European Journal of Pharmacology, 178 (1990) 151-159
151
Elsevier EJP 51215
Pre- and postjunctional effects of N-ethylmaleimide in the isolated mouse vas deferens Michael Kaschube and Helmut Brasch Department of Pharmacology, Medical University Liibeck, Lftbeck, F.R.G. Received 7 August 1989, revised MS received 23 November 1989, accepted 2 January 1990
The influence of N-ethylmaleimide (NEM) on contractions due to exogenously applied noradrenaline and bethanechol and on the inhibitory effects of clonidine, of the enkephalin derivative, FK 33-824, and 2-chloroadenosine (2-CLA) on field stimulation-response curves and [3H]noradrenaline ([3H]NA) release was studied in the isolated mouse vas deferens. Exposure to NEM (60/~M: 10 min) caused a 30% reduction of the maximal contraction due to NA but nearly abolished the response to bethanechol. NEM partially reversed the depression of the pulse width-response curves by clonidine and FK 33-824 but was without effect with 2-CLA. The contractions evoked by stimulation frequencies above 20 Hz were depressed by NEM both in presence and absence of the agonists. NEM diminished the inhibition of the stimulation-evoked release of [3H]NA by the three agonists. The prejunctional effect of NEM was markedly influenced by the stimulation parameters. These findings support the suggestion that the inhibition mediated by a2-adrenoceptors, /~- and Pl-receptors in the mouse vas deferens is NEM-sensitive and possibly transmitted by a pertussis toxin-sensitive G-protein. Vas deferens; (mouse); N-Ethylmaleimide; Clonidine; 2-Chloroadenosine; Idazoxan; FK 33-824; (Stimulation parameters)
1. Introduction The SH-alkylating agent, N-ethylmaleimide (NEM), has been reported to interrupt the signal transduction of inhibitory prejunctional receptors located on sympathetic neurons in the central nervous system (Allgaier et al., 1986). N E M has been claimed to display some specificity for pertussis toxin-sensitive G-proteins under certain conditions (Jakobs et al., 1982; Allgaier et al., 1987). Though the isolated rodent vas deferens has been used frequently to study drugs interfering with sympathetic transmission, the effect of N E M
Correspondence to: M. Kaschube, Institut fiir Pharmakologie, Medizinische Universit~it zu Ltibeck, Ratzeburger Allee 160, D-2400 Liibeck, F.R.G.
on prejunctional inhibition has not been studied in this tissue. Only Y a m a m o t o et al. (1973) reported on N E M effects on contractile responses to hypogastric nerve stimulation in the isolated vas deferens of the guinea pig. The present study was therefore designed to characterize the influence of N E M incubation on the inhibitory effects of the a2-agonist, clonidine, the tt-agonist, F K 33-824, and the Pl-agonist, 2-chloroadenosine (2-CLA), on the mechanical response and [3H]noradrenaline ([3H]NA) release in response to field stimulation of the mouse vas deferens. We made use of stimulation-response relations to analyse the N E M effect because prejunctional drug effects can be modulated by field stimulation parameters (Zetler and Kaschube, 1985; Kaschube and Zetler, 1989). Since the mechanical response to field stimulation can be affected by both pre- and postjunctional
0014-2999/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
152 actions of a drug, we included experiments on the N E M effect on contractions evoked by exogenously applied agonists acting directly on the smooth muscle.
2. Materials and methods The experimental conditions were essentially the same as described previously (Zetler and Kaschube, 1985; Kaschube and Zetler, 1989). In short, vasa deferentia dissected from freshly killed N M R I mice of 30-40 g body weight were carefully cut free from their mesenteric sheaths. For measuring mechanical responses the vasa were mounted isometrically under a pre-load of 1 m N in a 5-ml organ bath containing a modified Krebs solution (Hughes et al., 1975) gassed with carbogen (95% O 2 and 5% CO2). The bath temperature was kept at 32°C. In some experiments the normal Ca 2+ concentration of 2.54 m M was lowered to 0.2 m M without correction of osmolarity.
approximately 200 mA). Each experiment started with a 1-h equilibration period during which the vas was stimulated with 5 Hz, 0.1 ms (basic stimulation). Three stimulation-response curves (R1, R2 and R3) were obtained for each vas in one experiment by stepwise increases in either the frequency at a constant pulse width of 0.1 ms (3, 5, 10, 20, 50, 80 and 100 Hz) or the pulse width (0.04, 0.06, 0.1, 0.2, 0.3 and 0.5 ms) at a constant frequency of 15 Hz. It should be noted that, with this stimulation protocol with increasing frequency, the number of pulses in the 1-s train also had to be increased. Intervals of 10 min (R1-R2) or 15 rain (R2-R3) were allowed between the three stimulation series. N E M 60 /~M was added to the organ bath between R2 and R3 and was left in contact with the tissue for 10 min. The bath solution was changed by flushing immediately after R1 and R2 as well as after 10 min of N E M exposure. The drugs were added in microliter amounts of aqueous solutions after R1, R2 and N E M incubation. 2.3. [3H]NA release
2.1. Mechanical response to noradrenaline and bethanechol After a 1-h equilibration period two successive non-cumulative concentration-response curves for either agonist (NA: 0.6-240 /~M; bethanechol: 2200/~M) were determined, the first before and the second after exposure of the vas to 60/~M N E M for 10 min and subsequent washout of the alkylating agent. The concentration was increased at 5-min (with noradrenaline) or 10-min (with bethanechol) intervals and the agonist was left in contact with the organ for 0.5 min. The maxima of the resulting phasic contractions were evaluated. The ECs0 values and their 95% confidence limits for N A and bethanechol were calculated from log concentration-effect curves after linearization by logit transformation and estimation of the regression function by the least squares method. 2.2. Mechanical response to field stimulation Twitch responses were induced by field stimulation applied every 30 s as 1-s trains of monophasic square pulses of constant amplitude (28 V,
Both vasa from one mouse were tied in parallel and pre-incubated at 3 2 ° C for 1 h with DL-[73H]noradrenaline ([3H]NA; 16.7 C i / m m o l ) . The resulting activity was 1 # C i / m l . The vasa were then transferred to a 5-ml organ bath and mounted under a pre-load of 5 m N as described above. The modified Krebs solution contained in addition 3 /~M cocaine-HC1 to prevent neuronal re-uptake. The tissues were thoroughly washed six times during 1 h by flushing the organ bath. For the experiment itself, 1-ml samples of the bath fluid were taken every 2.5 min (exception: 10 min with N E M exposure) and added to 9 ml scintillation fluid (Hydroluma, Baker Chemicals B.V., Deventer, Netherlands). The bath fluid was renewed immediately after sample taking. During the 3rd, 7th and l l t h sampling period (S1, $2, $3) 10 twitch contractions were elicited by field stimulation with 1-s trains of pulses at 15-s intervals. The stimulation parameters were either 15 Hz and 0.5 ms pulse duration (VP) or 100 Hz and 0.1 ms pulse duration (VF), i.e. the sets of parameters that produced maximal contractions in the stimulation-response curves (v.s.). N E M 60 ~ M was ad-
153
ded at the start of the 9th collection period (with the exception of NEM-free control experiments) and was washed out after 10-min contact time. The drugs were added before (7th and 8th sampiing period) and after N E M exposure (10th and l l t h sampling period). In some experiments idazoxan was present throughout the experiment, starting from the 2nd collection period. Finally, the vasa were solubilized in 1 ml Soluene 350 (Packard Instruments, Frankfurt, F.R.G.) for determination of the residual tritium content. The outflow of tritiated compounds was calculated as fractional rate of outflow, i.e. the quotient of tritium-labeled compounds released during a collection period and the tissue content at the start of the collection period. The stimulation-evoked rise in outflow was determined as differences between the fractional rates of outflow of the 3rd, 7th and l l t h sampling period ($1, $2, $3) and the basal outflow of the corresponding pre-stimulation period. This rise in outflow over the base level was taken as an approximation of the stimulationevoked release of [3H]NA. It has been shown previously that after incubation of the isolated mouse vas deferens with [3H]NA the stimulationevoked release of tritiated compounds induced by trains of pulses reflects the release of intact [3H]NA (Marshall, 1983; Stj~irne and Astrand, 1985).
3. Results 3.1. Mechanical response to noradrenaline and bethanechol
To evaluate a possible direct influence on the smooth muscle we studied the effect of N E M exposure on the concentration-response relation of two vas-contracting agonists. N E M 60/zM for 10 min depressed the maximal response to noradrenaline by about 30% and caused a moderate rightward shift of the concentration-response curve (fig. 1A). The ECs0 and 95% confidence interval (n = 8) for N A were found to be 7.32 (5.97-8.97) ~tM before and 11.95 (9.68-14.75) /~M after exposure to NEM. A more marked effect of N E M was encountered with bethanechol (fig. 1B). N E M treatment reduced the maximal response by approximately 90%. ECs0 (n = 8) for bethanechol before N E M was 56.4 (50.1-63.5) /~M. The concentration-relS-
Nor ad re n a l i
mN
2.5. Drugs
DL-[7-3H]Noradrenaline • HC1, 16.7 C i / m m o l (Amersham Int., Amersham, U.K.), clonidine (Boehringer Ingelheim, Ingelheim, F.R.G.); 2chloroadenosine, N-ethylmaleimide, bethanechol chloride, cocaine. HC1 (Sigma); noradrenaline 1hydrogentartrate (Fluka, Buchs, Switzerland); [DAla2,MePhen,Met(0)-olS]enkephalin (FK 33-824; gift of Sandoz, Basel, Switzerland), idazoxan (gift of Reckitt & Colman, Hull, U.K.).
/'I' \J. /
÷
2.4. Statistics
The results are presented as arithmetic means + S.E.M. (n = 5-8). Differences between means were tested for statistical significance by means of Student's t-test. A probability level of P < 0.05 was chosen as threshold of significance.
1
ne
o
.
01
.
.
.
~'~/~ "1 1
........
I 10
........
/
I
,
.
100 ,uM
8rnN 6
./
Bethanechol
.L
/+~T
l.-
2-
~10--¢--¢ 0
, o.-7Y,,,, I
.
.
.
.
i
I0
100
.
.
.
.
I
1000 ~ulut
Fig. 1. Concentration-response curves for noradrenaline and bethanechol (filled symbols) as influenced by 60 /~M NEM (open symbols). Ordinate: force of contraction (mN): abscissa: concentration (/~M). The symbols represent the mean ( _+S.E.M.) of eight experiments performed in duplicate.
154
15
Clonidi ne
iL~.l... L
raN"
10-
//.+ '
' 11~"'1
........
I
ms
1
0.1
15 rnN
~/..~...~
FK 3 3 - 8 2 / ,
1SraN:
to-
10,
5,
|
m
,
,
0.1
,m,|
ms
/+--
IS. mN
l
o/
0¸ 0.1
1
rnN
10. .
0.2 mM CaCl~//J'-&-&
'
ms
1
ms
1
no agonist
10.
C
s. '
'
|'~ .... I 0.1
........ ms
I I
0
0.1
Fig. 2. Effect of exposure (10 min) to 60/~M NEM on pulse width-response curves (frequency: 15 Hz) without addition of agonist (E) and under the influence of clonidine (0.38/~M; A), FK 33-824 (1.36/zM; B), 2-CLA (80/~M; C) and lowered extracellular Ca2÷ (0.2 mM; D). The symbols represent the mean (+ S.E.M.) of five experiments in triplicate. Ordinates: force of contraction (mN). (A-C) al, Control; [3, agonist; A, agonist after NEM; (D) II, control; D, 0.2 mM Ca2+; &, 0.2 mM Ca2+ after NEM; (E) aa, control 1; [~, control 2; A, after NEM. sponse curve for bethanechol after application of N E M was too shallow to allow calculation of the ECs0 value with reasonable accuracy. I n control experiments without N E M , repetitive d e t e r m i n a tion of c o n c e n t r a t i o n - r e s p o n s e curves for either agonist i n one vas did n o t change the m a x i m a l response or the ECs0 (not shown).
3.2. Mechanical response to field stimulation The i n h i b i t o r y effects of clonidine (0.38 /~M), F K 33-824 (1.36 /~M), 2 - C L A (80 /~M) a n d a lower b a t h calcium c o n c e n t r a t i o n o n the stimulation-response curves were critically d e p e n d e n t o n the s t i m u l a t i o n p a r a m e t e r s (figs. 2 a n d 3). Raising
155
zo-: mN 15.
C(onidine
./ /Pff
T05' 0
'
~ "r[~'"'l 0.1
........ ms
I I
20-" mN 15.
20:
FK 3 3 - e Z , ~
J. lJ,
~/./~-'tt
10.
raN:
o
ll-,o,-i=r7 ........ i o
0
0.I
ms
j,...~l
° - ~ ° ~ i - - "
'
0.1
' .....
ms
I 1
20-' 2 - CLA
i_i.j,
e,/~/~ ~
15 10-
C /
0
'
I
zo"
5
coc12
..//;/°/'"/"
,o.
.
mN
0.2 mM
.......
mN
no a g o n i s t
i/~"
,/~',~
is,
,/ ~t-~4 lo. ,/~' s. E
s¢~/
i
I
'
' ' ' ('"1
0
,
,
b ,,,,,i
'
' ' '''"1
0.1 ms 1 0.1 ms 1 Fig. 3. Effect of exposure (10 min) to 60/~M NEM on frequency-response curves (pulse width: 0.1 ms) without addition of agonist (E) and under the influence of clonidine (0.38 #M; A), FK 33-824 (1.36 #M; B), 2-CLA (80/~M; C) and lowered extracellular Ca2+ (0.2 raM; D). Ordinates and symbols are the same as in fig. 2.
the frequency of stimulation at constant pulse width antagonized the inhibition of twitch so that, at 100 Hz, the twitch responses were nearly restored to their control values (fig. 3A-D). N o such effect occurred when the pulse width was increased at constant frequency. In this case the stimulation-response curves were fully depressed by the agonists (fig. 2A-C) and by low calcium (fig. 2D).
Exposing the vasa to 6 0 / ~ M N E M for 10 rain partially reversed the inhibitory effect of clonidine and F K 33-824 on the pulse width-response curves (fig. 2A, B) and on the left part of the frequencyresponse curves (i.e. with frequencies below 20 Hz, fig. 3A, B). However, in the pulse width-response curve, N E M was without apparent effect on the 2 - C L A - i n d u c e d inhibition (fig. 2C). The SH-alkylating agent even enhanced the depression of
156
the pulse width-response curve caused by low extracellular Ca 2+, nearly abolishing the twitch responses (fig. 2D). When the stimulation frequency was raised to 20 Hz and above, NEM diminished the force of contraction with all three agonists as well as with low bath Ca 2÷ (fig. 3A-D). When no agonist was added, N E M had a moderate depressive effect on the pulse width- and frequency-response curves (figs. 2E and 3E). Concentrations of NEM higher than those used in the above experiments ( > 100 /tM) and prolonged exposure to the alkylating agent led to permanent rigour of the vas and a gradual decline of the twitch, finally abolishing the contractions due to field stimulation (not shown) as had been found for skeletal muscle (Kirsten and Kuperman, 1970).
3-
2-
15 Hz, 0.5 ms
A
1-
$1
52
$3
FK 33- 824
FR321-
3.3. [3H]NA release When the stimulation protocol was of the VP type (15 Hz, 0.5 ms) clonidine, F K 33-824 and 2-CLA effectively reduced the stimulation-evoked release of tritiated compounds (fig. 4A-C). Low extracellular Ca 2÷ nearly abolished the evoked release of [3H]NA (not shown). Exposing the vasa to NEM (60 /~M) for 10 min between $2 and $3 antagonized this reduction. Surprisingly, the alkylating agent not only restored the stimulationevoked release but raised it to approximately twice the control level (fig. 4A-C). This latter effect was also found when no prejunctional agonist was added (fig. 4D), but not with low extracellular Ca 2÷ (not shown). In contrast to this strong effect on the stimulation-evoked release, N E M caused only a moderate increase (20-30%) of the spontaneous outflow of tritiated compounds in these experiments (not shown). When the VF type of stimulation was applied, neither the agonists alone nor the agonists applied after NEM exposure influenced the stimulationevoked release of [3H]NA to a statistically significant extent (not shown). To test the hypothesis that interruption of a 2autoinhibition is part of the N E M effect on N A release caused by the VP type of field stimulation we included experiments with the a2-adrenoceptor antagonist, idazoxan. The addition of idazoxan
CIonidine
FR-
FR
no ogonist
3-
B
~r
2-.
| $1
'$2
$3
$I
$2
$3
$2
$3
FR-
FR. 2-CLA
3-
3- control
21
$1
$2
$3
S1
Fig. 4. Effect of the 60 ~ M N E M (10 min) on field stimulation-evoked release of [3H]NA without agonists (D) and after addition of clonidine (A), F K 33-824 (B), 2-CLA (C). The concentrations are the same as in fig. 2. E shows a control series without NEM. Ordinates: stimulation-evoked release expressed as percentage of the vas content (FR). Each column represents the mean ( + S.E.M.) of five observations. (A-C) SI: control; $2: agonist; $3, agonist after NEM. (D) SI: control 1; $2: control 2: $3: after N E M . (E) SI: control 1; $2: control 2; $3: no NEM. Statistical significance of the difference between $2 and $3: ~ P < 0.05; o N.S.
roughly doubled the stimulation-evoked release of [3H]NA (fig. 5A, B). N E M 60/~M did not further enhance this augmentation. The release-enhancing effect of idazoxan on $2 and $3 in control experiments without N E M was not significantly different (fig. 5B). The release-inhibiting effect of F K
157
i
FR. 3-
A21-
jj 51
S2
,ill o
FR-
B 21
,0
$3
S1
$2
$3
ldozoxon
Idozoxon
FR,
3-
C
21-
Sl
oi $2
ldazoxon
S3
D
$1
S2
S3
Idozoxon
FK 33-824 2-CLA Fig. 5. Influence of idazoxan (0.83/~M) on [3HINA release (B) as affected by 60 ~M N E M without agonists (A) and after addition of F K 33-824 (1.36 #M; C) and 2-CLA (80/zM; D). Each column represents the mean ( + S.E.M.) of five observations. Ordinates are the same as in fig. 4. '~ P < 0.05; o N.S. (A) SI: control; $2: idazoxan; $3: idazoxan after NEM. (B) SI: control; $2: idazoxan; $3: idazoxan, no NEM. (C, D) $1: idazoxan; $2: idazoxan+ agonist; $3: idazoxan+agonist after NEM.
33-824 and 2-CLA could still be partially reversed by NEM in presence of the c%-adrenoceptor antagonist (fig. 5C, D).
4. Discussion The disruption of the signal transduction pathway of the az-adrenoceptor and adenosine (Aa)receptor by uncoupling of pertussis toxin-sensitive G-proteins from the receptors has been repeatedly reported for central noradrenergic neurons (hippocampus). These studies involved the application of the protein-alkylating agent, N E M (Allgaier et al., 1986; Fredholm and Lindgren, 1987), as well
as the more specific pertussis toxin (PTX, Allgaier et al., 1985) to inactivate the regulatory G-proteins. In contrast, contradictory results have been presented for the sympathetic neurons of the rat vas deferens, i.e. peripheral neurons, concerning the sensitivity to PTX of %-adrenoceptor-mediated inhibition. Lai et al. (1983) reported that the inhibitory effect of clonidine on twitch contractions is attenuated after PTX, whereas Docherty (1988) found that PTX pretreatment had no effect on the xylazine inhibition in this tissue and subsequently concluded that %-adrenoceptor-mediated inhibition does not involve G-protein signal transduction. Contradicting results have been published for other peripheral tissues as well: Musgrave et al. (1987) found no effect of PTX on a2-inhibition of NA release in mouse atria, while Weber (1989) reported that N E M attenuates the a2-mediated modulation in the rat tail artery. Nozaki and Sperelakis (1989) described PTX-sensitive (~2-inhibition in the guinea pig mesenteric artery. The primary objective of the present study was to add information touching this controversial issue by applying the protein-alkylating agent, NEM. The action of N E M on the stimulationevoked release of NA has been shown previously to parallel closely that of PTX and the effects of N E M and PTX proved to be non-additive, which can be taken as evidence that both act at the G-protein step (Hertting and Allgaier, 1988). Our results show that 60 /~M N E M partially reverses the inhibition by the az-agonist clonidine and F K 33-824 of the contractile response of the mouse vas to field stimulation (figs. 2A, B and 3A, B). This reversal was only observed when the vasa were stimulated with lower frequencies, i.e. with the left part of the frequency-response curves ( < 20 Hz) and with pulse width-response curves. However, when prejunctional inhibition was partially overridden by stimulation with higher frequencies, N E M incubation further reduced the force of contraction. A reduction was also found when N E M was applied in the absence of prejunctionai agonists with both types of stimulus-response curves. The latter findings can be attributed to postjunctional effects of NEM. N E M 60/zM moderately depressed the contractions due to exogenous noradrenaline and nearly abolished those
158
caused by the muscarinic agonist, bethanechol (fig. 1). NEM also decreased by about 50% the contractions due to 700 /zM ATP (not shown), the purinergic co-transmitter of noradrenaline in the rodent vas deferens. Reduction to 0.2 mM of the calcium concentration of the modified Krebs solution depressed the stimulus-response curves in a way that was very similar to the inhibition by the prejunctional agonists (not shown). However, in contrast to its effect on az-adrenoceptor and /~agonist-induced inhibition, N E M treatment was unable to reverse the depression caused by low extracellular calcium. The alkylating agent even further enhanced the depression. In view of these results it seems unlikely that N E M counteracts by an unspecific action on the smooth muscle the depression by clonidine and F K 33-824 of the contractile response to field stimulation. However, with the methods we now used it cannot be completely ruled out that NEM alkylation transforms receptor sites, thereby impeding agonist binding. But since the vasa in these experiments were exposed to NEM in the presence of the corresponding agonist, the receptors may have been protected from attack by the alkylating agent (Childers, 1984). In contrast to that with clonidine and F K 33824, the depression of the pulse width-response curve by the Pl-agonist, 2-CLA, was apparently not influenced by NEM. This is contrast to the results obtained with 2-CLA in [3H]NA release experiments. On the other hand, this result implies that the postjunctional depression of twitch by NEM (observed when no prejunctional agonist was added; fig. 2D) was compensated. One can speculate that a postjunctional enhancing effect of the purine agonist is possibly superimposed on the presynaptic inhibition (White, 1988), especially if one takes into account the high concentration of 2-CLA (80/~M) which was necessary to achieve a twitch depression comparable to that in the clonidine and F K 33-824 experiments. The results of the experiments with [3H]NA are consistent with our findings for contractile responses to field stimulation: with VP, i.e. low frequency and long pulse width, inhibition of the stimulation-evoked release by the three presynaptic agonists was prevented by exposing the vasa
to 60 /~M N E M (fig. 4). However, the alkylating agent increased the stimulation-evoked release of [3H]NA, both in presence and absence of the agonist, to nearly twice the control value whereas it only moderately increased (20-30%) the spontaneous outflow of tritiated compounds. Comparable results have recently been obtained with hippocampal slices (Allgaier et al., 1987). On the other hand, no release-enhancing effect by N E M was observed with high frequency stimulation (VF). Frequency dependence of the N E M effect has also been reported for hippocampal neurons (Fredholm and Lindgren, 1987). This latter finding is hard to explain if N E M enhanced the stimulation-evoked [3H]NA release by an unspecific mechanism. Furthermore it appears less probable that N E M acted by interfering with NA uptake 1 (Sch6mig et al., 1988) because the release experiments were conducted in presence of cocaine. Idazoxan enhanced [3H]NA release to roughly the same degree as did NEM treatment in the experiments without a2-blocker (fig. 5A, B). This enhancing effect of idazoxan is commonly assigned to interruption of the a2-feedback loop (Starke, 1987). N E M exposure proved to be unable to further enhance the release in presence of idazoxan. This finding can be taken as circumstantial evidence that N E M enhanced the release due to a VP type of stimulation in the absence of exogenous prejunctional agonists by interrupting the az-autoinhibition. An additional corresponding purinergic feedback loop consisting of released ATP or its metabolites acting on prejunctional Pl-receptors seems to be of no relevance (Blakeley et al., 1988). To isolate the inhibition mediated by F K 33-824 and 2-CLA from the allegedly superimposed a zfeedback inhibition, we applied the agonists in presence of idazoxan (fig. 5C, D). Under these conditions N E M again partially reversed the inhibition caused by the agonists. These results suggest that N E M specifically diminishes the/~- and Pl-receptor-mediated inhibition in the mouse vas deferens. Taken together, our findings support the suggestion that prejunctional inhibition by a 2adrenoceptor, /~- and P1-receptor activation in the mouse vas deferens is a NEM-sensitive mechanism
159 and possibly involves signal transduction by a G-protein. The reversal of prejunctional inhibition b y N E M , like o t h e r p r e j u n c t i o n a l effects, is c r u cially dependent on the stimulation parameters.
Acknowledgements The authors thank Mrs. G. v. Braun and Ms. E. Obst for technical assistance and Mrs. M.L. Stolte for the drawings.
References AligNer, C., T.J. Feuerstein and G. Hertting, 1986, N-Ethylmaleimide (NEM) diminishes alpha2-adrenoceptor mediated effects on noradrenaline release, Naunyn-Schmiedeb. Arch. Pharmacol. 333, 104. Allgaier, C., T.J. Feuerstein, R. Jackisch and G. Hertting, 1985, Islet-activating protein (pertussis toxin) diminishes a2adrenoceptor mediated effects on noradrenaline release, Naunyn-Schmiedeb. Arch. Pharmacol. 331, 235. Allgaier, C., G. Hertting and O. Von Kiigelgen, 1987, The adenosine receptor-mediated inhibition of noradrenaline release possibly involves an N-protein and is increased by alpha2-autoreceptor blockade, Br. J. Pharmacol. 90, 403. Blakeley, A.G., P.M. Dunn and S.A. Peterson, 1988, A study of the actions of Pl-purinoceptor agonists and antagonists in the mouse vas deferens in vitro, Br. J. Pharmacol. 94, 37. Childers, S.R., 1984, Interaction of opiate receptor binding sites and guanine nucleotide regulatory sites: selective protection from N-ethylmaleimide, J. Pharmacol. Exp. Ther. 230, 684. Docherty, J.R., 1988, Pertussis toxin and pre-junctional alpha2-adrenoreceptors in rat heart and vas deferens, J. Autonom. Pharmacol. 8, 197. Fredholm, B.B. and E. Lindgren, 1987, Effects of N-ethylmaleimide and forskolin on noradrenaline release from rat hippocampal slices. Evidence that prejunctional adenosine and alpha-receptors are linked to N-proteins but not to adenylate cyclase, Acta Physiol. Scand. 130, 95. Hughes, J., H.W. Kosterlitz and F.M. Leslie, 1975, Effect of morphine on the adrenergic transmission in the mouse vas deferens, Br. J. Pharmacol. 89, 371. Hertting, G. and C. Allgaier, 1988, Participation of protein kinase C and regulatory G proteins in modulation of evoked noradrenaline release in brain, Cell. Mol. Neurobiol. 8, 105. Jakobs, K.H., P. Lasch, M. Minuth, K. Aktories and G. Schulz,
1982, Uncoupling of alpha-adrenoceptor-mediated inhibition of human platelet adenylate cyclase by N-ethylmaleimide, J. Biol. Chem. 257, 2829. Kaschube, M. and G. Zetler, 1989, Noradrenergic, purinergic, and cholinergic transmission in the mouse vas deferens: influence of field-stimulation parameters, reserpinization, 6-hydroxydopamine and 4-aminopyridine, J. Neural Transm. 76, 39. Kirsten, E.B. and A.S. Kuperman, 1970, Effects of sulphydryl inhibitors on frog sartorius: N-ethylmaleimide, Br. J. Pharmacol. 40, 827. Lai, R.T., Y. Watanabe and H. Yoshida, 1983, Effect of islet-activating protein (IAP) on contractile responses of rat vas deferens: evidence for participation of N i (inhibitory GTP binding regulating protein) in the c%-adrenoceptormediated response, European J. Pharmacol. 90, 453. Marshall, 1983, Stimulation-evoked release of [3H]-noradrenaline by 1, 10, or 100 pulses and its modification through presynaptic alpha2-adrenoceptors , Br. J. Pharmacol. 78, 221. Musgrave, I., P. Marley and H. Majewski, 1987, Pertussis toxin does not attenuate alpha2-adrenoceptor mediated inhibition of noradrenaline release in mouse atria, NaunynSchmiedeb. Arch. Pharmacol. 336, 280. Nozaki, M. and N. Sperelakis, 1989, Pertussis toxin effects on transmitter release from perivascular nerve terminals, Am. J. Physiol. 256, H455. Sch/Smig, E., J. Michael-Hepp and H. B~Snisch, 1988, Inhibition of neuronal noradrenaline uptake (uptakel) and desipramine binding by N-ethylmaleimide (NEM), NaunynSchrniedeb. Arch. Pharmacol. 337, 633. Starke, K., 1987, Presynaptic alpha-autoreceptors, Rev. Physiol. Biochem. Pharmacol. 107, 73. Stjarne, L. and P. Astrand, 1985, Relative pre- and postjunctional roles of noradrenaline and adenosine 5'-triphosphate as neurotransmitters of the sympathetic nerves of guinea-pig and mouse vas deferens, Neuroscience 14, 929. Weber, H.D., 1989, Presynaptic a2-adrenoceptors at the terminals of perivascular sympathetic nerves are coupled to an N-ethylmaleimide-sensitive G-protein (abstract), NaunynSchmiedeb. Arch. Pharmacol. 339, R35. White, T.D., 1988, Role of adenine compounds in autonomic neurotransmission, Pharmacol. Ther. 38, 129. Yamamoto, H., T. Kitano, H. Nishino and T. Murano, 1973, Studies on pharmacodynamic action of N-ethylmaleimide (NEM), Jap. J. Pharmacol. 23, 151. Zetler, G. and M. Kaschube, 1985, Importance of frequency and pulse width in field stimulation of the mouse vas deferens: different behaviour of twitch-inhibiting drugs. Naunyn-Schmiedeb. Arch. Pharmacol. 331, 82.
151
Elsevier EJP 51215
Pre- and postjunctional effects of N-ethylmaleimide in the isolated mouse vas deferens Michael Kaschube and Helmut Brasch Department of Pharmacology, Medical University Liibeck, Lftbeck, F.R.G. Received 7 August 1989, revised MS received 23 November 1989, accepted 2 January 1990
The influence of N-ethylmaleimide (NEM) on contractions due to exogenously applied noradrenaline and bethanechol and on the inhibitory effects of clonidine, of the enkephalin derivative, FK 33-824, and 2-chloroadenosine (2-CLA) on field stimulation-response curves and [3H]noradrenaline ([3H]NA) release was studied in the isolated mouse vas deferens. Exposure to NEM (60/~M: 10 min) caused a 30% reduction of the maximal contraction due to NA but nearly abolished the response to bethanechol. NEM partially reversed the depression of the pulse width-response curves by clonidine and FK 33-824 but was without effect with 2-CLA. The contractions evoked by stimulation frequencies above 20 Hz were depressed by NEM both in presence and absence of the agonists. NEM diminished the inhibition of the stimulation-evoked release of [3H]NA by the three agonists. The prejunctional effect of NEM was markedly influenced by the stimulation parameters. These findings support the suggestion that the inhibition mediated by a2-adrenoceptors, /~- and Pl-receptors in the mouse vas deferens is NEM-sensitive and possibly transmitted by a pertussis toxin-sensitive G-protein. Vas deferens; (mouse); N-Ethylmaleimide; Clonidine; 2-Chloroadenosine; Idazoxan; FK 33-824; (Stimulation parameters)
1. Introduction The SH-alkylating agent, N-ethylmaleimide (NEM), has been reported to interrupt the signal transduction of inhibitory prejunctional receptors located on sympathetic neurons in the central nervous system (Allgaier et al., 1986). N E M has been claimed to display some specificity for pertussis toxin-sensitive G-proteins under certain conditions (Jakobs et al., 1982; Allgaier et al., 1987). Though the isolated rodent vas deferens has been used frequently to study drugs interfering with sympathetic transmission, the effect of N E M
Correspondence to: M. Kaschube, Institut fiir Pharmakologie, Medizinische Universit~it zu Ltibeck, Ratzeburger Allee 160, D-2400 Liibeck, F.R.G.
on prejunctional inhibition has not been studied in this tissue. Only Y a m a m o t o et al. (1973) reported on N E M effects on contractile responses to hypogastric nerve stimulation in the isolated vas deferens of the guinea pig. The present study was therefore designed to characterize the influence of N E M incubation on the inhibitory effects of the a2-agonist, clonidine, the tt-agonist, F K 33-824, and the Pl-agonist, 2-chloroadenosine (2-CLA), on the mechanical response and [3H]noradrenaline ([3H]NA) release in response to field stimulation of the mouse vas deferens. We made use of stimulation-response relations to analyse the N E M effect because prejunctional drug effects can be modulated by field stimulation parameters (Zetler and Kaschube, 1985; Kaschube and Zetler, 1989). Since the mechanical response to field stimulation can be affected by both pre- and postjunctional
0014-2999/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
152 actions of a drug, we included experiments on the N E M effect on contractions evoked by exogenously applied agonists acting directly on the smooth muscle.
2. Materials and methods The experimental conditions were essentially the same as described previously (Zetler and Kaschube, 1985; Kaschube and Zetler, 1989). In short, vasa deferentia dissected from freshly killed N M R I mice of 30-40 g body weight were carefully cut free from their mesenteric sheaths. For measuring mechanical responses the vasa were mounted isometrically under a pre-load of 1 m N in a 5-ml organ bath containing a modified Krebs solution (Hughes et al., 1975) gassed with carbogen (95% O 2 and 5% CO2). The bath temperature was kept at 32°C. In some experiments the normal Ca 2+ concentration of 2.54 m M was lowered to 0.2 m M without correction of osmolarity.
approximately 200 mA). Each experiment started with a 1-h equilibration period during which the vas was stimulated with 5 Hz, 0.1 ms (basic stimulation). Three stimulation-response curves (R1, R2 and R3) were obtained for each vas in one experiment by stepwise increases in either the frequency at a constant pulse width of 0.1 ms (3, 5, 10, 20, 50, 80 and 100 Hz) or the pulse width (0.04, 0.06, 0.1, 0.2, 0.3 and 0.5 ms) at a constant frequency of 15 Hz. It should be noted that, with this stimulation protocol with increasing frequency, the number of pulses in the 1-s train also had to be increased. Intervals of 10 min (R1-R2) or 15 rain (R2-R3) were allowed between the three stimulation series. N E M 60 /~M was added to the organ bath between R2 and R3 and was left in contact with the tissue for 10 min. The bath solution was changed by flushing immediately after R1 and R2 as well as after 10 min of N E M exposure. The drugs were added in microliter amounts of aqueous solutions after R1, R2 and N E M incubation. 2.3. [3H]NA release
2.1. Mechanical response to noradrenaline and bethanechol After a 1-h equilibration period two successive non-cumulative concentration-response curves for either agonist (NA: 0.6-240 /~M; bethanechol: 2200/~M) were determined, the first before and the second after exposure of the vas to 60/~M N E M for 10 min and subsequent washout of the alkylating agent. The concentration was increased at 5-min (with noradrenaline) or 10-min (with bethanechol) intervals and the agonist was left in contact with the organ for 0.5 min. The maxima of the resulting phasic contractions were evaluated. The ECs0 values and their 95% confidence limits for N A and bethanechol were calculated from log concentration-effect curves after linearization by logit transformation and estimation of the regression function by the least squares method. 2.2. Mechanical response to field stimulation Twitch responses were induced by field stimulation applied every 30 s as 1-s trains of monophasic square pulses of constant amplitude (28 V,
Both vasa from one mouse were tied in parallel and pre-incubated at 3 2 ° C for 1 h with DL-[73H]noradrenaline ([3H]NA; 16.7 C i / m m o l ) . The resulting activity was 1 # C i / m l . The vasa were then transferred to a 5-ml organ bath and mounted under a pre-load of 5 m N as described above. The modified Krebs solution contained in addition 3 /~M cocaine-HC1 to prevent neuronal re-uptake. The tissues were thoroughly washed six times during 1 h by flushing the organ bath. For the experiment itself, 1-ml samples of the bath fluid were taken every 2.5 min (exception: 10 min with N E M exposure) and added to 9 ml scintillation fluid (Hydroluma, Baker Chemicals B.V., Deventer, Netherlands). The bath fluid was renewed immediately after sample taking. During the 3rd, 7th and l l t h sampling period (S1, $2, $3) 10 twitch contractions were elicited by field stimulation with 1-s trains of pulses at 15-s intervals. The stimulation parameters were either 15 Hz and 0.5 ms pulse duration (VP) or 100 Hz and 0.1 ms pulse duration (VF), i.e. the sets of parameters that produced maximal contractions in the stimulation-response curves (v.s.). N E M 60 ~ M was ad-
153
ded at the start of the 9th collection period (with the exception of NEM-free control experiments) and was washed out after 10-min contact time. The drugs were added before (7th and 8th sampiing period) and after N E M exposure (10th and l l t h sampling period). In some experiments idazoxan was present throughout the experiment, starting from the 2nd collection period. Finally, the vasa were solubilized in 1 ml Soluene 350 (Packard Instruments, Frankfurt, F.R.G.) for determination of the residual tritium content. The outflow of tritiated compounds was calculated as fractional rate of outflow, i.e. the quotient of tritium-labeled compounds released during a collection period and the tissue content at the start of the collection period. The stimulation-evoked rise in outflow was determined as differences between the fractional rates of outflow of the 3rd, 7th and l l t h sampling period ($1, $2, $3) and the basal outflow of the corresponding pre-stimulation period. This rise in outflow over the base level was taken as an approximation of the stimulationevoked release of [3H]NA. It has been shown previously that after incubation of the isolated mouse vas deferens with [3H]NA the stimulationevoked release of tritiated compounds induced by trains of pulses reflects the release of intact [3H]NA (Marshall, 1983; Stj~irne and Astrand, 1985).
3. Results 3.1. Mechanical response to noradrenaline and bethanechol
To evaluate a possible direct influence on the smooth muscle we studied the effect of N E M exposure on the concentration-response relation of two vas-contracting agonists. N E M 60/zM for 10 min depressed the maximal response to noradrenaline by about 30% and caused a moderate rightward shift of the concentration-response curve (fig. 1A). The ECs0 and 95% confidence interval (n = 8) for N A were found to be 7.32 (5.97-8.97) ~tM before and 11.95 (9.68-14.75) /~M after exposure to NEM. A more marked effect of N E M was encountered with bethanechol (fig. 1B). N E M treatment reduced the maximal response by approximately 90%. ECs0 (n = 8) for bethanechol before N E M was 56.4 (50.1-63.5) /~M. The concentration-relS-
Nor ad re n a l i
mN
2.5. Drugs
DL-[7-3H]Noradrenaline • HC1, 16.7 C i / m m o l (Amersham Int., Amersham, U.K.), clonidine (Boehringer Ingelheim, Ingelheim, F.R.G.); 2chloroadenosine, N-ethylmaleimide, bethanechol chloride, cocaine. HC1 (Sigma); noradrenaline 1hydrogentartrate (Fluka, Buchs, Switzerland); [DAla2,MePhen,Met(0)-olS]enkephalin (FK 33-824; gift of Sandoz, Basel, Switzerland), idazoxan (gift of Reckitt & Colman, Hull, U.K.).
/'I' \J. /
÷
2.4. Statistics
The results are presented as arithmetic means + S.E.M. (n = 5-8). Differences between means were tested for statistical significance by means of Student's t-test. A probability level of P < 0.05 was chosen as threshold of significance.
1
ne
o
.
01
.
.
.
~'~/~ "1 1
........
I 10
........
/
I
,
.
100 ,uM
8rnN 6
./
Bethanechol
.L
/+~T
l.-
2-
~10--¢--¢ 0
, o.-7Y,,,, I
.
.
.
.
i
I0
100
.
.
.
.
I
1000 ~ulut
Fig. 1. Concentration-response curves for noradrenaline and bethanechol (filled symbols) as influenced by 60 /~M NEM (open symbols). Ordinate: force of contraction (mN): abscissa: concentration (/~M). The symbols represent the mean ( _+S.E.M.) of eight experiments performed in duplicate.
154
15
Clonidi ne
iL~.l... L
raN"
10-
//.+ '
' 11~"'1
........
I
ms
1
0.1
15 rnN
~/..~...~
FK 3 3 - 8 2 / ,
1SraN:
to-
10,
5,
|
m
,
,
0.1
,m,|
ms
/+--
IS. mN
l
o/
0¸ 0.1
1
rnN
10. .
0.2 mM CaCl~//J'-&-&
'
ms
1
ms
1
no agonist
10.
C
s. '
'
|'~ .... I 0.1
........ ms
I I
0
0.1
Fig. 2. Effect of exposure (10 min) to 60/~M NEM on pulse width-response curves (frequency: 15 Hz) without addition of agonist (E) and under the influence of clonidine (0.38/~M; A), FK 33-824 (1.36/zM; B), 2-CLA (80/~M; C) and lowered extracellular Ca2÷ (0.2 mM; D). The symbols represent the mean (+ S.E.M.) of five experiments in triplicate. Ordinates: force of contraction (mN). (A-C) al, Control; [3, agonist; A, agonist after NEM; (D) II, control; D, 0.2 mM Ca2+; &, 0.2 mM Ca2+ after NEM; (E) aa, control 1; [~, control 2; A, after NEM. sponse curve for bethanechol after application of N E M was too shallow to allow calculation of the ECs0 value with reasonable accuracy. I n control experiments without N E M , repetitive d e t e r m i n a tion of c o n c e n t r a t i o n - r e s p o n s e curves for either agonist i n one vas did n o t change the m a x i m a l response or the ECs0 (not shown).
3.2. Mechanical response to field stimulation The i n h i b i t o r y effects of clonidine (0.38 /~M), F K 33-824 (1.36 /~M), 2 - C L A (80 /~M) a n d a lower b a t h calcium c o n c e n t r a t i o n o n the stimulation-response curves were critically d e p e n d e n t o n the s t i m u l a t i o n p a r a m e t e r s (figs. 2 a n d 3). Raising
155
zo-: mN 15.
C(onidine
./ /Pff
T05' 0
'
~ "r[~'"'l 0.1
........ ms
I I
20-" mN 15.
20:
FK 3 3 - e Z , ~
J. lJ,
~/./~-'tt
10.
raN:
o
ll-,o,-i=r7 ........ i o
0
0.I
ms
j,...~l
° - ~ ° ~ i - - "
'
0.1
' .....
ms
I 1
20-' 2 - CLA
i_i.j,
e,/~/~ ~
15 10-
C /
0
'
I
zo"
5
coc12
..//;/°/'"/"
,o.
.
mN
0.2 mM
.......
mN
no a g o n i s t
i/~"
,/~',~
is,
,/ ~t-~4 lo. ,/~' s. E
s¢~/
i
I
'
' ' ' ('"1
0
,
,
b ,,,,,i
'
' ' '''"1
0.1 ms 1 0.1 ms 1 Fig. 3. Effect of exposure (10 min) to 60/~M NEM on frequency-response curves (pulse width: 0.1 ms) without addition of agonist (E) and under the influence of clonidine (0.38 #M; A), FK 33-824 (1.36 #M; B), 2-CLA (80/~M; C) and lowered extracellular Ca2+ (0.2 raM; D). Ordinates and symbols are the same as in fig. 2.
the frequency of stimulation at constant pulse width antagonized the inhibition of twitch so that, at 100 Hz, the twitch responses were nearly restored to their control values (fig. 3A-D). N o such effect occurred when the pulse width was increased at constant frequency. In this case the stimulation-response curves were fully depressed by the agonists (fig. 2A-C) and by low calcium (fig. 2D).
Exposing the vasa to 6 0 / ~ M N E M for 10 rain partially reversed the inhibitory effect of clonidine and F K 33-824 on the pulse width-response curves (fig. 2A, B) and on the left part of the frequencyresponse curves (i.e. with frequencies below 20 Hz, fig. 3A, B). However, in the pulse width-response curve, N E M was without apparent effect on the 2 - C L A - i n d u c e d inhibition (fig. 2C). The SH-alkylating agent even enhanced the depression of
156
the pulse width-response curve caused by low extracellular Ca 2+, nearly abolishing the twitch responses (fig. 2D). When the stimulation frequency was raised to 20 Hz and above, NEM diminished the force of contraction with all three agonists as well as with low bath Ca 2÷ (fig. 3A-D). When no agonist was added, N E M had a moderate depressive effect on the pulse width- and frequency-response curves (figs. 2E and 3E). Concentrations of NEM higher than those used in the above experiments ( > 100 /tM) and prolonged exposure to the alkylating agent led to permanent rigour of the vas and a gradual decline of the twitch, finally abolishing the contractions due to field stimulation (not shown) as had been found for skeletal muscle (Kirsten and Kuperman, 1970).
3-
2-
15 Hz, 0.5 ms
A
1-
$1
52
$3
FK 33- 824
FR321-
3.3. [3H]NA release When the stimulation protocol was of the VP type (15 Hz, 0.5 ms) clonidine, F K 33-824 and 2-CLA effectively reduced the stimulation-evoked release of tritiated compounds (fig. 4A-C). Low extracellular Ca 2÷ nearly abolished the evoked release of [3H]NA (not shown). Exposing the vasa to NEM (60 /~M) for 10 min between $2 and $3 antagonized this reduction. Surprisingly, the alkylating agent not only restored the stimulationevoked release but raised it to approximately twice the control level (fig. 4A-C). This latter effect was also found when no prejunctional agonist was added (fig. 4D), but not with low extracellular Ca 2÷ (not shown). In contrast to this strong effect on the stimulation-evoked release, N E M caused only a moderate increase (20-30%) of the spontaneous outflow of tritiated compounds in these experiments (not shown). When the VF type of stimulation was applied, neither the agonists alone nor the agonists applied after NEM exposure influenced the stimulationevoked release of [3H]NA to a statistically significant extent (not shown). To test the hypothesis that interruption of a 2autoinhibition is part of the N E M effect on N A release caused by the VP type of field stimulation we included experiments with the a2-adrenoceptor antagonist, idazoxan. The addition of idazoxan
CIonidine
FR-
FR
no ogonist
3-
B
~r
2-.
| $1
'$2
$3
$I
$2
$3
$2
$3
FR-
FR. 2-CLA
3-
3- control
21
$1
$2
$3
S1
Fig. 4. Effect of the 60 ~ M N E M (10 min) on field stimulation-evoked release of [3H]NA without agonists (D) and after addition of clonidine (A), F K 33-824 (B), 2-CLA (C). The concentrations are the same as in fig. 2. E shows a control series without NEM. Ordinates: stimulation-evoked release expressed as percentage of the vas content (FR). Each column represents the mean ( + S.E.M.) of five observations. (A-C) SI: control; $2: agonist; $3, agonist after NEM. (D) SI: control 1; $2: control 2: $3: after N E M . (E) SI: control 1; $2: control 2; $3: no NEM. Statistical significance of the difference between $2 and $3: ~ P < 0.05; o N.S.
roughly doubled the stimulation-evoked release of [3H]NA (fig. 5A, B). N E M 60/~M did not further enhance this augmentation. The release-enhancing effect of idazoxan on $2 and $3 in control experiments without N E M was not significantly different (fig. 5B). The release-inhibiting effect of F K
157
i
FR. 3-
A21-
jj 51
S2
,ill o
FR-
B 21
,0
$3
S1
$2
$3
ldozoxon
Idozoxon
FR,
3-
C
21-
Sl
oi $2
ldazoxon
S3
D
$1
S2
S3
Idozoxon
FK 33-824 2-CLA Fig. 5. Influence of idazoxan (0.83/~M) on [3HINA release (B) as affected by 60 ~M N E M without agonists (A) and after addition of F K 33-824 (1.36 #M; C) and 2-CLA (80/zM; D). Each column represents the mean ( + S.E.M.) of five observations. Ordinates are the same as in fig. 4. '~ P < 0.05; o N.S. (A) SI: control; $2: idazoxan; $3: idazoxan after NEM. (B) SI: control; $2: idazoxan; $3: idazoxan, no NEM. (C, D) $1: idazoxan; $2: idazoxan+ agonist; $3: idazoxan+agonist after NEM.
33-824 and 2-CLA could still be partially reversed by NEM in presence of the c%-adrenoceptor antagonist (fig. 5C, D).
4. Discussion The disruption of the signal transduction pathway of the az-adrenoceptor and adenosine (Aa)receptor by uncoupling of pertussis toxin-sensitive G-proteins from the receptors has been repeatedly reported for central noradrenergic neurons (hippocampus). These studies involved the application of the protein-alkylating agent, N E M (Allgaier et al., 1986; Fredholm and Lindgren, 1987), as well
as the more specific pertussis toxin (PTX, Allgaier et al., 1985) to inactivate the regulatory G-proteins. In contrast, contradictory results have been presented for the sympathetic neurons of the rat vas deferens, i.e. peripheral neurons, concerning the sensitivity to PTX of %-adrenoceptor-mediated inhibition. Lai et al. (1983) reported that the inhibitory effect of clonidine on twitch contractions is attenuated after PTX, whereas Docherty (1988) found that PTX pretreatment had no effect on the xylazine inhibition in this tissue and subsequently concluded that %-adrenoceptor-mediated inhibition does not involve G-protein signal transduction. Contradicting results have been published for other peripheral tissues as well: Musgrave et al. (1987) found no effect of PTX on a2-inhibition of NA release in mouse atria, while Weber (1989) reported that N E M attenuates the a2-mediated modulation in the rat tail artery. Nozaki and Sperelakis (1989) described PTX-sensitive (~2-inhibition in the guinea pig mesenteric artery. The primary objective of the present study was to add information touching this controversial issue by applying the protein-alkylating agent, NEM. The action of N E M on the stimulationevoked release of NA has been shown previously to parallel closely that of PTX and the effects of N E M and PTX proved to be non-additive, which can be taken as evidence that both act at the G-protein step (Hertting and Allgaier, 1988). Our results show that 60 /~M N E M partially reverses the inhibition by the az-agonist clonidine and F K 33-824 of the contractile response of the mouse vas to field stimulation (figs. 2A, B and 3A, B). This reversal was only observed when the vasa were stimulated with lower frequencies, i.e. with the left part of the frequency-response curves ( < 20 Hz) and with pulse width-response curves. However, when prejunctional inhibition was partially overridden by stimulation with higher frequencies, N E M incubation further reduced the force of contraction. A reduction was also found when N E M was applied in the absence of prejunctionai agonists with both types of stimulus-response curves. The latter findings can be attributed to postjunctional effects of NEM. N E M 60/zM moderately depressed the contractions due to exogenous noradrenaline and nearly abolished those
158
caused by the muscarinic agonist, bethanechol (fig. 1). NEM also decreased by about 50% the contractions due to 700 /zM ATP (not shown), the purinergic co-transmitter of noradrenaline in the rodent vas deferens. Reduction to 0.2 mM of the calcium concentration of the modified Krebs solution depressed the stimulus-response curves in a way that was very similar to the inhibition by the prejunctional agonists (not shown). However, in contrast to its effect on az-adrenoceptor and /~agonist-induced inhibition, N E M treatment was unable to reverse the depression caused by low extracellular calcium. The alkylating agent even further enhanced the depression. In view of these results it seems unlikely that N E M counteracts by an unspecific action on the smooth muscle the depression by clonidine and F K 33-824 of the contractile response to field stimulation. However, with the methods we now used it cannot be completely ruled out that NEM alkylation transforms receptor sites, thereby impeding agonist binding. But since the vasa in these experiments were exposed to NEM in the presence of the corresponding agonist, the receptors may have been protected from attack by the alkylating agent (Childers, 1984). In contrast to that with clonidine and F K 33824, the depression of the pulse width-response curve by the Pl-agonist, 2-CLA, was apparently not influenced by NEM. This is contrast to the results obtained with 2-CLA in [3H]NA release experiments. On the other hand, this result implies that the postjunctional depression of twitch by NEM (observed when no prejunctional agonist was added; fig. 2D) was compensated. One can speculate that a postjunctional enhancing effect of the purine agonist is possibly superimposed on the presynaptic inhibition (White, 1988), especially if one takes into account the high concentration of 2-CLA (80/~M) which was necessary to achieve a twitch depression comparable to that in the clonidine and F K 33-824 experiments. The results of the experiments with [3H]NA are consistent with our findings for contractile responses to field stimulation: with VP, i.e. low frequency and long pulse width, inhibition of the stimulation-evoked release by the three presynaptic agonists was prevented by exposing the vasa
to 60 /~M N E M (fig. 4). However, the alkylating agent increased the stimulation-evoked release of [3H]NA, both in presence and absence of the agonist, to nearly twice the control value whereas it only moderately increased (20-30%) the spontaneous outflow of tritiated compounds. Comparable results have recently been obtained with hippocampal slices (Allgaier et al., 1987). On the other hand, no release-enhancing effect by N E M was observed with high frequency stimulation (VF). Frequency dependence of the N E M effect has also been reported for hippocampal neurons (Fredholm and Lindgren, 1987). This latter finding is hard to explain if N E M enhanced the stimulation-evoked [3H]NA release by an unspecific mechanism. Furthermore it appears less probable that N E M acted by interfering with NA uptake 1 (Sch6mig et al., 1988) because the release experiments were conducted in presence of cocaine. Idazoxan enhanced [3H]NA release to roughly the same degree as did NEM treatment in the experiments without a2-blocker (fig. 5A, B). This enhancing effect of idazoxan is commonly assigned to interruption of the a2-feedback loop (Starke, 1987). N E M exposure proved to be unable to further enhance the release in presence of idazoxan. This finding can be taken as circumstantial evidence that N E M enhanced the release due to a VP type of stimulation in the absence of exogenous prejunctional agonists by interrupting the az-autoinhibition. An additional corresponding purinergic feedback loop consisting of released ATP or its metabolites acting on prejunctional Pl-receptors seems to be of no relevance (Blakeley et al., 1988). To isolate the inhibition mediated by F K 33-824 and 2-CLA from the allegedly superimposed a zfeedback inhibition, we applied the agonists in presence of idazoxan (fig. 5C, D). Under these conditions N E M again partially reversed the inhibition caused by the agonists. These results suggest that N E M specifically diminishes the/~- and Pl-receptor-mediated inhibition in the mouse vas deferens. Taken together, our findings support the suggestion that prejunctional inhibition by a 2adrenoceptor, /~- and P1-receptor activation in the mouse vas deferens is a NEM-sensitive mechanism
159 and possibly involves signal transduction by a G-protein. The reversal of prejunctional inhibition b y N E M , like o t h e r p r e j u n c t i o n a l effects, is c r u cially dependent on the stimulation parameters.
Acknowledgements The authors thank Mrs. G. v. Braun and Ms. E. Obst for technical assistance and Mrs. M.L. Stolte for the drawings.
References AligNer, C., T.J. Feuerstein and G. Hertting, 1986, N-Ethylmaleimide (NEM) diminishes alpha2-adrenoceptor mediated effects on noradrenaline release, Naunyn-Schmiedeb. Arch. Pharmacol. 333, 104. Allgaier, C., T.J. Feuerstein, R. Jackisch and G. Hertting, 1985, Islet-activating protein (pertussis toxin) diminishes a2adrenoceptor mediated effects on noradrenaline release, Naunyn-Schmiedeb. Arch. Pharmacol. 331, 235. Allgaier, C., G. Hertting and O. Von Kiigelgen, 1987, The adenosine receptor-mediated inhibition of noradrenaline release possibly involves an N-protein and is increased by alpha2-autoreceptor blockade, Br. J. Pharmacol. 90, 403. Blakeley, A.G., P.M. Dunn and S.A. Peterson, 1988, A study of the actions of Pl-purinoceptor agonists and antagonists in the mouse vas deferens in vitro, Br. J. Pharmacol. 94, 37. Childers, S.R., 1984, Interaction of opiate receptor binding sites and guanine nucleotide regulatory sites: selective protection from N-ethylmaleimide, J. Pharmacol. Exp. Ther. 230, 684. Docherty, J.R., 1988, Pertussis toxin and pre-junctional alpha2-adrenoreceptors in rat heart and vas deferens, J. Autonom. Pharmacol. 8, 197. Fredholm, B.B. and E. Lindgren, 1987, Effects of N-ethylmaleimide and forskolin on noradrenaline release from rat hippocampal slices. Evidence that prejunctional adenosine and alpha-receptors are linked to N-proteins but not to adenylate cyclase, Acta Physiol. Scand. 130, 95. Hughes, J., H.W. Kosterlitz and F.M. Leslie, 1975, Effect of morphine on the adrenergic transmission in the mouse vas deferens, Br. J. Pharmacol. 89, 371. Hertting, G. and C. Allgaier, 1988, Participation of protein kinase C and regulatory G proteins in modulation of evoked noradrenaline release in brain, Cell. Mol. Neurobiol. 8, 105. Jakobs, K.H., P. Lasch, M. Minuth, K. Aktories and G. Schulz,
1982, Uncoupling of alpha-adrenoceptor-mediated inhibition of human platelet adenylate cyclase by N-ethylmaleimide, J. Biol. Chem. 257, 2829. Kaschube, M. and G. Zetler, 1989, Noradrenergic, purinergic, and cholinergic transmission in the mouse vas deferens: influence of field-stimulation parameters, reserpinization, 6-hydroxydopamine and 4-aminopyridine, J. Neural Transm. 76, 39. Kirsten, E.B. and A.S. Kuperman, 1970, Effects of sulphydryl inhibitors on frog sartorius: N-ethylmaleimide, Br. J. Pharmacol. 40, 827. Lai, R.T., Y. Watanabe and H. Yoshida, 1983, Effect of islet-activating protein (IAP) on contractile responses of rat vas deferens: evidence for participation of N i (inhibitory GTP binding regulating protein) in the c%-adrenoceptormediated response, European J. Pharmacol. 90, 453. Marshall, 1983, Stimulation-evoked release of [3H]-noradrenaline by 1, 10, or 100 pulses and its modification through presynaptic alpha2-adrenoceptors , Br. J. Pharmacol. 78, 221. Musgrave, I., P. Marley and H. Majewski, 1987, Pertussis toxin does not attenuate alpha2-adrenoceptor mediated inhibition of noradrenaline release in mouse atria, NaunynSchmiedeb. Arch. Pharmacol. 336, 280. Nozaki, M. and N. Sperelakis, 1989, Pertussis toxin effects on transmitter release from perivascular nerve terminals, Am. J. Physiol. 256, H455. Sch/Smig, E., J. Michael-Hepp and H. B~Snisch, 1988, Inhibition of neuronal noradrenaline uptake (uptakel) and desipramine binding by N-ethylmaleimide (NEM), NaunynSchrniedeb. Arch. Pharmacol. 337, 633. Starke, K., 1987, Presynaptic alpha-autoreceptors, Rev. Physiol. Biochem. Pharmacol. 107, 73. Stjarne, L. and P. Astrand, 1985, Relative pre- and postjunctional roles of noradrenaline and adenosine 5'-triphosphate as neurotransmitters of the sympathetic nerves of guinea-pig and mouse vas deferens, Neuroscience 14, 929. Weber, H.D., 1989, Presynaptic a2-adrenoceptors at the terminals of perivascular sympathetic nerves are coupled to an N-ethylmaleimide-sensitive G-protein (abstract), NaunynSchmiedeb. Arch. Pharmacol. 339, R35. White, T.D., 1988, Role of adenine compounds in autonomic neurotransmission, Pharmacol. Ther. 38, 129. Yamamoto, H., T. Kitano, H. Nishino and T. Murano, 1973, Studies on pharmacodynamic action of N-ethylmaleimide (NEM), Jap. J. Pharmacol. 23, 151. Zetler, G. and M. Kaschube, 1985, Importance of frequency and pulse width in field stimulation of the mouse vas deferens: different behaviour of twitch-inhibiting drugs. Naunyn-Schmiedeb. Arch. Pharmacol. 331, 82.
