Regulation Induced by Crosstalk Between Endothelin-1 and
Norepinephrine in Dog Ventricular Myocardium
Li Chu, Reiko Takahashi, Ikuo Norota, Takuya Miyamoto, Yasuchika Takeishi, Kuniaki Ishii,
Isao Kubota, Masao Endoh
Abstract—In certain cardiovascular disorders, such as congestive heart failure and ischemic heart disease, several endogenous regulators, including norepinephrine (NE) and endothelin-1 (ET-1), are released from various types of cell. Because plasma levels of these regulators are elevated, it seems likely that cardiac contraction might be regulated by crosstalk among these endogenous regulators. We studied the regulation of cardiac contractile function by crosstalk between ET-1 and NE and its relationship to Ca2⫹signaling in canine ventricular myocardium. ET-1 alone did not affect the contractile function. However, in the presence of NE at subthreshold concentrations (0.1 to 1 nmol/L), ET-1 had a positive inotropic effect (PIE). In the presence of NE at higher concentrations (100 to 1000 nmol/L), ET-1 had a negative inotropic effect. ET-1 had a biphasic inotropic effect in the presence of NE at an intermediate concentration (10 nmol/L). The PIE of ET-1 was associated with an increase in myofilament sensitivity to Ca2⫹ions and a small increase in Ca2⫹ transients, which required the simultaneous activation of protein kinase A (PKA) and PKC. ET-1 elicited translocation of PKC⑀ from cytosolic to membranous fraction, which was inhibited by the PKC inhibitor GF 109203X. Whereas the Na⫹-H⫹ exchange inhibitor Hoe 642 suppressed partially the PIE of ET-1, detectable alteration of pHidid not occur during application of ET-1 and NE. The negative inotropic effect of ET-1 was associated with a pronounced decrease in Ca2⫹ transients, which was mediated by pertussis toxin–sensitive G proteins, activation of protein kinase G, and phosphatases. When the inhibitory pathway was suppressed, ET-1 had a PIE even in the absence of NE. Our results indicate that the myocardial contractility is regulated either positively or negatively by crosstalk between ET-1 and NE through different signaling pathways whose activation depends on the concentration of NE in the dog. (Circ Res. 2003; 92:1024-1032.)
Key Words: endothelin-1 䡲 norepinephrine 䡲 myocardial contractility 䡲 Ca2⫹ transients 䡲 protein kinase C
D
uring the course of cardiovascular disorders, such as congestive heart failure and ischemic heart disease, plasma levels of both endothelin-1 (ET-1) and norepinephrine (NE) tend to increase.1– 4The signal transduction processesthat are triggered by the activation of receptors for these endogenous agonists are different, and, thus, it seems likely that crosstalk between ET-1 and NE might play a critical role in the regulation of cardiac function, determining hemody-namic responses to antagonists of-adrenoceptors or endo-thelin receptors under various pathophysiological conditions. The available evidence implies that these endogenous regu-lators are engaged in crosstalk at different levels of their respective signaling pathways. For example, the positive feedback mechanism seems to exist at the level of the synthesis of NE by which ET-1 increases the plasma concen-tration of NE,5 whereas NE facilitates the expression of
mRNA that encodes the prepro-ET-16and the production of
ET-1.7
ET-1 has a positive inotropic effect (PIE) in ventricular myocardium of most mammals, but it has no inotropic effect on canine ventricular myocardium.8,9By contrast, ET-1 has a
negative inotropic effect (NIE) in the presence of catechol-amines and antagonizes the-adrenoceptor–mediated facili-tatory regulation of contractile function in several mamma-lian species, including the dog.10 –14Regulation of myocardial
contractility induced by crosstalk between ET-1 and NE has not been studied in detail. It has been reported that the acceleration of the hydrolysis of phosphoinositide and the subsequent generation of inositol 1,4,5-trisphosphate and 1,2-diacylglycerol might be responsible for the PIE of ET-1 in certain species,8but the subcellular mechanism involved in
the NIE of ET-1 has not been fully elucidated. We designed
Original received February 18, 2002; resubmission received February 14, 2003; revised resubmission received March 24, 2003; accepted March 27, 2003.
From the Department of Pharmacology (L.C., R.T., I.N., K.I., M.E.) and the First Department of Internal Medicine (T.M., Y.T., I.K.), Yamagata University School of Medicine, Yamagata, Japan.
This manuscript was sent to Richard A. Walsh, Consulting Editor, for review by expert referees, editorial decision, and final disposition.
Correspondence to Masao Endoh, MD, PhD, Department of Pharmacology, Yamagata University School of Medicine, 2-2-2 Iida-nishi, Yamagata 990-9585, Japan. E-mail mendou@med.id.yamagata-u.ac.jp
© 2003 American Heart Association, Inc.
Circulation Research is available at http://www.circresaha.org DOI: 10.1161/01.RES.0000070595.10196.CF
1024
um. We examined subcellular mechanisms responsible for such regulation using selective inhibitors of protein kinases and other types of enzyme. ET-1 had a PIE and NIE, depending on the concentration of NE present before admin-istration of ET-1. The effects were mediated by an increase in the myofilament sensitivity to Ca2⫹ or a decrease in Ca2⫹ transients, which were induced by activation of different signaling pathways in dog ventricular myocardium. Prelimi-nary accounts of this study have been published elsewhere.13,15–18
Materials and Methods
All manipulations of animals were performed in accordance with the Guide for Animal Experimentation, Yamagata University School of Medicine, and Japanese Governmental Law (No. 105). Approval for all experiments with animals was obtained from the Committee for Animal Experimentation, Yamagata University School of Medicine, before the experiments, and the study was also carried out in accordance with the Helsinki Declaration. Mongrel dogs (7 to 10 kg) of both sexes were used in these experiments, which were performed as described previously.11,14,19
carneae of the right ventricular wall (⬍1 mm in diameter) were isolated and mounted in 20-mL organ baths that contained Krebs-Henseleit solution.8,11 The ventricular trabeculae were stimulated
electrically with square-wave pulses of 5-ms duration and a voltage that was 20% above the threshold (⬇0.4 V) at a frequency of 0.5 Hz. The average length of muscle preparations was 7.39⫾0.52 mm, and the average cross-sectional area was 1.43⫾0.19 mm2(n⫽165, from
a total of 73 dogs).
ET-1 was administered at a single concentration to each muscle preparation. Selective inhibitors were administered 20 to 30 minutes before the addition of ET-1 and were present in the organ bath throughout respective experiments. Pertussis toxin (PTX) at 0.5
g/mL was allowed to act for 10 hours before experiments were
started.
Preparation and Analysis of Canine Ventricular Myocytes
A portion of the free wall of the left ventricle that is supplied via a branch of the left anterior descending artery was excised. The artery was cannulated and perfused with Tyrode’s solution that contained 1.0 mg/mL collagenase and 0.1 mg/mL protease via a recirculating system for 15 to 25 minutes at room temperature (24°C). Then the muscle was perfused with Tyrode’s solution that contained 0.2 mmol/L CaCl2 and cut into small pieces ⬇3⫻3 mm2 with a
Figure 1. Inotropic effects of ET-1 in the absence and presence of NE at different concentrations in isolated canine ventric-ular trabeculae. A and B, PIE of 10 and 100 nmol/L ET-1 in the presence of 1 nmol/L NE. C and D, Inotropic effects of 10 nmol/L ET-1 and their dependence on the concentration of NE applied before ET-1 (C refers to control; no NE applied before ET-1). A, Actual tracings. B, C, and D, Summary of data. Values are mean⫾SE. Where SE is not shown, it is smaller than the symbol. Numbers in parentheses indicate the numbers of preparations examined. ***P⬍0.001 vs the force before the addition of ET-1.
scalpel. The resultant cells in suspension were rinsed several times with Tyrode’s solution that contained gradually increasing concen-trations of Ca2⫹up to 1.8 mmol/L.
Procedures used for loading of indo-1, superfusion of myocytes, measurements of fluorescence, and cell length are presented in detail in online data supplement, available at http://www.circresaha.org.
Subcellular Localization of Protein Kinase C Isoforms
The subcellular fractionation procedures, antibodies, and Western blotting techniques are described in the online data supplement.
Statistical Analysis
Experimental values are presented as mean⫾SE. Significant differ-ences between mean values were estimated by a repeated-measures ANOVA or by Student’s t test with analytic software STATVIEW J-4.5 (Abacus Concepts). P⬍0.05 was judged to indicate a signifi-cant difference.
Results
Influence of NE on the Inotropic Effects of ET-1
Endothelin-1 at 1 to 100 nmol/L did not, by itself, affect the peak twitch force. However, in the presence of NE at the subthreshold concentration of 1 nmol/L, ET-1 at 10 and 100 nmol/L had a definite PIE in association with negative lusitropic and clinotropic (the effect on time to peak tension) effects (data not shown), and these effects of ET-1 were concentration-dependent (Figures 1A and 1B). The EC50
value for ET-1 in the presence of NE at 1 nmol/L was
34.0⫾5.90 nmol/L (determined in 37 preparations from 14 dogs, including data presented in Reference 15).
NE had a concentration-dependent PIE at 10, 100, and
1000 nmol/L equivalent to 11.5⫾4.1% (n⫽5), 80.4⫾8.4%
(n⫽5), and 248.3⫾27.4% (n⫽8) of the basal force, respec-tively, and the threshold concentration and EC50value were 3
nmol/L and 0.87⫾0.09mol/L, respectively. When the NE concentration before the administration of ET-1 was in-creased, the PIE of ET-1 was converted to a NIE, depending on the concentration of NE. In the presence of 10 nmol/L NE, 10 nmol/L ET-1 induced a biphasic inotropic response (ie, a transient NIE followed by a long-lasting PIE); in the presence of NE at higher concentrations (ⱖ100 nmol/L), ET-1 had a definite NIE (Figure 1C). In the presence of NE at 1000 nmol/L, the NIE of ET-1 was markedly reduced, and ET-1 did not have any inotropic effect in the presence of 10mol/L NE (Figure 1D).
Regulation of Ca2ⴙ Signaling by Crosstalk
Between ET-1 and NE
In canine ventricular myocytes, neither ET-1 (10 nmol/L) nor NE (0.1 and 1 nmol/L) by itself affected the cell shortening and Ca2⫹
transients (Figure 2A). When 10 nmol/L ET-1 was administered in the presence of 0.1 nmol/L NE, ET-1 induced an increase in cell shortening (Figures 2A and 2B, bottom) in association with a small increase in Ca2⫹transients (Figure 2B, top).
The increase in cell shortening induced by 10 nmol/L ET-1 was equivalent to that produced by an increase in
extracellu-Figure 2. Increases in cell shortening and in the ratio of indo-1 fluorescence at 405 and 500 nm in response to ET-1 in the presence of NE at a subthreshold concentration in isolated canine ventricu-lar myocytes. A, Actual tracings of the effects of 10 nmol/L ET-1 in the absence (top) and presence (bottom) of 0.1 nmol/L NE. B, Individual signals recorded at times a through c in A (bottom trac-ings). Individual tracings were obtained by averaging of 5 successive signals. Top, indo-1 fluorescence ratio; bottom, cell shortening. C, Relationship between the maximum ratio of indo-1 fluores-cence and maximum shortening in response to ET-1 in the presence of 0.1 nmol/L NE compared with the effects of an increase in [Ca2⫹]oto 3.6 mmol/L. Basal indicates basal value
([Ca2⫹]o⫽1.8 mmol/L) before administra-tion of drugs or elevaadministra-tion of [Ca2⫹]o. Numbers in parentheses indicate the numbers of cells; vertical and horizontal bars indicate SEM.
lar Ca2⫹concentration ([Ca2⫹]
o) to 3.6 mmol/L. However, the
increase in Ca2⫹transients induced by ET-1 was significantly (P⬍0.05) smaller than that induced by [Ca2⫹]
oof 3.6 mmol/L
(Figure 2C), an indication that the increase in cell shortening induced by ET-1 was attributable, at least in part, to an increase in the myofilament sensitivity to Ca2⫹.
NE at 100 nmol/L induced a pronounced increase in the maximum cell shortening together with a remarkable increase in Ca2⫹transients. ET-1 markedly decreased the NE-induced increase in cell shortening and Ca2⫹transients (Figures 3A and 3B). The NE-induced increases in cell shortening and Ca2⫹ transients were inhibited by ET-1 to an essentially similar extent (Figure 3C), and the relationship between the amplitudes of cell shortening and Ca2⫹ transients that was observed with NE alone was unaffected by ET-1 (data not shown).
In the presence of NE at an intermediate concentration of 10 nmol/L, ET-1 had a biphasic effect, inducing a transient decrease in cell shortening that was associated with a de-crease in Ca2⫹transients, followed by a long-lasting increase in cell shortening that was associated with a statistically insignificant alteration of Ca2⫹ transients (Figures 4A and 4B). Summary of these data are presented in Figure 4C. Our findings indicate that, in the presence of 10 nmol/L NE, the inotropic response to ET-1 involves a combination of facili-tatory and inhibitory effects.
Similar results were obtained in aequorin-loaded canine right ventricular trabeculae (see the online data supplement).
Signal Transduction Pathway for the ET-1–Induced PIE
Figure 5 shows the effects of selective inhibitors of cAMP-mediated and protein kinase C–cAMP-mediated (PKC-cAMP-mediated) pathways on the PIE of ET-1 in the presence of a subthresh-old concentration of 1 nmol/L NE. The PIE of ET-1 was abolished by treatment of trabeculae with timolol (1mol/L), which blocks-adrenoceptors, and with H-89 (1 mol/L), an inhibitor of protein kinase A (PKA; Figure 5A). These inhibitors, at the concentrations used, did not affect the PIE induced by an elevation of [Ca2⫹]
o(data not shown).
The inhibitors of PKC, staurosporine (10 nmol/L) and H-7 (10 mol/L), and an inhibitor of phospholipase C (PLC), neomycin (10 mol/L), abolished the PIE of ET-1 in the presence of 1 nmol/L NE (Figure 5A). These selective inhibitors, at the concentrations used, did not affect the PIE of NE that was induced by activation of-adrenoceptors (data not shown). Carbachol (0.1 mol/L), which selectively in-hibits the cAMP-mediated PIE,20 also reversed the PIE of
ET-1 (Figure 5B). These results indicate that the PIE of ET-1 in the presence of NE requires the simultaneous activation of PKA and PKC signaling pathways.
Figure 3. Decreases in cell shortening and indo-1 fluorescence ratio in response to ET-1 in the presence of 100 nmol/L NE in isolated canine ventricular myocytes. A, Actual tracings of the effects of 10 nmol/L ET-1 in the presence of 100 nmol/L NE. B, Individual signals recorded at times a through c in A. Individual tracings were obtained by averaging of 5 successive signals. Top, indo-1 fluorescence ratio; bottom, cell shortening. C, Summary of data in A and B. Basal indicates baseline Ca2⫹transients and cell shortening before administration of drugs. Numbers in parentheses indicate numbers of cells. ***P⬍0.001 vs 100 nmol/L NE alone.
Signal Transduction Pathway for the ET-1–Induced NIE
At concentrations that completely inhibited the PIE of 10 nmol/L ET-1, the inhibitors of PKC and PLC did not suppress but, in fact, enhanced the NIE of ET-1 that was induced in the presence of 100 nmol/L NE (Figure 6A).
Prior treatment with PTX (0.5 g/mL) and the treatment with LY83583 (10mol/L), an inhibitor of guanylyl cyclase
(GC), with KT5823 (0.3 mol/L), an inhibitor of
cGMP-dependent protein kinase (PKG), or with cantharidin (10
mol/L), an inhibitor of protein phosphatase (PP), almost
completely suppressed the NIE of ET-1 in the presence of 100 nmol/L NE (Figure 6B). These selective inhibitors had no effects on the basal force and on the PIE of NE mediated by
-adrenoceptors (data not shown).
Unmasking of the PIE of ET-1 by Suppression of Inhibitory Pathways
After the pretreatment of trabeculae with PTX (Figure 7A) and the treatment with KT5823 (Figure 7B) or with canthar-idin (Figure 7C), ET-1 (10 nmol/L) had a PIE even in the absence of NE.
Influence of GF 109203X and Hoe 642 on PKC⑀ Translocation
We had examined the subcellular distribution of 4 major PKC isoforms (␣, , ␦, and ⑀) by immunoblotting with the use of isoform-specific antibodies. We found that the dog right ventricle expressed␣ and ⑀ isoforms, whereas no significant immunoreactivity was detected for  and ␦. Because the subcellular localization of the PKC␣ isoform did not change
Figure 4. Biphasic effects of 10 nmol/L ET-1 in the presence of 10 nmol/L NE in isolated canine ventricular myocytes. A, Actual tracings of the effects of ET-1 at 10 nmol/L in the presence of 10 nmol/L NE. B, Individual signals recorded at times a through d in A. Individual trac-ings were obtained by averaging of 5 successive signals. Top, indo-1 fluores-cence ratio; bottom, cell shortening. C, Summary of data in A and B. Basal indi-cates baseline Ca2⫹transients and cell shortening before administration of drugs. Numbers in parentheses indicate numbers of cells. *P⬍0.05; ***P⬍0.001 vs 10 nmol/L NE alone.
in response to pharmacological stimuli, we reported data for the PKC⑀ isoform in the present study.
Representative immunoblots of the PKC⑀ isoform are
shown in Figure 8A. The membrane-associated immunoreac-tivity was markedly increased in response to phorbol
dibu-tyrate and ET-1⫹NE. The translocation of PKC⑀ to the
membranous fraction by ET-1⫹NE was completely blocked
by GF 109203X (a selective PKC inhibitor, 1mol/L) but not by Hoe 642 (a Na⫹/H⫹ inhibitor, 1 mol/L) (Figure 8B). Under the same experimental condition, Hoe 642 suppressed significantly (P⬍0.05) the PIE induced by combination of
ET-1 with NE by ⬇40% (control, 151⫾5.6%; Hoe 642,
132⫾4.8%; n⫽4 each).
Discussion
In the present study, we demonstrated that the extent and quality of the inotropic effects of ET-1 in canine ventricular myocardium are determined by the concentration of simulta-neously applied NE, with the functional outcome of coupling subsequent to stimulation of ET receptors being dependent,
apparently, on the extent of -stimulation by NE. The
regulation of contraction induced by ET-1 involves activation of PKC, of a Gsprotein– coupled cAMP/PKA pathway, and
of a PTX-sensitive Gi protein– coupled cGMP/PKG/PP
pathway.
By itself, ET-1 did not induce any inotropic response, but it had a definite PIE in the presence of NE at 0.1 to 1 nmol/L, which did not significantly affect the basal force. The PIE of ET-1 was converted to a NIE when the concentration of NE was increased. ET-1 had a prominent NIE in the presence of NE at 100 nmol/L and higher, and these observations are essentially consistent with previous reports that ET-1 atten-uated the PIE and positive chronotropic effect of
-stimulation in the dog,11rat,21and guinea pig.22Our results
indicate that in the presence of NE, positive and negative inotropic responses compete with one another and that, in the presence of NE at high concentrations, the PIE of ET-1 is completely replaced by a NIE. This crosstalk between ET-1 and NE might contribute significantly to the variable inotro-pic responses to ET-1; in previous studies, ET-1 had no effect,8,9,23 a PIE,4,8 and a NIE,24 –26 including a
time-dependent component to these phenomena.27
Differential Regulation of Ca2ⴙSignaling by ET-1
The PIE of ET-1 in the presence of a threshold concentration of NE was associated with a small increase in Ca2⫹transients.
on the PIE of ET-1 in isolated canine ventricular myocardium. A, Effects of treatment with timolol (1mol/L), H-89 (1 mol/L), staurosporine (STS; 10 nmol/L), H-7 (10mol/L), and neomycin (NEO; 10 mol/L) individually on the PIE of ET-1 in the presence of 1 nmol/L NE. ***P⬍0.001 vs 1 nmol/L NE alone. B, Effects of car-bachol (CCh; 0.1mol/L) on the PIE of ET-1 in the presence of 1 nmol/L NE. Numbers in parentheses indicate num-bers of preparations examined. ***P⬍0.001 vs 1 nmol/L NE plus 10 nmol/L ET-1.
Figure 6. Effects of inhibitors of PKC and Gipathways on the NIE of ET-1 in iso-lated canine ventricular myocardium. A, Effects of treatment with staurosporine (STS; 10 nmol/L), H-7 (10mol/L), and neomycin (NEO; 10mol/L) individually on the NIE of ET-1 in the presence of 100 nmol/L NE. B, Effects of prior treat-ment with PTX (0.5g/mL), LY83583 (LY; 10mol/L), KT5823 (KT; 0.3 mol/L), and cantharidin (Cant; 10mol/L) individu-ally on the NIE of ET-1. The force of con-traction before the addition of ET-1 was taken as 100% for each preparation, and changes in the force recorded 20 minutes after the application of ET-1 are expressed as a percentage relative to the maximum force. Numbers in columns indicate num-bers of preparations examined. ***P⬍0.001 vs 100 nmol/L NE alone.
This increase was significantly smaller than that induced by an increase in [Ca2⫹]
othat elicited a PIE equivalent to that of
ET-1 (Figure 2). These observations imply that the PIE of ET-1 is associated definitively with an increase in the myofilament sensitivity to Ca2⫹, being consistent with
previ-ous findings with ET-1 in other species.19,28 –30Because ET-1
did not have a PIE in the presence of inotropic interventions, such as dihydroouabain or an increase in [Ca2⫹]
o,13 ET-1
might require a weak-stimulation for induction of its PIE. This synergistic action of ET-1 and NE indicates that there is a critical difference in the regulation between the dog and other mammals.8,19,29 –31 In mice, -stimulation and ET-1
regulate cardiac contractility in opposite directions in part through phosphorylation of troponin I on distinct sites,32
indicating that the phosphorylation of contractile proteins plays a crucial role in the regulation, the role of which could not be determined in the present study.
In contrast to the PIE, the NIE of ET-1 was accompanied by a pronounced decrease in Ca2⫹transients (Figure 3), which may play a key role in the NIE of ET-1. We investigated in dog ventricular myocytes that ET-1 inhibited significantly the increase in L-type Ca2⫹ current (I
Ca) induced by
isoprotere-nol.14 Because the inhibitory action of ET-1 on the
isoproterenol-induced increase in ICa was suppressed by the
treatment with PTX in rabbit myocytes, the inhibition of the cAMP-mediated increase in ICavia the PTX-sensitive
inhib-itory pathway activated by ET-1 might contribute, to some
extent, to the ET-1–induced decrease in Ca2⫹ transients. Involvement of effects on other processes, such as PKA and SR Ca2⫹release, however, is not excluded.
Present observations with indo-1–loaded myocytes are consistent with findings obtained with aequorin-loaded dog ventricular trabeculae (online data supplement).18
Subcellular Mechanisms for the PIE of ET-1
Neomycin, staurosporine, and H-7, at a concentration that did
not affect the PIE of NE mediated by -adrenoceptors,
abolished the PIE of ET-1. Furthermore, timolol and H-89 completely suppressed the PIE of ET-1 (Figure 5A), whereas carbachol reversed the PIE of ET-1 (Figure 5B). These observations together imply that the ET-1–induced PIE and increase in the myofilament sensitivity to Ca2⫹ require the
simultaneous activation of PKC and PKA.
Although the Na⫹-H⫹exchange inhibitor Hoe 642 inhib-ited partially the PIE of ET-1, we could not detect an appreciable alteration of pHiin single myocytes loaded with
the fluorescent pHiprobe SNARF-1 (online data supplement).
We found recently that the Ca2⫹-sensitizing actions of OR-1896 and levosimendan were abolished by carba-chol,33–35 an indication that a Ca2⫹-sensitizing mechanism might exist that requires accumulation of cAMP. Although myosin-binding protein C, which is phosphorylated via a cAMP/PKA signaling pathway, might be a candidate for the source of increased sensitivity to Ca2⫹,36the target proteins
Figure 7. Effects of prior treatment with PTX (0.5g/mL) and treatment with KT5823 (KT; 0.3mol/L) and cantharidin (Cant; 30mol/L) individually on the ino-tropic effects of ET-1 in the absence of NE in isolated canine ventricular trabecu-lae. The basal force before the addition of ET-1 was taken as 100% for each preparation, and changes in the force recorded after the application of ET-1 are expressed as a percentage relative to the basal force. Numbers in parenthe-ses indicate numbers of preparations. *P⬍0.05, **P⬍0.01, ***P⬍0.001 vs 10 nmol/L ET-1 alone.
Figure 8. A, Representative immunoblots of the PKC⑀ isoform in canine ventricular trabeculae. Positions of the molecular mass markers (in kDa) are indicated on the left. B, Quantitative group data for PKC⑀ translocation. Data are reported as mean⫾SEM and were obtained from 5 separate experiments. *P⬍0.01 vs con-trol, #P⬍0.01 vs ET-1⫹NE. Details in the experimental procedures are described in the online data supplement.
on the ␣- and -mediated PIE,37,38 which is discussed in
detail in online data supplement.
Treatment with cantharidin and with KT5823 and pretreat-ment with PTX unmasked the PIE of ET-1 even in the absence of NE. These findings indicate that the signaling process that leads to activation of Gi-proteins, PKG, and
phosphatases might be highly effective even in the baseline state, counteracting the Gq-/Gs-mediated PIE via suppression
of cAMP-mediated signaling and leading to the absence of a PIE of ET-1; ie, ET-1 stimulates both Gq and Gi proteins,
accounting for the activation of facilitative and inhibitory (Figure 7).
Subcellular Mechanisms for the NIE of ET-1
The selective inhibitors that abolished the PIE enhanced the NIE of ET-1 (Figure 6), an indication that the NIE of ET-1 is mediated by signaling processes that are different from those involved in the induction of the PIE of ET-1. Because the NIE of ET-1 was almost completely inhibited by pretreatment with PTX, the accumulation of cAMP induced by
-stimulation might be suppressed by the Gi-mediated
deac-tivation of adenylyl cyclase. However, this scenario is un-likely, because the NIE of ET-1 is not accompanied by a significant reduction in the accumulation of cAMP that is mediated by-stimulation.10,11,39Although
2-adrenoceptors
that are coupled to Gi proteins are activated by NE40 and
could come into play in the NIE of ET-1 in the presence of high concentrations of NE, the findings with ET-1 in the presence of various -adrenoceptor agonists, including the
2-selective agonist zinterol, exclude the essential role of
2-subtype in the NIE of ET-141(online data supplement).
Both LY83583, an inhibitor of GC, and KT5823, an inhibitor of PKG, suppressed the NIE of ET-1, an indication that Gi-coupled cGMP/PKG signaling might be responsible
for the inhibitory action of ET-1, whereas nitric oxide is not involved in such regulation.10,11 It has been reported that
Gi-coupled receptors, such as muscarinic M242and adenosine
A1receptors,43counteract the effect of activation of PKA, in
part via the activation of PP: the muscarinic agonists acetyl-choline and carbachol inhibit protein phosphatase inhibitor-1 (PPI-1) through Gi-mediated stimulation of the PP activity
with resultant dephosphorylation of a variety of functional proteins that are phosphorylated by PKA.44 – 47 The
musca-rinic inhibition could occur without a concomitant decrease in the cAMP content or in the PKA activity.45,46Our observation
that an inhibitor of PP, cantharidin, abolished the NIE of ET-1 is consistent with the results of previous investigations described above. In this context, it is noteworthy that the NIE of ET-1 is more susceptible to cantharidin than the NIE of carbachol and the PIE of NE mediated by-adrenoceptors,48
an indication that the differential effect of cantharidin on various signaling processes is exerted in dog ventricular myocardium.
In summary, ET-1 has a positive effect, biphasic effect, or NIE, depending on the extent of -stimulation in canine
phosphatases via Gi-coupled cGMP/PKG signaling. The PIE
was associated with an increase in the myofilament sensitiv-ity to Ca2⫹, whereas the NIE was attributable to a decrease in Ca2⫹ transients. The crosstalk between ET-1 and NE might play a crucial role in the regulation of myocardial contractil-ity under pathophysiological conditions that are associated with elevated plasma levels of endogenous regulators.
Acknowledgments
This work was supported in part by Grants-in-Aid for Scientific Research (B) from the Ministry of Education, Culture, Sports, Science and Technology, Japan, and by a Research Grant for Cardiovascular Disease (11-1) from the Ministry of Health and Welfare, Japan.
References
1. Minami M, Yasuda H, Yamazaki N, Kojima S, Nishijima H, Matsumura N, Togashi H, Koike Y, Saito H. Plasma norepinephrine concentration and plasma dopamine--hydroxylase activity in patients with congestive heart failure. Circulation. 1983;67:1324 –1329.
2. Wei CM, Lerman A, Rodeheffer RJ, McGregor CG, Brandt RR, Wright S, Heublein DM, Kao PC, Edwards WD, Burnet JC Jr. Endothelin in human congestive heart failure. Circulation. 1994;89:1580 –1586. 3. McMurray JJ, Ray SG, Abdullah I, Dargie HJ, Morton JJ. Plasma
endo-thelin in chronic heart failure. Circulation. 1992;85:1374 –1379. 4. Sakai S, Miyauchi T, Sakurai T, Kasuya T, Ihara M, Yamaguchi I, Goto
K, Sugishita Y. Endogenous endothelin-1 participates in the maintenance of cardiac function in rat with congestive heart failure: marked increase in endothelin-1 production in the failing heart. Circulation. 1996;93: 1214 –1222.
5. Miller WL, Redfield MM, Burnett JC. Integrated cardiac, renal, and endocrine action of endothelin. J Clin Invest. 1989;83:317–320. 6. Masaki T, Kimura S, Yanagisawa M, Goto K. Molecular and cellular
mechanism of endothelin regulation: implication for vascular function.
Circulation. 1991;84:1457–1468.
7. Yanagisawa M, Kurihara H, Kimura S, Tomobe Y, Kobayashi M, Mitsui Y, Yazaki Y, Goto K, Masaki T. A novel potent vasoconstrictor peptide produced by vascular endothelial cells. Nature. 1988;332:411– 415. 8. Takanashi M, Endoh M. Characterization of the positive inotropic effect
of endothelin on mammalian ventricular myocardium. Am J Physiol. 1991;261:H611–H619.
9. Banyasz T, Magyar J, Kortvely A, Szigeti G, Szigligeti P, Papp Z, Mohacsi A, Kovacs L, Nanasi PP. Different effects of endothelin-1 on calcium and potassium currents in canine ventricular cells. Naunyn
Schmiedebergs Arch Pharmacol. 2001;363:383–390.
10. Zhu Y, Yang HT, Endoh M. Does nitric oxide contribute to the negative chronotropic and inotropic effects of endothelin-1 in the heart? Eur
J Pharmacol. 1997;332:195–199.
11. Zhu Y, Yang HT, Endoh M. Negative chronotropic and inotropic effects of endothelin isopeptides in mammalian cardiac muscle. Am J Physiol. 1997;273:H119 –H127.
12. Watanabe T, Endoh M. Characterization of the endothelin-1 induced regulation of L-type Ca2⫹current in rabbit ventricular myocytes. Naunyn Schmiedebergs Arch Pharmacol. 1999;360:654 – 664.
13. Chu L, Endoh M. Biphasic inotropic effects of endothelin-1 in the presence of sympathomimetic drugs at different concentrations in canine ventricular myocardium. Jpn J Pharmacol. 2000;82(suppl I):198P. Abstract.
14. Watanabe T, Endoh M. Antiadrenergic effects of endothelin-1 on L-type Ca2⫹current in canine ventricular myocytes. J Cardiovasc Pharmacol.
2000;36:344 –350.
15. Chu L, Endoh M. Biphasic inotropic response to endothelin-1 in the presence of various concentrations of norepinephrine in dog ventricular myocardium. J Cardiovasc Pharmacol. 2000;36(suppl 2):S9 –S14. 16. Endoh M, Chu L, Takahashi R, Norota I. Regulation of Ca2⫹signaling by
cross talk of endothelin-1 and norepinephrine. Jpn J Pharmacol. 2001; 85(suppl I):20P. Abstract.
17. Chu L, Endoh M. Regulation of Ca2⫹signaling induced by endothelin-1
in indo-1 loaded canine ventricular myocytes. Jpn J Pharmacol. 2001; 85(suppl I):83P. Abstract.
18. Takahashi R, Chu L, Endoh M. Dual inotropic responses to endothelin-1 in the presence of high or low concentrations of norepinephrine involve differential regulation of Ca2⫹signaling in aequorin-loaded canine right
ventricular myocardium. Jpn J Pharmacol. 2001;85(suppl I):81P. Abstract.
19. Yang HT, Sakurai K, Sugawara H, Watanabe T, Norota I, Endoh M. Role of Na⫹/Ca2⫹exchange in endothelin-1-induced increases in Ca2⫹transient
and contractility in rabbit ventricular myocytes: pharmacological analysis with KB-R7943. Br J Pharmacol. 1999;126:1785–1795.
20. Endoh M. Correlation of cyclic AMP and cyclic GMP levels with changes in contractile force of dog ventricular myocardium during cholinergic antagonism of positive inotropic actions of histamine, glucagon, theoph-ylline and papaverine. Jpn J Pharmacol. 1979;29:855– 864.
21. Reid JJ, Lieu AT, Rand MJ. Interactions between endothelin-1 and other chronotropic agents in rat isolated atria. Eur J Pharmacol. 1991;194: 173–181.
22. Reid JJ, Wong-Dusting HK, Rand MJ. The effect of endothelin on noradrenergic transmission in rat and guinea-pig atria. Eur J Pharmacol. 1989;168:93–96.
23. Volkmann R, Bokvist K, Wennmalm A. Endothelin has no positive inotropic effect in guinea-pig atria or papillary muscle. Acta Physiol
Scand. 1990;138:345–348.
24. Teerlink JR, Loffler BM, Hess P, Maire JP, Clozel M, Clozel JP. Role of endothelin in the maintenance of blood pressure in conscious rats with chronic heart failure: acute effects of the endothelin receptor antagonist Ro 47-0203 (Bosentan). Circulation. 1994;90:2510 –2518.
25. Shimoyama H, Sabbah HN, Borzak S, Tanimura M, Shevlygin S, Scicli G, Goldstein S. Short-term hemodynamic effects of endothelin receptor blockade in canines with chronic heart failure. Circulation. 1996;94: 779 –784.
26. Spinale FG, Walker JD, Mukherjee R, Iannini JP, Keever AT, Gallagher KP. Concomitant endothelin receptor subtype-A blockade during the progression of pacing-induced congestive heart failure in rabbits.
Circu-lation. 1997;95:1918 –1929.
27. Pi YQ, Streekumar R, Huang XP, Walker JW. Positive inotropy mediated by diacylglycerol in rat ventricular myocytes. Circ Res. 1997;81:92–100. 28. Wang JX, Paik G, Morgan JP. Endothelin-1 enhances myofilament Ca2⫹
-responsiveness in aequorin-loaded ferret myocardium. Circ Res. 1991; 69:582–589.
29. Fujita S, Endoh M. Effects of endothelin-1 on [Ca2⫹]
i-shortening
tra-jectory and Ca2⫹sensitivity in rabbit single ventricular cardiomyocytes
loaded with indo-1/AM: comparison with the effects of phenylephrine and angiotensin II. J Card Fail. 1996;2:S45–S57.
30. Kelly RA, Eid H, Kramer BK, O’Neill M, Liang BT, Reers M, Smith TW. Endothelin enhances the contractile responsiveness of adult rat ventricu-lar myocytes to calcium by a pertussis toxin-sensitive pathway. J Clin
Invest. 1990;86:1164 –1171.
31. Yang HT, Zhu Y, Endoh M. Species-dependent differences in inotropic effects and phosphoinositide hydrolysis induced by endothelin-3 in mam-malian ventricular myocardium. Br J Pharmacol. 1997;120:1497–1504. 32. Pi YQ, Kemnitz KR, Zhang D, Kranias EG, Walker JW. Phosphorylation of troponin I controls cardiac twitch dynamics: evidence from phosphor-ylation site mutants expressed on a troponin I-null background in mice.
Circ Res. 2002;90:649 – 656.
33. Sato S, Talukder MA, H, Sugawara H, Sawada H, Endoh M. Effects of levosimendan on myocardial contractility and Ca2⫹transient in
aequorin-loaded right-ventricular papillary muscles and indo-1-aequorin-loaded single ven-tricular cardiomyocytes of the rabbit. J Mol Cell Cardiol. 1998;30: 1115–1128.
34. Takahashi R, Talukder MA, H, Endoh M. Effects of OR-1896, an active metabolite of levosimendan, on contractile force and aequorin light transient in intact rabbit ventricular myocardium. J Cardiovasc
Pharmacol. 2000;36:118 –125.
35. Takahashi R, Talukder MA, H, Endoh M. Inotropic effects of OR-1896, an active metabolite of levosimendan, on canine ventricular myocardium.
Eur J Pharmacol. 2000;400:103–112.
36. Winegrad S. Cardiac myosin binding protein C. Circ Res. 1999;84: 1117–1126.
37. Endoh M, Norota I, Takanashi M, Kasai H. Inotropic effects of stauro-sporine, NA 0345 and H-7, protein kinase C inhibitors, on rabbit ven-tricular myocardium: selective inhibition of the positive inotropic effect mediated by␣1-adrenoceptor. Jpn J Pharmacol. 1993;63:17–26.
38. Hasson Talukder MA, Endoh M. Differential effects of protein kinase C activators and inhibitors on␣- and -adrenoceptor-mediated positive inotropic effect in isolated rabbit papillary muscle. J Cardiovasc
Pharmacol Ther. 1997;2:159 –170.
39. Walker CA, Ergul A, Zile MR, Zellner JL, Crumbley AJ, Spinale FG.
-Adrenergic and endothelin receptor interaction in dilated human
car-diomyopathic myocardium. J Card Fail. 2001;7:129 –137.
40. Xiao RP, Avdonin P, Zhou YY, Cheng H, Akhter SA, Eschenhagen T, Lefkowitz RJ, Koch WJ, Lakatta EG. Coupling of2-adrenoceptor to Gi
proteins and its physiological relevance in murine cardiac myocytes. Circ
Res. 1999;84:43–52.
41. Chu L, Endoh M. Biphasic inotropic effects of endothelin-1 in the presence of sympathomimetic drugs at different concentrations in canine ventricular myocardium. Jpn J Pharmacol. 2000;82(suppl I):198P. Abstract.
42. Gupta RC, Neumann J, Boknik P, Watanabe AM. M2-specific muscarinic cholinergic receptor-mediated inhibition of cardiac regulatory protein phosphorylation. Am J Physiol. 1994;266:H1138 –H1144.
43. Narayan P, Mentzer RM Jr, Lasley RD. Phosphatase inhibitor cantharidin blocks adenosine A1 receptor anti-adrenergic effect in rat cardiac myocytes. Am J Physiol. 2000;278:H1–H7.
44. Ahmad Z, Green F, Subuhi HS, Watanabe AM. Autonomic regulation of type 1 protein phosphatase in cardiac muscle. J Biol Chem. 1989;264: 3859 –3863.
45. Gupta RC, Neumann J, Watanabe AM. Comparison of adenosine and muscarinic receptor-mediated effects on protein phosphatase inhibitor-1 activity in the heart. J Pharmacol Exp Ther. 1993;266:16 –22. 46. Sakai R, Shen JB, Pappano AJ. Elevated cAMP suppresses muscarinic
inhibition of L-type calcium current in guinea pig ventricular myocytes.
J Cardiovasc Pharmacol. 1999;34:304 –315.
47. Herzig S, Meier A, Pfeiffer M, Neumann J. Stimulation of protein phosphatases as a mechanism of the muscarinic-receptor-mediated inhi-bition of the cardiac L-type Ca2⫹channels. Pflugers Arch. 1995;429:
531–538.
48. Chu L, Norota I, Ishii K, Endoh M. Inhibitory action of the phosphatase inhibitor cantharidin on the endothelin-1induced and the carbachol-induced negative inotropic effect in the canine ventricular myocardium.
J Cardiovasc Pharmacol. 2003;41(suppl 1)S89 –S92.
