Primary cultures of rat cortical neurons were prepared from embryonic 16 days old fetal Sprague-Dawley rats. In brief, brain tissues free of meninges were dissociated by triturating through a Pasteur pipette, and cell suspension (cell number: 5×10⁵ cells/ml) was added onto the culture plates pre-coated with poly-D-lysine and laminin. Cells were maintained in Dulbecco's modified Eagle's medium (DMEM) containing 10% heat-inactivated fetal bovine serum (FBS) in humidified 95% air/5% CO₂ at 37℃. Three days later, cytosine arabinoside (5 µM) was added for 1 day to block non-neuronal cell division. Subsequently medium was replaced twice a week. Experiments were performed on 16-18 days cultures.

Cells were exposed to HNE (1 µM) and/or NMDA (2 µM) in nominally Mg²⁺-free Hank's balanced salt solution (HBSS) containing 1.5 mM CaCl₂, 1 mM glycine, and 10 mM glucose. For the calcium-free condition, CaCl₂ was not included in the HBSS. Six hours later, cell injury or death was assessed by the morphological examination using a phase-contrast microscopy and quantified by measuring the amount of lactate dehydrogenase (LDH) released into the bathing medium. LDH activity was measured using a diagnostic kit (Sigma Chemical Co., St. Louis, MO). Cell death was expressed as percentage of total LDH, which was measured in sister cultures frozen and thawed after the experiments.

[Ca²⁺]_i was measured fluorometrically according to the fura-2 method, as we described before (Lim et al., 2006). Cells were cultured on poly-D-lysine and laminin-coated coverslips (diameter=13 mm). They were then washed twice with HBSS, incubated with 1 µM fura-2AM for 30 min at 37℃, and then washed twice with HBSS. A coverslip with cells was diagonally placed inside the cuvette, and the bathing medium was stirred with a teflon-coated magnetic bar to allow adequate mixing of added drugs. The fluorescence was stimulated at 340 and 380 nm of incident light and collected at 510 nm. Constants were determined for each coverslip with 0.1% Triton-X100 for Rmax and 40 mM EGTA for R_min. The ratios of [Ca²⁺]_i at 340 an 380 nm were calculated according to the formula: [Ca²⁺]_i=K_d(R-R_min)/(R_max-R)×S_f2/S_b2, where R is the 340/380 wavelength ratio of the sample, Rmin is the ratio obtained with fura-2 in the absence of Ca²⁺, and S_f2/S_b2 is the ratio of fluorescence values at 380 nm for free and Ca²⁺-bound dye. A Kd value of 224 nM was used in this study.

Cells were loaded with JC-1 (1.0 µg/ml; Molecular Probes, Eugene, OR) for 20 min at 37℃. Depolarization of ΔΨ_m was assessed by measuring fluorescence intensities at the excitation wavelength of 485 nm and dual emission wavelengths of 530 and 590 nm (for measuring JC-1 monomers and aggregates, respectively) using a fluorescence microplate reader (SpectraMax GeminiXS, Molecular Devices, CA). During measurements, cells were maintained at 37℃ and protected from light. Fluorescence intensity was measured every 15 or 30 min. All fluorescent measurements were corrected by autofluorescence that was determined in cells not loaded with JC-1. Autofluorescence was constant throughout the experiment. In control experiments, no photobleaching was observed during fluorescence monitoring.

In order to measure [Ca²⁺]_m, cultured cortical neurons were incubated with the mitochondrial Ca²⁺-sensitive fluorescent indicator, rhod-2 AM (5 µM, Molecular Probes), for 30 min at room temperature. Rhod-2 fluorescence was excited at 560 nm and emitted fluorescence was collected through a 590 nm long pass barrier filter using a fluorescence microplate reader (SpectraMax GeminiXS, Molecular Devices, CA). The rhod-2 fluorescence intensities were expressed as F/F₀ (F₀: fluorescence before drug addition, F: maximal fluorescence after drug addition).

Mitochondria were isolated from livers of male Sprague-Dawley rats (220-280 g) (Charles River Breeding Lab), as we described before [17]. Rats were starved overnight before sacrifice. All steps of mitochondria isolation were carried out on ice and in a cold medium containing 0.25 M sucrose, 2 mM K⁺-EDTA, and 3 mM HEPES, adjusted to pH 7.4 with KOH. Rat liver was washed, minced, and homogenized on ice using a ground-glass homogenizer. The homogenate was subjected to centrifugation at 600 g for 10 min, and the supernatant obtained to 8,000 g for 20 min. The subsequent mitochondrial pellet was washed in EDTA-free medium and used for experiments. Mitochondrial protein concentration was determined by the Bradford method using bovine serum albumin as the standard.

The ΔΨ_m in isolated mitochondria was measured using JC-1 [17] or rhodamin 123 [18]. Rhodamine 123 has spectral and metabolic properties as a probe of transmembrane potential in isolated rat-liver mitochondria. Thus, mitochondria (0.5 mg protein) were suspended with 2 ml of recording buffer containing 150 mM sucrose, 5 mM MgCl₂, 5 mM succinate, 2.7 µM rotenone, 200 nM Ca²⁺, 50 mM potassium phosphate, and 20 mM HEPES (pH 7.4). Before each experiment, mitochondria in the recording medium were incubated with each dye [0.2 µM rhodamine 123 (Molecular Probes) or JC-1 (1.0 µg/ml)], for 20 min in the dark in a 3 ml recording cuvette at 35℃. The emission signals at 590 and 530 nm elicited by the excitation at 485 nm for JC-1 and the emission signals at 529 nm elicited by excitation at 507 nm for rhodamin 123 were measured with a spectrofluorophotometer RF-5301PC (Shimadzu, Kyoto, Japan). The ratio of the signal at 590 nm over that at 530 nm (red/green ratio) of JC-1 or fluorescence intensity of rhodamin 123 was calculated to estimate ΔΨ_m. The recording chamber was magnetically stirred, and the measurements were carried out at 35℃.

MPTP opening was assessed by measuring mitochondrial swelling under energized and de-energized conditions, as described previously [19]. Mitochondrial swelling was determined spectrophotometrically (Shimadzu UV-2401PC, Kyoto, Japan) by measuring absorbance changes at 540 nm. Mitochondria (1 mg protein) were incubated at 25℃ under energized conditions in 2 ml of medium containing 13 mM mannitol, 70 mM sucrose, 3 mM HEPES (pH 7.4), 10 mM succinate, and 1 µM rotenone. When required, de-energized condition was achieved by incubating mitochondria in KSCN buffer (150 mM KSCN, 20 mM MOPS, 10 mM Tris, 0.5 µM rotenone, and 0.5 mM antimycin, pH 7.4).

The level of intracellular ATP was measured using the method described previously (Choi et al., 2002). Briefly, cells were lysed with 10% trichloroacetic acid, sonicated for 1 min on ice, and 2 mM EDTA and 2 mg/ml bovine serum albumin (BSA) were added with. After centrifugation, the supernatant was collected and neutralized with 4 N to the lysate. The ATP content was determined using a luminescence detection kit (Molecular Probes, Eugene, OR).

Data are expressed as mean ± standard error of mean (S.E.M.) and analysed for statistical significance using repeated measures of ANOVA or two-way ANOVA by running SAS Window v.9.1. program. If needed, post-hoc Scheffe's test was performed for multiple comparisons. A p value <0.05 was considered significant.

In the present study, neither HNE (1 µM) nor NMDA (2 µM) alone injured neuronal cells (Fig. 1A). However, combined treatment of the cells with HNE and NMDA evoked the release of LDH (Fig. 1A) with apparent morphological deterioration (data not shown). Furthermore, it is of a great interest to note that the synergistic death was observed only in the presence of calcium (Fig. 1). HNE alone neither increased the cytosolic calcium level ([Ca²⁺]_i) nor altered the NMDA-induced intracellular calcium influx (Fig. 2), which is consistent with our previous observation (Lim et al., 2006). Therefore, our previous and present results together suggest that NMDA-mediated [Ca²⁺]_i rise interplays with HNE to augment the neuronal cell death.

Mitochondrial dysfunction is one of the prominent features of oxidative stress-mediated cell death. Oxidative stress, including HNE, has been shown to depolarize the ΔΨ_m and facilitate Ca²⁺-evoked MPTP opening [20]. Similarly, NMDA receptor activation is also known to induce mitochondrial dysfunction [21]. As shown in Fig. 3A, however, neither 1 µM HNE nor 2 µM NMDA altered the ΔΨ_m in neuronal cells (Fig. 3A). On the other hand, when the cells were incubated for 3 h with 1 µM HNE together with 2 µM NMDA, there was complete depolarization of ΔΨ_m (Fig. 3A), which was comparable to the extent induced by 5 µM FCCP, a potent mitochondrial uncoupler (data not shown). Interestingly, the potentiated depolarization of ΔΨ_m by HNE and NMDA occurred only in the presence of extracellular calcium (Fig. 3A, B). Accumulation of calcium in mitochondria is well associated with the depolarization and dysfunction of mitochondria. Our results clearly showed that co-treatment with HNE and NMDA increased the [Ca²⁺]_m in cortical neurons (Fig. 3C). However, neither 1 µM HNE nor 2 µM NMDA alone showed any significant changes in [Ca²⁺]_m (Fig. 3C). The results imply that the synergistic cytotoxicity by HNE and NMDA may be related to the rise of [Ca²⁺]_m and mitochondrial dysfunction.

Our previous and present data showed that 2 µM NMDA elevated the [Ca²⁺]_i up to 200 nM. Therefore, we tested in isolated mitochondria whether HNE (1 µM) could induce ΔΨ_m depolarization and MPTP opening in the presence of 200 nM calcium. Neither 1 µM HNE nor 200 nM calcium alone affected ΔΨ_m in isolated mitochondria (data not shown). However, 1 µM HNE significantly depolarized ΔΨ_m in the presence of 200 nM calcium in the test buffer (Fig. 4). Mitochondrial swelling experiments also showed that 1 µM HNE strongly promoted the MPTP opening induced by 200 nM CaCl₂ in both energized (Fig. 5A) and de-energized (Fig. 5B) conditions. These findings indicate that HNE facilitates the calcium-induced opening of MPTP. The HNE/Ca²⁺-facilitated MPTP opening shown in Fig. 4 was completely blocked by two different MPTP blockers, cyclosporine A and ciclopirox (data not shown).

We previously reported that ciclopirox protected adenocarcinoma cells from H₂O₂ toxicity via complete prevention of ΔΨ_m decrease and MPTP opening [17]. Unlike cell membrane-impermeable cyclosporine A or bonkrekic acid, the cell membrane-permeable MPTP blocker ciclopirox rapidly and markedly reduced mitochondrial damage induced by H₂O₂ in cultured cells. In the present study, therefore, we further examined whether ciclopirox blocked the synergistic toxicity of HNE and NMDA on mitochondria, and found that combined treatment of HNE and NMDA strongly increased the [Ca²⁺]_m (Fig. 6A) and synergistically decreased the intracellular ATP level (Fig. 6B). However, ciclopirox completely blocked the mitochondrial calcium overload, restored intracellular ATP levels and ameliorated the neuronal cell death caused by HNE and NMDA (Fig. 6).