Nitric oxide-evoked [3H]γ-aminobutyric acid release is mediated by two distinct release mechanisms

MOLECULAR BRAIN RESEARCH ELSEVIER

Molecular Brain Research 36 (1996) 137-144

Research report

Nitric oxide-evoked [3H]y-aminobutyric acid release is mediated by two distinct release mechanisms Seitaro Ohkuma, Masashi Katsura, Da-Zhi Chen, Hidehiko Narihara, Kinya Kuriyama * Department of Pharmacology, Kyoto Prefectural University of Medicine, Kamikyo-Ku, Kyoto 602, Japan Accepted 2(1 September 1995

Abstract

Mechanisms underlying the release of [3H]y-aminobutyric acid (GABA) evoked by nitric oxide (NO) were investigated by use of primary cultured neurons prepared from the mouse cerebral cortex. NO generators such as sodium nitroprusside (SNP) and S-nitroso-Nacetylpenicillamine (SNAP) increased both [3H]GABA release from the neurons and [':5Ca2+ ] influx into the neurons in a dose-dependent manner, which was significantly diminished by hemoglobin. The removal of Ca 2" significantly reduced the NO-induced [3H]GABA release by about 50%. Nipecotic acid and •-(2-(((dipheny•methy•ene)amin•)•xy)ethy•)-••2•5•6-tetrahydr•-3-pyridinecarb•xy•ic acid (NO-711), GABA uptake inhibitors, dose-dependently inhibited the NO-evoked [3H]GABA release in either the presence or absence of Ca 2+. The concentration of these GABA uptake inhibitors to suppress the NO-induced release of [3H]GABA was sufficiently lower than that to exhibit the inhibition of [3H]GABA transport into the neurons. In addition, the NO-evoked [3H]GABA release was reduced by approximately 5(1% when total Na + in incubation buffer was replaced with equimolar choline, and was also completely abolished by the removal of both Ca 2' and Na t. These results indicate that the release of [3H]GABA evoked by NO is mediated by two release mechanisms, a Ca 2*-dependent release system and the reverse process of the Ca 2+-independent and Nat-dependent carrier-mediated GABA uptake system. Keywordw Nitrix oxide; y-Aminobutyric acid; Release mechanism

1. Introduction

The release of neurotransmitters, including amino acid neurotransmitters from neurons in the central nervous system (CNS), has been considered to be dominantly mediated by a Ca 2~-dependent exocytotic mechanism [2,23]. Such a type of mechanism of neurotransmitters release is also present in cells in the peripheral organ, the adrenal chromaffin cells [6]. On the other hand, several lines of evidence reported over the past two decades reveal the

Abbreviations: DMEM, Dulbecco's modified Eagle's medium; GABA, y-aminobutyric acid; KRB, Krebs-Ringer bicarbonate buffer; NO, nitric oxide; NO-711, 1-(2-(((diphenylmethylene)amino)oxy)ethyl)-l,2,5,6-tetrahydro-3-pyridinecarboxylic acid; SNAP, S-nitroso-N-acetylpenicillamine; SNP, sodium nitroprusside • Corresponding author. Fax: (81) (75) 241-0824. 0169-328X/96/$15.00 © 1996 Elsevier Science B.V. All rights reserved SSDI 0 1 6 9 - 3 2 8 X ( 9 5 ) 0 0 2 5 6 - 1

presence of a release mechanism of neurotransmitters different from Ca 2 +-dependent exocytotic release. This mechanism is a reverse process of the carrier-mediated transport system for neurotransmitters and is characterized by Ca 2~independent and Na +-dependent property [2,16]. NO is an unstable radical produced during the reaction of oxidative deamination catalyzed by NO synthase which converts L-arginine to L-citrulline in a Ca 2+- and NADPH-dependent manner [5]. Among the physiological roles of NO in the CNS, it is emphasized that NO induces releases of several neurotransmitters [ 11,12,17,26,29,32]. However, the mechanisms of neurotransmitter release evoked by NO have not been clear. In the present study, we have attempted to clarify whether NO-evoked release of G A B A is mediated by a Ca2+-dependent release system or by the reverse process of a Ca-'÷-independent and Na +-dependent carrier-mediated transport system using the cerebral cortical neurons possessing the Na÷-dependent carrier-mediated G A B A transport system [15].

138

S. Ohkuma et aL / Molecular Brain Research 36 (1996) 137-144

2. Materials and methods

2.1. Primary culture of cerebral cortical neurons The procedures to isolate and to culture neurons from mouse cerebral cortex were carried out according to the methods previously described [15,24]. In short, the neopallium free of meninges was removed from 15-day-old fetuses of ddY strain mice, minced, treated with trypsin and centrifuged. The resultant pellet was suspended with DMEM containing 15% fetal calf serum and filtered through a nylon mesh (mesh size: 60 ,u,m). One ml of the cell suspension containing 3 x 106 cells/ml were added into a culture dish (Falcon's 'Primaria ' * , diameter: 35 mm), pretreated with poly(L-lysine) for 24 h prior to use for culture, and cultured at 37°C in humidified 95% air/5% CO 2 for 3 days. After the cells were exposed to 10 p,M cytosine arabinoside dissolved in DMEM supplemented with 10% horse serum for 24 h to suppress the proliferation of non-neuronal cells, the neurons were cultured with DMEM supplemented with 10% horse serum. The culture medium was exchanged every 4 days and the neurons cultured for 13 days were used for the following experiments. The cerebral cortical neurons used here have been reported to possess the Na÷-dependent carrier-mediated GABA transport system [15].

2.2. Measurement of [3H]GABA release from neurons The measurement of [3H]GABA release was performed according to a previously reported method [27] with a minor modification. The neurons were incubated in KRB (137 mM NaC1, 4.8 mM KCI, 1.2 mM KH2PO4, 2.7 mM CaCI 2 • 2H20, 1.2 mM MgSO4 • 6H20, 25 mM NaHCO 3, 10 mM glucose, pH 7.4) at 37°C for 10 min after preincubation of the neurons in KRB containing 2 nM [3H]GABA (1.0 p, Ci [3H]GABA/dish) at 37°C for 30 min, followed by rinsing the neurons five times with ice-cold KRB (total volume: 7.5 ml). Warm KRB (37°C, 1.0 ml) was added into the dish and further incubation for 10 min was carried out. After discarding the incubation buffer, the neurons were incubated in KRB at 37°C at an interval of 5 min for a total of 25 min. Sodium nitroprusside (SNP) and Snitroso-N-acetylpenicillamine (SNAP) were added into the incubation buffer at the beginning of the 4th interval of the incubation. GABA uptake inhibitors, such as nipecotic acid and NO-711, were also added to the incubation buffer immediately before the addition of SNP and SNAP. An aliquot of the incubation buffer collected during each interval was used for measuring released [3H]GABA by liquid scintillation spectrometry. All incubation buffers contained 100 /zM aminooxyacetic acid to prevent the degradation of [3H]GABA released into KRB and incorporated into the neurons. The release of [3H]GABA during the 3rd interval of incubation was defined, as the basal release and the stimulated release was expressed as a

percentage of the basal release. To examine the Na" dependence of [3H]GABA release, Na- was replaced by equimolar choline. The basal releases of [3H]GABA determined by the use of Ca2 ~-free and Na+-free KRB were not significantly different from those by using KRB.

2.3. M e a s u r e m e n t o f [4SCa 2 +] influx into neurons The influx of [45Ca2+] into the neurons was measured according to a previously reported method [25] with a minor modification. The neurons were preincubated in Ca2÷-free KRB at 37°C for 10 min and the buffer was exchanged to fresh and warm (37°C) Ca2~-free KRB. The reaction was initiated by the addition of 1.26 mM [45Ca2 + ]CI 2 (1.0/zCi [45Ca2 + ]Cl2/dish). After incubation at 37°C for 2 min, the neurons were rinsed five times with ice-cold KRB and were scraped off with 0.5 M NaOH. Such washing procedures were carried out within 45 s. An aliquot of alkaline-digested cells was neutralized with 0.5 M acetic acid and transferred to a scintillation vial containing Bioflour for measuring [45Ca2+] incorporated into the neurons by liquid scintillation spectrometry. SNP and SNAP were added into the incubation buffer immediately before the addition of [45Ca2 + ]CI 2. In addition, hemoglobin was added to the incubation buffer immediately before the addition of SNP and SNAP.

2.4. Measurement of [~H]GABA uptake by neurons The uptake of [3H]GABA by the neurons was measured by a method previously described [15]. After a 10-min incubation of the neurons in KRB at 37°C, the incubation buffer was exchanged to fresh and warm (37°C) KRB, and the reaction was started by the addition of 2 nM [3H]GABA (0.01 /zCi [3H]GABA/dish). At the end of the incubation at 37°C for 5 min, the incubation buffer containing radiolabeled GABA was discarded by aspiration, and the neurons were washed five times with ice-cold KRB. An aliquot of the neurons digested with 0.5 M NaOH was neutralized with equimolar acetic acid, transferred to a scintillation vial with Triton-xylene scintillator [Triton X-100:xylene (4 g Omnifluor/litter of xylene) = 2:1], and the radioactivity accumulated in the neurons was measured by liquid scintillation spectrometry. The net amount of [3H]GABA transported into the neurons was calculated by correcting for extracellular space determined by [3H]inulin [15]. The incubation buffer used for the experiments to examine the uptake of [3H]GABA contained 100 /zM aminooxyacetic acid.

2.5. Measurement of protein content The protein content in the digested cells was determined by the method of Lowry et al. [18] using bovine serum albumin as standard.

S. Ohkuma et al. / Molecular Brain Research 36 ( I996) 137-144

13t)

'25 I

2.6. Statistical analysis The data obtained w e r e expressed as the m e a n + S E M and the statistical significance was determined by the m e t h o d described in each legend of the figures and tables f o l l o w i n g the application of the o n e - w a y A N O V A . Each statistical method such as D u n n e t t ' s test was e m p l o y e d for the statistical evaluation o f the data w h e n P values determ i n e d by the o n e - w a y A N O V A w e r e found to be less than 0.0001.

~ 17Sl-

2. 7. Materials

W//A

0

0.3

1

3

10

30

100

300

Concentration (I~M)

[3H]GABA (2.6 T B q / m m o l ) , 45CaCi., (0.3511 G B q / m g ) , and [3H]inulin (606.8 M B q / g ) w e r e the products of N e w England Nuclear (Boston, USA). Research B i o c h e m i c a l s , Inc. (Natick, U S A ) was the source of S N A P and NO-711. H e m o g l o b i n was purchased from S i g m a C h e m i c a l s (St. Louis, MO, U S A ) . Fetal b o v i n e and horse serums were obtained from Hazleton Research Products, Inc. (Lenexa, U S A ) . Other c h e m i c a l s used were locally available and of analytical grade.

3. Results 3.1. [~H]GABA release by NO generators The effects o f two different N O generators on [ 3 H ] G A B A release from cerebral cortical neurons in primary culture w e r e e x a m i n e d . A s s h o w n in Fig. !, these N O generators d o s e - d e p e n d e n t l y increased [ 3 H ] G A B A release. The m a x i m a l stimulatory effects of S N P and S N A P were found at 100 /.tM and 5 - 1 0 ~ M , respectively. Based on these results, 100 /zM o f S N P and 5 /zM o f S N A P w e r e e m p l o y e d to induce the release of [ 3 H ] G A B A in the f o l l o w i n g release experiments. The stimulatory actions o f S N P and S N A P w e r e inhibited by h e m o g l o b i n in a dosedependent m a n n e r (data not s h o w n ) and c o m p l e t e l y abolished in the presence of 40 /zM h e m o g l o b i n (Table 1), indicating that the induction of [ 3 H ] G A B A release by S N P and S N A P w e r e certainly mediated by N O liberated from the N O generators.

Fig. I. Effects of SNP and SNAP on [~H]GABA release fiom cerebral cortical neurons. Neurons cultured for 13 days wcrc preincubated in Ca"'-free KRB with 2 nM [~tl]GABA (containing 1.1) uCi [3H]GABA/dish) at 37°C for 311 min. After the preincubation, neurons were rinsed three timcs with ice-cold Ca-"-frce KRB and further incubated with the same incubation buffer at 370(.` t~.)r I1) min. Thereafter. fresh and warm (37°(") Ca:" -free KRB was added into a euhurc dish and neurons were incubated at an interval of 5 rain for a total ot 25 min. At the beginning of the 4th interval of the incubation. SNP and SNAP were added into the incubation buffer. The relcase of [~H]GABA detected during the 3rd interval of the incubation was defined as the basal relcase. The evoked release was expressed as a percentage of the basal release of [3H]GABA. Thc incubation buffer obtained during each interval of the incubation was pooled and an aliquot of sample was used to measure the radioactivity of released [~H]GABA by liquid scintillation spectrometry. The incubation buffer used throughout the experiment was supplemented with 100 ArM aminooxyacetic acid. The basal releases of [3H]GABA ll~r SNP and SNAP were 4127+556 and 4731)+ 5191 dpm/mg protein per interval, respectively. Each value represents the mean ±SI'M obtained from three separate experiments. " P < 0.05. " " /' < ().l)l. compared with each basal release (Dunnett's test)

duced [ 3 H ] G A B A release by the degree found in the case o f Ca-' "-free K R B without E G T A (data not shown). The release of a m i n o acid neurotransmitters has been reported to be mediated by the reverse process of a C a : independent and Na ÷-dependent carrier-mediated transport

Table 1 Effect of hemoglobin (Hb) on SNP- and SNAP-induced [XH]GABA release from cerebral cortical neurons [3ti]GABArelease (%of basal release)

3.2. Effects o f removal of Ca'-* and / or Na + and GABA uptake inhibitors on [ 3H/GABA release by NO generators In order to e x a m i n e the effect o f the r e m o v a l of Ca 2 + from the incubation buffer on [ 3 H ] G A B A release induced by N O generators such as S N P and S N A P , the neurons w e r e incubated in Ca 2+-free incubation buffer in the presence of 100 B,M S N P and 5 p,M S N A P . The incubation o f the neurons in Ca 2 +-free K R B induced a significant reduction o f the S N P - and S N A P - e v o k e d [ 3 H ] G A B A releases by 48% and 45%, respectively (Fig. 2). The addition of 1 m M E G T A into C a 2 ' - f r e e K R B also reduced the N O - i n -

None Hb (40 #M)

SNP (IOO pM)

SNAP (5 /xM)

184.0 ± 8.4 102.0±3.8 " "

21)9.1)+ 7.2 118.3 L 7.4 " "

Thc procedure to measure the release of [~H]GABA is dcscribcd in the legend of Fig. 1. Hb was added into the incubation buffer immediately before the addition of NO generators. Each value represents the mean ± SEM determined from three separate experiments and is expressed as a percentage of the basal release. The basal releases for SNP and SNAP were 5431 +463 and 4957 + 267 dpm/mg protein per interval, respectively. P < I).01, compared with each value obtained in the absence of Hb (Bonferroni's test). •

•

140

S. Ohkuma et al. / Molecular Brain Research 36 (1996) 137-144 200

i ~

SNP (100 pM ~t/t, ##

~'A', ##

A

SO

i

m

"6

o S N A P (5 p M ) '

g e

"0 rr <

* ~ , ##

--'r-

150

3.3. Effects of nipecotic acid and NO-711 on [3H]GABA uptake by cerebral cortical neurons

~ , * , ##

.-r-

m <

~t/t 100

P_.,

0

Na +

+

+

--

--

Ca2÷

+

--

'4"

--

Fig. 2. Effects of removal of Ca z" a n d / o r Na ~ on SNP- and SNAP-induced release of [3H]GABA from cerebral cortical neurons. The procedure to measure the release of [3H]GABA is described in the legend of Fig. 1. Sodium ion was replaced with equimolar choline when the effect of removal of Na " was examined. Each value represents the mean :t: SEM obtained from five separate experiments and is expressed as a percentage of the basal release. The basal relea~s of [3H]GABA for SNP and SNAP were 5550 ± 487 and 4545 + 397 d p m / m g protein per interval, respectively. " " P < 0.01, compared with each value determined in the presence of Ca"- and Na ~ (Bonferroni's test). ## P < 0.05, compared with each value determined in the absence of Ca 2- and Na ÷ (Bonferroni's test).

To compare the difference in the concentrations of these GABA transporter blockers to inhibit the release of GABA and to block the uptake of GABA, we examined the profile of GABA transport inhibitors, such as nipecotic acid and NO-711, to inhibit [3H]GABA incorporation into the neurons. As shown in Fig. 5, these two inhibitors dose-dependently inhibited [3H]GABA uptake. The minimal concentrations of nipecotic acid and NO-711 to exhibit their inhibitory actions on [3H]GABA transport were 10 /xM and 0.3 /xM, respectively. The respective K, values of nipecotic acid and NO-711 for [3H]GABA uptake were 25 /zM and 0.88 /xM (the mean of four separate experiments). It is noteworthy that the concentrations of these GABA transport blockers to show their maximal inhibitions of the NO-evoked [3H]GABA release were lower than those exhibiting their inhibitory effects on the uptake of [~H]GABA (Figs. 3-5).

(A) 1K

jj

leR

tt

14C

100 8~

--e--

ca(.)

i

i

i

i

o

O.Ol

0.1

I

C o n c e n t r a t k ~ of NA (I~M)

system [16]. To examine whether this release mechanism is involved in the release of [3H]GABA induced by NO, we checked the effects of the removal of Na 4 and GABA transport inhibitors on [3H]GABA release evoked by NO. Under the conditions that the incubation buffer contained 1.26 mM CaCI 2, the replacement of Na + with choline reduced the [3H]GABA release evoked by SNP and SNAP by approx. 48% and 48%, respectively (Fig. 2). In addition, the release of [3H]GABA by these NO generators was completely abolished in the absence of both C a 2+ and Na ÷ (Fig. 2). Nipecotic acid, an inhibitor of carrier-mediated GABA transport system, dose-dependently suppressed the release of [3H]GABA induced by SNP in either the presence or absence of Ca 2+ (Fig. 3A). NO-711, a GABA transporter blocker with higher potency than nipecotic acid [9], inhibited the SNP-evoked [3H]GABA release in a dose-dependent manner in both KRB and Ca2+-free KRB (Fig. 3B). These two GABA anti-transporters also dose-dependently inhibited the SNAP-induced [3H]GABA release as observed in the case of SNP (Fig. 4A and B).

2OO 180

i(.)

A

It't

140 12(] Ca 1,)

10(] N

-'-0-" i o

"" "" "" " ' " " " " I~ t t t

Ca(-) i 0.OOl

i o~o1

i o.1

C o n c e n t r a t i o n of NO-711 (~A)

Fig. 3. Effects of (A) nipecotic acid (NA) and (B) NO-711 on SNP (100 /xM)-evoked [3H]GABA release from cerebral cortical neurons. The procedure to measure the release of [3H]GABA is described in the legend of Fig. 1. Nipecotic acid and NO-711 were added into the incubation buffer immediately before the addition of SNP. The basal releases of [3H]GABA for (A) and (B) were 4789± 397 and 5007 ±477 d p m / m g protein per interval, respectively. Each value represents the mean + SEM obtained from 4 separate experiments and is expressed as a percentage of the basal release. " " P < 0.01, compared with each value determined in the absence of nipecotic acid or NO-7ll (Dunnett's test). ## P < 0.01, compared with the value determined in the presence of Na ~ and in the absence of nipecotic acid or NO-711 (Bonferroni's test).

S. Ohkuma et al. / Molecular Brain Research 36 (1996) 137-144 2OO

100

iI I=

~

'~}~i* **

A e"

(A)

•-0

160

140

"~.*..I...

120

100

oo

"'--...

t 0.01

--O-

'~-C rs "6

--'O'- .o-~.

¢

,. ,~.*._ -*--

,~p.o,
.././.0.... /*' /,

3.

--O-- Ca(.) t 0

ii ,0o

..... ~,..,.

Ca(~)

+

141

0.11

GI <

I 1

Concentration of NA (i~M)

P.., -2£

0

2oo

jj'

lOO

[;

120 lOO 80

0.01 0.03

I

0

""1} ~t~ I

o.ool

I

l

0.01

0.1

Co~ncentrlltlon of NO-711 (v.M)

Fig. 4. Effects of (A) nipecotic acid (NA) and (B) NO-711 on SNAP (5 /xM)-evoked [3H]GABA release from cerebral cortical neurons. The procedure to measure the release of[ 3H]GABA is described in the legend of Fig. 1. Nipecotic acid and NO-711 were added into the incubation buffer immediately before the addition of SNAP. The basal releases of [3H]GABA for (A) and (B) were 5159+527 and 5611 +497 d p m / m g protein per 5 min, respectively. Each value represents the mean + SEM obtained from four separate experiments and is expressed as a percentage of the basal release. The value without SEM means that the value of S.E.M. is smaller than the diameter of the circle. " P < 0.05, " " P < 0.01, compared with each value determined in the absence of nipecotic acid or NO-711 (Dunnett's test). # # P <1).111, compared with the value determined in the presence of Na- and in the absence of nipecotic acid or NO-711 (Bonferroni's test).

0.3

1

3

10

30

100

Concentration (pM)

(B)

--4F-- Cl(-)

0.1

Fig. 5. The effects of nipccotic acid (NA) and NO-71I on [~H]GABA transport into cerebral cortical neurons. Neurons were preincubated in KRB at 37°C for 10 min. and were further incubated with fresh and warm (37¢C) KRB. The reaction was initiated by thc addition of 2 nM [3H]GABA (0.01 /xCi [3H]GABA/dish). After the incubation at 37~C for 5 min, the reaction was terminated by rapid aspiration of the incubation buffer and subsequent washing of neurons five times with ice-cold KRB. After digesting neurons with 0.5 M NaOH, an aliquot of alkaline digested neurons was neutralized with equimolar acetic acid and then transferred to a scintillation vial to measure the radioactivity accumulated in neurons. GABA uptake inhibitors were added into a dish immediately before the addition of [3H]GABA. Each value representcd the mean +_SEM obtained from four separate experiments. " " P < t}.01, compared with the value determined in the absence of nipccotic acid and NO-711 (Bunnett's test).

generators on [45Ca2 + ] influx into the neurons were significantly depressed by hemoglobin (Table 2). These data indicate that NO liberated from NO generators is certainly A

100o0

[] •

SNP SNAP

.~* **

r

~) 750(]

3.4. Effects of NO generators on [45Ca2+] influx into rleurorls

We have examined whether the NO generators employed in this study have the ability to induce Ca 2- influx, since about half the amount of [3H]GABA released is mediated by the Ca 2+-dependent release system. Both SNP and SNAP increased the [45Ca2+] influx into the neurons in a dose-dependent manner (Fig. 6). The dose-dependent curve of the [45Ca2~'] influx against SNAP was considered to be similar to that of [3H]GABA release. On the other hand, the [45Ca2+] influx induced by SNP attained its maximal level at a concentration of 10 /xM SNP, whereas [3H]GABA release attained its plateau at 100 /xM SNP. When comparing the maximal [45Ca2+] influx evoked by these NO generators (SNP: 100 /xM; SNAP: 10 /xM), the level of [45Ca2+] influx induced by SNP was similar to that by SNAP (Fig. 6). These stimulatory effects of the NO

5000 0

0.1

1

3 10 C o n c e n t r a t i o n (tiM)

30

100

Fig. 6. Effects of SNP and SNAP on {45(_'a:-J influx into cerebral cortical neurons. After neurons were preincubated in C a " - f r e e KRB at 37°C for 10 rain, neurons were further incubated in warm (37°C) Ca 2*-free KRB following aspiration of the preincubation buffer. The reaction was initiated by adding 1.26 mM CaCl 2 containing l.I)/.zCi [45Ca2 ' ]CI 2 into the incubation buffer. After the incubation for 2 rain, the reaction was terminated by aspiration of the incubation buffer and subsequent washing of cells five times with ice-cold KRB and neurons were digested with 0.5 M NaOH. An aliquot of alkaline digested cells was neutralized with equimolar acetic acid and transferred to a scintillation vial to measure the radioactivity accumulated in neurons. The addition of SNP and SNAP was carried out immediately before the initiation of the reaction. Each value represents the mean + SEM obtained from three separate experiments. " P < 0.05, " " P < 0.01, compared with the value determined in the absence of SNP and SNAP (Dunnett's test).

142

S. Ohkuma et aL / Molecular Brain Research 36 (1996) 13 7-144

Table 2 Effect of hemoglobin (Hb) on NO-induced cortical neurons

[45Ca2÷ ] influx

into cerebral

[~SCa 2" ] influx ( c p m / m g protein per 2 min) -

SNP ( 100 p.M) SNP (5 /zM)

Hb

5443 + 263 5222 .+ 234

+

Hb ( 3 0 / x M )

4609 _+ 146 " " 4031 _+ 127 " "

The procedure to measure [4SCa "~" ] influx was described in the legend of Fig. 6. Hb was added into the incubation buffer immediately before the addition of NO generators. Each value represents the mean + SEM determined from three separate experiments. The basal influx of [4SCa2* ] for SNP and SNAP was 43811+78 and 3681 + 114 c p m / m g protein per 2 min• respectively. SNP, sodium nitroprusside; SNAP• S-nitroso-Nacctylpcnicillamine. • " P < II.{}l, compared with each value determined in the absence of Hb.

involved in the influx of [~5Ca2" ] into the cerebral cortical neurons.

4. Discussion In the present study using primary cultured cerebral cortical neurons with the Na+-dependent carrier-mediated GABA transport system [15], we have attempted to clarify the mechanisms underlying the release of [~H]GABA evoked by NO liberated from NO generators such as SNP and SNAP. Two NO generators, SNP and SNAP, dose-dependently evoked the release of [3H]GABA from the cerebral cortical neurons. The potency of SNAP to induce the release of [3H]GABA was about 10 times stronger than that of SNP, and this difference appeared to be similar to the difference in the potency to produce cyclic GMP [13], although cyclic GMP is considered to have no involvement in neurotransmitter release. The SNP- and SNAP-evoked releases of [3H]GABA were completely abolished by hemoglobin, indicating that SNP and SNAP induce [3H]GABA release from neurons by the liberation of NO from them. Such a stimulatory effect of NO on neurotransmitter release has been reported by several investigators [11,12,17,26,27,29, 32]. These [3H]GABA releases evoked by NO liberated from the molecules of SNAP and SNP, are also assumed to be partially mediated by peroxynitrite, since earlier we have reported that peroxynitrite formed by the reaction of NO with superoxide is partially involved in NO-evoked [3H]GABA release [26,27]. The removal of Ca 2+ from the incubation buffer reduced the SNP- and SNAP-evoked releases of [3H]GABA by about 50%, indicating that about half the amount of [3H]GABA release induced by NO is mediated by a Ca2+-dependent release mechanism. Such Ca 2~ dependence of GABA release by NO was reported by GuevaraGuzman et al. [11] using rat striatum. In the next step, we have examined whether the two NO donors employed in

this study induce the influx of ['*SCa2 ~] into the neurons. The present study revealed that both SNP and SNAP increased the influx of [4~Ca2"] into the cerebral cortical neurons in a dose-dependent manner. The NO-evoked [3H]GABA release induced by NO in a Ca2~-dependent manner is considered to be associated with the influx of Ca 2' into the neurons. Therefore, these two data, the Ca 2 dependence of [3H]GABA release evoked by NO and the increase in the influx of Ca 2~ by NO, are assumed to be not contradictory to each other. Furthermore, these increases in [45Ca2~ ] influx were significantly inhibited by hemoglobin. These data lead to the conclusion that NO has the ability to enhance the influx of Ca 2+ into neurons, although the mechanism of the stimulatory action of NO on Ca 2 ~ influx remains to be elucidated. Similar effect of NO on Ca 2~ influx in rat sympathetic neurons has been reported by Chen and Schofield [7]. In contrast, several investigators have demonstrated that NO itself does not alter the influx of Ca 2" [14,22] and that NO reduced Ca 2+ entry [8,19]. At present, the differences in the effects of NO on Ca e÷ influx are not clear. The dose-dependent curve of the SNAP-induced [3H]GABA release was very similar to that of [45Ca2'] influx stimulated by SNAP. Similarly, SNP dose-dependently induced [3H]GABA release and its maximal stimulatory effect was observed at concentrations higher than 100 /,tM. At these concentrations of SNP, the degree of [3H]GABA release was the same as that by SNAP at concentrations higher than 10 /xM, whereas the magnitude of [~SCa-~" ] influx induced by 100 /xM SNP was approx. 50% of that induced by 10 /zM SNAP. These discrepancies of SNP between the potency to evoke [3H]GABA and that to increase [45Ca2+] influx may be due to products, including CN ion, which are simultaneously formed from SNP with NO [10]. ttowever, this possibility is unlikely, because hemoglobin abolished both SNP-induccd [~H]GABA release and [45Ca 2~ ] influx, which implied that the stimulatory effects of SNP on [3H]GABA release and [4SCa2-] influx was certainly mediated by NO liberated from SNP. Therefore, the exact reasons for these discrepancies are not clear at present. In the cerebral cortical neurons used in this study, the release of [3H]GABA in response to NO was also Na ~-dependent in conditions with both the presence and absence of Ca 2 ÷. In addition, we have demonstrated that the release of [3H]GABA evoked by NO is dose-dependently inhibited by two distinct G A B A uptake inhibitors, nipecotic acid and NO-711 [9] in either the presence or absence of Ca 2~ in the incubation buffer. These results indicate that part of the [~H]GABA evoked by NO is operated by a Ca 2~-independent and Na+-dependent mechanism, which raises the possibility that this type of [3H]GABA release may be mediated by the reverse mechanism of a Na'-dependent carrier-mediated GABA transport system. In fact, several lines of evidence have recently revealed the presence of the reverse process of the C a 2 * - i n d e p e n d e n t and

S. Ohkuma et al. / Moh'cular Brain Research

Na +-dependent carrier-mediated neurotransmitter transport as a mechanism for neurotransmitter release. In retinal horizontal cells [3(I], striatal neurons [28], striatal slices [3], and growth cones derived from rat forebrain [31], G A B A is released by the reverse process of a Ca2+-independent and Na+-dependent carrier-mediated G A B A transport mechanism, and this type of neurotransmitter release is inhibited by nipecotic acid, a competitive inhibitor of G A B A transport, or by replacement of extracellular Na" with Li +. A similar release mechanism has also been reported to mediate the release of biogenic amines [20,21]. Nipecotic acid and NO-711 dose-dependently inhibited the NO-induced [3H]GABA release. At I /,LM of nipecotic acid and 0.1 /zM of NO-711, these G A B A uptake inhibitors completely abolished the NO-induced release of [3H]GABA in the absence of C a - " . Even in the presence of Ca: +, they showed their inhibitory effects on the release of [3H]GABA induced by NO generators. On the other hand, Bernath and his co-workers reported that nipecotic acid increased the release of G A B A in the presence of Ca ~-' but not in the absence of C a - " , when electrical stimulation was used as a manipulation to evoke G A B A release [3,4]. Therefore, the discrepancy of the effect of nipecotic acid on G A B A release observed in the study of Bernath et al. and the present investigation may be due to the difference in the methods to induce the release of G A B A . Another possibility causing such a difference is assumed to be due to the difference in the concentrations of nipecotic acid to examine its effect on G A B A release, because the concentration of nipecotic acid employed by Bernath et al. [3,4] has been about 100-1000 times higher than that used in this study, although the exact reason fi~r these differences is not clear at present. The G A B A uptake inhibitors used here did not exhibit the inhibition of [~H]GABA uptake at the concentrations showing the suppression of the NO-induced release of [~H]GABA. That is, the higher concentrations of the uptake inhibitors are necessary to inhibit the [~I-I]GABA uptake by the cerebral cortical neurons used in this study. Indeed, the K. values of nipecotic acid and NO-711 for [~H]GABA uptake were about 25- and 88-fold higher than the concentrations showing maximal inhibitory effects on the NO-induced [~H]GABA release. These results indicate that the system for the release of [SH}GABA evoked by NO is more sensitive to G A B A transport inhibitors than that for [~H]GABA uptake in the cerebral cortical neurons used in the present study. In contrast, S K F 89976-A, an analog of nipecotic acid, inhibited K ~-evoked G A B A release from cerebral cortical neurons [I] and growth cones prepared from the rat forebrain [31] in the presence of C a - " , and the concentration of S K F 89976-A to suppress K ' - s t i m u l a t e d G A B A release by 5(.1% was the same as that to inhibit G A B A uptake by 50% [31]. At present, the reasons that the sensitivity of Ca2'-independent and N a ' dependent process of G A B A release to the inhibitory actions of G A B A uptake inhibitors is higher than that of

.?6 (I 996) 13 7-144

143

carrier-mediated G A B A uptake system remains to be elucidated. In conclusion, we have attempted to clarify the mechanism of G A B A release evoked by NO by the use of mouse cerebral cortical neurons in primary culture. Two NO generators, SNP and SNAP, dose-dependently increased [:~H]GABA release from neurons, which was significantly inhibited by hemoglobin. The removal of Ca z' reduced the NO-induced [3H]GABA release by approx. 511%. Similarly, the total replacement of Na " to choline also induced about 50°£ reduction of the NO-evoked release of [~H]GABA from the neurons. The withdrawal of both Ca 2+ and Na ~ from the incubation buffer completely abolished the release of [~H]GABA evoked by NO. The G A B A uptake inhibitors, nipecotie acid and NO-71 I, inhibited the NO-induced release of [~H]GABA in a dosedependent manner in either the presence or absence of C a - " . These results indicate that NO-induced [ " H ] G A B A release from cerebral cortical neurons is mediated by two distinct release mechanisms, the Ca z ' - d e p e n d e n t release process and the reverse process of a ('a z "-independent and N a ' - d e p e n d e n t carrier-mediated G A B A uptake system.

References [1] Belhage. B., Hansen O.tl. and Schousoc. A.. Depolarization by K ÷ and glutamate activate different ncurotransmitter release mechanisms in GABAergic neurons: Vesicular versus non-vesicular release of GABA, Neuro~'cience, 54 (1993) IIHg-11)34. [2] Bcrnath. S., Calcium-independent release of amino acid neurotransmitters: fact or artifact?, Prog. Neurohiol.. 38 (I~,~c,~2)57 t~l [3] Bernath. S. and Zigmond, M.J.. Calcium-independentGAHA release from striatal slices: The role of c,dcium channel. Neuro~cten~c, 3(~ (IC,~9())677-682. [4] Bernath. S.. Keller, R.W. Jr. and Zigmond, MJ., Release of endogenous GABA can occur through CaZ'-dependent and ('a: '-independent processes, Neuroctwm Int., 14 (1(,~8~))43~) 445 [5] Bredt, D. and Snyder, S.H, Isolation i~l nitric t~xide svnthase, a calmodulin-requiringenzyme, l'roc. NatL Ac ad Sci. L"SA. 87 (1991)) 682-685. [6] Burgoyne. R.I)., Morgan. A., Robinson, [.. Pender. N. and Cheek, T.R., Exocytosis in adrenal chromaffin cell,,..L ,4nat.. Ig3 (19q3) 309- 314. [7] (.'hen, ('. and Schofield, G.D., Nitric oxide mt~dulates('a:* channel currents in rat sympathetic neurcms, Eur..L IJharmacol.. 243 (1~93) 83-86. [8] Desolc, M.S., Kim, W.-K.. Rabin. R.A. and Laychock, S.G., Nitric oxide reduced depolarization-induced calcium influx in P(.'I2 cells by a cyclic GMP-mediated mechanism...VeuropharmacohJ.~,~v. 33 (1994) 193-198. [9] Dong, C.-J., Picaud, S.A. and Werblin, F.,',;.,GABA transporters and GABAC-like receptors on catfish cone- but not rod-driven horizontal cells. J. Neurosci.. 14 (1Q84) 2648-265,~. [10] Feelish. M. and Noack, E.A.. ('orrclation between nitric oxide fi~rmation during degradation of organic nitrates and activation ol guanylate cyclase, Eur. J. Pharrnacol., 13~;' (19871 19-30. [11] Guevara-Guzman, R., Emson, P.C. and Kcndrick, K.M., Modulation of in vitro striatal neurotransmitter release bv nitric oxide and cyclic (JMP. J. Neurochem.. 62 (lC~q4)~()7-~10. [12] Hanbauer. 1.. Wink, D., Osawa. Y., Edelman. (;.M. and Gaily. J.A..

144

[13]

[14]

[15]

[16] [17]

[18]

[19]

[20]

[21]

[22]

S. Ohkuma et al. / Molecular Bruin Research 36 (1996) 137-144

Role of nitric oxide in NMDA-evokcd release of [3 H]dopamine from striatal slices, NeuroReport, 3 (1992) 409-412. Kiedrowski, L., Manev, H., Costa, E. and Wrobrewski, J.T., Inhibition of glutamate-induced cell death by sodium nitroprusside is not mediated by nitric oxide, Neuropharmacology, 30 (1991) 12411243. Kiedrowski, L., Costa, E. and Wroblewski, J.T., Sodium nitroprusside inhibits N-methyl-o-aspartate-evoked calcium influx via a nitric oxide- and cGMP-independent mechanism, Mol. Pharmacol., 41 (1992) 779-784. [ Kuriyama, K., Tomono, S., Kishi, M., Mukainaka, T. and Ohkuma, S., Development of y-aminobutyric acid (GABA)ergic neurons in cerebral cortical neurons in primary culture, Brain Res., 416 (1987) 7-21. Levi, G. and Raiteri, M., Carrier-mediated release of neurotransmitters, Trends Pharmacol. Sci., 16 (1993) 415-419. Lonart, G., Wang, J. and Johnson, K.M., Nitric oxide induces neurotransmitter release from hippocampal slices, Eur. J. Pharmacol., 220 (1992) 271-272. I.x~wry, O.H., Rosenbrough, N.J., Farr, A.L. and Randall, R.J.. Protein mcasurement with the Folin phenol reagent, d. Biol. Chem., 193 (1951) 265-275. Manzani, O., Prezeau, L., Marin, P., Deshager, S., Bockaert, J. and Fagni, L., Nitric oxide-induced blockade of NMDA receptors, Neuron, 8 (1992) 653-662. Marchi, M., Paudice, P., Bonanno, G. and Raiteri, M., Presynaptic muscarinic receptor activation enhances striatal dopamine release by depolarization but not that induced by non-depolarizing stimuli, Neurochem. Int., 7 (1985) 137-141. Maura, G., Gemignani, A., Versace, P., Martire, M. and Raiteri, M., Carrier-mediated and carrier-independent release of serotonin from isolated central nerve endings, Neurochem. Int., 4 (1982) 219-224. M~ry, P.-F., Pavoine, C., Belhasscn, L., Pecker, F. and Fis.chmeister, R., Nitric oxide regulates cardiac Ca 2- current. Involvement of cGMP-inhibited and cGMP-stimulated phosphodiesterases through

[23]

[24]

[25]

[26]

[27]

[28]

[29] [30]

[31]

[32]

guanylyl cyclase activation, J. Biol. Chem., 268 (1993) 2628626295. Nicholls, D.G., Release of glutamate, asparlate, and y-aminobutyric acid from isolated nerve terminals, J. Neurochem., 52 (1989) 331341. Ohkuma, S., Tomono, S., Tanaka, Y., Kuriyama, K. and Mukainaka, T., Development of taurine biosynthesizing system in cerebral cortical neurons in primary culture, Int. J. DeLl. Neurosci., 4 (1986) 383-395. Ohkuma, S., Chen, D.-Z., Katsura, M., Chen, S.-H. and Kuriyama, K., GABAA receptor stimulation enhances NMDA-induced Ca"influx in mouse cerebral cortical neurons in primary culture, Mol. Brain Res., 27 (1994) 145-151. Ohkuma, S., Katsura. M., Guo, J.-L., Hasegawa, T. and Kuriyama, K., Participation of peroxynitrite in acetylcholine release by nitric oxide generators, Neurosci. Lett., 183 (1995) 151-154. Ohkuma, S., Narihara, H., Katsura, M., Hasegawa, T. and Kuriyama, K., Nitric oxide (NO)-induced [3H]GABA release from cerebral cortical neurons is mediated by peroxynitrite, J. Neurochem., 65 (1995) 1109-1114. Pin, J.-P. and Bockaert, J., Two distinct mechanisms, differentially affected by excitatory amino acids, trigger GABA release from fetal mouse striatal neurons in primary culture, J. Neurosci., 9 (1989) 648-656. Prast, H. and Philippu, A., Nitric oxide releases acetylcholine in the basal forebrain, Eur. J. Pharmacol., 216 (1992) 139-140. Schwartz, E.A., Depolarization without calcium can release yaminobutyric acid from a retinal neuron, Science, 238 (1987) 350355. Taylor, J. and Gordon-Weeks, P.R., Calcium-independent y-aminobutyric acid relea~ from growth cones: Role of y-amino butyric acid transport, J. Neurochem., 56 (1991) 273-280. Zhu, X.-Z. and Luo, L.-G., Effect of nitroprusside (nitric oxide) on endogenous dopamine release from rat striatal slices, J. Neurochem.. 59 (1992) 932-935.