- Email: [email protected]
The release of acetylcholine and of catecholamine from the cat's adrenal gland
Neuroscience Vol. 13,No. 3, pp. 957-964,1984 Printed in Great Britain
0306-4522/84 $3.00+ 0.00 PergamonPressLtd 0 1984IBRO
THE RELEASE OF ACETYLCHOLINE AND OF CATECHOLAMINE FROM THE CAT’S ADRENAL GLAND B. COLLIER*,G. JOHNSON*,S. M. KIRPEKAR$+ and J. PRAT$ Department of Pharmacology, *McGill University, 3655 Drummond, Montreal H3G lY6, Canada; SDownstate Medical Center, Brooklyn, NY 11203, U.S.A. Abstract-The cat’s adrenal gland was perfused in situ with Krebs solution containing eserine; the amount of acetylcholine and of catecholamine released was measured. Splanchnic nerve stimulation (5 Hz for 2 min) increased the release of acetylcholine and catecholamine; the molar ratio of evoked release of catecholamine to acetylcholine was 122 f 8. It is suggested that this amplification is achieved because a chromaffin cell granule contains more mediator than does the acetylcholine quantum that releases it. The release per impulse of catecholamine during splanchnic nerve stimulation at 30 Hz was less than that released by stimulation at 1 or 5 Hz. This depression is attributed to a presynaptic failure, because the release of acetylcholine was similarly frequency dependent. The release of catecholamine was linearly related to the release of acetylcholine over the range tested, indicating that the input-output relationship at the splanchnic-adrenal medullary junction is linear. During continuous stimulation of the splanchnic nerve (5 Hz), catecholamine release declined to a level that was 32 + 2% of the initial output. This fatigue is attributed primarily to a postsynaptic depression, because the release of acetylcholine was maintained at 71 k 6% of its initial level. The presence of eserine in the perfusate was necessary for the release of acetylcholine to be detected, but in the presence of eserine catecholamine release was 90 &-10% that in the drug’s absence. It is concluded that released acetylcholine is hydrolysed at some distance from its site of release and action. Glands perfused with raised K+ released acetylcholine and catecholamine. During a 10 min exposure to K+, catecholamine release was less well maintained than was acetylcholine release. It is suggested that calcium channels on adrenal cells desensitize more readily than those on cholinergic nerve terminals. Muscarine or nicotine released catecholamine but not acetylcholine; acetylcholine antagonists depressed evoked catecholamine release but did not alter acetylcholine release. These findings are interpreted to indicate that compounds secreted from the adrenal medulla have no physiologically important feed-back effects on cholinergic nerve terminals.
The physiological secretion of hormones from the adrenal medulla occurs as the consequence of the release of acetylcholine (ACh) from splanchnic nerve
terminals. Following its first clear demonstration by Dreyer,9 many investigators have studied the release of catecholamines during splanchnic stimulation (see review by DouglaQ but there seem to have been no attempts to measure ACh release at the splanchnic-adrenal medullary junction since the early demonstrations by Feldberg and his co-workers that splanchnic stimulation released ACh from the adrenal gland.‘2,‘3 In particular, there has been no study in which both ACh and catecholamine release has been measured. The objective of the present work was to measure the amount of ACh and of catecholamine in the same sample collected from the cat’s adrenal gland during stimulation of its splanchnic nerves in an attempt to provide information about the following points. First, we sought quantitative information about the relationship of first mediator (ACh) and second *To whom correspondence should be addressed. ACh, acetylcholine.
Abbreviation:
957
mediator (catecholamine) release to describe the input-output relationship of this neuroendocrine system; as indicated above, there is not existing information about this. Second, we characterized the release of the two mediators during different rates of stimulation of the splanchnic nerve in order to determine whether the previously reported frequency dependence of catecholamine release’0,25,27,30’5 is the result of presynaptic or postsynaptic changes. Third, we characterized the pattern of release of ACh and catecholamine during prolonged splanchnic stimulation to test directly whether the decline of catecholamine release during such activity’0,“,24.25.30 is due to presynaptic or postsynaptic depression. Fourth, we measured mediator released by K+. The release of catecholamine in response to persistent depolarization is not maintained, mainly because of calcium channel inactivation;2*20 the present experiments test whether splanchnic nerve terminals behave similarly. And finally, we tested whether ACh agonists and antagonists altered presynaptic function in an attempt to assess whether agents secreted from the medullary cells have physiologically relevant feedback effects on the cholinergic nerve terminals. In
958 B. Collier et al.
addition to catecholamine, the adrenal medulla releases ATP6 and enkephalins;34 as reviewed by Starke3’ amines, purines and peptides can all affect ACh release as evidenced by pharmacological experiments, although the physiological consequence is less certain. EXPERIMENTAL
PROCEDURES
Adult cats were anaesthetized with ether or with halothane in N,O-oxygen. The abdomen was opened, the animals was eviscerated, and the left splanchnic nerve was isolated and cut so that it could be stimulated. The left adrenal gland was prepared for perfusion in situ as described by Douglas and Rubin’ and by Dixon et aL5The gland was perfused (0.5 ml/min) in the orthograde direction via a cannula placed in the abdominal aorta and perfusate was collected by a cannule fed retrogradely from the renal vein into the adrenolumbar vein. Anaesthesia was terminated, and the animal killed. The perfusion fluid was Krebs solution of the following composition (mM): NaCl 120, KC1 4.6, CaCI, 2.4, MgSO, 1.2, KH,PO, I .2, NaHCO, 25, glucose IO. ‘IL: medium was equilibrated with 5% CO, in 0, to maintain pIi 7.4, it was supplemented with choline chloride (5 x IO-’ M) to sustain ACh synthesis and, except where indicated, contained eserine sulphate (5 x IO-‘M). In some experiments, the amount of KC1 in the medium was increased to 35 mM with reduced NaCl to maintain the fluid iso-osmolar. All glands were perfused, at rest, for an initial 20 min before collections were made; then perfusate was collected as 2 min samples into ice-cold tubes. In each experiment, an initial sample was collected from the resting gland, following which secretion was evoked by stimulating the splanchnic nerve (IO V, 0.5 ms) or by switching perfusion to high K+ medium. Perfusate collected at rest or during activity was assayed for catecholamine and for ACh, using a separate aliquot of each sample for each assay. We were interested in total catecholamine release and this was measured by the fluorometric method described by Anton and Sayre’ using noradrenaline as standard. Acetylcholine was measured by a modification of the radio-enzymatic method of Goldberg and McCaman”’ as described by Kwok and Collier;” in each experiment representative samples were first subject to hydrolysis by acetylcholinesterase and then assay for ACh to ensure that the ACh assay measured ACh, not choline or some other base.
larly, the release of catecholamine was readily evoked by nerve stimulation, with some spill into the sample collected just after the period of activity. When the splanchnic nerve was stimulated a second time, the co-release of the two mediators was again similar in profile, and reasonably consistent with the amount released during the initial period of stimulation. It is evident from Fig. 1 that the amount of catecholamine measured in the perfusate collected during splanchnic nerve stimulation was considerably more than was the amount of ACh in the same sample. This point was investigated in 15 experiments which measured ACh and catecholamine released before and during splanchnic stimulation at 5 Hz. The results confirmed that an appreciable amplification of transmission is evident during nerve stimulation: the molar ratio of catecholamine to ACh collected varied from 68 to 197. In general, there was a reasonable relationship between the amount of ACh released during stimulation and the amount of catecholamine secreted by that ACh; the experiment with the lowest ACh release (123 pmol) had the lowest catecholamine release (15.6 nmol) and, conversely, the experiment with the most ACh released (457 pmol) had the highest catecholamine secretion (55.1 nmol). There was a less clear relationship between the release of the two mediators from resting glands with relatively high ACh values being associated with rather low catecholamine release and appreciable catecholamine release not necessarily coinciding with the higher ACh release values. From these experiments, the amplification factor for this neuroeffector system during activity was calculated as the molar ratio of catecholamine to ACh released during splanchnic nerve stimulation; for this we subtracted spontaneous mediator release from evoked release: the ratio was 122 f 8. To test whether the relationship between ACh release and catecholamine release was altered when
RESULTS
Release of acetylcholine and of catecholamine during splanchnic nerve stimulation Initial experiments were designed to test whether the co-release of ACh and catecholamine from the adrenal gland could be detected. For this, the adrenal gland was perfused with Krebs medium containing eserine to preserve any released ACh and choline to support ACh synthesis; perfusate was collected during rest and during periods of splanchnic nerve stimulation (5 Hz for 2 min). The result of one of these initial experiments is illustrated by Fig. 1. It is clear that a small amount of ACh could be detected in the perfusate collected during rest, and that this was much increased by nerve stimulation; there was some spill-over of ACh into the collection period that immediately followed the stimulation period. Simi-
Fig. I. The spontaneous and evoked release of ACh and catecholamine from a perfused adrenal gland. During the time indicated by the horizontal bar, the splanchnic nerve was stimulated (5 Hz).
959
ACh and catecholamine release in adrenal gland
the frequency of nerve stimulation was varied, eight experiments tested mediator release evoked by stimulating the splanchnic nerve at 1 Hz (for 2 min), 5 Hz (for 2 min) or 30 Hz (for 20 s). In each experiment, all three frequencies were tested, each test was separated from the next by 14min of rest, and the order of testing the various frequencies of stimulation was varied randomly from one experiment to the next. The amount of mediator released by stimulation was calculated from the value measured during stimulation plus that following stimulation minus twice the value for spontaneous release measured on a sample collected just before stimulation. (This was done in order to include mediator collected as wash-out when stimulation was for the entire 2 min collection period.) The results of these experiments are summarized by Table 1, which shows that the average amount of both ACh and catecholamine released per pulse was similar at 1 or at 5 Hz, but release at 30 Hz was much reduced. The decreased release of ACh at 30 Hz (55 If: 4%) relative to that at 5 Hz was very similar to the equivalent value for catecholamine (49 + 5%). Thus, the amplification of the chemical message at the splan~hnic-a~enal neuroeffector system was rather independent of frequency of stimulation: the ratio of catecholamine to ACh was 105 f 12, 121 + 7 and 128 + 10 at 1, 5 and 30Hz respectively.
Table 1. Evoked release of acetylcholine and catecholamine from adrenal glands during spianchnic nerve stimulation at various frequencies Average release of mediator per pulse ACh Catecholamine (fmol) (pmot)
Frequency (Hz)
1 3;
541 i: 35
56.8 F 5.0
485 224 & + 23 39
31.0 57.6 + 4.6 3.5
Values are mean f SEM of eight experiments.
measured in its usual amount (224 + 27 pmol). In contrast, the presence of eserine had little effect on the amount of catecholamine collected; release in the presence of eserine was 90 rt 10% of that in its absence (27.8 _t 3.4 and 31.8 + 3.0 nmol, respectively). Release of acetylcholine and catecholamine from pot~sium-st~ulated adrenal glands
In six experiments, the adrenal gland was stimulated by switching for 2 min the perfusion medium to one with rasied K+. This released both ACh and catecholamine (see Fig. 3 for result of a typical experiment). The release of catecholamine was much greater than was that of ACh: in 12 tests in the six Release of acetylcholine and catechoiamine during cats, K+-evoked release of ACh was 87 rt 8 pmol and proIonged stimulation of the splanchnic nerve that of catecholamine was 74 f 10nmol. When Six experiments measured the release of mediators glands were exposed to raised K+ for 10min and during nerve stimulation at 5 Hz for 12 min. Effluent mediator release measured in 2 min collections (four was collected as 2min samples, and the amount of experiments), the release of both ACh and catemediator present in the first sample following the cholamine declined with time (Fig. 4); the fade in onset of stimulation was given a value of 100 so that subsequent release could be expressed relative to that value (Fig. 2). During continuous stimulation, the release of catecholamine declined markedly to reach a plateau at which release was about one-third the value of the initial sample. The release of ACh was T clearly better maintained at all time intervals tested; T it fell also to achieve a plateau, but maintained release was at 70% of the initial value, clearly higher than that for catecholamine.
--__*_-
The effect ofeserine on the amount of acetylcholine and ~atecholamine collected during splanchnic nerve stimulation
In eight experiments, the adrenal gland was perfused with Krebs medium without eserine; effluent was collected during rest and during splanchnic nerve stimulation (5 Hz for 2 min). Then, perfusion was switched to medium containing eserine and 10 min later collections similar to those made in the absence of the esterase inhibitor were made. None of the samples collected before exposure of the gland to eserine contained detectable ACh, but following perfusion with the anti-cholinesterase agent spontaneous release was usually detected and evoked release was
Catecholamine
Q, 1
2
3 2 min.
Collection
4
5
6
Periods
Fig. 2. The release of ACh and catecholamine from perfused adrenal glands during continuous stimulation (5 Hz) of the splanchnic nerves. Perfusate was collected as 2 min samples, mediators measured in the first collected sample were given the value 100 and subsequent release was expressed relative to that initial value. Results are presented as mean f SEM six experiments. The initial of values were: ACh = 196 f 18 pmol; catecholamine = 33.0 + 8.3 nmol.
960
I
B. Collier
ef al.
70
-
60
2 E
50
2
40
z
30
f $
20
2
10 0 2
4
20
6
22
24”34
36
40mln
38
Fig. 3. The release of ACh and catecholamine from a perfused adrenal gland stimulated by K+ (applied
during the horizontal bars). catecholamine that of ACh.
secretion
was clearly greater than was
Effects of acetylcholine agonists and antagonists on mediator release
terminals. To test positive feed-back, we tested whether adrenal secretagogues resulted in an increase of the spontaneous ACh release. They did not: nicotine (three experiments) and muscarine (three experi-
The main purpose of these experiments was to test whether substances released from the medullary cells can alter ACh release from the splanchnic nerve
8
10
12
14
16
18m1n
8
10
12
14
16
18min
60
40 \
I
Catecholamine \
\
250
200
\ ‘;; s
20
150
E 9 $
100
2
2
3
2 min.
Collection
4
5
Periods
Fig. 4. The release of ACh and catecholamine from adrenal glands perfused with medium containing 35 mM KCI. Perfusate was collected as 2 min samples, mediators measured in the first such sample were given the value of 100, and subsequent release was expressed relative to that initial value. Results are presented as mean & SEM of five experiments. The initial values were: ACh = 96 k 11 pmol; catecholamine 74 k 14 nmol.
d 4
50
0
E
-
2
4
m SHZ
6
m Nicotine
Fig. 5. The release of ACh and catecholamine from a perfused adrenal gland during splanchnic nerve stimulation (5 Hz) or exposure to muscarine (6 x lo-’ M) in the upper part or exposure to nicotine (6 x lO-5 M) in the lower part.
ACh and catecholamine release in adrenal gland
961
Table 2. Effect of acetylcholine antagonists on the release of acetylcholine and catecholamine from adrenal glands during splanchnic nerve stimulation
Experiment (a) (b) (c) (d) (e)
ments)
stimulated
glands
to secrete
Release in presence: release in absence ACh Catecholamine
Antagonist Tetraethylammonium (6 x IO-‘M) Hexamethonium (1O-s M) Hexamethonium (1O-5 M) Hexamethonium (1O-4 M) Hexamethonium (lO-4 M) + Atropine (1O-6 M)
catecholamine,
but these compounds did not affect the release of ACh (Fig. 5 shows typical experiments). To test negative feed-back, we tested whether blocking secretion from the medullary cells resulted in enhanced ACh release during splanchnic stimulation. Thus, evoked ACh release was measured before and during perfusion with ACh antagonists. The results (Table 2) were negative: ganglionblocking drugs depressed secretion but did not change ACh release and a combination of nicotineand muscarine-blocking drugs almost completely blocked secretion without altering ACh release. DISCUSSION
The main objective of the present experiments was to measure the release of ACh from splanchnic nerve terminals and the catecholamines secreted from the adrenal medullary cells as the result of the action of that ACh. The quantitation of these two mediators clearly showed an appreciable amplification of the chemical signal at this neuroendocrine junction, because the amount of catecholamine released was more than 100 times greater than the amount of ACh released. This amplification was approximately constant over the range of mediator release evoked in the present experiments; in other words the input-output relationship at this neuroeffector junction is linear at least over the ten-fold range of mediator release explored in these experiments. This is illustrated in Fig. 6, in which the relationship between ACh and catecholamine release, in all experiments done in which nerve stimulation was with 600 or fewer impulses, is plotted. The mechanism by which amplification of the chemical message is achieved at the splanchnic-adrenal medullary junction was not explored in the present experiments, but the most likely explanation seems to be that it results from unequal storage capacity of ACh quanta and chromaffin
0.91
0.25
0.96
0.44
0.98
0.47
1.05
0.32
0.97
0.04
granules. If the splanchnic nerve terminals release their transmitter in response to a nerve impulse as quanta, and if such quanta contain as much ACh as estimated at other junctions (see e.g. Macintosh*)), then splanchnic terminals would release multiples of packages containing ACh in the order of lo4 molecules. The release of catecholamines from adrenal medullary cells occurs by exocytosis of chromaffin granules,6s33 each of which is estimated to contain catecholamine in the order of IO6 molecules.‘9.29a.36 Therefore, the measured ratio of catecholamine:ACh in the order of 100 would result if each ACh quantum released one chromaffin granule. Estimates of the quanta1 content The quanta1 content of ACh and of catecholamine released by the average impulse can be calculated if one accepts 3 x 10’ to be the number of chromaffin
. ioos /_I/// ; w600
I
3
/
.
d
$ 2
P s ;
.
.
300
0.
.
200
.
l
.
l
l*
.
l
,”
100
l
ov
10
Evoked
20
30
Catecholamine
40
Release
50
60
(nmoles)
Fig. 6. Relationship of evoked release of ACh and of catecholamine from perfused adrenal glands during stimulation of the splanchnic nerve. This is the collective data from all experiments where 600 impulses or less were delivered.
962
B. Collier et al.
cells in the cat’s adrenal gland.” The average output of catecholamine per pulse (1 Hz stimulation, 120 impulses) was some 57 pmol, so if each cell were activated, the release per impulse per cell was about lo6 molecules, or the content of one granule. Similarly, from the average output of ACh (some 500 fmol per pulse), the release per impulse onto each cell must have been lo4 molecules, or one quantum of ACh. It is entirely probable that not every medullary cell discharges amine in response to each splanchnic impulse, but it is clear that the quanta1 content of evoked ACh release from splanchnic terminals must be rather low. This is compatible with their morphology: splanchnic terminals are < 3 pm in diameter and the area of apparent synaptic contact is smalL4 it is known that the number of quanta released per impulse is related to synaptic area.‘5.” Effects of ,frequency of nerve stimulation
The release per pulse of catecholamine was similar during splanchnic stimulation at 1 or 5 Hz, but was reduced at 30 Hz. Thus, we did not measure any facilitation of amine release with increasing frequency of stimulation at low rates, such as has been shown before ‘8.24,25~35 but we did measure the depression these and other authors’0*27have reported at higher frequency of activation. This pattern of secretion of catecholamine was matched by that of ACh release from the splanchnic nerve terminals, which clearly shows the synaptic depression evident at 30 Hz stimulation is due to a presynaptic failure. The reduced release of ACh at the higher rate of activity was reflected very closely as reduced release of catecholamine, and this indicates that there is a rather low safety factor for transmission at this synapse. The narrow margin of safety for transmission at this junction is not surprising if, as suggested above, the quanta1 content of ACh released per pulse is low. The most likely cause of presynaptic depression would be intermittent failure of some terminals to release transmitter, possibly due to conduction block of impulses in splanchnic terminal branches, which are mostly unmyelinated.4 In contrast to the presynaptic depression measured during high frequency splanchnic stimulation, the depression of synaptic transmission during prolonged low-frequency stimulation clearly had a more pronounced postsynaptic component. Like others,‘“~“~24~25~30 we showed a progressive decline of the output of catecholamine during continuous splanchnic stimulation. This decrease of hormone secretion from adrenal cells was not matched by an equivalent decline of ACh release from the splanchnic terminals. This conclusion that postsynaptic factors are responsible for this phenomenon of fatigue differs from the suggestion previously made on the basis of rather less direct evidence.24,25*30 These authors concluded that the depression is the result of presynaptic failure, because ACh-like agonists could evoke an appreciable discharge of catecholamine at a time
when splanchnic stimulation could not. They recognized that this evidence was less than compelling, for they had no way to distinguish between agonist action on innervated and non-innervated portions of the chromaffin cells. The present evidence is more direct and indicates postsynaptic depression, which could be the result of ACh receptor desensitization or the result of Ca channel inactivation. This lastmentioned mechanism is a prominent feature of fatigue of catecholamine release to secretagogues applied to adrenal cells in vitro (see Knight and Kesteven2’) but its relative importance during splanchnic stimulation in situ remains to be determined. The results of Knight and Kesteven” indicate that the inactivation of Ca channels most likely explains the failure of catecholamine release to be maintained during perfusion of the adrenal gland with raised K+ (present results as well as, e.g. previous ones’.*). The present experiments show that ACh release from splanchnic terminals is better maintained during prolonged exposure to K+, presumably indicating that factors regulating Ca channels on neuronal membranes differ from those operating on non-neuronal membranes. Eserine
Eserine was present in the medium that perfused the adrenal glands in all of the experiments discussed above. It appears not likely that eserine’s presence seriously affects the conclusions reached on the basis of catecholamine secretion, because we measured the same amount of its release in the presence or in the absence of the anti-cholinesterase. This result is similar to that which seems evident from the measures made on glands perfused with Krebs solution by others,‘“,” although these authors did not comment on their results. However, this result showing no effect of eserine on catecholamine release differs considerably from the potentiation of catecholamine release by eserine shown with the blood-perfused gland. ‘3.25The present result suggests that ACh released from splanchnic terminals has its action terminated by diffusion and subsequent hydrolysis, rather than by hydrolysis at its site of action. It is now clear from cytochemistry that, although there is extracellular enzyme, much of the adrenal gland’s acetylcholinesterase is sequestered in cisternae of the endoplasmic reticulum within chromaffin cells,26,3’where it would appear not readily accessible to released ACh. In the blood-perfused organ, ACh hydrolysis may well be by esterases of blood, which would account for the observed potentiation of catecholamine by eserine. In the Krebs-perfused organ, there would be no such hydrolysis and no consequent potentiation by eserine. It is clear from the present experiments that released ACh is destroyed eventually because no ester was measurable in samples collected in the absence of eserine. This hydrolysis could proceed by cholinesterase located some distance from the sites of
ACh and catecholamine release in adrenal gland
ACh release and action, or it could be the consequence of the active secretion of acetylcholinesterase from the stimulated gland.3~28~29 Cholinergic agonists and antagonists
The present results showing no effect of ACh agonists or antagonists on ACh release from splanchnit nerve terminals argue against any prominent feed-back effects on nerve terminals of comoounds released from the medullary cells. The ACh agonists released catecholamine and the antagonists clearly decreased catecholamine secretion; ATP6 and enkephalin34 secretion parallel catecholamine secretion, so it can be presumed that any presynaptic effects of these compounds or their metabolites would have been detected if they were Prominent. negative result does not contradict the possible
This pres-
963
ence of presynaptic receptors capable of modulating ACh release, but it argues strongly against their physiological role in feed-back modulation. It is entirely probable that the chromaffin cell is polarized such that secreted product does not have ready access to the region innervated; thus released compounds would be carried away from the gland without cumulation to pharmacological concentration at the presynaptic nerve terminal. Acknowledgements-This work was supported by grants from the Medical Research Council of Canada (to B.C.) and from the USPHS (to S.M.K.). The experiments were ini-
tiated during the tenure by S.M.K. of a Visiting Scientist Award from the Medical Research Council of Canada. The preparation of this manuscript was helped by an exchange with D. Aunis (Strasbourg), by comments from D. Birks (Montreal) and by L. Wecker (New Orleans) who read the final draft.
REFERENCES 1. Anton A. H. and Sayre D. F. (1962) A study of the factors affecting the aluminum oxide-trihydroxyindole procedure for the analysis of catecholamines. J. Pharmac. exp. Ther. 138, 360-37.5. 2. Baker P. F. and Rink T. J. (1975) Catecholamine release from bovine adrenal medulla in response to maintained depolarization. J. Physiol., Lond. 253, 593-620. 3. Chubb I. W. and Smith A. D. (1975) Release of acetylcholinesterase into the perfusate from the ox adrenal gland. Proc. R. Sot. Ser. B 191, 263-269. 4. Coupland R. E. (1965) The Natural History of the ChromafJin Cell. Longmans, London. 5. Dixon W. R., Garcia A. G. and Kirpekar S. M. (1975) Release of catecholamines and dopamine /?-hydroxylase from the perfused adrenal gland of the cat. J. Physiol., Land. 244, 805-824. 6. Douglas W. W. (1975) Secretomotor control of adrenal medullary secretion: synaptic, membrane and ionic events in stimulus-secretion coupling. In Handbook of Physiology-Endocrinology (ed. Blaschko H., Sayers G. and Smith A. D.). Vol. VI, pp. 367-388. Am. Physiol. Sot., Washington, DC. 7. Douglas W. W. and Rubin R. P. (1961) Role of calcium in the secretory response of the adrenal medulla to acetylcholine. J. Physiol., Lond. 159, 40-57. 8. Douglas W. W. and Rubin R. P. (1963) The mechanism of catecholamine release from the adrenal medulla and the role of calcium in stimulus-secretion coupling. J. Physiol., Lond. 167, 288-310. 9. Dreyer G. P. (1899) On the secretory nerves to the suprarenal capsules. Am. J. Physiol. 2, 203-219. 10. Edwards A. V., Furness P. N. and Helle K. B. (1980) Adrenal medullary responses to stimulation of the splanchnic nerve in the conscious calf. J. Physiol., Lond. 308, 15-27. 11. Elliott T. R. (1912) The control of the suprarenal glands by the splanchnic nerves. J. Physiol., Lond. 44, 374409. 12. Feldberg W. and Minz B. (1933) Das Auftreten eines acetylcholinartigen Stoffes im Nebenierenvenenblut bei Reizung der Nervi splanchnici. Pjiigers Arch. ges. Physiol. 233, 657-682. 13. Feldberg W., Minz B. and Tsudzimura H. (1934) The mechanism of the nervous discharge of adrenaline. J. Physiol., Lond. 81, 286304. 14. Goldberg A. M. and McCaman R. E. (1973) The determination of picomole amounts of acetylcholine in mammalian brain. J. Neurochem. 20, 1-8. 15. Harris J. B. and Ribchester R. R. (1979) The relationship between end-plate size and transmitter release in normal and dystrophic muscles of the mouse. J. Physiol., Land. 296, 245-265. 16. Ito S., Nakazato Y. and Ohga A. (1980) Exocytotic release of catecholamines from perfused adrenal gland of guinea-pig induced by veratridine. Br. J. Pharmac. 70, 527-535.
17. Khan M. T. and Furchgott R. F. (1982) Interactions of Dhenoxvbenzamine and nicotinic agonists at the nicotinic receptor in the cat adrenal medulla. J. Pharmac. exp. The;. 221, -117-122. 18. Kiroekar S. M. and Cervoni P. (1963) Effect of cocaine. ohenoxvbenzamine and ohentolamine on the catecholamine ’ output from spleen and adrenal‘med;lla. J. Pharmac. bx>. Ther: 142, 59-70. 19. Kirshner N. and Viveros 0. H. (1970) Quanta1 aspects of the secretion of catecholamines and dopamine-/?-hydroxylase from the adrenal medulla. In New Aspects of Storage and Release Mechanisms of Catecholamines (ed. Schiimann H. J. and Kroneberg G.), pp. 78-88. Springer, Berlin. 20. Knight D. E. and Kesteven N. T. (1983) Evoked transient intracellular free Ca2+ changes and secretion in isolated bovine adrenal medullary cells. Proc. R. Sot. Ser. B. 218, 177-199. 21. Kuno M., Turkanis S. A. and Weakly J. N. (1971) Correlation between nerve terminal size and transmitter release at the neuromuscular junction of the frog. J. Physiol., Lond. 213, 545-556. 22. Kwok Y. N. and Collier B. (1982) Synthesis of acetylcholine from acetate in a sympathetic ganglion. J. Neurochem. 39, 1626. 23. Macintosh F. C. (1978) The present status of the vesicle hypothesis. pharmacology (ed. Jendon D. J.), pp. 297-322. Plenum, New York.
In Cholinergic Mechanisms and Psycho-
24. Marley E. and Paton W. D. M. (1961) The output of sympathetic amines from the cat’s adrenal gland in response to splanchnic nerve activity. J. Physiol., Lond. 155, l-27.
964
B. Collier et al.
25. Marley E. and Prom G. I. (1965) Physiology and pharmacology of the splanchnic-adrenal medullary junction. J. Physiol., Land. 180, 483-513. 26. Millar T. J. and Unsicker K. (1981) Catecholamine-storing cells in the adrenal medulla of the pre- and postnatal rat: acetylcholinesterase as a means for early discrimination of cell types. Cell Tissue Res. 217, 155-170. 27. Mirkin B. L. (1961) Factors influencing the selective secretion of adrenal medullary hormones. J. Pharmac. exp. Ther. 132, 218-225. 28. Mizobe F. and Livett B. G. (1980) Production and release of acetylcholinesterase by a primary cell culture of bovine adrenal medullary chromaffin cells. J. Neurochem. 35, 1469-1472. 29. Mizobe F. and Livett B. G. (1983) Nicotine stimulates secretion of both catecholamines and acetylcholinesterase from cultured adrenal chromatlin cells. J. Neurosci. 3, 871-876.
29a. Nordmann J. J. (1984) Combined stereological and biochemical analvsis of storage and release of catecholamines in the adrenal medulla of the rat. J. Neurochem. 42, 434437. 30. Silver M. (1960) The output of adrenaline and noradrenaline from the adrenal medulla of the calf. J. Physiol., Lond. 152, 14-29. 31. Somogyi P., Chubb I. W. and Smith A. D. (1975) A possible structural basis for the extracellular release of acetylcholinesterase. Proc. R. Sot. Ser. B 191, 271-283. 32. Starke K. (1981) Presynaptic receptors. A. Reu. Pharmac. Toxicol. 21, 7-30. 33. Viveros 0. H. (1975) Mechanism of secretion of catecholamines from adrenal medulla. In Handbook oj Physiology-Endocrinology (ed. Blaschko H., Sayers G. and Smith A. D.), Vol. VI, pp. 389426. Am. Physiol. Sot., Washington, DC. 34. Viveros 0. H. and Wilson S. P. (1983) The adrenal chromaffin cell as a model to study to co-secretion of enkephalins and catecholamines. J. auton. Net-c. Sysf. 7, 41-58. 35. Wakade A. R. (1981) Studies on secretion of catecholamines evoked by acetylcholine or transmural stimulation of the rat adrenal gland. J. Physiol., Lond. 313, 463480. 36. Winkler H. and Westhead E. (1980) The molecular organization of adrenal chromaffin granules. Neuroscience 5, 1803-1823. (Accepted 2 May 1984)
0306-4522/84 $3.00+ 0.00 PergamonPressLtd 0 1984IBRO
THE RELEASE OF ACETYLCHOLINE AND OF CATECHOLAMINE FROM THE CAT’S ADRENAL GLAND B. COLLIER*,G. JOHNSON*,S. M. KIRPEKAR$+ and J. PRAT$ Department of Pharmacology, *McGill University, 3655 Drummond, Montreal H3G lY6, Canada; SDownstate Medical Center, Brooklyn, NY 11203, U.S.A. Abstract-The cat’s adrenal gland was perfused in situ with Krebs solution containing eserine; the amount of acetylcholine and of catecholamine released was measured. Splanchnic nerve stimulation (5 Hz for 2 min) increased the release of acetylcholine and catecholamine; the molar ratio of evoked release of catecholamine to acetylcholine was 122 f 8. It is suggested that this amplification is achieved because a chromaffin cell granule contains more mediator than does the acetylcholine quantum that releases it. The release per impulse of catecholamine during splanchnic nerve stimulation at 30 Hz was less than that released by stimulation at 1 or 5 Hz. This depression is attributed to a presynaptic failure, because the release of acetylcholine was similarly frequency dependent. The release of catecholamine was linearly related to the release of acetylcholine over the range tested, indicating that the input-output relationship at the splanchnic-adrenal medullary junction is linear. During continuous stimulation of the splanchnic nerve (5 Hz), catecholamine release declined to a level that was 32 + 2% of the initial output. This fatigue is attributed primarily to a postsynaptic depression, because the release of acetylcholine was maintained at 71 k 6% of its initial level. The presence of eserine in the perfusate was necessary for the release of acetylcholine to be detected, but in the presence of eserine catecholamine release was 90 &-10% that in the drug’s absence. It is concluded that released acetylcholine is hydrolysed at some distance from its site of release and action. Glands perfused with raised K+ released acetylcholine and catecholamine. During a 10 min exposure to K+, catecholamine release was less well maintained than was acetylcholine release. It is suggested that calcium channels on adrenal cells desensitize more readily than those on cholinergic nerve terminals. Muscarine or nicotine released catecholamine but not acetylcholine; acetylcholine antagonists depressed evoked catecholamine release but did not alter acetylcholine release. These findings are interpreted to indicate that compounds secreted from the adrenal medulla have no physiologically important feed-back effects on cholinergic nerve terminals.
The physiological secretion of hormones from the adrenal medulla occurs as the consequence of the release of acetylcholine (ACh) from splanchnic nerve
terminals. Following its first clear demonstration by Dreyer,9 many investigators have studied the release of catecholamines during splanchnic stimulation (see review by DouglaQ but there seem to have been no attempts to measure ACh release at the splanchnic-adrenal medullary junction since the early demonstrations by Feldberg and his co-workers that splanchnic stimulation released ACh from the adrenal gland.‘2,‘3 In particular, there has been no study in which both ACh and catecholamine release has been measured. The objective of the present work was to measure the amount of ACh and of catecholamine in the same sample collected from the cat’s adrenal gland during stimulation of its splanchnic nerves in an attempt to provide information about the following points. First, we sought quantitative information about the relationship of first mediator (ACh) and second *To whom correspondence should be addressed. ACh, acetylcholine.
Abbreviation:
957
mediator (catecholamine) release to describe the input-output relationship of this neuroendocrine system; as indicated above, there is not existing information about this. Second, we characterized the release of the two mediators during different rates of stimulation of the splanchnic nerve in order to determine whether the previously reported frequency dependence of catecholamine release’0,25,27,30’5 is the result of presynaptic or postsynaptic changes. Third, we characterized the pattern of release of ACh and catecholamine during prolonged splanchnic stimulation to test directly whether the decline of catecholamine release during such activity’0,“,24.25.30 is due to presynaptic or postsynaptic depression. Fourth, we measured mediator released by K+. The release of catecholamine in response to persistent depolarization is not maintained, mainly because of calcium channel inactivation;2*20 the present experiments test whether splanchnic nerve terminals behave similarly. And finally, we tested whether ACh agonists and antagonists altered presynaptic function in an attempt to assess whether agents secreted from the medullary cells have physiologically relevant feedback effects on the cholinergic nerve terminals. In
958 B. Collier et al.
addition to catecholamine, the adrenal medulla releases ATP6 and enkephalins;34 as reviewed by Starke3’ amines, purines and peptides can all affect ACh release as evidenced by pharmacological experiments, although the physiological consequence is less certain. EXPERIMENTAL
PROCEDURES
Adult cats were anaesthetized with ether or with halothane in N,O-oxygen. The abdomen was opened, the animals was eviscerated, and the left splanchnic nerve was isolated and cut so that it could be stimulated. The left adrenal gland was prepared for perfusion in situ as described by Douglas and Rubin’ and by Dixon et aL5The gland was perfused (0.5 ml/min) in the orthograde direction via a cannula placed in the abdominal aorta and perfusate was collected by a cannule fed retrogradely from the renal vein into the adrenolumbar vein. Anaesthesia was terminated, and the animal killed. The perfusion fluid was Krebs solution of the following composition (mM): NaCl 120, KC1 4.6, CaCI, 2.4, MgSO, 1.2, KH,PO, I .2, NaHCO, 25, glucose IO. ‘IL: medium was equilibrated with 5% CO, in 0, to maintain pIi 7.4, it was supplemented with choline chloride (5 x IO-’ M) to sustain ACh synthesis and, except where indicated, contained eserine sulphate (5 x IO-‘M). In some experiments, the amount of KC1 in the medium was increased to 35 mM with reduced NaCl to maintain the fluid iso-osmolar. All glands were perfused, at rest, for an initial 20 min before collections were made; then perfusate was collected as 2 min samples into ice-cold tubes. In each experiment, an initial sample was collected from the resting gland, following which secretion was evoked by stimulating the splanchnic nerve (IO V, 0.5 ms) or by switching perfusion to high K+ medium. Perfusate collected at rest or during activity was assayed for catecholamine and for ACh, using a separate aliquot of each sample for each assay. We were interested in total catecholamine release and this was measured by the fluorometric method described by Anton and Sayre’ using noradrenaline as standard. Acetylcholine was measured by a modification of the radio-enzymatic method of Goldberg and McCaman”’ as described by Kwok and Collier;” in each experiment representative samples were first subject to hydrolysis by acetylcholinesterase and then assay for ACh to ensure that the ACh assay measured ACh, not choline or some other base.
larly, the release of catecholamine was readily evoked by nerve stimulation, with some spill into the sample collected just after the period of activity. When the splanchnic nerve was stimulated a second time, the co-release of the two mediators was again similar in profile, and reasonably consistent with the amount released during the initial period of stimulation. It is evident from Fig. 1 that the amount of catecholamine measured in the perfusate collected during splanchnic nerve stimulation was considerably more than was the amount of ACh in the same sample. This point was investigated in 15 experiments which measured ACh and catecholamine released before and during splanchnic stimulation at 5 Hz. The results confirmed that an appreciable amplification of transmission is evident during nerve stimulation: the molar ratio of catecholamine to ACh collected varied from 68 to 197. In general, there was a reasonable relationship between the amount of ACh released during stimulation and the amount of catecholamine secreted by that ACh; the experiment with the lowest ACh release (123 pmol) had the lowest catecholamine release (15.6 nmol) and, conversely, the experiment with the most ACh released (457 pmol) had the highest catecholamine secretion (55.1 nmol). There was a less clear relationship between the release of the two mediators from resting glands with relatively high ACh values being associated with rather low catecholamine release and appreciable catecholamine release not necessarily coinciding with the higher ACh release values. From these experiments, the amplification factor for this neuroeffector system during activity was calculated as the molar ratio of catecholamine to ACh released during splanchnic nerve stimulation; for this we subtracted spontaneous mediator release from evoked release: the ratio was 122 f 8. To test whether the relationship between ACh release and catecholamine release was altered when
RESULTS
Release of acetylcholine and of catecholamine during splanchnic nerve stimulation Initial experiments were designed to test whether the co-release of ACh and catecholamine from the adrenal gland could be detected. For this, the adrenal gland was perfused with Krebs medium containing eserine to preserve any released ACh and choline to support ACh synthesis; perfusate was collected during rest and during periods of splanchnic nerve stimulation (5 Hz for 2 min). The result of one of these initial experiments is illustrated by Fig. 1. It is clear that a small amount of ACh could be detected in the perfusate collected during rest, and that this was much increased by nerve stimulation; there was some spill-over of ACh into the collection period that immediately followed the stimulation period. Simi-
Fig. I. The spontaneous and evoked release of ACh and catecholamine from a perfused adrenal gland. During the time indicated by the horizontal bar, the splanchnic nerve was stimulated (5 Hz).
959
ACh and catecholamine release in adrenal gland
the frequency of nerve stimulation was varied, eight experiments tested mediator release evoked by stimulating the splanchnic nerve at 1 Hz (for 2 min), 5 Hz (for 2 min) or 30 Hz (for 20 s). In each experiment, all three frequencies were tested, each test was separated from the next by 14min of rest, and the order of testing the various frequencies of stimulation was varied randomly from one experiment to the next. The amount of mediator released by stimulation was calculated from the value measured during stimulation plus that following stimulation minus twice the value for spontaneous release measured on a sample collected just before stimulation. (This was done in order to include mediator collected as wash-out when stimulation was for the entire 2 min collection period.) The results of these experiments are summarized by Table 1, which shows that the average amount of both ACh and catecholamine released per pulse was similar at 1 or at 5 Hz, but release at 30 Hz was much reduced. The decreased release of ACh at 30 Hz (55 If: 4%) relative to that at 5 Hz was very similar to the equivalent value for catecholamine (49 + 5%). Thus, the amplification of the chemical message at the splan~hnic-a~enal neuroeffector system was rather independent of frequency of stimulation: the ratio of catecholamine to ACh was 105 f 12, 121 + 7 and 128 + 10 at 1, 5 and 30Hz respectively.
Table 1. Evoked release of acetylcholine and catecholamine from adrenal glands during spianchnic nerve stimulation at various frequencies Average release of mediator per pulse ACh Catecholamine (fmol) (pmot)
Frequency (Hz)
1 3;
541 i: 35
56.8 F 5.0
485 224 & + 23 39
31.0 57.6 + 4.6 3.5
Values are mean f SEM of eight experiments.
measured in its usual amount (224 + 27 pmol). In contrast, the presence of eserine had little effect on the amount of catecholamine collected; release in the presence of eserine was 90 rt 10% of that in its absence (27.8 _t 3.4 and 31.8 + 3.0 nmol, respectively). Release of acetylcholine and catecholamine from pot~sium-st~ulated adrenal glands
In six experiments, the adrenal gland was stimulated by switching for 2 min the perfusion medium to one with rasied K+. This released both ACh and catecholamine (see Fig. 3 for result of a typical experiment). The release of catecholamine was much greater than was that of ACh: in 12 tests in the six Release of acetylcholine and catechoiamine during cats, K+-evoked release of ACh was 87 rt 8 pmol and proIonged stimulation of the splanchnic nerve that of catecholamine was 74 f 10nmol. When Six experiments measured the release of mediators glands were exposed to raised K+ for 10min and during nerve stimulation at 5 Hz for 12 min. Effluent mediator release measured in 2 min collections (four was collected as 2min samples, and the amount of experiments), the release of both ACh and catemediator present in the first sample following the cholamine declined with time (Fig. 4); the fade in onset of stimulation was given a value of 100 so that subsequent release could be expressed relative to that value (Fig. 2). During continuous stimulation, the release of catecholamine declined markedly to reach a plateau at which release was about one-third the value of the initial sample. The release of ACh was T clearly better maintained at all time intervals tested; T it fell also to achieve a plateau, but maintained release was at 70% of the initial value, clearly higher than that for catecholamine.
--__*_-
The effect ofeserine on the amount of acetylcholine and ~atecholamine collected during splanchnic nerve stimulation
In eight experiments, the adrenal gland was perfused with Krebs medium without eserine; effluent was collected during rest and during splanchnic nerve stimulation (5 Hz for 2 min). Then, perfusion was switched to medium containing eserine and 10 min later collections similar to those made in the absence of the esterase inhibitor were made. None of the samples collected before exposure of the gland to eserine contained detectable ACh, but following perfusion with the anti-cholinesterase agent spontaneous release was usually detected and evoked release was
Catecholamine
Q, 1
2
3 2 min.
Collection
4
5
6
Periods
Fig. 2. The release of ACh and catecholamine from perfused adrenal glands during continuous stimulation (5 Hz) of the splanchnic nerves. Perfusate was collected as 2 min samples, mediators measured in the first collected sample were given the value 100 and subsequent release was expressed relative to that initial value. Results are presented as mean f SEM six experiments. The initial of values were: ACh = 196 f 18 pmol; catecholamine = 33.0 + 8.3 nmol.
960
I
B. Collier
ef al.
70
-
60
2 E
50
2
40
z
30
f $
20
2
10 0 2
4
20
6
22
24”34
36
40mln
38
Fig. 3. The release of ACh and catecholamine from a perfused adrenal gland stimulated by K+ (applied
during the horizontal bars). catecholamine that of ACh.
secretion
was clearly greater than was
Effects of acetylcholine agonists and antagonists on mediator release
terminals. To test positive feed-back, we tested whether adrenal secretagogues resulted in an increase of the spontaneous ACh release. They did not: nicotine (three experiments) and muscarine (three experi-
The main purpose of these experiments was to test whether substances released from the medullary cells can alter ACh release from the splanchnic nerve
8
10
12
14
16
18m1n
8
10
12
14
16
18min
60
40 \
I
Catecholamine \
\
250
200
\ ‘;; s
20
150
E 9 $
100
2
2
3
2 min.
Collection
4
5
Periods
Fig. 4. The release of ACh and catecholamine from adrenal glands perfused with medium containing 35 mM KCI. Perfusate was collected as 2 min samples, mediators measured in the first such sample were given the value of 100, and subsequent release was expressed relative to that initial value. Results are presented as mean & SEM of five experiments. The initial values were: ACh = 96 k 11 pmol; catecholamine 74 k 14 nmol.
d 4
50
0
E
-
2
4
m SHZ
6
m Nicotine
Fig. 5. The release of ACh and catecholamine from a perfused adrenal gland during splanchnic nerve stimulation (5 Hz) or exposure to muscarine (6 x lo-’ M) in the upper part or exposure to nicotine (6 x lO-5 M) in the lower part.
ACh and catecholamine release in adrenal gland
961
Table 2. Effect of acetylcholine antagonists on the release of acetylcholine and catecholamine from adrenal glands during splanchnic nerve stimulation
Experiment (a) (b) (c) (d) (e)
ments)
stimulated
glands
to secrete
Release in presence: release in absence ACh Catecholamine
Antagonist Tetraethylammonium (6 x IO-‘M) Hexamethonium (1O-s M) Hexamethonium (1O-5 M) Hexamethonium (1O-4 M) Hexamethonium (lO-4 M) + Atropine (1O-6 M)
catecholamine,
but these compounds did not affect the release of ACh (Fig. 5 shows typical experiments). To test negative feed-back, we tested whether blocking secretion from the medullary cells resulted in enhanced ACh release during splanchnic stimulation. Thus, evoked ACh release was measured before and during perfusion with ACh antagonists. The results (Table 2) were negative: ganglionblocking drugs depressed secretion but did not change ACh release and a combination of nicotineand muscarine-blocking drugs almost completely blocked secretion without altering ACh release. DISCUSSION
The main objective of the present experiments was to measure the release of ACh from splanchnic nerve terminals and the catecholamines secreted from the adrenal medullary cells as the result of the action of that ACh. The quantitation of these two mediators clearly showed an appreciable amplification of the chemical signal at this neuroendocrine junction, because the amount of catecholamine released was more than 100 times greater than the amount of ACh released. This amplification was approximately constant over the range of mediator release evoked in the present experiments; in other words the input-output relationship at this neuroeffector junction is linear at least over the ten-fold range of mediator release explored in these experiments. This is illustrated in Fig. 6, in which the relationship between ACh and catecholamine release, in all experiments done in which nerve stimulation was with 600 or fewer impulses, is plotted. The mechanism by which amplification of the chemical message is achieved at the splanchnic-adrenal medullary junction was not explored in the present experiments, but the most likely explanation seems to be that it results from unequal storage capacity of ACh quanta and chromaffin
0.91
0.25
0.96
0.44
0.98
0.47
1.05
0.32
0.97
0.04
granules. If the splanchnic nerve terminals release their transmitter in response to a nerve impulse as quanta, and if such quanta contain as much ACh as estimated at other junctions (see e.g. Macintosh*)), then splanchnic terminals would release multiples of packages containing ACh in the order of lo4 molecules. The release of catecholamines from adrenal medullary cells occurs by exocytosis of chromaffin granules,6s33 each of which is estimated to contain catecholamine in the order of IO6 molecules.‘9.29a.36 Therefore, the measured ratio of catecholamine:ACh in the order of 100 would result if each ACh quantum released one chromaffin granule. Estimates of the quanta1 content The quanta1 content of ACh and of catecholamine released by the average impulse can be calculated if one accepts 3 x 10’ to be the number of chromaffin
. ioos /_I/// ; w600
I
3
/
.
d
$ 2
P s ;
.
.
300
0.
.
200
.
l
.
l
l*
.
l
,”
100
l
ov
10
Evoked
20
30
Catecholamine
40
Release
50
60
(nmoles)
Fig. 6. Relationship of evoked release of ACh and of catecholamine from perfused adrenal glands during stimulation of the splanchnic nerve. This is the collective data from all experiments where 600 impulses or less were delivered.
962
B. Collier et al.
cells in the cat’s adrenal gland.” The average output of catecholamine per pulse (1 Hz stimulation, 120 impulses) was some 57 pmol, so if each cell were activated, the release per impulse per cell was about lo6 molecules, or the content of one granule. Similarly, from the average output of ACh (some 500 fmol per pulse), the release per impulse onto each cell must have been lo4 molecules, or one quantum of ACh. It is entirely probable that not every medullary cell discharges amine in response to each splanchnic impulse, but it is clear that the quanta1 content of evoked ACh release from splanchnic terminals must be rather low. This is compatible with their morphology: splanchnic terminals are < 3 pm in diameter and the area of apparent synaptic contact is smalL4 it is known that the number of quanta released per impulse is related to synaptic area.‘5.” Effects of ,frequency of nerve stimulation
The release per pulse of catecholamine was similar during splanchnic stimulation at 1 or 5 Hz, but was reduced at 30 Hz. Thus, we did not measure any facilitation of amine release with increasing frequency of stimulation at low rates, such as has been shown before ‘8.24,25~35 but we did measure the depression these and other authors’0*27have reported at higher frequency of activation. This pattern of secretion of catecholamine was matched by that of ACh release from the splanchnic nerve terminals, which clearly shows the synaptic depression evident at 30 Hz stimulation is due to a presynaptic failure. The reduced release of ACh at the higher rate of activity was reflected very closely as reduced release of catecholamine, and this indicates that there is a rather low safety factor for transmission at this synapse. The narrow margin of safety for transmission at this junction is not surprising if, as suggested above, the quanta1 content of ACh released per pulse is low. The most likely cause of presynaptic depression would be intermittent failure of some terminals to release transmitter, possibly due to conduction block of impulses in splanchnic terminal branches, which are mostly unmyelinated.4 In contrast to the presynaptic depression measured during high frequency splanchnic stimulation, the depression of synaptic transmission during prolonged low-frequency stimulation clearly had a more pronounced postsynaptic component. Like others,‘“~“~24~25~30 we showed a progressive decline of the output of catecholamine during continuous splanchnic stimulation. This decrease of hormone secretion from adrenal cells was not matched by an equivalent decline of ACh release from the splanchnic terminals. This conclusion that postsynaptic factors are responsible for this phenomenon of fatigue differs from the suggestion previously made on the basis of rather less direct evidence.24,25*30 These authors concluded that the depression is the result of presynaptic failure, because ACh-like agonists could evoke an appreciable discharge of catecholamine at a time
when splanchnic stimulation could not. They recognized that this evidence was less than compelling, for they had no way to distinguish between agonist action on innervated and non-innervated portions of the chromaffin cells. The present evidence is more direct and indicates postsynaptic depression, which could be the result of ACh receptor desensitization or the result of Ca channel inactivation. This lastmentioned mechanism is a prominent feature of fatigue of catecholamine release to secretagogues applied to adrenal cells in vitro (see Knight and Kesteven2’) but its relative importance during splanchnic stimulation in situ remains to be determined. The results of Knight and Kesteven” indicate that the inactivation of Ca channels most likely explains the failure of catecholamine release to be maintained during perfusion of the adrenal gland with raised K+ (present results as well as, e.g. previous ones’.*). The present experiments show that ACh release from splanchnic terminals is better maintained during prolonged exposure to K+, presumably indicating that factors regulating Ca channels on neuronal membranes differ from those operating on non-neuronal membranes. Eserine
Eserine was present in the medium that perfused the adrenal glands in all of the experiments discussed above. It appears not likely that eserine’s presence seriously affects the conclusions reached on the basis of catecholamine secretion, because we measured the same amount of its release in the presence or in the absence of the anti-cholinesterase. This result is similar to that which seems evident from the measures made on glands perfused with Krebs solution by others,‘“,” although these authors did not comment on their results. However, this result showing no effect of eserine on catecholamine release differs considerably from the potentiation of catecholamine release by eserine shown with the blood-perfused gland. ‘3.25The present result suggests that ACh released from splanchnic terminals has its action terminated by diffusion and subsequent hydrolysis, rather than by hydrolysis at its site of action. It is now clear from cytochemistry that, although there is extracellular enzyme, much of the adrenal gland’s acetylcholinesterase is sequestered in cisternae of the endoplasmic reticulum within chromaffin cells,26,3’where it would appear not readily accessible to released ACh. In the blood-perfused organ, ACh hydrolysis may well be by esterases of blood, which would account for the observed potentiation of catecholamine by eserine. In the Krebs-perfused organ, there would be no such hydrolysis and no consequent potentiation by eserine. It is clear from the present experiments that released ACh is destroyed eventually because no ester was measurable in samples collected in the absence of eserine. This hydrolysis could proceed by cholinesterase located some distance from the sites of
ACh and catecholamine release in adrenal gland
ACh release and action, or it could be the consequence of the active secretion of acetylcholinesterase from the stimulated gland.3~28~29 Cholinergic agonists and antagonists
The present results showing no effect of ACh agonists or antagonists on ACh release from splanchnit nerve terminals argue against any prominent feed-back effects on nerve terminals of comoounds released from the medullary cells. The ACh agonists released catecholamine and the antagonists clearly decreased catecholamine secretion; ATP6 and enkephalin34 secretion parallel catecholamine secretion, so it can be presumed that any presynaptic effects of these compounds or their metabolites would have been detected if they were Prominent. negative result does not contradict the possible
This pres-
963
ence of presynaptic receptors capable of modulating ACh release, but it argues strongly against their physiological role in feed-back modulation. It is entirely probable that the chromaffin cell is polarized such that secreted product does not have ready access to the region innervated; thus released compounds would be carried away from the gland without cumulation to pharmacological concentration at the presynaptic nerve terminal. Acknowledgements-This work was supported by grants from the Medical Research Council of Canada (to B.C.) and from the USPHS (to S.M.K.). The experiments were ini-
tiated during the tenure by S.M.K. of a Visiting Scientist Award from the Medical Research Council of Canada. The preparation of this manuscript was helped by an exchange with D. Aunis (Strasbourg), by comments from D. Birks (Montreal) and by L. Wecker (New Orleans) who read the final draft.
REFERENCES 1. Anton A. H. and Sayre D. F. (1962) A study of the factors affecting the aluminum oxide-trihydroxyindole procedure for the analysis of catecholamines. J. Pharmac. exp. Ther. 138, 360-37.5. 2. Baker P. F. and Rink T. J. (1975) Catecholamine release from bovine adrenal medulla in response to maintained depolarization. J. Physiol., Lond. 253, 593-620. 3. Chubb I. W. and Smith A. D. (1975) Release of acetylcholinesterase into the perfusate from the ox adrenal gland. Proc. R. Sot. Ser. B 191, 263-269. 4. Coupland R. E. (1965) The Natural History of the ChromafJin Cell. Longmans, London. 5. Dixon W. R., Garcia A. G. and Kirpekar S. M. (1975) Release of catecholamines and dopamine /?-hydroxylase from the perfused adrenal gland of the cat. J. Physiol., Land. 244, 805-824. 6. Douglas W. W. (1975) Secretomotor control of adrenal medullary secretion: synaptic, membrane and ionic events in stimulus-secretion coupling. In Handbook of Physiology-Endocrinology (ed. Blaschko H., Sayers G. and Smith A. D.). Vol. VI, pp. 367-388. Am. Physiol. Sot., Washington, DC. 7. Douglas W. W. and Rubin R. P. (1961) Role of calcium in the secretory response of the adrenal medulla to acetylcholine. J. Physiol., Lond. 159, 40-57. 8. Douglas W. W. and Rubin R. P. (1963) The mechanism of catecholamine release from the adrenal medulla and the role of calcium in stimulus-secretion coupling. J. Physiol., Lond. 167, 288-310. 9. Dreyer G. P. (1899) On the secretory nerves to the suprarenal capsules. Am. J. Physiol. 2, 203-219. 10. Edwards A. V., Furness P. N. and Helle K. B. (1980) Adrenal medullary responses to stimulation of the splanchnic nerve in the conscious calf. J. Physiol., Lond. 308, 15-27. 11. Elliott T. R. (1912) The control of the suprarenal glands by the splanchnic nerves. J. Physiol., Lond. 44, 374409. 12. Feldberg W. and Minz B. (1933) Das Auftreten eines acetylcholinartigen Stoffes im Nebenierenvenenblut bei Reizung der Nervi splanchnici. Pjiigers Arch. ges. Physiol. 233, 657-682. 13. Feldberg W., Minz B. and Tsudzimura H. (1934) The mechanism of the nervous discharge of adrenaline. J. Physiol., Lond. 81, 286304. 14. Goldberg A. M. and McCaman R. E. (1973) The determination of picomole amounts of acetylcholine in mammalian brain. J. Neurochem. 20, 1-8. 15. Harris J. B. and Ribchester R. R. (1979) The relationship between end-plate size and transmitter release in normal and dystrophic muscles of the mouse. J. Physiol., Land. 296, 245-265. 16. Ito S., Nakazato Y. and Ohga A. (1980) Exocytotic release of catecholamines from perfused adrenal gland of guinea-pig induced by veratridine. Br. J. Pharmac. 70, 527-535.
17. Khan M. T. and Furchgott R. F. (1982) Interactions of Dhenoxvbenzamine and nicotinic agonists at the nicotinic receptor in the cat adrenal medulla. J. Pharmac. exp. The;. 221, -117-122. 18. Kiroekar S. M. and Cervoni P. (1963) Effect of cocaine. ohenoxvbenzamine and ohentolamine on the catecholamine ’ output from spleen and adrenal‘med;lla. J. Pharmac. bx>. Ther: 142, 59-70. 19. Kirshner N. and Viveros 0. H. (1970) Quanta1 aspects of the secretion of catecholamines and dopamine-/?-hydroxylase from the adrenal medulla. In New Aspects of Storage and Release Mechanisms of Catecholamines (ed. Schiimann H. J. and Kroneberg G.), pp. 78-88. Springer, Berlin. 20. Knight D. E. and Kesteven N. T. (1983) Evoked transient intracellular free Ca2+ changes and secretion in isolated bovine adrenal medullary cells. Proc. R. Sot. Ser. B. 218, 177-199. 21. Kuno M., Turkanis S. A. and Weakly J. N. (1971) Correlation between nerve terminal size and transmitter release at the neuromuscular junction of the frog. J. Physiol., Lond. 213, 545-556. 22. Kwok Y. N. and Collier B. (1982) Synthesis of acetylcholine from acetate in a sympathetic ganglion. J. Neurochem. 39, 1626. 23. Macintosh F. C. (1978) The present status of the vesicle hypothesis. pharmacology (ed. Jendon D. J.), pp. 297-322. Plenum, New York.
In Cholinergic Mechanisms and Psycho-
24. Marley E. and Paton W. D. M. (1961) The output of sympathetic amines from the cat’s adrenal gland in response to splanchnic nerve activity. J. Physiol., Lond. 155, l-27.
964
B. Collier et al.
25. Marley E. and Prom G. I. (1965) Physiology and pharmacology of the splanchnic-adrenal medullary junction. J. Physiol., Land. 180, 483-513. 26. Millar T. J. and Unsicker K. (1981) Catecholamine-storing cells in the adrenal medulla of the pre- and postnatal rat: acetylcholinesterase as a means for early discrimination of cell types. Cell Tissue Res. 217, 155-170. 27. Mirkin B. L. (1961) Factors influencing the selective secretion of adrenal medullary hormones. J. Pharmac. exp. Ther. 132, 218-225. 28. Mizobe F. and Livett B. G. (1980) Production and release of acetylcholinesterase by a primary cell culture of bovine adrenal medullary chromaffin cells. J. Neurochem. 35, 1469-1472. 29. Mizobe F. and Livett B. G. (1983) Nicotine stimulates secretion of both catecholamines and acetylcholinesterase from cultured adrenal chromatlin cells. J. Neurosci. 3, 871-876.
29a. Nordmann J. J. (1984) Combined stereological and biochemical analvsis of storage and release of catecholamines in the adrenal medulla of the rat. J. Neurochem. 42, 434437. 30. Silver M. (1960) The output of adrenaline and noradrenaline from the adrenal medulla of the calf. J. Physiol., Lond. 152, 14-29. 31. Somogyi P., Chubb I. W. and Smith A. D. (1975) A possible structural basis for the extracellular release of acetylcholinesterase. Proc. R. Sot. Ser. B 191, 271-283. 32. Starke K. (1981) Presynaptic receptors. A. Reu. Pharmac. Toxicol. 21, 7-30. 33. Viveros 0. H. (1975) Mechanism of secretion of catecholamines from adrenal medulla. In Handbook oj Physiology-Endocrinology (ed. Blaschko H., Sayers G. and Smith A. D.), Vol. VI, pp. 389426. Am. Physiol. Sot., Washington, DC. 34. Viveros 0. H. and Wilson S. P. (1983) The adrenal chromaffin cell as a model to study to co-secretion of enkephalins and catecholamines. J. auton. Net-c. Sysf. 7, 41-58. 35. Wakade A. R. (1981) Studies on secretion of catecholamines evoked by acetylcholine or transmural stimulation of the rat adrenal gland. J. Physiol., Lond. 313, 463480. 36. Winkler H. and Westhead E. (1980) The molecular organization of adrenal chromaffin granules. Neuroscience 5, 1803-1823. (Accepted 2 May 1984)