Neuromelanin Modulates Heterocyclic Aromatic Amine-Induced Dopaminergic Neurotoxicity

Abstract

Heterocyclic aromatic amines (HAAs) are mutagens and potential human carcinogens. Our group and others have demonstrated that HAAs may also produce selective dopaminergic neurotoxicity, potentially relevant to Parkinson's disease (PD). The goal of this study was to elucidate mechanisms of HAA-induced neurotoxicity through examining a translational biochemical weakness of common PD models. Neuromelanin is a pigmented byproduct of dopamine metabolism that has been debated as being both neurotoxic and neuroprotective in PD. Importantly, neuromelanin is known to bind and potentially release dopaminergic neurotoxicants, including HAAs (eg, β-carbolines such as harmane). Binding of other HAA subclasses (ie, aminoimidazoaazarenes) to neuromelanin has not been investigated, nor has a specific role for neuromelanin in mediating HAA-induced neurotoxicity been examined. Thus, we investigated the role of neuromelanin in modulating HAA-induced neurotoxicity. We characterized melanin from Sepia officinalis and synthetic dopamine melanin, proposed neuromelanin analogs with similar biophysical properties. Using a cell-free assay, we demonstrated strong binding of harmane and 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) to neuromelanin analogs. To increase cellular neuromelanin, we transfected SH-SY5Y neuroblastoma cells with tyrosinase. Relative to controls, tyrosinase-expressing cells exhibited increased neuromelanin levels, cellular HAA uptake, cell toxicity, and oxidative damage. Given that typical cellular and rodent PD models form far lower neuromelanin levels than humans, there is a critical translational weakness in assessing HAA-neurotoxicity. The primary impacts of these results are identification of a potential mechanism by which HAAs accumulate in catecholaminergic neurons and support for the need to conduct neurotoxicity studies in systems forming neuromelanin.

Keywords: Parkinson’s disease; PhIP; harmane; neuromelanin.

Figure 1.

Biophysical analysis of Sepia melanin and dopamine melanin. Two established experimental neuromelanin analogs, melanin from S. officinalis (1 mg/ml) and DAM (1 mg/ml) were qualitatively and quantitatively evaluated. A, Representative images taken of Sepia melanin (left) and DAM suspension (1 mg/ml) using a dark-field microscope with 60× and 100× lenses. Scale bar = 30 µm. B, Finite track-length adjusted plot demonstrating graph of concentration versus size for Sepia melanin (left) and DAM particles obtained using NanoSight. The data represent the average of 5 independent trials of 1 mg/ml solution per trial.

Figure 2.

Electron microscopic images of Sepia melanin and DAM particles. A and C, Representative scanning electron microscope (SEM) images of Sepia melanin particles. B and D, Representative SEM images of DAM particles. Representative transmission electron microscope images of (E) Sepia melanin and (F) DAM. The scale bars in (A, B) denote 1 µm; (B, C) denote 500 nm; and in (E, F) denote 50 nm.

Figure 3.

Binding analysis of harmane and 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) to Sepia melanin and dopamine melanin. Concentration-dependent binding analysis of harmane with 1 mg of Sepia melanin (A) and dopamine melanin (DAM) (C); and PhIP with 1 mg Sepia melanin (E) and DAM (G). The graphs on the right represent % of harmane bound to 1 mg of Sepia melanin (B) and DAM (D); and % PhIP bound to 1 mg of Sepia melanin (F) and DAM (H). The data are presented as mean ± SEM for 3 independent experiments, n = 6–9/group.

Figure 4.

Formation of neuromelanin in tyrosinase-overexpressing SH-SY5Y neuroblastoma cells. A, Representative immunoblot showing expression of tyrosinase (TYR, 80 kDa) in SH-SY5Y cells transfected with empty vector (EV) or TYR constructs for 48 h. β-actin (43 kDa) used as a loading control. The bar histogram represents the normalized densitometry of tyrosinase in EV- and TYR-transfected cells. The data represented as mean ± SEM, n = 4/group, analyzed using unpaired t test. ****p < .0001 compared with EV group. B, Representative dark-field microscopic image of EV or TYR-transfected SH-SY5Y cells at 48 h, demonstrating the formation of melanin in TYR SH-SY5Y cells. Scale bar = 30 µm. C, Estimated amount of neuromelanin per EV and TYR SH-SY5Y cell after 36, 48, 60, 72, and 96 h post-transfection. Neuromelanin levels were quantified by solubilizing cells in basic buffer at 80°C for 2 h then measuring absorbance at 470 nm. Sepia Melanin was used to generate a standard curve. The data represented as mean ± SEM of pg/cell values (n = 6/group), analyzed by 2-way ANOVA followed by Sidak’s post hoc test. *p < .05, **p < .01, and ****p < .0001 comparing EV- versus TYR-*transfected SH-SY5Y cells at a given time post-transfection.

Figure 5.

Electron microscopic observation of SH-SY5Y cells shows neuromelanin pigments. SH-SY5Y neuroblastoma cells were transfected with empty vector (EV) or tyrosinase (TYR) constructs and 48 h post-transfection, cells were fixed for TEM. A, C, E, and G, Representative images of EV SH-SY5Y cells taken with an FEI/Philips CM-100 transmission electron microscope at different magnifications. B, D, F, and H, Representative TEM images of TYR SH-SY5Y cells taken using an FEI/Philips CM-100 transmission electron microscope shows the presence of dark-black pigments. The scale bars in denote 5 µm in (A, B); 2 µm in (C, D); 1 µm in (E, F); 500 nm in (G, H). The arrows point toward intracellular neuromelanin pigments in TYR-transfected SH-SY5Y cells.

Figure 6.

Increased intracellular 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) uptake in neuromelanin-forming SH-SY5Y cells. SH-SY5Y neuroblastoma cells were transfected with empty vector (EV) or tyrosinase (TYR) for 48 h followed by treatment with 100 nM radiolabeled PhIP (¹⁴C) for 24 h. Post-treatment, cell lysates and media supernatants were analyzed for ¹⁴C radioactivity using PerkinElmer TRI-CARB 4910TR 110V Liquid Scintillation Counter. The standard curve was obtained and PhIP concentrations were determined in cell lysates and media concentrations, based on relative cpms. A, Radiolabeled PhIP (¹⁴C) levels in cells versus in treatment media. B, Ratio of accumulated PhIP (¹⁴C) in cells to treatment media after 24 h of exposure. The data represented as mean ± SEM for experiment performed at least in triplicate. Statistical analysis performed using 2-way ANOVA and Sidak’s post hoc test for (A) and unpaired t test for (B). **p < .01 and ****p < .0001 comparing EV- versus TYR-transfected SH-SY5Y cells treated with ¹⁴C-PhIP (n = 6/group).

Figure 7.

Decreased cell viability in neuromelanin-forming, heterocyclic aromatic amine (HAA)-treated SH-SY5Y cells. Galactose supplemented SH-SY5Y neuroblastoma cells were transfected with empty vector (EV) or tyrosinase (TYR). Post 48 h of transfection, cells were exposed to increasing concentrations of (A) harmane, (B) 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), (C) 4′-OH PhIP, (D) N-OH-PhIP, or (E) NO₂-PhIP for 24 h. An MTT assay was performed to assess cell viability, and dose-response curves were obtained for EV or TYR cells. Calculated LC50 are mentioned in each graph for respective toxicants. The data represented as mean ± SEM for each treatment group performed in triplicate with n = 9. *p < .05, **p < .01, ***p < .001, and ****p < .0001 comparing EV- versus TYR-transfected SH-SY5Y cells at a given HAA concentration. F and G, Time-dependent cell death of EV and TYR SH-SY5Y cells treated with 10 µM harmane and 10 µM PhIP, respectively. The data represented as mean ± SEM for n = 6, analyzed by 2-way ANOVA followed by Sidak’s post hoc test. *p < .05, **p < .01, ***p < .001, and ****p < .0001 compared with SH-SY5Y cells transfected with EV at 24 h. Abbreviation: ns, nonsignificant.

Figure 8.

Neuromelanin formation increases heterocyclic aromatic amine-induced oxidative damage. SH-SY5Y cells were transfected with tyrosinase (TYR) to form neuromelanin in vitro, whereas empty vector (EV)-transfected cells were used as control cells. Empty vector and TYR SH-SY5Y cells were treated with varying concentrations of different toxicants. Post 24 h treatment, intracellular ROS levels were measured using CM-H2DCFDA fluorogenic dye. Dose-dependent alterations of ROS levels in EV and TYR SH-SY5Y cells exposed to (A) harmane, (B) 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), (C) 4′-OH PhIP, (D) N-OH-PhIP, and (E) NO2-PhIP were recorded as % control of observed relative fluorescence unit. The data presented as mean ± SEM for experiment performed at least twice independently, n = 6–8, analyzed by 2-way ANOVA followed by Sidak’s post hoc test to determine significance. ^ap < .05, ^aap < .01, ^aaap < .001, and ^aaaap < .0001 compared with EV at 0 h; ^bp < .05, ^bbp < .01, ^bbbp < .001, and ^bbbbp < .0001 compared with TYR at 0 h; whereas, *p < .05, **p < .01, ***p < .001, and ****p < .0001 demonstrate significant difference between EV- versus TYR-transfected SH-SY5Y cells at a given time-point. Changes in levels of reduced glutathione (GSH) at various time-points in EV and TYR SH-SY5Y cells exposed to (F) 10 µM harmane and (G) 10 µM PhIP, respectively. The data were calculated as % of EV control group and represented as mean ± SEM, n = 4 per group, analyzed by 2-way ANOVA followed by Sidak’s post hoc test. ^ap < .05, ^aap < .01, ^aaap < .001, and ^aaaap < .0001 compared with EV at 0 h; ^bp < .05, ^bbp < .01, ^bbbp < .001, and ^bbbbp < .0001 compared with TYR at 0 h; whereas, *p < .05, **p < .01, ***p < .001, and ****p < .0001 demonstrate significant difference between EV- versus TYR-transfected SH-SY5Y cells at a given time-point.

Figure 9.

Neuromelanin increases 3-nitrotyrosine (3-NT) in heterocyclic aromatic amine-treated SH-SY5Y cells. Post-transfection with empty vector (EV) or tyrosinase (TYR) for 48 h, SH-SY5Y neuroblastoma cells were treated with 10 µM 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) for 24 h. Post-treated, fixed cells were imaged using a dark-field microscope under a 60× lens. Cells were labeled with actin (red) and 3-NT (green) using specific antibodies. Neuromelanin images were converted to “magenta” color using ImageJ. Hoechst was used to mark the chromatin nucleus (blue). Merged images on the extreme right show all channels. Inset from each demonstrates marked regions. Scale bar = 30 µm. For quantification, 3-NT fluorescence intensities were measured in regions of interest surrounding each cell. Data presented as the mean ± SEM; analyzed using 2-way ANOVA with Sidak’s multiple comparison post hoc test, *p < .05 and ****p < .0001 compared with EV control; ^ΔΔp < .01 compared with TYR control; whereas ^##p < .01 compares EV and TYR cells treated with PhIP.

Figure 10.

N-acetyl cysteine (NAC) alleviates heterocyclic aromatic amine (HAA)-induced toxicity in neuromelanin forming SH-SY5Y cells. After 48 h of transfection, the tyrosinase (TYR) overexpressing SH-SY5Y cells were pretreated with NAC (100 µM) for 1 h followed by treatment with 10 µM 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) or 10 µM harmane for 24 h. A, Assessment of ROS levels in TYR SH-SY5Y cells pretreated with NAC (100 µM, 1 h) followed by HAA treatment. Changes in relative fluorescence unit values were plotted as % control and represented as mean ± SEM, n = 8/group. B, Cell death was assessed using MTT assay in PhIP or harmane-treated TYR SH-SY5Y cells, with or without NAC pretreatment. The cell viability was presented as % of vehicle-treated control group and represented as mean ± SEM, n = 6–8/group. ****p < .0001 compared with non-treated control group; ^ΔΔp < .001 denotes significance between harmane-treated groups with or without NAC pre-exposure; whereas ^ϕϕp < .001 and ^ϕϕϕp < .0001 denote significance between PhIP-treated groups with or without NAC pre-exposure.