Endothelin enhances delayed potassium current via phospholipase C in guinea pig ventricular myocytes YOSHIZUMI HABUCHI, HIDE0 TANAKA, TAIJI FURUKAWA, YOSHINORI HAKUO TAKAHASHI, AND MANABU YOSHIMURA Department of Laboratory Medicine and Third Department of Internal Medicine, Kyoto Prefectural University of Medicine, Kyoto 602, Japan Habuchi, Yoshizumi, Hideo Tanaka, Taiji Furukawa, Yoshinori Tsujimura, Hakuo Takahashi, and Manabu Yoshimura. Endothelin enhances delayed potassium current via phospholipase C in guinea pig ventricular myocytes. Am. J. Physiol. 262 (Heart Circ. Physiol. 31): H345-H354, 1992.-The effects of endothelin, a novel vasoconstrictive peptide, on the delayed rectifier K’ current (1k) were examined in single dialyzed cells from guinea pig ventricles. Either big endothelin or endothelin-1 enhanced 1k at a dissociation constant of 2 nM with L-type Ca2’ current being unaffected. Under intracellular perfusion with pCa 7.6 solution, 3 nM big endothelin increased 1k by 55 t 38.5%. Either pretreatment with 10 PM l-(5isoquinolinylsulfonyl) -2-methyl-piperazine (H 7) or a low Ca2’ [ 10 mM ethylene glycol-bis(P-aminoethyl ether)-N,N,N’,N’tetraacetic acid (EGTA) and minus CaC12] internal solution diminished the enhancement. Preceding stimulation of protein kinase C (PKC) by IO-20 nM 12-0-tetradecanoylphorbol-13acetate also reduced the degree of enhancement. When Na+ was eliminated from the solutions, endothelin increased 1k distinctively in cells internally dialyzed with a low Ca2+ solution. This enhancement was not abolished by either pretreatment with H 7 or by removal of Ca2+ from the external perfusate but by increasing the internal EGTA concentration to 40 mM. Preincubation with ryanodine or internal perfusion with heparin also reduced the 1k enhancement under Na+-free conditions. Intracellular application of 200 PM guanosine 5’-0-(3thiotriphosphate) effectively attenuated the effect of endothelin. It is concluded that endothelin enhances 1k via phospholipase C-mediated PKC activation and intracellular Ca2+ mobilization. GTP-binding protein is involved in these reactions. protein kinase C; inositol 1,4,5-trisphosphate; triphosphate-binding protein
guanosine
5’-
IS A PEPTIDE first derived from porcine aortic endothelial cells (53). Among various vasoactive agents, endothelin causes the most potent and longlasting contraction of vascular smooth muscle. Of several endothelin analogues, endothelin-1 consisting of 21 amino acids is understood to be the most potent active form, and big endothelin consisting of 48 amino acids is designated as a precursor form (27). Recent studies regarding the effects of endothelin on smooth muscle have been focused on phospholipase C (PLC) activation (34, 39, 42, 46). The resultant degradation of phosphatidylinositol-bis-phosphate produces inositol 1,4,5trisphosphate (IP3) and l,%diacylglycerol. The former product is known to release Ca2+ from the intracellular storage sites in numerous types of tissues (5), and the latter characteristically activates protein kinase C (PKC) in the cytoplasm. Besides Ca2+ release triggered by IP3, the activation of PKC also contributes to the contraction of vascular smooth muscle (39, 46). With regard to cardiac tissues, Ishikawa et al. (24) first demonstrated that there are endothelin receptors in
ENDOTHELIN
0363-6135/92
$2.00 Copyright
TSUJIMURA,
guinea pig atrium and that endothelin exerts a positive inotropic effect at concentrations in the order of 1 nM. A recent report has demonstrated the abundant distribution of mRNA of endothelin receptors in the heart (2). Based on action potential experiments, Ishikawa et al. (24,25) speculated that the endothelin-induced inotropic effect is attributable to an increase in the voltage-dependent L-type Ca2+ current (I&. Subsequent reports using the voltage-clamp method demonstrated contradictory results. Tohse et al. (50) reported that endothelin rather decreases Ica of the myocardium at concentrations in the order of 100 nM. On the other hand, Lauer and Clusin (31) reported that endothelin augments Ica under intracellular GTP-rich conditions. Aside from lca, the positive inotropic effects of endothelin in myocardium may be ascribed to an IP3-mediated increase in intracellular Ca2+ concentration ( [ Ca2+]i). However, little is known to date about the role of IP3 in the intracellular signaling systems in myocardium, and contradictory results have been reported regarding its Ca2+-releasing function (37, 38). On the other hand, activation of PKC is considered rather to reduce the developed tension or [Ca2+]i transient in myocardium (6, 32). Kramer et al. (28) proposed that the positive inotropic effect of endothelin is attributable in part to intracellular alkalization mediated by PKC and Na+-H+ exchanger. Of various ionic currents in myocardial cells, the action of PKC has been most extensively investigated in the delayed rectifier K+ current (1k) of guinea pig ventricular cells. Tohse et al. (51) first demonstrated 1k enhancement by phorbol ester, a well-known activator of PKC. Subsequent reports by Walsh and Kass (52) and Yazawa and Kameyama (54) showed that PKC phosphorylated the channel protein at a different site from that phosphorylated by adenosine 3’,5’-cyclic monophosphate (CAMP)-dependent protein kinase. In addition, an increase in [Ca2+]; also stimulates 1k directly (49, 51). In this regard, measurement of 1k of the ventricular myocytes could be a good indicator for either activation of PKC or an increase in [Ca2+];. The present study was aimed at elucidating 1) whether endothelin affects 1k in guinea pig ventricular cells and, if so, 2) which mechanism( s) is (are) involved. MATERIALS
AND
METHODS
Cell isoLution. The procedures for cell isolation and voltageclamp experiments were as described previously (15). Excised guinea pig hearts were suspended on a Langendorff apparatus to be retrogradely perfused. A 5- to 7min perfusion with a Ca2+-free Tyrode solution was followed by perfusion with an enzyme solution for 30-40 min. The enzyme solution was then washed out by a K’-rich, low-cl- recovery solution. All of these
0 1992 the American
Physiological
Society
H345
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (128.059.222.107) on January 16, 2019.
H346
ENDOTHELIN
ON
CARDIAC
solutions were oxygenated, and the temperature was maintained at 3537°C. The ventricles were then cut into small pieces, and single cells were isolated by gentle agitation. These cells became Ca2+-tolerant after a 20- to 30-min incubation at 32°C in the recovery solution. They were subsequently stocked in normal Tyrode solution containing 1% bovine serum albumin (Sigma, fraction V). Sohtions. Normal Tyrode solution contained (in mM) 145 NaCl, 5.4 KCl, 1.8 CaC12, 1.0 MgC12, 0.33 NaH2P04, 20 glucose, and 5 N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acid (HEPES) (pH 7.4 adjusted by NaOH). The enzyme solution was a Ca2+-free Tyrode solution containing 0.3-0.4 mg/ml collagenase (Sigma, type 1), 5 mM MgCl,, and 1% bovine albumin. In some cases, 1 mM ethylene glycol-bis(@-aminoethyl ether)-N,N,N’,N’-tetraacetic acid (EGTA) and 0.75-0.8 mM CaC12 were added to the enzyme solution to stabilize the Ca2+ concentration. The recovery solution contained (in mM) 110 glutamate, 10 oxalate, 25 KCl, 10 KH2P04, 2.0 MgS04, 20 taurine, 5 creatine, 0.5 EGTA-K2, 20 glucose, and 5.0 HEPES (pH 7.2 adjusted by KOH). The standard external test solution contained (in mM) 145 NaCl, 2.0 KCl, 1.0 CaC12, 1.0 MgC1,, 2.0 CoC12, 20 glucose, and 5 HEPES-NaOH buffer (pH 7.4), and the standard internal solution contained (in mM) 85 Kaspartate, 25 KCl, 1.0 MgC12, 1.0 CaC12, IO EGTA-K2, 5.0 ATPK2, 5.0 creatine phosphate-Na2, and 5 HEPES-KOH buffer (pH 7.2). When the experiments were carried out under Na+free conditions, NaCl in the external solution was replaced by equimolar choline-Cl with the pH adjusted by LiOH or tris(hydroxymethyl)aminomethane. Creatine phosphate-K2 instead of creatine phosphate-Na2 was used for the internal solution (0 Na+ solutions). According to the stabilizing constants proposed by Fabiato and Fabiato (9), the pCa of these internal solutions was -7.6. In some experiments, CaC12 was eliminated from the internal solution to lower [Ca”‘]i (low [Ca”+]i solution). In others, the chelating capacity for the intracellular Ca2+ was reinforced by increasing the concentration of EGTA-K2 to 40 mM with equivalent K-aspartate substituted. For measurements of -Tea, 1k was eliminated by the use of the following solutions. The external solution contained (in mM) 100 NaCl, 20 CsCl, 30 tetraethyl ammonium-Cl, 1.8 CaC12, 1.0 MgCl,, 20 glucose, and 5 HEPES-CsOH buffer (pH 7.4), and the internal solution contained 70 Cs-aspartate, 20 CsCl, 20 tetraethylammonium-Cl, 1.0 CaC12, 10 EGTA-Cs2, 5.0 ATPMg, 5.0 creatine-Na2, and 5 HEPES-CsOH buffer (pH 7.2). Materials. Porcine big endothelin (PET-48) and endothelin 1 (PET-21) were purchased from the Peptide Institute (Osaka, Japan) and dissolved in distilled water to make a frozen stock solution. In all experiments, a concentration of 3 nM was used unless otherwise specified. GTP-NaZ, 12-O-tetra-decanoylphorbol-13-acetate (TPA), 4cr-phorbol, l-(5-isoquinolinylsulfonyl)2-methyl-piperazine (H 7), heparin-Na and heparin-Li were purchased from Sigma Chemical (St. Louis, MO). Guanosine 5’-0-(3-thiotriphosphate) (GTP+) was from Boehringer Mannheim. TPA and 4cu-phorbol were dissolved in dimethyl sulfoxide as a 1 mM stock solution. Ryanodine from Wako Chemical (Tokyo) was dissolved in 5% methanol as a IO mM stock solution. Electrical measurement and data analysis. Voltage-clamp experiments were carried out using the single-pipette whole-cell clamp method (16). A whole cell-clamp amplifier (TM 1000, ACT ME, Tokyo) was used for voltage clamping. The heatpolished pipette electrodes had a tip resistance between 2 and 4 MQ when filled with an internal solution. A gigohm seal was attained while cells were perfused with normal Tyrode solution. After the cell membrane in the patch was ruptured by a negative pressure of 70-150 cmH20, the external solution was switched to a test solution. The volume of bathing chamber was 0.15 ml. A switch of external solutions was completed within 30 s when
POTASSIUM
CURRENT
the chamber was superfused at a rate of 2-3 ml/min. All experiments were carried out at temperatures of 35-37°C. The series resistance was compensated to minimize the duration of the capacitive surge. 1k was usually activated by a depolarizing pulse of 2-s duration applied from a -40 mV holding potential to a +40 mV test potential at a frequency of 0.05 Hz. The magnitude of 1k was measured from the tail current on return to the holding potential. After rupture of the patch membrane, 1k rapidly decreased to between 25 and 50% of the initial value during the first 5-7 min. This was followed by a relatively stable period of 5- to 15-min duration. In most experiments of this study, a rest period of 3-5 min at a holding potential of -40 mV was usually set before the application of test pulses, and data were obtained during the stable period. The current records were filtered at 5 kHz and stored on a digital cassette data recorder (TEAC, RD-lOIT). The data were subsequently reproduced on a digital oscilloscope (Nicolet 31OC) at a dwell time of l-2 ms. The digitized data were entered into a computer (NEC 9801, Tokyo) for analysis. Statistical analysis was by Student’s t test, and data are presented as means t SD. RESULTS
Endothelin effects on I& Figure IA shows membrane current recordings from a cell perfused with normal Tyrode solution. The data before and 3 min after exposure to big endothelin at a concentration of 3 nM are presented. In the top traces, a step depolarization from a holding potential of -40 to 0 mV initially evoked lca. The high-speed traces in the inset demonstrate that this concentration of endothelin hardly affected the inward peak. The top records in Fig. 1A concurrently revealed that the current during the latter part of the l,OOO-ms test depolarization increased outwardly after the endothelin application, being accompanied by an increase in the outward tail current (closed circles). This augmentation of the time-dependent outward current (1k) was more obvious when a depolarizing test pulse was applied to a more positive potential of +60 mV (bottom traces). The effect of endothelin on Ica is further demonstrated in Fig. lB, where the cell was perfused with K+-free solutions (see MATERIALS AND METHODS). Although lca decreased from -2.54 to -2.33 nA during the 5-min perfusion with 3 nM big endothelin, the reduction was not discernible from the run-down sequence. Similar results were obtained in six cells either with big endothelin (n = 4) or endothelin-1 (n = 2). Figure 2 shows a representative response of 1k to endothelin in a cell perfused with the standard test solutions. 1k exhibited run-down during the initial 3 min of the recording. After attaining a stable period, 3 nM big endothelin was added to the solution, which promptly increased the 1k with a maximal effect being attained 2 min after the start of perfusion. 1k enhancement by big endothelin was consistently observed in seven cells tested; however, the magnitude varied from cell to cell, ranging from 17 to 112% (55.0 t 38.5%). The peak response was usually followed by a small decrease of lo20% (see also Fig. 6B). However, the superimposing rundown sequence hindered the quantitative analysis of the effect of endothelin in a tonic phase. The 1k responses to various concentrations of endothelin are summarized in Fig. 3. The 1k enhancement is expressed by a doseresponse curve with a dissociation constant (&) of 2 nM
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (128.059.222.107) on January 16, 2019.
ENDOTHELIN
CARDIAC
POTASSIUM
H347
CURRENT
0
OmV 40
ON
l
mV
??---
0.6
Endothelin
0.5
.+,.*
&_I 2 set
3 nM
2
(3) %
(1) 3 =%
0.4- m ..A
(2) .
+c
5
0.3-
1
1 5
0
Time
O0
1
Time 5
(min)
I
10 I
1
-,
(5) -b-
0.2 1
B
(4)
e
-.-
200 msec
0
I 10 (min)
I 15
1 20
Fig. 2. IK enhancement induced by endothelin. Cell was perfused with standard test solutions. Test pulses of 2-s duration were applied from -40 mV holding potential to +40 mV. Top: current records obtained at time indicated by numbers in parentheses. Bottom: amplitude of tail current was measured between peak of tail and current 5 s after return to holding potential and is plotted as function of time. Horizontal bar, period of 3 nM endothelin perfusion.
2.0-I y -3 J
T
1.8-
(7)
Endothelin
Fig. 1. A: effects of endothelin on Ca2+ and K+ currents (lca and 1k) of single ventricular cell bathed in normal Tyrode solution. Internal solution was standard internal test solution. Holding potential was -40 mV, and l-s depolarizing pulses were applied to 0 mV every 20 s. At start and 3 min into endothelin perfusion, additional 0.5-s depolarizing pulse to +60 mV was applied. Concentration of endothelin (big endothelin) was 3 nM. Top and bottom: currents in response to test pulses to 0 and +60 mV, respectively. Inset: high-speed recordings of initial inward Ica at 0 mV are indicated by closed arrow. Currents before and 3 min after endothelin exposure are marked by open and closed circles, respectively. B: effect of endothelin on Ica. Internal and external solutions are K+ free. Test pulses were applied from -40 to 0 mV every 20 s. Magnitude of Ica was measured from holding current to inward peak and is plotted in diagram. Current records at time indicated by numbers in parentheses are shown in insets.
Fig. 3. Dose-response curve for IK enhancement by endothelin. Fractional change in IK responding to various concentrations of endothelin1 (open circles) and big endothelin (closed circles) are plotted. Circles and bars represent means t SD. Numbers in parentheses show number of experiments. Solid curve was fitted by equation: relative IK = E,,,/ (1 + &/[endothelin]), where E,,, denotes maximal response and & is dissociation constant. E,,, = 1.8. & = 2 nM.
and a maximal fractional response of 1.8. There was no statistical significance between the effects of endothelin1 and big endothelin at 3 nM concentration (P > 0.05). In the following experiments, data were obtained using big endothelin. The effects of endothelin on the voltage-dependent gating kinetics for 1k are shown in Fig. 4. As shown by the current traces in Fig. 4A or the current-voltage relationships in Fig. 4B, 3 nM endothelin doubled the 1k tail amplitude in this cell. When the activation curves were obtained by normalizing the tail amplitudes by the maximal one (Fig. 4C), they were superimposable before and after exposure to endothelin, suggesting that endothelin does not essentially alter the voltage dependence of the channel availability. The effect of endothelin on the decaying time course of 1k tail is shown in Fig. 40. The decay of the tail on return to a holding potential of -40 mV was well expressed by a sum of two exponentials. Although 3 nM endothelin increased the tail amplitude by 35% in this cell (left), it barely affected either the faster or slower time constant. From eight cells where
the deactivating time constants at -40 mV were measured, the faster and slower time constants amounted to 120 t 29 and 456 t 193 ms during control, respectively, and 122 +- 24 and 479 t 178 ms after exposure to 3 nM endothelin, respectively. The differences in these time constants were not statistically significant. Accordingly, endothelin is considered to enhance 1k by increasing the conductance of the macroscopic current. Involvement of GTP-binding protein in the actions of endothekn on 1K. Various types of receptors that regulate PLC activity are coupled with GTP-binding protein (G,) (33). In Fig. 5, the effect of endothelin was tested in cells internally perfused with 200 PM GTP+, a nonhydrolyzable GTP. In this cell, test pulses were applied immediately after rupture of the patch membrane under perfusion with the standard test solutions. The 1k showed a transient run-up during the initial l-3 min, which was followed by an overwhelming early run-down phase. A lo-30% initial increase in 1k was observed in three of five cells tested. This run-up phenomenon would indicate that the stimulatory effects of G, and G, proteins on 1k
0.3 1 Concentration
3 10 of endothelin
30 (nM)
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (128.059.222.107) on January 16, 2019.
H348
ENDOTHELIN Endothelin
A
ON
3 nM
CARDIAC
POTASSIUM
v) d 1 P
Control
1set
A
CURRENT l.O-
-
(2) JI, “0.
0.8-
- (1) zE. 0.6Y 0.4 -
t N
0.2 0
f
I
I
A
5
10
Rupture 0.8-
(min)
Fig. 5. Internal pretreatment with nonhydrolyzable GTP analogue (GTP+) abolishes 1k enhancement by endothelin. Test solutions were standard, but internal solution contained 200 PM GTP+. Recording was commenced just after rupture of patch membrane. Abscissa represents time after rupture of membrane. Concentration of endothelin was 10 nM.
0.6-
-40
Test potential
Time
I
15
-20
(mv)
0
Test potential
+20
+40
+60
(mv)
-WA
0.6
A
0 Control 0 Endothelin
l 0.2
lfest=
141
msec
set
Fig. 4. Effects of endothelin on kinetics of IK. A-C: effect on voltage dependence of Ix activation. Depolarizing pulses of 3-s duration were delivered to various potentials ranging from -30 to +60 mV in IO mV steps. A: resultant membrane currents. Tail amplitudes and their normalized values are plotted against test potentials in B and C, respectively. Smooth curves were drawn by eye. D: effect on deactivation time constants. Tail currents on repolarization from +40 test potential to -40 mV holding potential are shown on left. Tail currents are semilogarithmically plotted on right. Deactivation sequence was expressed as 2 exponential functions; straight lines represent leastsquares fit. Faster component was acquired by subtracting extrapolation of exponential fit of slower component from observed current. Faster and slower time constants are indicated. Open and closed circles, time constants before and after 3 nM endothelin perfusion, respectively.
overwhelm the inhibitory effect of Gi protein. Other researchers (54) as well as ourselves (48) have shown that 1K of the guinea pig ventricular cells is augmented when both G, and Gi proteins are fully activated. However, we could not quantitatively evaluate the effect of intracellular application of GTPyS in the present study because of the superimposing initial run-down of 1r+ As exemplified in Fig. 5, the preceding stimulation of G, by GTPyS abolished the 1K response to endothelin in two of five cells. In the remaining cells, endothelin increased 1K only by ~20%. In six other cells, an internal solution containing 200 PM GTP was used. Endothelin, 3 nM, increased 1K by 59.3 t 30.5% under standard conditions. The degree of the increase was not significantly different from that seen in cells without GTP in the pipette (Fig. 3) . PKC-related IK enhancement. Figures 6 and 7 show the contribution of PKC to the endothelin-induced 1K enhancement. TPA activates PKC with a Kd of -10 nM
4
Time
6 (min)
8
10
12
4 a-r>horbol
0
5
10 Time
15
(min)
Fig. 6. Effect of endothelin on Ik after decanoylphorbol-13-acetate (TPA) and nM endothelin was preceded by that phorbol (B). Inset: tail currents observed in parentheses. Test pulse duration was
pretreatment with 12-O-tetra4cu-phorbol. Perfusion with 3 with 20 nM TPA (A) or 4~ at times indicated by numbers 2 s, and potential was +40 mV.
(3). In Fig. 6A, 1K increased from 0.29 to 0.455 nA in response to 20 nM TPA. Subsequent addition of endothelin increased the 1K only to 0.49 nA, or by 7.7%. In similar experiments, lo-20 nM TPA increased 1K by 45.9 t 38.8% (n = 7). Further addition of 3 nM endothelin did not cause a discernible 1K increase in two cells, but a small increase of ~20% was observed in the remaining cells. We then examined the effects of 4a-phorbol, an inactive isomer of TPA. A representative result is shown in Fig. 6B. It is demonstrated that 20 nM 4cu-phorbol did not change the 1K and that its pretreatment did not attenuate the endothelin effect on 1i+ Namely, 3 nM endothelin increased 1K by 35.1 t 17.5% in the presence
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (128.059.222.107) on January 16, 2019.
ENDOTHELIN
A d
ON
CARDIAC
H-7
I
Endothelin
I
0.1
I
B
2+ LOW
I
tea
I
I
2
0
4 6 Time (min)
I
8
10
Ii Endothelin
0
2
4
6 Time
8
10
12
(min)
Fig. 7. A: effect of endothelin on 1K in presence of 10 PM 1-(5isoquinolinylsulfonyl)-2-methyl-piperazine (H 7). Internal and external solutions were standard. Periods of H 7 and endothelin perfusions are indicated by solid bars. B: effect of endothelin under low intracellular Ca2+ concentration ([Ca”‘];) conditions. Internal solution was CaC12free with 10 mM EGTA (low [ Ca2+]i solution), and external solution was standard test solution. Solid curve assumed to represent run-down sequence was estimated by eye.
POTASSIUM
H349
CURRENT
external and internal solutions (0 Na+ solutions) and the Na+-Ca2+ exchange system was blocked. In Fig. 8A, endothelin was applied to cells perfused internally with a low [Ca2+]; solution (n = 7). In contrast to the results presented in Fig. 7B, endothelin gradually increased 1k. The degree of 1k enhancement was again variable from cell to cell as suggested by the large SD values in the diagram. The data presented in Fig. 8B were from an experiment in which the endothelin application was preceded by perfusion with 10 PM H 7. It can be seen that even under conditions designed to minimize the activity of PKC, endothelin again induced a distinctive 1k enhancement (25.9 t 13.9%, n = 3). To elucidate the role of Ca2+ in the 1k enhancement under the Na+-free condition shown in Fig. 8, the effect of endothelin on 1k was tested in four cells that were internally perfused with 40 mM EGTA. In these experiments, pipette electrodes having a relatively large tip diameter or a tip resistance of 1.5-2 MS2 were used. As shown in Fig. 9A, 3 nM endothelin failed to increase 1k in these four cells tested. Use of a larger pipette electrode may have changed the membrane (receptor) properties or diluted intracellular essential substances such as GTP. Thus, in three other cells clamped through an electrode of a similar size, the effects of endothelin were tested with standard test solutions. In these cells, 3 nM endoEndothelin I””
-TT
160
of 20 nM 4a-phorbol (n = 3). The effect of endothelin on 1k in cells pretreated with H 7, a protein kinase inhibitor (17), is shown in Fig. 7A. It was reported that 10 PM of this agent abolished TPAinduced 1k augmentation in guinea pig ventricular cells (51). H 7 application usually reduced 1k by -10%. The addition of 3 nM endothelin produced only a small increase of 5-10%. Similar results were observed in two other cells among four cells tested, and in the remaining one, endothelin did not increase 1k to a distinguishable extent. The activity of PKC is profoundly dependent on [ Ca2+]i. In the experiments shown in Fig. 7B, the [Ca2+]i was lowered by the low [Ca2+]i solution. Although a careful inspection would discern a slight increase of Ifc from the run-down sequence (estimated by the solid curve), use of the low [Ca2+]i solution also resulted in a drastic attenuation of the endothelin-induced 1k enhancement. A similar small 1k enhancement of
guanosine
5’-
IS A PEPTIDE first derived from porcine aortic endothelial cells (53). Among various vasoactive agents, endothelin causes the most potent and longlasting contraction of vascular smooth muscle. Of several endothelin analogues, endothelin-1 consisting of 21 amino acids is understood to be the most potent active form, and big endothelin consisting of 48 amino acids is designated as a precursor form (27). Recent studies regarding the effects of endothelin on smooth muscle have been focused on phospholipase C (PLC) activation (34, 39, 42, 46). The resultant degradation of phosphatidylinositol-bis-phosphate produces inositol 1,4,5trisphosphate (IP3) and l,%diacylglycerol. The former product is known to release Ca2+ from the intracellular storage sites in numerous types of tissues (5), and the latter characteristically activates protein kinase C (PKC) in the cytoplasm. Besides Ca2+ release triggered by IP3, the activation of PKC also contributes to the contraction of vascular smooth muscle (39, 46). With regard to cardiac tissues, Ishikawa et al. (24) first demonstrated that there are endothelin receptors in
ENDOTHELIN
0363-6135/92
$2.00 Copyright
TSUJIMURA,
guinea pig atrium and that endothelin exerts a positive inotropic effect at concentrations in the order of 1 nM. A recent report has demonstrated the abundant distribution of mRNA of endothelin receptors in the heart (2). Based on action potential experiments, Ishikawa et al. (24,25) speculated that the endothelin-induced inotropic effect is attributable to an increase in the voltage-dependent L-type Ca2+ current (I&. Subsequent reports using the voltage-clamp method demonstrated contradictory results. Tohse et al. (50) reported that endothelin rather decreases Ica of the myocardium at concentrations in the order of 100 nM. On the other hand, Lauer and Clusin (31) reported that endothelin augments Ica under intracellular GTP-rich conditions. Aside from lca, the positive inotropic effects of endothelin in myocardium may be ascribed to an IP3-mediated increase in intracellular Ca2+ concentration ( [ Ca2+]i). However, little is known to date about the role of IP3 in the intracellular signaling systems in myocardium, and contradictory results have been reported regarding its Ca2+-releasing function (37, 38). On the other hand, activation of PKC is considered rather to reduce the developed tension or [Ca2+]i transient in myocardium (6, 32). Kramer et al. (28) proposed that the positive inotropic effect of endothelin is attributable in part to intracellular alkalization mediated by PKC and Na+-H+ exchanger. Of various ionic currents in myocardial cells, the action of PKC has been most extensively investigated in the delayed rectifier K+ current (1k) of guinea pig ventricular cells. Tohse et al. (51) first demonstrated 1k enhancement by phorbol ester, a well-known activator of PKC. Subsequent reports by Walsh and Kass (52) and Yazawa and Kameyama (54) showed that PKC phosphorylated the channel protein at a different site from that phosphorylated by adenosine 3’,5’-cyclic monophosphate (CAMP)-dependent protein kinase. In addition, an increase in [Ca2+]; also stimulates 1k directly (49, 51). In this regard, measurement of 1k of the ventricular myocytes could be a good indicator for either activation of PKC or an increase in [Ca2+];. The present study was aimed at elucidating 1) whether endothelin affects 1k in guinea pig ventricular cells and, if so, 2) which mechanism( s) is (are) involved. MATERIALS
AND
METHODS
Cell isoLution. The procedures for cell isolation and voltageclamp experiments were as described previously (15). Excised guinea pig hearts were suspended on a Langendorff apparatus to be retrogradely perfused. A 5- to 7min perfusion with a Ca2+-free Tyrode solution was followed by perfusion with an enzyme solution for 30-40 min. The enzyme solution was then washed out by a K’-rich, low-cl- recovery solution. All of these
0 1992 the American
Physiological
Society
H345
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (128.059.222.107) on January 16, 2019.
H346
ENDOTHELIN
ON
CARDIAC
solutions were oxygenated, and the temperature was maintained at 3537°C. The ventricles were then cut into small pieces, and single cells were isolated by gentle agitation. These cells became Ca2+-tolerant after a 20- to 30-min incubation at 32°C in the recovery solution. They were subsequently stocked in normal Tyrode solution containing 1% bovine serum albumin (Sigma, fraction V). Sohtions. Normal Tyrode solution contained (in mM) 145 NaCl, 5.4 KCl, 1.8 CaC12, 1.0 MgC12, 0.33 NaH2P04, 20 glucose, and 5 N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acid (HEPES) (pH 7.4 adjusted by NaOH). The enzyme solution was a Ca2+-free Tyrode solution containing 0.3-0.4 mg/ml collagenase (Sigma, type 1), 5 mM MgCl,, and 1% bovine albumin. In some cases, 1 mM ethylene glycol-bis(@-aminoethyl ether)-N,N,N’,N’-tetraacetic acid (EGTA) and 0.75-0.8 mM CaC12 were added to the enzyme solution to stabilize the Ca2+ concentration. The recovery solution contained (in mM) 110 glutamate, 10 oxalate, 25 KCl, 10 KH2P04, 2.0 MgS04, 20 taurine, 5 creatine, 0.5 EGTA-K2, 20 glucose, and 5.0 HEPES (pH 7.2 adjusted by KOH). The standard external test solution contained (in mM) 145 NaCl, 2.0 KCl, 1.0 CaC12, 1.0 MgC1,, 2.0 CoC12, 20 glucose, and 5 HEPES-NaOH buffer (pH 7.4), and the standard internal solution contained (in mM) 85 Kaspartate, 25 KCl, 1.0 MgC12, 1.0 CaC12, IO EGTA-K2, 5.0 ATPK2, 5.0 creatine phosphate-Na2, and 5 HEPES-KOH buffer (pH 7.2). When the experiments were carried out under Na+free conditions, NaCl in the external solution was replaced by equimolar choline-Cl with the pH adjusted by LiOH or tris(hydroxymethyl)aminomethane. Creatine phosphate-K2 instead of creatine phosphate-Na2 was used for the internal solution (0 Na+ solutions). According to the stabilizing constants proposed by Fabiato and Fabiato (9), the pCa of these internal solutions was -7.6. In some experiments, CaC12 was eliminated from the internal solution to lower [Ca”‘]i (low [Ca”+]i solution). In others, the chelating capacity for the intracellular Ca2+ was reinforced by increasing the concentration of EGTA-K2 to 40 mM with equivalent K-aspartate substituted. For measurements of -Tea, 1k was eliminated by the use of the following solutions. The external solution contained (in mM) 100 NaCl, 20 CsCl, 30 tetraethyl ammonium-Cl, 1.8 CaC12, 1.0 MgCl,, 20 glucose, and 5 HEPES-CsOH buffer (pH 7.4), and the internal solution contained 70 Cs-aspartate, 20 CsCl, 20 tetraethylammonium-Cl, 1.0 CaC12, 10 EGTA-Cs2, 5.0 ATPMg, 5.0 creatine-Na2, and 5 HEPES-CsOH buffer (pH 7.2). Materials. Porcine big endothelin (PET-48) and endothelin 1 (PET-21) were purchased from the Peptide Institute (Osaka, Japan) and dissolved in distilled water to make a frozen stock solution. In all experiments, a concentration of 3 nM was used unless otherwise specified. GTP-NaZ, 12-O-tetra-decanoylphorbol-13-acetate (TPA), 4cr-phorbol, l-(5-isoquinolinylsulfonyl)2-methyl-piperazine (H 7), heparin-Na and heparin-Li were purchased from Sigma Chemical (St. Louis, MO). Guanosine 5’-0-(3-thiotriphosphate) (GTP+) was from Boehringer Mannheim. TPA and 4cu-phorbol were dissolved in dimethyl sulfoxide as a 1 mM stock solution. Ryanodine from Wako Chemical (Tokyo) was dissolved in 5% methanol as a IO mM stock solution. Electrical measurement and data analysis. Voltage-clamp experiments were carried out using the single-pipette whole-cell clamp method (16). A whole cell-clamp amplifier (TM 1000, ACT ME, Tokyo) was used for voltage clamping. The heatpolished pipette electrodes had a tip resistance between 2 and 4 MQ when filled with an internal solution. A gigohm seal was attained while cells were perfused with normal Tyrode solution. After the cell membrane in the patch was ruptured by a negative pressure of 70-150 cmH20, the external solution was switched to a test solution. The volume of bathing chamber was 0.15 ml. A switch of external solutions was completed within 30 s when
POTASSIUM
CURRENT
the chamber was superfused at a rate of 2-3 ml/min. All experiments were carried out at temperatures of 35-37°C. The series resistance was compensated to minimize the duration of the capacitive surge. 1k was usually activated by a depolarizing pulse of 2-s duration applied from a -40 mV holding potential to a +40 mV test potential at a frequency of 0.05 Hz. The magnitude of 1k was measured from the tail current on return to the holding potential. After rupture of the patch membrane, 1k rapidly decreased to between 25 and 50% of the initial value during the first 5-7 min. This was followed by a relatively stable period of 5- to 15-min duration. In most experiments of this study, a rest period of 3-5 min at a holding potential of -40 mV was usually set before the application of test pulses, and data were obtained during the stable period. The current records were filtered at 5 kHz and stored on a digital cassette data recorder (TEAC, RD-lOIT). The data were subsequently reproduced on a digital oscilloscope (Nicolet 31OC) at a dwell time of l-2 ms. The digitized data were entered into a computer (NEC 9801, Tokyo) for analysis. Statistical analysis was by Student’s t test, and data are presented as means t SD. RESULTS
Endothelin effects on I& Figure IA shows membrane current recordings from a cell perfused with normal Tyrode solution. The data before and 3 min after exposure to big endothelin at a concentration of 3 nM are presented. In the top traces, a step depolarization from a holding potential of -40 to 0 mV initially evoked lca. The high-speed traces in the inset demonstrate that this concentration of endothelin hardly affected the inward peak. The top records in Fig. 1A concurrently revealed that the current during the latter part of the l,OOO-ms test depolarization increased outwardly after the endothelin application, being accompanied by an increase in the outward tail current (closed circles). This augmentation of the time-dependent outward current (1k) was more obvious when a depolarizing test pulse was applied to a more positive potential of +60 mV (bottom traces). The effect of endothelin on Ica is further demonstrated in Fig. lB, where the cell was perfused with K+-free solutions (see MATERIALS AND METHODS). Although lca decreased from -2.54 to -2.33 nA during the 5-min perfusion with 3 nM big endothelin, the reduction was not discernible from the run-down sequence. Similar results were obtained in six cells either with big endothelin (n = 4) or endothelin-1 (n = 2). Figure 2 shows a representative response of 1k to endothelin in a cell perfused with the standard test solutions. 1k exhibited run-down during the initial 3 min of the recording. After attaining a stable period, 3 nM big endothelin was added to the solution, which promptly increased the 1k with a maximal effect being attained 2 min after the start of perfusion. 1k enhancement by big endothelin was consistently observed in seven cells tested; however, the magnitude varied from cell to cell, ranging from 17 to 112% (55.0 t 38.5%). The peak response was usually followed by a small decrease of lo20% (see also Fig. 6B). However, the superimposing rundown sequence hindered the quantitative analysis of the effect of endothelin in a tonic phase. The 1k responses to various concentrations of endothelin are summarized in Fig. 3. The 1k enhancement is expressed by a doseresponse curve with a dissociation constant (&) of 2 nM
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (128.059.222.107) on January 16, 2019.
ENDOTHELIN
CARDIAC
POTASSIUM
H347
CURRENT
0
OmV 40
ON
l
mV
??---
0.6
Endothelin
0.5
.+,.*
&_I 2 set
3 nM
2
(3) %
(1) 3 =%
0.4- m ..A
(2) .
+c
5
0.3-
1
1 5
0
Time
O0
1
Time 5
(min)
I
10 I
1
-,
(5) -b-
0.2 1
B
(4)
e
-.-
200 msec
0
I 10 (min)
I 15
1 20
Fig. 2. IK enhancement induced by endothelin. Cell was perfused with standard test solutions. Test pulses of 2-s duration were applied from -40 mV holding potential to +40 mV. Top: current records obtained at time indicated by numbers in parentheses. Bottom: amplitude of tail current was measured between peak of tail and current 5 s after return to holding potential and is plotted as function of time. Horizontal bar, period of 3 nM endothelin perfusion.
2.0-I y -3 J
T
1.8-
(7)
Endothelin
Fig. 1. A: effects of endothelin on Ca2+ and K+ currents (lca and 1k) of single ventricular cell bathed in normal Tyrode solution. Internal solution was standard internal test solution. Holding potential was -40 mV, and l-s depolarizing pulses were applied to 0 mV every 20 s. At start and 3 min into endothelin perfusion, additional 0.5-s depolarizing pulse to +60 mV was applied. Concentration of endothelin (big endothelin) was 3 nM. Top and bottom: currents in response to test pulses to 0 and +60 mV, respectively. Inset: high-speed recordings of initial inward Ica at 0 mV are indicated by closed arrow. Currents before and 3 min after endothelin exposure are marked by open and closed circles, respectively. B: effect of endothelin on Ica. Internal and external solutions are K+ free. Test pulses were applied from -40 to 0 mV every 20 s. Magnitude of Ica was measured from holding current to inward peak and is plotted in diagram. Current records at time indicated by numbers in parentheses are shown in insets.
Fig. 3. Dose-response curve for IK enhancement by endothelin. Fractional change in IK responding to various concentrations of endothelin1 (open circles) and big endothelin (closed circles) are plotted. Circles and bars represent means t SD. Numbers in parentheses show number of experiments. Solid curve was fitted by equation: relative IK = E,,,/ (1 + &/[endothelin]), where E,,, denotes maximal response and & is dissociation constant. E,,, = 1.8. & = 2 nM.
and a maximal fractional response of 1.8. There was no statistical significance between the effects of endothelin1 and big endothelin at 3 nM concentration (P > 0.05). In the following experiments, data were obtained using big endothelin. The effects of endothelin on the voltage-dependent gating kinetics for 1k are shown in Fig. 4. As shown by the current traces in Fig. 4A or the current-voltage relationships in Fig. 4B, 3 nM endothelin doubled the 1k tail amplitude in this cell. When the activation curves were obtained by normalizing the tail amplitudes by the maximal one (Fig. 4C), they were superimposable before and after exposure to endothelin, suggesting that endothelin does not essentially alter the voltage dependence of the channel availability. The effect of endothelin on the decaying time course of 1k tail is shown in Fig. 40. The decay of the tail on return to a holding potential of -40 mV was well expressed by a sum of two exponentials. Although 3 nM endothelin increased the tail amplitude by 35% in this cell (left), it barely affected either the faster or slower time constant. From eight cells where
the deactivating time constants at -40 mV were measured, the faster and slower time constants amounted to 120 t 29 and 456 t 193 ms during control, respectively, and 122 +- 24 and 479 t 178 ms after exposure to 3 nM endothelin, respectively. The differences in these time constants were not statistically significant. Accordingly, endothelin is considered to enhance 1k by increasing the conductance of the macroscopic current. Involvement of GTP-binding protein in the actions of endothekn on 1K. Various types of receptors that regulate PLC activity are coupled with GTP-binding protein (G,) (33). In Fig. 5, the effect of endothelin was tested in cells internally perfused with 200 PM GTP+, a nonhydrolyzable GTP. In this cell, test pulses were applied immediately after rupture of the patch membrane under perfusion with the standard test solutions. The 1k showed a transient run-up during the initial l-3 min, which was followed by an overwhelming early run-down phase. A lo-30% initial increase in 1k was observed in three of five cells tested. This run-up phenomenon would indicate that the stimulatory effects of G, and G, proteins on 1k
0.3 1 Concentration
3 10 of endothelin
30 (nM)
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (128.059.222.107) on January 16, 2019.
H348
ENDOTHELIN Endothelin
A
ON
3 nM
CARDIAC
POTASSIUM
v) d 1 P
Control
1set
A
CURRENT l.O-
-
(2) JI, “0.
0.8-
- (1) zE. 0.6Y 0.4 -
t N
0.2 0
f
I
I
A
5
10
Rupture 0.8-
(min)
Fig. 5. Internal pretreatment with nonhydrolyzable GTP analogue (GTP+) abolishes 1k enhancement by endothelin. Test solutions were standard, but internal solution contained 200 PM GTP+. Recording was commenced just after rupture of patch membrane. Abscissa represents time after rupture of membrane. Concentration of endothelin was 10 nM.
0.6-
-40
Test potential
Time
I
15
-20
(mv)
0
Test potential
+20
+40
+60
(mv)
-WA
0.6
A
0 Control 0 Endothelin
l 0.2
lfest=
141
msec
set
Fig. 4. Effects of endothelin on kinetics of IK. A-C: effect on voltage dependence of Ix activation. Depolarizing pulses of 3-s duration were delivered to various potentials ranging from -30 to +60 mV in IO mV steps. A: resultant membrane currents. Tail amplitudes and their normalized values are plotted against test potentials in B and C, respectively. Smooth curves were drawn by eye. D: effect on deactivation time constants. Tail currents on repolarization from +40 test potential to -40 mV holding potential are shown on left. Tail currents are semilogarithmically plotted on right. Deactivation sequence was expressed as 2 exponential functions; straight lines represent leastsquares fit. Faster component was acquired by subtracting extrapolation of exponential fit of slower component from observed current. Faster and slower time constants are indicated. Open and closed circles, time constants before and after 3 nM endothelin perfusion, respectively.
overwhelm the inhibitory effect of Gi protein. Other researchers (54) as well as ourselves (48) have shown that 1K of the guinea pig ventricular cells is augmented when both G, and Gi proteins are fully activated. However, we could not quantitatively evaluate the effect of intracellular application of GTPyS in the present study because of the superimposing initial run-down of 1r+ As exemplified in Fig. 5, the preceding stimulation of G, by GTPyS abolished the 1K response to endothelin in two of five cells. In the remaining cells, endothelin increased 1K only by ~20%. In six other cells, an internal solution containing 200 PM GTP was used. Endothelin, 3 nM, increased 1K by 59.3 t 30.5% under standard conditions. The degree of the increase was not significantly different from that seen in cells without GTP in the pipette (Fig. 3) . PKC-related IK enhancement. Figures 6 and 7 show the contribution of PKC to the endothelin-induced 1K enhancement. TPA activates PKC with a Kd of -10 nM
4
Time
6 (min)
8
10
12
4 a-r>horbol
0
5
10 Time
15
(min)
Fig. 6. Effect of endothelin on Ik after decanoylphorbol-13-acetate (TPA) and nM endothelin was preceded by that phorbol (B). Inset: tail currents observed in parentheses. Test pulse duration was
pretreatment with 12-O-tetra4cu-phorbol. Perfusion with 3 with 20 nM TPA (A) or 4~ at times indicated by numbers 2 s, and potential was +40 mV.
(3). In Fig. 6A, 1K increased from 0.29 to 0.455 nA in response to 20 nM TPA. Subsequent addition of endothelin increased the 1K only to 0.49 nA, or by 7.7%. In similar experiments, lo-20 nM TPA increased 1K by 45.9 t 38.8% (n = 7). Further addition of 3 nM endothelin did not cause a discernible 1K increase in two cells, but a small increase of ~20% was observed in the remaining cells. We then examined the effects of 4a-phorbol, an inactive isomer of TPA. A representative result is shown in Fig. 6B. It is demonstrated that 20 nM 4cu-phorbol did not change the 1K and that its pretreatment did not attenuate the endothelin effect on 1i+ Namely, 3 nM endothelin increased 1K by 35.1 t 17.5% in the presence
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (128.059.222.107) on January 16, 2019.
ENDOTHELIN
A d
ON
CARDIAC
H-7
I
Endothelin
I
0.1
I
B
2+ LOW
I
tea
I
I
2
0
4 6 Time (min)
I
8
10
Ii Endothelin
0
2
4
6 Time
8
10
12
(min)
Fig. 7. A: effect of endothelin on 1K in presence of 10 PM 1-(5isoquinolinylsulfonyl)-2-methyl-piperazine (H 7). Internal and external solutions were standard. Periods of H 7 and endothelin perfusions are indicated by solid bars. B: effect of endothelin under low intracellular Ca2+ concentration ([Ca”‘];) conditions. Internal solution was CaC12free with 10 mM EGTA (low [ Ca2+]i solution), and external solution was standard test solution. Solid curve assumed to represent run-down sequence was estimated by eye.
POTASSIUM
H349
CURRENT
external and internal solutions (0 Na+ solutions) and the Na+-Ca2+ exchange system was blocked. In Fig. 8A, endothelin was applied to cells perfused internally with a low [Ca2+]; solution (n = 7). In contrast to the results presented in Fig. 7B, endothelin gradually increased 1k. The degree of 1k enhancement was again variable from cell to cell as suggested by the large SD values in the diagram. The data presented in Fig. 8B were from an experiment in which the endothelin application was preceded by perfusion with 10 PM H 7. It can be seen that even under conditions designed to minimize the activity of PKC, endothelin again induced a distinctive 1k enhancement (25.9 t 13.9%, n = 3). To elucidate the role of Ca2+ in the 1k enhancement under the Na+-free condition shown in Fig. 8, the effect of endothelin on 1k was tested in four cells that were internally perfused with 40 mM EGTA. In these experiments, pipette electrodes having a relatively large tip diameter or a tip resistance of 1.5-2 MS2 were used. As shown in Fig. 9A, 3 nM endothelin failed to increase 1k in these four cells tested. Use of a larger pipette electrode may have changed the membrane (receptor) properties or diluted intracellular essential substances such as GTP. Thus, in three other cells clamped through an electrode of a similar size, the effects of endothelin were tested with standard test solutions. In these cells, 3 nM endoEndothelin I””
-TT
160
of 20 nM 4a-phorbol (n = 3). The effect of endothelin on 1k in cells pretreated with H 7, a protein kinase inhibitor (17), is shown in Fig. 7A. It was reported that 10 PM of this agent abolished TPAinduced 1k augmentation in guinea pig ventricular cells (51). H 7 application usually reduced 1k by -10%. The addition of 3 nM endothelin produced only a small increase of 5-10%. Similar results were observed in two other cells among four cells tested, and in the remaining one, endothelin did not increase 1k to a distinguishable extent. The activity of PKC is profoundly dependent on [ Ca2+]i. In the experiments shown in Fig. 7B, the [Ca2+]i was lowered by the low [Ca2+]i solution. Although a careful inspection would discern a slight increase of Ifc from the run-down sequence (estimated by the solid curve), use of the low [Ca2+]i solution also resulted in a drastic attenuation of the endothelin-induced 1k enhancement. A similar small 1k enhancement of
