The Journal of Neurophysiology Vol. 79 No. 2 February 1998, pp. 605-621
Copyright ©1998 by the American Physiological Society

Identification and Characterization of Catecholaminergic Neuron B65, Which Initiates and Modifies Patterned Activity in the Buccal Ganglia of Aplysia

E. A. Kabotyanski, D. A. Baxter, and J. H. Byrne

Department of Neurobiology and Anatomy, The University of TexasHouston Medical School, Houston, Texas 77030

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Kabotyanski, E. A., D. A. Baxter, and J. H. Byrne. Identification and characterization of catecholaminergic neuron B65, which initiates and modifies patterned activity in the buccal ganglia of Aplysia. J. Neurophysiol. 79: 605-621, 1998. Catecholamines are believed to play an important role in regulating the properties and functional organization of the neural circuitry mediating consummatory feeding behaviors in Aplysia. In the present study, we morphologically and electrophysiologically identified a pair of catecholaminergic interneurons, referred to as B65, in the buccal ganglia. Their processes innervate both the ipsi- and contralateral neuropil, and separate branches of B65 appeared to innervate the somata of both ipsi- and contralateral B4/5 neurons. B65 exhibited patterned burst(s) of activity during spontaneous cycles of fictive feeding. Patterned activity in B65 also was elicited by stimulation of the radula nerve, by depolarization of the pattern initiating neurons B31/32 or B63, and by bath application of L-3,4-dihydroxyphenylalanine (DOPA). B65 appeared to be a member of the protraction group of neurons. Action potentials in B65 elicited fast one-for-one excitatory postsynaptic potentials (EPSPs) in neurons B4/5, B8A/B, B31/32, B63, and B64. In turn, B31/32 and B63 excited B65 and B64 inhibited B65. Some of the synaptic connections of B65 were plastic. For example, the fast EPSPs elicited in B4/5 and B64 decremented, whereas those in B31/32 andB8A/B facilitated. In addition to fast EPSPs, B65 elicited slow postsynaptic potentials in some of its follower cells. Depolarization of B65 elicited cycles of patterned activity indicative of fictive feeding in buccal neurons, including B65 itself. During series of B65-induced patterns, the properties of the buccal motor programs appeared to change. In particular, the activity of radula closure motor neurons B8A/B, which initially coincided mainly with the protraction phase of a cycle, gradually extended to overlap mostly with the retraction phase. This observation suggests that prolonged activity in B65 may play a role in transitioning from rejection-like to ingestion-like fictive feeding. The phase shift of the activity of B8A/B appears due, at least in part, to a decrease in activity of B4/5, and thus a reduction in inhibition from B4/5 onto B8A/B, during the retraction phase. The functional properties and synaptic connections of B65 suggest that it may play an important role in determining features of patterned neural activity in the buccal ganglia.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

Rhythmic behaviors in invertebrates provide excellent opportunities to understand, at the level of individual neurons and synapses, how such behaviors are generated, controlled, and modified. One useful preparation for such studies is feeding behavior in Aplysia. Aplysia usually start feeding with appetitive responses that lead to contact with food (Bablanian et al. 1987; Kupfermann 1974a,b; Teyke et al. 1990, 1992). The following consummatory phase of feeding consists of a series of well-coordinated rhythmic movements of the lips, jaws, odontophore, and esophagus, which result in either ingestion (biting or swallowing) or rejection of food (Howells 1942; Kandel 1979; Kupfermann 1974a; Morton and Chiel 1993a; Susswein et al. 1976). In general, a cycle of ingestion or rejection, starts with protraction of the odontophore, a muscular tongue-like organ covered with toothed radula, and concludes with its retraction (Hurwitz et al. 1995, 1996; Kupfermann 1974a; Morton and Chiel 1993a). If the two halves of the radula are open during the protraction phase and closed during the retraction phase, then food (seaweed) is grasped and transported into the buccal cavity (i.e., ingestion). Alternatively, if the radula is closed during protraction and open during retraction, then objects are expelled from the mouth (i.e., rejection) (Church and Lloyd 1994; Kupfermann 1974a; Morton and Chiel 1993b).

The rhythmic movements of the jaws and radula are believed to be under the control of a neural circuit located in the buccal ganglia that functions as a central pattern generator (CPG) (for recent reviews of CPGs, see Arshavsky et al. 1993; Cropper and Weiss 1996; Getting 1989; Harris-Warrick 1993; Kupfermann 1994; Selverston 1992). The understanding of the buccal CPG has progressed significantly in recent years. Many premotor and motor neurons expressing patterned activity that underlies aspects of feeding or are critical for generating different patterns have been identified (Church and Lloyd 1994; Church et al. 1991, 1993; Cohen et al. 1978; Cropper et al. 1990; Gardner 1971, 1977; Hurwitz et al. 1994; Kirk 1989; Plummer and Kirk 1990). The relative phase of spike activity in some of these neurons has been characterized and, accordingly, they have been classified into the protraction- or the retraction-phase groups or the radula closure group (Church and Lloyd 1994). For example, neurons B31/32, B35, and B63 appear to be critical for generating the protraction phase of the buccal motor programs, whereas B64 appears to be critical for generating the retraction phase (Hurwitz and Susswein 1996; Hurwitz et al. 1993, 1996; Susswein and Byrne 1988; Susswein et al. 1996; see also Kabotyanski et al. 1994b; Ziv et al. 1994), and motor neurons B8A/B, B16 control the closure of the radula (Church and Lloyd 1994; Cropper et al. 1990; Morton and Chiel 1993b). Finally, a distinction between ingestion and rejection-like motor programs was established based on the timing of activity in the radula closure group neurons relative to the activity of protraction and retraction-groups. An ingestion-like buccal motor program (fictive feeding) is characterized by the firing of radula closure neurons mainly in phase with retraction-group neurons, whereas a rejection-like fictive feeding is characterized by the firing of radula closure neurons mainly in phase with protraction-group neurons (Church and Lloyd 1994; Morton and Chiel 1993b; Nargeot et al. 1997). The neuronal mechanisms of this phase shift underlying switching between different types of motor programs are not well understood, however.

In other CPGs, for example in the crustacean stomatogastric CPG, functional reconfiguration accompanied by phase shifts in the patterned activity have been shown to be induced by modulatory transmitters (Harris-Warrick et al. 1995). A possible candidate for the transmitter that could change functional properties in the buccal CPG of Aplysia is dopamine. The buccal ganglia receive an abundant catecholaminergic input from the foregut via the esophageal nerves (Susswein et al. 1993) and from the cerebral ganglia via the cerebro-buccal connective (Rathouz and Kirk 1988; Rosen et al. 1991). In addition, several catecholamine-containing neurons have been observed in the buccal ganglia (Goldstein and Schwartz 1989; Hawkins 1989; Rathouz and Kirk 1988; Salimova et al. 1987; Teyke et al. 1993). The principal catecholamine in the CNS of Aplysia and other mollusks is dopamine (Franchini et al. 1985; Guthrie et al. 1975; Juorio and Killick 1972; McCaman et al. 1973, 1979; Sloley et al. 1990; Straub and Kuhlmann 1984). Finally, dopamine has been implicated in feeding in gastropod mollusks (Trimble and Barker 1984; Voronezhskaya and Kabotyanski 1991; Wieland and Gelperin 1983), including Aplysia (Rosen et al. 1989; Teyke et al. 1993). Previously, we found that dopamine or its precursor L-3,4-dihydroxyphenylalanine (DOPA) initiated or substantially accelerated rhythmic activity in the buccal CPG (Baxter et al. 1995; Kabotyanski et al. 1993, 1994b). Moreover, both dopamine or DOPA biased the phase relationships among the elements of the feeding circuit toward that characteristic of an ingestion-like motor program (Baxter et al. 1995; Kabotyanski et al. 1994b).

To examine further the cellular mechanisms of dopamine-dependent transitions from rejection to ingestion, the present study focused on finding catecholaminergic neurons that may contribute to these processes in the buccal ganglia. We identified and characterized a lateral pair of midsized catecholaminergic neurons that we refer to as B65. We also examined the possible role that B65 may play in generating different types of buccal motor programs. Repeated firing of B65 led to a gradual transition from rejection-like to intermediate and ingestion-like patterns. In addition, the activity of B4/5 neurons appeared to decrease gradually during B65-induced buccal motor programs, and we examined how this effect may contribute to the phase shift of activity in radula closure neurons B8A/B. Preliminary reports of aspects of this work have appeared previously (Kabotyanski et al. 1994a,b).

	METHODS

Abstract Introduction Methods Results Discussion References

Aplysia californica were obtained from the Aplysia Resource Facility, University of Miami (Miami, FL), and Marine Specimens Unlimited (Pacific Palisades, CA). Animals (150-350 g) were maintained in aquaria with aerated artificial seawater (Instant Ocean, Aquarium Systems, Mentor, OH) at 15°C on 12 h dark/light cycle and fed regularly with dried seaweed (Hang Loong Marine Products, Hong Kong).

Before dissection, animals were anesthetized by injection of a volume of isotonic MgCl₂ solution equal to 30% of their body volume, and then 10-20 ml of high [Mg²⁺] hemolymph was drawn. The buccal ganglia were excised, usually treated for 2.5 min with 0.4% solution of Pronase E (Sigma) at room temperature, then washed with cold hemolymph (15 min), pinned to the silicone elastomer (Sylgard, Dow Corning, Midland, MI)-covered bottom of a recording chamber in 1:1 mixture of artificial seawater and high [Mg²⁺] hemolymph, and surgically desheathed.

Glyoxylate-induced histofluorescence of catecholamines was obtained in whole-mount preparations as described in Kabotyanski and Sakharov (1990). Briefly, desheathed ganglia were incubated in freshly made glyoxylate-containing solution at 4°C for 2 h, placed on a glass slide, dried with air blower at room temperature for 1.5-2 h, heated in an oven at 66°C for 9 min, covered with paraffin oil, coverslipped, and viewed under an Axiophot fluorescence microscope (Carl Zeiss, Oberkochen, Germany) with filter set 05 (green-blue fluorescence for catecholamines and yellow fluorescence for indolamines, i.e., serotonin). The incubating solution contained (in mM) 500 sodium glyoxylate (Fluka Chemical, Ronkonkoma, NY), 40 N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid (HEPES) buffer, and 100 sucrose, dissolved in deionized water, pH 7.0 (Sigma Chemical, St. Louis, MO). The final pH of the solution was adjusted to 7.0 with either glyoxylic acid (Fluka) or sodium bicarbonate crystals. Control experiments used either an incubating solution in which sodium glyoxylate was substituted in equimolar amounts for sodium chloride or specific fluorescence was extinguished with water (Sakharov and Salimova 1980).

For electrophysiological experiments, ganglia were pinned either caudal side up or twisted along the commissure, allowing simultaneous access to the caudal surface of one ganglion and rostral surface of the other. The preparation was superfused continuously with artificial seawater at a flow rate of ~2 ml (the volume of the recording chamber) per minute. The composition of artificial seawater (in mM) was 450 NaCl, 10 KCl, 30 MgCl₂, 20 MgSO₄, 10 CaCl₂, 2.5 NaHCO₃, and 10 HEPES, pH adjusted to 7.5. In some experiments, artificial seawater contained DOPA (Calbiochem, La Jolla, CA) in concentration 40 µg/ml. In some experiments, the bathing medium was used that contained three times higher concentrations of divalent ions (3×[Ca²⁺], 3×[Mg²⁺] artificial seawater) with the following composition (in mM): 300 NaCl, 10 KCl, 130 MgCl₂, 20 MgSO₄, 30 CaCl₂, 2.5 NaHCO₃, and 10 HEPES, pH adjusted to 7.5. Intracellular recordings were performed using glass microelectrodes filled with a solution containing 3 M potassium acetate and 0.1 M potassium chloride (10-15 M) or with a fluorescent dye, which was used for double-labeling experiments (15-25 M). Intracellular recordings were performed with conventional current-clamp techniques. Neurons were identified based on their position and physiology. In addition, the identification of B8 was verified by its ability to produce axonal spikes in the radula nerve, and the identification of B63 was verified by its morphology. Extracellular recordings from one of two radula nerve branches were performed using plastic suction electrodes and a differential AC-coupled amplifier. When needed, the radula nerve was stimulated via the same electrodes with train of five pulses (3-6 V, 75 ms, 1 Hz) applied via stimulus isolator. The preparations were maintained at 15-16°C.

For double-labeling and morphological studies, neurons were filled ionophoretically with the vital dye Fast Green FCF (Sigma), processed for histofluorescence, and then viewed in the fluorescence microscope with filter set 00 for bright red fluorescence of Fast Green, as well as with filter set 05 for catecholamines. Preparations were photographed on Kodak Technical Pan film (Eastman Kodak, Rochester, NY) and drawn using the Neurolucida program (MicroBrightField, Baltimore, MD).

For statistical analysis, data were averaged within each experiment first and then averaged across experiments. When phase ratio was analyzed with Student's t-test (Figs. 13 and 14), data in each experiment were first transformed using arcsine square root transformation to meet assumption of normality because proportions are binomial (Zar 1984).

View larger version (35K): [in this window] [in a new window]   FIG. 13. Repeated firing of B65 changes phase distribution of the activity in radula-closure neurons B8A/B, resulting in a shift toward intermediate and ingestion-like fictive feeding. Shaded areas below the B65 traces indicate the time of application of the stimulus train. Dashed vertical lines mark point of transition from protraction phase (to the left) to retraction phase (to the right). Data in A1 and A2, are from an experiment in which B65 and B4 were located contralaterally. A1: rejection-like patterned activity induced by the 2nd train of pulses. A2: intermediate pattern induced by the 10th train of pulses. Majority of spikes in B8 occurred during the retraction phase (33 spikes vs. 26). B1: rejection-like patterned activity induced by the 2nd train of pulses in another preparation in which B65 and B4 were located ipsilaterally. In this preparation, the B8 neuron that was monitored did not exhibit spikes during the first 3 trains. Nevertheless, large unit activity in the radula nerve indicated that other B8 neurons were active exclusively during protraction phase. Sequence of gradually decrementing EPSPs (arrows) in B4 is a likely indicator of the beginning and the rate of firing in contralateral B65. It appears that firing in the contralateral B65 is delayed and less intense than that elicited by train of pulses in the impaled (ipsilateral) B65. B2: ingestion-like patterned activity during the 10th train of pulses. Spikes in B8 occurred almost exclusively during the retraction phase. C: summary data for the shift in the phase distribution of activity in neurons B8A/B as represented by the phase ratio (see text). In one experiment, the impaled B8A/B neuron was not spiking during first 3 cycles (e.g., B1), so the ratio was calculated based on large units activity in radula nerve. Data were averaged for a corresponding train number across all experiments and plotted. Numbers in bars show how many data points were included in each average (if a train in B65 elicited an incomplete cycle, i.e., lacking distinct protraction or retraction, it resulted in a missing point).

View larger version (22K): [in this window] [in a new window]   FIG. 14. Control of radula closure group neurons B8A/B by ipsilateral neurons B4/5. All data taken from the same experiment. A1: during a quiescent episode, B4 elicited weak IPSPs in ipsilateral B8. A2: during a spontaneous burst in B8, B4 elicited inhibition of B8 that was sufficiently strong to interrupt B8 activity. B: inhibiting the firing of neuron B4 increased the activity of B8 during retraction phase. B1: a cycle of spontaneous patterned activity with rejection-like phase distribution (i.e., activity in B8 occurs primarily during the protraction phase). B2: inhibiting B4 with hyperpolarizing current resulted in an increase in B8 activity during the retraction phase. B3: spontaneous activity returned to rejection-like after removal of inhibition in B4. C: summary data from 6 experiments identical to that illustrated in B. Phase ratio values were determined for each cycle in each experiment, and 2 groups of values were obtained: one for cycles in which B4/5 was not hyperpolarized (control) and another for cycles in which B4/5 was hyperpolarized (B4 hyperpolarized). Data were averaged across all experiments and plotted. Phase ratios in 2 groups are significantly different (paired-sample t-test; 2-tailed, P < 0.0002)

	RESULTS

Abstract Introduction Methods Results Discussion References

Localization of catecholamines

Before electrophysiological identification, we obtained preliminary information about the number, relative positions, sizes, and branching patterns of catecholaminergic neurons. The mapping technique in the present study had to be compatible with the double-labeling procedure in subsequent experiments. In pilot studies, a commercially available antidopamine immunostaining procedure was found incompatible with the use of fluorescent labels because preparations had intense nonspecific background fluorescence. Thus we used a modification of glyoxylate-induced fluorescence technique (Kabotyanski and Sakharov 1990), which revealed the anatomic features of monoaminergic cells in greater detail than other histofluorescent procedures such as the FaGlu method (Goldstein and Schwartz 1989).

In 46 of 47 preparations, the paired buccal ganglia contained five midsized (40-60 µm diam), morphologically identifiable cells, and two lateral clusters of small (5-10 µm diam) cells, that exhibited bright green-blue fluorescence associated with catecholamines (Figs. 1A and 2A). Of the five midsized catecholamine-containing neurons, one was unpaired and usually located in the right hemiganglion. The other four neurons formed two bilaterally symmetric pairs. These results are consistent with the previous findings of Goldstein and Schwartz (1989). In one of the 47 preparations, we observed only three midsized catecholamine-containing neurons. In 27 preparations, we also observed an additional relatively large (~100 µm diam), paired cell, exhibiting pale green-blue fluorescence and located in the dorso-medial quarter on the caudal side of the ganglia (not shown). This observation suggests that additional catecholamine-containing neurons probably could be found in the buccal ganglia using more sensitive techniques (e.g., immunocytochemistry) or using animals in different behavioral states.

View larger version (146K): [in this window] [in a new window]   FIG. 1. Catecholamine-containing neural elements in the buccal ganglia revealed by glyoxylate-induced histofluorescence (caudal view, ventral surface is at the top, and right side is at left of the micrograph). A: low-power view of a whole-mount preparation. Five midsized neurons (2 paired and 1 unpaired), 2 lateral clusters of small cells (arrows; some cells in the clusters are slightly out of the focal plane) and rich catecholamine-containing neuropil are seen. Most lateral pair of mid-sized neurons is B65. Arrowhead points at area enlarged on B. Inset: a Neurolucida drawing of the buccal ganglia from another preparation showing the topography of the catecholamine-containing neurons. B: higher magnification of the area marked with the arrowhead on A that contained the somata of B4/5 neurons. Fine catecholaminergic processes on the soma of one right B4/5 cell are seen. The soma of catecholamine-containing neuron B20 (Teyke et al. 1993) appears at the bottom. C: higher magnification microphotograph of the right ventromedial region of another preparation. Fine catecholaminergic processes form a web-like wrapping around the soma of B4/5. Calibration bar: 200 µm for A, 50 µm for B and C. Abbreviations: b.n.1, buccal nerve 1; c.-b. c., cerebro-buccal connective; e.n., esophageal nerve; r.n., radula nerve. Nerve nomenclature from Gardner (1971).

View larger version (162K): [in this window] [in a new window]   FIG. 2. Double labeling of B65. B65was identified electrophysiologically, stainedintracellularly with red-fluorescing dye FastGreen, and then processed to reveal the catecholamines. A: histofluorescence of catecholamines (same orientation as in Fig. 1) viewed with the appropriate filter set. Five midsized cell bodies and two lateral clusters (arrows) of small cell bodies of catecholamine-containing neurons are seen. Arrowhead, soma of the left B65. B: filter set appropriate for red fluorescence reveals that the dye injected into cell via the recording microelectrode colocalizes with catecholamines in B65 (arrowhead). Calibration bar: 200 µm. Abbreviations are as in Fig. 1.

The cells of the medial pair recently were characterized and designated B20 (Teyke et al. 1993). The unpaired neuron, like B20, is bipolar; it sends one of each processes into one of each cerebro-buccal connectives. The lateral pair of cells appeared to send processes only into the contralateral ganglion via the buccal commissure (see further). This lateral pair of neurons was identified anatomically and electrophysiologically in the present study and designated B65.

The somata of B65 usually were covered by at least two layers of neurons on the caudal side. Some of the outer cells were B31-B37 (Susswein and Byrne 1988), which were covered in turn by neurons B1 and B2. The somata of B65 were found consistently within an arc formed by two axonal bundles that enter the buccal neuropil from the esophageal nerve (e.n.) and buccal nerve 1 (b.n.1; Fig. 2A) (nerve nomenclature from Gardner 1971).

The histofluorescence technique also revealed catecholamine-containing processes in many nerves of the buccal ganglia and in the neuropil. The radula nerves were the only nerves that failed to consistently contain catecholaminergic processes. In 12 preparations, however, we observed one or two fine catecholamine-containing neurites with varicosities on the surface of the radula nerves. The most abundant catecholaminergic processes appeared in the esophageal nerves, which previously have been shown to contain numerous neurites of catecholamine-containing neurons located in esophagus (Susswein et al. 1993). In addition, at least five bright catecholamine-containing processes were observed consistently in each cerebro-buccal connective (Fig. 2A). Three of these processes were apparently from the two B20 cells and the unpaired catecholamine-containing neuron. The fourth axon was probably from the cerebral catecholaminergic neuron CBI-1 (Rosen et al. 1991). The origin of the remaining catecholamine-containing process(es) in the cerebro-buccal connective is unknown.

We regularly observed a catecholaminergic braiding covering some neurons in the buccal ganglia, similar to that reported by Salimova et al. (1987). Somata of visually identifiable cells B1, B2, B3, B4, and B5 and some other unidentified cells were surrounded by a dense web of fine catecholamine-containing varicose neurites. Most prominent was the braiding around cells B4 and B5 (Figs. 1, B and C, and 3, A and C). We show below that, in part, these neurites are from B65.

View larger version (87K): [in this window] [in a new window]   FIG. 3. Morphology of B65. A: Neurolucida drawing of the buccal ganglia in which the right B65 was injected with Fast Green (same orientation as in Fig. 1). Arrow points at smaller branch that extended toward the cell bodies of neurons B4/5 located in ventromedial quadrant of the ipsilateral ganglion. This branch forms a part of catecholamine-containing wrapping around the B4/5 somata. A similar branch(es) is sent to the somata of contralateral B4/5 (arrowhead). B: high-power microphotograph taken from the preparation drawn in A. Cell body and proximal axon with a host of fine neurites terminating in the neuropil are seen. C: high-power micrograph taken from the ipsilateral ventromedial quadrant of the same preparation that was marked with arrow in A. A fine varicose neurite surrounds the cell body of B4/5. Calibration bar: 200 µm for A, 40 µm for B, 50 µm for C. Abbreviations are as in Fig. 1.

In the control experiments (5 with chloride for glyoxylate in the incubating solution and 5 with specific fluorescence extinguished with water), no fluorescing neurons were found.

Double labeling and morphology of B65

Because the somata of B65 were covered by other cells, they were inaccessible for identification based on simple visual criteria. The double-labeling procedure was employed routinely to verify that data collected in the electrophysiological experiments were obtained from B65. For this approach, microelectrodes filled with Fast Green-containing solution were used to impale a candidate cell. After experiment, the neuron was filled ionophoretically with Fast Green (2 to 5 nA, 15-40 min) until the soma acquired a light-blue color. The preparation then was processed for histofluorescence of monoamines (see METHODS) to determine whether green-blue fluorescence of catecholamines colocalized with the red fluorescence of Fast Green in the candidate cell. All electrophysiological data presented in this paper were obtained in experiments with positive double labeling (n = 24). Figure 2 illustrates one such experiment. Figure 2A shows histofluorescence of catecholamines in the buccal ganglia with arrowhead pointing to the right B65. Switching to the red fluorescence filter set (Fig. 2B) demonstrated that the neuron was filled with Fast Green during the electrophysiological experiment.

Intracellular injections of Fast Green revealed details of the morphology of B65. B65 usually had a major process that extended into the contralateral ganglion (Figs. 2B and 3A). In some preparations (3 of 15 histofluorescent preparations, 16 of 24 double-labeling preparations), at least one B65 of the pair had a second relatively large process branching in the proximity of the soma (e.g., see Fig. 1A, inset). In all cases, B65 branched extensively in the ipsilateral neuropil (Fig. 3, A and B), forming varicosities and terminals mainly along the neural tracts that enter the ganglion from the esophageal nerve and buccal nerve 1. In 5 of 24 double-labeling preparations, we observed fine branches that terminated at the entrance point of the esophageal nerve. In the medial zone of ipsilateral ganglion, the major axon sent a distinct branch toward the cell bodies of ipsilateral B4 and B5 (Fig. 3A, arrow). This branch formed a fine covering around B4/5 (Fig. 3C). In the contralateral ganglion, another branch, or several branches, were sent toward and formed braiding around contralateral B4/5 somata (Fig. 3A, arrowhead). The major process terminated in the contralateral ganglion, forming many large varicosities within the neuropil, although it did not branch as extensively as the ipsilateral portion (Figs. 2B and 3A). No branches of B65 were found in the buccal nerves or connectives. Thus B65 appeared to be an interneuron. Its functional properties were studied in electrophysiological experiments (see further).

Electrophysiological properties of B65

B65 EXHIBITED PATTERNED ACTIVITY. The resting potential of B65 ranged from -44 to -60 mV, with an average of-52.2 ± 1.1 mV (mean ± SE; n = 24). Average spike amplitude was 70.3 ± 1.9 mV. B65 was typically silent, sometimes producing a burst of spikes on impalement, but in preparations that exhibited patterned activity, B65 fired bursts of spikes that were phase-locked with cycles of the buccal motor program (Fig. 4A).

View larger version (0K): [in this window] [in a new window]   FIG. 4. B65 fires phasically along with other neurons during patterned activity. A: spontaneous cycle of rejection-like patterned activity. Cycle begins with activation of protraction-group neurons B65 by a synaptic input from unknown source (see text for explanation of phase relationships of B65) and is followed by activation of retraction-group neuron B4 concomitantly with inhibition in B65. Activity in radula closure neuron B8 is in phase with B65 and antiphase with B4, which is indicative of a rejection-like buccal motor program. B: cycle of rejection-like patterned activity elicited in the same preparation by electrical stimulation () of the radula nerve. The radula nerve trace was blanked during stimulation of the nerve. C: an ingestion-like patterned activity that was induced in the same preparation by bath-applied L-3,4-dihydroxyphenylalanine (DOPA). Abbreviations: r.n., radula nerve; R, right; L, left.

In the buccal ganglia, a cycle of a buccal motor program usually appears as a stereotypic, recurring sequence, or pattern, of phasic bursts of activity in a number of neurons. The cycle starts with a burst of spikes in protraction-group neurons and is followed by synaptic inhibition in these neurons coincident with a burst in retraction-group neurons. Activity in closure-group neurons shifts between these two phases depending on whether an ingestion-like or a rejection-like buccal motor program is produced. A useful monitor of the buccal motor programs is the radula nerve, in which a cycle of a buccal motor program occurs as a distinct pattern of extracellular activity (e.g., Figs. 4 and 11). The largest units recorded in the radula nerve during the pattern (e.g., Figs. 12C and 13, A and B) are produced by two bilateral radula closure neurons B8A/B (Morton and Chiel 1993b). Therefore simultaneously recording from a protraction- or retraction-group neuron and from a closure-group neuron or the radula nerve allows for discriminating between ingestion- and rejection-like patterns.

View larger version (1K): [in this window] [in a new window]   FIG. 11. Depolarization of B65 accelerates ongoing rhythmic activity. Periodic patterned activity was elicited by bath application of DOPA (40 min). Driving B65 with depolarizing current led to a fourfold increase in the cycling rate.

View larger version (0K): [in this window] [in a new window]   FIG. 12. Brief activation of B65 reveals regenerative properties of the buccal central pattern generator (CPG). A: firing B65 for 4 s with depolarizing pulse of +1.2 nA was insufficient to elicit sustained spiking in the follower cells. B: after 2 min rest, B65 was fired for the same period of time but with a current pulse of +1.4 nA, which elicited more spikes in B65 and a brief burst in B63. C: 2 min later, +1.4-nA current pulse of longer duration in B65 elicited a regenerative burst in B63, which, in turn, was sufficient to initiate a cycle of patterned activity in other neurons of the circuit, including B65 itself. Arrows in C indicate that EPSPs in B4 during the protraction phase (top arrows) are synchronous with spikes in contralateral B65 (bottom arrows). These EPSPs in B4/5 could be used to monitor activity in the contralateral B65 (see Fig. 13).

In addition to firing during spontaneous buccal motor programs, B65 was also active phasically during patterned activity that was elicited by stimulation of the radula nerve (Fig. 4B; 30 observations in all 8 experiments in which this was tested) or by depolarization of the pattern initiating neurons B31/32 (5 observations in 5 experiments) or B63 (13 observations in 3 experiments). When continuous rhythmic activity was elicited by bath application of DOPA, B65 also exhibited patterned activity (Fig. 4C; 3 experiments). Note that the bursts in B65 were associated with, and preceded by, a strong excitatory synaptic input to B65 from an unidentified source. Patterns in Fig. 4, A and B, were rejection-like, and pattern in Fig. 4C was ingestion-like.

When B65 fired during spontaneous or induced patterned activity, it was active primarily in phase with protraction-group neurons (e.g., B31/32, B63) and in antiphase with retraction-group neurons (e.g., B4/5 and B64). Moreover, B65 received a strong hyperpolarizing input during the retraction phase of patterned activity (Figs. 4, 10, 11, 12, and 13). This strong inhibitory input during the retraction phase is a distinguishing characteristic of the protraction-group neurons (Church and Lloyd 1994). Thus B65 was classified as a protraction-group neuron.

View larger version (0K): [in this window] [in a new window]   FIG. 10. B65 is sufficient to induce patterned activity in the buccal ganglia. The pattern was characterized by a gradual decrease in the firing rate of B4 neurons during the retraction phase in successive cycles and by a gradual increase in the activity of B8. Note that B65 fired in phase with B63, which belongs to the protraction-group neurons, and in antiphase with the most intense bursts of activity in B4, which is a retraction-group neuron. The patterned activity was not sustained despite the continued depolarization of B65.

SYNAPTIC CONNECTIONS OF B65. We explored the synaptic connections among B65 and some of the neurons known to exhibit patterned activity in the buccal ganglia. The multifunctional neurons B4/5 commonly are used to monitor intracellularly patterned activity in the buccal ganglia (e.g., Gardner 1977; Hurwitz et al. 1996; Plummer and Kirk 1990; Rosen et al. 1991; Sossin et al. 1987; Susswein and Byrne 1988; Teyke et al. 1993). B4/5 were classified as the neurons of the retraction-group, although their activity often starts during protraction (Church and Lloyd 1994). Firing B65 with depolarizing current elicited excitatory postsynaptic potentials (EPSPs) in B4/5. Effects of B65 on the ipsilateral and contralateral B4/5 were asymmetrical. B65 always elicited relatively large EPSPs in the contralateral B4/5, and these EPSPs gradually decreased in amplitude (Fig. 5, A and C; 178 observations in all 24 experiments). Often, the first EPSP initiated a spike (Fig. 5C). These characteristic EPSPs could be used as method for identifying and monitoring the activity of the contralateral B65 (e.g., Fig. 12). In contrast, in the ipsilateral B4/5, B65 usually elicited much smaller EPSPs that typically did not change in amplitude (Fig. 5C; 24 observations in 4 out of 6 experiments in which we compared the effects of B65 on the ipsi- and contralateral B4). In one experiment, no postsynaptic potentials (PSPs) were observed in the ipsilateral B4/5 (4 observations), and in only one experiment did PSPs appear identical on both sides (4 observations). EPSPs in both ipsi- and contralateral B4/5 were fast, followed one-for-one spikes in B65 with a constant delay, no failures were observed under different rates of firing (Fig. 5B). These EPSPs remained in 3×[Ca²⁺], 3×[Mg²⁺] artificial seawater (7 observations in 3 experiments). Neither the ipsi- nor the contralateral B4/5 had synaptic effects on B65 (not shown; 53 observations in all 15 experiments in which this was tested). No electrical coupling between B65 and B4/5 was found.

View larger version (25K): [in this window] [in a new window]   FIG. 5. Synaptic connections between B65 and B4/5 and B65 andB8A/B. A: spikes in B65 elicit one-for-one excitatory postsynaptic potentials (EPSPs) in contralateral neurons B4 and B8. Note that EPSPs in B4 gradually decremented, whereas those in B8 were facilitated. B: shaded segment of experiment shown in A was played back from tape recorder and displayed on a storage oscilloscope. Superimposed traces were obtained by triggering the oscilloscope with the rising phase of spikes in B65. EPSPs in B4 (top trace) and B8 (middle trace) followed spikes in B65 (bottom trace) with constant delay and no failures. C: B65 has asymmetric effects on ipsilateral and contralateral B4/5. EPSPs in ipsilateral B4 (middle trace) were of lower amplitude and did not exhibit gradual depression typical of EPSPs in the contralateral B4 (top trace). D: B65 also has asymmetric effects on ipsilateral and contralateral B8. Firing B65 elicited fast EPSPs only in contralateral B8. Recordings in A and B were from the same experiment, and recordings in C and D were from a separate experiment.

The radula closure motor neurons B8A/B usually received fast facilitating EPSPs from the contralateral B65 (Fig. 5A; 154 observations in 16 experiments) that followed one-for-one spikes in B65. In one experiment, B65 elicited biphasic, excitatory followed by inhibitory PSPs in B8. EPSPs in the contralateral B8A/B also were observed when the concentration of divalent ions was elevated (7 observations in 2 experiments). B65 produced no fast one-for-one PSPs in the ipsilateral B8 (Fig. 5D; 9 observations in 3 experiments). B8 produced no synaptic input to B65 (not shown; 50 observations in 7 experiments). No electrical coupling was found.

We also examined the connections between B65 and the multifunctional neurons B31/32, which previously were shown to be critical for initiating patterned activity (Hurwitz et al. 1994; Susswein and Byrne 1988). These cells belong to the protraction-group neurons (Hurwitz et al. 1995, 1996). Firing B65 elicited fast EPSPs in contralateral B31/32 that followed one-for-one spikes in B65 with constant delay and no failures under different rates of firing (Fig. 6C; 71 observations in 10 experiments). These PSPs summated and their amplitude increased with the increased firing rates in B65 (Fig. 6, A and C). In addition to fast time-locked EPSPs,B65 usually evoked polysynaptic excitatory input to B31/32(Fig. 6, A and C). In turn, activation of B31/32 depolarized B65 (not shown, 7 observation in 5 experiments), but it was difficult to further assess the B31/32 to B65 connection because B31/32 generate plateau potentials and do not produce conventional spikes in their somata. We recorded from both cells ipsilaterally in only one experiment. In this case, B65 or B31/32 induced slow depolarization in each other, but we did not observe fast PSPs. B65 also was coupled electrically with B31/32 (Fig. 6B; 18 observations in 5 experiments). The coupling ratio between contralateral B31/32 and B65 cells was ~1:20 and was much lower between ipsilateral cells. Because EPSPs in B31/32 followed spikes in B65 with ~30-ms delay and electrotonic depolarization in B31/32 was negligible (Fig. 6C), the B65 to B31/32 synapse appears to be primarily chemical.

View larger version (33K): [in this window] [in a new window]   FIG. 6. Synaptic connections between B65 and pattern-initiating neurons B31/32. A: firing B65 elicited in B31/32 fast EPSPs that followed one-for-one spikes in B65 under different rates of firing. B: B65 and B31/32 were electrically coupled. C: data from shaded segments 1-3 of experiment in A were redisplayed in spike-triggering mode (C1, C2, and C3, correspondingly). EPSPs in B31/32 followed spikes in B65 with constant delay and no failures. Note that with higher rates of firing of B65, EPSPs in B31/32 increased in amplitude. Increase in amplitude of the EPSPs was associated with spike broadening in B65.

Recently, a new element of the buccal CPG has been identified as bilateral neuron B63, which is sufficient and necessary for the generation of buccal motor programs (Hurwitz et al. 1993; Susswein et al. 1996). We examined the connections between B63 and B65. Recordings were made in three experiments from a total of four B63 cells, which were identified based on location of their somata, morphology (all 4 were stained), and characteristic electrophysiological properties. A brief firing of B65 elicited fast, modestly facilitating, EPSPs in the contralateral B63 that followed one-for-one spikes in B65 with constant delay and no failures under different rates of firing (Fig. 7, A and B; 26 observations in 4 cells). These EPSPs summated and, with higher rates of firing in B65, led to the initiation of spikes in B63 (Fig. 12, A and B). Firing B63 produced a slow depolarization in the contralateral B65 (Fig. 7C; 26 observations in 4 cells). B65 and B63 were coupled electrically (Fig. 7D; 20 observations in 4 cells) with coupling ratio ~1:20.

View larger version (26K): [in this window] [in a new window]   FIG. 7. Synaptic connections between B65 and the protraction-group neuron B63. A: B65 elicited EPSPs in contralateral interneuron B63. B: data from experiment in A redisplayed in spike-triggering mode. EPSPs in B63 followed spikes in B65 with constant delay and no failures. C: B63 produced a slow depolarization but no fast PSPs in B65. D: B65 and B63 were coupled electrically. All data are from the same experiment.

B65 also elicited EPSPs in retraction-group interneurons B64. EPSPs in B64 followed one-to-one spikes in B65 and decremented (Fig. 8, A and B; 32 observations in 4 experiments). In turn, B64 produced fast summating inhibitory PSPs (IPSPs) in B65 (Fig. 8C; 53 observations in 4 experiments).

View larger version (14K): [in this window] [in a new window]   FIG. 8. Synaptic connections between B65 and the retraction-phase interneuron B64. A: B65 elicited in interneuron B64 EPSPs that exhibited homosynaptic depression. B: data from shaded segment of experiment in A redisplayed in spike-triggering mode. EPSPs in B64 followed spikes in B65 with constant delay and no failures. Waveform of the 1st and 2nd EPSPs was distorted by superimposed fast small PSPs that were elicited synchronously in B64 and B65 (see bottom trace) by an unidentified source. C: B64 elicited fast inhibitory postsynaptic potentials (IPSPs) in B65. Data in C are from a different experiment than data in A and B.

LONG-LASTING SYNAPTIC ACTIONS OF B65. In addition to fast synaptic potentials, B65 elicited slow and relatively persistent synaptic potentials in B8A/B (Figs. 5A and 9). In addition, the depression of B65-elicited activity in B4/5 appeared to be slow to recover (Fig. 9).

View larger version (25K): [in this window] [in a new window]   FIG. 9. B65 has slow- and long-lasting effects that are cell specific. Five bursts were elicited in B65 by five 1-s depolarizing current pulse injections with 1-s interpulse interval. In B4, these bursts produced EPSPs and spikes that gradually decreased from burst to burst until response in B4/5 stabilized. In contrast to B4, in B8 EPSPs did not decrement, and B8 gradually depolarized during the consecutive bursts in B65.

B65 ELICITED PATTERNED ACTIVITY. The observation that B65 exhibited synaptically driven patterned discharges during spontaneous or induced episodes of patterned activity and had synaptic connections with neurons implicated in the CPG for feeding suggested that B65 was an element of the CPG. An important test for direct involvement of this neuron in pattern generation is whether B65 is sufficient to induce a pattern. Indeed, in initially quiescent preparations, firing B65 elicited patterned activity in B65, as well as in neurons B4/5, B8A/B, B31/32, B63, and B64 (e.g., Fig. 10; multiple observations in all 18 experiments in which this was tested). Usually, threshold currents for eliciting patterned activity were +1.5 to +2 nA. We used long (1-3 min) or sustained depolarizing currents of twice the threshold to study B65-elicited patterns.

Although B65 invariably induced patterned activity, the number of cycles that were elicited varied. Most often, only two cycles of activity were elicited with an average period of 35.8 ± 4.0 s (11 experiments). Sometimes, multiple cycles of rhythmic activity were initiated (e.g., Fig. 10) with average cycle period of 13.7 ± 2.8 s (5 experiments). In only two experiments, the B65-induced rhythmic activity was sustained as long as B65 was depolarized. Firing B65 also modulated the frequency of ongoing rhythmic activity (Fig. 11). In this experiment, we used a relatively low concentration of DOPA (40 µg/ml) to induce sustained cyclic activity of the CPG. Depolarization of B65 resulted in an increase (~4 times) in the rate of cyclic activity.

Brief activation of B65 was also sufficient to induce a pattern. Firing B65 with a brief depolarizing current pulse could elicit activity in other neurons that outlasted the depolarizing pulse and resulted in a cycle of patterned activity (Fig. 12C; note that the depolarizing pulse was much shorter than the initial protraction phase of the cycle). This finding (multiple observations in all 12 experiments where this was tested) indicates that B65 recruits elements of the CPG with regenerative properties, such as B31/32 (Susswein and Byrne 1988). A possible mechanism of pattern initiation may involve B65-elicited facilitating EPSPs in B31/32 and B63 as well as triggering polysynaptic input to B31/32 (Figs. 6 and 7).

Figure 12 also illustrates that B65 appears to be the source of characteristic synaptic input to the contralateral neurons B4/5 (Fig. 12C, arrows) that usually is observed during a cycle of fictive motor activity. This input often evoked spiking in B4/5 during the protraction phase of patterned activity. Moreover, blocking spikes in B65 by hyperpolarizing current blocked this characteristic input to the contralateral B4/5 (not shown), suggesting that one B65 is sufficient and necessary to produce this synaptic input in the contralateral B4/5. This input to B4/5 therefore can serve as a monitor of activity in contralateral B65 (see further).

Blocking activity in one of the paired B65 neurons with hyperpolarizing current failed to prevent the spontaneous or radula nerve-elicited patterned activity in other buccal neurons (9 observations in 4 experiments) nor did it change the phase relationship of patterns. This result suggested that B65 may not be necessary for the expression of this type of patterned activity, which usually was rejection-like. Rather B65 may play a role in modifying the pattern (see further).

B65 ELICITS VARIOUS TYPES OF MOTOR PROGRAMS. What is the nature of buccal motor programs (ingestion-like or rejection-like) that are initiated by B65? The answer to this question depends, in part, on the time at which activity in the radula closure group neurons (e.g., B8A/B) occurs relative to the protraction and retraction phases. In the present study, activity of closure-group neurons was monitored in two ways: using either intracellular recordings from radula closure motor neuron B8A/B or extracellular recordings from the radula nerve, where spikes of B8A/B are recognized as the larger units (Morton and Chiel 1993b).

To determine whether ingestion-like, rejection-like or intermediate patterns were elicited by B65, we reanalyzed the 18 experiments in which depolarizing B65 produced patterned activity. In 13 of 18 experiments, it was possible to monitor activity of closure-group neurons. We used a criterion similar to that developed by Morton and Chiel (1993a) to establish the type of B65-elicited patterns. We estimated the portion of large unit activity in the radula nerve (Morton and Chiel 1993a), or the portion of B8A/B activity, that overlapped with the retraction phase of a cycle of B65-elicited buccal motor programs. Cycles with an overlap of 0.1-0.7 were considered intermediate, smaller overlap indicated rejection-like, and larger overlap indicated ingestion-like activity.

B65 was activated with long (1-3 min) depolarizing currents. Each depolarization elicited one or more cycles of patterned activity. An average percentage of rejection, ingestion-like and intermediate cycles was determined for each preparation and then for all 13 preparations. Most patterns (50 ± 11%) were classified as being rejection-like (e.g., 1st, 2nd, 3rd, and 6th cycles in Fig. 10); 38 ± 11% were intermediate (e.g., 4th and 5th cycles in Fig. 10); and 12 ± 7% were ingestion-like (e.g., last 5 cycles in Fig. 11). This result is in contrast with the effects of prolonged applications of dopamine and DOPA on the phase relationships within CPG, as these drugs bias the patterned activity toward ingestion-like motor programs (Baxter et al. 1995; Kabotyanski et al. 1994b, 1995).

REPEATED ACTIVATION OF B65 AFFECTS PHASE RELATIONSHIPS BETWEEN BUCCAL CPG NEURONS. In three experiments in which firing B65 with constant depolarizing currents elicited five or more cycles, there appeared to be a trend for the phase distribution of B8 activity to shift from rejection-like to intermediate (e.g., Fig. 10) or to ingestion-like. Similar shifts occurred when B65 was depolarized during application of DOPA (e.g., Fig. 11). Figure 11 illustrates, for example, that the phasing of B8A/B neurons changed with each cycle during B65 stimulation until it became ingestion-like.

Because sustained depolarizations of B65 usually did not provide sustained patterned activity (see preceding text), in a separate set of five preparations, B65 was stimulated with series of trains of depolarizing current pulses to investigate the development of the apparent phase changes. The brief depolarizing pulses (80-ms, +6 nA/pulse, 40-ms interpulse interval) produced one spike per pulse and thereby provided a constant firing rate of B65 in all experiments. The frequency of the stimulus train was chosen to be similar to the firing rate of B65 (~8 Hz) during bursts of patterned activity induced by the stimulation of radula nerve or buccal nerve 2. In these experiments, the stimulation of B65 was terminated at the beginning of retraction phase of each induced cyclea point in time when activity of B65 naturally is inhibited by synaptic input. Finally, we found that patterned activity could not be evoked repeatedly if the stimulation was delivered too often. A 4-min rest period was needed between stimulations in most preparations. In each preparation, stimulus trains were applied to B65 10 times.

Figure 13 illustrates two of these experiments: one in which B65 and B4 were recorded contralaterally (Fig. 13A1 and B2), and another in which B65 and B4 were recorded ipsilaterally (Fig. 13B1 and B2). Typically, the first one to three trains in B65 induced a rejection-like pattern because B8 activity, monitored either intracellularly or by extracellular recording from the radula nerve, occurred almost exclusively during protraction phase (Fig. 13, A1 and B1). During subsequent cycles, a shift in phase distribution of spike activity in B8 occurred in all experiments. By the 10th cycle, B8 was active either during both protraction and retraction phases (Fig. 13A2; 3 experiments) or primarily during the retraction phase (Fig. 13B2; 2 experiments). Although the search was not exhaustive, we did not observe reproducible changes in the phase distribution of activity of cells other than B8. There was noticeable variability of neuronal activity within and between experiments, however, and this variability underlined some differences in activities of some cells in Fig. 13.

Summary data for the phase shift in B8A/B are illustrated in Fig. 13C. Data were quantified by counting spikes in a single B8A/B that coincided with retraction phase and dividing this number by the total number of spikes generated in the cell during that cycle. These values represented a measure of phase distribution of B8A/B activity, referred here to as phase ratio. The phase ratio did not significantly change during first four trains, but it gradually increased about sixfold from the 5th to the 10th trains. Statistical analysis indicated that this increase was significant in trains 9-10 as compared with trains 1-2 (paired-sample t-test, 2-tail, P < 0.03, df = 3, t = 3.18). This phase shift in the activity of the radula closure neurons indicated that repeated activation of a single B65 can lead to a transition from rejection-like toward ingestion-like fictive feeding.

In control experiments, we examined whether repeated induction of patterned activity via neural elements other than B65 could cause the phase shift in B8A/B. In four experiments, the same mode of stimulation described above was delivered to the pattern initiating neurons B31/32, while spike activity in one of the paired B65 was blocked by hyperpolarizing current. In these cases, we did not observe changes in the phase ratio of B8 activity (not shown). In another set of three experiments, we repeatedly evoked patterned activity by stimulating the radula nerve. In these experiments, all induced patterns were rejection-like. In addition, the phase relationships in the CPG during spontaneous patterned activity in the isolated buccal ganglia do not significantly change over time (Baxter et al. 1995). These results suggest that different means of pattern initiation can preferentially elicit different types of buccal motor programs, and repeated activation of B65 appears to be associated with ingestion-like patterns.

In addition to the phase shift in B8A/B, there was an increase of the total number of spikes in B8. Moreover, the level of activity in B4/5 during retraction phase appeared to gradually decrease during successive cycles of B65-elicited patterned activity (e.g., Fig. 13, A and B). This effect was manifested most prominently in those B4/5 that were ipsilateral to the B65 that was being repeatedly fired (compare Fig. 13B, 1 and 2, top traces). This decrease of the activity in neurons B4/5 was unlikely to be due to spike adaptation because we observed full bursts of spikes in B4/5 when patterned activity was elicited by stimulating the radula nerve between and after B65-elicited patterns (not shown). Thus the changes in the level of activity of B4/5 during retraction phase appeared to be specific to B65-induced activity. At the same time, this effect was transient, and after a rest period of ~10 min, B4/5 regained a high frequency of action potentials in response to reactivation of B65.

B4/5 AFFECTS PHASE DISTRIBUTION OF B8A/B ACTIVITY. Previously, B4/5 were reported to produce IPSPs in motor neurons B8A/B (Gardner 1971, 1977; Gardner and Kandel 1977), thus a decrease in activity of B4/5 during the retraction phase may contribute to concurrent increase of B8A/B activity. To explore this possibility, we first examined the B4/5 to B8A/B connection during periods of spontaneous bursts of activity in B8. In these experiments, B4 was hyperpolarized continuously to prevent it from spiking during spontaneous bursts; the current was released briefly to allow B4 to fire a burst of rebound spikes. During quiescent periods, B4/5 elicited small IPSPs in the ipsilateral B8 (Fig. 14A1). On arrival of a spontaneous burst in B8, the postinhibitory rebound burst of spikes in B4 reduced or interrupted firing in the ipsilateral B8 (Fig. 14A2; 5 of 6 experiments where this was tested). This effect was noticeable even though the other three B4/5 neurons in the ganglia were not hyperpolarized. These experiments illustrated that activation of B4/5 can inhibit spiking in the ipsilateral B8A/B during bursts of activity in B8. We next examined whether decreasing the activity in the B4/5 cells could affect the phase ratio of activity in B8A/B during fictive-feeding cycles.

Normally, spontaneous patterned activity in isolated buccal ganglia is predominantly rejection-like, i.e., B8A/B firing primarily during the protraction phase (Fig. 14B1). When activity in one of the four B4/5 neurons was depressed by applying a strong hyperpolarizing current at the onset of the retraction phase of a spontaneous cycle, the ipsilateral B8 produced more spikes during the retraction phase, indicative of an intermediate pattern (Fig. 14B2). Rejection-like patterned activity was restored in subsequent cycles if B4 was not hyperpolarized (Fig. 14B3; 6 experiments). Similar effects were observed when patterned activity was elicited by depolarization of neurons B31/32 (not shown). One B4/5 controlled the phase distribution of activity in both ipsilateral B8A/B cells in a similar manner (not shown). A summary of these experiments is plotted in Fig. 14C. These experiments indicated that decreasing the level of activity in the B4/5 cells during the retraction phase led to a significant concomitant increase in the activity of B8A/B (paired-sample t-test, 2-tail, P < 0.0002, df = 5, t = 2.57) and suggested a mechanism underlying the B65-induced phase shift.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

Previous results showed that bath applied dopamine or DOPA elicited or increased the frequency of rhythmic activity in isolated buccal ganglia and biased the phase relationships of this activity toward an ingestion-like one (Baxter et al. 1995; Kabotyanski et al. 1993, 1994b). In addition, in semi-intact preparations dopamine or DOPA induced biting-like movements (Kabotyanski et al. 1994b, 1995). These observations suggested that the ingestive behaviors of Aplysia may be controlled by catecholaminergic neural elements in the buccal ganglia.

CATECHOLAMINE-CONTAINING CELLS IN THE BUCCAL GANGLIA. Estimates of the number of catecholamine-containing neurons within the buccal ganglia from earlier histochemical studies varied. For example, Salimova et al. (1987) reported a total of two catecholamine-containing neurons in both ganglia, Rathouz and Kirk (1988) observed three to five midsized cells, and Hawkins (1989) found eight cells. In the present study, we addressed this issue using a glyoxylate technique modified for marine invertebrates (Kabotyanski and Sakharov 1990). In agreement with Goldstein and Schwartz (1989) and Teyke et al. (1993), we found that the buccal ganglia contain two symmetrical, lateral clusters of small-sized catecholamine-containing neurons, and five midsized cells, two paired and one unpaired. One pair of these cells is lateral and the other pair is medial. In addition, in >50% of preparations we observed an additional pair of relatively large cells in the dorso-medial quarter. The variations of the number of catecholamine-containing neurons in these studies could be explained by variations of sensitivities of different histofluorescent techniques that were used. The variations observed in our study could be caused by seasonal or behavioral variations of the level of catecholamines in individual neurons. The lateral pair of catecholamine-containing neurons was identified in this study as B65. On the basis of intracellular staining, B65 appeared to be an interneuron that was restricted to the buccal ganglia and that apparently formed synapses in both the ipsi- and contralateral ganglia. Because of its histochemistry, morphology, and position, B65 appears to be homologous to a catecholaminergic interneuron N1a that was recently identified in the buccal ganglia of Helisoma (Quinlan et al. 1997).

The specific catecholamine associated with the buccal neurons remains to be determined because the histochemical reaction used to reveal them visualizes all catecholamines and DOPA (Falk et al. 1962; Lindval and Bjorklund 1974; Lindval et al. 1974), and no effort was made to distinguish among them. Nevertheless, various biochemical studies have indicated that dopamine is a major (Carpenter et al. 1971; Juorio and Killick 1972), if not the only (McCaman et al. 1973, 1979), catecholamine in Aplysia. Therefore, there is a high probability that these neurons are dopaminergic.

CENTRAL CATECHOLAMINE-CONTAINING CELLS SHARE SOME SIMILAR PROPERTIES. Another pair of catecholamine-containing neurons located within the buccal ganglia is B20 (Teyke et al. 1993). In addition, the buccal ganglia are innervated by the catecholamine-containing cerebral-to-buccal interneuron CBI-1 (Rosen et al. 1991). B65, B20, and CBI-1 have similar effects in that they can all drive patterned activity. However, the types of activity elicited by their activation differ. For example, CBI-1, unlike B20 and B65, fires tonically and does not exhibit rhythmic activity on depolarization (Rosen et al. 1991). In addition, depolarization of CBI-1 always elicited only one cycle (Rosen et al. 1991), whereas B20 drives sustained cyclic activity (Teyke et al. 1993) and B65 induces mostly double or multiple cycles. Patterns driven by B65 appeared to exhibit a progressive, cycle-by-cycle decay in the level of activity ofB4/5 (see also Kabotyanski et al. 1994b), which, judging from published data, apparently is not the case for B20-driven activity (Rosen et al. 1991; Teyke et al. 1993). In other respects, the features of B65 and B20 are similar. They both have similar connections with the neurons of the buccal CPG, they elicit patterned activity in those neurons, and they exhibit phase-locked rhythmic activity. In addition, both B20 (based on data from Teyke et al. 1993) and B65 belong to the protraction-group neurons. It would be interesting to find whether B20 also can bias buccal motor programs toward ingestion.

B65 MAY BE AN ELEMENT OF THE BUCCAL CPG. Several lines of evidence suggest that B65 may be an element of the buccal CPG for feeding. First, direct activation of B65 elicited patterned activity in B65 itself as well as in neurons that have been identified previously as elements of the CPG. Second, B65 exhibited patterned activity during both spontaneous and induced buccal motor programs, and the activity of B65 was in phase with other protraction-group neurons. Third, B65 affected the rate of ongoing rhythmic activity (e.g., Fig. 11). Fourth, B65 had strong, probably monosynaptic, connections with other CPG neurons, particularly with pattern-initiating neurons B31/32 and B63. On the other hand, blocking spikes in one B65 with hyperpolarizing current did not prevent rejection-like patterned activity from occurring spontaneously or in response to stimulation of the radula nerve. This observation may indicate that B65 is not necessary for generation of rejection-like patterns, but rather that it is required for a specific (e.g., ingestion-like) activity.

SOME EFFECTS OF B65 ARE LONG LASTING AND SITE SPECIFIC. In addition to fast PSPs, B65 produced longer-lasting effects in some of its follower neurons. Some of these effects, such as synaptic depression in B4/5 and facilitation in B8A/B of B65-elicited EPSPs, developed rather rapidly. Others, such as decay of activity in B4/5 during the retraction phase (in this case, B4/5 activity is driven by retraction phase interneurons, e.g., B64), developed more gradually, afterrepeated activation of B65. Although we cannot currently establish whether these effects are pre- or postsynaptic, direct or indirect, these observations raise the intriguing possibility that B65 may be an example of an intrinsic modulatory CPG element, similar to the previously reported DSI neurons in Tritonia (Katz et al. 1994; Katz and Frost 1995).

View larger version (25K): [in this window] [in a new window]   FIG. 15. Circuit diagram illustrating connections between B65 and other neurons that were reported in this study. Connections for which data were collected in this study are indicated by thick lines, connections not studied here are indicated by thin lines. Because this circuit is limited to neurons investigated in the present study, some elements are not included, nor did we incorporate the differences in responses of contralateral counterparts of paired neurons. To reflect functional properties of the neurons, those that fire simultaneously are aligned vertically. Dashed lines separate neurons that are active during the protraction phase (left) from those firing during the retraction phase (right). Usually, the circuit is configured as shown in A and produces rejection-like pattern. After repeated stimulations of B65, the circuit becomes functionally reconfigured as shown in B to produce ingestion-like pattern. Shading in B4/5 in B reflects the decrease in the firing rate of this cell.

The B65 activity-dependent changes in synaptic efficacy were site specific. Although connections from both ipsi- and contralateral B65 converge on ipsi- and contralateral B4/5, decrementing EPSPs are exhibited only in the contralateral B4/5. In contrast, the depression of the firing in B4/5 is most prominently expressed in the ipsilateral B4/5. The synaptic depression of B65-elicited EPSPs in the contralateral B4/5 neurons represents an example of site-specific plasticity of synaptic strength, which has been reported previously at other synapses in the buccal (Gardner 1993) and abdominal ganglia (Clark and Kandel 1993) of Aplysia and in crustacean neuromuscular junctions (Wojtowicz et al. 1994).

FUNCTIONAL ROLE OF B65. An important issue is the functional role of B65 in the feeding network. On the basis of the effects of dopamine and DOPA on this system (Baxter et al. 1995; Kabotyanski et al. 1993, 1994b), we expected that B65 could initiate ingestion-like buccal motor programs. However, a single activation of B65 usually elicited patterns that were indistinguishable from spontaneously occurring rejection-like or intermediate patterns. In contrast, repeated bursts of B65 activity produced a significant phase shift in the firing of closure-group neurons B8A/B that indicated a transition from rejection-like toward ingestion-like buccal motor programs. We suggest that repeated bursts of activity in B65 may be necessary to release sufficient levels of dopamine and thereby reconfigure phase relationships between elements of the CPG. This intensive stimulation of B65 may have been required simply to compensate for the lack of stimulation of the second B65. Alternatively, it may represent a natural mode of activation of B65 by food stimuli. The finding that repeated stimulation of B65 results in the production of intermediate and ingestion-like patterns supports our hypothesis that dopamine is involved in the control of ingestive feeding behavior in Aplysia (Kabotyanski et al. 1994b, 1995). In the stomatogastric CPG, dopamine also produces phase shifts in the pyloric motor pattern (e.g., Harris-Warrick et al. 1995), which rises a question of the homology of the roles that this neurotransmitter may play in these evolutionary distant feeding systems.

B65 also fired during repetitive radula nerve-elicited or spontaneous buccal motor programs in control experiments when no phase shift was observed. Under these conditions, however, bursts in B65 were associated with, and preceded by, the strong excitatory synaptic input from an unidentified source(s) (Fig. 4, A and B). Therefore the rejection-like phasing of these patterns could not be linked causally to B65 because they were initiated by a different source. On the other hand, when patterned activity was elicited by repeatedly firing B65, the phase shift in B8 was linked causally to B65. Apparently, the buccal CPG is a multifunctional circuit that underlies various types of motor programs, and specific elements may play different roles in eliciting different patterns under specific conditions.

B65-ELICITED CHANGES IN B4/5 ACTIVITY MAY UNDERLIE TRANSITIONS TOWARD INGESTION-LIKE PATTERNED ACTIVITY. During activity induced by repeated stimulation of B65, we observed a gradual decay of the rate of firing in B4/5. This apparent effect was particularly interesting because it was similar to the apparent decay of B4/5 firing that accompanied the transition to ingestion-like patterned activity in presence of dopamine or DOPA (Kabotyanski et al. 1994b, 1997). In addition, the firing frequency of B4/5 appeared to be more than two times lower during ingestion-like buccal motor programs than during rejection-like buccal motor programs (Church and Lloyd 1994). This effect may contribute to the phase shift in B8A/B that occurs in series of B65-induced patterns. Indeed, inhibiting activity in B4/5 during retraction phase led to a concomitant increase of B8A/B activity (Fig. 14B), which is believed to be critical for biasing the buccal CPG toward generating ingestion-like patterns. The results further suggest that control of B4/5 activity may be one of several mechanisms controlling phase distribution of closure-group neurons during switching between rejection and ingestion caused by food or inedible objects.

CIRCUIT DIAGRAM. Figure 15 illustrates a summary of connections between B65 and neurons that were examined in this study. The connections on which data were collected in this study, including the B4 to B8 connection (Gardner 1971, 1977), are depicted with thick lines, and connections reported elsewhere with thin lines. To reflect functional properties of these neurons, we aligned vertically those cells that fire simultaneously during two types of patterned activityrejection-like (Fig. 15A) and ingestion-like (Fig. 15B) patterns. There are no anatomic differences between the two. Dashed lines in Fig. 15 separate neurons active during protraction phase (left) and those active during the retraction phase (right). The excitatory connections between B65 and B31/32 and B63 underlie the ability of B65 to initiate patterned activity because B31/32 and B63 are critical for the initiation of buccal motor programs (Hurwitz et al. 1993, 1996; Susswein and Byrne 1988; Susswein et al. 1996). After the onset of the pattern, protraction-group neurons discharge followed by a burst in B64, which starts the retraction phase by inhibiting B31/32, B63 and B65, and activating B4/5. During a rejection-like pattern (Fig. 15A), the inhibition from B4/5 prevents firing of B8A/B during retraction phase. With repeated activation of B65, however, B4/5 fires fewer spikes, which reduces its inhibitory effects on B8A/B. Thus B8A/B now fires during retraction, and the circuit becomes physiologically reconfigured to produce ingestion-like buccal motor program (Fig. 15B).

Further studies of the modulatory abilities of B65 to alter synaptic interconnections and intrinsic properties of the elements of the buccal CPG may provide better insights into functional significance of this cell in the buccal CPG as well as into mechanisms by which this CPG may be reconfigured to produce ingestive-like feeding motor programs.

	ACKNOWLEDGEMENTS

  We thank Dr. L. Cleary for help with microscopy and use of the Neurolucida program and Dr. G. Phares for comments on an earlier draft of the manuscript.

  This research was supported by Air Force Office of Scientific Research Grant F49620-97-1-0049 and National Institute of Mental Health Award K05 MH-00649.

	FOOTNOTES

  Address for reprint requests: J. H. Byrne, Dept. of Neurobiology and Anatomy, The University of Texas-Houston Medical School, PO Box 20708, Houston, TX 77225.

  Received 13 June 1997; accepted in final form 14 October 1997.

	REFERENCES

Abstract Introduction Methods Results Discussion References

• ARSHAVSKY, YU I., ORLOVSKY, G. N., PANCHIN, YU. V., ROBERTS, A., AND SOFFE, S. R. Neuronal control of swimming locomotion: analysis of the pteropod mollusc Clione and embryos of the amphibian Xenopus. Trends Neurosci. 16: 227-233, 1993.[Medline]
• BABLANIAN, G. M., WEISS, K. R., KUPFERMANN, I. Motor control of the appetitive phase of feeding behavior in Aplysia. Behav. Neural Biol. 48: 394-407, 1987.[Medline]
• BAXTER, D. A., CUSHMAN, S. J., KABOTYANSKI, E. A., BYRNE, J. H. Characterization of motor programs generated in isolated buccal ganglia of Aplysia and their modulation by biogenic amines. Soc. Neurosci. Abstr. 21: 1766, 1995.
• CARPENTER, D., BREESE, G., SCHANBERG, S., KOPIN, I. Serotonin and dopamine: distribution and accumulation in Aplysia nervous and non-nervous tissues. Int. J. Neurosci. 2: 49-56, 1971.[Medline]
• CHURCH, P. J., COHEN, K. P., SCOTT, M. L., KIRK, M. D. Peptidergic motorneurons in the buccal ganglia of Aplysia californica: immunocytochemical, morphological, and physiological characterizations. J. Comp. Physiol. [A] 168: 323-336, 1991.[Medline]
• CHURCH, P. J., LLOYD, P. E. Activity of multiple identified motor neurons recorded intracellularly during evoked feeding-like motor programs in Aplysia. J. Neurophysiol. 72: 1794-1809, 1994.[Abstract/Free Full Text]
• CHURCH, P. J., WHIM, M. D., LLOYD, P. E. Modulation of neuromuscular transmission by conventional and peptide transmitters released from excitatory and inhibitory motor neurons in. Aplysia. J. Neurosci. 13: 2790-2800, 1993.[Abstract]
• CLARK, G. A., KANDEL, E. R. Induction of long-term facilitation in Aplysia sensory neurons by local application of serotonin to remote synapses. Proc. Natl. Acad. Sci. USA 90: 11411-11415, 1993.[Abstract/Free Full Text]
• COHEN, J. L., WEISS, K. R., KUPFERMANN, I. Motor control of buccal muscles in Aplysia. J. Neurophysiol. 41: 157-180, 1978.[Free Full Text]
• CROPPER, E. C., KUPFERMANN, I., WEISS, K. R. Differential firing patterns of the peptide-containing cholinergic motor neurons B15 and B16 during feeding behavior in. Aplysia. Brain Res. 522: 176-179, 1990.[Medline]
• CROPPER, E. C., WEISS, K. R. Synaptic mechanisms in invertebrate pattern generation. Curr. Opin. Neurobiol. 6: 833-841, 1996.[Medline]
• FALK, B., HILLARP, N.-A., THIEME, G., TORP, A. Fluorescence of catecholamines and related compounds condensed with formaldehyde. J. Histochem. Cytochem. 10: 348-354, 1962.[Abstract]
• FRANCHINI, A., OTTAVIANI, E., CASELGRANDI, E. Biogenic amines in the snail brain of Helicella virgata (Gastropoda, Pulmonata). Brain Res. 347: 132-134, 1985.[Medline]
• GARDNER, D. Bilateral symmetry and interneuronal organization in the buccal ganglia of Aplysia. Science 173: 550-553, 1971.[Abstract/Free Full Text]
• GARDNER, D. Interconnections of identified multiaction interneurons in buccal ganglia of Aplysia. J. Neurophysiol. 40: 349-361, 1977.[Abstract/Free Full Text]
• GARDNER, D. Static determinants of synaptic strength. In: The Neurobiology of Neural Networks, edited by and D. Gardner . Cambridge, MA: MIT Press, 1993, p. 21-70
• GARDNER, D., KANDEL, E. R. Physiological and kinetic properties of cholinergic receptors activated by multiaction interneurons in buccal ganglia of Aplysia. J. Neurophysiol. 40: 333-348, 1977.[Abstract/Free Full Text]
• GETTING, P. A. Emerging principles governing the operation of neural networks. Annu. Rev. Neurosci. 12: 185-204, 1989.[Medline]
• GOLDSTEIN, R. S., SCHWARTZ, J. H. Catecholamine neurons in Aplysia: improved light-microscopic resolution and ultrastructural study using paraformaldehyde and glutaraldehyde (FaGlu) cytochemistry. J. Neurobiol. 20: 203-218, 1989.[Medline]
• GUTHRIE, P. B., NEUHOFF, V., OSBORNE, N. N. Dopamine, noradrenaline, octopamine and tyrosine-hydroxylase in the gastropod Helix pomatia. Comp. Biochem. Physiol. C Pharmacol. Toxicol. Endocrinol. 52: 109-111, 1975.
• HARRIS-WARRICK, R. M. Pattern generation. Curr. Opin. Neurobiol. 3: 982-988, 1993.[Medline]
• HARRIS-WARRICK, R. M., CONIGLIO, L. M., BARAZANGI, N., GUCKENHEIMER, J., GUERON, S. Dopamine modulation of transient potassium current evokes phase shifts in a central pattern generator network. J. Neurosci. 15: 342-358, 1995.[Abstract]
• HAWKINS, R. D. Localization of potential serotonergic facilitator neurons in Aplysia by glyoxylic acid histofluorescence combined with retrograde fluorescent labeling. J. Neurosci. 9: 4214-4226, 1989.[Abstract]
• HOWELLS, H. H. The structure and function of the alimentary canal of Aplysia punctata. Q. J. Microsc. Sci. 83: 357-397, 1942.
• HURWITZ, I., GOLDSTEIN, R. S., SUSSWEIN, A. J. A network of cells initiating patterned activity in the buccal ganglia of Aplysia. Soc. Neurosci. Abstr. 19: 1700, 1993.
• HURWITZ, I., GOLDSTEIN, R. S., SUSSWEIN, A. J. Compartmentalization of pattern-initiation and motor functions in B31 and B32 neurons of the buccal ganglia of Aplysia. J. Neurophysiol. 71: 1514-1527, 1994.[Abstract/Free Full Text]
• HURWITZ, I., MORTON, D. W., CHIEL, H. J., SUSSWEIN, A. The B31/32 pattern initiators of Aplysia are active during protraction of biting, swallowing and rejection movements. Soc. Neurosci. Abstr. 21: 149, 1995.
• HURWITZ, I., NEUSTADTER, D., MORTON, D. W., CHIEL, H. J., SUSSWEIN, A. Activity patterns of the B31/32 pattern initiators innervating the I2 muscle of the buccal mass during normal feeding movements in Aplysia californica. J. Neurophysiol. 75: 1309-1326, 1996.[Abstract/Free Full Text]
• HURWITZ, I., SUSSWEIN, A. B64, a newly identified central pattern generator element producing a phase switch from protraction to retraction in buccal motor programs of Aplysia californica. J. Neurophysiol. 75: 1327-1344, 1996.[Abstract/Free Full Text]
• JUORIO, A. V., KILLICK, S. W. Monoamines and their metabolism in some molluscs. Comp. Gen. Pharmacol. 3: 283-295, 1972.[Medline]
• KABOTYANSKI, E. A., BAXTER, D. A., BYRNE, J. H. Metabolic precursors of dopamine and serotonin modulate rhythmic neural activity in the buccal ganglia of Aplysia. Soc. Neurosci. Abstr. 19: 350, 1993.
• KABOTYANSKI, E. A., BAXTER, D. A., BYRNE, J. H. Identification of a catecholaminergic neuron in the buccal ganglia of Aplysia that initiates rhythmic neuronal activity indicative of ingestion. Soc. Neurosci. Abstr. 20: 23, 1994a.
• KABOTYANSKI, E. A., BAXTER, D. A., BYRNE, J. H. Experimental and computational analyses of a central pattern generator underlying aspects of feeding behavior of Aplysia. Neth. J. Zool. 44: 357-373, 1994b.
• KABOTYANSKI, E. A., BAXTER, D. A., BYRNE, J. H. Dopamine (DA) and serotonin (5HT) coordinate different aspects of consummatory feeding behavior in Aplysia. Soc. Neurosci. Abstr. 21: 1766, 1995.
• KABOTYANSKI, E. A., BAXTER, D. A., BYRNE, J. H. Involvement of neuron B4/5 in biasing the buccal CPG toward ingestion-like fictive feeding in Aplysia. Soc. Neurosci. Abstr. 23: 1046, 1997.
• KABOTYANSKI, E. A., SAKHAROV, D. A. Catecholaminergic neurons in the pteropod mollusc Clione limacina. J. Evol. Biochem. Physiol. 25: 198-207, 1990.
• KANDEL, E. R. In: Behavioral Biology of Aplysia., . San Francisco: W. H. Freeman, 1979
• KATZ, P. S., FROST, W. N. Intrinsic neuromodulation in the Tritonia swim CPG: the serotonergic dorsal swim interneurons act presynaptically to enhance transmitter release from interneuron C2. J. Neurosci. 15: 6035-6045, 1995.[Abstract]
• KATZ, P. S., GETTING, P. A., FROST, W. N. Dynamic neuromodulation of synaptic strength intrinsic to a central pattern generator circuit. Nature 367: 729-731, 1994.[Medline]
• KIRK, M. D. Premotor neurons in the feeding system of Aplysia californica. J. Neurobiol. 20: 497-512, 1989.[Medline]
• KUPFERMANN, I. Feeding behavior in Aplysia: a simple system for the study of motivation. Behav. Biol. 10: 1-26, 1974a.[Medline]
• KUPFERMANN, I. Dissociation of the appetitive and consummatory phases of feeding behavior in Aplysia: a lesion study. Behav. Biol. 10: 89-97, 1974b.[Medline]
• KUPFERMANN, I. Neural control of feeding. Curr. Opin. Neurobiol. 4: 869-876, 1994.[Medline]
• LINDVALL, O., BJORKLUND, A. The glyoxylic acid fluorescence histochemical method: a detailed account of the methodology for the visualization of central catecholamine neurons. Histochemistry 39: 97-127, 1974.[Medline]
• LINDVALL, O., BJORKLUND, A., SVENSSON, L. Fluorophore formation from catecholamines and related compounds in the glyoxylic acid fluorescence histochemical method. Histochemistry 39: 197-227, 1974.[Medline]
• MCCAMAN, M. W., ONO, J. K., MCCAMAN, R. E. Dopamine measurements in molluscan ganglia and neurons using a new, sensitive assay. J. Neurochem. 32: 1111-1113, 1979.[Medline]
• MCCAMAN, M. W., WEINREICH, D., MCCAMAN, R. E. The determination of picomole levels of 5-hydroxytryptamine and dopamine in Aplysia, Tritonia and leech nervous tissues. Brain Res. 53: 129-137, 1973.[Medline]
• MORTON, D. W. AND CHIEL, H. J. In vivo buccal nerve activity that distinguishes ingestion from rejection can be used to predict behavioral transitions in Aplysia. J. Comp. Physiol. [A] 172: 17-32, 1993a.
• MORTON, D. W., CHIEL, H. J. The timing of activity in motor neurons that produce radula movements distinguishes ingestion from rejection in Aplysia. J. Comp. Physiol. [A] 173: 519-536, 1993b.[Medline]
• NARGEOT, R., BAXTER, D. A., BYRNE, J. H. Contingent-dependent enhancement of rhythmic motor patterns: an in vitro analog of operant conditioning. J. Neurosci. 17: 8093-8105, 1997.[Abstract/Free Full Text]
• PLUMMER, M. R., KIRK, M. D. Premotor neurons B51 and B52 in the buccal ganglia of Aplysia californica: synaptic connections, effects on ongoing motor rhythms, and peptide modulation. J. Neurophysiol. 63: 539-558, 1990.[Abstract/Free Full Text]
• QUINLAN, E. M., ARNETT, B. C., MURPHY, A. D. Feeding stimulants activate an identified dopaminergic interneuron that induces the feeding motor program in Helisoma. J. Neurophysiol. 78: 812-824, 1997.[Abstract/Free Full Text]
• RATHOUZ, M. M., KIRK, M. D. Localization of catecholamines in the buccal ganglia of Aplysia californica. Brain Res. 458: 170-175, 1988.[Medline]
• ROSEN, S. C., TEYKE, T., MILLER, M. W., WEISS, K. R., KUPFERMANN, I. Identification and characterization of cerebral-to-buccal interneurons implicated in the control of motor programs associated with feeding in Aplysia. J. Neurosci. 11: 3630-3655, 1991.[Abstract]
• ROSEN, S. C., TEYKE, T., WEISS, K. R., KUPFERMANN, I. Identification and characterization of putative dopamine-containing neurons implicated in the control of feeding behavior in Aplysia. Soc. Neurosci. Abstr. 15: 736, 1989.
• SAKHAROV, D. A., SALIMOVA, N. B. Serotonin neurons in the peripheral nervous system of the larval lamprey, Lampetra planeri. A histochemical, microspectrofluorometric and ultrastructural study. Zool. Jahrb. Physiol. 84: 231-239, 1980.
• SALIMOVA, N. B., SAKHAROV, D. A., MILOEVIC, I., TURPAEV, T. M., RAKIC, L. Monoamine-containing neurons in the Aplysia brain. Brain Res. 400: 285-299, 1987.[Medline]
• SELVERSTON, A. I. Pattern generation. Curr. Opin. Neurobiol. 2: 776-780, 1992.[Medline]
• SLOLEY, B. D., JUORIO, A. V., DURDEN, D. A. High-performance liquid chromatographic analysis of monoamines and some of their -glutamyl conjugates produced by the brain and other tissues of Helix aspresa (Gastropoda). Cell. Mol. Neurobiol. 10: 175-192, 1990.[Medline]
• SOSSIN, W. S., KIRK, M. D., SCHELLER, R. H. Peptidergic modulation of neuronal circuitry controlling feeding in Aplysia. J. Neurosci. 7: 671-681, 1987.[Abstract]
• STRAUB, H., KUHLMANN, D. Identification and quantitative measurements of biogenic monoamines in the central nervous tissue of some gastropods. Comp. Biochem. Physiol. C Pharmacol. Toxicol. Endocrinol. 78: 319-323, 1984.
• SUSSWEIN, A. J., BYRNE, J. H. Identification and characterization of neurons initiating patterned neural activity in the buccal ganglia of Aplysia. J. Neurosci. 8: 2049-2061, 1988.[Abstract]
• SUSSWEIN, A. J., KABOTYANSKI, E. A., HURWITZ, I., SAKHAROV, D. A. Catecholaminergic neurons in the esophagus of Aplysia: a peripheral brain? Soc. Neurosci. Abstr. 19: 350, 1993.
• SUSSWEIN, A. J., KUPFERMANN, I., HURWITZ, I. Buccal cerebral interneurons B63 and B34 play different roles during the protraction phase of buccal motor programs in Aplysia. Soc. Neurosci. Abstr. 22: 2044, 1996.
• SUSSWEIN, A. J., KUPFERMANN, I., WEISS, K. R. The stimulus control of biting in Aplysia. J. Comp. Physiol. 108: 75-96, 1976.
• TEYKE, T., ROSEN, S. C., WEISS, K. R., AND KUPFERMANN, I. Dopaminergic neuron B20 generates rhythmic neuronal activity in the feeding motor circuitry of Aplysia. Brain Res. 630 226-237: 1993.
• TEYKE, T., WEISS, K. R., KUPFERMANN, I. Appetitive feeding behavior of Aplysia: behavior and neural analysis of head turning. J. Neurosci. 10: 3922-3934, 1990.[Abstract]
• TEYKE, T., WEISS, K. R., KUPFERMANN, I. Orientation of Aplysia californica to distant food sources. J. Comp. Physiol. [A] 170: 281-289, 1992.[Medline]
• TRIMBLE, D. L., BARKER, D. L. Activation by dopamine of patterned motor output from the buccal ganglia of Helisoma trivolvis. J. Neurobiol. 15: 37-48, 1984.[Medline]
• VORONEZHSKAYA, E. E., KABOTYANSKI, E. A. A comparative study of dopaminergic activation of feeding movements in "model" gastropods. In: Signal Molecules and Behavior, edited by W. Winlow, O. V. Vinogradova, and D. A. Sakharov . Manchester, UK: Manchester Univ. Press, 1991, p. 141-145
• WIELAND, D. L., GELPERIN, A. Dopamine elicits feeding motor programs in Limax maximus. J. Neurosci. 3: 1735-1745, 1983.[Abstract]
• WOJTOWITZ, J. M., MARIN, L., ATWOOD, H. L. Activity-induced changes in synaptic release sites at the crayfish neuromuscular junction. J. Neurosci. 14: 3688-3703, 1994.[Abstract]
• ZAR, J. H. In: Biostatistical Analysis., . Englewood Cliffs, NJ: Prentice-Hall, 1984
• ZIV, I., BAXTER, D. A., BYRNE, J. H. Simulator for neural networks and action potentials: description and application. J. Neurophysiol. 71: 294-308, 1994.[Abstract/Free Full Text]

0022-3077/98 $5.00 Copyright ©1998 The American Physiological Society