Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Although the cellular mechanisms underlying neuronal death caused by cerebral ischemia are not fully understood, perturbation of intracellular Ca²⁺ homeostasis and subsequent activation of a variety of Ca²⁺-related enzymes, including nitric-oxide synthase (NOS), phospholipases, and proteases, may participate in the pathogenesis of ischemic brain damage (Orrenius et al., 1989; Choi, 1990, 1992; Kristián and Siesjö, 1998). Several lines of evidence have demonstrated that the elevation of intracellular free Ca²⁺ concentrations ([Ca²⁺]_i) under ischemic or hypoxic conditions is attributable to the activation of voltage-gated Na⁺ and Ca²⁺ channels (Buchan et al., 1994; Lysko et al., 1994; Lynch et al., 1995; Taylor and Meldrum, 1995). A variety of Na⁺ channel blockers, such as tetrodotoxin (Lysko et al., 1994) and phenytoin (Ratand et al., 1994); Ca²⁺ channel blockers, particularly an N-type Ca²⁺ channel blocker, SNX-111, a synthetic peptide derived from -conotoxin MVIIA (Buchan et al., 1994); and a P/Q-type Ca²⁺ channel blocker, -agatoxin IVA (-Aga) (Asakura et al., 1997), have been demonstrated to protect neurons against focal as well as global cerebral ischemia in rats. We have reported previously that the hypoxic injury induced in rat cerebrocortical slices by a transient exposure to hypoxia/glucose deprivation, followed by reoxygenation, is abolished by the removal of extracellular Ca²⁺ or an intracellular Ca²⁺ chelator, 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid/acetoxymethyl ester (BAPTA/AM) (Tatsumi et al., 1998b). Moreover, the hypoxic injury in rat cerebrocortical slices is suppressed by a variety of Na⁺ channel blockers and Ca²⁺ channel blockers, including an N-type blocker, -conotoxin GVIA (-CTX), and -Aga (Tatsumi et al., 1998b).

NS-7 [4-(4-fluorophenyl)-2-methyl-6-(5-piperidinopentyloxy) pyrimidine hydrochloride] is a novel Na⁺/Ca²⁺ channel blocker developed in our laboratories. It shows an almost equipotent blocking action on tetrodotoxin-sensitive Na⁺ current as well as L-type and N-type Ca²⁺ currents with IC₅₀ values of 4.5 to 7.8 µM in patch-clamp configuration using NG 108-15 cells (Suma et al., 1997), displaces [³H]batrachotoxin binding with K_i of 1 µM in rat brain membrane (Shimidzu et al., 1997), and inhibits the KCl-evoked nitric oxide (NO) synthesis mediated through L-type and P/Q-type Ca²⁺ channels with IC₅₀ values of 2.5 and 3.1 µM, respectively, in primary neuronal culture (Oka et al., 1999). This compound has been shown to reduce the size of cerebral infarction as well as activation of calpain, a Ca²⁺-activated protease, caused by permanent middle cerebral artery occlusion in rats (Takagaki et al., 1997) and to inhibit the hypoxic injury in rat cerebrocortical slices (Tatsumi et al., 1998b). We have found recently in primary neuronal culture that NS-7 inhibits the NO synthesis induced by high concentrations (1-10 µM) of Bay K 8644, a Ca²⁺ channel agonist, without affecting the responses of low concentrations (0.01-0.1 µM) of Bay K 8644 (Oka et al., 1999). Moreover, NS-7, unlike other Na⁺ and Ca²⁺ channel blockers, reduces the KCl (80 mM)-evoked or veratridine-stimulated dopamine release from rat striatum, measured by an intracerebral microdialysis, without changing spontaneous or evoked release induced by 25 to 50 mM KCl (Itoh et al., 1998). Taken together, it is suggested that the blockade of Na⁺ and Ca²⁺ channels by NS-7 is dependent on the activity of these ion channels in which the compound blocks preferentially the activated Na⁺ and Ca²⁺ channels.

During cerebral ischemia, depletion of ATP causes a dysfunction of Na⁺/K⁺-ATPase and leads to the elevation of intracellular Na⁺ concentration, which in turn depolarizes neuronal membranes and subsequent activation of voltage-gated Na⁺ and Ca²⁺ channels. Membrane depolarization occurring after exposure to hypoxia or hypoxia/glucose deprivation, which is known as an anoxic depolarization, has been considered to be closely related to the neuronal injury (Boening et al., 1989). Fowler and Li (1998) have reported that in CA1 pyramidal neurons of rat hippocampal slices hypoxia/glucose deprivation for 10 min causes a marked decrease in tissue ATP content and anoxic depolarization. They also have shown that tetrodotoxin (1 µM) inhibits the anoxic depolarization as well as the loss of tissue ATP, indicating a close relationship between the ATP depletion and anoxic depolarization. The extent of tissue ATP depletion depends largely on the severity of hypoxic insults. Therefore, it seems probable that NS-7 blocks Na⁺ and Ca²⁺ channels more effectively in brain tissues exposed to severe hypoxic insults than in those under mild hypoxic condition. To ascertain this idea, the severity of the hypoxic injury was controlled by changing the glucose concentration, and the effect of NS-7 on severe and mild hypoxic injury was subsequently examined in slices of rat cerebral cortex. The effect of this compound on the hypoxia-induced enhancement of NO synthesis in the absence or presence of low concentration of glucose was examined in rat cerebrocortical slices in relation to the action on the depolarization (KCl)-evoked NO synthesis. In addition, the mode of inhibition by NS-7 of KCl-induced elevation of [Ca²⁺]_i was investigated in primary neuronal culture of mouse cerebral cortex and compared with that of nifedipine.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals. NS-7 and nifedipine were synthesized in our chemical laboratories. The following chemicals and drugs were obtained from commercial sources: N^G-nitro-L-arginine methylester (L-NAME; Nacalai Tesque, Kyoto, Japan), 3-isobutyl-1-methylxanthine (IBMX; Sigma Chemical Co., St. Louis, MO), -CTX (Peptide Institute, Osaka, Japan), (+)MK-801 (Funakoshi, Tokyo, Japan), and cyclic GMP enzyme immunoassay system (Amersham Co., Buckinghamshire, UK). Other chemicals were all of guaranteed grade.

Animals. Male 10- to 14-week-old Sprague-Dawley rats (Japan SLC, Shizuoka, Japan) and pregnant ddY strain mice (Japan SLC) were used. Rats were housed in groups of five to six, and mice were individually bred in a plastic cage in a room controlled at 21-25°C, 45 to 65% humidity, and maintained in an alternating 12-h light/dark cycle (lights automatically on at 8:00 AM). Food and water were freely given. All experiments were carried out in accordance with the Guide for the Care and Use of Laboratory Animals written by the Japanese Pharmacological Society.

Hypoxia/Reoxygenation-Induced Tissue Injury in Slices of Rat Cerebral Cortex. After immersing the whole brain in ice-cold 5% glucose solution for 10 min, the cerebral cortex was dissected on ice, and serial coronal sections of 500-µm thickness were prepared using a McIlwain tissue chopper (Mickle Laboratory Engineering Co., Gomshall, UK). The chopped tissue then was transferred to a glass dish containing ice-cold Krebs-Ringer-bicarbonate solution (KRB; 118 mM NaCl, 4.7 mM KCl, 2.54 mM CaCl₂, 1.2 mM MgCl₂, 1.2 mM KH₂PO₄, 25 mM NaHCO₃, and 11.5 mM D-glucose, pH 7.4), and each slice was gently isolated with forceps. A piece of the slice was transferred to a 12-well plastic dish containing 2 ml of KRB and was preincubated at 37°C for 1 h under continuous bubbling with a gas mixture of 95% O₂ and 5% CO₂. After preincubation, the slice was incubated in glucose-deprived KRB (in case of severe injury) or 3 mM glucose-containing medium (in case of mild injury) at 37°C for 45 min under continuous bubbling with 95% N₂ and 5% CO₂, followed by reoxygenation for 5 h in normal KRB. Tissue damage was assessed by the leakage of lactate dehydrogenase (LDH) into the incubation medium. A 50-µl portion of the incubation medium was taken at 5 h after the reoxygenation, and LDH activity in the medium was determined by using an enzymatic assay kit (Wako Pure Chemical, Osaka, Japan). LDH leakage was expressed as the percentage of that in the normoxic group in which slices were incubated for 5 h and 45 min in KRB under continuous bubbling with 95% O₂ and 5% CO₂. At the end of experiments, brain slices were dissolved in 1 ml of 1 N NaOH, and protein content was determined by the method of Bradford (1976) using BSA as the standard.

Measurement of Tissue ATP Content and Extracellular Glutamate during Exposure of Brain Slices to Hypoxic Insults. Brain slices were preincubated at 37°C for 1 h in normal KRB solution under continuous bubbling with 95% O₂, 5% CO₂ and then exposed for 45 min to hypoxia in the absence or presence of 3 mM glucose. In the normoxia group, after 1 h of preincubation, slices were incubated at 37°C for 45 min in normal KRB solution under continuous bubbling with 95% O₂, 5% CO₂. At the end of the 45-min incubation under normoxic or hypoxic condition, slices were transferred to plastic tubes containing 1 ml of 0.4 N perchloric acid and homogenized, while the incubation medium was used for glutamate assay. The brain homogenates were centrifuged at 10,000g for 30 min, and the resultant supernatant was neutralized to pH 6.0 with a 10% K₂CO₃ solution. The ATP content was determined by HPLC with spectrophotometric detection as described previously (Tatsumi et al., 1998b). In brief, a portion of the neutralized supernatant was injected directly onto an HPLC system composed of a pump (LC-6A; Shimadzu, Kyoto, Japan), a reversed-phase separation column (250 × 4.6 mm, Wakosil II 5C18 RS; Wako Pure Chemical), a UV spectrophotomonitor (SPD-6A; Shimadzu), and a chromatocorder (Chromatopac C-R4A; Shimadzu). The column temperature was 35°C. The mobile phase was 0.2 M ammonium dihydrogen phosphate (pH 4.1) containing tetrabutylammonium hydrogen sulfate (50 mg/l) (Aldrich, Milwaukee, WI) and 1% acetonitrile. The flow rate was 0.8 ml/min. ATP was detected at A₂₆₀.

Measurement of Nitric Oxide (NO) Formation in Extracellular Fluids. The NO formation was estimated from cyclic GMP accumulation in the extracellular fluids after addition of GTP, IBMX, and the soluble fraction of rat cerebellum containing guanylate cyclase as described previously (Oka et al., 2000). Briefly, the cerebral cortex and cerebellum were dissected on ice, and serial coronal sections of the cerebral cortex were prepared as described above. The cerebellum was homogenized with 10 volumes of ice-cold buffer A (50 mM Tris-HCl containing 1 mM EDTA, 1 mM dithiothreitol, and 200 mM phenylmethylsulfonyl fluoride) using a Polytron homogenizer (PT 3000; Kinematica AG, Littau, Switzerland). After centrifugation at 30,000g for 30 min, the supernatant fraction was used for the source of guanylate cyclase. A cerebrocortical slice was transferred to a 24-well plastic plate in which 2 ml of KRB was included and preincubated at 37°C for 1 h under continuous bubbling with 95% O₂, 5% CO₂. After preincubation, the incubation medium was discarded, and the slice was further incubated in 2 ml of oxygen- and glucose-deprived KRB (severe hypoxic insults) or oxygen-deprived but 3 mM glucose-containing medium (mild hypoxic insults) for 45 min under continuous bubbling with 95% N₂, 5% CO₂. In a set of experiments, normoxic slices were exposed to 30 to 50 mM KCl for 45 min to examine the effects of drugs on the depolarization (KCl)-induced NO synthesis. To measure the extracellular cyclic GMP formation during hypoxia or KCl stimulation, 0.5 mM GTP, 1 mM IBMX, and a 20-µl aliquot of the supernatant fraction of rat cerebellum (containing 50-80 µg of protein) were added to the incubation medium immediately after the exposure to hypoxia or KCl stimulation. The cyclic GMP produced in the incubation medium was measured by using the Amersham cyclic GMP enzyme immunoassay kit.

Primary Neuronal Culture. Primary neuronal culture was prepared from the cerebral cortex of fetal mice as described previously (Ma et al., 1991; Tatsumi et al., 1998a). Briefly, the cerebral cortex was dissected from 15-day-old fetal mouse brain, and the meninges were carefully removed in a Ca²⁺-free Puck's solution (pH 7.4). Tissues were minced and washed with Ca²⁺-free Puck's solution, followed by treatment with 0.1% trypsin dissolved in Ca²⁺-free Puck's solution at 37°C for 5 min under a stream of 95% O₂, 5% CO₂. The trypsin digestion was terminated by the addition of ice-cold Dulbecco's modified Eagle's minimum essential medium (DMEM) supplemented with 30 mM D-glucose, 10 mM N-tris(hydroxymethyl) methyl-2-aminoethanesulfonic acid, 10 mM HEPES, carbenicillin (0.1 mg/ml), streptomycin (0.1 mg/ml), and 20% horse serum, and tissues were triturated with a Pasteur pipette. The dispersed cells were collected after centrifugation at 900g at 4°C for 2 min. The resultant pellet was resuspended in DMEM followed by trituration, and then the cell suspension was passed through a nylon sieve (mesh size, 60 µm). A portion (1.5 ml) of the cell suspension containing 1.45 × 10⁶ cells/ml was inoculated on a 35-mm glass-bottom dish (MatTek, Ashland, MA). The medium was replaced by freshly prepared DMEM containing 15% fetal calf serum, and cells were successively incubated for 2 days. On the third day, cells were exposed to 20 mM cytosine--D-arabinofuranoside dissolved in DMEM containing 15% horse serum for 24 h to prevent the growth of non-neuronal cells; then the culture medium was replaced by freshly prepared DMEM supplemented with 15% horse serum.

Measurement of [Ca²⁺]_i in Primary Neuronal Culture. The [Ca²⁺]_i was determined as described previously (Oka et al., 1999). Briefly, on the fifth day of in vitro culture, cells were preloaded with 10 µM fura-2/acetoxymethyl ester (AM) in 10 mM HEPES-NaOH (pH 7.05) buffer containing 5.4 mM KCl, 130 mM NaCl, 2.5 mM CaCl₂, and 5.5 mM D-glucose at 37°C for 30 min, followed by further incubation for 30 min to hydrolyze acetoxymethyl ester. The fura-2-loaded cells were imaged on an inverted microscope (IMT-2; Olympus, Tokyo, Japan), using a 40×, 0.6 numerical aperture immersion objective (Olympus) and a charge-coupled device camera (C-2400-08; Hamamatsu Photonics, Shizuoka, Japan) fitted with 500 television line. A 75-W xenon arc lamp was used to provide fluorescence excitation. Ratio images were obtained by acquiring pairs of images at alternate excitation wavelength (340/380 nm) and filtering the emission at 510 nm (OSP-3; Olympus). Images were recorded every 5 s with a video camera (Hamamtsu Photonics) equipped with an Argus-50/CA system (Hamamtsu Photonics), which controlled the image acquisition and display. The fluorescence images of fura-2-loaded neurons were corrected by subtracting the background images measured in a field lacking neuronal cells. The images at individual wavelengths were averaged over 32 frames. Cells were depolarized consecutively with 30, 40, and 50 mM KCl for 15 s at 5-min intervals in the absence or presence of test compounds. [Ca²⁺]_i was monitored in 32 to 51 neurons, the average was calculated in one experiment, and data were obtained from three to four experiments.

Statistical Analysis. Statistical analyses were all performed by using SAS program (SAS/STAT, Version 6; SAS Institute Inc., Cary, NC). Unless otherwise indicated, data (real values) were analyzed for statistical significance by Dunnett's test for multiple comparison. In the set of experiments where the effects of NS-7 and -CTX on KCl-evoked cyclic GMP formation were studied, data were analyzed first by two-way ANOVA followed by Student's t test to compare the data between the control and drug-treated groups at the respective KCl concentration.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effect of NS-7 on Severe and Mild Hypoxic Injury in Slices of Rat Cerebral Cortex. As shown in Fig. 1, when the rat cerebrocortical slices were exposed to oxygen- and glucose-deprived medium for 45 min, followed by reoxygenation for 5 h, a marked LDH leakage was observed (severe hypoxic injury: 574.6 ± 50.4% of that in normoxic slices, mean ± S.E., n = 5). In the case of exposure to the medium devoid of oxygen but containing 3 mM glucose, the LDH leakage was less marked (mild hypoxic injury: 214.5 ± 16.4% of that in normoxic slices, n = 5). NS-7 (3-30 µM) produced a marked and concentration-dependent reduction of LDH leakage caused by severe hypoxic insult. The significant effect was observed at 3 µM and greater (the values were 415.2 ± 48.0% at 3 µM, n = 5, P < .05; 283.3 ± 23.4% at 10 µM, n = 5, P < .01; and 245.8 ± 43.8% at 30 µM, n = 5, P < .01 by Dunnett's test). However, NS-7, even at 30 µM, did not significantly inhibit the mild hypoxic injury.

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Effect of NS-7 on severe (A) and mild hypoxic injury (B) in slices of rat cerebral cortex. After preincubation for 1 h, slices were exposed to hypoxia for 45 min in the absence (A) or presence of 3 mM glucose (B), followed by reoxygenation for 5 h. NS-7 was included during hypoxia and reoxygenation periods. The tissue injury was determined by LDH leakage. The ordinate represents the percentage of LDH leakage of that in normoxic slices. Each column represents the mean ± S.E. of five experiments. *P < .05, **P < .01 compared with the control group (Dunnett's test).

Effects of MK-801, -CTX, and L-NAME on Severe and Mild Hypoxic Injury in Slices of Rat Cerebral Cortex. In contrast to the action of NS-7, MK-801 (1-3 µM) and -CTX (5 µM) significantly inhibited the mild hypoxic injury (Fig. 2B). The values in MK-801 (1 µM)-, MK-801 (3 µM)-, and -CTX-treated groups exposed to mild hypoxic insults were 167.1 ± 7.8% (n = 5, P < .05), 149.8 ± 6.3% (n = 5, P < .01), and 96.0 ± 19.7% (n = 5, P < .01) of that in normoxic slices, respectively, as compared with the control value (245.6 ± 20.1%, n = 5). Both compounds, however, did not significantly affect the LDH leakage caused by severe hypoxic insult (Fig. 2A). The mild and severe hypoxic injury in rat cerebrocortical slices was attenuated by L-NAME in a concentration-dependent manner, although higher concentrations of L-NAME were needed to inhibit the severe injury than those for suppressing the mild hypoxic injury (Fig. 3).

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Effects of MK-801 and -CTX on severe (A) and mild hypoxic injury (B) in slices of rat cerebral cortex. After preincubation for 1 h, slices were exposed to hypoxia for 45 min in the absence (A) or presence of 3 mM glucose (B), followed by reoxygenation for 5 h. -CTX was included during hypoxia and reoxygenation periods, whereas MK-801 treatment was from 15 min before hypoxia to the end of reoxygenation. Each column represents the mean ± S.E. of five experiments. *P < .05, **P < .01 compared with the control group (Dunnett's test).

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Effect of L-NAME on severe (A) and mild hypoxic injury (B) in slices of rat cerebral cortex. After preincubation for 1 h, slices were exposed to hypoxia for 45 min in the absence (A) or presence of 3 mM glucose (B), followed by reoxygenation for 5 h. L-NAME was included during hypoxia and reoxygenation periods. Each column represents the mean ± S.E. of five experiments. *P < .05, **P < .01 compared with the control group (Dunnett's test).

Enhancement of Extracellular Cyclic GMP Formation during Exposure of Rat Cerebrocortical Slices to Severe and Mild Hypoxic Insults: Effects of NS-7, MK-801, and -CTX. The formation of cyclic GMP in the incubation medium containing the soluble fraction of rat cerebellum, 0.5 mM GTP, and 1 mM IBMX was measured as an index of NO synthesis. As shown in Fig. 4, the extracellular cyclic GMP formation was enhanced during hypoxia, although the enhancement was more marked in the absence of glucose (15.9 ± 2.5 pmol/mg protein/45 min, n = 8, in hypoxic slices; and 3.0 ± 0.4 pmol/mg protein/45 min, n = 10, in normoxic slices) than in the presence of 3 mM glucose (Fig. 4B) (5.7 ± 0.4 pmol/mg protein/45 min, n = 7). NS-7 (10 µM) strongly reversed the augmentation of the cyclic GMP formation induced by severe hypoxic insult (Fig. 4A, Experiment I: 534.9 ± 83.2% of normoxic slices, n = 8, in control group and 246.5 ± 48.0%, n = 10, in NS-7-treated group; P < .01) without significant inhibition of that caused by mild hypoxic insult (Fig. 4B, Experiment I: 190.1 ± 14.9%, n = 7, in control group; and 149.2 ± 27.9%, n = 7, in NS-7-treated group). Conversely, both MK-801 (3 µM) and -CTX (5 µM) showed preferential reduction of the cyclic GMP formation caused by mild hypoxic insult (Fig. 4B, Experiment II). The values in control, MK-801 (3 µM)-, and -CTX-treated groups were 190.7 ± 23.0% (n = 5), 96.9 ± 20.7% (n = 5), and 93.4 ± 13.6% (n = 5), respectively. Significant inhibition was observed in the MK-801 (3 µM)-treated group (P < .01) and -CTX-treated groups (P < .01). However, neither MK-801 nor -CTX produced any significant inhibition on the cyclic GMP formation induced by severe hypoxic insult (Fig. 4A, Experiment II).

View larger version (31K): [in this window] [in a new window]   Fig. 4.   Comparative effects of NS-7, MK-801, and -CTX on the enhancement of extracellular cyclic GMP formation induced during exposure of rat cerebrocortical slices to hypoxia in the absence (A) or presence of 3 mM glucose (B). After preincubation for 1 h, an aliquot of crude supernatant fraction of rat cerebellum, 0.5 mM GTP, and 1 mM IBMX were added to the incubation medium, and slices were exposed to hypoxia for 45 min in the absence or presence of glucose. In A, cyclic GMP levels in the normoxia group were 3.0 ± 0.4 pmol/mg protein/45 min in experiment I (mean ± S.E., n = 10) and 8.3 ± 0.6 pmol/mg protein/45 min in experiment II (n = 5), whereas in B, those values were 3.0 ± 0.4 pmol/mg protein/45 min in experiment I (n = 10) and 3.2 ± 0.5 pmol/mg protein/45 min in experiment II (n = 5). Each column represents the mean ± S.E. of 7-10 (experiment I) or 5 (experiment II) determinations. **P < .01 compared with the control group (Dunnett's test).

Depletion of Tissue ATP Content and Increase in Extracellular Glutamate Concentration during Exposure of Rat Cerebrocortical Slices to Hypoxia in the Absence or Presence of 3 mM Glucose. As shown in Fig. 5, the tissue ATP content in slices exposed to hypoxia/glucose deprivation for 45 min was reduced to 5.4% of that in normoxic slices (0.27 ± 0.01 nmol/mg protein, n = 6, in hypoxic slices; and 5.09 ± 0.35 nmol/mg protein, n = 6, in normoxic slices), whereas the value was 14.1% in slices exposed for 45 min to hypoxia with 3 mM glucose (0.72 ± 0.02 nmol/mg protein, n = 6). Significant differences were observed in both mild (P < .01 by Dunnett's test) and severe hypoxic group (P < .01 by Dunnett's test). Moreover, there was a significant (P < .01) difference in the ATP content between the mild and severe hypoxic groups (by Student's t test). However, the extracellular glutamate concentration was markedly elevated during exposure to hypoxia in the absence (46.03 ± 4.18 nmol/mg protein, n = 6; P < .01 by Dunnett's test) or presence of 3 mM glucose (9.20 ± 0.35 nmol/mg protein, n = 6; P < .05 by Dunnett's test) as compared with that in normoxic slices (0.32 ± 0.02 nmol/mg protein, n = 6). In addition, a significant (P < .01) difference in the glutamate concentration was observed between mild hypoxic and severe hypoxic slices (by Student's t test).

View larger version (20K): [in this window] [in a new window]   Fig. 5.   Depletion of tissue ATP (A) and increase in the extracellular glutamate concentration (B) after exposure of rat cerebrocortical slices to hypoxia in the absence or presence of 3 mM glucose. After preincubation for 1 h, slices were incubated at 37°C for 45 min in oxygen-deprived medium in the absence or presence of 3 mM glucose under continuous bubbling with 95% N2, 5% CO2. Each column represents the mean ± S.E. of six experiments. **P < .01 compared with values in the mild hypoxic group (Student's t test).

Comparative Effects of NS-7 and -CTX on Enhancement of Extracellular Cyclic GMP Formation Induced by Various Concentrations of KCl in Slices of Rat Cerebral Cortex. As shown in Fig. 6, KCl (30-50 mM) produced a concentration-dependent increase in the extracellular cyclic GMP formation. NS-7 (10 µM) significantly reduced 50 mM KCl-induced enhancement of the extracellular cyclic GMP formation without affecting the response of 30 to 40 mM KCl (Fig. 6A). The values at 50 mM KCl were 670.1 ± 120.8% of basal values in the control group (n = 6) and 329.0 ± 76.8% in the NS-7-treated group (n = 6; P < .05 by Student's t test), whereas those at 30 mM KCl were 145.0 ± 33.1% in the control group (n = 6) and 127.7 ± 31.8% in the NS-7-treated group (n = 6). In contrast, -CTX (5 µM) almost completely blocked 30 mM KCl-evoked cyclic GMP formation (Fig. 6B). The values at 30 mM KCl were 150.0 ± 15.9% in the control group (n = 6) and 81.2 ± 15.3% in the -CTX-treated group (n = 6; P < .05 by Student's t test). However, it produced only a slight and not significant inhibition of the response of 40 to 50 mM KCl.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Comparative effects of NS-7 (A) and -CTX (B) on KCl-evoked enhancement of extracellular cyclic GMP formation in slices of rat cerebral cortex. After 1 h of preincubation, an aliquot of crude supernatant fraction of rat cerebellum, 0.5 mM GTP, and 1 mM IBMX were added to the incubation medium, and slices were exposed to 5.9 to 50 mM KCl for 45 min in the absence or presence of NS-7 (10 µM) or -CTX (5 µM). The basal levels of cyclic GMP were 10.7 ± 1.6 pmol/mg protein/45 min (mean ± S.E., n = 5) in A and 4.1 ± 0.7 pmol/mg protein/45 min (n = 5-6) in B. Each point represents the mean ± S.E. of five to six experiments. *P < .05 compared with the respective control group (Student's t test).

Comparative Effects of NS-7 and Nifedipine on KCl-Induced Elevation of [Ca²⁺]_i in Primary Neuronal Culture. In this experiment, [Ca²⁺]_i was measured in neurons on the fifth day of culture in which the Ca²⁺ channels involved in the KCl-induced increase in [Ca²⁺]_i are mostly L-type (Oka et al., 1999). Therefore, the effect of NS-7 on [Ca²⁺]_i was compared with that of nifedipine. As shown in Fig. 7, 30 to 50 mM KCl increased [Ca²⁺]_i in a concentration-dependent manner. NS-7 (30 µM) significantly reversed the increase in [Ca²⁺]_i caused by 40 to 50 mM KCl without affecting the basal or KCl (30 mM)-evoked [Ca²⁺]_i. The inhibitory effect of NS-7 was more pronounced in the 50 mM KCl-evoked response. Nifedipine (1 µM), however, produced almost complete inhibition of the increase in [Ca²⁺]_i induced by 30 to 50 mM KCl.

View larger version (14K): [in this window] [in a new window]   Fig. 7.   Effect of NS-7 on KCl-induced increase in [Ca2+]i in primary cultured neurons of mouse cerebral cortex. On the fifth day of in vitro culture, neuronal cells were preloaded with fura-2/AM at 37°C for 30 min. [Ca2+]i was measured from the ratio of fluorescence images at the excitation wavelengths of 340 nm and 380 nm using an Olympus fluorescence microscope and Argus-50/CA system. Cultured neurons were exposed consecutively to 30, 40, and 50 mM KCl for 15 s at an interval of 5 min in the absence or presence of NS-7 (30 µM) or nifedipine (1 µM). [Ca2+]i was monitored in 32-51 neurons, the average was calculated in one experiment, and data were obtained from three to four experiments. Each point represents the mean ± S.E. The basal [Ca2+]i values in the control, NS-7-treated, and nifedipine-treated groups were 49.6 ± 1.7 nM (mean ± S.E., n = 4), 58.3 ± 7.7 nM (n = 3), and 63.8 ± 3.2 nM (n = 3), respectively. **P < .01 compared with the respective control group (Dunnett's test).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In our recent study in which hypoxic injury was determined by LDH leakage in rat cerebrocortical slices after exposure to oxygen- and glucose-deprived medium for 45 min, followed by reoxygenation for 6 h (Tatsumi et al., 1998b), NS-7 (10-30 µM) significantly attenuated the tissue injury, whereas -CTX, at a concentration (5 µM) that shows almost complete blockade of N-type Ca²⁺ channel (Regan et al., 1991), was effective only when the peptide was treated in combination with a P/Q-type Ca²⁺ channel blocker, -Aga. Consistently, NS-7 markedly inhibited the tissue injury caused by the severe hypoxic insult, and the significant effect was observed at 3 µM. However, NS-7, even at 30 µM, had negligible action on the mild hypoxic injury. In contrast, an N-methyl-D-aspartate (NMDA) receptor antagonist, MK-801, and an N-type Ca²⁺ channel blocker, -CTX, preferentially attenuated the mild hypoxic injury with little, if any, protective effect on severe hypoxic damage.

It has been demonstrated that the release of an excitatory neurotransmitter, glutamate, is elevated during cerebral ischemia. This amino acid transmitter stimulates postsynaptic NMDA receptors and -amino-3-hydroxy-5-methyl-4-isoxazole propionate (AMPA)/kainate receptors, which allows Ca²⁺ to enter neurons (Gleason and Spitzer, 1998). The resultant elevation of [Ca²⁺]_i activates a variety of Ca²⁺-dependent enzymes, including NOS, and causes an irreversible neuronal damage (Orrenius et al., 1989; Choi, 1990, 1992; Kristián and Siesjö, 1998).

In this study, the concentration of glutamate in the extracellular fluids was markedly elevated during 45 min of hypoxia and glucose deprivation, whereas in the presence of 3 mM glucose, the value was only one-fifth of that observed in the absence of glucose. Thus, it is likely that higher concentrations of glutamate receptor blockers, such as MK-801, are required to antagonize the actions of glutamate in the severe hypoxic model than are required to block the receptors in the mild hypoxic model.

In this study, L-NAME showed a cerebroprotective action, although the NOS inhibitor was less effective in the severe injury model. Dawson et al. (1991) have shown, in primary neuronal culture, that glutamate or NMDA stimulates NO formation, which is determined by cyclic GMP accumulation, leading to neuronal death. It also has been shown that the cerebral infarction caused by middle cerebral artery occlusion is inhibited by a variety of NOS inhibitors in mice (Nowicki et al., 1991), rats (Ashwal et al., 1993), and cats (Nishikawa et al., 1994). Huang et al. (1994) have demonstrated that the size of cerebral infarction after middle cerebral artery occlusion is smaller in mice genetically devoid of neuronal NOS. Moreover, selective inhibitors of neuronal NOS, such as 7-nitroindazole and ARL 1744, have been demonstrated to decrease the infarct volume after middle cerebral artery occlusion in rats (Dalkara et al., 1994; Yoshida et al., 1994; Zhang et al., 1996). Therefore, it is suggested that the ischemic brain damage or hypoxic injury is mediated, at least in part, by the enhancement of NO synthesis during ischemic or hypoxic condition. In this study, the cyclic GMP formation in the extracellular fluids was measured as an index of NO synthesis after addition of crude extract of rat cerebellum (as the source of guanylate cyclase), GTP, and IBMX to the incubation medium. We already have found that hypoxia/glucose deprivation caused an enhancement of the extracellular cyclic GMP formation, which was reduced by an NOS inhibitor, such as L-NAME, and an intracellular Ca²⁺ chelator, BAPTA/AM, thereby suggesting that the extracellular cyclic GMP formation results from the activation of Ca²⁺-dependent (type I or type II) NOS (Oka et al., 2000), although the origin of the NOS (neuron or glial) is yet unknown. In this study, the exposure to hypoxia/glucose deprivation elicited a more pronounced increase in the extracellular cyclic GMP formation than that measured in the presence of 3 mM glucose. Therefore, it seems likely that the concentration of the NOS inhibitor that is needed to block the NO synthesis and consequently inhibit tissue injury is lower in the mild hypoxic injury model than in the severe model.

Recently, we have found in primary neuronal culture that NS-7 (1-30 µM) inhibits the KCl-evoked NOS activity mediated through L-type and P/Q-type Ca²⁺ channels in which the blockade is almost complete at 30 µM. Moreover, the compound also suppresses the NOS activation induced by high concentrations of Bay K 8644 (1-10 µM), a Ca²⁺ channel agonist; however, it has little, if any, effect on the spontaneous activity or the responses of low concentrations (0.01-0.1 µM) of Bay K 8644 (Oka et al., 1999). Moreover, NS-7, unlike other Na⁺ and Ca²⁺ channel blockers, reduces the KCl (80 mM)-evoked dopamine release from rat striatum, measured by an intracerebral microdialysis, without changing the spontaneous or evoked release induced by 25 to 50 mM KCl (Itoh et al., 1998). Also, in this study, NS-7 inhibited the KCl-induced increase in [Ca²⁺]_i in primary neuronal culture in a manner dependent on the extent of KCl depolarization. Taken together, it is suggested that the blockade of Ca²⁺ channel by NS-7 is dependent on the channel activity in which the compound blocks preferentially the activated Ca²⁺ channel.

During cerebral ischemia, depletion of ATP causes the dysfunction of Na⁺/K⁺-ATPase and leads to the elevation of intracellular Na⁺ concentration, which in turn depolarizes neuronal membranes and subsequent activation of voltage-gated Na⁺ and Ca²⁺ channels. Thus, the depolarization of neuronal membranes may be more marked when the hypoxic insults are severe. The extent of ATP depletion is one of the determinants of the severity of ischemic insults. In this study, tissue ATP content in slices exposed to severe hypoxic condition (hypoxia/glucose deprivation) was reduced to 5.4% of that in normoxic slices, whereas the value was 14.1% in slices exposed to mild hypoxic condition (hypoxia/3 mM glucose). Therefore, it is probable that NS-7 blocks more effectively the highly activated Ca²⁺ channel during severe ischemic condition with little, if any, blocking action on the channel under nonischemic or mild ischemic condition. Interestingly, the extent of cyclic GMP formation during exposure to mild and severe hypoxic insult was similar to that induced by 30 to 40 mM KCl and 50 mM KCl, respectively. This may indicate that higher numbers of Ca²⁺ channels are activated during severe hypoxic insult. Under such a severe hypoxic condition, most of the Ca²⁺ channels may be in the activated and/or inactivated state. Several L-type Ca²⁺ channel blockers, such as verapamil and amlodipine, have been demonstrated to show higher affinity for the activated or inactivated state of Ca²⁺ channels and produce voltage-dependent and use-dependent block of multiple high-voltage activated Ca²⁺ channels, including L-type, N-type, and P/Q-type Ca²⁺ channels (Cai et al., 1997; Furukawa et al., 1997). It also has been demonstrated that the sites of action of these Ca²⁺ channel blockers for producing the state-dependent block are located within the sixth transmembrane segment (S6) in domain IV of the Ca²⁺ channel moiety (Hering et al., 1996). Taken together, the site of Ca²⁺ channel blockade by NS-7 may be near the binding sites of these Ca²⁺ channel blockers.

In conclusion, both severe and mild hypoxic injury were induced in rat cerebrocortical slices by an exposure to hypoxia for 45 min in the absence and presence of 3 mM glucose, respectively, followed by reoxygenation for 5 h. The extracellular cyclic GMP formation, a marker of NO synthesis, was enhanced by the hypoxic insult, although the enhancement was more marked in the absence of glucose than in the presence of 3 mM glucose. NS-7, unlike MK-801, -CTX, and L-NAME, preferentially suppressed the tissue injury as well as the enhancement of the cyclic GMP formation caused by severe hypoxic insult. In slices exposed to 30 to 50 mM KCl for 45 min, a concentration-dependent increase in the extracellular cyclic GMP formation was observed. NS-7 blocked the enhancement of the cyclic GMP formation induced by 50 mM KCl but not by 30 to 40 mM KCl. In primary neuronal culture, NS-7 reversed KCl-induced elevation of [Ca²⁺]_i in which the inhibition was marked when the KCl concentration was increased. These findings indicate that the cerebroprotective action of a novel Na⁺/Ca²⁺ channel blocker, NS-7, is largely dependent on the severity of the hypoxic insult in which the effect was more pronounced in severe injury. The highly voltage-dependent blockade of Ca²⁺ channels may participate in such a preferential action of NS-7.