INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

₁-ADRENOCEPTOR STIMULATION in the mammalian heart mediates alterations in ionic currents (17) in addition to the well-described increase in contractile force (48). ₁-Adrenoceptor stimulation has been shown, by electrophysiological methods, to decrease potassium outward currents (44). When total ionic fluxes were measured with isotope techniques, ₁-adrenoceptor stimulation concentration dependently increased the efflux of the potassium analog ⁸⁶Rb⁺ (2). The translocation mechanisms mediating this response are unknown but may include (either directly or indirectly) the Na⁺-K⁺ pump, Na⁺/H⁺ exchanger, K⁺ channels, or an electroneutral cation-chloride (Na-K-2Cl or K-Cl) cotransporter. Stimulation of myocardial ₁-adrenoceptors activates protein kinases, and different protein kinases play a role in the regulation of ion transport in the myocardium (51). The protein kinase C (PKC), Ca²⁺-calmodulin-dependent kinase II (CaM kinase II), and the extracellular signal-regulated kinase (ERK) member of the mitogen-activated protein (MAP) kinase family may all be involved in the effects of ₁-adrenoceptor stimulation. The aim of this study was to characterize the translocation mechanisms involved in the ₁-adrenoceptor-mediated increase in ⁸⁶Rb⁺ efflux from the isolated perfused rat heart and to elucidate a possible role of protein kinases in the activation of ⁸⁶Rb⁺ efflux. Phenylephrine-stimulated ⁸⁶Rb⁺ efflux was studied in the presence or absence of selective inhibitors against possible translocation mechanisms and the protein kinases potentially involved. Phenylephrine-stimulated ERK activity was measured in the presence or absence of inhibitors against protein kinases potentially involved in the activation of the ERK cascade in order to study more specifically the effect of ERK cascade inhibition on phenylephrine-induced increase of ⁸⁶Rb⁺ efflux. Control experiments studying the inotropic response to phenylephrine in the presence of some of the active inhibitors were included to elucidate if any effects of the inhibitors might be secondary to effects on the inotropic response. We report that ₁-adrenoceptor-stimulated ⁸⁶Rb⁺ efflux is mediated through a Na-K-2Cl cotransporter, probably involving the ERK cascade, and K⁺ channels.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Heart perfusion in ⁸⁶Rb⁺ efflux studies. Hearts from male Wistar rats (200-300 g body wt), fed ad libitum, were isolated under ether anesthesia. The aorta was cannulated, and the hearts were spontaneously beating and perfused in a nonrecirculating system. The flow was kept constant (10 ml/min) by a Bifok FIA 08 roller pump or a HiLoad Pump P-50 reciprocating piston pump (both pumps from Pharmacia Biotech, Uppsala, Sweden). The basic perfusion medium contained (in mmol/l) 119.3 NaCl, 3.0 KCl, 2.4 KH₂PO₄, 1.2 MgSO₄, 2.0 CaCl₂, 24.9 NaHCO₃, 10.0 glucose, and 2.2 mannitol and was equilibrated with 95% O₂-5% CO₂ at 31°C (pH 7.4). When used, ⁸⁶Rb⁺ was added to the perfusion medium to yield ~1-2 ×10⁵ counts/min (cpm)/ml.

Experimental design in ⁸⁶Rb⁺ efflux studies. All hearts were perfused with a basic perfusion medium containing a -adrenoceptor antagonist (timolol, 1 µmol/l) throughout the experiment to suppress the -adrenoceptor-stimulated component of phenylephrine. After we allowed a perfusion period of 10 min, the hearts were loaded with the K⁺ analog ⁸⁶Rb⁺ for 30 min, followed by a 25-min control period and a 30-min period of agonist stimulation. Perfusate fractions of 3 ml (collection period 18 s) were collected consecutively with an LKB 7000 Ultro Rac fraction collector (Pharmacia Biotech, Uppsala, Sweden) during the washout period. At the end of the experiment, the hearts were freeze-clamped with an aluminum clamp precooled in liquid nitrogen and pulverized in a cooled percussion mortar. The hearts were weighed (range 1.0-1.45 g wet wt). The tissue powder was stored in plastic vials at 80°C until ⁸⁶Rb⁺ tissue radioactivity was determined. When used, blockers of different ionic translocation mechanisms or protein kinases were added to the perfusion buffer at the start of the washout period. Corresponding and separate groups were exposed to the ₁-adrenoceptor agonist phenylephrine (30 µmol/l) in the presence or absence of bumetanide or furosemide (Na-K-2Cl and K-Cl cotransport inhibitors, respectively), 4-aminopyridine (nonselective K⁺ channel inhibitor), glibenclamide [ATP-sensitive K⁺ (K⁺_ATP) channel inhibitor], HOE-694 (Na⁺/H⁺ exchange inhibitor), ouabain (Na⁺-K⁺ pump inhibitor), or N-ethylmaleimide (activator of K-Cl cotransport), respectively. In addition, groups with phenylephrine in the presence of a combination of bumetanide and either 4-aminopyridine, HOE-694, or ouabain were included. Addition of N-ethylmaleimide (0.3 or 1.0 mmol/l) resulted in arrythmias and cardiac arrest and was thus not tolerated by the spontaneously beating heart preparation. A possible involvement of the K-Cl cotransporter could thus not be elucidated by the use of this compound. Separate groups were exposed to phenylephrine in the presence or absence of staurosporine (nonspecific protein kinase inhibitor) and chelerythrine (nonspecific protein kinase inhibitor), KN-62 (CaM kinase II), or PD-98059 [MEK (MAP kinase/ERK kinase) activation inhibitor] with or without bumetanide. Some of the inhibitors were dissolved in dimethyl sulfoxide (DMSO), and control experiments showed that DMSO in the concentrations used (maximum 0.1%) did not influence the phenylephrine-evoked response (data not shown). The maximal response to ₁-adrenoceptor stimulation varied to some degree between the experimental series probably due to variations in the ₁-adrenoceptor-mediated responses throughout the year. We have, as an example, noticed a low inotropic response to ₁-adrenoceptor stimulation during the summer time compared with the winter season (unpublished results). Because of this variation, the response in the presence of each blocker was compared with the response of separate and corresponding groups of hearts exposed to phenylephrine alone. The response in each individual heart was analyzed as a percentage of the extrapolated control values (before addition of agonist) in the same heart (see Calculations of ⁸⁶Rb⁺ washout kinetics and efflux rate).

Determination of ⁸⁶Rb⁺ radioactivity. Radioactivity in the collected perfusate fractions and in the heart tissue powder was measured with a model 1900 TR Tri-Carb liquid-scintillation spectrometer (Packard Instrument, Downers Grove, IL). The count rates were corrected for radioactivity decay. The radioactivity of the perfusate fractions was determined as Cerenkov radiation directly in the aqueous medium. The heart tissue powder (150 mg) was homogenized in 7.5% trichloroacetic acid at a dilution of 1:10 wt/vol. The radioactivity in 0.1 ml supernatant was determined as Cerenkov radiation after the addition of 2.9 ml basic perfusion medium. The contents of the heart radioactivity at time points corresponding to the sampling time of each perfusate fraction were calculated. To the ⁸⁶Rb⁺ contents in the hearts at the end of perfusion, the contents in each previous perfusate fraction were added sequentially and cumulatively.

Calculations of ⁸⁶Rb⁺ washout kinetics and efflux rate. The rate of ⁸⁶Rb⁺ efflux was calculated as radioactivity appearing in the perfusate per minute (cpm/min) during washout. The efflux rate index ("fractional efflux," in min¹) could be expressed as the ratio between perfusate ⁸⁶Rb⁺ radioactivity and ⁸⁶Rb⁺ contents of the heart at the corresponding time points. A curve-fit computer program was used to perform a regression analysis that described the course of the efflux rate index in the control period (25 min) before agonist stimulation. The equation of a hyperbola was used in the computer program { y = [c/(x  b)] + a} as described previously (2, 3). Regression lines were calculated individually from the control periods before intervention and were extrapolated into the stimulation period (Fig. 1A). In this way, every single heart was used as its own control. The response was expressed as a percentage of the extrapolated control index during the corresponding part of the washout period (Fig. 1B). Maximal response in each individual experiment was determined as the maximal increase in percentage irrespective of time within the 30-min observation period.

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Effects of bumetanide on 1-adrenoceptor-stimulated 86Rb+ efflux. A: 86Rb+ efflux rate index in rat hearts exposed to 1-adrenoceptor stimulation by phenylephrine (30 µmol/l) in the presence of 1 µmol/l timolol. Each data point represents mean values of 10 separate hearts. Hearts were exposed to phenylephrine in absence or presence of 50 µmol/l bumetanide (Bum). Values are means + SE (without Bum) and means  SE (with Bum). Regression lines given by the equation y = [0.66/(x + 84.2)] + 0.019 in absence and y = [1.21/(x + 116.1)] + 0.016 in presence of bumetanide, respectively. Regression lines are constructed from data points during 15-min period starting after a 10-min washout. Lines are extrapolated into the next 30-min observation time for comparison with the actually observed data. Lines are constructed from mean values shown, whereas separate regression lines for each experiment were used in actual calculations done (see text for details concerning curve-fit calculations). 86Rb+ efflux rate index (min1) was calculated by the following: cpm (86Rb+) perfusate × min1/cpm (86Rb+) heart. B: increase in 86Rb+ efflux rate index in percentage of basal efflux rate. Results from phenylephrine-stimulation experiments in A are expressed as increase in percentage of calculated control curves. Arrow, addition of phenylephrine.

Comparison of ⁸⁶Rb⁺ and ⁴²K⁺ efflux rate: effect of bumetanide. Because of recent concern (16) about the use of ⁸⁶Rb⁺ as a K⁺ analog in the study of bumetanide-sensitive efflux, dual-isotope experiments comparing the effect of bumetanide on the phenylephrine-stimulated increase of ⁸⁶Rb⁺ and ⁴²K⁺ efflux were done. These hearts were loaded with both isotopes for 30 min (cpm ⁸⁶Rb⁺:⁴²K⁺ ratio 1:3 in the perfusion medium) and were stimulated by 30 µmol/l phenylephrine after 50 min of washout. When used, bumetanide was added after 20 min of washout. Total radioactivity of the perfusate and the heart tissue was determined immediately after the end of the experiments following the same procedures as described above. After decay of ⁴²K⁺ (⁴²K⁺ activity within background level), the radioactivity in the samples was determined again, and the efflux rate indexes for ⁸⁶Rb⁺ and ⁴²K⁺ could be calculated after half-time corrections as described by several authors (2, 16). There was no statistically significant difference between the relative increase (in percentage of basal efflux values) in ⁸⁶Rb⁺ and ⁴²K⁺ efflux rate index after phenylephrine stimulation or the effect of bumetanide on the stimulated ⁸⁶Rb⁺ and ⁴²K⁺ efflux (Table 1).

View this table: [in this window] [in a new window]   Table 1.   Increase in 86Rb+ and 42K+ efflux rate index after 1-adrenoceptor stimulation with Phe in presence or absence of bumetanide

Measurement of ERK activity. ERK activity was measured in isolated hearts perfused in the presence or absence of phenylephrine and the presence or absence of different protein kinase inhibitors. Phenylephrine response in the presence of inhibitors was compared with separate control groups, which included the same inhibitors. After 5 min of phenylephrine stimulation, the hearts were freeze-clamped in liquid nitrogen, pulverized, and stored in plastic vials at 80°C in the same manner as described for the ⁸⁶Rb⁺ efflux studies, until ERK activity measurement was done essentially as described previously (53). The heart tissue powder (8 mg/ml buffer) was homogenized in a buffer containing 25 mmol/l tris(hydroxymethyl)aminomethane (Tris), pH 7.4, 25 mmol/l NaCl, 1.5 mmol/l EGTA, 1 mmol/l phenylmethylsulfonyl fluoride (PMSF), 1 mmol/l benzamidine, 1 µg/ml leupeptin, 1 µg/ml pepstatin A, 2 mmol/l dithiotreitol, 1 mmol/l NaVO₃, and 10% (vol/vol) ethylene glycol. The homogenate was centrifuged (15,000 g) for 10 min. The supernatant was mixed with phenyl-Sepharose and washed twice with each of the buffers containing 10 and 35% ethylene glycol before elution of the ERK with 60% ethylene glycol (4, 49). The eluate was assayed for ERK activity in the presence of an inhibitor of protein kinase A (protein kinase inhibitor, Sigma P-0300) using myelin basic protein (MBP) as the substrate. The reaction mixture (45 µl including 10 µl of eluate) contained 44 µmol/l ATP (with 50 µCi/ml of [-³²P]ATP), 9 mmol/l MgCl₂, and 0.44 mg/ml MBP. The enzyme incubation was performed at 30°C, and the reaction was stopped after 10 min with the addition of 0.9 mmol/l ATP and 90 µmol/l sodium pyrophosphate. The reaction mixture was spotted onto P81 paper (Whatman, Maidstone, UK); the papers were washed several times in phosphoric acid, dried, and counted in a liquid scintillation counter. Protein was measured with the bicinchoninic acid protein assay (Pierce, Rockford, IL).

Immunoblotting. Hearts were perfused as in Heart perfusion in ⁸⁶Rb⁺ efflux studies for 5 min with or without phenylephrine in the presence or absence of PD-98059 (10 µmol/l). Frozen heart samples were homogenized and boiled for 5 min in a 0.063 mol/l Tris · HCl buffer, pH 6.8, containing 10% glycerol, 3% SDS, 0.1 mmol/l EDTA, 0.1 mmol/l PMSF, 1 mmol/l NaVO₃, and 2.5% -mercaptoethanol. Aliquots with 20 µg cell protein were electrophoresed on 10% (wt/vol) polyacrylamide gels (acrylamide:N'N'-bis-methylene acrylamide 30:0.8) followed by electrotransfer of proteins to a nitrocellulose membrane (22) and incubation of the membranes with phospho-specific ERK antibodies against phosphorylated tyrosine (New England Biolabs, Beverly, MA) or against the dually phosphorylated threonine-tyrosine region (Promega, Madison, WI). Immunoreactive bands were visualized with enhanced chemiluminescence Western blotting detection reagents (Amersham International, Amersham, UK).

Preparation of rat papillary muscle. Rat heart papillary muscles were isolated as described previously with minor modifications (11). During coronary perfusion with the physiological salt solution mentioned below, left ventricular papillary muscles were excised and mounted in an organ bath with 18 ml of a slightly different physiological salt solution. During coronary perfusion, the solution contained the following (mmol/l): 118.3 NaCl, 3.0 KCl, 0.5 CaCl₂, 4.0 MgSO₄, 2.4 KH₂PO₄, 24.9 NaHCO₃, and 10.0 glucose and was continuously gassed with 95% O₂-5% CO₂ at 31°C (pH = 7.4). During incubation in the organ bath, the salt solution was identical to the salt solution used in the ⁸⁶Rb⁺ efflux studies (see Heart perfusion in ⁸⁶Rb⁺ efflux studies), except for mannitol, which was not included in the papillary muscle studies. The muscles were driven electrically, and tension was recorded as described previously (11). The direct current (DC) signals were analog-to-digital (AD) converted, logged, and stored in data files by software developed for the purpose in the visual programming language LabVIEW. Areas representative for the control (basal) period and the periods with ₁-adrenergic stimulation could be selected to calculate averaged contraction-relaxation cycles, which were representative for these periods. These cycles were then used to determine values for typical descriptive parameters, including maximal developed tension (T_max) and maximal development of tension (T'_max).

Experimental design of papillary muscle studies. The papillary muscles were allowed to equilibrate for 60 min before the experiments were started. The salt solution was changed after 45 min of equilibration. Bumetanide, chelerythrine, or PD-98059 when used, was added to the muscles 15 min before phenylephrine in order to study the effect of the inhibitors on the basal contractile force. The muscles in the concentration-response study groups were exposed to cumulatively increasing concentrations of the ₁-adrenoceptor agonist phenylephrine in the presence or absence of bumetanide, chelerythrine, or PD-98059, until the maximal response was obtained to evaluate the effect of the inhibitors on the concentration-response relationship to ₁-adrenoceptor stimulation. The effect of bumetanide on the inotropic response to phenylephrine was also evaluated by comparing the time course of the inotropic response in the presence or absence of bumetanide and, in separate experiments, by adding bumetanide after the maximal response to phenylephrine was achieved.

Calculations and statistics. Responses to agonist in the presence of different blockers were generally calculated as the percentage of the maximal response to agonist alone. The effect of each blocker was (if not noticed) compared with separate groups of agonist stimulation in the absence of blockers. The dose-response curves from the papillary muscle experiments were constructed according to Ariëns and Simonis (6), by estimating centiles (ED₁₀ to ED₁₀₀) for each single experiment and calculating the corresponding means. This calculation provides mean curves that express the response as a fractional response or as a percentage of maximum and display horizontal positioning and the correct mean slope of the curves. Horizontal positioning of the dose-response curves was expressed by pD₂ values (pD₂ = log ED₅₀, where ED₅₀ is the concentration dose giving half-maximal effect). The responses to phenylephrine were also recalculated and expressed as the percentage of control in order to yield curves that display differences in maxima as well.

The significance levels of differences between two groups were expressed by calculating P according to Student's one-sample test or two-sample test as appropriate. One-way analysis of variance (ANOVA), with Bonferroni's multiple- comparison test, was used to calculate P values when more than two groups were compared. A value of P  0.05 was considered to reflect statistically significant differences.

Materials. Phenylephrine hydrochloride, mannitol, furosemide, 4-aminopyridine, ouabain, staurosporine, N-ethylmaleimide, EGTA, PMSF, benzamidine, leupeptin, pepstatin, dithiotreitol, ethylene glycol, MBP, ATP, and sodium pyrophosphate were purchased from Sigma Chemical (St. Louis, MO). NaVO₃ was from Fluka Chemie (Buchs, Switzerland). Phosphoric acid was supplied from Riedel-deHSen (Seeize, Germany). Phenyl-Sepharose was from Pharmacia (Sweden). [-³²P]ATP (sp act 110 TBq/mmol), chelerythrine chloride, and KN-62 {1-[N,O-bis(5-isoquinolinesulfonyl)-N-methyl-L-tyrosyl]-4-phenyl piperazine} were from Biomol Research Laboratories (PA) or Calbiochem-Novabiochem (chelerythrine chloride) (San Diego, CA). PD-98059 [2-(2-amino-3-methoxyphenyl)-4H-1-benzopyran-4-one] came from New England Biolabs (Beverly, MA). HOE-694 and glibenclamide were kind gifts of Hoechst (Frankfurt, Germany). HOE-694 [(3-methylsulfonyl-4-piperidinobenzoyl)-guanidine methane sulfonate] was made available through Dr. W. Scholz. Bumetanide was the kind gift of P. Rasmussen at Leo Pharmaceutical Products (Denmark). Secondary antibodies (goat anti-rabbit horseradish peroxidase-labeled antibodies) were from Bio-Rad Lab (Richmond, CA). Stock solutions were prepared in double-distilled water and kept at 20°C. Further dilutions of the drugs were made fresh daily and kept in a cool (0-4°C) and dark atmosphere. ⁸⁶Rb⁺ (sp act in the range of 18.5-370 GBq/mg Rb) was purchased from DuPont or Amersham International (UK). ⁴²K⁺ (sp act of 5.8 GBq/mg K) was from the Danish Atomic Energy Commission Isotope Laboratory (Risø, Denmark).

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

₁-Adrenoceptor-mediated increase in ⁸⁶Rb⁺ efflux. The basal ⁸⁶Rb⁺ efflux rate decreased with time during the washout period, and the rate index was 3.2 ± 0.05 × 10² per minute (n = 10, Fig. 1A) after 25 min of washout. After phenylephrine stimulation, the rate index deviated from the control curve within a few minutes and reached a maximum value after 20 ± 1.4 min (n = 10). Phenylephrine (30 µmol/l) increased the ⁸⁶Rb⁺ efflux by 29 ± 1.6% (n = 36, P < 0.0001) compared with the control values (Fig. 1B). In the experimental series involving protein kinase inhibitors, the mean response to phenylephrine stimulation was 45 ± 1.7% (n = 21).

Effects of cation-chloride cotransport inhibition on stimulated ⁸⁶Rb⁺ efflux. To investigate whether the stimulated efflux of ⁸⁶Rb⁺ could involve an electroneutral cation-chloride cotransport mechanism, the experiments were performed in the absence or presence of bumetanide. The response was reduced to 73 ± 14.2% (n = 4) and to 58 ± 5.4% (n = 10, P < 0.01) of the maximal response in the presence of 5 and 50 µmol/l bumetanide, respectively (Figs. 1 and 2). A higher concentration of bumetanide (100 µmol/l) gave no further reduction in the response (Fig. 2B). The maximal increase in efflux rate index was 0.57 ± 0.073 × 10² per minute (n = 10) and 0.33 ± 0.043 × 10² per minute (n = 10) in the absence or presence of bumetanide (50 µmol/l), respectively (Fig. 1A, P < 0.01). The time to reach half-maximal response was reduced from 9 ± 1.1 to 5 ± 0.7 min (n = 10, Fig. 1B, P = 0.01) by 50 µmol/l bumetanide. The effect of bumetanide on the basal ⁸⁶Rb⁺ efflux was studied in separate experiments. The basal rate index decreased after the addition of bumetanide and was 92 ± 0.6% of the corresponding control value (n = 10, P < 0.0001) after 7 min. To compare the potency of bumetanide and furosemide for the stimulated ⁸⁶Rb⁺ efflux, phenylephrine stimulation in the presence of increasing concentrations of furosemide was performed (Fig. 2B). Furosemide reduced the response in a similar way as bumetanide did but at higher concentrations of antagonists. The response was reduced to 74 ± 7.6 (n = 7) and 56 ± 4.2% (n = 4, P < 0.01) of the maximal response in the presence of 50 and 100 µmol/l furosemide, respectively (Fig. 2B).

View larger version (27K): [in this window] [in a new window]   Fig. 2.   A: effect of Bum and 4-aminopyridine (4-AP) on 1-adrenoceptor-mediated increase in 86Rb+ efflux rate index (in presence of 1 µmol/l timolol). Groups of hearts were subjected to stimulation by phenylephrine (Phe, 30 µmol/l, n = 16) alone or in presence of 4-AP (n = 7), Bum (50 µmol/l, n = 16), or both 4-AP and Bum (Bum + 4-AP, n = 6). Responses in presence of 4-AP and Bum were compared with responses of separate groups exposed to Phe only (for 4-AP and Bum) and Phe + Bum (for 4-AP + Bum), which are shown as one group each. Response was expressed as increase of 86Rb+ efflux rate index in percentage of maximal response to phenylephrine stimulation in absence of inhibitors. Values are given as mean values + SE. * P  0.01 vs. Phe alone; ** P = 0.01 vs. Phe + Bum. B: concentration-response relationship between Phe-induced 86Rb+ efflux and Bum and furosemide. Figure shows that Bum reduced increased 86Rb+ efflux with higher potency than furosemide (n = 4-10 separate hearts/group). *P  0.01 vs. Phe alone.

Effects of Na⁺-K⁺ pump inhibition on stimulated ⁸⁶Rb⁺ efflux in presence or absence of cation-chloride cotransport inhibition. ₁-Adrenoceptor stimulation under certain conditions stimulates the Na⁺-K⁺ pump (46), and changes in Na⁺-K⁺ pump activity might indirectly affect the ⁸⁶Rb⁺ efflux. The phenylephrine-stimulated increase in ⁸⁶Rb⁺ efflux in the presence of 10 µmol/l ouabain was 82 ± 10.6% (n = 10) of the maximal response to phenylephrine in the absence of ouabain and did not represent a statistically significant reduction (P = 0.17). Higher concentrations of ouabain resulted in heart arrythmias and cardiac arrest. In the presence of ouabain (10 µmol/l), the addition of bumetanide (50 µmol/l), however, gave only a slight decrease in the stimulated ⁸⁶Rb⁺ efflux compared with the effect of ouabain alone [87 ± 9.1%, n = 14, vs. 100 ± 12.9%, n = 16, not significant (NS)] in contrast to the reducing effect of bumetanide alone (see Effects of cation-chloride cotransport inhibition on stimulated ⁸⁶Rb⁺ efflux). The response in the presence of both ouabain and bumetanide was, however, significantly different from the response to phenylephrine in the absence of inhibitors (71 ± 6.0% of maximal response without inhibitors, n = 14, P = 0.002). The reducing effect of bumetanide (50 µmol/l) on the basal ⁸⁶Rb⁺ efflux rate index was unchanged by the presence of 10 µmol/l ouabain. The basal efflux rate index dropped to 90 ± 1.3% (n = 4, P < 0.0001) of the corresponding control value in the presence of ouabain when bumetanide was added 15 min after addition of ouabain.

Effect of K⁺ channel inhibition on stimulated ⁸⁶Rb⁺ efflux. A possible role of K⁺ channels in the ₁-adrenoceptor-stimulated efflux was investigated by the use of two different K⁺ channel blockers. When 0.1 mmol/l 4-aminopyridine was added in addition to bumetanide, the bumetanide-insensitive part of the response was reduced by 55 ± 17.3% (Figs. 2A and 3). Increasing the concentration of 4-aminopyridine to 0.3 mmol/l did not reduce the response any further (data not shown), whereas 1 mmol/l 4-aminopyridine resulted in heart arrythmias. In the presence of bumetanide, the time to reach half-maximal response was not changed by 4-aminopyridine (2.7 ± 0.26 min, n = 6, vs. 3.3 ± 0.41 min, n = 6, with or without 4-aminopyridine, respectively; NS). When 4-aminopyridine was added in the absence of bumetanide, only a small reduction in the phenylephrine-stimulated ⁸⁶Rb⁺ efflux was seen (11 ± 6.2% reduction, P = 0.05, Fig. 2A).

View larger version (35K): [in this window] [in a new window]   Fig. 3.   Effect of 4-AP (0.1 mmol/l) in presence of Bum (50 µmol/l) on Phe-stimulated increase in 86Rb+ efflux rate index (in presence of 1 µmol/l timolol). Data points are means + SE (without 4-AP) or means  SE (with 4-AP) of 6 separate hearts and shown for every 10 value. Arrow, addition of agonist.

The addition of 3 µmol/l glibenclamide to block the K⁺_ATP channels did not influence the phenylephrine-mediated increase in ⁸⁶Rb⁺ efflux (27 ± 1.9%, n = 6, vs. 25 ± 2.1%, n = 8, without and with glibenclamide, respectively).

Effect of Na⁺/H⁺ exchanger inhibition on stimulated ⁸⁶Rb⁺ efflux. Possible indirect effects on the cation-chloride cotransporter and/or the K⁺ channels following activation of the Na⁺/H⁺ exchanger by phenylephrine were studied by the use of the Na⁺/H⁺ exchange inhibitor HOE-694. Neither the phenylephrine-mediated increase in ⁸⁶Rb⁺ efflux nor the inhibition by bumetanide was affected by 10 µmol/l HOE-694 (Fig. 4). Phenylephrine increased the ⁸⁶Rb⁺ efflux rate index by 28 ± 2.7 (n = 6) and 27 ± 2.1% (n = 6) in the presence or absence of HOE-694, respectively.

View larger version (34K): [in this window] [in a new window]   Fig. 4.   Effect of Na+/H+ inhibition by HOE-694 (HOE) in presence or absence of Bum on 1-adrenoceptor-mediated increase in 86Rb+ efflux rate index (in presence of 1 µmol/l timolol). Groups of hearts were subjected to stimulation by Phe (30 µmol/l, n = 6) alone or in presence of HOE (10 µmol/l, n = 6), Bum (50 µmol/l, n = 7), or both HOE and Bum (n = 10). Response was expressed as increase of 86Rb+ efflux rate index in percentage of maximal response to Phe stimulation in absence of inhibitors. Values are given as means + SE.

Effects of cation-chloride cotransport inhibition on inotropic response to phenylephrine. Possible effects of bumetanide on the inotropic response to phenylephrine stimulation might influence the ⁸⁶Rb⁺ efflux and thus possibly explain the inhibitory effects of bumetanide on phenylephrine-stimulated ⁸⁶Rb⁺ efflux. Control experiments in rat papillary muscles elucidated this aspect. In concentration-response experiments, the development of force expressed as T'_max increased by 35 ± 6.5 (n = 9) and 36 ± 3.2% (n = 9) after stimulation by phenylephrine in the presence or absence of bumetanide, respectively. Phenylephrine increased the T'_max with a pD₂ value of 4.9 ± 0.21 (n = 9) and 5.0 ± 0.18 (n = 9) in the presence or absence of bumetanide, respectively (Fig. 5). Bumetanide (50 µmol/l) did not influence the basal force development (1.7 ± 14% increase after bumetanide addition n = 12) and did not influence the phenylephrine-stimulated force development (1.0 ± 18% decrease, after addition of bumetanide n = 10). The time course of the response was evaluated with a phenylephrine concentration close to maximum (30 µmol/l). The time to half-maximal response after phenylephrine stimulation tended to be reduced by the presence of bumetanide, but the difference was not statistically significant (145 ± 11.3 s, n = 12, and 181 ± 25.7 s, n = 10, with or without bumetanide, respectively).

View larger version (20K): [in this window] [in a new window]   Fig. 5.   Concentration-response curve for inotropic response expressed as percentage increase of maximal rate of contraction (T'max) after 1-adrenoceptor stimulation in rat papillary muscle. Stimulation by Phe (in presence of 1 µmol/l timolol) in absence or presence of Bum (n = 9/group). Phe was the agonist used for log molar concentration. Means ± SE are given at half-maximal effect.

Effect of protein kinase inhibitors on stimulated ⁸⁶Rb⁺ efflux. Different inhibitors were used to elucidate the involvement of protein kinases in the ₁-adrenoceptor-stimulated ⁸⁶Rb⁺ efflux. The nonspecific protein kinase inhibitor staurosporine reduced the response to phenylephrine by 37 ± 5.5 and 76 ± 16.6% at 30 and 100 nmol/l, respectively (Fig. 6). The selective PKC inhibitor chelerythrine showed no inhibitory effect at concentrations up to 10 µmol/l (Fig. 6), and higher concentrations (50 µmol/l) resulted in cardiac arrest. The CaM-kinase II inhibitor KN-62 (1-3 µmol/l) also did not influence the response to phenylephrine (Fig. 6). To study the possible involvement of the ERK pathway, an inhibitor of the activation of MEK-1, PD-98059 (10 µmol/l), was used. This compound reduced the phenylephrine-evoked increase in ⁸⁶Rb⁺ efflux by 33 ± 6.4% (Figs. 6 and 7). We further tested whether the bumetanide-sensitive part of the response to phenylephrine was dependent on protein kinase activation. Staurosporine (30 nmol/l) and PD-98059 (10 µmol/l) reduced the bumetanide-sensitive part of the stimulated ⁸⁶Rb⁺ efflux to 27 ± 17.6 and 25 ± 12.6% of maximal response, respectively (Fig. 8A). The presence of chelerythrine did not influence the bumetanide-sensitive component of the phenylephrine response (Fig. 8A).

View larger version (34K): [in this window] [in a new window]   Fig. 6.   Effect of various protein kinase inhibitors on 1-adrenoceptor-mediated increase in 86Rb+ efflux rate index (in presence of 1 µmol/l timolol). Groups of hearts were subjected to stimulation by Phe (30 µmol/l, n = 21) alone or in presence of 30 nmol/l staurosporine [Stauro (30), n = 13], 100 nmol/l staurosporine [Stauro (100), n = 4], 10 µmol/l chelerythrine (Chel, n = 8), 3 µmol/l KN-62 (KN, n = 3), or 10 µmol/l PD-98059 (PD, n = 5). Response was expressed as increase of 86Rb+ efflux rate index in percentage of basal. Response in presence of inhibitors was compared with separate groups of Phe, which are shown as one group. Values are given as means + SE. * P  0.002 vs. Phe alone.

View larger version (33K): [in this window] [in a new window]   Fig. 7.   Effect of mitogen-activated protein (MAP) kinase/ extracellular signal-regulated kinase (ERK) (MEK) activation inhibitor PD-98059 (10 µmol/l) on Phe-stimulated increase in 86Rb+ efflux rate index (in presence of 1 µmol/l timolol). Data points represent means + SE (without PD-98059; ) or means  SE (with PD-98059; ) of 5 separate hearts.

View larger version (28K): [in this window] [in a new window]   Fig. 8.   Effect of protein kinase inhibitors on 1-adrenoceptor-mediated activation of Na-K-2Cl cotransport (A) and activation of MAP kinase (ERK) pathway (B). Effects are measured as responses in percentage of maximal responses to Phe (30 µmol/l, in presence of 1 µmol/l timolol) in absence of inhibitors. Protein kinase inhibitors: 10 µmol/l PD-98059, 30 or 100 nmol/l Stauro, 10 µmol/l Chel. A: Na-K-2Cl cotransport activation is measured as bumetanide-sensitive increase in 86Rb+ efflux rate index. Maximal response (100%) represents increase in Bum-sensitive part (42%) of 86Rb+ efflux. B: ERK-1/ERK-2 activity measured as increase in MBP kinase activity. Maximal response (100%) represents 340% increase in ERK-1/ERK-2 activity. Myelin basic protein (MBP) activity was measured in perfused hearts stimulated by Phe for 5 min. Each experimental condition consisted of 4-10 separate hearts in A and 3 separate hearts per group in B. Values are given as means + SE. * P < 0.01 compared with effect of Phe in absence of inhibitors. Response in presence of inhibitors was compared with separate groups of Phe, which are shown as one group.

Effect of protein kinase inhibitors on phenylephrine-stimulated ERK activity. Phenylephrine (30 µmol/l) increased the ERK activity (measured as MBP kinase activity) by 239 ± 37% above control (n = 3/group, P = 0.007). The phenylephrine-stimulated increase in ERK activity was reduced by 80 ± 16.2% in the presence of PD-98059 (10 µmol/l, Fig. 8B). Immunoblotting experiments using separate hearts were used to confirm that the MBP kinase activity represented ERK activity. Phosphorylation of both ERK isoforms, ERK-2 (p42^mapk) and ERK-1 (p44^mapk), were increased in hearts stimulated by phenylephrine for 5 min (Fig. 9) compared with control hearts. The presence of PD-98059 (10 µmol/l) reduced the phenylephrine-stimulated phosphorylation of both p42^mapk and p44^mapk (Fig. 9).

View larger version (17K): [in this window] [in a new window]   Fig. 9.   Effect of Phe ± PD on phosphorylated ERK-2 (p42mapk) and ERK-1 (p44mapk). Hearts were stimulated by Phe for 5 min. Each lane represents separate groups of hearts (n = 3 hearts for each experimental condition) different from hearts used in experiments shown in Fig. 8B. Results obtained with ERK specific anti-phosphotyrosine antibodies are shown. Similar results were obtained by using antibodies against the dually phosphorylated threonine/tyrosine region (data not shown).

The phenylephrine-stimulated ERK activity was not reduced by chelerythrine (10 µmol/l, Fig. 8B), whereas 30 and 100 nmol/l staurosporine reduced the stimulated ERK activity dose dependently (33.5 ± 16.6 and 89.8 ± 15.6% reduction, respectively, Fig. 8B).

Effect of ERK cascade inhibitor PD-98059 and the PKC inhibitor chelerythrine on inotropic response to phenylephrine. If PD-98059 had effects on the inotropic response to phenylephrine, such effects could per se influence both the phenylephrine-stimulated ⁸⁶Rb⁺ efflux and the stimulation of the ERK cascade. Control experiments in rat papillary muscles were performed to study the effect of PD-98059 both on basal force and on the inotropic response. No effect of PD-98059 was, however, found on the inotropic response to phenylephrine in papillary muscles. PD-98059 did not affect basal force development (3.5 ± 3.6% increase, n = 6), the maximal response (31 ± 6.4 and 39 ± 5.5% increase without and with PD-98059, respectively, n = 6 per group), or the dose-response relationship between phenylephrine and contractile force significantly (pD₂ value = 5.4 ± 0.08 and 5.4 ± 0.10 with and without PD-98059, respectively, Fig. 10B). The effect of chelerythrine on the inotropic response to phenylephrine was studied using the same batch of chelerythrine that was used in the ⁸⁶Rb⁺ efflux studies and the ERK measurement experiments to test whether the drug was active. Phenylephrine stimulation in the presence of 15 µmol/l chelerythrine resulted in a negative inotropic response (15 ± 3.5% decrease, n = 7, P = 0.003, Fig. 10A) at low concentrations and a positive inotropic response at higher concentrations (Fig. 10A). The positive component of the response per se was unaffected by chelerythrine, because neither the maximal response (29 ± 3.4, n = 8) nor the pD₂ value (5.25 ± 0.12, NS, Fig. 10B) differed from the results with phenylephrine without the inhibitor.

View larger version (19K): [in this window] [in a new window]   Fig. 10.   Concentration-response curve for inotropic response expressed as T'max after 1-adrenoceptor stimulation in rat papillary muscle. Stimulation by Phe (in presence of 1 µmol/l timolol). A: stimulation by Phe in absence or presence of 10 µmol/l PD-98059 or 10 µmol/l Chel (n = 8-9/group). Response to increasing concentrations of Phe in presence of Chel consisted of a negative component at low concentrations () and a positive component at high concentrations (). Negative and positive components of response in presence of Chel were calculated separately. Phe was the agonist for the log molar concentration. Means ± SE are given at maximal negative and positive response, respectively. B: stimulation by Phe in absence or presence of 10 µmol/l PD-98059 or 10 µmol/l Chel; n = 8-9/group. Means ± SE are given at half-maximal effect. Phe was the agonist for the log molar concentration.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

The main findings in this study are that ₁-adrenoceptor stimulation of the rat heart increase ⁸⁶Rb⁺ efflux through an activation of a bumetanide-sensitive cation-chloride cotransporter, in addition to increased efflux via 4-aminopyridine-sensitive K⁺ channels. The activation of a cation-chloride cotransporter was apparently dependent on the ERK cascade.

Involvement of a cation-chloride cotransport mechanism. In the present study, phenylephrine increased the ⁸⁶Rb⁺ efflux by ~30-40%, which is consistent with previous results from our laboratory (3, 21). Such a large movement of potassium out of the cells must involve electrobalanced transport mechanisms. The ₁-adrenoceptor-mediated increase in ⁸⁶Rb⁺ efflux was partially bumetanide sensitive, indicating an activation of a cation-chloride cotransporter by ₁-adrenoceptor stimulation in the rat heart. The time course of the responses in the presence and absence of bumetanide suggests that the total response is mainly mediated by a fast bumetanide-insensitive component and a slower bumetanide-sensitive component reflecting activation of a cation-chloride cotransport mechanism (Fig. 1B). This is illustrated by the time course of the bumetanide-sensitive part (i.e., difference between the curves in Fig. 1B).

The cation-chloride cotransporter involved in the response to phenylephrine could either be Na⁺ dependent (Na-K-2Cl cotransport) or Na⁺ independent (K-Cl cotransport). Both cotransporter proteins have been cloned (19, 57). The beating heart preparation makes it difficult to show whether the transport mechanism involved is Na⁺ dependent or not, because the intact heart preparation would not tolerate the removal of Na⁺. The K-Cl cotransporter can also be distinguished functionally from the Na-K-2Cl cotransporter by its 100-fold lower affinity for bumetanide compared with the Na⁺-dependent cotransporter (19, 27). We found a relatively high sensitivity to bumetanide (Fig. 2B) and no further inhibition at concentrations above 50 µmol/l, whereas bumetanide usually inhibits the K-Cl cotransporter at 100 µmol/l or higher concentrations (19), indicating that the cotransporter involved in the present study is the Na⁺-dependent transporter. The K-Cl cotransporter and the Na-K-2Cl cotransporter have different affinities for furosemide and bumetanide (19). Bumetanide is a more potent inhibitor of the Na-K-2Cl cotransporter than furosemide, whereas furosemide is the most potent inhibitor of the K-Cl cotransporter (18, 19). We found that bumetanide reduced the ₁-adrenoceptor-mediated increase in ⁸⁶Rb⁺ efflux more potently than furosemide (Fig. 2B), and thus the order of potency further supports that the increased ⁸⁶Rb⁺ efflux is mediated by the Na⁺-dependent cotransporter, although a contribution from the K-Cl cotransporter cannot be excluded. The K-Cl cotransport activator N-ethylmaleimide could not be used because of the toxic effects of the compound on the beating heart. Although the Na-K-2Cl cotransporter in most cells produces a net influx and the K-Cl cotransporter produces a net efflux, the salt transport via the cotransporters is in principle bidirectional and is determined by the sum of the chemical gradients for each of the transported ions (25). Thus the presently reported increase in loop-diuretic-sensitive K⁺ efflux suggests an ₁-adrenoceptor-mediated activation of the Na-K-2Cl cotransporter without determining the net direction of salt transport. The effect of bumetanide on the basal ⁸⁶Rb⁺ efflux confirmed the presence of a Na-K-2Cl cotransporter mediating a part of the basal ⁸⁶Rb⁺ transport in the adult rat heart as reported in chick heart cells (18, 26, 31), rabbit ventricle (13), and cultured rat cardiac myocytes (39). The Na-K-2Cl cotransporter is regulated by a variety of hormones in addition to extracellular tonicity (25, 29). The present findings with an ₁-adrenoceptor agonist suggest adrenergic regulation of Na-K-2Cl cotransport in the heart, although the Na⁺ dependency of the cation-chloride cotransporter involved could not be definitely proven in this model.

Recently, different effects of bumetanide on ⁸⁶Rb⁺ and K⁺ fluxes were reported (16). In control experiments, comparing the stimulated efflux of ⁸⁶Rb⁺ and ⁴²K⁺, we found that the bumetanide-sensitive part was of the same magnitude for the two isotopes (Table 1). Thus ⁸⁶Rb⁺ can be used as a K⁺ analog when studying ₁-adrenoceptor-mediated changes in bumetanide-sensitive efflux in the rat heart. Studies of ionic selectivity of Na-K-2Cl cotransport in different tissues found that Rb⁺ substitutes well for K⁺ both in flux studies and in bumetanide binding studies (23).

The physiological implications of an activation of Na-K-2Cl cotransport by ₁-adrenoceptor stimulation in the heart are currently unknown. A possible role of the cotransporters in cell volume regulation has been suggested from studies on chick heart cells (26). Previous studies from our laboratory showed an effect of ₁-adrenoceptor stimulation on both ⁸⁶Rb⁺ influx and ⁸⁶Rb⁺ efflux, with a net increased K⁺ content, but decreased K⁺ concentration due to increased cell volume (20, 21). ₁-Adrenoceptor stimulation induces myocardial growth under certain conditions (10, 47), and one could speculate that a possible effect of Na-K-2Cl cotransport activation is growth stimulation via changes in cell volume. Bumetanide was shown to inhibit cell proliferation in both human skin fibroblasts and vascular endothelial cells (38, 40).

A possible negative effect on contractile function of Na-K-2Cl cotransport inhibition in the rabbit heart has been proposed (13). Possible effects of bumetanide on contractile force could affect the ⁸⁶Rb⁺ efflux, and the present studies on force development in rat papillary muscle were undertaken to elucidate the effects of bumetanide on force development and on the inotropic response to ₁-adrenoceptor stimulation. Na-K-2Cl cotransport inhibition by bumetanide affected neither the positive inotropic response to phenylephrine nor the basal force development in the papillary muscle studies. The magnitude of the response and the pD₂ value were consistent with earlier studies (48). The inhibitory effect of bumetanide on the ⁸⁶Rb⁺ efflux seems to be a direct effect and not an indirect effect involving changes in contractile function. The bumetanide-sensitive ₁-adrenoceptor-mediated increase in ⁸⁶Rb⁺ efflux must be involved in aspects of myocardial ₁-adrenoceptor effects, different from regulation of contractile force.

Effect of Na⁺-K⁺ pump inhibition. ₁-Adrenoceptor stimulation has been shown to stimulate the Na⁺-K⁺ pump current (46, 56) and increased K⁺ influx due to Na⁺-K⁺ pump stimulation might indirectly affect the efflux of ⁸⁶Rb⁺. This mechanism was studied by inhibiting the Na⁺-K⁺ pump with ouabain, although the experimental model tolerated only a relatively low concentration of ouabain known to inhibit only about 20% of the Na⁺-K⁺-ATPase (35). The results indicate that the increased ⁸⁶Rb⁺ efflux is not an indirect reflection of Na⁺-K⁺ pump stimulation. Ouabain reduced the inhibitory effect of bumetanide on the stimulated ⁸⁶Rb⁺ efflux. There is no obvious explanation how ouabain should affect the sensitivity to bumetanide without changing the total response to phenylephrine, but these observations might indicate an interaction between ouabain and bumetanide on ⁸⁶Rb⁺ efflux mediated by the cotransporter. Previous work from our laboratory showed that ₁-adrenoceptor stimulation increased the cell volume in isolated rat hearts by 17% (P = 0.0006) (20). The presence of ouabain (10 µmol/l) gave a small nonstatistically significant increase in cell volume and abolished an increase during ₁-adrenoceptor stimulation. It is possible that the effect of ouabain in the present work is due to the influence of ouabain on the cell volume changes. A furosemide-sensitive K⁺ efflux activated by ouabain was found in chick heart cells (26), and ouabain was found to stimulate a bumetanide-sensitive ⁸⁶Rb⁺ influx in rat cardiac cells at concentrations (0.1 µmol/l) too low to affect the Na⁺-K⁺ pump (39).

Involvement of K⁺ channels. 4-Aminopyridine reduced the stimulated ⁸⁶Rb⁺ efflux in the presence of bumetanide; thus the K⁺ channels are apparently involved in addition to a cotransport mechanism in mediating the ₁-adrenoceptor-evoked response. Apparently, ₁-adrenoceptor stimulation activates a 4-aminopyridine-sensitive transport of ⁸⁶Rb⁺ out of the cells in addition to reducing outward current (17). These results are not contradictory because of the differences in measuring total ionic fluxes and electrical currents. 4-Aminopyridine is a nonselective blocker of different voltage-dependent and receptor-coupled K⁺ channels (55a). The use of maximal concentrations of 4-aminopyridine was prevented by the occurrence of heart arrythmias and could explain why the inhibition of the bumetanide-insensitive part was incomplete. The fact that 4-aminopyridine only had a minor effect on the increased efflux in the absence of bumetanide could reflect that the bumetanide-sensitive translocation mechanism is partly able to compensate for the inhibition of the channels. It is also possible that 4-aminopyridine, in the presence of phenylephrine, could affect the cell volume. A change in cell volume would influence the Na-K-2Cl cotransport activity and thus the increased ⁸⁶Rb⁺ efflux.

The K⁺_ATP channel is considered to mediate K⁺ efflux in early myocardial ischemia and hypoxia. The increase in ₁-adrenoceptor-mediated ⁸⁶Rb⁺ efflux was not sensitive to the K⁺_ATP channel inhibitor glibenclamide (7), indicating that these channels are not involved in the response to phenylephrine during normoxic conditions.

Effect of Na⁺/H⁺ exchanger inhibition. Stimulation of myocardial ₁-adrenoceptors activates the Na⁺/H⁺ exchanger (43). Stimulation of the Na⁺/H⁺ exchanger [resulting in increased intracellular Na⁺ and Ca²⁺ concentration via Na⁺/Ca²⁺ exchange] could affect K⁺ efflux via both (Ca²⁺ activated) K⁺ channels and bumetanide-sensitive Na-K-2Cl cotransport. To investigate a possible indirect role of the Na⁺/H⁺ exchanger in the ₁-adrenoceptor-mediated increase in ⁸⁶Rb⁺ efflux, we studied the influence of HOE-694, a selective inhibitor of the Na⁺/H⁺ exchanger (14), without any effects on the heart rate, contractile function, or electrocardiogram parameters (59). No effect of this drug was found at concentrations reported to inhibit the cardiac isoform (NHE-1) of the exchanger [inhibition constant (K_I) = 0.16 µmol/l (14)]. Thus the results show that ₁-adrenoceptor-mediated activation of a Na-K-2Cl cotransporter and K⁺ channels is not secondary to increased intracellular Na⁺ concentration as a result of activation of the Na⁺/H⁺ exchanger.

Involvement of ERK subgroup of MAP kinase family in activation of Na-K-2Cl cotransporter. We found that phenylephrine in an adult isolated heart preparation stimulated ERK activity by at least threefold, which is consistent with earlier reports from neonatal cardiomyocytes (9) and perfused hearts (28) with use of a different method for measuring ERK activity. The present results with the inhibitor PD-98059 indicate that the phenylephrine-evoked increase in ⁸⁶Rb⁺ efflux is dependent on activated ERK. PD-98059 is a selective inhibitor of MEK-1 activation, an upstream activator of the ERK with no effect on other kinases investigated so far (1). In the perfused heart, PD-98059 (10 µmol/l) inhibited both phenylephrine-stimulated increase in ⁸⁶Rb⁺ efflux and the increased ERK activity to about the same relative extent. The same concentration of PD-98059 was found to inhibit the phenylephrine-induced ERK activation in neonatal rat myocytes (41). The immunoblotting analysis confirmed the results on ERK activation, showing an increased level of phosphorylated p44^mapk (ERK-1) and p42^mapk (ERK-2), which was reduced by PD-98059. The presence of PD-98059 eliminated the inhibitory effect of bumetanide, thus indicating the involvement of the ERK in the activation of the Na-K-2Cl cotransporter. The inhibitory effects of PD-98059 were not secondary to changes in contractile force, since control experiments in papillary muscles showed no effect of the drug either on basal force development or on the inotropic response to phenylephrine stimulation.

The CaMK II inhibitor KN-62 (1 µmol/l) inhibited the spontaneous beating and intracellular Ca²⁺ transients in neonatal myocytes (37) but had no effect in the present study, indicating that the ₁-adrenoceptor-mediated increase in ⁸⁶Rb⁺ efflux is independent of the CaMK II.

The presence of the protein kinase inhibitor staurosporine and the PKC inhibitor chelerythrine affected the stimulated ⁸⁶Rb⁺ efflux differently. Staurosporine reduced the total response to phenylephrine and eliminated the bumetanide sensitivity, whereas chelerythrine had no effect on either the total increase in ⁸⁶Rb⁺ efflux or the bumetanide-sensitive part at concentrations that were tolerated by the isolated heart preparation. Higher concentrations of chelerythrine resulted in cardiac arrest, reflecting the concentration-related toxic effects of the compound. Chelerythrine at the same concentration (10 µmol/l) also influenced the inotropic response to phenylephrine. The lack of effect of chelerythrine on ⁸⁶Rb⁺ efflux and ERK activity, in contrast to staurosporine, could either indicate ineffective inhibition of PKC by chelerythrine, inhibition of different PKC isoforms, or that staurosporine also inhibits kinases other than the PKC. Chelerythrine is a selective inhibitor of PKC with an half-maximal inhibitory concentration of 0.6 µmol/l (24), and it blocked both ischemic preconditioning and ₁-adrenoceptor-mediated effects on the hearts at the concentrations used in the present study (12, 54). It also blocked phenylephrine- and phorbol 12-myristate 13-acetate (PMA)-induced activation of the Na⁺/H⁺ exchanger in cardiac myocytes (58). Staurosporine is a potent PKC inhibitor, but it will inhibit most protein kinases concentration dependently, including the MAP kinases at high concentrations (34). The marked increase in inhibitory effects of staurosporine when the concentration was increased to 100 nmol/l may thus result from inhibition of kinases other than the PKC [K_I for PKC inhibition = 0.7 nmol/l, (52)]. The results obtained with chelerythrine and staurosporine on the ERK measurement were consistent with the effect of the inhibitors on the stimulated ⁸⁶Rb⁺ efflux. The data obtained with PD-98059 and chelerythrine could suggest an ₁-adrenoceptor-mediated activation of the ERK-MAP kinase cascade, independently of PKC activation. Both PKC-dependent (51) and PKC-independent (15) activation of the MAP kinase pathway by Gq protein-coupled receptors (like ₁-adrenoceptors) have been reported. A possible role of a PKC-independent pathway, probably involving phosphatidylinositol 3-kinase and tyrosine kinases, linking Gq protein-coupled receptors to the MAP kinase cascade has been proposed (50, 55).

Recent evidence from noncardiac tissue has shown that modulation of the Na-K-2Cl cotransporter by stimulatory hormones involves direct phosphorylation of the transporter protein (33, 36). Regulation of the Na-K-2Cl cotransporter in epithelial tissue is mediated by a variety of protein kinases, including PKA, PKC, and CaM kinase (25, 30, 33, 36). Involvement of the MAP kinase pathway in the regulation of the Na-K-2Cl cotransporter has, to our knowledge, not been reported so far (32). Very little is known about the regulation of the myocardial Na-K-2Cl cotransporter, but involvement of Ca²⁺ and CaM kinase has been suggested (26). Stimulation of myocardial ₁-adrenoceptors activates PKC (42), CaM kinase II (45), and also the MAP kinase family (8, 51). The present results suggest that the ₁-adrenoceptor-mediated activation of the Na-K-2Cl cotransporter is dependent on the ERK subgroup of the MAP kinases.

An additional aspect of the results obtained with the protein kinase inhibitors is that ₁-adrenoceptor stimulation seems to regulate potassium fluxes and the inotropic response by different mechanisms. Whereas the Na-K-2Cl cotransporter activation seems to be dependent on the ERK cascade, the inotropic response to phenylephrine is apparently not dependent on this pathway.

In conclusion, ₁-adrenoceptor stimulation activates a Na-K-2Cl cotransporter and K⁺ channels in the rat heart. These translocation mechanisms mediate the major part of the ₁-adrenoceptor-mediated increase in ⁸⁶Rb⁺ efflux. The response is not dependent on activation of the Na⁺/H⁺ exchanger or the Na⁺-K⁺ pump. The activation of a myocardial Na-K-2Cl cotransporter is apparently dependent on the MAP kinase pathway.