Pleurobranchaea

• Prof. Rhanor Gillette, University of Illinois, Urbana, IL, USA

Pleurobranchaea californica

Specialists of Simplicity: Soft Bodies, Little Brains, and Low Cunning

The soft bodies and hydrostatic skeletons of Pleurobranchaea californica (Fig. 1) and other nudipleurans allow simple control of ciliary locomotion and turning. In the nervous system, only relatively small numbers of motorneurons and sparse central pattern generator (CPG) circuits are required to generate the simple flexions and stretches of the body needed in their behaviors. The animals are without need to coordinate segmented bodies or articulated appendages, or to keep track of where body parts are in space with proprioceptive input from any such articulations. Thus, smaller neuron numbers and relatively simple, multifunctional CPGs suffice for the activities of foraging, escape and reproduction.

Behaviors are also simple in their coordination: without an articulated skeleton, directional turns mostly involve asymmetric contracture of one side of the body relative to the other, powered by relatively few motor neurons. Even backwards substrate locomotion is unknown in these gastropods (as in most soft-bodied animals): retreat is effectively managed by body twisting and turning. For the nudipleurans (and pulmonate snails), the increased innervation needed by the expanding periphery during growth is not served by greater numbers of effector neurons, but by increasing neuron sizes that may serve to keep up with the synthetic and transport needs of the increased output synapses (Gillette, 1991; Yamagishi et al., 2011).

Figure 1: Two Pleurobranchaea californica near La Jolla Shores (photo, Tracy Clark).

Simplicity is markedly apparent in the multifunctional feeding network and its control of approach-avoidance decision. The encapsulation of multiple critical functions in the feeding network CPG of P. californica is highly efficient . The CPG is capable of assuming all of the needs for generating patterns of feeding, rejection and different modes of feeding suppression by prey avoidance learning or escape (Croll and Davis, 1984; Davis and Gillette, 1978). As to be described, the feeding CPG manifests appetitive state in its state of excitation, and its corollary output regulates approach-avoidance decision in the directional turn motor network. This simple model can be contrasted with the highly derived and markedly differentiated vertebrate brain, where feeding-related CPGs in cranial nuclei and spinal cord are subordinate to executive functions taken over by the basal ganglia and cortex (cost-benefit decision and premotor patterning) and the hypothalamus (appetitive state).

For these animals without vision, chemotactile information is used efficiently to identify odors previously associated with experience. Foraging success will be improved by combining that learning with hunger state and stimulus salience into appetitive state for approach-avoidance decisions. In particular, the abilities of the feeding and swim/turn CPGs to assume multiple configurations for quite varied expressions of behavior are marvels of biological engineering. The interactions of those CPGs lend adaptability to the foraging behavior and allow for best estimates in true cost-benefit decisions in foraging and escape. Reproduction has none of the frills of sexual competition or brooding the young. Complex hormonal relations between nutritive stores and appetitive state have not been found. The genus Pleurobranchaea shows by its broad distribution and speciations that with truly simple body plan, brain and behavior they are markedly successful in their simple lifestyles. Need for further complexity is apparently neutralized by their potent chemical defenses, which seems largely responsible for their marked success.

The Genus Pleurobranchaea

Members of the genus are actively predatory and around 60 species are distributed worldwide from subarctic to tropics. The Pleurobranchoidea – the “side-gilled slugs” – and the nudibranchs are considered sister clades in Nudipleura on the basis of shared anatomical and molecular characters (Wägele and Willan, 2000). Did some one or more drab Pleurobranchaea-like ancestors anciently engender the splendid nudibranchs of the dorids, aeolids and other families of flamboyant slugs, or did nudibranchs evolve in parallel from a common nudipleuran ancestor? Early in the history of molluscan systematics it was thought that the pleurobranchoids gave rise to the nudibranchs in evolution on the basis of shared characters in the basal nudibranchs. Like the nudibranchs, members of Pleurobranchaea’s subfamily Pleurobranchaeinae (three genera) are completely soft-bodied, with loss of adult shell. The other pleurobranchoid subfamily retains internal shells as a primitive character. While phylogenetic analysis presently does not have the resolution to answer the question, research in nudipleuran phylogeny is an interesting and ongoing mystery story; changing perspectives over 125 years are recounted by Wägele et al., (2014).

It seems likely that this genus, and the related genus [1] Euselenops, arose within the pleurobranchomorphs with actively predatory lifestyles of catholic tastes. Their relatively mobile lifestyle and wide-ranging foraging differ markedly from the other notaspids, which tend to patiently mumble sessile sponges and ascidians. Of the different species, most information is available for P. californica, which has been a significant model in neuroethological research since 1970. Much useful information is also available for P. japonica. Observations of feeding behavior for P. maculata (novazealandae) , P. brockii, P. bubala and P. meckelii indicate that opportunistic predatory and cannibal habits are strong characteristics of the genus.

P. californica and some of the other species have an escape swim consisting of cyclic dorso-ventral flexions; however, as will be pointed out, they may have few predators outside of their conspecifics.

Lifestyle

P. californica inhabits cool waters along the Pacific coast of North America from Baja California to Oregon, being collected and observed on muddy and rocky bottoms ranging to 1200 meters in depth. When cooler waters prevailed near the surface in the 1970s, they were collected in abundance in California in shallower depths up to 6 meters. Warmer waters during the El Niño of 1982/83 coincided with withdrawal of the population into deeper waters out of normal SCUBA range, where collection for study required use of otter trawls and traps set around 100 meters in depth. A circadian rhythm of nocturnal activity, inferred by scuba divers in the field when animals were in shallower depths, has also been observed in divers’ sightings of other Pleurobranchaea species [2] [3]. Eyes are internal, and their contributions, or those of extraocular photoreceptors, to diurnal rhymicity are unknown, but it is supposed that circadian photic entrainment is active only in the upper depths.

P. californica are opportunist and generalist feeders, as well as enthusiastic cannibals. In the lab, individuals must be separated to prevent losses. Gut contents of freshly caught animals have frequently contained smaller P. californica, other sea-slugs like Armina californica, Tritonia diomedea, lophophorate worms and the squid, Loligo opalescens. Squid parts often stuff the guts during the squid spawning season and are probably taken when squid sink exhausted to the sea-floor after spawning. P. californica have been sighted in the wild eating fish carrion, scyphozoan jellyfish and anemones.

Cannibalism for P. californica may be important in its foraging economy. In the laboratory larger animals readily attack smaller. However, studies in progress show that smaller animals typically recoil from the touch of a larger one, and react with avoidance turns and locomotion, and, not uncommonly, escape swimming. Cannibalism may well have been the ultimate cause of selection in the evolution of escape swimming in this genus. The triggering of avoidance in a smaller animal by a larger appears to be caused by a mucus factor. Avoidance habituates with repeated contact. It is hypothesized that animals secrete an aversive factor in proportion to their size and habituate to their own levels. Thus, smaller animals may respond to higher levels of factor in larger ones, and larger animals recognize smaller only as suitable prey. There are few other documented specific defenses to conspecific cannibalism (but see Zimmer et al., 2006).

No regular predators of P. californica are known outside of their conspecifics. This may be due in part to their defensive skin secretion of sulfuric acid in response to noxious stimuli, in which skin pH may drop from near that of sea-water to pH 1.5 (Gillette et al., 1991). Volunteer human tasters (!) report no more salient quality than a sour (acidic) taste.

Chemical Ecology

Many invertebrates of the pacific coast show escape responses to the touch of a P. californica, including abalones Haliotis kamschatkana and H. rufescens, various intertidal limpets (Acmaea spp. ), seaslugs Tritonia diomedea and Dendronotus spp., and the swimming anemone Stomphia coccinea, similar to the same animals’ responses to starfish (personal observations). In the video at https://vimeo.com/100379315 a nudibranch prey of a P. californica performs a vain escape jump a fraction of a second after being touched. The mucus factor recognized by these animals may be the same that triggers avoidance of small P. californica to larger individuals. The maladaptiveness of the predators broadcasting their identities to potential prey is perhaps offset by a defensive role of the aversive factor or in a somehow greater usefulness of their odor signatures in reproductive recognition.

The Nudipleura appear to have traded their protective external shells for effective chemical defenses (Cimino and Ghiselin, 2009). For the pleurobranchoids, defensive acid secretion (pH ~1.5) in response to noxious stimuli is a characteristic predator deterrent (Thompson, 1984). Acid secretion is accompanied by withdrawal and avoidance, behaviors that can also be elicited by acidified sea-water less acid than the animals’ secretions (Gillette et al. 1991). Thus, the animals’ defensive exocrine secretions must potentiate their defensive behaviors, similar to actions of endocrine adrenaline in mammals. However, acid secretion is unlikely to be the aversive factor causing escape behavior in conspecifics and other invertebrates, since such secretion is of longer latency by seconds than the observed behavioral responses.

It is notable that the common organic acid and osmolyte taurine is a stimulator of skin acid secretion applied anywhere on the body at concentrations of 10^-5-10^-2 M (Gillette et al., 1991). It may be that injury to the skin by a noxious stimulus causes taurine release and consequent acid secretion. Indeed, taurine is a large component of the acid secretion induced by mechanical stimulation in the related pleurobranchoid Berthellina citrina (Moustafa et al. 2014).

Tetrodotoxin (TTX), a blocker of fast Na+ currents in many vertebrates and invertebrates, has been identified in P. maculata in New Zealand and Australia (MacNabb et al., 2010) as a cause of dog poisonings from eating carcasses thrown up on the beach. Khor et al. (2013) found that P. maculata throve on and even preferred a TTX-containing diet. P. californica might also contain TTX at times; Na+ current in its neurons is unaffected by TTX at concentrations that largely block the current in Aplysia (W.F. Gilly and R. Gillette, unpublished observations), similar to animals that normally carry TTX. TTX is found in other actively predatory molluscs, such as the naticid moon snails (Hwang et al., 1991). Whether they concentrate TTX from the food web, or may sometimes harbor the symbiotic bacteria that synthesize it, remains an open question. Whatever the cause, it may be another reason why the sea-slug has so few known natural predators outside of cannibal opportunists.

Reproduction

P. californica is an approximately annual species that grows throughout its lifetime. The largest specimen observed by this writer, with a volume of 3.1 liters and slightly more than neutral density, was collected in Monterey Bay in 1977. More commonly, sexually mature animals collected are between 0.6-1.2 liters. Of this volume, about 65% is hemocoele blood, an unusually large fraction for a nudipleuran, and for which a biomechanical justification is obscure. Animals breed as simultaneous hermaphrodites and attach an attractive curtain-like egg ribbon to exposed surfaces. Egg ribbons tend to tip off the daytime diver of their presence before individuals are seen. Captives in aquaria have been observed to copulate and lay eggs almost nightly, losing appreciable body mass with each clutch deposited until eventual death. During the drastic El Niño event of 1982/83 most animals collected by trawl in both northern and southern California were less than 0.1 L, of which many were found to be sexually mature – a likely adaptation to a food supply made scarce.

Only a little is known about reproductive physiology. Reproductive organs receive the densest serotonergic innervation from the central nervous system of any peripheral site (Moroz et al., 1997). The significance is speculative. Ram and colleagues showed that an egg-laying hormone of similar molecular weight and actions to that of Aplysia was found in a cluster of neurons in a small lobe of each pedal ganglion. Injection of an homogenate into the hemocoele quickly suppressed feeding behavior and induced egg-laying (Ram et al., 1977).

Foraging

Feeding Behavior

Hungry P. californica react to food stimuli with a lunging bite/strike with a rapid eversion of the proboscis and seizure of prey with the radula, leading to rhythmic biting and swallowing. Live prey brought into the buccal cavity can be rapidly subdued by copious secretion of dilute sulfuric acid from the duct of the extensive acid gland (Morse, 1984). The active predatory habit of P. californica promoted its use as a model system for study of learning, motivation and decision-making (Hirayama et al., 2013). The readiness to feed is determined by nutritional status, hormonal status, sensory stimuli, health, and memory of experience. The most potent single appetitive chemical stimulus yet found is betaine (trimethylglycine), followed by glycine and cysteine (Gillette et al., 2000). Betaine and glycine are common osmolytes of marine invertebrates, and are detectable by hungry P. californica at micromolar concentrations. Protein (bovine serum albumin), trimethylamine oxide and 5’-AMP, appetents for some other marine invertebrate predators, are inactive. Glycine is also a potent appetent for P. japonica (Chiken et al., 2001).

For animals fed in the lab and held at 14-15 °C, satiation decays over a week (Davis et al., 1977). Satiation in gastropods following a meal is caused by bulk stretch of the gut, mediated by afferents to the feeding motor network that suppress feeding output (Susswein and Kupferman 1975; Croll et al., 1987; Elliot and Benjamin, 1989). Satiated animals show elevated feeding thresholds and may respond to food stimuli with active avoidance behavior (see further). Satiation effects of digestive hormones or digestive products like glucose do not appear to be important. However, likely longer-term satiation mechanisms involving serotonin (5 HT) have been identified in P. californica (see further).

P. californica readily learns the values of food stimuli. For animals trained to avoid food stimuli (e.g., squid homogenate) in a food-shock paradigm, feeding thresholds for proboscis extension and the bite/strike rise 100- to 1000-fold (Mpitsos and Collins, 1978). P. japonica can also be classically conditioned in a food-avoidance task (Chiken et al., 2009), and this ability may be widespread among the genus as an adaptation to an opportunistic predatory lifestyle.

The animal exploits many different prey, some of which can be costly to attack. When offered Flabellina iodinea , an aeolid nudibranch with a stinging defense, biting attack was followed by quick rejection and aversive turns as the sea slug rapidly learned to avoid its odor (Noboa and Gillette, 2013; http://www.youtube.com/watch?v=-EuVvkQ3VYM). Aposematic odor learning was selective: avoidance was not linked to change in feeding thresholds, and trained animals readily attacked and consumed a related aeolid, Hermissenda crassicornis. Notable exceptions were animals with extremely low and high appetites (extreme feeding thresholds) that either ignored F. iodinea or completely consumed it, respectively. Experienced slugs showed strong avoidance of F. iodinea for days after exposure. Thus, for P. californica, aposematic learning is a cognitive adaptation in which sensation, motivation and memory are integrated to direct cost–benefit choice, and thereby lends flexibility to the generalist’s foraging strategy.

Behavioral and Decision Elements of Foraging

Major parameters regulating foraging are rhythms of arousal and appetitive state. Animals are nocturnally active; divers report they tend to be cryptic in daytime, hiding under rocks and in old sunken tires. Hungry animals are more active and exploratory, while the satiated tend to a quiescent state where they may convert their meal to body mass and reproductive potential.

During exploratory locomotion animals repeatedly touch the substrate with tentacles and oral veil (http://www.life.illinois.edu/slugcity/movie/feeding2.mpg), which may aid in prey tracking (Yafremava et al., 2007). Both oral veil/tentacles and rhinophores may function in tracking odor plumes, with rhinophores acting in rheodetection of prevailing currents, as for other sea-slugs.

P. californica’s horizontal, chemotactile oral veil (Fig. 2) is like a functional composite of the mammalian tongue and olfactory systems. It has primary appetent receptors for amino acids (but not sweet or bitter chemicals) that may directly assess nutritive content, and others that must encode odors for associative learning of stimulus and consequence. For prey-tracking, the animal averages chemotactile stimulus strength and multiple stimulus sites on the oral veil into fairly precise turn angles that optimize the chase (Yafremava and Gillette, 2011).

Figure 2: Pleurobranchaea with oral veil and tentacles indicated, and the isolated nervous system showing cerebropleural ganglion (top), pedal ganglia, and the buccal ganglion.

Yafremava et al. (2007) studied directional turn responses to oral veil stimulation and showed that turn amplitudes and direction are affected by appetitive state. Moreover, a form of working memory is present, such that chemotactile stimuli can induce sequential turns in which the angles of the first and third turns are similar. Thus, animals tracking a trail or odor plume might turn away temporarily when the trail or plume is lost for a chance of finding it nearby, but turn back again if it is not found. These observations outline a framework for efficient tracking of odor trails, which is regulated by decision mechanisms that integrate sensation, internal state and experience.

Very hungry animals will not only attack very weak food stimuli, but will also vigorously attack mildly noxious stimuli (taurine, acidic solutions and mechanical stimuli). However, for less hungry animals with higher feeding thresholds, sub-threshold food stimuli may actually induce active avoidance (Gillette et al., 2000). These observations reflect a simple cost-benefit computation for decision in foraging, where food stimuli above or below the incentive level for feeding induce feeding or avoidance, respectively.

The Neuronal Elements of Foraging

Appetitive state, long a mostly inferred physiological construct, was found to be expressed in the excitation state of the homeostatic feeding network (Hirayama and Gillette, 2012). Feeding thresholds were closely related to spontaneous activity of neurons in the feeding network. The level of feeding network excitation state was found to regulate approach-avoidance decision, where avoidance turning motor activity was changed to orienting at higher levels of feeding network activity. It was inferred that the turning motor network is by default organized to respond to sensory input with avoidance, and that increasing feeding arousal state reconfigures the turn network to respond with approach behavior.

Satiation (an internal state) sums into appetitive state with effects of sensation and learning. Both satiation and general arousal mechanisms entail serotonin (5 HT), a modulator of the feeding network (Palovcik et al., 1982; Jing and Gillette, 2003; Hirayama et al., 2014). 5 HT from interneurons in the feeding network regulates excitation state and arousal, much like orexin in mammals (Gillette, 2006). 5 HT raises general arousal and reduces feeding thresholds, thereby promoting appetitive behavior. 5 HT levels in major serotonergic neurons of the feeding network rise and fall inversely with satiation state (Hatcher et al., 2008). Concomitantly, CNS isolated from hungry or satiated donors respond to oral veil nerve stimulation with fictive orienting or avoidance turn output, respectively. Thus, isolated CNS retain elements in their fictive behavior that conserve behavior of the intact donors, likely influenced by neuronal levels of 5 HT.

The feeding network is a final integration site for sensation, internal state and memory in cost-benefit decision. The effects of learning and satiation are found at strategic sites in the feeding network, where they control responses to aversively learned odors and appetitive stimuli (Davis and Gillette, 1978; Davis et al., 1982). In naïve hungry animals food stimuli induce network excitation, culminating in rhythmic feeding motor output . However, in odor-avoidance trained and in partially satiated animals the feeding network is inhibited, raising feeding thresholds by 100-1000 fold. In these cases, active feeding can be released by increasing the stimulus strength or by artificially providing excitation directly to the feeding network via feeding command neurons. Excitation state of the feeding network thus integrates the contributions of learning and motivation into its responsiveness to sensory input, and so embodies appetitive state in the animal’s moment-to-moment regulation of readiness to feed.

A neuronal sequel of food avoidance conditioning is inhibition by appetitive stimuli of feeding command neurons necessary and sufficient to active feeding (Fig. 3). This arises through a tonic, food-induced activation of retraction neurons in the protraction-retraction oscillator network of the feeding motor network, which is the origin of the synaptic inhibition of the command neurons – themselves part of the protraction neuron population (ibid.). Accordingly, the feeding network in the food-avoidance trained and satiated animals appears to be locked up in the retraction phase of the feeding cycle.

Figure 3: The PCp feeding command neuron in naïve (A) and food avoidance (B) trained animals stimulated with squid extract.

The general model of Figure 4 captures the cost-benefit computation that foraging animals make in the integration of sensation and internal state. The model provides general circuit relations to confer ability to react to likely costs and benefits of a given action in a given situation, and for deciding on appropriate behavior. Such observations join others to indicate that the behaviors of P. californica and other simple invertebrates are organized hedonically, such that they utilize affect as information in their decision-making like higher vertebrates (Schwarz and Clore, 1983). Thus, like more complex animals they do not use information per se in their approach-avoidance decision making, but they make those decisions on the basis of how information makes them feel.

Figure 4: A model of the relations regulating approach-avoidance decision in simple animals like P. californica. Excitation state in the feeding network integrates sensation, internal state and memory to express appetitive state and control turn direction. Sensory inputs for resource quality, sensory signatures,and nociception access sensory networks for Incentive and Somesthesia, which promote excitation of feeding and map stimulus position on the oral veil, respectively. The turn network is organized by default for an avoidance turn. Excitation in the feeding network suppresses avoidance and promotes orienting turns (approach). Active avoidance and satiation inhibit the feeding network. Modulatory feedback pathways from the Feeding and Avoidance networks potentiate learning of sensory signatures, mediating positive and negative reward. Modified from Hirayama and Gillette (2012).

In the predator’s natural environment these neural relations must sustain the essential behavioral economics of foraging. Thus, avoidance of appetitive stimuli below orienting thresholds represents a negotiation of expected nutritive benefit against costs of meal acquisition and risk of becoming prey to another predator attracted by the scent (e.g., a cannibal conspecific). Otherwise, orienting and attacking a stimulus reflects a decision in which estimated benefit offsets costs and risks. Here, appetitive state itself captures the cost-benefit computation as it is embodied in the excitatory configuration of the goal-directed feeding network. This decision mechanism can weigh the animal’s need for nutrient against potential risk from other predators or prey defenses and the cost of energy outlay in an attack on prey. The tuning of the decision mechanism need not come from conscious apprehension of risk, but simply the selective pressures in evolution affecting the balances of approach and avoidance output.

The interactions of the goal-directed feeding network of the mollusk with its turn network form a decision module acting at a simplest cognitive level in this solitary, cannibal predator. The uncomplicated module forms a potentially fundamental type of core circuitry around which the more complex neuronal circuit functions of valuation and comparison may have been elaborated in the social vertebrates. As such, it provides a useful starting point for considering the evolution of more complex systems, and it invites future modeling for adding neural and behavioral complexity.

Locomotion, Escape, Goal-Directed Turns and Behavioral Arousal

Substrate locomotion, escape swimming and turn responses to sensory stimuli are all thought to be driven by a multifunctional motor network capable of activity in multiple modes of coordination. Invertebrate studies of motor control provide diverse examples of neuronal networks able to function in multiple states of coordination (Kristan and Gillette, 2007). In particular for P. californica, elements of the multifunctional CPG for escape swimming and directional turning have been located in a group of neurons (the “A-cluster”) on the dorsal cerebral lobe of the cerebropleural ganglion. These are mostly interneurons with axons descending to the pedal ganglia and a small set of neurons with local cerebral connections. The neurons’ roles and connectivity in the CPG, and the state transitions of the multifunctional network, are identified and well documented (Jing and Gillette, 1995-2003). The A-cluster neurons compose a multi-functional network that drives quite different behaviors in different states of coordination. Several elements of the multifunctional CPG, including the As neurons, are easily recognizable as homologs in other molluscan species. The A-cluster neurons resemble in their connectivity and functions the neural architecture that controls the trunk locomotor pattern generators in annelid and vertebrate (cf. Tosches and Arendt, 2013). Are there some deep homologies? The general natures of the multifunctional network and its transitions are only beginning to be touched (also see Crow et al., 2013, for discussion of a highly derived homologous system for visual control of phototaxis).

i. Locomotion

Substrate locomotion is likely driven by the serotonergic As neurons, as Popescu and Frost (2002) have shown for the homologous DSI neurons of Tritonia to do. The As cells make excitatory, neuromodulatory synapses on the putative ciliolocomotor neurons of the pedal ganglia that innervate the foot, themselves serotonergic (Jing and Gillette, 1999).

ii. Escape Swimming

P. californica has an escape swim consisting of cyclic dorso-ventral flexions. Members of other pleurobranchoid genera like Pleurobranchus spp. and Berthella spp. contrast with Pleurobranchaea spp. in retaining a sizable internal shell that must inhibit their dorso-ventral and lateral flexibility. Accordingly, swimming behaviors involving those flexions are not known for those genera. However, Pleurobranchus membranaceus, which swims in the plankton as a mode of dispersal and concentration in coastal eddies, has an enlarged foot and swims on its back with an ungainly, rocking alternating flexion of the two sides of the foot (Thompson and Slinn, 1959) (see video ).

Escape swimming is an episode of alternating dorsal and ventral body flexions that overrides all other behaviors to lift the animal into prevailing current and away from danger. It closely resembles that described for Tritonia diomedea. However, most stimuli that induce escape in T. diomedea, touch by starfish or salt crystals dropped on the back, fail in P. californica; the predatory starfish Pycnopodia helianthoides avoids P. californica. Electric shock works well, as does the sting of the anemone Stomphia coccinea (unpublished observation). However, the most effective natural stimulus known is the touch or bite of another, usually larger, P. californica, perhaps partly because of the aversive factor described above.

The swim CPG comprises at least eight bilaterally paired interneurons (Fig. 5), each of which contributes to the swim rhythm. The motor pattern emerges from the integrated activity of four of the interneurons, which is driven by the action of one, the “swim command cell” A10. Driving a single A10 with injected current artificially induces the entire swim rhythm, which however ceases immediately upon stopping the injected depolarizing current. For the swim to occur in a natural sustained fashion, once triggered, the action of the four serotonergic As neurons is required. The swim CPG resembles other known CPGs where command cells like A10 act as potent positive feedback elements of the system to drive the motor output, which is patterned and sustained by recurrent excitation among interneuron partners and recurrent inhibition between antagonist populations, all supported by neuromodulatory action throughout the swim episode. The slow and sustained neuromodulatory depolarization of the collective CPG cells that supports multiple swim cycles only occurs when the serotonergic As cells are active in relatively intense activity, firing spikes during the cycle phase of dorsal flexion. In this instance the As cells act as positive feedback modulatory elements within the CPG, providing the slow motive excitation to drive it. The interested reader is referred to the original papers (Jing and Gillette, 1995-2003).

Figure 5: Patterned activity in the neurons of the escape swim central pattern generator (A) and the underlying neuronal circuitry (B).

It is of interest that only a fraction of the P. californica population performs escape swims. A salient difference between swimming and non-swimming animals is the extent of serotonergic modulation in the CPG neurons (Lillvis and Katz, 2013). When not swimming, the swim CPG neurons have other duties in turning, locomotion and likely other motor behaviors. The known neurons of the swim CPG of P. californica have likely homologs in the swim CPG cells of Tritonia, save the swim command neuron A10 and one of the serotonergic interneurons. The evolution of the swim CPG is a subject of interesting and ongoing investigations (Newcomb et al., 2012; Lilvis and Katz, 2013). The swim CPG of the multifunctional network is thought to have evolved on top of a neuron set with duties common to most molluscs, including locomotion, turning, righting and withdrawal.

iii. The Turn

The turn CPG is composed of several neurons that are alternatively active in the escape swim and at least one that is not (Figs. 6).

Figure 6: The turn network circuit of P. californica.

It is a bilaterally symmetrical neuronal network where turns to right or left are driven by higher activity in the contralateral side of the motor network. The asymmetrical activity sustains the turn through the duration of the movement by causing asymmetric contracture of the turn musculature of the body wall and foot (Jing and Gillette, 2003). The bilateral A4 “turn command neuron” pair drive contralateral turns when stimulated individually, but their activity (and the turn movement) requires the activity of the same serotonergic As neurons that participate in the swim network. However, the actual goal-orientation of the turn – approach or avoidance – is regulated by activity in the feeding motor network (Hirayama and Gillette, 2012).

A model for feeding network action in regulating approach-avoidance decision in the turn suggests that corollary outputs redirect sensory inputs from one side of the turn network to the other (Fig. 7; Hirayama et al., 2013). Thus, sensory inputs from the oral veil to the turn network are directed by default to elicit avoidance turn motor output, and feeding network activity causes a switch of the inputs to the contralateral side of the turn network. Thereby the approach-avoidance decision is regulated by appetitive state (combined sensory input, satiation, and learning) through the interactions of the feeding and turn networks via the simple switch mechanism.

Figure 7: The hypothetical circuit relations for control of the approach-avoidance decision by activity in the feeding motor network.

iv. Arousal

The serotonergic neurons throughout the CNS’s of Clione limacina, P. californica, and Aplysia californica have been shown to be coupled through excitatory chemical and electrical connections, so that they appear to form a “distributed arousal network” (reviewed in Jing et al., 2009) that may underlie the general behavioral arousal of the hungry animal, which can transit rapidly between orienting and avoiding, locomotion, feeding, and escape. The distributed network may be persistently activated by sensitizing stimuli, thereby regulating the learned sensitization of avoidance behavior (Marinesco and others 2004). The serotonergic As cells are persistently activated by swim-inducing stimuli and promote arousal state of the feeding and locomotor networks by neuromodulatory connections (Jing and Gillette, 1999). Figure 8 illustrates a central role for the As1-4 cells in the arousal network.

Figure 8: The serotonergic As1-4 interneurons have central roles in the swim, the turn and locomotion, and are key elements in the distributed serotonergic arousal network.

Overview

Pleurobranchaea, and nudipleurans in general, are relatively simple animals for their morphologies, sensory-motor abilities, reproductive strategies and behavioral economies. Their soft bodies and lack of articulated segmentation and appendages simplify motor control and proprioception so that behavioral choice within their limited repertories occurs largely through direct interactions of the motor networks involved. This completely avoids the organizational needs for executive controls evolved in vertebrate brains as specialized basal ganglia, pallium, cerebellum and hypothalamus; and in insect brains by the mushroom bodies and central complex.

Nudipleuran cognitive abilities, in terms of goal-directed use of knowledge, may extend no further than the learned influences in approach-avoidance decisions shown so far in very motile generalist foragers P. californica and Hermissenda. There are probably subtle aspects of learning that optimize the adaptations of other nudipleurans to their narrow niches that will be appreciated at some later time. Finally, the simple neural and behavioral characters of the genus Pleurobranchaea, and its success, aptly represent the advantages of potent chemical defenses in deterring predation. It may well be that the effectiveness of those chemical defenses is the major factor that has effectively functioned to reduce selective advantages of evolving further complexity in body form, cognition, or reproduction.

References

• Chiken S, Kuwasawa K, Kurokawa M and Ohsuga K (2001) Amino Acid-induced Reflexes and Their Neural Pathways in an Opisthobranch Mollusc Pleurobranchaea japonica. Zool. Sci. 18: 465–473.
• Chiken, S, Kuwasawa, K and Kurokawa, M. (2009) A neural analysis of avoidance conditioning with the feeding attractant glycine in Pleurobranchaea japonica. Comp. Biochem. Physiol. Part A 154:333–340.
• Cimino G and Ghiselin M (2009) Chemical Defense and the Evolution of Opisthobranch Gastropods. Proc Calif Acad Sci Series 4, Vol. 60, 175-422.
• Croll RP, Albuquerque T, Fitzpatrick L (1987) Hyperphagia resulting from gut denervation in the sea slug Pleurobranchaea. Behav Neural Biol 47:212–18.
• Davis WJ and Gillette R (1978) Neural correlates of behavioral plasticity in command neurons in Pleurobranchaea. Science 199:801 804.
• Crow T, Jin, NG, and Tian LM. Network interneurons underlying ciliary locomotion in Hermissenda. J Neurophysiol 109:640–648, 2013.
• Davis WJ, Gillette R. Kovac MP, Croll RP, and Matera EM (1983) Organization of synaptic inputs to paracerebral feeding command interneurons of Pleurobranchaea californica III. Modifications induced by experience. J. Neurophysiol. 49:1557 1572.
• Davis, WJ and Mpitsos GJ (1971) Behavioral choice and habituation in the marine mollusk Pleurobranchaea californica. Z. Vergl. Physiol. 75:207 232.
• Davis WJ, Mpitsos GJ, Pinneo JM and Ram JL (1977) Modification of the behavioral hierarchy of Pleurobranchaea. I. Satiation and feeding motivation. J Comp Physiol 117:99 125.
• Elliot CJH and Benjamin PR (1989). Esophageal mechanoreceptors in the feeding system of the pond snail, Lymnaea stagnalis. J Neurophysiol 61:727–36.
• Gillette, R (1991) On the significance of neuron giantism in gastropods. Biol Bull 180:234-240.
• Gillette R. Evolution and function in serotonergic systems. (2006) Integ. Comp. Biol. 46:838-846.
• Gillette R, Huang R-C, Hatcher N and Moroz LL (2000) Cost-benefit analysis potential in feeding behavior of a predatory snail by integration of hunger, taste and pain. Proc Natl Acad Sci, USA 97:3585-3590.
• Gillette R, Saeki M and Huang R-C (1991) Defense mechanisms in notaspidean snails: Acid humor and evasiveness. J Exp Biol 156:335-347.
• Göbbeler K and Klussmann-Kolb A (2010) Out of Antarctica? – New insights into the phylogeny and biogeography of the Pleurobranchomorpha show them as the sister group of the nudibranchs (Mollusca, Gastropoda). Molec Phylogen Evol 55 (2010) 996–1007.
• Hatcher NG, Zhang X, Potgieter K, Moroz LL, Sweedler JV. and Gillette, R. (2008) 5 HT and 5 HT-SO4, but not 5-HIAA, track hunger state in single feeding neurons. J Neurochem 104:1358-1363.
• Hwang DF, Tai KP, Chueh CH, Lin LC and Jeng SS. (1991)Tetrodotoxin and derivatives in several species of the gastropod Naticidae. Toxicon 29:1019-24.
• Hirayama K and Gillette R (2012) A neuronal network switch for approach/avoidance toggled by appetitive state. Curr Biol 22(2):118-123
• Hirayama K, Moroz LL, Hatcher NG, Gillette R (2014) Neuromodulatory Control of a Goal-Directed Decision. PLoS ONE 9(7):e102240. doi:10.1371/journal.pone.0102240
• Jensen KR (1997) Cannibalism in an opisthobranch mollusc, Pleurobranchaea brockii Bergh, 1897 from Hong Kong waters. Phuket Marine Biological Center Special Publication 17(1):41-46.
• Jing J and Gillette R (1995) Neuronal elements that mediate escape swimming and suppress feeding behavior in the predatory seaslug Pleurobranchaea. J Neurophysiol 74:1900-1910.
• Jing J and Gillette R (1999) The central pattern generator for escape swimming in the sea slug Pleurobranchaea californica. J Neurophysiol 81:654-657.
• Jing J and Gillette R (2000) Distinct corollary outputs of the escape swim motor network mediate a behavioral switch and excite a distributed serotonergic network in Pleurobranchaea. J Neurophysiol 83:1346-1355.
• Jing J and Gillette R. (2003) Directional avoidance turns encoded by single neurons and sustained by multifunctional serotonergic cells. J Neurosci 23:3039-3051.
• Jing J, Gillette R Weiss KR (2009) Evolving Concepts of Arousal: Insights from Simple Model Systems. Rev Neurosci 20:405-427.
• Khor S, Wood SA, Salvitti L, Taylor DI, Adamson J, McNabb P, Cary SC. (2013) Investigating diet as the source of tetrodotoxin in Pleurobranchaea maculata. Mar Drugs 12:1-16.
• Kristan WB and Gillette R. Behavioral choice. In, Invertebrate Neurobiology. North G and Greenspan RJ. Eds. Pub. Cold Spring Harbor Laboratory Press, 2007.
• Lillvis JL, Katz PS. (2013) Parallel evolution of serotonergic neuromodulation underlies independent evolution of rhythmic motor behavior. J Neurosci 33:2709-2717.
• Martynov AV and Schrödl M (2009) The new Arctic side-gilled sea slug genus Boreoberthella (Gastropoda, Opisthobranchia): Pleurobranchoidean systematics and evolution revisited. Polar Biol (2009) 32:53–70.
• McNabb P, Selwood AI, Munday R, Wood SA, Taylor DI, MacKenzie LA, van Ginkel R, Rhodes LL, Cornelisen C, Heasman K, Holland PT and King C. (2010) Detection of tetrodotoxin from the grey side-gilled sea slug – Pleurobranchaea maculata, and associated dog neurotoxicosis on beaches adjacent to the Hauraki Gulf, Auckland, New Zealand. Toxicon 56, 466-473.
• Moroz LL, Sudlow LC, Jing J and Gillette R. (1997) Serotonin-immunoreactivity in peripheral tissues of the opisthobranch molluscs Pleurobranchaea californica and Tritonia diomedea. J Comp Neurol 382:176-188.
• Morse MP (1984) Functional adaptations of the digestive system of the carnivorous mollusc Pleurobranchaea californica. MacFarland, 1966. J Morphol 180:253-262.
• Mpitsos GJ and Collins SD (1978) Learning: Rapid aversive conditioning in the gastropod mollusk Pleurobranchaea. Science 188:954-957 (1978).
• Moustafa AY, Wägele H, El behairi SEI (2014) Toxicological and Biological Activities of the Acid Secretion of Berthellina citrina (Heterobranchia, Pleurobranchidae) from the Red Sea. Internatl J Adv Res 2 (8), 13-24.
• Newcomb JM, Sakurai A, Lillvis JL, Gunaratne CA, and Katz PS. (2012). Homology and homoplasy of swimming behaviors and neural circuits in the Nudipleura (Mollusca, Gastropoda, Opisthobranchia). Proc Natl Acad Sci 109:10669–10676.
• Noboa V and Gillette R. (2013) Selective prey avoidance learning in the predatory sea-slug Pleurobranchaea californica. J Exp Biol 216:3231-3236. doi: 10.1242.
• Palovcik RR, Basberg BA and Ram JL (1982) Behavioral state changes induced in Pleurobranchaea and Aplysia by serotonin. Behav. Neural Biol. 35:383-394.
• Popescu IR and Frost WN. (2002). Highly dissimilar behaviors mediated by a multifunctional network in the marine mollusk Tritonia diomedea. J Neurosci 22:1985–1993.
• Ram JL, Salpeter SR and Davis WJ. (1977) Pleurobranchaea egg-laying hormone: localization and partial purification. J Comp Physiol 119:173 194.
• Schwarz N and Clore GL (1983) Mood, misattribution, and judgments of well-being: Informative and directive functions of affective states. J Pers Soc Psychol 45:513-523.
• Stuart JN, Ebaugh JD, Copes AL, Hatcher NG, Gillette R and Sweedler JV (2004) Systemic serotonin sulfate in opisthobranch mollusks. J. Neurochem. 90:734-742.
• Susswein AJ, Kupfermann I. (1975) Bulk as a stimulus for satiation in Aplysia. Behav Biol 13:203–9.
• Thompson TE & Colman JG (1984) Histology of acid glands in Pleurobranchomorpha. J Moll Stud 50, 66-67.
• Thompson TE and Slinn DJ (1959) On the biology of the opisthobranch Pleurobranchus membranaceus. J Mar Biol Ass UK 38:507 524.
• Tosches, MA and Arendt D (2013) The bilaterian forebrain: an evolutionary chimaera. Curr Opin Neurobiol 23:1080-1089.
• Wägele H, Klussmann-Kolb A, Verbeek E, Schrödl M (2014) Flashback and foreshadowing—a review of the taxon Opisthobranchia. Org Divers Evol 14:133–149.
• Wägele H and Willan RC (2000) Phylogeny of the Nudibranchia. Zool J Linn Soc 130:83–181.
• Yafremava LS Anthony CW, Lane L, Campbell JK and Gillette R (2007) Orienting and avoidance turning are precisely computed by the predatory sea-slug Pleurobranchaea californica McFarland. J Exp Biol 210:561-569.
• Yafremava LS and Gillette R (2011) Putative lateral inhibition in sensory processing for directional turns. J Neurophysiol 105:2885-2890.
• Yamagishi M, Ito E, Matsuo R (2011) DNA Endoreplication in the Brain Neurons during Body Growth of an Adult Slug. J Neurosci 31:5596-5604.
• Zimmer RK, Schar DW, Ferrer RP, Krug PJ, Kats, LB and Michel, WC (2006) The scent of danger: Tetrodotoxin (TTX) as an olfactory cue of predation risk. Ecol Monogr 76:585–600.

Sponsored by: Dr. Paul S. Katz, Georgia State University, Atlanta, GA, USA

Reviewed by: Paul R. Benjamin, University of Sussex

Reviewed by: James Newcomb, New England College

Accepted on: 2014-11-14 05:07:40 GMT