Ca²⁺ uptake in the sarcoplasmic reticulum from the systemic heart of octopod cephalopods

Coleoid cephalopods (octopus, cuttlefish and squid) have high metabolic rates and a relatively low oxygen-carrying capacity in the blood, which implies large cardiac outputs by the standards of invertebrate animals or fish (Wells, 1992). Furthermore, during periods of exercise, the increased rate of oxygen uptake (2.3-fold in Octopus vulgaris; Wells et al., 1983) is met by an increased cardiac output, since there is little or no scope for increasing oxygen extraction (already approximately 80 % in resting conditions; Houlihan et al., 1986).

The high demand on the cardiac pump is also reflected in the cellular characteristics of the systemic cardiac muscle. Isolated cardiac muscle strips from Octopus vulgaris show a regular twitch force development at stimulation frequencies at which the cardiac muscle of other ectothermic vertebrates, such as the rainbow trout, commonly fail (Gesser et al., 1997). Furthermore, ryanodine strongly inhibits twitch-force development and increases resting tension, suggesting that excitation–contraction coupling is highly dependent on Ca²⁺ cycling via the sarcoplasmic reticulum (SR), as is known to occur in the cardiac muscle of the rat.

The strong response of O. vulgaris systemic heart tissue to ryanodine is rather surprising, since it far exceeds that described for cardiac tissue of ectothermic vertebrates. In ectothermic vertebrates, heart tissue is either weakly responsive or unresponsive to ryanodine at physiological heart rates and temperatures (Driedzic and Gesser, 1988; Hove-Madsen, 1992; Møller-Nielsen and Gesser, 1992) although trout ventricular myocytes have recently been shown to possess an SR with a significant Ca²⁺-accumulating capacity and high Ca²⁺ maximal uptake rate (Hove-Madsen et al., 1998). Since the cellular aspects of cardiac muscle function have been poorly studied in cephalopods, the possibility that ryanodine affects mechanisms other than SR-dependent Ca²⁺ cycling in this tissue cannot be excluded.

In addition, ultrastructural studies suggest that in Octopus vulgaris the heart possesses a richly developed SR (Schipp, 1987; Dykens and Mangum, 1979), but no quantitative analysis has been performed. The importance of the SR in cardiac tissue of cephalopods was therefore examined using two alternative methods. In one approach, the Ca²⁺ uptake rate of the SR in crude cardiac homogenates from octopus and rat was compared in terms of the maximum rate of uptake and its Ca²⁺ dependence. Rat heart was chosen because its excitation–contraction coupling has been shown to be extremely dependent on the SR (Bers, 1985). It also has the advantage that this dependence on the SR is retained at lower temperatures, so that ryanodine has virtually the same effect at 20 °C and at 37 °C (Shattock and Bers, 1987). The comparison was therefore carried out at 20 °C, a temperature within the range experienced by the living octopus but below the body temperature of rat.

In a second approach, the importance of the SR was assessed by measuring cell shortening and the ionic current elicited by exposure of isolated cardiomyocytes to caffeine. This substance is known to induce Ca²⁺ release from the SR, which in turn gives rise to a cell contracture and a trans-sarcolemmal ionic current due to Ca²⁺ extrusion by the electrogenic Na⁺/Ca²⁺ exchanger. This approach has previously been used to quantify the Ca²⁺ content of the SR in single cardiac myocytes from both mammalian (Varro et al., 1993) and teleost (Hove-Madsen et al., 1998) species.

Three different species of octopuses were used in the study: Octopus vulgaris Cuvier, Eledone cirrhosa Lamarck and Eledone moschata Lamarck. The animals were caught with trawl-netting by commercial fishermen in fishing grounds near Barcelona (North West Mediterranean, Spain). While at sea, the animals were kept in clean aerated water. They were later transported to the aquarium facilities of the Universitat Autònoma of Barcelona (UAB), where they were maintained in a closed seawater system (3.6 %) at 16 °C and fed live crayfish (Carcinus maenas) every other day. Sprague-Dawley rats were housed in the Animal Facility at UAB and taken to the laboratory on the day of the study.

Thirteen octopuses (306±46 g) and seven rats (402±4 g) (means ± S.E.M.) were used in the study. Octopuses were killed by decerebration and rats by cervical dislocation. The systemic hearts of octopuses and the ventricles of rats were quickly dissected free and placed in chilled homogenization solution (see below). Subsequently, the tissue was cut into small pieces and homogenized using a Polytron homogenizer in 25 volumes (50 volumes for rat tissue) of ice-cold homogenization buffer. The final homogenate was treated with a ground-glass homogenizer. The homogenization solution contained (in mmol l⁻¹): 20 Tris, 2 MgCl₂, 0.01 leupeptin, 0.01 phenylmethylsulfonyl fluoride, 1 dithiothreitol, 2 benzamidine and 250 (rat) or 1000 (octopus) sucrose. The difference in sucrose concentration reflects the different osmolality of the body fluids, which in Octopus are nearly iso-osmotic with sea water (Wells, 1978). The pH of the homogenization solution was adjusted to 7.2 (at 20 °C) using HCl.

Cardiomyocytes from the systemic heart of E. cirhosa were obtained by perfusing the systemic heart with an extracellular solution containing collagenase and trypsin. The composition of this solution was 419 mmol l⁻¹ NaCl, 20 mmol l⁻¹ KCl, 20 mmol l⁻¹ MgCl₂, 10 mmol l⁻¹ Hepes, 0.05 mmol l⁻¹ EGTA, 0.0375 mmol l⁻¹ CaCl₂, 0.25 mg ml⁻¹ collagenase, 0.2 mg ml⁻¹ trypsin and 0.5 mg ml⁻¹ bovine serum albumin (BSA).

Portions of the crude homogenate obtained were added to different uptake solutions containing (in mmol l⁻¹): 80 imidazole, 8 potassium oxalate, 6 MgCl₂, 6 Na₂-ATP, 0.58 EGTA, 0.02 Ruthenium Red, 3 sodium azide and 120 (rat) or 360 (octopus) KCl. The pH was adjusted to 7.4 with HCl. These solutions also contained varying concentrations of CaCl₂, all with ⁴⁵Ca²⁺ (3.7×10⁻³ Bq ml⁻¹). The resulting Ca²⁺ activities were calculated with respect to the concentrations of EGTA, ATP and oxalate, ionic strength and pH using MaxChelator (Bers et al., 1994). Ruthenium Red, which blocks the Ca²⁺ efflux channels of the SR, was included to prohibit the escape of Ca²⁺ taken up by the SR vesicles (Feher et al., 1988). Sodium azide was applied to block mitochondrial accumulation of Ca²⁺ (Ito et al., 1974). To assess the SR-independent Ca²⁺ uptake, control experiments were performed in the presence of cyclopiazonic acid (30 μmol l⁻¹), a blocker of the SR Ca²⁺-ATPase (Takahashi et al., 1995).

At different times after the addition of the crude homogenate, 0.5 ml of the solution, directly followed by 2.5 ml of unlabelled uptake solution, were sucked through membrane filters (0.45 μm, HAWP 025 Millipore). The filter paper was incubated overnight in 4 ml of scintillation solution (Optiphase HiSafe 2, Wallac, Milton Keynes, UK) before counting ⁴⁵Ca²⁺ activity in a scintillation counter. To relate radioactivity to mmoles of Ca²⁺, 50 μl of ⁴⁵Ca²⁺-labelled uptake solution was counted under identical conditions. All the assays were carried out at 20 °C.

Protein-specific Ca²⁺ uptake was calculated after determining the protein concentration of the homogenate (Lowry et al., 1951). Ca²⁺ uptake is expressed as nmol Ca²⁺ min⁻¹ mg⁻¹ protein.

The uptake measurements involved two protocols. In the maximum uptake rate protocol, the uptake solution had a constant pCa of 4.96, and samples were taken every second minute for rat or third minute for octopus to assess the maximal uptake rate, up to a maximum time of 6 min for rat or 9 min for octopus. To assess the dependence of uptake rate on the SR, some of these experiments were performed in the presence of cyclopiazonic acid. The second protocol aimed to quantify the effect of Ca²⁺ availability, and uptake rate was determined after 8 min for rat and 12 min for octopus in a series of solutions with varying pCa.

Ionic currents were measured using the patch-clamp technique in the whole-cell configuration using an EPC-9 software-driven amplifier (HEKA, Germany). The extracellular medium contained (in mmol l⁻¹): 419 NaCl, 20 CsCl, 20 MgCl₂, 10 Hepes, 10 CaCl₂, 4.5 NaHCO₃, 1.5 NaH₂PO₄, 5 sodium pyruvate, 5 glucose. The pH was adjusted to 7.4 using NaOH. The pipette solution contained (in mmol l⁻¹): 450 CsCl, 3.1 Na₂ATP, 4 MgCl₂, 5 sodium phosphocreatine, 0.42 Na₂GTP, 0.025 EGTA, 10 Hepes, 20 tetraethylammonium chloride, and the pH was adjusted to 7.2 using CsOH. To measure the ionic current elicited by exposure to caffeine, the membrane potential was clamped at —80 mV, and the cell was exposed to 10 mmol l⁻¹ caffeine for a few seconds (Hove-Madsen et al., 1998). The charge carried by the caffeine-induced current was measured as the difference between the time integral of the ionic current elicited by caffeine and the time integral of the small constant leakage current at the holding potential (see solid line in Fig. 4A). Assuming that the ionic current was due to Ca²⁺ extrusion by the Na⁺/Ca²⁺ exchanger (q_NCX), with a stoichiometry of 1 Ca²⁺ to 3 Na⁺, one net charge would result per Ca²⁺ extruded. The amount of Ca²⁺ carried by the Na⁺/Ca²⁺ exchange current (Ca_NCX) was obtained as q_NCX/F, where F is Faraday’s constant. Ca²⁺ release from the SR was calculated as Ca_NCXC/V, where C is capacitance and V is non-mitochondrial cell volume, assuming a non-mitochondrial cell volume equivalent to 50 % of the cell volume (Dykens and Mangum, 1979).

Maximal cell shortening in isolated myocytes from the systemic heart was measured manually ‘off-line’ by replaying individual video frames from recordings of the caffeine-induced cell contractures. Maximal cell shortening was normalized to resting cell length.

The ratio C/V was used as a conversion factor to calculate the cell volume from capacitance measurements in patch-clamp experiments.

Values are given as mean ± S.E.M. Differences were analyzed for statistical significance using Student’s t-test with the level of significance set at P=0.05.

In the first set of experiments the rate of uptake of Ca²⁺ was measured in conditions of saturating Ca²⁺ activity (pCa=4.96) with or without cyclopiazonic acid (CPA). The amount of Ca²⁺ recovered in the particulate fraction with time was well described by linear regression. Furthermore, Ca²⁺ accumulation appeared to be exclusively related to the uptake by the SR as it was abolished in the presence of CPA (Fig. 1). The maximal uptake rate of the cardiac homogenate was 43±4 nmol min⁻¹ mg⁻¹ protein for octopus (N=7) and 28±2 nmol min⁻¹ mg⁻¹ protein for rat (N=4). Hence, at 20 °C, the maximal uptake rate is significantly higher for octopus than for rat (P<0.02).

To examine the dependence of the uptake rate on the free Ca²⁺ concentration, the uptake values of the individual data sets were calculated as a percentage of the maximal uptake value. As shown in Fig. 2, the normalized Ca²⁺ uptake has a similar Ca²⁺ dependence for octopus and rat and follows a typical sigmoidal relationship. The fit of individual data to the Hill equation showed that pCa₅₀ values were not significantly different: 6.04±0.02 for octopus and 6.02±0.05 for rat. The Hill coefficients (n_H) were not significantly different, being 2.5±0.2 for octopus and 3.5±0.5 for rat. The dimensions of myocytes isolated from the systemic heart are shown in Table 1.

The effect of caffeine was examined by exposing the cell to 10 mmol l⁻¹ caffeine. This induced a partially reversible contracture and an ionic current. Images of the cell before and during exposure to 10 mmol l⁻¹ caffeine are shown in Fig. 3A.

The time course of the caffeine contracture was relatively slow. In particular, the relaxation was slow and incomplete, although the cell eventually relaxed in the continuous presence of caffeine, suggesting that relaxation is possible even when Ca²⁺ reuptake in the SR is impeded by caffeine. The mean resting cell length was 227.7±18.8 μm (N=6). The caffeine contracture reduced the cell length to 120.8±10.7 μm (Fig. 3B). After washout of caffeine, the mean cell length was 170±22 μm.

Patch-clamp recording of trans-sarcolemmal ionic current permitted the quantification of the electrogenic extrusion of Ca²⁺ by the Na⁺/Ca²⁺ exchanger. Fig. 4A shows the ionic current elicited by exposure of a cardiomyocyte to 10 mmol l⁻¹ caffeine. A horizontal line was drawn through the resting membrane current to determine the net charge carried by the ionic current, which was obtained by subtracting the resting membrane current (solid line in Fig. 4A) from the ionic current trace. The time integral of this net ionic current is shown in Fig. 4B. The total amount of Ca²⁺ carried by the Na⁺/Ca²⁺ exchange current averaged 34.04±7.27 amol pF⁻¹, corresponding to an increase in total intracellular Ca²⁺ concentration due to the caffeine-induced Ca²⁺ release from the SR of 404±86 μmol l⁻¹ non-mitochondrial volume (Table 2). The SR Ca²⁺ release was compared with Ca²⁺ current elicited by a 200 ms depolarization to 0 mV. Fig. 5A shows a typical Ca²⁺ current elicited by a depolarization with 10 mmol l⁻¹ Ca²⁺ in the extracellular solution and Fig. 5B shows the amount of Ca²⁺ carried by this current. The average amount of Ca²⁺ carried by the current (Ca_ICa) is approximately 50 times smaller than that released from the SR by caffeine (Ca_NCX) (Table 2).

The strong inhibition of contractility by ryanodine in Octopus vulgaris cardiac muscle suggested that the SR is important in excitation–contraction coupling (Gesser et al., 1997). Since our knowledge of the cellular cardiac physiology in cephalopods is poor, it is possible that the effect of ryanodine in depressing contractility is not related to its effects on the SR. Instead, it could disrupt sarcolemmal Ca²⁺ influx (Valdivia and Coronado, 1989) or other cellular processes.

The results of this study support a prominent role for the SR in the cephalopod heart. The maximal Ca²⁺ uptake rate at an optimal Ca²⁺ activity was higher in heart homogenates from octopuses than in those from rat. Cardiac excitation– contraction coupling is highly dependent on the SR in the rat (Bers, 1985), even at non-physiological temperatures, given the prominent effects of ryanodine on tension development (Shattock and Bers, 1987). Extrapolating the uptake rate from 20 °C to 37 °C on the basis of a temperature coefficient of 2.21 (Lundblad et al., 1986) provides an uptake rate of 108 nmol min⁻¹ mg⁻¹ for the rat ventricle. While the SR Ca²⁺ uptake rate at 20 °C in octopus was approximately 1.5 times larger than in rat, the maximal uptake rate at the respective physiological temperatures is approximately 2.5 times lower in octopus than values reported for rat ventricle at body temperature.

Our values on oxalate-supported Ca²⁺ uptake are in agreement with those in the literature and also on measurements from permeabilised rat ventricular myocytes (Wimsatt et al., 1990; Hove-Madsen and Bers, 1993) if one assumes that the cardiac myocytes constitute approximately 50 % of the protein content in a crude homogenate (Lew et al., 1991). Our experimental approach, therefore, gives reproducible values for SR Ca²⁺ uptake rates, although higher values have been obtained in single rat ventricular myocytes (Balke et al., 1994).

The SR Ca²⁺ uptake measured for octopus and rat has a similar Ca²⁺ dependence, as is clear from the pCa₅₀ values and the Hill coefficients. The values are in agreement with other studies (Wu and Feher, 1996; Berrebi-Bertrand et al., 1997; Zucchi et al., 1996) and suggest that the SR operates at similar Ca²⁺ concentrations and is capable of participating in muscle relaxation in the octopus heart as in the rat heart. Indeed, if it is assumed that the protein concentration is approximately 120 mg ml⁻¹ in intact cardiac tissue (Pierce et al., 1985), the non-mitochondrial volume is approximately 50 % and that a total Ca²⁺ transient of 85 μmol l⁻¹ is needed to activate a normal contraction (Dykens and Mangum, 1979; Berlin et al., 1994; Delridge et al., 1996; Hove-Madsen et al., 1998), a maximal uptake rate of 43 nmol min⁻¹ mg⁻¹ enables the SR to remove the total Ca²⁺ transient in less than 1 s. In other words, the SR is able to remove the entire Ca²⁺ transient at heart rates of up to 60 beats min⁻¹.

Caffeine induced a strong transient cell contracture and a transient inward ionic current. Under the experimental conditions of the study, the ionic current is due to extrusion of the Ca²⁺ liberated from the SR by the Na⁺/Ca²⁺ exchanger. Therefore, the time integral of this current should be proportional to the amount of Ca²⁺ liberated from the SR. The mean SR Ca²⁺ content measured is lower than values obtained in trout ventricular myocytes (Hove-Madsen et al., 1998) but larger than values obtained in mammals (Varro et al., 1993; Overend et al., 1997; Terraciano and MacLeod, 1997).

Apart from electrogenic Na⁺/Ca²⁺ exchange, the caffeine-induced current could also originate from other Ca²⁺-activated currents such as Cl⁻ or K⁺ currents. Given that K⁺ currents are inhibited by tetraethylammonium and Cl⁻ concentrations are similar in the extracellular and the pipette solutions, activation of a Cl⁻ current by Ca²⁺ is expected to give rise to an inward current at a membrane potential of —80 mV. The observation that rapid caffeine application causes a very strong cell contracture (Table 2; Fig. 3) does, however, suggest that the amount of Ca²⁺ released from the SR is much larger than that needed to activate a normal contraction. The relative contribution of a Ca²⁺-activated Cl⁻ current to the caffeine-induced inward current is therefore likely to be small compared with that of the Na⁺/Ca²⁺ exchange current. The measured time integral of the caffeine-activated ionic current should, nevertheless, be considered as an upper limit for the SR Ca²⁺ content.

In contrast, simultaneous removal of Ca²⁺ from the cytosol by mitochondria and a sarcolemmal Ca²⁺-ATPase would diminish the amount of Ca²⁺ extruded by the Na⁺/Ca²⁺ exchanger. These mechanisms are, however, not predominant in myocyte relaxation in a number of vertebrate species (Hove-Madsen et al., 1998; Bassani et al., 1994). The estimated SR Ca²⁺ content of 404±86 μmol l⁻¹ non-mitochondrial volume is about four times larger than the total Ca²⁺ needed to activate a normal contraction in vertebrate cardiomyocytes, indicating that the SR Ca²⁺ content is more than sufficient to participate in the activation of contraction.

Aside from the qualitative evidence gathered from ultrastructural studies that suggest the presence of an extensive SR (Dykens and Mangum, 1979; Schipp, 1987), recent physiological data point out the critical role of this subcellular compartment in excitation–contraction coupling. Firstly, the fact that ryanodine has a substantial inhibitory effect in multicellular preparations at physiological temperatures and heart rates (Gesser et al., 1997) suggests that Ca²⁺ release from the SR contributes significantly to the activation of contraction.

Secondly, as demonstrated in this study, the strong caffeine-induced cell contracture, the maximal SR Ca²⁺ uptake rate, the Ca²⁺-dependence of the SR Ca²⁺ uptake rate and the SR Ca²⁺ content in the cephalopod heart also point to a central role for the SR in the regulation of contraction in these species.

It should, however, be kept in mind that the present results are concerned with the Ca²⁺-accumulating properties of the SR. The fact that the amount of Ca²⁺ carried by the Ca²⁺ current amounts to only a fraction of the total Ca²⁺ necessary to activate contraction does suggest that other Ca²⁺ sources must contribute significantly to the activation of contraction. This situation is similar to that found in trout cardiomyocytes (Hove-Madsen et al., 1998; Hove-Madsen and Tort, 1998). The observation that the time courses of both relaxation and Na⁺/Ca²⁺ exchange current are slow in the presence of caffeine in the cephalopod heart suggests that the Na⁺/Ca²⁺ exchanger may not dominate the regulation of intracellular [Ca²⁺], indicating that the SR is required to activate and relax contraction.

In conclusion, the measured maximal uptake rate at 20 °C is sufficient to enable the SR to remove the total Ca²⁺ transient at a heart rate of 60 beats min⁻¹. Furthermore, the SR Ca²⁺ content is sufficiently high to activate contraction while the Ca²⁺ carried by the Ca²⁺ current appears to be insufficient fully to activate contraction. A quantification of the amount of Ca²⁺ released from the SR during a contraction will, however, have to await simultaneous measurements of ionic currents and intracellular [Ca²⁺].

We are greatly indebted to José Antonio Caparrós, captain of the fishing vessel General Aranda, and his crew for their help in supplying live octopuses. We also acknowledge the expertise of Marc Puigcerver in taking care of the animals while at the UAB aquarium facility. H.G. and J.A. express their thanks to Dr Lluis Tort for allowing us the use of his research facilities.