Electrophysiology and Behavior of Cnidarian Nervous Systems
Summary and Keywords
Although the Cnidaria have evolved a wide range of body forms matched with an equally varied neural anatomy, individual species exhibit common patterns of behavior. For example, in all species a key challenge for the nervous system is to transfer food from the peripherally mounted tentacles to the centrally located stomach. Foraging movements, necessary to maintain the food supply, must be accomplished in such a way as to avoid interference with the primary objective of getting prey into the mouth. Furthermore, the hunt for prey must be balanced by a measured response to “threat.” Different species respond to threat in markedly different ways, but in each case foraging is inhibited, just as it is during transmission of food.
One hundred years ago, G. H. Parker questioned whether a centralized or a locally organized nervous system could best account for sea anemone behavior. Anatomical and electrophysiological studies now suggest that in most Cnidaria there is a degree of hierarchical control, with local reflexes coordinated by more condensed systems of neurons. This organization is highly developed in the nerve rings of hydrozoan medusae and takes the form of ganglion-like rhopalia in the Cubozoa. Even in hydrozoan polyps such as Hydra there are at least four separate neuronal systems. It is likely that the underlying mechanisms (containing both homologous and analogous elements) will be best revealed by a comparative approach that directly relates behavior with its molecular basis. Useful examples include comparisons between sea anemones with and without through-conducting systems; between hydra with and without oral rings; between medusae with and without coordinated escape swimming. Recent advances in transgenomic labeling have shown the way forward.
In a letter from H. M. S. Rattlesnake in October 1849, Thomas Huxley remarked on his astonishment at what seemed to be the widespread distribution of the hydroid polyp Tubularia “until I discovered one day that it was attached in large masses to the ship’s bottom!” (Huxley, 1850, p. 66). During the course of an extended cruise along the east coast of Australia, Huxley came to recognize the underlying unity of structure between sessile polyps, such as Tubularia, and the wide variety of motile medusae that he found in Australian waters. In a report to the Royal Society in London, he commented: “Perhaps no class of animals has been so much investigated with so little satisfactory and comprehensive result as the family of the Medusae” (Huxley, 1849, p. 413) and added:
. . . much has been said of the difficulties attending the examination of the Medusae. I confess I think that they have been greatly exaggerated; at least, with a good microscope and a good light (with the ship tolerably steady), I never failed in procuring all the information I required. The great matter is to obtain a good successive supply of specimens, as the more delicate oceanic species are usually unfit for examination within a few hours after they are taken.
(Huxley, 1849, p. 413)
Nevertheless, the difficulties of working onboard ship may account for Huxley failing to observe “any indubitable trace of a nervous system in the Medusae” (1849, p. 424)—and explain why it was not until 1865 that his friend Ernst Haeckel persuaded him otherwise.
Huxley’s reasons for adding the Hydroids and Medusae to the group of animals he called the “Nematophora” included the universal presence of “peculiar organs called thread-cells” (1849, p. 425), the cnidocytes or nematocytes, and the fact that their stomach was enclosed by two layers, the endoderm and ectoderm. Now known as the Cnidaria, the phylum is considered to include:
1. Anthozoa (Actinozoa)—polyps such as sea anemones and corals
2. Hydrozoa—some of which change from being an asexually reproducing polyp to a sexually reproducing medusa and may be either solitary or colonial in form
3. Staurozoa—stalked jellyfish, with a form that combines both polypoid and medusoid characteristics
4. Scyphozoa—medusoid in form and called “true” jellyfish
5. Cubozoa—box jellyfish, medusoid in form and capable of hunting and courtship
6. Myxozoa—obligate parasites, but containing key nematocyte proteins
Although they are “for all intents and purposes, little more than guts with tentacles” (Kass-Simon & Scappaticci, 2002, p. 1773), the range of body forms exhibited by the Cnidaria ensures that there is no such thing as a typical nervous system. This variability, although conducive to comparative studies, inevitably complicates the process of review. The approach used here is to examine the behavior Cnidaria have in common and reduce the problem to elements that Huxley would recognize. Thus, a key challenge for the nervous system is the transfer of food from the peripherally mounted, nematocyte-laden tentacles to the centrally located stomach. Figure 1 shows how this is accomplished in a variety of different species and demonstrates their overall similarity. To separate the homologous and analogous elements of the underlying neural mechanisms, it is necessary to examine the different forces that have constrained the design of the system, be they environmental, anatomical, cellular, molecular, or developmental.
One important restriction that applies to animals in an aqueous environment relates to the water flows associated with movement. Motile Cnidaria, the “medusae,” can forage widely so long as their swimming movements do not interfere with the primary objective of getting prey into the mouth. A similar constraint applies in static forms, where the whole body undergoes elaborate contortions while apparently fishing for food. In either case, the nervous system is organized in such a way as to ensure that these searching movements are inhibited during the food transmission process (Mackie, Meech, & Spencer, 2012).
Another key challenge for the nervous system is how to balance the need to hunt for prey with a measured response to threat. Many hydrozoan medusae respond to a noxious or damaging stimulus by ceasing to swim and withdrawing their delicate tissues, their tentacles and mouth, into the contracted bell. This “crumpling” response, which allows them to sink in the water column, presumably away from danger (Hyman, 1940), is brought about by the contraction of radially aligned bands of longitudinal muscle arranged around the bell of the animal. In species with flattened (oblate) swimming bells, such as Aequorea victoria, where the mouth is far away from the periphery, these same muscles are used to bring trapped food inward during feeding (see Figure 1C; Satterlie, 1985a). In Aglantha digitale, which has a more elongated (prolate) swimming bell, the radial muscle bands are confined to the mouth and used with great precision to reach trapped prey (see Figure 1A; Mackie, Marx, & Meech, 2003). The absence of radial muscle in the bell means that Aglantha does not crumple when attacked, but instead has evolved a number of adaptations that permit it to “escape-swim.” Crumpling is also absent in Nanomia, a colonial hydrozoan, and here escape consists of a coordinated contraction of its multiple swimming bells (Mackie, 1984).
In sea anemones, the response to threat is also species specific. Pantin could find no protective reaction at all in Anemonia viridis (Pantin, 1935b). In fact, a strong mechanical stimulus that produces withdrawal in Calliactis or Metridium evokes a feeding response in Anemonia. In Metridium, the oral disc is withdrawn by the contraction of the mesenteric retractor muscle, and it is only after this has been accomplished that the marginal sphincter contracts and covers the disc (Pantin 1935b). In Calliactis, on the other hand, the “thick mesogloea of the column is too inflexible to allow the disc to be rapidly pulled in” (Pantin, 1952, p. 150), and it is the sphincter that closes off the column and, once the tentacles have withdrawn, provides protection (Hall & Pantin, 1937; Pantin, 1935a). See Figure 2C for the disposition of the muscles involved.
The fact that feeding behavior may be favored over foraging, and defense favored over feeding, might indicate the existence of some centralized integrating region, and the complex behaviors, such as obstacle avoidance, courtship and mating, seen in some Cubomedusae (see Figure 1B; Garm, Lebouvier, & Tolunay, 2015; Lewis & Long, 2005) might lend support to this model. However such behavior might also arise from a hierarchy of semi-independent nerve circuits, or modules, each one being responsible for a separate subroutine. One purpose of this article is to show how these circuits are linked and how in more complex forms, in species such as Aglantha, individual modules have evolved the means to share components. In other complex forms, such as the Cubomedusae, the evolution of shared control provides a flexibility not always possible in a centrally driven system.
The structural basis of this complex integrated action is examined in “General Structural Features.” The article then divides into two sections; “Polypoid Nervous Systems” summarizes what is known of polypoid electrophysiology and how it relates to their behavior, with due prominence being given to recent discoveries of the role of peptides. There follows a section on the medusae, “The Role of the Nervous System in Medusoid Cnidaria,” which emphasizes the role of gap junctions and ion channels in generating complex behavior in the Hydrozoa. There are also short sections on the nervous systems of the Scyphozoa and the Cubozoa, but space constraints have forced the omission of some important topics, such as coral behavior and some aspects of sea anemone behavior such as swimming, climbing, and walking.
General Structural Features
The cnidarian body wall consists of two layers of cells, the ectoderm and endoderm, separated and supported by an acellular mesogloea. For many years, experimentalists depended on species that tolerate laboratory life, such as Hydra or the sea anemone Calliactis, both of which appeared to have nervous systems that were largely net based. As a result, the cnidarian nervous system became synonymous with the nerve net concept in spite of condensed neuronal structures being widely distributed in both polyps and medusae. In the hydrozoan medusae in particular, the bell margin may contain giant neurons, giant axons, nerve ganglia, and well-developed nerve tracts; even in polyps, some nerve cells are clustered into nerve rings, and some are condensed into what are described as “through-conducting systems.” The nerve net in Hydra is now known to consist of at least four separate, non-overlapping nerve networks, each one linked to a specific movement (Dupre & Yuste, 2017).
In hydrozoan medusae, the swimming bell consists of subumbrella and exumbrella layers innervated by two rings of neurons located in the ectoderm on either side of a projection of the mesogoea that supports the velum (see Figure 3). Pacemaker cells within the inner nerve ring maintain the swimming rhythm. In some species, such as Polyorchis, the pacemaker cells directly drive the circular muscles of the subumbrella, but in others, such as Aglantha, motor nerves intervene (see Figure 7C). The endoderm, which lines the stomach and radial canals, also contains circular muscles, which are presumed to regulate the distribution of the products of digestion (Singla, 1978). Also in the endoderm is a plexus of nerve cells (Jha & Mackie, 1967), which is active during feeding and provides an inhibitory synaptic input to the swimming pacemaker neurons in the ectoderm (Mackie et al., 2012).
In polypoid forms, as with the medusae, neurons are present in both ectoderm and endoderm, with the endoderm generally having fewer nerve elements than the ectoderm. In Hydra, where they can be separated by maceration in a mixture of acetic acid and glycerine (David, 1973), the ratio of nerve cells to epithelial cells in the ectoderm is 2–7 times that in the endoderm (Epp & Tardent, 1978). Generally, nerve cells may be multipolar, bipolar, or unipolar; bipolar cells are generally symmetrical in the ectoderm but may be asymmetrical in the endoderm. That said, neurons are continually being formed from interstitial cells, and forms may be present that reflect degeneration or differentiation (Epp & Tardent, 1978).
In Hydra, endodermal cells innervate the circular muscles that control column thickness, and the more elaborate two-dimensional nerve net in the ectoderm innervates the longitudinal muscles that determine column length. In the mouth region a single hypostome-tentacle junction contains as many as 194 nerve cells in a 80 μm wide space (Kinnamon & Westfall, 1982). A three-dimensional reconstruction (see Figure 5C) suggests that neurons at apical positions transduce the sensory information required for feeding whereas neuronal clusters at the base of the tentacles are responsible for tentacle coordination and the generation of rhythmic activity (Kinnamon & Westfall, 1981; see Spontaneous Activity in Hydra). The more centrally located neurons are described as ganglion cells (see Figure 5E).
Polypoid Nervous System
Electrophysiology of Anthozoan Nervous System
“In studying sea anemones, one soon becomes aware of the paradox between their functional and behavioural complexity on the one hand, and the apparently simple structure of their nervous systems on the other.” (Lawn, 1976b, p. 311). Earlier attempts to come to terms with the Anthozoan nervous system led to a somewhat different, yet no less conflicted, view as summarized by G. H. Parker in 1917:
In examining the literature . . . two tendencies are quite obvious. One emphasises the diffuse non-centralized nature . . . and deals with the behavior of these animals in terms of relatively simple reflexes and the like; the other . . . interprets their behavior from the standpoint of the whole organism. (pp. 223–224)
Parker went on to say:
The facts that the pedal half . . . may creep normally without the oral half and that the oral half is responsible for the feeding activities . . ., make it quite obvious that the organic unity of the animal as a whole is very weak. (pp. 225–226)
To Parker, clear-cut behavioral observations of this kind made it unlikely that sea anemones had anything that could be described as a central nervous system, a conclusion that was in stark contrast to the findings of O. and R. Hertwig (1879) as noted by Parker and Titus (1916b). According to the Hertwigs the epithelia contained many branching sense cells that gave rise to a nervous meshwork which, supplemented by fibers arising from large deep-seated ganglion cells, exerted control over the muscles. This nervous structure was said to be best developed in the ectoderm of the oral disc and to form there a primitive central nervous organ from which connections pass to the tentacles, oesophagus, etc. As Parker and Titus (1916a, p. 3) pointed out, however, a host of other authors (Bethe, 1903; Havet, 1901; Jordan, 1908, 1912; Loeb, 1895; Nagel, 1892), “have declared the nervous system to be diffuse and not centralized at all.” In essence, the argument was less about central versus peripheral control and more about how much of a central organizer the Hertwigs’ collection of neurons really was. After all, there was little agreement among the centralizers as to the site of the main concentration of neurons. “Grošelj . . . maintains that the ectoderm of the oesophagus, not that of the oral disc, is the region of chief concentration” (see Parker & Titus, 1916b, p. 443).
The disagreement between those who focused on the non-centralized nature of sea anemone behavior and those anatomists who pointed to areas of condensed neuronal tissue might have been resolved had it been realized that sea anemones possess multiple nerve networks. In such a case, diffusely distributed networks could mediate the local reflexes highlighted by Parker and Titus (1916a), whereas a more condensed system might provide a means of distributing commands more widely. Such a system has been identified as the “through-conduction system,” whose signals spread in all directions from the point of stimulation and travel at high speed (rates of up to 120 cm/s). In a sea anemone like Anemonia sulcata, where through-conduction is absent, electrical stimuli give only local responses wherever they are applied (Pantin, 1935b).
It is the fast “through-conduction system” that is responsible for the rapid defensive withdrawal of Calliactis parasitica, the subject of classical studies by Carl Pantin (1935a, 1935b, 1935c). It consists of a lattice of large bipolar cells, which are especially obvious in the mesenteries, and connect with a fast-conducting neuronal ring near the sphincter. Pantin realized that although conduction was fast the impulse frequency in this system was “100 to 1000 times slower than that which characterises the nervous organisation of the Vertebrates” (1935b, p. 154). This meant that in the circular muscle of the column, for example, full contraction could eventually be achieved even with only one stimulus every 6 to 10 seconds. By taking care to use physiological stimulus rates of this kind, Pantin showed that a single impulse, although not necessarily causing contraction by itself, would facilitate transmission at neuromuscular and neuron/neuron synapses (see Figures 4C and 6B).
It turned out that individual muscles produce facilitated responses over a characteristic range of stimulus frequencies, and so the nature of the contraction depends on the time course of facilitation compared to the time course of muscle shortening. In the sphincter muscle at the top of the column in Calliactis (see Figure 2A–C), the two processes develop at about the same rate, so that shortening has a stepped form; in the more slowly contracting mesenteric muscle, facilitation is damped out, and shortening develops more smoothly. A stimulus applied once every 3 seconds produces column shortening (caused by contraction of the parietal muscles); at faster rates, there is withdrawal of the oral disc (caused by shortening of the mesenteric retractor muscles); at even higher rates, the sphincter muscles dominate the response.
Although through-conduction is absent in Anemonia sulcata, in many species slow and fast systems exist side-by-side. Although there is known to be at least two slow systems: a slow ectodermal system (SS1) and a slow endodermal system (SS2), their nature remains somewhat unclear. At one time the more labile ectodermal pathway was thought to be non-nervous (McFarlane, 1969)—but more recent immunohistochemical studies (see McFarlane, Graff, & Grimmelikhuijzen, 1989) make this seem less likely. Instead, “earlier failures to detect neurons in places where physiological evidence shows a conducting system exists were due simply to the limitations of previous staining techniques” (McFarlane, Graff, & Grimmelikhuijzen, 1989, p. 114). The endoderm may also contain a local conducting system of multipolar nerve cells (Josephson, 1966), which is responsible for delayed activity in the ectoderm (Jackson & McFarlane, 1976; see Figure 4C).
Electrical impulses originating from both slow and fast systems may be elicited by electrical stimuli all over the body surface, but they are recorded most readily from the tentacles (McFarlane, 1982; see Figure 4A). The low threshold of the through-conducting system makes it difficult to stimulate the slow systems alone, although McFarlane (1970) has had some success with Tealia feline. By cutting a flap in its column and stimulating the mesogloea below (see Figure 4D), he was able to excite the slowly conducting system in the ectoderm and show that it promoted oral disc expansion by relaxing its radial muscles. The endodermal system can be stimulated alone using prolonged low-intensity stimuli (McFarlane, 1974b), and it is activity in this system that is associated with mouth opening and pharynx protrusion.
Spontaneous movement is a characteristic feature of cnidarian systems, and in sea anemones the pacemakers that drive these bursts of activity appear to be widely distributed in the through-conducting system (McFarlane, 1974a). The pacemakers involved are inhibited by activity in the endodermal slow system (McFarlane, 1974a) and excited by the ectodermal slow system (McFarlane, 1976). Activity in the endodermal slow system also inhibits spontaneous contractions in the endodermal muscles of the body wall so that the animal exhibits a marked loss of muscle tone with maintained stimulation (McFarlane, 1974b). The ectodermal slow system excites endodermal circular muscle contractions (McFarlane, 1976) and inhibits spontaneous contractions of ectodermal muscles (McFarlane & Lawn, 1972). These interactions are summarized in Figure 4C, and the anatomical relationship of some of the muscles involved is shown in Figures 2 and 4B.
Josephson (1966) found that in Calliactis polypus, in contrast to C. parasitica, a single supra-threshold shock to the oral disk always activates the whole disk as if its neuromuscular junctions are permanently facilitated. Stronger electric shocks evoke a local response, however, as if there were two conducting systems in the same tissue, a facilitating through-conducting system and a non-facilitating local system. In Tealia, impulses in the slow system do not facilitate the muscle responses to the through-conducting system as if the two systems innervate separate sets of muscles, a suggestion supported by the presence of two classes of fiber (phasic and tonic) with different myofilament patterns and different responses to fixing and staining (Amerongen and Peteya, 1976; 1980).
Subpopulations of Nerve Cells
A range of different techniques have been used to characterize subpopulations of nerve cells. These include use of the following:
1. Silver staining (Holmes, 1947).
3. Intracellular recording from larger units such as rhythmically firing “pacemaker” cells, photosensitive cells, cnidocytes, and motor axons (Anderson & Mackie, 1977; Anderson & McKay, 1987; Roberts & Mackie, 1980; Spencer & Satterlie, 1980).
6. A transgenic Hydra line that expresses calcium indicator (GCaMP6s) in its neurons, revealing calcium-based action potentials in separate nerve systems (Dupre & Yuste, 2017).
7. A fluorescent reporter in a transgenic Nematostella line to show the consistency of neuronal locations, numbers, and neurite projection patterns and indicating hidden levels of organization (Havrilak, Faltine-Gonzalez, Wen, Fodera, Simpson, Magie, & Layden, 2017).
8. Application of inhibitory transmitter agents (Lauro & Kass-Simon, 2018).
In spite of these widely different approaches, much of the detailed circuitry remains uncertain. For example, it is clear from physiological experiments that signals must pass between the ectodermal and endodermal nerve systems, (Lawn, 1976a) but to do this they would have to cross the mesogloea and according to Batham, Pantin, and Robson (1961) “the nervous system in the mesenteries and column of Metridium follows the epithelial layers and does not penetrate the mesogloea” (p. 143). Batham (1965) subsequently observed silver stained neurons “lying partly in the endoderm, partly in the ectoderm” p. 400), but nerves may cross the mesogloea to connect the two surfaces of the ectoderm—the external ectoderm and the internal, invaginated ectoderm—without making any connection with the endodermal system (Mackie, 1960). It remains to be seen whether electron microscopical studies will reveal this mysterious pathway.
Could the endoderm and ectoderm be connected without the participation of nerves? In the Hydrozoa, epithelial conduction plays an important role as an adjunct to the nervous system (Mackie, 1965; Mackie & Passano, 1968; Mackie, Passano, & Pavans de Ceccatty, 1967) with gap junctions between cells providing the electrical pathway. It is therefore possible that communication between endoderm and ectoderm occurs via electrically coupled epithelial bridges. Although gap junctions appear to be absent from other Cnidarians, other forms of electrical junction may be present. In the anthozoan Rennila koellikeri, immunostaining reveals a connexin-43–like protein in endodermal cells and threadlike processes within the mesoglea (Germain & Anctil, 1996). There are also “tiny zones of close membrane apposition . . . with gaps of 2–4 nm” (Germain & Anctil, 1996, p. 353), which might provide for electrical continuity. Connexin-43, a gap junction component, appears to be present in the tentacles of the sea anemone Haliplanella luciae at junctions between different cell types (see Figure 5A, B; Mire, Nasse, & Venable-Thibodeaux, 2000), whereas “areas of close membrane apposition” (with clefts of 4–7 nm) are found between supporting cells and between nerve net neurons. Dye injection suggests these junctions provide for signal transmission between vibration detectors (hair cells made up of supporting cells and sensory neurons) and effector cells (cnidocytes). The marked electrical coupling between Nematostella blastomeres (Popova, Vornov, Kosevich, & Panchin, 2012) and the presence of an innexin homologue in the genome (Chapman et al., 2010) provides further circumstantial evidence.
Molecular Basis of Neuromuscular Transmission
Peptides Mediate Neuromuscular Transmission in Hydra
The use of neuropeptide antisera to localize specific subsets of neurons (Grimmelikhuijzen, 1985) might suggest that the Cnidaria utilize peptides such as RFamide (with an Arg-Phe-NH2 motif at its C-terminal) as neurotransmitters, but it always seemed possible that the antisera might simply bind to structural elements. Reviews by Anctil (2017) and Grimmelikhuijzen, Williamson, and Hansen (2004) give a full account of the evidence and provide references to the literature. Anctil (2017) writes that “the phylum Cnidaria was a laboratory for the early evolution of neuropeptides” (p. 290) and he points to the “presence in cnidarian neurons of several peptide families . . . [that] probably represent evolutionary dead ends in the sense that they vanished in the bilaterian lineage” (p. 290).
The evidence for peptide-gated ion channels in Hydra has been summarised by Gründer and Assmann (2015). These channels, expressed in myoepithelial cells at the base of the tentacles and found in the vicinity of RF-amide–containing neurons (Koizumi, Wilson, Grimmelikhuijzen, & Westfall, 1989; Westfall, 1996), are part of the degenerin/epithelial Na+ channel family (Dürrnagel et al., 2010; Golubovic et al., 2007). Members of this family have a variety of functions that range from mechanoreception in Caenorhabditis elegans to Na+ reabsorption in mammalian kidney distal tubules (see Kellenberger & Schild, 2002, for a review). In spite of its name, the Hydra Na+ channel (HyNaC) is, broadly speaking, non-selective, although it is highly Ca2+ permeable (as expressed in Xenopus oocytes; Dürrnagel, Falkenburger, & Gründer, 2012).
A total of 12 different HyNaC subunits have been cloned (HyNaC1–HyNaC12; Assmann, Kuhn, Dürrnagel, Holstein, & Gründer, 2014). Oocytes expressing either HyNaC2 or HyNaC3 generate large inward currents in response to Hydra-RFamides I and II (pyro-Glu-Trp-Leu-Gly-Gly-Arg-Phe-NH2 and pyro-Glu-Trp-Phe-Asn-Gly-Arg-Phe-NH2). Co-expression with HyNaC5 strongly increases both current amplitude and peptide affinity, and in the case of Hydra-RFamide II, the affinity increases 100 times (Dürrnagel et al., 2010; Golubovic et al., 2007). Co-expression of HyNaC2, HyNaC9, and HyNaC7 gave the most sensitive channel, being activated by nanomolar concentrations of peptide.
The HyNaC subunits that co-assemble to form peptide-gated ion channels in myoepithelial cells at the base of the tentacles appear to be involved with feeding. The diuretic amiloride, which is a relatively low-affinity HyNaC blocker, delays the feeding response to glutathione (GSH, in its reduced form) (Dürrnagel et al., 2010), which is present in the body fluids released from trapped prey. Diminazene, which blocks HyNaC currents at micromolar levels, is also effective (Assmann et al., 2014).
A role for neuropeptides in neuromuscular transmission is supported by the finding that Hym-370, a member of the GLWamide peptide family (Lys-Pro-Asn-Ala-Tyr-Lys-Gly-Lys-Leu-Pro-Ile-Gly-Leu-Trp-NH2), isolated from Hydra magnipapillata, evoked concentration-dependent contractions in both H. magnipapillata and the retractor muscle of a sea anemone, Anthopleura fuscoviridis (Takahashi et al., 2003). A slightly shorter GLWamide (Hym-248) promoted lengthening of H. magnipapillata by inducing contraction in its endodermal muscles.
Spontaneous Activity in Hydra
Many species of Cnidaria are active even in the apparent absence of external stimuli, and this is true for polypoid forms as much as for the medusae (see Josephson, 1974). In Hydra, electrophysiological studies using external electrodes (Passano & McCullough, 1963), supplemented by analysis of calcium transients monitored in a transgenic line expressing a calcium indicator (Dupre & Yuste, 2017), have led to the identification of four neuronal systems, each of which exhibits spontaneous activity.
1. The ectodermal contraction burst (CB) system, which is associated with transient shortening of the body column. Its pacemakers are located within the circumferential nerve ring at the base of the tentacles (Kass-Simon, 1972; Kinnamon & Westfall, 1981, 1982; Koizumi et al., 1992), and its somata are larger (10.73 ± 0.62 μm [SEM]; Dupre & Yoste, 2017) than those of the other neuronal categories.
2. The tentacle pulse (TP) system, which is associated with tentacle contraction. Its pacemakers are located in individual tentacles (Kass-Simon, 1972, 1973; Rushforth & Burke, 1971), and its activity may lead to excitation in the CB system. A component of the TP system (called the subtentacle network [STN] by Dupre and Yoste ) is associated with “nodding,” which is described as “a gentle swaying of the hypostome . . . and its tentacles to one side but without turning the rest of the body” (Dupre & Yoste, 2017, p. 1092).
3. The ectodermal rhythmic pulse system (RP1; see Dupre & Yuste, 2017) is associated with elongation as a result of the relaxation of longitudinal muscles; that is, it acts in opposition to the CB system and is less active when CB frequency rises. It appears to receive an input from photosensors in the tentacles and is responsible for the elongation induced by photic stimuli.
4. The endodermal rhythmic pulse system (RP2; see Dupre & Yuste, 2017) is also associated with elongation through contractions of the endodermal circular muscle (Kass-Simon & Passano, 1978; Passano & McCullough, 1965; Shibley, 1969). Its pacemakers originate near the base of the body column (Passano & McCullough, 1965), but it also appears to receive an input from sensory cells exposed to the gastric environment. When elongated, most of the gastrovascular fluid is stored in the lower 25% of the body column; upon contraction it is pumped into the rest of the cavity (Shimizu & Fujisawa, 2003).
The nature of the connectivity between the individual neurons is unknown, and either electrical or chemical transmission is possible (Dupre & Yuste, 2017). RP1 activity is not directly correlated with elongation; neither is RP2 activity directly correlated with radial contraction, and it seems that the two networks might integrate sensory information prior to triggering a behavioral response. Any contribution provided by epithelial conduction would be invisible because expression of the calcium indicator is limited to neurons.
Nerve-free (colchicine-treated) hydra exhibit no tentacle contraction pulses or rhythmic potentials—although contraction burst pulses may be elicited with electrical stimuli (Campbell, Josephson, Schwab, & Rushforth, 1976). Thus, although spontaneous activity depends on nerve cells, impulses can propagate in non-neuronal cells (albeit at a slower rate than in normal hydra) and produce contraction of the column (Fraser & Bode, 1981).
The CB and RP systems act on antagonistic muscle systems—the CB system causing shortening, the RP system producing lengthening. Upon mechanical agitation, the CB system produces single pulses and defensive withdrawal. This suppresses the burst of CB pulses associated with spontaneous contraction. With repeated mechanical agitation, the system habituates, and the CB bursts are restored. Inhibition of burst activity can be also accomplished by single electrical shocks (Passano & McCullough, 1964), light or GSH. Rushforth (1973) has suggested that GSH directly inhibits the CB system but that light activates the RP system (now known to be the RP1 system; Dupre & Yuste, 2017), which is inhibitory to the CB system. The CB and RP systems are both inhibited by gamma-amino butyric acid (GABA), whereas the output of the TP system is unaffected (Kass-Simon, Pannaccione, & Pierobon, 2003).
Electrophysiology of Sea Anemones
Polypoid cnidarians provide a considerable technical challenge to electrophysiologists on all fronts. Intracellular recording is difficult because the nerve elements are small and embedded in spontaneously contracting tissue. The nerves are either too deep within the body tissue or too diffusely distributed to make extracellular recording manageable except at the tentacles. For patch clamping, the problem is that pipette seals are hard to achieve because of the presence of large amounts of mucous. The amount of mucous can be minimized by treatment with protease, but the problem remains (Cho & McFarlane, 1995). One approach is to isolate the cells by digesting tissue with activated papain (Holman & Anderson, 1991) and then use a patch pipette or single-electrode voltage clamp to examine their electrical properties.
When isolated in this way, myoepithelial cells from the mesenteries of Calliactis tricolor (with their fast-contracting retractor muscles) repeatedly fire overshooting action potentials when depolarized (see Figure 6C), and under voltage clamp there is at least one, and possibly two, inward calcium currents as well as four different outward currents (Holman & Anderson, 1991). When the more slowly contracting endodermal myoepithelial cells from C. parasitica were examined, the currents recorded were qualitatively similar, but the calcium-based inward currents were elicited at rather more negative potentials. The tissue had two distinct types of muscle cell (Type 1 and Type 2) that differed in the time to peak of their inward current (Cho & McFarlane, 1996b). The cells had a similar resting level to C. tricolor (about –60 mV) but did not fire repetitive action potentials upon depolarization (Cho & McFarlane, 1996c). However, there were marked differences in the recording techniques used; Cho and McFarlane were able to record from in situ rather than isolated cells, for example. Other differences included the composition of the pipette-filling solution and whether the tissue was pretreated with high levels of magnesium ions.
The neuropeptides Antho-RWamide I (<Glu-Ser-Leu-Arg-Trp-NH2) and II (<Glu-Gly-Leu-Arg-Trp-NH2), isolated from sea anemones by Grimmelikhuijzen et al. (1992), induce slow contractions in endodermal muscle cells from C. parasitica without acting on any of the neuronal systems (McFarlane, Anderson, & Grimmelikhuijzen, 1991). Neurons associated with both gastrodermal and oral sphincter muscle cells contain vesicles with an Antho-RWamide–like material (Westfall, Sayyar, Elliott, & Grimmelikhuijzen, 1995) and furthermore, Antho-RWamide I produces a dose-dependent, reversible enhancement of the inward calcium currents recorded from endodermal muscle cells (Cho & McFarlane, 1996a). Even more significant may be the finding that another neuropeptide, Antho-RNamide (L-3-phenyllactyl-Leu-Arg-Asn-NH2), differs in its actions on the two types of myoepithelial cells mentioned above (Cho & McFarlane, 1996b); it enhances inward currents in Type 1 cells and outward currents in Type 2 cells. Both cell types are subject to a barrage of what may be excitatory synaptic potentials, but the amplitude of these small events (about 5 mV) hardly seems sufficient to depolarize the muscle membrane into the range at which the calcium channels activate.
A question that naturally arises is whether the electrophysiology provides an explanation for the facilitation of contraction described by Pantin (1935b). Figure 6 summarizes some of the difficulties associated with answering this question. Robson and Josephson (1969) used external recording techniques to relate changes in the electrical events in the Metridium nerve net and muscle epithelium with the mechanical responses of the muscle. They showed (Figure 6A) that a single shock produced a complex electrical event they identified as a nerve net potential; a second shock 1 second later produced an identical nerve net potential plus a delayed response that they interpreted as a muscle potential. During a series of six shocks, 1 second apart, the muscle potential and the associated mechanical contraction both increased in amplitude (Figure 6B). The increase in muscle potential is probably due to an increase in action potential duration because the spike amplitude decreases rather than increases in size during repetitive firing (Holman & Anderson, 1991; Figure 6C). However, when electrical and mechanical responses are compared at different stimulus intervals, there is little correlation between them (Figure 6D), for even if the muscle potential increases in amplitude during the first three shocks, it may then decrease while the mechanical response continues to increase.
To explain this long-standing puzzle, it may be necessary to separate sea anemone facilitation into two components. One, which may account for the observation that the muscle epithelium is unresponsive to single shocks, is probably associated with the rapidly activating (and inactivating) outward current revealed by voltage-clamp experiments (Holman & Anderson, 1991). This current tends to prevent spiking (see Byrne, 1980) and is probably responsible for the small delay before the first action potential in the series (Figure 6C). If the outward current were to be inactivated by the first shock in the series, subsequent shocks would be able to elicit a muscle spike. The other aspect of facilitation, the increasing amplitude of the mechanical responses, is most likely to reflect summation of the sarcoplasmic Ca2+ increase, together with a sigmoidal relationship between the level of Ca2+ and the response of the contractile proteins.
The Role of the Nervous System in Polypoid Feeding Behavior
Cnidocytes and the Feeding Response
Cnidocytes are an essential part of the feeding process, discharging in response to the combined mechanical and chemical stimulus that the prey provides (Pantin, 1942). In so doing, they not only secure and paralyze it but also initiate the full feeding response. Once the cnidocil, the cnidocyte sensory apparatus, has been deflected and the cnidocyst discharged (Kass-Simon & Scappaticci, 2002), the tentacles move toward the vibration source. Agents such as reduced GSH or phosphatidylcholine that may be released from wounded prey (Grosvenor, Bellis, Kass-Simon, & Rhoads, 1992; Loomis, 1955; Venturini, 1987), lower the threshold for mechanical stimulation (Ewer, 1947).
Although T. H. Huxley recognized that cnidocytes were a characteristic feature of what we now call the Cnidaria, he can hardly have appreciated their structural variety; Mariscal (1974) has identified 25 different types of cnidocyst. Cnidocyte function separates into four categories: “those that pierce, ensnare, or adhere to prey, and those that adhere to the substrate” (Kass-Simon & Scappaticci, 2002, p. 1772). In Hydra, they are responsible for locomotion as well as defense and prey capture (Ewer, 1947). Individual cnidocytes discharge their cnidocysts once only, in an all-or-nothing Ca2+-dependent process. In the colonial hydrozoan Physalia, there is a fully functional Ca2+ channel β subunit at the apical (sensory) end (Bouchard & Anderson, 2014; Bouchard et al., 2006), and although voltage-gated Ca2+ currents are missing in cnidocytes under whole-cell recording conditions, this may be because of Ca2+ channel “washout”—often present with this recording configuration (Anderson & McKay, 1987). The discharge event is controlled by an elaborate neuronal network (Hadzi, 1909) that connects with a variety of sensory cells and is coordinated by areas of densely populated ganglion cells at the top and bottom of the body column (see Figure 5C; Bode et al., 1973; Grimmelikhuijzen, 1985).
According to Oliver, Brinkmann, Sieger, and Thurm (2008), cnidocytes receive multiple inputs arising from the mechanochemical stimulation of their neighbors, tentacle deformation, and contractions of the tentacular shaft as well as movements of the prey pulling at discharged cnidocysts. In the athecate hydroids Coryne, Dipurena and Stauridiosarsia, intracellular recordings from piercing cnidocytes (often called stenotele cells) reveal two kinds of postsynaptic electrical event (Oliver et al., 2008; Thurm et al., 2004). The longer-lasting L‑potentials were elicited by deflecting the cnidocil apparatus of a neighbor, whereas the more transient T—potentials were associated with spontaneous contractions of myoepithelial cells in the tentacular shaft some distance away. Stimulated cells rarely discharged their cnidocysts with purely mechanical stimuli, however, unless provided by phosphatidylcholine-coated probes. All cells that did discharge their cnidocysts induced L—potentials in their neighbors, with a delay short enough to indicate a single synapse pathway.
Depolarized stenotele cells generate Na+—dependent action potentials with a threshold somewhat more positive than that achieved by individual L—potentials. Similar impulses occur in cnidocytes of Cladonema and Physalia (Hydrozoa) and Chrysaora (Scyphozoa) (Anderson & McKay, 1987), and they generally cause cnidocyst discharge (Brinkmann, Oliver, & Thurm, 1996). Occasionally, the mechanoreceptor potential drives the cnidocyte membrane directly to threshold, but generally it is the first step in a two- or three-stage cascade (Brinkmann et al., 1996). The summating L—potentials, which arise in response to discharge events in neighboring cnidocytes, also contribute and facilitate the spread of excitation. Distant discharge events may, after a slight delay, set off a series of Ca2+-dependent T—potentials (Oliver et al., 2008; Thurm et al., 2004). Similar events, preceded by what appears to be a synaptic depolarization, have also been recorded in Cladonema cnidocytes after they were exposed to Artemia extracts (Price & Anderson, 2006).
Tentacle, Mouth and Column Responses During Feeding
Descriptions of feeding in different polypoid species indicate that random movements seen prior to feeding turn into coordinated contractions once the prey has been seized. In the snakelocks anemone Anemonia sulcata,-
Immediately after contact, the tentacles clasp the food and bend strongly towards the mouth. The edge of the disc carrying these tentacles contracts, so that they bunch together round the food, and then [it] rises up and turns inwards thereby folding tentacles and food towards the mouth. After a few seconds, the part of the oral disc between the food and the mouth slowly contracts and sinks downwards, so that the mouth turns towards the food. The mouth then protrudes and begins to open, the food is gradually thrust in, and muscular action of the pharynx pulls it into the gastral cavity.
(Pantin & Pantin, 1943, p. 6).
Some of the operational steps that make up feeding can be initiated by electrically stimulating separate pathways within the tentacles and body column. McFarlane (1970) divides the process into pre-feeding and feeding responses. Pre-feeding occurs once food extract has been added to the water, and it has the effect of increasing the “food catching” space around the animal. It involves relaxing the endodermal sphincter muscle at the top of the column so as to expand the oral disc while also lowering its margin by relaxing the radial muscles in the ectoderm. These actions go together with mouth opening and pharynx protrusion and are associated with the relaxation of the circular muscles of the mouth and pharynx (see Figure 1D). The feeding response itself involves coordinated tentacle movements as well as the ingestion of solid food.
Hydra’s feeding response was described by Abraham Trembley as early as 1742:
They are voracious animals: their arms extended into the water, are so many snares which they set for numbers of small insects that are swimming there. As soon as any of them touches one of the arms, it is caught. The Polypus being seized of a prey, conveys it to his mouth, by contracting or bending his arm. If the prey be strong enough to make resistance, he makes use of several arms. (p. v).
After the prey becomes attached to the tentacle, there is a delay, which can last for more than a minute (Josephson, 1965) before its proximal end contracts and brings the prey toward the mouth. As it does so, concerted contractions in nearby tentacles can bring them into contact with the prey, making it more secure and taking it closer to the mouth region. Muscles in this region pull the epithelial cells into a thin lamella until the cell attachments (septate junctions) separate, and a round opening is formed (Campbell, 1987). Once the open mouth spreads around the prey, the Hydra will engulf metallic objects, a process thought to result from the mechanical stimulation of receptors on the inside of the mouth (see Figure 1F).
The concerted movements of the tentacles can continue long after the prey has been engulfed. Even at rest, tentacles give random contractions and sometimes bursts of contractions, often just before a spontaneous contraction of the entire column (Rushforth, 1973).
“Between these periodic [column] contractions the animals usually remain extended, seemingly inert, in what appears to be a ‘fishing’ attitude. Occasionally, however, they may take a locomotor step or they may circle slowly about the attached basal disk (Haug, 1933), sweeping the substratum with their tentacles” (Passano & McCullough, 1964 p. 648).
Passano and McCullough (1964 p. 648) also describe a “contraction burst” that compresses the column into a “tight contracted ball” for 30–60 seconds. The frequency of such endogenously driven activity is altered by changes in ambient light such that it is synchronized to the likely presence of prey; the contraction burst frequency under natural daylight is 1.5-2 times that in the dark. In “Spontaneous Activity in Hydra,” the electrophysiological basis for this behavior was described, which included the action of light on the RP1 system and the inverse relationship between activity in the RP1 and CB systems (see Dupre & Yuste, 2017).
Han, Taralova, Dupre, & Yuste (2018) have devised a system of machine learning to separate Hydra’s spontaneous activity into six recognizable subroutines. The analysis begins with a video record of individual Hydra and provides a statistical summary of the changes in body form, as reflected for example by the angle of an edge at specific points on a tentacle. The subroutines include: elongation, contraction and bending of the body, body swaying and tentacle swaying. Elongation and bending are usually long continuous movements while body contractions and tentacle and body swaying are usually executed in short bursts. More complex behaviors such as feeding and somersaulting, consist of smooth transitions between a series of subroutines. When feeding, Hydra undergoes tentacle writhing, ball formation, and mouth opening while the locomotory movement known as somersaulting, includes sequences of tentacle and body swaying, bending, elongation and contraction. The duration of each subroutine is quite variable.
Surprisingly Han, Taralova, Dupre, and Yuste (2018) found Hydra’s basal repertoire to be highly stable; lighting conditions had no effect on the average time spent performing each subroutine and there was no significant difference between starved and well-fed animals. The authors ask whether the robustness they observe reflects a “passive stability” because of the animal’s unresponsiveness to the environment, or a homeostatic “active stability” that ensures survival under different environmental conditions. On the face of it, this finding appears to be at odds with the observations of Passano and McCullough (1964) who recorded a lower frequency of contraction bursts (as determined by electrical recording) at night compared with daytime. Han, Taralova, Dupre, and Yuste (2018) suggest that this might be because, thus far, they have only measured the total time spent performing a given subroutine and not the frequency of switching between subroutines, which would more closely correspond with the measurements made by Passano and McCullough (1964).
It may be that the contraction burst and associated ball-formation are connected in some way with foraging. The impulses associated with a contraction burst are conducted throughout the column and appear to originate at the tentacle attachment zone, where Kinnamon and Westfall (1981) have identified large ganglionic clusters of neurons. In Hydra oligactis, these clusters are stained by antisera to RFamide (Davis, Burnett, & Haynes, 1968; Grimmelikhuijzen, 1985; Koizumi et al., 1992). In the stalked hydra Pelmatohydra robusta, Matsuno and Kageyama (1984) found a nerve ring made up of neurites and ganglion cells at the apical edge of each tentacle base, whereas in Hydra vulgaris there is a more ganglion-like structure with “numerous sensory cells located in a region around the mouth opening and a dense plexus of processes which project mostly radially towards the bases of the tentacles” (Grimmelikhuijzen, 1985, p. 171).
Koizumi (2007) has examined GSM (s-methyl-glutathione)-induced feeding movements in four species of Hydra and finds a significant difference between the behavior of common and green hydra, on the one hand, and stalked hydra, on the other. All four species exhibit tentacle writhing and mouth opening, but the tentacles in H. vulgaris, H. magnipapillata, and H. viridissima quickly become tightly coiled around the hypostome, whereas in the stalked (brown) hydra, H. oligactis, the tentacles stay extended and wave up and down. As H. oligactis was the only species examined that had an RFamide-positive nerve ring, this structure might provide the basis for such activity.
Pierobon (2015) has used the GSH response of H. vulgaris (by measuring the time interval between mouth opening and mouth closing) as a quantitative assay to compare different neurotransmitter candidates (Concas et al., 1998; Pierobon et al., 1995, 2001, 2004). She finds that GABA increases the response duration in whole animals, reduces it in isolated heads, and has no effect in animals in which the peduncle and foot have been amputated. These observations suggest that GABA can inhibit mouth closing through a circuit located in the gastric region (i.e., peduncle and/or foot) but that it also has an effect on circuits in the head region. (Note that the delay to mouth opening remains relatively constant, and so it seems likely that GABA affects the underlying neuromuscular circuitry rather than the GSH receptor itself; reviewed in Pierobon, 2012). Changes in electrical activity in the head region in response to GSH are inhibited by gamma-aminobutyric acid type B (GABAB) ligands suggesting that the GSH-receptor circuitry responsible for mouth opening is located in hypostomal tissue proximal to hydra's mouth. On the basis of these recordings (and the presence of many sensory cells), Lauro and Kass-Simon (2018) suggest that the dome of the hypostome (rather than the mouth itself) bears most of the GSH receptors and that GABA acts via GABAB receptors in the proximal nerve ring. Putative GABAB receptors also modulate nematocyst discharge (Scappaticci & Kass-Simon, 2008) and are present in the tentacles, nematocytes and ganglion cells of the sea fan Eunicella cavolini (Girosi et al., 2007).
Interactions Between Foraging, Feeding and the Response to Threat
Polyps, such as Hydra, search their fishing space by periodically contracting, and moving their tentacles around in an exploratory fashion. These movements, which appear endogenous and driven by a variety of pacemakers, are inhibited during feeding and in the presence of GSH (Rushforth & Hofman, 1972). The duration of the inhibition depends on the animal’s nutritional state. Food extracts also inhibit locomotion in Hydra by suppressing the discharge of adhesive nematocysts (Ewer, 1947). Concerted tentacle movements are suppressed in feeding Tubullaria (Josephson & Mackie, 1965) and are inhibited by food extract (Rushforth, 1969; 1976). Food extracts inhibit rhythmic contractions in the sea anemone Metridium (Pantin, 1950) and they also suppress Stomphia’s swimming response to starfish (Ross & Sutton, 1964).
Foraging movements of this kind may be affected by strong mechanical stimuli such as might be provided by a predator (Rushforth, Burnett, & Maynard, 1963). In Hydra pirardi, an animal that lives in a fast-moving environment, mechanical agitation produces a defensive withdrawal, but the effect habituates over a period of 8 minutes, and foraging restarts in spite of continued mechanical activity (Rushforth, 1973). On the other hand, a species found in still waters, such as H. pseudoligatis, shows no such habituation to threat.
The Role of the Nervous System in Medusoid Cnidaria
Although condensed sets of neurons exist in polypoid Cnidaria, intracellular recording is difficult, and knowledge of their biophysical properties relatively limited. Even the mechanism of facilitation, which Pantin proposed to explain their “diffuse” properties, remains something of a mystery. Many of the medusae, on the other hand, have large neuronal components and are easier to study using intracellular techniques. This has led to a better understanding of the integrative processes that give rise to their behavior. The nerve circuits may be simple, but this is more than compensated for by the complex biophysical properties of the units involved. One well-studied example is the innervation of the subumbrella of the trachymedusa Aglantha digitale. The myoepithelium that makes up the bell of Aglantha is a uniform sheet of electrically coupled contractile cells, but its innervation is such that it can operate in two distinct modes, one specialized for escape, the other for foraging. This dual function arises directly from the biophysical properties of the myoepithelium and the nerves that innervate it. The rest of the article contains a description of this and other examples of medusan behavior such as feeding, defense, and tentacle management.
Neurobiology of Swimming in Aglantha digitale
Aglantha draws attention to itself by a surprisingly rapid escape response initiated by vibrations in the local environment (see Figure 7B; Donaldson, Mackie, & Roberts, 1980; Roberts & Mackie, 1980; Singla, 1978). The receptors concerned, which resemble vertebrate “hair cells,” are located around the bell margin, and their activity excites the large “ring giant axon” in the outer nerve ring (see Figure 3; Arkett, Mackie, & Meech, 1988; Mackie & Meech, 1995b). The ring giant is in indirect synaptic contact with eight large diameter motor axons that run up the inside of the bell (Meech & Mackie, 1995). During an escape swim, action potentials in the ring giant conduct at such a high speed that they excite all eight motor axons more or less coincidentally (see Figure 7C). The Na+-based action potentials generated travel from the base to the apex of the bell in about 5 milliseconds, and so the entire muscle sheet contracts as a unit, forcing a jet of water through the narrowed base of the bell and producing considerable thrust (Kerfoot et al., 1985; Mackie & Meech, 1985; Meech, 2015). As a consequence, the animal jets forward at a speed of 12 body lengths/second (Donaldson et al., 1980).
There are neuromuscular synapses all the way up the bell between the motor axons and the subumbrella myoepithelium. Intracellular recordings show that the synaptic delay (0.7 ms; see Figure 7C) is faster than at the frog neuromuscular junction at the same temperature (10°C; Kerfoot et al., 1985). The depolarization at the synaptic junction (the junction potential) takes the myoepithelium to a point at which a small regenerative Ca2+ spike develops (see Figure 7C). The Ca2+ influx that occurs during the spike is necessary for muscle contraction. In other medusae, such as Polyorchis, regenerative impulses spread excitation through the electrically coupled myoepithelium (Spencer, 1978), but in Aglantha the spread of excitation depends on the numerous lateral neurons that innervate the myoepithelium and divide it into distinct fields (Figure 7C; Kerfoot et al., 1985).
The speed of response of the myoepithelial sheet is remarkable enough, but what is even more remarkable is that the entire neuromuscular system is capable of operating at two levels of power. Aglantha has a second form of swimming—slow swimming, which takes place when the animal is foraging for food (see Figure 7A; Mackie, 1980). Foraging is a cyclical activity, starting with a series of slow swims that drive the animal up the water column. Once Aglantha stops swimming, it turns over and sinks passively, spreading its cloud of tentacles so as to trap any prey in the vicinity. During each slow swim, the animal moves forward no more than one body length, usually less. In full-sized animals, the upper part of the bell gives only a partial contraction, and there is almost no contraction at the base of the bell (Meech, 2015). As a consequence, the thrust generated is minimized and this provides a much more efficient form of jet-propelled swimming, one that can be maintained for long periods.
The neural basis of slow swimming is a low amplitude, Ca2+-dependent spike that propagates along the same set of motor axons as the Na+-dependent action potential responsible for escape swimming (Figure 7C; Mackie & Meech, 1985). This means that these remarkable axons operate with two separate thresholds. The synaptic potential that elicits a slow swim spike arises from pacemaker activity in neurons of the inner nerve ring (Meech & Mackie, 1995). Although small in amplitude, each synaptic potential is large enough to reach the Ca2+-spike threshold (about –55 mV). When fully formed, the Ca2+ spike has such a low amplitude that it can propagate up the motor axon without reaching the escape-swim threshold (which is at about –25 mV). However, near the base of the bell (where the pacemaker cells synapse with the motor axons) the spike is effectively “short-circuited” by the high conductance associated with the synaptic potential and the associated contractions are much weaker than in the mid-bell region (Meech, 2015; 2017; see Figure 7D). The synaptic potential decays exponentially as it spreads electrotonically along the axon until the full regenerative Ca2+ spike is recorded at distances beyond about a third of the way up the bell.
The properties of the ion channels concerned with these two forms of propagating impulse have been analyzed under different forms of voltage clamp (Meech & Mackie, 1993a, 1995). Sampling of membrane patches, using large tipped pipettes (up to 10 μm diameter), suggests that inward currents are focused into small areas of membrane containing a high density of channels. The Na+ current that generates the escape-swim action potential is tetrodotoxin resistant like many invertebrate Na+ channels, whereas the inward current that forms the basis for the slow swim spike flows through T—type Ca2+ channels. The fact that T-type channel inactivation develops slowly near the resting potential, during a slowly rising Ca2+ spike, but quickly during a rapidly rising synaptic potential, may mean that the resulting differences in Ca2+ influx produce the different strengths of neuromuscular transmission observed at different positions on the bell.
When the membrane was explored with 1–2 μm tipped pipettes, only K+-conducting channels were observed. These channels are particularly interesting because they ensure that the peak of the slow swim spike remains below the threshold for escape swimming. The Aglantha axon turns out to have three different classes of K+ channel, each class being found in homogenous clusters. Classes could be distinguished on the basis of their kinetics and voltage dependence, but not by their selectivity or by differences in their single channel conductance (Meech & Mackie, 1993b). The uniformity of their unitary conductance is striking but whether it has a functional significance is not known. Perhaps the channels are generated by alternative splicing.
To explain how the different characteristics of the Ca2+ and Na+ spikes in the motor axon translate into different strengths of contraction of the bell, it was necessary to examine the current/voltage relationship of the myoepithelium (Meech & Mackie, 2006). Although it was clear that Ca2+ influx occurred over a wide range of membrane potentials, it was a surprise to find that two different sets of channels were involved; those activated during escape swimming had a more positive operating range than those activated during slow swimming. Perhaps the system evolved in two stages with slow swimming appearing before escape swimming.
Feeding in Aglatha digitale
Aglantha has 60–80 fine tentacles distributed around the rim (margin) of its bell. Suitable prey, a copepod for example, is trapped by a single tentacle or by a group of tentacles. These contract so as to bring the trapped animal inward to the bell margin. The food object is then transferred to the manubrium (mouth), which bends across to engulf it (see Figure 1A). This behavior, called pointing, is reproduced when electric shocks are applied to the rim of the bell (Mackie, Marx & Meech, 2003). Such electrically induced pointing is highly accurate (to within 15°) and is mediated by impulses (F—potentials) in bundles of small F-system axons that run in the ectoderm near each radial canal. F—potentials excite radial muscles in the manubrium, causing them to contract. Those muscle bands closest to the food site are the first to be excited, but excitation also spreads around the nerve ring to generate impulses in the remaining small axon bundles. The arrival of a single F-potential at the manubrium appears insufficient to cause flexion, but two or three arriving close together produce a visible response. Examination of pairs of F—potentials shows that the second of a pair travels more slowly than the first. This means that impulses travelling from the margin to the manubrium will be closer together if they have travelled by the most direct route. Thus the summed input to the manubrial muscle band closest to the food site will be greater than the summed inputs to other muscle bands. This will determine the direction of pointing.
We may attribute the precision of the pointing mechanism not to any complex nerve circuitry or to any sophisticated neurotransmitter action, but to the simple biophysical properties of the axon membrane itself; the fact that the second of a pair of impulses travels more slowly than the first. For an explanation of why this is so, consider the properties of the axon as it recovers from inactivation. The passage of a single action potential causes the axon to enter a short-lived refractory state. For a short time, the refractoriness is absolute, and no impulse can be elicited, no matter how strong the stimulus. This property is generally accepted as the mechanism by which the directionality of the nerve impulse is determined, as impulses are unable to go back on themselves. Excitability slowly returns as the ion channels recover from inactivation. At this stage, the axon is described as being in a relatively refractory state, because there are fewer channels available to carry inward current. In this condition, action potential propagation will occur at a reduced rate.
Some specimens of Aglantha exhibit tentacle coiling and manubrial searching movements when in the presence of GSH (25 mmol/L). This resembles normal feeding behavior and is reminiscent of the feeding movements of Hydra under these conditions (Loomis, 1955—see Cnidocytes and the Feeding Response above). However, this is not seen in all Aglantha specimens, and responsiveness may depend on the animals’ physiological state. Nitric oxide promotes foraging in both Hydra (i.e., tentacle movements without the typical mouth opening; Colasanti et al., 1997) and Aglantha (slow swimming; Moroz, Meech, Sweedler, & Mackie, 2004).
Inhibition of Swimming During Feeding
In many hydromedusae, the process of feeding brings about a long-term inhibition of swimming. In the case of species with a more oblate bell, such as Aequorea forskalea, swimming inhibition is associated with the contraction of the subumbrella radial muscles, bringing the bell margin closer to the mouth (see Figure 1C; Horridge, 1955; Satterlie, 1985). This action is absent in more prolate forms, such as Aglantha, because the bell is narrow enough for the mouth to reach the margin (see Figure 1A). Nevertheless, swimming is also inhibited in Aglantha, perhaps because the flow of water from the bell during a swim contraction might dislodge the prey.
Once food makes contact with the manubrial lips, bursts of propagated impulses (E-system potentials) are elicited in a nerve plexus within the walls of the radial canals (Mackie et al., 2003; Mackie & Meech, 2008). These impulses travel to an area of the inner nerve ring where the pacemaker neurons that initiate swimming are located (Mackie et al., 2003). In Polyorchis, the swim pacemaker cells, which are particularly large in this species and permit intracellular recording, exhibit long-lasting inhibitory potentials in response to E-system stimulation (Mackie, Meech & Spencer, 2012). The pacemaker cells are electrically coupled so that the entire system behaves like a low-pass filter (Bennett, 1966; Bennett & Zukin, 2004), such that input signals are progressively attenuated and slowed (Spencer, 1981). It seems likely that the prolonged inhibitory potentials produced, directly affect the availability of the different voltage-gated channels in the pacemaker cell membrane. The effect will be predominantly on the voltage-gated K+ channels because channels carrying Ca2+ and Na+ currents are largely insensitive to voltage in the range covered by the inhibitory potentials (Grigoriev, Spafford, Przysiezniak, & Spencer, 1996; Meech, 2015; Przysiezniak & Spencer, 1992, 1994). The prolonged membrane hyper-polarization during the inhibition provides enough time for the K+ channels to recover, at least partially, from inactivation so that they are available to delay the next pacemaker spike. An as yet unexplained feature of the mechanism is its apparent lability. This may be related to the mechanism that couples events in the E-system nerve plexus, which is in the endoderm, with the pacemaker cells in the ectoderm.
Inhibition of Swimming During “Crumpling”—an Avoidance Response
The escape swims exhibited by Aglantha are a highly unusual form of avoidance; most medusae react to agitation by “crumpling” (Hyman, 1940). This involves contracting the subumbrella radial muscles, as well as the tentacles and manubrium, so as to draw the margin and tentacles up into the bell cavity. It can be elicited by mechanical or electrical stimulation and is associated with impulses that are conducted from cell to cell over the entire surface of the bell (Mackie & Passano, 1968; Spencer, 1971, 1975). In Polyorchis, these epithelial action potentials elicit long-lasting inhibitory potentials in the pacemaker system, and swimming is inhibited (Spencer, 1981).
The mechanism by which excitation spreads from exumbrellar ectoderm to the radial muscles in the subumbrellar ectoderm was unclear for many years. However, a key study by King and Spencer (1981) established the likely presence of an intermediate nervous pathway that relays the epithelial excitation to the effector muscle. This is certainly the situation in other animal tissues such as amphibian skin (Roberts, 1971) and the skin of the pelagic tunicates Oikopleura labradoriensis (Bone & Mackie, 1975) and Salpa fusiformis (Anderson, Bone, Mackie, & Singla, 1979). How the epithelial impulse invades the nervous system remains to be discovered, but the most likely mechanism is that gap junctions between epithelial cells and neurons provide a low-resistance pathway.
Graded Responsiveness and Piggybacking
The ability to produce both fast-twitch and slow postural responses is widespread in the Cnidaria, and in some cases it is clear that both occur in the same muscle sheet (see Josephson, 1974). According to George Mackie (1976) one way in which local twitch contractions are coordinated with slower postural ones depends on what he describes as “piggybacking.” Examples of piggybacking occur in both Aglantha and Nanomia.
Nanomia cara is a colonial Cnidarian organized such that the different units of the colony are connected by a contractile stem. Within the stem, the ectodermal myoepithelium is innervated by two large but slightly different-sized motor axons (Mackie, 1973). Action potentials in these axons conduct at sufficiently different speeds that, even if elicited together, they quickly become separated in time. Responses summate at the muscle and produce a contraction that is strongest in the region near the stimulus site, where the action potentials are still close together (see “Feeding in Aglatha digitale”). Contractions cannot propagate in the muscle itself but may be augmented by slowly conducting epithelial impulses within the endoderm (Spencer, 1971; Mackie, 1976, 1978). It appears that current can spread from the endoderm into the muscle by way of transmesogloeal processes (seen in sections prepared for EM) so that impulses conducting in the endoderm alone can bring about a slowly spreading and sustained contraction of the stem.
If axonal and endodermal systems are stimulated together, propagation in the endodermal system (normally 30 cm/s) is greatly speeded up. It is as if the axonal impulse carries the endodermal impulse along—piggyback style. Mackie (1984) suggests that axonal current passes through the transmesogloeal pathway into the endodermal epithelium but will only contribute to the impulse in the vicinity of the stimulus because it is only here that it has been sufficiently summated. Further away from the stimulus, the endodermal impulse will “fall off the piggyback,” and conduction will revert to its normal rate.
The practical effect of piggybacking is that in regions of the stem where the muscle contractions are strong, piggybacking ties the slow postural changes associated with the endodermal impulse to the faster contractions associated with axonal conduction. Further from the stimulus, where the fast contraction fades, the postural movements revert to spreading at a normal speed. This mechanism makes it possible for the giant axon through-conducting system to be involved with producing more localized contractions.
A similar piggybacking phenomenon plays an important role in tentacle management in Aglantha. During slow swims the tentacles show slow postural movements, which range from a slight curling at the tip of the tentacle to complete flexion. The system that controls these movements receives an input from a specific set of neurons (the “relay system”) that run in the inner nerve ring (Mackie & Meech, 1995a). This relay system is itself driven by the “pacemaker system” responsible for slow swimming. As excitation passes around the inner nerve ring, the delay between the pacemaker impulse and the impulse in the relay system remains constant, a relationship ensured by piggybacking, so that the tentacle movements and the slow swims are locked together.
During an escape swim, an outspread cloud of tentacles would provide a significant drag on movement but for the presence of a fast system that ensures the tentacles are contracted before the swim begins. As already discussed (see Neurobiology of Swimming in Aglantha digitale), impulses in the ring giant axon propagate so rapidly around the bell of the animal that the eight motor giant axons fire almost synchronously. Ring giant spikes are almost always matched one for one with spikes in each tentacle giant axon, but the connection is indirect and involves the “carrier system”, another neuronal set running in the inner nerve ring. Conduction in the carrier system appears to be piggybacked on the ring giant axon so that fast tentacle contraction during an escape swim occurs synchronously around the bell (Mackie & Meech, 1995b).
Occasionally, during slow swims, the tentacles exhibit twitch contractions as a result of activity in the tentacle giant axon. These contractions arise from summated responses in the ring giant axon, representing input from three different sets of neuronal systems—the pacemaker, relay, and carrier systems, firing in that order. It appears that the pacemaker system triggers the relay system, and the relay system triggers the carrier system. The ring giant spike then sets off a generalized twitch response in the tentacles without eliciting an escape swim. This is an example of a circuit that normally sets off escape behavior being co-opted to contribute to tentacle management during slow swimming (Mackie & Meech, 1995a).
In Aglantha, tentacles are quickly withdrawn before a fast swim but contract more gradually during a slow swim sequence. Another example of this gradual approach to tentacle management is provided by the siphonophore Chelophyes (Inoue, Tsutsui, & Bone, 2005). The initial swims in a series are quite weak and have the effect of drawing the tentacles together into a more streamlined shape. As the tentacles are reconfigured, and their drag effect becomes less, the swims get stronger. The mechanism for this facilitation lies within the muscle itself because the strength of contraction is related to action potential duration. During a swim sequence, the action potential duration increases progressively, and so more Ca2+ enters the muscle with each impulse. The reason for this is that action potential repolarization depends on an inactivating K+ current which recovers from inactivation only slowly, and so during a sequence of swims more and more repolarizing K+ channels become unavailable (Inoue et al., 2005).
Another siphonophore, Muggiaea, has a contrasting strategy. A disturbance in the surrounding water can lead to swimming, but only after it slowly withdraws its tentacles. Thus swimming can be delayed by 3 seconds or more. However, once swimming begins, each contraction is of the same duration, and the animal moves forward by a standard length. Examination of the ion currents generated in the muscle membrane shows the presence of a non-inactivating K+ current in addition to an inactivating one. Thus the availability of the repolarising K+ current remains relatively constant.
Two hydromedusae for which we have a comprehensive wiring diagram are Aglantha digitale (Mackie et al., 2003) and Polyorchis penicillatus (Arkett & Spencer, 1986; Spencer, 1978, 1979; Spencer & Arkett, 1984).
1. Aglantha digitale: The principal pathways involved in swimming, tentacle control and feeding include (a) the tentacle nerves that report the presence of trapped prey; (b) a system that mediates “pointing” during transfer of prey from tentacles to mouth (F-system; Mackie et al., 2003); (c) a system that mediates swimming inhibition when the mouth has engulfed prey (E-system; Mackie et al., 2012); (d) an epithelial conduction system which inhibits swimming (Mackie & Singla, 1997); (e) a swim pacemaker system (Mackie & Meech, 2000); (f) the nitric oxide pathway which appears to modulate slow swimming (Moroz et al., 2004); (g) a ring giant axon, which receives an input from vibration receptors at the bell margin and distributes excitation to the motor giant axons in the bell and the twitch contraction system in the tentacles (Arkett et al., 1988); (h) the motor giant axons that mediate both fast and slow swimming (Mackie & Meech, 1985; 1995a; 1995b); (i) the lateral neurons that spread excitation through the myoepithelium (Kerfoot et al., 1985); (j) the carrier system, which is interposed between the ring giant axon and the twitch contraction system in the tentacles, and also between the ring giant and the motor giant axon (Mackie & Meech, 1995b); (k) the relay system that links the neurons responsible for fast and slow swimming (Mackie & Meech, 1995a) and which not only excites tonic tentacle contractions during slow swimming but also contributes to ring giant axon excitation, leading the tentacles to give twitch contractions; (l) the tentacle giant axon, responsible for twitch contractions; and (m) the slow tentacle system, responsible for tonic contractions.
2. Polyorchis penicillatus: The muscle epithelium on the inner surface of the bell is directly excited by a ring of electrically coupled swim motor neurons. This nerve ring, which behaves like an endogenous pacemaker, receives an excitatory synaptic input from a second ring of coupled neurons, called the “B” system, which runs around the margin of the bell and along each tentacle. The “B” system neurons receive an excitatory input from photoreceptors (absent from Aglantha). There is also a third coupled network called the “O” system, which receives excitation from the swim motor neurons. Its function is unknown (see Figure 8A; Spencer & Arkett, 1984).
In Polyorchis, the “B” system becomes active prior to swimming, causing the tentacles to contract before swimming starts. In Aglantha, tentacle posture is controlled by the slow tentacle system, but this in turn is excited by the relay system. Thus the functions of the Polyorchis “B” system have been partitioned into distributor (the relay system) and operator (the slow tentacle system) in much the same way as the functions of the pacemaking swim motor neurons in Polyorchis have become separated into the pacemaker system (distributor) and motor axon/lateral neuron systems (operator) in Aglantha. The effect of this partition is twofold: it increases the flexibility of slow swimming and makes escape swimming possible.
Another important adaptation in Aglantha is the ring giant axon. This structure has no equivalent in any other medusa. It is there to spread excitation all around the bell so that the escape contraction is uniform and fast. However, it appears to depend for its function on a network of small carrier system neurons for its interactions with other neuronal subsystems. Micrographs show that although the ring giant receives an input from the many small neurites, probably hair cell processes running beside it in the outer nerve ring, its only obvious output consists of symmetrical synapses onto what seem likely to be carrier system units. Once again the Aglantha system is divided into distributor (the ring giant axon) and operator (the carrier system) components.
Horridge’s (1955) work on the ephyra larva of Aurellia aurita (moon jellyfish), which showed that feeding and swimming could operate independently, provided a satisfying function for each of its two anatomically distinct nerve nets (see Figure 8B). The ephyra, a juvenile form of medusa, is a small transparent disc with eight arms—each arm being capable of trapping protozoans or small copepods and bending so as to bring them toward the mouth. As with Aglantha, the mouth moves to meet it. Each arm is divided into two lappets, and at the point of division there is a rudimentary ganglion located within a hollow structure called a tentaculocyst. As shown in Figure 8B, the tentaculocysts provide a meeting point for the two nerve nets, and it is here that they are likely to interact.
One net (the giant fiber nerve net), is associated with sheets of circular and radial muscle, and consists of large bipolar cells (6 to 10 μm long), which provide the rapid coordination essential for symmetrical swimming. Swimming appears to be initiated within the tentaculocysts because separated arms continue to beat so long as the tentaculocyst is intact. Even when locomotor activity is stopped, the whole arm will still make a feeding movement or a sustained defense response (spasm), in which all the eight arms fold over the oral surface (see Figure 1E). A more dispersed system of neurons (the diffuse nerve net) distributed all over the epithelium appears to initiate the radial muscle contraction that brings an arm toward the mouth during feeding and coordinates this with the movement of the mouth. A mechanism like that described for pointing in Aglantha could provide an explanation for the precision with which this is effected. Besides this, the diffuse nerve net also inhibits swimming during the feeding response, presumably by acting on pacemaker neurons within the tentaculocyst. Horridge (1956) draws attention to the similarity in function between the diffuse net and the oral disk conducting system that controls the tentacles and the feeding reaction in sea anemones (Pantin, 1935a). The “B” system has a similar function in Polyorchis.
In adult Aurellia, the feeding response is very different—prey are drawn in by the continuous beat of the swimming bell, and food is transferred to the mouth by cilliary action (see Figure 1E). The swim pacemakers are once again gathered into ganglion-like structures called rhopalia, and once again there are two distinct nerve nets. Excitation spreads from the rhopalia (Romanes, 1877) to the swimming muscles via the giant fiber nerve net, which also transmits excitation from one rhopalium to another (Anderson & Schwab, 1981, 1983). The giant fiber nerve net transmits activity in any direction. In Cyanea capillata, neuronal communication depends on bidirectional synapses, with either side of the synaptic junction able to release chemical transmitter (see Figure 9; Anderson, 1985). These synapses show no interneuronal facilitation, and in fact the amplitude of the postsynaptic potential declines with repeated stimuli. It should be noted that synapses with arrays of vesicles on either side of a junction also occur in hydromedusae (Jha & Mackie, 1967) and hydroids (Westfall, 1973) as well as in the cubomedusae (Satterlie, 1979).
The role of the diffuse nerve net is to set off tentacle contraction (Romanes, 1885) and trigger slow manubrial reflexes (Passano, 1973). In some species, it can set off a swim, but in Cassiopeia it simply facilitates the effect of the giant fiber nerve net (see Passano, 1965). Although the characteristics of the neurons in the diffuse network are not entirely clear, there is no evidence to suggest they behave any differently to neurons in the giant fiber network. Ultrastructural studies indicate that all synapses in Cyanea are morphologically similar (Anderson & Grunert, 1988).
In Cassiopea, activity in the diffuse nerve net can advance the timing of the swim pacemaker system. The normal swim interval lasts 4–10 seconds, but this can be significantly reduced by a single stimulus to the diffuse nerve net (Passano, 1965). Paired stimuli were even more effective, with the delay being reduced to 0.5 second or less. The effect resembles that of stimulating the “B” system in the hydrozoan jellyfish Sarsia (Passano, 1973). In each case, there is a significant delay between what we could call the priming stimulus and the swim response. In hydrozoan nerve circuits, electrical coupling prolongs synaptic events and provides a possible mechanism for this slow process (see Inhibition of Swimming During Feeding). However, electrical coupling may be absent in the Scyphozoa (Mackie, Anderson, & Singla, 1984), and so it is necessary to seek an alternative explanation. The only electrical event long enough to contribute to the phenomenon is the post-spike hyperpolarization in Cyanea neurons. Particularly interesting is the fact that paired spikes markedly enhance its duration (Anderson & Schwab, 1983). It is possible that such prolonged periods of hyperpolarization have a profound effect on the recovery from inactivation of the different ion channels involved in setting the pacemaker frequency.
Another possibility is that the bidirectional synapses may be recurrently active. According to Anderson (Anderson, 1985), the bidirectional synapses in Cyanea produce brief (about 20 ms) synaptic potentials only because they have an unusually high threshold for transmission (+20 mV or more). If the diffuse network were to have a lower transmission threshold, low-amplitude depolarizations would evoke transmitter release in a postsynaptic cell; this would in turn depolarize the presynaptic cell and evoke further transmitter release there, and so on.
Cubomedusae have four ganglion-like rhopalia each, with about 1,000 neurons and six eyes (two large lens-bearing eyes and two pairs of ocelli). Swimming is driven by four pacemakers (Satterlie, & Nolen, 2001), one in each rhopalia, and is influenced by the input from the eyes in such a way as to make the animal capable of avoiding obstacles (Garm & Mori, 2009). Some species, including Tripedalia cystophora and Copula sivickisi, also exhibit complex courtship and mating behavior (Bentlage et al., 2010; Lewis & Long, 2005; Werner, 1973). As described by Cheryl Lewis for Carybdea sivickisi (Lewis & Long, 2005): courtship begins with both animals swimming, their tentacles fully extended. Once a male attaches a tentacle to one of the females, the tentacle contracts so as to bring their manubriums into direct contact. “Upon contact, the area below each of the male’s hemigonads darkened and a strand of red-pigmented sperm was released . . . into the subgastric sacs. These eight strands of sperm . . . finally coalesced into one thick strand in the manubrium.” (Lewis & Long, 2005, p. 479). Eventually, a globular spermatophore (Werner, 1973) is formed and transferred to one of the female’s tentacles (see Figure 1B). Following transfer of the spermatophore, the male releases the female, who then inserts the spermatophore into her manubrium. After mating, the female stops feeding for a couple of days, and in Copula sivickisi the sperm are partly digested and the nuclei released onto the surface of the female gonad (Garm et al., 2015).
When feeding, species such as Carybdea marsupialis are capable of immobilizing small fish with one of their four tentacles. The tentacle contracts, becoming rigid, and then its base, the pedalium, bends inward to bring the prey into the center of the bell cavity and within reach of the short manubrium (see Figure 1B; Larson, 1976). Animals exposed to GSH (10–1 mg/ml) exhibit a similar feeding reaction while a similar concentration of proline produces no response. GSH also stimulates the manubrium to make searching movements with flared lips. The presence of food near the manubrium inhibits swimming, as in Aglantha.
The normal swim frequency is faster than the natural rhythm of individual pacemakers because the system is driven by the first pacemaker to reach threshold. However, although the rhopalia are linked inasmuch as pacemaker impulses in one can suppress activity in others, the linkage is not absolute. The pacemakers have a level of independence that permits asymmetric swimming when the animal is tracking prey or avoiding obstacles. Turning is achieved through facilitated local contractions (Gladfelter, 1973) and variations in the shape of the valarium (Petie, Garm, & Nilsson, 2013).
Cnidaria as a Basis for Exploring Neural Integration
In medical textbooks, the Cnidaria are often presented as oddities—primitive attempts by evolution to put together a nervous system, with little significance other than as an “origin story” (see Satterlie, 2011 for examples of standard textbook treatments). Such casual misrepresentations are a mistake; the Cnidaria are far more valuable than that. As Michel Anctil (2015) points out in his superb survey of the pioneering work of the early neurozoologists, the “‘footprints’ of nervous system emergence are missing from the paleontological record” (p. ix), and we are left to study animals “presumed to possess the ‘likeness’ of the earliest nervous systems” (p. x). The first nervous systems are beyond reach perhaps, but at our disposal are over nine thousand examples of cnidarian species that have successfully adapted to ever-changing environmental conditions. The secret of their success is written in their every gene, every adaptive shift in behavior or breakthrough neuronal strategy. So, although unlikely to provide hints as to nervous system origins, the Cnidaria provide a model phylum for comparative and genomic studies designed to uncover links between genes and behavior and between neural structure and neural processing.
The following is a list of some of their advantages:
1. The behavioral repertoire consists of a limited number of clear-cut and measurable operations.
2. The neural systems are set in a simple body form with two main layers and a single cavity.
3. The relationship between neural components and behavior can be accessed directly.
4. Behavior patterns can be assessed across diverse forms and varied neural organizations.
5. Asexually reproducing polyps may alternate with sexually reproducing medusae.
6. A colonial organization may compensate for the absence of specialized organs.
7. Hydra and Nematostella have already proved amenable to transgenic manipulation, and provide a model for the application of such techniques to other species.
While presenting part of the historical background to the neuroscience of the Cnidaria, this review has attempted to highlight some of the topics currently discussed by workers in the field. The following is a brief summary:
1. The limited movements available to polypoid forms can be difficult to interpret. Hydra, for example, appears to undergo similar spontaneous contractions when pumping gastric fluid from one region to another, or when foraging for food, or in avoiding mechanical agitation. Fortunately the machine learning approach used by Han, Taralova, Dupre, & Yuste, (2018) to precisely identify behavioral subroutines, may resolve this particular difficulty. Especially now that it is possible to monitor active nerve circuits by recording their Ca2+ transients (Dupre & Yuste, 2017).
2. The question of whether the Cnidaria exhibit long-term behavioral modification (see Rushforth, 1973) has been little studied, but there is evidence for associative learning in some sea anemones: (a) during the swimming response in Stomphia (Ross, 1965), (b) during habituation of aggression in Actinia equina (Brace & Santer, 1991), and (c) during habituation and conditioning in Condylactis gigantean (Hodgson & Hodgson, 1966).
3. Carl Pantin’s view was that sea anemone behavior is a function of the contractile properties of individual muscles (i.e., their rates of contraction and relaxation) and the facilitatory properties of the nerves that excite them (i.e., the rise and fall of facilitation). He thought that an increase in the general excitability of the nerve net, acting through the filter of facilitation and contractility, might translate into specific action by selected muscle systems. Knowing more about facilitation on the one hand and contractility on the other should provide a better understanding of what Pantin (1952, p. 165) called the “behavior machine.” Although, many decades later, the nature of facilitation remains something of a mystery, the use of optical techniques to monitor synaptic activity promises to be a way forward.
4. A means of activating effectors more or less coincidentally over a wide surface is a necessary condition for symmetrical movement in a radially organized animal. Neuronal solutions exist; these include the ring giant axon in Aglantha, with its associated vibration-sensitive hair cells, and the ring of electrically coupled pacemaker neurons in Polyorchis. Conducting epithelia also provide a solution. For example, the radial muscles necessary for crumpling are located all around the bell but can be activated from any point via an epithelial impulse. Epithelial conduction effectively converts the whole exumbrella into a sensory surface. On the face of it, a pathway based on epithelial conduction lacks the precision of a point-to-point neuronal connection, but an intermediate neural link would overcome this disadvantage (see King & Spencer, 1981). The precise mechanism whereby excitation passes from epithelium to nerve remains unknown.
5. In the Hydrozoa, many aspects of behavior depend on the electrical coupling afforded by gap junctions. Examples include the role of prolonged synaptic potentials in the pacemaker circuit of Polyorchis, the spread of currents through the myoepithelium during swimming in Aglantha, and through simple epithelia during crumpling in Sarsia. In view of this, the challenge is to determine how other forms of junction might perform the same function in the Scyphozoa and Cubozoa.
According to Schlosser and Wagner (2004), organisms are not “tightly integrated wholes,” but are “composed of quasi-independent parts that are tightly integrated within themselves” (p. 1). They acknowledge, however, that although the component parts can exhibit degrees of modularity, they are not fully autonomous. This is so in the nervous systems of the Cnidaria, where the more highly evolved forms with the most complex behaviors contain modules with shared nerve circuits. One hundred years ago, Parker debated whether a centralized or locally organized nervous system can best account for sea anemone behavior. Nowadays it is pretty clear that most Cnidaria exhibit a degree of central control, even though how the animal generates a unified response to its surroundings is unclear. The medusan solution is to coordinate local reflexes with condensed, hierarchically organized systems. Hydroids appear to have the same hierarchical organization, although, in most cases, the systems appear less condensed and the command less centralized.
The advantage of a modular system for the experimentalist is that individual circuits have “electrical signatures” that may be identified using external recording electrodes. Techniques such as these combined with transgenomic labeling allow us to examine what highly centralized species can do that less centralized species cannot. Possible comparisons include sea anemones with through-conducting systems versus those without; hydra with oral rings versus those without; and medusae with coordinated escape swimming versus those without. It is hard to exaggerate the great opportunities on offer to researchers prepared to devote themselves to studying the phylum Cnidaria.
Amerongen, H. M., & Peteya, D. J. (1976). The ultrastructure of the muscle system of Stomphia coccinea. In G. O. Mackie (Ed.), Coelenterate ecology and behavior (pp. 541–547). New York, NY: Plenum Publishing.Find this resource:
Amerongen, H. M., & Peteya, D. J. (1980). Ultrastructural study of two kinds of muscle in sea anemones: The existence of fast and slow muscles. Journal of Morphology, 166, 145–154.Find this resource:
Anctil, M. (2015). Dawn of the neuron. Montreal, QB, and Kingston, ON: McGill-Queen’s University Press.Find this resource:
Anctil, M. (2017). Neurotransmission—evolving systems. In S. V. Shepherd (Ed.), The Wiley handbook of evolutionary neuroscience. New York, NY: John Wiley & Sons.Find this resource:
Anderson, P. A. V. (1985). Physiology of a bidirectional, excitatory chemical synapse. Journal of Neurophysiology, 53, 821–835.Find this resource:
Anderson, P. A. V., Bone, Q., Mackie, G. O., & Singla, C. L. (1979). Epithelial conduction in salps: II. The role of nervous and non-nervous conduction system interactions in the control of locomotion. Journal of Experimental Biology, 80, 241–250.Find this resource:
Anderson, P. A. V., & Grunert, U. (1988). Three-dimensional structure of bidirectional, excitatory chemical synapses in the jellyfish Cyanea capillata. Synapse, 2, 606–613.Find this resource:
Anderson, P. A. V. & Mackie, G. O. (1977). Electrically coupled, photosensitive neurons control swimming in a jellyfish. Science, New Series, 197, 186–188.Find this resource:
Anderson, P. A. V., & McKay, M. C. (1987). The electrophysiology of cnidocytes. Journal of Experimental Biology, 133, 215–230.Find this resource:
Anderson, P. A. V., & Schwab, W. E. (1981). The organization and structure of nerves and muscles in the jellyfish Cyanea capillata (Coelenterata: Scyphozoa). Journal of Morphology, 170, 383–399.Find this resource:
Anderson, P. A. V., & Schwab, W. E. (1983). Action potential in neurons of motor nerve net of Cyanea (Coelenterata). Journal of Neurophysiology, 50, 671–683.Find this resource:
Arkett, S., Mackie, G. O., & Meech R. W. (1988). Hair-cell mechanoreception in the jellyfish Aglantha digitale. Journal of Experimental Biology, 135, 329–342.Find this resource:
Arkett, S. A., & Spencer, A. N. (1986). Neuronal mechanisms of a hydromedusan shadow reflex: I. Identified reflex components and sequence of events. Journal of Comparative Physiology A, 159, 201–213.Find this resource:
Assmann, M., Kuhn, A., Dürrnagel, S., Holstein, T. W., & Gründer, S. (2014). The comprehensive analysis of DEG/ENaC subunits in Hydra reveals a large variety of peptide-gated channels, potentially involved in neuromuscular transmission. BMC Biology, 12, 84–98.Find this resource:
Batham, E. J. (1965). The neural architecture of the sea anemone Mimetridium cryptum. American Zoologist, 5, 395–402.Find this resource:
Batham, E. J., Pantin, C. F. A., & Robson, E. A. (1960). The nerve-net of the sea-anemone Metridium senile: The mesenteries and the column. Quarterly Journal of Microscopial Science, 101, 487–510.Find this resource:
Batham, E. J., Pantin, C. F. A., & Robson, E. A. (1961). The nerve-net of Metridium senile:Artifacts and the nerve-net. Quarterly Journal of Microscopial Science, 102, 143–156.Find this resource:
Bennett, M. V. L. (1966). Physiology of electrotonic junctions. Annals of the New York Academy of Sciences, 137, 509–539.Find this resource:
Bennett, M. V. L., & Zukin, R. S. (2004). Electrical coupling and neuronal synchronization in the mammalian brain. Neuron, 41, 495–511.Find this resource:
Bentlage, B., Cartwright, P., Yanagihara, A. A., Lewis, C., Richards, G. S., & Collins, A. G. (2010, February 7). Evolution of box jellyfish (Cnidaria: Cubozoa), a group of highly toxic invertebrates. Proceedings of the Royal Society B: Biological Sciences, 277, 493–501.Find this resource:
Bethe, A. (1903). Allgemeine Anatomie und Physiologie des Nervensystems. Leipzig: G.Thieme. Bode, H., Berking, S., David, C. N., Gierer, A., Schaller, H., & Trenkner, E. (1973). Quantitative analysis of cell types during growth and morphogenesis in Hydra . Wilhelm Roux’s Archives of Developmental Biology, 171, 269–285.Find this resource:
Bone, Q., & Mackie, G. O. (1975). Skin impulses and locomotion in Oikopleura (Tunicata: Larvacea). Biological Bulletin, 149, 267–286.Find this resource:
Bouchard, C., & Anderson, P. A. V. (2014). Immunolocalization of a voltage-gated calcium channel subunit in the tentacles and cnidocytes of the Portuguese man-of-war, Physalia physalis. Biological Bulletin, 227, 252–262.Find this resource:
Bouchard, C., Moneypenny, C. G., Price, R. B., Thompson, L. F., Stalheim, L., & Anderson, P. A. V. (2006). Cloning, functional expression and localization of voltage-gated calcium channel subunits from cnidocytes of the Portuguese Man O’War, Physalia physalis. Journal of Experimental Biology, 209, 2979–2989.Find this resource:
Brace, R. C., & Santer. S,-J. (1991). Experimental habituation of aggression in the sea anemone Actinia equina. Hydrobiologia, 216/217, 533–537.Find this resource:
Brinkmann, M., Oliver, D., & Thurm, U. (1996). Mechanoelectric transduction in nematocytes of a hydropolyp (Corynidae). Journal of Comparative Physiology A, 178,125–138.Find this resource:
Byrne, J. H. (1980). Quantitative aspects of ionic conductance mechanisms contributing to firing pattern of motor cells mediating inking behavior in Aplysia californica. Journal of Neurophysiology, 43, 651–668.Find this resource:
Campbell, R. D., Josephson, R. K., Schwab, W. E., & Rushforth, N. B. (1976). Excitability of nerve-free hydra. Nature, 262, 388–390.Find this resource:
Chapman. J. A., Kirkness, E. F., Simakov, O., Hampson, S. E., Mitros, T., Weinmaier, T., . . . Steele, R. E. (2010). The dynamic genome of Hydra. Nature, 464, 592–596.Find this resource:
Cho, K., & McFarlane, I. D. (1995). A method for recording electrical activity from the body wall nerve nets in sea anemones. Journal of Experimental Biology, 198, 817–820.Find this resource:
Cho, K., & McFarlane, I. D. (1996a). The anthozoan neuropeptide Antho-RWamide I modulates Ca2+ current in sea anemone myoepithelial cells. Neuroscience Letters, 209, 53–56.Find this resource:
Cho, K., & McFarlane, I. D. (1996b). Physiological actions of the neuropeptide Antho-RNamide on antagonistic muscle systems in sea anemones. Neuroscience Letters, 219, 171–174.Find this resource:
Cho, K., & McFarlane, I. D. (1996c). Synaptic potentials recorded from sea anemone muscle cells in situ. Journal of Neuroscience Research, 44, 184–190.Find this resource:
Colasanti, M., Venturini, G., Merante, A., Musci, G., & Lauro, G. M. (1997). Nitric oxide involvement in Hydra vulgaris very primitive olfactory-like system. Journal of Neuroscience, 17, 493–499.Find this resource:
Campbell, R. D. (1987). Structure of the mouth of Hydra spp. A breach in the epithelium that disappears when it closes. Cell and Tissue Research, 249, 189–197.Find this resource:
Concas, A., Pierobon, P., Mostallino, M. C., Marino, G., Minei, R., & Biggio, G. (1998). Modulation of γ-aminobutyric acid (GABA) receptors and the feeding response by neurosteroids in Hydra vulgaris. Neuroscience, 85, 979–988.Find this resource:
David, C. N. (1973). A quantitative method for maceration of Hydra tissue. Wilhelm Roux’s Archives of Developmental Biology, 171, 259–268.Find this resource:
Davis, L. E., Burnett, A. L., & Haynes, J. F. (1968). Histological and ultrastructural study of the muscular and nervous systems in Hydra. Journal of Experimental Zoology, 167, 295–332.Find this resource:
Donaldson, S., Mackie, G. O., & Roberts, A. (1980). Preliminary observations on escape swimming and giant neurons in Aglantha digitale (Hydromedusae: Trachylina). Canadian Journal of Zoology, 58, 549–552.Find this resource:
Dupre, C., & Yuste, R. (2017). Non-overlapping neural networks in Hydra vulgaris. Current Biology, 27, 1085–1097.Find this resource:
Dürrnagel, S., Falkenburger, B. H., & Gründer, S. (2012). High Ca2+ permeability of a peptide-gated DEG/ENaC from Hydra. Journal of General Physiology, 140, 391–402.Find this resource:
Dürrnagel, S., Kuhn, A., Tsiairis, C. D., Williamson, M., Kalbacher, H. Grimmelikhuijzen, C. J., . . . Gründer, S. (2010). Three homologous subunits form a high affinity peptide-gated ion channel in Hydra. Journal of Biological Chemistry, 285, 11958–11965.Find this resource:
Epp, L., & Tardent, P. (1978). The distribution of nerve cells in Hydra attenuata Pall. Wilhelm Roux’s Archives of Developmental Biology, 185, 185–193.Find this resource:
Ewer, D. W. (1947). On the functions and mode of action of the nematocysts of Hydra. Proceedings of the Zoological Society of London, 117, 365–376.Find this resource:
Fraser, S. E., & Bode, H. R. (1981). Epithelial cells of Hydra are dye-coupled. Nature (London), 294, 356–358.Find this resource:
Garm, A., Lebouvier, M., & Tolunay, D. (2015). Mating in the box jellyfish Copula sivickisi—novel function of cnidocytes. Journal of Morphology, 276, 1055–1064.Find this resource:
Garm, A., & Mori, S. (2009). Multiple photoreceptor systems control the swim pacemaker activity in box jellyfish. Journal of Experimental Biology, 212, 3951–3960.Find this resource:
Germain, G., & Anctil, M. (1996). Evidence for intercellular coupling and connexin-like protein in the luminescent endoderm of Renilla koellikeri (Cnidaria, Anthozoa). Biology Bulletin, 191, 353–366.Find this resource:
Girosi, L., Ferrando, S., Beltrame, F., Ciarcia, G., Diaspro, A., Fato, M., . . . Tagliafierro, G. (2007). Gammaaminobutyric acid and related molecules in the sea fan Eunicella cavolini (Cnidaria: Octocorallia): A biochemical and immunohistochemical approach. Cell and Tissue Research, 329, 187–196.Find this resource:
Gladfelter, W. B. (1973). A comparative analysis of the locomotory systems of medusoid Cnidaria. Helgolaender Wissenschaftliche Meeresunters, 25, 228–272.Find this resource:
Golubovic, A., Kuhn, A., Williamson, M., Kalbacher, H., Holstein, T. W., Grimmelikhuijzen, C. J. P., & Gründer, S. (2007). A peptide-gated ion channel from the freshwater polyp Hydra. Journal of Biological Chemistry, 282, 35098–35103.Find this resource:
Grigoriev, N. G., Spafford, J. D., Przysiezniak, J., & Spencer, A. N. (1996). A cardiac-like sodium current in motor neurons of a jellyfish. Journal of Neurophysiology, 76, 2240–2249.Find this resource:
Grimmelikhuijzen, C. J. P. (1983). FMRFamide immunoreactivity is generally occurring in the nervous system of coelenterates. Histochemistry 78, 361–381.Find this resource:
Grimmelikhuijzen, C. J. P. (1985). Antisera to the sequence ArgPhe-amide visualize neuronal centralization in hydroid polyps. Cell and Tissue Research, 241, 171–182.Find this resource:
Grimmelikhuijzen, C. J. P., Darmer, D., Schmutzler, C., Carstensen, K., Moosler, A., Nothacker, H.-P., . . . & McFarlane, I. D. (1992). The peptidergic nervous system of coelenterates, Progress in Brain Research, 92, 137–148.Find this resource:
Grimmelikhuijzen, C. J. P., Williamson, M., & Hansen, G. N. (2004). Neuropeptides in cnidarians. In I. Fairweather (Ed.), Cell signaling in prokaryotes and lower metazoans (pp. 115–139). Dordrecht, Netherlands: Kluwer Academic.Find this resource:
Grosvenor, W., Bellis, S. L., Kass-Simon, G., & Rhoads, D. E. (1992). Chemoreception in hydra: specific binding of glutathione to a membrane fraction. Biochimica et Biophysica Acta (BBA)—General Subjects, 1117, 120–125.Find this resource:
Gründer, S., & Assmann, M. (2015). Peptide-gated ion channels and the simple nervous system of Hydra. Journal of Experimental Biology, 218, 551–561.Find this resource:
Hadzi, J. (1909). Uber das nervensystem von Hydra. Arbeiten aus dem Zoologischen Instituten der Universität, 17, 225–268.Find this resource:
Hall, D. M., & Pantin, C. F. A. (1937). The nerve net of the Actinozoa: V. Temperature and facilitation in Metridium senile. Journal of Experimental Biology, 14, 71–78.Find this resource:
Han, S., Taralova, E., Dupre, C., & Yuste, R. (2018). Comprehensive machine learning analysis of Hydra behavior reveals a stable basal behavioral repertoire. eLife 7:e32605.Find this resource:
Haug, G. (1933). Die lichtreaktionen der Hydren. (Chlorohydra viridissima und Pelmatohydra oligactis). Zeitschrift für Vergleichende Physiologie, 19, 246–303.Find this resource:
Havet, J. (1901). Contribution a 1’etude du systeme nerveux des Actinies. La Cellule, 18, 385–419.Find this resource:
Havrilak, J. A., Faltine-Gonzalez, D., Wen, Y., Fodera, D., Simpson, A. C., Magie, C. R., & Layden, M. J. (2017). Characterization of NvLWamide-like neurons reveals stereotypy in Nematostella nerve net development. Developmental Biology, 431, 336–346.Find this resource:
Hertwig, O., & Hertwig, R. (1879). Die Actinien anatomisch und histologisch mit besonderer beriicksichtigung des nervenmuskelsystems untersucht. Jena Zeit, 13, 457–640.Find this resource:
Hobmeyer, E., Holstein, T., & David, C. N. (1990). Tentacle morphogenesis in hydra: II. Formation of a complex between a sensory nerve cell and a battery cell. Development, 109, 897–904.Find this resource:
Hodgson, V. S., & Hodgson, E. E. (1966). Habituation and discrimination in sea anemones. American Zoologist, 6, 542.Find this resource:
Holman, M. A., & Anderson, P. A. V. (1991). Voltage-activated ionic currents in myoepithelial cells isolated from the sea anemone Calliactis tricolor. Journal of Experimental Biology, 161, 333–346.Find this resource:
Holmes, W. (1947). The peripheral nerve biopsy. In S. C. Dyke (Ed.), Recent advances in clinical pathology (pp. 402–417). London, U.K.: J&A Churchill.Find this resource:
Horridge, G. A. (1955). The nerves and muscles of medusae IV. Inhibition in Aequorea forskalea. Journal of Experimental Biology, 32, 642–648.Find this resource:
Horridge, A. (1956). The nervous system of the ephyra larva of Aurellia aurita. Quarterly Journal of Microscopical Science, 97, 59–74.Find this resource:
Huxley, T. H. (1849). On the anatomy and affinities of the family of the medusae. Philosophical Translations of the Royal Society of London, 139, 413–434.Find this resource:
Huxley, T. H. (1850). Notes on medusae and polypes. Annals and Magazine of Natural History, 6, 66–67.Find this resource:
Hyman, L. H. (1940). Observations and experiments on the physiology of medusae. Biological Bulletin, 79, 282–296.Find this resource:
Inoue, I., Tsutsui, I., & Bone, Q. (2005). Long-lasting potassium channel inactivation in myoepithelial fibres is related to characteristics of swimming in diphyid siphonophores. Journal of Experimental Biology, 208, 4577–4584.Find this resource:
Jackson, A. J., & McFarlane I. D. (1976). Delayed initiation of SSI pulses in the sea anemone Calliactis parasitica: Evidence for a fourth conducting system. Journal of Experimental Biology, 65, 539–552.Find this resource:
Jha, R. K., & Mackie, G. O. (1967). The recognition, distribution and ultrastructure of hydrozoan nerve elements. Journal of Morphology, 123, 43–61.Find this resource:
Jordan, H. (1908). Ueber reflexarme Tiere. II. Zeit. allg. Physiol. 8, 222–266.Find this resource:
Jordan, H. (1912). Ueber rcflexarme Tiere. III. Zeit. wiss. Zool. 101, 116–138.Find this resource:
Josephson, R. K. (1965). Mechanisms of pacemaker and effector integration in coelenterates. Symposia of the Society for Experimental Biology, 20, 33–47.Find this resource:
Josephson, R. K. (1966). Neuromuscular transmission in a sea anemone. Journal of Experimental Biology, 45, 305–319.Find this resource:
Josephson, R. K. (1974). Cnidarian neurobiology. In L. Muscatine and H. M. Lenhoff (Eds.), Coelenterate biology: Reviews and new perspectives. New York, NY: Academic Press.Find this resource:
Josephson, R. K., & Mackie. G. O. (1965). Multiple pacemakers and the behaviour of the hydroid Tubularia. Journal of Experimental Biology, 43, 292–332.Find this resource:
Kass-Simon, G. (1972). Longitudinal conduction of contraction burst pulses from hypostomal excitation loci in Hydra attenuata. Journal of Comparative Physiology, 80, 29–49.Find this resource:
Kass-Simon, G. (1973). Transmitting systems in Hydra. In T. Tokioka & S. Nishimura (Eds.), Recent trends in coelenterate biology: Proceedings of the Second International Symposium on Cnidaria (pp. 583–594). Kushimoto, Japan: Seto Marine Biological Laboratory.Find this resource:
Kass-Simon, G., Pannaccione, A., & Pierobon, P. (2003). GABA and glutamate receptors are involved in modulating pacemaker activity in Hydra. Comparative Biochemistry and Physiology A: Molecular & Integrative Physiology, 136, 329–342.Find this resource:
Kass-Simon, G., & Passano, L. M. (1978). A neuropharmacological analysis of the pacemakers and conducting tissues of Hydra attenuata. Journal of Comparative Physiology A: Molecular & Integrative Physiology, 128, 71–79.Find this resource:
Kass-Simon, G., & Scappaticci, A. A., Jr. (2002). The behavioral and developmental physiology of nematocysts. Canadian Journal of Zoology, 80, 1772–1794.Find this resource:
Kellenberger, S., & Schild, L. (2002). Epithelial sodium channel/degenerin family of ion channels: A variety of functions for a shared structure. Physiological Reviews, 82, 735–767.Find this resource:
Kerfoot, P. A. H., Mackie, G. O., Meech, R. W., Roberts, A., & Singla, C. L. (1985). Neuromuscular transmission in the jellyfish Aglantha digitale. Journal of Experimental Biology, 116, 1–25.Find this resource:
Kinnamon, J. C., & Westfall, J. A. (1981). A three dimensional serial reconstruction of neuronal distributions in the hypostome of a Hydra. Journal of Morphology, 168, 321–329.Find this resource:
Kinnamon, J. C., & Westfall, J. A. (1982). Types of neurons and synaptic connections of hypostome-tentacle junctions in Hydra. Journal of Morphology, 173, 119–128.Find this resource:
King, M. G., & Spencer A. N. (1981). The involvement of nerves in the epithelial control of crumpling behaviour in a hydrozoan jellyfish. Journal of Experimental Biology, 94, 203–218.Find this resource:
Koizumi, O. (2007). Nerve ring of the hypostome in Hydra: Is it an origin of the central nervous system of bilaterian animals? Brain, Behavior and Evolution, 69, 151–159.Find this resource:
Koizumi, O., & Bode, H. R. (1986). Plasticity in the nervous system of adult Hydra: I. The position-dependent expression of FMRFamide-like immunoreactivity. Developmental Biology, 116, 407–421.Find this resource:
Koizumi, O., Wilson, J. D., Grimmelikhuijzen, C. J., & Westfall, J. A. (1989). Ultrastructural localization of RFamide-like peptides in neuronal dense-cored vesicles in the peduncle of Hydra. Journal of Experimental Zoology, 249, 17–22.Find this resource:
Koizumi, O., Itazawa, M., Mizumoto, H., Minobe, S., Javois, J. C., Grimmelikhuijzen, C. J. P., & Bode, H. R. (1992). The nerve ring of the hypostome in Hydra. I. Its structure, development and maintenance. Journal of Comparative Neurology, 326, 7–21.Find this resource:
Larson, R. J. (1976). Cubomedusae: Feeding—functional morphology, behavior and phylogenetic position. In G. O. Mackie (Ed.), Coelenterate ecology and behavior (pp. 237–245). New York, NY: Plenum Press.Find this resource:
Lauro, B. M., & Kass-Simon, G. (2018). Hydra's feeding response: Effect of GABAB ligands on GSH-induced electrical activity in the hypostome of H. vulgaris. Comparative Biochemistry and Physiology, Part A 225, 83–93.Find this resource:
Lawn, I. D. (1976a). Chemoreception and conduction systems in sea anemones. In G. O. Mackie (Ed.), Coelenterate ecology and behavior (pp. 581–590). New York, NY: Plenum Press.Find this resource:
Lawn, I. D. (1976b). The marginal sphincter of the sea anemone Calliactis parasitica. II. Properties of the inhibitory response. Journal of Comparative Physiology 105, 301–311.Find this resource:
Lenhoff H.M. (1961). Activation of the feeding reflex in Hydra littoralis: I. Role played by reduced glutathione, and quantitative assay of the feeding reflex. Journal of General Physiology, 45, 331–344.Find this resource:
Lewis, C., & Long, T. A. F. (2005). Courtship and reproduction in Carybdea sivickisi (Cnidaria: Cubozoa). Marine Biology, 147, 477–483.Find this resource:
Loeb., J. (1895). Zur Physiologie und Psychologie der Actinien. Arch. ges. Physiol. 59, 415–420.Find this resource:
Loomis, W. F. (1955). Glutathione control of the specific feeding reactions of Hydra. Annals of the New York Academy of Science, 62, 211–227.Find this resource:
Mackie, G. O. (1960). The structure of the nervous system in Velella. Quarterly Journal of Microscopial Science, 101, 119–131.Find this resource:
Mackie, G. O. (1965). Conduction in the nerve-free epithelia of Siphonophores. American Zoologist, 5, 439–453.Find this resource:
Mackie, G. O. (1973). Report on giant nerve fibres in Nanomia. In T. Tokioka (Ed.) Recent Trends in Research in Coelenterate Biology. Publ. Seto Mar. Biol. Lab. 20, 745–756.Find this resource:
Mackie, G. O. (1976). The control of fast and slow muscle contractions in the siphonophore stem. In G. O. Mackie (Ed.), Coelenterate ecology and behavior (pp. 647–659). New York: Plenum Press.Find this resource:
Mackie, G. O. (1978). Coordination in physonectid Siphonophores. Marine Behaviour and Physiology, 5, 325–346.Find this resource:
Mackie, G. O. (1980). Slow swimming and cyclical “fishing” behavior in Aglantha digitale (Hydromedusae, Trachylina). Canadian Journal of Fisheries and Aquatic Sciences, 37, 1550–1556.Find this resource:
Mackie, G. O. (1984). Fast pathways and escape behavior in Cnidaria. In R. C. Eaton (Ed.), Neural mechanisms of startle behavior (pp. 15–42). New York, NY: Plenum Press.Find this resource:
Mackie, G. O., Anderson, P. A. V., & Singla, C. L. (1984). Apparent absence of gap junctions in two classes of cnidaria. Biological Bulletin, 167, 120–123.Find this resource:
Mackie, G. O., Marx, R. M., & Meech, R. W. (2003). Central circuitry in the jellyfish Aglantha digitale IV. Pathways coordinating feeding behaviour. Journal of Experimental Biology, 206, 2487–2505.Find this resource:
Mackie, G. O., & Meech, R. W. (1985). Separate sodium and calcium spikes in the same axon. Nature, 313, 791–793.Find this resource:
Mackie, G. O., & Meech, R. W. (1995a). Central circuitry in the jellyfish Aglantha digitale: I. The relay system. Journal of Experimental Biology, 198, 2261–2270.Find this resource:
Mackie, G. O., & Meech, R. W. (1995b). Central circuitry in the jellyfish Aglantha digitale: II. The ring giant and carrier systems. Journal of Experimental Biology, 198, 2271–2278.Find this resource:
Mackie, G. O., & Meech, R. W. (2000). Central circuitry in the jellyfish Aglantha digitale: III. The rootlet and pacemaker systems. Journal of Experimental Biology, 203, 1797–1807.Find this resource:
Mackie, G. O., & Meech, R. W. (2008). Nerves in the endodermal canals of hydromedusae and their role in swimming inhibition. Invertebrate Neuroscience, 8, 199–209.Find this resource:
Mackie, G. O., Meech, R. W., & Spencer, A. N. (2012). A new inhibitory pathway in the jellyfish Polyorchis penicillatus. Canadian Journal of Zoology, 90, 172–181.Find this resource:
Mackie, G. O., & Passano, L. M. (1968). Epithelial conduction in hydromedusae. Journal of General Physiology, 52, 600–621.Find this resource:
Mackie, G. O., Passano, L. M., & Pavans de Ceccatty, M. (1967). Physiologie du comportement de l’hydromeduse Sarsia tubulosa Sars: Les systemes a conduction aneurale. Comptes rendus de l’Académie des Sciences, 264, 466–469.Find this resource:
Mackie, G. O., & Singla, C. L. (1997). The role of epithelial conduction in the behaviour of Aglantha digitale (O. F. Müller, 1776) (Hydromedusae: Rhopalonematidae). In J. C. den Hartog (Ed.), Proceedings of the 6th International Conference on Coelenterate Biology (pp. 307–313). Leiden: National Natuurhistorisch Museum.Find this resource:
Mariscal, R. N. (1974). Nematocysts. In L. Muscatine & H. M. Lenhoff (Eds.), Coelenterate biology (pp. 129–178). New York: Academic Press.Find this resource:
Matsuno, T., & Kageyama, T. (1984). The nervous system in the hypostome of Pelmatohydra robusta: The presence of a circumhypostomal nerve rng in the epidermis. Journal of Morphology, 182, 153–168.Find this resource:
McFarlane, I. D. (1969). Two slow conduction systems in the sea anemone Calliactis parasitica. Journal of Experimental Biology, 51, 377–385.Find this resource:
McFarlane, I. D. (1970). Control of preparatory feeding behaviour in the sea anemone Tealia felina. Journal of Experimental Biology, 53, 211–220.Find this resource:
McFarlane, I. D. (1974a). Control of the pacemaker system of the nerve net in the sea anemone Calliactis parasitica. Journal of Experimental Biology, 61, 129–143.Find this resource:
McFarlane, I. D. (1974b). Excitatory and inhibitory control of inherent contractions in the sea anemone Calliactis parasitica. Journal of Experimental Biology, 60, 397–422.Find this resource:
McFarlane, I. D. (1976). Two slow conduction systems coordinate shell-climbing behaviour in the sea anemone Calliactis parasitica. Journal of Experimental Biology, 64, 431–446.Find this resource:
McFarlane, I. D. (1982). Calliactis parasitica. In G. A. B. Shelton (Ed.), Electrical conduction and behaviour in “simple” invertebrates (pp. 243–265). Oxford, U.K.: Oxford University Press.Find this resource:
McFarlane, I. D. (1984). Nerve nets and conducting systems in sea anemones pathways excite tentacle contractions in Calliactis parsasitica, 108, 137–149.Find this resource:
McFarlane, I. D., Graff, D., & Grimmelikhuijzen, C. J. P. (1989). Evolution of conducting systems and neurotransmitters in the Anthozoa. In P. A. V. Anderson (Ed.), Evolution of the first nervous systems (pp. 111–127). New York and London: Plenum Press.Find this resource:
McFarlane, I. D., Anderson, P. A. V., & Grimmelikhuijzen, C. J. P. (1991). Effects of three anthozoan neuropeptides, Antho-RWamide I, Antho-RWamide II and Antho-RFamide, on slow muscles from sea anemones, Journal of Experimental Biology, 156, 419–431.Find this resource:
McFarlane, I. D., & Lawn I. D. (1972). Expansion and contraction of the oral disc in the sea anemone Tealia felina. Journal of Experimental Biology, 57, 633–649.Find this resource:
Meech, R. W. (2015). Electrogenesis in the lower metazoa and implications for neuronal integration. Journal of Experimental Biology, 218, 1–14.Find this resource:
Meech, R. W. (2017). The evolution of neurons. In S. V. Shepherd (Ed.), Handbook of evolutionary neuroscience (pp. 88–124). Hoboken, NJ: Wiley-Blackwell.Find this resource:
Meech, R. W., & Mackie, G. O. (1993a). Ionic currents in giant motor axons of the jellyfish, Aglantha digitale. Journal of Neurophysiology, 69, 884–893.Find this resource:
Meech, R. W., & Mackie, G. O. (1993b). Potassium channel family in giant motor axons of Aglantha digitale. Journal of Neurophysiology, 69, 894–901.Find this resource:
Meech, R. W., & Mackie, G. O. (1995). Synaptic potentials and threshold currents underlying spike production in motor giant axons of Aglantha digitale. Journal of Neurophysiology, 74, 1662–1670.Find this resource:
Meech, R. W., & Mackie, G. O. (2006). Ionic currents in the myoepithelium of Aglantha digitale. Federation of American Societies of Experimental Biology (FASEB) Journal, 20, A826.Find this resource:
Moroz, L. L., Meech, R. W., Sweedler, J. V., & Mackie, G. O. (2004). Nitric oxide regulates swimming in the jellyfish Aglantha digitale. Journal of Comparative Neurology, 471, 26–36.Find this resource:
Mire, P., Nasse, J., & Venable-Thibodeaux, S. (2000). Gap junctional communication in the vibration-sensitive response of sea anemones. Hearing Research, 144, 109–123.Find this resource:
Nagel, W. A. (1892). Der geschmachssinn der Actinien. Zool. Anz., 15, 334–338.Find this resource:
Oliver, D., Brinkmann, M., Sieger, T., & Thurm, U. (2008). Hydrozoan nematocytes send and receive synaptic signals induced by mechanochemical stimuli. Journal of Experimental Biology, 211, 2876–2888.Find this resource:
Pantin, C. F. A. (1935a). The nerve net of the Actinozoa: I. Facilitation. Journal of Experimental Biology, 12, 119–138.Find this resource:
Pantin, C. F. A. (1935b). The nerve net of the Actinozoa: II. Plan of the nerve net. Journal of Experimental Biology,12, 139–155.Find this resource:
Pantin, C. F. A. (1935c). The nerve net of the Actinozoa: III. Polarity and after-discharge. Journal of Experimental Biology, 12, 156–164.Find this resource:
Pantin, C. F. A. (1942). The excitation of nematocysts. Journal of Experimental Biology, 19, 294–310.Find this resource:
Pantin, C. F. A. (1950). Behaviour patterns in lower invertebrates. Symposia of the Society for Experimental Biology, 4, 175–195.Find this resource:
Pantin, C. F. A. (1952). The elementary nervous system. Croonian Lecture. Proceedings of the Royal Society B: Biological Sciences, 140, 147–168.Find this resource:
Pantin, C. F. A., & Pantin, A. M. P. (1943). The stimulus to feeding in Anemonia sulcata. Journal of Experimental Biology, 20, 6–13.Find this resource:
Parker, G. H. (1917). Actinian nervous system. Journal of Experimental Zoology, 22, 193–229.Find this resource:
Parker, G. H., & Titus, E. G. (1916a). The neuromuscular structure of sea-anemones. Proceedings of the National Academy of Sciences of the United States of America, 2, 339–341.Find this resource:
Parker, G. H., & Titus, E. G. (1916b). The structure of Metridium (Actinoloba) marginatum Milne-Edwards with special reference to its neuro-muscular mechanism. Journal of Experimental Zoology, 21, 433–459.Find this resource:
Passano, L. M. (1965). Pacemakers and activity patterns in medusae: Homage to Romanes. American Zoologist, 5, 465–481.Find this resource:
Passano, L. M. (1973). Behavioural control systems in medusae; a comparison between hydro- and scyphomedusae. Publications of the Seto Marine Biological Laboratory, 20, 615–645.Find this resource:
Passano, L. M., & McCullough, C. B. (1963). Pacemaker hierarchies controlling the behaviour of hydras. Nature, 199, 1174–1175.Find this resource:
Passano, L. M., & McCullough, C. B. (1964). Co-ordinating systems and behaviour in Hydra: I. Pacemaker system of the periodic contractions. Journal of Experimental Biology, 41, 643–664.Find this resource:
Passano, L. M., & McCullough, C. B. (1965). Co-ordinating systems and behaviour in Hydra: II. The rhythmic potential system. Journal of Experimental Biology, 43, 205–231.Find this resource:
Petie, R., Garm, A., & Nilsson, D.-E. (2013). Velarium control and visual steering in box jellyfish. Journal of Comparative Physiology, 199, 315–324.Find this resource:
Pierobon, P. (2012). Coordinated modulation of cellular signaling through ligand gated ion channels in Hydra vulgaris (Cnidaria, Hydrozoa). Int. Journal of Developmental Biology, 56, 551–565.Find this resource:
Pierobon, P. (2015). Regional modulation of the response to glutathione in Hydra vulgaris. Journal of Experimental Biology, 218, 2226–2232.Find this resource:
Pierobon, P., Concas, A., Santoro, G., Marino, G., Minei, R., Pannaccione, A., . . . & Biggio, G. (1995). Biochemical and functional identification of GABA receptors in Hydra vulgaris. Life Sciences, 56, 1485–1497.Find this resource:
Pierobon, P., Minei, R., Porcu, P., Sogliano, C., Tino, A., Marino, . . . Concas, A. (2001). Putative glycine receptors in Hydra: a biochemical and behavioural study. European Journal of Neuroscience, 14(10), 1659–1666.Find this resource:
Pierobon, P., Tino, A., Minei, R., & Marino, G. (2004). Different roles of GABA and glycine in the modulation of chemosensory responses in Hydra vulgaris (Cnidaria, Hydrozoa). Hydrobiologia, 530–531, 59–66.Find this resource:
Popova, L. B., Vornov, D. A., Kosevich, I. A.,& Panchin, I. V. (2012). Gap junctions in sea anemone, Nematostella vectensis, embryo. Zhurnal Obshcheĭ Biologii, 73, 83–87.Find this resource:
Price, R. B., & Anderson, P. A. V. (2006). Chemosensory pathways in the capitate tentacles of the hydroid Cladonema. Invertebrate Neuroscience, 6, 23–32.Find this resource:
Przysiezniak, J., & Spencer, A. N. (1992). Voltage-activated calcium currents in identified neurons from a hydrozoan jellyfish, Polyorchis penicillatus. Journal of Neuroscience, 12, 2065–2078.Find this resource:
Przysiezniak, J., & Spencer, A. N. (1994). Voltage-activated potassium currents in isolated motor neurons from the jellyfish Polyorchis penicillatus. Journal of Neurophysiology, 72, 1010–1019.Find this resource:
Roberts, A. (1971). The role of propagated skin impulses in the sensory system of young tadpoles. Zeitschrift für Vergleichende Physiologie, 75, 388–401.Find this resource:
Roberts, A., & Mackie, G. O. (1980). The giant axon escape system of a hydrozoan medusa, Aglantha digitale. Journal of Experimental Biology, 84, 303–318.Find this resource:
Robson, E. A., & Josephson, R. K. (1969). Neuromuscular properties of mesenteries from the sea-anemone Metridium. Journal of Experimental Biology, 50, 151–168.Find this resource:
Romanes, G. J. (1885). Jelly-fish, star-fish and sea-urchins: Being a research on primitive nervous systems (International Scientific Series). (Vol. 49). New York, NY: D. Appleton.Find this resource:
Romanes, G. J. (1877). Further observations on the locomotor system of medusae. Philosophical Transactions of the Royal Society: B 166, 659–752.Find this resource:
Ross, D. M. (1965). Complex and modifiable behavior patterns in Calliactis and Stomphia. American Zoologist, 5, 573–580.Find this resource:
Ross, D. M., & Sutton, L. (1964). Inhibition of the swimming response by food and of nematocyst discharge during swimming in the sea anemone Stomphia coccinea. Journal of Experimental Biology, 41, 751–757.Find this resource:
Rottini Sandrini, L., & Avian, M. (1989). Feeding mechanism of Pelagia noctiluca (Scyphozoa: Semaeostomeae); laboratory and open sea observations. Marine Biology, 102, 49–55.Find this resource:
Rushforth, N. B. (1969). Electrophysiological correlates of feeding behavior in Tubularia. American Zoologist. 9, 1114.Find this resource:
Rushforth, N. B. (1973). Behavioral modifications in Coelenterates. In W. C. Corning, J. A. Dyal, A. O. Dennis Willows (Eds.), Invertebrate learning (pp. 123–169). New York, NY: Plenum Press.Find this resource:
Rushforth, N. B. (1976). Electrophysiological correlates of feeding behavior in Tubularia. In G. O. Mackie (Ed.), Coelenterate ecology and behavior (pp. 729–738). New York, NY: Plenum Press.Find this resource:
Rushforth, N. B., & Burke, D. S. (1971). Behavioral and electrophysiological studies of Hydra. II. Pacemaker activity of isolated tentacles. Biological Bulletin, 140, 502–519.Find this resource:
Rushforth, N. B., Burnett, A. L., & Maynard, R. (1963). Behavior in Hydra: Contraction responses of Hydra pirardi to mechanical and light stimuli. Science, 139, 760–761.Find this resource:
Rushforth, N. B., & Hofman F. (1972). Behavioral and electrophysiological studies of Hydra. III. Components of feeding behavior. Biological Bulletin, 142, 110–131.Find this resource:
Satterlie, R. A. (1979). Central control of swimming in the cubomedusan jellyfish Carybdea rastonii. Journal of Comparative Physiology, 133, 357–367.Find this resource:
Satterlie, R. A. (1985). Central generation of swimming activity in the hydrozoan jellyfish Aequorea aequorea. Journal of Neurobiology, 16, 41–55.Find this resource:
Satterlie, R. A. (2011). Do jellyfish have central nervous systems? Journal of Experimental Biology, 214, 1215–1223.Find this resource:
Satterlie, R. A., & Nolen, T. G. (2001). Why do cubomedusae have only four swim pacemakers? Journal of Experimental Biology, 204, 1413–1419.Find this resource:
Scappaticci, A. A., & Kass-Simon, G. (2008). NMDA and GABAB receptors are involved in controlling nematocyst discharge in hydra. Comparative Biochemistry and Physiology, Part A: Molecular & Integrative Physiology, 150, 415–422.Find this resource:
Schlosser, G., & Wagner, G. P. (2004). Modularity in development and evolution. Chicago, IL: University of Chicago Press.Find this resource:
Shibley, G. A. (1969). Gastrodermal contractions correlated with rhythmic potentials and prelocomotor bursts in Hydra. American Zoology, 9, 586.Find this resource:
Shimizu, H., & Fujisawa, T. (2003). Peduncle of Hydra and the heart of higher organisms share a common ancestral origin. Genesis, 36, 182–186.Find this resource:
Singla, C. L. (1978). Locomotion and neuromuscular system of Aglantha digitale. Cell and Tissue Research, 188, 317–327.Find this resource:
Spencer, A. N. (1971). Myoid conduction in the siphonophore Nanomia bijuga. Nature, 233, 490–491.Find this resource:
Spencer, A. N. (1975). Behavior and electrical activity in the hydrozoan Probostidactyla flavitirrata (Brandt): II. The medusa. Biological Bulletin, 149, 236–250.Find this resource:
Spencer, A. N. (1978). Neurobiology of Polyorchis: I. Function of effector systems. Journal of Neurobiology, 9, 143–157.Find this resource:
Spencer, A. N. (1979). Neurobiology of Polyorchis: II. Structure of effector systems. Journal of Neurobiology, 10, 95–117.Find this resource:
Spencer, A. N. (1981). The parameters and properties of a group of electrically coupled neurons in the central nervous system of a hydrozoan jellyfish. Journal of Experimental Biology, 93, 33–50.Find this resource:
Spencer, A. N., & Arkett, S. A. (1984). Radial symmetry and the organization of central neurones in a hydrozoan jellyfish. Journal of Experimental Biology, 110, 69–90.Find this resource:
Spencer, A. N., & Satterlie, R. A. (1980). Electrical and dye coupling in an identified group of neurons in a coelenterate. J Neurobiol 11, 13–19.Find this resource:
Stephenson, T. A. (1928). The British sea anemones. (Vol. 1). London, U.K.: Ray Society.Find this resource:
Takahashi, T., Kobayakawa, Y., Muneoka, Y., Fujisawa, Y., Mohri, S., Hatta, M., . . . Koizumi, O. (2003). Identification of a new member of the GLWamide peptide family: Physiological activity and cellular localization in cnidarian polyps. Comparative Biochemistry and Physiology B, 135, 309–324.Find this resource:
Thurm, U., Brinkmann, M., Golz, R., Holtmann, M., Oliver, D., & Sieger, T. (2004). Mechanoreception and synaptic transmission of hydrozoan nematocytes. Hydrobiologia, 530/531, 97–105.Find this resource:
Trembley, A. (1742–1743). Observations and experiments upon the freshwater Polypus, by Monsieur Trembley, at the Hague. (P. H. Z. F. R. S, Trans.). Philosophical Transactions, 42, iii–xi.Find this resource:
Venturini, G. (1987). The hydra GSH receptor: Pharmacological and radioligand binding studies. Comparative Biochemistry and Physiology, 87C, 321–324.Find this resource:
Werner, B. (1973). Spermatozeugmen und paarungsverhahen bei Tripedalia cystophora (Cubomedusae). Marine Biology, 18, 212–217.Find this resource:
Westfall, J. A. (1973). Ultrastructural evidence for a granule containing sensory-motor interneuron in Hydra littoralis. Journal of Ultrastructure Research, 42, 268–282.Find this resource:
Westfall, J. A. (1973). Ultrastructure of synapses in the first-evolved nervous systems. Journal of Neurocytology, 25, 735–746.Find this resource:
Westfall, J. A. (1996). Ultrastructure of synapses in the first-evolved nervous systems. Journal of Neurocytology, 25, 735–746.Find this resource:
Westfall, J. A. (2004). Neural pathways and innervation of cnidocytes in tentacles of sea anemones. Hydrobiologia, 530, 117–121.Find this resource:
Westfall, J. A., Sayyar, K. L., Elliott, C. F., & Grimmelikhuijzen, C. J. P. (1995). Ultrastructural localization of Antho-RWamides I and II at neuromuscular synapses in the gastrodermis and oral sphincter muscle of the sea-anemone Calliactis parasitica. Biological Bulletin, 189, 280–287.Find this resource: