- Email: [email protected]
Distinct regulations by calcium of cyclic GMP levels and catecholamine secretion in isolated bovine adrenal chromaffin cells
Distinct Catecholamine
Regulations Secretion
Simon Lemaire,
by Calcium of Cyclic GMP Levels and in Isolated Bovine Adrenal Chromaffin Cells
Gilles Derome,
Richard Tseng,
Paulette
Mercier,
and Irma Lemaire
The effects of various calcium-dependent secretagogues on cyclic GMP levels and catecholamine (CA) secretion were measured in a preparation of bovine adrenal chromaffin cells. The secretory effect of acetylcholine (ACh; 8-10 fold stimulation) was mimicked by nicotine but not muscarine. Three-five fold stimulations of cyclic GMP levels were also obtained with ACh and muscarine but not nicotine. High concentration of K+, and the ionophore A23187, also elevated cyclic GMP levels. However, secretion produced by veratridine, ouabain, and the ionophore X537A was not accompanied by any rise in cyclic GMP levels. Removal of extracellular calcium significantly decreased both basal levels of CA secretion and of cyclic GMP and completely abolished their stimulation by ACh. The half-maximal effects of calcium on the cholinergic stimulations of cyclic GMP levels and of CA secretion were observed at 0.2 and 2.5 mM, respectively. Substitution of Ca** by S? was more effective in maintaining the cyclic GMP response than the secretory response. The calcium channel blockers Co’+, Mg” and Ni2+ .Inhibited the cholinergic stimulation of cyclic GMP more than that of CA release. On the other hand, the organic calcium channel blockers. verapamil and methoxyverapamil (O-800) were more effective antagonists of the secretory response. These data indicate that the cholinergic stimulations of CA secretion and of cyclic GMP levels in bovine adrenal chromaffin cells are regulated by calcium via two distinct mechanisms.
T
HE ROLE OF CYCLIC AMP as an effector of both CA synthesis’** and CA secretion3*4 in the adrenal medulla, has been extensively examined, but only limited data are available concerning the role of cyclic GMP.’ Studies in other tissues have indicated that cyclic GMP regulates numerous biological functions’ and it has been suggested that it may have a neuronal function and mediate postsynaptic transmission.6*7 Recently, using a preparation of bovine adrenal chromaffin cells, we have observed that the cholinergic stimulation of CA secretion by ACh is accompanied by an important increase in cyclic GMP levels8 and a muscarinic receptor was involved in this effect.’ Since calcium plays a key function in stimulus-secretion coupling in the adrenal medulla,‘O~” it was important to study its role in the cholinergic stimulation of cyclic GMP in our cell preparation and to establish the relations between the effects of calcium on cyclic GMP levels and on CA secretion. The present data indicate that calcium is required both for stimulation of CA secretion and for the regulation of cyclic GMP levels in bovine adrenal chromaffin cells. Furthermore, the use of specific calcium-dependent secretagogues, calcium substitutes and calcium-channel blockers has distinguished the effects of calcium in both phenomena.
From the Department of Pharmacology, Centre Hospitalier Universitaire. Sherbrooke. Quebec, Canada JIH 5N4. Received for publication June 20, I980. Address reprint requests to Simon Lemaire, Department of Pharmacology, Centre Hospitalier Universitaire. Sherbrooke, Quebec, Canada JIH 5N4. (c I981 by Grune & Stratton. Inc. 002&0495/81/3005-0009~01.00/0
462
MATERIALS
AND
METHODS
Materials Radioimmune assay kits for the measurement of cyclic nucleotides were purchased from Schwarz-Mann (Boston, Mass.). Cyclic GMP and cyclic AMP were obtained from Boehringer Mannheim (W. Germany). ACh, nicotine, muscarine, veratridine and ouabain were purchased from Sigma Chemical Co. (St. Louis. MO.). Verapamil and methoxyverapamil (D-600) were generously supplied by A.G. Knoll Ltd (Chemische Fabriske) Ludwigshafen am Rhein. A23 I87 and X537A were kindly provided by Eli Lilly & Co. and Hoffmann LaRoche Ltd, respectively. Fresh bovine adrenal glands were obtained in a local slaughter house.
Cell Preparation Bovine adrenal glands were rapidly removed from the animal after exsanguination and transported to the laboratory on ice. Chromaffin cells were isolated with a few modifications of the procedure of Hochman and Perlman.‘* Briefly, the medullae of 6-8 glands were dissected free of cortical tissue and placed in ice-cold calcium-free Krebs-Ringer buffer, pH 7.4 (buffer I: NaCI, 1I8 mM; KH,PO,, 1.2mM; MgSO,, I .2 mM; NaHCO,; 25 mM; glucose, IO mM: bovine serum albumin, 0.5%). The medullary tissue was cut into small pieces (-2 mm), washed three times with 30 ml of buffer I. and then submitted to five successive 30 min digestions with collagenase (1.5 mg/ml of collagenase CLS Type I from Worthington) in the same buffer at 37°C. The dispersed cells were removed after each digestion and filtered through four layers of cheese cloth. The first harvest, containing mostly red blood cells and debris, was discarded and the four others were washed three times with 30 ml of buffer I supplemented with 2.2 mM CaC12 (Buffer II) and collected by centrifugation at I IO x g for 10 min at room temperature. The isolated cells were pooled and added to 40 ml of buffer 11. Chromaffin cells were purified by differential sedimentation at room temperature and unit gravity for 2 hr. During this process, the chromaffin cells sedimented as a white pellet leaving most debris and red blood cells in the supernatant. The supernatant was then discarded and the chromaffin cells (pellet) were resuspended in 6 ml of buffer II containing 100 units of penicillin and 100 pg of streptomycin. The cells were gassed with a mixture of O,/CO, (95%/5%) and stored overnight in a capped plastic tube at room temperature. The next day, the cells were washed with 30 ml of buffer II and counted in a
Metabolism, Vol. 30, No. 5 (May). 1981
ROLE OF CALCIUM IN ADRENOMEDULLAAY
hemocytometer.
Viability
(97%) was determined
trypan blue. This procedure chromaffin ceils.
Stimulation
usually
of Catecholamine
yielded
463
CELLS
by the
million
Secretion
CA secretion was assayed as described.* Briefly, the secretion was started by the addition of 2.5 x lo5 chromafhn cells (0.1 ml in buffer II) to a prewarmed (37OC) solution (0.9 ml) of buffer II containing the various drugs to be tested. Secretion was stopped after 5 min by transferring the tubes to an ice-water bath. After cooling the samples for 3 to 5 min. the cells were centrifuged at I10 x g for 10 min and 0.7 ml aliquots of the supernatants were collected. Proteins were precipitated by the addition of 70 ~1 of cold 50% trichloroacetic acid (tinal concentration: 5%) and removed by centrifugation at 900 x g for 30 min. The clear supernatants were collected and stored at -20°C for subsequent CA determination. Secretory experiments were performed in triplicate and controls were obtained by the incubation of the cells in the absence of drugs at 0°C (control to be subtracted) and 37°C (basal level of CA secretion).
Measurement
of Catecholamines
For total CA determination
(epinephrine plus norepinephrine), aliquots (200 ~1) of the above samples were added to 800 ~1 of I M sodium borate buffer, pH 8.0 (final pH: 7.0) and oxidized by the procedure of Miura et al.” Fluorescence was monitored on a spectrofluorometer (Carl Zeiss. model ZFM 4) at excitation/ emission wavelengths of 4 I O/5 I5 nm. respectively. At these wavelengths, the emission intensities of epinephrine and norepinephrine were equal. Each experiment was repeated three times in triplicate and values are expressed as nmole of CA secreted per IO6 cells per 5 min of incubation.
Cyclic Nucleotide
Effect of Various Drugs on Catecholamine and Cyclic Gh4P Levels
Secretion
Various drugs can induce the release of CA from perfused adrenal glands in the presence of calcium. Among these drugs are cholinergic agonists, calcium ionophores, and depolarizing agents. Table 1 shows the effects of some of these drugs on CA secretion and cyclic GMP levels in bovine adrenal chromaffin cells. ACh (SO PM) induced an 8.1-fold stimulation of CA release and a 3.1-fold increase in cyclic GMP levels. The secretory effect of ACh was mimicked by nicotine (7.2-fold stimulation), but not muscarine. Conversely, cyclic GMP levels were elevated by muscarine (3-fold stimulation), but not by nicotine. The secretory effect of increased K’ (56 mM; 5.1-fold stimulation) was accompanied by a 2.1-fold increase in cyclic GMP levels. On the other hand, veratridine and ouabain, also known to depolarize nerve cells,‘5,‘h stimulated CA secretion (6.18 and 2.52-fold stimulation, respectively) without affecting cyclic GMP levels. The involvement of calcium ions in both responses was further supported by the use of the calcium ionophores, X53lA and A23187.“,” X537A was a more potent secretagogue than A23 187 (5.09- and 2.13-fold stimulation of CA release, respectively). Only A23187, however, elevated cyclic GMP levels (2.87fold stimulation).
Assuy
The cyclic nucleotide assay was started by the addition of 5 x IO5 chromaffin cells (0.1 ml in buffer II) to a pre-warmed (37OC) solution of bulfer II (0.9 ml) containing the various drugs to be tested. The reaction was stopped after 3 min by the addition of 1 ml of cold 20X trichloroacetic acid and freezing and thawing three times respectively in liquid nitrogen and a water bath at 37°C. The protein precipitate was then removed by centrifugation at 900 x g for 30 min at W”C. The supernatants were collected and transfered into a boiling-water bath for 3 min. The samples were extracted five times with 4 ml portions of water-saturated ether, evaporated under vacua and dissolved with 0.3 ml of water. They were then stored at ~~2Q°C for subsequent cyclic nucleotide determination. Assays were performed in triplicate and controls were obtained by the incubation of the cells in the absence of drugs. Maximal stimulation of cyclic G MP levels was obtained at approximately 30 set of incubation with ACh and was maintained for the first 10 min.’ The 3 min incubation time was choosen mainly because it represents a maximal time for stimulation. Furthermore. measurements at 3 min were more accurate then at 30 sec.
Determination
RESULTS
exclusion of
fifty to sixty
of Cyclic GMP
Cyclic GMP was measured from 30 /rl aliquots following the radioimmune techniques of Frandsen and Krishna.14 Recovery (over 95%) was calculated by the addition of [‘HI-cyclic GMP (2000 CPM) to the cyclic nucleotide assay as described previously. Nonspecific binding (determined after digestion wrth phosphodiesterase, Sigma) was estimated to be smaller than 5%. Values are expressed as picomoles of cyclic GMP per IO6 cells.
Effect of Calcium
and Sodium
Removal
To verify the importance of extracellular calcium and sodium ions on the stimulatory effects of some of these drugs, cells were incubated in calcium or sodium-free buffers (Table 1). Removal of extracellular calcium significantly decreased both basal levels of cyclic G MP (from 0.4 to 0.3 pmoles/ 1O6cells) and CA secretion (from 0.65 to 0. I6 nmole/ IO’ cells). Furthermore, the absence of calcium completely abolished the stimulations of cyclic GMP levels and of CA secretion by ACh, high K ’ and A23 187. On the other hand, removal of sodium ions from the incubation medium did not modify the basal levels of cyclic GMP and of CA release but significantly depressed their stimulation by ACh (Table 2). Sodium removal also decreased the secretory response to high K +; however the cyclic GMP response to high K + was much less affected by sodium deprivation. Effect of Calcium
Concentration
To verify the sensitivity of both cyclic GMP and secretory responses to extracellular calcium, cells were incubated in the presence of increasing concentrations of calcium with or without ACh (50 PM; Fig 1).
464
LEMAIRE
ET AL.
Table 1. Effect of Various Secretagogues on CA Secretion and Cyclic GMP Levels in Isolated Bovine Adrenal Chromaffin Cells Catecholamine
Secretion
Cyclic GMP PsrC@nt
Secretagogue Normal
nmoles/ 1Oh Cells
P.WCl?fX
Cnntr0l
pmoles/ 106 Cells
Control
Buffer
Control
0.65 k 0.12
100
0.40
ACh (50 pM)
5.27
+ 0.32
810
1.25 t 0.10
Nicotine (50 /.LM)
4.71
+ 0.22
724
0.38
Muscarine (50 fiLMI
0.71
+ 0.18’
109
1.21 + 0.07
302
KCI (56 mM)
4.34
+ 0.43
667
0.82
-t 0.08
205
3.31
* 0.19
509
0.43
* 0.09*
1.39 +- 0.23
213
1.15 * 0.12
lonophore X537A
I10 FM)
lonophore A23187
(10 PM)
+ 0.06
100 312
k 0.05*
95
107 287
Veratridine I10 PM)
4.02
? 0.31
618
0.42
+ 0.05’
105
Ouabain (10 PM)
1.64 i 0.16
252
0.41
* 0.04*
102
Calcium-free Buffer Control
0.16
k 0.05
-
0.30
+ 0.05
100
ACh (50 PM)
0.19
+ 0.10’
-
0.32
+ 0.07,
106
KCI (56 mM)
0.13
+ 0.08*
-
0.30
* 0.04,
100
0.08
t 0.04*
-
0.33
k 0.09+
110
lonophore A23187
(10 PM)
Sodium-free Buffer Control ACh
(50
KCI (56
(56
mM)
0.65
k 0.10
100
0.42
k 0.02
100
1.06
5 0.13
160
0.96
k 0.05
228
1.88
* 0.16
290
0.78
t
185
0.12
experiments were repeated three times in triplicate and values are mean t SD (statistically different of control: P -c .05 Student’s
The when
/.rM)
indicated*). mM),
tris (143
Catecholamine
an equivalent mM,
pH 7.2,
amount HCI)
secretion of NaCl
to replace
and cyclic (56
mM)
the sodium
GMP
was
levels
removed
ion. Total
were from
measured the
CA content:
as described
incubation 34.9
Calcium induced a dose-dependent increase in basal levels of cyclic GMP (from 0.3 to 0.4 pmoles between 0 and 5 mM CaCl,) and of CA secretion (from 0 to 1 nmole of CA). The stimulatory effects of ACh were also calcium-dependent. The half-maximal effects of calcium on ACh-stimulated cyclic GMP levels and CA secretion were observed at 0.2 and 2.5 mM CaCl,, respectively. Maximal stimulation of cyclic GMP levels was also obtained at a lower concentration of calcium (-5 mM) than the maximal secretory response (a 10 mM).
under
medium
k 3.4
“Materials
to maintain
nmoles/106
Replacement
and Methods.”
the osmolality.
t test,
For stimulation
Sodium-free
buffer
except with
K+
contained
cells.
of Ca’+ by Ba” and Sr2+
In perfused adrenal glands, both Ba2+ and Sr” are good substitutes for Ca2+ for the stimulation of CA release.‘9,20 It was of interest to verify if such replacements could also maintain cyclic GMP and secretory responses in our preparation of bovine adrenal chromaffin cells. Table 2 shows that either Ba2’ or Sr2+ can replace Ca2+ for stimulations of cyclic GMP levels and CA secretion by ACh. However, Ba*+ itself induced important increases in basal levels of CA
Table 2. Effect of the Replacement of Calcium by Barium or Strontium Ions on the ACh-induced Stimulations of CA Secretion and Cyclic GMP Levels in Isolated Bovine Adrenal Chromaffin Cells Catecholamine
Secretion
Cyck
GMP Percent
Percent Addition Normal
(50
Calcium-free
PM) buffer
+
B&I,
(2.2
(50
Calcium-free
fiMI buffer
+
SrCI,
(2.2
The assays
(50 were
to the incubations. test).
Control
PM) performed
as described
The experiments
were
0.54
* 0.12
5.57
* 0.43
100 1,031
0.36
k 0.03
100
1.24
+ 0.07
344
6.36
_c 0.38
100
0.54
+ 0.04
100
9.56
+ 0.44
150
0.98
t
181
0.06
mM)
Control ACh
1O6 Cells
mM)
Control ACh
pm&s/
Buffer
Control ACh
Control
nmoles/ 1O6 Cells
0.76
k 0.16
100
0.34
+ 0.03
100
2.81
+ 0.19
370
0.85
+ 0.05
250
under repeated
“Materials 3 times
and Methods, in triplicate
” except
and values
that
the cells were
are mean
washed
k SD (statistically
with different
15 ml of the indicated of control:
buffer
prior
P rc .05 Student’s
t
ROLE OF CALCIUM
IN ADRENOMEDULLARY
465
CELLS
secretory response respectively).
(from
10.3- to 1.5- and
3.7-fold
EJect of Calcium Channel Blockers Many compounds such as the divalent cations Co”, Ni” and Mg” or the organic drugs, verapamil and methoxyverapamil (D-600) block the entry of calcium in nerve cells through voltage-sensitive calcium channels.*’ 24 Figure 2 and Table 3 illustrate the effect of these compounds on the cholinergic stimulations of cyclic GMP levels and of CA secretion in our cell preparation. Cobalt (Fig. 2) inhibited both responses, but the cyclic GMP response was six times more
1
w I
0
I
I
2 4 6 CALCIUM
I
I
8 IO (mM1
0
0.2 0.4 0.6 0.8 1.0
Fig. 1. Effect of increasing concentrations of calcium on basal 1-1 and ACh-stimulated G---O1 cyclic GMP levels IA) and CA secretion fB). The experiments were performed as described under “Materials and Methods” and values are mean t SD fstatisticslly different of the zero Cai ’ concentration control: P < .05. Student’s t test).
secretion (from 0.54 to 6.36 nm&s/ ItI6 cells) and of cyclic GMP (from 0.36 to 0.54 pmole/106 ceils). In the presence of Ba”, the stimulatory effects of ACh were much depressed (from 10.3- to 1.5-fold stimulation of CA release and from 3.44- to I .8 I -fold stimulation of cyclic GMP levels). However, this large decrease in the efficiency of ACh to stimulate CA secretion or cyclic GMP levels may depend on the stimulatory effects of Ba” itself. Sr2+ did not affect basal values and was a better substitute for maintaining both responses to ACh. Interestingly, the cyclic GMP response to ACh was less affected by the replacement of Ca” with Ba2+ or Sr’+ (from 3.44- to 1.81- and 2.5-fold stimulation, respectively) than the
o-
0
0.2 0.4 0.6 0.8 1.0 COBALTcmM)
Fig. 2. Effect of increasing concentrations of cobalt on basal (U) and ACh-stimulated (0-Q cyclic GMP levels (Al and CA secretion (5). The experiments were performed as described under “Materials and Methods” and values are mean 2 SD (variations in the presence of ACh were statistically different of the zero Co’ ’ concentration: P c .05, Student’s t test whereas those observed in its absence were nonsignificant).
466
LEMAIRE ET AL.
Table 3.
Effect of MgZf. Ni’+, Verapamil
and D-600
on the Cholinergic
and Cyclic GMP Levels in Isolated Bovine Adrenal Catecholamme
AddWon
Secretion
Inmoles/1O6cells)
InhibitIon
(%)
Stimulations
Chromaffin
of CA Secretion
Cells Cyclic GMP (pm&s/ 1O6cellsl
Inhibition 1%)
Ach (50 /JM)
4.70
+ 0.35
0
0.79
* 0.04
0
Ach + MgCI, (10 mM)
2.91
? 0.23
38
0.04
* 0.02
95
Ach + MgCI, (20 mM)
0.73
-t 0.15
84
0 + 0.01
100
Ach + NiCI, (1 mM)
2.67
% 0.19
43
0.03
i 0.01
Ach + verapamil (10 PM)
1.66 + 0.20
65
0.62
+ 0.02
21
Ach + D-600
1.55 + 0.14
66
0.60
+ 0.01
24
(10 PM)
96
The experiments were repeated 3 times in triplicate and values are mean + SD (statistically different of the assay with ACh alone: P < .05 Student’s t test). “Catecholamine secretion” represents the amount of CA released by ACh after subtraction of the basal release of 0.5
+ 0.1 nmole/lOB cells.
“Cyclic GMP” represents the increase in cyclic GMP levels evoked by ACh after the subtraction of the basal level of 0.4 * 0.08 pmoles/108 cells. The experiments were performed as described under “Materials and Methods.”
sensitive to Co’+ (ED,,: 0.05 mM) than the secretory response (EDSo: 0.33 mM). Increasing concentrations of Co*+ (up to 1 mM) had however no effects on the basal levels of cyclic GMP and of CA secretion. Stimulation of cyclic GMP levels by ACh was also more sensitive to the presence of Mg2+ (10 and 20 mM) and Ni’+ (1 mM) than the secretory response (Table 3). Interestingly, the organic calcium channel blockers verapamii and D-600 (at 10 FM) inhibited the secretory response more completely (65%-66% inhibition) than the stimulation of cyclic GMP levels (2 I %-24% inhibition). None of these calcium channel blockers affected the basal levels of CA secretion and of cyclic GMP (legend of Table 3). It then appears that each response can be selectively blocked by one or the other type of calcium channel blockers. DISCUSSION
The present studies indicate that calcium is required for the regulation of CA secretion and of cyclic GMP levels in adrenal chromafhn cells. This concept was first supported by the stimulation of CA release with the calcium ionophores X537A and A23 187 and of cyclic GMP levels with A23 187 (Table 1). Furthermore, removal of calcium completely abolished both responses to ACh, high K+ and A23187 (Table 1). However, many distinctions were observed between the regulation of both responses by calcium. First, the half-maximal effect of calcium on the cholinergic stimulation of cyclic GMP levels was observed at a much lower calcium concentration (0.2 mM) than the half-maximal effect of calcium on the stimulation of CA release (2.5 mM; Fig. 1). Second, both CA secretion and cyclic GMP levels were stimulated by a depolarizing concentration of K’ in the presence of calcium; however, the stimulation of CA secretion by the depolarizing drugs veratridine and ouabain was not accompanied by any rise in cyclic GMP levels (Table 1). Third, although the calcium ionophore X537A was a much more potent secretagogue than
A23187, it had no effect on cyclic GMP levels (Table 1). Fourth, the substitution of calcium by Ba2+ and Sr2+ was more effective in maintaining the cyclic GMP increase than the stimulation of CA release in response to ACh (Table 2). Finally, both responses could be selectively and independently inhibited by the use of distinct types of calcium channel blockers, that is the divalent cations CO’+, Mg2+ and Ni” or the organic compounds verapamil and D-600 (Table 3). These data clearly indicate that the stimulations of CA secretion and of cyclic GMP levels in our cell preparation are regulated by calcium via distinct mechanisms. The requirement of calcium in the cholinergic stimulation of cyclic GMP levels agrees with the data obtained from other tissues.25-28 Calcium could regulate cyclic GMP levels by stimulating guanylate cyclase29 or by inhibiting cyclic GMP phosphodiesterase.” The direct stimulation of guanylate cyclase could occur either at the membrane level or inside the chromaffin cell. However, since Aunis et a1.3’ have demonstrated that 46% of guanylate cyclase in the adrenal medulla is located in plasma membranes while only 40% is found in the soluble fraction. the stimulation of guanylate cyclase by calcium may well occur at the membrane level. The divalent cations Mg2+, Co’+ and Ni2+ block voltage-sensitive calcium channels2’-23 and inhibit the release CA from perfused adrenal glands.32-33 However, these same ions may also compete with calcium at its sites and inhibit a direct stimulation of guanylate cyclase by calcium ions. On the other hand, although verapamil and D-600 are also well-known as specific blockers of calcium channels,24 they are less likely to interact directly with calcium sites. The inability of these two latter compounds to interact with calcium sites on the chromaffin cell membrane could explain their relative inefficiency in antagonizing the cyclic GMP response. One must not exclude however the importance of calcium influx through voltage-
ROLE OF CALCIUM
IN ADRENOMEDULLARY
467
CELLS
sensitive calcium channels for the cholinergic stimulation of cyclic GMP levels in the chromaffin cell preparation since verapamil and D-600 (at IO-’ M) blocked 20’525% of the cyclic GMP response (Table 3). The finding that X537A stimulated CA secretion more efficiently than A23187 (Table I) was in accordance with the data obtained with the perfused However, it is difficult to explain adrenal gland.“.” why X537A did not stimulate cyclic GMP levels while A23187 was effective. X537A may have other effects that prevent the accumulation of cyclic GMP; however, in our experiments, it did not inhibit the elevation of cyclic GMP produced by ACh or A23 187 (data not shown). A23187 also raises cyclic GMP levels in other tissues.26m’7 This effect was related to an increase in Cal’ influx since A23 187 facilitates calcium transport through plasma membrane in the direction of the calcium concentration gradient.““’ The stimulatory effect of A23187 on cyclic GMP levels in our cell preparation may also be caused by an increase in Ca” influx. The depolarizing effects of K’, veratridine and ouabain also stimulate the release of CA from the adrenal medulla. However, these agents act via distinct mechanisms. High concentrations of extracellular K’ depolarize cells and, in the presence of extracellular calcium, stimulate the release of CA.34 Veratridine depolarizes nerve cells by opening sodium channels,‘” and also induces the Ca”-dependent release of CA from the adrenal medulla.35 Finally, the inhibition of (Na + K)-ATPase by ouabain16 is also a potent stimulus for CA secretion.36 In our cell preparation, although all these compounds were potent secre-
tagogues, only high K’ could stimulate cyclic GMP levels (Table I). Such discrepancies between the stimulation of cyclic GMP levels and that of CA secretion favor the concept of distinct mechanisms for the regulation of both responses. The present data do not distinguish between differences in the mechanism of calcium permeability and differences in the mechanism of action of calcium. Since both actions of ACh are mediated by distinct receptors, that is the nicotinic receptors for CA secretion and the muscarinic receptors for cyclic G MP increase (Table 1, ref. 9), any differences in the sensitivity of these actions of ACh to divalent cations or to inhibitors of calcium permeability may be due to differences in the properties of these two classes of cholinergic receptors, or their associated channels. It is quite interesting to notice for instance that in chick sympathetic ganglia, muscarinic actions of acetylcholine were sensitive to lower concentrations of calcium than were nicotinic actions.” The particular sensitivity of the cyclic GMP response in our cell preparation to low concentrations of calcium (ED,,: 0.2 mM, compared with 2.5 mM for stimulation of CA secretion, Fig. I) seems to indicate that the rise in cyclic GMP possibly participates in an important physiological role. The modulation of CA release by the muscarinic stimulation of cyclic GMP levels as previously suggested’.” remains a matter to be investigated. ACKNOWLEDGMENT We are
greatly
indebted
to Michelle
Lemieux
support. This work was supported by the Medical of Canada,
The Canadian
Heart
Foundation
for secreterial
Research Council
and “Les Fondations
des maladies du Rein.”
REFERENCES I. Guidotti
A, Hanbauer
I, Costa E: Role of cyclic nucleotides in
the induction of tyrosine hydroxylase..
in GI Drummond,
P Green-
gard, GA Robinson (eds): Advances in Cyclic Nucleotide
Research,
vol 5. pp 619-639,
New York, Raven Press, 1975
2. Otten V, Mueller
RA, Thoenen
between increase in c-AMP
hydroxylase
in the rat adrenal
Arch Phnrmacol285:2333242, 3. Peach MJ: Stimulation adenosive
3’. 5’cyclic
Naunyn-Schmiedeberg’s
Ca*+.
and
Proc Natl
theophylline
Acad Sci USA
by
in the 69:834-
4. Gutman
Y.
Boonyaviroj
P: Activation
cyclase and catecholamine
berg’s Arch Pharmacol 5. Goldberg
ND,
307:3944,
Haddox
of adrenal
medulla
secretion. Naunyn-Schmiede-
8. Schneider GMP
AS,
accompanies
9. Yanagihara tor-mediated
Cyclic
GMP
metabolism
and
Ann Rev Biochem 46:823-896,
P: Cyclic nucleotides, protein phosphorylation
function.
in Cl
Drummond.
P Greengard,
GA
1977 HT.
Lemaire
catecholamine
S: Rapid
secretion
in
rise in cyclic suspensions
N, lsosaki M, Ohuchi T, et al: Muscarinic
increase
release
from
and
Robinson
in cyclic
of
GMP
ievels in isolated
Rahwan
medulla
and
the
coupling. J Physiol 167:288 RG.
recepbovine
1979
Rubin RP: The mechanism of catecholamine
the adrenal
stimulus-secretion
Borowitz
JL:
role of calcium
in
3 16, 1963
Mechanisms
secretion coupling in adrenal medulla. Pharmaceutical
of
stimulus-
Sci 62:191 I-
1923, 1973 12. Hochman
1977 6. Greengard
Cline
adrenal medullary cells. FEBS Lett 105:296298,
1 I.
1979
MK:
involvement in biological regulation.
neuronal
JA: Cyclic nucleotides and nervous system func-
IO. Douglas WW,
836, 1972 adenylate
7. Nathanson
Res. vol 5. New York. Raven
1975
adrenal chromaffin cells. Life Sci 24: I38991 394, 1979
1974 of release of adrenal catecholamine
monophosphate
absence of extracellular
and induction of tyrosine
medulla.
Press, pp 5855602,
tion. Physiol Rev 57: 157-240,
H: Evidence against a causal
relationship
(eds): Advances in Cyclic Nucleotide
J,
Perlman
RL:
Catecholamine
secretion
by
isolated adrenal cells. Biochim Biophys Acta 421: 16% 175. 1976 13. Miura
Y. Campese V. De Quattro
V et al: Plasma catecho-
468
lamines via an improved fluorimetric assay: comparison with an enzymatic assay. J Lab Clin Med 89:421-427, 1977 14. Frandsen EK, Krishna G: A simple ultrasensitive method for the assay of cyclic AMP and cyclic GMP in tissues. Life Sci 18:529-542 15. Ohta M. Narahashi T, Keeler RF: Effects of veratrum alkaloids on membrane potential and conductance of squid and crayfish giant axons. J Pharmac Exp Ther 184: 143-l 54, 1973 16. Banks J: The effect of ouabain on the secretion of catecholamines and on the intracellular concentration of potassium. J Physiol 193:63 l-637, 1967 17. Reed PW, Lardy HA: A23 187: a divalent cation ionophore. J Biol Chem 24716970-6977, 1972 18. Entman ML, Gillette PC, Wallick ET, et al: A study of calcium binding and uptake by isolated cardiac sarcoplasmic reticulum: the use of a new ionophore (X537A). Biochem Biophys Res Commun 48:847-853, 1972 19. Douglas WW, Rubin RP: The effects of alkaline earth and other divalent cations on adrenal medullary secretion. J Physiol 175:231-241, 1964 20. Douglas WW, Rubin RP: Stimulant action of barium on the adrenal medulla. Nature 203:305-307, 1964 21, Hagiwara S, Nakajima S: Differences in Na and Ca spikes as examined by application of tetrodotoxin, procaine and manganese ions. J Gen Physiol 49:793-806, 1966 22. Baker PF, Meves H, Ridgeway EB. Effects of manganese and other agents on the calcium uptake that follows depolarization of squid axons. J Physiol 23 1:5 1l-526, 1973 23. Kohlhardt M, Bauer B, Krause H. et al: Selective inhibition of the transmembrane Ca-conductivity of mammalian myocardial fibres by Ni, Co and Mn ions. Pflugers Arch 338:115-123, 1973 24. Fleckenstein A: Specific inhibitors and promotors of calcium action in the excitation-contraction coupling of heart muscle and their role in the prevention of production of my cardial lesions, in P Harris, L Ophe (eds): Calcium and the heart. New York and London, Academic Press, 197 1, pp 1355188 25. Schultz G, Hardman JG, Schultz K. et al: The importance of calcium ions in the regulation of guanosine 3’,5’-cyclic monophosphate levels. Proc Nat1 Acad Sci 70:3889-3893, 1973 26. DeRubertis FR, Craven PA: Calcium-dependent regulation of guanosine 3’,5’-monophosphate in renal cortex: effects of ionophore A23 187 and tetracaine and evidence for independent control of adenosine 3’,5’-monophosphate. Metabolism 25: I 112-l 127, 1976 27. Butcher FR: The role of calcium and cyclic nucleotides in
LEMAIRE
ET AL.
cu-amylase release from slices of rat parotid: studies with the divalent cation ionophore A23187. Metabolism 24:409-418, 1975 28. Van Sande J, Decoster C, Dumont JE. Control and role of cyclic 3’,5’-guanosine monophosphate in the thyroid. Biochem Biophys Res Commun 62: 168-I 75, 1975 29. Hardman JG, Beavo JA, Gray JP. Chrisman TD, Patterson WD, Sutherland EW. The formation and metabolism of Cyclic GMP. Ann NY Acad Sci 185:27-35, 1971 30. Kakiuchi S, Yamazaki R, Teshima Y. Regulation of brain phosphodiesterase activity: Ca2’ plus Mg”-dependent phosphodiesterase and its activating factor from rat brain. in P Greengard, R Paoletti, GA Robison (eds): Advances in cyclic Nucleotide Research, vol. I. New York, Raven Press, 1972, p 455 3 I. Aunis D. Pescheloche M. Zwiller J: Guanylate cyclase from bovine adrenal medulla: subcellular distribution and studies on the effect of lysolecithin on enzyme activity. Neuroscience 3:83-93, 1978 32. Rubin RP, Feinstein MB, Jaanus SD. et al: Inhibition of catecholamine secretion and calcium exchange in perfused adrenal glands by tetracaine and magnesium. J Pharmacol Exp Ther 155:4633471, 1967 33. Pinto JEB, Trifaro JM. The different effects of D-600 (methoxyverapamil) on the release of adrenal catecholamines induced by acetylcholine. high potassium or sodium deprivation. Brit J Pharmacol57:127~132, 1976 34. Smith AD, Winkler H. Fundamental mechanism in the release of catecholamines. in Blaschko H, Muscholl E. (eds): Handbuck Exptl Pharmak, vol 33. Berlin-Heidelberg-New York, Springer.
1972. pp 53886 17
35. Kirpekar SM. perfused cat adrenal 76:2081-2083, 1979
Prat JC: Release of catecholamines from gland by veratridine. Proc Nat1 Acad Sci
36. Garcia AG. Kirpekar SM. Release of noradrenaline from the cat spleen by sodium deprivation. Brit J Pharmacol 47:729-747. 1973 37. Green LA, Rein G: Release of norepinephrine from neurons in dissociated cell cultures of chick sympathetic ganglia via stimulation of nicotinic and muscarinic acetylcholine receptors. J Neurothem 30:579-586, 1978 38. Derome G, Tseng R, Mercier P. et al: Possible muscarinic regulation of catecholamine secretion mediated by cyclic GMP in isolated bovine adrenal chromaffin cells. Biochem Pharmacol (in press)
Regulations Secretion
Simon Lemaire,
by Calcium of Cyclic GMP Levels and in Isolated Bovine Adrenal Chromaffin Cells
Gilles Derome,
Richard Tseng,
Paulette
Mercier,
and Irma Lemaire
The effects of various calcium-dependent secretagogues on cyclic GMP levels and catecholamine (CA) secretion were measured in a preparation of bovine adrenal chromaffin cells. The secretory effect of acetylcholine (ACh; 8-10 fold stimulation) was mimicked by nicotine but not muscarine. Three-five fold stimulations of cyclic GMP levels were also obtained with ACh and muscarine but not nicotine. High concentration of K+, and the ionophore A23187, also elevated cyclic GMP levels. However, secretion produced by veratridine, ouabain, and the ionophore X537A was not accompanied by any rise in cyclic GMP levels. Removal of extracellular calcium significantly decreased both basal levels of CA secretion and of cyclic GMP and completely abolished their stimulation by ACh. The half-maximal effects of calcium on the cholinergic stimulations of cyclic GMP levels and of CA secretion were observed at 0.2 and 2.5 mM, respectively. Substitution of Ca** by S? was more effective in maintaining the cyclic GMP response than the secretory response. The calcium channel blockers Co’+, Mg” and Ni2+ .Inhibited the cholinergic stimulation of cyclic GMP more than that of CA release. On the other hand, the organic calcium channel blockers. verapamil and methoxyverapamil (O-800) were more effective antagonists of the secretory response. These data indicate that the cholinergic stimulations of CA secretion and of cyclic GMP levels in bovine adrenal chromaffin cells are regulated by calcium via two distinct mechanisms.
T
HE ROLE OF CYCLIC AMP as an effector of both CA synthesis’** and CA secretion3*4 in the adrenal medulla, has been extensively examined, but only limited data are available concerning the role of cyclic GMP.’ Studies in other tissues have indicated that cyclic GMP regulates numerous biological functions’ and it has been suggested that it may have a neuronal function and mediate postsynaptic transmission.6*7 Recently, using a preparation of bovine adrenal chromaffin cells, we have observed that the cholinergic stimulation of CA secretion by ACh is accompanied by an important increase in cyclic GMP levels8 and a muscarinic receptor was involved in this effect.’ Since calcium plays a key function in stimulus-secretion coupling in the adrenal medulla,‘O~” it was important to study its role in the cholinergic stimulation of cyclic GMP in our cell preparation and to establish the relations between the effects of calcium on cyclic GMP levels and on CA secretion. The present data indicate that calcium is required both for stimulation of CA secretion and for the regulation of cyclic GMP levels in bovine adrenal chromaffin cells. Furthermore, the use of specific calcium-dependent secretagogues, calcium substitutes and calcium-channel blockers has distinguished the effects of calcium in both phenomena.
From the Department of Pharmacology, Centre Hospitalier Universitaire. Sherbrooke. Quebec, Canada JIH 5N4. Received for publication June 20, I980. Address reprint requests to Simon Lemaire, Department of Pharmacology, Centre Hospitalier Universitaire. Sherbrooke, Quebec, Canada JIH 5N4. (c I981 by Grune & Stratton. Inc. 002&0495/81/3005-0009~01.00/0
462
MATERIALS
AND
METHODS
Materials Radioimmune assay kits for the measurement of cyclic nucleotides were purchased from Schwarz-Mann (Boston, Mass.). Cyclic GMP and cyclic AMP were obtained from Boehringer Mannheim (W. Germany). ACh, nicotine, muscarine, veratridine and ouabain were purchased from Sigma Chemical Co. (St. Louis. MO.). Verapamil and methoxyverapamil (D-600) were generously supplied by A.G. Knoll Ltd (Chemische Fabriske) Ludwigshafen am Rhein. A23 I87 and X537A were kindly provided by Eli Lilly & Co. and Hoffmann LaRoche Ltd, respectively. Fresh bovine adrenal glands were obtained in a local slaughter house.
Cell Preparation Bovine adrenal glands were rapidly removed from the animal after exsanguination and transported to the laboratory on ice. Chromaffin cells were isolated with a few modifications of the procedure of Hochman and Perlman.‘* Briefly, the medullae of 6-8 glands were dissected free of cortical tissue and placed in ice-cold calcium-free Krebs-Ringer buffer, pH 7.4 (buffer I: NaCI, 1I8 mM; KH,PO,, 1.2mM; MgSO,, I .2 mM; NaHCO,; 25 mM; glucose, IO mM: bovine serum albumin, 0.5%). The medullary tissue was cut into small pieces (-2 mm), washed three times with 30 ml of buffer I. and then submitted to five successive 30 min digestions with collagenase (1.5 mg/ml of collagenase CLS Type I from Worthington) in the same buffer at 37°C. The dispersed cells were removed after each digestion and filtered through four layers of cheese cloth. The first harvest, containing mostly red blood cells and debris, was discarded and the four others were washed three times with 30 ml of buffer I supplemented with 2.2 mM CaC12 (Buffer II) and collected by centrifugation at I IO x g for 10 min at room temperature. The isolated cells were pooled and added to 40 ml of buffer 11. Chromaffin cells were purified by differential sedimentation at room temperature and unit gravity for 2 hr. During this process, the chromaffin cells sedimented as a white pellet leaving most debris and red blood cells in the supernatant. The supernatant was then discarded and the chromaffin cells (pellet) were resuspended in 6 ml of buffer II containing 100 units of penicillin and 100 pg of streptomycin. The cells were gassed with a mixture of O,/CO, (95%/5%) and stored overnight in a capped plastic tube at room temperature. The next day, the cells were washed with 30 ml of buffer II and counted in a
Metabolism, Vol. 30, No. 5 (May). 1981
ROLE OF CALCIUM IN ADRENOMEDULLAAY
hemocytometer.
Viability
(97%) was determined
trypan blue. This procedure chromaffin ceils.
Stimulation
usually
of Catecholamine
yielded
463
CELLS
by the
million
Secretion
CA secretion was assayed as described.* Briefly, the secretion was started by the addition of 2.5 x lo5 chromafhn cells (0.1 ml in buffer II) to a prewarmed (37OC) solution (0.9 ml) of buffer II containing the various drugs to be tested. Secretion was stopped after 5 min by transferring the tubes to an ice-water bath. After cooling the samples for 3 to 5 min. the cells were centrifuged at I10 x g for 10 min and 0.7 ml aliquots of the supernatants were collected. Proteins were precipitated by the addition of 70 ~1 of cold 50% trichloroacetic acid (tinal concentration: 5%) and removed by centrifugation at 900 x g for 30 min. The clear supernatants were collected and stored at -20°C for subsequent CA determination. Secretory experiments were performed in triplicate and controls were obtained by the incubation of the cells in the absence of drugs at 0°C (control to be subtracted) and 37°C (basal level of CA secretion).
Measurement
of Catecholamines
For total CA determination
(epinephrine plus norepinephrine), aliquots (200 ~1) of the above samples were added to 800 ~1 of I M sodium borate buffer, pH 8.0 (final pH: 7.0) and oxidized by the procedure of Miura et al.” Fluorescence was monitored on a spectrofluorometer (Carl Zeiss. model ZFM 4) at excitation/ emission wavelengths of 4 I O/5 I5 nm. respectively. At these wavelengths, the emission intensities of epinephrine and norepinephrine were equal. Each experiment was repeated three times in triplicate and values are expressed as nmole of CA secreted per IO6 cells per 5 min of incubation.
Cyclic Nucleotide
Effect of Various Drugs on Catecholamine and Cyclic Gh4P Levels
Secretion
Various drugs can induce the release of CA from perfused adrenal glands in the presence of calcium. Among these drugs are cholinergic agonists, calcium ionophores, and depolarizing agents. Table 1 shows the effects of some of these drugs on CA secretion and cyclic GMP levels in bovine adrenal chromaffin cells. ACh (SO PM) induced an 8.1-fold stimulation of CA release and a 3.1-fold increase in cyclic GMP levels. The secretory effect of ACh was mimicked by nicotine (7.2-fold stimulation), but not muscarine. Conversely, cyclic GMP levels were elevated by muscarine (3-fold stimulation), but not by nicotine. The secretory effect of increased K’ (56 mM; 5.1-fold stimulation) was accompanied by a 2.1-fold increase in cyclic GMP levels. On the other hand, veratridine and ouabain, also known to depolarize nerve cells,‘5,‘h stimulated CA secretion (6.18 and 2.52-fold stimulation, respectively) without affecting cyclic GMP levels. The involvement of calcium ions in both responses was further supported by the use of the calcium ionophores, X53lA and A23187.“,” X537A was a more potent secretagogue than A23 187 (5.09- and 2.13-fold stimulation of CA release, respectively). Only A23187, however, elevated cyclic GMP levels (2.87fold stimulation).
Assuy
The cyclic nucleotide assay was started by the addition of 5 x IO5 chromaffin cells (0.1 ml in buffer II) to a pre-warmed (37OC) solution of bulfer II (0.9 ml) containing the various drugs to be tested. The reaction was stopped after 3 min by the addition of 1 ml of cold 20X trichloroacetic acid and freezing and thawing three times respectively in liquid nitrogen and a water bath at 37°C. The protein precipitate was then removed by centrifugation at 900 x g for 30 min at W”C. The supernatants were collected and transfered into a boiling-water bath for 3 min. The samples were extracted five times with 4 ml portions of water-saturated ether, evaporated under vacua and dissolved with 0.3 ml of water. They were then stored at ~~2Q°C for subsequent cyclic nucleotide determination. Assays were performed in triplicate and controls were obtained by the incubation of the cells in the absence of drugs. Maximal stimulation of cyclic G MP levels was obtained at approximately 30 set of incubation with ACh and was maintained for the first 10 min.’ The 3 min incubation time was choosen mainly because it represents a maximal time for stimulation. Furthermore. measurements at 3 min were more accurate then at 30 sec.
Determination
RESULTS
exclusion of
fifty to sixty
of Cyclic GMP
Cyclic GMP was measured from 30 /rl aliquots following the radioimmune techniques of Frandsen and Krishna.14 Recovery (over 95%) was calculated by the addition of [‘HI-cyclic GMP (2000 CPM) to the cyclic nucleotide assay as described previously. Nonspecific binding (determined after digestion wrth phosphodiesterase, Sigma) was estimated to be smaller than 5%. Values are expressed as picomoles of cyclic GMP per IO6 cells.
Effect of Calcium
and Sodium
Removal
To verify the importance of extracellular calcium and sodium ions on the stimulatory effects of some of these drugs, cells were incubated in calcium or sodium-free buffers (Table 1). Removal of extracellular calcium significantly decreased both basal levels of cyclic G MP (from 0.4 to 0.3 pmoles/ 1O6cells) and CA secretion (from 0.65 to 0. I6 nmole/ IO’ cells). Furthermore, the absence of calcium completely abolished the stimulations of cyclic GMP levels and of CA secretion by ACh, high K ’ and A23 187. On the other hand, removal of sodium ions from the incubation medium did not modify the basal levels of cyclic GMP and of CA release but significantly depressed their stimulation by ACh (Table 2). Sodium removal also decreased the secretory response to high K +; however the cyclic GMP response to high K + was much less affected by sodium deprivation. Effect of Calcium
Concentration
To verify the sensitivity of both cyclic GMP and secretory responses to extracellular calcium, cells were incubated in the presence of increasing concentrations of calcium with or without ACh (50 PM; Fig 1).
464
LEMAIRE
ET AL.
Table 1. Effect of Various Secretagogues on CA Secretion and Cyclic GMP Levels in Isolated Bovine Adrenal Chromaffin Cells Catecholamine
Secretion
Cyclic GMP PsrC@nt
Secretagogue Normal
nmoles/ 1Oh Cells
P.WCl?fX
Cnntr0l
pmoles/ 106 Cells
Control
Buffer
Control
0.65 k 0.12
100
0.40
ACh (50 pM)
5.27
+ 0.32
810
1.25 t 0.10
Nicotine (50 /.LM)
4.71
+ 0.22
724
0.38
Muscarine (50 fiLMI
0.71
+ 0.18’
109
1.21 + 0.07
302
KCI (56 mM)
4.34
+ 0.43
667
0.82
-t 0.08
205
3.31
* 0.19
509
0.43
* 0.09*
1.39 +- 0.23
213
1.15 * 0.12
lonophore X537A
I10 FM)
lonophore A23187
(10 PM)
+ 0.06
100 312
k 0.05*
95
107 287
Veratridine I10 PM)
4.02
? 0.31
618
0.42
+ 0.05’
105
Ouabain (10 PM)
1.64 i 0.16
252
0.41
* 0.04*
102
Calcium-free Buffer Control
0.16
k 0.05
-
0.30
+ 0.05
100
ACh (50 PM)
0.19
+ 0.10’
-
0.32
+ 0.07,
106
KCI (56 mM)
0.13
+ 0.08*
-
0.30
* 0.04,
100
0.08
t 0.04*
-
0.33
k 0.09+
110
lonophore A23187
(10 PM)
Sodium-free Buffer Control ACh
(50
KCI (56
(56
mM)
0.65
k 0.10
100
0.42
k 0.02
100
1.06
5 0.13
160
0.96
k 0.05
228
1.88
* 0.16
290
0.78
t
185
0.12
experiments were repeated three times in triplicate and values are mean t SD (statistically different of control: P -c .05 Student’s
The when
/.rM)
indicated*). mM),
tris (143
Catecholamine
an equivalent mM,
pH 7.2,
amount HCI)
secretion of NaCl
to replace
and cyclic (56
mM)
the sodium
GMP
was
levels
removed
ion. Total
were from
measured the
CA content:
as described
incubation 34.9
Calcium induced a dose-dependent increase in basal levels of cyclic GMP (from 0.3 to 0.4 pmoles between 0 and 5 mM CaCl,) and of CA secretion (from 0 to 1 nmole of CA). The stimulatory effects of ACh were also calcium-dependent. The half-maximal effects of calcium on ACh-stimulated cyclic GMP levels and CA secretion were observed at 0.2 and 2.5 mM CaCl,, respectively. Maximal stimulation of cyclic GMP levels was also obtained at a lower concentration of calcium (-5 mM) than the maximal secretory response (a 10 mM).
under
medium
k 3.4
“Materials
to maintain
nmoles/106
Replacement
and Methods.”
the osmolality.
t test,
For stimulation
Sodium-free
buffer
except with
K+
contained
cells.
of Ca’+ by Ba” and Sr2+
In perfused adrenal glands, both Ba2+ and Sr” are good substitutes for Ca2+ for the stimulation of CA release.‘9,20 It was of interest to verify if such replacements could also maintain cyclic GMP and secretory responses in our preparation of bovine adrenal chromaffin cells. Table 2 shows that either Ba2’ or Sr2+ can replace Ca2+ for stimulations of cyclic GMP levels and CA secretion by ACh. However, Ba*+ itself induced important increases in basal levels of CA
Table 2. Effect of the Replacement of Calcium by Barium or Strontium Ions on the ACh-induced Stimulations of CA Secretion and Cyclic GMP Levels in Isolated Bovine Adrenal Chromaffin Cells Catecholamine
Secretion
Cyck
GMP Percent
Percent Addition Normal
(50
Calcium-free
PM) buffer
+
B&I,
(2.2
(50
Calcium-free
fiMI buffer
+
SrCI,
(2.2
The assays
(50 were
to the incubations. test).
Control
PM) performed
as described
The experiments
were
0.54
* 0.12
5.57
* 0.43
100 1,031
0.36
k 0.03
100
1.24
+ 0.07
344
6.36
_c 0.38
100
0.54
+ 0.04
100
9.56
+ 0.44
150
0.98
t
181
0.06
mM)
Control ACh
1O6 Cells
mM)
Control ACh
pm&s/
Buffer
Control ACh
Control
nmoles/ 1O6 Cells
0.76
k 0.16
100
0.34
+ 0.03
100
2.81
+ 0.19
370
0.85
+ 0.05
250
under repeated
“Materials 3 times
and Methods, in triplicate
” except
and values
that
the cells were
are mean
washed
k SD (statistically
with different
15 ml of the indicated of control:
buffer
prior
P rc .05 Student’s
t
ROLE OF CALCIUM
IN ADRENOMEDULLARY
465
CELLS
secretory response respectively).
(from
10.3- to 1.5- and
3.7-fold
EJect of Calcium Channel Blockers Many compounds such as the divalent cations Co”, Ni” and Mg” or the organic drugs, verapamil and methoxyverapamil (D-600) block the entry of calcium in nerve cells through voltage-sensitive calcium channels.*’ 24 Figure 2 and Table 3 illustrate the effect of these compounds on the cholinergic stimulations of cyclic GMP levels and of CA secretion in our cell preparation. Cobalt (Fig. 2) inhibited both responses, but the cyclic GMP response was six times more
1
w I
0
I
I
2 4 6 CALCIUM
I
I
8 IO (mM1
0
0.2 0.4 0.6 0.8 1.0
Fig. 1. Effect of increasing concentrations of calcium on basal 1-1 and ACh-stimulated G---O1 cyclic GMP levels IA) and CA secretion fB). The experiments were performed as described under “Materials and Methods” and values are mean t SD fstatisticslly different of the zero Cai ’ concentration control: P < .05. Student’s t test).
secretion (from 0.54 to 6.36 nm&s/ ItI6 cells) and of cyclic GMP (from 0.36 to 0.54 pmole/106 ceils). In the presence of Ba”, the stimulatory effects of ACh were much depressed (from 10.3- to 1.5-fold stimulation of CA release and from 3.44- to I .8 I -fold stimulation of cyclic GMP levels). However, this large decrease in the efficiency of ACh to stimulate CA secretion or cyclic GMP levels may depend on the stimulatory effects of Ba” itself. Sr2+ did not affect basal values and was a better substitute for maintaining both responses to ACh. Interestingly, the cyclic GMP response to ACh was less affected by the replacement of Ca” with Ba2+ or Sr’+ (from 3.44- to 1.81- and 2.5-fold stimulation, respectively) than the
o-
0
0.2 0.4 0.6 0.8 1.0 COBALTcmM)
Fig. 2. Effect of increasing concentrations of cobalt on basal (U) and ACh-stimulated (0-Q cyclic GMP levels (Al and CA secretion (5). The experiments were performed as described under “Materials and Methods” and values are mean 2 SD (variations in the presence of ACh were statistically different of the zero Co’ ’ concentration: P c .05, Student’s t test whereas those observed in its absence were nonsignificant).
466
LEMAIRE ET AL.
Table 3.
Effect of MgZf. Ni’+, Verapamil
and D-600
on the Cholinergic
and Cyclic GMP Levels in Isolated Bovine Adrenal Catecholamme
AddWon
Secretion
Inmoles/1O6cells)
InhibitIon
(%)
Stimulations
Chromaffin
of CA Secretion
Cells Cyclic GMP (pm&s/ 1O6cellsl
Inhibition 1%)
Ach (50 /JM)
4.70
+ 0.35
0
0.79
* 0.04
0
Ach + MgCI, (10 mM)
2.91
? 0.23
38
0.04
* 0.02
95
Ach + MgCI, (20 mM)
0.73
-t 0.15
84
0 + 0.01
100
Ach + NiCI, (1 mM)
2.67
% 0.19
43
0.03
i 0.01
Ach + verapamil (10 PM)
1.66 + 0.20
65
0.62
+ 0.02
21
Ach + D-600
1.55 + 0.14
66
0.60
+ 0.01
24
(10 PM)
96
The experiments were repeated 3 times in triplicate and values are mean + SD (statistically different of the assay with ACh alone: P < .05 Student’s t test). “Catecholamine secretion” represents the amount of CA released by ACh after subtraction of the basal release of 0.5
+ 0.1 nmole/lOB cells.
“Cyclic GMP” represents the increase in cyclic GMP levels evoked by ACh after the subtraction of the basal level of 0.4 * 0.08 pmoles/108 cells. The experiments were performed as described under “Materials and Methods.”
sensitive to Co’+ (ED,,: 0.05 mM) than the secretory response (EDSo: 0.33 mM). Increasing concentrations of Co*+ (up to 1 mM) had however no effects on the basal levels of cyclic GMP and of CA secretion. Stimulation of cyclic GMP levels by ACh was also more sensitive to the presence of Mg2+ (10 and 20 mM) and Ni’+ (1 mM) than the secretory response (Table 3). Interestingly, the organic calcium channel blockers verapamii and D-600 (at 10 FM) inhibited the secretory response more completely (65%-66% inhibition) than the stimulation of cyclic GMP levels (2 I %-24% inhibition). None of these calcium channel blockers affected the basal levels of CA secretion and of cyclic GMP (legend of Table 3). It then appears that each response can be selectively blocked by one or the other type of calcium channel blockers. DISCUSSION
The present studies indicate that calcium is required for the regulation of CA secretion and of cyclic GMP levels in adrenal chromafhn cells. This concept was first supported by the stimulation of CA release with the calcium ionophores X537A and A23 187 and of cyclic GMP levels with A23 187 (Table 1). Furthermore, removal of calcium completely abolished both responses to ACh, high K+ and A23187 (Table 1). However, many distinctions were observed between the regulation of both responses by calcium. First, the half-maximal effect of calcium on the cholinergic stimulation of cyclic GMP levels was observed at a much lower calcium concentration (0.2 mM) than the half-maximal effect of calcium on the stimulation of CA release (2.5 mM; Fig. 1). Second, both CA secretion and cyclic GMP levels were stimulated by a depolarizing concentration of K’ in the presence of calcium; however, the stimulation of CA secretion by the depolarizing drugs veratridine and ouabain was not accompanied by any rise in cyclic GMP levels (Table 1). Third, although the calcium ionophore X537A was a much more potent secretagogue than
A23187, it had no effect on cyclic GMP levels (Table 1). Fourth, the substitution of calcium by Ba2+ and Sr2+ was more effective in maintaining the cyclic GMP increase than the stimulation of CA release in response to ACh (Table 2). Finally, both responses could be selectively and independently inhibited by the use of distinct types of calcium channel blockers, that is the divalent cations CO’+, Mg2+ and Ni” or the organic compounds verapamil and D-600 (Table 3). These data clearly indicate that the stimulations of CA secretion and of cyclic GMP levels in our cell preparation are regulated by calcium via distinct mechanisms. The requirement of calcium in the cholinergic stimulation of cyclic GMP levels agrees with the data obtained from other tissues.25-28 Calcium could regulate cyclic GMP levels by stimulating guanylate cyclase29 or by inhibiting cyclic GMP phosphodiesterase.” The direct stimulation of guanylate cyclase could occur either at the membrane level or inside the chromaffin cell. However, since Aunis et a1.3’ have demonstrated that 46% of guanylate cyclase in the adrenal medulla is located in plasma membranes while only 40% is found in the soluble fraction. the stimulation of guanylate cyclase by calcium may well occur at the membrane level. The divalent cations Mg2+, Co’+ and Ni2+ block voltage-sensitive calcium channels2’-23 and inhibit the release CA from perfused adrenal glands.32-33 However, these same ions may also compete with calcium at its sites and inhibit a direct stimulation of guanylate cyclase by calcium ions. On the other hand, although verapamil and D-600 are also well-known as specific blockers of calcium channels,24 they are less likely to interact directly with calcium sites. The inability of these two latter compounds to interact with calcium sites on the chromaffin cell membrane could explain their relative inefficiency in antagonizing the cyclic GMP response. One must not exclude however the importance of calcium influx through voltage-
ROLE OF CALCIUM
IN ADRENOMEDULLARY
467
CELLS
sensitive calcium channels for the cholinergic stimulation of cyclic GMP levels in the chromaffin cell preparation since verapamil and D-600 (at IO-’ M) blocked 20’525% of the cyclic GMP response (Table 3). The finding that X537A stimulated CA secretion more efficiently than A23187 (Table I) was in accordance with the data obtained with the perfused However, it is difficult to explain adrenal gland.“.” why X537A did not stimulate cyclic GMP levels while A23187 was effective. X537A may have other effects that prevent the accumulation of cyclic GMP; however, in our experiments, it did not inhibit the elevation of cyclic GMP produced by ACh or A23 187 (data not shown). A23187 also raises cyclic GMP levels in other tissues.26m’7 This effect was related to an increase in Cal’ influx since A23 187 facilitates calcium transport through plasma membrane in the direction of the calcium concentration gradient.““’ The stimulatory effect of A23187 on cyclic GMP levels in our cell preparation may also be caused by an increase in Ca” influx. The depolarizing effects of K’, veratridine and ouabain also stimulate the release of CA from the adrenal medulla. However, these agents act via distinct mechanisms. High concentrations of extracellular K’ depolarize cells and, in the presence of extracellular calcium, stimulate the release of CA.34 Veratridine depolarizes nerve cells by opening sodium channels,‘” and also induces the Ca”-dependent release of CA from the adrenal medulla.35 Finally, the inhibition of (Na + K)-ATPase by ouabain16 is also a potent stimulus for CA secretion.36 In our cell preparation, although all these compounds were potent secre-
tagogues, only high K’ could stimulate cyclic GMP levels (Table I). Such discrepancies between the stimulation of cyclic GMP levels and that of CA secretion favor the concept of distinct mechanisms for the regulation of both responses. The present data do not distinguish between differences in the mechanism of calcium permeability and differences in the mechanism of action of calcium. Since both actions of ACh are mediated by distinct receptors, that is the nicotinic receptors for CA secretion and the muscarinic receptors for cyclic G MP increase (Table 1, ref. 9), any differences in the sensitivity of these actions of ACh to divalent cations or to inhibitors of calcium permeability may be due to differences in the properties of these two classes of cholinergic receptors, or their associated channels. It is quite interesting to notice for instance that in chick sympathetic ganglia, muscarinic actions of acetylcholine were sensitive to lower concentrations of calcium than were nicotinic actions.” The particular sensitivity of the cyclic GMP response in our cell preparation to low concentrations of calcium (ED,,: 0.2 mM, compared with 2.5 mM for stimulation of CA secretion, Fig. I) seems to indicate that the rise in cyclic GMP possibly participates in an important physiological role. The modulation of CA release by the muscarinic stimulation of cyclic GMP levels as previously suggested’.” remains a matter to be investigated. ACKNOWLEDGMENT We are
greatly
indebted
to Michelle
Lemieux
support. This work was supported by the Medical of Canada,
The Canadian
Heart
Foundation
for secreterial
Research Council
and “Les Fondations
des maladies du Rein.”
REFERENCES I. Guidotti
A, Hanbauer
I, Costa E: Role of cyclic nucleotides in
the induction of tyrosine hydroxylase..
in GI Drummond,
P Green-
gard, GA Robinson (eds): Advances in Cyclic Nucleotide
Research,
vol 5. pp 619-639,
New York, Raven Press, 1975
2. Otten V, Mueller
RA, Thoenen
between increase in c-AMP
hydroxylase
in the rat adrenal
Arch Phnrmacol285:2333242, 3. Peach MJ: Stimulation adenosive
3’. 5’cyclic
Naunyn-Schmiedeberg’s
Ca*+.
and
Proc Natl
theophylline
Acad Sci USA
by
in the 69:834-
4. Gutman
Y.
Boonyaviroj
P: Activation
cyclase and catecholamine
berg’s Arch Pharmacol 5. Goldberg
ND,
307:3944,
Haddox
of adrenal
medulla
secretion. Naunyn-Schmiede-
8. Schneider GMP
AS,
accompanies
9. Yanagihara tor-mediated
Cyclic
GMP
metabolism
and
Ann Rev Biochem 46:823-896,
P: Cyclic nucleotides, protein phosphorylation
function.
in Cl
Drummond.
P Greengard,
GA
1977 HT.
Lemaire
catecholamine
S: Rapid
secretion
in
rise in cyclic suspensions
N, lsosaki M, Ohuchi T, et al: Muscarinic
increase
release
from
and
Robinson
in cyclic
of
GMP
ievels in isolated
Rahwan
medulla
and
the
coupling. J Physiol 167:288 RG.
recepbovine
1979
Rubin RP: The mechanism of catecholamine
the adrenal
stimulus-secretion
Borowitz
JL:
role of calcium
in
3 16, 1963
Mechanisms
secretion coupling in adrenal medulla. Pharmaceutical
of
stimulus-
Sci 62:191 I-
1923, 1973 12. Hochman
1977 6. Greengard
Cline
adrenal medullary cells. FEBS Lett 105:296298,
1 I.
1979
MK:
involvement in biological regulation.
neuronal
JA: Cyclic nucleotides and nervous system func-
IO. Douglas WW,
836, 1972 adenylate
7. Nathanson
Res. vol 5. New York. Raven
1975
adrenal chromaffin cells. Life Sci 24: I38991 394, 1979
1974 of release of adrenal catecholamine
monophosphate
absence of extracellular
and induction of tyrosine
medulla.
Press, pp 5855602,
tion. Physiol Rev 57: 157-240,
H: Evidence against a causal
relationship
(eds): Advances in Cyclic Nucleotide
J,
Perlman
RL:
Catecholamine
secretion
by
isolated adrenal cells. Biochim Biophys Acta 421: 16% 175. 1976 13. Miura
Y. Campese V. De Quattro
V et al: Plasma catecho-
468
lamines via an improved fluorimetric assay: comparison with an enzymatic assay. J Lab Clin Med 89:421-427, 1977 14. Frandsen EK, Krishna G: A simple ultrasensitive method for the assay of cyclic AMP and cyclic GMP in tissues. Life Sci 18:529-542 15. Ohta M. Narahashi T, Keeler RF: Effects of veratrum alkaloids on membrane potential and conductance of squid and crayfish giant axons. J Pharmac Exp Ther 184: 143-l 54, 1973 16. Banks J: The effect of ouabain on the secretion of catecholamines and on the intracellular concentration of potassium. J Physiol 193:63 l-637, 1967 17. Reed PW, Lardy HA: A23 187: a divalent cation ionophore. J Biol Chem 24716970-6977, 1972 18. Entman ML, Gillette PC, Wallick ET, et al: A study of calcium binding and uptake by isolated cardiac sarcoplasmic reticulum: the use of a new ionophore (X537A). Biochem Biophys Res Commun 48:847-853, 1972 19. Douglas WW, Rubin RP: The effects of alkaline earth and other divalent cations on adrenal medullary secretion. J Physiol 175:231-241, 1964 20. Douglas WW, Rubin RP: Stimulant action of barium on the adrenal medulla. Nature 203:305-307, 1964 21, Hagiwara S, Nakajima S: Differences in Na and Ca spikes as examined by application of tetrodotoxin, procaine and manganese ions. J Gen Physiol 49:793-806, 1966 22. Baker PF, Meves H, Ridgeway EB. Effects of manganese and other agents on the calcium uptake that follows depolarization of squid axons. J Physiol 23 1:5 1l-526, 1973 23. Kohlhardt M, Bauer B, Krause H. et al: Selective inhibition of the transmembrane Ca-conductivity of mammalian myocardial fibres by Ni, Co and Mn ions. Pflugers Arch 338:115-123, 1973 24. Fleckenstein A: Specific inhibitors and promotors of calcium action in the excitation-contraction coupling of heart muscle and their role in the prevention of production of my cardial lesions, in P Harris, L Ophe (eds): Calcium and the heart. New York and London, Academic Press, 197 1, pp 1355188 25. Schultz G, Hardman JG, Schultz K. et al: The importance of calcium ions in the regulation of guanosine 3’,5’-cyclic monophosphate levels. Proc Nat1 Acad Sci 70:3889-3893, 1973 26. DeRubertis FR, Craven PA: Calcium-dependent regulation of guanosine 3’,5’-monophosphate in renal cortex: effects of ionophore A23 187 and tetracaine and evidence for independent control of adenosine 3’,5’-monophosphate. Metabolism 25: I 112-l 127, 1976 27. Butcher FR: The role of calcium and cyclic nucleotides in
LEMAIRE
ET AL.
cu-amylase release from slices of rat parotid: studies with the divalent cation ionophore A23187. Metabolism 24:409-418, 1975 28. Van Sande J, Decoster C, Dumont JE. Control and role of cyclic 3’,5’-guanosine monophosphate in the thyroid. Biochem Biophys Res Commun 62: 168-I 75, 1975 29. Hardman JG, Beavo JA, Gray JP. Chrisman TD, Patterson WD, Sutherland EW. The formation and metabolism of Cyclic GMP. Ann NY Acad Sci 185:27-35, 1971 30. Kakiuchi S, Yamazaki R, Teshima Y. Regulation of brain phosphodiesterase activity: Ca2’ plus Mg”-dependent phosphodiesterase and its activating factor from rat brain. in P Greengard, R Paoletti, GA Robison (eds): Advances in cyclic Nucleotide Research, vol. I. New York, Raven Press, 1972, p 455 3 I. Aunis D. Pescheloche M. Zwiller J: Guanylate cyclase from bovine adrenal medulla: subcellular distribution and studies on the effect of lysolecithin on enzyme activity. Neuroscience 3:83-93, 1978 32. Rubin RP, Feinstein MB, Jaanus SD. et al: Inhibition of catecholamine secretion and calcium exchange in perfused adrenal glands by tetracaine and magnesium. J Pharmacol Exp Ther 155:4633471, 1967 33. Pinto JEB, Trifaro JM. The different effects of D-600 (methoxyverapamil) on the release of adrenal catecholamines induced by acetylcholine. high potassium or sodium deprivation. Brit J Pharmacol57:127~132, 1976 34. Smith AD, Winkler H. Fundamental mechanism in the release of catecholamines. in Blaschko H, Muscholl E. (eds): Handbuck Exptl Pharmak, vol 33. Berlin-Heidelberg-New York, Springer.
1972. pp 53886 17
35. Kirpekar SM. perfused cat adrenal 76:2081-2083, 1979
Prat JC: Release of catecholamines from gland by veratridine. Proc Nat1 Acad Sci
36. Garcia AG. Kirpekar SM. Release of noradrenaline from the cat spleen by sodium deprivation. Brit J Pharmacol 47:729-747. 1973 37. Green LA, Rein G: Release of norepinephrine from neurons in dissociated cell cultures of chick sympathetic ganglia via stimulation of nicotinic and muscarinic acetylcholine receptors. J Neurothem 30:579-586, 1978 38. Derome G, Tseng R, Mercier P. et al: Possible muscarinic regulation of catecholamine secretion mediated by cyclic GMP in isolated bovine adrenal chromaffin cells. Biochem Pharmacol (in press)