Second messengers of octopamine receptors in the snail Lymnaea.

We describe octopamine responses of 3 large buccal neurons of Lymnaea and test the hypothesis that these are cAMP-dependent. The B1 neuron is excited by octopamine and the depolarisation is significantly enlarged (P < 0.05) by application of the blocker of cAMP breakdown, 3-isobutyl-1-methylxanthine (IBMX). The B1 neuron is also depolarised by forskolin, an activator of adenylyl cyclase. The B2 and B3 neurons are inhibited by octopamine, and the response is not affected by IBMX. Both cells are excited by forskolin. We conclude that the B1 neuron response to octopamine is likely to be mediated by cAMP, while the B2 and B3 responses are cAMP-independent.

Octopaminergic modulation of the membrane currents in the central feeding system of the pond snail Lymnaea stagnalis.

Octopamine is released by the intrinsic OC interneurons in the paired buccal ganglia and serves both as a neurotransmitter and a neuromodulator in the central feeding network of the pond snail Lymnaea stagnalis. The identified B1 buccal motoneuron receives excitatory inputs from the OC interneurons and is more excitable in the presence of 10 microM octopamine in the bath. This modulatory effect of octopamine on the B1 motoneuron was studied using the two electrode voltage clamp method. In normal physiological saline depolarising voltage steps from the holding potential of -80 mV evoke a transient inward current, presumably carried by Na(+) ions. The peak values of this inward current are increased in the presence of 10 microM octopamine in the bath. In contrast, both the transient (IA) and delayed (IK) outward currents are unaffected by octopamine application. Replacing the normal saline with a Na(+)-free bathing solution containing K(+) channel blockers (50 mM TEACl, 4 mM 4AP) revealed the presence of an additional inward current of the B1 neurons, carried by Ca(2+). Octopamine (10 microM) in the bath decreased the amplitudes of this current. These results suggest that the membrane mechanisms which underlie the modulatory effect of octopamine on the B1 motoneuron include selective changes of the Na(+)- and Ca(2+)-channels.

Multilevel inhibition of feeding by a peptidergic pleural interneuron in the mollusc Lymnaea stagnalis.

The pleural interneuron PlB is a white neuron in the pleural ganglion of the snail Lymnaea. We test the hypothesis that it inhibits neurons at all levels of the feeding system, using a combination of anatomy, physiology and pharmacology. There is just one PlB in each pleural ganglion. Its axon traverses the pedal and cerebral ganglia, running into the buccal ganglia. It has neuropilar branches in the regions of the cerebral and buccal ganglia where neurons that are active during feeding also branch. Activation of the PlB blocks fictive feeding, whether the feeding rhythm occurs spontaneously or is driven by a modulatory interneuron. The PlB inhibits all the neurons in the feeding network, including protraction and retraction motoneurons, central pattern generator interneurons, buccal modulatory interneurons (SO, OC), and cerebral modulatory interneurons (CV1, CGC). Only the CV1 interneuron shows discrete 1:1 IPSPs; all other effects are slow, smooth hyperpolarizations. All connections persist in Ca(2+)/Mg(2+)-rich saline, which reduces polysynaptic effects. The inhibitory effects are mimicked by 0.5 to 100 micromol l(-1) FMRFamide, which the PlB soma contains. We conclude that the PlB inhibits neurons in the feeding system at all levels, probably acting though the peptide transmitter FMRFamide.

Heterosynaptic modulation by the octopaminergic OC interneurons increases the synaptic outputs of protraction phase interneurons (SO, N1L) in the feeding system of Lymnaea stagnalis.

We examined the cholinergic synapses between protraction phase interneurons (SO or N1L) and their targets (N1M interneuron, B1 motoneuron) in the buccal ganglia of the pond snail Lymnaea stagnalis. We have tested the hypothesis that the OC (octopamine-containing) interneuron, an intrinsic modulator of the feeding network, can increase the synaptic efficacy from the SO or N1L to their targets. Prestimulation of the OC interneuron, 4 s before the activation of the SO or N1L increases the strength of their output synapses by 75% (SO)-110% (N1L). The individual excitatory postsynaptic potentials evoked by SO or N1L stimulation increase in size. OC prestimulation also produces an increase in the firing rate of these presynaptic interneurons: SO 40%; N1L 33%. The facilitation lasts up to 6 s after the end of the OC burst. The enhancement of PSPs is seen at all the output synapses (both excitatory and inhibitory) of the SO and N1L interneurons. The output synapses of the non-cholinergic swallowing phase N3p interneuron are not affected, even when the same postsynaptic target is selected. The SO-->N1M, SO-->B1 and N1L-->N1M synapses are also strengthened by bath application of 1-5 microM octopamine (average increase 60%). The major effect is an increased excitability of the SO; the B1 motoneuron response to the main transmitter of the SO, acetylcholine, is unaffected. Increased synaptic outputs of the protraction phase SO and N1L interneurons is functionally significant for generation of feeding pattern in the Lymnaea CNS. Strengthening the connections of SO and N1L to the central pattern generator (N1M) interneurons enhances their ability to drive fictive feeding. Thus heterosynaptic facilitation by the octopaminergic OC interneurons in the central pattern generator network may contribute to the behavioral plasticity of feeding in the intact animal.

Comparative neuroethology of feeding control in molluscs.

Over the last 30 years, many laboratories have examined, in parallel, the feeding behaviour of gastropod molluscs and the properties of the nervous system that give rise to this behaviour. Equal attention to both behavioural and neurobiological issues has provided deep insight into the functioning of the nervous system in generating and controlling behaviour. The conclusions derived from studies on gastropod feeding are generally consistent with those from other systems, but often provide more detailed information on the behavioural function of a particular property of the nervous system. A review of the literature on gastropod feeding illustrates a number of important messages. (i) Many of the herbivorous gastropods display similarities in behaviour that are reflected in corresponding similarities in neural anatomy, pharmacology and physiology. By contrast, the same aspects of the behaviour of different carnivorous species are quite variable, possibly because of their specialised prey-capture techniques. Nonetheless, some aspects of the neural control of feeding are preserved. (ii) Feeding in all species is flexible, with the behaviour and the physiology adapting to changes in the current environment and internal state and as a result of past experience. Flexibility arises via processes that may take place at many neural sites, and much of the modulation underlying behavioural flexibility is understood at a systems and at a cellular level. (iii) Neurones seem to have specific functions that are consistent with their endogenous properties and their synaptic connections, suggesting that individual neurones code specific pieces of information (i.e. they are 'grandmother cells'). However, the properties of a neurone can be extremely complex and can be understood only in the context of the complete neural circuit and the behaviour that it controls. In systems that are orders of magnitude more complex, it would be impossible to understand the functional properties of an individual neurone, even if it also coded specific information. (iv) Systems such as gastropod feeding may provide a model for understanding the functional properties of more complex systems.

Activation and reconfiguration of fictive feeding by the octopamine-containing modulatory OC interneurons in the snail Lymnaea.

We describe the role of the octopamine-containing OC interneurons in the buccal feeding system of Lymnaea stagnalis. OC neurons are swallowing phase interneurons receiving inhibitory inputs in the N1 and N2 phases, and excitatory inputs in the N3 phase of fictive feeding. Although the OC neurons do not always fire during feeding, the feeding rate is significantly (P < 0.001) higher when both SO and OC fire in each cycle than when only the SO fires. In 28% of silent preparations, a single stimulation of an OC interneuron evokes the feeding pattern. Repetitive stimulation of the OC interneuron increases the proportion of responsive preparations to 41%. The OC interneuron not only changes both the feeding rate and reconfigures the pattern. Depolarization of the OC interneurons increases the feeding rate and removes the B3 motor neuron from the firing sequence. Hyperpolarization slows it down (increasing the duration of N1 and N3 phases) and recruits the B3 motor neuron. OC interneurons form synaptic connections onto buccal motor neurons and interneurons but not onto the cerebral (cerebral giant cell) modulatory neurons. OC interneurons are electrically coupled to all N3 phase (B4, B4Cl, B8) feeding motor neurons. They form symmetrical connections with the N3p interneurons having dual electrical (excitatory) and chemical (inhibitory) components. OC interneurons evoke biphasic synaptic inputs on the protraction phase interneurons (SO, N1L, N1M), with a short inhibition followed by a longer lasting depolarization. N2d interneurons are hyperpolarized, while N2v interneurons are slowly depolarized and often fire a burst after OC stimulation. Most motor neurons also receive synaptic responses from the OC interneurons. Although OC and N3p interneurons are both swallowing phase interneurons, their synaptic contacts onto follower neurons are usually different (e.g., the B3 motor neurons are inhibited by OC, but excited by N3p interneurons). Repetitive stimulation of OC interneuron facilitates the excitatory component of the biphasic responses evoked on the SO, N1L, and N1M interneurons, but neither the N2 nor the N3 phase interneurons display a similar longer-lasting excitatory effect. OC interneurons are inhibited by all the buccal feeding interneurons, but excited by the serotonergic modulatory CGC neurons. We conclude that OC interneurons are a new kind of swallowing phase interneurons. Their connections with the buccal feeding interneurons can account for their modulatory effects on the feeding rhythm. As they contain octopamine, this is the first example in Lymnaea that monoaminergic modulation and reconfiguration are provided by an intrinsic member of the buccal feeding network.

Comparative pharmacology of feeding in molluscs.

1. This paper reviews the role of transmitters in identified neurons of gastropod molluscs in generating and modulating fictive feeding. 2. In Lymnaea and Helisoma the 3 phase rhythm is generated by sets of interneurons which use acetylcholine for the N1 (protraction) phase, glutamate for the N2 (rasp) phase interneurons. The N3 interneurons are likely to use several different transmitters, of which one is octopamine. 3. In all the species examined, serotonin (5-HT) is released from giant cerebral cells. Other amines, including dopamine and octopamine, are present in the buccal ganglia and all these amines activate or enhance feeding. 4. Nitric oxide (NO), mostly originating from sensory processes, can also activate fictive feeding, but (at least in Lymnaea) may also be released centrally from buccal (B2) and cerebral neurons (CGC). 5. The central pattern generator for feeding is also modulated by peptides including APGWamide, SCP(B) and FMRFamide. 6. There is increasing evidence that most of these transmitters/modulators act on feeding neurons through second messenger systems--allowing them to act as longer-lasting neuromodulators of the feeding network. 7. Many of the transmitters are used in similar ways by each of the gastropods examined so far, so that their function in the CNS seems to have been conserved through evolution.

The octopamine-containing buccal neurons are a new group of feeding interneurons in the pond snail Lymnaea stagnalis.

In the pond snail, Lymnaea stagnalis, the paired buccal ganglia contain 3 octopamine-immunoreactive neurons, which have previously been shown to be part of the feeding network. All 3 OC cells are electrically coupled together and interact with all the known buccal feeding motoneurons, as well as with all the modulatory and central pattern generating interneurons in the buccal ganglia. N1 (protraction) phase neurons: Motoneurons firing in this phase of the feeding cycle receive either single excitatory (depolarising) synaptic inputs (B1, B6 neurons) or a biphasic response (hyperpolarisation followed by depolarisation) (B5, B7 motoneurons). Protraction phase feeding interneurons (SO, N1L, NIM) also receive this biphasic synaptic input after OC stimulation. All of protraction phase interneurons inhibit the OC neurons. N2 (retraction) phase neurons: These motoneurons (B2, B3, B9, B10) and N2 interneurons are hyperpolarised by OC stimulation. N2 interneurons have a variable (probably polysynaptic) effect on the activity of the OC neurons. N3 (swallowing) phase: OC neurons are strongly electrically coupled to both N3 phase (B4, B4cluster, B8) motoneurons and to the N3p interneurons. In case of the interneuronal connection (OC<->N3) the electrical synapse is supplemented by reciprocal chemical inhibition. However, the synaptic connections formed by the OC neurons or N3p interneurons to the other members of the feeding network are not identical. CGC: The cerebral, serotonergic CGC neurons excite the OC cells, but the OC neurons have no effect on the CGC activity. In addition to direct synaptic effects, the OC neurons also evoke long-lasting changes in the activity of feeding neurons. In a silent preparation, OC stimulation may start the feeding pattern, but when fictive feeding is already occurring, OC stimulation decreases the rate of the fictive feeding. Our results suggest that the octopaminergic OC neurons form a sub-population of N3 phase feeding interneurons, different from the previously identified N3p and N3t interneurons. The long-lasting effects of OC neurons suggest that they straddle the boundary between central pattern generator and modulatory neurons.

Octopamine is the synaptic transmitter between identified neurons in the buccal feeding network of the pond snail lymnaea stagnalis.

We report the pharmacological properties of synaptic connections from the three octopamine-containing OC interneurons to identified buccal feeding neurons in the pond snail, Lymnaea stagnalis. Intracellular stimulation of an OC interneuron evokes inhibitory postsynaptic potentials in the B3 motoneurons and N2 (d) interneurons, while the synapse between OC and N3 (phasic) interneurons has two components: an initial electrical excitation followed by chemical inhibition. All synaptic responses persist in a saline with elevated calcium and magnesium suggesting that the connections are monosynaptic. Local perfusion of 10(-4) M octopamine produces the same inhibitory membrane responses from these buccal neurons as OC stimulation. These responses also persist in high Mg(2+)/Ca(2+) saline indicating direct membrane effects. The similarities in reversal potentials for the synaptic hyperpolarization evoked on B3 neurons after OC stimulation (-89.0 mV, S.E.M.=14.1, n=10) and the octopamine response of the B3 neurons (-84.7 mV, S.E.M.=6.6, n=6) indicate that increased K(+)-conductance underlies both responses. Bath application of the octopaminergic drugs phentolamine (10(-6) M), epinastine (10(-6) M) or DCDM (10(-4) M) blocks the inhibitory synapse onto B3 or N2 neurons and the chemical component of the N3 response. They also block the octopamine-evoked inhibition of B3, N2 and N3 neurons. NC-7 (2x10(-5) M) has a hyperpolarizing agonist effect (like octopamine) on these neurons and also blocks their chemical synaptic input from the OC interneurons. These results provide pharmacological evidence that the neurotransmitter between the octopamine-immunopositive OC interneurons and its followers is octopamine. This is the first example of identified octopaminergic synaptic connections within the snail CNS.

Polycyclic neuromodulation of the feeding rhythm of the pond snail Lymnaea stagnalis by the intrinsic octopaminergic interneuron, OC.

We have examined the role of the octopamine-containing buccal OC interneuron in the fictive feeding rhythm generated by depolarizing a modulatory interneuron, SO, in the isolated central nervous system (CNS) of Lymnaea stagnalis. Before stimulating the SO, the initial fictive feeding rate was 2.0+/-0.37 bites/min (mean+/-S.E.). When the SO was stimulated, the fictive feeding rate more than doubled, increasing by 5.4+/-2.6 bites/min. Prestimulation of OC facilitates the ability of the modulatory neuron SO to drive fictive feeding 4 s later. Following OC stimulation, the increase in SO-driven feeding rate was 10.8+/-1.6 bites/min, significantly more than when only the SO was stimulated (P<0.02, paired t-test on five preparations). OC activity is not required during the SO stimulation for this enhancement. The maximum of the SO driven rhythm occurs between 6 and 12 s after the end of the OC stimulation at 20 bites/min. This is the maximum feeding rate of intact Lymnaea in sucrose. Facilitation is mimicked by bath applied octopamine at 5 microM. Facilitation is specific to OC interneurons, as the same prestimulation of the electrically coupled neuron N3P (central pattern generator) interneurons does not affect the feeding rhythm. The OC interneuron acts as a long term, polycyclic modulator, which peaks several feeding cycles after the OC activity.

Enhancement of chemotherapeutic drug toxicity to human tumour cells in vitro by a subset of non-steroidal anti-inflammatory drugs (NSAIDs).

The effect on cytotoxicity of combining a range of clinically important non-steroidal anti-inflammatory drugs (NSAIDs) with a variety of chemotherapeutic drugs was examined in the human lung cancer cell lines DLKP, A549, COR L23P and COR L23R and in a human leukaemia line HL60/ADR. A specific group of NSAIDs (indomethacin, sulindac, tolmetin, acemetacin, zomepirac and mefenamic acid) all at non-toxic levels, significantly increased the cytotoxicity of the anthracyclines (doxorubicin, daunorubicin and epirubicin), as well as teniposide, VP-16 and vincristine, but not the other vinca alkaloids vinblastine and vinorelbine. A substantial number of other anticancer drugs, including methotrexate, 5-fluorouracil, cytarabine, hydroxyurea, chlorambucil, cyclophosphamide, cisplatin, carboplatin, mitoxantrone, actinomycin D, bleomycin, paclitaxel and camptothecin, were also tested, but displayed no synergy in combination with the NSAIDs. The synergistic effect was concentration dependent. The effect appears to be independent of the cyclo-oxygenase inhibitory ability of the NSAIDs, as (i) the synergistic combination could not be reversed by the addition of prostaglandins D2 or E2; (ii) sulindac sulphone, a metabolite of sulindac that does not inhibit the cyclooxygenase enzyme, was positive in the combination assay: and (iii) many NSAIDs known to be cyclo-oxygenase inhibitors, e.g. meclofenamic acid, diclofenac, naproxen, fenoprofen, phenylbutazone, flufenamic acid, flurbiprofen, ibuprofen and ketoprofen, were inactive in the combination assay. The enhancement of cytotoxicity was observed in a range of drug sensitive tumour cell lines, but did not occur in P-170-overexpressing multidrug resistant cell lines. However, in the HL60/ADR and COR L23R cell lines, in which multidrug resistance is due to overexpression of the multidrug resistance-associated protein MRP, a significant increase in cytotoxicity was observed in the presence of the active NSAIDs. Subsequent Western blot analysis of the drug sensitive parental cell lines, DLKP and A549, revealed that they also expressed MRP and reverse-transcription-polymerase chain reaction studies demonstrated that mRNA for MRP was present in both cell lines. It was found that the positive NSAIDs were among the more potent inhibitors of [3H]-LTC4 transport into inside-out plasma membrane vesicles prepared from MRP-expressing cells, of doxorubicin efflux from preloaded cells and of glutathione-S-transferase activity. The NSAIDs did not enhance cellular sensitivity to radiation. The combination of specific NSAIDs with anticancer drugs reported here may have potential clinical applications, especially in the circumvention of MRP-mediated multidrug resistance.

Novel interneuron having hybrid modulatory-central pattern generator properties in the feeding system of the snail, Lymnaea stagnalis.

1. We used intracellular recording techniques to examine the role of a novel type of protraction phase interneuron, the lateral N1 (N1L) in the feeding system of the snail Lymnaea stagnalis. 2. The N1Ls are a bilaterally symmetrical pair of electrotonically coupled interneurons located in the buccal ganglia. Each N1L sends a single axon to the contralateral buccal ganglia. Their neurite processes are confined to the buccal neuropile. 3. In the isolated CNS, depolarization of an N1L is capable of driving a full (N1-->N2-->N3), fast (1 cycle every 5 s) fictive feeding rhythm. This was unlike the previously described N1 medial (N1M) central pattern generator (CPG) interneurons that were only capable of driving a slow, irregular rhythm. Attempts to control the frequency of the fictive feeding rhythm by injecting varying amounts of steady current into the N1Ls were unsuccessful. This contrasts with a modulatory neuron, the slow oscillator (SO), that has very similar firing patterns to the N1Ls, but where the frequency of the rhythm depends on the level of injected current. 4. The N1Ls' ability to drive a fictive feeding rhythm in the isolated preparation was due to their strong, monosynaptic excitatory chemical connection with the N1M CPG interneurons. Bursts of spikes in the N1Ls generated summating excitatory postsynaptic potentials (EPSPs) in the N1Ms to drive them to firing. The SO excited the N1M cells in a similar way, but the EPSPs are strongly facilitatory, unlike the N1L-->N1M connection. 5. Fast (1 cycle every 5 s) fictive feeding rhythms driven by the N1L occurred in the absence of spike activity in the SO modulatory neuron. In contrast, the N1L was usually active in SO-driven rhythms. 6. The ability of the SO to drive the N1L was due to strong electrotonic coupling, SO-->N1L. The weaker coupling in the opposite direction, N1L-->SO, did not allow the N1L to drive the SO. 7. Experiments on semintact lip-brain preparations allowed fictive feeding to be evoked by application of 0.1 M sucrose to the lips (mimicking the normal sensory input) rather than by injection of depolarizing current. Rhythmic bursting, characteristic of fictive feeding, began in both the SO and N1L at exactly the same time, indicating that these two cell types are activated in "parallel" to drive the feeding rhythm. 8. The N1L is also part of the CPG network. It Excited the N2s and inhibited the N3 phasic (N3p) and N3 tonic (N3t) CPG interneurons like the N1Ms.(ABSTRACT TRUNCATED AT 400 WORDS)

The hybrid modulatory/pattern generating N1L interneuron in the buccal feeding system of Lymnaea is cholinergic.

This study examines neurotransmission between identified buccal interneurons in the feeding system of the snail Lymnaea stagnalis. We compare the pharmacology of the individual synaptic connections from a hybrid modulatory/pattern generating interneuron (N1L) to a pattern generating interneuron (N1M) with that from a modulatory interneuron (SO) to the same follower cell (N1M). The pharmacological properties of the N1L to N1M and the SO to N1M connections closely resemble each other. Both interneurons produce fast cholinergic EPSPs as judged by the blocking effects of cholinergic antagonists hexamethonium, d-tubocurarine and the cholinergic neurotoxin AF-64A. A slower, more complex but non-cholinergic component of the synaptic response is also present after stimulating either the presynaptic N1L or SO interneurons. This second component of the postsynaptic response is not dopaminergic, on the basis of its persistence in the presence of dopaminergic antagonists ergometrine and fluphenazine and the dopaminergic neurotoxin MPP+. We conclude that, although there has been an evolutionary divergence in function, the modulatory SO and the hybrid modulatory/pattern generating N1L are pharmacologically similar. Neither of them contributes directly to dopaminergic modulation of the feeding activity. These neurons also resemble the N1M protraction phase pattern generating neurons which are cholinergic (Elliott and Kemenes, 1992).

Modulatory role for the serotonergic cerebral giant cells in the feeding system of the snail, Lymnaea. II. Photoinactivation.

1. Photoinactivation of dye-filled neurons was used to examine the modulatory role of the paired cerebral giant cells (CGCs) in the Lymnaea feeding system. 2. Both CGCs were filled with fluorescent dyes. Lucifer yellow was used for "soma" kills and injected via intracellular microelectrodes. CGC axons were retrogradely filled with 5 (6)-carboxyfluorescein (5-CF), through the cut ends of the ventro- and lateral buccal nerves, for "axonal" kills. 3. Irradiation of the CGC soma with a blue laser light (0.5 MW/m2) led to a loss of their recorded membrane potentials and the synaptic responses with their postsynaptic cells (feeding motor neurons). CGC coupling and axonal fluorescence were lost after axonal irradiation. 4. The tonic firing rate of CGC axon spikes in peripheral nerve roots following bilateral soma kills was reduced to approximately 15% of preirradiation levels (n = 2; from 52.5 +/- 3.75 spikes/min to 8.2 +/- 0.95 spikes/min; mean +/- SE) but spike activity was not completely eliminated. 5. The fictive feeding rhythm was evoked by depolarizing a modulatory neuron, the slow oscillator (SO), before and after laser irradiation. Thirty minutes after both the CGCs were irradiated (n = 8), the frequency of the SO-driven feeding rhythm was reduced. Mean fictive feeding rates were reduced from 8.3 to 4.5 cycles/min for soma kills (n = 3) and from 16.2 to 9.6 cycles/min for axonal kills (n = 5; P < 0.05). 6. The results suggest that the CGCs play a modulatory role in controlling the frequency of oscillation of the feeding central pattern generator (CPG) in Lymnaea. The SO could still drive a full fictive feeding rhythm after irradiation but at a reduced rate. At least in the soma kills, the residual spike activity retained in the distal branches of the CGCs appeared sufficient to allow the SO to drive this slow rhythm.

Analysis of the feeding motor pattern in the pond snail, Lymnaea stagnalis: photoinactivation of axonally stained pattern-generating interneurons.

We have photoinactivated identified feeding interneurons known as N1 and N2 neurons. These are pattern-generating neurons that are active in the protraction of the radula and rasping phases, respectively, of the feeding cycle of the pond snail. The N1 or N2 feeding interneurons in the buccal ganglia were filled with the fluorescent dye 5(6)-carboxyfluorescein (5-CF) from the cut end of the nerve that contains their axon. Filling the cerebrobuccal connective (N = 151) stained just one N1 cell in the contralateral buccal ganglion. Filling the postbuccal nerve stained neurons symmetrically in both buccal ganglia (N = 75): only one labeled cell in each ganglion is an N2 interneuron. The feeding rhythm was evoked by depolarizing a modulatory neuron, the SO, located in the buccal ganglia. The axonally filled N1 interneuron was irradiated at its axon in the buccal commissure with blue laser light (intensity of 0.5 MW.m-2). Irradiation of just one N1 completely blocked the feeding rhythm (seven preparations). In seven further preparations, N1 ablation slowed the SO-driven feeding rhythm and weakened the N1 input to the feeding neurons. Irradiation of the cell bodies of both the filled left and right N2 interneurons killed the cells but did not produce any consistent change in the feeding rate (15 preparations). The feeding interneurons and motoneurons still showed the characteristic N2 phase synaptic inputs, so more, as yet unidentified, N2 neurons must be located in other parts of the buccal ganglia. We conclude that the participation of the identified N1 interneurons is essential for the normal feeding pattern while other, still to be identified N2 neurons must be present and must contribute to the feeding rhythm. We suggest that the extra redundancy of the N2 network may be related to the greater necessity of sensory feedback control during rasping than during protraction of the radula.

Cholinergic interneurons in the feeding system of the pond snail Lymnaea stagnalis. I. Cholinergic receptors on feeding neurons.

All the identified feeding motoneurons of Lymnaea respond to bath or iontophoretically applied acetylcholine (ACh). Three kinds of receptors (one excitatory, one fast inhibitory and one slow inhibitory) were distinguished pharmacologically. The agonist TMA (tetramethylammonium) activates all three receptors, being weakest at the slow inhibitory receptor. PTMA (phenyltrimethylammonium) is less potent than TMA and is ineffective at the slow inhibitory receptor, which is the only receptor sensitive to arecoline. At 0.5 mM the antagonists HMT (hexamethonium) and ATR (atropine) selectively block the excitatory response, while PTMA reduces the response to ACh at all three receptors. d-TC (curare) antagonizes only the fast excitatory and the fast inhibitory responses, but MeXCh (methylxylocholine) blocks the fast excitatory and slow inhibitory responses solely. For each of the feeding motoneurons, the sign of the cholinergic response (excitation or inhibition) is the same as the synaptic input received in the N1 phase of the feeding rhythm.

Cholinergic interneurons in the feeding system of the pond snail Lymnaea stagnalis. II. N1 interneurons make cholinergic synapses with feeding motoneurons.

The N1 neurons are a population of interneurons active during the protraction phase of the feeding rhythm. All the N1 neurons are coupled by electrical synapses which persist in a high Mg/low Ca saline which blocks chemical synapses. Individual N1 spikes produce discrete electrotonic postsynaptic potentials (PSPS) in other N1 cells, but the coupling is not strong enough to ensure 1:1 firing. Bursts of N1 spikes generate compound PSPS in the feeding motoneurons. The sign (excitation or inhibition) of the N1 input corresponds with the synaptic barrage recorded during the protraction phase. Discrete PSPS are only resolved in a Hi-Di saline. Their variation in latency and number can be explained by variation in electrotonic propagation within the electrically coupled network of N1 cells. The excitatory postsynaptic potentials (ESPS) in the 1 cell are reduced by 0.5 mM antagonists hexamethonium (HMT), atropine (ATR), curare (d-TC) and by methylxylocholine (MeXCh), all of which block the excitatory cholinergic receptor (Elliott et al. (Phil. Trans. R. Soc. Lond. 336, 157-166 (Preceding paper.) (1992)). The 1 cell EPSPS were transiently blocked by phenyltrimethylammonium (PTMA), which is both an agonist and antagonist at the 1 cell excitatory acetylcholine (ACh) receptor (Elliott et al. 1992). The inhibitory postsynaptic potential (IPSP) in the 3 cell is blocked by bath applications of MeXCh and PTMA, which both abolish the response of the 3 cell to ACh (Elliott et. al. 1992). The effects of the cholinergic antagonists on the response of 4 cluster and 5 cells to N1 stimulation matches their response to ACh (Elliott et al. 1992). It is concluded that the population of N1 cells are multiaction, premotor cholinergic interneurons.

Cholinergic interneurons in the feeding system of the pond snail Lymnaea stagnalis. III. Pharmacological dissection of the feeding rhythm.

The feeding activity of the pond snail Lymnaea stagnalis was stimulated by depolarization of a modulatory interneuron (SO) or of a N1 pattern-generating interneuron. The cholinergic antagonists phenyltrimethylammonium (PTMA), methylxylocholine (MeXCh), hexamethonium (HMT) and atropine (ATR) were applied at 0.5 mM in the bath and their effects on the rhythmic feeding pattern were monitored. Each of the antagonists slowed or blocked the feeding rhythm. The block was due to interference in the pattern generating network, not to disturbance of modulatory inputs. The experimental results favour a model in which the alternation of protraction (N1) and retraction (N2) phases occurs by recurrent inhibition. The results would be more difficult to explain on the reciprocal inhibition model. When all the N1 output was blocked, the N1 neurons fired rhythmic bursts endogenously.

Vital staining with 5(6)-carboxyfluorescein in the feeding system of Lymnaea and the selective photoinactivation of the filled neurones.

Lymnaea feeding interneurones were stained with 5(6)-carboxyfluorescein by cutting the nerves containing their axons. They were located under low-intensity blue light and were healthy with normal pharmacology. The interneurones were selectively killed by high-intensity blue light.

Photoinactivation of neurones axonally filled with the fluorescent dye 5(6)-carboxyfluorescein in the pond snail, Lymnaea stagnalis.

We describe a new, simple and reliable technique to fill molluscan neurones from their cut axons with sufficient fluorescent dye for photoinactivation experiments. The fluorescent dye 5(6)-carboxyfluorescein (5-CF) travels quickly up the nerves of the gastropod mollusc, Lymnaea stagnalis into the buccal ganglia and fills the cell bodies in 1-3 h. 5-CF filled neurones can be located in the intact ganglia with low intensity blue light. Impalement shows that they are alive and show normal resting, action and synaptic potentials. Intense laser light (wavelength 442 nm, intensity 0.5 MW.m-2) kills all the 5-CF filled cells in less than 5 min in laboratory reared snails. Unstained neurones are not killed. 5-CF fills neurones quicker than Lucifer yellow (LY) when the dye is applied axonally. Neurones stained with Lucifer yellow do not contain sufficient dye to be killed with 5 min laser illumination, but this irradiation reduces the membrane resistance to less than 25%.