Neuromuscular actions of sodium selenite on chick biventer cervicis nerve-muscle preparation

Neuropharmacology Vol. 29, No. 5, pp. 493-501, Printed in Great Britain. All rights reserved

002%3908/90$3.00+ 0.00 Copyright 0 1990Pergamon Press plc

1990

NEUROMUSCULAR ACTIONS OF SODIUM SELENITE CHICK BIVENTER CERVICIS NERVE-MUSCLE PREPARATION Pharmacological

ON

S. Y. LIN-SHIAU,* S. H. LIU and W. M. Fu Institute, College of Medicine, National Taiwan University, Taipei, Taiwan, R.O.C. (Accepted 21 November 1989)

Summary-Sodium selenite was found to be toxic to chicks, with an LD, of 8.5 pg/g, which was increased to 16.3 pg/g by NaCN. The major symptoms of chicks, treated with selenite, were sedation and then dyspnea and paralysis. The cause of death by selenite was apparently due to the respiratory failure. The possible mechanism of toxicity was explored in the isolated chick biventer cervicis nerve-muscle preparation. Selenite initially increased the amplitude of the twitch, reversed the suppression of the twitch caused by d-tubocurarine, Mg2+, Cd*+ or Mn2+ and significantly increased the quanta1 content and amplitude of endplate potentials. Subsequently, selenite depressed the amplitude of the twitch, blocked the axonal conduction and inhibited excitatory postsynaptic potentials. Both NH: and K+ enhanced the action of selenite in depressing the twitches. In addition, selenite induced a sustained contracture of the muscle, which was partially inhibited by removal of external Ca2+ and markedly blocked by EGTA. Entry of Ca*+ and release of the internal Ca2+ were considered to be responsible for inducing contracture by selenite. Pretreatment with trypsin, glutathione (GSH) and cyanide profoundly inhibited the effects of selenite, indicating that the site of action of selenite was on the outer membrane and the binding of selenite to the sulfhydryl groups of membrane proteins was proposed to be an essential step for selenite-induced contracture and neuromuscular action. These findings suggest that neuromuscular blockade and tetanic spasm, produced by selenite in chicks, may play a role in causing respiratory failure in vivo. Ke): words-selenite,

Selenium has been known poisonous. In 1936, Franke

neuromuscular transmission, chick muscle.

for many years to be and his co-workers first

reported effects of “toxic wheat” grown in South Dakota on high selenium soil (Franke, Tully and Paley, 1936). Feeding the toxic wheat or selenium to hens has produced deformed embryos, with eyes and beaks missing, and with distorted wings and feet. High levels of selenium are carcinogenic (Nelson, Fitzhugh and Calvery, 1943; Seifter, Ehrlich, Hudyma and Mueller, 1946; Medina, Lane and Tracey, 1983). However, recognition of selenium as an essential trace element to animals began in 1957; selenium deficiency in rats, characterized by liver necrosis, could be prevented by a dietary supplement of selenite (Schwarz and Foltz, 1957). Selenium had meanwhile emerged as an anticarcinogen (Birt, Lawson , Julius, Runice and Salmasi, 1982; Ip and Sinha. 1981; Lane and Medina, 1985; Horvath and Ip, 1983) and as a component of glutathione peroxidase, which is important in the metabolism of injurious hydroperoxides (Pope, Ganther, Swanson, Hafeman and Hoekstra, 1973; Sunde and Hoekstra, 1980; Stadtman, 1980). At present, much is known about the biochemical and nutritional effects but the toxic and pharmacological actions of selenium are poorly understood. Preliminary toxicity tests of selenite in chicks showed that selenite caused dyspnea and *To whom all correspondence

should be addressed.

paralysis. The toxic mechanism of selenite was therefore studied in the isolated chick biventer cervicis nerve-muscle preparation. The results suggest that blockade of neuromuscular transmission and tetanic spasm, induced by selenite, may play a role in causing respiratory failure. METHODS Assay of toxicity

Sodium selenite in saline solution was injected intrathoracically into chicks aged 3-5 days. More than 6 chicks were used for each test dose. The time from injection to death was recorded and the number of deaths within 24 hr was used for the calculation of the dose causing 50% lethality (LD,,) and slope function, according to the method of Litchfield and Wilcoxon (1949). Chick biventer cervicis nerve-muscle preparations

Chicks aged 3-5 days were used in all experiments. The biventer cervicis nerve-muscle preparation was isolated according to the method of Ginsborg and Warriner (1960). The muscle was suspended in 10 ml of modified Krebs’ solution (130.6 mM NaCl, 4.8mM KCl, 1.2mM MgSO,, 12SmM NaHCO,, 2.5 mM CaCl, and 11.1 mM glucose), which was constantly aerated with 95% 0, + 5% CO, at 37°C. The nerve was stimulated through the tendon with

493

494

S. Y.

LIN-SHIAU et

supramaximal rectangular pulses of 0.5 msec at a rate of 0.1 Hz. In some experiments, the muscle was directly stimulated by a pair of circular platinum electrodes in the presence of 7 PM d-tubocurarine. The contraction of the muscle was recorded isometrically, with a force-displacement transducer (Grass FT.03), on a Grass Model 7 polygraph. Effects of sodium seIenite on neuromuscular mission in the chick muscle

trans-

The paired muscles were isolated from one chick; one for the control treated with selenite alone and the other for the test pretreated for 20 min with various compounds, e.g. 1 mM (NH,)$O,, 1 mM Na,SO,, 10 mM and 15 mM KCl, 2.5 mM glutathione (GSH), 0.005-O. 1 mM NaCN, 1.1 p M d-tubocurarine, 10m4g/ml trypsin, Ca *+-free Krebs’ with or without 1 or 10 mM ethyleneglycol-bis-( /?-amino-ethyl ether)N,N’-tetraacetic acid (EGTA), prior to the application of 0.1 mM Na,SeO, . The EGTA was washed out with Krebs’ solution before application of selenite. The inhibitory action of selenite on neuromuscular transmission was measured by the depression of amplitude of the twitch and the time required for the complete blockade of twitches, as described previously (Lin-Shiau and Fu, 1980). The potency of the contracture-inducing activity of selenite was judged by the latent period, peak tension and the time required to reach the peak tension. The mechanism of action of selenite was elucidated by pretreatment of the chick muscle with the various specific inhibitors mentioned above. The modification of the effects of selenite by the specific inhibitors was statistically analyzed by the changes in the time for complete blockade of the twitches, latent period and the time to reach peak tension of the contracture, as well as the peak tension. Intracellular

al.

Axonal

conduction of the mouse phrenic nerve

The experiments were performed with a modification of the RandiC and Straughan technique (1964). Because the nerve innervating on chick biventer cervicis is surrounded by tendon, it is difficult to record compound axonal action potentials. Therefore, the mouse phrenic nerve-diaphragm preparation was used to explore the action of selenite on axonal conduction. The mouse phrenic nervediaphragm was isolated in an acrylate chamber, containing Krebs’ solution, through which 95% 0, and 5% CO, was bubbled continuously. Connective tissue was carefully dissected from the nerve. The phrenic nerve, with thread attached, was carefully drawn through a needle into an adjacent chamber, filled with liquid paraffin, the nerve was held just above the Krebs’ solution with a micromanipulator. Circular platinum stimulating electrodes were placed in contact with the proximal end of the nerve near the diaphragm in Krebs’ solution for stimulation (with supramaximal rectangular pulses of 0.02 msec duration). The distal (cut) end of the nerve rested on platinum recording electrodes in liquid paraffin. Compound nerve axonal potentials were amplified with a Grass p-16 preamplifier and recorded with a Data 6000 waveform analyzer. Chemicals

All of the chemical compounds listed above for the test experiments were purchased from Sigma Chemical Company, St Louis, Missouri, U.S.A. Statistics

The number of the experiments for each group was more than four and significance was assessed by Scheffe’s test for the pair control and each test group after analysis of variance (ANOVA), for total groups of each experiment.

recording

Conventional microelectrode recording techniques (Fatt and Katz, 1951) were used. Glass microelectrodes, filled with 3 M KCl, had resistances in the range of 6-20 MR. The chick biventer cervicis muscle was placed in modified Krebs’ solution at 37.0 f 0.5”C and gassed with 95% O2 + 5% CO,. An Axoclamp-2 preamplifier and Hitachi V-352 oscilloscope were used for recordings. The amplitude of endplate potentials (EPPs), elicited at 1 Hz was determined in a medium containing d-tubocurarine (1.5-l .8 PM), with or without 0.1 mM sodium selenite. The quanta1 content of endplate potentials was calculated according to the variance method of de1 Castillo and Katz (1954) with the following formula: Variance = Standard deviation of amplitude of EPP Mean of amplitude of EPP 1 Quanta1 content = (Variance)2

RESULTS

Toxic action of selenite in vivo

Administration of a large dose of sodium selenite (20-30 mg/kg) to chicks produced sedation, followed by dyspnea and paralysis in 1 hr. All the chicks apparently died of respiratory failure in 3-5 hr. A smaller dose of 10-15 mg/kg prolonged the time to death to 8-10 hr. From the dose-lethality curve, the LD, was obtained as 8.5 pg/g and 95% confidence limits ranged from 7.93 to 9.11 pg/g. Sodium cyanide (NaCN), at a sublethal dose of 2 pg/g, increased the LD,, of selenite to 16.3 pg/g with 95% confidence limits in the range 15.8616.75 pg/g. Effects of selenite on neuromuscular

transmission

Selenite slightly increased the amplitude of the twitch of the chick biventer cervicis muscle, evoked by electrical stimulations of the nerve (Figs 3A and 5A). The transient potentiation of the twitch was

495

Neuromuscular actions of selenite

-_.--,--673 g*o 110 150 lb0

45

220

rain

6.03mN NaZSeO3 _ _ B

io

rain

-!!!I59’0

4’5

-- -6’0 7-oio

160

Go min

y-c------85

s’s

9’5

ilo li0

min

Na3Se03

Fig. 1. Effects of NH: and high level of K+ on the neuromuscular blocking action and contractureinducing activity of sodium selenite in the chick biventer cervicis muscle. The nerve was stimulated through the tendon and the contraction of the muscle was recorded. The Krebs’ solution was used throughout the experiments. A. Selenite alone initially inhibited the twitches and then induced a contracture. B. Ammonium sulphate alone transiently increased the amplitude of the twitch. The twitches remained unchanged or were only slightly depressed, after prolonged incubation for 145min. C. Ammonium sulphate markedly enhanced the inhibition of the twitch induced by selenite, and the response to acetylcholine (ACh) remained unaffected after complete inhibition of the twitch. D. Sodium sulphate had no significant effect on inhibition of the twitch induced by selenite. E. Potassium chloride (10 mM) alone markedly potentiated the twitches for more than 160 min. F. Pretreatment with KC1 (10 mM) enhanced the action of selenite in inhibiting the twitches but not in inducing the contracture. G. Potassium chloride (15 mM) inhibited the twitches but had no effect on the selenite-induced contracture. W denotes washout with Krebs’ solution. by depression of the twitch. The time required for inhibition of the twitch was inversely proportional to the concentration of selenite (Fig. 2). The inhibition of the twitch by selenite was enhanced by 1 mM ammonium sulphate [(NH,),SOJ and 10 mM potassium (K+ ) but not by 1 mM sodium sulphate (Na,SO,) (Figs lC, D and F). In the control experiments, both 1 mM (NH,),SO, and 10 mM K+ markedly increased the amplitude of the twitch which then declined slightly (Figs 1B and E). After complete blockade of the twitch to indirect nerve stimulation by selenite, the response to acetylcholine remained unaffected (Fig. 1C). followed

The increase in the amplitude of the twitch by selenite was more evident right after the suppression of the twitch induced by 1.1 PM d-tubocurarine, 3 mM magnesium (Mg*+ ), 0.03 mM cadmium (Cd2+ ) or 0.5 mM manganese (Mn*+) (Fig. 3). It was noted that twitch suppression by Cd*+ and Mg2+ was completely reversed but that by d-tubocurarine and Mn2+ was only partially reversed by selenite. This reversing effect lasted for only 20-30 min and then the inhibitory action of selenite on twitches and contracture of the muscle appeared. By contrast, caffeine, which markedly potentiated the amplitude of twitches to 171.6 f 6.7% of control, did not or

S. Y. LIN-SHIAUet al.

496

selenite- and caffeine-induced contracture to about the same extent (Table 2). However, a greater concentration of EGTA (lOmM), which almost abolished the caffeine-induced contracture, only depressed selenite-induced contracture to 33% of control (Fig. 5 and Table 2). Prolonged treatment with 30 mM EGTA for 2 hr further suppressed the seleniteinduced contracture to 26% of control (peak tension, 0.6 f 0.1 g, n = 4). No,SeO,

trnF.4)

Fig. 2. Inhibition of the twitch by sodium selenite (Na,SeO,) in the chick biventer cervicis muscle. The twitch was evoked by electrical stimulation of the nerve of the chick muscle. The time required for complete blockade of the twitches, induced by sodium selenite, was inversely proportional to the concentration, ranging from 0.03 to 3 mM. The significance (*) was tested by ANOVA and then by Scheffe’s test, as compared with that of either 0.03 or 3 mM Na,SeO,. only slightly, reversed the inhibitory

action of Cd*+ (n = 4); a subsequent application of selenite dramatically increased the twitches to 168.7 k 24.7% of control (n = 4, Fig. 3F). Electrophysiological studies showed that 0.3 mM selenite significantly increased the amplitude, from a control value before treatment with selenite of 2.6 f 0.2 mV (n = 62 from 4 preparations) to 5.7 f 0.4 mV (n = 20 from 4 preparations) after treatment with selenite for 10-15 min and the quanta1 content from 44.3 + 1.4 to 84.2 &-2.9 of endplate potentials (EPPs), respectively. After 20 min, selenite gradually inhibited the endplate potentials. Measurement of compound action potentials of the mouse phrenic nerve revealed that 0.1 mM selenite depressed the amplitude to about 50% of the control in 40 min. Moreover, 0.3 mM selenite decreased the membrane potential of the chick muscle from 71.1 f 1.2 mV (n = 50 from 4 preparations) to 58.7 f 1.4 mV (n = 30 from 4 preparations) in 15 min. Selenite-induced contracture of the chick muscle Selenite not only affected electrically-evoked twitches but also induced contracture of the chick biventer cervicis muscle (Fig. 1A). The initial phase of the contracture was induced after incubation for 34.8 + 2.9 min with 0.1 mM selenite, followed by a bigger second phase of contracture, which reached a peak tension at 126 f 11 min (Fig. 3A and Table 1). Selenite induced the contracture of chick muscles in a concentration-dependent manner (Fig. 4). High levels of K+ (10 mM) enhanced the contracture by shortening the latent period (Table l), while removal of extracellular Ca2+ from the Krebs’ solution inhibited the first phase of selenite-contracture but not the caffeine-induced contracture (Fig. 5 and Table 2). In the paired study, as listed in Table 2, selenite-induced contracture appeared to be more resistant to the antagonistic action of EGTA than caffeine-induced contracture. The drug EGTA (1 mM) inhibited both

Eflect of selenite on caffeine- and high K +-contracture As shown in Figure 6, selenite markedly increased the peak tension of caffeine-induced contracture from 2.3 + 0.5 g to 3.2 + 0.3 g (n = 4). By contrast, contracture induced by high K+ (35 mM) was not significantly affected by selenite; peak tensions of the control and selenite-pretreated preparations were 2.7 + 0.1 g and 2.8 f 0.1 g, respectively (n = 4). On the other hand, neither caffeine nor K+ suppressed selenite-induced contracture, which appeared normally after treatment with caffeine and K (Figs 6C and E). Eflects of treatment NaCN

with trypsin, glutathione and

Pretreatment with trypsin (0.1 mg/ml) for 50 min and then washout of the trypsin only slightly depressed the amplitude of the twitch. Treatment with trypsin did not affect selenite-induced depression of the twitch but markedly inhibited selenite-induced contracture (Fig. 5). On the other hand, pretreatment with glutathione (2.5 mM), not only antagonized the neuromuscular blocking action of 0.1 mM selenite (the time to complete blockade of twitch responses to nerve stimulation was greater than 160 min, as compared with that of control, 41.8 f 2.3 min, n = 4) but also abolished the contracture-inducing activity of selenite. Sodium cyanide (0.005-O. 100 mM) was even more potent than glutathione in antagonizing the effects of selenite. Figure 7 and Table 1 show that NaCN antagonized selenite in a concentrationdependent manner. DISCUSSION

In this study, it was demonstrated that selenite was toxic to chicks and caused dyspnea and paralysis. In order to explore the possible mechanism of the in uivo effects of selenite, the effect of selenite was investigated on the isolated chick biventer cervicis nervemuscle preparation. Selenite produced an initial transient increase in the amplitude of the twitch, followed by a gradual depression of the twitch. In addition, selenite was capable of inducing a sustained contracture of the chick muscle. The increase in the amplitude of the twitch by selenite was more prominent right after the suppression of the twitch, caused by d-tubocurarine, Mg*+, Cd*+ or Mn2+. d-Tubocurarine is known to compete with acetylcholine for the receptors (Zaimis, 1959)

Neuromuscular actions of selenite

A 16 210 %.3 Ml Na2SeO3 l

io s 0.5 xl@4 8.3

i3 io

Qo ab lie i20

35’ 45

45

70

60

90

120

mM m2sef33

-2

min

min

--

li5.185

min

5mMcEtffeine Fig. 3. Reversal by sodium selenite of suppression of the twitch induced by d-tubocurarine (d-TC), Mgz+, Cd*+ and Mn’+ in the chick biventer cervicis muscle. The twitches were evoked by electrical stimulation of the nerve of the chick muscle in Krebs’ solution. Note that inhibition of the twitch by d-TC (B), Mg’+ (C), Cd’+ (D) and Mn*+ (E) was rapidly r eversed by sodium seienite. Aithou~ caffeine itself increased the amplitude of the twitch but was not as effective as selenite in reversing the suppression of the twitch induced by Cd2+ (F). W denotes washout with Krebs’ solution.

and Mg*+, Cd* and Mn*+ all inhibited release of transmitter. Therefore, the reversal by selenite of the suppression of the twitch was apparently mediated by a presynaptic action, increasing the release of trans-

mitter (Balnave and Gage, 1973; Forshaw, 1977; Toda, 1976; Lin-Shiau and Fu, 1980). In accordance with this finding, selenite significantly increased the amplitude and quanta1 content of the endpiate

Table I. Effects of high levels of K’, cyanide and trypsin on the selenite-induced contractwe of the chick biventer cervicis muscle Treatment

”

Control

7

High K- (1OmM) High K’ (15 mM) Cyanide (0.005 mM) Cyanide (0.01 mM) Cyanide (0.1 mM) Trypsin (0.1 mg/ml)

4 3 4 4 3 4

Peak tension (a)

Latent period (mint

2.1 i.O.3 2.2 * 0.4 2.2 i_ 0.3 2.0 + 0.1 1.9 + 0.2* 0.6 + 0.02’ 0.6 rt 0.07’

34.8 + 2.9 19.3 i: 2.9 5.8 * 0.4* 44.8*2.i* 48.0 * I .5* 202.0 2 23.2’ 12.0 + 3.11

Time to peak tension (min) 126.0 t 116.0 + 98.3 2 150.0 rt 171.3 + 377.3 + 46.1+

Il.5 3.1 10.9 5.8 .5.9* I .8* 1.8’

Data are presented as Means k SE. ‘P < 0.05 as compared with control, by Scheffe’s test after analysis of variance for all groups.

Time to block of twitch (min) 41.8 + 2.3 24.6 T 1.2’ 77.0 * 10.4’ 107.3 f 3.7’ 228.0+6.l* -

S. Y.

498 240

r

LIN-SHIAU

4

al.

et

Table 2. Inhibitory action of low levels of Ca2+ and EGTA on selenite- and caffeine-induced contractme of the chick biventer cervicis muscle

E CL

Peak tension (g)

6

Treatment

n

s

Control

1

P

Low Ca2+ (10h6M) EGTA (I mM) EGTA (10 mM) EGTA (30 mM)

3 3 4

2.g

0

Fig. 4. Contracture-inducing activity of sodium selenite (Na,SeO,) in the chick biventer cervicis muscle. A-A,

Peak tension; m-0, time required for peak tension; O-0, latent period for inducing a contracture. The bar indicates SE of the means of 6 experiments. The significance (*) of difference between the data of each curve was tested by ANOVA and then Scheffe’s test, as compared with that of either 0.03 or 3 mM Na, SeO,.

By contrast, caffeine was more potent than selenite in increasing the amplitude of the twitch, but caffeine was not so effective in reversing the suppression of the twitch caused by Cd2+, also indicating that the site of action of selenite might be different from that of caffeine. It is known that Cd’+ suppresses the amplitude of the twitch by the inhibition of release of potential.

-_

~~~~b--2b

Caffeine (15 mM) 2.3 f 0.5 2.5 f 0.3 I.1 fO.2’ 0.04 f 0.02’ 0

transmitter. Thus, selenite inhibited the action of Cd2+ through an increase in release of transmitter at a presynaptic site. The depression of the twitch induced by selenite, was enhanced by NH: and KC. After complete depression of the twitch, responses to acetylcholine remained unaffected, suggesting that selenite blocked the twitches presynaptically. Ammonium has been shown to be a cation possessing a considerable permeability to the K+ channel and can carry an ionic current across the membrane (Binstock and Lecar, 1969; Benjamin, Okamoto and Quastel, 1978; Iles and Jack, 1980). Thus, similar to K+, NH: is a depolarizing agent. In the present study, it was shown

--

S

2.1 f 0.3 1.3 * 0.3* 1.0f0.1’ 0.7 + 0.031 0.6+O.l*

Data are presented as Means + SE. ‘P < 0.05 as compared with control by Scheffe’s test after analysis of variance for all groups.

Naa Se03 (mtvl)

A

Selenite (0. ImM)

j5

5b

60 is

do i20

i30 i42

--

--

iso

i7o tin

1;5

rain

i35

min

Q.ltnM Na2SeO3

:o

i

5t)

lO-+tl Txypsin

‘0 6.0 %.llnM Na2Se03

9’5

ii5

-C ---

40

6’0

io iooilo

i20

Na$%O3

il

;0

CZ-f ree g 1omM (Casfree) EGTA

;o 65 t

ii0

1.30

150

ieo

;OO ruin

0. lrnbl Na2SeO3

Fig. 5. Inhibition by trypsin, low levels of Cal+ and EGTA of selenite-induced contracture in the chick biventer cervicis muscle. Twitches were elicited by electrical stimulation of the nerve of the chick muscle. Pretreatment with trypsin (B), Car+-free Krebs’ solution without (C) or with IOmM EGTA (D) all inhibited the selenite-induced contracture. Note that Ca 2+-free medium plus EGTA, initially induced a contracture (D). W denotes washout with Krebs’ solution.

Neuromuscularactions of selenite

194-

a/----= t.f2b

Sti. off

;s

499

. rf--&

80

6b

5~ Caffeine

ja_free t P ld

70 55

,

min

15lNl EGTA(Ca-fr==)Caffeine

Na2SeOj Caffeine /--

Fig. 6. Effects of sodium selenite on caffeine- and high levels of KC- of the chick biventer cervicis muscle. The twitches were elicited by electrical stimulation of the nerve of the chick muscle and the stimulation was turned off prior to the addition of either caffeine or high levels of K+. Caffeine-induced contracture (A) was almost abolished by 10 mM EGTA (IS) and markedly potentiated by selenite (C). By contrast, contracture induced high levels of K+ (compare D and E) was not affected by selenite.

that both 1 mM NH: and IOmM K+ prominently increased the twitch responses of the chick muscle, evoked by electrical stimulation of the nerve and a greater concentration of 15 mM K+ initially increased and finally depressed the twitches. These observations implicate that selenite, NH: and K+ exerted a similar effect at the presynaptic site, probably depolarizing the nerve membrane. Release of transmitter could be increased by the depolarization of the nerve terminals but persistent depolarization caused by the failure in conduction in reponse to the electrical stimulation of the nerve. Since the motor nerve of the chick biventer cervicis muscle was not accessible for the recording of action potentials, the effect of selenite was tested on the compound action potentials of the mouse phrenic nerve. The result obtained was in agreement with the proposition that

selenite blocked the axonal conduction, possibly due to the persistent depolarization. Although it was not known whether selenite could depolarize the nerve terminals, it was shown that seienite significantly depolarized the muscle membrane. In addition to the effects on the nerve, selenite also induced a sustained contracture of the muscle through a direct action on the muscle membrane. Neither d-tubocurarine, Mg2’, Cd2+ nor MnZ+ inhibited the selenite-induced contracture, indicating that the ~lenite-induced contracture was not produced through the release of acetylcholine. Predepotarization of the muscle membrane by high levels of K+ also had no effect on the selenite-induced contracture, suggesting that selenite did not induce the contracture through the depolarization of the muscle membrane. Removal of external Ca2+ and

S.Y. LIN-SHIAU et al.

500

.

lp

I Smin

Fu-

-

--

-_

--c-

~~~ io

35

50

6'0 is 9b ii0 i30

1h2

ii0 rain

%.lmM Na2SeO3

b 2'0 s .5mM GSH

6

d0

6b

2' 4.0 6b 8 .lmM Na2SeO3

E2.5mM GSH

S'O

100

i20

140

lko

rain

i~~~~~.~~------

.

4b

8b ilo i4piolE;o

260 ;2gio 2'80 j20 370 iO0 min

8 .lmM NoCN O.lmM

Na2SeO3

Fig. 7. Antagonism by glutathione and NaCN of inhibition of the twitch and contracture of the chick biventer cervicis muscle, induced by selenite. Glutathione (2.5 mM GSH) gradually depressed the twitches (B) and selenite did not exert a further effect on the amplitude of the twitch (C). Sodium cyanide (0.1 mM) markedly inhibited the effects of selenite. treatment with EGTA inhibited the selenite-induced contracture; the first phase of the selenite-induced contracture appeared to be induced by an increase in entry of Ca2+, since it was inhibited by removal of external Ca2+. In the concentrations ranging from 1 to 30mM, EGTA inhibited the selenite-induced contracture to a less extent than it inhibited the caffeine-induced contracture. Moreover, pretreatment with selenite markedly potentiated the caffeineinduced contracture. Contracture induced by caffeine is known to be mainly induced by releasing Cal+ from the sarcoplasmic reticulum (Weber and Herz, 1968; Endo, 1977). Thus, it is conceivable that selenite induced the contracture of the muscle, partly by releasing Ca 2c from the sarcoplasmic reticulum. Unlike caffeine, selenite may also cause the release of more firmly bound Ca2+, since the selenite-induced contracture was more resistant to the inhibitory action of EGTA, than was the caffeine-induced contracture. The possible site of action of selenite was explored by treatment with trypsin, glutathione and cyanide. Trypsin is a proteolytic enzyme, capable of modifying the membrane proteins on the superficial sites (Sellinger, Borens and Nordrum, 1969). Both glutathione and cyanide are capable of interacting with the sulthydryl groups of the membrane and binding

to the membrane proteins, respectively (Kraus and Ganther, 1980; Kraus, Prohaska and Ganther, 1980). Indeed, treatment with trypsin markedly inhibited the selenite-induced contracture, suggesting that the site of action of selenite may be on the superficial membrane. Profound antagonistic actions of glutathione and cyanide, on the effects of selenite also supported this idea that the binding of selenite to the sulfhydryl groups of membrane proteins was essential for eliciting the pharmacological actions of selenite. In this study, the potent antagonistic action of cyanide on the action of selenite, both in vivo and in vitro was demonstrated. In conclusion, selenite was toxic to chicks both in vivo and in vitro. Studies on the isolated nerve-muscle preparation showed that selenite transiently increased and then decreased the amplitude of the twitch and finally induced a sustained contracture of the muscle. A presynaptic site, for modifying the release of transmitter and a direct action on the muscle membrane, for inducing a contracture possibly, at least in part, account for the respiratory failure induced by selenite in vivo. Acknowledgement-This work was supported by research grants from National Science Council of the Republic of China (NSC75-0412-B002-62 and NSC74-0412-B002-101).

Neuromuscular REFERENCES Balnave R. J. and Gage P. W. (1973) The inhibitory effect of manganese on transmitter release at the neuromuscular junction of the toad. Br. J. Pharmac. 47: 339-352. Benjamin A. M., Okamoto K. and Quastel J. H. (1978) Effects of ammonium ions on spontaneous action potentials and on contents of sodium, potassium and chloride ions in brain in vitro. J. Neurochem. 30: 131-143. Binstock L. and Lecar H. (1969) Ammonium ion currents in the squid giant axon. J. gen. Physiol. 53: 342-361. Birt D. F., Lawson T. A., Julius A. D., Runice C. E. and Salmasi S. (1982) Inhibition by dietary selenium of colon cancer induced in the rat by bis (2-oxopropyl) nitrosamine. Cancer Res. 42: 44552459. Del Castillo J. and Katz B. (1954) Ouantal comuonents of the endplate potential. J. >hysiol.-124: 560-573. Endo M. (1977) Calcium release from the sarcoplasmic reticulum. Physiol. Rev. 57: 71-108. Fatt P. and Katz B. (1951) An analysis of the endplate potential recorded with an intracellular electrode. J. Physiol. 115: 320-370. Forshaw P. J. (1977) The inhibitory effect of cadmium on neuromuscular transmission in the rat. Eur. J. Pharmac. 42: 371-377. Franke K. W., Tully W. C. and Paley W. E. (1936) Monstrosities produced by injection of selenium salts into hens’ eggs. Anat. Rec. 65: 15-22. Ginsborg B. L. and Warriner J. (1960) The isolated chick biventer cervicis nerve muscle preparation. Br. J. Pharmac. 15: 410-411. Horvath P. M. and Ip C. (1983) Synergistic effect of vitamin E and selenium in the chemoprevention of mammary carcinogenesis in rats. Cancer Res. 42: 533555341. Iles J. F. and Jack J. J. B. (1980) Ammonia: assessment of its action on postsynaptic inhibition as a cause of convulsions. Brain 103: 5555578. Ip C. and Sinha D. (1981) Anticarcinogenic effect of selenium in rats treated with dimethylbenz(a)anthracene and fed different levels and types of fat. Carcinogenesis (Lond.) 2: 435438. Kraus R. J. and Ganther H. E. (1980) Reaction of cyanide with glutathione peroxidase. Biochem. biophys. Res. Commun. 96: 1116-I 122. Kraus R. J., Prohaska J. R. and Ganther H. E. (1980) Oxidized forms of bovine erythrocyte glutathione peroxidase. Cyanide inhibition of a 4-glutathione-4-selenoenzyme. Biochem. biophys. Acta 615: 19-22.

actions

of selenite

501

Lane H. W. and Medina D. (1985) Mode of action of selenium inhibition of 7,12-dimethylbenz(a)anthraceneinduced mouse mammary tumorgenesis. J. natn. Cancer Inst. 75: 675679. Lin-Shiau S. Y. and Fu W. M. (1980) Effects of divalent cations on neuromuscular transmission in the chick. Eur. J. Pharmac. 64: 259-269. Litchfield J. T. and Wilcoxon F. (1949) A simplified method of evaluating doseeffect experiments. J. Pharmac. exp. Ther. %: 99-113. Medina D., Lane H. W. and Tracey C. M. (1983) Selenium and mouse mammary tumorgenesis: an investigation of possible mechanism. Cancer Res. 43: 24605-24645. Nelson A. A., Fitzhugh 0. G. and Calvery H. 0. (1943) Liver tumors following cirrhosis caused by selenium in rats. Cancer Res. 3: 230-236. Pope A. L., Ganther H. E., Swanson A. B., Hafeman D. G. and Hoekstra W. G. (1973) Selenium: Biochemical role as a component of glutathione peroxidase. Science 179: 588-590. Randic M. and Straughan D. W. (1964) Antidromic activity in the rat phrenic nerve-diaphragm preparation. J. Physiol. 173: 13&148. Schwarz K. and Foltz C. M. (1957) Selenium as an integral part of factor 3 against dietary necrotic liver degeneration. J. Am. Chem. Sot. 79: 3292-3293. Seifter J.. Ehrhch W. E.. Hudvma G. and Mueller G. (1946) Thyroid adenomas in’rats receiving selenium. Science 103: 762. Sellinger D. Z., Borens R. N. and Nordrum L. M. (1969) The action of trypsin and neuraminidase on the synaptic membrane of brain cortex. Biochim. biophys. Acta 173: 185-191. Stadtman T. C. (1980) Selenium-dependent enzymes. Ann. Rev. Biochem. 49: 93-110. Sunde R. A. and Hoekstra W. G. (1980) Structure, synthesis and function of glutathione pkroxidase. Nutr. Rev. 38: 265-273. Toda N. (1976) Neuromuscular blocking action of cadmium and manganese in isolated frog striated muscle. Eur. J. Pharmac. 40: 67-75. Weber A. and Herz R. (1968) The relationship between caffeine contracture of intact muscle and the effect of caffeine on reticulum. J. gen. Physiol. 53: 750-759. Zaimis E. J. (1959) Mechanisms of neuromuscular blockade. In: Curare and Curarelike Agents (Bovet D., Bovel-Nitt F. and Marini-Bettolo G. B., Eds), pp. 191-203. Elsevier, Amsterdam.