Protection by diphenyliodonium against glutamate neurotoxicity due to blocking ofN-methyl-d -aspartate receptors

~ ) Pergamon

Neuroscience Vol. 76, No. 2, pp. 459 466, 1997 Copyright :~: 1996 IBRO. Published by ElsevierScience Ltd Printed in Great Britain PII: S0306-4522(96)00375-2 03064522/97 $17.00+0.00

PROTECTION BY DIPHENYLIODONIUM AGAINST GLUTAMATE NEUROTOXICITY DUE TO BLOCKING OF N-METHYL-D-ASPARTATE RECEPTORS Y. N A K A M U R A , * ¶ K. T S U J I , * ? M. S H U T O , ++ K. OGITA,:~ Y. Y O N E D A , ++ K. S H I M A M O T O , § T. S H I B A T A t and K. K A T A O K A * *Department of Physiology, Ehime University School of Medicine, Shigenobu, Ehime, 791-02, Japan CDepartment of Orthopaedic Surgery, Ehime University School of Medicine, Shigenobu, Ehime, 791-02, Japan {Department of Pharmacology, Setsunan University, Hirakata, Osaka, Japan §Suntory Institute for Bioorganic Research, Osaka, Japan Abstract--The protective effect of diphenyliodonium, known as an inhibitor of flavin enzymes including nitric oxide synthases, was examined against the neurotoxicity of excitatory amino acids on cultured spinal neurons of the rat. Diphenyliodonium reduced the neuronal damage induced by 15-rain exposure to glutamate or N-methyl-D-aspartate in a dose-dependent manner; half effective concentrations (Ecs0) were about 3 gM for both. Protection was only observed when diphenyliodonium was added into the exposure medium. Diphenyliodonium showed no effect on the toxicity induced by 24 h exposure to non-N-methylD-aspartate receptor agonists. Using a microfluorometry technique with Fura 2, we observed that diphenyliodonium reversibly inhibited the N-methyI-D-aspartate-evoked intracellular Ca 2+ elevation. The amount of 45Ca2+ influx induced by N-methyl-D-aspartate was also inhibited by diphenyliodonium in a dose-dependent manner; ECsowas about 3 gM. Furthermore, we examined the effect of diphenyliodonium on an opening activity of the Nomethyl-D-aspartate receptors estimated by binding of dizocilpine maleate to membrane fractions from whole brain of adult rat and from cultured spinal neurons. Diphenyliodonium inhibited the binding of dizocilpine maleate dose-dependently; Ecs0 was 5-8 gM. These results suggest that diphenyliodonium is a new antagonist to the N-methyl-l)-aspartate receptors and that diphenyliodonium protects neurons against glutamate toxicity due to a direct blocking of the Ca 2+ influx. This conclusion is supported by the similarity of the stereochemical structures predicted by computer between diphenyliodonium and dizocilpine maleate. Copyright ',~ 1996 IBRO. Published by Elsevier Science Ltd. Key words: diphenyleneiodonium, dizocilpine maleate (MK-801), NMDA receptor antagonist, NOS

inhibitor, Ca > infux, Ca 2+ microfluorometry.

A growing body of evidence indicates that the neurotoxicity of the excitatory transmitter glutamate contributes to the brain damages associated with various acute insults, including hypoxia, ischemia, hypoglycemia, prolonged seizures and trauma. 3"22'23'27 These insults all disturb cellular homeostasis, leading to an elevation of extracellular glutamate level] '17 which is large enough to destroy vulnerable neurons in the brain. 5 It is also well known that a brief exposure to glutamate induces cell death in cultured neurons. 4"s 10,25.28

~To whom correspondence should be addressed. Abbreviations: AMPA, (RS)-c~-amino-3-hydroxy-5-methyl4-isoxazolepropionate; HEPES, N-2-hydroxyethylpiperazine-N'-ethanesulfonic acid; HKR, HEPESbuffered Krebs Ringer solution; LDH, lactate dehydrogenase; MK-801, dizocilpine maleate; NADH, reduced nicotinamide adenine dinucleotide; NADPH, reduced nicotinamide adenine dinucleotide phosphate; NO, nitric oxide; NOS, nitric oxide synthase. 459

Recently, it has been reported that N-methyl-Daspartate ( N M D A ) neurotoxicity is attenuated by inhibitors for nitric oxide synthase (NOS) in primary cell cultures of the b r a i n ] Nitric oxide (NO) donors such as sodium nitroprusside induced neuronal cell death and N O S inhibitors such as N~-nitro c-arginine and diphenyleneiodonium protected the neurons against N M D A - i n d u c e d cell death. On the basis of these and other results, it was proposed that N M D A - i n d u c e d neuronal death is mediated by NO, which is produced by NOS in N A D P H diaphorase-positive neurons. Thereafter, a number of studies have reported that N O plays either harmful or beneficial roles in neuronal viability in various experimental conditions. ~3 Inhibitors for NOS do not always protect neurons against toxicity of excitatory amino acids. 12A<24'26 In order to clarify whether NO causes the neuronal damage after glutamate exposure, we examined the effect of NOS inhibitors using neurons cultured from spinal cords of rat embryos.

Y. Nakamura et al.

460 EXPERIMENTAL PROCEDURES

Neuron culture Spinal neurons were prepared as described previously. 19"32 Spinal cords of 14-day embryos of Wistar rats (Clea Japan, Tokyo, Japan) were used. Briefly, after trypsinization and trituration, the dissociated cells were cultured on polyethyleneimine-coated plastic coverslips at a density of 3 ~ x 105 cells/ll x 11 mm 2 with Dulbecco's modified Eagle medium (Gibco) containing 5% fetal calf serum and 10% horse serum. After 24 h, the medium was replaced with a serum-free medium which consisted of Dulbecco's modified Eagle medium and supplements.31 On the fourth cultured day, the cells were exposed to 20 ~tM cytosine arabinoside for 24 h. Cells were cultured totally for 12 days with changing medium every other day, and then used for experiments. Approximately 70% of the cells were immunoreactively positive to a neuronal marker, microtubule-associated protein 2.

Neurotoxicity assay For the glutamate-induced neurotoxicity assay, the neurons on coverslips were preincubated with HEPES-buffered Krebs Ringer solution (HKR; composition in mM: NaC1, 130; KCI, 5.4; MgSO4, 0.8; NaH2PO 4, 1; HEPES, 50; CaCI2, 1.8; glucose, 5.5; pH was adjusted to 7.4 with NaOH) in 24-well plates for 30 min. The neurons on coverslips were then exposed for 15 rain to glutamate in HKR, returned to serum-free medium and postincubated for 24 h. The cells on the coverslips were taken out and sonicated in 0.15 ml of 0.2 M HC1. Aliquots were neutralized with NaOH and used for the assay of cell protein by the method of Bradford. 2 The postincubation media were subjected to a lactate dehydrogenase (LDH) release assay as described below. To examine the neurotoxicity of NMDA, we used Mg2+-free HKR instead of normal HKR. In order to assess the neurotoxicity of non-NMDA receptor agonists, kainate or (RS)-u-amino-3-hydroxy-5methyl-4-isoxazolepropionate (AMPA), the neurons were exposed for 24 h to either of the agonists dissolved in serum-free medium and the incubation medium was subjected to the LDH release assay, since exposure of the neurons to these agonists for the short period of 15 min induces little neuronal damage. 3~ In this case, we added dizocilpine maleate (MK-801) to the incubation medium to eliminate possible damage mediated by NMDA receptors. The cell damage was estimated by the measurement of the activity of LDH released into the medium from degenerated cells, as reported by Koh and Choi.15 Usually, 100 gl of the postincubation medium was added to a 100 gl mixture of 0.2 mM NADH and 0.2 M potassium phosphate at pH 7.5. The LDH activity was determined as the difference of oxidation rate of NADH, monitored by absorbance at 340 nm, between the presence and the absence of 1 mM sodium pyruvate using a microplate reader MTP-32 (Corona Electric, Japan).

The extent of Ca 2÷ influx was measured using 45Ca. The neurons on coverslips were preincubated with MgZ+-free HKR for 30 min, then exposed to NMDA in the presence of 1.8 mM 45CAC12 (about 3 gCi/ml). After 5 min, the cells on coverslip were washed three times with 1 ml ice-cold HKR containing 200 gM LaCI 3 and then dissolved in 0.3 ml of 0.15 M NaOH. The amounts of 45Ca2+ in the solution were determined by a scintillation counter.

[3H]Dizocilpine maleate bhMing assay [3H]MK-801 binding assay was carried out as described previously. 34 In brief, crude synaptic membrane fractions obtained from the whole brains of male Wistar rats weighing 200-250 g were washed with 40 vols of 50 mM Trisacetate buffer (pH 7.4) by centrifugation at 50,000g for 20 min, and then the resultant pellets were suspended in 8 vols of buffer followed by storage at -80°C until use. Spinal neurons cultured in a 90-ram dish were collected with a policeman and a centrifugation, then the cells were frozen. The thawed samples were homogenized in 40 vols of buffer using Physcotron at setting no. 6 for 1 min at 4°C. The homogenates were then centrifuged, suspended and stored as described above. Aliquots of these membrane preparations (about 0.15 mg protein/assay) were incubated with 5 nM [3H]MK-801 in 0.5 ml of 50 mM Tris-acetate buffer (pH 7.4) at 30°C for 30 min in either the presence or absence of at least four different concentrations of diphenyliodonium from 10 nM to 0.1 mM. Incubation was terminated by the addition of 3 ml ice-cold buffer and subsequent filtration through a Whatman GF/B glass fiber filter under constant vacuum of 15mmHg. The filter was rinsed four times with 3 ml buffer within 10 s, and radioactivity retained on the filter was measured by a liquid scintillation counter. Assays were carried out in triplicate with variations of less than 10%. Non-specific binding was defined by the addition of both 2-amino-5-phosphonovalerate and 7-chlorokynurenate at 0.1 mM, and accounted for 15 20% of the total binding determined in their absence. Under these conditions, binding reached complete equilibrium within 30 rain in the absence of any added agonists. Since these membrane preparations contained endogenous agonists such as L-glutamate and glycine at concentrations sufficient to maximally saturate each recognition domain, the addition of glutamate and glycine did not potentiate binding at 0.1 mM. 2~ However, spermidine at 1 mM almost doubled binding. Therefore, binding was determined in either the presence or absence of added spermidine at 1 mM in this study.

Materials' Diphenyliodonium chloride and diphenyleneiodonium chloride were purchased from Aldrich Chemical Company Inc. and Alexis Corp. (Lfiufelfingen, Switzerland) through Wako Chemicals (Tokyo, Japan), respectively. They were dissolved in dimethylsulfoxide, whose concentration in test solutions was kept constant at 1% (v/v).

Measurements of [Ca2+]i and Ca 2+ influx Intracellular Ca 2÷ concentrations ([Ca2+]i) were measured as described previously.2° Briefly, the cells cultured on a glass coverslip were incubated with 5 laM Fura-2 acetoxymethyl ester (Dojin, Japan) dissolved in Mg2+-free HKR at 37°C for 30 min. After washing with Mg2+-free HKR, the Fura-2-1oaded cells were placed on the stage of an inverted fluorescence microscope (Olympus IMT-2, x 20 objective; DApo20UV/340) and were perfused in thermostatted (37°C) Mg2+-free HKR at a rate of 300 gl/min. Drugs were applied into the perfusion medium. Fluorescence with wavelength greater than 450 nm was monitored by an SIT camera (C2741 Hamamatsu, Japan) with dual excitation wavelengths at 340 and 360 nm, and the [Ca2+]~ in each cell was calculated using an FC-300 image processor (Furusawa Laboratory, Japan).

RESULTS

Neuroprotective effect oJ diphenyliodonium To evaluate neurotoxicity, the neurons were exposed to glutamate for 15 min and then returned to the n o r m a l serum-free medium, and we measured the activity o f L D H released into the m e d i u m during 24 h postincubation. The exposure to 300 g M glutamate induced m a r k e d L D H release from 33 :k I nmol/ min/mg (n=3) b a c k g r o u n d level to 167 + 9 nmol/min/ m g (n=3). W h e n the neurons were exposed to glutamate together with various concentrations o f

Diphenyliodonium protects against glutamate neurotoxicity E" .~ 200 -

~

these and the other receptors, we used mainly N M D A in the following studies in order to simplify the interpretation of the results. To examine the mechanism of protection by diphenyliodonium against NMDA-induced neuronal damage, we first changed the application periods of diphenyliodonium. We measured the LDH release induced by 1 mM NMDA exposure in the following three different methods of diphenyliodonium (10/aM) treatment: diphenyliodonium was added into the preincubation medium, exposure medium and postincubation medium (Table 2). Only when diphenyliodonium was added to the exposure medium was a significant protection observed. The addition of diphenyliodonium to the pre- or postincubation media afforded little protection from neuronal damage. Although the reason for this is not clear, the addition of diphenyliodonium to preincubation medium tended to slightly enhance neuronal damage. Diphenyliodonium (10/aM) by itself did not induce significant cell damage in each period of application. We also examined the effect of diphenyleneiodonium on NMDA neurotoxicity. The LDH release was not reduced by diphenyleneiodonium but increased further in a dose-dependent manner (data not shown). Even in the absence of NMDA, diphenyleneiodonium alone induced neurotoxicity significantly and the LDH release increased dosedependently (data not shown).

Glu

CI.

0")

150

.__ ___E 100 O

E

m c3 ..-,I

50

_O T I 0 0

El

'

'

' ' ' ' ' [

,

,

,

,,,,,l

1 DI conc. (pM)

r

10

Fig. 1. Protective effects of diphenyliodonium against the neuronal damage induced by glutamate and NMDA in cultured spinal neurons. The neurons were exposed for 15 min to 300 ~tM glutamate (open circles) and 1 mM NMDA (open triangles) combined with various concentrations of diphenyliodonium, and then returned to serumfree medium. Twenty-four hours later, the activity of LDH released in the medium was measured. Control values in the treatment without glutamate and NMDA in the sister culture are indicated with closed circle and closed triangle, respectively. Data shown are means 4- S.D. of triplicate measurements. Similar results were obtained in three other experiments. diphenyliodonium, the extent of the LDH release decreased in a dose-dependent manner; the half effective concentration (ECso) was about 3/aM (Fig. 1). NMDA-induced neuronal damage was also tested. The exposure to 1 mM N M D A for 15 rain increased the LDH release from 55±8 nmol/min/mg (n--3) background level to 1954-10nmol/min/mg (n=3). Diphenyliodonium protected the neuronal damage induced by 1 mM N M D A in almost the same dose dependence as that induced by 300/aM glutamate; ECso was about 3/aM (Fig. 1). We also examined the effects of diphenyliodonium on the neurotoxicity induced by the n o n - N M D A receptor agonists, kainate or AMPA, in which the neurons were exposed to these agonists for 24 h with 2 gM MK-801 to eliminate possible damage mediated by the NMDA receptors (see Experimental Procedures). Discordantly with NMDA, 10/aM diphenyliodonium did not show a protective effect on the neurotoxicity induced by exposure to 300 gM kainate or to 100 gM AMPA (Table 1). Since glutamate is an agonist common for

Blockade of Ca 2+ influx by diphenyliodonium To clarify further the mechanisms of protection by diphenyliodonium, we examined the effects of diphenyliodonium on NMDA-induced [Ca:+]i elevation using the method of microfluorometry, with Fura-2 as a [Ca2+]i indicator. A typical result is shown in Fig. 2A. When 25 gM NMDA was applied for 30 s, [Ca2+]± was elevated from background level (96+34 nM, n=22) to a peak value (244± 107 nM, n=22). When 10/aM diphenyliodonium was applied together with 25 jaM NMDA, the extent of [Ca2+]~ elevation decreased to about one-third (140 ± 59 nM, n=22). The effect of diphenyliodonium was reversible; the extent of [Ca2+]~ elevation was returned to the control level after washing out of diphenyliodonium (221 ± 67 nM, n=22). There was no significant

Table 1. Effects of diphenyliodonium on the neurotomcity induced by kainate and AMPA LDH release (nmol/min/mg)

Control 300 gM kainate 100 pM AMPA

461

0 pM diphenyliodonium

10 gM diphenyliodonium

40.64- 3.5 313.44- 17.4 183.44-4.1

53.7 4-5.4 324.5 ± 10.4 231.1 ± 8.5

The neurons were exposed to 300gM kainate and 1001aM AMPA for 24h in the absence or presence of 10 gM diphenyliodonium. Activity of LDH released in the exposure medium was measured. Data shown are mean ± S.D. of triplicate measurements. Similar results were obtained in two other experiments.

462

Y. Nakamura et al. Table 2. Protective effect of diphenyliodonium with different application periods against N-methylD-aspartate-induced neurotoxicity LDH release (nmol/min/mg) 0 mM NMDA

1 mM NMDA

36.0 -4-3.4

141.0 ± 10.4

29.5 ± 5.4 30.1 ± 5.6 51.3 ± 6.3

190.2 ± 10.8 66.8 ± 3.8" 131.4:1- 8.4

Control l0 llM diphenyliodonium in: preincubation medium exposure medium postincubation medium

The neurons were exposed to l mM NMDA for 15 min, then returned to serum-free medium. Activity of LDH released in the postincubation medium was measured. Diphenyliodonium (1011M) was added to either preincubation medium, exposure medium or postincubation medium. Data shown are means+S.D, of quadruplicate measurements. Similar results were obtained in two other experiments. Asterisk indicates a statistically significant difference compared with the value with NMDA and without diphenyliodonium (one-way ANOVA followed by Scheffe's multiple comparison procedure, *P<0.01).

A [:3 NMDA

~" 150

~

NMDA

~

NMDA

C v

2 o

100

._c O r-

x:

o

50

~

0

I

;

I

4

I

I

I

8

Time (rain)

B IC

5O

£ Q.

40 tO En

3o X

-~ 20 o kO

II

0T 0

,

i

i

tr

trl

i

i

1

i

ii

till

10

Dt (~M) Fig. 2. Effects of diphenyliodonium on the [Ca2+]i elevation (A) and 4SCa2+ influx (B) induced by NMDA exposure in cultured spinal neurons. (A) Datum indicates a typical recording of (Ca2+)i change measured with Fura-2. The open bars indicate the application period (30 s) of 25 FM NMDA alone. The hatched bar indicates the application period of 25 laM NMDA together with 10 laM diphenyliodonium. (B) The neurons were incubated for 5 min with 1.8 mM 45Ca2+ in the presence of 1 mM NMDA and various concentrations of diphenyliodonium (open circles). 45Ca2+ influx was measured after washing with LaCl3-containing solution (see details in the text). The value of the treatment without NMDA is indicated with a closed circle. Data shown are means 4-S.D. of a representative experiment performed in triplicate. Similar data were obtained in two other experiments.

Diphenyliodonium protects against glutamate neurotoxicity o 120 E

o

o 100

,+_ 0

8o c

~5 60 cUz 40 ,2 7-

SPD(+)

20

-~_ l

0 0

II . . . . . . . . . . . . . . . . . . . . . . . . 0.01 0.1 1 DI c0nc. ~ M )

.-r

10

100

Fig. 3. Blocking effects of diphenyliodonium on opening activity of the NMDA receptor, which was estimated by binding of [3H]MK-801 to the membrane fractions from whole brains of adults rats. Aliquots of Triton-treated membrane fractions were incubated with [3H]MK-801 in buffer containing diphenyliodonium at concentrations from 10nM to 0.1 mM in either the absence (open circles) or presence (closed circles) of 1 mM spermidine (SPD). Data are meaniS.E.M, values from four independent determinations.

difference between the peaks of [Ca2+]i induced by the first and third applications of N M D A ; however, the second peak with diphenyliodonium was significantly lower than the first and third peaks (P<0.01, one-way A N O V A followed by Scheffe's multiple comparison procedure). Similar data were obtained in two other experiments. We also examined the effects of diphenyliodonium on NMDA-induced 45Ca 2+ influx. In the absence of N M D A , the amount of 4 5 C a 2+ in the cells increased from the background level of 9.84-0.8 nmol/mg to 2 3 . 3 i 1.8 nmol/mg (n=3) during 5 min incubation with 45CaZ+. In the presence of 1 mM N M D A , the amount of 45Ca 2+ influx increased to 43.8-4-2.6 nmol/ mg (n=3). As shown in Fig. 2B, this enhancement was completely blocked by diphenyliodonium in a dose-dependent manner; ECso was about 3 gM.

Diphenyliodonium's binding

effects on dizocilpine maleate

To clarify whether diphenyliodonium acts directly on N M D A receptors, we examined the effects of diphenyliodonium on an opening activity of the Ca 2+ channel in N M D A receptor molecules that was estimated by binding capacities of MK-801, an agent known as a specific and potent open channel blocker of N M D A receptors. 33 In the synaptic membrane fraction from whole brains of adult rats, diphenyliodonium inhibited the [3H]MK-801 binding in a dose-dependent manner, and 100gM diphenyliodonium completely blocked [3H]MK-801 binding (Fig. 3). The half inhibitory concentration (ICso) of diphenyliodonium was 5.8gM in this membrane fraction; the Hill coefficient value was about 0.8 in the absence of spermidine (Table 3). The direct effect

463

of diphenyleneiodonium on N M D A receptors seems to be much less potent than that of diphenyliodonium, since 10 gM diphenyleneiodonium showed almost no effect on [3H]MK-801 binding (data not shown). In the presence of spermidine, [3H]MK-801 binding was similarly inhibited by diphenyliodonium in a dose-dependent manner; the lCso was 7.0 gM and the Hill coefficient was about 1.5. We also examined the effect of diphenyliodonium using the membrane fraction from cultured spinal neurons. Although the extents of [3H]MK-801 binding per protein in this membrane fraction were about one-tenth of those from whole brains, diphenyliodonium inhibited similarly the [3H]MK-801 binding in cultured spinal neurons in a dose-dependent manner; lCso values were 8.0 and 5.1 gM, respectively, in the absence and presence of spermidine, and Hill coefficients were also around one (Table 3). DISCUSSION

In the present study, we found that diphenyliodonium, which is known as an inhibitor of flavin enzymes, protects cultured spinal neurons against damage induced by glutamate and NMDA, but not by non-NMDA receptor agonists. We clearly demonstrated that diphenyliodonium inhibited N M D A evoked Ca 2+ influx by two different methods using 45Ca2+ and microfluorometry. Furthermore, we observed that diphenyliodonium decreased the extent of [3H]MK-801 binding on N M D A receptors in membrane fractions from cultured spinal neurons and also from whole brain. All these effects of diphenyliodonium were observed in an identical concentration range. These results thus strongly suggest that the mechanism of neuroprotection by diphenyliodonium is due to blocking Ca 2+ influx by its direct effect on N M D A receptors. Molecules of diphenyliodonium and diphenyleneiodonium are composed of two benzene rings with a positively charged iodine in between (see upper part of Fig. 4). This structure is basically similar to that of MK-801, in which the positive charge is an imino group located in the center of the middle cycloheptene ring. Interestingly, in the stereochemical structure (lower part of Fig. 4), the most stable conformation of diphenyliodonium shows a 100 ° angle between two benzene rings. This angle is quite similar to that in MK-801. On the other hand, the most stable conformation of diphenyleneiodonium shows a planar structure. Therefore, it is thus highly likely that diphenyliodonium affects N M D A receptors at the same site as does MK-801. Further studies are under way to elucidate whether diphenyliodonium inhibits MK-801 binding in a competitive manner and whether diphenyliodonium affects the other domains, such as glutamate and polyamine. NO has recently been recognized as an essential factor for controlling intracellular signaling mechanisms, even in the nervous system. NO should also be

464

Y. Nakamura et al. Table 3. Effects of diphenyliodonium on [3H]dizocilpine maleate binding of N-methyl-D-aspartate receptors [3H]MK-801 binding Synaptic membrane of whole brains Sperm±dine

Cultured spinal neurons Sperm±dine

fmol/mg

Ics0 (gM)

Hill coefficient

0mM lmM

356±5 690±9

5.8t0.6 7.0±0.5

0.8±0.1 1.5±0.2

0mM lmM

29.8±2.1 84.5±4.0

8.0±1.4 5.1±0.7

1.6±0.6 1.l±0.2

Aliquots of the membrane fractions of whole brains of adult rats and cultured spinal neurons were incubated with 5 nM [3H]MK-801 in buffer containing four different concentrations of diphenyliodonium, ranging from 1 ~tM to 0.1 mM, in either the absence or presence of 1 mM sperm±dine. Data are mean ± S.E.M. values from four independent determinations.

OH 3


Diphenyleneiodonium (DPI) MK-801 Fig. 4. Chemical structures of diphenyliodonium (DI), MK-801 and diphenyleneiodonium (DPI). Conventional chemical structures are shown in the upper part. In the lower part, two directional views of the stereochemical structures of the compounds are shown, which were calculated using the computer program CAChe MOPAC (calculation type, Optimized Search; optimization method, Eigenvector Following; parameters, PM3). Diphenyliodonium (DI)

involved in the mechanism of neuronal death induced by excitatory amino acids; however, there are several controversial reports on the effect of NO, whether it is harmful or beneficial to neurons (see Ref. 13). In our cultured spinal neurons, contrary to the observations of Dawson e t al., 7 diphenyleneiodonium showed no protective effect on N M D A - i n d u c e d neuronal damage, but it was toxic. Another NOS inhibitor, NG-nitro-h-arginine (100 ~tM), also failed to protect spinal neurons (data not shown). The reason for the discrepancy is not clear; however, our spinal neurons may contain much less or practically no NOS. The difference may be due to the difference in culture conditions, and/or to the tissue sources from the cerebral cortex or spinal cord.

Diphenyliodonium and diphenyleneiodonium are both known as inhibitors of flavin enzymes, 6,1L~4'3° some of which are essential for general cell metabolism. Therefore, a high concentration of these drugs should be cytotoxic. In our spinal neurons, diphenyleneiodonium seems to be harmful, which is possibly due to such general cytotoxicity. Recently, Hewett e t al. ~3 reported observations suggesting that astrocytes and microglia play roles in the N O mediation mechanism of N M D A neurotoxicity. They showed that addition of ant±oxidant enzymes (superoxide dismutase/catarase) prevented the potentiation of N M D A neurotoxicity seen following astrocytic inducible NOS, suggesting that the toxicity of N O resulted from interaction with oxygen

Diphenyliodonium protects against glutamate neurotoxicity

radicals, producing a more toxic compound, such as peroxynitrite, which could enhance neuronal damage. Diphenyleneiodonium and diphenyliodonium are also known as inhibitors of superoxide-generating N A D P H oxidase by macrophages ~ and neutrophils, 6 in which flavin adenine dinucleotide and cytochrome b mediate the electron flow from NADPH to 02. It is possible, therefore, that diphenyliodonium and diphenyleneiodonium protect neurotoxicity through inhibition of superoxide generation from unavoidably contaminating microglia or brain macrophages. However, this is unlikely to be a major reason for the neuronal protection by diphenyliodonium in the present study, because the potency of diphenyliodonium is reported to be 100 times less than that of diphenyleneiodonium in inhibition of superoxide production. 6'~ Microglial proliferation and gliosis of astrocytes are seen after an insult of transient ischemia in vivo. 18.29 In addition to the well-known excitotoxicity of glutamate, functions of glial cells, such as production of NO and superoxide by astrocytes and microgila, are inevitably involved in the mechanisms of

465

ischemia-induced neuronal death. Diphenyliodonium has three potential actions of blocking Ca 2+ entry, through N M D A channels, inhibition of NO synthesis and inhibition of superoxide generation. We are now in the process of examining the possible protection by

its triple action against neuronal damage induced by ischemia in vivo. CONCLUSIONS

Diphenyliodonium showed a protective effect against glutamate neurotoxicity on cultured spinal neurons. Diphenyliodonium is known as an inhibitor of flavin enzymes; however, it seems to act as an antagonist for N M D A receptors and to block Ca 2+ influx directly. This conclusion is supported by the similarity of the stereochemical structures predicted by computer between diphenyliodonium and MK-801. A c k n o w l e d g e m e n t ~ T h i s investigation was supported in

part by Grant-in-Aids for Scientific Research, 06680760 to Y.N. and 06454700 to K.K., from the Ministry of Education, Science and Culture of Japan.

REFERENCES

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. t8. 19. 20.

Benveniste H., Drejer J., Schousboe A. and Diemer N. H. (1984) Elevation of the extracellular concentrations of glutamate and aspartate in rat hippocampus during transient cerebral ischemia monitored by intracerebral microdialysis. J. Neurochem. 43, 1369 1374. Bradford M. M. (1976) A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Analyt. Biochem. 72, 248 254. Choi D. W. (1988) Glutamate neurotoxicity and diseases of the nervous system. Neuron 1,623 634. Choi D. W., Maulucci-Gedde M. and Kriegstein A. R. (1987) Glutamate neurotoxicity in cortical cell culture. J. Neurosci. 7, 357 368. Choi D. W. and Rothman S. M. (1990) The role of glutamate neurotoxicity in hypoxic ischemic neuronal death. A. Rev. Neurosci. 13, 171 182. Cross A. R. and Jones O. T. G. (1986) The effect of the inhibitor diphenyleneiodonium on the superoxide generating system of neutrophils: specific labeling of a component polypeptide of the oxidase. Biochem. J. 237, 111-116. Dawson V. L., Dawson T. M., Bartley D. A., Uhi G. R. and Snyder S. H. (1993) Mechanisms of nitric oxide-mediated neurotoxicity in primary brain cultures. J. Neurosci. 13, 2651 2661. Finkbeiner S. and Stevens C. F. (1988) Applications of quantitative measurements for assessing glutamate neurotoxicity. Proc. natn. Acad. Sci. U.S.A. 85, 40714074. Frandsen A. and Schousboe A. (1987) Time and concentration dependency of the toxicity of excitatory amino acids on the cerebral neurons in primary culture. Neurochem. Int. 10, 583 591. Frandsen A. and Schousboe A. (1990) Development of excitatory amino acid induced cytotoxicity in cultured neurons. hzt. J. devl Neurosci. 8, 209 216. Hancock J. T. and Jones O. T. G. (1987) The inhibition by diphenyleneiodonium and its analogues of superoxide generation by macrophages. Biochem. J. 242, 103-107. Hewett S. J., Corbett J. A., McDaniel M. L. and Choi D. W. (1993) Inhibition of nitric oxide formation does not protect murine cortical cell cultures from N-methyl-D-aspartate neurotoxicity. Brain Res. 625, 337-341. Hewett S. J., Csernansky C. A. and Choi D. W. (1994) Selective potentiation of NMDA-induced neuronal injury following induction of astrocytic iNOS. Neuron 13, 487494. Holland P. C., Clark M. G,, Bloxham D. P. and Lardy H. A. (1973) Mechanism of action of the hypoglycemic agent diphenyleneiodonium. J. biol. Chem. 248, 6050-6056. Koh J. Y. and Choi D. W. (1987) Quantitative determination of glutamate mediated cortical neuronal injury in cell culture by lactate dehydrogenase efflux assay. J. Neurosci. Meth. 20, 83-90. Lernner-Natoli M., Rondouin G., Debock F. and Bockaert J. (1992) Chronic NO synthase inhibition fails to protect hippocampal neurons against NMDA toxicity. NeuroReport 3, 1109-1112. Mitani A., Kubo H., lga K., Imon H., Kadoya F. and Kataoka K. (1990) A new enzymatic cycling technique for glutamate determination in brain microdialysates. J. Neuroehem. 54, 709 711. Morioka T., Kalehua A. N. and Streit W. J. (1991) The microglial reaction in the rat dorsal hippocampus following transient forebrain ischemia. J. eerebr. Blood Flow Metab. 11,966-973. Ogata T., Nakamura Y., Tsuji K., Shibata T. and Kataoka K. (1993) Steroid hormones protect spinal cord neurons from glutamate toxicity. Neuroseience 55, 445449. Ogata T., Nakamura Y., Tsuji K., Shibata T., Kataoka K. and Schubert P. (1994) Adenosine enhances intracellular Ca 2+ mobilization in conjunction with metabotropic glutamate receptor activation by t-ACPD in cultured hippocampal astrocytes. Neurosei. Lett. 170, 5-8.

466 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34.

Y. Nakamura et al. Ogita K., Ohkawara A., Suzuki T., Ohgaki T., Uchida S., Meguri H. and Yoneda Y. (1992) Effects of ifenprodil on the N-methyl-D-aspartate receptor ionophore complex in rat brain. Neurochem. Int. 21, 135 147. Olney J. W. (1969) Brain lesions, obesity and other disturbances in mice treated with monosodium glutamate. Science 164, 719-721. Olney J. W., Ho O. L. and Rhee V. (1971) Cytotoxic effects of acidic and sulfur-containing amino acids on the infant mouse central nervous system. Expl Brain Res. 14, 61 76. Pauwels P. J. and Leysen J. E. (1992) Blockade of nitric oxide formation does not prevent glutamate-induced neurotoxicity in neuronal cultures from rat hippocampus. Neurosci. Lett. 143, 27 30. Regan R. F. and Choi D. W. (1991) Glutamate neurotoxicity in spinal cord cell culture. Neuroscience 43, 585-591. Regan R., Renn K. and Pnter S. S. (1993) NMA neurotoxicity in murine cortical cell cultures is not attenuated by hemoglobin or inhibition of nitric oxide synthesis. Neurosci. Lett. 153, 53-56. Rothman S. M. and Olney J. W. (1987) Excitoxicity and NMDA receptor. Trends Neurosci. 10, 299-302. Rothman S. M., Thurston J. H. and Hauhart R. E. (1987) Delayed neurotoxicity of excitatory amino acids in vitro. Neuroscience 22, 471~480. Schumidt-Kastner R., Szymas J. and Hossmann K.-A. (1990) Immunohistochemical study of glial reaction and serum-protein extravasation in relation to neuronal damage in rat hippocampus after ischemia. Neuroscience 38, 527-540. Stuehr D. J., Fasehun O. A., Kwon N. S., Gross S. S., Gonzalez J. A., Levi R. and Nathan C. F. (1991) Inhibition of macrophage and endothelial cell nitric oxide synthase by diphenyleneiodonium and its analogs. Fedn Proc. Fedn Am. Socs exp. Biol. 5, 98-103. Tsuji K., Nakamura Y., Ogata T., Mitani A., Kataoka K., Shibata T., Ishida M. and Shinozaki H. (1995) Neurotoxicity of acromelic acid in cultured neurons from rat spinal neurons. Neuroscience 68, 585-591. Tsuji K., Nakamura Y., Ogata T., Shibata T. and Kataoka K. (1994) Rapid decrease in ATP content without recovery phase during glutamate-induced cell death in cultured spinal neurons. Brain Res. 662, 289-292. Wong E. H. F., Kemp J. A., Priestley T., Knight A. R., Woodruff G. N. and Iversen L. L. (1986) The anticonvulsant MK801 is a potent N-methyl°D-aspartate antagonist. Proc. natn. Acad. Sci. U.S.A. 83, 7104 7108. Yoneda Y., Enomoto R. and Ogita K. (1994) Supporting evidence for negative modulation by protons of an ion channel associated with the N-methyl-D-aspartate receptor complex in rat brain using ligand binding techniques. Brain Res. 636, 298-307. (Accepted 13 June 1996)