Dopamine stimulates snail albumen gland glycoprotein secretion through the activation of a D1-like receptor

The catecholamine dopamine has a widespread distribution in the molluscan central nervous system (CNS; Hetherington et al., 1994; Trimble et al.,1984; Audesirk,1985; Muneoka et al.,1991), where it is a neurotransmitter regulating numerous physiological processes including learning and memory(Lukowiak and Syed, 1999), the regulation of neurite outgrowth and synaptic connectivity(Spencer et al., 1998),embryonic neural development (Croll,2000) and the regulation of neural circuits involved in feeding(Elliott and Susswein, 2002)and respiration (Taylor and Lukowiak, 2000). Dopamine is also found in peripheral tissues of molluscs, where it controls physiological processes such as smooth muscle contraction (Gies,1986), motility of the gastrointestinal tract(Hernadi et al., 1998) and salivary secretion (Kiss et al.,2003).

Catecholaminergic axons have been shown to innervate the reproductive organs of various molluscs (Hartwig et al., 1980; Smith et al.,1998; Croll et al.,1999; Croll, 2001; Kiehn et al., 2001), and histochemical analyses have revealed that catecholaminergic cell bodies and axon processes are concentrated in the region of the exocrine albumen gland(AG) of the freshwater snails Bulinus truncatus(Brisson and Collin, 1980; Brisson, 1983) and Helisoma duryi (Kiehn et al.,2001). The AG is a female accessory reproductive gland that secretes a viscous substance known as the perivitelline fluid (PVF) around the individual eggs as they enter the carrefour, the area where the main duct of the AG empties. The PVF consists mainly of glycoproteins and galactogen (a highly branched galactose polymer), which provide the main source of nutrients to the developing embryos (Duncan,1975; Geraerts and Joosse,1984). Therefore, the timely release (secretion) of PVF is a key regulatory process governing egg production in freshwater snails.

Although the neuroendocrine control of polysaccharide (galactogen)synthesis in the AG has been studied extensively in freshwater pulmonate molluscs (de Jong-Brink et al.,1982; Wijdenes et al.,1983; Miksys and Saleuddin, 1985, 1988; Mukai et al., 2001b), little is known with respect to the regulation of PVF release. We have identified the major protein produced by the AG of the freshwater snail H. duryi as a 288 kDa glycoprotein, which is composed of several 66 kDa subunits(Morishita et al., 1998). The gene encoding for the 66 kDa glycoprotein subunit has recently been cloned(Mukai et al., 2004) and the corresponding protein was given the name Helisoma duryi albumen gland protein (HdAGP). The release of HdAGP is known to be stimulated by a novel brain peptide (Morishita et al.,1998) through a cAMP signalling pathway(Mukai et al., 2001a).

The secretion of HdAGP coincides with the arrival of the eggs at the carrefour, and the existence of a control mechanism over AG secretory activity has been postulated (Mukai et al.,2004). Kiehn et al.(2001) showed that the AG and carrefour region of H. duryi is innervated by dopaminergic nerve fibres. Moreover, dopamine has also been shown to induce the secretion of total protein from isolated AGs of H. duryi(Mukai, 1998; Saleuddin et al., 2000) and Biomphalaria glabrata(Santhanagopalan and Yoshino,2000). Subsequent studies by Boyle and Yoshino(2002) showed that B. glabrata AG dopamine levels increased during the initial stage of egg mass production (the period during which the AG is secreting PVF), whereas protein levels in the AG decreased during the latter stages of egg mass production. Collectively, these results suggest that the secretion of protein by the AG of freshwater pulmonate snails is regulated by dopamine and, by inference, specific dopamine receptors in the AG.

Dopamine receptors were originally classified into two categories, D1 and D2, based on their ability when activated to either stimulate or inhibit,respectively, adenylate cyclase activity (reviewed by Civelli et al., 1993). The genes for five distinct dopamine receptor subtypes (D1, D2, D3, D4, D5) have been cloned in mammals and placed into one of the two dopamine receptor groups based on their gene structure and pharmacology(O'Dowd, 1993). The dopamine D1-like receptors, which stimulate cAMP formation, comprise the D1 and D5 subtypes, whereas the dopamine D2-like receptors, comprising the D2, D3 and D4 subtypes, either inhibit or have no effect on adenylate cyclase(Missale et al., 1998). Using a number of dopamine receptor agonists and antagonists, Saleuddin et al.(2000) suggested that a dopamine D1-like receptor mediates total protein secretion by AGs of H. duryi. Here, we extend our previous study and examine the intracellular signalling pathway activated after AGs were treated with dopamine. Addition of exogenous dopamine to AG explants stimulates secretion of HdAGP viathe activation of the cAMP signalling pathway. A number of D1-selective and D2-selective agonists and antagonists were used to obtain a pharmacological profile of the AG dopamine receptor. It is concluded that the AG dopamine receptor regulating protein secretion is distinct but functionally similar to vertebrate D1-like receptors and mediates secretion of HdAGP through an elevation in glandular cAMP, possibly through a protein kinase A(PKA)-independent pathway.

Helisoma duryi were reared in 2-litre plastic containers with dechlorinated tap water at 22°C and maintained under a 16 h:8 h L:D photoperiod. The snails were fed a diet of boiled lettuce every 2–3 days, occasionally supplemented with fish pellets. The water was changed at least once per week. Two weeks prior to experimentation, adult snails (15 mm shell diameter) were selected and then individually placed in plastic cups to monitor their egg-laying activity. Only snails that had not laid eggs within the previous 12 h were used for experiments.

Dopamine (3-hydroxytyramine), serotonin, acetylcholine,γ-aminobutyric acid, norepinephrine, histamine, octopamine, glutamate,forskolin (7β-acetoxy-8,13-epoxy-1α, 6β,9α-trihydroxylabd-14-en-11-one), IBMX (3-isobutyl-1-methylxanthine),apomorphine hydrochloride, bromocriptine (2-bromo-α-ergocryptine methanesulfonate salt), (R)(+)-SCH23390 hydrochloride, (R)(±)-SKF-38393 hydrochloride, R(+)-SKF81297 hydrobromide, (±)-SKF83566 hydrochloride,dihydrexidine hydrochloride{(±)-trans-10,11-dihydroxy-5,6,6a,7,8,12b-hexahydrobenzo[a]phenanthridine},6,7-ADTN [(±)-2-amino-6,7-dihydroxy-1,2,3,4-tetrahydronapthalene],haloperidol{4-(4-[4-chlorophenyl]-4-hydroxy-1-piperidinyl)-1-(4-fluorophenyl)-1-butanone},(–)-butaclamol hydrochloride, cis-flupenthixol, chlorpromazine,eticlopride, Rp-cAMP and H-89 were purchased from Sigma-Aldrich Canada,Oakville, Ontario, Canada.

Albumen glands were dissected free from surrounding tissue under Helisoma saline (51.3 mmol l^-1 NaCl, 1.7 mmol l^-1 KCl, 4.1 mmol l^-1 CaCl₂, 1.5 mmol l^-1 MgCl₂, 5.0 mmol l^-1 Hepes, 1 mmol l^-1 glucose, pH 7.4, 120 mOsm H₂O), cut into halves and then washed in several changes of saline. One half served as a control gland while the other was treated with test compound. Each AG piece(∼0.5–1.5 mg) was placed in a separate well of a 96-well culture plate (Becton-Dickinson and Co., Lincoln Park, NJ, USA) containing 100 μl of saline. The saline surrounding the AG was removed and replaced with another 100 μl of fresh saline every 20 min. The collected saline was placed in 1.7 ml polypropylene microtubes (Brinkmann Instruments Inc., Westbury, NY, USA)and centrifuged at 2000 g for 1 min. An 80 μl sample was removed and added directly to a 1.5 ml polystyrene cuvette containing 420μl of Triton X-100 (0.0095% in water). Total protein was determined by adding 125 μl Bio-Rad Protein Dye Reagent Concentrate (Bio-Rad Laboratories Canada Ltd, Mississauga, Ontario, Canada) and measuring the absorbance at 595 nm with a Zeiss PM2DL spectrophotometer (Carl Zeiss, Oberkochen, Germany). Basal HdAGP secretion was measured for the first 60 min. A stock solution of dopamine (10 mmol l^-1) was prepared in deoxygenated-demineralized water and diluted in saline to a final concentration of 10 μmol l^-1 immediately before use. All the test compounds were dissolved in Helisoma saline (with or without dopamine) to their final concentrations and applied to the AGs at 60 min. The test agents were removed at 80 min and replaced with normal saline for another 60 min. Dopamine receptor antagonists (D1-selective or D2-selective) or PKA inhibitors (Rp-cAMP and H-89) were first preincubated with AGs at 40 min, then removed and replaced with antagonist plus dopamine at 60 min. The amount of HdAGP secreted between 40 and 60 min (control) was compared with the amount secreted between 60 and 80 min (treated).

To qualitatively determine the proteins secreted by the AG in vitro after dopamine stimulation, the saline surrounding the AG was collected and analyzed by sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE). The collected saline was first evaporated to dryness using a Savant SVC 100 Speed-Vac (Instruments Inc., Farmingdale, NY,USA), then resuspended in SDS-PAGE sample buffer and separated on a 9%mini-gel apparatus (Bio-Rad) according to Laemmli(1970). Following electrophoresis, the gel was stained overnight with 0.2% Coomassie Brilliant Blue R-250 in 50% methanol/10% acetic acid. The gel was destained the next day and dried onto Whatman 3MM filter paper using a slab gel dryer (Hoefer Scientific Instruments, San Francisco, CA, USA).

Albumen glands were dissected under snail saline, then quartered and rinsed in normal saline. Prior to treatment, each AG piece was preincubated for 20 min with 1 mmol l^-1 IBMX in saline to allow the phosphodiesterase inhibitor to penetrate the AG cells. All test compounds were dissolved in IBMX saline and applied to the AG for 10 min unless indicated otherwise. For the time course experiment, one AG piece served as a control, the second piece was treated with dopamine (10 μmol l^-1), the third piece was treated with forskolin (10 μmol l^-1) and the fourth piece was treated with both dopamine and forskolin. At the end of each time point, the AGs were immediately plunged into liquid nitrogen, then stored in 1.7 ml microtubes at–80°C. To extract cAMP, AGs were homogenized in 3% ice-cold perchloric acid using a motor-driven Teflon pestle and then centrifuged at 10 000 g (10 min at 4°C). The precipitates were kept for subsequent protein determinations and resuspended in 0.1 mol l^-1NaOH (60°C for 1 h) prior to use. The acidic supernatant was transferred to a new tube and neutralized to pH 6 with 2.6 mol 1^-1 potassium bicarbonate. The resultant potassium perchlorate precipitate was discarded and the recovered supernatant was evaporated to dryness. The dried residue was resuspended in cAMP assay buffer and the concentration of AG cAMP was measured with a commercial [3H]cAMP assay kit (Diagnostic Products Corp., Los Angeles,CA, USA). For the pharmacological characterization of the AG dopamine receptor, AGs were quartered: one AG piece served as a control; the second piece was treated with 10 μmol l^-1 dopamine (positive control);the third piece was treated with dopamine receptor agonist or antagonist (10μmol l^-1); and the fourth piece was treated with dopamine plus agonist or antagonist. The AGs were extracted for cAMP as described above. Data were expressed as pmol cAMP mg^-1 AG protein.

Dopamine was previously demonstrated to induce the release of total protein from H. duryi AGs maintained in vitro; however, the identity of proteins released was not confirmed. To this end, we collected the saline surrounding the AG and identified the major protein secreted in vitroas the 66 kDa subunit of HdAGP (Fig. 1). The time course of basal protein secretion was measured for 60 min, followed by the addition of 10 μmol l^-1 dopamine for 20 min. Dopamine induced a rapid increase in the amount of protein released in vitro, as quantified by protein assay. The SDS-PAGE separation of the secreted material confirmed that the major protein secreted was the 66 kDa subunit of HdAGP (Fig. 1). After removal of the dopamine stimulus at 80 min, HdAGP secretion showed a precipitous decline and then maintained a secretory rate comparable with that observed before stimulation. The time course of HdAGP secretion as identified by SDS-PAGE was identical to the amount of total protein released as quantified by protein assay.

The specificity of the AG response to dopamine was tested by applying other known neurotransmitters found in molluscan nervous tissue to the AG and then measuring in vitro secretion of HdAGP. Serotonin, acetylcholine,γ-aminobutyric acid (GABA), norepinephrine, histamine, octopamine and l-glutamate were tested at 0.1, 1 and 10 μmol l^-1. With the exception of dopamine, none of the other neurotransmitters were capable of inducing the secretion of HdAGP from the AG(Table 1). A significant stimulation of HdAGP secretion was induced at a concentration of 1 and 10μmol l^-1 dopamine.

Since the existence of a dopamine D1-like receptor on the AG was indicated by previous protein secretion bioassays(Saleuddin et al., 2000) and since forskolin (an adenylate cyclase activator) is known to be a potent stimulator of HdAGP secretion (Morishita et al., 1998), we tested for the effect of dopamine on intracellular cAMP production. The time course of AG cAMP production was measured over a 20 min period after application of 10 μmol l^-1dopamine, 10 μmol l^-1 forskolin or both of these compounds together. Both dopamine and forskolin caused a significant increase in cAMP production by the AG between 5 and 20 min after their application. Furthermore, dopamine-stimulated cAMP levels were augmented in the presence of forskolin and increased linearly for at least 20 min(Fig. 2). Subsequent cAMP determinations were performed using a 10 min treatment unless otherwise indicated. The addition of dopamine (Fig. 3A) or forskolin (Fig. 3B) to AGs increased cAMP production in a dose-dependent manner. Significant elevation of cAMP levels were observed at 10 μmol l^-1 and 100 μmol l^-1 for both compounds; however,forskolin was a more potent activator of cAMP production than dopamine.

To assess the effect of various dopamine receptor agonists on AG cAMP production, D1-selective agonists (dihydrexidine, SKF81297 and 6,7-ADTN) and D2-selective agonists (bromocriptine and apomorphine) were tested. Treatment of AGs with dopamine (10 μmol l^-1) increased AG cAMP levels 3–6-fold above basal AG cAMP levels(Fig. 4). The D1-selective agonists dihydrexidine (Fig. 4A) and 6,7-ADTN (Fig. 4B) stimulated AG cAMP production significantly (3-fold and 5-fold, respectively), whereas SKF81297 had a modest (2-fold) non-significant stimulatory effect on AG cAMP production(Fig. 4C). No augmentation of cAMP production was detected when both dopamine and the D1-selective agonists were added to the AGs. By contrast, neither of the two D2-selective agonists apomorphine (Fig. 5A) nor bromocriptine (Fig. 5B) had a statistically significant effect on basal cAMP production. Bromocriptine had no effect on dopamine-stimulated cAMP production(Fig. 5B) but apomorphine unexpectedly caused a significant inhibition of dopamine-stimulated cAMP production (Fig. 5A).

Various D1-selective antagonists were tested for their ability to inhibit dopamine-stimulated cAMP production. At the concentrations (10 μmol l^-1) tested in this study, the benzazepines SCH23390(Fig. 6A) and SKF83566(Fig. 6B) suppressed dopamine-stimulated cAMP production by 62% and 48%, respectively. The compound flupenthixol had a slight inhibitory effect (22%) on dopamine-stimulated cAMP production but it was not statistically significant. None of the D1-selective antagonists affected basal cAMP levels significantly. The D2-selective antagonists chlorpromazine (Fig. 7A), eticlopride (Fig. 7B) or haloperidol (Fig. 7C) and the mixed D1/D2 antagonist butaclamol(Fig. 7D) had no effect on either basal or dopamine-stimulated cAMP production.

To confirm our previous results on the presence of a dopamine D1-like receptor, which mediates protein secretion, in the AG of H. duryi, we extended the number of D1-like receptor agonists and antagonists used. The D1-selective agonists SKF81297 and dihyrexidine were tested for their ability to induce protein secretion from the AGs whereas the D1-selective antagonists SKF83566 and flupenthixol were tested for their ability to inhibit dopamine-induced protein secretion. Both SKF81297(Fig. 8A) and dihyrexidine(Fig. 8B) were capable of inducing HdAGP secretion by AG explants, although dihydrexidine was the more potent stimulator. A concentration of 10 μmol l^-1 dihyrexidine stimulated HdAGP secretion 5-fold above control levels whereas 100 μmol l^-1 of SKF81297 was required to increase protein secretion 3.9-fold. By contrast, the D2-selective agonists apomorphine and bromocriptine had no effect on protein secretion (Fig. 9). Previous studies showed that SCH23390 (100 μmol l^-1), a D1-selective antagonist, was a potent inhibitor of dopamine-stimulated protein secretion(Saleuddin et al., 2000). In the present study, protein secretion was also inhibited by two D1-selective antagonists, SKF83566 (50 μmol l^-1) and flupenthixol (10 μmol l^-1). Both antagonists inhibited dopamine-stimulated HdAGP secretion by 50% (Fig. 10A)and 37% (Fig. 10B),respectively.

The involvement of PKA in mediating AG protein secretion was determined using the PKA antagonists Rp-cAMP (1 mmol l^-1) and H-89 (10 μmol l^-1). Both Rp-cAMP (Fig. 11A) and H-89 (Fig. 11B) did not significantly inhibit dopamine-stimulated protein secretion and neither inhibitor affected basal protein secretion.

The results of the present study support the existence of a specific dopamine D1-like receptor in the AG of the freshwater snail H. duryithat controls glycoprotein secretion. Addition of exogenous dopamine induced the secretion of the 66 kDasubunit of HdAGP, the major protein produced by the AG of H. duryi. Although other minor proteins are released when the AG is stimulated with dopamine, HdAGP is clearly the most prominent. Similar results were obtained when AGs were stimulated with either forskolin(Morishita et al., 1998) or brain peptide (Mukai et al.,2001a). Furthermore, the time course and profile of HdAGP secretion mirrors that of total protein secretion. Hence, total protein secretion by the AG consists primarily of HdAGP. Other neurotransmitters commonly found in the CNS of molluscs, including serotonin, acetylcholine,norepinephrine, GABA, histamine, octopamine and glutamate(Audesirk, 1985; Hetherington et al., 1994),had no effect on AG protein secretion, suggesting that the effect of dopamine on the AG was specific. Indeed, dopamine was the only neurotransmitter detected in significant quantity in the AG of H. duryi by HPLC(Kiehn et al., 2001).

Despite the widespread occurrence of dopamine in the peripheral tissues of molluscs, little information is available about the pharmacological properties of peripheral dopamine receptors, particularly in reproductive organs. In the salivary duct muscle of Helix pomatia, dopamine-induced contraction was inhibited by the D1-selective antagonists flupenthixol and fluphenazine,whereas the D1-selective agonist SKF38393 mimicked the effect of dopamine(Kiss et al., 2003). In the AG of B. glabrata, the D2-selective antagonist chlorpromazine inhibited dopamine-induced protein secretion(Santhanagopalan and Yoshino,2000) but it was without effect in H. duryi, as shown in an earlier report (Saleuddin et al.,2000) and in the present study. Previous studies have indicated the presence of a D1-like receptor that mediated AG protein secretion in H. duryi (Saleuddin et al.,2000). Since D1-like receptors are known to stimulate adenylate cyclase activity (Missale et al.,1998) and since forskolin is a potent stimulator of protein secretion (Morishita et al.,1998), we tested for the effect of dopamine on AG cAMP production. Addition of dopamine or forskolin to AG explants increased cAMP production in a time- and concentration-dependent manner. In the presence of forskolin,dopamine-stimulated cAMP production was enhanced because forskolin bypasses the receptor and directly activates the catalytic subunit of adenylate cyclase(Insel and Ostrom, 2003).

Using a number of different dopamine receptor agonists and antagonists it was possible to obtain a pharmacological characterization for the dopamine receptor in the AG of H. duryi. The D1-selective benzazepine agonist SKF81297 showed a slight stimulatory effect on AG cAMP production at a concentration of 10 μmol l^-1, indicating that it is a weak agonist at the AG dopamine receptor. This result is consistent with its effect on the secretion of HdAGP, where it was only half as potent as dopamine. In vertebrates, SKF81297 is considered to be a partial D1 agonist, i.e. it modestly activates adenylate cyclase when compared with dopamine(Andersen and Jansen, 1990). Saleuddin et al. (2000) found that SKF38393, another vertebrate D1-selective agonist, also had little effect on AG protein secretion, suggesting that benzazepines are relatively weak agonists on the AG dopamine receptor. In locust salivary glands, SKF38393 was also inactive in stimulating cAMP production(Ali and Orchard, 1994) and caused a weak hyperpolarization of the acinar cells as compared with dopamine(Keating and Orchard, 2001). In vertebrates, benzazepine analogues are effective dopamine receptor antagonists but display only limited effectiveness as cAMP agonists(Andersen and Jansen, 1990). Therefore, the H. duryi AG dopamine receptor is functionally similar to vertebrate D1-like receptors since its activation leads to cAMP formation but its structure is probably different because of its specificity to various dopamine receptor agonists.

The dopamine receptor agonist that was most effective in elevating AG cAMP levels was 6,7-ADTN. In vertebrates, this tetraline compound does not discriminate between D1-like and D2-like receptors. The activation characteristics of the AG dopamine receptor resembled that of a cloned D1-like receptor in Drosophila. This primordial Drosophila dopamine receptor also displayed poor affinity for benzazepines and was activated by the tetraline 6,7-ADTN (Sugamori et al.,1995). In the CNS of Lymnaea stagnalis, 6,7-ADTN mimicked the effect of dopamine at synapses between right pedal ganglion 1 (RPeD1) and its follower cells; however, this neuronal dopamine receptor was pharmacologically identified as a D2-like receptor(Magoski et al., 1995). The presence of a D2-like receptor that mediates dopamine-induced hyperpolarization in the growth hormone-producing light green cells of L. stagnalis has also been shown (de Vlieger et al., 1986; Werkman et al., 1987). Together, these results imply the existence of different dopamine receptor subtypes in molluscan neural and peripheral tissues. The other D1-selective agonist used in the present study was dihydrexidine. Dihydrexidine is a phenanthridine analogue that is considered to be a full agonist at vertebrate D1-like receptors(Brewster et al., 1990; Mottola et al., 1992);however, it was only modestly effective in elevating AG cAMP levels in H. duryi. Although dihyrexidine was not as effective as 6,7-ADTN in stimulating AG cAMP production, it was a more potent agonist in inducing HdAGP secretion when compared with 6,7-ADTN(Saleuddin et al., 2000). Dihydrexidine-stimulated HdAGP secretion was dose dependent and its activity was comparable with equivalent concentrations of dopamine. Finally, no additive or synergistic effects on AG cAMP production were observed with any of the agonists used in this study, indicating that these compounds probably act on a single receptor system.

The D2-selective agonists apomorphine and bromocriptine had no effect on AG protein secretion or basal cAMP levels. However, apomorphine exhibited an unusual effect by attenuating dopamine-stimulated cAMP production. The reason for this is unclear but it is possible that apomorphine occupies the same binding site on the receptor as dopamine, thereby reducing the latter's overall effect on AG cAMP formation.

The effect of various dopamine receptor antagonists was tested on AGs for their ability to inhibit dopamine-stimulated AG cAMP production and dopamine-induced HdAGP secretion. In the present study, the most effective inhibitor of dopamine-stimulated cAMP production was the benzazepine SCH23390,followed by SKF83566. In support of these findings, SCH23390 was also the most effective antagonist inhibiting dopamine-induced protein secretion in the AG of H. duryi (Saleuddin et al.,2000). Flupenthixol at a concentration of 10 μmol l^-1 suppressed dopamine-stimulated AG cAMP production only marginally but significantly inhibited dopamine-induced protein secretion. In the salivary duct muscle of the snail H. pomatia, SCH23390 was inactive in suppressing dopamine-induced contractions whereas flupenthixol was shown to be the most potent antagonist(Kiss et al., 2003). This indicates that the peripheral dopamine receptor in the salivary duct muscle of H. pomatia displays characteristics of a D1-like receptor and that its pharmacological properties are distinct from the D1-like receptor in the AG of H. duryi.

The AG dopamine receptor might represent a member of a novel class of dopamine receptors that is substantially different from those characterized in vertebrates and even other invertebrates. In support of this notion, the dopamine receptor in the corpus allatum of the tobacco hornworm, Manduca sexta, exhibits a pharmacological profile distinct from other insect dopamine receptors and has both D1- and D2-like properties(Granger et al., 2000). In the nematode Caenorhabditis elegans, a novel dopamine D1-like receptor that did not bind [³H]SCH23390 was recently cloned(Suo et al., 2002). In addition, the D1-like receptor from C. elegans also contained introns in the coding region, a feature not present in mammalian or previously characterized invertebrate D1-like receptors, providing further evidence that some invertebrate dopamine receptors represent a structurally distinct group. However, definitive characterization of the H. duryi AG dopamine receptor can only be achieved by molecular cloning and functional expression in a heterologous system.

To investigate intracellular signalling events further downstream of adenylate cyclase and cAMP, we tested for the involvement of PKA. In the vast majority of cAMP-mediated signalling, the primary effector regulated by cAMP is PKA. Cyclic AMP binds to the regulatory subunits of PKA, which in turn releases the catalytic subunits of PKA(Francis and Corbin, 1996). The activated catalytic subunits can then phosphorylate specific intracellular target proteins and mediate the appropriate physiological response. The PKA inhibitors Rp-cAMP and H-89 suppress PKA-mediated signalling in a number of invertebrates, including leeches (Ali et al., 1998), crustaceans(Locatelli et al., 2002),insects (Hazel et al., 2003)and molluscs (Malagoli et al.,2000). Our results demonstrate that dopamine-mediated secretion of HdAGP might be a PKA-independent process because inhibitors of both the regulatory (Rp-cAMP) and catalytic (H-89) subunits of PKA failed to inhibit protein secretion. PKA-independent signalling has recently become a recognized phenomenon due to the identification of novel cAMP-binding proteins, such as guanine exchange factors and cyclic nucleotide-gated channels (reviewed by Dremier et al., 2003; Kopperud et al., 2003), and the establishment of `cross-talk' among other signalling pathways(Cooper et al., 1995; Houslay and Milligan, 1997; Schwartz, 2001). However, it is possible that these PKA inhibitors were not able to fully penetrate into the glandular cells to exert their effects. Alternative approaches such as microinjection of PKA inhibitors into dissociated glandular cells combined with electrophysiological measurements are required to rule out this possibility.

In addition to dopamine, a putative neuropeptide from the CNS of H. duryi also appears to be involved in regulating HdAGP secretion through the cAMP signalling system (Morishita et al., 1998; Mukai et al., 2001). Whether these two molecules are co-released or perhaps affect each other's release is currently unknown. Identifying specific interactions between these two molecules can only be done after the H. duryi brain peptide is sequenced. The intracellular target of cAMP action in the AG of H. duryi is not known but it is possible that cAMP interacts with other intracellular signalling pathways such as Ca²⁺. In support of this, we have recently shown that, in addition to cAMP, the influx of Ca²⁺ into the AG cells is an important regulator of protein secretion(Kiehn et al., 2004). Structural elucidation of the brain peptide as well as the identification of downstream targets of cAMP and Ca²⁺ is in progress.

In summary, we have identified a specific receptor to the neurotransmitter dopamine in the AG of H. duryi that participates in HdAGP secretion. Based on the pharmacological profile obtained using dopamine receptor agonists and antagonists, it is concluded that the AG dopamine receptor displays functional characteristics of a vertebrate D1-like receptor because of the following: dopamine stimulates cAMP formation and protein secretion in a dose-dependent fashion; D1-selective agonists stimulate basal cAMP formation and induce protein secretion in isolated AGs; D2-selective agonists had no effect on cAMP production or protein secretion; D1-selective antagonists suppress dopamine-stimulated cAMP production and dopamine-induced protein secretion; and D2-selective antagonists had no inhibitory effect on dopamine-stimulated cAMP production. However, the AG dopamine receptor of H. duryi also displays some unique pharmacological characteristics in that its activity is stimulated by the non-selective agonist 6,7-ADTN and is attenuated by the D2-selective agonist apomorphine. Finally, dopamine-induced protein secretion might occur through a PKA-independent mechanism involving Ca²⁺ or, perhaps, some unidentified protein that is regulated by cAMP.

Symbols and abbreviations

 
• 6,7-ADTN

  (±)-2-amino-6,7-dihydroxy-1,2,3,4-tetrahydronapthalene

 
• AG

  albumen gland

 
• cAMP

  adenosine 3′,5′-cyclic monophosphate

 
• CNS

  central nervous system

 
• HdAGP

  Helisoma duryi albumen gland protein

 
• H-89

  N-(2-[p-bromocinnamylamino]ethyl)-5-isoquinolinesulfonamide

 
• IBMX

  3-isobutyl-1-methylxanthine

 
• PKA

  protein kinase A

 
• PVF

  perivitelline fluid

 
• Rp-cAMP

  Rp-adenosine 3′,5′-cyclic monophosphorothioate triethylammonium salt

 
• SCH23390

  R-(+)-7-chloro-8-hydroxy-3-methyl-1-phenyl-2,3,4,5-tetrahydo-1H-3-benzazepine hydrochloride

 
• SKF81297

  R-(+)-6-chloro-7,8-dihydroxy-1phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine hydrobromide

 
• SKF38393

  (±)-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine-7,8-diol hydrochloride

 
• SKF83566

  (±)-7-bromo-8-hydroxy-3methyl-2,3,4,5-tetrahydro-1H-3-benzazepine hydrochloride

The authors wish to thank Dr Barry Loughton for his comments on the manuscript. This work was supported by an NSERC Discovery Grant awarded to A.S.M.S. and a York University Contract Faculty Research Grant to S.T.M.