Hypoxic dilation of coronary arteries is mediated by ATP-sensitive potassium channels.

The function of the heart depends critically on an adequate oxygen supply through the coronary arteries. Coronary arteries dilate when the intravascular oxygen tension decreases. Hypoxic vasodilation in isolated, perfused guinea pig hearts can be prevented by glibenclamide, a blocker of adenosine triphosphate (ATP)-sensitive potassium channels, and can be mimicked by cromakalim, which opens ATP-sensitive potassium channels. Opening of potassium channels in coronary smooth muscle cells and the subsequent drop in intracellular calcium is probably the major cause of hypoxic and ischemic vasodilation in the mammalian heart.

Long-chain acyl-coenzyme A esters and fatty acids directly link metabolism to K(ATP) channels in the heart.

ATP-sensitive K (K(ATP)) channels are inhibited by cytosolic ATP, a defining property that implicitly links these channels to cellular metabolism. Here we report a direct link between fatty acid metabolism and K(ATP) channels in cardiac muscle cells. Long-chain (LC) acyl-coenzyme A (CoA) esters are synthesized from fatty acids and serve as the principal metabolic substrates of the heart. We have studied the effects of LC acyl-CoA esters and LC fatty acids on K(ATP) channels of isolated guinea pig ventricular myocytes and compared them with the effects of phosphatidylinositol 4,5-bisphosphate (PIP(2)). Application of oleoyl-CoA (0.2 or 1 micromol/L), a naturally occurring acyl-CoA ester, to the cytosolic side of excised patches completely prevented rundown of K(ATP) channels, but not of Kir2 channels. The open probability of K(ATP) channels measured in the presence of oleoyl-CoA or PIP(2) was voltage dependent, increasing with depolarization. Oleoyl-CoA greatly reduced the ATP sensitivity of K(ATP) channels. At a concentration of 2 micromol/L, oleoyl-CoA increased the half-maximal inhibitory concentration of ATP >200-fold. The time course of the decrease in ATP sensitivity was much faster during application of oleoyl-CoA than during application of PIP(2). The effects of PIP(2), but not of oleoyl-CoA, were inhibited by increasing Ca(2+) to 1 mmol/L. Oleate (C18:1; 10 micromol/L), the precursor of oleoyl-CoA, inhibited K(ATP) channels activated by oleoyl-COA: Palmitoleoyl-CoA and palmitoleate (C16:1) exerted similar reciprocal effects. These findings indicate that LC fatty acids and their CoA-linked derivatives may be key physiological modulators of K(ATP) channel activity in the heart.

The effect of hyperosmolality on the rate of heat production of quiescent trabeculae isolated from the rat heart.

We have measured the rate of heat production of isolated, quiescent, right ventricular trabeculae of the rat under isosmotic and hyperosmotic conditions, using a microcalorimetric technique. In parallel experiments, we measured force production and intracellular calcium concentration ([Ca2+]i). The rate of resting heat production under isosmotic conditions (mean +/- SEM, n = 32) was 100 +/- 7 mW (g dry wt)-1; it increased sigmoidally with osmolality, reaching a peak that was about four times the isosmotic value at about twice normal osmotic pressure. The hyperosmotic thermal response was: (a) abolished by anoxia, (b) attenuated by procaine, (c) insensitive to verapamil, ouabain, and external calcium concentration, and (d) absent in chemically skinned trabeculae bathed in low-Ca2+ "relaxing solution." Active force production was inhibited at all osmolalities above isosmotic. Passive (tonic) force increased to, at most, 15% of the peak active force developed under isosmotic conditions while [Ca2+]i increased, at most, 30% above its isosmotic value. We infer that hyperosmotic stimulation of resting cardiac heat production reflects, in large part, greatly increased activity of the sarcoplasmic reticular Ca2+ ATPase in the face of increased efflux via a procaine-inhibitable Ca(2+)-release channel.

The sensitivity of coronary vascular tone to glibenclamide: a study on the isolated perfused guinea pig heart.

OBJECTIVE: The aim was to obtain an estimate of the affinity of glibenclamide binding to ATP sensitive potassium channels of coronary smooth muscle cells in the intact guinea pig heart, in order to assess possible cardiovascular risks for diabetic patients receiving sulphonylurea drugs. METHODS: The arrested isolated perfused heart was used as a bioassay for KATP channels in coronary resistance vessels. The decrease of coronary perfusion pressure induced by hypoxia and by various concentrations of the potassium channel opener levcromakalim was measured and the inhibitory effect of glibenclamide on hypoxic and drug induced vasodilatation was analysed. RESULTS: The delay between application of hypoxic solution and half maximum vasodilatation was on average 105(SEM 4) s. It was increased by 17.6(2.0) s in the presence of 5 nM glibenclamide. The dose-response curve of the effects of levcromakalim on coronary perfusion pressure was shifted to the right in the presence of 20 nM to 1 microM glibenclamide. The concentrations of levcromakalim required to produce half maximum vasodilatation in the presence and absence of glibenclamide were determined, and the data were analysed using a Schild plot. The measured concentration ratios could be fitted by a straight line with a slope of 0.93(SD 0.04), which was not significantly different from 1. This is consistent with competitive inhibition of the vasodilator effects of levcromakalim by glibenclamide. The apparent dissociation constant of glibenclamide binding to KATP channels in coronary smooth muscle cells was 35 nM. CONCLUSIONS: The KATP channels in coronary resistance arteries are almost as sensitive to inhibition by glibenclamide as the KATP channels in pancreatic beta cells. It is likely that the plasma concentrations reached after oral ingestion of glibenclamide may inhibit a fraction of the KATP channels in coronary smooth muscle cells and thus impair hypoxic dilatation of coronary resistance arteries.

A glibenclamide sensitive potassium conductance in terminal arterioles isolated from guinea pig heart.

OBJECTIVE: The aim was to isolate terminal arterioles from guinea pig heart, to describe their basic electrical properties, and to obtain evidence for the presence of KATP channels in microvascular coronary smooth muscle cells. METHODS: Terminal arterioles of 20-50 microns diameter were obtained by enzymatic digestion of isolated perfused hearts. The isolated arterioles were viable for up to 8 h and constricted upon application of high potassium solution. Whole cell clamp experiments on smooth muscle cells of these arterioles were performed at room temperature. RESULTS: The resting potential of coronary smooth muscle cells in terminal arterioles showed a bimodal distribution with one peak at -23(SD 8) mV and the other peak at -61(4) mV. Application of glibenclamide (50 microM) to the latter group of arterioles produced a depolarisation to -37(8) mV; application of cromakalim (1 microM) produced a hyperpolarisation to -71(1) mV. The current changes associated with voltage steps and slow voltage ramps in the range -120 to +40 mV indicated that the smooth muscle cells in the arterioles were coupled electrically. The steady state current-voltage relation was sigmoid with a flat region in the range -50 to -30 mV. In the presence of 2-50 microM glibenclamide the slope resistance at potentials negative to -50 mV and positive to -30 mV was markedly increased. In the presence of 1 microM cromakalim the slope resistance was decreased and the current-voltage relation at negative potentials became nearly linear. The crossover point of the current-voltage relations measured under control conditions and in the presence of glibenclamide was near the calculated potassium equilibrium potential. CONCLUSIONS: Glibenclamide closes and cromakalim opens potassium channels in smooth muscle cells of coronary arterioles. The voltage dependence of the steady state current changes suggests that the current activated by cromakalim is not carried by the same channels as the current inhibited by glibenclamide. The glibenclamide sensitive channels make a significant contribution to the membrane potential of isolated arterioles.

KATP channels and basal coronary vascular tone.

After reviewing recent experimental work from various laboratories we have come to the following conclusions. (1) An increase in transmural pressure causes depolarisation of coronary arterioles, which increases smooth muscle tone. Under these conditions the opening of KATP channels can induce a much larger change in membrane potential than in relaxed arteries. Furthermore, the rate of ATP hydrolysis by contractile proteins, and thus the submembrane nucleotide concentrations, might also be changed in the presence of myogenic tone. Therefore care should be taken when extrapolating patch clamp results from isolated coronary smooth muscle cells to the function of KATP channels in vivo. (2) The opening of KATP channels is increased in situations related to energy imbalance, such as hypoxia, adenosine release, intracellular acidification, and lactate accumulation. However, there is increasing evidence that KATP channels also contribute to the setting of the membrane potentials of coronary smooth muscle cells under normoxic conditions. Thus the modulation of KATP channels by intracellular metabolites and by vasoactive autacoids may play an important role in the regulation of coronary blood flow even in the presence of normal intracellular ATP concentrations. (3) The smooth muscle cells of coronary terminal arterioles form an electrical syncytium. The opening of a new KATP channels in smooth muscle cells of a terminal arterioles might induce a spatially homogeneous hyperpolarisation of the entire arteriole. The resulting homogeneous decrease in the tone of the coronary smooth muscle cells of the arteriole may induce a considerable change in vascular resistance.(ABSTRACT TRUNCATED AT 250 WORDS)

Genetic and functional linkage of Kir5.1 and Kir2.1 channel subunits.

We have identified several cDNAs for the human Kir5.1 subunit of inwardly rectifying K(+) channels. Alternative splicing of exon 3 and the usage of two alternative polyadenylation sites contribute to cDNA diversity. The hKir5.1 gene KCNJ16 is assigned to chromosomal region 17q23.1-24.2, and is separated by only 34 kb from the hKir2.1 gene (KCNJ2). In the brain, Kir5.1 mRNA is restricted to the evolutionary older parts of the hindbrain, midbrain and diencephalon and overlaps with Kir2.1 in the superior/inferior colliculus and the pontine region. In the kidney Kir5.1 and Kir2.1 are colocalized in the proximal tubule. When expressed in Xenopus oocytes, Kir5.1 is efficiently targeted to the cell surface and forms electrically silent channels together with Kir2.1, thus negatively controlling Kir2.1 channel activity in native cells.

The passive electrical properties of guinea-pig ventricular muscle as examined with a voltage-clamp technique.

1. A voltage-clamp technique was developed for stable recording of small currents in guinea-pig ventricular muscle. Small cylindrical preparations were impaled with three micro-electrodes, one for measuring the feed-back potential and two for injecting current. 2. The longitudinal potential profile resulting from current injection at one point was measured. It agreed well with the theoretical predictions for a linear cable which is sealed at both ends ('healing over'), with a length constant (lambda) of 580 +/- 145 micron. 3. When the clamp current was injected symmetrically into each half of the preparation via two electronic current pumps a spatially homogeneous clamp could be achieved in preparations with a diameter of less than or equal to 250 micron and a length of less than or equal to 2 lambda. 4. The membrane capacity and the membrane resistance of the preparations at the resting potential were measured with small voltage-clamp pulses. Assuming a specific membrane capacity (Cm) of 1 microF/cm2 a specific membrane resistance (Rm) of 6.7 +/- 1.8 k omega cm2 was obtained in Tyrode solution containing 3 mM-K. 5. The total surface area was calculated from the measured capacity of the preparation assuming a Cm of 1 microF/cm2. The total cellular volume was estimated from optical measurement of the external dimensions of the preparation assuming an extracellular space of 25%. From these data the average surface/volume ratio of individual cells was calculated to be 7200 cm2/cm3. 6. From the measured electrical constants the specific resistance of the intracellular space (Ri) was calculated to be 200-250 omega cm. With small constant current pulses a membrane time constant of 6.6 +/- 1.3 ms was measured. 7. The influence of the extracellular potassium concentration ([K]o) on Rm was studied in the range 1.5-6 mM-[K]o. Rm was found to depend on [K]o less than predicted by the constant field theory.

Inhibition of the sodium pump in guinea-pig ventricular muscle by dihydro-ouabain: effects of external potassium and sodium.

The inhibition of the electrogenic pump current in quiescent guinea-pig ventricular muscle by dihydro-ouabain (DHO) was studied with the three-micro-electrode voltage-clamp technique described previously (Daut, 1982c). From dose-response curves of the drug-induced current change (ID) the equilibrium dissociation constant of the binding of DHO to the Na-K pump (KD) and the electrogenic pump current flowing in the steady state (Ip) were inferred (Daut & Rüdel, 1982b). The external K concentration ([K]o) was varied between 2 and 4.5 mM (substituted by Na). KD was found to increase with increasing [K]o. A plot of log KD versus log [K]o gave a straight line with a slope of about 1.5. The time constants of the onset (tau on) and decay (tau off) of ID are supposed to represent the chemical kinetics of binding and unbinding of the drug (Daut & Rüdel, 1981, 1982b). Tau on was found to be inversely related to [K]o whereas tau off was found to be independent of [K]o. Ip was found to be independent of [K]o. This was interpreted to indicate that in the steady state Ip is mainly determined by the passive influx of Na into the cell, which may be relatively insensitive to small changes in [K]o. The effects of [K]o on the drug-induced current change are consistent with competitive inhibition of the binding of DHO to the Na-K pump. It is suggested that K ions and cardiac glycosides compete for extracellular binding sites on the same conformation of the Na-K pump. The external Na concentration ([Na]o) was varied between 147 and 49 mM (substituted by choline or Tris). Reduction of [Na]o produced a proportional decrease of Ip. This may be a consequence of the accompanying reduction of passive Na influx and the resulting decrease in intracellular Na activity (alpha iNa). Reduction of [Na]o markedly increased KD. This effect may be mediated by competition between Na and K at the K-loading sites of the pump and/or by separate modulatory Na-binding sites. It is concluded that the well known effects of external Na and K on the positive inotropic action of cardiac glycosides can be fully accounted for by the marked changes in the apparent binding affinity of the drug reported here.

Substrate dependence of energy metabolism in isolated guinea-pig cardiac muscle: a microcalorimetric study.

1. The effects of glucose, pyruvate and lactate on basal metabolism and on contraction-related energy expenditure of thin trabeculae isolated from guinea-pig heart were studied using a microcalorimetric technique. 2. Resting heat rates of cardiac ventricular muscle measured in the presence of substrate-free solution (56 +/- 20 mW (g dry weight)-1), 10 mM-lactate (54 +/- 12 mW (g dry weight)-1) and 10 mM-glucose (63 +/- 24 mW (g dry weight)-1) did not differ significantly. Increasing the external glucose concentration (up to 100 mM) and/or adding insulin (up to 80 units l-1) had virtually no effect on the measured resting heat rate. 3. With 10 mM-pyruvate as substrate resting heat rate was substantially larger (106 +/- 40 mW (g dry weight)-1) than with glucose, lactate or substrate-free solution. The concentrations of pyruvate producing a half-maximal increase in resting heat rate as compared to substrate-free solution ranged between 0.4 and 1.2 mM. 4. In order to test whether the development of an anoxic core contributed to the substrate dependence of resting heat production the critical PO2 (i.e. the PO2 that produced a just-noticeable decrease in heat rate) was determined in cylindrical preparations of various diameters. It was found that none of the preparations had an anoxic core at rest in a solution equilibrated with 100% oxygen. 5. From the dependence of the critical PO2 on the diameter of the preparation the diffusion coefficient of oxygen through cardiac muscle was calculated using a modification of Hill's equation (Hill, 1928). The O2 diffusion coefficient was found to be 1.09 X 10(-5) cm2 s-1. 6. Contraction-related heat production was also found to be dependent on the substrate used. In the presence of 10 mM-pyruvate it was about 60% larger than in the presence of 10 mM-glucose, 10 mM-lactate or with substrate-free solution. 7. Isometric force of contraction showed the same substrate dependence as contraction-related heat production and increased with a similar time course during repetitive stimulation. 8. The possible mechanisms underlying the substrate dependence of myocardial energy metabolism are discussed. It is suggested that the increased energy expenditure observed in the presence of pyruvate may be related to a decrease in intracellular phosphate and/or to an increase in intracellular pH.

Heat production of quiescent ventricular trabeculae isolated from guinea-pig heart.

1. A new calorimetric technique has been developed which allows continuous measurement of the rate of energy expenditure in superfused preparations of cardiac muscle. Thin trabeculae of guinea-pig ventricular muscle were mounted in a Perspex tube of 0.8 mm inner diameter and the temperature difference of the perfusate upstream and downstream of the preparation was measured. 2. The resting heat rate of trabeculae of 240-575 microns diameter from guinea-pig heart was determined repeatedly for up to 6 h after cardiectomy. It did not vary with time during the course of the experiment. 3. The average resting heat rate measured in HEPES-buffered Tyrode solution containing 20 mM-glucose and 2 mM-pyruvate as substrates was 130 +/- 29 mW/g dry weight or 36 +/- 8 mW/cm3 of tissue (n = 15). This is an order of magnitude larger than the resting heat rate reported in the literature for isolated cardiac preparations. 4. After omitting the pyruvate from the superfusate the resting heat rate decreased to 60-70% of its steady value within 4 min. After readmission of pyruvate this effect was reversed. The average resting heat rate with glucose as sole substrate was 23 +/- 4 mW/cm3. 5. Uncoupling of the mitochondria by 50 microM-2,4-dinitrophenol (DNP) increased the heat rate up to 170 mW/cm3. This effect could be maintained for several minutes and was fully reversible. Raising the external K+ concentration to 150 mM (NaCl replaced by KCl) induced a transient rise in the rate of heat production up to 115 mW/cm3. 6. The heat production during uncoupling of the mitochondria and during potassium contractures was inversely related to the diameter of the preparation. Calculation based on Hill's equation (Hill, 1928) indicated that this was caused by the development of anoxia at the core of the preparation. 7. In contrast, the rate of heat production of quiescent preparations was not correlated with diameter and calculation indicated that at rest there was no anoxic core. The high value of resting heat rate found in the present study is discussed within the context of the large variation of 1.7-25 mW/g reported in the literature for resting metabolic rate of cardiac muscle.

Sodium-potassium movement and the regulation of cardiac muscle activity.

The changes in intracellular sodium activity and contractility produced by short-lasting application of a fast-acting cardiac glycoside were measured in guinea-pig ventricular muscle. It was found that under certain conditions the change in twitch tension paralleled the change in sodium activity. It is suggested that the electrogenic sodium pump may be involved in the regulation of both the mechanical and the electrical activity of cardiac muscle.

Hyperpolarization of isolated capillaries from guinea-pig heart induced by K+ channel openers and glucose deprivation.

1. The present study was designed to test if microvascular coronary endothelial cells express ATP-sensitive K+ channels (KATP channels). We performed microfluorometric measurements of the membrane potential of freshly isolated guinea-pig coronary capillaries equilibrated with the voltage-sensitive dye bis-oxonol (bis-[1,3-dibutylbarbituric acid] trimethineoxonol, [DiBAC4(3)]). 2. The resting membrane potential of capillaries in physiological salt solution was -46 +/- 4.2 mV (n = 8) at room temperature (22 degrees C) as determined after calibration of the fluorescence using the Na(+)-K+ ionophore gramicidin in the presence of different K+ concentrations. Spontaneous membrane potential fluctuations of 10-20 mV amplitude were often observed. 3. A reversible, sustained hyperpolarization to a new membrane potential close to the K+ equilibrium potential (EK) could be induced by application of the K+ channel openers HOE 234 (100 nM to 1 microM), diazoxide (10 PM to 100 nM) or pinacidil (100 nM). Subsequent addition of glibenclamide (200 nM to 2 microM) reversed this hyperpolarization. 4. A glibenclamide-sensitive hyperpolarization of coronary capillaries to values near EK was also observed upon omission of D-glucose (10 mM) from the superfusing solution or by substituting L-glucose for D-glucose. Maximum hyperpolarization was reached in less than 10 min. 5. Our results suggest that microvascular coronary endothelial cells express KATP channels which may be activated during hypoglycaemia.

Voltage-dependent K(+) current in capillary endothelial cells isolated from guinea pig heart.

Capillary fragments were isolated from guinea pig hearts, and their electrical properties were studied using the perforated-patch and cell-attached mode of the patch-clamp technique. A voltage-dependent K(+) current was discovered that was activated at potentials positive to -20 mV and showed a sigmoid rising phase. For depolarizing voltage steps from -128 to +52 mV, the time to peak was 71 +/- 5 ms (mean +/- SE) and the amplitude of the current was 3.7 +/- 0.5 pA/pF in the presence of 5 mM external K(+). The time course of inactivation was exponential with a time constant of 7.2 +/- 0.5 s at +52 mV. The current was blocked by tetraethylammonium (inhibitory constant approximately 3 mM) but was not affected by charybdotoxin (1 microM) or apamin (1 microM). In the cell-attached mode, depolarization-activated single-channel currents were found that inactivated completely within 30 s; the single-channel conductance was 12.3 +/- 2.4 pS. The depolarization-activated K(+) current described here may play a role in membrane potential oscillations of the endothelium.

The electrogenic sodium pump in guinea-pig ventricular muscle: inhibition of pump current by cardiac glycosides.

1. The inhibition of the electrogenic sodium pump in guinea-pig ventricular muscle by cardiac glycosides was studied with a voltage-clamp technique.2. Superfusion of the preparation with dihydro-ouabain (DHO) produced a reversible depolarization of up to 7 mV. When the membrane potential was clamped to a constant value near the resting potential application of DHO produced a corresponding current change in the inward direction which reached a steady state in less than 1 min.3. The drug-induced current change (I(D)) was found to be the result of a parallel shift of the current-voltage relation. The contributions of a change in extracellular K or intracellular Na to the measured I(D) were shown to be very small. From these findings and the results summarized below it was concluded that I(D) represents the blockage of the electrogenic pump current by DHO and that it is proportional to the number of drug molecules bound to the Na-K-ATPase in the intact cell.4. The dependence of I(D) on the concentration of DHO applied (5 x 10(-6)-8 x 10(-4) M) was found to be consistent with the predictions of the law of mass action for reversible one-to-one binding of the drug to the Na-K pump under equilibrium conditions. From a Scatchard-type plot the equilibrium dissociation constant (K(D)) of DHO was determined to be 4.6 (+/-2.3) x 10(-5) M.5. The steady-state pump current in the resting preparation was calculated to be 0.81+/-0.26 muA/cm(2). It contributed 6.4+/-0.9 mV to the resting potential in Tyrode solution containing 3 mM-K.6. In the smallest preparations used the measured time course of the onset and decay of I(D) agreed with the chemical kinetics of binding and unbinding calculated for various DHO concentrations. The rate constant of unbinding (k(2)) was found to be 3.4 (+/-0.7) x 10(-2) S(-1) and the average rate constant of binding (k(1)) was 7.4 x 10(2) M(-1) S(-1).7. By comparing the effects of ouabain and DHO in the same preparation the following estimates of the chemical constants of ouabain binding to the Na-K pump were obtained: K(D) approximately 1.5 x 10(-6) M; k(1) approximately 4 x 10(3) M(-1) S(-1); k(2) approximately 6 x 10(-3) S(-1).8. An analysis of the transmembrane movements of Na and K in the steady state showed that the measured pump current density is consistent with a counter-transport of 3 Na and 2 K ions.

The electrical response of cultured guinea-pig coronary endothelial cells to endothelium-dependent vasodilators.

1. Primary cultures of coronary endothelial cells were obtained by enzymatic dispersion of isolated guinea-pig hearts and separation of different cardiac cell types by density gradient centrifugation. The cells were grown to confluency and the membrane potential of the monolayer was recorded using the whole-cell-clamp mode of the patch-clamp technique. 2. When superfused with physiological salt solution at 37 degrees C the average resting potential of the monolayers was -35 +/- 9 mV. A 2 min application of bradykinin (0.1-20 nM) induced a transient hyperpolarization of up to 40 mV (median 33 mV), which was followed by a sustained depolarization of up to 28 mV (median 10 mV). The average duration of the hyperpolarization, measured midway between the resting potential and peak negativity, was 48 s with 20 nM-bradykinin. 3. The concentration of bradykinin producing a half-maximal hyperpolarization was 2.5 nM. When high concentrations of bradykinin (greater than 10 nM) were applied for several minutes voltage oscillations of low amplitude with periodicity of 2-3 min were observed. 4. The peak of the hyperpolarization depended on the extracellular potassium concentration ([K+]o). The limiting slope of the relation between membrane potential and log [K+]o was 52 mV per 10-fold change in [K+]o. With 50 mM [K+]o the hyperpolarization was abolished and with 100 mM [K+]o the hyperpolarization turned into a depolarization. 5. After removal of external Ca2+ the first transient hyperpolarization elicited by bradykinin had the same amplitude as under control conditions, but its duration was reduced to about 72%. The second application of bradykinin in Ca2(+)-free solution produced only a depolarization. The hyperpolarizing response to bradykinin could be re-primed by exposing the preparation to Ca2(+)-containing solution for 2 min. 6. The transient hyperpolarization elicited by 4 nM-bradykinin could be inhibited by d-tubocurarine, a blocker of Ca2(+)-activated potassium channels. On average, 1 mM-tubocurarine reduced the hyperpolarization by 49 +/- 18%. Apamin (10 microM) reduced the hyperpolarization by 15 +/- 11%. 7. ATP (1 microM) produced a hyperpolarization of similar amplitude to that produced by bradykinin, but of shorter duration (average 29 s), and a very small (less than 5 mV) or no sustained depolarization. Histamine (10 microM) produced an even shorter transient hyperpolarization, followed by a depolarization of up to 15 mV. 8. Most of the monolayers of coronary endothelial cells responded to adenosine in a similar way as to bradykinin.(ABSTRACT TRUNCATED AT 400 WORDS)

Effects of vasoactive agonists on the membrane potential of cultured bovine aortic and guinea-pig coronary endothelium.

1. The effects of bradykinin, ATP, adenosine, histamine and thrombin on the membrane potential of confluent monolayers of cultured bovine aortic endothelial cells (BAECs) and guinea-pig coronary endothelial cells (GCECs) were studied at 37 degrees C using the whole-cell mode of the patch-clamp technique. 2. The amplitude histogram of the resting potentials of BAEC monolayers showed a bimodal distribution with one peak around -25 mV and another peak around -85 mV. Transitions from one potential level to the other were observed. The bistable membrane potential can be explained by an N-shaped current-voltage relation of the endothelial cell membrane. 3. When BAECs with a low resting potential (-10 to -30 mV) were superfused with maximally effective concentrations of ATP (2-10 microM) an initial hyperpolarization of -80 to -90 mV was observed which decayed to a plateau of about -60 mV within 1 min. When ATP was removed after 2-3 min the membrane potential returned to control level within 1 min. This was followed by a second hyperpolarization of 10-20 mV, which decayed within 15 min. 4. In the absence of extracellular calcium, ATP produced only a brief transient hyperpolarization in aortic endothelium. The plateau and the secondary hyperpolarization were abolished. These findings are consistent with the idea that the changes in membrane potential reflect changes in intracellular free Ca2+ and that the initial peak is due to release of Ca2+ from intracellular stores, whereas the plateau and the secondary hyperpolarization depend on transmembrane Ca2+ influx. 5. Bradykinin evoked potential changes similar to ATP in BAECs, except that the secondary hyperpolarization during wash-out was absent. When the membrane potential was more negative than -80 mV, ATP and bradykinin induced only a small initial hyperpolarization followed by a depolarization of up to 20 mV. 6. In aortic endothelium, ADP (10 microM) evoked a much smaller response than ATP. Adenosine (10 microM), thrombin (2 units/ml), acetylcholine (10 microM) and histamine (10 microM) had only a very small effect on the membrane potential, if any. 7. The amplitude histogram of the membrane potential of GCECs showed only one peak around -35 mV. In coronary endothelium, application of bradykinin, ATP, histamine, thrombin, acetylcholine and adenosine all evoked a transient hyperpolarization of 10-40 mV lasting 1 min or less, which then turned into a depolarization. 8. The K+ channel openers cromakalim (BRL 34915) and lemakalim (BRL 38227) did not affect the membrane potential of GCECs or BAECs.(ABSTRACT TRUNCATED AT 400 WORDS)

Oxidative phosphorylation in myocardial mitochondria 'in situ': a calorimetric study on permeabilized cardiac muscle preparations.

A novel flow calorimetric technique was developed to study the energy turnover of myocardial mitochondria. Cylindrical strands of cardiac muscle (trabeculae) weighing 100-500 micrograms were isolated from guinea-pig heart and mounted in a tubular recording chamber which was continuously perfused with physiological salt solution at 37 degrees C. The temperature difference between the upstream and the downstream side of the chamber, which is proportional to the rate of heat production of the trabecula, was measured at high resolution. In this way the rate of energy expenditure of isolated cardiac muscle could be recorded continuously for several hours. When the preparations were superfused with an 'intracellular' solution containing 5 mM pyruvate and 2 mM malate as substrates, permeabilization of the sarcolemma with 25 microM digitonin induced a marked increase in the measured heat rate in the presence of 2 mM ADP. The major fraction of the ADP sensitive heat production (83%) could be blocked with 400 microM atractyloside, an inhibitor of the adeninenucleotide translocase, and by 600 microM alpha-cyano-4-hydroxycinnamate, an inhibitor of monocarboxylate/H+ co-transport. The atractyloside sensitive heat production was abolished in anoxic solution. These results suggest that the atractyloside-sensitive heat production (21.8 +/- 3.5 mW cm-3 of tissue) was attributable to oxidative phosphorylation. The mitochondria apparently remained intact after treatment with digitonin, since application of the uncoupler 2,4-dinitrophenol (DNP) produced a very large increase in heat rate. A minor fraction of the heat rate induced by ADP in permeabilized cardiac muscle preparations (17%) was not sensitive to atractyloside. This component was also seen before application of digitonin and was probably related to ectonucleotidases. In conclusion, our calorimetric technique allows investigation of the energy metabolism of myocardial mitochondria 'in situ', i.e. without destroying the microarchitecture of cardiac muscle cells.

A new cardiotonic drug reduces the energy cost of active tension in cardiac muscle.

The novel thiadiazinone EMD 57033 (EMD) increases the calcium responsiveness of the contractile proteins in cardiac muscle. In skinned ventricular trabeculae isolated from guinea-pig heart, application of 10 microM EMD shifted the curve relating isometric tension to the applied calcium concentration to the left and increased maximal tension by 15%. In intact trabeculae, the rate of heat production, an indicator of the rate of ATP hydrolysis in the steady state, and isometric tension were measured at 37 degrees C. Both the thiadiazinone (EMD; 2.5, 5, and 10 microM) and the cardiac glycoside dihydro-ouabain (DHO; 5, 10, and 20 microM) produced a concentration-dependent increase in contraction-related heart production (Hc) and in the tension time integral of isometric contractions (Tti). In the presence of EMD the energy cost of active tension (Hc/Tti) was substantially decreased as compared to control conditions. The energy cost of the positive inotropic effect of EMD (43.8 mW N-1 cm-1) was only about half as large as the energy cost of the positive inotropic effect of DHO (88.4 mW N-1 cm-1). It is concluded that EMD causes a change in cross-bridge kinetics that increases the contractility of cardiac muscle and improves the economy of chemo-mechanical energy transduction. Our results suggest that EMD 57033 represents a prototype of a new class of cardiotonic agents that might be potentially useful in the therapy of congestive heart failure.