0013.7227/92/1302-1017$03.00/0 Endocrinology Copyright 0 1992 by The
Vol. 130, No. 2 Endocrine
Society
Printed
Calcium: Its Role in al-Adrenergic Natriuretic Peptide Secretion* RICK
J. SCHIEBINGER,
HARRY
G. PARR,
AND
EDWARD
Stimulation J. CRAGOE,
in U.S.A.
of Atria1
JR.
Wayne State University, Detroit, Michigan 48201; the Veterans Administration Medical Center, Allen Park, Michigan 48101; and P.O. Box 631548 (E.J.C.), Nacogdoches, Texas 75963
porter activity. To determine whether the resulting rise in intracellular sodium may alter Na-Ca exchange to raise intracellular calcium levels, atria were superfused with the Na-H antiporter inhibitor, 5-(N,N-hexamethylene)amiloride. Superfusion with 25 FM 5-(N,N-hexamethylene)amiloride did not inhibit phenylephrine-stimulated ANP secretion. Lastly, the calcium dependency of the maintenance of an established response to phenylephrine was examined. Atria were superfused with phenylephrine in buffer containing 1.8 mM calcium for 45 min, followed by superfusion with phenylephrine in 0.2 mM calcium for 30 min. There was no fall in phenylephrine-stimulated secretion by atria superfused in 0.2 mM calcium. In contrast, addition of the ol,-adrenergic antagonist phentolamine induced an immediate fall in phenylephrine-stimulated ANP secretion. We conclude that 1) calcium influx is necessary to initiate ai-agonist-stimulated ANP secretion; 2) calcium release from the SR does not play a role in a,-agonist-stimulated secretion; 3) calcium entry through L-type calcium channels is responsible for half of the calcium influx; 4) enhanced Na-H antiporter activity does not play a role in a,-agonist-stimulated secretion; and 5) maintenance of ocl-agonist-stimulated secretion is not dependent on calcium influx. (Endocrinology 130: 1017-1023, 1992)
ABSTRACT.
ai-Adrenergic agonists increase atria1 natriuretic peptide (ANP) secretion. The mechanism of ol,-adrenergic-stimulated secretion is not known. In this study we examine the calcium dependency of Lu,-agonist-stimulated ANP secretion. Isolated superfused rat left atria paced at 2 Hz were used for study. Superfusion with 10 pM phenylephrine increased ANP secretion by P-fold. Lowering the superfusate calcium concentration from 1.8 to 0.2 mM totally negated the secretory response to phenylephrine. To determine whether this reflected a reduction in calcium influx, reduced calcium release from the sarcoplasmic reticulum (SR), or both, atria were superfused with 1 pM ryanodine, an inhibitor of SR calcium release. Ryanodine had no effect on phenylephrine-stimulated ANP secretion. Atria were superfused with 10 pM nitrendipine to determine whether calcium influx through voltage-dependent calcium channels was a mechanism of calcium entry for stimulation. Nitrendipine inhibited phenylephrine-stimulated ANP secretion by 49% without interfering with n,-adrenergic antagonist receptor binding. This finding was supported by the observation that phenylephrine-stimulated secretion was 52% lower in nonbeating atria. (YeAdrenergic agonists have been reported to enhance Na-H anti-
T
HE ai-ADRENERGIC agonists increase atria1 natriuretic peptide (ANP) secretion in vitro (l-3). The calcium dependency of this response has not been clearly defined. In this study we have examined the role of calcium in cq-agonist-stimulated ANP secretion. Stimulation with al-adrenergic agonists increases calcium influx (4). This results from a prolongation of the action potential, which delays inactivation of the calcium current, leading to a rise in calcium influx (4-6). Thus, calcium influx through voltage-dependent calcium channels is enhanced by al-adrenergic agonists. ai-Adrenergic agonists may also increase calcium release from the sarcoplasmic reticulum (SR). Stimulation with ai-adrenergic agonists increases inositol-1,4,&trisphosphate (7),
which has been shown to release calcium or potentiate the release of calcium from the SR (8-10). Thus, calcium influx and/or calcium release from the SR may play a participatory role in al-agonist-stimulated ANP secretion. In this study we examined the dependency of cylagonist-stimulated ANP secretion on calcium influx and calcium release from the SR. We also investigated the possibility that the calcium dependency of the initiation of the ANP secretory response to q-adrenergic agonists may differ from that necessary to maintain the secretory response.
Received August 15, 1991. Address all correspondence and requests for reprints to: Rick J. Schiebinger, M.D., Division of Endocrinology and Hypertension, Wayne State University School of Medicine, University Health Center 4H, 4201 St. Antoine, Detroit, Michigan 48201. * This work was supported by a Student Research Fellowship from the American Heart Association of Michigan (to H.G.P.), NIH Grant ROl-HL-42209, and V.A. Medical Research Funds.
Materials ylephrine
Materials
and Methods
Materials were purchased and propranolol
from from
the following Sigma (St.
sources: phenLouis, MO); rat
aANP and ANP antibodies from Peninsula Laboratories (Belmont, CA); ryanodine from Progressive Agri-Systems (Wind Gap, PA); medium 199 from GIBCO (Grand Island, NY); and [furanyl-5-3H]prazosin from Heights, IL). 5-(N,N-Hexamethylene)amiloride
Amersham
(Arlington (HMA) was
1017
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 05:48 For personal use only. No other uses without permission. . All rights reserved.
CALCIUM
1018
AND
al-STIMULATED
synthesized specifically for this study as previously described (11). Nitrendipine was a gift from A. Scriabine, Miles Institute of Preclinical Pharmacology (New Haven, CT). Phentolamine was a gift from Ciba-Geigy (Suffern, NY), and prazosin was a gift from Pfizer (New York, NY). Experimental
design
Atria1 superfusion experiments were performed, as previously described,using isolated rat left atria superfusedwith modified medium 199 containing Earle’s salts with 20 mM NaHCO:$and 4 mM KC1 (12). Low superfusatecalcium experiments were performed by lowering the calcium concentration from 1.8to 0.2 mM 20 min before the addition of phenylephrine. This was achieved by superfusion with modified medium 199 containing 0.2 mM CaC12.Nitrendipine and ryanodine were addedto the superfusate55 min before beginning samplecollection and continued until the termination of the experiment. Both were addedin ethanol, with final concentrations of 0.1% or 0.05%, respectively. HMA was added in dimethylsulfoxide (DMSO; final concentration, 0.1%) and was added to the superfusate 10 min before the addition of phenylephrine. All control atria were superfusedwith the sameethanol or DMSO concentrations as the atria receiving the inhibitor. All experimentswereperformed with 1 pM propranolol, sincephenylephrine hasa minimal amount of /3-adrenergicagonist activity. We have previously demonstrated that 1 FM propranolol and 10 pM phentolamine negetethe ANP secretory responseto 10 j.~cM phenylephrine (3). Tension measurementswere obtained from a strip chart recording of atria1 performance using a Gould sixchannel rectilinear oscillographicrecorder.
ANP SECRETION
Endo. Vol130.
1992 No 2
a lowered buffer calcium concentration. Atria were superfused with 0.2 or 1.8 mM calcium. Atria superfused with 0.2 mM calcium failed to mount a secretory response to 10 PM phenylephrine (Fig. 11, suggesting that calcium influx is necessary to initiate the secretory response to CQ-adrenergic agonists. Developed tension also failed to rise in response to phenylephrine in the presence of 0.2 mM calcium, whereas developed tension rose 1.7-fold with 1.8 mM calcium (Table 1). Blunting calcium influx by superfusion with 0.2 mM calcium results in a commensurate lowering of calcium release from the SR. To distinguish the importance of calcium influx us. intracellular calcium release, atria were superfused with ryanodine, an inhibitor of SR calcium release (14-16). Superfusion with 1 PM ryanodine did not block the ANP secretory response to phenylephrine (Fig. 2). The biological activity of ryanodine was demonstrated by a fall in developed tension to an undetectable level. No detectable rise in developed tension occurred with the addition of phenylephrine (Table 1). These results suggest that SR calcium release does not play an important role in the secretory response. To determine whether calcium influx was occurring through voltage-dependent calcium channels, atria were superfused with the calcium channel blocker nitrendi-
ANP RIA
ANP measurementswere performed on timed fractions of the superfusateby RIA, as previously described(13). Binding
studies
Three a,-adrenergic ligand binding studies were performed to determinewhether nitrendipine interfered with binding. Rat atria were homogenizedin 250 mM sucrose,1 mM EDTA, and 50 mM Tris, pH 7.4. A crude membranefraction was obtained by differential centrifugation, usingthe pellet betweenthe 1,500 x g lo-min spin and the 200,000x g 30-min spin. Membrane (290 pg) was incubated with 1 nM [“Hlprazosin (22 Ci/mmol) in 120 mM NaCl, 1 mM EDTA, and 50 mM Tris, pH 7.4, in a 0.4-ml volume for 1 h at room temperature with or without nitrendipine. The nitrendipine concentrations tested were 5 or 50 pM. Nonspecific binding was determined by incubation with 1 FM prazosin. Bound and free ligand were separatedby filtration through 13-mmglassfiber filters. Data analysis
Statistical analyseswere performed by paired or unpaired t test. Resultsare expressedasthe mean + SE. Results The dependency of ol,-agonist-stimulated ANP secretion on calcium influx was examined by superfusion with
200 5 i= Id 5 LIJ
150
v, CL z 6 100
501
' 0
I 5
t 10
I 15
TIME
I 20
1 25
I 30
I 35
I 40
I 45
(min)
FIG. 1. Effect of a lowered extracellular calcium concentration on phenylephrine-stimulated ANP secretion. Rat left atria paced at 2 Hz were superfused with 0.2 mM (0, n = 9; A, n = 6) or 1.8 mM calcium (0; n = 7). Atria were continuously superfused with 10 pM phenylephrine (0 and 0) from 15-45 min or superfused without phenylephrine (A). Results are expressed as a percentage of the mean of seven baseline measurements obtained between O-15 min. Basal ANP secretion was 269 k 35 (A), 294 + 44 (0), and 324 f 42 pg/ml (0).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 05:48 For personal use only. No other uses without permission. . All rights reserved.
CALCIUM TABLE 1. Effect
of calcium-lowering Phenylephrine (10 PM)
Group Control
Calcium
(0.2
mM)
Control
Ryanodine
(1
GM)
(10
on atria1
pM)
“P < 0.01 compared * No change.
aI-STIMULATED
function Resting tension (g)
Before After
0.10 * 0.01
0.18 f 0.02"
0.34 * 0.02 NC*
Before After
0.02 + 0.01 0.01 * 0.01
0.29 + 0.02 NC
Before After
0.10 + 0.01 0.17 + 0.01”
0.32 + 0.02 NC
0 0
SECRETION
1019
1
0.32 k 0.02 NC
Before After
0.12 f 0.02 0.20 k 0.03”
0.46 rk 0.04 NC
Before After
0.06 + 0.01 0.06 + 0.01
0.30 + 0.03 NC
to before
ANP 250
Developed tension (g)
Before After
Control
Nitrendipine
agents
AND
11 0
measurements.
’ 5
’ 10
’ 15
’ 20
’ 25
TIME
’ 30
’ 35
’ 40
’ 45
’ 50
’ 55
J 60
(min)
FIG. 3. Effects
6
of nitrendipine on phenylephrine-stimulated ANP secretion. Rat left atria paced at 2 Hz were superfused with 10 pM nitrendipine (0, n = 6; A, n = 6) or vehicle alone (e, n = 5). Atria were continuously superfused with 10 pM phenylephrine (0 and 0) from 1560 min or superfused without phenylephrine (A). Results are expressed as described in Fig. 1. Basal ANP secretion was 253 k 41 (A), 244 + 35 (0), and 246 + 35 pg/ml (0).
200
ifi t!s
150
w U-J LL 5
100
50 1
I
I
I
0
5
10
15
I
I
I
I
I
20
25
30
35
40
TIME
11
45
11
50
55
60
(min)
FIG. 2. Effect of ryanodine on phenylephrine-stimulated ANP secretion. Rat left atria paced at 2 Hz were superfused with 1 @M ryanodine (0, n = 8; A, n = 6) or vehicle alone (e, n = 8). Atria were continuously superfused with 10 FM phenylephrine (0 and 0) from 15-60 min or superfused without phenylephrine (A). Results are expressed as described in Fig. 1. Basal ANP secretion was 163 + 27 (A), 221 f 21 (0), and 256 f 25 pg/ml (0).
pine. Superfusion with 10 PM nitrendipine phenylephrine-stimulated secretion by 49% veloped tension was lower in atria superfused nitrendipine, and it did not rise with the phenylephrine (Table 1). Nitrendipine did [3H]prazosin binding to atria1 membranes.
inhibited (Fig. 3). Dewith 10 PM addition of not inhibit Binding in
the presence of 5 or 50 PM nitrendipine was 96 k 6% of the control value (n = 3 experiments). Thus, phenylephrine-stimulated ANP secretion is partially dependent on calcium influx through nitrendipine-inhibitable calcium channels. To corroborate the findings with nitrendipine, the effect of phenylephrine on nonbeating atria was examined. In nonbeating left atria, calcium channels are quiescent, due to the absence of membrane depolarization. Thus, a lower ANP secretory response to phenylephrine should occur in nonbeating atria us. paced atria. Indeed, the response to phenylephrine was 52% lower in nonbeating atria (Fig. 4), similar to the lower response of atria superfused with nitrendipine (Fig. 3). Thus, it appears that half of the calcium influx is derived from calcium entry through voltage-dependent calcium channels. It has recently been reported that cui-adrenergic stimulation activates the Na-H antiporter, resulting in a rise in intracellular calcium (17). The increase in intracellular calcium was abolished in the absence of extracellular sodium or calcium and by the Na-H antiporter inhibitor HMA (18), suggesting that the rise in intracellular sodium from enhanced Na-H exchange may increase calcium influx through Na-Ca exchange. To determine whether calcium influx via Na-Ca exchange was occur-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 05:48 For personal use only. No other uses without permission. . All rights reserved.
CALCIUM
1020
AND
al-STIMULATED
ANP
SECRETION
Endo. Voll30.
1992 No 2
200
250 oz
150
cl i!5 ti
100
2 50
0
5
10
15
20
25
TIME
30
35
40
45
50
55
60
(min)
FIG. 4. Effect of phenylephrine on paced and nonbeating rat left atria. Atria were continuously superfused with 10 pM phenylephrine from 15-60 min and were either paced at 2 Hz (0; n = 6) or were nonbeating (0, n = 6). Half of the control atria (A; n = 6), which were not superfused with phenylephrine, were paced at 2 Hz, and half were nonbeating. Results were expressed as described in Fig. 1. Basal ANP secretion was 195 + 37 (A), 122 f 14 (0), and 202 f 31 pg/ml (0).
ring as the result of an increase in Na-H exchange and intracellular sodium, we examined the effect of HMA on phenylephrine-stimulated ANP secretion by nonbeating left atria. Using this experimental paradigm, we could determine whether calcium channel-independent calcium influx was occurring by this mechanism. HMA (25 pM) did not inhibit phenylephrine-stimulated ANP secretion (Fig. 5). Thus, enhanced calcium influx via NaCa exchange resulting from a rise in intracellular sodium due to increased Na-H antiporter activity does not appear to be involved in phenylephrine-stimulated ANP secretion. As illustrated in Fig. 1, calcium influx is necessary for initiating the ANP secretory response to phenylephrine. We designed an experiment to determine the dependency on calcium influx of the maintained secretory response to phenylephrine. Atria were superfused in 1.8 mM calcium and stimulated with phenylephrine for 45 min. ANP secretion rose 1.9-fold in response to 10 PM phenylephrine. Atria were either maintained in 1.8 mM calcium or switched to medium 199 containing 0.2 mM calcium during continuous superfusion with 10 PM phenylephrine. Lowering the buffer calcium concentration did not affect the continued secretory response to phenylephrine (Fig. 6). Developed tension fell from a maximum of 0.13 f 0.02 to 0.01 + 0.00 g in atria superfused with
1
1
t
I
I
I
1
I
I
I
I
0
5
10
15
20
25
30
35
40
45
TIME (min) FIG. 5. Effect of HMA on phenylephrine-stimulated ANP secretion. Nonbeating rat left atria were superfused with 25 pM HMA from 5-45 min (0, n = 6; 0, n = 6) or vehicle alone (A; n = 6). Atria were continuously superfused with 10 pM phenylephrine (0 and A) from 1545 min or superfused without phenylephrine (0). Results are expressed as described in Fig. 1. Upper panel, The graph of phenylephrine plus HMA (0) is presented after correction for the rise in ANP secretion by superfusion with HMA alone. Lowel panel, Graphs illustrate the ANP secretary response to HMA alone (0) and to phenylephrine plus HMA (0) without corrections. Basal ANP secretion was 110 f 12 (0), 125 f 28 (O), and 123 + 14 pg/ml (A).
0.2 mM calcium compared to only a slight fall in control atria from 0.14 f 0.02 to 0.11 f 0.01 g over the same time period. In contrast, when the a-adrenergic antagonist phentolamine (10 PM) was added to the superfusate, phenylephrine-stimulated ANP secretion fell to near-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 05:48 For personal use only. No other uses without permission. . All rights reserved.
CALCIUM
0
5
10
15
AND
20
al-STIMULATED
25
JO
TIME (min) FIG. 6. Effect of lowering extracellular calcium from 1.8 to 0.2 mM on phenylephrine-stimulated ANP secretion. Rat left atria paced at 2 Hz were superfused with 10 pM phenylephrine for 45 min before time zero and for 30 min thereafter. At time zero, the superfusate calcium concentration was lowered to 0.2 calcium (0, n = 10) or remained in 1.8 mM calcium (e n = 9). As a positive control, 10 FM phentolamine was added to the superfusate at time zero for those atria superfused with 1.8 mM calcium (A; n = 6). Results are expressed as a percentage of the net increase in ANP secretion, using the mean of two measurements obtained at -2.5 and 0 min as 100%.
basal levels (Fig. 6), and developed tension fell to baseline. Initially, developed tension was 0.17 f 0.01 g, which rose to 0.29 + 0.05 g with phenylephrine and then fell to 0.14 f 0.02 g with the addition of phentolamine. Thus, once the secretory response to phenylephrine is established, calcium influx appears to play a markedly reduced role in the maintenance of the response over 30 min. Discussion
Atria1 contraction results from a rise in cytosolic calcium. The sources of calcium responsible for this rise are both extracellular and intracellular. With membrane depolarization, calcium influx occurs primarily through voltage-dependent calcium channels and to a limited degree via Na-Ca exchange (19). The rise in cytosolic calcium, resulting from calcium influx releases calcium from the SR (20). Atria1 contraction then occurs as a result of these events. Lowering the extracellular calcium concentration in the present study totally blocked the initiation of phenylephrine-stimulated ANP secretion. This may be the result of decreased calcium influx and/ or lack of calcium release from the SR. The fact that ryanodine did not inhibit phenylephrine-stimulated se-
ANP
SECRETION
1021
cretion suggests that the source of calcium necessary for the initiation of the response is extracellular and not intracellular in origin. Calcium influx primarily occurs by the opening of voltage-dependent calcium channels and Na-Ca exchange. Superfusion with 10 PM nitrendipine inhibited phenylephrine-stimulated ANP secretion by 49%. The dose of 10 PM nitrendipine has been previously shown by us to totally negate isoproterenoland (Bu)~AMPstimulated ANP secretion (21). Inhibition was not due to nitrendipine interfering with al-adrenergic binding sites. Thus, we conclude that half of the calcium necessary for initiating phenylephrine-stimulated secretion is derived from calcium influx through L-type calcium channels. In nonbeating atria, where calcium channels are relatively quiescent, phenylephrine-stimulated ANP secretion was 52% lower than in paced atria. This corroborates the nitrendipine studies and also suggests that calcium influx through T-type calcium channels is probably of negligible importance, because there was no difference in the lowering of phenylephrine-stimulated secretion by nitrendipine and nonbeating atria. Thus, these findings are consistent with the observation that CQadrenergic agonists increase calcium influx through voltage-dependent calcium channels by prolonging the action potential (4-6). In contrast, our study does not support the speculation that a,-adrenergic agonists increase SR calcium release. As mentioned in the introduction, several investigators have found that inositol-1,4,5-trisphosphate either releases or enhances the release of calcium from the SR (8-10); however, there is not uniform agreement on this point (22). If it were true that inositol-1,4,5-trisphosphate enhances SR calcium release, ryanodine should inhibit phenylephrine-stimulated secretion to some degree, and it did not. Ryanodine was biologically active in this study, as demonstrated by a fall in developed tension to undetectable levels. The dose of 1 ~.LM ryanodine has been previously shown by us to totally negate isoproterenol- and (Bu)2AMP-stimulated ANP secretion (21). This latter observation is consistent with the known effect of protein kinase A on the SR to increase the pool of stored calcium by phosphorylation of phospholamban (23). Therefore, the present study does not support the premise that inositol-1,4,5-trisphosphate generated by phenylephrine stimulation significantly enhances SR calcium release due to the failure of ryanodine to inhibit phenylephrine-stimulated secretion. We conclude that calcium influx, not calcium release from the SR, is the primary source of calcium necessary for initiating phenylephrine-stimulated secretion. Our results differ from those of Sei and Glembotski (24), who found that lowering the buffer calcium concentration to 2 /IM did not inhibit phenylephrine-stimulated
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 05:48 For personal use only. No other uses without permission. . All rights reserved.
1022
CALCIUM
AND
(u~STIMULATED
secretion by cultured rat neonatal atria1 cardiocytes. When the extracellular calcium level was lowered further to 2 nM with EGTA, they observed a 40% lower phenylephrine-stimulated response. In contrast, results similar to ours were found with the calcium channel inhibitor nifedipine. Phenylephrine-stimulated secretion was inhibited by a little greater than 50% with nifedipine. Thus, their results suggest that phenylephrine-stimulated secretion is partially independent of calcium influx, whereas our results indicate a complete dependency on calcium influx. However, we both agree that calcium influx through L-type calcium channels plays a significant role in phenylephrine-stimulated secretion. The differences in these findings may reflect differences in calcium handling by neonatal cells us. adult atria1 cells or differences between cultured cells and intact atria1 tissue. Calcium influx by a calcium channel-independent mechanism appears to account for 50% of the calcium influx. Residual calcium influx is most likely achieved by Na-Ca exchange. Na-Ca exchange activity may be altered by increasing intracellular sodium secondary to enhanced Na-H antiporter activity (17). A rise in intracellular sodium increases cytosolic calcium by either increasing calcium influx or decreasing its efflux by NaCa exchange. We examined this possibility by using HMA to inhibit the Na-H antiporter. HMA did not inhibit phenylephrine-stimulated secretion, suggesting another mechanism by which Na-Ca exchange is altered. It is unlikely that higher concentrations of HMA would have demonstrated an inhibitory effect. We have previously shown that 25 j.~tM HMA totally negates frequencystimulated ANP secretion (25). Secondly, the Ki for HMA inhibition of Na-H antiporter activity is estimated to be 0.2 PM (18). Thus, we chose a concentration that was more than loo-fold higher than the Ki. Na-Ca exchange activity may be increased by cr,-adrenergic agonists by activation of phospholipase C (26). Also, phosphatidic acid, which may be a product of phospholipase C by conversion of diacylglycerol to phosphatidic acid, has been shown to increase Na-Ca exchanges (26). However, from these studies it is not clear in which direction calcium is being moved. The orientation of the exchanger is not known in these experiments, since they were performed with sarcolemmal vesicles, which may be inside out or right side out. In addition, the exchanger can move calcium either into or out of the cells depending on the relative concentrations of sodium and calcium across the membrane. Thus, activation of phospholipase C by phenylephrine may increase calcium influx through Na-Ca exchange. This may be the explanation for calcium channel-independent calcium influx. In contrast to the marked dependency on calcium influx to initiate the ANP secretory response to phenyl-
ANP SECRETION
Endo.
Vol130.No
1992
2
ephrine, calcium influx appears to play a minimal role in the maintenance of an established response, as suggested by the lack of an inhibitory effect of lowering the buffer calcium concentration over 30 min. During this time, developed tension fell to almost undetectable levels after prior stimulation with phenylephrine. With the addition of phentolamine, phenylephrine-stimulated secretion fell to near baseline, demonstrating that removal of the stimulatory signal does indeed induce a fall in phenylephrine-stimulated ANP secretion. In addition, we have previously demonstrated that lowering the buffer calcium Concentration to 0.2 mM lowers isoproterenolstimulated ANP secretion to baseline (21). These results suggest a stimulatory pathway whose initiation is dependent on calcium influx; however, once the signal is set in motion, calcium influx is no longer necessary at least for the short term. The mechanism of this observation may be due to activation of multifunctional Ca2+/calmodulin-dependent protein kinase (CaM kinase). CaM kinase undergoes autophosphorylation after activation by calcium and calmodulin. The result of autophosphorylation is the release of the kinase from its dependence on calcium and calmodulin for activity. This autonomous kinase is then capable of phosphorylating exogenous substrates in the absence of calcium and calmodulin (27). Thus, autophosphorylation of CaM kinase renders it active in spite of a fall in cytoplasmic calcium. Activated CaM kinase may explain the persistence of phenylephrine-stimulated ANP secretion in the presence of low extracellular calcium concentrations. Basal ANP secretion varied at times between groups within experiments or between studies. This may be due to paced vs. nonpaced atria or the addition of organic solvents, such as DMSO or ethanol. It may also reflect differences in atria1 size or shape of the atria, which determines the amount of atria1 tissue stretched in a single plane. Atria were randomly selected for all study groups. The calcium dependency of ai-adrenergic agoniststimulated ANP secretion is unique from that of /3adrenergic or endothelin-stimulated secretion (Table 2). As we have previously reported, P-agonist-stimulated secretion is totally dependent on calcium influx and is TABLE
agonist-,
2. Comparison of the dependency and endothelin-stimulated ANP
on calcium secretion
by P-agonist-,
Stimulus Exp
variable Isoproterenol
0.2 mM calcium 2 mM lanthanum 10 pM nitrendipine 1 pM ryanodine Values
shown
are percent
Phenylephrine
100
100
100 100
50 0
inhibition.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 05:48 For personal use only. No other uses without permission. . All rights reserved.
Endothelin 50 100 50 0
crl-
CALCIUM
AND
,X,-STIMULATED
negated by 10 PM nitrendipine or 1 PM ryanodine (21). These results suggest that the primary source of cytosolic calcium responsible for stimulation is from the expanded calcium storage pool in the SR. In contrast, endothelinstimulated secretion is only partially inhibited by 0.2 mM calcium (12). However, superfusion with 2 InM lanthanum, an inhibitor of calcium transport, in the presence of 1 mM calcium completely inhibited endothelinstimulated ANP secretion (our personal observation). A similar observation was made with ouabain-stimulated ANP secretion. Again, superfusion with 0.2 mM calcium only partially inhibited stimulated secretion, whereas 2 mM lanthanum completely negated ouabain-stimulated secretion (our personal observation). Thus, when calcium influx is aggressively inhibited, endothelin stimulation also appears to be totally dependent on calcium influx. Calcium influx is partially achieved through L-type calcium channels, similar to that in response to al-adrenergic stimulation. Ryanodine had no effect on endothelinstimulated secretion. These results suggest that calcium influx, but not SR calcium release, is required for endothelin-stimulated secretion. Thus, the calcium dependency of phenylephrine-stimulated secretion is similar, but not identical, to that of endothelin (Table 2). This is consistent with the observation that both endothelin and al-adrenergic agonists activate the phosphoinositide pathway and do not activate adenylyl cyclase. However, endothelin is more resistant to inhibition by superfusion with 0.2 InM calcium. This may be due to stimulation by endothelin of phospholipase AZ, which may give it unique stimulating properties (28). Thus, these three secretogogues of ANP release display different dependencies on calcium, which may reflect differences in the second messenger systems evoked by each. Acknowledgments The excellent technical support by Jennifer Harmon Terri Carswell for
is greatly typing the
appreciated. manuscript.
ANP 6. Miura
on the
also
Sullivan thank
and Ms.
8.
9.
10.
11.
12.
13.
14. 15.
16. 17.
18.
19. 20.
22.
23.
References 1. Sonnerberg H, Veress AT 1984 Cellular mechanism of release of atria1 natriuretic factor. Biochem Biophys Res Cummun 124:443449 Currie MG, Newman WH 1986 Evidence for a-1 adrenergic receptor regulation of atriopeptin release from the isolated rat heart. Biochem Biophvs Res Commun 137:94-100 Schiebinger I&J,-Baker MZ, Linden J 1987 Effect of adrenergic and muscarinic cholinergic agonists on atria1 natriuretic neptide secretion by isolated rat atria-J Clin Invest 80:1687-1691’ Fedida D. Shimoni Y. Giles WR 1990 a-Adrenereic modulation of the transient outward current in rabbit atria1 myocytes. J Physiol
24.
25
26.
27.
stimulation Pharmacol
Otani H, Das DK 1988 a,-Adrenoceptor-mediated phosphoinositide breakdown and inotropic response in rat left ventricular papillary muscles. Circ Res 62:8-17 Hirata M, Suematsu E, Hashimoto T, Hamachi T, Koga T 1984 Release of Ca’+ from a non-mitochondrial store site in peritoneal macrophages treated with saponin by inositol 1,4,5-trisphosphate. Biochem J 223:229-236 Nosek TM, Williams MF, Zeigler ST, Godt RE 1986 lnositol trisphosphate enhances calcium release in skinned cardiac and skeletal muscle. Am J Physiol 25O:C807-C811 Fabiato A 1989 Comparison and relation between inositol (1,4,5)trisphosphate-induced release and calcium-induced release of calcium from the cardiac sarcoplasmic reticulum. In: Yamanada K, Shibata S (eds) Recent Advances in Ca’+-Channels and Ca2+Antagonists. Proceedings of the Japan-U.S. Symposium on Cardiovascular Drugs. Pergamon Press, Elmsford, pp 35-39 Cragoe Jr EJ, Woltersdorf OW, Bicking JB, Kwong SF, Jones JH 1967 Pyrazine diuretics. Il. N-Amidino+amino-5-substituted-6halopyrazine-carboxamides. J Med Chem 10:66-75 Schiebinger RJ, Gomez-Sanchez CE 1990 Endothelin: a notent stimulus-of atria1 natriuretic peptide secretion by superfused rat atria and its dependency on calcium. Endocrinology 127:119-125 Schiebinger RJ, Linden J 1986 The influence of resting tension on immunoreactive atria1 natriuretic peptide secretion by rat atria superfused in uitro. Circ Res 59:105-109 Fabiato A 1985 Effects of ryanodine in skinned cardiac cells. Fed Proc 44:2970-2976 Wier WG, Yue DT, Marban E 1985 Effects of ryanodine on intracellular Ca”’ transients in mammalian cardiac muscle. Fed Proc 44:2989-2993 Rousseau E, Smith JS, Meissner G 1987 Ryanodine modifies conductance and gating behavior of single Ca*+ release channel. Am J Physiol253:C364-C368 lwakura K, Hori M, Watanabe Y, Kitabatake A. Craeoe Jr EJ. Yoshida H, Kamada T 1990 al-Adrenocepter stimulation increases intracellular PH and Ca2+ in cardiomvocvtes through Na+/H+ and Na+/Ca’+ exchange. Eur J Pharmacol186:29-40 ’ Simchowitz L, Cragoe EJ 1986 Inhibition of chemotactic factoractivated Na+/H+ exchange in human neutrophils by analogues of amiloride: structure-activity relationships in the amiloride series. Mol Pharmacol30:112-120 Leblanc N, Hume JR 1990 Sodium current-induced release of calcium from cardiac sarcoplasma reticulum. Science 248:372-376 Nabauer M, Callewaert G, Cleemann L, Morad M 1989 Regulation of calcium release is gated by calcium current, not gating-charge, in cardiac myocytes. Science 244:800-803 Schiebinger RJ 1989 Calcium, its role in isoproterenol-stimulated atria1 natriuretic peptide secretion by superfused rat atria. Circ Res 65:600-606 Movsesian MA, Thomas AP, Selak M, Williamson JR 1985 lnositol trisphosphate does not release Ca’+ from permeabilized cardiac myocytes and sarcoplasmic reticulum. FEBS Lett 185:328-332 Tsien RW 1977 Cyclic AMP and contractile activity in heart. Adv Cyclic Nucleotide Res 8:363-420 Sei CA, Glembotski CC 1990 Calcium dependence of phenylephrine-, endothelin-, and potassium chloride-stimulated atria1 natriuretic factor secretion from long term primary neonatal rat atria1 cardiocytes. J Biol Chem 265:7166-7172 Schiebinger RJ 1988 Mechanism of enhanced atria1 natriuretic peptide (ANP) secretion by increased atria1 contraction frequency. Hypertension 12:354 (Abstract) Philipson KD, Frank JS, Nishimoto AY 1983 Effects of phospholipase C on the Na+-Ca*+ exchange and Ca’+ permeability of cardiac sarcolemmal vesicles. J Biol Chem 258:5905-5910 Schulman H, 1988 The multifunctional Ca’+/calmodulin-dependent protein kinase. Adv Second Messenger Phosphoprotein Res
22:X-l
4231257-277
Dukes ID, Vaughan Williams EM 1984 Effects of selective /&-, P2-, (Y,- and al-adrenoceptor stimulation on potentials and contractions in the rabbit heart. J Physiol355:523-546
Y, lnui J 1984 Multiple effects of n-adrenocepter action potential of the rabbit atrium. Arch
1023
325:47-53 7. Otani H,
21.
William We
SECRETION
28.
12
Resnick TJ, Scott-Burden T, Buhler FR 1989 Activation pholipase A, by endothelin in cultured vascular smooth cells. Biochem Biophys Res Commun 158:279-286
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 05:48 For personal use only. No other uses without permission. . All rights reserved.
of phosmuscle
Vol. 130, No. 2 Endocrine
Society
Printed
Calcium: Its Role in al-Adrenergic Natriuretic Peptide Secretion* RICK
J. SCHIEBINGER,
HARRY
G. PARR,
AND
EDWARD
Stimulation J. CRAGOE,
in U.S.A.
of Atria1
JR.
Wayne State University, Detroit, Michigan 48201; the Veterans Administration Medical Center, Allen Park, Michigan 48101; and P.O. Box 631548 (E.J.C.), Nacogdoches, Texas 75963
porter activity. To determine whether the resulting rise in intracellular sodium may alter Na-Ca exchange to raise intracellular calcium levels, atria were superfused with the Na-H antiporter inhibitor, 5-(N,N-hexamethylene)amiloride. Superfusion with 25 FM 5-(N,N-hexamethylene)amiloride did not inhibit phenylephrine-stimulated ANP secretion. Lastly, the calcium dependency of the maintenance of an established response to phenylephrine was examined. Atria were superfused with phenylephrine in buffer containing 1.8 mM calcium for 45 min, followed by superfusion with phenylephrine in 0.2 mM calcium for 30 min. There was no fall in phenylephrine-stimulated secretion by atria superfused in 0.2 mM calcium. In contrast, addition of the ol,-adrenergic antagonist phentolamine induced an immediate fall in phenylephrine-stimulated ANP secretion. We conclude that 1) calcium influx is necessary to initiate ai-agonist-stimulated ANP secretion; 2) calcium release from the SR does not play a role in a,-agonist-stimulated secretion; 3) calcium entry through L-type calcium channels is responsible for half of the calcium influx; 4) enhanced Na-H antiporter activity does not play a role in a,-agonist-stimulated secretion; and 5) maintenance of ocl-agonist-stimulated secretion is not dependent on calcium influx. (Endocrinology 130: 1017-1023, 1992)
ABSTRACT.
ai-Adrenergic agonists increase atria1 natriuretic peptide (ANP) secretion. The mechanism of ol,-adrenergic-stimulated secretion is not known. In this study we examine the calcium dependency of Lu,-agonist-stimulated ANP secretion. Isolated superfused rat left atria paced at 2 Hz were used for study. Superfusion with 10 pM phenylephrine increased ANP secretion by P-fold. Lowering the superfusate calcium concentration from 1.8 to 0.2 mM totally negated the secretory response to phenylephrine. To determine whether this reflected a reduction in calcium influx, reduced calcium release from the sarcoplasmic reticulum (SR), or both, atria were superfused with 1 pM ryanodine, an inhibitor of SR calcium release. Ryanodine had no effect on phenylephrine-stimulated ANP secretion. Atria were superfused with 10 pM nitrendipine to determine whether calcium influx through voltage-dependent calcium channels was a mechanism of calcium entry for stimulation. Nitrendipine inhibited phenylephrine-stimulated ANP secretion by 49% without interfering with n,-adrenergic antagonist receptor binding. This finding was supported by the observation that phenylephrine-stimulated secretion was 52% lower in nonbeating atria. (YeAdrenergic agonists have been reported to enhance Na-H anti-
T
HE ai-ADRENERGIC agonists increase atria1 natriuretic peptide (ANP) secretion in vitro (l-3). The calcium dependency of this response has not been clearly defined. In this study we have examined the role of calcium in cq-agonist-stimulated ANP secretion. Stimulation with al-adrenergic agonists increases calcium influx (4). This results from a prolongation of the action potential, which delays inactivation of the calcium current, leading to a rise in calcium influx (4-6). Thus, calcium influx through voltage-dependent calcium channels is enhanced by al-adrenergic agonists. ai-Adrenergic agonists may also increase calcium release from the sarcoplasmic reticulum (SR). Stimulation with ai-adrenergic agonists increases inositol-1,4,&trisphosphate (7),
which has been shown to release calcium or potentiate the release of calcium from the SR (8-10). Thus, calcium influx and/or calcium release from the SR may play a participatory role in al-agonist-stimulated ANP secretion. In this study we examined the dependency of cylagonist-stimulated ANP secretion on calcium influx and calcium release from the SR. We also investigated the possibility that the calcium dependency of the initiation of the ANP secretory response to q-adrenergic agonists may differ from that necessary to maintain the secretory response.
Received August 15, 1991. Address all correspondence and requests for reprints to: Rick J. Schiebinger, M.D., Division of Endocrinology and Hypertension, Wayne State University School of Medicine, University Health Center 4H, 4201 St. Antoine, Detroit, Michigan 48201. * This work was supported by a Student Research Fellowship from the American Heart Association of Michigan (to H.G.P.), NIH Grant ROl-HL-42209, and V.A. Medical Research Funds.
Materials ylephrine
Materials
and Methods
Materials were purchased and propranolol
from from
the following Sigma (St.
sources: phenLouis, MO); rat
aANP and ANP antibodies from Peninsula Laboratories (Belmont, CA); ryanodine from Progressive Agri-Systems (Wind Gap, PA); medium 199 from GIBCO (Grand Island, NY); and [furanyl-5-3H]prazosin from Heights, IL). 5-(N,N-Hexamethylene)amiloride
Amersham
(Arlington (HMA) was
1017
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 05:48 For personal use only. No other uses without permission. . All rights reserved.
CALCIUM
1018
AND
al-STIMULATED
synthesized specifically for this study as previously described (11). Nitrendipine was a gift from A. Scriabine, Miles Institute of Preclinical Pharmacology (New Haven, CT). Phentolamine was a gift from Ciba-Geigy (Suffern, NY), and prazosin was a gift from Pfizer (New York, NY). Experimental
design
Atria1 superfusion experiments were performed, as previously described,using isolated rat left atria superfusedwith modified medium 199 containing Earle’s salts with 20 mM NaHCO:$and 4 mM KC1 (12). Low superfusatecalcium experiments were performed by lowering the calcium concentration from 1.8to 0.2 mM 20 min before the addition of phenylephrine. This was achieved by superfusion with modified medium 199 containing 0.2 mM CaC12.Nitrendipine and ryanodine were addedto the superfusate55 min before beginning samplecollection and continued until the termination of the experiment. Both were addedin ethanol, with final concentrations of 0.1% or 0.05%, respectively. HMA was added in dimethylsulfoxide (DMSO; final concentration, 0.1%) and was added to the superfusate 10 min before the addition of phenylephrine. All control atria were superfusedwith the sameethanol or DMSO concentrations as the atria receiving the inhibitor. All experimentswereperformed with 1 pM propranolol, sincephenylephrine hasa minimal amount of /3-adrenergicagonist activity. We have previously demonstrated that 1 FM propranolol and 10 pM phentolamine negetethe ANP secretory responseto 10 j.~cM phenylephrine (3). Tension measurementswere obtained from a strip chart recording of atria1 performance using a Gould sixchannel rectilinear oscillographicrecorder.
ANP SECRETION
Endo. Vol130.
1992 No 2
a lowered buffer calcium concentration. Atria were superfused with 0.2 or 1.8 mM calcium. Atria superfused with 0.2 mM calcium failed to mount a secretory response to 10 PM phenylephrine (Fig. 11, suggesting that calcium influx is necessary to initiate the secretory response to CQ-adrenergic agonists. Developed tension also failed to rise in response to phenylephrine in the presence of 0.2 mM calcium, whereas developed tension rose 1.7-fold with 1.8 mM calcium (Table 1). Blunting calcium influx by superfusion with 0.2 mM calcium results in a commensurate lowering of calcium release from the SR. To distinguish the importance of calcium influx us. intracellular calcium release, atria were superfused with ryanodine, an inhibitor of SR calcium release (14-16). Superfusion with 1 PM ryanodine did not block the ANP secretory response to phenylephrine (Fig. 2). The biological activity of ryanodine was demonstrated by a fall in developed tension to an undetectable level. No detectable rise in developed tension occurred with the addition of phenylephrine (Table 1). These results suggest that SR calcium release does not play an important role in the secretory response. To determine whether calcium influx was occurring through voltage-dependent calcium channels, atria were superfused with the calcium channel blocker nitrendi-
ANP RIA
ANP measurementswere performed on timed fractions of the superfusateby RIA, as previously described(13). Binding
studies
Three a,-adrenergic ligand binding studies were performed to determinewhether nitrendipine interfered with binding. Rat atria were homogenizedin 250 mM sucrose,1 mM EDTA, and 50 mM Tris, pH 7.4. A crude membranefraction was obtained by differential centrifugation, usingthe pellet betweenthe 1,500 x g lo-min spin and the 200,000x g 30-min spin. Membrane (290 pg) was incubated with 1 nM [“Hlprazosin (22 Ci/mmol) in 120 mM NaCl, 1 mM EDTA, and 50 mM Tris, pH 7.4, in a 0.4-ml volume for 1 h at room temperature with or without nitrendipine. The nitrendipine concentrations tested were 5 or 50 pM. Nonspecific binding was determined by incubation with 1 FM prazosin. Bound and free ligand were separatedby filtration through 13-mmglassfiber filters. Data analysis
Statistical analyseswere performed by paired or unpaired t test. Resultsare expressedasthe mean + SE. Results The dependency of ol,-agonist-stimulated ANP secretion on calcium influx was examined by superfusion with
200 5 i= Id 5 LIJ
150
v, CL z 6 100
501
' 0
I 5
t 10
I 15
TIME
I 20
1 25
I 30
I 35
I 40
I 45
(min)
FIG. 1. Effect of a lowered extracellular calcium concentration on phenylephrine-stimulated ANP secretion. Rat left atria paced at 2 Hz were superfused with 0.2 mM (0, n = 9; A, n = 6) or 1.8 mM calcium (0; n = 7). Atria were continuously superfused with 10 pM phenylephrine (0 and 0) from 15-45 min or superfused without phenylephrine (A). Results are expressed as a percentage of the mean of seven baseline measurements obtained between O-15 min. Basal ANP secretion was 269 k 35 (A), 294 + 44 (0), and 324 f 42 pg/ml (0).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 05:48 For personal use only. No other uses without permission. . All rights reserved.
CALCIUM TABLE 1. Effect
of calcium-lowering Phenylephrine (10 PM)
Group Control
Calcium
(0.2
mM)
Control
Ryanodine
(1
GM)
(10
on atria1
pM)
“P < 0.01 compared * No change.
aI-STIMULATED
function Resting tension (g)
Before After
0.10 * 0.01
0.18 f 0.02"
0.34 * 0.02 NC*
Before After
0.02 + 0.01 0.01 * 0.01
0.29 + 0.02 NC
Before After
0.10 + 0.01 0.17 + 0.01”
0.32 + 0.02 NC
0 0
SECRETION
1019
1
0.32 k 0.02 NC
Before After
0.12 f 0.02 0.20 k 0.03”
0.46 rk 0.04 NC
Before After
0.06 + 0.01 0.06 + 0.01
0.30 + 0.03 NC
to before
ANP 250
Developed tension (g)
Before After
Control
Nitrendipine
agents
AND
11 0
measurements.
’ 5
’ 10
’ 15
’ 20
’ 25
TIME
’ 30
’ 35
’ 40
’ 45
’ 50
’ 55
J 60
(min)
FIG. 3. Effects
6
of nitrendipine on phenylephrine-stimulated ANP secretion. Rat left atria paced at 2 Hz were superfused with 10 pM nitrendipine (0, n = 6; A, n = 6) or vehicle alone (e, n = 5). Atria were continuously superfused with 10 pM phenylephrine (0 and 0) from 1560 min or superfused without phenylephrine (A). Results are expressed as described in Fig. 1. Basal ANP secretion was 253 k 41 (A), 244 + 35 (0), and 246 + 35 pg/ml (0).
200
ifi t!s
150
w U-J LL 5
100
50 1
I
I
I
0
5
10
15
I
I
I
I
I
20
25
30
35
40
TIME
11
45
11
50
55
60
(min)
FIG. 2. Effect of ryanodine on phenylephrine-stimulated ANP secretion. Rat left atria paced at 2 Hz were superfused with 1 @M ryanodine (0, n = 8; A, n = 6) or vehicle alone (e, n = 8). Atria were continuously superfused with 10 FM phenylephrine (0 and 0) from 15-60 min or superfused without phenylephrine (A). Results are expressed as described in Fig. 1. Basal ANP secretion was 163 + 27 (A), 221 f 21 (0), and 256 f 25 pg/ml (0).
pine. Superfusion with 10 PM nitrendipine phenylephrine-stimulated secretion by 49% veloped tension was lower in atria superfused nitrendipine, and it did not rise with the phenylephrine (Table 1). Nitrendipine did [3H]prazosin binding to atria1 membranes.
inhibited (Fig. 3). Dewith 10 PM addition of not inhibit Binding in
the presence of 5 or 50 PM nitrendipine was 96 k 6% of the control value (n = 3 experiments). Thus, phenylephrine-stimulated ANP secretion is partially dependent on calcium influx through nitrendipine-inhibitable calcium channels. To corroborate the findings with nitrendipine, the effect of phenylephrine on nonbeating atria was examined. In nonbeating left atria, calcium channels are quiescent, due to the absence of membrane depolarization. Thus, a lower ANP secretory response to phenylephrine should occur in nonbeating atria us. paced atria. Indeed, the response to phenylephrine was 52% lower in nonbeating atria (Fig. 4), similar to the lower response of atria superfused with nitrendipine (Fig. 3). Thus, it appears that half of the calcium influx is derived from calcium entry through voltage-dependent calcium channels. It has recently been reported that cui-adrenergic stimulation activates the Na-H antiporter, resulting in a rise in intracellular calcium (17). The increase in intracellular calcium was abolished in the absence of extracellular sodium or calcium and by the Na-H antiporter inhibitor HMA (18), suggesting that the rise in intracellular sodium from enhanced Na-H exchange may increase calcium influx through Na-Ca exchange. To determine whether calcium influx via Na-Ca exchange was occur-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 05:48 For personal use only. No other uses without permission. . All rights reserved.
CALCIUM
1020
AND
al-STIMULATED
ANP
SECRETION
Endo. Voll30.
1992 No 2
200
250 oz
150
cl i!5 ti
100
2 50
0
5
10
15
20
25
TIME
30
35
40
45
50
55
60
(min)
FIG. 4. Effect of phenylephrine on paced and nonbeating rat left atria. Atria were continuously superfused with 10 pM phenylephrine from 15-60 min and were either paced at 2 Hz (0; n = 6) or were nonbeating (0, n = 6). Half of the control atria (A; n = 6), which were not superfused with phenylephrine, were paced at 2 Hz, and half were nonbeating. Results were expressed as described in Fig. 1. Basal ANP secretion was 195 + 37 (A), 122 f 14 (0), and 202 f 31 pg/ml (0).
ring as the result of an increase in Na-H exchange and intracellular sodium, we examined the effect of HMA on phenylephrine-stimulated ANP secretion by nonbeating left atria. Using this experimental paradigm, we could determine whether calcium channel-independent calcium influx was occurring by this mechanism. HMA (25 pM) did not inhibit phenylephrine-stimulated ANP secretion (Fig. 5). Thus, enhanced calcium influx via NaCa exchange resulting from a rise in intracellular sodium due to increased Na-H antiporter activity does not appear to be involved in phenylephrine-stimulated ANP secretion. As illustrated in Fig. 1, calcium influx is necessary for initiating the ANP secretory response to phenylephrine. We designed an experiment to determine the dependency on calcium influx of the maintained secretory response to phenylephrine. Atria were superfused in 1.8 mM calcium and stimulated with phenylephrine for 45 min. ANP secretion rose 1.9-fold in response to 10 PM phenylephrine. Atria were either maintained in 1.8 mM calcium or switched to medium 199 containing 0.2 mM calcium during continuous superfusion with 10 PM phenylephrine. Lowering the buffer calcium concentration did not affect the continued secretory response to phenylephrine (Fig. 6). Developed tension fell from a maximum of 0.13 f 0.02 to 0.01 + 0.00 g in atria superfused with
1
1
t
I
I
I
1
I
I
I
I
0
5
10
15
20
25
30
35
40
45
TIME (min) FIG. 5. Effect of HMA on phenylephrine-stimulated ANP secretion. Nonbeating rat left atria were superfused with 25 pM HMA from 5-45 min (0, n = 6; 0, n = 6) or vehicle alone (A; n = 6). Atria were continuously superfused with 10 pM phenylephrine (0 and A) from 1545 min or superfused without phenylephrine (0). Results are expressed as described in Fig. 1. Upper panel, The graph of phenylephrine plus HMA (0) is presented after correction for the rise in ANP secretion by superfusion with HMA alone. Lowel panel, Graphs illustrate the ANP secretary response to HMA alone (0) and to phenylephrine plus HMA (0) without corrections. Basal ANP secretion was 110 f 12 (0), 125 f 28 (O), and 123 + 14 pg/ml (A).
0.2 mM calcium compared to only a slight fall in control atria from 0.14 f 0.02 to 0.11 f 0.01 g over the same time period. In contrast, when the a-adrenergic antagonist phentolamine (10 PM) was added to the superfusate, phenylephrine-stimulated ANP secretion fell to near-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 05:48 For personal use only. No other uses without permission. . All rights reserved.
CALCIUM
0
5
10
15
AND
20
al-STIMULATED
25
JO
TIME (min) FIG. 6. Effect of lowering extracellular calcium from 1.8 to 0.2 mM on phenylephrine-stimulated ANP secretion. Rat left atria paced at 2 Hz were superfused with 10 pM phenylephrine for 45 min before time zero and for 30 min thereafter. At time zero, the superfusate calcium concentration was lowered to 0.2 calcium (0, n = 10) or remained in 1.8 mM calcium (e n = 9). As a positive control, 10 FM phentolamine was added to the superfusate at time zero for those atria superfused with 1.8 mM calcium (A; n = 6). Results are expressed as a percentage of the net increase in ANP secretion, using the mean of two measurements obtained at -2.5 and 0 min as 100%.
basal levels (Fig. 6), and developed tension fell to baseline. Initially, developed tension was 0.17 f 0.01 g, which rose to 0.29 + 0.05 g with phenylephrine and then fell to 0.14 f 0.02 g with the addition of phentolamine. Thus, once the secretory response to phenylephrine is established, calcium influx appears to play a markedly reduced role in the maintenance of the response over 30 min. Discussion
Atria1 contraction results from a rise in cytosolic calcium. The sources of calcium responsible for this rise are both extracellular and intracellular. With membrane depolarization, calcium influx occurs primarily through voltage-dependent calcium channels and to a limited degree via Na-Ca exchange (19). The rise in cytosolic calcium, resulting from calcium influx releases calcium from the SR (20). Atria1 contraction then occurs as a result of these events. Lowering the extracellular calcium concentration in the present study totally blocked the initiation of phenylephrine-stimulated ANP secretion. This may be the result of decreased calcium influx and/ or lack of calcium release from the SR. The fact that ryanodine did not inhibit phenylephrine-stimulated se-
ANP
SECRETION
1021
cretion suggests that the source of calcium necessary for the initiation of the response is extracellular and not intracellular in origin. Calcium influx primarily occurs by the opening of voltage-dependent calcium channels and Na-Ca exchange. Superfusion with 10 PM nitrendipine inhibited phenylephrine-stimulated ANP secretion by 49%. The dose of 10 PM nitrendipine has been previously shown by us to totally negate isoproterenoland (Bu)~AMPstimulated ANP secretion (21). Inhibition was not due to nitrendipine interfering with al-adrenergic binding sites. Thus, we conclude that half of the calcium necessary for initiating phenylephrine-stimulated secretion is derived from calcium influx through L-type calcium channels. In nonbeating atria, where calcium channels are relatively quiescent, phenylephrine-stimulated ANP secretion was 52% lower than in paced atria. This corroborates the nitrendipine studies and also suggests that calcium influx through T-type calcium channels is probably of negligible importance, because there was no difference in the lowering of phenylephrine-stimulated secretion by nitrendipine and nonbeating atria. Thus, these findings are consistent with the observation that CQadrenergic agonists increase calcium influx through voltage-dependent calcium channels by prolonging the action potential (4-6). In contrast, our study does not support the speculation that a,-adrenergic agonists increase SR calcium release. As mentioned in the introduction, several investigators have found that inositol-1,4,5-trisphosphate either releases or enhances the release of calcium from the SR (8-10); however, there is not uniform agreement on this point (22). If it were true that inositol-1,4,5-trisphosphate enhances SR calcium release, ryanodine should inhibit phenylephrine-stimulated secretion to some degree, and it did not. Ryanodine was biologically active in this study, as demonstrated by a fall in developed tension to undetectable levels. The dose of 1 ~.LM ryanodine has been previously shown by us to totally negate isoproterenol- and (Bu)2AMP-stimulated ANP secretion (21). This latter observation is consistent with the known effect of protein kinase A on the SR to increase the pool of stored calcium by phosphorylation of phospholamban (23). Therefore, the present study does not support the premise that inositol-1,4,5-trisphosphate generated by phenylephrine stimulation significantly enhances SR calcium release due to the failure of ryanodine to inhibit phenylephrine-stimulated secretion. We conclude that calcium influx, not calcium release from the SR, is the primary source of calcium necessary for initiating phenylephrine-stimulated secretion. Our results differ from those of Sei and Glembotski (24), who found that lowering the buffer calcium concentration to 2 /IM did not inhibit phenylephrine-stimulated
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 05:48 For personal use only. No other uses without permission. . All rights reserved.
1022
CALCIUM
AND
(u~STIMULATED
secretion by cultured rat neonatal atria1 cardiocytes. When the extracellular calcium level was lowered further to 2 nM with EGTA, they observed a 40% lower phenylephrine-stimulated response. In contrast, results similar to ours were found with the calcium channel inhibitor nifedipine. Phenylephrine-stimulated secretion was inhibited by a little greater than 50% with nifedipine. Thus, their results suggest that phenylephrine-stimulated secretion is partially independent of calcium influx, whereas our results indicate a complete dependency on calcium influx. However, we both agree that calcium influx through L-type calcium channels plays a significant role in phenylephrine-stimulated secretion. The differences in these findings may reflect differences in calcium handling by neonatal cells us. adult atria1 cells or differences between cultured cells and intact atria1 tissue. Calcium influx by a calcium channel-independent mechanism appears to account for 50% of the calcium influx. Residual calcium influx is most likely achieved by Na-Ca exchange. Na-Ca exchange activity may be altered by increasing intracellular sodium secondary to enhanced Na-H antiporter activity (17). A rise in intracellular sodium increases cytosolic calcium by either increasing calcium influx or decreasing its efflux by NaCa exchange. We examined this possibility by using HMA to inhibit the Na-H antiporter. HMA did not inhibit phenylephrine-stimulated secretion, suggesting another mechanism by which Na-Ca exchange is altered. It is unlikely that higher concentrations of HMA would have demonstrated an inhibitory effect. We have previously shown that 25 j.~tM HMA totally negates frequencystimulated ANP secretion (25). Secondly, the Ki for HMA inhibition of Na-H antiporter activity is estimated to be 0.2 PM (18). Thus, we chose a concentration that was more than loo-fold higher than the Ki. Na-Ca exchange activity may be increased by cr,-adrenergic agonists by activation of phospholipase C (26). Also, phosphatidic acid, which may be a product of phospholipase C by conversion of diacylglycerol to phosphatidic acid, has been shown to increase Na-Ca exchanges (26). However, from these studies it is not clear in which direction calcium is being moved. The orientation of the exchanger is not known in these experiments, since they were performed with sarcolemmal vesicles, which may be inside out or right side out. In addition, the exchanger can move calcium either into or out of the cells depending on the relative concentrations of sodium and calcium across the membrane. Thus, activation of phospholipase C by phenylephrine may increase calcium influx through Na-Ca exchange. This may be the explanation for calcium channel-independent calcium influx. In contrast to the marked dependency on calcium influx to initiate the ANP secretory response to phenyl-
ANP SECRETION
Endo.
Vol130.No
1992
2
ephrine, calcium influx appears to play a minimal role in the maintenance of an established response, as suggested by the lack of an inhibitory effect of lowering the buffer calcium concentration over 30 min. During this time, developed tension fell to almost undetectable levels after prior stimulation with phenylephrine. With the addition of phentolamine, phenylephrine-stimulated secretion fell to near baseline, demonstrating that removal of the stimulatory signal does indeed induce a fall in phenylephrine-stimulated ANP secretion. In addition, we have previously demonstrated that lowering the buffer calcium Concentration to 0.2 mM lowers isoproterenolstimulated ANP secretion to baseline (21). These results suggest a stimulatory pathway whose initiation is dependent on calcium influx; however, once the signal is set in motion, calcium influx is no longer necessary at least for the short term. The mechanism of this observation may be due to activation of multifunctional Ca2+/calmodulin-dependent protein kinase (CaM kinase). CaM kinase undergoes autophosphorylation after activation by calcium and calmodulin. The result of autophosphorylation is the release of the kinase from its dependence on calcium and calmodulin for activity. This autonomous kinase is then capable of phosphorylating exogenous substrates in the absence of calcium and calmodulin (27). Thus, autophosphorylation of CaM kinase renders it active in spite of a fall in cytoplasmic calcium. Activated CaM kinase may explain the persistence of phenylephrine-stimulated ANP secretion in the presence of low extracellular calcium concentrations. Basal ANP secretion varied at times between groups within experiments or between studies. This may be due to paced vs. nonpaced atria or the addition of organic solvents, such as DMSO or ethanol. It may also reflect differences in atria1 size or shape of the atria, which determines the amount of atria1 tissue stretched in a single plane. Atria were randomly selected for all study groups. The calcium dependency of ai-adrenergic agoniststimulated ANP secretion is unique from that of /3adrenergic or endothelin-stimulated secretion (Table 2). As we have previously reported, P-agonist-stimulated secretion is totally dependent on calcium influx and is TABLE
agonist-,
2. Comparison of the dependency and endothelin-stimulated ANP
on calcium secretion
by P-agonist-,
Stimulus Exp
variable Isoproterenol
0.2 mM calcium 2 mM lanthanum 10 pM nitrendipine 1 pM ryanodine Values
shown
are percent
Phenylephrine
100
100
100 100
50 0
inhibition.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 05:48 For personal use only. No other uses without permission. . All rights reserved.
Endothelin 50 100 50 0
crl-
CALCIUM
AND
,X,-STIMULATED
negated by 10 PM nitrendipine or 1 PM ryanodine (21). These results suggest that the primary source of cytosolic calcium responsible for stimulation is from the expanded calcium storage pool in the SR. In contrast, endothelinstimulated secretion is only partially inhibited by 0.2 mM calcium (12). However, superfusion with 2 InM lanthanum, an inhibitor of calcium transport, in the presence of 1 mM calcium completely inhibited endothelinstimulated ANP secretion (our personal observation). A similar observation was made with ouabain-stimulated ANP secretion. Again, superfusion with 0.2 mM calcium only partially inhibited stimulated secretion, whereas 2 mM lanthanum completely negated ouabain-stimulated secretion (our personal observation). Thus, when calcium influx is aggressively inhibited, endothelin stimulation also appears to be totally dependent on calcium influx. Calcium influx is partially achieved through L-type calcium channels, similar to that in response to al-adrenergic stimulation. Ryanodine had no effect on endothelinstimulated secretion. These results suggest that calcium influx, but not SR calcium release, is required for endothelin-stimulated secretion. Thus, the calcium dependency of phenylephrine-stimulated secretion is similar, but not identical, to that of endothelin (Table 2). This is consistent with the observation that both endothelin and al-adrenergic agonists activate the phosphoinositide pathway and do not activate adenylyl cyclase. However, endothelin is more resistant to inhibition by superfusion with 0.2 InM calcium. This may be due to stimulation by endothelin of phospholipase AZ, which may give it unique stimulating properties (28). Thus, these three secretogogues of ANP release display different dependencies on calcium, which may reflect differences in the second messenger systems evoked by each. Acknowledgments The excellent technical support by Jennifer Harmon Terri Carswell for
is greatly typing the
appreciated. manuscript.
ANP 6. Miura
on the
also
Sullivan thank
and Ms.
8.
9.
10.
11.
12.
13.
14. 15.
16. 17.
18.
19. 20.
22.
23.
References 1. Sonnerberg H, Veress AT 1984 Cellular mechanism of release of atria1 natriuretic factor. Biochem Biophys Res Cummun 124:443449 Currie MG, Newman WH 1986 Evidence for a-1 adrenergic receptor regulation of atriopeptin release from the isolated rat heart. Biochem Biophvs Res Commun 137:94-100 Schiebinger I&J,-Baker MZ, Linden J 1987 Effect of adrenergic and muscarinic cholinergic agonists on atria1 natriuretic neptide secretion by isolated rat atria-J Clin Invest 80:1687-1691’ Fedida D. Shimoni Y. Giles WR 1990 a-Adrenereic modulation of the transient outward current in rabbit atria1 myocytes. J Physiol
24.
25
26.
27.
stimulation Pharmacol
Otani H, Das DK 1988 a,-Adrenoceptor-mediated phosphoinositide breakdown and inotropic response in rat left ventricular papillary muscles. Circ Res 62:8-17 Hirata M, Suematsu E, Hashimoto T, Hamachi T, Koga T 1984 Release of Ca’+ from a non-mitochondrial store site in peritoneal macrophages treated with saponin by inositol 1,4,5-trisphosphate. Biochem J 223:229-236 Nosek TM, Williams MF, Zeigler ST, Godt RE 1986 lnositol trisphosphate enhances calcium release in skinned cardiac and skeletal muscle. Am J Physiol 25O:C807-C811 Fabiato A 1989 Comparison and relation between inositol (1,4,5)trisphosphate-induced release and calcium-induced release of calcium from the cardiac sarcoplasmic reticulum. In: Yamanada K, Shibata S (eds) Recent Advances in Ca’+-Channels and Ca2+Antagonists. Proceedings of the Japan-U.S. Symposium on Cardiovascular Drugs. Pergamon Press, Elmsford, pp 35-39 Cragoe Jr EJ, Woltersdorf OW, Bicking JB, Kwong SF, Jones JH 1967 Pyrazine diuretics. Il. N-Amidino+amino-5-substituted-6halopyrazine-carboxamides. J Med Chem 10:66-75 Schiebinger RJ, Gomez-Sanchez CE 1990 Endothelin: a notent stimulus-of atria1 natriuretic peptide secretion by superfused rat atria and its dependency on calcium. Endocrinology 127:119-125 Schiebinger RJ, Linden J 1986 The influence of resting tension on immunoreactive atria1 natriuretic peptide secretion by rat atria superfused in uitro. Circ Res 59:105-109 Fabiato A 1985 Effects of ryanodine in skinned cardiac cells. Fed Proc 44:2970-2976 Wier WG, Yue DT, Marban E 1985 Effects of ryanodine on intracellular Ca”’ transients in mammalian cardiac muscle. Fed Proc 44:2989-2993 Rousseau E, Smith JS, Meissner G 1987 Ryanodine modifies conductance and gating behavior of single Ca*+ release channel. Am J Physiol253:C364-C368 lwakura K, Hori M, Watanabe Y, Kitabatake A. Craeoe Jr EJ. Yoshida H, Kamada T 1990 al-Adrenocepter stimulation increases intracellular PH and Ca2+ in cardiomvocvtes through Na+/H+ and Na+/Ca’+ exchange. Eur J Pharmacol186:29-40 ’ Simchowitz L, Cragoe EJ 1986 Inhibition of chemotactic factoractivated Na+/H+ exchange in human neutrophils by analogues of amiloride: structure-activity relationships in the amiloride series. Mol Pharmacol30:112-120 Leblanc N, Hume JR 1990 Sodium current-induced release of calcium from cardiac sarcoplasma reticulum. Science 248:372-376 Nabauer M, Callewaert G, Cleemann L, Morad M 1989 Regulation of calcium release is gated by calcium current, not gating-charge, in cardiac myocytes. Science 244:800-803 Schiebinger RJ 1989 Calcium, its role in isoproterenol-stimulated atria1 natriuretic peptide secretion by superfused rat atria. Circ Res 65:600-606 Movsesian MA, Thomas AP, Selak M, Williamson JR 1985 lnositol trisphosphate does not release Ca’+ from permeabilized cardiac myocytes and sarcoplasmic reticulum. FEBS Lett 185:328-332 Tsien RW 1977 Cyclic AMP and contractile activity in heart. Adv Cyclic Nucleotide Res 8:363-420 Sei CA, Glembotski CC 1990 Calcium dependence of phenylephrine-, endothelin-, and potassium chloride-stimulated atria1 natriuretic factor secretion from long term primary neonatal rat atria1 cardiocytes. J Biol Chem 265:7166-7172 Schiebinger RJ 1988 Mechanism of enhanced atria1 natriuretic peptide (ANP) secretion by increased atria1 contraction frequency. Hypertension 12:354 (Abstract) Philipson KD, Frank JS, Nishimoto AY 1983 Effects of phospholipase C on the Na+-Ca*+ exchange and Ca’+ permeability of cardiac sarcolemmal vesicles. J Biol Chem 258:5905-5910 Schulman H, 1988 The multifunctional Ca’+/calmodulin-dependent protein kinase. Adv Second Messenger Phosphoprotein Res
22:X-l
4231257-277
Dukes ID, Vaughan Williams EM 1984 Effects of selective /&-, P2-, (Y,- and al-adrenoceptor stimulation on potentials and contractions in the rabbit heart. J Physiol355:523-546
Y, lnui J 1984 Multiple effects of n-adrenocepter action potential of the rabbit atrium. Arch
1023
325:47-53 7. Otani H,
21.
William We
SECRETION
28.
12
Resnick TJ, Scott-Burden T, Buhler FR 1989 Activation pholipase A, by endothelin in cultured vascular smooth cells. Biochem Biophys Res Commun 158:279-286
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 15 November 2015. at 05:48 For personal use only. No other uses without permission. . All rights reserved.
of phosmuscle
