0013.7227/92/1305-2455$03.00/O Endocrinology Copyright 0 1992 by The Endocrine

Vol. Printed

Society

130, No. 5 in U.S.A.

Endothelin-Induced Atria1 Natriuretic Peptide Release from Cultured Neonatal Cardiac Myocytes: The Role of Extracellular Calcium and Protein Kinase-C* PAAVO A. UUSIMAA, OLLI VUOLTEENAHO,

ILMO AND

Departments of Medical Biochemistry (H.R.), University of Oulu, SF-90220

E. HASSINEN, HEIKKI RUSKOAHO (P.A. U., I.E.H.), Oulu, Finland

Physiology

ABSTRACT. Regulation of atria1 natriuretic peptide (ANP) secretion from neonatal rat myocytes cultured on microcarriers was studied using endothelin-1 (ET-l) as a secretagogue. Myocytes were cultured for 3 days on microcarriers, packed in a chromatography column, and perifused with Krebs-Henseleit bicarbonate buffer. ANP secretion was measured by RIA, and the cytosolic free calcium concentration ([Ca”],) was measured continuously during secretion by the fluorescent calcium indicator fura-2: In perifused atria1 and ventricular cells, basal values for lCa’+l< were 146 and 167 nM. and immunoreactive ANP (IRANP) secretion rates were 61 and 65 pg/min. mg protein, respectively. ET-1 at concentrations of 1, 10, and 100 nM caused a concentration-dependent increases in [Ca*+]r and IR-ANP secretion in atria1 myocytes. The maximal increases in [Ca’+], and IR-ANP secretion were 30% and lOO%, respectively. Diltiazem (1 PM), an inhibitor of voltage-sensitive Ca*+ channels, inhibited [Ca*+h increments, but had no effect on ET-induced

(0. V.), and Pharmacology

and Toxicology

IR-ANP secretion. Staurosporine (10 nM), a protein kinase-C inhibitor, augmented [Ca’+]rchanges, but inhibited the sustained phase of ET-induced IR-ANP secretion (P < 0.05). Diltiazem abolished the stimulatory effect of staurosporine on [Ca*+]r and its inhibitory effect on IR-ANP secretion. ET-1 caused increases in [Ca2+lr and IR-ANP secretion in ventricular myocytes similar to those in atria1 myocytes. Peptides corresponding in size to pro-ANP and ANP-(1-28) were detected in the original cell culture medium and perifusion effluent, and ET-1 did not change their concentration ratio in the eluate. Lactate dehydrogenase was not detected in the effluents before or during ET infusion, showing that the increase in IR-ANP secretion was not due to cell damage. This study shows that ET stimulates atria1 and ventricular ANP secretion. The results also suggest that sustained ET-induced atria1 ANP secretion is dependent on protein kinase-C, but does not require the influx of extracellular calcium. (Endocrinology 130: 2455-2464, 1992)

A

TRIAL natriuretic peptide (ANP), a hormone that regulates salt and water balance and blood pressure, is synthesized and secreted by atria1 and ventricular myocytes. Atria1 stretch stimulates ANP secretion in uiuo and in uitro, whereas the physiological stimulus for ventricular ANP release is unknown (1, 2). Most studies suggest that calcium-activated protein kinase-C regulates ANP release (3-9), but the direct role of calcium in ANP secretion is controversial. In perfused rat heart (3, 4, lo), isolated atria (ll), and short term (5, 8) and long term myocyte cultures (12), extracellular calcium influences ANP secretion. In contrast, ANP secretion from cultured neonatal myocytes (9), freshly isolated myocytes (13), and quiescent adult atria1 myocyte cultures (6) was not dependent on Ca’+, and Ca2+ inhibited ANP secre-

tion from isolated atria (14). Moreover, increasing Ca2+ influx inhibited stretch-induced ANP release from the perfused rat heart (15). Endothelins (ETs) are a family of 21-amino acid vasoactive peptides that act in a paracrine fashion on nearby smooth muscle and connective tissue cells (16). They bind to membrane receptors that are coupled to Gproteins (17, 18). ETs activate phospholipase-C and promote the hydrolysis of phosphatidylinositol, with subsequent production of diacylglycerol and inositol 1,4,5triphosphate (IP,) (19). Mobilization of intracellular Ca’+ by IPB, Ca*+ influx through the plasma membrane, and activation of protein kinase-C by diacylglyceroi may then mediate the intracellular effects of ET. Transmembrane signalling by ETs also involves arachidonic acid cascade, Na+-H+ antiport, Na+-K+-ATPase, and changes in membrane potential (19). ET has been shown to stimulate ANP gene expression in cultured myocytes (20-22). It also releases ANP in uiuo (23, 24) and from isolated atria (25-27) and increases basal ANP release from cultured atria1 (12, 20, 28, 29) and ventricular (21,

Received October 25, 1991. Address all correspondence and requests for reprints to: Dr. Heikki Ruskoaho, Department of Pharmacology and Toxicology, University of Oulu, Kajaanintie 52 D, SF-90220 Oulu, Finland. * This work was supported by grants from the Medical Research Council of the Academy of Finland (Helsinki, Finland), the Sigrid Juselius Foundation (Helsinki, Finland). and the Paavo Ilmari Ahvenainen Foundation (Helsinki, Finland). 2455

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 09:22 For personal use only. No other uses without permission. . All rights reserved.

ET-l-INDUCED

2456

22) myocytes. Moreover, both basal and stretch-induced ANP secretion from isolated perfused hearts (30) and cultured atria1 myocytes (22) are increased by ET. Because shear stress releases ET from endothelial cells (31), ET may contribute to the stretch-induced ANP secretion. However, the mechanism of ET-induced ANP release is unknown. In the present experiments the mechanism of ANP secretion from cultured neonatal cardiomyocytes was studied by simultaneous measurements of the intracellular free calcium concentration ([Ca”],) and immunoreactive ANP (IR-ANP) secretion (9) using ET-l as a secretagogue. Because previous studies suggested that Ca2+ and protein kinase-C are involved in ANP release, the interactions of diltiazem, an inhibitor of L-type voltage-sensitive calcium channels (32), and staurosporine, an inhibitor of protein kinase-C (33), with ET were tested. Our results show that ET stimulates both atria1 and ventricular ANP secretion and that the effect of ET on atria1 ANP secretion may be mediated by protein kinase-C, whereas an influx of extracellular calcium is not necessary for ET-induced ANP release Materials

and Methods

Materials

Fetal calf serum (FCS), RPMI-1640 cell culture medium, Dulbecco’s PBS, and glutamine were obtained from Gibco Europe (Paisley, Scotland); antibiotics and dexamethasone from Sigma Chemicals (St. Louis, MO); staurosporine from Fluka AG (Buchs, Switzerland); collagenase,type Worthington (CLS II), from Millipore (Freehold, NJ); ET-l from Peninsula Laboratories, Inc. (Belmont, CA); indomethacin from Laakefarmos (Turku, Finland); diltiazem hydrochloride from Medipolar (Oulu, Finland); ionomycin from Calbiochem (La Jolla, CA); fura- acetoxymethyl ester (fura- AM) from Molecular Probes(Eugene,OR); regular insulin (Insulin Velosulin) from Nordisk Insulinlaboratorium (Gentofte, Denmark); BioRad protein assayreagent and Econo-Columns from Bio-Rad Laboratories (Richmond, CA); Cytodex 3 microcarriers from Pharmacia Fine Chemicals (Uppsala, Sweden); spinner flasks from Bellco Glass (Wineland, NJ); Sep-Pak Cl8 cartridges, a Protein-Pak I-125 HPLC gel filtration column, and reverse phaseC4 and phenyl columnsfrom Waters (Milford, MA); rat [‘251]ANP-(1-28) from Amersham (Aylesbury, Buckinghamshire, United Kingdom); and other chemicals from E. Merck (Darmstadt, Germany). Microcarrier

cell culture

Neonatal rat heart myocytes were grown on Cytodex 3 microcarriers, as describedpreviously (9, 34). The experimental protocol was approved by the Committee for Animal Experimentation of the University of Oulu. About 20 3- to 5-day-old Sprague-Dawley rats from the stocks of the Departments of Medical Biochemistry and Physiology (Oulu, Finland) were decapitated,and the thorax was openedaseptically. The aorta

ANP RELEASE

Endo. Vol130.

1992 No 5

was clamped, the right atrium was opened,and the heart was perfused with the disaggregationmedium (2 g collagenaseCLS II and 50 pmol CaCl*/liter PBS) through a needleinserted into the left ventricle through the apex. The apicesand basesof the hearts were dissectedseparately and disaggregatedby repeatedly incubating in collagenasemedium and pipetting. The supernatants were filtered, and the cells were washedby repeated centrifugation and resuspensionin fresh RPMI-1640 cell culture mediumsupplementedwith 2 mmol glutamine, lo5 U benzylpenicillin, 100 mg streptomycin sulfate, and 200 ml FCS/liter (medium 1). Most of the fibroblasts were eliminated by adhesionduring 30-min incubation on dishesat 37 C in a humidified atmosphere of 5% CO, in air. The unattached myocytes were transferred to the spinner flasks containing 5 ml preswollen Cytodex 3 microcarriers in 50 ml medium 1 supplementedwith 100nmol dexamethasone/liter. The suspension was stirred for 30 set intermittently during the first 6 h, starting with 15-min intervals and ending with a 1.5-h interval and then continuously at 25 rpm. The cultures were maintained for 3 days. Fifty milliliters of fresh culture medium RPMI-1640 supplementedwith 2 mmol glutamine, lo5 U benzylpenicillin, 100mg streptomycin sulfate, 100 nmol dexamethasone,and 100 ml FCS/liter (medium 2) were added after 24 h, and half of the medium was replaced with medium 2 after 48 h. After 3 days in culture, myocardial cells grown on microcarriers were washedby sedimentation at 1 x g in fresh culture medium 2 to remove dead cells from the cultures and incubated in culture medium2 at 37 C for at least 60 min before the experiments. Measurement

of [Ca2+], in perifused

myocytes on microcarriers

Calcium was measuredessentially as previously described (9). A chromatography column (length, 4 cm; id, 7 mm; volume, 1.54 ml) modified for cell perifusion was filled with myocytes on microcarriers and inserted into the cell holder in a Farrand spectrofluorometer (Farrand Optical Co., Valhalla, NY). The fluorometer cell holder, the perifusion fluid reservoir, and most of the connecting tubing were thermostated at 37 C with a recirculating water bath. The fluorescenceexcitation light beam had a wavelength maximum at 340 nm, with a half-intensity band width of 10 nm, and the emissionwas recorded with a half-intensity band width of 20 nm centered at 510 nm. The myocytes were perifused at a flow rate of 1 ml/min with Krebs-Henseleit bicarbonate buffer containing 1.25 mmol CaCl*, 10 mmol glucose,and 12 U insulin/liter and equilibrated with 02-CO, (19:l) at 37 C. Five-minute fractions of the outflow were collected in tubes kept on ice and stored at -20 C until analyzed. The myocytes were loadedwith fura- AM (3.3 mM stock solution in dimethylsulfoxide added to the perifusion mediumto a final concentration of l-2 pM) for 30-40 min until the fluorescenceemissiontrebled comparedwith the autofluorescenceof the packed cells. Unhydrolyzed fura- AM was washed away, and the fluorescence reading was allowed to stabilize for IO-15 min. After recording the basal fluorescence for 20 min, the myocytes were exposedto 1, 10, and 100 nM ET-l (40 pM stock solution in 0.9% NaCl), 10 and 100 nM staurosporine (2.1 mM stock solution in dimethylsulfoxide), 1 pM diltiazem hydrochloride (1.8 mM stock solution in water), and 0.1 mM indomethacin (28 mM stock solution in ethanol)

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 09:22 For personal use only. No other uses without permission. . All rights reserved.

ET-l-INDUCED or their different combinations. At the end of each experiment maximum fluorescence (F,,,) was obtained after a lo-min infusion of 10 pM ionomycin (10 mM stock solution in ethanol) and minimum fluorescence (F,J after a lo-min infusion of 22 mM EGTA (0.5 M stock solution in water). [Ca’+]r was calculated according to the equation [Ca’+]r = KJF - F,,,)/(F,,, F), assuming an effective Kd of 224 nM (35). After perifusion, the column was emptied, and the packed volume of the microcarriers was measured. ANP immunoreactivity, the molecular form of ANP, the activity of lactate dehydrogenase, and protein determinations

The perifusate samplesfor the ANP RIA were incubated in duplicatesof 100 ~1with 100~1specific rabbit ANP antiserum in a final dilution of 1:25,000(36). Synthetic rat ANP-(l-28) in the range 0- 500 pg/tube was used as a standard, and rat (““I]ANP-(1-28) as an ANP tracer. After incubation for 48 h at 4 C, the immunocomplexeswere precipitated with sheep antiserum against rabbit y-globulin in the presenceof 500 ~1 1.2 M ammonium sulfate, pH 7, followed by centrifugation at 3000 x g for 40 min. The sensitivity of the assaywas 0.8 pg/ tube. The 50% displacement of the standard curve was at 20 pg/tube. The inter- and intraassay variations were 14% and 5%, respectively. Serial dilutions of the perifusate were parallel to those of the synthetic ANP standard. Samples(1.5 ml) with a low concentration of IR-ANP were acidified to pH 4 with 300 ~1 1 M HCl in 1.6% glycine, as describedpreviously for plasma (37). The acidified samplewas applied to a Sep-Pak Cl8 cartridge that had been activated previously with propanol, followed by 10 ml 0.1% trifluoroacetic acid (TFA). After washing with 0.1% TFA, the peptide was eluted with 60% CH&N in 0.1% TFA. The eluateswere evaporated in a Speed-Vat concentrator (Savant, Hicksville, NY), and the residue was reconstituted with 400 ~1 RIA buffer containing 0.1 M NaCl, 0.05 M sodiumphosphate,0.02%sodium azide, and 0.1% BSA, pH 7.0. The recovery of the added rat ANP-(l-28) was81.7 + 3.3% (meanf SD; n = 6). The molecularform of the IR-ANP releasedinto culture and perifusion media was determined by HPLC gel filtration. The sampleswere acidified with HCl to pH 2, extracted with SepPak C18, dried in a Speed-Vat concentrator, and redissolved in 500 ~140% acetonitrile in aqueous0.1% TFA. The samples were applied to a Protein-Pak I-125 (3.9 x 300 mm) HPLC gel filtration column and eluted with the samesolvent at 1 ml/ min. Fractions of 0.5 ml were collected, dried in a Speed-Vat, and subjectedto ANP RIA, as describedabove. BSA and iZ511 wereusedto calibrate the void and total volumesof the column. Pro-ANP purified for calibration of the column from SpragueDawley rat auricles by acid extraction, gel filtration, and two stepsof reverse phaseHPLC (C4 and phenyl) was chromatographically identical (gel filtration and reverse phase HPLC) with freshly extracted immunoreactive pro-ANP from rat atria. It did not show any detectable degradation (as assessed by gel filtration, followed by RIA) when kept overnight in physiological salt solutions (pH 7.4), 0.1% TFA (pH 2.1), or 40% acetonitrile-0.1% TFA at 4 or 22 C and was stable for weekswhen kept at -20 C in the samesolutionsand apparently indefinitely if lyophilized. The recovery was similar for purified rat proANP and synthetic rat ANP-(l-28) and varied between 75

ANP RELEASE

2457

and 94% when 1 ng of each was loaded. The ANP antiserum recognizedsynthetic rat ANP-( l-28) and purified rat pro-ANP equally well on a molar basis. Lactate dehydrogenase(LDH) was chosen as the cytosolic marker and assayedspectrophotometrically (38). Bio-Rad reagent wasusedfor protein determinations in a separatesample from each culture, asdescribedpreviously (34). Statistical

analysis

The values are expressedin proportion to the mean basal value (mean f SE) during the first 20 min. The statistical significanceswere tested by two-way analysis of variance for repetitive measurements (ANOVA), followed by NewmanKeuls test. Results Basal and ET-induced changes release in atria1 and ventricular

in [Ca2+], myocytes

and IR-ANP

In perifused atria1 and ventricular cells, basal values for [Ca2+lf were 145.8 + 16.3 (n = 24) and 166.9 f 18.2 (n = 10) nM, respectively, during the 20-min basal period. IR-ANP secretion rates during the sameperiod were 60.8 + 11.0 (n = 24) and 64.9 f 8.2 (n = 10) pg/min.mg protein. Both [Ca’+]r and IR-ANP secretion remained stable during 60-min perifusion of the control cells (Figs. 1 and 2). LDH activity was not detected in the effluent fluids of atria1 or ventricular cells during the basal period or perifusion with effecters (data not shown). Peptides corresponding to both pro-ANP and ANP-(l-28) were detected in gel filtration of the culture medium and perifusion effluent; the major form was pro-ANP (Fig. 3). The cells that had been perifused for 2 h appeared intact using phase contrast microscopy after dismantling the column. ET-l caused a concentration-dependent increase in [Ca’+]r of atria1 cells (Fig. 1). ET-l at a concentration of 100 nM caused an initial 30% increase in [Ca2+lf, followed by a rapid return to a level 10% above the baseline and to the basal level after a 30-min stimulation. There was no significant change in [Ca*+]r at concentrations of 1 and 10 nM (P = 0.946 and 0.076, respectively). ET-l significantly increased IR-ANP secretion from atria1 myocytes at all concentrations used in the experiments (Fig. 1). ET-l (1 nM) caused a slowly developing increase in IR-ANP release. When the concentration of ET was increased, the initial rise in IR-ANP release was more rapid. ET-l (100 nM) induced a biphasic secretory response, with a steep initial increase followed by a slower decline to a level 40% above the baseline. The infusion of 10 nM ET-l did not change the ratio of the peptides corresponding to pro-ANP and ANP-( l-28) in gel filtration (Fig. 3). Similarly, ET-l at concentrations of 1 and 100 nM did not affect the processing of ANP (data not shown). ET-l at a concentration of 100 nM also increased

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 09:22 For personal use only. No other uses without permission. . All rights reserved.

2458

ET-l-INDUCED

ANP

RELEASE

Endo. Voll30.

1992 No 5

2.0-

I

0.6 1 0

I

%

1.6..

1

1.6..

2

1.4..

.:

1.2..

i

;I;..

20

30

40

50

*+q++q-+

'% 0.6.. ; :: 0.4-y P 0.2.

I

10

B

0.0 7 0

60

10

20

30

40

50

60

30

40

50

60

2.4-

FIG. 1. The effect of ET-1 on [Ca’+]r and ANP secretion by cultured atria1 myocytes. The duration of the infusion is shown by the horizontal bar. The values are expressed in proportion to the mean basal value during the first 20 min and as the mean + SE for five independent experiments. A and B, Vehicle; C and D, 1 nM ET-l; E and F, 10 nM ET-l; G and H, 100 nM ET-l. The variations in [Ca*+]r during the infusions of 1 and 10 nM ET-l were not statistically different from values in the control experiments, whereas during the infusion of 100 nM ET-l, the variation in [Ca*+]r was significant at the level of P c 0.001 (by ANOVA). The variation in ANP secretion during the infusion of 1 nM ET-1 was different from that of the control experiments at the level of P < 0.05 and during 10 and 100 nM ET-1 at the level of P < 0.001 (by ANOVA).

0.61

6

2.2..

P

2.0..

2

1.6..

2

1.6..

1:

1.2.. 1.4..

i

;:;-

s

0.6-b

I 0

10

20

30

40

50

D

A,l+h +-p

f

0.4,

60

0

10

20

2.4 -

0.6

I 1 0

II 10

20

30

40

50

6 P::

2.2..

54

1.6..

,j

1.6..

2s 'Z L

1.2.. 1.4.. 1.0..

P

0.6..

2

0.6.. 0.4+

2.0..

60

F

0

1.4 1

I

p'k@,d

/

10

20

30

40

50

4 60

10

20

30

40

50

Iml

248 P

2.2.. 2.0..

2 'zt

1.6..

3

1.4..

1.6 . .

5

1.2..

P

1.0..

P 0.6.. 2 0.6

7 0

10

20

30 Time (min)

[Ca2+lf and IR-ANP secretion in ventricular cells (Fig. 2). The patterns of calcium changes and IR-ANP release resembled those in atria1 cells with 10 nM ET-l. Effects of diltiazem, staurosporine, and indomethacin on ET-induced [Ca*+], changes and IR-ANP release in atria1 myocytes

The mechanism of ET-induced IR-ANP secretion from atria1 cells was studied using diltiazem, an inhibitor of L-type voltage-sensitive calcium channels (32), and staurosporine, a protein kinase-C inhibitor (33). A dose of 10 nM ET-l was chosen for these experiments because

40

0.6.. 0.44

50

60

O

Time (min)

it caused a statistically significant steady increase in IRANP secretion. Diltiazem and staurosporine alone had no effect on [Ca2+lf and IR-ANP secretion (Fig. 4). The combination of 10 nM ET-l and 1 FM diltiazem had no significant effect on [Ca2+lf (Fig. 5). However, 1 PM diltiazem inhibited the increase in [Ca*+]f induced by 100 nM ET-l (data not shown). Combining 10 nM staurosporine and 10 nM ET-l significantly augmented the stimulatory effect of ET-l on [Ca2+lf (Fig. 5), and increasing the concentration of staurosporine further to 100 nM caused a permanent increase in [Ca*+]f by 25%. When 1 PM diltiazem was added to ET-l and stauros-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 09:22 For personal use only. No other uses without permission. . All rights reserved.

ET-l-INDUCED

ANP

RELEASE

2459

l.J-

2.4.

=P 1.2.. B 2 !.I--

FIG. 2. The effect of ET-l on [Ca’+]rand ANP secretion by cultured ventricular myocytes. The duration of the infusion is shown by the horizontal bar. The values are expressed in proportion to the mean basal value during the first 20 min and as the mean + SE for five independent experiments. A and B, Vehicle; C and D, 100 nM ET-l. The variations in [Ca2+lr and ANP secretion during the infusion of ET-1 were different from those in the control experiments at the significance levels of P < 0.001 and P < 0.05, respectively (by ANOVA).

z ;

A

fj .k

l-O'r~+&)-~

2 ;: H

0.9..

f

' 0

10

20

30

I

f 'ii9 5 rF z4 2

20

15

proANP I

1.6..

4

:::::

f

;;p::

f

0.4-w

~-~/'~i-n~~/n-~~-~~-~

40

50

0.2 7 0

60

10

20

10

20

30

40

50

I 60

40

50

60

2.2.. = P x 2

1.2..

C

I.?-1.0,+0.9.0.6..

ANqm2S I

&

o.a--

5

0.6-m 0.4.. 0.2,

10

20

30

4

“0

a

2.4-

“0

10

.f

l.J-

lime

5

2.0.. 1.6..

0.7.l

0

0

2

0.6..

0.77

A

2.2..

25

"tot

;30

I

Fraction

FIG. 3. The molecular form of ANP released by atria1 myocytes. Protein-Pak I-125 gel filtration chromatograms of the original culture medium after 3 days of incubation (A) and pooled perifusates before (0) and during (0) the infusion of 10 nM ET-1 (B). The migration of standards is indicated in the figures. V,, Void volume; V,,, total volume. Typical experiments are shown.

40 (min)

50

60

0

30 lime

(min)

porine, the increase in [Ca2+lf was abolished (P < 0.05; Fig. 5). Diltiazem at a concentration of 1 pM had no significant effect on IR-ANP release induced by 10 nM ET-l (Fig. 5). Staurosporine at a concentration of 10 nM did not affect the initial ET-l-induced increase in IR-ANP release, but caused a decline in the secretion to the basal level in 40 min compared with the sustained effect of ET-l only (P < 0.05, by ANOVA; P < 0.05 after 25-40 min infusion of ET-l and staurosporine us. ET-l, by Newman-Keuls test). The pattern of IR-ANP secretion with 10 nM ET-l and staurosporine (Fig. 5) resembled the biphasic response induced by 100 nM ET-l (Fig. 1). Staurosporine did not completely prevent ET-induced IR-ANP release even if the concentration of staurosporine was increased to 100 nM or the infusion was started 20 min before the infusion of ET-l. The dose of 10 nM staurosporine also inhibited IR-ANP secretion induced by 1 nM ET-l as it did the secretion induced by 10 nM ET-l, but had no effect on IR-ANP secretion induced by 100 nM ET-l (data not shown). Diltiazem abolished the inhibitory effect of staurosporine on ET-induced IRANP release (Fig. 5). Because staurosporine and diltiazem failed to inhibit the initial rise in IR-ANP secretion, and arachidonic acid cascade has been shown to be involved in the effects of ET, the effect of indomethacin, a cyclooxygenease inhibitor (39) on ET-induced IR-ANP secretion was tested. Indomethacin (0.1 IIIM) did not inhibit ET-induced IRANP secretion, and calcium measurement was hampered

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 09:22 For personal use only. No other uses without permission. . All rights reserved.

ET-l-INDUCED

2460

ANP

RELEASE

Endo. Voll30.

1992 No 5

1.4-

F1c.4. The effects of diltiazem and staurosporine on [Ca*+]r and ANP secretion by cultured atria1 myocytes. The duration of the infusion is shown by the horizontal bar. The values are expressed in proportion to the mean basal value during the first 20 min and as the mean k SE for four independent experiments. A and B, 1 PM diltiazem; C and D, 10 nM staurosporine. The variations in [Ca*+]f and ANP secretion during the infusions of diltiazem and staurosporine were not different from those in the control experiments in Fig. 1 (by ANOVA).

0.s _I 0

10

20

30

40

50

60

0

1.4. = p 6

1.3 -.

c

B

10

20

30

40

50

60

2.0s 1.6..

x 2: !

1.6..

E

1.2..

0

1.4..

y5 0.8.1.0..

,wq$p~,*

/

+

5 E 0.6-b f

0.4..

9 0.8 -I 0

10

20

30 Time (min)

by the ability of indomethacin to absorb light at the wavelengths used in the experiments (data not shown). Discussion Atria1 wall stretch is a physiological stimulus of ANP release (40), but the mechanism of secretion is not known. Recent studies have shown that ET stimulates ANP secretion from the cardiac atrium. In the present experiments [Ca*+h and IR-ANP secretion of cultured myocardial cells were measured simultaneously during infusion of ET-l. ET-l increased [Ca*+h and was a potent stimulus of IR-ANP release in atria1 and ventricular neonatal myocytes. Staurosporine, a protein kinaseC inhibitor, inhibited sustained secretion of JR-ANP from atria1 cells induced by ET. Diltiazem, an inhibitor of voltage-sensitive calcium channels, did not affect ETinduced IR-ANP secretion, but abolished the staurosporine-induced inhibition of IR-ANP release. These results suggest that the effects of ET on ANP secretion in atria1 cells are dependent on protein kinase-C, whereas transplasmalemmal calcium influx is not necessarily required for ET-induced ANP release. ANP secretion from cultured myocytes was studied by a method that enabled us to measure [Caz+lf simultaneously with IR-ANP secretion (9). The basal values of [Ca*+h and IR-ANP secretion rate in the atria1 cells (146 nM and 61 pg/min . mg protein, respectively) are close to our previously reported values of 128 IIM and 88 pg/min . mg protein (9). ET-l caused a rapid rise in [Ca*+h in atria1 myocytes, which agrees with previous data on cultured cardiocytes (41) and freshly isolated myocytes (42). However, in isolated guinea pig heart cells, ET has

40

50

60

, ;I;‘. 1 0

10

20

4 30 Time (min)

40

50

60

been reported to decrease the amplitude of the Ca*+ current (43). The biphasic pattern of calcium changes resembles that reported in cultured neonatal ventricular myocytes (41) and freshly isolated neonatal atria1 myocytes (42). The maximal changes in [Ca*+]r were half of those reported by Hirata et al. (41) with the same concentrations of ET, probably because in the present experimental model the cells were not stimulated simultaneously (-90-set difference between the upper and lower ends of the column). ET-l was a potent stimulus of IRANP release from cultured atria1 myocytes in the same concentration range as that observed in isolated rat atria (25, 27), isolated perfused rat hearts (30), and cultured rat atria1 myocytes (12, 28). LDH activity was not found in the effluent, indicating that the increase in IR-ANP release was not caused by leakage from damaged cells. Peptides corresponding in size to pro-ANP and ANP(l-28) were detected in the culture and perifusion media. The difference in the molecular forms of the secreted peptides between cultured cells and intact organ has been observed previously (7,22) and may indicate that cardiac mesenchymal cells or extracellular processes are involved in the final maturation of ANP (1). The ratio of the peptides was not changed by stimulation with ET, as reported by Gardner et al. (22). Diltiazem, an inhibitor of L-type voltage-sensitive calcium channels (32), was used to test the involvement of calcium influx in ET-induced IR-ANP secretion. As a compound of benzothiazepine structure, it is nonfluorescent and, thus, more suitable for experiments involving calcium measurements than nifedipine, a dihydropyridine compound, which causes quenching of fluorescense

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 09:22 For personal use only. No other uses without permission. . All rights reserved.

ET-l-INDUCED

ANP

RELEASE

2461

f x 2 t ‘g

2.4 2.2.. 2.0.. 1.6.. 1.6..

f 5 1.2.. 1.4.. ‘Zr 1.0.. 0a? 0.8.. z

5. The effects of diltiazem and staurosporine on ET-l-induced changes in [Ca*+]r and ANP secretion by cultured atria1 myocytes. The duration of the infusion is shown by the horizontal bar. The values are expressed in proportion to the mean basal value during the first 20 min and as the mean f SE for five independent experiments. A and B, 10 nM ET-l; C and D, 10 nM ET-l plus 1 pM diltiazem; E and F, 10 nM ET-l plus 10 nM staurosporine; G and H, 10 nM ET-1 plus 10 nM staurosporine plus 1 pM diltiazem. The variations in [Ca*+]r and ANP secretion during the infusion of 10 nM ET-l plus 10 nM staurosporine were different from the variations in the experiments with 10 nM ET-1 at the levels of P < 0.001 and P < 0.05, respectively, whereas the variations in [Ca2+lr and ANP secretion during the infusion of 10 nM ET-1 plus 1 NM diltiazem and during the infusion of 10 nM ET-l, 10 nM staurosporine, and 1 pM diltiazem were not different from those in the experiments with 10 nM ET-l (by ANOVA).

B

/ Iwg

0.6.. 0.41

I 0

FIG.

1.20 6 %

I

II

0.90 4 0

10

20

30

40

50

60

1.15

: 0

1.10

E

f ‘Z _o 1.05 5 4 + “0 0

4b

1.00

0.6.. 0.4,

20

30

40

50

60

10

20

30

40

50

60

50

I60

50

+ 60

D

0

2.4.

‘.201 6

2

2.4. 2.2.. 2.0..

10

---H--q.

% d

2.2.. 2.0..

’ t ‘g

1.6-b 1.6..

5 f;n

1.2.. 1.4.. o.s.1.0..

2

0.6.. 0.4+

F

\

”

*‘+A*

/I

&-d-t\{

-h

by

0.95 I

0.90 I 0

’ 10

I

20

30

40

50

60

O

I

I

z 8 a 2 Y ‘3 $

1 10

20

30

40

Time (min)

at 340 nm, as shown for the dihydropyridine calcium agonist Bay K8644 in our previous study (9). Diltiazem inhibited the increase in [Ca2+lf caused by ET, which supports the view that voltage-sensitive calcium channels are involved in the actions of ET in atria1 cells. ET has been shown to mobilize calcium from caffeine- and ryanodine-insensitive intracellular pools and stimulate its entry through non-L-type calcium channels in dispersed neonatal rat atria1 cells (42), but a benzothiazepine interaction domain has not been found in non-Ltype calcium channels (44). Although diltiazem inhibited the ET-induced [Ca2+lf increase in cultured atria1 myocytes, it had no effect on ET-induced IR-ANP release,

50

60

20

30

40

2.4. 2.2.. 2.0.. 1.6..

H

1.6.. 1.4..

\

A---H

‘Z: :::.. Y 0.6.. 2 0.6.. 0.90 J 0

10

‘-$

_/’

0.47 0

10

20

30

40

Time (min)

showing that stimulated ANP secretion is not dependent on transplasmalemmal calcium influx. This supports our previous findings that the calcium channel agonist Bay K8644, ionomycin, and depolarization with KC1 have no effect on atria1 ANP secretion in the same experimental model (9). Our findings are also supported by the studies of Iida and Page (6) in quiescent cultured atria1 myocytes and Greenwald et al. (13) in suspensions of freshly isolated atria1 myocytes. However, in perfused rat heart (3, 4, lo), isolated atria (ll), and short term neonatal atria1 myocyte cultures (5, B), extracellular calcium affects ANP secretion. Enhanced calcium influx also appears to play a significant role in ET-induced ANP secretion from

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 09:22 For personal use only. No other uses without permission. . All rights reserved.

2462

ET-l-INDUCED

isolated atria (27) and in the sustained ET-induced ANP response from long term atria1 myocyte cultures (12). The difference between the results of Sei and Glembotski (12) and ours may be due to the different ages of the cultures, but it is not clear why their short term cultures are unresponsive to ET, while ours and those of Fukuda et al. (28) responded. Next, we studied the role of protein kinase-C in ETinduced ANP secretion using staurosporine, a potent and selective inhibitor of protein kinase-C with an inhibition constant in the nanomolar range (33). Staurosporine augmented ET-induced [Ca2+lf increment, but alone had no effect on calcium homeostasis. Because diltiazem abolished this effect of staurosporine, it is probable that staurosporine influences transplasmalemmal calcium fluxes. This agrees with the previous findings that protein kinase-C is able to activate the Ca*+ pump of the plasma membrane by phosphorylation (45). Staurosporine inhibited sustained IR-ANP secretion induced by ET-l, which supports the findings of Pitkanen et al. (46) in the perfused rat heart. The initial increase in ETinduced IR-ANP secretion was not inhibited by increasing the concentration of staurosporine to 100 nM, which is needed to abolish vasoconstrictor effects of TPA and partially inhibit ANP secretion induced by 2.3 nM ET-l in the perfused rat heart (46). Our results, thus, support the role of protein kinase-C in the regulation of ANP release, as has been reported in several in uitro models (3-9). There are differences in the calcium sensitivities of protein kinase-C subtypes in the myocardium (47), and at least in quiescent adult atria1 myocytes, phorbol ester-stimulated ANP secretion is independent of calcium (6). Surprisingly, diltiazem abolished the inhibitory effect of staurosporine on ET-induced IR-ANP secretion, although it had no effect on basal or ET-induced IRANP secretion. These results suggest that calcium influx may even have an inhibitory effect on stimulated ANP secretion, as has been shown for stretch-induced ANP secretion from the perfused rat heart (15). Intracellular calcium stores released by IP3, arachidonic acid cascade, Na+-H+ antiport, Na+-K+-ATPase, and membrane potential may also mediate intracellular effects of ET. The involvement of intracellular calcium stores in ET-induced ANP release from cultured myocytes has been suggested (12, 22). In the present study ET, especially at the concentration of 100 nM, caused a steep initial increase in [C!a2+lf, and increases in [Ca2+lf and ANP secretion during the first 10 min became steeper when the concentration of ET increased. Therefore, intracellular Ca2+ stores released by IP, may be involved in the initial phase of ANP secretion, whereas the staurosporine-sensitive pathway appears to regulate the sustained phase of ANP secretion. However, more work is required to clarify the role of intracellular calcium

ANP

RELEASE

Endo. Voll30.

1992 No 5

stores in ET-induced ANP release. Since 100 nM ET-l caused a biphasic secretion of IR-ANP, and staurosporine converted the response curve of 10 nM ET-l to one very similar to that with 100 nM ET-l alone, but did not significantly inhibit IR-ANP release induced by 100 nM ET, it is possible that the predominant signalling pathway may be dependent on the concentration of ET (48), and signalling may be directed through other pathways when protein kinase-C is inhibited. Prostaglandins F2a and E have been shown to promote ANP release from cultured neonatal atria1 myocytes (49). However, the finding that indomethacin had no effect on ET-induced IR-ANP secretion shows that arachidonic acid metabolites are not responsible for the observed secretory response. This supports the results by Gardner et al. (22) showing that meclophenemate, another inhibitor of cyclooxygenase, does not affect ET-induced ANP secretion. Application of other fluorescent intracellular probes to cell culture systems may help in studying the role of Na+-H+ antiport, Na+-K+-ATPase, and membrane potential in ET-induced ANP secretion. In addition to cardiac atria, the ANP gene is expressed in ventricles and several extracardiac tissues. Ventricular ANP expression is high during the fetal period, rapidly declines during neonatal life to nearly undetectable levels, and is increased if ventricular hypertrophy occurs (50). The basal [Ca’+]r of 167 nM in the ventricular cells is close to our previously reported value of 137 nM (9). The basal IR-ANP secretion rate of 65 pg/min. mg protein in ventricular cells is lower than the value of 116 pg/min . mg protein we reported previously (9). The ages of rats used in these two series of experiments were identical, but in the present study ventricular cultures were grown 3 days before perifusion experiments, while 2-day-old cultures were predominantly used in our previous study. This may explain the lower ventricular IRANP secretion, since ANP release rapidly declines during the neonatal period (51). Our results also show that ET1 causes a similar rise in [Ca2+lf in ventricular cells and atria1 cells and is able to stimulate ANP secretion from ventricular myocytes. This is surprising, since ventricles were originally suggested to secrete ANP constitutively without storage in granules (51), and ANP secretion of cultured ventricular myocytes does not respond to exogenous stimuli (7, 9). However, our finding agrees with the recent data showing that ET increases ANP secretion from cultured ventricular myocytes (21,22), and phorbol esters increase ANP secretion from hypertrophic heart ventricles (52). In conclusion, we have for the first time measured changes in [Ca2+lf during ET-induced IR-ANP secretion. ET-l was a potent secretagogue of ANP release from both atria1 and ventricular myocytes. ET-induced sustained ANP secretion from atria1 myocytes seemed to be

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 09:22 For personal use only. No other uses without permission. . All rights reserved.

ET-l-INDUCED

ANP

dependent on protein kinase-C, whereas transplasmalemma1 calcium influx was not required for ANP release. Acknowledgments We thank Ms. tonen for technical

Minna Hietala assistance.

and

Mrs.

Maija-Leena

Leh-

References 1. Needleman P, Blaine EH, Greenwald JE, Michener ML, Saper CB, Stockmann PT, Tolunay E 1989 The biochemical pharmacology of atria1 peptides. Annu Rev Pharmacol Toxic01 29:23-54 H, Kinnunen P, Mdntymaa P, Uusimaa P, Taskinen T, 2. Ruskoaho Vuolteenaho 0 Lenndluoto J 1991 Cellular signals regulating the release of ANF. Can J Physiol Pharmacol69: