Ethyl Pyruvate Rescues Nigrostriatal Dopaminergic Neurons by Regulating Glial Activation in a Mouse Model of Parkinson’s Disease

Parkinson’s disease (PD) is a neurodegenerative disorder characterized by motor symptoms, including resting tremor, muscle rigidity, bradykinesia (slowness of movement), and postural instability (1, 2). The pathological hallmark of PD is the loss of dopaminergic (DA) neurons in the substantia nigra pars compacta (SNpc) that project to the striatum (STR), which plays an essential role in normal motor function (2). Clinical parkinsonian symptoms arise when a majority (∼60–70%) of SNpc DA neurons are lost, resulting in reduced dopamine levels in the nigrostriatal system (3). Although PD is a sporadic disease of unknown pathogenesis, accumulating evidence suggests that glial activation-derived oxidative stress increases the risk of developing PD (4). In vivo and in vitro 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) models of PD have shown that key enzymes involved in the production of reactive oxygen species (ROS)/reactive nitrogen species (RNS), such as microglial NAPDH oxidase and inducible NO synthase (iNOS), and astroglial myeloperoxidase (MPO), are upregulated in damaged areas and contribute to DA neuronal death (5–8).

Ethyl pyruvate (EP), a derivative of pyruvate, is an effective scavenger of ROS, especially hydrogen peroxide, by virtue of its nonenzymatic oxidative decarboxylation reaction. Because it is a simple ester, EP might be more stable in aqueous solution than the pyruvate anion (9). EP has been reported to exert pharmacological effects, such as scavenging of ROS, suppression of inflammation, inhibition of apoptosis, and support of cellular ATP synthesis (10). Several lines of evidence have shown that systemic administration of EP decreases infarct volume in the ischemic brain (11) and decreases kainic acid-induced neuronal cell death in the hippocampus (12). EP was also found to protect PC12 cells from dopamine-induced apoptosis (13). However, little is known about the effects of EP related to inflammation in the nigrostriatal DA system in the context of PD. Thus, the current study sought to determine whether EP could prevent the degeneration of nigrostriatal DA neurons by regulating glial activation and reducing oxidative stress in the MPTP model of PD.

Experimental protocol (KHUASP[SE]-10-030) was approved by the Institutional Animal Care and Use Committee of the Kyung Hee University. All experiments conducted with 8- to 10-wk-old male C57BL/6 mice (22–24 g; Charles River Breeding Laboratory) in a room maintained at 20–22°C on a 12-h light/dark cycle with food and water available ad libitum. For MPTP intoxication, mice received four i.p. injection of MPTP-HCl (20 mg/kg; Sigma-Aldrich) dissolved in PBS at 2-h intervals in accordance with published guidelines (14). For EP, in a Ringer’s-type Ca²⁺- and K⁺-containing balanced salt solution (9), mice received a single injection per day of EP (10, 25, and 50 mg/kg body weight; Sigma-Aldrich) into the peritoneum at 12 h after the last MPTP injection. Some mice were injected with EP alone as a control. For sodium pyruvate (Py) in PBS, mice received a single injection per day of Py (500 mg/kg body weight; Sigma-Aldrich) into the peritoneum at 12 h after the last MPTP injection. Some mice were injected with Py alone or vehicle as a control.

Animals were transcardially perfused with a saline solution containing 0.5% sodium nitrate and heparin (10 U/ml) and then fixed with 4% paraformaldehyde dissolved in 0.1 M phosphate buffer. Brains were dissected from the skull and postfixed overnight in buffered 4% paraformaldehyde at 4°C. And then, the brains were stored in a 30% sucrose solution for 2–3 d at 4°C until they sank and were frozen sectioned on a sliding microtome in 30-μm-thick coronal sections. All sections were collected in six separate series and processed for immunohistochemical staining as described previously (15). The primary Abs included those directed against tyrosine hydroxylase (TH, 1:2000; Pel-freez), macrophage Ag complex-1 (MAC-1, 1:200; Serotec), ionized calcium-binding adaptor molecule 1 (Iba-1, 1:1000; Wako Chemicals), CD68 (clone ED1 CD68 microglial/macrophage marker [ED-1] 1:1000; Serotec), glial fibrillary acidic protein (GFAP, 1:5000; Neuromics), MPO (1:500; Thermo Scientific), 8-hydroxy-2′-deoxyguanosine (8-OHdG, 1:200; JaICA), and nitrotyrosine (1:100; Millipore). Stained cells were viewed and analyzed under a bright-field microscope (Nikon) or viewed with a confocal laser-scanning microscope (Olympus). For Nissl staining, substantia nigra (SN) tissues immunostained with 8-OHdG or nitrotyrosine Ab were mounted on gelatin-coated slide and dried for 1 h at room temperature and stained with 0.5% cresyl violet (Sigma-Aldrich).

At 2 d after the last MPTP injection, animals were sacrificed, and the brains were harvested for fluorojade-C staining (Chemicon). The staining was performed according to the manufacturer’s protocol. Briefly, 30-μm-thick coronal sections were mounted on gelatin-coated slides, dried for 20 min at room temperature, immersed in 1% sodium hydroxide in 80% ethanol, rinsed in 70% ethanol for 2 min and in distilled water for 2 min, and then incubated in 0.06% potassium permanganate solution for 10 min. After the rinse with distilled water, slides were incubated in 0.0001% fluorojade-C solution for 10 min. After the rinse with distilled water, slides were coverslipped with Vectashield medium and viewed using an LSM confocal laser scanning microscope (Carl Zeiss). To assess the extent of neuronal injury in the sections, the number of fluorojade-C–immunopositive (ip) cells per area was counted.

For double-immunofluorescence staining, sections were processed as described recently (15). Briefly, 30-μm-thick coronal sections were mounted on gelatin-coated slides, dried for 20 min at room temperature, and washed twice in PBST (PBS containing 1% Tween 20). Slides were incubated overnight at 4°C in a combination of an anti-GFAP Ab (1:5000; Neuromics), anti–MAC-1 Ab (1:200; Serotec) and anti-MPO Ab (1:500; Thermo Scientific). After washing in PBST, the sections were incubated for 1 h at room temperature with a mixture of FITC-conjugated anti-mouse IgG (Vector Laboratories) and/or Cy3-counjugated anti-rabbit IgG (Millipore). Slides were coverslipped with Vectashield medium and viewed using an LSM confocal laser scanning microscope (Carl Zeiss). To analyze the localization of different Ags in double-stained samples, images were obtained from the same area and merged using interactive software. For Nissl and GAD67 double-immunofluorescence staining, SN tissues were obtained at 7 d after last MPTP injection and 30-μm-thick coronal sections were mounted on gelatin-coated slides, dried for 20 min at room temperature, and washed twice in PBST (PBS containing 1% Tween 20). Slides were incubated overnight at 4°C in a combination of an anti-GAD67 Ab (1:200; Chemicon). After washing in PBST, the sections were incubated for 1 h at room temperature with a Cy3-counjugated anti-mouse IgG (Millipore). After washing in PBS, the sections were incubated for 1 h at room temperature with NeuroTrace staining solution (1:100; Molecular Probes). The sections were washed with PBST and then incubated overnight at 4°C in PBS. Slides were coverslipped with Vectashield medium and viewed using an LSM confocal laser scanning microscope (Carl Zeiss). To analyze the localization of different Ags in double-stained samples, images were merged the same area and obtained by the computed mask of the colocalized pixels.

The unbiased stereological estimation of the total number of the TH-ip neurons, MAC-1–ip cells, Iba-1–ip cells, and ED-1–ip cells in the substantia nigra (SN) was made using the optical fractionator method performed on an Olympus Computer-Assisted Stereological Toolbox system version 2.1.4 (Olympus) as described recently (15). The sections used for counting covered the entire SN from the rostral tip of the pars compacta back to the caudal end of the pars reticulate (anterioposterior, −2.06 to −4.16 mm from bregma) (16). The SN was delineated at a ×1.25 objective and generated counting grid of 150 × 150 μm. An unbiased counting frame of known area (47.87 × 36.19 μm = 1733 μm²) superimposed on the image was placed randomly on the first counting area and systemically moved through all counting areas until the entire delineated area was sampled. Actual counting was performed using a ×100 objective. The estimate of the total number of neurons was calculated according to the Optical Fractionator Equation (17).

As previously described (15), an average of 17 coronal sections of the STR, starting from the rostral anteroposterior (+1.60 mm) to anteroposterior (0.00 mm), according to bregma of the brain atlas (16), were examined at a ×5 magnification using the IMAGE PRO PLUS system (version 4.0; Media Cybernetics) on a computer attached to a light microscope (Zeiss Axioskop) interfaced with a charge-coupled device video camera (Kodak Mega Plus model 1.4 I). To determine the density of the TH-immunoreactive staining in the STR, a square frame of 700 × 700 μm was placed in the dorsal part of the STR. A second square frame of 200 × 200 μm was placed in the region of the corpus callosum to measure background values. To control for variations in background illumination, the average of the background density readings from the corpus callosum was subtracted from the average of density readings of the STR for each section as described recently (15). Then, the average of all sections of each animal was calculated separately before the data were processed statistically.

An accelerating rotarod (five-lane accelerating rotarod; Ugo Basile) was used to measure forelimb and hindlimb motor dysfunctions as described recently (15). The rotarod unit consisted of a rotating spindle (diameter, 3 cm) where mice were challenged for speed. For training, mice were placed on the rotarod at 10 rpm for 20 min, 7 consecutive days before MPTP treatment. Mice that stayed on the rod without falling during training were selected and randomly divided into experimental groups. Seven days after the last MPTP treatment, animals receiving various treatment regimes were placed in a separate compartment on the rotating rod and tested at 20 rpm for 20 min. The latency to fall was automatically recorded by magnetic trip plates.

To measure contents of dopamine in STR, we performed the reversed-phase HPLC with electrochemical detector as described recently (15). Dissected striatal tissues were homogenized with 0.1 M perchloric acid and 0.1 mM EDTA buffer and centrifuged at 9000 rpm for 20 min. The supernatant was injected into an autosampler at 4°C (Waters 717 plus autosampler) and eluted through μBondapak C18 column (3.9 × 300 mm × 10 μm, ESA) with mobile phase for catecholamine analyses (Chromosystems). The peaks of dopamine content were analyzed by ESA Coulochem II electrochemical detector and integrated using a commercially available program (Breeze). All samples were normalized for protein content as spectrophotometrically determined using the Bio-Rad protein assay kit (Bio-Rad).

The extracted striatal tissues were frozen and homogenized with a tissue crusher and sonicated for 10 min in an ice bath. After sonication, the samples were centrifuged at 13,000 × g for 15 min. The upper layers were taken for solid-phase extraction cleanup. The process of solid-phase extraction was performed using mixed-mode strong cation-exchange (1 ml, 30 mg) cartridges. Norepinephrine was eluted with 2 ml 5% NH₄OH, and the extracts were dried under the gentle stream of nitrogen gas and reconstituted with 70 μl ethyl acetate. Prior to injection, extracts were derivatived. The sample analysis was carried out with an Agilent GC/MS (Palo Alto, CA) equipped with DB-5MS fused-silica capillary column (30 × 0.25 mm i.d., film thickness 0.25 μm; J&W Scientific, Folsom, CA). The oven temperature was held at 150°C for 2 min, elevated to 195°C at 10°C/min and to 210°C at 5°C/min, increased to 250°C at 20°C/min and elevated from 250 to 300°C at 40°C/min, and held for 3 min. The mass spectra of derivatives were obtained in the mass range from 100 to 600 amu. For the monitoring and confirmation analysis, the selected ion monitoring mode was used, and the dwell time of each ion was set at 50 ms.

Hydroethidine histochemistry was performed for in situ visualization of O₂⁻ and O₂⁻-derived oxidants (15). Three days after the last MPTP injection, hydroethidine (1 mg/ml in PBS containing 1% DMSO; Molecular Probes) was administered i.p. After 15 min, the animals were transcardially perfused with a saline solution containing 0.5% sodium nitrate and heparin (10 U/ml) and then fixed with 4% paraformaldehyde in 0.1 M phosphate buffer. After fixation, the brains were cut into 30-μm slices using a sliding microtome. Sections were mounted on gelatin-coated slides, and the oxidized hydroethidine product ethidium was examined by IX71 confocal microscop (Olympus Optical). To quantify, obtained images were analyzed by the Computer Imaging Device and accompanying software (Fujifilm).

For immunoprecipitation, we dissected SN tissues from the animals at 2 d after injection of MPTP. As recently described (15), SN samples were homogenized with using a glass homogenizer in ice-cold buffer consisting of the following: 50 mM Tris (pH 8), 150 mM NaCl, 2 mM EDTA, 1% Nonidet P-40, 0.1% SDS, 0.5% deoxycholate, and protease inhibitor mixture (Roche). Homogenates were centrifuged for 5 min at 500 × g at 4°C, and supernatants were collected and centrifuged for 20 min at 13,000 × g at 4°C. Equal amounts of proteins (240 μg) were immunoprecipitated with anti-p47^phox Ab (2.4 μg; Santa Cruz Biotechnology) or anti-TH Ab (2.4 μg; Millipore). The immunoprecipitated proteins were separated on a 12% SDS-PAGE gel and were then electroblotted onto polyvinylidene difluoride membrane and blocked for 1 h at room temperature with TBS containing 5% nonfat dried milk. The membranes were immunoblotted with anti-phosphoserine Ab (1:1000; Invitrogen Life Technologies), anti-p47^phox Ab (1:1000; Santa Cruz Biotechnology), anti-gp91^phox Ab (1:1000; BD Biosciences), anti-nitrotyrosine Ab (1:1000), and anti-TH Ab (1:2000). For semiquantitative analyses, the densities of whole bands on immunoblots were measured with the Computer Imaging Device and accompanying software (Fujifilm).

We dissected SN tissues from the animals at 3 d after injection of MPTP. As previously described (18, 19), SN samples were homogenized with using a glass homogenizer in ice-cold buffer consisting of the following: 20 mM HEPES, 250 mM sucrose, 10 mM KCl, 1.5 mM MgCl₂, 2 mM EDTA, and protease inhibitor mixture (Roche). Homogenates were centrifuged for 5 min at 500 × g at 4°C, and supernatants were collected and centrifuged for 20 min at 13,000 × g at 4°C. Equal amounts of proteins (30 μg) were separated by 10% SDS-PAGE gels, transferred to polyvinylidene difluoride membrane (Millipore) using an electrophoretic transfer system (Bio-Rad). The membranes were immunoblotted with anti–β-tubulin Ab (1:1000; Santa Cruz Biotechnology), anti-nitrotyrosine Ab (1:1000; Millipore), anti-MPO Ab (1:000; Thermo Scientific), and anti-iNOS Ab (1:500; BD Biosciences). For semiquantitative analyses, the densities of bands on immunoblots were measured with the Computer Imaging Device and accompanying software (Fujifilm).

Neuron-enriched mesencephalic cultures were prepared as described previously (20, 21). In brief, dissociated cells from the ventral mesencephalons of embryonic day (E) 14 Sprague–Dawley rats were seeded on 12-mm round aclar plastic coverslips (1.2 × 10⁵ cells/coverslip) housed in 24-well culture plates. To suppress the proliferation of glial cells, at 2 d in vitro (DIV), the medium was replaced with a chemically defined serum-free medium. At DIV 4, cultures were pretreated with EP for 30 min and then treated with vehicle or 20 μM 1-methyl-4-phenyl-pyridinium (MPP⁺) in RF medium ([DMEM; Life Technologies, Rockville, MD] supplemented with 6 mg/ml gllucose, 204 μg/ml l-glutamine, 100 U/ml penicillin/streptomycin, and 2% FBS [Hyclone]) for 48 h. These cultures were processed for immunocytochemistry. When cultures were immunostained with cell-type specific Abs (see above immunohistochemistry), the cell composition included ∼5% astrocytes and less than ∼1% microglia (21). The remaining cells were presumed to be neurons, 1–1.2 and 9.5–11% of which were TH-ip and γ amino butyric acid-ip neurons, respectively (22). For methyl green staining, cells were stained for 5 min at room temperature with 0.5% methyl green solution (Sigma-Aldrich) after immunostained with TH Ab.

Mesencephalic microglia cultures were prepared from the ventral mesencephalons of E14 Sprague–Dawley rat brains as described previously (23). Tissues were triturated and plated in 75-cm² T flasks precoated with poly-d-lysine at a density of 1 × 10⁷ cells/flask and then maintained in DMEM supplemented with 10% FBS. After 2–3 wk, microglia were detached from the flasks, applied to a nylon mesh to remove astrocytes, and then plated into 24-well plates at a density of 5 × 10⁴ cells/well. At 1 h after plating, unattached cells were removed, and at 24 h after plating, the cells were pretreated with EP for 30 min and then treated with MPP⁺ or vehicle.

At DIV 3, neuron-enriched mesencephalic cultures plated on 12-mm round aclar plastic coverslips (1.2 × 10⁵ cells/coverslip) housed in 24-well plates were supplemented with 5 × 10⁴ mesencephalic microglia/well. After 1 h, the culture medium was replaced with neuronal culture medium (DMEM containing 2% FBS); 24 h later, the cocultures were pretreated with, EP for 30 min at indicated doses and then treated with MPP⁺ or vehicle for 48 h and were processed for measurement of dopamine uptake or for immunocytochemistry.

Measurement of dopamine uptake was performed as described previously (24). Briefly, cultures were washed twice with an incubation solution (HBSS containing 10 mM HEPES, 0.6% glucose, 0.2 mM pargyline, and 0.01% ascorbic acid), tritiated dopamine ([³H]DA; 444 GBq/mmol) was added to a final concentration of 83.3 nM, and the cultures were incubated at 37°C for 20 min. Blanks were obtained by incubating cultures at 0°C. The reaction was stopped by rinsing the cultures with ice-cold incubation solution. The cultures were then lysed with 0.2 M NaOH containing 0.2% Triton X-100 and transferred to scintillation vials for counting.

Intracellular ATP contents were measured using luciferin–luciferase reaction with CellTiter-Glo Luminescent Cell ViabilityAssay kit (Promega) according to the manufacturer’s instructions. Briefly, 20 μl cell media was mixed with 20 μl CellTiter-Glo Luminescent Cell ViabilityCellTiter-Glo Reagent and incubated at 20°C for 10 min. We measured the luminescence signal with an LB 9501 Lumatluminometer (Berthold, German). We subtracted a background luminescence value in control wells containing medium without cells from the signal. The amounts of ATP contents were normalized to protein concentration. The ROS generation was measured using 5,6-chloromethyl-2′,7′-dichlorodihydrofluorescein diacetate, acetyl ester (Molecular Probes). The cells in phenol red-free media without serumin black 96-well culture plate were incubated with 1 μM 5,6-chloromethyl-2′,7′-dichlorodihydrofluorescein diacetate, acetyl ester at 37 °C for 1 h and then incubated with 0.5 μM Hoechst 33342 at room temperature for 10 min. The fluorescence intensities at 485/535 nm for diacetate, acetyl ester were normalized by Hoechst intensity at 355/460 nm. All data were presented as percent of control after calculation.

All values are expressed as mean ± SEM. Statistical significance (p < 0.05 for all analyses) was assessed by ANOVA using Instat 3.05 (GraphPad, San Diego, CA), followed by Student–Newman–Keuls analyses.

The mice in each group received four i.p. injections of MPTP (20 mg/kg) or PBS (control) at 2-h intervals. Seven days later, brains were removed, and sections were immunostained for TH to specifically detect DA neurons. Consistent with our recent report (15), there was a significant loss of TH-ip cell bodies in the SN (Fig. 1B) and fibers in the STR (Fig. 2B) at 7 d in MPTP-injected mice compared with those in PBS-treated control mice (Figs. 1A, 2A). TH-ip cells in the SN and their nerve terminals in the STR were quantified by stereological counts and densitometric analyses, respectively. MPTP treatment decreased the number of TH-ip neurons by 69% in the SN (Fig. 1E) and reduced the OD of TH-ip fibers by 68% in the STR (Fig. 2E) compared with those of PBS-treated mice.

To investigate whether EP altered MPTP-induced neurotoxicity of nigrostriatal DA neurons, we administered EP (10–50 mg/kg body weight) 12 h after the last MPTP or PBS injection. We chose a 12-h time point because MPTP is completely converted into MPP⁺ and sufficiently accumulates in DA neurons prior to 12 h posttreatment (15); thus, its effects could not be attributed to the diminished metabolism of MPTP to MPP⁺ or uptake of MPP⁺ into DA neurons. The results of TH immunohistochemistry showed that treatment with 50 mg/kg EP effectively reduced MPTP-induced loss of DA neurons in the SN (Fig. 1C) and their nerve terminals in the STR (Fig. 2C). Quantification of TH immunostaining showed that 25 and 50 mg/kg EP increased the number of TH-ip neurons in the SN by 13% (p < 0.01; Fig. 1E) and 25% (p < 0.001; Fig. 1E), respectively, and produced a corresponding increase in the OD of TH-ip fibers in the STR of 23% (p < 0.01; Fig. 2E) and 35% (p < 0.001; Fig. 2E) compared with controls. Treatment with 10 mg/kg EP did not increase the number of TH-ip neurons. EP alone had no effect on the number of neurons in the SN (Fig. 1E) or DA fibers in the STR (Fig. 2E).

To investigate whether Py altered MPTP-induced neurotoxicity of nigrostriatal DA neurons, we administered 500 mg/kg Py 12 h after the last MPTP or PBS injection. Quantification of TH immunostaining showed that 500 mg/kg Py increased the number of TH-ip neurons in the SN by 32% (p < 0.001; Fig. 1D, 1E) and produced a corresponding increase in the OD of TH-ip fibers in the STR of 49% (p < 0.001; Fig. 2D, 2E) compared with controls.

To verify neuroprotective effect of EP on GABA neurons, SN tissues were double immunostaining with Nissl and GAD67 Ab. MPTP did not induce loss of Nissl⁺/GABA neurons. In addition, EP had no effect on the number of Nissl⁺/GABA neurons with or without MPTP (Supplemental Fig. 2J–N).

We next determined whether EP improves MPTP-induced motor deficits by testing performance on a rotarod apparatus, as described recently (15). Animals receiving various treatment regimens were evaluated 7 d after the last MPTP injection by measuring the latency to falling. MPTP treatment decreased sustained rotarod time to 11.04 ± 0.4 min, a 61% decrease compared with that of PBS treatment (p < 0.01; Fig. 2F). The behavioral dysfunction in MPTP-treated mice was recovered by EP treatment, which increased the latency to falling to 14.59 ± 0.83 min (p < 0.05; Fig. 2F). EP alone had no effect on motor behavior.

After rotarod performance tests, the mice were sacrificed for biochemical assessments. In parallel with behavioral dysfunction, HPLC analyses revealed that MPTP reduced striatal dopamine levels to 71% of those in PBS controls (Fig. 2F). By contrast, treatment of MPTP-injected mice with EP increased dopamine levels by 25% compared with MPTP only (p < 0.05; Fig. 2F). EP alone had no effect on striatal dopamine levels.

The loss of norepinephrine may aggravate the MPTP-mediated damage of the nigrostriatal DA system (25, 26). gas chromatography–mass spectrometry analysis showed that norepinephrine levels were unchanged in the STR with or without MPTP (Supplemental Fig. 1H). This observation is comparable to a recent report that striatal norepinephrine level was decreased by acute MPTP treatment (40 mg/kg per dose, twice), and these effects of acute MPTP on norepinephrine may be transient (26).

Accumulating evidence suggests that activated microglia play an important role in DA neuronal cell death in the MPTP model (27). Thus, we next examined whether the neuroprotective effect of EP resulted from inhibition of microglial activation in the SN. At 3 d after the last MPTP injection, brain tissues from mice treated with or without EP were processed for immunostaining with Abs against the markers for activated microglia, MAC-1, Iba-1, and ED-1. In contrast to the SN of control mice treated with PBS (Fig. 3A) or EP (Fig. 3B) alone, where relatively few positive microglia were observed, the SN of MPTP-treated mice contained numerous robustly immunoreactive MAC-1–positive (activated) microglia (Fig. 3C). EP treatment dramatically reduced the number of MAC-1–ip microglia in the MPTP-treated SN (Fig. 3D, 3M), indicating that EP suppressed MPTP-induced microglial activation.

Consistent with MAC-1 staining result, Iba-1–positive microglia exhibited the typical ramified morphology of resting microglia in the PBS-treated (Fig. 3E) and EP-treated (Fig. 3F) control SN. In contrast, the majority of Iba-1–ip microglia in the MPTP-treated SN displayed an activated morphology, including larger cell bodies with short and thick process (Fig. 3G). Treatment with EP mitigated these effects of MPTP, dramatically decreasing the number of activated microglia in the MPTP-treated SN (Fig. 3H, 3M).

Microglia in the MPTP-treated SN appeared to reach a state similar to that of active phagocytes (Fig. 3K), as determined by ED-1 immunostaining, which labels phagocytes (28, 29). The majority of ED-1–ip microglia in the MPTP-treated SN displayed an activated morphology, including larger cell bodies with short and thick process or no processes (Fig. 3K). Treatment with EP mitigated these effects of MPTP, dramatically decreasing the number of activated microglia in the MPTP-treated SN (Fig. 3L, 3M). The numbers of ED-1–ip cells were not detectable in the SN of mice treated with PBS (Fig. 3I) or EP (Fig. 3J) only.

Few such ED-1–ip cells were observed in the MPTP and EP-treated SN (Fig. 3L), similar to the results of MAC-1 immunostaining. ED-1–ip cells were not detectable in the SN of mice treated with PBS (Fig. 3I) or EP (Fig. 3J) only.

The production of ROS, such as H₂O₂ and O₂⁻, is increased in the midbrain of MPTP-treated mice, and oxidants originating from microglia are thought to mediate the loss of DA neurons in the SN (30). Thus, we examined whether EP rescued nigral DA neurons by inhibiting MPTP-induced oxidant production. The fluorescent product of oxidized hydroethidine (i.e., ethidium) was significantly increased in the MPTP-injected SN 3 d after administration (Fig. 4C, 4E) compared with that of the PBS-injected SN (Fig. 4A, 4E). EP dramatically decreased MPTP-induced oxidant production in the SN in vivo (Fig. 4D, 4E) but had no effect alone (Fig. 4B, 4E).

We then examined the assembly and activation of NADPH oxidase as a source of ROS production and resultant oxidative damage in the SN 2 d after the last MPTP injection by immunoprecipitating with an anti-p47^phox Ab and then immunoblotting with anti-phosphoserine, anti-gp91^phox and anti-p47^phox Abs. In MPTP-treated SN tissues, p47^phox phosphorylation and p47^phox–gp91^phox complexes were significantly increased compared with PBS-treated SN samples (p < 0.01; Fig. 4F, 4G), indicating activation of NADPH oxidase. By contrast, treatment with EP after the MPTP injection significantly decreased phosphorylation of p47^phox and the level of p47^phox–gp91^phox complexes in the SN compared with MPTP alone (p < 0.01; Fig. 4F, 4G). EP alone had no effect on the level of phospho-p47^phox or p47^phox–gp91^phox complex formation (Fig. 4F, 4G). These results suggest that MPTP-induced activation of NADPH oxidase is inhibited by EP. As shown in our recent report (15), p47^phox immunoreactivity was detected exclusively in activate microglia but not astrocytes and neurons (data not shown).

We have recently shown that, in MPTP mice, DA neuronal cell death is accompanied by increased levels of 8-OHdG, a marker of oxidative nucleic acid damage (15). To determine whether EP prevents MPTP-induced oxidative damage to DNA in the SN, we immunostained for 8-OHdG. The present results showed that the levels of 8-OHdG were dramatically increased in the SN 3 d after MPTP injection (Fig. 4J) compared with those in the PBS-treated SN (Fig. 4H). This MPTP-induced oxidative DNA damage was dramatically inhibited by EP (Fig. 4K). EP alone had no effect on DNA damage (Fig. 4I).

It has been shown that transgenic mice deficient in iNOS are resistant to MPTP-induced neurotoxicity (5). Thus, we examined whether EP might modulate DA neuronal survival by affecting MPTP-induced expression of iNOS in the SN. Two days after the final MPTP injection, SN tissues were dissected and prepared for Western blot analyses. The results showed that EP attenuated MPTP-induced expression of iNOS protein in the SN (p < 0.0.1; Fig. 5A, 5B).

Nitration of protein tyrosine residues, a well-known marker of oxidative stress in PD patients (31), is mediated by iNOS-derived oxidants (32). To measure the extent of iNOS-mediated oxidative damage to proteins, we immunostained the SN for nitrotyrosine 3 d after MPTP injection, with or without EP treatment. We found that the levels of nitrotyrosine were dramatically increased in the MPTP-injected SN (Fig. 5E) compared with the PBS-treated SN (Fig. 5C). This MPTP-induced nitration of protein was dramatically inhibited by EP (Fig. 5F), which had no effect alone (Fig. 5D). Consistent with this immunocytochemical staining, Western blot analyses also demonstrated that the levels of nitrotyrosine were significantly increased in the MPTP-treated SN compared with those in the PBS-treated SN (p < 0.01; Fig. 5A, 5B). EP significantly reduced the levels of MPTP-induced nitrotyrosine and increased expression of TH proteins in the MPTP-treated SN (p < 0.01; Fig. 5A, 5B). TH is a target for MPTP-induced nitrotyrosine, leading to loss of TH (33, 34). Additional Western blot analysis revealed that EP significantly attenuated MPTP-induced nitrotyrosine of TH (Supplemental Fig. 1G), indicating that EP protects TH from nitrosative damage.

Because EP was administered 12 h after the final injection of MPTP in all in vivo experiments, its effects could not be attributed to reduced metabolism of MPTP to MPP⁺ or MPP⁺ uptake into DA neurons (15). However, the possibility remained that EP might promote neuronal survival by preventing MPP⁺-induced blockade of mitochondrial respiration in neurons. To test this hypothesis, we performed additional experiments with mesencephalic neurons cultured alone or cocultured with microglia. In microglia-free, neuron-enriched mesencephalic cultures, pretreatment with 0.5–1 mM EP (30 min before MPP⁺ treatment) had no protective effects (Fig. 6A–D, 6I), and 5 mM EP alone was neurotoxic (data not shown). By contrast, in cocultures of mesencephalic neurons and microglia, 1 mM EP blocked MPP⁺-induced death of DA neurons (p < 0.05; Fig. 6E–H, 6I).

GFAP-immunostained tissues adjacent to those used for MAC-1 immunostaining confirmed that resting astrocytes (small somas with thin processes; Fig. 7A, 7B) were transformed into active astrocytes (enlarged cell bodies with short or thick processes; Fig. 7C, 7D) in MPTP-treated mice. Interestingly, treatment with EP (50 mg/kg) after MPTP injection sustained the activated state of GFAP–ip astrocytes (Fig. 7E, 7F), but alone had no effect (data not shown).

It has been shown that MPO is upregulated in astrocytes in the SN of PD patients and MPTP-treated mice; furthermore, mice deficient in MPO are resistant to MPTP neurotoxicity (35). Thus, we examined whether EP inhibited MPTP-induced expression of MPO in the SN, which would be predicted to enhance neuronal survival. Western blot analyses showed that the levels of MPO were 4-fold higher in the MPTP-treated SN than that in the SN of PBS-treated controls (p < 0.001; Fig. 7G, 7H). Treatment with EP after MPTP injection dramatically decreased the level of MPO protein expression by 61% (p < 0.001; Fig. 7G, 7H) but alone had no effect. Double immunostaining showed MPO expression in GFAP–ip astrocytes (Supplemental Fig. 1A–C) but not in microglia (Supplemental Fig. 1D–F). The number of MPO and GFAP double-ip (MPO⁺/GFAP⁺) cells in the MPTP-treated SN was 4-fold higher than in PSB- or EP-treated controls (p < 0.001; Fig. 7H).

In this study, we demonstrated that EP protects nigrostriatal DA neurons from MPTP neurotoxicity in vivo by regulating glial activation and the resultant oxidative stress. We showed that EP suppressed MPTP-induced ROS/RNS generation and reduced oxidative damage to nucleic acids by inhibiting microglia-derived NADPH oxidase and/or iNOS and astrocyte-derived MPO, leading to survival of nigrostriatal DA neurons, recovery of striatal dopamine depletion in vivo, and reversal of motor dysfunction. To our knowledge, this is the first study to show that EP rescues nigrostriatal DA neurons through regulation of glial activation in the MPTP model of PD.

It is well-known that EP increases the level of intracellular Py and tricarboxylic acid metabolites because EP is cleaved into ethanol and Py by numerous esterases (36). Several lines of evidence have shown that Py decreases infarct volume in the ischemic brain (11) and protects neuroblastoma cells against MPP⁺ or 6-hydroxydopamine neurotoxicity (37). To verify whether neuroprotective effects of EP are associated with metabolic recovery, additional experiments were performed to assess neuroprotective effects of Py in MPTP mice. In the present study, we showed that Py protects nigrostriatal DA neurons from MPTP neurotoxicity as evidenced by TH immunostaining, indicating that neuroprotective effects of EP may be due to metabolic protection.

Microglia, resident immunocompetent, and phagocytic cells play an important role in supporting neurons in the CNS (38). However, in the presence of adverse stimuli, they may become activated and contribute to chronic, damaging inflammation and, ultimately, to neurodegeneration. Many studies have reported the presence of reactive microglia in the SN of PD patients (39, 40) and in the SN of animal models of PD produced by administration of MPTP (15, 41). Activated microglia produce NADPH oxidase- and/or iNOS-derived ROS/RNS, which trigger or exacerbate neurotoxicity by inducing oxidative stress to neurons (19, 35).

Activated NADPH oxidase is evident in the SN of PD patients (7) and in the SN of MPTP mouse PD models (15). NADPH oxidase is composed of the cytosolic components p47^phox, p67^phox, and Rac1 as well as the membrane components gp91^phox and p22^phox (42, 43). Upon stimulation, the oxidase complex is activated, a process that involves phosphorylation of p47^phox on serine residues and recruitment of the cytosolic components to the membrane-bound components, including gp91^phox, to assemble the active NADPH oxidase (44, 45). Thus, both phosphorylation of p47^phox and binding of p47^phox and gp91^phox are thought to be important aspects of NADPH oxidase activation (46); the resulting activation leads to ROS generation (18). The generation of ROS by NADPH oxidase is increased in the SN of PD patients (7) and in the SN of MPTP-treated mice (15), and oxidants originating from microglia are thought to mediate the loss of DA neurons in the SN (47). Several studies have provided evidence for oxidative stress in PD patients and in MPTP-treated mice, including high levels of oxidative nucleic acid damage, as visualized by 8-OHdG immunostaining in the SN (15, 48). The results of the current study show that MPTP activated NADPH oxidase, as demonstrated by increases in phosphorylation of p47^phox serine residue and binding of phospho-p47^phox and gp91^phox. NADPH oxidase activation resulted in increased O₂⁻ and O₂⁻-derived oxidants and DNA damage, as visualized by hydroethidine staining and 8-OHdG immunostaining in the SN, respectively. Treatment with EP not only inhibited microglial NADPH oxidase activation but also mitigated ROS production and nucleic acid oxidation. These results verify that EP inhibited MPTP-induced activation of NADPH oxidase and oxidative damage, thereby resulting in neuroprotection in the MPTP model.

RNS, another major contributor to oxidative stress, has been implicated in the pathogenesis and progression of PD (49). Moreover, NO, synthesized by iNOS, reacts with superoxide to form peroxynitrate, which subsequently induces oxidative stress to proteins by nitration of tyrosine residues (31, 32, 50). Several studies have demonstrated that reactive glia that express iNOS and/or increased levels of nitrotyrosine are present in the midbrains of PD patients (51, 52). In the MPTP mouse model of PD, iNOS contributes to DA neuronal death through MPTP-induced RNS and increases in nitrotyrosine levels (5, 53), although the existence of iNOS in human microglia remains controversial (54). Treatment with EP not only reduced iNOS expression but also decreased the levels of TH nitrotyrosine (Supplemental Fig. 1G) in the SN of MPTP-treated mice. Collectively, these results suggest that EP decreases MPTP-induced expression of iNOS and oxidative stress, leading to increased neuronal survival.

Although the role of activated astrocytes remains controversial, accumulating evidence obtained from the MPTP model suggests that activation of astrocytes can contribute to the loss of nigrostriatal DA neurons via production of neurotoxic substances (6, 55, 56). MPO, which is expressed in activated astrocytes, is the key enzyme involved in the generation of cytotoxic ROS/RNS (57, 58). These astroglial MPO-derived oxidants can be released and subsequently exert toxic effects on adjacent mesencephalic DA neurons in the MPTP model (15). A role for this mechanism is supported by evidence that increased levels of 3-chlorotyrosine- and hypochlorous acid-modified proteins—MPO-derived oxidative stress markers—are present in the MPTP-treated SN (6). Alternatively, MPO secretion may participate in DA neuronal death in the MPTP model of PD via microglia activation (59). Although potential cytokine-like effects of MPO are still not well-known, this enzyme may play an important role in the signaling pathway of microglial activation that results in neuronal death (15, 59). This is consistent with our present data showing that MPO expression in activated astrocytes was increased in the MPTP-treated SN, as evidenced by an increase in the number of MPO⁺/GFAP⁺ cells. Western blot analyses also confirmed increases in MPO expression. Notably, EP significantly attenuated the levels of MPO expression in the SN. Similar to our recent report (15), the present findings suggest that EP rescues DA neuronal death via suppression of MPO expression in activated astrocytes. Interestingly, in spinal cord injury, EP reduces MPO expression originating in neutrophils but not in astrocytes (60). This discrepancy may reflect differences in the animal models used (PD versus spinal cord injury) and/or regions studied (SN versus spinal cord) in each experiment.

In the current study, we found that EP failed to protect DA neurons from MPP⁺ neurotoxicity in neuron-enriched mesencephalic cultures, as assessed by TH immunostaining. By contrast, recent data have demonstrated that the same dose of EP as used in our experiments (1 mM) decreases neuronal cell death caused by MPP⁺ neurotoxicity in SH–SY5Y neuroblastoma cells, as assessed using the tetrazolium-based MTT assay (61). The apparent discrepancy in experimental results between these two studies is probably a result of differences in the cell types used (primary mesencephalic neurons versus neuroblastoma cells). The neuroprotective effects of EP in cocultures of mesencephalic microglia and neurons also provide further evidence that EP acts through microglia to mediate its neuroprotective effects under our experimental conditions. Although not tested directly, these results provide indirect evidence that the neuroprotective effects of EP do not involve blocking MPP⁺ entry into DA neurons.

EP has been known to reinforce mitochondrial functions such as the production of ATP in cerebrocortical slice cultures (36) and ROS in PC12 cells and cortical microglia cultures (13, 62), which results in neuronal survival. Taking this into consideration, the current data show that EP aids neuronal survival by affecting mitochondrial functions (increased ATP production and decreased ROS levels) in the cocultures of mesencephanlic microglia and neurons (Supplemental Fig. 3A, 3B). The data carefully suggest that the effect of EP on mitochondria is possibly an additional protective mechanism. However, this might be irrelevant for our study because even though EP can produce similar amounts of ATP and ROS (Supplemental Fig. 3A, 3B), it fails to show neuroprotection in neuron-enriched cultures. Therefore, it is likely that under our experimental conditions, the effect of EP on mitochondrial functions is an indirect contribution to neuronal survival. Although we do not provide direct evidence, the speculation is that it does so by regulating microglial-mediated inflammation.

The most prominent biochemical change in the STR of PD patients and MPTP-treated mice is a decrease in the levels of dopamine (2, 14). In our study, such deficits in striatal dopamine in MPTP-treated mice led to a decreased latency to fall from an accelerating rotarod apparatus, reflecting diminished coordination and balance (15). EP was found to increase striatal dopamine levels and to ameliorate motor deficits in MPTP-treated mice. These behavioral and in vivo biochemical effects of EP on the lesioned nigrostriatal DA system, together with the present finding of inhibitory actions of EP on glia-mediated oxidative stress, suggest that EP may be beneficial in treating PD related to neuroinflammation.

It is worth considering whether the neuroprotection afforded by EP results in the partial restoration or prevention of further decreases in DA neurons in the nigrostriatal pathway. Our recent study revealed that MPTP reduced DA neurons by 17% in the SN and the OD of striatal TH-ip fibers by 15% at 12 h after the last MPTP injection, indicating that a majority of nigrostriatal TH-ip neurons are viable (15). Therefore, it is likely that the observed neuroprotective effects are due to the prevention of further decreases in nigrostriatal DA neurons. Although we did not provide direct evidence, it is also possible that partial restoration may account for the neuroprotective actions of EP.

In conclusion, in view of the clinical safety of EP and its ability to effectively penetrate the blood–brain barrier (63), we propose that EP may have therapeutic value in the treatment of aspects of PD and other disorders associated with neuroinflammation and glial activation-mediated oxidative damage.

The authors have no financial conflicts of interest.