ScienceDirect - Brain Research Bulletin: The role of serotonin in reflex modulation and locomotor rhythm production in the mammalian spinal cord Subject View: - All Sciences - Agricultural and Biological Sciences Arts and Humanities Biochemistry, Genetics, and Molecular Biology Business, Management and Accounting Chemical Engineering Chemistry Civil Engineering Computer Science Decision Sciences Earth and Planetary Sciences Economics, Econometrics and Finance Energy and Power Engineering and Technology Environmental Science Immunology and Microbiology Materials Science Mathematics Medicine Neuroscience Pharmacology, Toxicology and Pharmaceutics Physics and Astronomy Psychology Social Science Quick Search: within All Sources This document SummaryPlus Article Journal Format-PDF (693 K) Actions Cited By Save as Citation Alert Export Citation Brain Research Bulletin Volume 53, Issue 5 15 November 2000 Pages 689-710 PII: S0361-9230(00)00402-0 Copyright © 2001 Elsevier Science Inc. All rights reserved. The role of serotonin in reflex modulation and locomotor rhythm production in the mammalian spinal cord Brian J. Schmidt, , a and Larry M. Jordana a Department of Physiology, Faculty of Medicine, University of Manitoba, Winnipeg, Manitoba, Canada Received 5 June 2000; revised 16 August 2000; accepted 20 August 2000. Available online 6 February 2001. Abstract Over the past 40 years, much has been learned about the role of serotonin in spinal cord reflex modulation and locomotor pattern generation. This review presents an historical overview and current perspective of this literature. The primary focus is on the mammalian nervous system. However, where relevant, major insights provided by lower vertebrate models are presented. Recent studies suggest that serotonin-sensitive locomotor network components are distributed throughout the spinal cord and the supralumbar regions are of particular importance. In addition, different serotonin receptor subtypes appear to have different rostrocaudal distributions within the locomotor network. It is speculated that serotonin may influence pattern generation at the cellular level through modulation of plateau properties, an interplay with N-methyl-D-aspartate receptor actions, and afterhyperpolarization regulation. This review also summarizes the origin and maturation of bulbospinal serotonergic projections, serotonin receptor distribution in the spinal cord, the complex actions of serotonin on segmental neurons and reflex pathways, the potential role of serotonergic systems in promoting spinal cord maturation, and evidence suggesting serotonin may influence functional recovery after spinal cord injury. Author Keywords: Central pattern generator; Neuromodulator; Reflex; Motoneuron; In vitro; In vivo Article Outline Introduction Origin of 5-HT in the spinal cord Maturation of bulbospinal serotonergic projections and role in development 5-HT receptor distribution and maturation in the spinal cord Serotonergic modulation of spinal reflex pathways Historical overview Different receptor subtypes mediate the actions of 5-HT on reflex pathways 5-HT has presynaptic actions on segmental afferent fibers 5-HT modulates group II muscle afferent pathways Serotonergic activation and modulation of locomotor networks In vivo studies¯¯cat and rabbit In vitro studies¯¯rat and mouse The 5-HT-sensitive network is distributed throughout the supralumbar cord 5-HT7 receptors may have a critical role in activating locomotor rhythms Locomotor networks display rostrocaudal gradients of responsiveness to 5-HT Why do different preparations display different responsiveness to 5-HT? 5-HT exerts excitatory and inhibitory modulation of the locomotor network Role of 5-HT in locomotor network maturation Possible cellular actions of 5-HT on locomotor network neurons 5-HT facilitates the expression of plateau potentials 5-HT modulates spike afterhyperpolarization 5-HT modulates the voltage-sensitivity of NMDA receptor channels Other effects of 5-HT on membrane properties Role of 5-HT in regeneration and recovery of motor function Summary Acknowledgements References Introduction The possibility that aminergic mechanisms may have a role in mammalian spinal motor control emerged after the demonstration of bulbospinal monoamine systems in the 1960s [9, 56, 57, 88, 89 and 236]. Interestingly, these early experiments included the use of an in vitro mammalian spinal cord to show serotonin (5-HT) release in response to electrical stimulation [5 and 6], two decades before this preparation was popularized by Kudo and Yamada [208] and Smith and Feldman [339] for the study of locomotion. Over the past 13 years, the neonatal rat in vitro spinal cord model has proven to be a powerful instrument for examining the neurochemical substrate of locomotor networks in mammals. Our focus on 5-HT is not intended to suggest that no other monoamines are involved in the activation or modulation of locomotor circuitry. In fact, there is ample evidence to the contrary (e.g., [25, 60, 61, 122, 181, 182, 186, 196, 197 and 199]). However, the potential role of 5-HT in locomotor system activation, modulation and functional recovery after spinal cord lesions has received considerable attention and continues to be an area of major interest and investigation. A recent review of the rapidly evolving subject of 5-HT receptor distribution, structure and pharmacology is provided by Barnes and Sharp [27]. The following discussion is limited to 5-HT actions in the mammalian spinal cord with an emphasis on the modulation of reflexes and locomotor networks. Where relevant, observations obtained from some of the non-mammalian literature are also cited. Within the scope of this review, we do not attempt to address the extensive literature on serotonergic modulation of pain pathways, spinal autonomic systems, or respiration. Origin of 5-HT in the spinal cord The anatomy and development of neuronal projections to the rat spinal cord, including 5-HT fibers, has been reviewed in detail elsewhere (see [210]). Only those observations most relevant to spinal reflexes and locomotor systems are highlighted here. Almost all 5-HT in the rat spinal cord originates supraspinally [56 and 57] from cells located in three of the nine 5-HT-containing brainstem regions described by Dahlstrom and Fuxe [88]. These projections descend from the medullary raphe pallidus (B1), raphe obscuris (B2) and the raphe magnus (B3), as well as from part of the reticular formation immediately surrounding the pyramidal tract at these levels [88 and 89]. 5-HT axon terminals are found at all levels of the spinal cord. Although multiple regions of the spinal cord gray matter are supplied by any given raphe nucleus [210], in general raphe magnus neurons project predominantly to the dorsal horn via the dorsolateral funiculus whereas the raphe obscuris and pallidus project mainly to the motoneuronal cell group and intermediate gray via the ventral and ventrolateral funiculi, respectively [28, 89, 162, 244 and 336]. A single raphe neuron can send axon collaterals to both the cervical and lumbar cord [38, 175 and 220]. Approximately 50% of raphe magnus neurons, 80% of raphe pallidus neurons, and the majority of raphe obscuris neurons contain 5-HT [38, 39 and 40]. However, more recent studies indicate only 47 to 28% of spinal-projecting neurons in the raphe nuclei and surrounding para-raphe zone, respectively, contained 5-HT [189 and 220]. 5-HT and substance P are colocalized in approximately 40% of the raphe-spinal projections [63, 157, 41 and 382] and almost all of these colabelled neurons also contain glutamate [280]. Colocalization of these three substances is associated predominately with 5-HT projections to the ventral horn where 99% of serotonergic axons also contain substance P. The colocalized neurotransmitters are found in large boutons surrounding the dendrites and cell bodies of motoneurons [280 and 382]. Substance P is found in approximately 50% of the 5-HT fibers projecting to the intermediolateral cell column and in only 3% of 5-HT projections to the superficial layers of the dorsal horn [382]. Several other neuroactive substances have been found colocalized with 5-HT, including enkephalin [158, 260 and 354], -aminobutyric acid (GABA) [33, 250 and 259], thyrotropin-releasing hormone [12, 161 and 188], galanin [13, 250 and 255] and other peptides (e.g., [250]). The distribution of 5-HT fiber terminals in the ventral and dorsal horn has been examined in the cat and rat spinal cord (e.g., [4, 12, 183, 344, 360 and 382]). Axons containing both 5-HT and glutamate decarboxylase (GAD) project exclusively to lamina I and II in the superficial dorsal horn [250]. The density of contacts is uniform over the motoneuronal surface and includes over 1500 dendritic contacts and an average of 52 somal contacts per cell; this represents 3¯6% of the total number of synapses on a motoneuron [4]. The ventral horn of one L7 hemisegment of the cat cord contains approximately 55¯110 million serotonergic nerve terminals, which amounts to approximately 4,000 boutons per serotonergic fiber [12]. The concentration of 5-HT in each bouton is approximately 3¯6 mM [12]. In the dorsal horn, the termination of 5-HT axons on two functionally identifiable cell types, the dorsal spinocerebellar tract cell (DSCT) and interneurons receiving group II muscle afferent input, has been examined. Single 5-HT axon terminals were in apposition with the cell bodies of group II interneurons, only a small proportion of which were synaptic contacts [183]. In contrast, rings of 5-HT-containing varicosities apposed DSCT cell bodies, the majority of which formed synaptic contacts [184 and 248]. Jankowska and colleagues [183] proposed that the variation in structural relationship of 5-HT contacts with different cells types may correspond to functional differences in the effect of 5-HT on the different cell types. For instance, the presence of somatic 5-HT synaptic contacts on some cell types, such as motoneurons and dorsal spinocerebellar tract neurons, may be associated with excitatory effects in the form of plateau properties (also see Skydsgaard and Hounsgaard [337]). Other neurons, such as group II afferent neurons, receive few somatic synaptic contacts, and display inhibition by 5-HT [183]. Complete transection of the cat, rabbit, or rat spinal cord spinal cord eliminates all but 2¯15% of the 5-HT content in the cord below the lesion [7, 24, 57, 72, 139 and 235]. Intraspinal 5-HT-containing neurons, which number only 3¯9 in total in the adult rat cord and are located dorsal to the central canal, account for at least part of the residual 5-HT [36, 278 and 279]. The majority (73%) of these cells are found between T3 and L1. Intraspinal 5-HT-containing bipolar neurons appear 1¯2 days after birth while multipolar 5-HT neurons are not found until after day 45 [277]. These cells may have a role in sympathetic and/or nociceptive function [278]. Serotonin-containing intraspinal cells are more common in other mammals such as the opossum [96 and 228] and the monkey [211]. Similarly, the spinal cord of some non-mammalian species such as urodeles [43 and 190] reptiles [198], lamprey [144, 325, 365 and 397], and fish [311 and 367] have a more abundant supply of intraspinal 5-HT-containing neurons than the rat. There is sufficient variation in the number and distribution of intraspinal 5-HT neurons among the different species studied thus far to prevent recognition of any clear evolutionary or phylogenetic pattern that might suggest function. Maturation of bulbospinal serotonergic projections and role in development Rat raphe cells are generated between embryonic days 11 and 15 (E11 and E15) [210]. The axons of 5-HT-containing neurons enter the cervical cord, via the ventral and lateral funiculi, at E13¯14 and a few fibers begin to reach the lumbar cord by E15¯16 [207, 306 and 398]. Invasion of the anterior horn and intermediolateral column by 5-HT axon collaterals, or sharp angulation of axons, starts in the cervical and upper thoracic levels at E15, with more profuse innervation at these levels being present by E16¯17 [306]. The lower thoracic and lumbar gray matter is not invaded until E18 [306 and 398]. Synapses are found in the anterior horn and intermediolateral column by E17¯18 [306]. There are no 5-HT fibers in the dorsal horn, throughout the spinal cord, until E19 when fibers begin to invade the dorsal horn from the lateral funiculus [306]. The intensity of innervation progressively increases. By postnatal day 3 the diffuse pattern of 5-HT innervation is starting to organize into more restricted zones involving the motoneuron area in the ventral horn of the cervical and lumbar enlargement, and the intermediolateral column of the thoracic cord [306]. The intensity and pattern of gray matter innervation starts to resemble that seen in the adult by postnatal day 7¯9 but does not fully mimic the adult pattern until postnatal day 14 in the cervical cord and day 21 in the lower thoracic and lumbar cord [46 and 306]. However, Newton and colleagues showed that the adult pattern of 5-HT innervation of sympathetic nuclei in laminae VII and X is not established until postnatal day 60 [277]. At birth, approximately half of the quadriceps motoneuron population shows close apposition with 5-HT fiber terminals and varicosities, whereas all quadriceps motoneurons have close appositions by postnatal day 5 [353]. Serotonin may serve as a differentiating or stabilizing signal for various neuron populations in early development [218, 219 and 306] and contribute to the regulation of cell division, differentiation and control of growth cones [384]. Whatever the target signals are that attract 5-HT fibers in the fetus, they also appear to be present in the adult. Serotonergic neurons transplanted into the adult cord innervate the gray matter by a process similar to the developmental pattern observed in the fetus [98, 134, 299 and 305]. Embryonic raphe neurons transplanted below the level of an adult rat cord transection also suppress post-lesional changes that normally occur in spinal GABAergic neurons, suggesting that 5-HT, among other factors, may regulate the expression of the GABAergic phenotype [98]. Depletion of 5-HT in newly hatched or adult chickens, using intraperitoneal para-chlorophenylalanine (pCPA) injections, decreases the number of nonserotonergic synapses on cells that are the normal targets of 5-HT fibers, consistent with a role for 5-HT in the normal increase and maintenance of synapses during development [287]. The trophic effect of 5-HT on synaptic density is mediated, at least in part, by 5-HT2A receptors [281] and disappears in the ventral, but not dorsal horn of aged chickens [64]. Intraperitoneal injection of pregnant rats with pCPA produces in the pups smaller motoneuron somas compared with control animals and poor development of dendrites [271 and 272] 5-HT receptor distribution and maturation in the spinal cord Initial autoradiographic studies of 5-HT receptors in the mammalian spinal cord reported multiple 5-HT1 subtypes but failed to detect 5-HT2 receptors [262]. Subsequent investigations demonstrated 5-HT2 receptors, but a generally higher density of 5-HT1 receptors. 5-HT1 receptors dominate the dorsal horn whereas 5-HT2 receptors are found mainly in the ventral horn and intermediolateral nucleus [117, 120, 135, 174, 241, 261, 291, 292, 293, 297, 303 and 356]. Similar to the rostrocaudal gradient of 5-HT in the spinal cord [74 and 139], the distribution of 5-HT1 receptors displays a rostrocaudal gradient with the more caudal regions having a higher density of receptors than rostral regions [241]. However, not all studies have observed a rostrocaudal gradient of 5-HT1 receptor distribution [135]. An immunohistochemical study using an antibody directed against both 5-HT1 and 5-HT2 receptor subtypes demonstrated intense labelling of small cells in the superficial laminae of the dorsal horn as well as motoneurons [310]. Antibody directed specifically against 5-HT1A receptors localized to the axon hillock region (but not dendrites) of primate cervical motoneurons and labelled neurons in the superficial laminae of the dorsal horn as well as around the central canal [192]. In the ventral horn, 5-HT receptor labelling occurs on the active zone of synapses facing presynaptic membrane, whereas in the dorsal horn immunoreactive structures are found on dendrites and cell bodies without synaptic differentiation [310]. These observations are consistent with other studies showing that, in contrast to the ventral horn, most 5-HT projections to the dorsal horn do not make synaptic contacts, suggesting non-junctional (volume) neurotransmission [126, 249, 240 and 310]. The rat ventral horn has high levels of 5-HT2A receptor mRNA [296] and strong immunoreactivity for 5-HT2A receptors, compared to relatively weak labelling in the dorsal horn [78 and 234]. Interestingly, 5-HT2A receptors, unlike other 5-HT receptor subtypes (e.g., 5-HT1A and 5-HT1B), localize predominantly intracellularly instead of on the plasma membrane. The 5-HT2A receptor is transported in axons and dendrites, both antero- and retrogradely, and therefore it is postulated that one of its functions is to mediate retrograde signaling activated by 5-HT [78]. 5-HT2A receptors are also found in dorsal root axons [78] and dorsal root ganglia [234]. High levels of 5-HT2C receptor mRNA are found throughout the grey matter of the spinal cord, except in the dorsal laminae [296]. There is evidence that 5-HT2C receptors may be located mainly presynaptically on the nerve terminals of descending 5-HT fibers in the dorsal thoracic cord. Thus 5-HT2C receptors may function as autoreceptors, decreasing the release of 5-HT, as has been demonstrated for 5-HT1B receptors in the spinal cord [49, 263 and 267]. In the ventral horn, 5-HT2C receptors are located post-synaptically [327]. 5-HT3 receptors are present on both dorsal horn neurons and primary afferents [212 and 264] where they may be involved in pain modulation [141 and 212]. High levels of 5-HT3 receptors are found by immunohistochemical [264] but not by radioligand methods [213] in the ventral horn. The role of 5-HT3 receptors in the spinal motor control remains to be determined [70]. Little is known about the distribution and function of 5-HT7 receptors in the spinal cord. However, we have recently found evidence that this receptor may be involved in the activation of locomotor rhythms (see below). 5-HT receptors are also present on glial cells [234, 150, 142 and 347] in both white and gray matter (especially the dorsal horn) of the spinal cord [192 and 310]. Motoneurons respond to 5-HT application starting at E16, despite the absence of 5-HT projections into the anterior horn until E18 [398]. Experimental suppression of 5-HT synthesis during the perinatal period, using pCPA, results in an increased number or affinity of 5-HT receptors suggesting that 5-HT is not required for receptor expression in developing neurons and may contribute to receptor down-regulation [128]. Electrical stimulation of the dorsolateral fascicle of the spinal cord fails to inhibit dorsal horn neurons before postnatal day 10 [118], even though 5-HT fibers are relatively well established in the dorsal horn by day 6 [118 and 306]. Before E18, 5-HT induces slow-rising potentials (1¯4 mV) that are replaced at E18 by long-lasting depolarizations (mean 6 mV) [398]. The amplitude of these long-lasting potentials increases during the first 3 days after birth (mean = 19 mV) and superimposed repetitive firing also occurs. At embryonic days 18¯21, motoneurons are depolarized in response to agonists that stimulate 5-HT1A, 5-HT2, and 5-HT3 receptor subtypes [398]. Neonatal rat spinal motoneurons also depolarize in response to 5-HT1A receptor activation [351]. However, motoneurons of older animals (aged 2¯3 weeks) hyperpolarization in response to 5-HT1A agonists and depolarization in response to 5-HT2 receptor activation (see below). 5-HT1A receptor activation of neonatal rat hypoglossal motoneurons results in afterhyperpolarization (AHP) suppression, whereas after 3 weeks of age no AHP suppression is produced [352]. In the embryonic chick spinal cord, 5-HT1 and 5-HT2 receptor activation hyperpolarizes motoneurons at E12 but depolarizes cells by E18 [148]. Serotonergic modulation of spinal reflex pathways Historical overview Beginning in 1955, several reports suggested 5-HT had an influence on segmental reflex activity in the spinal cord [86, 201, 202, 225 and 338]. However, initial studies involving iontophoretically applied tryptamines on spinal cord neurons produced negative results leading to the conclusion, in 1962, that tryptamine receptors were not likely to be of much significance with respect to synaptic transmission in the spinal cord [85 and 87]. Marley and Vane [238] suggested the negative results were related to the barbiturate anesthetic used in these experiments. Indeed, subsequent experiments involving decerebrate or ether anesthetized cats demonstrated inhibition (and sometimes excitation) of interneurons, and predominantly inhibition of motoneurons [111, 294 and 381]. In contrast to what one might predict from the early experiments showing predominantly inhibitory effects, systemically administered serotonergic agents or the 5-HT precursor 5-hydroxytryptophan (5-HTP), which unlike 5-HT readily crosses the blood¯¯brain barrier, facilitated monosynaptic reflexes in acutely spinalized animals (but inhibited or excited polysynaptic reflexes) [10, 11, 17, 57, 239, 329 and 366]. Some studies indicated 5-HTP inhibited transmission from flexor reflex afferents to motoneurons, as well as primary afferents and ascending pathways, despite increased motoneuron excitability [8] and gamma motoneuron discharge [2, 106 and 268]. Others reported that serotonergic agents facilitated the flexor reflex [284] and this effect was more prominent in chronic than acute spinalized rats [92]. Electrical stimulation of the raphe spinal projections enhanced the monosynaptic reflex, but depressed the polysynaptic reflex response to dorsal root stimulation in the rat [18]. Thus, by the end of 1970s, the cumulative literature indicated 5-HT had a complex range of effects on spinal neurons and reflex pathways. In the early 1980s, the in vitro neonatal rat spinal cord emerged as a popular preparation for pharmacological investigations of spinal cord function. Initially, ventral root reflexes were used to determine that 5-HT had a depolarizing action on spinal motoneurons [3, 203 and 317]. Subsequent studies used intracellular recordings of motoneurons and synaptic isolation with tetrodotoxin (TTX) to confirm that 5-HT can directly depolarize spinal [75, 107, 108, 276, 351 and 380] and facial [217] motoneurons. Consistent with increased motoneuronal excitation was the observation that 5-HT application or raphe stimulation enhanced antidromic field potentials produced by ventral root stimulation [18, 69, 290, 312 and 387]. Raphe stimulation was also shown to elicit excitatory postsynaptic potentials in cat lumbar motoneurons [124]. Stimulation of the neonatal rat thoracic cord in vitro elicited slow 5-HT receptor-mediated excitatory postsynaptic potentials in lumbar motoneurons [109]. Despite the repeated demonstration of a direct excitatory effect on motoneurons, the complexity of 5-HT actions in the spinal cord was reflected by other studies indicating 5-HT depressed spinal reflexes [71, 82, 83, 140, 270, 286, 317 and 395]. Aghajanian and colleagues [251 and 362] highlighted an important concept when they demonstrated that 5-HT-induced depolarization of rat facial motoneurons modulates the actions of other synaptic inputs, even though the depolarization itself was subthreshold for activating spikes. Similarly, it was shown that 5-HT facilitates the response of rat spinal motoneurons to glutamate-evoked activity [177 and 388], for review see [386]. 5-HT-induced depolarization is associated with a decrease in resting membrane conductance, thought to be related to a decrease in potassium currents in rat facial [217 and 363] and spinal [108, 380, 385 and 396] motoneurons. Consistent with the proposed reduction in potassium current, 5-HT also produced a short-lasting (1¯2 min) decrease in AHP amplitude in rat spinal motoneurons [385], analogous to the 5-HT-induced suppression of AHP observed in lamprey spinal neurons [364]. Although most reports indicated a predominantly depolarizing action of 5-HT on motoneurons, a more complex influence of iontophoretically applied 5-HT on cat motoneurons was observed by Zhang [396], although this study did not involve synaptically isolated neurons. In 50% of the motoneurons examined, 5-HT-induced membrane depolarization was followed by a long-lasting 5-HT receptor-mediated hyperpolarization [396]. Different receptor subtypes mediate the actions of 5-HT on reflex pathways In the in vitro frog spinal cord, low concentrations of 5-HT acting on motoneuronal 5-HT1A receptors produce membrane hyperpolarization as well as an enhancement of the depolarizing response to N-methyl-D-aspartate (NMDA) application and polysynaptic reflex input, whereas higher concentrations of 5-HT produced direct motoneuron depolarization via 5-HT2 receptors [159 and 160]. Similarly, in the rat spinal cord, although initially it was unclear which 5-HT receptor subtype mediated the depolarizing action of 5-HT on motoneurons [75 and 76], it was later shown that 5-HT2 receptors on rat motoneurons mediate the depolarizing response (decreased potassium conductance) noted in 81% of rat motoneurons, whereas 5-HT1A receptors mediate hyperpolarization (increased potassium conductance) which occurs in about 9% of rat lumbar motoneurons [108, 380 and 390]. However, one study concluded that 5-HT1-like receptors are responsible for increased motoneuron excitability [312]. Phrenic motoneurons also display competing responses to 5-HT; depolarization is mediated by 5-HT2 receptors while 5-HT1B receptor activation decreases (possibly presynaptically) inspiratory modulated synaptic current [95, 152 and 224]. Consistent with the concept that 5-HT2 receptors mediate excitation in the spinal cord, 5-HT2 receptor activation restores extensor tone and stretch reflex excitability in the cat spinal cord after cord transection [256], increases the excitability of motoneurons [394], facilitates polysynaptic reflex transmission in the rat [269], increases the sural-gastrocnemius reflex in the rabbit [70], and combined depletion of 5-HT and noradrenaline in the spinal cord of intact rats decreases tonic discharge in the soleus muscle [195]. Synaptic transmission from segmental afferents to motoneurons, including mono- and polysynaptic reflex pathways, is suppressed by exogenously applied 5-HT, or in response to 5-HT1A, 5-HT1B, or 5-HT2 receptor agonists [82, 83, 108, 269 and 390]. Similarly, 5-HT also contributes to suppression of descending glutamatergic responses evoked by stimulation of the ventrolateral thoracic cord [377]. However, facilitation of segmental reflex transmission via presynaptic 5-HT1A receptors [177] or 5-HT1A and 5-HT2 receptors in polysynaptic (but not monosynaptic) reflex pathways [269] has also been reported. 5-HT1A receptors mediate tonic inhibition of transmission in the sural-gastrocnemius reflex pathway in the rabbit, although the precise site of action is unknown [70]. Interestingly, inhibition of the monosynaptic reflex by exogenously applied 5-HT is insensitive to 5-HT2 receptor blockade with ketanserin [82], whereas segmental reflex inhibition produced by endogenously released 5-HT is sensitive to ketanserin [82, 378, 379 and 395]. These observations led to the suggestion that endogenous inhibition produced by descending 5-HT pathways is mediated via subsynaptic 5-HT2 receptors, whereas exogenously applied 5-HT activated a combination of extra-synaptic 5-HT1A or 5-HT1B receptors, or a novel 5-HT receptor subtype, possibly located on the afferent fibers themselves [82, 378 and 379]. Subsequently it was shown that combined blockade of 5-HT1A, 5-HT1B, 5-HT2, and 5-HT3 receptors failed to abolish the inhibitory effect of exogenously applied 5-HT on the monosynaptic reflex, suggesting that a novel 5-HT receptor subtype was involved [94 and 237]. With respect to the effect of endogenously released 5-HT from descending fibers, the concept of 5-HT2 receptor-mediated inhibition of reflex pathways conflicts with the fact that direct activation of motoneuronal 5-HT2 receptors has been shown to produce membrane depolarization and increased excitation (see above). This discrepancy was reconciled by the demonstration that the inhibitory influence of descending 5-HT projections on the monosynaptic reflex is more likely mediated by 5-HT1D than 5-HT2 receptors [237]. Analysis of 5-HT actions on reflex pathways not only needs to consider the specific receptor subtypes but also the enantiomer components of the applied agonist. For instance, the racemic mixture of the `selective' 5-HT1A agonist 8-hydroxy-2-(di-n-propylamino)tetralin hydrobromide (8-OH-DPAT) is not truly selective [163]. In the intact rat, (R)-8-OH-DPAT activates 5-HT1A receptors located supraspinally on serotonergic cells that project to the spinal cord and mediate enhancement of the monosynaptic reflex [163 and 146]. However, when (R)-8-OH-DPAT is administered in high concentrations to intact rats, or given to spinalized rats, the monosynaptic reflex is suppressed suggesting a direct inhibitory action at the segmental level. On the other hand, the (S) enantiomer of 8-OH-DPAT enhances the monosynaptic reflex through supraspinal actions on both descending serotonergic and noradrenergic systems and has no direct effect on reflex modulation at the segmental level when given to spinalized animals [163]. The exact receptor(s) involved in the inhibitory effect of (R)-8-OH-DPAT at the segmental level is unclear, but the participation of 5-HT7 or 5-HT1D receptors has been postulated [163]. 5-HT has presynaptic actions on segmental afferent fibers Electrical stimulation of 5-HT-containing neurons in the brainstem that project to the spinal cord through the dorsolateral fasciculi produces primary afferent depolarization (PAD) [243 and 302], and there is autoradiographic evidence of 5-HT receptors on primary afferent fibers [91 and 141]. The excitability of primary afferent fibers is modulated by 5-HT [58 and 149] and 5-HT application can induce TTX-resistant PAD [227]. 5-HT2 and 5-HT3 receptors mediate depolarization of dorsal root ganglion cell bodies [357]. 5-HT presynaptically inhibits sensory afferents in the Xenopus [330] and lamprey [105]. However, neither direct 5-HT receptor-mediated afferent polarization nor changes in the membrane properties of dorsal horn neurons readily account for the inhibitory influence of 5-HT on synaptic transmission in the rat dorsal horn [226 and 227]. Other mechanisms that result in decreased transmitter release, such as 5-HT1 receptor-mediated blockade of a calcium conductance, has been proposed (e.g., [226]). In their study of presynaptic inhibition of transmission to hypoglossal motoneurons, Berger and colleagues showed that glycinergic transmission was partly inhibited by activation of 5-HT1A receptors located on the soma of the glycinergic interneuron; this in turn produced a decrease of calcium influx at the presynaptic terminal [361]. However, the dominant source of presynaptic inhibition of glycinergic currents was mediated via 5-HT1B receptors using a mechanism independent of inhibition of calcium influx [361]. Glutamatergic synaptic transmission to hypoglossal motoneurons was inhibited by activation of presynaptic 5-HT1B receptors which resulted in a decreased probability of vesicle release [335]. 5-HT modulates group II muscle afferent pathways Jankowska and colleagues have examined the modulatory effect of 5-HT and other monoamines on transmission from group II muscle afferents to cat lumbar motoneurons [44, 45, 282 and 185]. Group II afferent stimulation in the cat elicits a variety of reflex actions including disynaptic excitation of motoneurons [180]. Some midlumbar interneurons with group II input receive projections from the mesencephalic locomotor region [101], are rhythmically active during locomotion [328], and have been postulated to have a role in the transition from stance to swing phase during locomotion [100 and 101]. Before spinalization, group II stimulation inhibits contralateral extensor and flexor motoneurons [14]. After spinalization, group II stimulation produces excitation of contralateral extensor motoneurons (crossed extension reflex). The effect of spinalization on group II transmission is due to bilateral interruption of descending monoaminergic fibers in the dorsolateral fasciculus and can be reversed by application of 5-HT agonists [1]. Stimulation of the raphe nuclei (or locus coeruleus/subcoeruleus nuclei) depresses the group II interneurons in the intermediate zone that mediate excitation of ipsilateral motoneurons [185], and also depresses transmission to dorsal horn neurons [282]. Serotonergic activation and modulation of locomotor networks In vivo studies¯¯cat and rabbit In 1967, Jankowska and colleagues provided the first evidence that monoamines may have a role in the generation of locomotion when they showed that intravenous L-3,4-dihydroxyphenylalanine (L-DOPA) administration to acute spinal cats evoked rhythmic alternating discharge in flexor and extensor efferents [181 and 182]. Two years later, Viala and Buser reported that intravenous 5-HTP administration to lightly anesthetized, paralyzed rabbits with intact nervous systems facilitated the locomotor-like discharge of hindlimb flexor nerves [369 and 371]. They also showed that intravenous 5-HTP induced fictive rhythmic activity, especially in flexor nerves, when administered to acute spinalized (<2 days after cord transection) as well as decerebrate rabbits [370]. Compared to the rabbit, attempts to initiate locomotion in the spinalized cat using 5-HT precursors or agonists have been generally less successful (although see ref [99]). For instance, in preliminary experiments involving 3 low spinal decerebrate cats, Grillner and Shik were unable to elicit locomotor-like alternating flexor and extensor activity in response to 5-HTP injection, although increased extensor and flexor tone was observed [137]. We observed that neurochemical reduction of 5-HT (and noradrenaline) in the spinal cord had no effect on locomotion induced in decerebrate cats using brain stem electrical stimulation [346]. However, the maximum level of spinal cord 5-HT reduction achieved in these experiments was only in the range of 60¯80%, and therefore the results do not allow definitive conclusions about the role of 5-HT in locomotion. Barbeau and Rossignol examined the effect of intravenous 5-HT agonists and antagonists, as well as other monoamines on the initiation and modulation of locomotion in the adult chronic spinal (T13) cat [24 and 25]. They reported that although 5-HT alone failed to initiate locomotion, serotonergic substances modulated treadmill-induced locomotor patterns. In particular, 5-HT precursors and agonists increased the (a) step length, (b) duration and amplitude of hindlimb flexor and extensor EMG activity, and (c) excursion of the hip, knee and ankle joints, especially in the flexion direction. This effect was observed in chronic (>3 months) but not in acutely spinalized animals. Edgerton and colleagues reported a similar modulatory effect of 5-HT agonists on treadmill-induced locomotion in the spinalized cat [99]. In contrast to the effect of 5-HT, administration of either the noradrenergic precursor L-DOPA or agonist clonidine induced locomotion in acutely spinalized (<1 week) preparations (e.g., [22, 25, 51, 122 and 138]). In vitro studies¯¯rat and mouse The development of the in vitro neonatal rat spinal cord preparation for the study of mammalian locomotion offered a new approach for examining the neurochemical substrate of locomotor circuits [208 and 339]. In addition to a critical role for excitatory amino acids, initial investigations using this preparation indicated that other agents including cholinergic, dopaminergic, and noradrenergic substances could evoke rhythmic activity [340 and 341]. It was then shown that 5-HT also induced locomotor-like discharge of lumbar ventral roots [61 and 62], the frequency of which was concentration-dependent [30 and 61]. In fact, of the various agents used to induce locomotor activity in this preparation, we observed that 5-HT alone most reliably evokes a locomotor-like pattern of bilateral flexor and extensor nerve activity (approximately 60% of preparations), whereas NMDA alone induced a locomotor-like pattern in only 10% of preparations (although see [208]) and acetylcholine-induced rhythms were only rarely consistent with hindlimb stepping [80]. Kiehn and Kjaerulff observed that dopamine, more readily than 5-HT, produced an in vitro spatiotemporal profile of flexor and extensor electromyographic signals resembling locomotion in intact adult rats; however 5-HT produced a faster and more stable rhythm overall compared with dopamine [197]. The effect of bath applied 5-HT is blocked by 5HT1 and 5-HT2 receptor antagonists but not by 5-HT3 antagonists [62]. We observed that the rhythmic activity evoked by application of NMDA or acetylcholine (in the absence of bath applied 5-HT) is also abolished by 5-HT antagonists suggesting an essential role for endogenous 5-HT receptor activation in the neurochemical induction of rhythmic activity in this preparation (Fig. 1) [230]. However, Bracci and colleagues observed that locomotor rhythms induced by increasing the extracellular potassium concentration are resistant to 5-HT receptor blockade with ritanserin [42], in which case, direct excitation (depolarization) of locomotor elements achieved by the increased potassium level may bypass the need for 5-HT receptor-mediated effects. (16K) FIG. 1. N-methyl-D-aspartate (NMDA)-induced rhythmic activity in the left (L) and right (R) second (L2) and fifth (L5) ventral roots of the neonatal rat spinal cord (A) is blocked by the addition of the serotonin (5-HT) receptor antagonist mianserin (B). Adapted with permission from [230]. 5-HT produces synchronous bursting on all ventral roots at E16.5, side-to-side alternation with coactivation of ipsilateral ventral roots at E18.5, and both side-to-side as well as ipsilateral ventral root alternation at E20.5 [176]. The functionally mature (postnatal day 12) in vitro mouse spinal cord does not display locomotor activity in response to 5-HT, NMDA, or dopamine alone, but does generate rhythmic alternation of ventral root discharge after combined application of all three neurochemicals [186]. However, in the neonatal mouse (postnatal day 1¯7) either NMDA, 5-HT, or noradrenaline alone can induce rhythmic activity [60 and 281a]. The 5-HT-sensitive network is distributed throughout the supralumbar cord We examined the distribution of the 5-HT-sensitive rhythmogenic network in the neonatal rat spinal cord [81]. If the cord was transected at the T13/L1 junction (Fig. 2B), or if the bath was partitioned at this level, application of 5-HT to the lumbosacral cord alone elicited only tonic hindlimb nerve activity. On the other hand, 5-HT applied selectively to the cervicothoracic cord (i.e., either transected or partitioned at T13/L1) produced rhythmic activity in cervical and thoracic roots, but not in the lumbar region (in the case of the partitioned cord). However, if the cord was transected at the T12/13 level or higher (Fig. 2A), then 5-HT application to the lumbosacral region, in continuity with one or more supralumbar segments, induced locomotor-like discharge in hindlimb nerves. In the case of the cord partitioned at T13/L1, lumbar rhythmic activity was recorded only if 5-HT was applied to both the supralumbar and lumbosacral sides of the partition. (29K) fig. 2. Effect of complete transverse spinal cord lesions on serotonin (5-HT)-induced locomotion in the neonatal rat spinal cord. (A) A locomotor-like pattern of left (L) and right (R) alternating tibial (Tib) and peroneal (Per) nerve activity, produced by bath application of 5-HT, continued after transection at the T4/T5 junction. (B) Locomotion also continued after transection at the T12/T13 junction, but was permanently abolished after transection at the T13/L1 level. The increased tonic discharge observed in response to the latter transection completely subsided after several minutes (not shown). Adapted with permission from [81]. The results of the recent study by Gimenez y Ribotta and colleagues [134] are consistent with our observation that 5-HT promotes locomotor activity only if it is applied to both the lumbar cord and at least a few segments of the supralumbar cord. In their experiments, the locomotor performance of adult rats spinalized at T8 and transplanted with embryonic raphe neurons at T11 improved, while T8 spinalized animals receiving transplants at the T9 level did not [134]. Immunohistochemical analysis of T11 transplanted animals showed serotonergic reinnervation of both the lower thoracic (T8¯13) and upper lumbar cord (L1¯2). In contrast, serotonergic fibers reinnervated the lower thoracic cord but not the lumbar segments in T9 transplanted animals. These findings raise the question whether 5-HT-sensitive elements located specifically in the caudal thoracic region are essential for 5-HT induced rhythmic activity in the lumbar cord. Therefore, we performed double bath partition experiments (barrier at T9 and L3); 5-HT application to the entire cord, excluding the caudal thoracic-upper lumbar region, was still able to induce lumbar rhythmic activity [81]. Thus, it appears that the 5-HT-sensitive oscillatory network, capable of producing a locomotor-like pattern of activity, is diffusely distributed throughout the supralumbar spinal cord and mediates descending rhythmic drive to lumbar motor centers. The tonic discharge observed during application of 5-HT to the lumbosacral region alone is likely due, at least in part, to the direct excitatory effect of 5-HT on lumbar motoneurons. The suggestion that a 5-HT-sensitive network is distributed throughout the supralumbar cord contrasts with the proposal that central pattern generator (induced by a combination of 5-HT and NMDA) is restricted to the L1¯L2 segments [59]. However, a distributed network is consistent with data obtained from other studies using this preparation [205 and 206] and other species [73, 93, 136, 153, 191 and 265], as well as human data indicating that the neuronal circuits generating stepping are distributed throughout the thoracic cord [94]. The important role of supralumbar regions in generating rhythmic drive of the lumbar region is also illustrated by the observation that left-right hindlimb and flexor-extensor coordination can be coordinated and maintained by the supralumbar cord after complete midsagittal section of the lumbosacral cord (Fig. 3A). In addition, transverse hemisection of the cord at L1/2 abolished ipsilateral lumbar rhythmic activity (Fig. 3B), despite continued whole cord exposure to 5-HT. (24K) fig. 3. Role of supralumbar spinal cord regions in generating locomotor rhythms in the neonatal rat spinal cord. (A) A serotonin (5-HT)-induced locomotor-like pattern of rhythmic hindlimb flexor (Per) and extensor (Tib) nerve activity persists despite midsagittal separation of the left (L) and right (R) sides of the cord from T13/L1 to the conus, inclusive. (B) Transverse hemisection of the rostral lumbar cord, at the left L1/L2 level, abolished 5-HT-induced rhythmic activity in ipsilateral (left) but not contralateral segments caudal to the lesion. Note that the left side of the cord is on the right side of the drawing (cord is depicted ventral side up) and Iliac refers to the iliacus muscle, a hip flexor (i.e., electromyographic rather than neurographic activity is shown for this trace only). Adapted with permission from [81]. The results of these experiments using 5-HT alone should be distinguished from the effects of acetylcholine alone, and NMDA used alone or in combination with 5-HT. In particular, our observations suggest that these different neurochemicals activate rhythmogenic substrates with distinct anatomical organizations [81]. For instance, in contrast to 5-HT alone, either NMDA or acetylcholine applied alone to the lumbar region induces rhythmic activity, albeit often non-locomotor-like patterns. 5-HT7 receptors may have a critical role in activating locomotor rhythms We have also explored the role of endogenous 5-HT release using electrical stimulation of the medioventral medulla to activate locomotion in the rat in vitro brainstem¯spinal cord preparation [127]. The medioventral medulla receives projections from the mesencephalic locomotor region [129 and 345], and bulbospinal serotonergic neurons in this region display increased firing rates during locomotor activity [121, 151, 178 and 368]. Stimulation of the medioventral medulla activates locomotion in the cat [129] and in vitro rat [15 and 16]. We placed microdialysis probes at the T12¯L4 levels to sample 5-HT, norepinephrine and dopamine levels, during brainstem-evoked locomotion. A rostrocaudal gradient of release for all three monoamines was observed [127]. 5-HT levels during locomotion were 10-fold higher in the caudal thoracic cord compared with the lumbar cord. Bath partitions enabled application of various serotonergic, noradrenergic, and dopaminergic receptor antagonists to selected spinal cord regions during brainstem-evoked locomotion [127a]. The combined results suggested that 5-HT7 receptors located in the lower thoracic/upper lumbar region, and 5-HT2A receptors located in the caudal (below L3) portion of the lumbar cord, are likely to mediate 5-HT actions on the locomotor network. These results are consistent with the preliminary report by Cina and Hochman [67]. We have recently started to explore the role of 5-HT7 receptors in the adult cat locomotor network. Our preliminary findings indicate that intrathecal administration of the nonselective 5-HT7 antagonist, clozapine, abolishes locomotion evoked by brainstem stimulation in the cat (Fig. 4), analogous to the effect of clozapine in the in vitro rat cord. Further experiments using other 5-HT receptor antagonists, as well as dopaminergic and noradrenergic antagonists, are needed to clarify whether 5-HT7 receptors are involved. Our recent mapping studies indicate that cells double-labelled for activity (using c-fos immunohistochemistry) during treadmill walking and for the presence 5-HT7 receptors, are particularly numerous in the caudal thoracic cord (unpublished observations). Over 80% of the c-fos positive cells are also labelled with the 5-HT7 antibody. (27K) fig. 4. Intrathecal application of the serotonin (5-HT)7 receptor antagonist clozapine blocks fictive locomotion induced by electrical stimulation of the mesencephalic locomotor region in the cat. (A) Rhythmic hindlimb nerve discharge [shown on an expanded time scale in (a1)] was not affected by low concentrations of clozapine (0.1 mM) applied intrathecally to the caudal thoracic and lumbar spinal segments. (B) In contrast, clozapine (1 mM) was effective in blocking fictive locomotion [shown on an expanded time scale in (b1)] within approximately 30 s of application. Recovery 116 min after clozapine (1 mM) administration is shown in (b2). The 20-s scale bar applies to the left panels in (A) and (B). The 2-s scale bar applies to (a1), (b1), and (b2). Locomotor networks display rostrocaudal gradients of responsiveness to 5-HT The combined results of the experiments outline above involving cord partitioning and lesioning, microdialysis sampling, antagonist application, and immunohistochemical techniques suggest there exists a rostrocaudal gradient of locomotor network responsiveness to 5-HT and bulbospinal release of 5-HT in the cord. A rostrocaudal gradient of spinal 5-HT immunoreactivity thought to be related to the axial locomotor network has also recently been described in the urodele [43]. In addition, both rostrocaudal and caudorostral gradients of 5-HT influence have been observed in the lamprey spinal cord depending on the excitatory amino acid used to induce locomotion [397]. 5-HT release has been examined in rats during free-running or treadmill exercise using microdialysis probe implanted in either the ventral funiculus white matter or the ventral horn itself (at the L4 level) [130, 131 and 132]. 5-HT levels increased in the white matter but not in the ventral horn. These investigators speculated that the failure to detect an increase in 5-HT during exercise in this study may have been because the release was masked by an equally effective activation of the 5-HT uptake system [131], or feedback inhibition of 5-HT release via presynaptic 5-HT autoreceptors [132]. However, an additional factor may be that that sampling at the L4 level alone was too caudal to detect significant increases in 5-HT release. Cina and Hochman used sulforhodamine to identify thoracolumbar (T11¯L6) neurons that are active during 5-HT-induced locomotion [68]. Of course, in this type of experiment cell labelling does not necessarily indicate that the neuron is also 5-HT sensitive. Labelled neurons were diffusely scattered in the thoracolumbar cord without evidence of a rostrocaudal gradient. Why do different preparations display different responsiveness to 5-HT? It appears there is a discrepancy between the results of in vitro (rat) studies showing 5-HT activation of locomotor-like activity and in vivo (cat) experiments where attempts to initiate locomotion using serotonergic drugs have been generally unsuccessful. The level of the cord lesion may be relevant. In vivo cat studies have generally used preparations spinalized at the low thoracic (T13) level. If a 5-HT-sensitive network with important supralumbar components exists in the adult cat model, as is the case for the in vitro neonatal rat [81], then predictably 5-HT will fail to induce locomotion in the lumbosacral cord isolated from the thoracic cord by a T13 transection (e.g., [24, 25 and 137]). Interestingly, the only study examining the effect of 5-HTP administration to high spinalized (C1) cats reported that "brief alternating movements of the forelimbs in register with those of the hindlimbs were occasionally seen" [257 and 258]. Table 1 indicates that successful serotonergic activation of locomotor rhythms in adult in vivo mammals has involved preparations that preserve at least a few thoracic segments. One exception to this trend, with respect to the level of spinalization, is the study by Edgerton and colleagues. They noted that "on occasion" treadmill locomotion could be induced by the serotonergic agonist quipazine in cats transected at the T12/13 level [99]. Otherwise quipazine usually modulated stepping in cats that had already acquired, through training, at least some degree of locomotor ability on the treadmill [99]. Table 1. Spinal cord transection level vs. initiation of locomotion in response to serotonergic (5-ht) substances, administered to in vivo mammalian preparationslegend, legend, legend (21K) Certainly other factors, including species type, developmental issues, and route of application may contribute to the variable effectiveness of 5-HT in different models of locomotion. In fact, considering the complex and competing actions of 5-HT on spinal cord neurons and reflexes, mediated by different receptor subtypes differentially distributed in the cord, perhaps it is not surprising that 5-HT induces locomotion with variable effectiveness among different experimental preparations. The capacity to initiate locomotion from a quiescent state in the in vitro preparation may be the result of non-specific excitation of the cord by 5-HT, an effect not otherwise achievable in adult in vivo preparations. In particular, the increased responsiveness of the neonatal rat in vitro cord may be a consequence of its immaturity and/or the ability to rapidly achieve critical concentrations of 5-HT throughout the entire cord during bath application. In the intact animal, 5-HT may serve primarily as a modulator, facilitating the appropriate timing and rhythmicity of various network components, rather mediating activation of the network. 5-HT exerts excitatory and inhibitory modulation of the locomotor network In addition to the apparent network activating function of 5-HT in the in vitro rat spinal cord preparation, 5-HT also has a modulatory role in this preparation. However, there is evidence of competing excitatory and inhibitory influences. Thus, 5-HT has an inhibitory effect on the frequency of rhythms induced by NMDA or increased extracellular potassium concentration but enhances burst amplitude and duration [31, 42 and 343]. This twofold modulatory action of 5-HT on rhythm frequency versus burst intensity is similar to that described in several other preparations including the lamprey [66 and 143] and urodele [43 and 190]. The slow stable rhythm induced by 5-HT contrasts with the faster but more labile rhythms evoked by NMDA alone, and may account for the observation that combined NMDA/5-HT application produces a more robust and stable rhythm than use of either agent alone in the rat cord [79, 80, 204 and 343]. Beato and Nistri showed that 5-HT-induced decrease in the frequency of NMDA-induced rhythm in the neonatal rat is facilitated by 5-HT2 receptor blockade and partially mimicked by 5-HT1A and 5-HT1B/D receptor agonists [31]. Role of 5-HT in locomotor network maturation In addition to a neuromodulatory function, 5-HT may have an influence on locomotor network maturation. In the Xenopus, the locomotor network develops in a rostrocaudal fashion in concert with the descent of the bulbospinal 5-HT projections, and in parallel with a rostrocaudal progression of locomotor responsiveness to 5-HT application [333 and 334]. Reduction of 5-HT in the prenatal and neonatal period, by intraperitoneal injection of pregnant rats with pCPA, produces a marked delay in the development of the 5-HT system in the spinal cord of the pups and the swimming pattern is disorganized [273]. However, between postnatal days 17 and 22 the spinal 5-HT system and swimming pattern recovers to normal. When the 5-HT levels are increased in the perinatal period, using a monoamine oxidase A-deficient mouse, there is also a transient delay in the development of the swimming pattern which recovers by postnatal day 14 [60]. Thus, normal development of the spinal 5-HT system appears to be critical for proper operation of the locomotor network during the first 2 weeks of life. However, because normal swimming is ultimately achieved by postnatal day 14¯22, despite experimental perturbations of the serotonergic system, the question is raised whether 5-HT is in fact essential for the development of a functional network in the adult. Spinal cord circuitry in the immature cord may be able to adapt to experimentally induced anomalies of the serotonergic system such that functional locomotor circuitry is assembled by postnatal day 14¯22 regardless of the perinatal insult. Possible cellular actions of 5-HT on locomotor network neurons 5-HT facilitates the expression of plateau potentials Plateau potentials are slow regenerative depolarizations typically lasting longer than sodium spikes [145]. Plateau properties have been examined extensively in cat and turtle spinal motoneurons [77, 84, 102, 103, 164, 165, 166, 167, 168, 169, 170, 171, 194, 221, 222, 289, 313, 314, 315, 324, 337, 349 and 350] and have been postulated to help shape motor output during locomotion ([for review, see [193]). Plateau potentials in motoneurons are a latent property; 5-HT facilitates their expression [84, 167, 164 and 165]. Plateau potentials are generated by calcium influx through L-type calcium channels, which are normally curtailed by several potassium conductances [168 and 171]. Serotonergic facilitation of plateau properties has been linked to suppression of the calcium-dependent potassium current responsible for AHPs [168], although other mechanisms including direct 5-HT facilitation of L-type calcium channels have not been excluded [350]. Similar plateau properties have been identified in a subpopulation of turtle ventral horn interneurons [170] and dorsal horn neurons [313]. However, 5-HT is not required for plateau expression in these two groups of interneurons. In some dorsal horn neurons the plateaus are inactivated by a slow AHP, thus leading some cells to develop voltage oscillations. Russo and Hounsgaard speculated that such cells were good candidates for participation in a central pattern generator [313]. 5-HT facilitates plateau potentials in turtle motoneurons when applied at the soma but not when applied in the dendritic field [337]. Brownstone and colleagues recently provided evidence of a parallel postnatal development of motoneuronal calcium channels, bistable membrane properties, and locomotor maturation in the mouse [54, 55, 186 and 187]. Similar to the turtle, L-type calcium channel currents underlie the production of plateau potentials in this preparation [53]. However, it remains to be determined whether 5-HT modulates plateau potentials in the mouse preparation. 5-HT modulates spike afterhyperpolarization In addition to the possible role of AHP reduction in promoting plateau potentials, we observed that AHPs are suppressed in motoneurons during locomotion in the cat [50] and in vitro rat [322]. This may be important in the regulation of cell firing during locomotion. In the neonatal rat, activation of 5-HT1A receptors inhibits high voltage-activated calcium channels on hypoglossal (but not spinal) motoneurons, which in turn leads to AHP suppression [29]. 5-HT also inhibits AHPs in guinea pig trigeminal motoneurons [173]. Although it has been shown that iontophoretically applied 5-HT reduces AHPs in cat spinal motoneurons [385], it remains to be determined whether AHP modulation in spinal neurons during locomotion in the cat and rat is indeed 5-HT-dependent. On the other hand, Grillner and colleagues provided evidence that AHP modulation by 5-HT1A-like receptors can explain the modulatory influence of 5-HT on locomotor rhythm, spike frequency, and burst termination in the lamprey [104, 245, 364, 374, 375 and 389]. However, apamin-induced suppression of AHPs has little effect on swimming suggesting additional mechanisms of 5-HT action must be important in the lamprey [254]. For instance, recently it was shown that 5-HT (and substance P) has an activity-dependent metaplastic influence on synaptic transmission in the lamprey locomotor network, consistent with the effect of 5-HT (and substance P) on locomotor rhythm [288]. 5-HT modulates the voltage-sensitivity of NMDA receptor channels We hypothesized that 5-HT may exert some of its influence on mammalian locomotor networks via an interaction with NMDA receptors for several reasons. First, both NMDA [62, 90, 97, 114, 208, 339, 340 and 383] and 5-HT [62 and 230] receptors have a critical role in the operation of the locomotor network (although see [30]). Second, combined bath application of NMDA and 5-HT produces a more stable locomotor pattern than either substance alone [79, 80, 204 and 343]. Third, both 5-HT and NMDA receptor activation was required for the appearance of intrinsic voltage oscillations in spinal neurons in the embryonic amphibian cord [331 and 332]. Fourth, we demonstrated earlier that mammalian spinal motoneurons [155, 156 and 233] and interneurons [154] generate intrinsic voltage oscillations in the presence of NMDA. Fifth, colocalization of 5-HT and glutamate-like immunoreactivity in synaptic boutons surrounding motoneuron somata suggests there may be an interplay of these receptor systems [280]. Finally, multiple actions of 5-HT on neuronal responses to excitatory amino acids have been described throughout the central nervous system (e.g., [110, 266, 274, 275, 304, 309 and 359]). We first examined whether 5-HT modulates the voltage-sensitive conductance associated with NMDA receptor activation [246]. This non-linear property, derived from the voltage-dependent blockade of the NMDA receptor channel by Mg2+ ions [247 and 283], produces a region of negative slope conductance (RNSC) in the current-voltage (I-V) relationship of the cell [119 and 229] and contributes to the generation of voltage oscillations in synaptically isolated rat [154, 155, 231 and 323], lamprey [376], and amphibian [331 and 332] spinal neurons. We found that 5-HT receptor antagonists block network rhythmic activity and abolish TTX-resistant NMDA receptor mediated voltage oscillations [230]. Conversely, 5-HT application facilitates the expression of NMDA-receptor-mediated rhythmic voltage oscillations in Xenopus [298, 308, 326, 331 and 332] and rat [230] motoneurons. However, unlike rat sympathetic preganglionic neurons [295], 5-HT alone did not induce voltage oscillations. Thus, given the appropriate neurochemical milieu, which includes 5-HT, at least some neurons are able to express membrane properties well-suited to the needs of a rhythmogenic network. We also examined whether the central role of NMDA receptors in locomotion was specifically linked to the production of a RNSC in the current-voltage relationship of neurons, or whether NMDA receptor activation simply supported overall excitatory synaptic transmission (i.e., in conjunction with non-NMDA excitatory amino acid receptors) within the network, independent of any requirement to generate a RNSC. When Mg2+ is removed from the bath, thus abolishing the RNSC while still allowing for NMDA receptor activation, the locomotor pattern became poorly organized [231]. This was also observed in the lamprey [48 and 358] and Xenopus [342]. In TTX-treated preparations, 5-HT application promoted the appearance of a RNSC in the I-V relationship of healthy motoneurons that failed to develop a RNSC in NMDA alone [232]. In motoneurons that did display a RNSC in response to NMDA alone, 5-HT caused a negative shift (i.e., hyperpolarizing direction) of the RNSC, whereas the 5-HT antagonist mianserin reversed this effect. The negative shift of the RNSC by 5-HT was simulated by lowering the concentration of Mg2+ in the bath [232]. 5-HT2 receptor actions are mediated through the protein kinase C pathway (PKC) [242], and PKC produces a negative shift of the NMDA receptor-dependent RNSC in conjunction with decreased Mg2+ blockade of the NMDA channel [37 and 65]. Thus, the influence of 5-HT on the RNSC may be due to a decrease in the Mg2+ block of the NMDA ionophore. One powerful form of 5-HT modulation, shown to be present in some dorsal horn neurons in the rat, is the 5-HT-induced transformation of silent glutamatergic synapses into functional synapses [223]. Silent synapses are NMDA receptor-mediated excitatory postsynaptic currents detected experimentally by depolarization of the cell from -70 mV (absent at this potential) to a more depolarized level (e.g., +40 mV). Application of 5-HT in low concentration (5 M) acts via activation of a G-protein-coupled 5-HT receptor to allow previously silent synaptic events to appear at the clamped voltage of -70 mV. Silent glutamatergic synapses were more commonly detected in the neonatal rat dorsal horn compared with older animals [223]. At this point, it remains to be determined whether 5-HT exerts any of its effect on locomotor circuitry through activation of silent synapses. Other effects of 5-HT on membrane properties In addition to the currents described above, 5-HT modulates a variety of other membrane currents in vertebrate spinal neurons. Depending on the neuron population and preparation under study, these include enhancement of Ih, low voltage-activated calcium, and persistent inward currents, and inhibition of leak potassium, high voltage-activated calcium, and calcium-dependent potassium currents [29, 34, 52, 172, 173, 214, 215, 216, 316, 348, 351, 376 and 380]. In the dorsal horn of the rat cord, 5-HT2 receptor activation enhances glycine and GABAA-induced currents via second messenger systems [391 and 392]. Unfortunately, information on the relevance of these currents to the operation of the locomotor network is largely unavailable. However, there is some evidence that Ih may contribute to the firing pattern of motoneurons during locomotion in the neonatal rat spinal cord [35]. In addition, and of particular interest for future exploration, are the persistent sodium and high voltage-activated calcium currents. These currents are enhanced, either directly or indirectly, by 5-HT and have been implicated in the development of bistable membrane behaviours in guinea pig trigeminal motoneurons [172]. It will also be of interest to learn whether 5-HT modulates presynaptic transmission within locomotor network. In the Xenopus, 5-HT and noradrenaline have been shown to have opposing presynaptic effects on glycine release from inhibitory neurons involved in the swimming circuitry [252]. Role of 5-HT in regeneration and recovery of motor function The observation that 5-HT modulates, and in some cases initiates, locomotion in spinalized animals suggests that therapeutic interventions aimed at augmenting 5-HT levels in the spinal cord, caudal to a lesion, may have a beneficial effect on recovery of locomotor function. One experimental approach using this strategy has been to transplant 5-HT-containing neurons from the embryonic brainstem into the spinal cord of adult rats, below the level of the transection or compression injury [116, 123, 134, 285, 299, 300, 305 and 307]. As in the intact animal, synaptic contacts are established in the ventral horn and intermediolateral cell column, while nonsynaptic contacts predominate in the dorsal horn [301 and 305]. Privat and colleagues showed that rats transected at the T8/9 level and subsequently injected with embryonic 5-HT-containing neurons at the T11¯13 level had improved recovery of hindlimb locomotor activity and this activity was further facilitated by the 5-HT re-uptake inhibitor zimelidine [116, 133, 134 and 393]. Instead of injecting 5-HT-containing cells, other studies have transplanted fetal spinal cord tissue (which contains virtually no serotonergic cells) into the site of a spinal cord transection in the adult rat. This technique also facilitates recovery of locomotor function [47, 200, 209 and 355]. Recently it has been shown that systemic administration of 5-HT agonists further enhances locomotor function in preparations with fetal spinal cord grafts, but not the non-transplanted group of lesioned animals [200 and 355]. Kim and colleagues concluded that the effect of 5-HT observed in transplanted animals depends on remodeling of spinal circuitry by the graft [200]. They also noted that 5-HT re-uptake inhibitors failed to improve locomotor function suggesting that endogenous spinal 5-HT likely did not mediated the improved recovery observed in animals receiving spinal tissue transplants [200]. 5-HT may have a biphasic influence on spinal cord recovery after trauma. 5-HT levels increase acutely in and around the site of spinal cord injury, presumably as a result of release from platelets [319]. This release of 5-HT is postulated to contribute to local adverse effects on the injured tissue [318]. A single dose of the 5-HT2 receptor antagonist mianserin given at the time of the injury improves recovery outcomes [319]. On the other hand, 5-HT may facilitate rather than impede recovery, once the acute injury phase is over. Thus, administration of mianserin to rats during the days following thoracic cord hemisection transiently worsened locomotor scores measured between the 2nd and 4th weeks post-transection, suggesting a beneficial influence of 5-HT in the recovery process [320]. However, the study by Hashimoto and Fukuda raised questions about the apparent beneficial effects of 5-HT on recovery from spinal cord injury [147]. They observed that 14 days after a moderately severe cord compression injury in the rat, the level of 5-HT decline in the lumbar cord correlated with worsening of the neurological score. Similar results in response to compression injury have been documented using immunohistochemical analysis of 5-HT changes [112]. Further reduction of the 5-HT level by injection of 5,7 dihydroxytryptamine 24 h after the injury resulted in further decline of the neurological score [147]. However, pCPA had no influence on the score despite lowering 5-HT concentrations to levels similar to those achieved using 5,7 dihydroxytryptamine. 5,7-Dihydroxytryptamine destroys 5-HT containing neurons; these neurons also contain other substances, such as thyroid releasing hormone (TRH) and substance P. pCPA selectively depletes 5-HT alone by inhibition of tryptophan hydroxylase. Therefore, Hashimoto and Fukuda concluded that 5-HT does not influence recovery. Instead, they proposed that other substances, co-localized in 5-HT neurons and their projections, may be important [147]. Saruhashi and colleagues examined the recovery of locomotor function in rats with T8 spinal cord hemisections in relation to the reappearance of 5-HT fibers and terminals in the lumbosacral cord [321]. After the hemisection, 5-HT fibers were markedly reduced in the ipsilateral, but not contralateral, lumbosacral cord and locomotor scores were substantially impaired in both hindlimbs, worse on the lesioned side. Over the next 2¯3 weeks 5-HT fibers crossed the midline in both the thoracic and lumbosacral cord. Locomotor recovery coincided with the establishment of 5-HT fiber density that reached 20% of control values or higher on the lesioned side. As expected, the capacity for 5-HT fiber regeneration is greater in the neonatal animal than in the adult. One to 12 months after a mid-thoracic hemisection of the spinal cord, the 5-HT content in the ventral horn of the ipsilateral lumbar cord is reduced to 8% and 43% of the levels measured on the intact side in the adult and neonatal rat, respectively [46]. Transplantation of fetal spinal cord into the hemisection lesion site in neonatal animals results in a 5-HT level that is 83% of control values [46]. In contrast to the general facilitatory effects of 5-HT on locomotor function in animal models, Barbeau and colleagues have shown that 5-HT receptor antagonists, not agonists, facilitate locomotor function in humans with chronic spinal cord injury [26, 23, 125, 372 and 373]. In particular, they demonstrated that 5-HT receptor antagonist cyproheptadine facilitates locomotion by reducing spasticity that interferes with stepping. They proposed that the spasticity is associated with 5-HT receptor denervation supersensitivity, which increases motoneuronal and segmental reflex excitability. The number of 5-HT1A receptors increases during the first month after spinal cord transection in the cat and then returns to normal, despite pharmacological evidence of continued denervation hypersensitivity beyond the first month [135]. It was hypothesized that supersensitivity in the presence of normal numbers of receptors in the chronic state may be due to the development of a more efficient rate of operation of the receptor and its effector complex [135]. Summary This review has outlined the complex and sometimes competing array of effects 5-HT has on spinal cord reflexes and locomotor function. However, a consistent theme is that 5-HT helps shape the pattern of motor output. This observation is consistent with the hypothesis forwarded by Jacobs and Fornal, which states that the primary function of 5-HT is to facilitate motor output and inhibit sensory input [179]. Hopefully, further research and a more detailed understanding of the role of 5-HT and other monoaminergic systems, with special attention to receptor subtypes and their regional distribution, will ultimately contribute to the development of rationale therapeutic strategies aimed at restoring motor function after spinal cord injury. Acknowledgements We wish to thank Drs. Brent Fedirchuk and Doreen Fyda for their assistance in providing the data shown in Fig. 4, and Dr. Susan Shefchyk for her valuable comments. This work was supported by the Medical Research Council of Canada. References 1. N.C. Aggelopoulos, M.J. Burton, R.W. Clarke and S.A. Edgley, Characterization of a descending system that enables crossed group II inhibitory reflex pathways in the cat spinal cord. J. Neurosci. 16 (1996), pp. 723¯729. Abstract-EMBASE | Abstract | Abstract-Elsevier BIOBASE | $Order Document 2. H. Ahlman, S. Grillner and M. Udo, The effect of 5-HTP on the static fusimotor activity and tonic stretch reflex of an extensor muscle. Brain Res. 27 (1971), pp. 393¯396. 3. H. Akagi, S. Konishi, M. Otsuka and M. Yanagisawa, The role of substance P as a neurotranmsitter in the reflexes of slow time courses in the neonatal rat spinal cord. Br. J. Pharmacol. 84 (1985), pp. 663¯673. Abstract-EMBASE | $Order Document 4. F.J. Alvarez, D. Pearson, D. Harrington, D. Dewey, L. Torbeck and R.E.W. Fyffe, Distribution of 5-hydroxytryptamine-immunoreactive boutons on -motoneurons in the lumbar spinal cord of adult cats. J. Comp. Neurol. 393 (1998), pp. 69¯83. Abstract-EMBASE | Abstract | $Order Document | Full-text via CrossRef 5. N.E. Anden, A. Carlsson, N.A. Hillarp and T. Magnusson, 5-Hydroxytryptamine release by nerve stimulation of the spinal cord. Life Sci. 3 (1964), pp. 473¯478. 6. N.E. Anden, A. Carlsson, N.A. Hillarp and T. Magnusson, Noradrenaline release by nerve stimulation of the spinal cord. Life Sci. 4 (1965), pp. 129¯132. 7. N.E. Anden, T. Haggendal, T. Magnusson and E. Rosengren, The time course of the disappearance of noradrenaline and 5-hydroxytryptamine in the spinal cord after transection. Acta Physiol. Scand. 62 (1964), pp. 115¯118. 8. N.E. Anden, M.G.M. Jukes and A. Lundberg, Spinal reflexes and monoamine liberation. Nature 202 (1964), pp. 1222¯1223. 9. N.E. Anden, A. Lundberg, E. Rosengren and L. Vyklicky, The effect of DOPA on spinal reflexes from the FRA (flexor reflex afferents). Experentia 19 (1963), pp. 654¯655. 10. E.G. Anderson, Bulbospinal serotonin-containing neurons and motor control. Fed. Proc. 31 (1972), pp. 107¯112. 11. E.G. Anderson and T. Shibuya, The effect of 5-hydroxytryptophan and l-tryptophan on spinal synaptic activity. J. Pharm. Exp Ther. 153 (1966), pp. 352¯360. 12. U. Arvidsson, S. Cullheim, B. Ulfhake, G.W. Bennett, K.C.F. Fone, A.C. Cuello, A.A.J. Verhofstad, T.J. Visser and T. Hokfelt, 5-Hydroxytryptamine, substance P, and thyrotropin-releasing hormone in the adult cat spinal cord segment L7: Immunohistochemical and chemical studies. Synapse 6 (1990), pp. 237¯270. Abstract-EMBASE | Abstract-BIOTECHNOBASE | $Order Document 13. U. Arvidsson, B. Ulfhake, S. Cullheim, A. Bergstrand, E. Theodorson and T. Hokfelt, Distribution of 125I-galanin binding sites, immunoreactive galanin, and its coexistence with 5-hydroxytryptamine in the cat spinal cord: Biochemical, histochemical, and experimental studies at the light and electron microscopic level. J. Comp. Neurol. 308 (1991), pp. 115¯138. Abstract-EMBASE | Abstract-BIOTECHNOBASE | $Order Document 14. T. Arya, S. Bajwa and S.A. Edgley, Crossed reflex actions from group II muscle afferents in the lumbar spinal cord of the anesthetized cat. J. Physiol. 444 (1991), pp. 117¯131. Abstract-EMBASE | $Order Document 15. Y. Atsuta, E. Garcia-Rill and R.D. Skinner, Electrically induced locomotion in the in vitro brainstem-spinal cord preparation. Dev. Brain Res. 42 (1988), pp. 309¯312. Abstract-EMBASE | $Order Document 16. Y. Atsuta, E. Garcia-Rill and R.D. Skinner, Characteristics of electrically induced locomotion in rat in vitro brain stem-spinal cord preparation. J. Neurophysiol. 64 (1990), pp. 727¯735. Abstract-EMBASE | $Order Document 17. N.R. Banna and E.G. Anderson, The effect of 5-hydroxytryptamine antagonists on spinal neuronal activity. J. Pharm. Exp. Ther. 162 (1968), pp. 319¯325. 18. S. Barasi and M.H.T. Roberts, The modification of lumbar motoneurone excitability by stimulation of a putative 5-hysdroxytryptamine pathway. Br. J. Pharmacol. 52 (1974), pp. 339¯348. Abstract-EMBASE | $Order Document 19. H. Barbeau and P. Bedard, Similar motor effects of 5-HT and TRH in rats following chronic spinal transection and 5,7-dihyroxytryptamine injection. Neuropharmacology 20 (1981), pp. 477¯481. Abstract-EMBASE | $Order Document 20. H. Barbeau and P. Bedard, Denervation supersensitivity to 5-hydroxytryptamine in rats following spinal transection and 5,7-dihydroxytryptamine injection. Neuropharmacology 20 (1981), pp. 611¯616. Abstract-EMBASE | $Order Document 21. H. Barbeau, M. Filion and P. Bedard, Effects of agonists and antagonists of serotonin on spontaneous hindlimb EMG activity in chronic spinal rats. Neuropharmacology 20 (1981), pp. 99¯107. Abstract-EMBASE | $Order Document 22. H. Barbeau, C. Julien and S. Rossignol, The effects of clonidine and yohimbine on locomotion and cutaneous reflexes in the adult chronic spinal cat. Brain Res. 437 (1987), pp. 83¯96. Abstract-EMBASE | $Order Document 23. H. Barbeau, C.L. Richards and P. Bedard, Action of cyproheptadine in spastic paraparetic patients. J. Neurol. Neurosurg. Psychiatry 45 (1982), pp. 923¯926. Abstract-EMBASE | $Order Document 24. H. Barbeau and S. Rossignol, The effects of serotonergic drugs on the locomotor pattern and on cutaneous reflexes of the adult chronic spinal cat. Brain Res. 514 (1990), pp. 55¯67. Abstract-EMBASE | $Order Document 25. H. Barbeau and S. Rossignol, Initiation and modulation of the locomotor pattern in the adult chronic spinal cat by noradrenergic, serotonergic and dopaminergic drugs. Brain Res. 546 (1991), pp. 250¯260. Abstract-EMBASE | $Order Document 26. H. Barbeau and S. Rossignol, Enhancement of locomotor recovery following spinal cord injury. Curr. Opin. Neurol. 7 (1994), pp. 517¯524. Abstract-EMBASE | $Order Document 27. N.M. Barnes and T. Sharp, A review of central 5-HT receptors and their function. Neuropharmacology 38 (1999), pp. 1083¯1152. SummaryPlus | Article | Journal Format-PDF (4453 K) 28. A.I. Basbaum, C.H. Clanton and H.L. Fields, Three bulbospinal pathways from the rostral medulla of the cat: An autoradiographic study of pain modulating systems. J. Comp. Neurol. 178 (1978), pp. 209¯224. Abstract-EMBASE | $Order Document 29. D.A. Bayliss, M. Umemiya and A.J. Berger, Inhibition of N- and P-type calcium channels and the after-hyperpolarization in rat motoneurones by serotonin. J. Physiol. 485 (1995), pp. 635¯647. Abstract-EMBASE | Abstract | Abstract-Elsevier BIOBASE | $Order Document 30. M. Beato, E. Bracci and A. Nistri, Contribution of NMDA and non-NMDA glutamate receptors to locomotor pattern generation in the neonatal rat spinal cord. Proc. R. Soc. Lond. B 264 (1997), pp. 877¯884. Abstract-EMBASE | Abstract | $Order Document 31. M. Beato and A. Nistri, Serotonin-induced inhibition of locomotor rhythm of the rat isolated spinal cord is mediated by the 5-HT1 receptor class. Proc. R. Soc. Lond. B 265 (1998), pp. 2073¯2080. Abstract-EMBASE | Abstract | $Order Document | Full-text via CrossRef 32. P. Bedard, H. Barbeau, G. Barbeau and M. Filion, Progressive increase of motor activity induced by 5-HTP in the rat below a complete section of the spinal cord. Brain Res. 169 (1979), pp. 393¯397. Abstract-EMBASE | $Order Document 33. M.F. Belin, D. Nanopoulos, D. Didier, M. Aguera, H. Steinbusch, A. Verhofstad, M. Maitre and J.F. Pujol, Immunohistochemical evidence for the presence of gamma-aminobutyric acid and serotonin in one nerve cell. A study of the raphe nuclei of the rat using antibodies to glutamate decarboxylase and serotonin. Brain Res. 275 (1983), pp. 329¯339. Abstract-EMBASE | $Order Document 34. A.J. Berger and T. Takahashi, Serotonin enhances a low-voltage-activated calcium current in rat spinal motoneurons. J. Neurosci. 10 (1990), pp. 1922¯1928. 35. S. Bertrand and J.-R. Cazalets, Postinhibitory rebound during locomotor-like activity in neonatal rat motoneurons in vitro. J. Neurophysiol. 79 (1998), pp. 342¯351. Abstract-EMBASE | Abstract | $Order Document 36. A. Bjorklund, B. Falck and V. Stenevi, On the possible existence of a new intraneuronal monoamine in the spinal cord of the rat. J. Pharmacol. Exp. Ther. 175 (1970), pp. 525¯532. 37. T. Blank, R. Zwart, I. Nijholt and J. Spiess, Serotonin 5-HT2 receptor activation potentiates NMDA receptor-mediated ion currents by protein kinase C-dependent mechanism. J. Neurosci. Res. 45 (1996), pp. 153¯160. Abstract-EMBASE | Abstract-Elsevier BIOBASE | $Order Document | Full-text via CrossRef 38. R.M. Bowker and L.C. Abbott, Quantitative re-evaluation of descending serotonergic and non-serotonergic projections from the medulla of the rodent: Evidence for extensive co-existence of serotonin and peptides in the same spinally projecting neurons, but not from the nucleus raphe magnus. Brain Res. 512 (1990), pp. 15¯25. Abstract-EMBASE | $Order Document 39. R.M. Bowker, H.W.M. Steinbusch and J.D. Coulter, Serotonergic and peptidergic projections to the rat spinal cord demonstrated by a combined retrograde HRP histochemical and immunocytochemical staining method. Brain Res. 211 (1981), pp. 412¯417. Abstract-EMBASE | $Order Document 40. R.M. Bowker, K.N. Westlund, M.C. Sullivan and J.D. Coulter, Organization of descending serotonergic projections to the spinal cord. Prog. Brain Res. 57 (1982), pp. 239¯265. 41. R.M. Bowker, K.N. Westlund, M.C. Sullivan, J.F. Wilber and J.D. Coulter, Descending serotonergic, peptidergic, and cholinergic pathways from the raphe nuclei: A multiple transmitter complex. Brain Res. 288 (1983), pp. 33¯48. Abstract-EMBASE | $Order Document 42. E. Bracci, M. Beato and A. Nistri, Extracellular K+ induces locomotor-like patterns in the rat cord in vitro: Comparison with NMDA or 5-HT induced activity. J. Neurophysiol. 79 (1998), pp. 2643¯2652. Abstract-EMBASE | Abstract | Abstract-Elsevier BIOBASE | $Order Document 43. P. Branchereau, J. Rodriquez, I. Delvolve, D.N. Abrous, M. Le Moal and J.-M. Cabelguen, Serotonergic systems in the spinal cord of the amphibian urodele Pleurodeles waltl. J. Comp. Neurol. 419 (2000), pp. 49¯60. Abstract-Elsevier BIOBASE | Abstract-EMBASE | Abstract | $Order Document | Full-text via CrossRef 44. H. Bras, P. Cavallari, E. Jankowska and D. McCrea, Comparison of effects of monoamines on transmission in spinal pathways from group I and II muscle afferents in the cat. Exp. Brain Res. 76 (1989), pp. 27¯37. Abstract-EMBASE | $Order Document 45. H. Bras, E. Jankowska, B. Noga and B. Skoog, Comparison of effects of various types of NA and 5-HT agonists on transmission from group II muscle afferents in the cat. Eur. J. Neurosci. 2 (1990), pp. 1029¯1039. Abstract-EMBASE | $Order Document 46. B.S. Bregman, Development of serotonergic immunoreactivity in the rat spinal cord and its plasticity after neonatal spinal cord lesions. Dev. Brain Res. 34 (1987), pp. 245¯263. Abstract-EMBASE | $Order Document 47. B.S. Bregman, Recovery of function after spinal cord injury: Transplantation strategies. In: S.B. Dunnett and A. Bjorklund, Editors, Functional neural transplantation, Raven Press, New York (1994), pp. 489¯529. 48. L. Brodin and S. Grillner, Effects of magnesium on fictive locomotion induced by activation of N-methyl-D-aspartate (NMDA) receptors in the lamprey spinal cord in vitro. Brain Res. 380 (1986), pp. 244¯252. Abstract-EMBASE | $Order Document 49. L. Brown, J. Amedro, G. Williams and D. Smith, A pharmacological analysis of the rat spinal cord serotonin (5-HT) autoreceptor. Eur. J. Pharmacol. 145 (1988), pp. 163¯171. Abstract-EMBASE | $Order Document 50. R.M. Brownstone, L.M. Jordan, D.J. Kriellaars, B.R. Noga and S. Shefchyk, On the regulation of repetitive firing in lumbar motoneurones during fictive locomotion in the cat. Exp. Brain Res. 90 (1992), pp. 441¯455. Abstract-EMBASE | $Order Document 51. N.N. Budakova, Stepping movements in the spinal cat due to DOPA administration. Fiziol. Zh. SSSR 59 (1973), pp. 1190¯1198. Abstract-EMBASE | $Order Document 52. C.G. Cardenas, L.P. Del Mar and R.S. Scroggs, Variation in serotonergic inhibition of calcium channel currents in four types of rat sensory neurons differentiated by membrane properties. J. Neurophysiol. 74 (1995), pp. 1870¯1879. Abstract-EMBASE | Abstract | Abstract-Elsevier BIOBASE | $Order Document 53. Carlin, K. P. Voltage-gated calcium currents in mouse spinal motoneurons: Possible role in plateau potentials. Doctoral thesis, University of Manitoba, Winnipeg, Canada; 2000. 54. K.P. Carlin, Z. Jiang and R.M. Brownstone, Characterization of calcium currents in functionally mature mouse spinal motoneurons. Eur. J. Neurosci. 12 (2000), pp. 1624¯1634. Abstract-Elsevier BIOBASE | Abstract-EMBASE | Abstract | $Order Document | Full-text via CrossRef 55. K.P. Carlin, K.E. Jones, L.M. Jordan and R.M. Brownstone, Dendritic L-type calcium currents in mouse spinal motoneurons: Implications for bistability. Eur. J. Neurosci. 12 (2000), pp. 1635¯1646. Abstract-Elsevier BIOBASE | Abstract-EMBASE | Abstract | $Order Document | Full-text via CrossRef 56. A. Carlsson, B. Falk, K. Fuxe and N.A. Hillarp, Cellular localization of monoamines in the spinal cord. Acta Physiol. Scand. 60 (1964), pp. 112¯119. 57. A. Carlsson, T. Magnusson and E. Rosengren, 5-Hydroxytryptamine of the spinal cord normally and after transection. Experientia 19 (1963), p. 359. 58. E. Carstens, H. Gilly, H. Schreiber and M. Zimmermann, Effects of midbrain stimulation and iontophoretic application of serotonin, noradrenaline, morphine and GABA on electrical thresholds of afferent C- and A- fibre terminals in cat spinal cord. Neuroscience 21 (1987), pp. 395¯406. Abstract-EMBASE | $Order Document 59. J.-R. Cazalets, M. Borde and F. Clarac, Localization and organization of the central pattern generator for hindlimb locomotion in the newborn rat. J. Neurosci. 15 (1995), pp. 4943¯4951. Abstract-EMBASE | Abstract | Abstract-Elsevier BIOBASE | $Order Document 60. J.-R. Cazalets, M. Gardette and G. Hilaire, Locomotor network maturation is transiently delayed in the MAOA-deficient mouse. J. Neurophysiol. 83 (2000), pp. 2468¯2470. Abstract | Abstract-Elsevier BIOBASE | Abstract-EMBASE | $Order Document 61. J.R. Cazalets, P. Grillner, J. Menard and F. Clarac, Two types of motor rhythm induced by NMDA and amines in an in vitro spinal cord preparation of neonatal rat. Neurosci. Lett. 111 (1990), pp. 116¯121. Abstract-EMBASE | $Order Document 62. J.R. Cazalets, Y. Sqalli-Housaaini and F. Clarac, Activation of the central pattern generator for locomotion by serotonin and excitatory amino acids in neonatal rat. J. Physiol. 455 (1992), pp. 187¯204. Abstract-EMBASE | $Order Document 63. V. Chan-Palay, V.G. Jonsson and S.L. Palay, Serotonin and substance P coexist in neurons of the rat's central nervous system. Proc. Natl. Acad. Sci. USA 75 (1978), pp. 1582¯1586. Abstract-EMBASE | $Order Document 64. L. Chen, K. Hamaguchi, S. Hamada and N. Okado, Regional differences of serotonin-mediated synaptic plasticity in the chicken spinal cord with development and aging. J. Neural. Transplant. Plasticity 6 (1997), pp. 41¯48. Abstract-EMBASE | Abstract | $Order Document 65. L. Chen and L.M. Huang, Protein kinase C reduces Mg2+ block of NMDA-receptor channels as a mechanism of modulation. Nature 356 (1992), pp. 521¯523. Abstract-EMBASE | $Order Document 66. Christenson, J.; Franck, J.; Grillner, S. Increase in endogenous 5-hydroxytryptamine levels modulate the central network underlying locomotion in the lamprey spinal cord. Neurosci. Lett. 188¯192; 1989. 67. C. Cina and S. Hochman, Serotonin receptor pharmacology of the mammalian locomotor CPG: Activation by a 5-HT7 receptor agonist in the isolated rat spinal cord. Soc. Neurosci. Abstr. 25 (1998), p. 1669. 68. C. Cina and S. Hochman, Diffuse distribution of sulforhodamine-labeled neurons during serotonin-evoked locomotion in the neonatal rat thoracolumbar spinal cord. J. Comp. Neurol. 423 (2000), pp. 590¯602. Abstract | Abstract-Elsevier BIOBASE | $Order Document | Full-text via CrossRef 69. K.A. Clarke, A.J. Parker and G.C. Stirk, Potentiation of motoneurones excitability of 5-HT agonists and TRH analogue. Neuropeptides 6 (1985), pp. 269¯282. Abstract-EMBASE | $Order Document 70. R.W. Clarke, J. Harris and A.K. Houghton, Spinal 5-HT-receptors and tonic modulation of transmission through a withdrawal reflex pathway in the decerebrated rabbit. Br. J. Pharmacol. 119 (1996), pp. 1167¯1176. Abstract-EMBASE | Abstract | Abstract-Elsevier BIOBASE | $Order Document 71. B.V. Clineschmidt and E.G. Anderson, The blockade of bulbospinal inhibition by 5-hysdroxytryptamine antagonists. Exp Brain Res. 11 (1970), pp. 175¯186. 72. B.V. Clineschmidt, J.E. Pierce and W. Lovenberg, Tryptophan hydroxylase and serotonin in spinal cord and brain before and after chronic transection. J. Neurochem. 18 (1971), pp. 1593¯1596. 73. A.H. Cohen and P. Wallen, The neuronal correlates of locomotion in fish: "Fictive swimming" induced in an in vitro preparation of the lamprey spinal cord. Exp. Brain Res. 41 (1980), pp. 11¯18. Abstract-EMBASE | $Order Document 74. M.I. Colada, A. Arnedo, E. Peralta and J. Del Rio, Unilateral rhizotomy decreases monoamine levels in the rat spinal cord. Neurosci. Lett. 87 (1988), pp. 302¯306. 75. L.A. Connell and D.I. Wallis, Responses to 5-hydroxytryptamine evoked in the hemisected spinal cord of the neonate rat. Br. J. Pharmacol. 94 (1988), pp. 1101¯1114. Abstract-EMBASE | $Order Document 76. L.A. Connell and D.I. Wallis, 5-Hydroxytryptamine depolarizes neonatal rat motoneurones through a receptor unrelated to an identified binding site. Neuropharmacology 28 (1989), pp. 625¯634. Abstract-EMBASE | $Order Document 77. B.A. Conway, H. Hultborn, O. Kiehn and I. Mintz, Plateau potentials in -motoneurones induced by intravenous injection of L-DOPA and clonidine in the spinal cat. J. Physiol. 405 (1988), pp. 369¯384. Abstract-EMBASE | $Order Document 78. V. Cornea-Hebert, M. Riad, C. Wu, S.K. Singh and L. Descarries, Cellular and subcellular distribution of the serotonin 5-HT2A receptor in the central nervous system of adult rat.J. Comp. Neurol. 409 (1999), pp. 187¯209. Abstract-EMBASE | Abstract-Elsevier BIOBASE | Abstract | $Order Document | Full-text via CrossRef 79. K.C. Cowley and B.J. Schmidt, Some limitations of ventral root recordings for monitoring locomotion in the in vitro neonatal rat spinal cord preparation. Neurosci. Lett. 171 (1994), pp. 142¯146. Abstract-EMBASE | Abstract | Abstract-Elsevier BIOBASE | $Order Document 80. K.C. Cowley and B.J. Schmidt, A comparison of motor patterns induced by N-methyl-D-aspartate, acetylcholine and serotonin in the in vitro neonatal rat spinal cord. Neurosci. Lett. 171 (1994), pp. 147¯150. Abstract-EMBASE | Abstract | Abstract-Elsevier BIOBASE | $Order Document 81. K.C. Cowley and B.J. Schmidt, Regional distribution of the locomotor pattern-generating network in the neonatal rat spinal cord. J. Neurophysiol. 77 (1997), pp. 247¯259. Abstract-EMBASE | Abstract | $Order Document 82. H. Crick and D.I. Wallis, Inhibition of reflex responses of neonate lumbar spinal cord by 5-hydroxytryptamine. Br. J. Pharmacol. 103 (1991), pp. 1769¯1775. Abstract-EMBASE | $Order Document 83. H. Crick, N.A. Manuel and D.I. Wallis, A novel 5-HT receptor or a combination of 5-HT receptor subtypes may mediate depression of a spinal monosynaptic reflex in vitro. Neuropharmacology 33 (1994), pp. 897¯904. Abstract-EMBASE | Abstract | Abstract-Elsevier BIOBASE | $Order Document 84. C. Crone, H. Hultborn, O. Kiehn, L. Mazieres and H. Wigstrom, Maintained changes in motoneuronal excitability by short-lasting synaptic inputs in the decerebrate cat. J. Physiol. 405 (1988), pp. 321¯343. Abstract-EMBASE | $Order Document 85. D.R. Curtis, Action of 3-hydroxytryptamine and some tryptamine derivatives on spinal neurones. Nature 194 (1962), p. 292. 86. D.R. Curtis, J.C. Eccles and R.M. Eccles, Pharmacological studies on reflexes. Am. J. Physiol. 183 (1955), p. 606. 87. D.R. Curtis, J.W. Phillis and J.C. Watkins, Cholinergic and non-cholinergic transmission in the mammalian spinal cord. J. Physiol. 158 (1961), pp. 296¯323. 88. A. Dahlstrom and K. Fuxe, Evidence for the existence of monoamine neurons in the central nervous system. I. Demonstration of monoamines in cell bodies of brainstem neurons. Acta Physiol. Scand. 62 (1964), pp. 3¯55. 89. A. Dahlstrom and K. Fuxe, Evidence for the existence of monoamine neurons in the central nervous system. II. Experimentally induced changes in the intraneuronal amine levels of bulbospinal neurons systems. Acta Physiol. Scand. 64 (1965), pp. 1¯36. 90. N. Dale and A. Roberts, Excitatory amino acid receptors in Xenopus embryo spinal cord and their role in the activation of swimming. J. Physiol. 348 (1984), pp. 527¯543. Abstract-EMBASE | $Order Document 91. G. Daval, D. Verge, A.I. Basbaum, S. Bourgoin and H. Hamon, Autoradiographic evidence of serotonin1 binding sites on primary afferent fibres in the dorsal horn of the rat spinal cord. Neurosci. Lett. 83 (1987), pp. 71¯76. Abstract-EMBASE | $Order Document 92. G. Defenu and P. Mantegazzini, Effects of the 5-hydroxytryptamine on spinal reflexes of the cat. Psychopharmacology 10 (1966), pp. 96¯102. 93. T.G. Deliagina, G.N. Orlovsky and G.A. Pavlova, The capacity for generation of rhythmic oscillations is distributed in the lumbosacral spinal cord of the cat. Exp. Brain Res. 53 (1983), pp. 81¯90. Abstract-EMBASE | $Order Document 94. V. Dietz, K. Nakazawa, M. Wirz and Th. Erni, Level of spinal cord lesion determines locomotor activity in spinal man. Exp. Brain Res. 128 (1999), pp. 405¯409. Abstract-EMBASE | Abstract | Abstract-Elsevier BIOBASE | $Order Document | Full-text via CrossRef 95. E. Di-Pasquale, A. Lindsay, J. Feldman, R. Monteau and G. Hilaire, Serotonergic inhibition of phrenic motoneuron activity: An in vitro study in neonatal rat. Neurosci. Lett. 230 (1997), pp. 29¯32. SummaryPlus | Article | Journal Format-PDF (166 K) 96. F.J. DiTirro, G.F. Martin and R.H. Ho, A developmental study of substance-P, somatostatin, enkephalin, and serotonin immunoreactive elements in the spinal cord of the North American opossum. J. Comp. Neurol. 213 (1983), pp. 241¯261. Abstract-EMBASE | $Order Document 97. J.R. Douglas, B.R. Noga, X. Dai and L.M. Jordan, The effects of intrathecal administration of excitatory amino acid agonists on the initiation of locomotion in the adult cat. J. Neurosci. 13 (1993), pp. 990¯1000. Abstract-EMBASE | $Order Document 98. A. Dumoulin, A. Privat and M. Gimenez y Ribotta, Transplantation of embryonic raphe cells regulates the modifications of the GABAergic phenotype occurring in the injured spinal cord. Neuroscience 95 (2000), pp. 173¯182. 99. V.R. Edgerton, R.D. de Leon, N. Tillakaratne, M.R. Recktenwald, J.A. Hodgson and R.R. Roy, Use-dependent plasticity in spinal stepping and standing. In: F.J. Seil, Editor, Neuronal regeneration, reorganization, and repair. Adv. Neurol. 72, Lippincott-Raven Press, Philadelphia (1997), pp. 233¯247. 100. S.A. Edgley and E. Jankowska, An interneuronal relay from group I and II muscle afferents in the midlumbar segments of the cat spinal cord. J. Physiol. 389 (1987), pp. 675¯690. 101. S.A. Edgley, E. Jankowska and S. Shefchyk, Evidence that interneurones in reflex pathways from group II afferents are involved in locomotion in the cat. J. Physiol. 403 (1988), pp. 57¯73. 102. Eken, T.; Hultborn, H.; Kiehn, O. Possible functions of transmitter-controlled plateau potentials in -motoneurones. Prog. Brain Res. 257¯267; 1989. 103. T. Eken and O. Kiehn, Bistable firing properties of soleus motor units in unrestrained rats. Acta Physiol. Scand. 136 (1989), pp. 383¯394. Abstract-EMBASE | $Order Document 104. A. El Manir, J. Tegner and S. Grillner, Calcium-dependent potassium channels play a critical role for burst termination in the locomotor network in lamprey. J. Neurophysiol. 72 (1994), pp. 1852¯1861. 105. A. El Manira, W. Zhang, E. Svensson and N. Bussieres, 5-HT inhibits calcium current and synaptic transmission from sensory neurons in lamprey. J. Neurosci. 17 (1997), pp. 1786¯1794. 106. P.H. Ellaway, J.E. Pascoe and J.R. Trott, On the descending 5-hydroxytryptaminergic pathway controlling the stretch reflex. J. Physiol. 230 (1973), pp. 17P¯18P. 107. P. Elliot and D.I. Wallis, The actions of 5-hydroxytryptamine on lumbar motoneurones in neonatal rat spinal cord in vitro. J. Physiol. 426 (1990), p. 54P. 108. P. Elliot and D.I. Wallis, Serotonin and L-norepinephrine as mediators of altered excitability in neonatal rat motoneurons studied in vitro. Neuroscience 47 (1992), pp. 533¯544. 109. P. Elliott and D.I. Wallis, Glutamatergic and non-glutamatergic responses evoked in neonatal rat lumbar motoneurons on stimulation of the lateroventral spinal cord surface. Neuroscience 56 (1993), pp. 189¯197. Abstract | Abstract-EMBASE | $Order Document 110. I. Engberg, J.A. Flatman and J.D.C. Lambert, Bistable behavior of spinal cord motoneurones during amino-acid excitation. Acta Physiol. Scand. 121 (1984), p. 6A. 111. I. Engberg and R.W. Ryall, The inhibitory action of noradrenaline and other monoamines on spinal neurones. J. Physiol. 185 (1966), pp. 298¯322. 112. A.I. Faden, A. Gannon and A.I. Basbaum, Use of serotonin immunocytochemistry as a marker of injury severity after experimental spinal trauma in rats. Brain Res. 450 (1988), pp. 94¯100. Abstract-EMBASE | Abstract-BIOTECHNOBASE | $Order Document 113. B. Fedirchuk, J. Nielsen, N. Petersen and H. Hultborn, Pharmacologically evoked fictive motor patterns in the acutely spinalized marmoset monkey (Callithrix jacchus). Exp. Brain Res. 122 (1998), pp. 351¯361. Abstract-EMBASE | Abstract | $Order Document | Full-text via CrossRef 114. F. Fenaux, M. Corio, R. Palisses and D. Viala, Effects of an NMDA-receptor antagonist, MK-801, on central locomotor programming in the rabbit. Exp. Brain Res. 86 (1991), pp. 393¯401. Abstract-EMBASE | $Order Document 115. D. Feraboli-Lohnherr, J.-Y. Barthe and D. Orsal, Serotonin-induced activation of the network for locomotion in adult spinal rats. J. Neurosci. Res. 55 (1999), pp. 87¯98. Abstract-EMBASE | $Order Document | Full-text via CrossRef 116. D. Feraboli-Lohnherr, D. Orsal, A. Yakovleff, M. Gimenez y Ribotta and A. Privat, Recovery of locomotor activity in the adult chronic spinal rat after sublesional transplantation of embryonic nervous cells: Specific role of serotonergic neurons. Exp. Brain Res. 113 (1997), pp. 443¯454. Abstract-EMBASE | Abstract | $Order Document 117. C.T. Fischette, B. Nock and K. Renner, Effects of 5,7-dihydroxytryptamine on serotonin1 and serotonin2 receptors throughout the rat central nervous system using quantitative autoradiography. Brain Res. 421 (1987), pp. 263¯279. Abstract-EMBASE | $Order Document 118. M. Fitzgerald and M. Koltzenburg, The functional development of descending inhibitory pathways in the dorsolateral funiculus of the newborn rat spinal cord. Dev. Brain Res. 24 (1986), pp. 261¯