- Email: [email protected]
Retrograde labelling of serotonergic projections onto the neuroendocrine bag cells of Aplysia
148
Ncurosciencc Lcltvr~ i2~ t i~)gl ) 14~, t ~: : 1991 Elsevier Scientific Publishers Ireland Ltd. i)~(14-3940/91/$ 03,5!i ADONIS 03(1439409100115H
NSL 07551
Retrograde labelling of serotonergic projections onto the neuroendocrine bag cells of Aplysia D u a n e R. McPherson* and James E. Blankenship Marine Biomedical Institute. University of Texas Medical Branch, Galveston, TX (U.S.A.) (Received 12 October 1990; Revised version received 26 October 1990; Accepted 8 November 1990) Key words." Aplysia; Neuroendocrine bag cell; Serotonin; Retrograde labelling; Rhodamine-conjugated latex microsphere Injection of rhodamine-conjugated latex microspheres into the right bag cell cluster of Aplysia brasiliana yielded retrograde labelling of a small number of cells in the cerebral and abdominal ganglia. Subsequent staining for serotonin immunoreactivity demonstrated consistent double-labelling in specific cerebral and abdominal ganglion serotonergic cells. The double-labelled populations were also stained in vivo by prior treatment with 5,7-dihydroxytryptamine. These retrogradely labelled serotonergic neurons may represent sources of inhibitory input to the neuroendocrine bag cells.
Egg-laying behavior in the marine mollusc Aplysia is initiated by episodes of long-lasting electrical activity in the neuroendocrine bag cells of the abdominal ganglion [16]. Bag cell firing causes the release of a family of peptides that influence the gonad and central neurons which presumably mediate the behavior [14]. The natural excitatory trigger for inducing bag cell activity is unknown, but most evidence points toward a source in the cerebral and/or pleural ganglia (see refs. 3, 5, 15). Less is known about inhibitory input to the bag cells; but inhibitory control is likely to be important, not only because it may contribute acutely to the cessation of a bag cell discharge and subsequent refractoriness [8, 9], but also because it may serve tonically to prevent a bag cell discharge and subsequent prolonged egg-laying activity that would be disruptive or inappropriately timed in the context of other ongoing behaviors. It has been reported that serotonin can shorten or inhibit a bag cell discharge, apparently by activating a tetraethyl ammonium (TEA)-sensitive potassium current [8, 9]. There is serotonin immunoreactivity in nerve endings around the bag cell bodies [6, 12], and there exists a high concentration of serotonin receptors in the region of bag cell neurites [4]. To determine the physiological significance of serotonin in the natural regulation of bag cell activity, it is necessary to identify the serotenergic neurons hypothesized to project *Present address: Department of Biology, Marquette University, Milwaukee, WI 53233, U.S.A. Correspondence: J.E. Blankenship, Marine Biomedical Institute, 200 University Blvd., Galveston, TX 77550, U.S.A.
onto the bag cells. We identify in this report two populations of serotonergic neurons which are retrogradely labelled by rhodamine-conjugated latex microspheres injected into a bag cell cluster. The experiments were performed on mature (100-200 g) specimens of Aplysia brasiliana, collected near Port Isabel, TX, and maintained at room temperature in artifical seawater (ASW) and fed dried seaweed daily. Rhodamine-conjugated latex microspheres (Luma-Fluor, New City, NY) were used as retrograde tracers [10]. For injection of the microspheres, animals were first anesthetized by injecting isotonic MgC12 into the foot sinus (2535 ml/100 g b. wt.). The abdominal ganglion was exposed through an incision in the body wall, and the microspheres, suspended in isotonic NaC1, were injected into the right bag cell cluster through a micropipette under visual control. All injections avoided the pleurovisceral connective and appeared to remain within the envelope of glia and connective tissue which surrounds each bag cell cluster. Incisions were sutured in two layers and any loss of body fluid was restored by injection of filtered ASW. Two animals were treated with 5,7-dihydroxytryptamine (5,7-DHT, Sigma) 3-4 weeks before rhodamine microsphere injection. For this, an injection of 5,7-DHT (1 mg/ml in ASW containing 0.5 mg/ml ascorbic acid) was made into the foot at a dose of 10 mg/ kg. In Aplysia brasiliana a single injection of 5,7-DHT is adequate to stain serotonergic neurons red in 4-5 weeks [7, 1 l, 17]. In 2-5 weeks after bead injection, animals were reanesthetized and the cerebral, pedal, pleural, and abdo-
149
minal ganglia of each animal were removed and fixed in 4% paraformaldehyde in 0.1 M phosphate buffer (PB, pH 7.2) for 5 12 h at 4°C. Ganglia were sectioned at 16 /~m on a cryostat ( - 18°C) and the sections were collected on gelatin-coated slides. Rabbit anti-serotonin antibodies (lncstar Corp., Stillwater, MN) were applied at 1:500 to 1:1000 dilution. Immunostaining began with a 1 2 h presoak in a diluting buffer (phosphate-buffered saline (PBS), 0.9% NaC1 in 0.1 M PB), 0.4% Triton X100, 3% normal goat serum). This was followed by incubation in primary antibody solution for 2-3 h, a wash in PBS for 1/2 h, incubation in secondary antibody solution for 1 h, and 1/2 h wash in PB. The secondary antibody was fluoreiscin isothiocyanate (FITC)-conjugated goat anti-rabbit (Cappel) applied at 1:50 dilution. Slides were coverslipped with PB/glycerol (1:6) including 0.5% n-propyl gallate to retard fading. All incubations were at room temperature. Specificity of the primary antibody was tested on sections from the nervous system of an untreated animal, fixed and sectioned in the same manner. The antibody was preabsorbed with 1 mg/ml of 5HT (creatinine sulfate complex) for 1 h. Preabsorption of the anti-5-HT antibody consistently reduced the intensity of specific staining, although slight fluorescence sometimes persisted in the neuropil and axons, and in known serotonergic cell bodies (e.g., the metacerebral giant cells). Similar results have been reported in the lobster nervous system [2]. Substitution of serum from a non-immune rabbit for the primary antibody resulted in no fluorescence above background. Selected sections were photographed on a Leitz photomicroscope with Tri-X Pan, Ektachrome 400, or T-MAX 3200 film. The cerebral, pedal, and pleural ganglia of seven animals were examined for the presence of rhodamine microspheres in serotonergic neurons. The punctate appearance of rhodamine microspheres in Aplysia neurons is similar to that in cats [10]. Two of the animals had been treated with 5,7-DHT, and were examined without immunocytochemistry. In one of these animals, a symmetrical pair of 5,7-DHT-stained neurons in the posterior part of the cerebral ganglion was labelled with rhodamine microspheres (Fig. 1A,B). To confirm that the neurons stained by 5,7-DHT are serotonergic, abdominal ganglion sections from the same animals were treated with anti-5-HT antibodies. Serotonin immunoreactivity was consistently observed in 5,7-DHT-stained cells (Fig. IC,D), supporting the assertion that 5,7-DHT treatment of molluscs selectively stains serotonergic cells [1]. Ganglia from the other five animals all showed 5-HT immunoreactivity (5-HT-IR) together with rhodamine microspheres in a similar posterior pair of neurons in the cerebral ganglion, and no animal showed rhodamine microspheres in other serotonergic neurons in the cere-
B
%
Fig. 1. Neuron in posterior region of right cerebral ganglion that showed dark reaction product (arrow, A) from previous exposure to 5,7-dihydroxytryptamine (5,7-DHT) also was labelled with rhodamine microspheres (B) injected into the right bag cell cluster. C: section of right caudal quadrant of dorsal surface of abdominal ganglion showing 3 neurons (RB cluster) stained dark after exposure to 5,7-DHT. D: same section with fluorescent illumination for FITC-labelled anti-serotonin antibody. E: whole mount of cerebral ganglion showing neurons that are serotonin-immunoreactive (peroxidase anti-peroxidase/diaminobenzidine reaction). The pair of cells indicated by arrows are the most caudal 5-HT-positive neurons in the cerebral ganglion, homologous to the neuron shown in A and B. Bar (A, B) 50 l~m; (C, D) 50 /~m; width of cerebral ganglion in E is approximalely 1.5 ram.
bral or pedal ganglia (there are no serotonergic cell bodies in the pleural ganglia). The double-labelled neurons were always the most posterior pair of serotonergic cerebral cell (Fig. I E). Rhodamine microsphere injections into the right bag cell cluster labelled relatively few neurons within the abdominal ganglion. We never observed labelling in the contralateral bag cell cluster. There was, however, a consistent population of 15-20 small cells just caudal to the left bag cell cluster, on either side of the PVC as it continues into the ganglion. Besides these cells there were
150
Fig. 2. Relationship of abdominal ganglion serotonergic neurons to bag cell clusters. A: fluorescence micrograph of section through left bag cell cluster showing staining of 5-HT-IR endings surrounding bag cell bodies. B: section of abdominal ganglion stained (with same antibody) as in A. Rostral is toward top. Stained are several small neurons near center of ganglion, below the ventral surface and slightly to the right of the midline. Note also the dense network of 5-HT-IR processes. C: higher magnification of same labelled 5-HT cells as in B. The neuron marked with arrow in B and C is also visible with rhodamine fluorescence in D, and contains microspheres retrogradely transported from the right bag cell cluster. E,F. Section from another animal showing similar anti-serotonin immunolabelling (E) of neuron in abdominal ganglion simil~r to that in C, and rhodamine microsphere labelling of same neuron (F). Bar (A, also for B) 50 pm; (D, also for C. E, F) 50/m~.
typically 10-15 small neurons near the midline at middle depth, a few small cells close to the right bag cell cluster, and occasionally several medium-to-large cells
( > 150/am) near the right dorsal surface. A small amount ( < 10%) of the dye leaked or diffused out of the bag cell envelope, so there may have been some uptake by neu-
151
tons which do not project onto the bag cells. When abdominal ganglion sections were examined for 5-HTIR (Fig. 2), double-labelled cells (rhodamine microspheres and 5-HT-IR) were found in 6 of 7 animals (several sections at the appropriate depth were lost in one animal). The number of cells ranged from 1 to 6, and all were small cells (20-30/tin diameter) in the region of the right ventral subsurface layer described by Kistler et al. [12]. Double-labelled neurons were present in the abdominal ganglion in both of the 5,7-DHT-treated animals. In one of these, double-labelling was only apparent after applying 5-HT immunocytochemistry, consistent with reports that 5,7-DHT, while specific to serotonergic neurons, labels fewer cells than immunocytochemistry Ill]. Thus, rhodamine microsphere injection into the right bag cell cluster consistently labels a posterior pair of serotonergic neurons in the cerebral ganglion and a polulation of small serotonergic neurons within the abdominal ganglion. Interestingly, the cerebral 5-HT cells appear to correspond to a pair of B-cluster neurons (LCB1 and RCBI) described by Mackey et al. [13] as serotonergic facilitator neurons projecting to siphon sensory cells in the abdominal ganglion. The small number of cerebral ganglion neurons labelled by our rhodamine microsphere injections (range: 2-15) suggests that our experiments labelled bag cell-specific inputs rather than general cerebral-to-abdominal ganglion projections, For example, Mackey et al. [13] reported that several cerebral B-cluster neurons on each side project to the abdominal ganglion, but we typically saw no other rhodamine microsphere-labelled neurons close to the labelled serotonergic cells. The same authors also reported that several serotonergic cells in the pedal ganglia were labelled by backfills of the pleurovisceral connective; we never observed such labelling after dye injection into a bag cell cluster. The simplest explanation is that cells LCB1 and RCBI project to the bag cells as well as to siphon sensory neurons. If 5-HT facilitates reflex withdrawal from noxious stimuli and also inhibits a bag-cell discharge, it is behaviorally consistent for one population of neurons to be involved with both functions: egg laying would be prevented when the animal is sensitized to potentially harmful stimuli. Regarding the double-labelled cells in the abdominal ganglion, we have no basis for behavioral speculation, but their consistent labelling, and the absence of rhodamine microsphere labelling in any other abdominal serotonergic neurons, suggest that they represent a specific projection onto the bag cells. Neurons in this population and the LCB1 and RCBI neurons were labelled in vivo by 5,7-DHT, so it is feasible to explore the physiological effects of these small neurons by staining them beforehand.
We thank Dr. R.D. Hawkins for comments on this manuscript. Supported by NSF BBS 8711368, NIH NS 27314, 11255 and 07185. I Balaban, P.M., Zakharow, J.S. and Matz, V.N., Method of vital selective staining of serotonergic nerve cells by 5,7-dihydroxytryptamine, Dokl. Akad. Nauk USSR, 282 (1985) 735 738 (in Russian). 2 Beltz, B.S. and Kravitz, E.A., Mapping of serotonin-like immunoreactivity in the lobster nervous system, J. Neurosci., 3 (1983) 585 602. 3 Brown, R.O., Pulst, S.M. and Mayeri, E., Neuroendocrine bag cells of Aplysia are activated by bag cell peptide-containing neurons in the pleural ganglia, J. Neurophysiol., 61 (1989) 1142 1152. 4 Drummond, A.H., Bucher, F. and Levitan, I.B., Distribution of serotonin and dopamine receptors in Aplvsia tissue: analysis by [~H]LSD binding and adenylate cyclase stimulation, Brain Res., 184(1980) 163 177. 5 Ferguson, G.P., ter Maat, A. and Pinsker, H.M., Egg laying in Aplysia. II. Organization of central and peripheral pathways for initiating neurosecretory activity and behavioral patterns, J. Comp. Physiol., A 164 (1989) 849-857. 6 Hopkins, W.E., Stone, L.S., Rothman, B.S., Basbaum, A.I. and Mayeri, E., Egg-laying hormone, leucine-enkephalin and serotonin immunoreactivity in the abdominal ganglion of ApO,sia, Soc. Neurosci. Abstr., 8 (1982) 587. 7 Jahan-Parwar, B., S.-Rosza, K., Salanki, J., Evans, M.L. and Carpenter, D.O., In vivo labelling of serotonin-containing neurons by 5,7-dihydroxytryptamine in Aplvsia, Brain Res., 426 (1987) 173 178. 8 Jennings, K.R., Host, J.J., Kaczmarek, L.K. and Strumwasser, F., Serotonergic inhibition of afterdischarge in peptidergic bag cells, J. Neurobiol., 12 (1981) 579-590. 9 Kaczmarek, L.K., Jennings, K. and Strumwasser, F., Neurotransmitter modulation, phosphodiesterase inhibitor effects, and cyclic AMP correlates of afterdischarge in peptidergic neurons, Proc. Natl. Acad. Sci. U.S.A. 75 (1978) 5200 5204. 10 Katz, L.C., Burkhalter, A. and Dreyer, W.J., Fluorescent latex microspheres as a retrograde neuronal marker for in vivo and in vitro studies of visual cortex, Nature, 310 (1984) 498 500. 11 Kemenes, G., Elekes, K., Hiripi, L. and Benjamin, P.R., A comparison of four techniques for mapping the distribution of serotonin and serotonin-containing neurons in fixed and living ganglia of the snail Lymnaea, J. Neurocytol., 18 (1989) 193 208. 12 Kistler, H.B., Jr., Hawkins, R.D., Koester, J., Steinbusch, H.W.M., Kandel, E.R. and Schwartz, J.H. Distribution of serotonin-immunoreactive cell bodies and processes in the abdominal ganglion of mature Aplysia, J. Neurosci., 5 (1985) 72 80. 13 Mackey, S.L., Kandel, E.R. and Hawkins, R.D. Identified serotonergic neurons LCBI and RCBI in the cerebral ganglia of Aplvsia produce presynaptic facilitation of siphon sensory neurons, J. Neurosci., 9 (1989) 4227~4235. 14 Mayeri, E. and Rothman, B.S. Neuropeptides and the control ot" egg-laying behavior in Aplvsia. In A.I. Selverston (Ed.), Model Neural Networks and Behavior, Plenum, New York, 1985, pp. 285 -30 I. 15 Painter, S.D., Rock, M.K., Nagle, G.T. and Blankenship, J.E., Peptide B induction of bag-cell activity in Aplysia: localization of sites of action to the cerebral and pleural ganglia, J. Neurobiol., 8 (1988) 695-706. 16 Pinsker, H.M. and Dudek, F.E., Bag cell control of egg laying in freely behaving Aplvsia, Science, 197 (I977)490 493. 17 S.-Rosza, K., Hernadi, L. and Kemenes, G., Selective in viw~ labelling of serotonergic neurones by 5,6-dihydroxytryptamine in the snail Helix pomatia g., Comp. Biochem. Physiol., 85 (1986) 419 425.
Ncurosciencc Lcltvr~ i2~ t i~)gl ) 14~, t ~: : 1991 Elsevier Scientific Publishers Ireland Ltd. i)~(14-3940/91/$ 03,5!i ADONIS 03(1439409100115H
NSL 07551
Retrograde labelling of serotonergic projections onto the neuroendocrine bag cells of Aplysia D u a n e R. McPherson* and James E. Blankenship Marine Biomedical Institute. University of Texas Medical Branch, Galveston, TX (U.S.A.) (Received 12 October 1990; Revised version received 26 October 1990; Accepted 8 November 1990) Key words." Aplysia; Neuroendocrine bag cell; Serotonin; Retrograde labelling; Rhodamine-conjugated latex microsphere Injection of rhodamine-conjugated latex microspheres into the right bag cell cluster of Aplysia brasiliana yielded retrograde labelling of a small number of cells in the cerebral and abdominal ganglia. Subsequent staining for serotonin immunoreactivity demonstrated consistent double-labelling in specific cerebral and abdominal ganglion serotonergic cells. The double-labelled populations were also stained in vivo by prior treatment with 5,7-dihydroxytryptamine. These retrogradely labelled serotonergic neurons may represent sources of inhibitory input to the neuroendocrine bag cells.
Egg-laying behavior in the marine mollusc Aplysia is initiated by episodes of long-lasting electrical activity in the neuroendocrine bag cells of the abdominal ganglion [16]. Bag cell firing causes the release of a family of peptides that influence the gonad and central neurons which presumably mediate the behavior [14]. The natural excitatory trigger for inducing bag cell activity is unknown, but most evidence points toward a source in the cerebral and/or pleural ganglia (see refs. 3, 5, 15). Less is known about inhibitory input to the bag cells; but inhibitory control is likely to be important, not only because it may contribute acutely to the cessation of a bag cell discharge and subsequent refractoriness [8, 9], but also because it may serve tonically to prevent a bag cell discharge and subsequent prolonged egg-laying activity that would be disruptive or inappropriately timed in the context of other ongoing behaviors. It has been reported that serotonin can shorten or inhibit a bag cell discharge, apparently by activating a tetraethyl ammonium (TEA)-sensitive potassium current [8, 9]. There is serotonin immunoreactivity in nerve endings around the bag cell bodies [6, 12], and there exists a high concentration of serotonin receptors in the region of bag cell neurites [4]. To determine the physiological significance of serotonin in the natural regulation of bag cell activity, it is necessary to identify the serotenergic neurons hypothesized to project *Present address: Department of Biology, Marquette University, Milwaukee, WI 53233, U.S.A. Correspondence: J.E. Blankenship, Marine Biomedical Institute, 200 University Blvd., Galveston, TX 77550, U.S.A.
onto the bag cells. We identify in this report two populations of serotonergic neurons which are retrogradely labelled by rhodamine-conjugated latex microspheres injected into a bag cell cluster. The experiments were performed on mature (100-200 g) specimens of Aplysia brasiliana, collected near Port Isabel, TX, and maintained at room temperature in artifical seawater (ASW) and fed dried seaweed daily. Rhodamine-conjugated latex microspheres (Luma-Fluor, New City, NY) were used as retrograde tracers [10]. For injection of the microspheres, animals were first anesthetized by injecting isotonic MgC12 into the foot sinus (2535 ml/100 g b. wt.). The abdominal ganglion was exposed through an incision in the body wall, and the microspheres, suspended in isotonic NaC1, were injected into the right bag cell cluster through a micropipette under visual control. All injections avoided the pleurovisceral connective and appeared to remain within the envelope of glia and connective tissue which surrounds each bag cell cluster. Incisions were sutured in two layers and any loss of body fluid was restored by injection of filtered ASW. Two animals were treated with 5,7-dihydroxytryptamine (5,7-DHT, Sigma) 3-4 weeks before rhodamine microsphere injection. For this, an injection of 5,7-DHT (1 mg/ml in ASW containing 0.5 mg/ml ascorbic acid) was made into the foot at a dose of 10 mg/ kg. In Aplysia brasiliana a single injection of 5,7-DHT is adequate to stain serotonergic neurons red in 4-5 weeks [7, 1 l, 17]. In 2-5 weeks after bead injection, animals were reanesthetized and the cerebral, pedal, pleural, and abdo-
149
minal ganglia of each animal were removed and fixed in 4% paraformaldehyde in 0.1 M phosphate buffer (PB, pH 7.2) for 5 12 h at 4°C. Ganglia were sectioned at 16 /~m on a cryostat ( - 18°C) and the sections were collected on gelatin-coated slides. Rabbit anti-serotonin antibodies (lncstar Corp., Stillwater, MN) were applied at 1:500 to 1:1000 dilution. Immunostaining began with a 1 2 h presoak in a diluting buffer (phosphate-buffered saline (PBS), 0.9% NaC1 in 0.1 M PB), 0.4% Triton X100, 3% normal goat serum). This was followed by incubation in primary antibody solution for 2-3 h, a wash in PBS for 1/2 h, incubation in secondary antibody solution for 1 h, and 1/2 h wash in PB. The secondary antibody was fluoreiscin isothiocyanate (FITC)-conjugated goat anti-rabbit (Cappel) applied at 1:50 dilution. Slides were coverslipped with PB/glycerol (1:6) including 0.5% n-propyl gallate to retard fading. All incubations were at room temperature. Specificity of the primary antibody was tested on sections from the nervous system of an untreated animal, fixed and sectioned in the same manner. The antibody was preabsorbed with 1 mg/ml of 5HT (creatinine sulfate complex) for 1 h. Preabsorption of the anti-5-HT antibody consistently reduced the intensity of specific staining, although slight fluorescence sometimes persisted in the neuropil and axons, and in known serotonergic cell bodies (e.g., the metacerebral giant cells). Similar results have been reported in the lobster nervous system [2]. Substitution of serum from a non-immune rabbit for the primary antibody resulted in no fluorescence above background. Selected sections were photographed on a Leitz photomicroscope with Tri-X Pan, Ektachrome 400, or T-MAX 3200 film. The cerebral, pedal, and pleural ganglia of seven animals were examined for the presence of rhodamine microspheres in serotonergic neurons. The punctate appearance of rhodamine microspheres in Aplysia neurons is similar to that in cats [10]. Two of the animals had been treated with 5,7-DHT, and were examined without immunocytochemistry. In one of these animals, a symmetrical pair of 5,7-DHT-stained neurons in the posterior part of the cerebral ganglion was labelled with rhodamine microspheres (Fig. 1A,B). To confirm that the neurons stained by 5,7-DHT are serotonergic, abdominal ganglion sections from the same animals were treated with anti-5-HT antibodies. Serotonin immunoreactivity was consistently observed in 5,7-DHT-stained cells (Fig. IC,D), supporting the assertion that 5,7-DHT treatment of molluscs selectively stains serotonergic cells [1]. Ganglia from the other five animals all showed 5-HT immunoreactivity (5-HT-IR) together with rhodamine microspheres in a similar posterior pair of neurons in the cerebral ganglion, and no animal showed rhodamine microspheres in other serotonergic neurons in the cere-
B
%
Fig. 1. Neuron in posterior region of right cerebral ganglion that showed dark reaction product (arrow, A) from previous exposure to 5,7-dihydroxytryptamine (5,7-DHT) also was labelled with rhodamine microspheres (B) injected into the right bag cell cluster. C: section of right caudal quadrant of dorsal surface of abdominal ganglion showing 3 neurons (RB cluster) stained dark after exposure to 5,7-DHT. D: same section with fluorescent illumination for FITC-labelled anti-serotonin antibody. E: whole mount of cerebral ganglion showing neurons that are serotonin-immunoreactive (peroxidase anti-peroxidase/diaminobenzidine reaction). The pair of cells indicated by arrows are the most caudal 5-HT-positive neurons in the cerebral ganglion, homologous to the neuron shown in A and B. Bar (A, B) 50 l~m; (C, D) 50 /~m; width of cerebral ganglion in E is approximalely 1.5 ram.
bral or pedal ganglia (there are no serotonergic cell bodies in the pleural ganglia). The double-labelled neurons were always the most posterior pair of serotonergic cerebral cell (Fig. I E). Rhodamine microsphere injections into the right bag cell cluster labelled relatively few neurons within the abdominal ganglion. We never observed labelling in the contralateral bag cell cluster. There was, however, a consistent population of 15-20 small cells just caudal to the left bag cell cluster, on either side of the PVC as it continues into the ganglion. Besides these cells there were
150
Fig. 2. Relationship of abdominal ganglion serotonergic neurons to bag cell clusters. A: fluorescence micrograph of section through left bag cell cluster showing staining of 5-HT-IR endings surrounding bag cell bodies. B: section of abdominal ganglion stained (with same antibody) as in A. Rostral is toward top. Stained are several small neurons near center of ganglion, below the ventral surface and slightly to the right of the midline. Note also the dense network of 5-HT-IR processes. C: higher magnification of same labelled 5-HT cells as in B. The neuron marked with arrow in B and C is also visible with rhodamine fluorescence in D, and contains microspheres retrogradely transported from the right bag cell cluster. E,F. Section from another animal showing similar anti-serotonin immunolabelling (E) of neuron in abdominal ganglion simil~r to that in C, and rhodamine microsphere labelling of same neuron (F). Bar (A, also for B) 50 pm; (D, also for C. E, F) 50/m~.
typically 10-15 small neurons near the midline at middle depth, a few small cells close to the right bag cell cluster, and occasionally several medium-to-large cells
( > 150/am) near the right dorsal surface. A small amount ( < 10%) of the dye leaked or diffused out of the bag cell envelope, so there may have been some uptake by neu-
151
tons which do not project onto the bag cells. When abdominal ganglion sections were examined for 5-HTIR (Fig. 2), double-labelled cells (rhodamine microspheres and 5-HT-IR) were found in 6 of 7 animals (several sections at the appropriate depth were lost in one animal). The number of cells ranged from 1 to 6, and all were small cells (20-30/tin diameter) in the region of the right ventral subsurface layer described by Kistler et al. [12]. Double-labelled neurons were present in the abdominal ganglion in both of the 5,7-DHT-treated animals. In one of these, double-labelling was only apparent after applying 5-HT immunocytochemistry, consistent with reports that 5,7-DHT, while specific to serotonergic neurons, labels fewer cells than immunocytochemistry Ill]. Thus, rhodamine microsphere injection into the right bag cell cluster consistently labels a posterior pair of serotonergic neurons in the cerebral ganglion and a polulation of small serotonergic neurons within the abdominal ganglion. Interestingly, the cerebral 5-HT cells appear to correspond to a pair of B-cluster neurons (LCB1 and RCBI) described by Mackey et al. [13] as serotonergic facilitator neurons projecting to siphon sensory cells in the abdominal ganglion. The small number of cerebral ganglion neurons labelled by our rhodamine microsphere injections (range: 2-15) suggests that our experiments labelled bag cell-specific inputs rather than general cerebral-to-abdominal ganglion projections, For example, Mackey et al. [13] reported that several cerebral B-cluster neurons on each side project to the abdominal ganglion, but we typically saw no other rhodamine microsphere-labelled neurons close to the labelled serotonergic cells. The same authors also reported that several serotonergic cells in the pedal ganglia were labelled by backfills of the pleurovisceral connective; we never observed such labelling after dye injection into a bag cell cluster. The simplest explanation is that cells LCB1 and RCBI project to the bag cells as well as to siphon sensory neurons. If 5-HT facilitates reflex withdrawal from noxious stimuli and also inhibits a bag-cell discharge, it is behaviorally consistent for one population of neurons to be involved with both functions: egg laying would be prevented when the animal is sensitized to potentially harmful stimuli. Regarding the double-labelled cells in the abdominal ganglion, we have no basis for behavioral speculation, but their consistent labelling, and the absence of rhodamine microsphere labelling in any other abdominal serotonergic neurons, suggest that they represent a specific projection onto the bag cells. Neurons in this population and the LCB1 and RCBI neurons were labelled in vivo by 5,7-DHT, so it is feasible to explore the physiological effects of these small neurons by staining them beforehand.
We thank Dr. R.D. Hawkins for comments on this manuscript. Supported by NSF BBS 8711368, NIH NS 27314, 11255 and 07185. I Balaban, P.M., Zakharow, J.S. and Matz, V.N., Method of vital selective staining of serotonergic nerve cells by 5,7-dihydroxytryptamine, Dokl. Akad. Nauk USSR, 282 (1985) 735 738 (in Russian). 2 Beltz, B.S. and Kravitz, E.A., Mapping of serotonin-like immunoreactivity in the lobster nervous system, J. Neurosci., 3 (1983) 585 602. 3 Brown, R.O., Pulst, S.M. and Mayeri, E., Neuroendocrine bag cells of Aplysia are activated by bag cell peptide-containing neurons in the pleural ganglia, J. Neurophysiol., 61 (1989) 1142 1152. 4 Drummond, A.H., Bucher, F. and Levitan, I.B., Distribution of serotonin and dopamine receptors in Aplvsia tissue: analysis by [~H]LSD binding and adenylate cyclase stimulation, Brain Res., 184(1980) 163 177. 5 Ferguson, G.P., ter Maat, A. and Pinsker, H.M., Egg laying in Aplysia. II. Organization of central and peripheral pathways for initiating neurosecretory activity and behavioral patterns, J. Comp. Physiol., A 164 (1989) 849-857. 6 Hopkins, W.E., Stone, L.S., Rothman, B.S., Basbaum, A.I. and Mayeri, E., Egg-laying hormone, leucine-enkephalin and serotonin immunoreactivity in the abdominal ganglion of ApO,sia, Soc. Neurosci. Abstr., 8 (1982) 587. 7 Jahan-Parwar, B., S.-Rosza, K., Salanki, J., Evans, M.L. and Carpenter, D.O., In vivo labelling of serotonin-containing neurons by 5,7-dihydroxytryptamine in Aplvsia, Brain Res., 426 (1987) 173 178. 8 Jennings, K.R., Host, J.J., Kaczmarek, L.K. and Strumwasser, F., Serotonergic inhibition of afterdischarge in peptidergic bag cells, J. Neurobiol., 12 (1981) 579-590. 9 Kaczmarek, L.K., Jennings, K. and Strumwasser, F., Neurotransmitter modulation, phosphodiesterase inhibitor effects, and cyclic AMP correlates of afterdischarge in peptidergic neurons, Proc. Natl. Acad. Sci. U.S.A. 75 (1978) 5200 5204. 10 Katz, L.C., Burkhalter, A. and Dreyer, W.J., Fluorescent latex microspheres as a retrograde neuronal marker for in vivo and in vitro studies of visual cortex, Nature, 310 (1984) 498 500. 11 Kemenes, G., Elekes, K., Hiripi, L. and Benjamin, P.R., A comparison of four techniques for mapping the distribution of serotonin and serotonin-containing neurons in fixed and living ganglia of the snail Lymnaea, J. Neurocytol., 18 (1989) 193 208. 12 Kistler, H.B., Jr., Hawkins, R.D., Koester, J., Steinbusch, H.W.M., Kandel, E.R. and Schwartz, J.H. Distribution of serotonin-immunoreactive cell bodies and processes in the abdominal ganglion of mature Aplysia, J. Neurosci., 5 (1985) 72 80. 13 Mackey, S.L., Kandel, E.R. and Hawkins, R.D. Identified serotonergic neurons LCBI and RCBI in the cerebral ganglia of Aplvsia produce presynaptic facilitation of siphon sensory neurons, J. Neurosci., 9 (1989) 4227~4235. 14 Mayeri, E. and Rothman, B.S. Neuropeptides and the control ot" egg-laying behavior in Aplvsia. In A.I. Selverston (Ed.), Model Neural Networks and Behavior, Plenum, New York, 1985, pp. 285 -30 I. 15 Painter, S.D., Rock, M.K., Nagle, G.T. and Blankenship, J.E., Peptide B induction of bag-cell activity in Aplysia: localization of sites of action to the cerebral and pleural ganglia, J. Neurobiol., 8 (1988) 695-706. 16 Pinsker, H.M. and Dudek, F.E., Bag cell control of egg laying in freely behaving Aplvsia, Science, 197 (I977)490 493. 17 S.-Rosza, K., Hernadi, L. and Kemenes, G., Selective in viw~ labelling of serotonergic neurones by 5,6-dihydroxytryptamine in the snail Helix pomatia g., Comp. Biochem. Physiol., 85 (1986) 419 425.