J Comp Physiol A (1991) 169:541-555

Jk~mal

elf Neural,

Physiology A

~ " Physiology

9 Springer-Verlag 1991

Respiratory behavior in the pond snail Lymnaea stagnalis I. Behavioral analysis and the identification of motor neurons N.I. Syed*, D. Harrison, and W. Winlow Department of Medical Physiology, University of Leeds, Leeds, LS2 9NQ, England, U.K. Accepted August 6, 1991

Summary. This study describes the neural basis of respiratory behavior in a pulmonate mollusc, Lymnaea stagnalis. We describe and identify muscles of the respiratory orifice (pneumostome) and mantle cavity as well as relevant motor neurons innervating these muscles. All of these identified motor neurons are active during spontaneously occurring respiratory behavior and a sporadically occurring synaptic input, termed Input 3, controls the activities of these motor neurons. This spontaneous input can also be recorded from isolated brain preparations, suggesting that the respiratory motor program is generated centrally. However, evidence is also presented that in semi-intact preparations the role of peripheral feedback is important for the initiation and termination of respiratory behavior in Lymnaea. Key words: Central pattern generator Motor neurons - Mollusc - Respiratory behavior - Neuronal circuits Peripheral feedback

Introduction

Invertebrates have proved to be useful models for gaining insights into the neuronal basis of respiratory behavior. In contrast to vertebrate preparations where it has not been possible to identify specific individual motor neurons involved in ventilatory or respiratory behavior, individual respiratory motor neurons and relevant interneurons have been identified in several invertebrates, including locust (Burrows 1975a, b, 1982), and molluscs (Byrne 1983; Janse et al. 1985; van der Wilt et al. 1987; Koester 1989; Alevizos et al. 1989). In pulmonate molluscs, respiration occurs partially across the somatic epidermis but mainly via the lung. * P r e s e n t a d d r e s s (and to whom offprint requests should be sent): Department of Medical Physiology University of Calgary, HSC 3330 Hospital Drive NW Calgary, Alberta, T2N 4N1, Canada

Lung ventilation is an irregular but rhythmic behavior which involves opening and closing the lung orifice (the pneumostome) and evacuating the lung. The pneumostome of the freshwater pond snail Lymnaea stagnalis (L.) is situated on the underside of the right part of the mantle and consists of at least 3 muscle systems that control its opening and closing movements (de Vlieger et al. 1976). These snails make periodic visits to the water surface in order to replenish their air supply. Upon contacting the surface, the pneumostome opens slowly and remains open for several seconds. Air is expelled from the lung cavity during these opening movements. After expiration the muscles surrounding the mantle cavity relax and air is drawn into the lung cavity by negative pressure. Closure of the pneumostome soon follows. Pneumostome opening and closing can be repeated several times before the animal submerges again. The overall rate of oxygen uptake in pond snails such as Lymnaea stagnalis and Planorbis corneus is directly related to the pOz of the water in which they are contained (Jones 1961). In both of these snails the total oxygen consumption is approximately constant over a wide range of dissolved pO2. This is due to the fact that pulmonary ventilation is increased at low pO 2 and skin respiration is increased when water p O 2 is high. Planorbis, which occupies a deeper ecological habitat than Lymnaea, uses its lung mainly as an oxygen store (in combination with hemoglobin), while Lymnaea maintains its pulmonary p O 2 at a higher level by making frequent dives of shorter duration (Bekius 1972). To study the neural basis of respiratory behavior in Lymnaea the effects of changes in the external p O 2 o n central neurons were tested by Janse et al. (1985). The interneuron visceral dorsal 4 (V.D.4) and visceral E group (V.E group) neurons, both located on the dorsal surface of the visceral ganglion of Lymnaea were implicated in the opening movements of the pneumostome (Janse et al. 1985; van der Wilt et al. 1987). These authors also reported that electrical stimulation of a neuron located in the right parietal ganglion caused closure of the pneumostome. However, they did not record from these central neurons

N.I. Syed et al. : Respiratory behavior in Lymnaea

542

during spontaneously occurring pneumostome opening and closing movements, nor did they examine any interactions between these identified neurons. In the present study we have analyzed the ventilatory behavior of freely moving Lymnaea using video recordings and still photography. A newly developed semiintact preparation has enabled us to identify the motor neurons which are involved in the pneumostome opening and closing movements. We have also dentified those motor neurons that cause contractions of the mantle cavity musculature. We show that these motor neurons, driven by an identified input (Input 3), are active during spontaneously occuring respiratory behavior and that their activation in a semi-intact preparation causes contraction of the appropriate muscles. We also demonstrate A

B

/

TENTACL

a role for peripheral feedback in this behavior. In the companion paper (Syed and Winlow 1991) we describe specific interneurons which control the rhythmic activities of the respiratory motor neurons. Materials and methods Specimens of Lymnaea stagnalis obtained from animal suppliers (Blades Biological, Sussex) were maintained at 10-16 oC in aerated, filtered natural pond water and fed lettuce. All experiments were performed on snails of 1-4 g wet weight. Preparations were bathed in standard Lymnaea saline buffered to pH 7.9 using HEPES as described (Benjamin and Winlow 1981). Several modified salines were also used. For zero Ca2+/high Mg z+ saline, calcium in the normal saline was replaced by magnesium, the latter being raised from 2 to 6 raM. For high Ca2+/high Mg z+ saline, concentrations of both divalent cations were raised sixfotd: Ca z+ to 24 mM, Mg z§ to t2 mM. Intracellular recordings from either isolated brains or semi-intact preparations were made using standard electrophysiological techniques. Isolated brain preparations were made according to the methods described earlier by Benjamin and Winlow (198 I), except that either no protease or very little protease was used and the inner perineurat sheath was removed by microdisseetion.

Freely movin9 intact animals

FOOT ~

In some cases intact animals were anaesthetized in 2% halothane and fine nickel wires attached to the pneumostome. Alter recovery the animals moved freely around the tank with the wires attached loosely to tension transducers. However, whenever the animal surfaced the nickel wire was tightened and pneumostome movements were recorded.

~,~j"

The semi-intact preparation To prepare semi-intact preparations, animals were first anaesthetized by bubbling 2% halothane into pond water as described by Girdlestone et al. (1989). Snails were assumed to be anaesthetized when they either dropped to the bottom of the chamber or exhibited no withdrawal response when stimulated with a glass rod. Anaes-

P,C.M

V.J Celt

PNEUMOSTOME OPENING MOVEMENTS

V.KCell

tllit

,~ ~ ~ ~ .......... r "~

tiI}i{{fill

,

m~H,

~~

~,

Fig. 1A-C, A diagrammatic representation of the semi-intact preparation, A An undissected Lymnaea. B The anterior half of the shell was removed from an anaesthetized snail and fine nickel suspension wires (dotted lines) were attached to various parts of the animals' body. A dorsal midline incision (arrow) was made which divided the body wall musculature into halves, i.e., left and right body wall (L.B.W. and R,B. IV.). C Simultaneous muscle tension recordings from the pneumostome muscles and intracellular recordings from central neurons were made. The tension transducer is shown hooked to the pneumostome muscle and the locations of two intracellular microelectrodes are also indicated. Sample recordings are shown on the right. For the sake of simplicity, some parts of the apparatus have been omitted, e.g., the nickel suspension wires and the wax-covered spatula used to stabilize the brain during recording. P.C,M. (Pneumostome Closer Muscle)

J300mg B

15s

Fig. 2A, B, Muscle tension recordings from the pneumostome of freely' moving snails, The tension transducer was attached by fine nickel wires to the pneumostome of anaesthetized animals which were then allowed to move freely on recovery. Opening movements of the pneumostome (upward deflections) were monitored by tightening the nickel wire whenever the animal came to the water surface, A Pneumostome opening movement from a freely moving snail kept in well aerated pond water. Note a single opening movement was carried out, B Pneumostome opening and dosing movements recorded from a snail kept in an unaerated environment. Upon surfacing, this snail carried out several opening movements which were of short duration compared to A

N.I. Syed et al. : Respiratory behavior in Lymnaea

Fig. 3A-F. Photographs of Lymnaea during various stages of the respiratory behavior. A Snails make frequent visits to the water surface in order to ventilate their lungs. Having reached the water surface, the pneumostome opens (arrow) and air is expelled from the lung cavity. B The open orifice of the pneumostome viewed from above the water surface. Arrow indicates area where loops of nickel wires were attached to monitor opening movements. C - F Various stages during the opening and closing movements of the pneumo-

543

stome. The intact snail was immobilized by attaching its shell to a wax platform and the respiratory movements were observed. In order to open and close the pneumostome, the foot of the animal must first be extended out of the shell (arrow). Panels C - E show the gradual opening of the pneumostome; panel F shows a fully opened pneumostome. Time from C - F 15 s. P Pneumostome; S shell; F foot; L lip; T tentacle. Scale bar 5 mm

544

N.I. Syed et al. : Respiratory behavior in Lymnaea

thetized animals were transferred to a chamber designed for these experiments (Syed 1988). The anterior half of the shell and body wall were dried using Kleenex laboratory tissues or warm air from a blow dryer. Nickel wires were attached to the junction where the body wall meets the foot musculature using tissue glue (cyanoacrylate adhesive: Loctite (U.K.) Ltd) (Fig. 1A). Wires were also glued to the shell. Fine nickel wires (diameter 0.01 mm: California Fine Wire Company, California) were wrapped in loops around the animal's shell in a manner similar to the Tritonia preparation described by Willows et al. (1973). A second wire loop was made around the mouth and tentacles. Wires were then attached to the loops encircling the mouth and the shell to restrain the animal anteriorly and posteriorly. Several small pieces of wire were also glued to the loop of wires around the top edge of the foot (Fig. 1B). All these wires were then extended and attached to the wall of the chamber either by Plasticine or other adhesive materials, allowing the animals to be suspended in the saline. Under these circumstances rhythmic, locomotor and respiratory behaviors, similar to those seen in freely moving animals, could be observed. Once the animal was firmly suspended, a small mid-dorsal body incision was made and the body wall was gently retracted with blunt hooks attached to the nickel wires. The connective tissue sheath which attaches the brain to the musculature of the animal was carefuly microdissected. The ventral (subesophageal) ring of ganglia was then raised by inserting a wax-covered spatula held on a micromanipulator (Syed 1988). In some experiments, either the cerebral commissure or the esophagus and penis were ablated to allow access to the pedal, visceral or parietal neurons. In all other experiments, however, visceral organs (e.g., heart, kidney and lung) were left intact. Depending on the experiments, one end of the nickel wire was glued to either the insertions of the pneumostome opener muscle, or closer muscle, or to the external side of the opening of the pneumostome (Fig. 1C). A cotton thread was attached to the posterior end of the shell and connected to a micromanipulator. It was then possible to expose the opening of the pneumostome to air by lifting the shell above the water level. CENTRAL GANGLIONIC RING OF LYMNAEA STAGNALIS DORSAL VIEW

6

Fig. 4. Diagrammatic representation of the CNS of Lymnaea stagnalis showing the location of central neurons used in the present study. The figure shows the dorsal view of the CNS except that buccal ganglia have been omitted and the cerebral commissure has been sectioned. The cerebral ganglia are displayed ventral side up. Individually identifiable neurons are numbered (e.g., R.Pe.D.I) whereas identified neuronal clusters are given a letter (e.g., R.P.A Group, cells) according to the convention of Slade et al. (1981). Shaded areas represent the degree of whiteness exhibited by specific neurons. 1, 2 left and right cerebral ganglia; 3, 4 left and right pedal ganglia; 5, 6 left and right pleural ganglia; 7, 8 left and right parietal ganglia; 9 visceral ganglion; St statocyst; R.Pe.D.1 right pedal dorsal 1; R.P.A right parietal A group neuron; V.G visceral G group neuron; V.H, L J, K visceral H, I, J, K cells

Upon completion of surgery, the animals were allowed to recover slowly from the anaesthesia. After several washes in normal saline, all snails recovered completely in 10-20 min. Intracellular recordings were made from neurons and pneumostome muscle tension was recorded via tension transducers during the course of respiratory, locomotor, or whole body withdrawal behaviors. However, due to anatomical constraints, it was not possible to record the tension simultaneously from both pneumostome opener and closer muscles. Intracellular recordings were made from individual muscle fibres with glass microelectrodes (40-50 Mfl). During the course of spontaneously occurring respiratory behavior, as recorded from semi-intact preparations, either mechanical or photic stimuli (see Ferguson 1984) were applied to the opening of the pneumostome to test its effects on central neurons. Soap bubbles were placed at the opening of the pneumostome to monitor the movement of the air in and out of the lung cavity. Deflation and inflation of the lung cavity were monitored visually.

Results

Behavioral investigations Our observations of the respiratory behavior in freely moving Lymnaea revealed the following pattern of respiratory cycles: >

Fig. 5A-F. The comparative morphology of respiratory motor and non motor neurons as revealed by Lucifer yellow. Lucifer yellow (CH) filled electrodes were used to impale various neurons in the visceral and parietal ganglia while making simultaneous intracellular recordings from various muscle fibers. These neurons were first electrically stimulated to observe their effects on target muscles and then filled with Lucifer yellow according to the methods of Syed and Winlow (1989). The V.J.V.K and R.P.A group neurons were identified on the basis of positions in the ganglia and their previously identified synaptic inputs (Benjamin and Winlow 1981). A Photomicrograph of a Lucifer yellow injected visceral J cell, which when stimulated, caused 1 : 1 EJPS in the pneumostome opener muscle fibers (P.O.M.). This electrophysiological data supports the morphological data since the axonal branches of this neuron project via genital nerve (12) and right internal parietal nerves (13) which innervate the pneumostome and mantle cavity. B Morphology of another previously identified V.J cell, which when stimulated intracellularly, had no effects on pneumostome or mantle cavity musculature. Furthermore, when filled with Lucifer yellow, this V.J cell did not have axonal projections in nerves innervating either pneumostome or mantle cavity muscles. C V.K cell which produced 1 : 1 EJPS in the pneumostome closer muscle fiber (P.C.M.) is shown here to project via nerves innervating the pneumostome musculature. In addition this cell had axonal projections via the intestinal nerve (10) which innervates the heart; its effects on heart muscles were not investigated in the present study. D This V.K cell had no effect on pneumostome closer muscle fibers, nor did it project to the periphery via nerves innervating these muscles. E Photomicrograph of a R.P.A group neuron which caused 1 : 1 EJPS in mantle cavity muscle fibers. The cell is shown here to project via nerves innervating the mantle cavity and pneumostome musculature. In addition, it was also found to project via other nerves, such as the intestinal nerve (10), which innervates the heart. F Morphology of another R.P.A group neuron that, although had axonal projections in nerves innervating the mantle cavity and pneumostome musculature, did not have motor neuronal effects on these muscles. The exact function of this particular neuron is unknown. Scale bars 50 gm. Abbreviations: V visceral ganglion; R.P right parietal ganglion. Nerves are numbered according to the convention of Slade et al. (1981): 10 intestinal nerve; 11 anal nerve; 12 genital nerve; 13 right internal parietal nerve

N.I. Syed et al. : Respiratory behavior in Lymnaea

545

N.I. Syed et al. : Respiratory behavior in Lymnaea

546

A R.P.A Group

R.Pe.D1

V.H Cell

V.J C e l l

=p.3

30mY 15s

B V.G Group

V.K Cell

V.I Cell

V.J Cell

ip.3

30rnV 15s

Fig. 6A, B, A spontaneously occurring compound postsynaptic potential, Input 3 (Ip. 3), causes a rhythmic discharge in its follower cells located in the visceral, right parietal, and right pedal ganglia. In freshly isolated brain preparations, the Ip. 3 discharge is often spontaneously active followed by periods of quiescence. A Input 3 has repetitive excitatory effects on all of the cell types recorded in this preparation. The first of these discharges is indicated by a bar. The mantle cavity motor neuron (R.P.A group, top trace), giant dopamine neuron (R.Pe.D.1, second trace), visceral H cell (V.H cell, third trace) and pneumostome opener muscle motor neuron (V.J cell, bottom trace) are all excited by Ip. 3. Although the V.H cell receives excitation from Ip. 3, the motor neuronal role of this

cell in the respiratory behavior is as yet unknown. B Simultaneous intracellular recordings were made from pneumostome closer and opener muscle motor neurons. Spontaneously occurring Ip. 3 inhibited a V.G. group neuron and a V.K cell while exciting a V.J Cell, which are pneumostome closer and opener muscle motor neurons, respectively. In addition V.I cell is excited concurrently with a V.J cell and is also suspected of being motor neuron. The bar represents the first of the series of Ip. 3 discharges. The spontaneous activity of the V.J cell in this preparation (bottom trace) was reduced by its selective hyperpolarization to augment the excitatory effects of Ip. 3, and therefore, the attenuation of spikes normally seen during Ip. 3 is absent

N.I. Syed et al.: Respiratory behavior in Lymnaea

547

(1) When snails were maintained in welt aerated pond water, with continuous air perfusion, the frequency o f respiratory cycles (i.e,, opening and closing movements o f the pneumostome) was low. Under these environmental conditions, Lymnaea visited the water surface at an average rate o f once every 40 rain (n = 200 trials), U p o n reaching the water surface, the pneumostome opened slowly and gradually and remained open for at least 30 s ( n = 173 trials) (Fig. 2A). This opening movement, during which the air was expelled from the lung cavity (expiration), was followed by a gradual, albeit faster, closure. Prior to the closure o f the pneumostome air rushed into the lung cavity (inspiration). If the duration o f this opening and closing cycle was less than 20 s it was often repeated before the animal left the water surface. (2) When kept in an unaerated environment, snails usually either stayed close to the water surface or made frequent visits at an average rate o f once every 20 rain (n = 153 trials). Having reached the water surface, they

A

carried out several respiratory movements. These opening and closing cycles had short intervals, averaging about 20 s (Fig. 2B). (3) When kept out o f water, the opening and closing cycles were very frequent and o f shorter intervals than described above. Under these experimental conditions the excessive amount o f mucus and the constant contact o f the nickel wires with the pneumostome disturbed the behavior (in an aquatic environment these wires are supported by the water and therefore do not interfere with the pneumostome opening and closing). Thus, it was not possible to obtain tension recordings for this category of respiratory behavior. Furthermore, locomotion was regularly interrupted by the respiratory movements because due to anatomical constraints the animals cannot locomote and breath at the same time. In order for pneumostome opening and closure to occur, the head-foot complex o f the animal must be extended (Fig. 3A-F). Mechanical stimulation o f the

B

P.O.M

P.C.M | 200g

__~~_.=~__

V.J Cell

t

V.K Cell

,,,.•30mV 15s

t

C P.O.M

200g R.P.A Group Ip.3

1p.3

Ip.3

30mV 15s D M.C.M

R.P.

~

_

1200rag

....I 30mV 15s

Fig. 7A-D. Effects of pneumostome opener, closer, and mantle cavity muscle motor neurons on the appropriate muscles recorded from semi-intact preparations. A Intraceltutar stimulation of a V.J Ceil in a semi-intact preparation (at arrows) caused the contraction of the pneumostome opener muscle (P.O.M.) as recorded by a tension transducer. B Similar intracellular stimulation of a V,K Cell Jn another preparation caused contraction of the pneumostome closer muscle (P.C.M.). Scale bar for A same as in B. C Simultaneous intracellular and pneumostome muscle tension recordings were made from P.O.M. (top trace) and a R.P.A group neuron during spontaneously occurring respiratory drive. Input 3 (Ip. 3) caused the opening of the pnemnostome (arrows) via its effects on motor neurons, and then caused the activation of the R.P.A group neuron, which in turn caused contraction of the mantle cavity (not shown here). D The direct effects of an R.P.A group neuron on the mantle cavity muscle (M.C.M.) were tested. The electrical stimulation of this neuron (at arrow) caused the contraction of the mantle cavity muscle

548

N.I. Syed et al. : Respiratory behavior in

our preparations, we have called these muscle layers the pneumostome opener muscle (P.O.M.) and the pneumostome closure muscle (P.C.M.), respectively. Similarly, the dorsal musculature, which consists of a plate of muscle bundles and is inserted at the column of the shell, has been previously described by Bekius (1972). This dorsal musculature runs through the lung roof and terminates in the dorsal wall of the head-foot. Since these muscle bundles run through most of the mantle cavity, we have termed them mantle cavity muscles (M.C.M.) in the present study. The P.O.M., P.C.M., and M.C.M. are multiply innervated via 3 nerves: the right superior and inferior parietal nerves emanating from the right parietal ganglion, and the anal nerve emanating from the visceral ganglion (Slade et al. 1981). It is not yet clear whether individual muscle fibres are multiply innervated.

foot causes contraction of the head-foot complex, resulting in abrupt closure of the pneumostome. Careful examination of the animal during respiratory movements revealed that during the opening movements of the pneumostome locomotion is temporarily halted. Cilia present on the sole of the foot, which provide the main propulsive locomotor force, stop beating during the opening movements; ciliary beating is resumed soon after pneumostome closure (Syed, unpublished data). Mechanical stimulation of the pneumostome following removal of the central nervous system causes very weak and localized contractions of the pneumostome musculature. This response suggests that the peripheral nervous system is also involved in controlling some aspects of the respiratory behavior. Muscles and motor neurons involved in the respiratory behavior

Pneumostome opener (V.J cells), closer (V.K cells) and mantle cavity (R.P.A 9roup) motor neurons

Two muscle groups described earlier by de Vlieger et al. (1976) as toplayer and underlayer muscles are involved in the pneumostome opening and closing movements. In

A

NORMAL

8ALINE

LOW

24" Ca

Lymnaea

In this study we found that the respiratory motor neurons of Lymnaea lie in a number of previously identified

.,G.

M .=§

WAS.

P.O.M

~_~____~

110mV

V.J Cell

/~

B

NORMAL

,

I 20mV 10s

5s

SALINE

LOW

24' Ca HIGH

2+ Mg

WASH

P.C.M _ _

~

1lomv

~ l l 0 m V

V.Ce,, 20mV 1S Fig. 8A, B. Chemical nature of connections between a V.J cell and P.O.M. fibers and a V.K cell and P.C.M. fibers. A Induced action potentials in a V.J cell evoked 1 : 1 EJPS in P.C.M. fibers, which were reversibly reduced in amplitude when Ca 2 § dependent release of neurotransmitters was blocked by replacing the Ca 2 § in the bathing saline with high Mg 2+. B The V.J motor neuron evoked

I 10mV lOs EJPS in P.C.M. fibers. Spontaneous action potentials in a V.J cell evoked EJPS in P.C.M. fibers. These EJPS were reversibly reduced in amplitude when preparations were bathed in Lymnaea saline containing low Ca 2+ and high Mg 2§ concentrations. The first and last panels represent pre- and post-treatment saline controls

N.I. Syed et al. : Respiratory behavior in

Lymnaea

549

groups of visceral and parietal neurons. The H, I, J and K cells of the visceral ganglion were first identified by Benjamin and Winlow (1981). These neurons form a heterogeneous cluster on the dorsal surface of the visceral ganglion near the visceral-parietal connectives (Fig. 4). Their somata range in diameter from 40-100 ~tm and their color varies from orange to white. Recently, we were able to identify H, I, J and K cells individually on the basis of their color, size, axonal projections and electrophysiological properties (Syed 1988). Only those cells among the H, I, J, K cell clusters that have axon projections in the anal nerve, and in the right superior and inferior parietal nerves which innervate the pneumostome and mantle cavity musculature, were found to be respiratory motor neurons (Fig. 5A, C and E). All the other cells did not project via nerves innervating respiratory musculature (Fig. 5B, D, F) and are perhaps heart motor neurons as described by Buckett et al. (1990). We found that in freshly dissected isolated brain preparations an identified, well characterized Input, the input 3 (Ip. 3) (Benjamin and Winlow 1981) caused a rhythmic patterned discharge in all the motor neurons described above and also in some other identified neurons. When spontaneously active, Ip. 3 excited visceral H, I and J cells, parietal A group cells and the giant

NORMAL

SALINE

LOW

Ca2 + H I G H

dopamine cell R.Pe.D. 1 (Fig. 6A and B) while inhibiting visceral K cells and V.G group neurons (Fig. 6B). The discharge pattern induced in these cells by Ip. 3 is unique and no other such input has so far been discovered in the CNS of Lymnaea. To ascertain the function of visceral H, I, J, K cells and parietal A group neurons, we used the semi-intact preparation described earlier. Electrical stimulation of a V.J cell and a V.K cell in semi-intact preparations caused the contractions of P.O.M. and P.C.M., respectively (Fig. 7A, B). Furthermore, when intracellular recordings were made from motor neurons and muscle fibers, action potentials in motor neurons evoked 1 : 1 excitatory junctional potentials (EJPS) in appropriate muscles (Figs. 8A, B, 9A). The effects of V.J and V.K cells on appropriate muscles were significantly reduced when Ca 2+ in the normal bathing solution was replaced with other divalent cations such as Mg 2§ or Co 2+ (Fig. 8A, B). In isolated brain preparation, we showed (Fig. 6) that all the respiratory motor neurons are rhythmically driven by Ip. 3. We next tested in semi-intact preparations the role of Ip. 3 in motoneuronal control of the respiratory musculature. Spontaneously occurring Ip. 3 discharges induced the opening of the pneumostome via its follower V.J cells. When the pneumostome was fully opened

2+ Mg

WASH

i10mv

M.C.M

R.P.AGroup

, omv lOs

Fig. 9. Chemical nature of connections between R.P.A group neuron and M.C.M. fibers. Spontaneously occurring or induced action potentials in a R.P.A group neuron also produced 1:1 EJPS in

HIGH

M.C.M. fibers. These were blocked when the Ca 2+ in the normal saline was replaced by Mg 2 +. The first and last panels represent preand post-treatment controls

2§ Ca SALINE

A

B

C

P.O.M

P.C.M

M.C.M

~

V.JCell j~

IlOmV

V.KCell__~ _ R.P.AG r o ~

t

30mY

30mV los

10s

Fig. 10A-C. Monosynaptic nature of connections between motor neurons and appropriate muscles. The connections between a V.J cell and P.O.M., a V.K cell and P.C.M. and a R.P.A group neuron and M.C.M., fibers persisted when normal Lymnaea saline was replaced with high Ca / + and high Mg 2 + saline

55O

N.I. Syed et al. : Respiratory behavior in Lymnaea

A P.O.M

1200mg

V.J C e l l

lJllllrll.

r F Ff.r.f.,,

Ip.3 V.K Cell

|

r,...............

1IlILI

,,l~flllllHllimrlrllrlr]l . . . . . .

I1[ I/HIHlll

~r..................... IIII~IfflTIVIIIlIlIIYIIll ~'~

B P.C.M f ~

200mg

V.J.Cell

iA

111

V.K Cell

lllIHIlfrll ~-~

f ~

I1NIII' I l l l l l '

lrlll r

I11111 I

Fig. l l h , B. Cyclical motor neuronal and muscle tension activity recorded during spontaneously occurring respiratory cycles, in a minimally dissected semi-intact preparation. A Simultaneous muscle tension and intracellular recordings were made from the pneumostome opener muscle (P.O.M.), a V.J cell. Spontaneous activity in a V.J cell maintains the muscle tonus in P.O.M, however, when the V.J cell was further activated, by Ip. 3 (see bar) it caused pneumostome opening, (as presented by the upward deflection)

during respiratory cycles, one of the R.P.A. group neurons activated by Ip. 3 fired a burst of action potentials (Fig. 7C) and caused the contraction of the M.C.M. (not shown here). Mantle cavity muscle contraction was closely followed by contraction of the lung, which resulted in air being expelled. The movement of air into and out o f the lung was monitored by placing soap bubbles at the opening of the pneumostome. Electrical stimulation o f two specific neurons o f the R.P.A group (the morphology of one o f these neurons is shown in Fig. 5E) described by Winlow and Benjamin (1976) resulted in

15s whereas the V.K cell received inhibition from Ip. 3. Scale bars same as B. B Due to anatomical constraints it was not possible to record simultaneously from P.O.M. and P.C.M. Therefore, in this preparation, muscle tension was recorded only from P.C.M. During the course of Ip. 3 (bar) the V.J cell receives excitation from Ip. 3, whereas the V.K cell is activated by V.D.4 (see companion paper), which leads to the contraction of P.C.M.

contraction of the M.C.M. (Fig. 7D), suggesting the direct involvement of these neurons in controlling the mantle cavity musculature. This was further confirmed by making simultaneous intracellular recordings from a R.P.A group neuron and M.C.M. fiber. Action potentials in the R.P.A group neuron induced 1:1 EJPS in individual M.C.M. fibers which were reversibly blocked by replacing Ca 2§ in the saline with Mg 2+ (Fig. 9B). Furthermore, 1 : 1 EJPS were maintained for all neuron types in high Ca 2 + saline, suggesting that the connections are monosynaptic (Fig. 10). To summarize, it appears

N.I. Syed et al. : Respiratory behavior in Lymnaea

551

A P.O.M 200mg

V.J Cell

V.K Cell

Ip.3

t B P.O.M I 200mg

V.J Cell

V.K Cell

q l II.

Ip.3

rllllTl~[[lllTrll[rlll[ ~' " e"

mill

hi[Ill

III IlHH H

!',.i. . . . . ,,,,~,I~IIIIIIIIIHIIIIIIIIIIIHI1NIIIITIIII[IIlIItlIIITI[II[II[

IllpIlllll

| 20mY

Fig. 12A, B. The role of peripheral sensory feedback in the initiation and termination of the respiratory cycle. A In a previously quiescent semi-intact preparation, lifting of the pneumostome above the saline surface (at arrow), initiated respiratory activity. The evidence for this activation was obtained from Ip. 3 (bar) and subsequent muscle tension recordings made from the P.O.M. B Respiratory movements were terminated when the pneumostome was lowered once again below the saline surface (open ar-

row)

15s

that only some specific cells located a m o n g the V.J cells are P.O.M. m o t o r neurons, visceral K cells are P.C.M. m o t o r neurons, and right parietal A group neurons are M.C.M. m o t o r neurons.

interneuron(s), causes contraction of the P.O.M. which contributes to the opening o f the p n e u m o s t o m e (Fig. 11A). During these opening movements a P.C.M. m o t o r neuron (V.K cell) is inhibited, but fires shortly after the Ip. 3 discharge to cause the contraction o f the P.C.M., and thus the closure of the p n e u m o s t o m e (Fig. l lB).

Motor neurons are active during spontaneously occurring respiratory cycles The results described above imply a role for Ip. 3 in the opening movements of the p n e u m o s t o m e since it excites P.O.M. m o t o r neurons (V.J cells) and M.C.M. m o t o r neurons (R.P.A group) while inhibiting P.C.M. m o t o r neurons (V.K cells). The identified interneuron V.D.4, has effects on these m o t o r neurons exactly opposite to those of Ip. 3. V.D.4 is implicated in the closure movement of the pneumostome. Ip. 3 interneuron(s) and V.D.4 have reciprocally inhibitory connections and a detailed account of their interactions and functional significance is described in the c o m p a n i o n paper (Syed and Winlow 1991). Here we demonstrate that during the active respiratory cycle, a V.J cell, driven by the Ip. 3

The role of peripheral feedback in the inith~tion and termination of respiratory behavior Neither Ip. 3, nor p n e u m o s t o m e opening movements, were observed in those semi-intact preparations where the mantle cavity and p n e u m o s t o m e were covered with saline. In preparations where the p n e u m o s t o m e was lifted above the surface o f the bathing saline via a thread attached to the shell, spontaneous respiratory movements occurred (Fig. 12A). These movements were terminated once the p n e u m o s t o m e was submerged (Fig. 12B). This result indicates that sensory feedback has an important role in initiating and terminating the respiratory movements.

552 Discussion

Respiratory behavior has been investigated in a number of molluscan species, but few of these studies have attempted to identify the motor neurons and interneurons involved. Perhaps the most successful of these attempts was a recent description of the neural circuitry underlying respiratory pumping in the opisthobranch Aplysia californica by Koester and co-workers (Byrne and Koester 1978; Koester 1989; Alevizos et al. 1989). These authors suggest that respiratory pumping in this marine snail is controlled by a neuronal network consisting of chemically and electrically coupled interneurons. These interneurons make direct chemical connections to motor neurons that mediate the respiratory pumping (Koester 1989). Respiratory pumping across the gill of Aplysia may be similar in terms of the underlying neural circuitry but is not directly comparable to respiration that occurs in pulmonate (lung breathing) snails such as Lymnaea stagnalis or Planorbis corneus. Understanding the neural basis of respiratory behavior in these pulmonate snails may enable us to gain further insights into the respiratory behavior of vertebrates, particularly that of diving mammals (e.g., whales, seals), which make visits to the water surface to satisfy their respiratory demands. Some aspects of the respiratory behavior of pulmonate snails were investigated earlier by Jones (1961). In examining ventilation in Lymnaea and Planorbis, he found species variation in both the volume and composition of pulmonary air during submersion when simultaneous pulmonary and cutaneous oxygen uptake (as a function of dissolved pOz) were compared. In both of these species, the lung was found to be of considerable importance. A major difference between the two species, however, was that hemoglobin present in the blood of Planorbis helped the animal to store pulmonary oxygen and allowed deeper and longer dives as compared with Lymnaea, which uses hemocyanin as its oxygen carrier. Since the lack of hemoglobin in Lymnaea forces this animal to make frequent visits to the water surface, the increased frequency of respiratory movements make Lymnaea a better model than Planorbis to investigate spontaneously occurring respiratory behavior. Jones (1961) limited his study to behavioral and biochemical parameters associated with respiration in these snails and did not undertake neurophysiological experiments. In the present study, we investigated the neural correlates of the respiratory behavior in Lymnaea by identifying the motor neurons involved in this behavior.

The respiratory behavior of Lymnaea Rather than making simple observations of respiring snails, as Jones (1961) had done, we filmed the behavior of freely moving snails while recording movements of different parts of the animal body with tension transducers (e.g., see Fig. 1). This enabled us to analyze quantitatively the spatial and temporal relationships between different parts of the animal body during movements,

N.I. Syed et al. : Respiratory behavior in Lymnaea and also allowed us to identify various muscles involved in respiration. These behavioral observations revealed that respiration in Lymnaea is a discontinuous act which depends upon environmental conditions. Nevertheless, when it occurs, it is manifested as rhythmical and cyclical movements. Even in terrestrial molluscs, such as Limax maximus, the opening and closing movements of the pneumostome are rhythmical and are affected by such parameters as hemolymph osmolarity and the presence of peptide hormones (Dickinson et al. 1988). It is logical to expect that having reached the water surface the snail would first expel air (principally CO2) from the lung cavity, i.e., expiration, which would then be followed by inspiration. In Aplysia, respiratory pumping is a stereotyped behavior that consists of synchronous gill, siphon, mantle shelf and parapodial contractions (Hening 1982; Koester 1989). Similarly, in Lymnaea the ventilatory movements follow a rhythmic pattern. Upon surfacing, the pneumostome opens and air is expelled from the lung. This is facilitated by the contraction of the pneumostome and mantle cavity muscles. The absence of mantle cavity muscle contraction indicates that the inspiratory phase is a passive phenomenon. Presumably, air rushes into the lung cavity due to the negative pressure created following active air expulsion. Once the lung is inflated, the pneumostome closure muscles are activated, which in turn causes the closure of the pneumostome. No experiments were carried out to directly test that the filling of the lung is a passive phenomenon. Three muscle groups were described earlier by de Vlieger et al. (1976) which were found to be involved in the opening and closing movements of the pneumostome. Tactile stimulation of these muscles caused either the opening or closing movements of the pneumostome depending upon the muscle layer stimulated (de Vlieger et al. 1976). According to these authors, the top layer of the pneumostome musculature caused the opening movement, whereas the underlayer was found to cause the closure of the pneumostome. We designated these muscles as P.O.M. and P.C.M., respectively. Furthermore, we found that extracellular stimulation of a portion of the mantle cavity musculature (not shown here) caused contraction of the entire muscle bundle. This contraction in turn resulted in the compression of the lung and air expulsion. During respiration, locomotion is interrupted briefly. The pedal cilia present on the foot sole of the animal, which provide the main propulsive force, cease beating especially during pneumostome opening movements (Syed, unpublished observation). In an earlier study, we demonstrated that Ip. 3, which induces the pneumostome opening movements, has inhibitory effects on putative ciliary motor neurons (Syed and Winlow 1989).

Pneumostome and mantle cavity motor neurons Previously, de Vlieger et al. (1976) characterized the effects of some identified visceral and parietal ganglion neurons on the pneumostome musculature of Lymnaea.

N.I. Syed et al.: Respiratory behavior in Lymnaea Instead of obtaining direct intracellular recordings, however, the central neurons were touched with steel microelectrodes and the effects of this stimulation were observed on various muscles of the pneumostome. Various visceral and parietal nerves were similarly stimulated. According to these authors, stimulation of the right parietal external nerve and the anal nerve (see Slade et al. 1981) caused the closure and the opening of the pneumostome, respectively. No direct evidence for the effects of central neurons on the pneumostome musculature was obtained. Furthermore, neither de Vliegers' study nor those of several others (Janse et al. 1985; van der Wilt et al. 1987, 1988) showed the involvement of various neurons during spontaneously occurring respiratory behavior. In the present study, intracellular injections of Lucifer yellow into visceral G, H, I, J and K cells and into right parietal A group neurons (see Fig. 5) revealed that some, but not all, of these neurons had axonal projections in at least one of the nerve trunks innervating the pneumostome and mantle cavity muscles. Electrical stimulation of only specific cells present in the H, I, J, K cell cluster in semi-intact preparations could cause contractions of the appropriate muscles and produce 1:1 EJPS in the muscle fibers recorded intracellularly. Due to anatomical constraints and to the powerful contractions of the muscle bundles it was not possible to obtain stable recordings from the individual muscle fibers for long periods. We most often obtained tension recordings from the entire muscle bundle. Furthermore, in order to test the motor neuronal role of these neurons and to test the chemical nature of their connections, we substituted either Mg 2§ or Co 2§ for Ca 2§ in our saline. The connections between motor neurons and appropriate muscle fibers were reversibly blocked by these manipulations suggesting that these connections are chemical in nature. Furthermore, since a 6-fold increase in Ca 2§ in the bathing solution did not affect the connections between motor neurons and appropriate muscles, we suggest that these connections are monosynaptic. However, these experiments do not rule out the possibility of interposed electrical connections between the various motor neurons. Recently, some of the neurons located within the H, I, J, K cell cluster have been identified as motor neurons to the heart musculature (Buckett et al. 1990). These heart motor neurons have peripheral projections via the intestinal nerve which innervates the heart (Buckett et al. 1990). In the present study we have shown that P.O.M. and P.C.M. motor neurons are also located within the H, I, J, K cells and that these project to the periphery via the anal nerve, and the internal and external parietal nerves. Although the motor neurons reported by Buckett et al. (1990) and those described here are situated in the same cell cluster and receive common synaptic inputs (e.g., Ip. 3), they are morphologically and functionally distinct cells. On the basis of these observations, it seems safe to infer that most neurons located within the H, I, J, K cell cluster are involved in cardiorespiratory behavior. A coordination between mammalian cardiorespiratory neurons has also been reported (Feldman and Ellenberger 1988). Unfortunately, in these vertebrate prepara-

553 tions the degree of this interaction and coordination can not, at present, be resolved at the level of single cells. In Lymnaea by contrast, individual cardiorespiratory neurons have now been identified. Elucidating the intrinsic and network properties of this neural ensemble may help to understand how this coordination is maintained in mammals. In addition to their direct effects on appropriate muscles, the motor neurons receive various synaptic inputs. One of these inputs is Ip. 3 (Benjamin and Winlow 1981). Neither the source nor the function of Ip. 3 was previously known, but its effects on follower cells were very well characterized (Benjamin and Winlow 1981 ; Winlow et al. 1981). Recently, at least one source of Ip. 3, an Input 3 interneuron (Ip. 3. I), has been identified (Syed et al. 1990) and its effects on P.O.M., M.C.M., and P.C.M. motor neurons have been confirmed directly (Syed and Winlow 1991; Syed et al., unpublished). The identified Ip. 3 interneuron is located on the ventral surface of the right parietal ganglion (Syed et al. 1990), whereas all the respiratory motor neurons described here are located on the dorsal surface. In the present study, therefore, this anatomical constraint made it nearly impossible to obtain simultaneous intracellular recordings from the Ip. 3 interneuron and motor neurons. In the companion paper (Syed and Winlow 1991) however, we made these simultaneous recordings from isolated brain preparations and directly confirmed the role of Ip. 3 described here. In our semi-intact preparations we found Ip. 3 to be directly involved in the respiratory behavior: it caused the opening of the pneumostome by exciting P.O.M. motor neurons. Once the pneumostome was opened, Ip. 3 then caused a delayed contraction of the mantle cavity musculature by exciting R.P.A group neurons leading to compression of the lung and air expulsion. The latter effect could easily be monitored by placing soap bubbles at the opening of the pneumostome. Other methods of monitoring air movement into and out of the pneumostome, such as by the insertion ofa cannula into the lung, resulted in total disruption of the respiratory behavior; animals withdrew entirely into their shells.

Role of peripheral feedback in the respiratory behavior While most rhythmic behaviors of animals have been shown to be under the control of central neurons, peripheral inputs have also been found to be important. Examples of the interaction between peripheral and central elements include locomotory movements of the limbs in cats and cockroaches (Pearson and Duysens 1976; also see Pearson 1985), swimming in the dogfish (Grillner and Wallen 1982), mammalian respiration (von Euler 1985), wing movements in locust (Wendler 1983), swimming in tadpoles (Stehouwer and Farel 1980), gastropod feeding (Benjamin 1983) and lobster escape swimming (Siller and Heitler 1985; also see Delcomyn 1980; Getting 1988 and Selverston 1980). In Lymnaea isolated brain preparations a respiratory motor pattern similar to that seen in semi-intact preparations is regularly observed. In contrast, in those semi-

554 i n t a c t p r e p a r a t i o n s w h e r e the a n i m a l s were p i n n e d d o w n to the b o t t o m o f the dish, n e i t h e r the m o t o r b e h a v i o r n o r the m o t o r p a t t e r n (as r e c o r d e d f r o m c e n t r a l n e u r o n s ) was observed. This led us to d e v e l o p the s e m i - i n t a c t p r e p a r a t i o n d e s c r i b e d a b o v e , w h e r e the a n i m a l was s u s p e n d e d in saline, w i t h m o r e f r e e d o m o f m o v e m e n t a n d less interference with the r e s p i r a t o r y b e h a v i o r . F u r t h e r m o r e , in these p r e p a r a t i o n s it was p o s s i b l e to either lift o r l o w e r the p n e u m o s t o m e a b o v e o r b e l o w the level o f the saline. O n c e again, in t h o s e cases w h e r e the p n e u m o s t o m e was l o w e r e d b e l o w the saline level n o r e s p i r a t o r y b e h a v i o r was o b s e r v e d . This is in a g r e e m e n t with o u r o b s e r v a t i o n s o f r e s p i r a t o r y b e h a v i o r in u n t e t h e r e d a n i m a l s , w h e r e the p n e u m o s t o m e m u s t c o n t a c t the w a t e r surface b e f o r e r e s p i r a t i o n c a n begin. I n o u r s e m i - i n t a c t p r e p a r a t i o n s , lifting the p n e u m o s t o m e a b o v e the saline level i n i t i a t e d the r e s p i r a t o r y b e h a v i o r , w h e r e a s l o w e r i n g the p n e u m o s t o m e b e l o w the saline level t e r m i n a t e d the b e h a v i o r . These e x p e r i m e n t s suggest a n i m p o r t a n t role for p e r i p h eral f e e d b a c k in i n i t i a t i n g a n d t e r m i n a t i n g r e s p i r a t o r y b e h a v i o r in Lymnaea. I n c o n c l u s i o n , this s t u d y describes the r e s p i r a t o r y b e h a v i o r o f Lymnaea, defines the muscles i n v o l v e d in this b e h a v i o r a n d identifies r e l e v a n t m o t o r neurons. W e p r e s e n t evidence t h a t these m o t o r n e u r o n s are p a r t o f a n e t w o r k w h i c h c o m p r i s e s the r e s p i r a t o r y c e n t r a l p a t t e r n g e n e r a t o r ( C P G ) . Since the m o t o r n e u r o n s a r e d r i v e n b y v a r i o u s i n t e r n e u r o n s , c o m p r e h e n s i o n o f the i n t e r a c t i o n s b e t w e e n i n t e r n e u r o n s a n d m o t o r n e u r o n s is essential if we are to fully u n d e r s t a n d the n e u r a l basis o f r e s p i r a t o r y b e h a v i o r in these animals. This is a c h i e v e d in the c o m p a n i o n p a p e r (Syed a n d W i n l o w 1991).

Acknowledgements. We thank Dr. R.L. Ridgway, Dr. A.G.M. Bulloch and Ms. K. McKenney for their intelligent and helpful comments during preparation of this manuscript, Dr. R.L. Ridgway for photographic assistance and Caroline Collins for typing the manuscript. This work was supported by a SERC grant to WW., N.I.S. was a University of Leeds Scholar and held an O.R.S. award.

References Alevizos A, Weiss KR, Koester J (1989) SCP-containing R20 neurons modulate respiratory pumping in Aplysia. J Neurosci 9:3058-3071 Bekius R (1972) The circulatory system of Lymnaea stagnalis (L.). Neth J Zool 22 : 1-58 Benjamin PR (1983) Gastropod feeding: Behavioral and neuronal analysis of a complex multicomponent system. In : Roberts A, Roberts BL (eds) Neural origin of rhythmic movements. Soc Exp Biol Symp 37:159-193 Benjamin PR, Winlow W (1981) The distribution of three wideacting synaptic inputs to identified neurons in the isolated brain ofLymnaea stagnalis (L.). Comp Biochem Physio170A: 293-307 Buckett KJ, Peters M, Benjamin PR (1990) Excitation and inhibition of the heart of the snail, Lymnaea by non-FMRFamidergic motoneurons. J Neurophysiol 63:1436-1447 Burrows M (1975a) Co-ordinating interneurons of the locust which convey two patterns of motor commands: Their connexions with flight motor neurons. J Exp Biol 63:713-733 Burrows M (1975b) Co-ordinating interneurons of the locust which convey two patterns of motor commands: Their connexions with ventilatory motor neurons. J Exp Biol 63:735-753 Burrows M (1982) Interneurons co-ordinating the ventilatory

N.I. Syed et al.: Respiratory behavior in Lymnaea movements of the thoracic spiracles in the locust. J Exp Biol 97:385-400 Byrne J (1983) Identification and initial characterization of a cluster of command and pattern generating neurons underlying respiratory pumping in Aplysia californica. J Neurophysiol 49: 482-508 Byrne J, Koester J (1978) Respiratory pumping: Neuronal control of a centrally command behavior in Aplysia. Brain Res 143 : 87-105 Delcomyn F (1980) Neural basis of rhythmic behavior in animals. Science 210:492-498 de Vlieger TA, Lever-de Vries CH, Plesch BEC (1976) Peripheral and central control of the pneumostome in Lymnaea stagnalis. In: Salanki J (ed) Neurobiology of invertebrates, Gastropoda brain, Akad~miai Kiad6, Budapest, pp 624-634 Dickinson PS, Prior D J, Avery C (1988) The pneumostome rhythm in slugs: a response to dehydration controlled by hemolymph osmolarity and peptide hormones. Comp Biochem Physiol 89A: 579-585 Feldman JL, Ellenberger HH (1988) Central coordination of respiratory and cardiovascular control in mammals. Annu Rev Physiol 50: 593~506 Ferguson GP (1984) Neurophysiological analysis of whole body withdrawal in Lymnaea stagnalis. Ph.D. Dissertation, University of Sussex Getting P (1988) Comparative analysis of invertebrate central pattern generators. In: Cohen AH, Rossignol S, Grillner S (eds) Neural control of rhythmic movements. Wiley, New York, pp 101-128 Girdlestone D, Cruickshank SGH, Winlow W (1989) A system for the application of general anaesthetics and other volatile agents to superfused, isolated tissue preparations. Comp Biochem Physiol 92C: 35-37 Grillner S, Wallen P (1982) On peripheral control mechanism acting on the central pattern generators for swimming in the dogfish. J Exp Biol 98:1-22 Hening WA (1982) Central generation and coordination of a complex behavioral sequence in Aplysia californica: Locomotion and respiratory pumping. Ph.D. Dissertation, New York University Janse C, van der Wilt CJ, van der Plas J, van der Roest M (1985) Central and peripheral neurons involved in oxygen perception in the pulmonate snail Lymnaea stagnalis (Mollusca, Gastropoda). Comp Biochem Physiol 82A:459-469 Jones HD (1961) Aspects of respiration in Planorbis corneus (L.) and Lymnaea stagnalis (L.) (Gastropoda: Pulmonata). Comp Biochem Physiol 4:1-29 Koester J (1989) Chemically and electrically coupled interneurons mediate respiratory pumping in Aplysia. J Neurophysiol 62:1113-1126 Pearson KG (1985) Are there central pattern generators for walking and flight in insects? In: Barnes WJP, Gladden MH (eds) Feedback and motor control in invertebrates and vertebrates. Croom Helm, London, pp 307-315 Pearson KG, Duysens J (1976) Function of segmental reflexes in the control of stepping in cockroaches and cats. In: Herman RM, Grillner S, Stein PSG, Stuart DG (eds) Neural control of locomotion. Plenum Press, New York, pp 519-539 Selverston AI (1980) Are central pattern generators understandable? Behav Brain Sci 3:535-571 Siller KT, Heitler WJ (1985) The neural basis of escape swimming behavior in the squat lobster Galathea striyosa. II. Mechanism of burst production. J Exp Biol 117:291-300 Slade CT, Mills J, Winlow W (1981) The neuronal organization of the paired pedal ganglia of Lymnaea stagnalis (L.). Comp Biochem Physiol 69A: 789-803 Stehouwer DJ, Farel PB (1980) Central and peripheral controls of swimming in anuran larvae. Brain Res 195:323-335 Syed NI (1988) Neural control of locomotion in Lymnaea. Ph.D. Dissertation, University of Leeds, U.K Syed NI, Winlow W (1989) Morphology and electrophysiology of

N.I. Syed et al. : Respiratory behavior in Lymnaea neurons innervating the ciliated locomotor epithelium of Lymnaea stagnalis (L.). Comp Biochem Physiol 93A: 633-644 Syed NI, Bulloch AGM, Lukowiak K (1990) In vitro reconstruction of the respiratory central pattern generator of the mollusk Lymnaea. Science 250: 282-285 Syed NI, Winlow W (1991) Respiratory behavior in the pond snail Lymnaea stagnalis II. Neural elements of the central pattern generator (CPG). J Comp Physiol 169:557-568 van der Wilt CJ, van der Roest M, Janse C (1987) Neuronal substrates of respiratory behavior and related functions in Lymnaea stagnalis. In: Boer HH, Geraerts WPM, Joosse J (eds) Neurobiology, molluscan models. North Holland, Amsterdam, pp 282-290 van der Wilt CJ, van der Roest M, Janse C (1988) The role of two peptidergic neurons in modulation of respiratory behavior in the pond snail Lymnaea stagnalis. Symp Biol Hung 26:377-386

555 von Euler C (1985) Central pattern generation during breathing. In: Evarts EV, Wise SP, David B (eds) The motor system in neurobiology. Elsevier Biomedical Press, Amsterdam, pp 47-51 Wendler G (1983) The locust flight system: functional aspects of sensory input and methods of innervation. In: Gustav Fischer (ed) Nachtigall W Biona Report 2, Stuttgart, pp 113 125 Willows AOD, Dorsett DA, Hoyle G (1973) The neuronal basis of behavior in Tritonia. I. Functional organization of the central nervous system. J Neurobiol 4:207-237 Winlow W, Benjamin PR (1976) Neuronal mapping of the brain of the pond snail, Lymnaea stagnalis (L.) In: Salanki J (ed) Neurobiology of invertebrates, Gastropoda brain. Akad~miai Kiad6, Budapest, pp 41-59 Winlow W, Haydon PG, Benjamin PR (1981) Multiple postsynaptic actions of the giant dopamine-containing neuron R.Pe.D.1 of Lymnaea stagnalis. J Exp Biol 94:137-148