INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

COMPARISON of hypoxia-sensitive and hypoxia-resistant ectothermic vertebrates has not revealed a simple relation between in vivo anoxia tolerance of the species and anoxic performance of in vitro cardiac preparations. In fish, isometric force production of ventricular muscle deteriorates faster in the hypoxia-sensitive trout than in the hypoxia-resistant eel (12, 13). In contrast, studies on freshwater turtles have demonstrated that heart rate and cardiac output are relatively strongly but reversibly depressed during anoxia without significant difference between the hypoxia-sensitive and hypoxia-resistant species (30). The variability in contractile responses might reflect phylogenetic differences in the energy supply/demand ratio of cardiac muscle or in the strategy that different species use to overcome the anoxic period. Maintenance of high cardiac activity requires enhancement of glycolytic energy production with danger of metabolic poisoning of the tissue and depletion of cardiac energy reserves. On the other hand, depression of cardiac contractility decreases the energy demands of the heart, but sets strict limits on the activity of the animal. Because glycolysis yields only a small portion of the ATP produced by aerobic metabolism, an adjustment of energy-producing and energy-consuming processes becomes increasingly important with prolongation of the anoxic period. Adjustment of cardiac function to a limited availability of metabolic energy can occur through peripheral sensing of environmental oxygen content and subsequent neurally mediated inhibition of cardiac contractility (17) or by locally produced metabolites [e.g., adenosine (Ado)] (22). Complete depletion of tissue oxygen content might directly modify contractile function by disturbing the intracellular Ca²⁺ homeostasis of cardiac myocytes (3). The inhibitory vagal reflex, mediated through muscarinic cholinergic receptors, induces anoxic bradycardia in several fish species and reduces the contractility of atrial tissue in some teleosts (17). Purinergic control of the heart is well characterized in mammals but poorly understood in fish. In mammals Ado, the major purine compound, tends to balance cardiac function under energy-limited conditions by reducing the work of the heart and maintaining myocardial energy supply through glycolysis by stimulating cellular glucose uptake (22). Although the purinergic and cholinergic systems have their own receptors on the sarcolemma of the cardiac myocyte, the two systems converge at the level of sarcolemmal ion channels by coupling to a specific class of inwardly rectifying potassium channels. Activation of these channels generates an inward current (I_{K Ado/ACh}) that curtails action potentials (AP) and thereby attenuates sarcolemmal Ca²⁺ influx and the contractile force (4).

The crucian carp is a teleost species that has adapted to live in environments where long-lasting oxygen shortage is a regular seasonal phenomenon. The extraordinary anoxia-tolerance of this fish species is based on huge glycogen stores, specialized anaerobic metabolism, and especially a reduction in activity (for references see Ref. 27). This seasonal anoxia tolerance is improved by a low ambient temperature that effectively suppresses the metabolic rate of different tissues. For example, the activity of the crucian carp heart is strongly depressed by low winter temperatures in the absence of positive thermal compensation. Winter torpor is manifested as a low heart rate and a limited ability to produce force at high contraction frequencies (20). At the cellular level, seasonal adaptation of heart function involves temperature-induced reductions in myofibrillar ATPase activity and Ca²⁺-uptake rate of sarcoplasmic reticulum, temperature-induced changes in myosin heavy chain composition, and anoxia-induced depression of sarcolemmal Na⁺-K⁺-ATPase activity (1, 2, 28). All of these changes occur during either long-term acclimation in the laboratory or seasonal acclimatization in nature. The physiological importance of these gradual, long-term adaptations is evident when considering the long and steady anoxic conditions to which the fish are subjected in winter. It is not clear, however, how the crucian carp heart function changes, or how its contractile performance is regulated when the heart is subjected to anoxia without a prior acclimation period. The acute anoxic response should be particularly important during summer months when the fish are likely to encounter hypoxic or anoxic water on the bottom of the pond during feeding bouts. Therefore, the aim of the present study was to examine the acute anoxic response of the crucian carp heart, and to clarify the importance of purinergic and muscarinic cholinergic systems in contractile and electrical activity of cardiac tissue. The contractile function of the crucian carp heart is relatively sensitive to anoxia. The anoxic depression of crucian carp heart function can occur directly at the myocyte level, or can be mediated by cholinergic receptors at the level of pacemaker cells and atrial muscle. In contrast, the effects of Ado on crucian carp heart are relatively weak, suggesting that the purinergic system is not involved in the acute anoxia response of the crucian carp heart.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animals. Crucian carp (n = 166), with a mean body mass of 25 ± 3 g, were used in these experiments. They were captured from local ponds during the summer (June-September) and were kept in 500-l metal tanks at 22°C in the laboratory. Aerated tap (ground) water circulated constantly through the tanks at a rate of 0.5 l/min. Fish were fed ad libitum with nutrient pellets (Ewos, Turku, Finland). The photoperiod was 12:12-h light-dark.

In vitro experiments with whole hearts. All in vitro experiments were conducted at room temperature (22 ± 1°C), i.e., at the acclimation temperature of the fish. Spontaneously beating hearts were used to measure the effects of Ado and carbachol (Cch) on the heart rate and the effects of NaCN on heart rate, force production, and pumping capacity of the crucian carp heart in vitro. Fish were stunned by a sharp blow to the head. Whole hearts, with an intact sinoatrial pacemaker, were carefully excised and placed in an oxygenated physiological solution (in mM: 150 NaCl, 5.0 KCl, 1.2 MgSO₄, 1.2 NaH₂PO₄, 1.8 CaCl₂, 10 HEPES, and 10 glucose at pH 7.6). The heart was mounted on the bottom of the tissue chamber (15 ml, covered with a layer of silicon) at the bulbus arteriosus with small needles and connected at the apex of the ventricle with a small hook and a short braided silk suture (Davis Geck 6-0) to the force transducer. A resting tension of 0.3 g was carefully applied by adjusting the position of the force transducer. The heart was allowed to equilibrate for ~1 h to reach a stable heart rate and force production before the experiments were started. The wet weight of the whole intact hearts varied between 15.1 and 37.9 mg. Contraction frequency and peak developed force (F_max) were recorded on paper with a polygraph. The preparation was also visually observed to watch for possible atrioventricular conduction blockade.

In vitro experiments with paced atrial and ventricular preparations. Inotropic effects of NaCN, Cch and Ado were examined using paced atrial and ventricular preparations. Whole atria, between 1.2 and 7.5 mg in wet weight, were used. One corner of the atrial muscle was punctured by a small silver hook, which was attached to the force transducer. The opposite corner of the atrium was fixed to the bottom of the chamber with a fine stimulating needle electrode (platinum; Grass). Ventricular preparations were between 9.5 and 26.7 mg in wet weight and included approximately two thirds of the whole ventricle; one third of the ventricular muscle close to the atrial orifice was removed to avoid spontaneous contractions of atrial origin and to allow free access of physiological saline to the preparation. The ventricular wall of the crucian carp heart is exclusively trabecular (no compact layer) and <1 mm thick in this size heart, so there should have been no limitation in oxygen delivery to muscle cells under control conditions. Ventricular muscle was fixed to the bottom of the tissue bath from its base with a stimulating needle electrode and connected from the apex to the transducer with a hook and a silk suture. The other stimulus electrode (platinum plate; Grass) was placed close to the muscle. Muscles were induced to contract at a rate of 1.0 Hz with square-wave pulses 5 ms in duration and 1.5× the threshold voltage from a Grass SD 6 stimulator. The frequency output of the stimulator was controlled by a computer via digital-to-analog converter (DigiData 1200; Axon Instruments, Fremont, CA) and a custom-made signal conditioner. With the aid of micromanipulators, the muscles were stretched to the length at which force production was close to maximal. Contractions were recorded with the polygraph on paper and stored on the hard disk of a computer after analog-to-digital conversion. The time course of single twitches was analyzed off-line with AxoScope or PClamp 6 software (Axon Instruments). Time to F_max (T_PF) is the duration of the contraction phase from the stimulus pulse to F_max. Time to half relaxation (T_0.5R) is the duration of relaxation from the F_max to the point where developed force has declined to half of its maximum value.

Electrophysiological recordings on enzymatically isolated atrial and ventricular cells. APs and ionic currents were measured with the whole-cell patch-clamp technique as described in detail elsewhere (29). For isolation of the cells, a cannula was inserted through the bulbus arteriosus into the ventricle, and the heart was retrogradely perfused from a height of 50 cm, first with a nominally Ca²⁺-free solution to disrupt Ca²⁺ dependent bonds between cells and then with proteolytic enzymes to dissolve intercellular connective tissue. The composition of the Ca²⁺-free saline was as follows (in mM): 100 NaCl, 10 KCl, 1.2 KH₂PO₄, 4 MgSO₄, 50 taurine, 20 glucose, and 10 HEPES at pH 6.9 with KOH. The Ca²⁺-free perfusion lasted for 10 min and was followed by 30 min of enzymatic digestion with collagenase (Sigma Type IA; 1.0 mg/ml) and trypsin (Sigma Type IX; 0.75 mg/ml). The enzymes were dissolved in the Ca²⁺-free medium that also contained 0.75 mg/ml fatty-acid-free BSA. Both solutions were continuously oxygenated (100% O₂), and the enzyme solution was recirculated with a peristaltic pump. After digestion, atrium and ventricle were separated, and the tissues were chopped with scissors into small pieces. Single cells were liberated by agitation of the tissue pieces through the opening of a Pasteur-pipette. Myocytes were suspended in physiological saline and stored at room temperature. Small aliquots of cells were taken for electrophysiological recordings on the bottom of a small chamber (0.5 ml) with a constant flow of physiological solution (2 ml/min). Patch pipettes were prepared from borosilicate glass (Modulohm, Denmark) by means of a Narishige PB-83 pipette puller. Pipettes were filled with potassium-based electrode solution (in mM: 140 KCl, 5 Na₂ATP, 1 MgCl₂, 0.03 Na₂GTP, and 10 HEPES at pH 7.2) for the recording of APs and potassium currents. Appropriate stimulus protocols were generated with PClamp-software and delivered after digital-to-analog conversion (TL-1 DMA, Axon Instruments) through the amplifier (AxoPatch 1D) to the cell. APs and potassium currents were recorded in current-clamp and voltage-clamp modes of the AxoPatch amplifier, respectively. Signals were analyzed off-line with the Clampfit portion of the PClamp-software package. Electrophysiological responses to cyanide (0.1 mM) or oxygen deprivation (sodium dithionate, 0.1 mM) did not normally reach steady state during the experiment but proceeded progressively and resulted finally in cell death. Only recordings obtained before any visible contracture of the cells were included in the data analysis.

In vivo experiments. Heart rates of living fish were recorded in vivo using two fine metal electrodes that punctured through the belly on both sides of the pericardium, with the connecting leads attached on the ventral surface of the fish with adhesive tape (27). After attaching the electrodes, the fish were placed in a 2-l perforated plastic beaker that was submerged in a bigger water-filled container (10 l). Electrocardiograms were observed on the screen of an oscilloscope. Hypoxia was induced by switching from oxygen (100%) to nitrogen (100%) gassing. This procedure reduced the oxygen content of the water to ~0.8 mg/l within 5 min but did not produce true anoxia even when continued for several hours (27). Thus in these experiments severe hypoxia was induced rather than actual anoxia. After a control run, cholinergic (atropine, 2 mg/kg) and purinergic (aminophylline, 50 mg/kg) receptor blockers were injected into the fish intraperitoneally. Receptor blockers were dissolved in 0.9 M NaCl.

Statistics. The data are given as means ± SE. Comparisons within treatment groups were accomplished with paired t-tests. The significance of any differences between atrial and ventricular preparations was assessed with a repeated-measures ANOVA with Scheffe's post hoc test for multiple comparisons. The percentage values of F_max recordings were arcsine transformed before analysis (SigmaPlot, Jandel Scientific). P values < 0.05 were considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Heart rate in vitro. The effects of NaCN (an inhibitor of mitochondrial cytochrome-c oxidase), Cch (a muscarinic cholinergic agonist), and Ado (an A₁-purinergic agonist) on the spontaneous beating rate of the crucian carp heart are shown in Fig. 1. Each of these treatments caused bradycardia that was very strong for cyanide anoxia and Cch but much weaker in the case of Ado. It is notable that Cch exerted its effect at much lower concentrations than did Ado. The nonspecific purinergic blocker aminophylline (1 mM) induced a sustained increase in the basal heart rate from 61.3 ± 5.8 to 79.2 ± 4.7 beats/min (P < 0.05) but did not block the effect of cyanide (3 mM), which depressed the heart rate down to 39.5 ± 8.4 beats/min (n = 5).

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Effects of sodium cyanide, adenosine (Ado), and carbachol (Cch) on spontaneous beating rate of excised crucian carp hearts in vitro. A: time-dependent effect of cyanide on heart rate. Oxygenation of solution was stopped at moment of cyanide addition (arrow). Inhibition of heart rate reached level of statistical significance (P < 0.05) 8 min after cyanide addition. Results are means ± SE of 6 preparations. B: dose-dependent effects of Cch (3 × 109 to 3 × 107 M) and Ado (106 to 3 × 103 M). Drugs were added cumulatively at 10-min intervals. * Significantly different from controls (C). Results are means ± SE of 6 preparations for both treatment groups.

Heart rate in vivo. Decreasing the water oxygen content with nitrogen gassing induced an almost immediate bradycardic reflex that attained steady state within 15 min. Aminophylline (50 mg/kg), injected intraperitoneally, exerted an effect similar to that observed in the in vitro experiments; it induced a prominent increase in the basal heart rate, from 56.2 ± 8.0 to 84.0 ± 2.0 beats/min (P < 0.05), but did not block the effect of oxygen deprivation, which depressed the heart rate down to 64.5 ± 1.5 beats/min (n = 5). These findings suggest that the purinergic system is not involved in hypoxic bradycardia. In contrast to aminophylline, atropine (2 mg/kg; a muscarinic cholinergic receptor blocker) completely prevented hypoxic bradycardia (Fig. 2). The latter finding indicates that the bradycardic reflex in the crucian carp heart is totally mediated by muscarinic cholinergic receptors. Furthermore, atropine induced marked tachycardia, indicating the presence of significant cholinergic tone in the resting fish.

View larger version (29K): [in this window] [in a new window]   Fig. 2.   Atropine block of hypoxic bradycardia in crucian carp in vivo. Hypoxic bradycardia was induced by nitrogen gassing, which reduced the oxygen content of water to ~0.8 mg/l. On restoration of normoxia, heart rate completely recovered. Next, fish were given atropine (2 mg/kg) by intraperitoneal injection. Atropine induced immediate tachycardia and totally blocked the hypoxic bradycardia. Sequences of oxygen/nitrogen gassing and atropine treatment are shown by horizontal bars in lower part of figure. Results are means ± SE of 6 fish. Temperature was 22 ± 1°C.

Force production of atrial and ventricular tissue. Cyanide (3 mM) caused an abrupt depression of force development (negative inotropic effect), both in atrial and ventricular tissue paced to contract at a rate of 1 Hz. In the atrium, the negative inotropic effect was associated with a slight decrease in the contraction duration, whereas in the ventricle T_0.5R was significantly increased by cyanide treatment (Fig. 3A). Aminophylline (1 mM) slightly decreased the contraction duration but did not antagonize the effects of cyanide, suggesting that purinergic receptors are not involved in the response (Fig. 3B). In the absence of aminophylline, cyanide-induced depression of F_max was stronger in ventricular tissue than in atrial tissue. This difference was, however, smaller and not statistically significant in the presence of aminophylline.

View larger version (33K): [in this window] [in a new window]   Fig. 3.   Effect of cyanide (3 mM) on contractility of paced atrial and ventricular preparations of crucian carp heart in absence (A) and presence (B) of 1 mM aminophylline. Time-dependent effects of cyanide on peak force (Fmax), time to peak force (TPF) and time to half relaxation (T0.5R) of atrial () and ventricular () tissue are indicated. * P < 0.05 vs. control value; # P < 0.05 between atrial and ventricular preparations. Horizontal solid and dashed lines indicate application of aminophylline and cyanide, respectively. Pacing frequency 1 Hz; temperature 22 ± 1°C.

In the above experiments, cardiac preparations were paced at normoxic heart rates (1 Hz), and the contractile failure might indicate an imbalance between the energy supply and demand at the high work load. Therefore, some experiments were conducted on spontaneously beating hearts where changes in heart rate might allow a reduction of the work load under anoxic stress. As in paced cardiac preparations, cyanide initially induced a strong depression of force development in the spontaneously beating whole hearts, even though the heart rate was markedly reduced (Fig. 4). In some preparations, however, a significant recovery of the contractile force occurred later on (in the sustained presence of cyanide), and in one case this secondary force recovery exceeded the initial control level (Fig. 4). Thus it seems that if the work load is sufficiently reduced by lowering the heart rate, contractile failure can be avoided. In all seven hearts, cyanide induced a rapid and sustained inhibition of the pumping capacity (20), indicating that the function of the heart was set to a lower functional level during anoxia.

View larger version (24K): [in this window] [in a new window]   Fig. 4.   Effect of cyanide on contraction force and frequency of spontaneously beating crucian carp hearts in vitro. Whole hearts were mounted for recording to determine how force production is altered by cyanide when heart rate is allowed to change freely. A: there were two types of responses to cyanide (3 mM). In all preparations there was an initial depression of both heart rate and force of contraction after 1 h of cyanide application (1 h). Later on, there was a progressive decrease in heart rate and either recovery of force generation (left; CN 4 h) or further depression of force generation (right; CN 4 h). Note more prominent decline of heart rate and marked prolongation of contraction duration in preparation where recovery of force production occurred. B: mean results (± SE) from 7 hearts indicating marked depression of heart rate (HR), force production, and pumping capacity (calculated as product of heart rate and Fmax) by cyanide.

Ado, at a concentration of 1 mM, slightly inhibited the contractile force in atrial tissue and clearly enhanced force production in the ventricle (Fig. 5). These modest inotropic effects were associated with a decrease in the contraction duration in the atrium and a slight but insignificant prolongation of the twitch in the ventricle. Cch induced a strong and dose-dependent inhibition of the contractile force in atrial tissue, which was associated with decrease in the contraction duration. In the ventricle, a reduction of ~20% in the maximal force was evident; this small and linear force decline was largely a result of the time-dependent deterioration of the preparation (Fig. 5).

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Dose-dependent effects of Ado and Cch on Fmax, TPF, and T0.5R of paced atrial () and ventricular () preparations. Results are means ± SE of 6 preparations. * P < 0.05 vs. control value. Pacing frequency 1 Hz; temperature 22 ± 1°C.

Electrophysiological responses. A series of experiments was conducted to determine the mechanisms that underlie the anoxia-associated changes in contractile function of the crucian carp heart. We noticed some prominent differences between the basic electrophysiological characteristics in atrial and ventricular cells. The atrial AP was much shorter in duration than the ventricular AP, and the density of inwardly rectifying potassium currents (I_K1) was much smaller in atrial than in ventricular cells (Figs. 6 and 7). In these two respects, atrioventricular differences in crucian carp heart are very similar to those in the mammalian heart (11). Cyanide (0.1 mM) caused strong depolarization of the resting membrane potential (RMP) both in the atrial and ventricular cells (Fig. 6). The RMP was practically abolished after 15-min exposure to cyanide, and the cell entered into contracture. There was also a notable difference between atrial and ventricular cells in that the AP duration was prolonged by cyanide in ventricular myocytes but not in atrial myocytes.

View larger version (9K): [in this window] [in a new window]   Fig. 6.   Effects of cyanide (100 µM) on resting membrane potential and action potential (AP) of enzymatically isolated ventricular (A) and atrial (B) cells of crucian carp heart. Recordings are representative examples for 3 atrial and 4 ventricular cells.

View larger version (28K): [in this window] [in a new window]   Fig. 7.   Effects of Ado (A) and Cch (B) on AP duration (top) and inwardly rectifying potassium current (IK ACh/Ado; bottom) in enzymatically isolated atrial (b, d) and ventricular (a, c) cells of crucian carp heart. APs are representative examples of 3 cells. Current recordings are means ± SE of 5-11 myocytes as indicated. * P < 0.05 vs. control (C) value. pA/pF, current density.

The ionic mechanisms underlying the cyanide-induced changes in RMP and AP were then examined. The RMP is determined by I_K1 flowing through the inwardly rectifying potassium channels. To record potassium currents, we used potassium-based intra- and extracellular solutions and blocked Ca²⁺ and Na⁺ channels with Cd²⁺ (200 µM) and tetrodotoxin (1 µM), respectively (29). It is also notable that the pipette solution contained 5 mM ATP. If the cyanide-induced depolarization was caused by inhibition of background potassium conductance, both an inward shift of the holding current (HP) and a reduction of I_K1 should be seen. In accordance with its depolarizing effect, cyanide (0.1 mM) induced a prominent inward shift of HP (40 mV; Fig. 8). This was not, however, associated with inhibition of the inward rectifier; rather, quite in contrast, the total inward current was enhanced by cyanide. This strongly suggests that the cyanide-induced depolarization was not caused by inhibition of background potassium conductance. Furthermore, voltage-dependence of the cyanide-induced current is quite different from that expected for a potassium current: it reversed the zero current level at slightly negative voltages (20 mV; reversal potential for K⁺ = 85 mV) and reached maximum at approximately 80 mV (Fig. 8). Because intracellular Ca²⁺ was not buffered, a likely candidate for the cyanide-induced current was some Ca²⁺-activated conductance, e.g., a Na⁺-Ca²⁺ exchange current. We tested this hypothesis by trying Ni²⁺, a Na⁺-Ca²⁺ exchange blocker. The cyanide-induced inward current was strongly inhibited by 2 mM NiCl₂ (Fig. 9A). The Ni²⁺ sensitivity suggests that this current component arises from the operation of Na⁺-Ca²⁺ exchange in its forward mode (Ca²⁺ efflux). Thus we have to suppose that the aerobic block causes first a rise of intracellular Ca²⁺ concentration, and subsequently the extrusion of intracellular Ca²⁺ through Na⁺-Ca²⁺ exchange generates the inward current and depolarizes the RMP. Prolongation of ventricular AP is also consistent with this hypothesis (9). The electrophysiological responses to cyanide were obtained in the presence of 5 mM ATP in the pipette solution, suggesting that the changes in membrane potential and current are not caused by the depletion of cytosolic ATP (perfusion of the cell with 5 mM ATP should maintain the mean cytosolic ATP level at a fairly constant level) or that the depletion of ATP occurs in some diffusion-restricted space not freely equilibrated with the bulk cytoplasm. It is also possible that the responses were unrelated to poisoning of the aerobic metabolism, and might represent some direct effects of cyanide on sarcolemmal ion channels. To check for the latter possibility, we conducted some experiments with sodium dithionate (Na₂S₂O₄), an oxygen scavenger, which removes molecular oxygen from the solution. Fig. 9C shows that 0.1 mM Na₂S₂O₄ induced an inward current similar to cyanide. Thus it seems very likely that the electrophysiological changes in the presence of cyanide are closely related to inhibition of the mitochondrial respiratory chain and the disruption of mitochondrial function somehow leads to the generation of an inward Na⁺-Ca²⁺ exchange current in crucian carp cardiac myocytes.

View larger version (21K): [in this window] [in a new window]   Fig. 8.   Effects of cyanide (0.1 mM) on whole-cell membrane current when Ca2+ current and Na+ current are blocked by Cd2+ (200 µM) and TTX (1 µM), respectively. A: in control conditions, main current component is background potassium conductance (IK1) flowing through inwardly rectifying potassium channels (left). Cyanide induces inward shift of holding current (40 mV) and increases total inward current (middle). Note that effect of cyanide does not reach steady state but develops during recording. This response is reversible on washout of cyanide (right). Bar on left side of each recording shows zero current level. Stimulus voltage protocol is shown in B. Mean results (± SE) from 7 cells are shown in C. *P < 0.05 vs. control value. All recordings were made in presence of 5 mM ATP in pipette solution.

View larger version (18K): [in this window] [in a new window]   Fig. 9.   Nickel (2 mM) blocks major part of cyanide-induced inward current. Because cyanide response changes rather rapidly with time (see Fig. 7), a voltage ramp protocol (B, inset) was used to record current-voltage relationship. A: original current recordings in absence and presence of 0.1 mM NaCN. Cyanide-induced current is largely blocked by 2 mM NiCl2. Incomplete block is at least partially caused by progressive development of cyanide response. B: mean (± SE) density of Ni2+-sensitive current from 3 cyanide-treated ventricular cells of crucian carp heart. C: sodium dithionate, an oxygen scavenger, induces similar response in membrane conductance as does cyanide. C, Control.

The in vivo recordings of heart rate revealed that the parasympathetic system is activated during hypoxia, and the in vitro recordings indicated that the cholinergic system is an effective regulator of atrial contractility in crucian carp heart. Therefore, it is important to know how activation of cholinergic receptors affects the electrical activity of atrial and ventricular myocytes. Cch did not have any effects on electrical activity of ventricular cells: AP and potassium conductance of the sarcolemma remained unchanged (Fig. 7). These results are consistent with the findings on intact tissue showing no response to Cch (Fig. 5). In contrast to the effects on ventricular cells, Cch (1 µM) induced very prominent effects in atrial myocytes. The duration of the AP was dramatically decreased, and this was associated with a marked increase in I_{K Ado/ACh}. These findings suggest that Cch activates inwardly rectifying potassium channels, which increase potassium conductance of the sarcolemma and decrease the AP duration. This will subsequently lead to a reduction in the sarcolemmal Ca²⁺ influx and a decline in the contractile force, as was recorded in intact atrial tissue (Fig. 5). The decreased AP duration is also consistent with the decreased contraction duration in atrial tissue by Cch (Fig. 5). The electrophysiological responses of crucian carp atrial and ventricular tissue to muscarinic cholinergic activation are very similar to those reported for the mammalian heart (4).

Ado at a concentration of 100 µM did not significantly affect atrial or ventricular AP (Fig. 7), and did not change the I_{K Ado/ACh}. These findings are consistent with the absence of a force response of atrial and ventricular tissue to Ado at this drug concentration (Fig. 5).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Effects of anoxia. The present results indicate that blockade of aerobic metabolism induces relatively rapid depression of contractile function in the crucian carp heart. Heart rate and force development were decreased so much that the pumping capacity of the anoxic heart was only a small percentage of its normoxic value. In another hypoxia-resistant fish species, the flounder, contractile activity of the heart is also severely compromised under hypoxic conditions (19), suggesting that this kind of response might be common to several hypoxia- and anoxia-resistant fish species. The prominent decline in heart rate and force production of crucian carp heart under anoxia is also similar to the response of working heart preparations from freshwater turtles (30). Thus our findings on crucian carp are consistent with the concept that under oxygen deprivation the hearts of ectothermic vertebrates intrinsically reduce their energy demand either by reducing pressure development or by conversion to intermittent beating (21). The reduction in demand might provide protection for cardiac myocytes against anoxic damage.

The depression of force production in anoxia is probably the result of elevations in intracellular concentrations of inorganic phosphate and protons (30), which inhibit force production by reducing Ca²⁺ sensitivity of cardiac myofilaments (8). When aerobic metabolism is blocked, inorganic phosphate is generated mainly in the creatine kinase-catalyzed hydrolysis of phosphocreatine. In the crucian carp heart, the content of total creatine is relatively high, which might indicate an efficient means for downregulation of contractility by liberation of inorganic phosphate from phosphocreatine (6). Owing to the effects of protons and inorganic phosphates on myofibrillar function, the decrease in developed force can occur without any changes in the Ca²⁺ transient, as noted in mammalian cardiac muscle (3). In some preparations of crucian carp heart in the continued presence of cyanide, the initial depression of developed force was followed by partial or complete recovery of force generation, suggesting that the intracellular Ca²⁺ transient was not decreased and might even be increased under anoxia. Although the rate of relaxation was decreased, there were no changes in resting tension, suggesting that diastolic Ca²⁺ was not markedly changed at this phase. The appearance of a Ni²⁺-sensitive inward current, an indication of increased Ca²⁺ efflux through the Na⁺-Ca²⁺ exchange denotes that, at a later phase of anoxia, cytosolic Ca²⁺ concentration rises significantly. The source of this "extra" Ca²⁺ is not completely clear, but it is unlikely to be of extracellular origin because sarcolemmal Ca²⁺ influx through L-type channels was blocked with Cd²⁺. The electrophysiological effects of aerobic block were recorded in the presence of exogenous ATP, and therefore the increase in intracellular Ca²⁺ is also unlikely to be a result of the inhibition of ATP-dependent processes, e.g., sarcolemmal Na⁺-K⁺ pump or sarcoplasmic reticulum Ca²⁺-pump. In mammalian cardiac myocytes, NaCN disrupts the mitochondrial potential, which results in release of mitochondrial Ca²⁺ and elevation of the resting cytosolic Ca²⁺ concentration (5). A similar sequence of events could explain the electrophysiological effects of cyanide poisoning in crucian carp ventricular myocytes under conditions where sarcolemmal Ca²⁺ influx was blocked and intracellular ATP was buffered.

There were some distinct differences between atrial and ventricular tissues in response to aerobic block. F_max decreased more in ventricular muscle than in atrial muscle during anoxia, and the suppression of force generation was associated with decreased contraction duration in atrial tissue, but with a prolongation of contraction in ventricular preparations. Furthermore, the electrophysiological effects of anoxia were different in atrial and ventricular cells: both myocyte types were depolarized by cyanide, whereas the prolongation of AP occurred only in ventricular myocytes. The RMP is normally maintained by background I_K1 flowing in an outward direction. Depolarization can be induced by either reduction of the background potassium conductance or by induction of an additional inward current. The depolarization of RMP in crucian carp myocytes is explained by the cyanide-induced inward current, carried probably by the forward mode of the Na⁺-Ca²⁺ exchange (Ca²⁺ efflux). Although the effects of cyanide on membrane currents were systematically studied only in ventricular cells, a cyanide-induced inward current was also found in atrial cells. The absence of a significant AP prolongation in atrial cells by cyanide might be a result of the preponderance of outward potassium currents in the rapid repolarization of atrial AP or a lower density of Na⁺-Ca²⁺-exchange current in atrial myocytes. Depolarization of RMP and prolongation of AP with a simultaneous depression of force production were previously observed in hypoxic flounder heart (19).

Effects of carbachol. Anoxia- and hypoxia-induced depression of the heart rate was completely blocked by atropine, which indicates that the anoxic bradycardia in crucian carp is exclusively mediated by muscarinic cholinergic receptors as in other teleost species (10). Our results indicate that the atrial tissue of the crucian carp heart is very sensitive to muscarinic inhibition, whereas ventricular tissue is totally insensitive in this respect. Inhibition of atrial contraction by acetylcholine was previously demonstrated in the cod, Gadus morhua (14). Vagal activity is probably a very efficient mechanism of regulating cardiac function under anoxia; first by inducing bradycardia and second by depressing force production by the atrial tissue. The latter effect will also weaken ventricular contractions owing to the importance of atrial contraction to end-diastolic volume and filling of the fish ventricle (16). The atrial inhibition, and possibly the bradycardic response as well, can be explained by an increase in the density of I_{K Ado/ACh}, which causes the decreased AP duration and consequently reduces Ca²⁺ entry through L-type Ca²⁺ channels. The ACh- and Cch-sensitive channel is different from I_K1 but identical to the channel activated by Ado (15). Activation of this potassium current is thought to underlie the effects of acetylcholine as well as the direct effects of Ado that are observed in pacemaker cells and atrial tissue (4). Although Cch-sensitive potassium channels are clearly present in crucian carp atrial cells, the weak responses of atrial preparations to Ado suggest that A₁-receptors are not effectively coupled to this channel or that the purinergic system is weakly developed in the cardiac tissue of this teleost species.

Effects of adenosine. Ado is an endogenous nucleoside that has a significant role in modulating electrophysiological and contractile activity of the vertebrate heart. Ado is released in the circulation in situations where the oxygen demand of the cardiac muscle exceeds the circulatory oxygen supply, i.e., during heavy exercise and in hypoxia or anoxia. The effects of Ado occur at the smooth muscle cells of coronary vessels, the conductive tissue of the heart, and at working cardiac cells of atria and ventricles (22). Ado improves blood flow in the coronary vessels (vasodilatation), reduces the velocity of impulse conduction (negative dromotropic effect), and weakens contractile force in atrial and ventricular cells (negative inotropic effect). By these actions, Ado is thought to rebalance the oxygen demand and oxygen supply of the heart and provide protection for the heart during energy-limited conditions (for review see Ref. 22). Theoretically, Ado is an almost ideal candidate for anoxic protection of the heart because it depresses the rate of energy use and at the same time promotes energy production. Against this background it was rather surprising that the effects of Ado in the heart of anoxia-tolerant crucian carp were very weak. In the atrial tissue of common carp, Cyprinus carpio, Ado depresses cardiac work by having both negative chronotropic and inotropic effects (7, 24). Similarly, in crucian carp heart Ado reduced the beating rate and contraction force of atrial tissue in vitro, but these effects required much higher concentrations than in the common carp. In vivo, Ado is not involved in anoxic bradycardia, because the response was unaffected by the purinergic blocker, aminophylline, but was completely abolished by atropine. The slight negative inotropic effect in atrial tissue is probably secondary to AP-shortening induced by increased potassium efflux through I_{K Ado/ACh}. The physiological importance of this effect is probably minor because it requires millimolar concentrations of Ado to occur. In ventricular tissue the effects of Ado were even weaker than in the atrium. In the ventricle, Ado (1 mM) slightly increased the contractile force, which was associated with an increased twitch duration. In the flounder ventricle, Ado induced a positive inotropic effect at a micromolar concentration (18). The indirect anti--adrenergic effects of Ado are usually less species-specific than the direct effects of Ado, at least in mammals (25). We have not studied the anti-adrenergic effects of Ado, but even these should be rather small because the stimulatory -adrenergic response itself is rather weak in the crucian carp heart (26). Thus the present results suggest that Ado is a weak modulator of cardiac function in crucian carp and unlikely to have any major role in anoxic protection of the heart. We cannot, however, exclude the beneficial effects of Ado on metabolic energy production due to activation of cellular glucose uptake or on preconditioning of the heart to anoxia.

Perspectives