Recent studies have implicated the role of dopamine-induced oxidative stress, proteasomal inhibition and mitochondrial dysfunction in destruction of dopaminergic neurons in the substantia nigra in PD (Kosel et al., 1999;McNaught et al., 2003;Munch et al., 0378-4274/$ -see front matter © 2006Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.toxlet.2006.07.340 1998). The toxicity of DA has been attributed to its electrophilic reactive metabolites such as quinones or generation of reactive oxygen species (ROS), both of which can lead to cellular damage and toxicity (Smythies, 1999).

[2] 331w Recently, marked increases in the astroglial and neuronal expression of NAD(P)H quinone oxidoreductase 1 (NQO1) in Parkinsonian substantia nigra have been observed. NQO1 (DT-diaphorase), is an inducible, primarily cytosolic enzyme which can reduce quinones via a two-electron reduction process to the corresponding hydroquinone. The null NQO1*2 genetic polymorphism has been reported to be over represented in PD patients (Jiang et al., 2004;Shao et al., 2001) but conflicting evidence has also been published (Okada et al., 2005). The reasons for the differences are unclear but it cannot be discounted that a null NQO1 genotype is associated with increased PD in some ethnic groups. It is known that DA can be autoxidized or metabolized by tyrosinase/MAO to initiate a complex series of metabolic and chemical transformations eventually resulting in the biosynthesis of melanin. Initial products in this pathway are the uncyclized dopamine o-quinone and at physiological pH, the cyclized quinone, aminochrome. Dopaminederived quinones (DAQ) are reactive electrophiles and can also participate in the generation of ROS. NQO1 has been shown to catalyze the two-electron reduction of aminochrome to its leuko or hydroquinone form (Segura-Aguilar and Lind, 1989). Since the hydroquinone form of aminochrome is redox unstable and can give rise to oxygen radical formation, it is an interesting question whether NQO1 protects against or potentiates dopamineinduced toxicity. In view of the elevated expression of NQO1 in Parkinsonian substantia nigra and its implication in the metabolism of aminochrome, we have investigated the toxic effects of DA in human SK-N-MC neuroblastoma cells, which have been engineered to overexpress human NQO1. Although SK-N-MC cells express the D1 dopamine receptor and downstream signaling pathways they are not dopaminergic and do not express plasma membrane dopamine transporters. They have been used previously as a model for postsynaptic striatal neurons (Chen et al., 2004) and are used in this work purely as a model neuroblastoma system to examine the effects of NQO1 on DA toxicity. Our study demonstrates that overexpression of NQO1 in SK-N-MC cells protects against dopamine-induced toxicity.

[1] 19w To investigate the effects of NQO1 on dopamineinduced cell death, SK-N-MC neuroblastoma cells were used as a model system.

[2] 79w Cell death induced by dopamine was initially assessed by morphological changes using Hoechst 33342 and PI dyes as described in Section 2. Control cells essentially showed homogenous staining and no nuclear condensation was observed (Fig. 1.). In contrast, nuclear condensation was significantly increased in DA treated SK-N-MC cells showing both early and late changes of apoptosis. The nuclear changes of apoptosis were further confirmed using staurosporine (STS) as a positive control where most of the cells exhibited apoptotic morphology.

[3] 220w DA (500 M, 24 h) induced both apoptosis and necrosis. The total cell death induced by DA 500 M for 24 h as reflected by both apoptosis and necrosis in wt SK-N-MC cells was 85 ± 4% with apoptosis and necrosis reflecting 54 ± 2% and 31 ± 1.5%, respectively. Lower doses of dopamine exhibited considerably less toxicity with 250 M DA exhibiting total apoptosis/necrosis of 24% at 24 h. It has previously been reported that the DA D1 receptor antagonist SCH 23390 abrogated the toxicity of DA 50 M to SK-N-MC cells (Moussa et al., 2006). We did not observe any inhibition of DAinduced apoptosis in our studies using SCH 23390. The reasons for this difference and the different concentrations of DA used to effect toxicity are unclear but may reflect the endpoints used. Our work used assessment of apoptosis and necrosis by flow cytometry whereas previous work used either trypan blue exclusion (Chen et al., 2004) or the MTT assay (Moussa et al., 2006), a mitochondrial dye which measures growth inhibition. The IC 50 determined previously for DA in SK-N-MC cells using trypan blue exclusion was approximately 200 M employing a 16 h treatment, which is not inconsistent with the extent of apoptosis and necrosis observed in this study after treatment with 500 M DA for 24 h.

[4] 123w To address the role of NQO1 in PD, we transfected SK-N-MC cells with human wt hNQO1 cDNA using neomycin as the selection marker. As shown in Fig. 2, the wt SK-N-MC cells had only a low level of NQO1 protein and the NQO1 enzymatic activity of these cells was 6 nmol/mg min. In contrast, the NQO1-transfected cells (clone C2) had a marked increase in the NQO1 protein level as detected by western blotting with an NQO1 enzymatic activity of 245 nmol/mg min, approximately 40 times higher compared to the wt cells. The NQO1 activity in vector control cells (clone C9), which received only the empty vector without hNQO1 cDNA had an NQO1 activity (12 nmol/min mg protein) similar to the wt SK-N-MC cells.

[5] 110w DA would autoxidize at physiological pH to form reactive quinones and oxygen radicals (Clement et al., 2002). We measured the increase of oxygen consumption when dopamine (DA 500 M) was added to cell incubation/growth media polarographically and this was calculated to be 2.0 ± 0.4 (n = 3) nmol O 2 /min. We examined the potential role of extracellular oxygen radicals generated from dopamine by adding superoxide dismutase, catalase or combinations of both to the medium prior to addition of dopamine. None of these additions had any effect on DA-induced cell death in wt SK-N-MC cells (Fig. 6) arguing against a role for extracellular oxygen radicals in induction of cytotoxicity.

[1] 127w Our data implicate NQO1 as a protective system against dopamine-induced mitochondrial damage and cell death induced in the SK-N-MC neuroblastoma cell line. NQO1 has been demonstrated to be elevated in the substantia nigra pars compacta of parkinsonian brains and it was suggested that the enzyme represented a protective mechanism against dopamine-induced cell death (van Muiswinkel et al., 2004). However, these authors also acknowledged that NQO1 could represent a bioactivation pathway for dopamine and that the elevated expression of NQO1 could actually contribute to dopamine-induced cell loss. Bioactivation of dopamine or its secondary metabolites via NQO1 has to be considered since the hydroquinone generated via NQO1 reduction of aminochrome is redox unstable and undergoes redox cycling with generation of ROS (Graham, 1978;Graham and Klintworth, 1978;Segura-Aguilar and Lind, 1989).

[2] 229w Our data demonstrates that in the SK-N-MC cellular system, DA-induced cell death was not affected by inclusion of either superoxide dismutase and/or catalase in the cellular medium prior to addition of DA arguing against a role for extracellular oxygen radicals. Since hydrogen peroxide is freely membrane-permeable (Halliwell et al., 2000), this observation also argues against a role for intracellular hydrogen peroxide generation in induction of cytotoxicity. In contrast, NQO1 provided a protective mechanism against dopamineinduced cell death. The mechanism of NQO1 mediated protection against dopamine-induced toxicity can most probably be ascribed to its ability to detoxify potential electrophilic intermediates such as DAQ. This is in agreement with our recent data where we have implicated DAQ such as aminochrome in dopamine-induced proteasomal inhibition (Zafar et al., 2003). Reactive quinones formed from dopamine have previously been implicated in induction of apoptosis in SH-SY5Y neuroblastoma cells via modulation of p53 and bcl-2 levels (Emdadul Haque et al., 2003) and our data are consistent with these observations. NQO1 can also directly scavenge ROS when expressed at high levels in cells (Siegel et al., 2004). This mechanism of NQO1-mediated protection appears less likely, however, since metabolism of aminochrome by either rat (Segura-Aguilar and Lind, 1989) or human NQO1 (Zafar et al., 2006) produces redox unstable hydroquinones leading to more oxygen consumption and oxygen radical generation while at the same time protecting against dopamine-induced toxicity.

[3] 175w It has previously been suggested that extracellular oxidation of dopamine to reactive metabolites including ROS, quinones and semiquinones contributes to dopamine-induced cytotoxicity (Berman and Hastings, 1997;Clement et al., 2002). The ability of DA to elicit effects via activation of the D1 receptor in SK-N-MC cells with subsequent intracellular MAP kinase activation provides an additional mechanism of toxicity (Chen et al., 2004). The mechanisms underlying DA-induced apoptosis in our work remain unclear. We did not observe any protective effects of SOD and/or catalase on DA-induced cell death, suggesting that extracellular superoxide and hydrogen peroxide appear unlikely to play a major role. In addition, under these conditions we did not observe any protective effect of the D1 receptor antagonist SCH23390 on DA-induced cell death. Since we observed autoxidation of DA extracellularly in medium, reactive DAQ will be formed but whether they enter the cell or exert their toxicity extracellularly at the level of the cell membrane are both possibilities. However, the protective effects of transfected NQO1 against DA-induced cell death argue for an intracellular mechanism of protection.

[4] 18w In summary, our data demonstrates that overexpression of NQO1 in SK-N-MC neuroblastoma cells protects against dopamine-induced cell death.

[1] 70w Apoptosis detection kits were purchased from Molecular Probes (Eugene, OR). The kits contained fluorescein isothiocyanate (FITC)-conjugated anti-annexin-V antibody, propidium iodide (PI) and binding buffer (10 mM HEPES-NaOH, pH 7.4, 140 mM NaCl, 5 mM CaCl 2 ). Tetramethylrhodamine ethyl ester (TMRE), a fluorescent dye used for the assessment of mitochondrial membrane potential (MMP) in whole cells was also purchased from Molecular Probes (Eugene, OR). All other reagents were from Sigma.

[2] 94w SK-N-MC human neuroblastoma cells were purchased from ATCC and grown in RPMI 1640 medium supplemented with 10% fetal calf serum, 2 mM l-glutamine, 50 units/ml penicillin and streptomycin (100 g/ml). Cells were grown in a humidified incubator with 5% CO 2 at 37 • C. 5 × 10 5 cells were cultured in a six-well plate and left overnight. Before treatment with DA, the medium was removed and fresh medium was added to the six-well plate. SK-N-MC cells were treated with 500 M dopamine for 24 h before assessment of apoptotic and necrotic endpoints.

[3] 99w Stable transfection of SK-N-MC cells was carried out essentially as described previously (Winski et al., 2001). Briefly, the CMV-driven mammalian expression vector pcDNA3, containing human wild-type (wt) hNQO1 cDNA was transfected into SK-N-MC cells using neomycin as the selection marker. Clones were selected after 7-10 days and were examined for both NQO1 activity and NQO1 immunoreactive protein using monoclonal antibodies to NQO1 generated in our laboratory. The NQO1-transfected cell lines (SK-N-MC clone C2) are routinely screened for NQO1 expression by activity assay and immunoblotting. The vector control is clone C9 where only empty vector was transfected in SK-N-MC cells.

[4] 69w Samples (1 × 10 6 cells) were mixed with Laemmli's loading buffer, boiled for 5 min and subjected to 12% SDS-PAGE at 130 V followed by electroblotting to polyvinylidene fluoride membranes for 2 h at 100 V. Standard immunoblotting procedure was carried out using an anti NQO1 monoclonal antibody, which was developed in our lab as described previously (Winski et al., 2001). ␤-Actin was used as a loading control.

[5] 33w One-way analysis of variance with Tukey post hoc test for multiple comparisons was used for statistical analysis in these studies. Statistical analyses were performed using Prism software (GraphPad Software Inc., San Diego, CA).

[6] 212w To quantify cell death induced by DA, we have utilized two flow cytometric assays which measure the externalization of phosphatidyl serine and loss of MMP using annexin-V/PI and TMRE, respectively. We initially investigated the effects of STS as an apoptotic inducer, which kills cells irrespective of NQO1 status in the wt, C2 and C9 clones of SK-N-MC cells. A typical cytogram is shown in Fig. 2, where cells stained negative for both annexin-V and PI (PS -/PI -) were live cells as shown in the lower left quadrant (R3). Annexin-Vpositive and PI-negative (PS + /PI -) stained cells undergo early stages of apoptosis where the plasma membrane was still intact and excluded PI (lower right quadrant, R4). In late stages of apoptosis, dying cells can no longer exclude PI and the upper right region (R2) displayed both annexin-V-positive and PI-positive (PS + /PI + ). PI- positive and annexin-V-negative (PS -/PI + ) stained cells in the upper left region (R1) were necrotic cells. Our data confirms that STS was capable of inducing cell death in the wt SK-N-MC cells, NQO1-transfected C2 clone and the vector control C9 clone (Fig. 3). The increase of cell death over control in STS treated wt, C2 and C9 clones were 36%, 28% and 37%, respectively.

[7] 79w Exposure to DA resulted in a significant increase in cell death (88%) in wt SK-N-MC cells as assessed by the annexin-V/PI flow cytometric method (Fig. 4). Similarly, DA-treated C9 cells also exhibited a marked increase in cell death (77%). In contrast, NQO1-transfected SK-N-MC cells showed significant resistance to DA-induced toxicity as less than 40% of the cells were affected. DA induced toxicity after subtraction of control values was 70%, 13% and 56% in wt, C2 and C9 clones respectively.

[8] 87w To confirm our observation with the annexin-V/PI method, we also determined the loss of MMP, an upstream event occurring during cell death. As shown in Fig. 5, the wt and C9 clone of SK-N-MC cells treated with DA showed en masse loss of MMP. In agreement with the cell death data (Fig. 4), less than 40% of DA treated C2 clone overexpressing NQO1 showed loss of MMP. Our data demonstrates a good correlation between loss of MMP and cell death using the flow cytometric techniques described above.

[1] 27w NQO1 activity measurements were performed as spectrophotometric assays by measuring the dicoumarolinhibitable reduction of dichlorophenol indophenol (DCPIP) at 600 nm as described previously (Benson et al., 1980).

[2] 98w Both apoptosis and necrosis can be measured by fluorescence microscopy using assessment of morphology and dual staining with Hoechst 33342 and PI as described previously (Moran et al., 1996). Briefly, 0.5-1 × 10 6 cells were collected and stained with Hoechst 33342 (0.5 g/ml) at 37 • C for 10 min. Cells were resuspended in 5 g/ml PI in phosphate buffered saline and visualized via fluorescent microscopy. Cells with condensed and fragmented chromatin and high intensity staining were scored as apoptotic. Necrotic cells were identified as those that failed to exclude PI and do not demonstrate condensed chromatin.

[3] 156w SK-N-MC cells were harvested by trypsinization and any floating cells were then added back to the trypsinized cells, and pelleted by centrifugation (450 g/5 min). Cell pellets were washed once in ice-cold PBS, gently resuspended in 0.1 ml annexin-V binding buffer and incubated with FITC-conjugated annexin-V (2.5 l) and PI (1 l of 100 g/ml stock in PBS) for 12 min at room temperature in the dark. Samples were immediately kept on ice and analyzed on a BD Biosciences FACS Calibur flow cytometer measuring fluorescence emission at 530 nm (FL1) for FITC, and red PI at above 600 nm (FL2). FL-1 and FL-2 were collected on log-scale with voltages of 450 and 458, respectively. Signal overlap was compensated for electronically. Data was acquired and analyzed using Cellquest software (Becton-Dickenson, Mountainview, CA). Acquisition gates were set using the forward and side light scatter of the cells and a minimum of 10,000 events were collected for each sample.

[4] 105w The loss of MMP ( ψ m ) in DA-treated cells was detected using TMRE. Briefly, 1 l of 50 M TMRE (dissolved in ethanol) was added to SK-N-MC cells (5 × 10 5 ) in 1 ml of medium and incubated for 10 min at 37 • C. The samples were immediately analyzed by flow cytometry using FL2. Acquisition gates were set using the forward and side light scatter of the cells and a minimum of 10,000 events were collected for each sample. Results were presented as mean ± S.E.M. and the experiments were repeated at least five times for the flow cytometric assays.

[5] 36w Cell death in SK-N-MC cells was measured by flow cytometric measurement. SK-N-MC cells were pre-incubated with superoxide dismutase (1 mg/ml) and catalase (1 mg/ml) 10 min prior to addition of DA 500 M for 24 h.