To examine the distribution of nitric oxide (NO)-generative cells and NO-responsive cells in the tentacles and procerebral lobes (olfactory processing center) of terrestrial slugs, we applied NADPH diaphorase (NADPH-d) histochemistry and NO-induced cyclic GMP (cGMP)-like immunohistochemistry. We found that NADPH-d reactive cells/fibers and cGMP-like immunoreactive cells/fibers were different, but they were localized adjacent to each other, in both the tentacles and the procerebral lobes. Then, we measured the concentration of NO that was generated around the procerebral lobes using an NO sensitive electrode, when the olfactory nerve was electrically stimulated as a replacement for an odorant stimulus. Stimulation of the olfactory nerve evoked an increase in NO concentration at nanomolar levels, suggesting that binding of nanomolar concentrations of NO to the prosthetic heme group activates soluble guanylyl cyclase. Taken together with previously reported physiological data, our results, therefore, showed that the NO/cGMP pathways are involved in slug olfactory processing.
Nitric oxide (NO) is an unconventional neurotransmitter because it diffuses easily from its production site through cell membranes and moves readily through the brain in three dimensions (Sadamoto et al., 1998; Kobayashi et al., 2000a, b; Philippides et al., 2000). NO is generated by catalysis of nitric oxide synthase (NOS), which employs L-arginine, O2, and NADPH-derived electrons (Marletta et al., 1988). This indicates that formation of NO requires the presence of NADPH as a cofactor of NOS. Therefore a simple histochemical method to localize NOS-containing cells is to stain fixed nervous tissues for NADPH diaphorase (NADPHd) (Gelperin, 1994a; Sadamoto et al., 1998; Gelperin et al., 2000, 2001). The principle function of NO in many systems is to activate soluble guanylyl cyclase, resulting in production of cyclic GMP (cGMP), which can act on cGMP-activated protein kinase to phosphorylate downstream target proteins and thus evoke cellular responses (Bredt and Snyder, 1992; Garthwaite and Boulton, 1995; Aonuma et al., 2000, Aonuma and Newland, 2001; Bicker, 2001).
Many lines of evidence indicate that NO plays important roles in olfactory processing as seen in both invertebrates and vertebrates (Breer and Shepherd, 1993; Gelperin, 1994a, b, 1999; Robertson et al., 1994, 1995; Müller and Hildebrandt, 1995; Gelperin et al., 1996, 2000, 2001; Müller, 1996; Okere et al., 1996, Teyke, 1996; Hildebrand and Shepherd, 1997; Kendrick et al., 1997; Koh and Jacklet, 1999; Teyke and Gelperin, 1999; Moroz, 2001). Previous observations suggested that NO-mediated signaling in olfactory systems operates in parallel with conventional synaptic transmission to synchronize the neural activity. For example, field potential oscillations, thought to underlie synchrony of odor-responding interneurons, are dependent on NO levels in the olfactory system of the terrestrial slug Limax maximus (Gelperin, 1994a, b, 1999; Gelperin et al., 1996, 2000, 2001; Teyke and Gelperin, 1999). The olfactory processing center, the procerebral lobe, of slugs has a relatively simple, regular structure and displays robust spontaneous oscillations, offering many advantages for understanding the functional roles of synchronized neural activity for olfactory processing (Kleinfeld et al., 1994; Gervais et al., 1996; Ermentrout et al., 1998, 2001; Ito et al., 1999, 2001; Teyke et al., 1999; Cooke and Gelperin, 2001; Wang et al., 2001). Despite the fact that NO signaling and synchronized neural activity appear to be characteristic for olfactory processing in slugs (Gelperin, 1994a, b, 1999; Gelperin et al., 1996, 2000, 2001; Teyke and Gelperin, 1999), the detailed cellular pathways of NO/cGMP signaling remain to be examined with histological techniques.
In the present study, we applied NADPH-d histochemistry and NO-induced cGMP-like immunohistochemistry to examine the distribution of NO-generative cells and NO-responsive cells, respectively, in the tentacles and procerebral lobes of the terrestrial slug L. marginatus. Then, we measured the concentration of NO that is generated around the procerebral lobes using an NO sensitive electrode, when the olfactory nerve was electrically stimulated as a replacement for an odorant stimulus. We found that NADPH-d reactive cells/fibers and cGMP-like immunoreactive cells/fibers were different, but adjacent to each other, in both the tentacles and the central nervous system (CNS). The NO measurements showed that stimulation of the olfactory input pathway evoked an increase in NO concentration at nano-molar levels, suggesting that binding of nanomolar concentrations of NO to the prosthetic heme group activates soluble guanylyl cyclase.
MATERIALS AND METHODS
Adult specimens of the terrestrial slug L. marginatus weighing 0.5-1.5 g were used. Slugs were obtained from our laboratory colony or collected in the open field, maintained in our laboratory in a 14:10 h light:dark cycle at 19°C and fed on a paste of rat chow. Slugs were anesthetized by injection of 0.2-0.3 ml of an anesthetizing solution (see Ito et al., 1999 for composition) before dissection.
NADPH diaphorase histochemistry on whole mount preparations and cryostat-section preparations
The entire CNS together with the tentacles was removed from anesthetized slugs. For whole mount preparations, these tissues were fixed with 3.7% formaldehyde for 20 min at 4°C, rinsed twice in 0.1 M Tris-HCl (pH 8.0) with 0.2% Triton X-100 for 20 min at 4°C, incubated with a staining solution (0.2 mM β-NADPH (tetrasodium; Sigma, St. Louis, MO), 0.2 mM nitro blue tetrazolium (NBT, Sigma), and 0.2% Triton X-100 in 0.1 M Tris-HCl (pH 8.0)) overnight at 4°C, washed twice in distilled water for 10 min at room temperature, post-fixed in methanol and acetic acid (3:1) for 30 min, dehydrated in 100% methanol for 40 min, and cleared in cedarwood oil (Biken, Tokyo, Japan) for 10 min. The preparations were examined under a light microscope (MICROPHOTO-FXA; Nikon, Tokyo, Japan). To prepare cryostat-sections, the tissues were fixed with 4% paraformaldehyde in phosphate buffer (PB, pH 7.6) for 1 hr at 4°C, cryo-protected in 30% sucrose in PB overnight at 4°C, embedded in Tissue-Tek O.C.T. compound (Sakura Finetek, Torrance, CA), and sectioned with a cryostat (CM 3000; Leica, Nossloch, Germany) at a thickness of 20 μm. The sections were collected on silane-coated slides (Muto Pure Chemicals, Tokyo, Japan), air-dried at room temperature, rinsed three times in 0.1 M Tris-HCl (pH 8.0) for 15 min, incubated with the staining solution for 1 hr, washed in distilled water, and mounted in glycerol (Nakarai, Kyoto, Japan). The cryostat sections were also examined under a light microscope.
Cyclic GMP-like immunohistochemistry on cryostat-sectioned preparations
Staining methods were modified from Ott et al. (2000). The entire CNS together with the tentacles was removed from anesthetized slugs. The cerebral commissure was cut to facilitate penetration of reagents. For detection of NO-induced cGMP responses, the tissues were first incubated with 1 mM 3-isobutyl-1-methyxanthine (IBMX; Sigma) in Limax saline (see Ito et al., 1999 for composition) for 30 min at 4°C, and then they were incubated with 20 mM sodium nitroprusside (SNP; Wako Pure Chemical Industries, Osaka, Japan) and 1 mM IBMX in Limax saline for 15 min at room temperature. IBMX is an endogenous phosphodiesterase inhibitor and SNP is an NO-generative agent, respectively. The tissues were fixed with 4% paraformaldehyde in 0.1 M PB (pH 7.6) overnight at 4°C, cryo-protected with 30% sucrose in PB, embedded in Tissue-Tek O.C.T. compound, and sectioned with a cryostat at a thickness of 20 μm. The sections were collected on silane-coated slides, air-dried at room temperature, rinsed three times in 0.1 M PBS (pH 7.4) with 0.2% Triton X-100 for 15 min, and incubated with 5% normal donkey serum in 0.1 M PBS with 0.2% Triton X-100 for 2 hr at room temperature. The preparations were incubated with primary antibody (sheep anti-cGMP antiserum, diluted 1:20000 in 0.1 M PBS with 0.2% Triton X-100 and 5% donkey serum) overnight at 4°C. The antiserum was a gift from Dr. Jan De Vente (Rijksuniversiteit Limburg, The Netherlands, see Tanaka et al., 1997 for its specificity and characterization). They were washed in 0.1 M PBS with 0.2% Triton X-100 and incubated with horseradish peroxidase-conjugated secondary anti-sheep IgG antibody (diluted 1:500 in 0.1 M PBS with 0.2% Triton X-100 and 5% donkey serum; Jackson ImmunoRe-search Laboratories, West Grove, PA) for 2–3 hr at room temperature. After the development of staining with diaminobenzidine, secntions were washed in distilled water, dehydrated in an ethanol series, cleared in xylene, and mounted in Bioleit (Takaken, Tokyo, Japan). The preparations were examined under a light microscope. Control preparations were not incubated in IBMX or SNP.
Double staining for NADPH diaphorase and cyclic GMP on cryostat-sectioned preparations
Double staining methods were performed according to Bicker et al. (1996, 1997). The binding of the cGMP antiserum was first detected by immunohistochemistry and then the staining for NADPH-d was done subsequently by histochemistry on the same cryostat-section preparations. The exceptions from the above methods are as follows. The CNS and tentacles were fixed with a 4% paraformaldehyde solution for 2 hr. For the subsequent visualization of NADPH-d reactivity, the preparations were washed in 0.1 M PBS with 0.2% Triton X-100, and incubated with the staining solution for 30 min. They were finally mounted in glycerol.
Superior tentacles were clipped at the base from the bodies of anesthetized slugs with a pair of fine scissors. They were fixed with 1% paraformaldehyde and 1% gluteraldehyde in 0.1 M PB (pH 7.6) for 30 min at room temperature, rinsed three times in PB, and postfixed with 1% osmium tetroxide in distilled water for 1 hr at 4°C. The fixed preparations were then dehydrated in an ethanol series and embedded in Luft's epon embedding medium according to a standard method. The embedded preparations were cut horizontally with a diamond knife (Diatome ultra 45°; Diatome, Biel, Switzerland) on an ultramicrotome (Porter-Blum MT-1; Ivan Sorvall, Newtown, CT), and stained by uranyl acetate for 9 min and by lead citrate for 3 min at room temperature. The stained ultrathin sections were observed under a transmission electron microscope (JEM-100S; JEOL, Tokyo, Japan).
The methods for NO measurement were modified from Kobayashi et al. (2000a, b). NO concentration was measured with an NO meter (ISO-NO; World Precision Instruments, Sarasota, FL) using an NO electrode with a tip diameter of 100 μm (ami-NO100; Innovative Instruments, Tampa, FL). The NO meter was provided by courtesy of World Precision Instruments Japan. The specificity of the NO measurement methods has previously been determined by Weyrich et al. (1994) and Kobayashi et al. (2000b). In addition, we examined the selectivity of the NO electrode. Sensitivity for NO2− and NO3− was tested by applying KNO2 and KNO3 (1 mM each) several times to the NO electrode in HEPES-buffered saline (pH 7.6). Dopamine, acetylcholine (1 mM each), and serotonin (0.5 mM), all electrically active monoamines associated with activation of olfactory signal transduction, were also applied to the NO electrode in HEPES-buffered saline. Furthermore, we tested the NO electrode response to pH changes (pH 4.0, 7.0, and 9.0).
The CNS and the superior tentacular nerves, that project to the procerebral lobes in the CNS, were dissected from anesthetized slugs and maintained in a Sylgard-coated recording dish perfused with Limax saline. The tip of the NO electrode was placed on the ventral surface of the internal mass from which the outer ganglionic sheath had been removed. NO signals recorded by the NO meter were monitored with an oscilloscope, and recorded with a computer. We set a criterion level of response such that responses over the noise level could be clearly distinguished as due to nerve stimulation. To remove NO from the recording dish, an NO scavenger, 2-phenyl-4, 4, 5, 5-tetramethylimidazoline-1-oxyl-3-oxide (PTIO, 1 mM; Sigma) was applied in Limax saline. A 3 sec pulse of 10 μA current was delivered to the superior tentacular nerve end, ipsilateral to the recording region of procerebral lobe, by a suction electrode made of a syringe body and a polyvinylchloride tip whose diameter (ca. 200 μm) was adjusted to fit the nerve cross-section. This method of electrical stimulation was used as a replacement for the odorant stimulus because it could be readily controlled in both time and amplitude.
The NO electrode was calibrated by decomposition of an NO-generative agent, S-nitroso-N-acetylpenicillamine (SNAP; Dojindo, Kumamoto, Japan), according to the Instruction Manual of World Precision Instruments (Sarasota, FL). Although the theoretical concentration of generated NO should be equal to the final concentration of SNAP in the calibration vial, the practical conversion ratio of SNAP to NO is approximately 53.9% (Askew et al., 1995). We therefore took this ratio into account for the calibration. Briefly, 1.3 mg SNAP was first dissolved in 50 ml of 0.02% EDTA (pH 9.0). Then, this SNAP solution was gradually added to 0.1 M copper sul-fate in distilled water (pH 4.0) to generate NO at concentrations of 0, 2.2, 4.4, 6.6, 8.7, 10.9, 13.4, 22.0, 32.8, 43.5, 65.1, 87.0, and 109 nM. SNAP decomposes to NO and a disulfide byproduct in the presence of copper sulfate as a catalyst. The calibration was performed before and after NO measurements in tissue.
NADPH diaphorase histochemistry in the superior and inferior tentacles
When the whole mount preparations were examined, the tentacular ganglia had strong reactivity for NADPH-d (Fig. 1a). This evidence was supported by the control experiments in which the preparations made without application of NADPH did not show this reactivity (Fig. 1a). The NADPHd reactivity was observed in both the superior and inferior tentacular ganglia and there were no differences in the staining pattern between these two tentacles. To investigate the details of NADPH-d reactivity, we prepared cryostat sections. The NADPH-d reactivity was found throughout the tentacular ganglia and in some cells underneath the sensory epithelia (Fig. 1b). These details were the same in both the superior and inferior tentacles. The NADPH-d reactivity in sensory neurons underneath the sensory epithelia was localized in cytoplasm in the cell bodies (Fig. 1c). The dendrites of these neurons were confirmed to extend to the sensory epithelia. Two types of interneurons, monopolar and bipolar ones, were stained by NADPH-d as well, both of which were located between the sensory epithelia and the tentacular ganglia (Fig. 1d, e; see Ito et al., 2000 for identification of cells). The collar cells were identified around the surfaces of the tentacular ganglia (see Hatakeyama et al., 2001 for details) and some of them were stained by NADPH-d (Fig. 1f).
Cyclic GMP-like immunohistochemistry in the superior and inferior tentacles
From low-power views of cryostat-section preparations, we observed strong cGMP-like immunoreactivity in the tentacular ganglia, which were incubated in IBMX (phosphodiesterase inhibitor) and SNP (NO generating agent) (Fig. 2a). This immunoreactivity was eliminated when both IBMX and SNP were omitted during incubation (Fig. 2a). Control preparations without application of antiserum against cGMP had no signals, either (data not shown). The staining pattern by cGMP antibody was the same in the superior and inferior tentacles. Two subtypes of interneurons were stained between the sensory epithelia and the tentacular ganglia. One was a monopolar interneuron (Fig. 2b), and the other was a bipolar one (Fig. 2c; see Ito et al., 2000). In the tentacular ganglia, some stained fibers were located near the tentacular nerves (Fig. 2d). The immunoreactive cells on the lateral surfaces of the tentacular ganglia formed lines (Fig. 2e). Near these immunoreactive cells, the immunoreactive fibers were assembled (Fig. 2e). In the regions of tentacular ganglia near the digits, some immunoreactive spots (2–3 μm in diameter) forming lines were found (Fig. 2f). We could not identify whether they were glial cells or cut ends of fibers.
Double staining for NADPH diaphorase and cyclic GMP in the superior and inferior tentacles
In some cryostat-section preparations, strongly immunoreactive regions recognized by cGMP antibody were found in the tentacular ganglia, whereas those structures stained by NADPH-d were found as fibers in the tentacular nerves (Fig. 3a). These fibers projected to the procerebral lobes. In other preparations, strongly cGMP-like immunore-active fibers were also found to project to the procerebral lobes through the tentacular ganglia and the tentacular nerves (Fig. 3b). As far as we observed, the NADPH-d reactive fibers and the cGMP-like immunoreactive ones were completely different. The NADPH-d stained fibers in the neuropil passed among the cGMP-like immunoreactive cells that were located at the boundary between the tentacular ganglia and the tentacular nerves (Fig. 3c). These cGMP-like immunoreactive cells were identified as glia-like cells by electron microscopy (Fig. 3d).
NADPH diaphorase histochemistry and cyclic GMP-like immunohistochemistry in the central nervous system
First, we examined the NADPH-d reactivity in the CNS. Positively stained cell bodies were found in the major cell body regions in all the central ganglia, including the procerebral lobes (Fig. 4a). The NADPH-d reactive fibers were found consistently in some specific connectives (Fig. 4a). In particular, the fibers in the neuropil regions of the procerebral lobes were strongly reactive. The internal mass neuropil in the procerebral lobes exhibited by far the most intense NADPH-d reactivity of any region of the CNS (Fig. 4b). The terminal mass, which contains neurites of the procerebral neurons and afferents from olfactory receptors located in the superior and inferior tentacles, also exhibited strong NADPH-d reactivity (Fig. 4b). Tracts of NADPH-d reactive fibers extended between the terminal mass and the cell mass of the procerebral lobes, and some punctate staining was observed in the periphery of many cell bodies in the cell mass (Fig. 4c). The NADPH-d reactive cell bodies in the cell mass were more frequently observed in the basal and middle regions than in the apical region (data not shown).
Second, we used an immunohistochemical technique for detection of cGMP to determine which regions in the CNS are affected by NO. In the procerebral lobes, the terminal mass was strongly immunoreactive and some cells in the cell mass were slightly stained, whereas no staining was observed in the internal mass (Fig. 4d). When we did not apply IBMX and SNP, or when we excluded the cGMP antibody from the preparations, no detectable staining of cGMP was observed (Fig. 4d).
Third, to analyze the relationship between NO-generative cells and NO-responsive cells, we performed double-staining experiments. Cryostat sections were processed for cGMP-like immunohistochemistry with subsequent NADPHd histochemistry. In the procerebral lobes, the internal mass was stained by only NADPH-d histochemistry, whereas the terminal mass and the cell mass were stained by both NADPH-d histochemistry and cGMP-like immunohistochemistry (Fig. 4e). Some cells in the cell mass had intense cGMP-like immunoreactivity (Fig. 4f), and these cells were found to be different from the NADPH-d reactive cells by careful observations of double-stained and single-stained preparations. The finding that the NADPH-d reactive cells were different from the cGMP-like immunoreactive ones was also observed in other ganglia. For example, in the cerebral ganglia except the procerebral lobes and in the visceral ganglia, the NADPH-d reactive regions including cell bodies and fibers were adjacent to those showing cGMP-like immunore-activity but they were completely different (Fig. 4g, h).
Nitric oxide measurement around the procerebral lobes
Because we found that the internal mass had the most intense NADPH-d reactivity (Fig. 4b), we decided to place an NO electrode on the ventral surface of the internal mass to measure the NO responses induced by electrical stimulation of the superior tentacular nerve (Fig. 5a). We paid special attention to the NO calibration, and therefore the NO electrode was calibrated by the chemical titration method before and after each experiment (Fig. 5b). There were no significant differences between the calibration curves before and after experiments (n = 4 each, two-factor factorial ANOVA). The electrical stimulation of the superior tentacular nerve evoked a significant and slow increase in NO concentration around the internal mass (Fig. 5c). The NO response reached a peak 1-5 min after the electrical stimulation, and decreased to the basal level within 20-30 min after the stimulation. The estimated mean peak of NO concentration was 4.8 ± 1.7 nM (mean ± SEM, n = 7 different preparations). To remove NO in the recording dish, PTIO (NO scavenger) was administered in Limax saline. The increase in NO concentration, which was elicited by the electrical stimulation of the superior tentacular nerve, was consistently suppressed by PTIO (Fig. 5d). The NO electrode was confirmed not to be sensitive to NO2−, NO3−, dopamine, acetylcholine, serotonin, PTIO, or pH changes (pH 4.0, 7.0, 9.0) (data not shown).
In the present study, we stained the tentacles and CNS of L. marginatus by NADPH-d histochemistry and NO-induced cGMP-like immunohistochemistry to examine the distribution of NO-generative cells and NO-responsive cells, respectively. Our results showed that these two kinds of cells and fibers were localized adjacent to each other in the olfactory system, suggesting the involvement of NO/cGMP pathways in slug olfactory processing.
In the tentacles, some NADPH-d reactive cells appeared to be sensory cells underneath the sensory epithelia and the interneurons between the sensory epithelia and the tentacular ganglia (Fig. 1c-e), whereas cGMP-like immunoreactive cells were all interneurons (Fig. 2b, c). Because the distance between the sensory neurons and interneurons (tens of micrometer) is small enough for rapid diffusion of NO in vivo (Sadamoto et al., 1998; Philippides et al., 2000), our results suggest that NO generation from the sensory neurons and interneurons can increase the cGMP concentration in the adjacent interneurons. From our preliminary experiments with γ-aminobutyric acid (GABA) immunohistochemistry, some interneurons in the tentacles were found to contain GABA, similar to GABAergic cells observed in mouse olfactory bulb (Kishimoto et al., 1993). The relation between NO-generative neurons and GABAergic ones will be examined in Limax tentacles in future work. Because the collar cells showed NADPH-d reactivity (Fig. 1f), they may be involved in NO generation as well as in hormone secretion. In the tentacular ganglia, the NADPH-d reactivity was very intense (Figs. 1a, b, 3a, b, c), and the cGMP-like immunoreactive cells and fibers (Figs. 2d, e, f, 3a, b, c) were adjacent to, but different from, these NADPH-d reactive regions. In the tentacular nerves as well, the cGMP-like immunore-active fibers were adjacent to the NADPH-d reactive regions (Fig. 3a, b, c). These results suggest that cGMP as well as NO is responsible for the procerebral lobe responses to inputs from the tentacular nerves.
We should also note that NO can modulate cyclic nucleotide-gated channels directly (Broillet and Firestein, 1996, 1997) or may affect receptor adaptation (Breer and Shepherd, 1993). Therefore, NO produced by sensory neurons underneath the sensory epithelia and by interneurons between the sensory epithelia and the tentacular ganglia (Fig. 1b-d) may directly influence ion channels of the olfac-tory neurons in slugs. NO also affects growth and migration of neurons and glial cells (Cramer and Sur, 1999; Renteria and Constantine-Paton, 1999; Van Wagenen and Rehder, 1999; Chen et al., 2000). Several types of cells including glial cells and receptor cells are continuously generated in the snail tentacles (Chase and Rieling, 1986), and therefore the procerebral lobes continually receive new input fibers (Zakharov et al., 1998). By observing ultrathin sections, we also found single cell bodies that contained two nuclei in the central regions of neuropil of tentacular ganglia (data not shown). Because the whole neuropil of the tentacular ganglia showed strong NADPH-d reactivity (Fig. 1a, b), these dividing cells can be influenced by NO. Therefore, NO may play a role in the processes required to integrate these new synapses into the existing circuits in the procerebral lobes and the tentacular ganglia. The role of NO and cGMP in glial cells (Fig. 3c, d) is one of the liveliest topics in neurobiology (Willmott et al., 2000).
Both NADPH-d reactivity and cGMP-like immunoreactivity were observed in the terminal mass and the cell mass in the procerebral lobes (Fig. 4b, d, e), whereas only the NADPH-d staining was observed in the internal mass (Fig. 4b, e). The result that the procerebral lobes were stained well by NADPH-d histochemistry is in good agreement with observations in L. maximus (Gelperin, 1994; Gelperin et al., 2000). Because the tentacular nerves project only to the terminal mass (Kawahara et al., 1997; Watanabe et al., 1998), olfactory information transmitted from the tentacles is certainly modulated by NO/cGMP interaction in the olfactory central organ. In addition, the internal mass was strongly stained by NADPH-d histochemistry (Fig. 4b). Because the input pathway from the tentacles runs over the internal mass to terminate at the terminal mass (Kawahara et al., 1997), NO signals from the tentacles may be strengthened in the internal mass.
In the cell mass, we note three previous results and one new one: (1) NADPH-d immunoreactive cells are most likely to be nonbursting cells (Gelperin et al., 2000); (2) the inhibitory postsynaptic potential (IPSP) in the nonbursting cells is considered to be driven by multiple spikes from the bursting cells (Kleinfeld et al., 1994); (3) exogenous NO application induced an increase in the neural oscillatory frequency recorded in the procerebral lobes (Gelperin et al., 2000); and (4) both NADPH-d reactive cells and cGMP-like immunoreactive cells were found in the cell mass, and the cGMP-like immunoreactive cells appeared to be bursting cells because of their size and population. These results suggest that cGMP concentration in the bursting cells of the cell mass is increased by NO generated from nonbursting cells and that cGMP-mediated responses play a role in an increase in the neural oscillatory frequency, even though we should still examine other possibilities, for example NO generated from nonbursting cells regulates the activity in non-bursting cells or bursting cells without cGMP cascade.
Taken together, the NADPH-d reactive fibers and the cGMP-like immunoreactive ones in the terminal mass are expected to be the projections of nonbursting cells and the afferents from tentacular nerves, respectively. However, the staining pattern in the cell bodies and projections from the same cells is not usually consistent (e.g. in Kenyon cells in the mushroom body of honeybee; Müller, 1996). We therefore cannot rule out a possibility that the projections of non-bursting cells in the terminal mass are cGMP-like immunore-active.
The results that the NADPH-d regions were surrounded by NO-induced cGMP-like immunoreactive cells and fibers in the cerebral ganglia and the visceral ganglia as well (Fig. 4g, h) indicate that NO/cGMP pathways are also involved in other brain functions. The observation of whole mount preparations in which the connectives were stained by NADPHd histochemistry (Fig. 4a), as similarly observed in Helix (Cooke, 1994; Sánchez-Alvarez et al., 1994), indicates that the regions receiving output from the procerebral lobes may also employ NO/cGMP pathways.
We finally addressed the question: Do NO-generative regions stained by NADPH-d histochemistry release NO (Fig. 5)? We found an NO increase to a concentration of about 5 nM was recorded adjacent to the procerebral lobe after electrical stimulation of the superior tentacular nerve. The validity of the NO measurement methods was confirmed from electrode calibrations described in Materials and Methods and Results. The recorded value of the NO increase suggests that binding of nanomolar concentrations of NO to the prosthetic heme group activates soluble guanylyl cyclase.
In the slug olfactory system, the present results suggest that NO generated from the NO-generative cells by stimulation of an odor input pathway can increase the intracellular cGMP concentration in some of the cells and fibers adjacent to the NO-generative regions. Using this knowledge, we will next investigate the localization of soluble guanylyl cyclase and effects of cGMP on neural oscillations and odor learning.
We thank Dr. De Vente (Rijksuniversiteit Limburg, The Netherlands) for his generous gift of the anti-cGMP antibody and thank World Precision Instruments Japan for providing us with the NO meter. We are grateful to Dr. Oka (Keio University, Japan) for his recommendation to perform cGMP immunohistochemistry. This work was supported in part by Grants-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology of Japan and by grants from Yamada Science Foundation and Itoh Science Foundation to E.I. and by Research Fellowship (No. 05725) of the Japan Society for the Promotion of Science for Young Scientists to I.I.