An increase in activity of serotonergic Retzius neurones may not be necessary for the consummatory phase of feeding in the leech Hirudo medicinalis

Serotonin as an integrator of leech behavior and muscle mechanical performance

The obliquely striated muscle in the leech body wall has a broad functional repertoire; it provides power for both locomotion and suction feeding. It also operates over an unusually high strain range, undergoing up to threefold changes in length. Serotonin (5-HT) may support this functional flexibility, integrating behavior and biomechanics. It can act centrally, promoting motor outputs that drive body wall movements, and peripherally, modulating the mechanical properties of body wall muscle. During isometric contractions 5-HT enhances active force production and reduces resting muscle tone. We therefore hypothesized that 5-HT would increase net work output during the cyclical contractions associated with locomotion and feeding. Longitudinal strains measured during swimming, crawling and feeding were applied to body wall muscle in vitro with the timing and duration of stimulation selected to maximize net work output. The net work output during all simulated behaviors significantly increased in the presence of 100μM 5-HT relative to the 5-HT-free control condition. Without 5-HT the muscle strips could not achieve a net positive work output during simulated swimming. The decrease in passive tension associated with 5-HT may also be important in reducing muscle antagonist work during longitudinal muscle lengthening. The behavioral and mechanical effects of 5-HT during locomotion are clearly complementary, promoting particular behaviors and enhancing muscle performance during those behaviors. Although 5-HT can enhance muscle mechanical performance during simulated feeding, low in vivo activity in serotonergic neurons during feeding may mean that its mechanical role during this behavior is less important than during locomotion.

Neural circuits controlling behavior and autonomic functions in medicinal leeches

In the study of the neural circuits underlying behavior and autonomic functions, the stereotyped and accessible nervous system of medicinal leeches, Hirudo sp., has been particularly informative. These leeches express well-defined behaviors and autonomic movements which are amenable to investigation at the circuit and neuronal levels. In this review, we discuss some of the best understood of these movements and the circuits which underlie them, focusing on swimming, crawling and heartbeat. We also discuss the rudiments of decision-making: the selection between generally mutually exclusive behaviors at the neuronal level.

Behavioral choice: a novel role for presynaptic inhibition of sensory inputs

A recent study of how incompatible behaviors are suppressed during feeding in the medicinal leech Hirudo provides new insights into the mechanisms of decision-making in neuronal networks.

Behavioral choice by presynaptic inhibition of tactile sensory terminals

When presented with multiple stimuli, animals generally choose to respond only to one input. The neuronal mechanisms determining such behavioral choices are poorly understood. We found that the medicinal leech had greatly diminished responses to moderate mechanosensory input as it fed. Feeding dominated other responses by suppressing transmitter release from mechanosensory neurons onto all of their neuronal targets. The effects of feeding on synaptic transmission could be mimicked by serotonin. Furthermore, the serotonin antagonist mianserin blocked feeding-induced decreases in synaptic transmission. These results indicate that feeding predominates behaviors by using serotonin at an early stage of sensory processing, namely on presynaptic terminals of mechanosensory neurons.

Activation of two forms of locomotion by a previously identified trigger interneuron for swimming in the medicinal leech

Higher-order projection interneurons that function in more than one behavior have been identified in a number of preparations. In this study, we document that stimulation of cell Tr1, a previously identified trigger interneuron for swimming in the medicinal leech, can also elicit the motor program for crawling in isolated nerve cords. We also show that motor choice is independent of the firing frequency of Tr1 and amount of spiking activity recorded extracellularly at three locations along the ventral nerve cord prior to Tr1 stimulation. On the other hand, during Tr1 stimulation there is a significant difference in the amount of activity elicited in the ventral nerve cord that correlates with the motor program activated. On average, Tr1 stimulation trials that lead to crawling elicit greater amounts of activity than in trials that lead to swimming.

Postactivation inhibition of spontaneously active neurosecretory neurons in the medicinal leech

Spontaneously active neurosecretory neurons in vertebrate and invertebrate nervous systems share similarities in firing frequencies, spike shapes, inhibition by the transmitters they themselves release and postactivation inhibition, an intensity-dependent period of suppressed spontaneous generation of action potentials following phases of high-frequency activity. High-frequency activation of spontaneously active serotonin-containing Retzius cells in isolated ganglia of the leech Hirudo medicinalis induced prolonged membrane hyperpolarisations causing periods of postactivation inhibition of up to 33 s. The duration of the inhibitory periods was directly related to both the number and rate of spikes during activation and was inversely proportional to a cell's spontaneous firing frequency. The periods of postactivation inhibition remained unaffected by both serotonin depletion through repeated injections of 5,7-dihydroxytryptamine and suppressing the afterhyperpolarisation following each action potential with tetraethylammonium (TEA), iberiotoxin or charybdotoxin, suggesting that neither autoinhibition by synaptic release of serotonin nor calcium-activated potassium channels contribute to the underlying mechanism. In contrast, the postactivation inhibitory period was significantly affected both by differential electrical stimulation of the same Retzius cells via microelectrodes filled with molar concentrations of either Na(+)-acetate or K(+)-acetate, and by partial inhibition of Na(+)/K(+)-ATPase with ouabain. Thus, postactivation inhibition in Retzius cells results from prolonged hyperpolarising activity of Na(+)/K(+)-ATPase stimulated by the accumulation of cytosolic Na(+ )during phases of high-frequency spike activity.

Neuronal control of leech behavior

The medicinal leech has served as an important experimental preparation for neuroscience research since the late 19th century. Initial anatomical and developmental studies dating back more than 100 years ago were followed by behavioral and electrophysiological investigations in the first half of the 20th century. More recently, intense studies of the neuronal mechanisms underlying leech movements have resulted in detailed descriptions of six behaviors described in this review; namely, heartbeat, local bending, shortening, swimming, crawling, and feeding. Neuroethological studies in leeches are particularly tractable because the CNS is distributed and metameric, with only 400 identifiable, mostly paired neurons in segmental ganglia. An interesting, yet limited, set of discrete movements allows students of leech behavior not only to describe the underlying neuronal circuits, but also interactions among circuits and behaviors. This review provides descriptions of six behaviors including their origins within neuronal circuits, their modification by feedback loops and neuromodulators, and interactions between circuits underlying with these behaviors.

Synaptic and Extrasynaptic Secretion of Serotonin

Serotonin is a major modulator of behavior in vertebrates and invertebrates and deficiencies in the serotonergic system account for several behavioral disorders in humans. The small numbers of serotonergic central neurons of vertebrates and invertebrates produce their effects by use of two modes of secretion: from synaptic terminals, acting locally in "hard wired" circuits, and from extrasynaptic axonal and somatodendritic release sites in the absence of postsynaptic targets, producing paracrine effects. In this paper, we review the evidence of synaptic and extrasynaptic release of serotonin and the mechanisms underlying each secretion mode by combining evidence from vertebrates and invertebrates. Particular emphasis is given to somatic secretion of serotonin by central neurons. Most of the mechanisms of serotonin release have been elucidated in cultured synapses made by Retzius neurons from the central nervous system of the leech. Serotonin release from synaptic terminals occurs from clear and dense core vesicles at active zones upon depolarization. In general, synaptic serotonin release is similar to release of acetylcholine in the neuromuscular junction. The soma of Retzius neurons releases serotonin from clusters of dense core vesicles in the absence of active zones. This type of secretion is dependent of the stimulation frequency, on L-type calcium channel activation and on calcium-induced calcium release. The characteristics of somatic secretion of serotonin in Retzius neurons are similar to those of somatic secretion of dopamine and peptides by other neuron types. In general, somatic secretion by neurons is different from transmitter release from clear vesicles at synapses and similar to secretion by excitable endocrine cells.

Second messenger cascade of glial responses evoked by interneuron activity and by a myomodulin peptide in the leech central nervous system

The giant glial cell in the neuropil of segmental ganglia of the leech Hirudo medicinalis responds to the activity of the Leydig interneuron and to a peptide of the myomodulin family, the presumed transmitter mediating the Leydig neuron-to-giant glial cell transmission, with a membrane hyperpolarization due to an increased membrane K+ conductance [Britz et al. (2002) Glia, 38, 215-227]. We have now studied the second messenger cascade initiated by Leydig neuron stimulation and by the endogenous myomodulin (MMHir) in the voltage-clamped giant glial cell. Glial responses to both stimuli are mediated by a G-protein-coupled receptor linked to adenylyl cyclase by the following criteria: (i) injection of GDP-beta-S, but not GDP, resulted in an irreversible decrease of the glial responses to both stimuli; (ii) the responses to both stimuli were reversibly inhibited by the adenylyl cyclase inhibitor SQ22,536; and (3) bath-applied di-butyryl-cyclic AMP, but not di-butyryl-cyclic GMP, elicited an outward current, which reduced the responses elicited by neuronal stimulation or myomodulin. A cocktail of protein kinase (PK) inhibitors (H-8, KT5720), the PKA antagonist Rp-cAMPS, or presumed inhibitors of cyclic nucleotide channels, LY83583 and l-cis-diltiazem, had no effect on the glial responses. Our results suggest that Leydig neuron stimulation and MMHir activate a cAMP-mediated K+ conductance in the glial cell, which appeared neither to be due to the activation of PKA nor of known cyclic nucleotide-gated channels directly.

Single-cell analysis reveals cell-specific patterns of expression of a family of putative voltage-gated sodium channel genes in the leech

To understand the molecular basis of nervous system function in the leech, Hirudo medicinalis, we have isolated four novel cDNAs encoding putative voltage-gated sodium (Na) channel alpha subunits, and have analyzed the expression of these genes in individual neurons of known function. To begin, degenerate oligonucleotide primers were used in combination with pre-existing cDNA libraries and reverse transcriptase-coupled polymerase chain reactions (RT-PCR). The putative leech Na channel cDNAs (LeNas) exhibit a higher degree of sequence homology to Na channel genes in other species than to voltage-gated calcium or potassium channel genes, including those expressed in leech. All LeNa cDNAs contain sequences corresponding to regions of functional importance in Na channel alpha subunits, including the "S4 region" involved in activation, the "pore loops" responsible for ion selectivity, and the "inactivation loop" between the third and fourth domains, though the latter lacks the highly conserved "IFM" motif critical for mammalian Na channel inactivation. Sequences corresponding to important determinants of tetrodotoxin sensitivity are found in some, but not all, LeNa cDNAs, consistent with prior electrophysiological evidence of Na channel heterogeneity in the leech with respect to this toxin. Subsequently, two different sets of isoform-specific primers and methods of RT-PCR, including a sensitive, fluorescence-based "real time" RT-PCR, were used to analyze LeNa isoform expression in functionally distinct neurons. The results from both approaches were consistent, and not only demonstrated that individual neurons often express more than one LeNa isoform, but also revealed cell-specific patterns of Na channel isoform expression in the leech nervous system.

Convergence of mechanosensory inputs onto neuromodulatory serotonergic neurons in the leech

By the frequency-dependent release of serotonin, Retzius neurons in the leech modulate diverse behavioral responses of the animal. However, little is known about how their firing pattern is produced. Here we have analyzed the effects of mechanical stimulation of the skin and intracellular stimulation of mechanosensory neurons on the electrical activity of Retzius neurons. We recorded the electrical activity of neurons in ganglia attached to their corresponding skin segment by segmental nerve roots, or in isolated ganglia. Mechanosensory stimulation of the skin induced excitatory synaptic potentials (EPSPs) and action potentials in both Retzius neurons in a ganglion. The frequency and duration of responses depended on the strength and duration of the skin stimulation. Retzius cells responded after T and P cells, but before N cells, and their sustained responses correlated with the activity of P cells. Trains of five impulses at 10 Hz in every individual T, P, or N cell in isolated ganglia produced EPSPs and action potentials in Retzius neurons. Responses to T cell stimulation appeared after the first impulse. In contrast, the responses to P or N cell stimulation appeared after two or more presynaptic impulses and facilitated afterward. The polysynaptic nature of all the synaptic inputs was shown by blocking them with a high calcium/magnesium external solution. The rise time distribution of EPSPs produced by the different mechanosensory neurons suggested that several interneurons participate in this pathway. Our results suggest that sensory stimulation provides a mechanism for regulating serotonin-mediated modulation in the leech.

Somatic exocytosis of serotonin mediated by L-type calcium channels in cultured leech neurones

We studied somatic exocytosis of serotonin and its mediation by L-type calcium (Ca2+) channels in cultured Retzius neurones of the leech. Exocytosis was induced by trains of impulses at different frequencies or by depolarisation with 40 mM potassium (K+), and was quantified by use of the fluorescent dye FM 1-43. Stimulation increased the membrane fluorescence and produced a pattern of FM 1-43 fluorescent spots of 1.28 +/- 0.01 microm in diameter, provided that Ca2+ was present in the bathing fluid. Individual spots lost their stain during depolarisation with 40 mM K+. Electron micrographs showed clusters of dense core vesicles, some of which were in contact with the cell membrane. Presynaptic structures with clear vesicles were absent from the soma. The number of fluorescent spots per soma, but not their diameter or their fluorescence intensity, depended on the frequency of stimulation. Trains at 1 Hz produced 19.5 +/- 5 spots per soma, 77.9 +/- 13.9 spots per soma were produced at 10 Hz and 91.5 +/- 16.9 spots per soma at 20 Hz. Staining patterns were similar for neurones in culture and in situ. In the presence of the L-type Ca2+ channel blocker nimodipine (10 microM), a 20 Hz train produced only 22.9 +/- 6.4 spots per soma, representing a 75 % reduction compared to control cells (P < 0.05). Subsequent incubation with 10 mM caffeine to induce Ca2+ release from intracellular stores increased the number of spots to 73.22 +/- 12.5. Blockers of N-, P-, Q- or invertebrate Ca2+ channels did not affect somatic exocytosis. Our results suggest that somatic exocytosis by neurones shares common mechanisms with excitable endocrine cells.

Membrane responses of the leech giant glial cell to the peptide transmitter myomodulin

A myomodulin peptide has been suggested to mediate the response of the giant glial cells to stimulation of the Leydig interneuron in the central nervous system of the leech Hirudo medicinalis [Eur. J. Neurosci. 11 (1999) 3125]. We have now studied the glial response to the endogenous leech MM peptide (GMGALRL-NH(2), MMHir). The peptide evokes a membrane outward current (EC(50) approximately 2 microM), which neither desensitizes nor shows any sign of run-down, and elicits a K(+) conductance increase of the glial cell membrane. The peptidase inhibitor phenylmethylsulfonyl fluoride (PMSF) enhances the glial current response, suggesting the presence of endogenous extracellular peptidases.

Characterization of a synaptiform transmission between a neuron and a glial cell in the leech central nervous system

The cross-talk between neurons and glial cells is receiving increased attention because of its potential role in information processing in nervous systems. Stimulation of a single identifiable neuron, the neurosecretory Leydig interneuron in segmental ganglia of the leech Hirudo medicinalis, which modulates specific behaviors in the leech, evokes membrane hyperpolarization directly in the giant glial cell (Schmidt and Deitmer. Eur J Neurosci 11:3125-3133, 1999). We have studied the neuron-to-glia signal transmission in the voltage-clamped giant glial cell to determine whether this interaction exhibits properties of a chemical synapse. The glial response had a mean latency of 4.9 s and was dependent on the action potential frequency; the glial cell responded to as few as five Leydig neuron action potentials in 50% of the trials. The glial current was sustained for minutes during repetitive Leydig neuron activity without any sign of desensitization. The current was sensitive to tetraethylammonium, and its reversal potential of -78 mV shifted with the external K+ concentration. The glial response increased with the duration of the neuronal action potentials and was sensitive to the external Ca2+/Mg2+ concentration ratio. The results suggest that Leydig neuron activity leads to a Ca2+-dependent release of transmitter from the neuronal dendrites, evoking an K+ outward current in the giant glial cell, implying a synapse-like transmission between a neuron and a glial cell.

Mixtures of Octopamine and Serotonin Have Nonadditive Effects on the CNS of the Medicinal Leech

It is well established that neural networks respond to a wide variety of modulatory substances by which they can become reconfigured, yet few studies have examined the effects of neurotransmitter mixtures on such networks. In a previous study of the medicinal leech using triple intracellular recordings, we found that stimulation of identified mechanosensory neurons activated both the serotonergic cell 21 (a swimgating neuron) and the dorsal lateral octopamine (DLO) cell. Because these findings suggested that serotonin (5-HT) and octopamine (OA) may be released together, we investigated the effects of 5-HT and OA mixtures on isolated nerve cords of Hirudo medicinalis (which contained both head and tail brains). Fifty micromolar OA, 50 μM 5-HT, or a mixture of 50 μM OA and 50 μM 5-HT was bath applied to the nerve cord under constant perfusion conditions. Additional experiments were performed with combinations of either 25 or 100 μM OA and 5-HT. Neural activity was examined specifically in the segmentally repeated dorsal posterior (DP) nerve because it has been shown to contain identified swim motor units. Nonadditive effects of amine combinations were most apparent in their ability to decrease overall activity in the DP nerve and to alter patterned motor activity in the form of fictive swimming. Whereas swim burst activity has been previously shown to increase in nerve cords bathed in either 5-HT or OA solutions alone, we demonstrated that a mixture of the two amines resulted in a robust decrease in the number of swim bursts expressed and an inhibition of swim activity in preparations already swimming. Most compelling was the observation that when the amine mixture was replaced with normal saline, swim burst activity increased dramatically. We discuss that the effects of amine mixtures may be due to their interaction with descending interneurons known to trigger and inhibit swimming as the mixture-induced effects were not observed in nerve cords lacking the head and tail brains. Because the net effect of the two amines was not simply additive (i.e., 5-HT or OA is known to activate swimming, yet the mix inhibits swimming), this result reveals yet another layer of complexity inherent in “simpler” invertebrate nervous systems.

Postingestive chemosensation and feeding by leeches

Distinct chemical cues acting on chemosensory structures on the dorsal lip of bloodsucking leeches activate the entire complement of appetitive and ingestive feeding behaviors. However, it is not known whether the maintenance of ingestion depends on continuous stimulation of these peripheral chemosensors. Leeches of the species Hirudo medicinalis and Macrobdella decora were fed for 2 min on an artificial blood solution containing 150 mM NaCl/1 mM arginine before switching the feeding solution to various experimental mixtures. Leeches did not start to feed on, but continued to ingest solutions in which equiosmolar KCl or lysine substituted for NaCl or arginine, respectively, until sated. In contrast, they rejected water and dropped off the feeding apparatus within 30 s of the exchange. Leeches also detached from the feeding tube when quinine, denatonium, or caffeine were added to the NaCl/arginine solution during an ongoing feeding bout. The duration of ingestion following the switch correlated inversely with the concentration of the drugs (0.1-10 mM). Superfusion of the dorsal lip with high concentrations of the bitter chemicals, while feeding was in progress, had no effect on the duration of ingestion. However, injections of the bitter substances directly into the gut, during a feeding bout, abruptly stopped ingestion. The results suggest that while leeches continue to sample their food once ingestion has begun, secondary chemosensory mechanisms situated downstream from the dorsal lip may be involved in the maintenance of ingestion and the rapid postingestive rejection of noxious foods.

Behavioral hierarchy in the medicinal leech, Hirudo medicinalis: feeding as a dominant behavior

The effect of feeding behavior on other behaviors (swimming, crawling and shortening) was investigated in the leech, Hirudo medicinalis. The stimulus locations and intensities required to produce mechanically elicited behaviors were first determined in the non-feeding leech. Stimuli were delivered while the leech was in various body positions to determine whether stimulus location affected behavioral response. Response thresholds were determined for the mechanically elicited behaviors. The same stimuli were then applied to feeding leeches to determine if response thresholds had changed. A solution with NaCl and arginine was used to elicit feeding. The same sets of stimuli were applied at intervals for an hour after feeding, to determine the duration of feeding-induced changes in behavior. Depending on the body position and stimulus location, stimuli produced different combinations of behaviors that included shortening, swimming and crawling. Anterior stimuli generally elicited shortening, whereas posterior stimuli generally elicited crawling and swimming, with swimming more likely to ventral stimulation than to dorsal stimulation. Having the front sucker attached changed these behavioral patterns. During feeding, the response thresholds changed dramatically, from 3-5 V to greater than 9 V. This increase in threshold began with the start of feeding, even before ingestion commenced. Suppression of the behaviors lasted up to 1 h after the end of feeding, with the effect on swimming being the most pronounced and longest lasting.

Feeding stimulants activate an identified dopaminergic interneuron that induces the feeding motor program in Helisoma

The neurotransmitter dopamine is shown to play a fundamental role in the generation of the feeding motor pattern and resultant feeding behavior in Helisoma. Application of exogenous dopamine triggered the fictive feeding motor pattern in the isolated CNS and triggered feeding movements in semi-intact preparations. Application of feeding stimulants to the oral cavity excited the putatively dopaminergic buccal interneuron N1a, and depolarization of interneuron N1a triggered the production of the fictive feeding motor pattern. The ability of dopamine superfusion and of interneuron N1a stimulation to activate the fictive feeding motor pattern was blocked by the dopamine antagonist sulpiride. The phase of the fictive feeding motor pattern was reset by brief hyperpolarization of interneuron N1a, demonstrating that interneuron N1a is an integral component of the buccal central pattern generator (CPG). During spontaneous fictive feeding patterns, prolonged hyperpolarizations of interneuron N1a inhibited the production of patterned activity. Exogenous dopamine maintained the fictive feeding motor pattern in the absence of interneuron N1a activity. Interneuron N1a was labeled by the formaldehyde-glutaraldehyde histochemical technique, which is indicative of the presence of dopamine in mollusks. These data suggest that interneuron N1a is an endogenous source of the neuromodulator dopamine, intrinsic to the buccal CPG, and that interneuron N1a has a prominent role in the sensory-motor integration triggering the consummatory response.