Search

The effects of 10 mM caffeine (CAF) on intramembrane charge movements (ICM) were studied in isolated guinea-pig ventricular heart cells with the whole-cell patch-clamp technique. In the presence of CAF, the properties (voltage dependence, maximum QON[Qmax], availability with voltage) of QON charge activated from −110 mV were barely affected. Following a 100 ms prepulse to −50 mV to decrease the participation of charges originating from Na channels, the voltage dependence of QON was shifted by 5 mV (negative component) and by 10 mV (positive component) towards negative potentials, and Qmax was depressed by 16.5%. CAF drastically reduced in a time- and voltage-dependent manner QOFF on repolarization to −50 mV, the effects being greater at positive potentials. CAF-induced QOFF immobilization could be almost entirely removed by repolarization to voltages as negative as −170 mV. In these conditions, the voltage-dependence of QOFF (repolarization to +30 to −170 mV) was shifted by 17 mV (negative component) and 30 mV (positive component) towards negative potentials, suggesting an interconversion into charge 2. Most of CAF effects were suppressed when the sarcoplasmic reticulum (SR) was not functional or when the cells were loaded with BAPTA-AM. We conclude that CAF effects on ICM are likely due to Ca2+ ions released from the SR, and which accumulate in the subsarcolemmal fuzzy spaces in the vicinity of the Ca channels. Because CAF effects were more pronounced on QOFF than on QON the channels have likely to open before Ca2+ ions could affect their gating properties. It is speculated that such an effect on gating charges might contribute to the Ca-induced inactivation of the Ca current. British Journal of Pharmacology (2002) 135, 721–734; doi:10.1038/sj.bjp.0704520

The effects of a monoclonal antibody (B8E5) directed against the second extracellular loop of the muscarinic M2receptor were studied on the L-type Ca2+currents ( ICa,L) of guinea pig ventricular myocytes using the whole cell patch-clamp technique. Similar to carbachol, B8E5 reduced the isoproterenol (ISO)-stimulated ICa,Lbut did not significantly affect basal ICa,L. Atropine blocked the inhibitory effect of B8E5. The electrophysiological parameters of ISO-stimulated ICa,Lwere not modified in presence of B8E5. Inhibition of ICa,Lby B8E5 was still observed when intracellular cAMP was either enhanced by forskolin or maintained constant by using a hydrolysis-resistant cAMP analog (8-bromoadenosine 3′,5′-cyclic monophosphate) or by applying the phosphodiesterase inhibitor IBMX. The effect of B8E5 was mimicked by 8-bromoguanosine 3′,5′-cyclic monophosphate, a potent stimulator of cGMP-dependent protein kinase, and prevented by a selective inhibitor of nitric oxide-sensitive guanylyl cyclase {1 H-(1,2,4)oxadiazolo[4,3-a]quinoxaline-1-one}. These results indicate that the antibody B8E5 inhibits the β-adrenergic-stimulated ICa,Lthrough activation of the M2muscarinic receptor and further suggest that the antibody acts not via the classical pathway of decreasing intracellular cAMP, but rather by increasing cGMP.

The possibility that the class III antiarrhythmic drugs clofilium and d-sotalol might affect delayed rectifier potassium channels at the level of their gating currents was assessed with the whole-cell patch-clamp technique in guinea-pig isolated ventricular heart cells. Clofilium (up to 20 μM) and d-sotalol (1 μM) did not decrease the Na current, the L-type Ca current or the background K current iKl, but significantly depressed the time-dependent delayed outward K current iK. Clofilium partially decreased in a dose-dependent manner (1–20 μM) QON of intramembrane charge movements (ICM) elicited by a depolarizing pulse applied from a holding potential of −110 mV or following a 100 ms inactivating prepulse to −50 mV. D-sotalol (1 μM) also decreased QON. Channel density estimated from the clofilium-sensitive ICM closely matched that of the delayed rectifier channels. Clofilium and d-sotalol decreased QOFF seen on repolarization in a dose- and voltage-dependent manner. The kinetics of the decay of the OFF gating currents were not affected, and only the fast phase was depressed. In control conditions, QON availability with voltage was most of the time well described by two inactivating components. In the presence of clofilium and d-sotalol, a complex behaviour of QON availability was observed, unmasking additional components. The reactivation kinetics of QON after a 500 ms inactivating pulse to 0 mV was not affected. We conclude that delayed rectifier K channels significantly contribute to QON and QOFF of ICM in guinea-pig ventricular heart cells, besides Na and Ca channels, and that clofilium and d-sotalol directly interact with these K channels proteins by affecting their gating properties. Keywords: Intramembrane charge movements, K+ channels, native cardiac cells, class III antiarrhythmic drugs, clofilium, d-sotalol Introduction Intramembrane charge movements (ICM) in excitable cells result from molecular rearrangements of ion channel proteins under the impetus changes in the membrane potential. In mammalian cardiac cells, only sodium and calcium channels have been thought to give rise to ICM recorded with short depolarizing pulses, because of the much slower activation kinetics of the delayed rectifier potassium current (see Field et al., 1988; Bean & Rios, 1989; Hadley & Lederer, 1989, 1991a; Hanck et al., 1990; Shirokov et al., 1992; 1993). However, we have found more than two components of ICM in guinea-pig ventricular myocytes (Malecot & Argibay, 1996) which could not be entirely suppressed by the Na and Ca channels blockers lidocaine, tetracaine, nifedipine and D600 (Malecot et al., 1997a,1997b), although the underlying currents are blocked. However, compounds such as D600 or nitrendipine have been shown to be able to suppress all of the gating currents in heart cells, but at high concentrations that also affect potassium currents (e.g., Hadley & Lederer, 1991a). These results suggest that other channels might also contribute to ICM in heart cells. Moreover, recent studies of different expressed cloned potassium channels have revealed their fast gating currents (Larsson et al., 1996; Fedida, 1997), which are very similar in shape to those recorded in isolated heart cells. Therefore, we have re-investigated the pharmacological properties of ICM in freshly isolated guinea-pig heart cells and, in particular, we have tested the hypothesis of the presence of a component of ICM originating from the voltage-activated delayed rectifier K channels. Because several antiarrhythmic drugs have been shown to affect gating currents in many preparations, we have tested in this paper the possibility that the class III antiarrythmic drugs, clofilium and d-sotalol, blockers of iKs and of iKr (for review see Baro & Escande, 1993; Snyders, 1995), might affect delayed rectifier potassium channels at the level of their gating currents. We show here that ICM activating at positive potentials are sensitive to clofilium and d-sotalol at concentrations not inhibiting the Na or the L-type Ca channels but depressing the delayed outward K current in our recording conditions. To our knowledge, this is the first time that K channels are shown to significantly contribute to ICM in freshly isolated heart cells, besides Na and Ca channels. This new finding opens the door to a better understanding of the relationships between ICM and membrane excitability and to the future development of more powerful pharmacological compounds. A preliminary note on these experiments has been published (Malecot & Argibay, 1998).

Monoclonal antibodies were produced against a peptide corresponding to the second extracellular loop of the human beta2-adrenoceptor. One of these monoclonals, inducing an agonist-like effect in neonatal rat cardiomyocytes, was used to define the structural and physiological basis of this activity. The epitope recognized by the antibody corresponds to the sequence Trp-Tyr-Arg-Ala-Thr-His-Gln-Glu as determined by peptide scanning. Analysis by alanine modification of the peptide epitope showed the importance of the Trp, and Glu residues in antibody recognition The apparent affinity of the antibody assessed either by surface plasmon resonance or by functional titration on its agonist-like activity showed a similar value (10(8) M(-1)). The antibody recognized the receptor in its native form as shown by immunofluorescence experiments on A431 cells but not in its denatured form as shown by its absence of staining in immunoblots. The positive chronotropic effect in vitro was specifically blocked by both the antigenic peptide and the specific beta2-antagonist (+/-)-1-[2,3-(Dihydro-7-methyl1H-inden-4-yl)oxy]-3-[(1-methy lethyl)amino]-2-butanol hydrochloride (ICI1118,551). This activity was mediated through activation of Ca2+ L-type channels as assessed in guinea pig cardiomyocytes. These results suggest that the epitope is located in an extracellular alpha-helix, whose recognition by the antibody could stabilize the receptor in its 'active' conformation.

Sera from T. cruzi infected mice were tested in an enzyme immunoassay on peptides corresponding to the second extracellular loops of the beta 1-, the beta 2-adrenergic receptor and the M2 muscarinic receptor. All sera of mice (4/4) in the acute phase recognized the beta 1-adrenergic receptor and the M2 muscarinic receptor peptides but not the beta 2-adrenergic receptor peptide. The same peptides were recognized during the chronic phase in half of the mice (6/12). The immunoglobulin fractions of the mice were tested for their activity on L-type Ca++ channels of isolated guinea-pig cardiomyocytes using the whole-cell patch clamp technique. The immunoglobulin fractions of acute phase mice were able to activate the Ca++ channels by stimulation of the beta-adrenergic receptors, as assessed by inhibition with propranolol. Those of the chronic phase mice reduced the Ca++ current by stimulation of the muscarinic receptors, as assessed by inhibition with atropine. These results confirm the existence of functional epitopes on the second extracellular loops of both receptors. They suggest that, as in humans, the parasite is able to elicit functional autoantibodies against these epitopes. They give evidence that these autoantibodies mediate their physiological effects by modulating the cAMP activated Ca++ channels.

The effects of 100 microM ryanodine on the L-type calcium channel were studied using the pacth-clamp technique in isolated guinea pig ventricular myocytes. The inactivation kinetics of the calcium current were slowed down in the presence of ryanodine in agreement with the blockade of the release of calcium from the sarcoplasmic reticulum by the drug. The I-V and steady-state inactivation curves of the calcium current were shifted to negative values by ryanodine. A similar shift was observed in the activation and inactivation curves of the intramembrane charge movement associated with the calcium channel. Due to this shift, ryanodine slightly reduced the maximal amount of displaced charge although it did not modify the transition from the inactivated to the activated state (i.e., charge movement repriming). This result is in notable contrast with that obtained in skeletal muscle, where it has been found that ryanodine interferes with charge movement repriming. These results provide additional evidence of the postulated differences between the architecture of the excitation-contraction coupling system in cardiac and skeletal muscle.

We show that gadolinium (Gd3+) is a potent calcium channel blocker in guinea-pig isolated ventricular myocytes. A dose-dependent inhibition of ICaL was found with an EC50 of 1.4 microM and a complete inhibition at 10 microM Gd3+. When compared with Cd2+, it appeared that the blockade of ICaL is a complex phenomenon probably involving more than one site of interaction (a Hill coefficient of 1.6 was found for Gd3+ vs. 1.0 for Cd2+). It is concluded that Gd3+ ions completely block ICaL at concentrations used to block stretch-activated channels (SAC), rendering its use as a specific SAC inhibitor problematic.

Until recently the investigation of length-dependent effects in cardiac muscle was restricted to multicellular preparations. We describe our experimental set-up which for the first time, in single cardiac myocytes, permits the effects of changes in cell length on auxotonic contractions (measured by carbon fibre transducers) to be simultaneously recorded with the effects on membrane potential and/or changes in intracellular calcium concentration (using indo-1 AM, acetoxylmethyl form). Consistent with previous findings (in experiments at 20-25 degrees C and 0.25 Hz) we report that following a stretch there was an increase in passive tension and contraction. A stretch which increased sarcomere length by approximately 3% had no significant effect on resting membrane potential or action potential amplitude. There was, however, a significant decrease in the action potential duration (P < 0.01, n = 8). No significant change in the amplitude of the intracellular calcium transient was seen following a stretch but a reduction in its duration was observed (P < 0.025, n = 11). Our observations on intracellular calcium transients are consistent with the hypothesis that, in mechanically loaded preparations, their time course is more dependent on changes in tension than changes in length.

IF : 4,96; International audience; 5-hydroxytryptamine-4 (5-HT(4)) receptors have been proposed to contribute to the generation of atrial fibrillation in human atrial myocytes, but it is unclear if these receptors are present in the hearts of small laboratory animals (e.g. rat). In this study, we examined presence and functionality of 5-HT(4) receptors in auricular myocytes of newborn rats and their possible involvement in regulation of gap junctional intercellular communication (GJIC, responsible for the cell-to-cell propagation of the cardiac excitation). Western-blotting assays showed that 5-HT(4) receptors were present and real-time RT-PCR analysis revealed that 5-HT(4b) was the predominant isoform. Serotonin (1 microM) significantly reduced cAMP concentration unless a selective 5-HT(4) inhibitor (GR113808 or ML10375, both 1 microM) was present. Serotonin also reduced the amplitude of L-type calcium currents and influenced the strength of GJIC without modifying the phosphorylation profiles of the different channel-forming proteins or connexins (Cxs), namely Cx40, Cx43 and Cx45. GJIC was markedly increased when serotonin exposure occurred in presence of a 5-HT(4) inhibitor but strongly reduced when 5-HT(2A) and 5-HT(2B) receptors were inhibited, showing that activation of these receptors antagonistically regulated GJIC. The serotoninergic response was completely abolished when 5-HT(4), 5-HT(2A) and 5-HT(2B) were simultaneously inhibited. A 24 h serotonin exposure strongly reduced Cx40 expression whereas Cx45 was less affected and Cx43 still less. In conclusion, this study revealed that 5-HT(4) (mainly 5-HT(4b)), 5-HT(2A) and 5-HT(2B) receptors coexisted in auricular myocytes of newborn rat, that 5-HT(4) activation reduced cAMP concentration, I(Ca)(L) and intercellular coupling whereas 5-HT(2A) or 5-HT(2B) activation conversely enhanced GJIC.

1. The effects of heptaminol on calcium current amplitude and characteristics were studied in single ventricular myocytes of guinea-pig by use of the whole cell configuration of the patch clamp technique. 2. A concentration-dependent decrease in ICa amplitude was observed. At heptaminol concentration as low as 10(-6) M, this effect was observed in only two cells (n = 6). At 10(-5) M the reduction of ICa was of 30 +/- 15% (n = 11). 3. The current recovery from inactivation at -40 mV holding potential (HP) seemed less sensitive to perfusion with heptaminol (greater than 10(-6) M). However, at -80 mV HP the overshoot of the recovery curve was decreased by heptaminol. 4. Both at -40 mV and -80 mV HP, heptaminol (10(-5) M) significantly increased the steady state inactivation of ICa. 5. As previously proposed by others to explain the effects of membrane active substances, the effects of heptaminol may result from alterations in cell membrane properties and possibly from an increase in intracellular free calcium ion concentration.

Single guinea-pig ventricular myocytes were loaded with the fluorescent Ca2+ indicator Indo-1 AM and stretched by carbon fibres. Stretching increased resting tension. Sarcomere lengths were increased by 2-18%. It was observed that a stretch increased resting [Ca2+]i in seven out of eight cells. The change in [Ca2+]i increased with the size of the stretch and returned to pre-stretch levels on return to resting cell length. These observations suggest a means by which changes in resting muscle length can modify the contractile state of cardiac muscle.

International audience; In human atrial myocytes, serotonin rather than sympathetic, stimulation is more frequently associated with atrial fibrillation. So does the arrhythmogenic effect of serotonin result from the mechanism of action of the receptor or the context of its action upon cardiac myocytes? The capacity of agonists to produce cAMP followed the sequence 5-HT < Iso < Forskolin to increase ICaL with 5-HT = Iso = Forskolin. The simultaneous application of threshold concentrations of 5-HT and Iso maximally increased ICaL. We will show that the effect of 5-HT upon human atrial myocytes is an imbalance between low production of cAMP and maximal activation of ICaL.

International audience; We have previously reported that antipeptide antibodies directed against the second extracellular loop of the cardiac h5-HT4 receptor could block the activation of the L-type Ca channel in human atrial cardiomyocytes. In this paper we investigate the immunological and physiological activity of these antibodies, in a cell system expressing a larger amount of receptors than the atrial cells. The recombinant receptor was expressed at the surface of COS-7 cells under an active form (serotonin, EC50 = 1.81 x 10(-7) M), at a high level (375 +/- 25 fmol receptor/mg total protein) and was able to bind a specific ligand (GR113808) with a high affinity (Kd = 0.28 +/- 0.05 nM). In this system, the same anti-peptide antibodies used for the cardiac cells induced an "agonist-like" effect on the recombinant h5-HT4 receptor. These results are in line with those shown for others G-protein coupled receptors, as adrenoreceptors. In addition, this work showed that the effect of the antibodies is not only dependent on the epitopic region recognised but also on the molecular density and/or the cellular environment of the target receptors. Finally, our results support the hypothesis that the h5-HT4 receptor could be a new target for autoantibodies in patients with atrial arrhythmia.

Antibodies directed against the second extracellular loop of G protein-coupled receptors have been shown to exert "agonist-like" activities. In order to test the hypothesis that this is a general phenomenon, antibodies were raised in rabbits against a synthetic peptide corresponding to the second extracellular loop of the newly sequenced human cardiac 5-HT(4)receptor. The antibodies were affinity-purified and shown to recognize the 5-HT(4)receptor in immunoblots of Chinese hamster ovary (CHO) cells expressing the receptor. The antibodies had no intrinsic effect but could depress the activation of L -type calcium channel induced by serotonin in human atrial cells. This antagonist-like effect was exerted both by intact IgG and by Fab fragments. These results are physiologically important since it has been shown that the 5-HT(4)receptor could be a target for autoantibodies in mothers at risk of giving birth to children with neonatal atrio-ventricular block.

The 52-kDa SSA/Ro (Ro52) ribonucleoprotein is an antigenic target strongly associated with the autoimmune response in mothers whose children develop neonatal lupus and congenital heart block. When sera from patients with systemic lupus erythematosus were used as autoimmune controls in an enzyme immunoassay to screen for antibodies against the human serotoninergic 5-HT4-receptor, a high correlation was found between the presence of anti-Ro52 protein antibodies in such sera and antibodies reacting with a synthetic peptide, corresponding to the second extracellular loop of the human 5-HT4 receptor (amino acid residues 165-185). Homology scanning between the 5-HT4 peptide and the sequence of the Ro52 protein indicated two potential common epitopes located between residues 365 and 396 of the Ro52 protein. Cross-reactivity was found between the peptide derived from the 5-HT4 receptor, and a peptide corresponding to residues 365-382 of the Ro52 protein. Autoantibodies, affinity-purified on the 5-HT4 receptor peptide, specifically recognized both the Ro52 protein and the 5-HT4 receptor protein in immunoblots. The affinity-purified antibodies antagonized the serotonin-induced L-type Ca channel activation on human atrial cells. This effect could explain the electrophysiological abnormalities in neonatal lupus.

A mixture of crude collagenase and non-specific proteases has been used to isolate guinea pig ventricular heart cells. Measurements of collagenase activity with Wunsch's substrate and protein content with sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) suggest that collagenase enzymes do not play a major role in heart cell isolation. On the other hand, an important factor in heart digestion seems to consist of some fractions of the proteases present in crude collagenase. It is also noted that crude collagenases do not present any sensitivity to added calcium but because this ion is important to obtain isolated cells its role is discussed. According to our results, the SDS-PAGE method can be used to determine the appropriate enzyme concentrations to obtain calcium-tolerant myocytes. These myocytes have electrophysiological properties as reported in the literature.

In small (less than 300 microns diameter) pulmonary arterial (PA) rings isolated from the cat, hypoxia induced a transient contraction (250 +/- 120 mg, n = 7), whereas in rings of rabbit PA of the same size, hypoxia had no significant effect (n = 19). Precontraction by 40 mmol KCl.1(-1), noradrenaline (NA) 10(-6) mol.1(-1), or histamine (His 10(-5) mol.1(-1)) did not modify this difference between the two species and did not potentiate the hypoxic contraction of small rings of the cat PA. Large rabbit pulmonary arterial segments (300-2000 microns) exhibited no response to hypoxia before precontraction (n = 15). In the presence of procaine (2%) rabbit PA rings (n = 6, small) exhibited no hypoxic contraction. These results in vitro reflect previous in vivo observations.

The study of the Frank-Starling's law in mammalian single cells has been hindered by a lack of an easily performed method of stretching cells. Some authors have succeeded in this but their methods required a great deal of technical expertise and in most cases they have not had much success. We have developed an easy method of stretching mammalian ventricular cells from slack sarcomere length (S.L.) (Lo, 1.77 +/- 0.05 microns) to about 117% of this length. Thin carbon fibers (12 microns in diameter) which can be bound electrochemically to the cell membrane surface have been used. A flexible long fiber of known compliance (80 microns/microN) was attached to one end of the cell and a stiff double fiber (4 microns/microN) to the other end. The cell attachment was relatively easy to perform and successful results were obtained in 80% of the attempts. The displacement of the flexible fiber allows the quantitative measurements of the resting tension in a group of non-stimulated cells and of auxotonic contractions developed upon stimulation in another group of cells. Increasing S.L. from Lo to 105-106% of Lo, an increase in active tension from 0.21 +/- 0.03 mN/mm to 0.26 +/- 0.01 mN/mm (n = 4) could be noticed with a stimulation frequency of 0.5 Hz. An increase in active tension was also observed at 1 Hz. Staircase kinetics were accelerated with stretching; this confirms at the single cell level the hypothesis of an effect of length-dependent activation on the staircase. Eulerian differential stiffness constant was calculated and found to be 13.5 +/- 1.2, a value which is comparable to that described in intact heart. Thus the important stiffness found in the whole heart may be due to intracellular component(s) such as myofilament and/or connectin.

Contractile responses of left extrapulmonary artery segments with either intact or damaged endothelium were examined with changes in PO2 at constant pH. Hypoxia consistently reduced noradrenaline-induced contractions. This hypoxia-induced relaxation was sometimes followed by a contraction and a second relaxation. Hypoxia-induced relaxations were also obtained if precontraction was elicited with KCl (40 mM), but no triphasic response was observed. Relaxations faded away with time and only contractions were then observed. Relaxations were more considerable in rings from young animals and, although always present, decreased with the age of the animal. Endothelium damage reduced hypoxia-induced relaxations. Indomethacin, a potent blocker of cyclooxygenase, increased hypoxia-induced contraction and reduced relaxation in segments with intact endothelium. Without endothelium the indomethacin effect was less significant. It is concluded that the response to hypoxia of the pulmonary artery is similar to that of systemic vessels. Endothelium seems to play a modulatory role in the hypoxia-induced response in extrapulmonary artery. Prostaglandin metabolism seems to play a minor role in this modulation.

Antibodies directed against a peptide corresponding to the second loop of the human beta2-adrenergic receptor were induced in rabbits by immunisation with the free peptide in complete Freund's adjuvant. The resulting antibodies were affinity-purified and shown to be monospecific for the target receptor. They were able to stimulate the L-type Ca2+ channels in whole-cell patch-clamp experiments on isolated adult guinea-pig cardiomyocytes. This effect was similar to that obtained by the specific beta2-adrenergic agonist zinterol. The antibody effects could be blocked with the specific beta2-adrenergic inverse agonist ICI118,551 but not with the neutral antagonist alprenolol. These results suggest that the antibodies recognise the active conformer of the beta2-adrenergic receptor.

1. The cardiotonic effect of heptaminol hydrochloride (Hept-a-myl, Delalande) was studied using 31P-nuclear magnetic resonance (n.m.r.) spectroscopy and left ventricular pressure (LVP) measurements in rat isolated hearts. The possibility of this effect being mediated by an intracellular realkalinisation was tested. 2. Isolated hearts were perfused at 10 ml min-1 by the Langendorff method with Krebs-Henseleit solution at 37 degrees C and stimulated at 5 Hz. Mechanical activity was measured as variations of left ventricular pressure (LVP). 31P-n.m.r. spectra were recorded every 2 min. Changes in cardiac adenosine triphosphate (ATP), phosphocreatine (PCr) and inorganic phosphate (Pi) were followed and intracellular pH (pHi) was estimated from the chemical shift of Pi. 3. The effects of heptaminol were tested in different conditions: normoxia, moderate ischaemia, severe ischaemia, and moderate ischaemia in the presence of amiloride or guanidinium chloride as inhibitors of the Na-H exchange. 4. In normoxia, heptaminol induced a cyclic increase of systolic LVP, associated with an increase in Pi. No significant effect on pHi was observed. In changing from normoxia to moderate ischaemia, PCr and systolic LVP decreased; a mild intracellular acidification (pHi 6.96) was obtained. Heptaminol induced a restoration of pHi and increased LVP. In severe ischaemia, the realkalinization effect and the restoration of LVP induced by heptaminol were no longer observed. During moderate ischaemia, Na-H exchange inhibitors decreased pHi and LVP. Heptaminol applied in the presence of these inhibitors was unable to restore pHi and LVP. In severe ischaemia, the realkalinization effect and the restoration of LVP induced by heptaminol were no longer observed. During moderate ischaemia, Na-H exchange inhibitors decreased pHi and LVP. Heptaminol applied in the presence of these inhibitors was unable to restore pHi and LVP. 5. These results suggest that the positive inotropic effect of heptaminol during moderate ischaemia could be related to a restoration of internal pH, possibly mediated by a stimulation of the Na-H exchange.