Neuronal control of catecholamine secretion from chromaffin cells in the rainbow trout (Oncorhynchus mykiss)

In teleosts, the principal sources of circulating catecholamines are chromaffin cells located within the walls of the posterior cardinal vein (Nandi, 1961; Gallo et al., 1993). The primary mechanism leading to their secretion is believed to be increased neuronal stimulation by preganglionic nerve fibres (from the sympathetic nervous system) innervating the chromaffin cells (Nilsson, 1983). Upon stimulation, the release of acetylcholine stimulates cholinergic receptors, ultimately initiating a series of Ca²⁺-dependent events leading to the secretion of catecholamines (Nilsson et al., 1976; Burgoyne et al., 1993; Fritsche et al., 1993; Furimsky et al., 1996). In teleost fish, it appears that nicotinic receptors are the predominant cholinoceptor present on chromaffin cells (for a review, see Reid et al., 1998).

The nicotinic nature of cholinergic catecholamine secretion has been demonstrated in several teleosts, including rainbow trout (Oncorhynchus mykiss; Fritsche et al., 1993; Julio et al., 1998), American eel (Anguilla rostrata; Reid and Perry, 1995; Al-Kharrat et al., 1997), Atlantic cod (Gadus morhua; Nilsson et al., 1976) and carp (Cyprinus carpio; Gfell et al., 1997). In the Atlantic cod, electrical stimulation of the sympathetic fibres or application of acetylcholine to the head kidney caused the secretion of catecholamines from in situ perfused preparations (Nilsson et al., 1976). Furthermore, bolus injections of the non-selective cholinergic agonist carbachol to in situ perfused preparations of American eels caused the secretion of both noradrenaline and adrenaline (Reid and Perry, 1995). In those studies, the secretion of catecholamines in response to either stimulus was prevented by prior treatment with the nicotinic receptor antagonist hexamethonium.

In rainbow trout, nicotinic receptor stimulation is believed to be the primary mediator of cholinergic-induced secretion of catecholamines (Reid et al., 1998). Recent evidence, however, suggests that muscarinic cholinergic stimulation may also contribute to catecholamine release. For example, in isolated trout chromaffin cells, hexamethonium reduced the increase in intracellular Ca²⁺ concentration ([Ca²⁺]_i) during nicotine administration, but was without effect on [Ca²⁺]_i in the presence of carbachol (Furimsky et al., 1996). Furthermore, Fritsche et al. (1993) demonstrated, using an in situ saline-perfused posterior cardinal vein preparation of rainbow trout, that pre-treatment with hexamethonium prevented the carbachol-induced release of noradrenaline but failed to block ‘completely’ the release of adrenaline. Together, the results of these studies suggest a role for muscarinic receptors in the control of catecholamine secretion from trout chromaffin cells.

More recently, Julio et al. (1998) demonstrated that the muscarinic receptor agonist pilocarpine was capable of directly eliciting the release of adrenaline and noradrenaline from in situ perfused posterior cardinal vein preparations of trout. Selective muscarinic receptor antagonists (e.g. atropine), however, were not used in any of these studies, thus casting doubt on a specific involvement of muscarinic receptor stimulation. Studies designed to evaluate the contribution of the muscarinic receptor to catecholamine secretion in other teleosts (eel and carp) have produced conflicting results (Reid and Perry, 1995; Al-Kharrat et al., 1997; Gfell et al., 1997; Abele et al., 1998). Moreover, while previous experiments have indirectly (Fritsche et al., 1993; Furimsky et al., 1996) or directly (Julio et al., 1998) suggested the presence of muscarinic receptors, no studies have yet been performed to evaluate the relative involvement of muscarinic receptors in catecholamine secretion during neuronal stimulation of rainbow trout chromaffin cells. Thus, the goal of the present investigation was to determine whether neuronal cholinergic stimulation of catecholamine secretion in rainbow trout involves muscarinic receptors. It was first necessary to develop and validate a nerve-stimulating protocol able specifically to elicit catecholamine secretion from the chromaffin tissue.

Rainbow trout Oncorhynchus mykiss (Walbaum) of both sexes were obtained from Linwood Acres Trout Farm (Campbellcroft, Ontario, Canada) and held in large fibreglass tanks supplied with dechlorinated city of Ottawa tapwater maintained at 15 °C. They were allowed to acclimate to the aquaria for at least 3 weeks before experimentation. In total, 164 sexually mature fish ranging in mass between 180 and 337 g (213±35 g; mean ± S.E.M.) were used between February and September. Fish were maintained on a 12 h:12 h L:D photoperiod and fed daily to satiation with a commercial fish diet.

Fish were killed by a sharp blow to the head, weighed and placed on ice. A ventral incision was made along the entire length of the animal, and the tissue overlying the heart was removed by blunt dissection to expose the ventricle and the bulbus arteriosis. The posterior cardinal vein and ventricle were catheterized (Clay-Adams, PE 160 polyethylene tubing) and served as the inflow and outflow, respectively, of the perfusion fluid. Each preparation was perfused for 20 min with modified aerated Cortland saline (Wolf, 1963; 125 mmol l⁻¹ NaCl, 2.0 mmol l⁻¹ KCl, 2.0 mmol l⁻¹ MgSO₄, 5.0 mmol l⁻¹ NaHCO₃, 7.5 mmol l⁻¹ glucose, 2.0 mmol l⁻¹ CaCl₂ and 1.25 mmol l⁻¹ KH₂PO₄, final pH 7.8) to allow catecholamine secretion to stabilize. Perfusion was accomplished by siphon resulting from a positive pressure difference between the surface of the saline and the outflow cannula, which resulted in a flow rate of approximately 1.5 ml min⁻¹.

After the stabilization period, a control pre-stimulation sample was collected in a pre-weighed microcentrifuge tube to assess basal catecholamine secretion rate. After collection of the pre-stimulation sample, electrical stimulation or a bolus injection of agonist was given to the preparation in accordance with the treatments described below. Agonists were delivered by bolus injection (final volume 0.3 ml) through a three-way valve connected to the inflow catheter. A period of 1 min was allowed for the drug to be delivered to the chromaffin tissue before post-stimulation samples were collected in pre-weighed tubes. In total, five post-stimulation samples were collected 1, 2, 3, 4 and 5 min after intervention. All samples were frozen in liquid N₂ after collection and stored at −80 °C until determination of catecholamine levels. Perfusate samples were reweighed before catecholamine analysis to permit an estimation of perfusion flow rates and thus to allow the calculation of catecholamine secretion rates.

Fish were electrically stimulated using a pair of electrodes (connected to a Grass SD-9 stimulator) sutured to the body wall on either side of the fish in the anterior region of the posterior cardinal vein. After collection of the pre-stimulation sample, each preparation was stimulated only once at 0, 10, 30, 45, 60 or 80 V. Field stimulation was carried out at a frequency of 20 pulses s⁻¹, 1 ms in duration for a period of 30 s, and collection of post-stimulation samples began immediately at the onset of stimulation. Stimulation at 60 V was determined to be the lowest stimulus amplitude capable of eliciting maximal catecholamine secretion (Fig. 1) and was subsequently utilized for all experiments involving field stimulation.

To assess whether field stimulation causes non-specific depolarization of chromaffin cells or neuronally mediated catecholamine secretion, the following series of experiments was performed.

Using the same procedures as described above, the effects of the anticholinesterase neostigmine on electrically evoked catecholamine secretion were assessed in preparations perfused with control saline or with saline containing neostigmine bromide (10⁻⁴ mol l⁻¹).

Two sets of experiments were performed in this series using the protocol of series 1. In the first, fish were perfused with control saline or with saline containing the nicotinic receptor antagonist hexamethonium (10⁻³ mol l⁻¹). In the second, fish were perfused with saline or with Na⁺-free saline containing 125 mmol l⁻¹N-methyl-D-glucamine (NMDG), 2 mmol l⁻¹ KCl, 2 mmol l⁻¹ MgSO₄, 5 mmol l⁻¹ KHCO₃, 7.5 mmol l⁻¹ glucose, 2 mmol l⁻¹ CaCl₂ and 1.25 mmol l⁻¹ KH₂PO₄.

Using the same procedures as in series 1, electrical stimulation was applied to fish in which the anterior third of the spinal cord had been removed. For sham preparations, the spinal cord was exposed but not removed.

The following series of experiments was performed to study the contributions of nicotinic and muscarinic receptor stimulation to catecholamine secretion from the chromaffin tissue.

After collection of the pre-stimulation sample, bolus injection (0.3 ml) of the non-selective cholinergic agonist carbachol (10⁻⁵ mol kg⁻¹ body mass) or electrical stimulation (as in series 1) was administered to preparations perfused with control saline or with saline containing the muscarinic receptor antagonist atropine (10⁻⁵ mol l⁻¹), the nicotinic receptor antagonist hexamethonium (10⁻³ mol l⁻¹) or both hexamethonium (10⁻³ mol l⁻¹) and atropine (10⁻⁴ mol l⁻¹).

Furthermore, a bolus injection (0.3 ml) of nicotine (10⁻⁷ mol kg⁻¹) or the muscarinic receptor agonist metacholine (10⁻³ mol l⁻¹) was administered to preparations pefused with control saline or saline containing hexamethonium (10⁻⁴ mol l⁻¹) or atropine (10⁻⁵ mol l⁻¹) to confirm the specificity of the cholinergic receptor antagonists.

After collection of the pre-stimulation sample, preparations were given a bolus injection (0.3 ml) of the nicotinic receptor agonist nicotine (10⁻⁷ mol kg⁻¹ body mass), the muscarinic receptor agonists oxotremorine sesquifumarate (10⁻⁴ mol kg⁻¹ body mass) or methacholine (10⁻³ mol kg⁻¹ body mass) or a cocktail of nicotine (10⁻⁷ mol kg⁻¹ body mass) and oxotremorine (10⁻⁴ mol kg⁻¹ body mass) or nicotine and methacholine (10⁻³ mol kg⁻¹ body mass).

Bolus injections (0.3 ml) of methacholine (10⁻³ mol kg⁻¹ body mass) or oxotremorine (10⁻⁴ mol kg⁻¹ body mass) were administered to preparations perfused with control saline, with saline containing atropine (10⁻⁴ mol l⁻¹) or with saline containing hexamethonium (10⁻³ mol l⁻¹).

Perfusate catecholamine levels were determined on alumina-extracted samples using high-pressure liquid chromatography (HPLC) with electrochemical detection (Woodward, 1982). 3,4-Dihydroxybenzalamine hydrobromide was used as an internal standard.

The data are presented as the mean ±1 standard error of the mean (S.E.M.). Where appropriate, the data were analysed statistically using a one-way repeated-measures analysis of variance (ANOVA) followed by Dunnett’s test for comparison with pre-stimulation values, or a one-way ANOVA followed by Dunnett’s test for multiple comparisons; if the normality test failed, an analysis of covariance (ANOVA) on ranks was performed followed by Dunnett’s multiple-comparison test. In other instances, data were analysed by Student’s t-test or paired t-test, and if the normality test failed, a Mann–Whitney rank sum test was performed. The fiducial limits of significance were set at 5 %. All statistical tests were performed using a commercial statistical software package (SigmaStat version 2.03).

Field stimulation ranging from 0 to 80 V elicited a voltage-dependent secretion of both catecholamines (Fig. 1). Significant elevation of the rate of adrenaline and noradrenaline secretion occurred at 45 V and reached a maximum at 60 V.

Field stimulation caused a significant increase in the rate of adrenaline secretion in both saline-perfused and neostigmine-perfused preparations (Fig. 2A). However, in neostigmine-treated preparations, secretion remained elevated at 2 and 3 min post-stimulation; over the 5 min post-stimulation period, total adrenaline secretion was nearly doubled (Fig. 2A). Pre-treatment with neostigmine did not affect noradrenaline secretion (Fig. 2B).

Perfusion with saline containing hexamethonium reduced the rate of adrenaline and noradrenaline secretion in response to field stimulation (Fig. 3). Adrenaline secretion, although reduced in the presence of hexamethonium, was significantly increased from pre-stimulation values (Fig. 3A). In contrast, perfusion with Na⁺-free saline eliminated both adrenaline and noradrenaline secretion during field stimulation (Fig. 4).

Removal of the spinal cord totally inhibited the secretion of both adrenaline and noradrenaline (Fig. 5) in response to field stimulation.

A set of experiments was performed to confirm the specificity of the cholinergic receptor antagonists hexamethonium and atropine; the results are presented in Table 1. Nicotine-evoked catecholamine secretion was abolished by pre-treatment with hexamethonium and unaffected by pre-treatment with atropine. Conversely, methacholine (cholinergic receptor agonist) -evoked catecholamine secretion was unchanged in the presence of hexamethonium but blocked by atropine. These results indicate that, at the concentrations employed in the present study, hexamethonium (10⁻³ mol l⁻¹) and atropine (10⁻⁵ mol l⁻¹) were acting as specific nicotinic and muscarinic receptor antagonists, respectively.

Both adrenaline and noradrenaline secretion rates were increased over baseline levels in response to field stimulation or bolus injections of carbachol (Figs 6A, 7A). In fish given carbachol, secretion of both catecholamines remained above resting rates for the entirety of the experiment (Fig. 7A). Pre-treatment with atropine significantly reduced the rate of adrenaline secretion in response to both stimuli (Figs 6B, 7B). Pre-treatment with hexamethonium significantly reduced the rate of adrenaline secretion in response to field stimulation (Fig. 6C). However, hexamethonium entirely prevented adrenaline secretion in fish given carbachol (Fig. 7C). Similar results were obtained in the presence of both antagonists (Figs 6D, 7D). The rate of adrenaline secretion was markedly reduced during electrical stimulation (Fig. 6D) and abolished during the application of carbachol (Fig. 7D). Pre-treatment with any of the antagonists was without effect on the rate of noradrenaline secretion during electrical stimulation (Fig. 6B–D). However, atropine reduced the rate of noradrenaline secretion in response to bolus injections of carbachol, while hexamethonium totally prevented its release (Fig. 7B–D).

Fig. 8 illustrates the effects of bolus injections of nicotine, oxotremorine (muscarinic agonist) and a cocktail of both agonists on the rate of catecholamine secretion by the perfused posterior cardinal vein preparation. Nicotine or oxotremorine administration predominantly affected adrenaline secretion. Noradrenaline secretion was only slightly affected, reaching approximately 0.1 nmol min⁻¹ in response to nicotine or oxotremorine. When administered together, nicotine and oxotremorine appeared to have a simple additive effect on the rate of adrenaline secretion. The effect on the rate of noradrenaline secretion, however, was greater than additive and the rate was more than tripled in the presence of both agonists. Application of another muscarinic agonist, methacholine (10⁻³ mol kg⁻¹), had similar effects on the rate of adrenaline secretion (see Table 1). Nicotine-evoked adrenaline and noradrenaline secretion occurred transiently. Oxotremorine administration caused a sustained secretion of both adrenaline and noradrenaline, albeit at a lower rate.

Pre-treatment with atropine or hexamethonium had no effect on maximal catecholamine secretion rates in response to a bolus injection of oxotremorine (data not shown). In contrast, atropine totally prevented methacholine-evoked secretion (Table 1).

In this study, we have developed and validated a nerve-stimulating procedure to study the neuronal control of catecholamine secretion from rainbow trout chromaffin tissue. Using this technique, we have demonstrated for the first time an involvement of muscarinic receptors in the cholinergic control of catecholamine secretion from rainbow trout chromaffin cells.

To demonstrate conclusively a role for muscarinic receptors in the stimulation of catecholamine secretion during neuronal excitation of chromaffin tissue, it was necessary to compare rates of catecholamine secretion during electrical activation of the nerves innervating the chromaffin cells in the presence and absence of a selective muscarinic receptor antagonist (e.g. atropine). Repeated efforts to stimulate the nerves innervating the chromaffin cells of trout by direct application of electrical current to the spinal cord or brainstem were unsuccessful: procedures known to induce neuronal catecholamine secretion in the cod Gadus morhua (Nilsson et al., 1976) or eel (Abele et al., 1998) were ineffective in trout. For this reason, a field stimulation protocol was developed and validated to elicit neuronally induced catecholamine secretion from the chromaffin cells.

Field stimulation of in situ saline-perfused posterior cardinal vein preparations could induce catecholamine secretion in one of three possible ways. First, conduction of electrical current to the spinal cord could generate action potentials in the preganglionic sympathetic fibres known to innervate the chromaffin cells and ultimately lead to the release of acetylcholine (among other neurotransmitters and neuromodulators). Second, the electrical current could cause direct excitation of the preganglionic nerve terminals without the intervention of the spinal cord. Third, catecholamine secretion could cause non-specific depolarization of the chromaffin cells in response to the electrical current. In the present study, data obtained using neostigmine, hexamethonium, Na⁺-free saline and from fish in which a portion of the spinal cord had been removed demonstrated conclusively that catecholamine secretion in response to field stimulation was neuronally induced and dependent upon an intact spinal cord. Furthermore, the contribution of non-specific depolarization of chromaffin cells to secretion appeared to be negligible.

The specificity of the field stimulation technique was confirmed by demonstrating that cholinergic catecholamine secretion was reduced in the presence of the nicotinic receptor antagonist hexamethonium, and that adrenaline secretion was prolonged and more than doubled in the presence of neostigmine over the 5 min period of sample collection. The absence of an effect of neostigmine on noradrenaline secretion during field stimulation was unexpected. However, if the the rate of secretion of noradrenaline were already maximal during field stimulation, then any effects of neostigmine on stabilizing acetylcholine levels would presumably go unnoticed.

Nevertheless, these results, together with the observation that secretion was prevented in fish in which the spinal cord had been removed, indicate that catecholamine secretion elicited by field stimulation is neuronally mediated.

To address the possibility that catecholamine secretion occurred as a result of non-specific depolarization, fish were perfused with Na⁺-free saline. It is believed that nicotinic receptor stimulation is coupled to the opening of receptor-linked Na⁺ channels and that the resultant membrane depolarization causes an opening of voltage-dependent Ca²⁺ channels (Brandt et al., 1976; Wada et al., 1985; Liu and Kao, 1990). The subsequent influx of Ca²⁺ is thought to trigger the cascade of events leading to the secretion of catecholamines (Burgoyne et al., 1993). The absence of a response during field stimulation in preparations perfused with Na⁺-free saline and from which the spinal cord had been removed indicates that the electrical current, in itself, was unable to elicit sufficient chromaffin cell membrane depolarization to initiate catecholamine secretion. Therefore, on the basis of these findings, we propose that field stimulation causes specific neuronally mediated secretion of catecholamines from trout chromaffin cells and that this protocol can be used as a tool to study the neuronal mechanisms regulating the secretion of catecholamines in fish.

During periods of acute severe stress, teleost fish release noradrenaline and adrenaline into the circulation (for reviews, see Randall and Perry, 1992; Thomas and Perry, 1992; Wendelaar Bonga, 1997; Reid et al., 1998). The subsequent elevation of plasma catecholamine levels is believed to initiate a series of physiological adjustments aimed at enhancing blood O₂-transport capacity and gill O₂-diffusing capacity (for reviews, see Randall and Perry, 1992; Epple, 1993). In teleosts, the secretion of catecholamines from chromaffin cells is under the control of numerous physiological stimuli. These include stimulation by preganglionic sympathetic nerve fibres (Nilsson et al., 1976; Abele et al., 1998), localized changes in blood chemistry (Opdyke et al., 1983; Perry et al., 1991), activation of the renin–angiotensin system (Opdyke et al., 1981; Bernier and Perry, 1997), serotonin (Fritsche et al., 1993; Reid et al., 1996) and adrenocorticotropic hormone (Reid et al., 1996; Bernier and Perry, 1996). Of these, increased neuronal stimulation of the chromaffin cells by preganglionic cholinergic fibres is thought to be the predominant mechanism promoting catecholamine secretion (for reviews, see Epple et al., 1995; Reid et al., 1998).

On the basis of the results of the present investigation, it appears that cholinergic control of catecholamine secretion from trout chromaffin cells is not confined to nicotinic-receptor-evoked events. Indeed, this is the first study to demonstrate conclusively a role for muscarinic receptors in catecholamine secretion from rainbow trout chromaffin cells. An involvement of muscarinic receptors in the catecholamine secretion process is not without precedent. In the common carp (Cyprinus carpio), the muscarinic receptor agonists muscarine and pilocarpine elicited catecholamine secretion in perifused head kidney preparations (Gfell et al., 1997). Previous studies using American eels (Anguilla rostrata) have provided conflicting evidence. Reid and Perry (1995) were unable to demonstrate any involvement of muscarinic receptors, whereas other studies have documented an inhibitory role of basal secretion (Al-Kharrat et al., 1997) or a positive modulatory role under electrical stimulation of the sympathetic fibres (Abele et al., 1998). Moreover, there are numerous reports that demonstrate a role for muscarinic receptors in catecholamine secretion in non-piscine vertebrates, although the results are highly variable. Indeed, these investigations have demonstrated that stimulation of muscarinic receptors can enhance nicotinic-evoked release (Role and Perlman, 1983; Accordi, 1991; Nassar-Gentina et al., 1997), inhibit nicotinic-evoked release (Swilem and Hawthorn, 1983; Cheek and Burgoyne, 1985; Nassar-Gentina et al., 1991), stimulate secretion in the absence of nicotinic agonists (Wakade and Wakade, 1983; Knight and Baker, 1986; Ballesta et al., 1989; Xu et al., 1991; Tobin et al., 1992; Chowdhury et al., 1994) or exert no effects on catecholamine secretion (Liang and Perlman, 1979; Mckay et al., 1991). The results of the present investigation suggest that, in rainbow trout, stimulation of muscarinic receptors enhances nicotine-induced secretion and that this, under intense stimulation, may directly elicit catecholamine secretion. This was shown by the fact that catecholamine secretion elicited by either field stimulation or carbachol was inhibited by atropine but was not further decreased when atropine was added with hexamethonium. The direct effect of muscarinic receptor stimulation on catecholamine secretion was confirmed using the specific muscarinic agonists oxotremorine and methacholine.

This is the first study in rainbow trout to use a muscarinic receptor antagonist (atropine) and the muscarinic receptor agonists methacholine and oxotremorine to investigate the cholinergic control of catecholamine secretion. Methacholine- and oxotremorine-induced catecholamine secretion and its inhibition by atropine have been described for avian (Knight and Baker, 1986) and mammalian (Ballesta et al., 1989; Xu et al., 1991; Yamagami et al., 1991; Nassar-Gentina et al., 1997) chromaffin cells. In the present study, atropine inhibited methacholine-evoked secretion but did not reduce the rate of catecholamine secretion during administration of oxotremorine. The inability of atropine to block oxotremorine-evoked secretion is not understood. However, resistance to muscarinic antagonists has been demonstrated in other systems (Robitaille et al., 1997). The possibility that oxotremorine-evoked secretion was mediated by activation of nicotinic receptors can be discounted because the response was insensitive to hexamethonium. Taken together, the fact that methacholine and oxotremorine elicited secretion and that atropine reduced the rate of secretion in response to field stimulation and bolus injections of carbachol and methacholine clearly indicates the presence of muscarinic receptors on rainbow trout chromaffin cells with a role in the control of catecholamine secretion.

Although high concentrations of cholinergic receptor antagonists were used in the present study (10⁻³ mol l⁻¹ hexamethonium; 10⁻⁴ mol l⁻¹ atropine), the effective levels in the vicinity of the chromaffin cells may have been considerably lower owing to diffusion limitations of the perfused posterior cardinal vein preparation. Indeed, the actual concentration of antagonist (or agonist) achieved in the extracellular fluid bathing the chromaffin cells is unknown. Nevertheless, the specificity of the cholinergic receptor antagonists was confirmed in a separate set of experiments (see Table 1). Furthermore, the results of previous studies utilising the same perfused preparation have demonstrated that hexamethonium (10⁻⁴ to 10⁻³ mol l⁻¹) does not modify catecholamine secretion elicited by angiotensin II (Bernier and Perry, 1997), adrenocorticotropic hormone (ACTH) (Reid et al., 1996), serotonin (Fritsche et al., 1993) or depolarizing levels of KCl (Reid and Perry, 1995). In addition, the secretion of adrenaline caused by vasoactive intestinal peptide (VIP) (10⁻⁹ mol kg⁻¹) in the perfused preparation is unaffected by a cocktail of hexamethonium and atropine (C. Montpetit and S. Perry, unpublished data). Thus, there is no evidence that hexamethonium or atropine elicits non-specific side effects on the catecholamine secretion process at the concentrations employed in the present study.

Modulation of presynaptic acetylcholine release via muscarinic receptor stimulation is an alternative explanation for the observations. Muscarinic cholinergic modulation of presynaptic activity has been demonstrated in the central nervous system of non-piscine vertebrates (Raiteri et al., 1990). It is unlikely, however, that the muscarinic effect on catecholamine secretion observed in the present study was mediated through modulation of presynaptic acetylcholine release from nerve fibres innervating the chromaffin cells. First, presynaptic muscarinic receptors generally regulate acetylcholine release through inhibitory actions (Raiteri et al., 1990). Second, the nicotinic receptor antagonist hexamethonium did not affect the oxotremorine-evoked secretion of catecholamines. If, indeed, stimulation of presynaptic muscarinic receptors was causing acetylcholine release, then catecholamine secretion should have been reduced by hexamethonium. Third, Furimsky et al. (1996) demonstrated that, when hexamethonium was added to isolated trout chromaffin cells after stimulation with carbachol, there was no significant decrease in [Ca²⁺]_i as was observed after nicotine administration, thus implicating the presence of hexamethonium-insensitive muscarinic receptors.

The stimulation of muscarinic receptors is known to activate phospholipase C to produce inositol trisphosphate, which subsequently causes the release of Ca²⁺ from intracellular stores within chromaffin cells (Eberhard and Holtz, 1987; O’Sullivan and Burgoyne, 1989). In some species, the increase in [Ca²⁺]_i is not sufficient to trigger catecholamine secretion (Cheek and Burgoyne, 1985; Abad et al., 1992; Sorimachi et al., 1992). However, in other species, muscarinic receptor stimulation can cause sufficient release of Ca²⁺ from intracellular stores to elicit the secretion of catecholamines (Burgoyne et al., 1993).

In accordance with other studies, high doses of muscarinic agonists were required to elicit catecholamine secretion. In rainbow trout, muscarinic receptor stimulation during field stimulation does not appear to be able to induce secretion in the absence of nicotinic receptor activation. However, under more intense conditions of stimulation, muscarinic receptor stimulation may have direct positive effects on secretion. Using the field stimulation technique, the muscarinic-cholinergic-evoked increase in [Ca²⁺]_i was presumably not sufficient to trigger secretion but probably enhanced post-nicotinic receptor Ca²⁺-dependent events, thereby increasing the nicotinic-induced secretion of catecholamines. However, under more intense muscarinic receptor stimulation, intracellular Ca²⁺ concentrations may reach threshold and trigger the secretion process.

Furthermore, muscarinic cholinergic stimulation of catecholamine release may help to sustain secretion in the event of nicotinic receptor desensitization (Boksa and Livett, 1984). In the present study, oxotremorine, methacholine and carbachol administration caused a sustained and prolonged secretion of both catecholamines. This response might be useful during prolonged stressful situations when the nicotinic secretory response may be desensitized (Boksa and Livett, 1984).

It is possible that catecholamine secretion is also controlled by non-cholinergic neurotransmitters. Reid et al. (1995) identified a number of neurotransmitters or neuropeptides in nerve fibres in the vicinity of chromaffin cells in teleost fish, including VIP and pituitary adenylate cyclase activating polypeptide (PACAP). Their role in the secretion process in fish has yet to be investigated, although both are known to have stimulatory effects on catecholamine secretion in non-piscine vertebrates (Watanabe et al., 1995; Yamaguchi, 1993). Similar studies are required to determine their relative contributions to catecholamine secretion in teleost fish. A line of evidence for this occurrence is that, in response to field stimulation, pre-treatment with hexamethonium plus atropine failed to inhibit completely the secretion of both adrenaline and noradrenaline. In conclusion, using a ‘new’ field stimulating technique, this is the first study to investigate directly the possible involvement of muscarinic receptor stimulation of catecholamine secretion from the chromaffin tissue in the rainbow trout. The results of the study suggest the presence of muscarinic receptors with a functional role in the cholinergic control of catecholamine secretion. More specifically, muscarinic receptor stimulation enhances the nicotinic-evoked secretion of catecholamines and this, under intense stimulation, may directly elicit secretion.

This work was financed by National Science and Engineering Research Council of Canada Research and Equipment grants to S.F.P. C.J.M. was the recipient of an Ontario Graduate Scholarship in Science and Technology.