Show Summary Details

Page of

PRINTED FROM OXFORD HANDBOOKS ONLINE ( (c) Oxford University Press, 2015. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy).

date: 19 February 2018

Neural Control of Swimming in Nudipleura Mollusks

Abstract and Keywords

This article compares the neural basis for swimming in sea slugs belonging to the Nudipleura clade of mollusks. There are two primary forms of swimming. One, dorsal/ventral (DV) body flexions, is typified by Tritonia diomedea and Pleurobranchaea californica. Although Tritonia and Pleurobranchaea evolved DV swimming independently, there are at least two homologous neurons in the central pattern generators (CPGs) underlying DV swimming in these species. Furthermore, both species have serotonergic neuromodulation of synaptic strength intrinsic to their CPGs. The other form of swimming is with alternating left/right (LR) body flexions. Melibe and Dendronotus belong to a clade of species that all swim with LR body flexions. Although the swimming behavior is homologous, their swim CPGs differ in both cellular composition and in the details of the neural mechanisms. Thus, similar behaviors have independently evolved through parallel use of homologous neurons, and homologous behaviors can be produced by different neural mechanisms.

Keywords: evolution, homologous neurons, neuromodulation, central pattern generator, mollusk, serotonin, behavior, swimming

Sea slugs are generally considered slow-moving creatures, but several have evolved rapid means of propulsion through water. Animals belonging to the larger clade of sea slugs called Heterobranchia exhibit a wide variety of swimming behaviors that differ in mode of propulsion, directionality, and function (see reviews: Farmer, 1970; Willows, 2001). The neural basis of rhythmic swimming has been studied in depth in species belonging to the clade Nudipleura (Fig. 1), which includes Pleurobranchaea californica and the nudibranchs: Tritonia diomedea, Melibe leonina, Dendronotus iris, and Hermissenda crassicornis (Newcomb et al., 2012). These studies were facilitated by the simple swimming behaviors and the large, identifiable neurons that constitute relatively uncomplicated circuits. Researchers can record stable motor patterns in isolated nervous systems and manipulate the firing of individual neurons. This has provided valuable insight into how neural networks produce rhythmic activity. Nudipleura are also notable because swimming evolved independently several times. Furthermore, homologs of identified neurons can be identified across species based on neuroanatomical and neurochemical characteristics. Thus, one can directly compare the organization of circuits that produce rhythmic movements in different species.

Neural Control of Swimming in Nudipleura MollusksClick to view larger

Figure 1 Abbreviated phylogenetic tree of the Nudipleura species. Nudipleura is a monophyletic clade that contains Pleurobranchaea and the subclade Nudibranchia. Nudibranchia has two subclades, Doridacea and Cladobranchaea. Species that swim with left/right (LR) body flexions are shown in red; those that swim with dorsal/ventral (DV) body flexions are shown in blue. The colored vertical bars represent the appearance of the behavioral trait in the phylogeny. Each behavior evolved independently several times. Dendronotus, Scyllaea, and Melibe form a clade of LR swimming species. The phylogeny is based on RNA sequencing data (Goodheart et al., 2015) and morphological traits (Newcomb et al., 2012).

Swimming Behaviors and Phylogeny

There are more than 2,000 different species in the monophyletic Nudipleura clade. Of those, fewer than 70 species have been observed to swim (Newcomb et al., 2012). The distribution of swimming in the phylogeny suggests that nonswimming (NS) is the ancestral behavioral state. Within Nudibranchia, swimming falls into two general categories, dorsal-ventral (DV) and left-right (LR) whole-body flexions. Both are rhythmic behaviors. DV swimming consists of the animal repeatedly bending at the anterior-posterior midpoint and having the head and tail meet above and below the horizontal plane. For LR swimming, the animal flattens in the sagittal plane and bends alternately from side to side.

Both DV and LR swimming appear to have evolved independently several times (Fig. 1). RNA sequencing has resolved much of the phylogeny of the subclade called Cladobranchia (Goodheart et al., 2015). Examination of the phylogenetic relationships among the swimming species shows that they, for the most part, do not fall into clades with other swimming species; each belongs to a clade where swimming is uncommon. An exception to this is the clade that contains Dendronotus, Scyllaea, and Melibe (Fig. 1); all three of these genera are comprised of LR swimming species. This suggests that LR swimming is homologous in these species, whereas it evolved independently in other nudibranchs.

Dorsal Ventral Flexion Swimmers: Tritonia diomedea

The Swimming Behavior

The nudibranch Tritonia diomedea (synonymous with Tritonia tetraquetra, Pallas 1788) produces a DV swim as an escape response to contact with a predatory starfish (Willows & Hoyle, 1969; Wyeth & Willows, 2006). Swimming is a transient behavior that lasts about 1 minute. The period between dorsal flexions lengthens from 6 seconds to 10 seconds over the course of a swim episode. The first flexion is in the ventral direction, lifting the animal off the substrate. The swim always ends on a sustained dorsal flexion, causing it to settle foot-first back on the bottom.

The number of flexion cycles can vary depending on stimulus strength and other factors. In particular, the number of cycles decreases if the animal is repeatedly stimulated, a form of behavioral habituation (Frost et al., 1996). If “habituation training” is repeated over several days, the animal exhibits long-term habituation, which lasts at least 2 days beyond training (Frost et al., 2006). The behavioral response can also exhibit sensitization in the form of a decreased latency following a single strong noxious stimulus (Frost et al., 1998). Furthermore, the behavioral response can be suppressed by providing vibratory tactile stimulus prior to stimulating with a strong noxious stimulus (Mongeluzi et al., 1998). The mechanism for the “prepulse inhibition” involves presynaptic inhibition of the sensory neurons as well as downstream inhibition of the swim neural circuit (Frost et al., 2003).

Basic Central Pattern Generator Circuitry

Work on the central pattern generator (CPG) underlying Tritonia swimming was pioneering in several respects and led to important new concepts such as the network oscillator (Getting, 1989a) and intrinsic neuromodulation (Katz & Frost, 1996). Early papers on Tritonia by A. O. Dennis Willows were some of the first to demonstrate the behavioral functions of individual identified neurons (Willows, 1967). Using intracellular recording and stimulation, Willows and colleagues showed that the DV swimming behavior correlated with the firing pattern of particular neurons in the brain (Willows & Hoyle, 1968). Based on this semi-intact preparation in which the animal was free to move while the brain was immobilized for recording, it was possible to then use the pattern of neuron firing to recognize the swim motor pattern in an isolated brain (Dorsett et al., 1969, 1973).

The neurons composing the swim CPG were gradually recognized through a methodical survey of neurons in brain using intracellular microelectrodes. There are three neuronal types in the CPG, the Dorsal Swim Interneurons (DSIs) (Getting, 1977), Cerebral Neuron 2 (C2) (Getting, 1977; Taghert & Willows, 1978), and Ventral Swim Interneuron B (VSI-B) (Fig. 2A, B) (Getting, 1983b). All of the CPG neurons are bilaterally paired and electrically coupled to their contralateral counterparts. Initially, it was thought that VSI-A constituted the third member of the swim CPG because of its activity pattern and connections to DSI and C2 (Getting et al., 1980; Lennard et al., 1980). However, modeling and electrophysiology studies suggested that the actions of VSI-A were not sufficient to account for the activity of the CPG (Getting 1983a, 1989b). VSI-B, which was discovered later, accounted for many of the missing properties (Getting, 1983b).

Neural Control of Swimming in Nudipleura MollusksClick to view larger

Figure 2 Dorsal/ventral flexion swim central pattern generators (CPGs) and motor patterns. (A) The Tritonia swim CPG consists of DSIa,b,c, C2, and VSI. DRI serves as a command/gating neuron for the swim CPG. The shaded box represents the CPG. (B) Three simultaneous intracellular microelectrode recordings from neurons in the Tritonia swim CPG show the motor pattern in response to stimulation of a body wall nerve (arrow). (C) The swim CPG circuit of Pleurobranchaea consists of AS1-4, A1, A3, and A10. Ivs is a hypothetical neuron that has not been identified. The colors of the neurons indicate their homology to Tritonia neurons. (D) Simultaneous intracellular microelectrode recording from As2/3, A10, A1, and A3 show the swim motor pattern.

(Recording is modified from Jing & Gillette, 1999)

DSI and C2 project their axons contralaterally through the cerebral commissure toward the contralateral pedal ganglion, whereas VSI-B sends its axon ipsilaterally toward the ipsilateral pedal ganglion. Their axons pass through one of two pedal commissures that circle around the esophagus, forming synapses in both left and right pedal ganglia (Sakurai & Katz, 2009a). The excitation of VSI-B by C2 occurs primarily in the pedal ganglion that is contralateral to the VSI-B soma (i.e., distal to the pedal commissure); however, the balance of C2-to-VSI excitation between the two pedal ganglia is highly variable among individuals (see later discussion).

Sequence of Events

The Tritonia swim CPG is gated by a “command neuron,” as defined by Kupfermann and Weiss (1978); the Dorsal Ramp Interneuron (DRI) is necessary and sufficient to elicit the behavior and has “preferred” access to sensory information (Frost & Katz, 1996). DRI may also be considered a “gating neuron” because it is active throughout the motor pattern. If it is hyperpolarized, the motor pattern ceases. Gating neurons have also been identified in the leech swim system (Brodfuehrer et al., 1995).

Our current understanding of how Tritonia produces its swim motor pattern is as follows (Katz, 2009, 2010). Sensory neurons are excited by starfish tube feet, causing them to fire more than a single spike (Getting, 1976). This excites a trigger neuron (Tr1), which relays excitation to DRI (Frost & Katz, 1996; Frost et al., 2001). DRI simultaneously excites all six DSIs (three on each side of the brain). The DSIs excite C2. After a delay, C2 excites VSI-B, which then feeds back inhibition to DSI and C2, ending the cycle and removing excitation from VSI-B. DRI continues to fire because of recurrent excitation from C2, maintaining the motor pattern. The period increases gradually until a C2 burst is insufficient to cause DRI to fire and C2 fails to trigger a VSI-B burst. At this point, the motor pattern halts. The DSIs continue to fire tonically after a swim motor pattern, which excites efferent neurons that contribute to ciliary locomotion (Popescu & Frost, 2002).

None of the neurons in the Tritonia swim CPG exhibits endogenous rhythmic bursting. The motor pattern derives from the synaptic interactions of the component neurons. Thus, this CPG is an example of a network oscillator (Getting, 1986).

Intrinsic Neuromodulation

An important finding made in this system is that the CPG is not hard-wired; rather, synaptic strengths are modified upon the onset of the motor pattern generation. The DSIs are serotonergic (McClellan et al., 1994) and use serotonin (5-HT) to enhance the strength of synapses made by C2 and VSI-B (Katz et al., 1994; Sakurai & Katz, 2003) and alter their membrane properties (Katz & Frost, 1997; Sakurai et al., 2006). This “intrinsic neuromodulation” is needed for the CPG to produce a rhythmic output (Calin-Jageman et al., 2007). Intrinsic neuromodulation has subsequently been found in many other motor circuits such as the mammalian respiratory CPG in the pre-Bötzinger complex (Lieske & Ramirez, 2006), in the periaqueductal gray (Jansen et al., 1998), and in vertebrate spinal CPGs (Sillar et al., 1998; El Manira et al., 2002, 2008).

The effect of DSI on C2 is mediated by 5-HT causing a presynaptic enhancement of neurotransmitter release (Katz & Frost, 1995) and is associated with an increase in presynaptic Ca2+ (Hill et al., 2008). DSI also reduces C2 spike frequency adaptation (Katz & Frost, 1997), which further contributes to enhancing the participation of C2 in the motor pattern. In addition, DSI stimulation causes short-term enhancement of VSI-evoked synaptic currents (Sakurai & Katz, 2003) through an increase in the fraction of the readily releasable pool of transmitter from VSI-B (Sakurai et al., 2007). Finally, 5-HT released from DSI depotentiates VSI synapses from homosynaptic potentiation, restoring them to their basal strength (Sakurai & Katz, 2009b). Thus, 5-HT released from DSI during the swim motor pattern has profound effects on the synaptic strength and firing of the other two swim interneurons, transforming a nonfunctional circuit into a CPG. Modeling studies suggest that this intrinsic neuromodulation is necessary for this network oscillator to produce a rhythmic pattern of activity (Calin-Jageman et al., 2007).

Individual Variability

There is variation in the strength of connections made by C2 onto VSI-B (Sakurai, Tamvacakis, & Katz, 2014). Under normal conditions, these individual differences do not affect the performance of the swim CPG. However, when the pedal commissure is cut, then the extent to which C2 inhibits VSI-B correlates with the decrease in the number of bursts produced by VSI-B. Such individual differences also have been observed in other CPGs with little effect on normal function (Goaillard et al., 2009; Roffman et al., 2012; Ransdell et al., 2013; Hamood & Marder, 2014).

After cutting the pedal commissure, Tritonia can functionally recover without regeneration of the axon (Sakurai & Katz, 2009a). There is also variability in the extent to which individuals recover from the lesion (Sakurai et al., 2016). However, unlike differences in the initial susceptibility to the lesion, the differences in recovery correlate with the recruitment of polysynaptic input. Although this CPG is small and relatively well defined, there are additional neurons that may mediate connections between neurons. These connections may assume a greater importance under certain conditions, such as during recovery from injury.

Recruitment of neurons may also occur during memory formation. There are neurons whose roles in motor pattern generation are not known, but whose activity was observed to vary from trial to trial when eliciting a motor pattern using voltage-sensitive dyes (Hill et al., 2012). Following sensitization of the swimming response, there is a recruitment of variably active neurons into the motor pattern (Hill et al., 2015). Thus, although there is a core group of CPG neurons, there are likely additional neurons that influence swimming under conditions such as recovery from injury or memory formation.

Dorsal Ventral Flexion Swimmers: Pleurobranchaea californica

Basic CPG Circuit

Pleurobranchaea is a DV swimmer, like Tritonia. There are both similarities and differences in the CPGs underlying their swimming behaviors. Two of the neuronal types in the Pleurobranchaea swim CPG are homologous to neurons in the Tritonia swim CPG (Fig. 2C, D): Neuron A1 is homologous to C2 (Jing & Gillette, 1995; Lillvis et al., 2012) and the serotonergic As1-3 neurons are homologous to the DSIs (Jing & Gillette, 1999; Katz et al., 2001). Another serotonergic neuron, As4, was identified as a member of the swim CPG. This neuron seems to be homologous to a very small serotonergic neuron in Tritonia, which has not yet been recorded from. Other neurons in the Pleurobranchaea swim CPG do not have correlates in the Tritonia swim CPG (A3, A10). This may be because these neurons do not have homologs or because their Tritonia homologs are not rhythmically active or connected to C2 in Tritonia. Furthermore, the Tritonia VSI-B neuron has not yet been identified in Pleurobranchaea, but it is represented in the circuit diagram by the putative neuron, IVS. The ventral phase of the motor pattern seems to be initiated by the A3 neuron, which inhibits A1 and A10.

A1 was shown to be a member of the swim CPG in whole-animal preparations similar to those performed in Tritonia, where the neuron was recorded from and stimulated while the animal moved (Jing & Gillette, 1995). A1 activity consistently preceded dorsal body flexions. Depolarization of A1 was sufficient to elicit rhythmic bursting activity, and hyperpolarization of A1 during a swim motor pattern halted the movement.

Initiation of the motor pattern involves A10, which is strongly coupled to A1. Depolarization of a single A10 can initiate the motor pattern. This differs from Tritonia once again, where swim motor pattern initiation is dependent upon DRI, which excites the DSIs, not C2 (Frost & Katz, 1996). However, depolarization of C2 is sometimes sufficient to initiate and maintain a motor pattern (Getting, 1977; Taghert & Willows, 1978). In Pleurobranchaea, the As1-4 neurons are not activated by A10 and may have their own access to the sensory input, which has not been discovered yet; perhaps it is the homolog of DRI.

The maintenance of the motor pattern once initiated is due to the recurrent excitation among the As1-4 neurons (Jing & Gillette, 1999). This was initially proposed to be the mechanism for maintenance of the motor pattern in Tritonia (Lennard et al., 1980) until it was discovered that the recurrent excitation arose from C2 exciting DRI (Frost & Katz, 1996).

Intrinsic Neuromodulation

As in Tritonia, the As1-3 cells use 5-HT to enhance the strength of C2 synapses (Lillvis & Katz, 2013). However, the modulation is not as reproducible as in Tritonia. Furthermore, unlike Tritonia, Pleurobranchaea exhibits daily variations in its propensity to produce the swimming behavior. The extent to which DSI modulates C2 synaptic strength correlates with the number of body flexions produced by the animal on the day of testing (Lillvis & Katz, 2013). The cause of this variability is still under investigation. Interestingly, neither DSI stimulation nor 5-HT application enhances C2 synaptic strength in another species, Hermissenda crassicornis, which does not swim with DV flexions (Lillvis & Katz, 2013). This suggests that the expression of 5-HT receptors in C2 might play a role in the tendency of a species or an individual to exhibit DV swimming.

Multifunctionality of CPG Interneurons

Although the neurons are described as swim CPG neurons, they have other functions. In Tritonia, for example, DSI accelerates mucociliary gliding by exciting serotoninergic and peptidergic efferent neurons in the pedal ganglion (Popescu & Frost, 2002). Both VSI-A and VSI-B stimulation suppress gliding by inhibiting those same pedal neurons. Furthermore, tactile stimulation inhibits DSI while exciting VSI, suggesting that this mechanism would be responsible for halting gliding upon contact with a noxious mechanical stimulus (Popescu & Frost, 2002).

In Pleurobranchaea, in addition to their role in the swim CPG, the A1 neuron suppresses feeding; stimulation of A1 caused proboscis retraction in whole-animal preparations and a temporary cessation of the swim motor pattern in isolated brains (Jing & Gillette, 1995). The A1, A3, and A10 neurons indirectly inhibit PCP neurons, which are coordinating interneurons that play an important role in maintaining feeding activity (Jing & Gillette, 1995, 2000). They also inhibit so-called feeding command neurons, the PSEs (Jing & Gillette, 2000). In contrast, the serotonergic As1-4 neurons enhance feeding by evoking slow, long-lasting excitatory potentials in those same targets as well as the giant serotonergic neuron (MCG) and the other nearby serotonergic G neurons (Jing & Gillette, 2000). Thus, the DV swim CPG neurons have other, often opposing, actions when the swim motor pattern is not being generated (Gillette & Jing, 2001).

Left-Right Flexion Swimmers: Melibe leonina


Melibe swims by gracefully flexing from side to side (Watson et al., 2001; Newcomb, 2008). Although swimming can be used as an escape response, Melibe has been observed to swim spontaneously and the response has some directionality; the animal moves in the direction of the vertically aligned foot (Lawrence & Watson, 2002). The threshold for evoking the swimming behavior in Melibe is much lower than for Tritonia or Pleurobranchaea. Melibe lives on eel grass, and when it simply loses contact with the substrate, it begins to swim. This rhythmic movement can continue for hours if the animal is not disturbed. However, as soon as it touches a surface, it ceases swimming. Even the surface tension of the water is sufficient to turn off swimming.

Light inhibits the swimming behavior (Newcomb et al., 2004), and animals tend to swim more at night (Newcomb et al., 2014). This appears to be due to a circadian oscillator located in the eyes. However, extraocular photoreceptors can entrain an eyeless Melibe to a light/dark cycle. The motor pattern is inhibited by nitric oxide (NO); NO donors increase the period and cause the motor pattern to become more erratic (Newcomb & Watson, 2002). These effects might be mediated by a pair of nitrergic cerebral ganglion neurons that project to the pedal ganglia, but their behavioral function is not known (Newcomb & Watson, 2001).

Basic CPG Circuit

The swim CPG in Melibe consists of four bilaterally represented neurons (Si1-4), thus, eight total (Fig. 3A). Each neuron mutually inhibits its contralateral counterpart (Thompson & Watson, 2005; Sakurai, Gunaratne, & Katz, 2014). In addition, there is strong electrical coupling between the ipsilateral Si1 and Si2 and the contralateral Si4, causing them to burst in phase with each other. Si3 is excited by the Si1,2,4 kernel and feeds back inhibition onto it, helping to terminate that phase of the motor pattern. Thus, although the behavior is a simple, left-right alternation, the motor pattern has four phases with Si3 falling 25% behind the contralateral Si1,2,4 bursts (Fig. 3B). The function of Si3 is to provide feedback inhibition to the Si1,2,4 kernel, helping to terminate that phase of bursting.

Neural Control of Swimming in Nudipleura MollusksClick to view larger

Figure 3 Left/right flexion swim central pattern generators (CPGs) and motor patterns. (A) The Melibe swim CPG has four bilaterally represented neuron types, Si1, Si2, Si3, and Si4. The open circles represent neurons whose cell bodies are on the left side of the brain; closed circles are on the right. The ipsilateral Si1 and Si2 and the contralateral Si4 form a kernel of neurons that are strongly coupled and fire together. (B) The Melibe swim motor pattern was recorded from six neurons simultaneously. The shaded bars show the duration of three L-Si3 bursts. The dashed rectangle highlights the delay from the R-Si2 burst to the L-Si3 burst. (C) The swim CPG for Dendronotus (shaded box) consists of just four neurons: the left and right Si2 and Si3. Si1 does not make or receive reciprocal inhibition across the midline. (D) The swim motor pattern differs from that of Melibe. Si1 is not rhythmically bursting. Si3 bursts lead those of the contralateral Si2, but they are mostly synchronous with them (dashed rectangle).

Left-Right Flexion Swimmers: Dendronotus iris


Dendronotus swims like Melibe, with left-right flexions (Sakurai et al., 2011), but the ecological significance of its swimming is less well known. Unlike Melibe, Dendronotus is a bottom dweller. It feeds on burrowing sea anemone. Although Dendronotus will swim in response to contact with a starfish, it has also been observed to swim spontaneously and will swim when it loses contact with the substrate.

Basic CPG Circuit

The swim CPG for Dendronotus contains potential homologs of Si1-3 (Fig. 3C), but they are wired up differently (Sakurai & Katz, 2016; Sakurai et al., 2011). In Dendronotus, Si1 does not make or receive contralateral inhibition. As a result, Si1 does not fire rhythmic bursts during the swim motor pattern, but instead is irregularly active (Fig. 3D). Depolarization of Si1 speeds up the motor pattern, whereas hyperpolarization slows it down (Sakurai et al., 2011). Si2 and Si3 each inhibit their contralateral counterparts. However, unlike in Melibe, Si3 is electrically coupled to and excites the contralateral Si2 (Sakurai & Katz, 2016). Si3 provides feedforward excitation to Si2 that helps initiate the next burst. Thus, the swim CPG in Dendronotus is organized more like a conventional half-center oscillator than is the Melibe swim CPG. However, the CPG is twisted, so that the contralateral Si2 and Si3 fire in phase with each other.

Even though Dendronotus and Melibe exhibit homologous LR swimming behaviors and even though the CPGs are comprised of similar sets of neurons, in which at least one of the neurons (Si1) has been determined to be homologous, the mechanism underlying swimming differs. In Dendronotus, Si3 initiates the bursts of the contralateral Si2, whereas in Melibe Si3 helps terminate the bursts of the contralateral Si1 and Si2. Furthermore, Si1 acts as an extrinsic modulator of cycle period in Dendronotus, whereas in Melibe it is part of the CPG. Because the LR swimming behavior is found across many, if not all species in their clade (Fig. 1), it is thought to have been present in the common ancestor of those species. This means that Melibe and Dendronotus have diverged in their neural mechanisms for swimming.

Different Functions for Homologous Neurons in Species With Different Behaviors

The CPGs underlying DV swimming and LR swimming are composed of nonoverlapping sets of neurons. In other words, the homologs of DV swim CPG neurons are not in LR swim CPGs and vice versa. However, animals that are not DV swimmers have neurons that are homologous to DSI and C2 (Newcomb & Katz, 2007; Lillvis et al., 2012). DSI homologs in a variety of species with different behaviors make similar synaptic connections. In particular, the DSI homologs in several species synapse onto pedal efferent neurons (Newcomb & Katz, 2007). Even in more distantly related species, such as Aplysia californica, these neurons activate pedal neurons (Marinesco et al., 2004). This suggests that there is a conserved function of the DSIs as activating an arousal system, whereas their incorporation into a DV swim CPG is derived (Katz et al., 2001; Katz, 2016b).

Although the DSI homologs are not part of the LR swim CPG in Melibe, they modulate the swim motor pattern. Brief stimulation of a DSI initiates the motor pattern in quiescent preparations. Furthermore, DSI stimulation or serotonin application will speed up and regularize ongoing motor patterns (Newcomb & Katz, 2009). However, unlike in Tritonia or Pleurobranchaea, the DSI homologs are not necessary for the motor pattern and do not fire rhythmic bursts of action potentials in phase with it.


Work on nudipleuran swim CPGs highlights how simple rhythmic neural circuits can be. Furthermore, the phylogeny suggests that these CPGs have evolved several times independently. The work on Tritonia and Pleurobranchaea suggests that although some of the same neurons (i.e., homologous neurons) have been incorporated into the CPG circuit for an analogous behavior, other neurons in the circuit differ. Yet both species converged on the use of 5-HT modulation of synaptic strength within the CPG as a necessary mechanism. This could suggest that the serotonergic neuromodulation is a pathway for creating an oscillator from a nonoscillatory circuit. It will be of interest to test this hypothesis by inducing the modulation in nonswimming species through changes in gene expression.

The differences in the mechanisms underlying swimming in Melibe and Dendronotus demonstrate that species that share a homologous behavior can have divergent neural mechanisms for producing it. This is a particularly important cautionary tale when extrapolating the results from one species to another (Katz, 2016a).

The variability seen between individuals and across species is also indicative of the fact that there are many solutions to producing rhythmic activity (Sakurai & Katz, 2015). There are many examples of similar behaviors evolving independently using different circuit mechanisms (Katz, 2016a). There are also examples of similar circuit mechanisms evolving independently. This suggests that during the course of evolution, there are certain constraints and possibilities open to selection that together affect the evolvability of neural circuitry and behavior (Katz, 2011). Comparing neural mechanisms and the functions of homologous neurons gives us a better understanding of the important parameters underlying the production of behavior.


Brodfuehrer, P. D., Debski, E. A., O’Gara, B. A., & Friesen, W. O. (1995). Neuronal control of leech swimming. Journal of Neurobiology, 27, 403–418.Find this resource:

Calin-Jageman, R. J., Tunstall, M. J., Mensh, B. D., Katz, P. S., & Frost, W. N. (2007). Parameter space analysis suggests multi-site plasticity contributes to motor pattern initiation in Tritonia. Journal of Neurophysiology, 98, 2382–2398.Find this resource:

Dorsett, D. A., Willows, A. O. D., & Hoyle, G. (1969). Centrally generated nerve impulse sequences determining swimming behavior in Tritonia. Nature, 224, 711–712.Find this resource:

Dorsett, D. A., Willows, A. O. D., & Hoyle, G. (1973). The neuronal basis of behavior in Tritonia. IV. The central origin of a fixed action pattern demonstrated in the isolated brain. Journal of Neurobiology, 4, 287–300.Find this resource:

El Manira, A., Kettunen, P., Hess, D., & Krieger, P. (2002). Metabotropic glutamate receptors provide intrinsic modulation of the lamprey locomotor network. Brain Research Reviews, 40, 9–18.Find this resource:

El Manira, A., Kyriakatos, A., Nanou, E., & Mahmood, R. (2008). Endocannabinoid signaling in the spinal locomotor circuitry. Brain Research Reviews, 57, 29–36.Find this resource:

Farmer, W. M. (1970). Swimming gastropods (Opisthobranchia and Prosobranchia). The Veliger, 13, 73–89.Find this resource:

Frost, W. N., Brandon, C. L., & Mongeluzi, D. L. (1998). Sensitization of the Tritonia escape swim. Neurobiology Learning and Memory, 69, 126–135.Find this resource:

Frost, W. N., Brandon, C. L., & Van Zyl, C. (2006). Long-term habituation in the marine mollusc Tritonia diomedea. Biology Bulletin, 210, 230–237.Find this resource:

Frost, W. N., Brown, G. D., & Getting, P. A. (1996). Parametric features of habituation of swim cycle number in the marine mollusc Tritonia diomedea. Neurobiology Learning and Memory, 65, 125–134.Find this resource:

Frost, W. N., Hoppe, T. A., Wang, J., & Tian, L. M. (2001). Swim initiation neurons in Tritonia diomedea. American Zoologist, 41, 952–961.Find this resource:

Frost, W. N., & Katz, P. S. (1996). Single neuron control over a complex motor program. Proceedings of the National Academy of Sciences USA, 93, 422–426.Find this resource:

Frost, W. N., Tian, L. M., Hoppe, T. A., & Mongeluzi, D. L. & Wang, J. (2003). A cellular mechanism for prepulse inhibition. Neuron, 40, 991–1001.Find this resource:

Getting, P. A. (1976). Afferent neurons mediating escape swimming of the marine mollusc, Tritonia. Journal of Comparative Physiology, 110, 271–286.Find this resource:

Getting, P. A. (1977). Neuronal organization of escape swimming in Tritonia. Journal of Comparative Physiology A, 121, 325–342.Find this resource:

Getting, P. A. (1983a). Mechanisms of pattern generation underlying swimming in Tritonia. II. Network reconstruction. Journal of Neurophysiology, 49, 1017–1035.Find this resource:

Getting, P. A. (1983b). Mechanisms of pattern generation underlying swimming in Tritonia. III. Intrinsic and synaptic mechanisms for delayed excitation. Journal of Neurophysiology, 49, 1036–1050.Find this resource:

Getting, P. A. (1986). Understanding central pattern generators: Insights gained from the study of invertebrate systems. In S. Grillner, P. S. G. Stein, D. G. Stuart, H. Forssberg, & R. M. Herman (Eds.), Neurobiology of vertebrate locomotion (pp. 231–244). Houndmills, UK: MacMillan.Find this resource:

Getting, P. A. (1989a). Emerging principles governing the operation of neural networks. Annual Review of Neuroscience, 12, 185–204.Find this resource:

Getting, P. A. (1989b). Reconstruction of small neural networks. In C. Koch & I. Segev (Eds.), Methods in neuronal modeling: From synapses to networks (pp. 171–194). Cambridge, MA: MIT Press.Find this resource:

Getting, P. A., Lennard, P. R., & Hume, R. I. (1980). Central pattern generator mediating swimming in Tritonia. I. Identification and synaptic interactions. Journal of Neurophysiology, 44, 151–164.Find this resource:

Gillette, R., & Jing, J. (2001). The role of the escape swim motor network in the organization of behavioral hierarchy and arousal in Pleurobranchaea. American Zoologist, 41, 983–992.Find this resource:

Goaillard, J. M., Taylor, A. L., Schulz, D. J., & Marder, E. (2009). Functional consequences of animal-to-animal variation in circuit parameters. Nature Neuroscience, 12, 1424–1430.Find this resource:

Goodheart, J. A., Bazinet, A. L., Collins, A. G., & Cummings, M. P. (2015). Relationships within Cladobranchia (Gastropoda: Nudibranchia) based on RNA-Seq data: An initial investigation. Royal Society Open Science, 2, 150–196.Find this resource:

Hamood, A. W., & Marder, E. (2014). Animal-to-animal variability in neuromodulation and circuit function. Cold Spring Harbor Symposia on Quantitative Biology, 79, 21–28.Find this resource:

Hill, E. S., Sakurai, A., & Katz, P. S. (2008). Transient enhancement of spike-evoked calcium signaling by a serotonergic interneuron. Journal of Neurophysiology, 100, 2919–2928.Find this resource:

Hill, E. S., Vasireddi, S. K., Bruno, A. M., Wang, J., & Frost, W. N. (2012). Variable neuronal participation in stereotypic motor programs. PLoS One, 7, e40579.Find this resource:

Hill, E. S., Vasireddi, S. K., Wang, J., Bruno, A. M., & Frost, W. N. (2015). Memory formation in Tritonia via recruitment of variably committed neurons. Current Biology, 25, 2879–2888.Find this resource:

Jansen, A. S. P., Farkas, E., Mac Sams, J., & Loewy, A. D. (1998). Local connections between the columns of the periaqueductal gray matter: A case for intrinsic neuromodulation. Brain Research, 784, 329–336.Find this resource:

Jing, J., & Gillette, R. (1995). Neuronal elements that mediate escape swimming and suppress feeding behavior in the predatory sea slug Pleurobranchaea. Journal of Neurophysiology, 74, 1900–1910.Find this resource:

Jing, J., & Gillette, R. (1999). Central pattern generator for escape swimming in the notaspid sea slug Pleurobranchaea californica. Journal of Neurophysiology, 81, 654–667.Find this resource:

Jing, J., & Gillette, R. (2000). Escape swim network interneurons have diverse roles in behavioral switching and putative arousal in Pleurobranchaea. Journal of Neurophysiology, 83, 1346–1355.Find this resource:

Katz, P. S. (2009). Tritonia swim network. Scholarpedia, 4, 36–38.Find this resource:

Katz, P. S. (2010). The Tritonia central pattern generator. In G. Shepherd & S. Grillner (Eds.), Handbook of microcircuits (pp. 443–449). New York, NY: Oxford University Press.Find this resource:

Katz, P. S. (2011). Neural mechanisms underlying the evolvability of behaviour. Philosophical Transactions of the Royal Society of London B Biological Sciences, 366, 2086–2099.Find this resource:

Katz, P. S. (2016a). Evolution of central pattern generators and rhythmic behaviours. Philosophical Transactions of the Royal Society of London B Biological Sciences, 371, 20150057.Find this resource:

Katz, P. S. (2016b). Phylogenetic plasticity in the evolution of molluscan neural circuits. Current Opinion in Neurobiology, 41, 8–16.Find this resource:

Katz, P. S., Fickbohm, D. J., & Lynn-Bullock, C. P. (2001). Evidence that the swim central pattern generator of Tritonia arose from a non-rhythmic neuromodulatory arousal system: Implications for the evolution of specialized behavior. American Zoology, 41, 962–975.Find this resource:

Katz, P. S., & Frost, W. N. (1995). Intrinsic neuromodulation in the Tritonia swim CPG: the serotonergic dorsal swim interneurons act presynaptically to enhance transmitter release from interneuron C2. Journal of Neuroscience, 15, 6035–6045.Find this resource:

Katz, P. S., & Frost, W. N. (1996). Intrinsic neuromodulation: Altering neuronal circuits from within. Trends in Neuroscience, 19, 54–61.Find this resource:

Katz, P. S., & Frost, W. N. (1997). Removal of spike frequency adaptation via neuromodulation intrinsic to the Tritonia escape swim central pattern generator. Journal of Neuroscience, 17, 7703–7713.Find this resource:

Katz, P. S., Getting, P. A., & Frost, W. N. (1994). Dynamic neuromodulation of synaptic strength intrinsic to a central pattern generator circuit. Nature, 367, 729–731.Find this resource:

Kupfermann, I., & Weiss, K. R. (1978). The command neuron concept. Behavior Brain Science, 1, 3–39.Find this resource:

Lawrence, K. A., & Watson, W. H. (2002). Swimming behavior of the nudibranch Melibe leonina. Biology Bulletin, 203, 144–151.Find this resource:

Lennard, P. R., Getting, P. A., & Hume, R. I. (1980). Central pattern generator mediating swimming in Tritonia. II. Initiation, maintenance, and termination. Journal of Neurophysiology, 44, 165–173.Find this resource:

Lieske, S. P., & Ramirez, J. M. (2006). Pattern-specific synaptic mechanisms in a multifunctional network. II. Intrinsic modulation by metabotropic glutamate receptors. Journal of Neurophysiology, 95, 1334–1344.Find this resource:

Lillvis, J. L., Gunaratne, C. A., & Katz, P. S. (2012). Neurochemical and neuroanatomical identification of central pattern generator neuron homologues in nudipleura molluscs. PLoS ONE, 7, e31737.Find this resource:

Lillvis, J. L., & Katz, P. S. (2013). Parallel evolution of serotonergic neuromodulation underlies independent evolution of rhythmic motor behavior. Journal of Neuroscience, 33, 2709–2717.Find this resource:

Marinesco, S., Kolkman, K. E., & Carew, T. J. (2004). Serotonergic modulation in Aplysia: I. A distributed serotonergic network persistently activated by sensitizing stimuli. Journal of Neurophysiology, 92, 2468–2486.Find this resource:

McClellan, A. D., Brown, G. D., & Getting, P. A. (1994). Modulation of swimming in Tritonia: Excitatory and inhibitory effects of serotonin. Journal of Comparative Physiology A, 174, 257–266.Find this resource:

Mongeluzi, D. L., Hoppe, T. A., & Frost, W. N. (1998). Prepulse inhibition of the Tritonia escape swim. Journal of Neuroscience, 18, 8467–8472.Find this resource:

Newcomb, J. M. (2008). Melibe. Scholarpedia, 3, 3965.Find this resource:

Newcomb, J. M., & Katz, P. S. (2007). Homologues of serotonergic central pattern generator neurons in related nudibranch molluscs with divergent behaviors. Journal of Comparative Physiology A, 193, 425–443.Find this resource:

Newcomb, J. M., & Katz, P. S. (2009). Different functions for homologous serotonergic interneurons and serotonin in species-specific rhythmic behaviours. Proceedings of the Royal Society B, 276, 99–108.Find this resource:

Newcomb, J. M., Kirouac, L. E., Naimie, A. A., Bixby, K. A. Lee, C., Malanga, S., … Watson, W. H. (2014). Circadian rhythms of crawling and swimming in the nudibranch mollusc Melibe leonina. Biological Bulletin, 227, 263–273.Find this resource:

Newcomb, J. M., Lawrence, K. A., & Watson, W. H. (2004). The influence of light on locomotion in the gastropod Melibe leonina. Marine and Freshwater Behavior and Physiology, 37, 253–267.Find this resource:

Newcomb, J. M., Sakurai, A., Lillvis, J. L., Gunaratne, C. A., & Katz, P. S. (2012). Homology and homoplasy of swimming behaviors and neural circuits in Nudipleura molluscs. Proceedings of the National Academy of Sciences USA, 109(Suppl 1), 10669–10676.Find this resource:

Newcomb, J. M., & Watson, W. H. (2001). Identifiable nitrergic neurons in the central nervous system of the nudibranch Melibe leonina localized with NADPH-diaphorase histochemistry and nitric oxide synthase immunoreactivity. Journal of Comparative Neurology, 437, 70–78.Find this resource:

Newcomb, J. M., & Watson, W. H. (2002). Modulation of swimming in the gastropod Melibe leonina by nitric oxide. Journal of Experimental Biology, 205, 397–403.Find this resource:

Popescu, I. R., & Frost, W. N. (2002). Highly dissimilar behaviors mediated by a multifunctional network in the marine mollusk Tritonia diomedea. Journal of Neuroscience, 22, 1985–1993.Find this resource:

Ransdell, J. L., Nair, S. S., & Schulz, D. J. (2013). Neurons within the same network independently achieve conserved output by differentially balancing variable conductance magnitudes. Journal of Neuroscience, 33, 9950–9956.Find this resource:

Roffman, R. C., Norris, B. J., & Calabrese, R. L. (2012). Animal-to-animal variability of connection strength in the leech heartbeat central pattern generator. Journal of Neurophysiology, 107, 1681–1693.Find this resource:

Sakurai, A., Calin-Jageman, R. J., & Katz, P. S. (2007). The potentiation phase of spike timing-dependent neuromodulation by a serotonergic interneuron involves an increase in the fraction of transmitter release. Journal of Neurophysiology, 98, 1975–1987.Find this resource:

Sakurai, A., Darghouth, N. R., Butera, R. J., & Katz, P. S. (2006). Serotonergic enhancement of a 4-AP-sensitive current mediates the synaptic depression phase of spike timing-dependent neuromodulation. Journal of Neuroscience, 26, 2010–2021.Find this resource:

Sakurai, A., Gunaratne, C. A., & Katz, P. S. (2014). Two interconnected kernels of reciprocally inhibitory interneurons underlie alternating left-right swim motor pattern generation in the mollusc Melibe leonina. Journal of Neurophysiology, 112(6), 1317–1328.Find this resource:

Sakurai, A., & Katz, P. S. (2003). Spike timing-dependent serotonergic neuromodulation of synaptic strength intrinsic to a central pattern generator circuit. Journal of Neuroscience, 23, 10745–10755.Find this resource:

Sakurai, A., & Katz, P. S. (2009a). Functional recovery after lesion of a central pattern generator. Journal of Neuroscience, 29, 13115–13125.Find this resource:

Sakurai, A., & Katz, P. S. (2009b). State-, timing-, and pattern-dependent neuromodulation of synaptic strength by a serotonergic interneuron. Journal of Neuroscience, 29, 268–279.Find this resource:

Sakurai, A., & Katz, P. S. (2015). Phylogenetic and individual variation in gastropod central pattern generators. Journal of Comparative Physiology A: Neuroethology, Sensory, Neural, and Behavioral Physiology, 201, 829–839.Find this resource:

Sakurai, A., & Katz, P. S. (2016). The central pattern generator underlying swimming in Dendronotus iris: A simple half-center network oscillator with a twist. Journal of Neurophysiology, 116, 1728–1742.Find this resource:

Sakurai, A., Newcomb, J. M., Lillvis, J. L., & Katz, P. S. (2011). Different roles for homologous interneurons in species exhibiting similar rhythmic behaviors. Current Biology, 21, 1036–1043.Find this resource:

Sakurai, A., Tamvacakis, A. N., & Katz, P. S. (2014). Hidden synaptic differences in a neural circuit underlie differential behavioral susceptibility to a neural injury. eLife. doi: 10.7554/eLife.02598.Find this resource:

Sakurai, A., Tamvacakis, A. N., & Katz, P. S. (2016). Recruitment of polysynaptic connections underlies functional recovery of a neural circuit after lesion. eNeuro, 3, ENEURO.0056-16.2016.Find this resource:

Sillar, K. T., Reith, C. A., & McDearmid, J. R. (1998). Development and aminergic neuromodulation of a spinal locomotor network controlling swimming in Xenopus larvae. Annals of the New York Academy of Sciences, 860, 318–332.Find this resource:

Taghert, P. H., & Willows, A. O. D. (1978). Control of a fixed action pattern by single, central neurons in the marine mollusk, Tritonia diomedea. Journal of Comparative Physiology, 123, 253–259.Find this resource:

Thompson, S., & Watson, W. H. (2005). Central pattern generator for swimming in Melibe. Journal of Experimental Biology, 208, 1347–1361.Find this resource:

Watson, W. H., Lawrence, K. A., & Newcomb, J. M. (2001). Neuroethology of Melibe leonina swimming behavior. American Zoologist, 41, 1026–1035.Find this resource:

Willows, A. O. (1967). Behavioral acts elicited by stimulation of single, identifiable brain cells. Science, 157, 570–574.Find this resource:

Willows, A. O. D. (2001). Costs and benefits of opisthobranch swimming and neurobehavioral mechanisms. American Zoologist, 41, 943–951.Find this resource:

Willows, A. O. D., & Hoyle, G. (1968). Correlation of behavior with the activity of single identifiable neurons in the brain of Tritonia. In J. Salanki (Ed.), Neurobiology of invertebrates (pp. 443–461). New York, NY: Plenum Press.Find this resource:

Willows, A. O. D., & Hoyle, G. (1969). Neuronal network triggering of fixed action pattern. Science, 166, 1549–1551.Find this resource:

Wyeth, R. C., & Willows, A. O. (2006). Field behavior of the nudibranch mollusc Tritonia diomedea. Biology Bulletin, 210, 81–96.Find this resource: