Modulatory control of multiple task processing in the stomatogastric nervous system

Eve Marder and James M. Weimann

Publisher Summary

This chapter focuses on the modulary control of multiple task processing in the stomatogastric nervous system. The stomatogastric nervous system of decapod crustaceans controls the coordinated movements of the oesophagus and stomach and consists of four ganglia. There are two commissural ganglia, each consisting of several hundred neurons, a single oesophageal ganglion containing 16–18 neurons, and a single stomatogastric ganglion consisting of approximately 30 neurons. The modulatory inputs aid in understanding how biological wetware can act as a multiple task processor (MTP). The modulatory substances program the MTP by selecting or activating neurons from the ensemble and by changing the gain of the synaptic connections in the ensemble. In a nervous system operating as a multiple task processor, a given neuron may operate at different times in entirely different roles in different functional and neural circuits. The phenotype of such neurons would be that they would switch their activity patterns, but a better understanding of the extent to which they subserve different roles requires more knowledge about the organization of the neural networks’ underlying behavior than is currently available in any preparation. Thus it is important to understand both how network structure constrains the inherent flexibility produced by multiple modulatory processes and how modulation itself is controlled. Research on the changes produced by modulatory substances and neurons in the stomatogastric nervous system enables the understanding of the detailed mechanisms underlying different kinds of modulatory change.

1.1 Introduction

Starting with the pioneering work of Don Maynard (1972), the stomatogastric nervous system of decapod crustaceans has provided numerous fundamental insights into the production of motor patterns by rhythmic neural networks. The earliest work focused on attempts to understand the neural mechanisms by which a rhythmic motor pattern can be generated (Maynard, 1972; Mulloney and Selverston, 1974a,b; Eisen and Marder, 1982; Miller and Selverston 1982a,b). Subsequently the study of the stomatogastric nervous system was instrumental in providing an understanding of the mechanisms by which neuromodulators and modulatory neurones can influence the forms of a single motor pattern (Dickinson and Nagy, 1983; Nagy and Dickinson, 1983; Hooper and Marder, 1984, 1987; Marder, 1984; Flamm and Harris-Warrick, 1986a,b; Heinzel and Selverston, 1988; Nagy et al., 1988; Nusbaum and Marder, 1989a,b; Katz and Harris-Warrick, 1990). Most recently, this preparation has allowed us to study interactions among neurones and the neural networks that together produce coordinated movements. In this chapter we will first review the evidence that demonstrates that a single group of neurones can produce multiple forms of a single motor pattern. Then we will summarise recent data which demonstrate that the same neurones can participate in the generation of several different motor patterns, and show how behavioural selection could be achieved by modulating their membrane properties and the connections between them.

The stomatogastric nervous system of decapod crustaceans controls the coordinated movements of the oesophagus and stomach (Fig. 1.1), and consists of four ganglia. There are two commissural ganglia (CGs), each consisting of several hundred neurones, a single oesophageal ganglion (OG) containing 16–18 neurones, and the single stomatogastric ganglion (STG) which consists of approximately 30 neurones. Together the neurones found in these ganglia produce at least four rhythmic motor patterns which govern the movement of food through the foregut region.

Fig. 1.1 Diagram of the side view of the foregut region and the position of the ganglia and nerves of the stomatogastric nervous system. Abbreviations in this and subsequent figures are as follows: CG, commissural ganglion; OG, oesophageal ganglion; STG, stomatogastric ganglion; dvn, ivn, lvn, mvn, dorsal, inferior, lateral and medial ventricular nerves, respectively; pdn, pyloric dilator nerve; pyn, pyloric nerve; son, superior oesophageal nerve; stn, stomatogastric nerve.

Food moves from the mouth into the oesophagus (Fig. 1.1). The oesophageal rhythm has a characteristic period of 10 s, and consists of alternating bursts in the oesophageal dilator and constrictor motoneurones. Relatively little is known concerning the mechanisms underlying the generation of the oesophageal rhythm, although it is known that many of the oesophageal motoneurones are found in the CGs and several are found in the OG. From the oesophagus food moves into the cardiac sac region of the stomach (Fig. 1.1). The cardiac sac rhythm has a characteristic period of 30 s to several minutes (Fig. 1.2). It involves motoneurones that are found in both the OG and the STG, and again relatively little is known about the pattern-generating network that underlies it.

Fig. 1.2 Motor patterns that control the movements of the foregut. Simultaneous extracellular (top two traces) and intracellular (bottom two traces) recordings. The cardiac sac rhythm is seen as slow bursts in the cardiac sac dilator 2 motoneurone (CD2) and bursts of the ivn fibres. The gastric rhythm is shown as bursts of gastric mill (GM) neurone activity recorded on the anterior lateral nerve (aln). The pyloric rhythm is shown as rapid bursts of activity recorded in the pyloric dilator (PD) neurone. (Modified from Dickinson and Marder, 1989.)

The gastric mill receives food from the cardiac sac (Fig. 1.1). The two lateral teeth and the single medial tooth of the gastric mill shred and chew food. The gastric rhythm (Fig. 1.2) has a characteristic period of 5–10 s, and is generated by neurones found in the STG. The gastric mill rhythm has been extensively studied, and is thought to be an example of a rhythmic motor pattern that depends critically for its production on emergent network properties. The pylorus is the last region of the decapod foregut (Fig. 1.1). The movements of the pyloric region are controlled by the pyloric rhythm (Fig. 1.2), which occurs with a characteristic frequency of approximately 1 Hz. The pyloric network has been extensively studied and depends on the membrane properties of a conditionally bursting neurone for its intrinsic rhythmicity, and on the synaptic interactions among the pattern-generating neurones for the phase relationships of the pyloric rhythm (Marder et al., 1987a).

1.2 Identification of modulatory inputs and substances

The earliest studies of the pyloric and gastric rhythms used preparations in which the STG was isolated from the more anterior ganglia of the stomatogastric nervous system by cutting either the stomatogastric nerve (stn) or the inferior and superior oesophageal nerves (ion and son, respectively) (Maynard, 1972; Mulloney and Selverston, 1974a,b). Although these reduced preparations showed rhythmic activity, robust and vigorous pyloric and gastric mill rhythms are most reliably produced by preparations in which the CGs and OG are left attached (Russell, 1976, 1979; Nagy and Moulins, 1987; Hartline et al, 1988). This suggests that neurones of the CGs and OG were likely to function as modulatory inputs to the network in the STG.

The search for these modulatory extrinsic inputs to the STG took two paths. The Moulins laboratory undertook a major attempt to identify specific neurones in the CGs and OG with modulatory actions on the neurones of the STG. This work has been reviewed by Nagy and Moulins (1987). A complementary approach has been to identify as many as possible of the modulatory substances present in fibres that project into the STG from extrinsic sources (Fig. 1.3). This latter approach has led to the demonstration of a large number of different substances in inputs to the STG, as shown in Fig. 1.3. These include serotonin (Beltz et al., 1984), dopamine and octopamine (Barker et al., 1979; Kushner and Barker, 1983), histamine (Claiborne and Selverston, 1984), GABA (Cazalets et al., 1987; Cournil et al., 1990; Mulloney and Hall, 1990), proctolin (Hooper and Marder, 1984; Marder et al., 1986), several FMRFamide-like peptides that we know now are likely to be extended FLRFamide-like peptides (Hooper and Marder, 1984; Marder et al., 1987b), red pigment concentrating hormone- (RPCH) like peptides (Nusbaum and Marder, 1988; Dickinson and Marder, 1989), crustacean cholecystokinin (CCK)-like peptides (Turrigiano and Selverston, 1989, 1990), a substance-P-like immunoreactivity (Goldberg et al., 1988), a β-pigment dispersing hormone (β-PDH)-like peptide (Mortin and Marder, 1989, 1991), and a myomodulin-like peptide (Lockhart, Hall, Oshinsky and Marder, unpublished results).

Fig. 1.3 Diagram summarising the neuromodulatory substances found in neuronal projections from...