|Allen Selverston (2008), Scholarpedia, 3(4):1661.||doi:10.4249/scholarpedia.1661||revision #91832 [link to/cite this article]|
The stomatogastric ganglion is a collection of about thirty neurons that sits on the dorsal surface of the foregut (stomach) in decapod crustaceans. The neurons it contains form two central pattern generators (CPGs), namely the pyloric and gastric mill CPGs. The pyloric CPG controls striated muscles that dilate and constrict the pyloric region of the stomach in a cyclic three phase rhythm. The gastric mill CPG produces a slower six phase rhythm that control muscles that produce chewing by three ossicles in the gastric mill. The system has special advantages that make it one of the most well-studied invertebrate neural circuits available. Its importance lies in the fact that the cellular and synaptic properties used by this ganglion are the same as those found in all nervous systems including the mammalian brain but because there are only a few neurons in each circuit the entire "wiring diagram" for the ganglion has been determined. Besides, neurons of the stomatogastric ganglion produce cell type-specific voltage output and firing patterns with high consistency among preparations. This feature makes their identification straightforward and helps revealing their connectivity. The stomatogastric ganglion can therefore serve as a model to understand how the synaptic interactions of individual nerve cells produce two functionally different behaviors. It is as a result of these cell-to-cell interactions that sequential spatiotemporal patterns are formed, patterns similar to those produced by CPGs in all nervous systems.
Anatomy and Behavior
The gastric mill muscles open and close a set of three teeth in a kind of chewing behavior. This activity can be observed in live animals with videoendoscopy and has been shown to usually occur in two different patterns (Heinzel, 1988). In one case the two lateral teeth and the single medial tooth all come together simultaneously in a squeeze movement. In the second form the two lateral teeth come together to grasp the food and the medial tooth is pulled over the held food in what has been called a cut and grind motion. The movements can be shown to be correlated with burst of spikes in the appropriate motor neurons when recorded from isolated preparations.
The pyloric CPG produces movements that are much simpler in function. Operating at a frequency of 1-2 Hz, four dilator muscles, pulling in opposite directions, open the pylorus. Then constrictor muscles at the front of the pylorus close the pyloric chamber and this is followed by further contraction of more rostral constrictors all in all producing a wave of peristaltic contraction which move food particles toward the gut.
The stomatogastric ganglion (STG) (Figure 4) lies in the ophthalmic artery that carries blood from the heart to the brain. The neurons are relatively large, about 30-50 microns in diameter and therefore easily impaled with microelectrodes for recording or stimulation. Motor nerves that leave the ganglion innervate the muscles shown in Figure 1 and can be recorded extracellularly. Neurons are identified by correlating their intracellularly recorded electrical activity with extracellularly recorded spikes from the motor nerves as they enter the muscles. Several of the neurons, especially in the pyloric circuit can be readily identified by visually inspecting their voltage waveforms alone. This has allowed researchers to determine the neuronal components of each each CPG and to describe the synaptic connections between them (Selverston et al. 1976).
|# of copies||Name||Abbr.||Muscle||Transmitter|
|2||pyloric dilator||(PD)||cpv 1&2||ach|
|1||ventricular dilator||(VD)||cv 2||ach|
|1||inferior cardiac||(IC)||cv 3||glu|
|8||pyloric||(PY)||p 2-4, 7-8, 10-11||glu|
The connections between these neurons shown in a simplified schematic are mostly inhibitory although there are many electrical connections as well. When left connected to the commissural ganglia and receiving modulatory inputs, they produce a characteristic oscillatory burst pattern shown in Figure 5. This is a three phase rhythm running at about 2 hz.
The synapses in the gastric mill circuit are also mostly inhibitory as shown in Figure 6B but there are a few excitatory synapses and electrical connections. They produce the 0.1 hz rhythm shown in Figure 6D.
|# of copies||Name||Abbr.||Muscle||Transmitter|
|1||Interneuron 1||(Int 1)||interneuron||glu|
|4||Gastric Mill||(GM)||gm 1b,2a,b||ach|
|1||Dorsal Gastric||(DG)||gm 4a,b,c||ach|
|1||Anterior Median||(AM)||c 6, c7||glu|
|1||Lateral Gastric||(LG)||gm 5b, 6a||glu|
|1||Medial Gastric||(MG)||gm 9a, 9c||glu|
|2||Lateral Posterior Gastric||(LPG)||gm 3||ach|
<review>The placement of these figures needs to be fixed </review>
The activity patterns of the pyloric and gastric mill neurons recorded intracellularly are shown in Figure 7A and B. All of the neurons have membrane oscillations with spike potentials on their depolarized phase. Most of the neurons, in addition to generating the patterns, send action potentials to the peripheral muscles. Only two neurons, AB and Int 1 are interneurons. Note that the spikes shown intracellularly correspond to the burst patterns recorded from the motor nerves ( Figure 5 and Figure 6).
A complete description of the neural circuit for the stomatogastric system includes the two independent CPGs, the gastric and pyloric circuits, the functional connections between them as well as ancillary and sensory neurons that affect the two patterns. These are illustrated schematically in Figure 8.
The patterns for both CPGs belong to the general class of cyclic sequential patterns that underlie such diverse behaviors as walking, chewing and flying. Since the CPGs do not require any patterned input or feedback from sensory receptors in order to produce their sequential rhythmic patterns, the question of what internal mechanisms are responsible is the most fundamental to address. It is generally assumed that the range of possible mechanisms is limited to intrinsic bursting (pacemaker activity), network properties or a combination of both. The two stomatogastric CPGs appear to fall into the last category.
Evidence for pacemakers
Pyloric CPG - All neurons of the pyloric system burst intrinsically when neuromodulatory inputs from the more anterior commissural ganglia (CG) are present, the cells therefore are termed as conditional bursters. When inputs from the CGs are blocked the pyloric rhythm ceases. The AB neuron displays the most robust and periodic bursting even with complete synaptic isolation from the other pyloric neurons. The AB neuron is electrically connected to the two PD neurons and this trio of neurons, also known as the pacemaker group, acts as the timer for the whole pyloric rhythm. Depolarizing the AB neuron increases the frequency of bursting and hyperpolarization decreases the frequency, a characteristic of pacemaker-driven activity.
Gastric CPG - Like the pyloric system, neuromodulatory inputs are necessary for the pattern to be initiated and maintained. While the mechanisms for pattern formation are less well-defined, current research suggests that the LG/MG group and the DG/AM group have intrinsic burst capability. It is not yet clear whether or not Int1 is an intrinsic burster, but its principle action appears to be to synchronize the lateral teeth neurons with the medial tooth neurons which without Int 1, burst at different frequencies and in an uncoupled fashion.
Role of the network
Pyloric CPG - The extensive inhibitory connections determine the sequence of firing by the PD/AB group inhibiting the other neurons which then recover at different rates. It can be shown by modeling experiments that neurons with reciprocal connections can generate alternate bursts, but there is no evidence that this mechanism is responsible for burst generation in vitro. It can be shown experimentally that two nonbursting pyloric neurons connected with reciprocal inhibition can generate out of phase bursting. In the pyloric system, the principal role of the network is to produce the correct pattern in terms of sequence but not to generate the bursts.
Gastric CPG - The exact role of the networks synaptic topology is not yet fully determined but synaptic interactions play a more prominent role in the gastric CPG than in the pyloric system. As described above, when the LG/MG pair bursts, they inhibit the LPGs which otherwise fire tonically. When the DG/AM group bursts, they inhibit the GM neurons. The rebound from this inhibition probably plays a significant role. The synaptic interactions between these two groups and Int 1 appears to be the main cause of the two subgroups becoming synchronized. Other synaptic interactions further separate some of these bursts. In particular, LG and MG are weakly connected with reciprocal inhibitory synapses and are linked with an electrotonic connection. This causes one of the two cells to fire earlier than the other and before the electrical connection brings them together. In addition, the action of the excitatory synapses from Int 1 to DG and AM causes one cell to fire before the other. As a result, the gastric mill rhythm is composed of six separate phases when in the normal experimental configuration i.e., with the connections from the CGs intact.
Role of voltage dependent conductances
The intrinsic properties of stomatogastric neurons also play a major role in determining the motor pattern sequences. Each neuron expresses a large number of different voltage sensitive ionic channels that, acting in concert, confer on each neuron it's characteristic behavior and unique identity. The conductances that have been described in stomatogastric neurons and their effect on the pattern are:
INa - the normal depolarization-activated transient sodium current that produces the rising phase of the action potential.
IK - the common delayed rectifier potassium current that is responsible for the falling phase of the action potential.
INaP - a persistent sodium current that can initiate and maintain tonic or burst firing.
Ih- a hyperpolarization-activated inward current sometimes called the pacemaker current because it can initiate bursting or affect postinhibitory rebound.
ICAN - a calcium activated non-selective cation current that maintains a depolarized plateau.
IK(Ca) - a calcium activated potassium current that could be responsible for burst termination.
IA - a transient potassium current that is activated by depolarization and acts to delay spike and burst initiation.
The stomatogastric ganglion contains many different types of neuromodulatory substances. Some of these are released from terminals of extrinsic neurons while others reach the ganglion in the blood. At this time, over twenty different modulators have been identified by immunohistochemistry or spectroscopy. In Figure 9, each color represents a different neuromodulator carried to the ganglion in a separate axon. These axons come from the brain and other ganglia, particularly the commissurals. Little is known about the mechanisms of their release or their synergistic or antagonistic effects on each other. But when applied individually to the ganglion in vitro, they produce fundamental changes in the circuits and therefore the output patterns. Modulators allow this small ganglion to produce many more behaviors than was initially thought because of this ability to functionally modify the circuitry.
Role of modulation
The action of neuromodulators is fundamental to understanding how the stomatogastric system works in vivo. In general they alter the properties of the channels listed above in characteristic ways, they affect the strength of gap junction (electrical) connections, and they can alter the strength and time course of chemical synapses. This allows them to turn the CPGs on or off or to modify the ongoing rhythms in specific repeatable ways in terms of frequency, phase relationships, duty cycle etc. In addition they can produce more complicated effects such as:
- Switch a neuron from one CPG to another CPG
- Combine the patterns from two CPGs to form a blended pattern
- Form an entirely different pattern de novo from components of separate CPGs
Early attempts at understanding how behavior is produced sought to examine the underlying circuitry as it were hardwired by the genome to perform only one task. We know now that within certain constraints, neuromodulation can allow these small and well-defined circuits to multitask into many stable patterns simply by changing the chemical milieu surrounding them.
Cellular mechanisms of neuromodulation
The cellular mechanisms by which various modulators functionally reconfigure the stomatogastric circuitry to achieve the phenomena described above has been well- studied. In general, the neuromodulators act via G protein pathways to activate second messenger cascades that culminate in the phosphorylation of particular channels. This changes the biophysical properties of the neurons and synapses and thus alters the function of the entire network for a time period that depends on the continued presence of the modulator and the rate at which the channels are dephosphorylated. Detailed studies of identified stomatogastric neurons have shown that each has a unique combination of channels and receptors so that each cell will respond differently but consistently to various modulators. For example in Figure 10 taken from Nusbaum and Beenhaker (2002), each of the pyloric neurons illustrated has a different compliment of channel responses to a particular modulator, in this case dopamine. When this is extended to the general response of each identified neuron to eight different modulators, it is quickly apparent that this mechanism for obtaining a large number of patterns from the same circuit is very effective, and we have not considered the effects on the chemical and electrical synapses which also can be altered.
Heinzel HG (1988) Gastric mill activity in the lobster I. Spontaneous modes of chewing. 59:528-550.
Marder E, Bucher D (2001) Central pattern generators and the control of rhythmic movements. Curr Biol 11:R986-R996.
Nusbaum M, Beennhakker MP (2002) A small-systems approach to motor pattern generation. Nature 417:343-350.
Selverston AI, Russell DF, Miller JP, King DG (1976) The stomatogastric nervous system: Structure and function of a small neural network. Prog Neurobiol 7:215-290.
Some General References for the Stomatogastric Ganglion
Bal T, Nagy F, Moulins M (1988) The pyloric central pattern generator in crustacea: A set of conditional neuronal oscillators. 163:715-727.
Baro D, Cole C, Harris-Warrick R (1996) RT-PCR analysis of shaker, shab, shaw, and shal gene expression in single neurons and glial cells. Receptors and Channels 4:149-159.
Golowasch JP (1991) Characterization of a stomatogastric ganglion neuron. A biophysical and a mathematical description. In: Brandeis University, Waltham, MA.
Harris- Warrick RM, E. Marder. (1991) Modulation of neural networks for Behaivor. Ann Rev Neurosci 14:39-57.
Harris-Warrick R. M. E. Marder AS, A.I. Selverston, M. Moulin, ed (1992) Dynamic Biological Networks: The Stomatogastric Nervous System. Cambridge: MIT Press.
Le Masson G, Le Masson S, Moulins M (1995) From conductances to neural network properties: Analysis of simple circuits using the hybrid network method. Prog Biophys molec Biol 64:201-220.
Russell DF (1979) CNS control of pattern generation in the lobster stomatogastric ganglion. 177:598-602.
Selverston, A.I. A neural infrastructure for rhythmic motor patterns. Cell Mol Neurobiol. 25, 223-244 (2005).
- Eugene M. Izhikevich (2006) Bursting. Scholarpedia, 1(3):1300.
- Jonathan E. Rubin (2007) Burst synchronization. Scholarpedia, 2(10):1666.
- Peter Jonas and Gyorgy Buzsaki (2007) Neural inhibition. Scholarpedia, 2(9):3286.
- Jeff Moehlis, Kresimir Josic, Eric T. Shea-Brown (2006) Periodic orbit. Scholarpedia, 1(7):1358.
- Philip Holmes and Eric T. Shea-Brown (2006) Stability. Scholarpedia, 1(10):1838.
- Arkady Pikovsky and Michael Rosenblum (2007) Synchronization. Scholarpedia, 2(12):1459.