N-Acetylneuraminic Acid (NANA) Stimulates in situ cyclic AMP Production in Tentacles of Sea Anemone (Aiptasia pallida): Possible Role in Chemosensitization of Nematocyst Discharge

Cnidarians capture swimming prey by discharging nematocysts located on their fishing tentacles. Nematocyst discharge is one of the most dramatic and rapid of single-cell events. Discharge is completed in less than 3ms (Holstein and Tardent, 1984). Despite a large number of studies on nematocyst discharge, the mechanism and control of discharge are poorly understood. In situ nematocyst discharge requires sensory stimulation that involves intracellular and intercellular events, indicating that it is a well-orchestrated process. In general, a combination of appropriate prey-derived chemical and mechanical stimuli initiate maximum discharge (Pantin, 1942).

In sea anemones, the unit of nematocyst discharge is an ectodermal cellular receptor/effector complex called the cnidocyte/supporting cell complex (CSCC) (Thorington and Hessinger, 1988b; Watson and Hessinger, 1989a; Watson and Hessinger, 1989b). The CSCCs consist of individual cnidocytes surrounded by two or more supporting cells (Fig.1). Two general classes of chemoreceptor detect substances derived from prey and predispose a subpopulation of CSCCs (type B CSCCs) to discharge nematocysts in response to suitable mechanical stimuli. One class of receptor detects free and conjugated N-acetylated sugars and another detects amino compounds such as certain amino acids (Thorington and Hessinger, 1988a). Chemoreceptors for N-acetylated sugars (e.g. N-acetylneuraminic acid, NANA) are located at the apical plasma membrane of supporting cells (Watson and Hessinger, 1988), as are chemoreceptors for proline (Watson and Roberts, 1994).

Two classes of mechanoreceptor have been implicated in nematocyst discharge: contact-sensitive mechanoreceptors (CSMs; Thorington and Hessinger, 1988a), which presumably trigger discharge and are common to all CSCCs, and vibration-sensitive mechanoreceptors (VSMs; Watson and Hessinger, 1989a; Watson and Hessinger, 1989b), which are frequency-tuned by chemoreceptors and presumably detect the swimming movements of nearby prey (Watson and Hessinger, 1991), thereby preparing a subpopulation of CSCCs (type A CSCCs) for discharge.

Agents that increase intracellular cAMP levels (e.g. dibutyryl-cAMP, forskolin and cholera toxin) sensitize the CSMs of type B CSCCs and tune the VSMs of type A CSCCs to lower frequencies, as does NANA. Furthermore, NANA appears to stimulate adenylyl cyclase activity in supporting cells (Watson and Hessinger, 1992). In the present study, we demonstrate that NANA increases in situ cAMP levels in a dose-dependent manner in the ectodermal layer of tentacles from the sea anemone Aiptasia pallida, while having no effect on the cAMP content of tentacle endoderm. Furthermore, NANA-induced cAMP appears to activate endogenous cAMP-dependent protein kinase (PKA) in a NANA chemosensory signaling pathway that may sensitize nematocyst discharge.

All chemical reagents, including N-acetylneuraminic acid (NANA, type VI), 3-isobutyl-1-methylxanthine (IBMX), dimethyl sulfoxide and MgCl₂·6H₂O were obtained from Sigma Chemical Co. (St Louis, MO, USA). Kemptide was obtained from Calbiochem Corp. (San Diego, CA, USA) and γ-[³²P]-ATP from Amersham Life Sciences (Arlington Heights, IL, USA). The Kerckoff Marine Laboratory of the California Institute of Technology generously provided natural sea water at Corona del Mar, CA, USA. Encysted embryos of Artemia salina were purchased from San Francisco Bay Brand (Newark, CA, USA).

Monoclonal specimens of the sea anemone (Aiptasia pallida, Miami strain) were reared en-masse in flat-bottomed, glass trays and maintained individually in glass finger bowls containing natural sea water at 23±1°C. The animals were fed daily with freshly hatched brine shrimp nauplii and cleaned 4–6h after feeding. Anemones were kept on a 12h:12h light:dark daily photoperiod using white fluorescent lights at an intensity of 5.5×10³lx (Hessinger and Hessinger, 1981).

Anemones of the same size were selected and starved for 72h prior to experiments to maximize responsiveness (Thorington and Hessinger, 1988a). During each day of starvation, the natural sea water in which they were kept was replaced. During the last 48h of starvation, anemones were kept under continuous fluorescent light at 4.5×10³lx. This provided optimal uniformity of anemone behavior (Thorington and Hessinger, 1988a; Thorington and Hessinger, 1988b).

Immediately before each experiment, sea anemones were gently rinsed with natural sea water to remove soluble wastes, and the medium was then exchanged with the test solution. Test solutions of NANA at specified concentrations, containing 10⁻⁴moll⁻¹ IBMX, were prepared in natural, filtered (type 1, Whatman) sea water adjusted to pH7.65 with 1moll⁻¹ HC1 or 1moll⁻¹ NaOH. Artificial sea water (ASW) consisted of 423mmoll⁻¹ NaC1, 10mmoll⁻¹ KC1, 10mmoll⁻¹ CaC1₂, 24mmoll⁻¹ MgC1₂, 25mmoll⁻¹ MgSO₄ and 1.2mmoll⁻¹ NaHCO₃, adjusted to pH7.65. High-Mg²⁺/seawater (Mg-ASW) was prepared by diluting (1:1) ASW with 0.6moll⁻¹ MgCl₂ in distilled water.

Individual anemones were chemosensitized with NANA solutions containing 10⁻⁴moll⁻¹ IBMX for 15min after preincubation with 10⁻⁴moll⁻¹ IBMX alone for 20min in filtered, natural sea water at room temperature (22±2°C). Individual primary tentacles were rapidly excised and transferred using fine forceps onto the polished ends of aluminum rods (2.2cm in diameter) standing in liquid nitrogen. The snap-frozen tentacles were then placed individually in marked Eppendorf tubes (1.5-ml) and stored in racks at −80°C until extracted and analyzed for cAMP.

Each frozen tentacle was extracted individually by adding 1ml of ice-cold 1moll⁻¹ formic acid, pH2.0 (Payne and Ames, 1982). After incubation for 15min with formic acid on ice followed by brief vortexing, samples were centrifuged at 4000g for 5min to separate the still intact endoderm from the soluble extract. The extraction method inactivates tissue enzymes, such as phosphodiesterases, and solubilizes the ectoderm of freeze-thawed tentacles, as judged by enzymatic assay, light microscopy and the protein content of the supernatant medium. The endoderm remains intact and is surrounded by the denuded, intact, sac-like mesoglea. To measure endodermal cAMP content, the endoderm with the mesoglea was homogenized in 1ml of ice-cold 1moll⁻¹ formic acid using a pestle connected to a motor-driven mixer (Kontes Glass Co., Vineland, NJ, USA). Formic acid was removed from the supernatant and homogenate samples by drying overnight on a Speed-Vac (Savant Instruments Inc., Farmingdale, NY, USA). Each dried extract was then dissolved in 300μl of 50mmoll⁻¹ acetate buffer (pH5.8). Individual samples of 100μl were taken for separate measurements of cAMP and protein content.

The cAMP content of samples was measured using a commercial competitive radioimmunoassay kit (RPA 509, Amersham Life Sciences, Arlington Heights, IL, USA) using 5.9×10⁴Bq of adenosine 3′,5′-cyclic phosphoric acid 2′-O-succinyl-3-[¹²⁵I]iodotyrosine methyl ester and a gamma counter. The protein content of tissue extracts was measured spectrophotometrically by the enhanced BCA protein assay with bovine serum albumin as the standard (Pierce Chemical Co., Rockford, IL, USA).

[³H]cAMP (adenosine-8-[³H]3′,5′-cyclic monophosphate; Sigma Chemical Co.; 3.7×10⁷Bqml⁻¹) was used to measure the efficiency of recovery of cAMP from tentacle extraction. [³H]cAMP (3.7×10⁴Bq≈2000ctsmin⁻¹) was added to replicate single tentacles, extracted normally and then counted by liquid scintillation. As a control for percentage recovery, the same amount of [³H]cAMP, without extraction and without tentacle, was counted by liquid scintillation, and the percentage recovery was calculated [(ctsmin⁻¹ in extract/total ctsmin⁻¹)×100]. The percentage recovery of [³H]cAMP was consistently between 90 and 95%. Assayed tentacle samples were corrected for recovery from extraction in each experiment.

Measuring adhesive force and counting discharged nematocysts quantifies the in situ cnidocyte response to combined chemical and mechanical stimulation. These measurements in response to NANA sensitization were determined using previously described methods (Giebel et al., 1988; Thorington and Hessinger, 1988a; Thorington and Hessinger, 1988b). Briefly, test probes consisted of insect pins, the nylon head (0.8±0.01mm in diameter) of which was coated with 0.06mm of 30% (w/v) gelatine. They were stored at 4°C under 100% humidity and used within 24h. Single bowls containing individual sea anemones in various concentrations of NANA were raised by hand until the distal third of a tentacle contacted the probe attached to a force transducer (FT-03, Grass Instruments, Quincy, MA, USA) and a strip-chart recorder. The resistance springs were removed from the transducer to maximize sensitivity, and the transducer was calibrated gravimetrically. The bowl was lowered gently and smoothly after 5s of contact until the tentacle separated from the probe. The force necessary to separate the probe from the tentacle was recorded and expressed in units of newtons (N) since there is negligible acceleration (Miller, 1959).

After the probes had been used once to measure adhesive force, they were processed for counting nematocysts. Individual gelatine-coated probes were placed in separate flat-bottomed microtiter wells (Microtest 11, Falcon Plastics) each containing 40μl of 1% enzyme/detergent mixture (Trizyme; Amway Products, Ada, MI, USA). After incubation for 4h at room temperature, the probes were removed and the nematocysts released from the hydrolyzed-gelatine were visually counted using an inverted light microscope at a final magnification of 512×.

The cAMP-dependent protein kinase (PKA) activity was assayed both in cell-free tentacle supernatants and in broken cell preparations of tentacles. For the preparation of cell-free supernatant solutions, 10 freshly harvested tentacles were pooled and homogenized by mortar and pestle under liquid nitrogen. Ground tentacles were quickly transferred into an Eppendorf tube with 1ml of ice-cold extraction buffer (10mmoll⁻¹ sodium phosphate buffer, pH6.8, containing 5mmoll⁻¹ EDTA, 0.5mmoll⁻¹ IBMX, 75mmoll⁻¹ NaCl, 10mmoll⁻¹ NaVO₄ and 10μmoll⁻¹ leupeptin). The sample was sonicated for 2–4s on ice (Sonic Dismembrator, Fisher model 60) followed by centrifugation at 20000g for 20min at 4°C. The supernatant was removed and immediately used as the source of PKA activity.

For tentacle broken cell preparations, 20 tentacles were harvested and transferred by Pasteur pipette into a Dounce glass homogenizer kept on ice. After removing most of the sea water, 100μl of extraction buffer was added. Following gentle cell rupture by hand with 10 strokes on ice, the total volume was increased to 2ml. GTP and ATP at final concentrations of 0.1mmoll⁻¹ were added to the preparation, which was then divided into two 1ml fractions. An 18μl volume of 10⁻³moll⁻¹ NANA was added to one fraction to a final concentration of 1.8×10⁻⁵moll⁻¹, while the other fraction received 18μl of the extraction buffer. Broken cell preparations with and without NANA were incubated on ice for 15min, and samples were then taken for the assay of PKA.

PKA activity was measured according to modifications of the assay described by Cherrington et al. (Cherrington et al., 1976; Byus and Fletcher, 1982). PKA activity was evaluated with kemptide as the substrate, and assays were performed with and without cAMP, with and without PKA inhibitor (PKI), with and without kemptide and in the presence of various doses of kemptide. Positive controls consisted of purified catalytic subunit from bovine heart and negative controls of PKI purified from rabbit skeletal muscle (Byus and Fletcher, 1982).

The assay mixture contained 35μmoll⁻¹ kemptide, 0.2mmoll⁻¹ ATP, γ-[³²P]ATP (total specific activity 7.4×10⁶Bqpmol⁻¹) and 200mmoll⁻¹ Mes buffer, pH6.8, containing 40mmoll⁻¹ MgCl₂ and 28mmoll⁻¹ β-mercaptoethanol. Volumes of 40μl of either cell-free supernatant or tentacle broken cell preparation were added to reaction tubes containing 40μl of assay mixture to start the enzyme reaction. The reaction was incubated for 10min (unless indicated otherwise) at 30°C. The reaction was terminated by pipetting 40μl of the reaction volume onto Whatman P81 filters (2cm in diameter) and then immediately placing the filter papers in ice-cold 30% acetic acid. The assay filters were washed in cold 30% acetic acid at 4°C followed by a cold wash in 15% acetic acid and then 15% acetic acid at room temperature for 10min. Filter papers were then rinsed in 100% acetone for 5min, dried and counted in a liquid scintillation counter (3800 Beckman scintillation counter).

Protein concentrations were determined spectrophotometrically using the enhanced BCA protein assay (Wiechelman et al., 1988) with bovine serum albumin as the standard (Pierce Chemical Co., Rockford, IL, USA).

Results of cAMP experiments were determined by generating a standard curve with each experiment. Data were corrected for recovery from extraction of [³H]cAMP and then normalized as cAMP content per microgram of protein (fmolcAMPμg⁻¹protein). PKA activity was expressed as pmoles of phosphate transferred per minute. All results are given as means ± s.e.m. Statistical analyses of paired data were performed using Student’s t-test. Two-way analysis of variance (ANOVA) was used to compare multiple groups of means with repeated measures and post-hoc Fisher PLSD tests (Stat-View Software, Abacus Concepts Inc., Berkeley, CA, USA). The differences were considered to be significant at P<0.05. The total number of replicate samples (n) and the number of replicate experiments (N) are indicated in the figure legends. The number of sea anemones used in each experimental condition ranged between four and 20.

The basal cAMP content of individual A. pallida tentacles was measured following excision and rapid freezing. Tentacles were thawed by adding either 1ml of ice-cold 0.05moll⁻¹ acetate buffer (pH5.8) or 1moll⁻¹ formic acid (pH2.0). The soluble extracts were of ectodermal origin, while the insoluble tentacle residues consisted of intact endodermal layers encased within intact mesoglea. The ectodermal cAMP content of the formic-acid-extracted tentacles yielded approximately five times the cAMP content of the acetate-extracted tentacles (Fig.2; P<0.0001). The endodermal cAMP contents prepared by the two extraction methods, however, were indistinguishable (P=0.98). In view of the higher cAMP recovery from the ectoderm extract and because the formic-acid-extracted tentacles exhibited complete removal of the ectodermal layer, as determined by light microscopic examination, we elected to use formic acid extraction in all subsequent experiments.

The basal cAMP content of tentacles from animals treated with filtered, natural sea water containing 10⁻⁴moll⁻¹ IBMX, a non-specific inhibitor of cAMP phosphodiesterases, was approximately 30% higher than that of animals incubated in sea water alone (Fig.3; P=0.04). The cAMP content of tentacles from animals treated with 1.8×10⁻⁵moll⁻¹ NANA plus 10⁻⁴moll⁻¹ IBMX was approximately 35% higher than that of animals incubated in sea water containing NANA alone (Fig.3; P<0.002). Because IBMX enhanced cAMP levels, we pre-incubated anemones in 10⁻⁴moll⁻¹ IBMX in all subsequent experiments.

In normal ASW, 1.8×10⁻⁵moll⁻¹ NANA stimulates an almost twofold increase in in situ cAMP content of tentacle ectodermal layers (Fig.4; P<0.0001). We tested the effects of anesthetizing levels of Mg²⁺ on NANA-stimulated cAMP content because high levels of Mg²⁺, usually achieved by mixing equal volumes of ASW and 0.6moll⁻¹ MgCl₂, are commonly used to anesthetize and immobilize spontaneously contracting excised tentacles. In Mg-ASW containing 312mmoll⁻¹ MgCl₂ and 12.5mmoll⁻¹ MgSO₄, the NANA-stimulated increase in cAMP content was inhibited, with cAMP content not being significantly different from that of controls (Fig.4; P=0.17).

The effect of NANA concentration on in situ nematocyst discharge was biphasic and consisted of two regions (Fig.5): sensitization and desensitization. The sensitization region occurred at the lower tested concentrations of NANA. Half-maximal discharge (EC₅₀) occurred at approximately 10⁻⁷moll⁻¹ NANA. Maximum discharge was 3.4 times that of the seawater controls and occurred at 1.8×10⁻⁵moll⁻¹ NANA (EC₁₀₀). The effect of NANA on adhesive force (Fig.5) was also biphasic and coincided with the nematocyst discharge response curve. Maximum adhesive force was approximately 1.2 times that of the seawater controls and occurred at 1.8×10⁻⁵moll⁻¹ NANA (EC₁₀₀).

NANA also increased the cAMP content of tentacle ectodermal layers in a dose-dependent manner (Fig.5). The dose/response curve of NANA-stimulated cAMP content was biphasic and coincided with the effects of NANA on nematocyst discharge and adhesive force. Maximum cAMP levels in tentacle ectodermal layers of NANA-stimulated anemones occurred at 1.8×10⁻⁵moll⁻¹ NANA, approximately twice that of seawater-treated anemones (P<0.0001). In contrast, endodermal layers showed no significant change in cAMP content over the tested range of NANA concentrations. The mean cAMP content of endodermal layers from all NANA doses was 5.2±0.5fmolμg⁻¹protein (Fig.5), a value approximately one-seventh that of ectodermal layers from control anemones and one-fourteenth that of ectodermal layers from anemones exposed to 1.8×10⁻⁵moll⁻¹ NANA.

The ectodermal cAMP content of individual tentacles was measured from sea anemones incubated in 1.8×10⁻⁵moll⁻¹ NANA for different times (Fig.6). No significant increase in cAMP levels occurred during the first 3min of incubation with NANA. At 4min, the in situ cAMP content increased significantly (P<0.01), and between 4 and 10min the cAMP content did not change. The maximum increase in cAMP content occurred at 15min; approximately 1.8 times that averaged for 0–3min. For incubation times longer than 15min, the cAMP content declined steadily, reaching basal levels at 30min and declining to below basal levels by 45min (P<0.05).

Basal cAMP-dependent protein kinase (PKA) activity was measured in cell-free supernatants prepared from pooled, excised tentacles. The activity was measured both in the presence and in the absence of 0.1mmoll⁻¹ cAMP (Fig.7). Approximately 40% more protein kinase activity was measured in the presence of added cAMP than in its absence. In addition, 1:1 dilutions of supernatants were also assayed, yielding PKA activities approximately 40% of those of the undiluted supernatants.

To evaluate the effect of kemptide concentration on PKA activity, different concentrations of kemptide were tested with and without 0.1mmoll⁻¹ cAMP. The enzyme activity increased with increasing substrate concentration both with and without added cAMP. Activities with added cAMP were consistently higher than activities without added cAMP at the same substrate concentrations (Fig.8). The inhibitor PKI at 20nmoll⁻¹ totally inhibited PKA activity both in the presence and in the absence of exogenously added cAMP (data not shown).

Pooled tentacles had their cells gently ruptured as described in the Materials and methods section. Broken cell preparations were assayed for PKA activity. The enzyme activity in the absence of NANA was 14.08±0.46pmolphosphatemin⁻¹. In the presence of NANA, the activity averaged 15.57±0.33pmolphosphatemin⁻¹. The difference between the two groups was statistically significant (P=0.02) (Fig.9). PKA inhibitor (PKI) completely inhibited the enzyme activities, both with and without added NANA.

Cnidarians, such as sea anemones, jellyfish, hydra and corals, employ complex secretory products, termed nematocysts, for such functions as prey capture, defense and aggression. The ectodermal layer of sea anemone tentacles is armed with these stinging organelles. Housed within a specialized cell called the cnidocyte, the nematocyst consists of an intracellular capsule containing a highly folded, contiguous and eversible tubule. Eversion of the tubule, through which the venomous contents of the capsule are conveyed, is called nematocyst discharge. In this manner, the discharging nematocyst injects potent toxins into the prey (Hessinger, 1988).

In situ nematocyst discharge is initiated by appropriate stimulation of chemo- and mechanoreceptors (Pantin, 1942; Thorington and Hessinger, 1988a; Thorington and Hessinger, 1988b). Activated chemoreceptors are alleged to predispose contact-sensitive mechanoreceptors to trigger discharge in response to prey contact with the tentacle. Chemical and mechanical stimuli originating from prey also regulate subsequent feeding behavior and ingestion of prey by sea anemones. In the sea anemone Anthopleura elegantissima, concerted tentacle movements towards the mouth are controlled by asparagine, which is presumed to leak from nematocyst-inflicted wounds, while prey-derived reduced glutathione controls the ingestion of food into the mouth (Lindstedt, 1971).

In A. elegantissima, reduced glutathione activates adenylyl cyclase in oral disc membrane preparations (Gentleman and Mansour, 1974). In another cnidarian, the sea pansy Renilla koellikeri, adenylyl cyclase activity in membrane preparations is stimulated by GTP, GTP-γS, NaF and cholera toxin (Awad and Anctil, 1993). Adenylyl cyclase activity is also found in membrane preparations of hydra (Venturini et al., 1984). Furthermore, cAMP plays an important role in mediating the effect of head activator and in head regeneration in hydra (Fenger et al., 1994; Galliot et al., 1995).

In the sea anemone H. luciae, the membrane-permeant cAMP analogue dibutyryl-cAMP biphasically sensitizes in situ nematocyst discharge, as do forskolin, cholera toxin and caged GTP-γS. Furthermore, endogenous adenylyl cyclase activity in H. luciae is detectable cytochemically at the apical plasma membranes of the tentacle-supporting cells, but only in the presence of N-acetylated sugars, such as NANA (Watson and Hessinger, 1992). These findings indirectly implicate cAMP as the second messenger for activated supporting cell chemoreceptors involved in the NANA sensitizing pathway.

In our present experiments, we directly demonstrate that the chemosensitizer NANA increases the in situ cAMP content of A. pallida tentacle ectoderm. The effects of NANA concentration on cAMP content were biphasic and yielded a dose/response curve that coincided with those of NANA-sensitized nematocyst discharge and nematocyst-mediated adhesive force (Fig.5). These dose/response curves have similar regions of sensitization, EC₁₀₀ values and regions of desensitization at higher concentrations. The EC₁₀₀ values of NANA-sensitized nematocyst discharge and nematocyst-mediated adhesive force in the present study (Fig.5) confirm those published previously (Thorington and Hessinger, 1988a; Thorington and Hessinger, 1990; Thorington and Hessinger, 1998).

The relevance of NANA-stimulated increases in cAMP concentration is confirmed at the tissue level by the fact that the effect is confined to the ectoderm, where the CSCCs are located. NANA does not affect cAMP levels in the tentacle endoderm, where cAMP contents averaged 14% of ectodermal control levels and 7% of ectodermal NANA-stimulated content. That NANA optimally increases cAMP levels in the ectoderm only twofold is probably because the extracted tentacle ectoderm includes contributions from many cells that may not respond to externally applied NANA, including those of the neural plexus and the longitudinal muscle layer (Fig.1). The fact that the ectodermal cAMP content decreases at desensitizing levels of NANA suggests that desensitization is regulated between the chemoreceptor and cAMP in the NANA chemosensory signaling pathway. Thus, our data strongly suggest that NANA-stimulated cAMP functions as a second messenger in NANA-sensitized nematocyst discharge.

The present direct findings in A. pallida confirm and extend the indirect evidence previously provided by us that, in H. luciae, the NANA-chemosensitizing system of the anemone employs cAMP as a second messenger. Among higher animals, several examples of chemosensory pathways using cAMP as an intracellular mediator are known, including mammalian olfaction (Nakamura and Gold, 1987), sweet gustation (Avenet et al., 1988) and lobster olfaction (Boekhoff et al., 1994).

High levels of Mg²⁺ prepared by mixing equal volumes of ASW and 0.6moll⁻¹ MgCl₂ to make Mg-ASW have been used to anesthetize H. luciae and to immobilize excised tentacles for in situ studies of NANA-stimulated effects (Watson and Hessinger, 1991; Watson and Hessinger, 1992; Watson and Hessinger, 1994; Thibodeaux-Mire and Watson, 1993). We find that Mg-ASW blocks the in vivo NANA-stimulated increase in cAMP concentration in the tentacle ectoderm (Fig.4), and it also blocks NANA-sensitized nematocyst discharge in A. pallida and H. luciae tentacles (G. U. Thorington and D. A. Hessinger, unpublished data). However, in excised H. luciae tentacles immobilized with Mg-ASW, stereociliary bundles, which comprise the vibration-sensitive mechanoreceptors (VSMs) of type A CSCCs, are reported to elongate in response to NANA and to agents that commonly increase intracellular cAMP levels (Watson and Hessinger, 1992). Thus, either NANA-induced elongation of stereociliary bundles occurs by a cAMP-independent pathway or A. pallida and H. luciae differ with respect to the effects of high-Mg²⁺ ASW on NANA-stimulated increases in cAMP concentration. The practice of using Mg-ASW is also questionable because it is a hyperosmotic solution.

The time course of NANA-stimulated cAMP shows that NANA increases cAMP levels above control levels only after 3min. In contrast, unpublished experiments with gelatine-coated probes pre-equilibrated with solutions of NANA at various concentrations indicate that the sensitizing effect of NANA on nematocyst discharge can occur within a few seconds. In tentacles of H. luciae, chemosensitization of nematocyst discharge increases within 1min, reaches its highest level at 5min, and then gradually declines by 25min. The 25min minimum is then followed by an increase with a smaller maximum at 35min, which is followed by a lower minimum at 45min (Watson and Hessinger, 1989a). In experiments in which whole anemones are bathed in medium containing NANA, specimens of A. pallida, which are found subtidally in nature, require several minutes to recover from the disturbance of exchanging the medium, whereas H. luciae, which are found intertidally in nature, appear to be undisturbed by medium exchanges. Thus, it is possible that the physical disturbance caused by changing the medium in our time course experiments with A. pallida causes the NANA-stimulated cAMP pathway to be temporarily inhibited, favoring defensive withdrawal behavior over prey capture, whereas H. luciae are relatively unaffected.

After 45min, NANA-stimulated cAMP levels fall to a level below that of control ectodermal layers. This suggests that chemosensitization is subject to adaptation in response to prolonged NANA stimulation. Furthermore, the decline in cAMP levels to below control levels by prolonged exposure to NANA indicates that a significant proportion of the basal cAMP content of approximately 40pmolμg⁻¹protein is maintained by contributions from the NANA-responsive CSCCs.

Our data also indicate the presence of endogenous PKA activity in A. pallida tentacles. The monospecific PKA inhibitor protein PKI, or Walsh protein, blocks 98% of the measured kinase activity, indicating that the activity is due to PKA (Fig.9). This is further confirmed by the observation that added cAMP increases activity (Fig.7, Fig.8). In broken cell preparations of tentacles to which ATP, but not cAMP, has been added, NANA stimulates PKA activity by approximately 10% (P<0.02). This suggests that stimulated NANA chemoreceptors activate adenylyl cyclase to synthesize enough additional cAMP to stimulate PKA significantly above untreated control levels. That higher levels of NANA-stimulated PKA activity were not measured may be because of problems associated with the PKA assay and the preparation of the broken cells. For instance, diluting the broken cell preparations 40-fold for the PKA assay probably lowered levels of endogenously produced cAMP to values that less than optimally activate PKA. Furthermore, partial physical uncoupling of the membrane-associated chemosensory components of the NANA signaling pathway may have occurred during the preparation of the broken cells.

Our results document for the first time that physiologically relevant levels of NANA stimulate cAMP production in anemone tentacles. This effect of NANA is restricted to cells of the tentacle ectoderm and does not affect cAMP levels in the tentacle endoderm. Our finding that high levels of Mg²⁺ in the bathing sea water block the NANA-stimulated increase in cAMP concentration is consistent with the fact that high levels of Mg²⁺ block nematocyst discharge, but raises questions about the purported role of cAMP in NANA-induced elongation of the stereociliary bundles of vibration-sensitive hair bundles in type A CSCCs. In conclusion, our findings are consistent with cAMP acting as a second messenger in NANA-sensitized discharge of nematocysts from type B CSCCs.

We wish to thank J. A. Hessinger for help in culturing anenomes and Drs J. Buckholz and R. Alburqueque for helpful discussions. Funded in part by a BSRC grant.