Monoamines and the isolated auricle of Sepia officinalis: are there β-like receptors in the heart of a cephalopod?

Within the Mollusca, the cardiovascular system of the dibranchiate cephalopods is uniquely complex. A single median ventricle, which receives oxygenated blood from the gills via two auricles, forms the main cardiac pump, supplying blood to the body through the cephalic, abdominal and gonadial aortas. After the blood has passed through a well-developed system of arteries and capillaries, it is moved by the peristaltic contractions of several pulsatile veins to the laterally sited branchial hearts. From the paired branchial hearts, the blood is moved through the gills to the auricles of the systemic heart.

Although numerous investigations have been carried out on the physiology of the cephalopod ‘heart’ (for reviews, see Wells, 1983; Kling and Jakobs, 1987; Messenger, 1996), little is known about the auricles of the systemic heart; in fact, only the ventricle has been examined and the physiological function of the auricles has yet to be investigated.

Some indirect indications that the auricles might play an active role in the regulation of cardiac performance were reported by Smith (1981a) for Eledone cirrhosa and by Foti et al. (1985) for Octopus vulgaris, who found that the preload pressure in systemic heart preparations, perfused via one or two auricles, has a direct effect on stroke volume and the beating rate of the ventricle. Smith (1981a) suggested that the ‘auricle might be playing a more active role in the filling of the ventricle than has been proposed (Johansen and Martin, 1962)’. Recent morphological findings from the auricle of Sepia officinalis L. support this assumption, suggesting that the special arrangement of circularly, transversely and longitudinally orientated muscle fibres in the auricular myocardium enables the auricle to push the haemolymph actively towards the ventricle (Versen and Schipp, 1997).

In isolated systemic hearts of Sepia officinalis perfused via both auricles (Jakobs, 1991a,b), the auricles and ventricle contract as in vivo, alternately and with the same frequency. Following the suggestion that the coordination of the ventricle and the branchial hearts depends not on the innervation but on a haemodynamic interaction (Eledone cirrhosa, Smith, 1979; Smith and Boyle, 1983; Octopus vulgaris, Wells, 1980; Wells and Smith, 1987), it was suggested that the coordination of auricular and ventricular systole also depends on a haemodynamic interaction (Versen et al., 1997).

This intrinsic mechanism in the systemic heart seems to be under extensive nervous control. The ventricle of the heart in Sepia officinalis is innervated by a single nerve trunk (nervus cardiacus), whereas the auricles are innervated (1) by nerve fibres issuing directly from the visceral nerves, (2) by fibres originating from the cardiac ganglia and (3) by unipolar ganglion cells, most of which are found in the auricles near the entrance of the efferent branchial vessels (Alexandrowicz, 1960). Transmission electron microscopy of the auricles (Versen and Schipp, 1997) and the ventricle (Schipp and Schäfer, 1969; Kling and Schipp, 1987) provided evidence that both compartments of the heart possess a dual innervation, but whereas numerous pharmacological investigations have indicated an antagonistic cholinergic–catecholaminergic innervation of the ventricle (for reviews, see Kling and Jakobs, 1987; Messenger, 1996), there have been no investigations on the extrinsic control of the auricles.

As a first step in analyzing the putative role of the monoaminergic transmitter/receptor system in the control of auricle function, we investigated the inotropic and chronotropic effects of different monoaminergic substances on the isolated auricle of Sepia officinalis under standardized conditions. These pharmacological examinations were supplemented by histofluorescence and HPLC investigations on the presence of monoamines in the auricles.

Seventy juvenile animals of both sexes of the cuttlefish Sepia officinalis L. (mantle length 90–130 mm; mass 90–190 g) from the Bassin d’Arcachon (Atlantic Ocean) were used for this study. The animals were acclimated for 1 week in tanks of circulating sea water before being used in experiments. Before surgery, animals were anaesthetized using 1–2 % ethanol in sea water or, for HPLC analysis, sea water with a double content of Mg²⁺ (Messenger et al., 1985; Kabotyanskii and Sakharov, 1989). Animals were considered anaesthetized when the tentacles ceased to move and no postural reflex occurred when the animal was turned on its back.

For the histochemical detection of biogenic monoamines, the SPG method of de la Torre and Surgeon (1976) as modified by Bolstad et al. (1979) was followed. Cryostat sections of frozen tissue (15–20 μm thick) were immersed in glyoxylic acid solution, dried under cold air and placed in an oven at 80–90 °C for 5 min. The sections were mounted in liquid paraffin and examined under a Leitz fluorescence microscope equipped with a fluorescence spectrophotometer (MPV compact). Photographs were taken on Fujichrome 100 film. A detailed description of the method has been given elsewere (Fiedler and Schipp, 1991).

The systemic heart was quickly removed from the animal and the auricles were cut off the ventricle, 1–2 mm distal of the atrioventricular junction, so that the two parts of the organ could be investigated separately. The tissues were briefly dried on blotting paper, weighed and placed in glass tubes containing 1 ml of ice-cold perchloric acid (0.1 mol l⁻¹ ). Homogenization was performed by repeated freeze–thaw procedures, immersing the tubes alternately in liquid nitrogen and warm water (Mercer et al., 1983). After vortexing, the homogenate was centrifuged at 4 °C for 10 min at 12 000 g. The supernatant was transferred to cryo-tubes and stored frozen at −24 °C until it was analysed (within 6 weeks). To ascertain the external recovery rate, reference solutions of catecholamines were treated as described above.

For extraction and analysis of catecholamines, we used an analysis kit (ClinRep) purchased from Recipe Chemicals and Instruments. Between 10 and 500 μl of each sample was transferred into reaction tubes containing 1 ml of Tris buffer (1 mol l⁻¹, pH 8.6) and 50 μl of an internal standard (dihydroxybenzylamine, DHBA). The catecholamines were extracted by adsorption onto Al₂O₃ and eluted with acetic acid after frequent washing with Tris buffer (Anton and Sayre, 1962). Reverse-phase HPLC was performed on a Waters catecholamine analysis system which included an automated sample injector (WISP710B), a programmable solvent delivery system (M600), a data module (M730) and an electrochemical detector (M460). For separation, we used a Resolve C₁₈ 5 μm column (3.9 mm×150 mm) and a mobile phase from Recipe Chemicals (flow rate 1 ml min⁻¹ ; pressure 6.2×10⁶ to 1.4×10⁷ Pa). A potential of 0.6 V was chosen for the detection of catecholamines. Under the conditions employed, the retention times were as follows: noradrenaline 5.09 min, adrenaline 6.23 min, DHBA 8.93 min and dopamine 14.63 min. The internal recovery (70–90 % according to the producer) was calculated and corrected using computer-aided calibration. The recovery of the external standards tested amounted to 85–90 %.

The auricle (left or right) was isolated by a cut proximal to the efferent branchial vessel (at the auricular valves) and 1–2 mm distal to the atrio-ventricular junction. After removing connective tissue, the tubular organ was mounted on stainless-steel clamps and isometrically suspended in a 50 ml water-jacketed organ bath (18 °C) with one clamp anchored and the other fixed to a strain gauge (Statham UC 2). The pressure transducer was connected to a direct-current bridge amplifier (HSE type 300), and its signals were registered on a thermographic recorder (Watanabe mark V). The bathing medium consisted of iso-osmotic filtered and aerated sea water containing 0.17 % (w/v) glucose at pH 8.2.

Depending on the size of the auricle, the initial resting tension was adjusted to 8±1 mN. All the auricles were equilibrated at 18 °C until they achieved regular and constant activity (15–20 min). During this time, the initial resting tension was reduced by dilatation of the organs to a constant value of 5±1 mN.

Drugs were added to the bath solution cumulatively. Immediately after application, the effects of each drug concentration were recorded for at least 10 min or until a stable organ response had occurred. The mechanical traces of stable organ responses were evaluated and concentration/response curves prepared. The contractions of the untreated organs (sea water + glucose) were recorded as a reference for the calculation of the concentration/response curves (based on fitting the data to Hill’s four-parameter equation; Endrenyi, 1981). The drug responses are expressed as the percentage deviation from the reference values (means ± S.E.M.; N is the number of preparations). For each concentration/response curve, the EC₅₀ (half-maximum excitatory concentration) and pD₂ (−logEC₅₀) values were calculated as described by van Rossum (1963). Significant differences between values were estimated, where applicable, by Student’s t-test for paired comparisons. The statistical evalution and the drawing of the concentration/response curves were carried out using Prism 2.0 (GraphPad software). Values are presented in the figures as percentage deviation from the control values in sea water + glucose.

The following drugs were used: agonists, adrenaline (Suprarenin, Hoechst), albuterol (=salbutamol, RBI), dobutamine (RBI), clonidine (Catapresan, Boehringer Ingelheim), dopamine (Serva), isoprenaline (Aludrin, Boehringer Ingelheim), methoxamine (Wellcome), noradrenaline (Arterenol, Hoechst), phenylephrine (RBI), serotonin (5-HT-creatinine sulphate, Sigma); antagonists, cyproheptadine (Serva), mianserin (RBI), phentolamine (Regitin, Ciba-Geigy), pindolol (Visken, Wander Pharma), prazosin (Pfizer), propranolol (Sigma), urapidil (Ebrantil, Byk Gulden), yohimbine (Sigma); and adenylate cyclase inhibitors, MDL-12,330A HCL (RBI), SQ-22,536 (RBI).

Values are presented as means ± S.E.M.

Serial cryostat sections of the auricle, treated with the glyoxylic-acid-induced fluorescence method, showed low background fluorescence with luminiscent cells (haemocytes) of a pale green colour. This autofluorescence was clearly distinguishable from specific reaction products because it was also present in control sections.

Fluorescent fibres of different sizes and brightness were detected in all parts of the auricular myocardium. Most of these fibres were bluish-green in colour (emission maximum 420 nm), which is typical of catecholamines, but a yellow fast-fading fluorescence (emission maximum 520 nm) typical of 5-HT was also seen. In the peripheral connective tissue, strongly fluorescent nerve trunks were observed, often showing ramifying branches that penetrated into the auricular wall (Fig. 1). Most of the fibres appeared to be profiles of single axons with varicosities, the latter being visible as intensely fluorescent spots. Specific fluorescent structures associated with cell somata of ganglion cells (Alexandrowicz, 1960) were not observed.

The auricles from three of four animals investigated contained appreciable amounts of dopamine (23.9±7.2 ng g⁻¹ ) and noradrenaline (1.3±0.8 ng g⁻¹ ). In one sample, noradrenaline was lacking. In the ventricle extracts examined (N=4), the concentration of noradrenaline (5.1±3.4 ng g⁻¹ ) tended to be higher (but P>0.05) than that of the auricles. Dopamine was not found in the ventricle, nor was adrenaline detectable in any of the tissue samples. Typical chromatograms are shown in Fig. 2A,B. In addition to the elution peaks with retention times (t_R) corresponding to those of noradrenaline, dopamine and the internal standard dihydroxybenzylamine, both chromatograms show two elution profiles (t_R=3.6 and 7.7 min) that did not occur in the standard chromatograms and are, therefore, unidentified.

The auricles contracted spontaneously after dissection and transfer to sea water with glucose. All the auricles contracted without any pressure or tension being exerted.

The contraction rate of the auricles was not influenced by isometric stretching of the organ. A gradual elevation of the resting tension (of 0–15 mN) evoked a transient acceleration of the frequency, but the previous rhythm of the organs reappeared after several minutes (3–4 beats min⁻¹ ). The extensibility of the auricle meant that preloads higher than 7 mN could only be kept constant when the resting tension was repeatedly adjusted to the chosen value. This led to a considerable enlargement of the auricular diameter; e.g. during an equilibration period of 30–40 min, the diameter of the auricles (length 7 mm) enlarged from 1.5–2 mm (at 0 mN) to 4.5–5.5 mm until the preload remained constant at 10 mN. These preparations often showed irregular contractions, varying in beating rate and contraction force, and only weak positive inotropic effects occurred after catecholamines were applied. In view of the range of diastolic diameter (2.5–3.5 mm) of the middle part of the auricles in vivo (Sepia officinalis mantle length 90–130 mm, mass 90–190 g) or in perfused systemic hearts (preload 490 Pa, afterload 196 Pa; Jakobs, 1991a,b), we assumed that this extension of the auricle was excessive. Therefore, the preload of the isometrically fixed preparations was adjusted to a minimum value (5±1 mN) corresponding to an extension of 3–4 mm. After an equilibration period of 15–20 min, the contraction frequency of the auricles was 3.5±1.0 beats min⁻¹ (N=100) and the contraction force was 8±2 mN (N=100).

Serotonin (5-HT) caused dose-dependent positive inotropic and chronotropic effects on auricle preparations (N=6, Fig. 3A). The threshold concentration for both effects ranged between 5×10⁻⁸ and 10⁻⁷ mol l⁻¹. Maximum reactions were obtained with 10⁻⁵ mol l⁻¹ (EC₁₀₀ amplitude) and 5×10⁻⁶ mol l⁻¹ (EC₁₀₀ frequency). Serotonin enhanced the tension developed to a maximum of 102.1±10.4 %. The frequency was accelerated by approximately 138.9±33.1 %. The concentration/response curves, from which EC₅₀ values of 1.1×10⁻⁶ mol l⁻¹ (pD₂ amplitude=5.96) and 5.5×10⁻⁷ mol l⁻¹ (pD₂ frequency=6.26) were deduced (Fig. 4A,B; Table 1), take nearly the same course.

The effects of 5-HT were not influenced by the 5-HT₂ blocking agent cyproheptadine (10⁻⁵ mol l⁻¹ ), but in four tests the antagonist showed a slight inhibitory effect when applied alone at concentrations greater than 10⁻⁵ mol l⁻¹. Comparable inhibitory effects occurred when the 5-HT receptor antagonist (5-HT₁ and 5-HT₂) mianserin (10⁻⁶ mol l⁻¹ ) was applied to the untreated organs (N=6). Added at a 10-fold lower concentration than cyproheptadine, mianserin virtually inhibited the excitatory effects of 5-HT up to a serotonin concentration of 5×10⁻⁷ mol l⁻¹ (Fig. 4A,B). The positive chronotropic and inotropic effects induced by serotonin concentrations between 5×10⁻⁷ mol l⁻¹ and 5×10⁻⁶ mol l⁻¹ were significantly (P<0.05) lower in the presence of the 5-HT receptor antagonist. The maximum effect of 5-HT was obtained when a concentration of 5×10⁻⁵ mol l⁻¹ serotonin was applied (Table 1). The chronotropic effects of 5-HT were inhibited to a lesser degree (EC₅₀ 2.7×10⁻⁶ mol l⁻¹ ; pD₂=5.57) than the inotropic effects (EC₅₀ 5.4×10⁻⁵ mol l⁻¹ ; pD₂=4.27) but, in comparison with the concentration/response curves for the chronotropic and inotropic effects of serotonin, both concentration/response curves are right-shifted, indicating that serotonin is antagonized competitively by mianserin.

Noradrenaline primarily induced concentration-dependent positive chronotropic effects on auricle preparations (N=12), whereas the contraction amplitude was enhanced to a lesser degree (Fig. 3B). The threshold concentration was approximately 10⁻⁸ mol l⁻¹ and the maximum chronotropic response (133.7±13.2 %) was produced by a concentration of 10⁻⁵ mol l⁻¹ (EC₁₀₀ frequency). With an EC₁₀₀ value of 2×10⁻⁵ mol l⁻¹, the amplitude was enhanced by 43.1±5.0 %. EC₅₀ values, evaluated arithmetically from the concentration/response curves (Fig. 5A,B; Table 1), were 2.3×10⁻⁷ mol l⁻¹ (pD₂ frequency=6.65) and 8.1×10⁻⁷ mol l⁻¹ (pD₂ amplitude=6.09).

Comparable with the results obtained with noradrenaline, adrenaline and dopamine evoked mainly concentration-dependent increases in beating frequency of the investigated organs (adrenaline, N=6; dopamine, N=5), but both catecholamines only partially mimicked the noradrenaline response. In the concentration range from 10⁻⁹ to 5×10⁻⁸ mol l⁻¹, adrenaline induced a slight negative chronotropic effect (−3.5±1 %), but this was not statistically significant (P>0.05). The concentration/response curves for the chronotropic effects induced by adrenaline and dopamine (Fig. 5A) are shallower than those for noradrenaline and fail to reach the same maxima (adrenaline, 91.6±27 %, EC₁₀₀=10⁻⁵ mol l⁻¹ ; dopamine, 82.6±5 %, EC₁₀₀=2×10⁻⁵ mol l⁻¹ ). With EC₅₀ values of 2.1× 10⁻⁶ mol l⁻¹ (adrenaline, pD₂=5.69) and 4.6×10⁻⁶ mol l⁻¹ (dopamine, pD2=5.34), the concentration/response curves for the chronotropic effects are clearly shifted to the right, indicating that adrenaline and dopamine have a lower affinity for the receptor than does noradrenaline. In contrast, adrenaline mimicked (Fig. 5B) the positive inotropic effects (EC₅₀=1.2×10⁻⁶ mol l⁻¹, pD₂=5.91) of noradrenaline and exhibited the same maximum effect (40.8±4.0 %, EC₁₀₀= 2×10⁻⁵ mol l⁻¹ ; Table 1). The positive inotropic effects of dopamine, which possesses a lower affinity for the receptor (EC₅₀=4.7×10⁻⁶, pD₂=5.33), were less prominent (24.4±5.4 %, EC₁₀₀=2×10⁻⁵ mol l⁻¹ ; Fig. 5B; Table 1). Dopamine concentrations higher than 2×10⁻⁵ mol l⁻¹ caused no further significant (P>0.05) increase in frequency or contraction force.

Clonidine, which is known to stimulate α₂-receptors in mammals, induced positive chronotropic and inotropic effects on all auricles tested (N=7). Clonidine caused a maximum increase in frequency (56.2±5.6 %) and contraction force (47.4±3.8 %) at a concentration of 2×10⁻⁵ mol l⁻¹ (EC₁₀₀; Table 1). The EC₅₀ values evaluated arithmetically from the concentration/response curves are 2.4×10⁻⁶ mol l⁻¹ (pD₂=5.63) for the chronotropic and 4.6×10⁻⁶ mol l⁻¹ (pD₂=5.34) for the inotropic effects (Fig. 6A,B).

The α₁-agonist methoxamine, added to four preparations at concentrations up to 10⁻⁴ mol l⁻¹, had no significant effect, but the α₁-agonist phenylephrine induced slight but significant (P<0.05) positive chronotropic and inotropic effects (N=6). The EC₅₀ values, calculated with 7.2×10⁻⁶ mol l⁻¹ (pD₂=5.14) for the chronotropic and 2.4×10⁻⁵ mol l⁻¹ (pD₂=4.62) for the inotropic effect (Fig. 6A,B), indicate that the α₁-agonist has a lower affinity for the receptor than does clonidine. The maximum increase in frequency (37.4±12.1 %) and amplitude (45.4±10.3 %) occurred at a phenylephrine concentration of 5×10⁻⁵ mol l⁻¹ (EC₁₀₀; Table 1).

The β₁-agonist dobutamine induced no chronotropic effects (Fig. 6A) on the four auricle preparations tested. At a concentration of 5×10⁻⁵ mol l⁻¹ (EC₁₀₀), the only effect of dobutamine was to increase the force of contraction (34.8±7.7 %; P<0.05; Table 1). The EC₅₀ value calculated from the concentration/response curve (Fig. 6B) was 5.3×10⁻⁵ mol l⁻¹ (pD₂=4.27).

The β-adrenergic agonist (β₂>β₁, Merck Index 12217) albuterol had no effect on contraction frequency (Fig. 6A; N=6), but caused significant (P<0.05) enhancement of the contraction force. Albuterol mimics the inotropic effects of noradrenaline (Fig. 6B), but must be added at a fivefold higher concentration (EC₁₀₀=5×10⁻⁵ mol l⁻¹ ) than noradrenaline to induce a comparable increase in the amplitude (46.1±18.1 %). From the calculated EC₅₀ value (1.9×10⁻⁶ mol l⁻¹, pD₂=5.73), the β-agonist possesses a lower affinity for the receptor than does noradrenaline, but has the highest affinity among the mimetics tested (Table 1).

The negligible difference between the EC₅₀ values for the inotropic effects of albuterol and the β-agonist isoprenaline (2.01×10⁻⁶ mol l⁻¹, pD₂=5.69) suggests an almost identical affinity of these two β-agonists to this receptor. However, in contrast to the other β-agonists used in this study, isoprenaline induced not only the strongest positive inotropic effects (122.3±9.1 %; EC₁₀₀=5×10⁻⁵ mol l⁻¹ ; Fig. 6B) on the twelve auricles tested, but also mimicked the positive chronotropic effects of noradrenaline quite strongly (Figs 3C, 6A). At concentrations higher than 5×10⁻⁸ mol l⁻¹ (threshold concentration), isoprenaline evoked a continuous acceleration of the contraction frequency until it exerted its maximum effect (103.2±10.2 %; EC₁₀₀=10⁻⁵ mol l⁻¹ ). From the EC₅₀ value for the chronotropic effects, calculated at 8.8×10⁻⁷ mol l⁻¹ (pD₂=6.06), only noradrenaline has a higher affinity for this receptor (Table 1).

Comparing the affinities of the mimetics and natural transmitters tested (potency range for inotropic effects: noradrenaline ⩾ adrenaline ⩾ albuterol = isoprenaline > clonidine = dopamine > phenylephrine > dobutamine; potency range for chronotropic effects: noradrenaline > isoprenaline > adrenaline ⩾ clonidine > dopamine > phenylephrine) suggests that noradrenaline is the natural adrenergic transmitter in the auricles of Sepia officinalis.

The β-adrenergic antagonist pindolol, tested on three preparations, induced some quite strong excitatory activity, increasing the contraction rate by approximately 83 % at a concentration of 10⁻⁵ mol l⁻¹. Pindolol has been reported to have similar agonistic activity in a number of mammalian organs (Prichard, 1978). The β-antagonist propranolol, applied at a concentration of 10⁻⁵ mol l⁻¹, did not alter the normal rhythm of isolated contracting auricles (N=8). Added in combination with isoprenaline (10⁻⁸ to 2×10⁻⁵ mol l⁻¹ ; N=4) or noradrenaline (10⁻⁸ to 2×10⁻⁵ mol l⁻¹ ; N=4), propranolol did not reduce the positive chronotropic effects of either agonist significantly (P>0.05). Propranolol also failed to block the positive inotropic effects of noradrenaline and isoprenaline, but in two preparations it enhanced (15±5 %) the effects of isoprenaline on the contraction force.

When the α-antagonist phentolamine (α₁+α₂) was applied at a concentration of 10⁻⁵ mol l⁻¹, the auricles (N=3) started to contract irregularly with a varying beat rate. Phentolamine (2×10⁻⁵ mol l⁻¹ ) was therefore only used on preparations prestimulated with 5×10⁻⁶ mol l⁻¹ noradrenaline (N=6) or 5×10⁻⁶ mol l⁻¹ isoprenaline (N=5). Phentolamine (Fig. 7) did not significantly (P>0.05) influence the positive inotropic effects evoked by noradrenaline (34.7±5.6 %; after phentolamine 32.2±6.6 %) or isoprenaline (112.1±10.13 %; after phentolamine 110.3±13.5 %), but almost completely inhibited the positive chronotropic effects (noradrenaline 100.9±10.3 %, after phentolamine 8.1±10.7 %; isoprenaline 94.9±2.6 %, after phentolamine 12.6±17.3 %).

The α₁-antagonist urapidil did not alter the contractions of the auricles tested (N=10) when it was added at a concentration of 10⁻⁵ mol l⁻¹. Applied in combination with noradrenaline (10⁻⁹ to 5×10⁻⁵ mol l⁻¹, N=5) or isoprenaline (10⁻⁹ to 5×10⁻⁵ mol l⁻¹, N=5), urapidil did not influence the positive inotropic effects of either agonist (Fig. 8A,B; Table 1), but inhibited significantly (P<0.05) the positive chronotropic effects of noradrenaline and isoprenaline in the concentration range between 10⁻⁸ and 5×10⁻⁶ mol l⁻¹. The concentration/response curves for the chronotropic effects of noradrenaline (EC₅₀=1.4×10⁻⁶ mol l⁻¹, pD₂=5.76) and isoprenaline (EC₅₀=2.5×10⁻⁶ mol l⁻¹, pD₂=5.6) in the presence of urapidil are clearly shifted to the right (Fig. 8A,B), indicating a competitive inhibitory effect of the α₁-antagonist. The α₁-antagonist prazosin (N=3) and the α₂-antagonist yohimbine (N=3) showed no influence on the known effects of noradrenaline or isoprenaline, even when applied at a concentration of 10⁻⁴ mol l⁻¹.

To characterize further the assumed β-like receptor, the influence of the adenylate cyclase inhibitors MDL-12,330A (MDL) and SQ-22,536 (SQ) on the effects of isoprenaline on the isolated auricles were examined.

When applied at concentrations higher than 5×10⁻⁶ mol l⁻¹, MDL simultaneously caused a continuous reduction of the amplitude and an acceleration of the contraction frequency (N=3). To avoid this effect, MDL was used only at a concentration of 2×10⁻⁶ mol l⁻¹. The application of MDL caused minor fluctuations of the contraction frequency and contraction force of the auricles tested (N=6), but the previous rhythm and amplitude reappeared after 5–10 min. Applied in combination with isoprenaline (10⁻⁸ to 5×10⁻⁵ mol l⁻¹ ), MDL caused a slight, but insignificant (P>0.05), reduction in the positive chronotropic effects evoked by isoprenaline (95.93±9.0 %; EC₁₀₀ isoprenaline + MDL=10⁻⁵ mol l⁻¹ ; Fig. 9A), but reduced significantly (P<0.05) the positive inotropic effects (71.45±20.9 %; EC₁₀₀ isoprenaline + MDL=5×10⁻⁵ mol l⁻¹ ; Fig. 9B). The EC₅₀ values for the chronotropic (6.8×10⁻⁷ mol l⁻¹, pD₂=6.17) and inotropic (1.6×10⁻⁶ mol l⁻¹, pD₂=5.8) effects induced by isoprenaline in the presence of MDL indicate that the adenylate cyclase inhibitor did not influence the affinity of isoprenaline to the receptor (Table 1).

The normal rhythm of the auricles was not affected by application of SQ at a concentration of 10⁻⁵ mol l⁻¹ (N=4). Added in combination with isoprenaline (10⁻⁸ to 5×10⁻⁵ mol l⁻¹ ), SQ reduced slightly, but not significantly (P>0.05), the positive chronotropic effects of isoprenaline (92.8±4.7 %, EC₁₀₀ isoprenaline + SQ=10⁻⁵ mol l⁻¹ ; Fig. 9A). The positive inotropic effects of isoprenaline on the auricles were significantly (P<0.05) lower in the presence of the adenylate cyclase inhibitor (54.1±15.5 %, EC₁₀₀ isoprenaline + SQ=5×10⁻⁵ mol l⁻¹ ; Fig. 9B). The EC₅₀ values for the inotropic (2.3×10⁻⁶ mol l⁻¹, pD₂=5.64) and chronotropic (5.5×10⁻⁷ mol l⁻¹, pD₂=6.25) effects indicate that SQ had no effect on the affinity of isoprenaline for the receptor (Table 1).

In accordance with the theory that pacemakers within the cephalopod circulatory system are triggered by pressure (Wells, 1980), examinations of auricular–ventricular interactions (Versen et al., 1997) indicate that the activity of the ventricular pacemaker, which is thought to be located in the region of the atrio-ventricular junction (Wells, 1979, 1983; Smith, 1981b; Wells and Smith, 1987; Versen et al., 1997), is dependent on hydro-mechanical stimulation. On the basis of these observations, it was suggested that the auricles might act as the pacemakers of the systemic heart (Versen et al., 1997; Versen and Schipp, 1997).

This assumption seems to be contradicted by our findings that isolated auricles contract with a very low frequency. However, the activity of isolated contractile cephalopod organs is often impaired by the experimental conditions (Wells and Smith, 1987). Thus, it might be assumed that sea water could affect auricular activity in vitro, but in organ cultures of embryonic (stage XX; Naef, 1928) and post-embryonic (Versen, 1998; Versen and Schipp, 1998) systemic hearts, the auricles contracted at the same frequency as in sea water, even though the osmolality and ion composition of the culture medium (pH 7.2–7.4) were adjusted to the known values for the haemolymph of S. officinalis (Abbott et al., 1985). Considering that Agnisola and Houlihan (1991) have shown in Octopus vulgaris that the performance of the isolated octopod heart depends directly on the oxygen content of the perfusion fluid, it might be expected that auricular performance in culture media as well as in aerated sea water would be reduced because of the low oxygen content. Although the effects of a low oxygen content on the activity of isolated auricles could not be ruled out, auricles of systemic heart preparations (Sepia officinalis) (Jakobs, 1991a,b; Versen et al., 1997) perfused only with aerated sea water contract at an obviously higher frequency (17±7 beats min⁻¹ ) than isometrically stretched auricles (3±1 beats min⁻¹ ). Given the complex organization of the circular, transverse and longitudinal muscle fibres within the auricular myocardium (Versen and Schipp, 1997), it appears more probable that the lack of luminal hydrostatic pressure, which in vivo and during perfusion stretches all muscle fibres simultaneously, is the main reason for the weak performance of isometrically stretched auricle preparations. Previous examinations of isolated perfused auricles from Sepia officinalis (20±8 beats min⁻¹ ; B. Versen, unpublished observations) and of auricles perfused in situ (35±10 beats min⁻¹ ; R. Schipp, unpublished observations), point in the same direction. Although these auricle preparations could not be standardized and were, therefore, not applicable to our pharmacological investigations, they gave the first hints that auricular myogenicity could depend on the internal pressure. The findings presented in this study suggest an excitatory innervation of the auricles, and it might be expected that in vivo these intrinsic mechanisms would be subject to extrinsic neuronal control.

Using the glyoxylic acid method of Bolstad et al. (1979), we observed bluish-green fluorescent fibres in all parts of the auricular myocardium, which suggested a typical catecholamine emission peak maximum at 480 nm (Lindvall and Björklund, 1974). As with similar examinations of the ventricle (Kling, 1984, 1986), the branchial hearts (Fiedler and Schipp, 1991) and vessels (Andrews and Tansey, 1983; Schipp et al., 1997) of Sepia officinalis and Octopus vulgaris, these findings indicate that the auricular myocardium has a catecholaminergic innervation. However, although nerves containing catecholamines have been found in every part of the cephalopod circulatory system that has been examined, neither immune-histochemical nor fluorescence-histochemical investigations have yet given any indication of a serotonergic innervation of the circulatory organs (Messenger, 1996).

However, in contrast to the branchial hearts of Sepia officinalis, where 5-HT was not only absent but without any effect in vitro (Fiedler and Schipp, 1990, 1991), 5-HT relaxed the dopamine-precontracted aorta (Schipp et al., 1991) and excited the isolated systemic heart of Sepia officinalis (Florey and Florey, 1954; Kling, 1985; Kling and Schipp, 1987) as well as of Octopus vulgaris (Bacq et al., 1952; Wells and Mangold, 1980; Foti et al., 1985). Although these findings indicated that the systemic heart has a serotonergic innervation, the fact that 5-HT could not be detected in the cephalopod heart suggested that excitatory effects on the ventricle, induced by externally applied 5-HT, are mediated by unknown peptidergic receptors (Kling and Schipp, 1987).

Our fluorescence-histochemical results suggest another possibility. Yellow-coloured fast-fading fluorescent structures were detected in all parts of the auricular wall. The emission peak maximum (E_max=520 nm) corresponds to that of serotonin (E_max=515 nm), providing evidence that 5-HT-containing nerves are present in the auricle of Sepia officinalis. Given that the auricular 5-HT receptor possesses a 100-fold lower affinity for 5-HT (pD₂ frequency=6.26; pD₂ amplitude=5.96) than the ventricular 5-HT receptor (pD₂ amplitude=8.2; Kling and Schipp, 1987), it seems probable that serotonin is released in vivo into the auricular lumen and reaches the ventricle via the bloodstream. Although we have not yet been able to prove this hypothesis, because our HPLC equipment was not able to detect 5-HT, this assumption corresponds to pharmacological results.

In contrast to ‘Straub-cannulated’ ventricles of Sepia officinalis, in which 5-HT largely induced strong positive inotropic reactions (Kling, 1985; Kling and Schipp, 1987), serotonin not only increased the contraction force (+102.1±10.4 %) of isometrically suspended auricles but also caused a marked acceleration of the contraction frequency (+138.9±33.1 %). However, in the ventricle of isolated systemic hearts of Sepia officinalis, perfused via both auricles, 5-HT caused significant (P<0.05) positive inotropic and chronotropic effects (U. J. Tschesche, unpublished results). 5-HT-induced excitatory effects on contraction force and contraction frequency have also been reported from in vitro (perfusion via the auricles; Foti et al., 1985) and in vivo studies of the heart of Octopus vulgaris (Johansen and Huston, 1962; Wells and Mangold, 1980). These observations could support the above-mentioned hypothesis that serotonergic stimulation of the ventricle might occur via the bloodstream, since 5-HT always induced positive inotropic effects on the ventricle regardless whether it was applied directly into the ventricular lumen or reached the ventricle with the perfusion fluid via the auricles. However, it is remarkable that 5-HT caused only positive chronotropic effects on the ventricle when it was applied over the auricles. Following the theory that the ventricular pacemaker is stimulated by auricular contractions (Versen et al., 1997), it might be supposed that the observed positive chronotropic effects of 5-HT on the ventricle (see Kling and Jakobs, 1987) represent the reflex-like response of the ventricle to contractions of the serotonin-stimulated auricles.

The different effects induced by 5-HT on the ventricle and the auricle suggest that different 5-HT receptor subtypes are present in the two compartments of the systemic heart. This possibility is supported by the fact that the 5-HT₂ antagonist cyproheptadine blocked the effects of 5-HT on the ventricle (Kling, 1985; Kling and Schipp, 1987), but not those on the auricles. Since the effects of serotonin on the auricles could be competitively antagonized by mianserin (which affects both 5-HT₁ and 5-HT₂ receptors), it is assumed that the auricular serotonin receptor more closely resembles the 5-HT₁ type than the 5-HT₂ type (Göthert, 1990). The different inhibitory potencies of mianserin on the chronotropic (pD₂=5.57) and inotropic (pD₂=4.27) effects of serotonin indicate that there may be two different 5-HT receptor subtypes in the auricle. Although this must be confirmed by further studies, the fact that six 5-HT receptor subtypes were detected in one neurone of Aplysia californica and Helix aspersa (Gerschenfeld and Paupardin-Tritsch, 1974) suggests that it is possible that different 5-HT receptor subtypes may also exist in other molluscan organs.

The HPLC analysis of tissue samples from the auricles and the ventricle revealed the presence of noradrenaline (1.3±0.8 ng g⁻¹ ) and dopamine (23.9±7.2 ng g⁻¹ ) in the auricles, while only noradrenaline (5.1±3.4 ng g⁻¹ ) was found in the ventricle. No adrenaline was detected in either the auricles or the ventricle. Comparable results were reported by Fiedler et al. (1992; S. officinalis), who found only dopamine and/or noradrenaline in the heart organs and blood vessels they examined, while in the haemolymph only adrenaline was detected. The authors suggested that noradrenaline plays an important role as a neurotransmitter in the central circulatory system, whereas adrenaline may have a humoral function in cephalopods and is probably released into the blood by the neurosecretory system of the vena cephalica. Since dopamine and noradrenaline are widely distributed in the sense organs and viscera (for a review, see Messenger, 1996) and since both catecholamines are present in the posterior suboesophageal lobes, i.e. the origin of the innervation of the systemic heart (Kime and Messenger, 1990), there is ample evidence that noradrenaline and/or dopamine, but not adrenaline, function as neurotransmitters in the heart of Sepia officinalis in vivo.

Pharmacological data (Kling and Schipp, 1987) and our HPLC results suggest that the ventricle has a noradrenergic innervation, but the question of whether noradrenaline or dopamine acts as a neurotransmitter in the auricles is more difficult to answer. In conjunction with the findings of Fiedler et al. (1992), who detected only dopamine (1433.0±480.5 ng per organ) but no noradrenaline in the cardiac ganglion of Sepia officinalis, the high dopamine concentrations in the auricles suggest that the auricles may be innervated by dopaminergic neurones. However, it must be remembered that dopamine may be present in the auricles not as a transmitter itself but as a precursor of noradrenaline.

Although the site of dopamine-β-hydroxylation has not been identified, the findings of our pharmacological investigations on isolated auricles of Sepia officinalis support this hypothesis. The catecholamines noradrenaline and dopamine, which may act as neurotransmitters on the auricles in vivo, produced mainly concentration-dependent positive chronotropic effects and induced less prominent positive inotropic effects. The pD₂ values for the inotropic and chronotropic effects of these catecholamines indicate that noradrenaline has a 10-fold higher affinity for the catecholaminergic receptor than does dopamine. Given that noradrenaline also induced the strongest effects on contraction frequency and force, it appears more probable that noradrenaline, and not dopamine, is the natural catecholaminergic transmitter of the auricles. These findings are consistent with results from previous studies on the ventricle (Kling, 1985; Kling and Schipp, 1987) and the branchial hearts (Fiedler and Schipp, 1990) of Sepia officinalis. However, in contrast to the auricles, both the ventricle and the branchial hearts show largely positive inotropic reactions to noradrenaline in vitro. The different responses of the auricle and ventricle to noradrenaline indicate that different catecholamine receptors may occur in the different compartments of the systemic heart.

The catecholamine receptor of the ventricle and the branchial heart was characterized as an α-like receptor, since both organs were insensitive to isoprenaline and noradrenaline-induced effects could only be blocked by phentolamine and not by the β-blocker propranolol (Kling and Schipp, 1987; Fiedler and Schipp, 1990). Similar results were obtained from investigations on the aorta of Sepia officinalis (Schipp et al., 1991) as well as on the statocysts and the chromatophores of Octopus vulgaris (Andrews et al., 1981, 1983; Messenger, 1996; Williamson, 1989; Williamson and Chrachri, 1994). On the basis of these findings, it was stated that, in cephalopods ‘the noradrenaline receptors are like the α-adrenergic subtype of vertebrates’ (Messenger, 1996).

Supporting this suggestion, the α₂-agonist clonidine and the α₁-agonist phenylephrine partially mimicked the positive chronotropic effects of noradrenaline on the auricles and almost completely mimicked the positive inotropic effects. However, since clonidine induced no effect on the ventricle or the branchial hearts (Kling and Schipp, 1987; Fiedler and Schipp, 1990), it seems likely that the catecholamine receptor in the auricle is not identical to the α-adrenergic receptor in these organs. Further evidence for the existence of another α-receptor subtype in the auricles is provided by the finding that isoprenaline closely mimicked the positive chronotropic effects of noradrenaline on the auricles and exhibited a high affinity for the receptor. Although it might be suggested that isoprenaline interacts with an adrenoreceptor of the β-subtype, two results disagree with this. The actogram recordings of the noradrenaline- and isoprenaline-induced effects indicate that the auricular response to noradrenaline application occurred 1–2 min faster than that to isoprenaline. Furthermore, the positive chronotropic effects of noradrenaline, as well as isoprenaline, were significantly blocked by the α-antagonists phentolamine (α₁ and α₂) and urapidil (α₁), whereas the β-antagonist propranolol showed no clear influence on the induced effects. This suggests that in the auricle of Sepia officinalis the β-agonist isoprenaline interacts unspecifically with an α-like adrenoreceptor subtype.

However, this α-like receptor seems to mediate only the positive chronotropic effects of the applied agonists, because both α-antagonists inhibited only the positive chronotropic effects of noradrenaline and isoprenaline, whereas the positive inotropic effects were not influenced. Given that the β-agonists albuterol and dobutamine also induced only positive inotropic effects, it appears likely that there is a further, β-like, adrenoreceptor in the auricle which largely mediates positive inotropic effects. The inefficiency of propranolol in blocking the inotropic effects evoked by noradrenaline or isoprenaline does not argue against this suggestion. Since examinations of Hydra japonica and the hydromedusa Polyorchis penicillatus have shown that dopamine-induced effects could be blocked by propranolol but not by dopamine antagonists (Hanai and Kitajima, 1984; Chung and Spencer, 1991), there is evidence that propranolol does not act as a specific β-blocking agent in invertebrates. Nevertheless, since albuterol (β₂>β₁) completely mimicked the positive inotropic effect of noradrenaline and had the highest affinity of any of the inotropic-effect-inducing adrenomimetics for the receptor, it is assumed that the β-like receptor in the auricle of Sepia officinalis resembles the β₂-receptor subtype of vertebrates.

The fact that the adenylate cyclase inhibitors MDL-12,330A and SQ-22,536 significantly blocked the positive inotropic effects of isoprenaline on the auricles supports this suggestion. In conjunction with preliminary examinations on isotonically suspended auricles, which have shown that the adenylate cyclase activator forskolin enhanced the auricular contraction force by up to 42±6.9 % (N=8; B. Versen, unpublished results) but did not influence the contraction frequency, our findings indicate that the signal transduction of the β-like receptor in the auricle of Sepia officinalis occurs by activation of the cyclic AMP pathway, in a manner analogous to that of the β-receptors of vertebrates. Sequences homologous to hamster and human β-receptors have been detected in many invertebrates, with the human β₂-adrenoreceptor being the most strongly conserved (Palacious et al., 1989). From these findings, and from the detection of a β-like adrenergic receptor in the cnidarian Renilla koellikeri, Awad and Anctil (1993) suggested that an ancestral form of the β-adrenoreceptor may have been conserved throughout evolution from the early metazoans to mammals. This theory suggests that a β-like adrenergic receptor will also exist in cephalopods.

In conclusion, our investigations substantiate a ‘double-aminergic’ excitatory innervation of the auricles of Sepia officinalis by serotonergic and noradrenergic neurones. A further excitatory stimulation might be caused by adrenaline, which is thought to exert a humoral function in cephalopods (Fiedler et al., 1992). Our results suggest that the auricles have a distinct regulatory function in the circulatory system of Sepia officinalis, probably modulating the ventricular input pressure which determines the beat rate and the contraction force of the ventricle (Smith, 1981a; Foti et al., 1985). Further studies should attempt to clarify whether there are inhibitory components to the system, e.g. acetylcholine, which might counteract the excitatory elements.

We would like to thank the Department of Anästhesie und Operative Intensivmedizin, Klinikum of the JLU, Giessen, for providing the technical facilities for the HPLC analysis. We also thank J. P. Labourg, C. Cazaux and M. Caumette who provided the technical facilities for work in the Laboratoire d’Océanographie Biologique Arcachon and A. Guille and S. von Boletzky for similar support at the Observatoire Océanologique de Banyuls. We thank J. Schmandt, A. Hudel and B. Fronk for their valuable technical assistance and are also grateful to R. Lawson for reading and correcting the manuscript. This work was supported by grants from the Deutsche Forschungsgemeinschaft (Schi 99/7-2; Schi 99/7-4).