MN9D cells were cultivated on the poly-D-Lysine (25 µg/ml PDL: Sigma Chemical Co., St. Louis, MO, USA)-coated p-100 plate in Dulbecco's modified Eagle's medium (DMEM: Sigma) supplemented with heat inactivated 10% fetal bovine serum (FBS: BioWhittaker, Walkersville, MD, USA) in an atmosphere of 10% CO₂ at 37℃. Prior to drug treatment, culture medium was switched to serum-free, chemically-defined N2 supplement [13] containing the predetermined concentrations of D-glucose (Sigma, 5-35 mM) plus 200 µM MPP⁺. If necessary, cells were co-treated with Boc-aspartyl(OMe)-fluoromethylketone (BAF; Enzyme Systems Products, Dublin, CA, USA), N-benzyloxycarbonyl-Val-Ala-Asp-fluoromethylketone (zVAD; Enzyme Systems Products), N-acetylcystein (NAC, Sigma), or SP600125 (Cell Signaling; Boston, MA, USA).

Following drug treatment, cells were washed twice with ice cold PBS and dissolved in a buffer containing 50 mM Tris, pH 8.0, 2 mM EDTA, 1% triton X-100, 2 mM phenymethylsulfonylfluoride (PMSF), and 50 µg/ml aprotinin (all from Sigma). Lysates were homogenized in a Dounce homogenizer on ice, followed by centrifugation at 13,000×g for 30 min at 4℃. Protein concentrations of the supernatant were measured using Bio-Rad protein assay kit (Herculus, CA, USA). Approximately 50 µg of protein was separated on the 12.5% SDS-polyacrylamide gel and blotted onto pre-wet polyvinylidene difluoride (PVDF; Bio-Rad) membrane. The membrane was blocked with Tris-buffered saline (TBS) containing 5% nonfat milk for 60 min at RT, followed by incubation at 4℃ with the indicated primary antibodies. These include a rabbit polyclonal anti-activated caspase 3 antibody (1: 1,000; Cell Signaling), a rabbit polyclonal anti-activated caspase 9 antibody (1: 1,000; Cell Signaling), a rabbit polyclonal anti-phospho-JNK antibody (1: 1,000; Cell Signaling), a rabbit polyclonal anti-JNK antibody (1: 1,000; Cell Signaling), a rabbit polyclonal anti-phospho-p38 antibody (1: 1,000; Cell Signaling), a rabbit polyclonal anti-p38 antibody (1: 1,000; Cell Signaling). Blots were further incubated with an appropriated secondary antibody conjugated with horse radish peroxidase for 60 min at RT. Signals were visualized using Enhanced Chemiluminescence (ECL) kit (Amersham Phamarcia, San Diego, CA, USA).

The rate of cell viability was assessed by a colorimetric measurement of 3-[4,5-dimethythioazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT: Sigma) reduction assay [14]. Briefly, cells at various time points were incubated with 1 µg/ml of MTT solution for 2 hr and then further incubated with 20% SDS in 50% dimethylformamide (DMF: Sigma) for 24 hr. The optical density of the dissolved formazan grains within the cell were measured at 570 nm. Values of each treatment were calculated as a percentage over the untreated control (100% survival).

Following drug treatment, MN9D cells on P-100 plate were washed twice with ice cold PBS and subjected to lysis in a buffer containing 500 µl 1% Triton X-100, 50 mM Tris buffer, PH7.4, and 2 mM EDTA for 30 min on ice. After microcentrifugation at 4℃ for 15 min, the supernatants were subjected to phenol/chloroform extraction and ethanol precipitation. Residual RNA contaminants were digested by incubation with DNase-free RNAse A. The resulting samples were loaded onto 1.2% agarose gel and subsequently electrophoresed at 50V for 4 hr. After staining with ethidium bromide, DNA gel was visualized on a UV-transilluminator and photographed using Polaroid 667 film.

Levels of ROS in the MN9D cell were measured with ROS-sensitive fluorescent dyes: dihydroethidium and 2, 7-dichlorofluorescin diacetate (DCF: Molecular Probes, Eugene, OR, USA). Following drug treatment, MN9D cells were washed twice with a saline containing 144 mM NaCl, 10 mM HEPES, 2 mM CaCl₂, 1 mM MgCl₂, 1 mM KCl, and 10 mM D-glucose. Cells were loaded with 1 µM dihydroethidium and incubated at 37℃ for 15 min. The cells were then washed twice with a saline and photographed under a fluorescence microscope (Carl Zeiss, Zena, Germany). To improve the permeability of DCF dye, cells were loaded with 5 µM DCF dye containing 0.1% pluronic F-127 (Molecular Probes) for 30 min prior to MPP⁺ treatment.

To prepare primary cultures of DA neurons, the ventral mesencephalon was isolated from the 14d gestation Sprague Dawley rat embryo (Daehan; Daejon, Korea) as described previously [12]. The dissected tissues were incubated with 0.01% trypsin in HBSS for 10 min at 37℃ and triturated with a constricted Pasteur pipette. The dissociated cells were plated at 1.0×10⁵ cells per 8-mm diameter Aclar embedded film (Electron Microscopy Sciences, Fort Washington, PA, USA) pre-coated with 100 µg/ml poly-D-lysine (Sigma) and 4 µg/ml laminin (Invitrogen, San Diego, CA, USA) in DMEM containing 2 mM glutamine plus varying concentrations of glucose.

For a double staining of tyrosine hydroxylase (TH; a rate-limiting enzyme of dopamine biosynthesis) and activated caspase 3, DA neurons exposed to 3 µM MPP⁺ for 48 hr were fixed with 4% paraformaldehyde for 20 min at RT and blocked for 1 hr in PBS containing 5% normal goat serum and 0.1% Triton X-100 at 37℃ overnight. Primary antibodies used were a mouse monoclonal anti-TH antibody (1: 7500; Pel-Freez, Rogers, AR, USA) and a rabbit polyclonal anti activated caspase-3 antibody (1: 200; Cell Signaling). After washing with PBS, the DA neurons were incubated with Alexa Flour 568-conjugated goat anti-rabbit IgG and Alexa Fluor 488-conjugated goat anti-mouse IgG (1: 200; Molecular Probes) at RT for 1 hr. Cells were then examined under an Axiocam Digital Camera (Carl Zeiss). To determine the number of the surviving TH-positive neurons after MPP⁺ treatment in the presence or absence of various inhibitors, DA neurons were fixed, blocked, and incubated with a monoclonal anti-TH as described above. After washing with PBS, cells were incubated with a biotinylated goat anti-mouse Ig G at RT for 1 hr, followed by peroxidase-conjugated avidin-biotin complex (1: 200; Vector Laboratories, Burlingame, CA, USA) for 30 min at RT. Staining was visualized by applying 3, 3'-diaminibenzidine (DAB, substrate for peroxidase; Vector Laboratories). Manual counts from each condition present in each 8 mm Aclar embedding film were made of all of the TH-positive cells having neurites twice the length of the soma [12]. A total of 300~400 TH positive neurons were counted in the untreated control.

The data were expressed as the means±S.D. or the means±S. E.M. Significance between groups was determined by one-way analysis of variance (ANOVA) and the Tukey's post hoc test using GraphPad Prism 6. Values of ^**p<0.01 or ^*p<0.05 were considered statistically significant.

To examine whether change of extracellular D-glucose concentrations may contribute to inducing a distinct cell death mode after MPP⁺ treatment, MN9D cells were incubated in N2 culture medium containing varying concentrations of D-glucose (5~35 mM) plus 200 µM MPP⁺. As shown in Fig. 1A, MPP⁺ caused a quite similar rate of cell death regardless of extracellular concentration of glucose. However, we observed that morphology of dying cells was quite differed and it depended on extracellular glucose levels. At lower concentrations of glucose (5~12.5 mM), for example, MPP⁺ led to typical of apoptotic cell death accompanying with shrinkage of cell body and a phase-bright appearance (Fig. 1B). At higher concentrations of glucose (15~35 mM), vacuolelike structures and flat appearance were apparent in cells after treatment with MPP⁺. Immunoblot analysis clearly indicated that activation of both caspase-9 and -3 was only detected in cells cultivated in lower concentrations of glucose (5~12.5 mM) following exposure to MPP⁺ (Fig. 1C). This is quite contrary to our previous study [11,12] demonstrating that no obvious signs of apoptosis were detected in MN9D cells and primary cultured DA neurons following exposure to MPP⁺. Since these cells were cultivated in culture medium containing 17.5 mM or higher, we further attempted to confirm that MN9D cells indeed die of apoptosis at lower concentrations of glucose. Therefore, we selected two doses of extracellular glucose (5 mM vs 17.5 mM). Again, MPP⁺ caused a quite similar kinetic of cell death regardless of extracellular levels of glucose (Fig. 2A). DNA fragmentation assay showed that MPP⁺-induced DNA fragmentation was occurred in a time-dependent manner only in MN9D cells cultivated in media containing 5 mM glucose (Fig. 2B). Immunoblot analysis also demonstrated that a time-dependent appearance of activated capsase-9 and -3 is detected only in MN9D cells at 5 mM glucose after MPP⁺ treatment (Fig. 2C). Although activated caspase-3 appeared as double bands in most cases, a single band of the activated caspase-3 was detected depending on passages of culture. Consequently, co-treatment with one of two widely used pan-caspase inhibitor (zVAD and BAF) rescued cells from MPP⁺-induced death when it was cultivated at 5 mM glucose (Fig. 2D). In contrast, this protective effect was not detected in cells cultivated at 17.5 mM (Fig. 2E). Taken together, we clearly demonstrated that MPP⁺ can induce either caspase-dependent or caspase-independent cell death and this is largely depended on extracellular levels of glucose in the medium.

Previously, we demonstrated that intracellular surge of calcium and its concurrent signaling is responsible for MPP⁺-induced cell death whereas ROS-mediated activation of MAPK significantly contributes to 6-OHDA-induced apoptotic cell death [15,16,17,18,19]. Therefore, we attempted to investigate whether generation of ROS at early phase of cell death and concurrent ROS-mediated MAPK signaling are involved and play an essential role in MPP⁺-induced apoptotic cell death. First, we measured levels of ROS using cell-permeable, ROS-sensitive fluorescent dyes: dihydroethidium (HET) and 2, 7-dichlorofluorescin diacetate (DCF) following treatment of MN9D cells with MPP⁺. Following MPP⁺ treatment, HET-positive cells appeared in a time-dependent manner when cells were cultivated at 5 mM glucose but not at 17.5 mM (Fig. 3A). Quantitative measurement indicated that levels of HET-sensitive ROS increased up to 2.5 fold in MN9D cells at 5 mM glucose. This increase was significantly blocked when cells were co-treated with an ROS scavenger, N-acetylcystein (NAC). When cells were stained with DCF, quite similar pattern of ROS surge was detected (Fig. 3B). Consequently, co-treatment of cells with NAC protected cells from MPP⁺-induced death when cells were cultivated only at 5 mM glucose (Fig. 3C).

Next, we wondered whether MPP⁺-induced surge of ROS is linked to activation of MAPK. As shown in Fig. 4A, phospho-JNK was first seen at 4 hr after MPP⁺ treatment in cells cultivated at 5 mM glucose and reached at its maximum at 8 to 12 hr and decreased thereafter. Levels of JNK remained the same at all time periods tested. However, any discernible signs of MPP⁺-mediated JNK activation were not detected when cells were incubated at 17.5 mM glucose. Interestingly, phosphorylated forms of p38 MAPK were not detected at any conditions tested (data not shown). Consequently, co-treatment with SP600125 (a JNK-specific inhibitor) rescued cells from MPP⁺-induced death when cells were cultivated at 5 mM glucose (Fig. 4B). Again, co-treatment with p38 inhibitor (PD 169316) or ERK inhibitor (U0126) had no effect on cell viability (data not shown). In sum, we clearly demonstrated that MPP⁺ can induce apoptotic cell death via ROS generation and subsequent activation JNK when MN9D cells were incubated at 5 mM glucose.

Finally, we chose primary cultures of DA neurons to firmly establish a correlation between levels of extracellular glucose and cell death mode. As we previously demonstrated [12], we prepared primary cultured DA neurons from the ventral mesencephalon of E14 rat embryos. We empirically first tested ranges of glucose concentrations in the medium that did not cause any detectable signs of cell death without drug treatment. We found that DA neurons were healthy when they are cultivated in medium containing 5~33.3 mM glucose for up to 10 days (data not shown). Double immunofluorescent staining showed that, following MPP⁺ treatment, active forms of caspase-3 appeared in TH-positive neurons when cells were cultivated at 5 mM glucose but not at higher concentrations of glucose (e.g., 20 mM or 33.3 mM glucose; Fig. 5A). Quantitative measurement assay showed that approximately 6.6% of TH-positive cells showed the active forms of caspase-3 in cells at 5 mM glucose (Fig. 5B). As shown in Fig. 6A, neurites from TH-positive cells were retracted and fragmented after MPP⁺ treatment. However, retraction and/or fragmentation of neurites were less obvious in cells co-treated with NAC (ROS scavenger), zVAD (a pan-caspase inhibitor) or SP600125 (JNK inhibitor). Indeed, numbers of TH-positive cells having neurites twice the length of the soma were increased when DA neurons were cultivated in the presence of one of these inhibitors (Fig. 6B). Taken all together, we showed an interesting scenario in which MPP⁺ can induces either ROS-mediated, caspase-activated apoptosis or caspase-independent cell death (Fig. 7). We have reported that MPP⁺ induces a distinct form of cell death that quite resembles necrosis in morphology [11]. In addition, we have demonstrated that MPP⁺-induced cell death cannot be rescued by a pan-caspase inhibitor [12]. In these previous reports, cells were cultivated in a medium containing higher levels of glucose (17.5 mM or higher). Several contradicting reports from other laboratories have indicated that ROS is involved in MPP⁺-induced cell death [20,21,22]. Our present results explain why and how cells die of either ROS-mediated, caspase-dependent pathway or -independent cell death pathway. To our knowledge, it is not known whether glucose level may be linked to DA neuronal death in patients with PD. Based on our present results, however, we propose that determination of cell death mode seems to be relied on levels of glucose in extracellular milieu. Therefore, our data explain how both caspase-dependent or -independent cell death are detected in postmortem brains of patient with PD.