Membrane properties and intracellular biochemical processes during vasopressin-induced bursting activity in snail neurons

Neuroscienee" Research, 4 (1986) 37-50

37

Elsevier Scientific Publishers Ireland Ltd. NSR 00150

Membrane properties and intracellular biochemical processes during vasopressin-induced bursting activity in snail neurons Minoru Onozuka, Hiroyasu Furuichi*, Kenichi Kishii** and Shizuko Imai*** Department of Physiology, Kanagawa Dental College, Yokosuka (Japan)

(Received 10 March 1986; Accepted 12 May 1986) Key words: Vasoprcssin; Bursting activity; Cyclic AMP; Protein kinase; Calcium; Adenylate cyclase; Snail neuron

SUMMARY To elucidate the mechanism generating bursting activity, the effect of arginine vasopressin (AVP) was studied electrophysiologically and biochemieally in ganglionic preparations from the snail, Euhadra peliomphala. AVP caused bursting activity which is accompanied by the development of a negative slope resistance (NSR) region in the current-voltage (I-V) curve of the identified neurons. Similar effects were observed by application ofveratridine, dibutyryl eye!ie AMP and isobutylmethylxanthine. Both the bursting activity and the I-V relation induced by AVP were markedly inhibited by reduction of extracellular Na + but not by Co2+-substituted Ca2+-free saline. This hormone also caused the following intracellular biochemical alterations: (i) elevation in the cyclic AMP levels; (ii)stimulation of adenylate cyclase and Ca2+ -dependent protein kinase activities; and (iii)promotion of Ca 2§ release from the intracellular reservoir, lysosomc-like granules. These results suggest that AVP-induced bursting activity is mediated through intracellular biochemical processes.

INTRODUCTION

Although bursting activity underlying epileptic seizure has been extensively studied2-4.1 ~.20.2s-3~.33.a4,the ultimate mechanism of its generation has not been fully * Present address: Central Research Laboratories, Zeria Pharmaceutical Co. Ltd., 2512-10shikiri,

Konan-machi, Osato-gun, Saitama 360-01, Japan. ** Present address: Antibiotics Rcsearch Laboratory, Kayaku Industry Co. Ltd., 27-12 Higashi-lkebukuro

l-chome, Toshimaku, Tokyo 170, Japan. *** Present address: Department of Pathology, Asahi University School of Dentistry, Gifu 501-02, Japan. Correspondence at present address: M. Onozuka, Department of Anatomy, Gifu University School of

Medicine, Gifu 500, Japan. 0168-0 02/86$03.50 9 1986 Elscvier Scientific Publishers Ireland Ltd.

38 elucidated. Our previous studies on the snail neurons (Euhadra peliomphala) demonstrated that the bursting activity induced by a convulsant agent, pentylenetetrazole (PTZ), is caused by a direct action of this agent on the neuronal membrane 2s, and accompanied by the development of a negative slope resistance (NSR) region in the current-voltage (I-V) relation of the neurons 29, the I-V curve being characteristic of the bursting neurons of Aplysia 34. In addition to such effects on the membrane properties, we found that PTZ promotes Ca 2 + release from the intracellular storage sites, the lysosome-like granules (LLGs) 3~ by mediating the increased intracellular cyclic AMP via an activation of adenylate cyclase25, and that it stimulates protein kinase activity 25 and protein phosphorylation, in a Ca 2 +-dependent manner 26. It has been shown that vasopressin, a neurohypophysial hormone synthesized in the vertebrate hypothalamus, is capable of causing bursting activity in molluscan neurons 2-4. The mechanism has been considered as the result ofdeveloping NSR region in the steady-state I-V curve measured under voltage clamp 4. These findings suggest the presence of a certain pharmacological or biochemical effect common to both PTZ and AVP. This assumption led us to repeat with AVP the experiments done with PTZ in Euhadra neurons. In the present paper, we describe the effects of AVP on the intracellular cyclic AMP level, adenylate cyclase and protein kinase activities and Ca 2+ release from the LLGs, as well as its effects on membrane properties. MATERIALS AND METHODS

Arginine vasopressin (AVP), calf thymus histone, ATP, dibutyryl cyclic AMP and isobutylmethylxanthine (IBMX) were obtained from Sigma Chemicals, [45Ca]CaCI2 (39.4 mCi/mg Ca) and [),-32p]ATP (8.1 Ci/mmol) from New England Nuclear, Whatman GFC filter from Bio-Rad and cyclic AMP assay kits and ACS Scintillator from Amersham. All other chemicals were of analytical grade. The suboesophageal ganglia were isolated from the snail, Euhadra peliomphala, and the overlying sheath was removed by microdissection. The dissected ganglia were used for all subsequent experiments. The normal saline used was the same composition as that described by Kerkut and Gardner ~9. Na § -reduced saline was made by substituting Tris hydroxymethyl aminomethane for Na + and Ca2+-free saline was made by replacing Ca 2 + with an equimolar amount of Co 2 +.

Electrical recording Conventional electrophysiological techniques were used. Experiments were performed on neurons RC-2 and RC-3 in the right caudal cluster of the ganglion 2s. In voltage clamp e~periments, an identified neuron was impaled with two independent 3 M KCl-filled microelectrodes (1-5 Mfl): one for recording membrane potential and the other for passing current. The membrane potential of these neurons was usually clamped at - 50 mV and voltage step commands of 5 s duration were applied. Clamp current was measured at the virtual ground through an I-V converter. Membrane potential was kept at the holding potential for at least 20 s before changing to different potential levels.

39

Measurement of cyclic AMP levels in the ganglia Ganglia were preincubated in normal saline for 20 min at 22 ~ This is necessary to allow cyclic AMP metabolism to return to basal levels after the change caused by dissection and handling 23. Following this incubation, AVP was added and the incubation was continued for an indicated period of time. The subsequent procedure was carried out as previously described 25. Determination of Ca content of LLGs After a 20 min preineubation period at 22 ~ ganglia were incubated in ice-cold normal saline containing 20 FCi of [45Ca]CaCI2 for 40 rnin. AVP was then added and the ganglia were incubated for the indicated time at 22 ~ Following this incubation, the LLG fraction was obtained by our previous method 25. The fraction (about 0.5 ml) was then transferred into scintillation vials containing 0.5 ml of 2% SDS and allowed to stand for 30 rain before adding 10 ml of ACS scintillatbr. Determination of radioactivity was performed by a liquid scintillation counter. "4aCa content", which was calculated from the 45Ca remaining in the LLG fi'action, was used as a rough estimate of calcium content in the LLGs. Protein kinase assay Protein kinase activity was determined by measurement of the incorporation of 32p phosphate from [?-32p]ATP into exogenously added calf thymus histone, according to the method of Levitan and Norman 24. The ganglia were homogenized in 25 mM TrisHCi, pH 7.4 (0.3 ml/ganglion), and the homogenate was centrifuged at 800 x g for 10 rain at 4 ~ to remove nuclei and cell debris. The supernatant (1"--2mg protein/ml) served as the sample for protein kinase. A fresh enzyme preparation was used for each experiment. The assay mixture (200 pl) in test tubes contained at final concentration 50 mM Tris-HCI, 10 mM MgCI 2, 30 FM ATP (cold), 10 mM NaF, 50 Fg calf thymus histone, and about 3 x 105 cpm of [7-32p]ATP (final pH 7.4). This reaction was terminated by addition of 1 ml of ice-cold 2 0 ~ TCA, together with 500 Fg of bovine serum albumin as a co-precipitant. The tubes were allowed to stand at 0 ~ for 60 min, and the precipitate was collected on Whatman GFC filters which were washed 4 times with 5 ml of ice-cold 5% TCA. Radioactivity on the filters was counted with the scintillation counter using an ACS scintillator. Adenylate cyclase assay The ganglia were homogenized in 50 vols. of 2 mM Tris-maleate buffer (pH 7.8) containing 2 mM EGTA using a glass homogenizer. Non-disrupted pieces of connective tissue were removed with forceps, and the homogenate was centrifuged at 50,000 x g for 15 rain. The pellet was washed twice with 2 mM Tris-maleate buffer, pH 7.8 (without EGTA), and finally resuspended in this buffer at a concentration of 0.5 mg of protein per mi. Adenylate cyclase activity in this crude membrane preparation was assayed by the method of Hegstrand et al.17, except that GTP was not added.

40 Proteht assay

Protein was estimated by the method of Hartree I~' using bovine serum albumin as the standard.

I~,ESUH'S

EJfi,ct of A VP on membrane properties

In normal saline, RC-2 and RC-3 neurons usually generated spontaneous regular firing and did not generate bursting activity even with the application of extrinsic current (Fig. IA, top trace). In these neurons, application of 50-500 nM AVP caused bursting activity which is characteristic of slow membrm)e poteiltial oscillations associated with bursts of spikes during the depolarization phase within 1-2 rain (Fig. IA, second to fourth traces). Under voltage clamp such neurons exhibited steady-state inward currcnts over a wide range of membrane potentials as shown in Fig. 1B. The I-V curve in the hyperpohirizing direction from a holding potential was typically monotonous and always had a positive slope. Currents generated by depolalizing conimand steps when examined in the presence of AVP, however, became more inward than that in the A

B

control I(nA) O control 9 50 nH AVP 9 IO0 nM AVP 9 500 nMAVP rinse

SOnMAVP (5min)

_If iIJJml II H IIl ,[I!JrlJl ll!UU IIl I -IO01

100 nM AVP 15 mtn)

500nMAVP (5minl _

lllllmwul l[

!lIlllliilllwUill

rinse (3 hi

-801r

/A

-60

-40

"

I !

0 !

V(mV)

II

III]IIIIIIIIIIJJlIII]III]IIIIIII]]III]I]IHIIIIII]IIIIIIII]IH Fig. I. F,II~cls of AVP on membrane properties ofEuhadra identilicd neurons. At the indicated conccntrathins, AVP was added to the bathing nledium. Resting mcn]brane potentials of the neurons used in tile present and subsequent illustrations, were between - 4 5 and - 4 0 inV. A: .~.VP effect on the electrical activity. B: AVP effect oq the steady-state I - V curve. Current recordings were taken before (control, open circle) and between 5 and 15 min after addition of 50 nM (filled circle), I00 nM (lilled triangle) or 500 nM AVP (filled square), and 3 h aP,er rinsing with the hormone-free normal saline (open triangle). Ilolding potential (Vh) was - 50 inV.

41 control, giving the I-V curve an N-shaped appearance. The magnitude of the inward current was progressively increased with the AVP concentration. The I-V curves more negative than the holding potential of - 50 mV were parallel to the control curve. These effects of AVP were nearly reversible upon rinsing with hormone-free saline (bottom trace in Fig. IA and open triangle in Fig. IB). Below 50 nM, AVP did not affect either the membrane potential or the I-V curve. In subsequent experiments, 100 nM AVP was used unless otherwise specified.

A control

B control

AVP

AVP (Ca-free Col

AVP (ll2Ha)

AVP (1/4Na)

rinse

AVP (Ha-free)

rinse

AVP

l(nA)

4

O control 9 AVP 9 AVP (C,-freeCo)

2~

~/

0

0

Z Vllu

-Z

.

ff/i

4

O control 9 9 AVP (i/2Nl)

AVP

~Itll

9 AW Ii/,,,) o ^,~ ,,.-,~..)

/ff~

. r,.,.

i/ll/J

O~

-o"ffl

-2

f

/~ //~11 II~/I

I1 "

\',,ff

..

Fig. 2. Effects of varying [Ca 2+ },, and INn ~ },, on AVP-induced changes in membrane properties of the neurons. Ca z + .free saline was made by replacing Ca 2 + with Co 2 * and Na + -reduced saline was made by substituting Tris" for Na + , as described in Materials and Methods. A: The effect of Ca 2 ' -free saline. Current recordings were takeu before (control, o p e . circle) and between 4 and 10 rain after perfusion with norm;d saline (filled circle) or Ca 2 * -free saline (filled triangle) containing 100 nM AVP, and 3 h after rinsing with the hormone-free normal saline (open triangle). Vh, - 5 0 inV. B: The effects of stepwise reductions of INn * I,,. Current recordings were taken before (control, open circle) and between 4 and 10 rain after perfusion with normal saline (filled circle), saline reducing [Na + I. to one-second (filled square), to one-fi)urth (filled triangle) or Na * -free saline (open square) containing 100 nM AVP, and 3 h after rinsi*]g with the hormone-free normal saline (open triangle). Vh, - 50 inV. Note: Ca spikes are seen in Na ' -free saline.

42 The ionic mechanism generating the development of NSR in the I-V curve and the bursting activity was studied by reducing extraeellular concentration ofCa 2 § ([Ca 2 + ],,) or Na § ([Na § ]~,), since both Na § and Ca 2§ influxes contribute to the rising phase of the action potential in molluscan neurons'. As shown in Fig. 2A, when Ca 2 +-free saline (substituted by Co 2§ was introduced, AVP-induced bursting activity was followed by prolonged depolarization (compare top right with the second row in A). In the I-V curve this saline apparently increased the amplitude of the NSR (filled triangle in A). In contrast, stepwise replacement of extracellular Na § with equimolar amounts of Tris § decreased the NSR amplitude in a manner corresponding to reduction of [Na § ]o (Fig. 2B, filled symbols and open square). The NSR, however, did not completely disappear even after total substitution of [Na ~ ]o, whereas bursting activity was completely abolished (Fig. 2B: open square; and third trace in left column). If the NSR is essential for the generation of bursting activity and Na § is the main charge carrier for this inward current, the NSR nnd bursting activity will be caused by application of an adequate concentration of veratridine which activates sodium channels 7'22. As expected, veratridine of 1-10/tM initiated bursting activity within 1-5 min after its application (Fig. 3A). Above 10 yM, the predominant effect of this agent was only an increase in depolarization of the membrane potential without bursting activity. In addition, the I-V curve, measured 5-15 min after application of 1-10/~tM veratridine, showed the NSR whose amplitude was dose-dependent (Fig. 3B). Thes'e data suggest that an adequate and slow influx of Na § may be necessary for the generation of bursting activity accompanied by the NSR. A

B

control

4 lull veratridlne i i ~ ii i

,J!I:

l(nA)

,'r

0 control 9 lUM vermtridtne 9 101~ veratrldlne a rinse

,,,,, ,,

lfhlflHhlh1111 l lililj llllllrlrlllUIlilUl , -100 I

lOpH veratrldlne

-80 ,I

-60 -40 I I ~ ~ . 1 ~ / /

~) v(.v)

rinse

Jill Ill llllJJJIll lllll/l l l l Ih JIll ..................................

j ~5~V Ssec

-4 ,./.4/"

e/

k/

-6

Fig. 3. Effects of veratridine ota membrane properties of the neurons. At the indicated concentrations, veratridine was added to the bathing medium. A: veratridine effect on the electrical activity. B: veratridine effect on the steady-state I-V curve. Current recordings were taken before (control, open circle) and between 5 and 12 min after addition of I/aM (filled circle) or 10 AcM veratridine (filled triangle), and 30 s i n after rinsing with the veratridine-free normal saline (open triangle), Vh, - 50 mV.

43 A

B ! (hA)

control

II

!,!

dibutyryl

I!)

l

:I

O control 9 dtbutyru

A,IBMx

cyclic AHP

cyclic AMP (lOmln)

" ~:,',)'JUIIUII

'll) lll)

~ -mTmmrlf/'r ~'-

; $ ....... ~ ,~iml"' i II

.?o

'!

11

-~o

0'

-?o ~" -47

;l/ ~ // ,o'17 ?

~,l.......... ii~If -2

JBMX (10 rntn) -4

-6

12smV2sec

/

Fig. 4. Effectsof the increased cyclicAMP levelon membrane properties of the neurons. To increase the cyclic AMP level,dibutyryl cyclicAMP or IBMX were added to bathing medium at a concentration of 100/4M. A: effectofdibutyryl cyclicAMP or IBMX on the electrical activity. B: effectof dibutyrylcyclic AMP (filledcircle)or IBMX(filledtriangle)on the steady-state I-V curve. The curve linked by open circles indicates recordings taken before addition and the other curves between 10 and 20 rain after addition.

The action of hormones on a variety of target tissue is mediated by cyclic AMP as an intracellular second messenger ~5. Moreover, in Euhadra neurons, the treatments which elevate intracellular levels of this messenger, caused bursting activity 25. We thus examined the effect of such treatments on membrane properties of the neurons. When either 100 llM dibutyryl cyclic AMP or 100/~M isobutylmethylxanthine (IBMX) 24.32 was added to the bathing medium, the neuron exhibited the bursting activity (Fig. 4A) and NSR (Fig. 4B) as did with AVP.

Effect of A VP on cyclic AMP level and adenylate cyclase activity in the ganglia The previous 25 and above findings (Fig. 4) obtained by application of dibutyryl cyclic A M P and IBMX, indicate that cyclic AMP may be a mediator for bursting activity elicited by AVP. In order to confirm this postulation, the effect of AVP on the cyclic A M P levels in the ganglia was examined. The results are shown in Fig. 5. When the ganglia were incubated with I00 nM AVP, the cyclic A M P level increased gradually to reach a maximum between 5 and 10 rain, and then decreased to near the control level by 30 rain (Fig. 5A, open circle). In contrast, its level in the ganglia incubated in the hormone-free saline, showed a gradual and slight increase but there was no significant change throughout the incubation period. We also examined the effect of various concentrations of this hormone on the cyclic A M P level. The AVP of 10-100 nM

44

A 9

L_ I2,

8

.E O

E "5

B

7 6 5 4

I

.,! I

3

oT , o

10

20

30

0

time (mini

10-8

10 - 7

10- 6

concentration (M)

Fig. 5. Effect of AVP on cyclic AMP levels in Euhadra ganglia. A: ganglia were incubated for varing I~eriods oftime in normal saline either with (open circle) or without (filled circle) 100 nM AVP prior to measurement of cyclic AMP levels. B: ganglia were incubated for 10 min in normal saline containing various concentrations of AVP prior to measurement of cyclic A M P levels. Values indicate means + S.E.M. of four experiments.

increased the cyclic AMP in a dose-dependent manner (Fig. 5B). Below I0 nM, ,~VP caused no significant effect and above 500 riM, the cyclic AMP level tended to decrease. The adenylate cyclase activity of a crude membrane preparation was measured in the absence or presence of AVP. AVP markedly stimulated adenylate cyclase activity as shown in Fig. 6. The enzyme activity increased linearly with the increase in time of Avp

.=_ .oo

|

~/ I

r~ I:n

I

"o w

E ~o O.

._ "6

200

100

1o " ~ " ' ~ n

/

0

E D.

0

~'''~ 0

,

,

,-,

5

, 10

9

, 15

time (rain) Fig. 6. Effect of AVP on adenylate cyclase activity in crude membrane preparation from Euhadra ganglia. Adenylate eyelase activity was measured, using 50 Fg ofmembrane protein in a total assay volume of 100 Id, in the absence (filled circle) or presence (open circle) of 100 nM AVP. Values indicate means + S,E,M, of que.druplicate determinations.

45 A

•

0.9;

o E =

0.8

B

[ 1 l T=:g--~.~._.

,

controt

c

0J 0,7

E

0.6

g uO.6

~ 0.4 M

m 0.3

U

"l"

* 0

=

I 9 5

time (rain)

I 10

T

I~, | 0 10 - 8

I

I 10 - 7

concentration

,

I

I 10 - 6

(M)

Fig. 7. Effectof AVPon the releaseofcalciumfromthe LLGs.A: gangliawereincubatedfor varyingperiods of limein normal salineeither with(open circle)or without(filledcircle) 100 nM AVP prior to determination of 4~Ca content in the LLG fraction. B: ganglia were incubated for I0 min in normal saline containing various concentrationsof AVP prior to determinationof 4SCacontent in the LLG fraction. Values indicate means + S.E.M. of four experiments.

incubation with AVP, until it reached a maximal plateau showing significantly higher values than controls. In contrast, AVP caused no significant effect on phosphodiesterase activity (data not shown).

Effect of A VP oll the release of calcium from the LLGs Application ofeither dibutyryl cyclic AMP or PTZ stimulates release of calcium from the LLGs in Euhadra neurons 2s'3~ After preineubation of the ganglia with [45Ca]CaC12, they were incubated either in the absence or presence of AVP. The ganglia were then homogenized and the LLG fraction was obtained by the method previously described 2s. The 45Ca content of the L L G fraction markedly decreased to reach about a half of the control within 5 min after incubation with 100 nM AVP (Fig. 7A). When the time of incubation with AVP was fixed at 5 min, the 45Ca content of this fraction steeply decreased with the increase in AVP concentration up to 100 riM, but no significant decrease with further increase in AVP (Fig. 7B). Effect of A VP on protein kinase activity A possibility that AVP may stimulate some kind o f protein kinase activity was examined since Ca2+-dependent protein kinase activity is stimulated during PTZinduced similar bursting activity25. Protein kinase of the homogenate from the ganglia treated with 100 nM AVP showed a marked increase in activity from 200 to 545 pmol/min/mg protein (Fig. 8A). This stimulatory effect was almost reversible after

46

~ooI. A ~0.

.~ ~

"~ c =

~

"~

m

[ ~ control soo

~

~

rinse B

300 F

2oo~

T

.:::.:::::

Fig. 8. Effect of AVP on protein kinase activity in the absence (A) and presence (B) of I mM EGTA in the reaction mixture. Ganglia were incubated for 10 rain in normal saline either with (shaded column) or without (white column) I00 nM AVP, and rinsed with the hormone-free normal saline for 3 h after AVP treatment. Values indicate means _+ S.E.M. of four experiments.

rinsingwith the hormone-free saline for 3 h. In contrast, when EGTA was added to the reaction mixture, the stimulatory effect of AVP was depressed significantly (Fig. 8B). Similar depressible effect was observed in the presence of the .Ca2 +/calmodulininhibiting agents, N-(6-aminohexyl)-5-chloro-l-naphthalenesulfonamide (W-7) and chlorpromazine (data not shown).

DISCUSSION

In the present experiments attention was focused to compare the changes in the electrical membrane properties and intracellular biochemical changes during AVPinduced bursting activity with those previously observed during PTZ-induced bursting activity. Our results show that the occurrence of AVP-induced bursting activity and the development of the NSR region in the I-V curve, were dependent on [Na + ]o but not on [Ca 2§ ]o. Moreover, this hormone also caused the following intracellular biochemical alterations: (i) elevation in the cyclic AMP levels; (ii)increase in the adenylate cyclase and protein kinase activities; and (iii) activation of the calcium release from the LLGs. In the same neurons used in the present experiments, PTZ has been found to manifest bursting activity with the NSR 29. This effect was also abolished by Na + -free saline (unpublished data). Similarly, Baker et al. 2-4 reported that lysine vasopressin can either induce or enhance bursting activity with the NSR in the neurons of Otala and Aplysia. They proved that sodium ions were a main charge carrier during the bursting activity. It is, therefore, suggested that the increase in sodium conductance leads to a depolarizing phase of the bursting cycle induced by AVP. This assumption is also

47 supported by the observation that veratridine produced similar bursting activity with the NSR in the same neurons. On the other hand, recent studies using the same neuron of Aplysia indicate that the slow oscillation of the membrane potential is caused by two different conductance mechanisms, a voltage-dependent calcium conductance and Ca 2 § potassium conductance linked through oscillation of[Ca z § ]i 13'14. In the present experiments, the replacement of extracellular Ca 2 + with Co z § converted AVP-induced bursting activity to a burst followed by a prolonged depolarizing phase and potentiated AVPinducing NSR in the I-V curve. This observation would imply a progressive decrease in the CaZ+-dependent potassium current during augmented depolarization. In addition, when the ganglia were incubated with AVP, calcium content in the LLG fraction was markedly decreased as seen during PTZ-induced bursting activity2~,3~ Since there is evidence that the post-burst hyperpolarization phase during the bursting cycle is caused by an increase in [Ca 2+ ]i j2, the intracellular Ca 2 § released from LLGs as well as the increased Ca 2 + influx during the depolarizing phase are responsible for the generation of the post-burst hyperpolarization. The question is how AVP generates bursting activity. There are two possible explanations: first, AVP may directly bring about the above-mentioned changes of the membrane properties. Second, AVP induces intraeellular biochemical changes which may lead to the developmen't"~f NSR and bursting activity. Hormones have been shown to modify cyclic nucleotide metabolism in a variety of cells 1~, including brain 5""~ and tissue culture cells derived from brain 6,2~. In the present experiments, AVP caused an elevation i,1 the cyclic AMP levels in the intact nervous system and activated the adenylate cyclase in cell-free membrane preparations. Furthermore, either dibutyryl cyclic AMP or IBMX mimicked both bursting activity and NSR produced by AVP. Since our results thus satisfy most of Greengard's criteria ~5for a physiological response to be attributed to cyclic AMP, we propose that AVP-indueed membrane electrical properties are generated by increased cyclic AMP resulting from AVP-aetivated adenylate cyelase. Drake and Treistman 9 also demonstrated that the external application of cyclic AMP derivatives or phosphodiesterase inhibitor could activate the voltage-dependent sodium conductance of the Aplysia neurons, thus generating the bursting activity. In addition, AVP markedly enhanced protein kinase activity, but this stimulatory effect was decreased nearly to the control level by addition of either EGTA or Ca 2 +/calmodulin-inhibiting agents to the reaction mixture. These facts suggest that AVP stimulated Ca 2 § protein phosphorylation. Previously, we found that similar effects were observed with PTZ 25, and that phosphorylation of 34- and 50-kDa proteins was stimulated by PTZ in a Ca 2 § manner 26. Judging from the observations that calcium ionophore A23187 causes bursting activity followed by long-lasting hyperpolarization and change in protein phosphorylation similar to that induced by PTZ 26, AVP-induced enhancement of protein kinase activity may be due to the increase in [Ca 2+ ]i during the depolarization phase of the bursting

48 cycle, and may be related to the generation of post-burst hyperpolarization. This postulation is also supported by the findings that the perfusion of snail neurons with the catalytic subunit of cyclic AMP-dependent protein kinase increases Ca 2 +-dependent potassium conductance 8, indicating an increase in the affinity of the Ca 2 + receptor protein by phosphorylation. Our recent observation of the Ca 2 ~" receptor protein in Euhadra neurons reveals that the effect of intracellular Ca 2 + on the activation of potassium channels is not due to a direct action of Ca 2 § but is mediated through a Ca 2 § complex (manuscript in preparation). From the above findings, we suggest that AVP-induced bursting activity may be indirectly generated by the following intraceUular processes similar to those described previously on the mechanism generating PTZ-induced bursting activity25"26:AVP binds the receptors in the membrane and consequently activates the adenylate cyclase. The activation ofthis enzyme catalyzes the conversion of ATP to cyclic AMP and inorganic pyrophosphate, followed by an increase in the sodium conductance leading to membrane depolarization which causes the increase in the intracellular release of Ca 2 § and the entry of Ca 2 § across the membrane. The increased calcium ions activate a protein kinase which in turn causes the post-burst hyperpolarization, resulting from the increase in potassium conductance. However, the detailed intraceUular biochemical processes associated with the modification of membrane properties which leads to bursting activity, remain to be elucidated.

ACKNOWLEDGEMENTS

We acknowledge Professor K. Yagi and Professor Y. Fukami for valuable advice in the course of this work and a critical reading of our manuscript. This work was partially supported by a grant from Kanagawa Dental College. REFERENCES I Adams, D.J., Smith, S.J. and Thompson, S.H., Ionic currents in molluscan soma, Annu. Rev. NeuroscL, 3 (1980) 141-167. 2 Barker, J.L. and Gainer, H., Peptide regulation of bursting pacemaker activity in a molluscan neurosecretory cell, Science, 184 (1974) 1371-1373. 3 Barker, J.L., lfshin, M.S. and Gainer, H., Studies on bursting pacemaker potential activity in molluscan neurons. 111. Effects of hormones, Brain Res., 84 (1975) 501-513. 4 Barker, J.L. and Smith, T.G., Peptide regulation of neuronal membrane properties, Brain Res., 103 (1976) 167-170. 5 Borgeat, P., Chavancy, G., Dupont, A., Labrie, F., Arimura, A. and Schally, A.V., Stimulation of adenosine 3' : 5'-cyclic monophosphate accumulation in anterior pituitary gland in vitro by synthetic luteinizing hormone-releasing hormone, Proc. Natl. Acad. Sci. U.S.A., 69 (1972) 2677-2681. 6 Brandt, M., Gu!lis, R.J., Fischer, K., Buchen, C., Hamprecht, B., MorOdr L. and Wflnsch, E., Enkephalin regulates the levels of cyclic nucleotides in neuroblastoma x glioma hybrid cells, Nature (London), 262 (1976) 311-313.

49 7 Catterall, W.A., Activation of the action potential Na § ionophore of cultured neuroblastoma cells by veratridine and batrachotoxin, J, BioL Chem., 250 (1975)4053-4059. 8 DcPcycr, J.E., Cachelin, A,B., Levitan, I.B. and Reuter, H., Ca2+-activated K + conductance in internally perfused snail neurons is enhanced by protein phosphorylation, Proc. Natl. Acad. ScL U.S.A., 79 (1982) 4207-421 I. 9 Drake, P.F. and Trcistman, S.N., Mechanisms of action of cyclic nuclcotides on a bursting pacemaker and silent neuron in Aplysia, Brain Res., 218 (1981) 243-254. 10 Duffy, M.J., Wong, J. and Powell, D., Stimulation of adenylatc cyclase activity in different areas of human brain by substance P, Neurophamlacology, 14 (1975) 615-618. 11 Faugier-Grimaud, S., Extrasynaptic mechanisms of cardiazol-induccd epileptiform activity of invertebrate neurons, Brain Res., 69 (1974) 354-360. 12 Gorman, A.L.F. and Thomas, M.V., Changes in the intracellular concentration of free calcium ions in a pace-maker neurone, measured with the mctallochromic indicator dye arsenazo Ill, J. Physiol. (LondmO, 275 (1978) 357-376. 13 Gorman, A.L.F., Hermann, A. and Thomas, M.V., Intracellular calcium and the control of neuronal pacemaker activity, Fed. Proc. Fed. Am. Soc. Exp. BioL, 40 (1981) 2233-2239. 14 Gorman, A.L.F., Hermann, A. and Thomas, M.V., Ionic requirements for membrane oscillations and their dependence on the calcium concentration in a molluscan pace-maker neurone,,/. Physiol. (LondotO, 327 (1982) 185-217. 15 Grccngard, P., Possible role for cyclic nucleotides and phosphorylatcd membrane proteins in postsynaptic actions of neurotransmitters, Nature (London), 260 (1976) 101-108. 16 Hartree, E.F., Determination of protein: a modification of the Lowry method that gives a linear photometric response, Anal. Biochem., 48 (1972) 422-427. 17 Hegstrand, L.R., Kanof, P.D. and Grcengard, P., Histamine-sensitive adenylate cyclasc in mammalian brain, Nature (London), 260 (1976) 163-165. 18 Jard, S. and Boekaert, J., Stimulus-response coupling in neurohypophysial peptide target cells, Physiol. Rev., 55 (1975) 489-536. 19 Kerkut, G.A. and Gardner, D.R., The role of calcium ions in the action potentials of Helix axpersa neurones, Comp. Biochem. Physiol., 20 (1967) 147-162. 20 KIte, M.R., Faber, D.S. and Heiss, W.D., Strychnine- and pentylenetetrazol-induced changes ufcxcitability in Aplysia neurons, Science, 179 (1973) 1133-1136. 21 KIee, W.A. and Nirenbcrg, M., Mode of action of endogenous opiate peptidcs, Nature (London), 263 (1976) 609-612. 22 Krueger, B.K. and Blaustein, M.P., Sodium channels in presynaptic nerve terminals, J. Gen. Physiol., 76 (1980) 287-313. 23 Levitan, I.B. and Treistman, S.N., Diurnal rhythms in cyclic nucleotide metabolism in Helix nervous system, J. Neurobiol., 8 (1977) 265-272. 24 Levitan, I.B. and Norman, J., Different effects of cAMP and eGMP derivatives on the activity of an identified neuron: biochemical and electrophysiological analysis, Brain Res., 187 (1980) 415-429. 25 Onozuka, M., Kishii, K., Furuichi, H. and Sugaya, E., Behavior of intracellular cyclic nucleotide and calcium in pentylenetetrazole-induced bursting activity in snail neurons, Brain Res., 269 (1983) 277-286. 26 Onozuka, M., Imai, S. and Sugaya, E., Pentylenetetrazole-induced bursting activity and cellular protein phosphorylation in snail neurons, Brain Res., 362 (1986) 33-39. 27 Piggott, S.M., Kerkut, G.A. and Walker, RJ., The action ofpierotoxin, strychnine, bicuculline and other convulsants and antagonists on the responses to acetylcholine glutamic acid and gamma-aminobutyric acid on Helix ncurones, Comp. Biochem. Physiol., 57C (1977) 107-116. 28 Sugaya, A., Sugaya, E. and Tsujitani, M., Pentylenetetrazol-induced intracellular potential changes of the neuron of the Japanese land snail Euhadra peliomp~ala. Jpn. J. PhysioL, 23 (1973) 261-274. 29 Sugaya, E., Onozuka, M., Usami, M. and Sugaya, A., Ionic distribution changes during bursting activity induced by pentylenetetrazol in a single isolated snail neuron: tentative application of electron probe

50

30 31 32 33 34

X-ray microanalyzer (o a single isolated neuron, In N. Chalazonitis and M. Boisson (Eds.), Abnormal Neuronal Discharges, Raven Press, New York, 1978, pp. 407-418. Sugaya, E. ~,nd Onozuka, M., Imracellular calcium: its movement during pcntyh'nctctrazolc-induced bursting ttctivity, Science, 200 (1978) 797-799. Sugaya, E. arid Onozuka, M., lntraccllular calcimn: its release frOlll granules during bursting activity in snail neurons, Scietlce, 202 (1978) IL95-1197. Trcistman, S.N. and Levitan, I.B., Alteration of electrical activity in molluscan neurollcs by cyclic nucJeotides alld peptidr factors, Nature (LmulotO, 261 (1976) 62-64. Wilson, W.A. and Escucta, A.V., Common synaptic effects of pcntylcnctetrazole and pctficillin, Brain Res., 72 (1974) 168-171. Wilsoa. W.A. at~d Wachtel, H., Negative resistance characteristic esscutial for the mait~tcuancc olslow oscillations in bursting neurons, Science, 186 (1974) 932-934.