Antidromic discharges of dorsal root afferents in the neonatal rat

Physiological effects of cathodal electrode configuration for transspinal stimulation in humans

Transspinal stimulation modulates neuronal excitability and promotes recovery in upper motoneuron lesions. The recruitment input-output curves of transspinal evoked potentials (TEPs) recorded from knee and ankle muscles, and their susceptibility to spinal inhibition, were recorded when the position, size, and number of the cathode electrode were arranged in four settings or protocols (Ps). The four Ps were the following: 1) one rectangular electrode placed at midline (KNIKOU-LAB4Recovery or K-LAB4Recovery; P-KLAB), 2) one square electrode placed at midline (P-2), 3) two square electrodes 1 cm apart placed at midline (P-3), and 4) one square electrode placed on each paravertebral side (P-4). P-KLAB and P-3 required less current to reach TEP threshold or maximal amplitudes. A rightward shift in TEP recruitment curves was evident for P-4, whereas the slope was increased for P-2 and P-4 compared with P-KLAB and P-3. TEP depression upon single and paired transspinal stimuli was pronounced in ankle TEPs but was less prominent in knee TEPs. TEP depression induced by single transspinal stimuli at 1.0 Hz was similar for most TEPs across protocols, but TEP depression induced by paired transspinal stimuli was different between protocols and was replaced by facilitation at 100-ms interstimulus interval for P-4. Our results suggest that P-KLAB and P-3 are preferred based on excitability threshold of motoneurons. P-KLAB produced more TEP depression, thereby maximizing the engagement of spinal neuronal pathways. We recommend P-KLAB to study neurophysiological mechanisms underlying transspinal stimulation or when used as a neuromodulation method for recovery in neurological disorders.NEW & NOTEWORTHY Transspinal stimulation with a rectangular cathode electrode (P-KLAB) requires less current to produce transspinal evoked potentials and maximizes spinal inhibition. We recommend P-KLAB for neurophysiological studies or when used as a neuromodulation method to enhance motor output and normalize muscle tone in neurological disorders.

Activity-dependent decline and recovery of synaptic transmission in central parts of surviving primary afferents after their peripheral cut in crayfish

Axons deprived of their nucleus degenerate within a few days in mammals but survive for several months in crustaceans. However, it is not known whether central synapses from sensory axons may preserve their molecular machinery in the absence of spiking activity. To assess this, we used peripheral axotomy, which removes their nuclei combined with electrophysiology techniques and electron microscopy imaging. We report the following. (1) Electron microscopy analysis confirms previous observations that glial cell nuclei present in the sensory nerve proliferate and migrate to axon tubes, where they form close contacts with surviving axons. (2) After peripheral axotomy performed in vivo on the coxo-basipodite chordotonal organ (CBCO), the sensory nerve does not convey any sensory message, but antidromic volleys are observed. (3) Central synaptic transmission from the CBCO to motoneurons (MNs) progressively declines over 200 days (90% of monosynaptic excitatory transmission is lost after 3 weeks, whereas 60% of disynaptic inhibitory transmission persists up to 6 months). After 200 days, no transmission is observed. (4) However, this total loss is apparent only because repetitive electrical stimulation of the sensory nerve in vitro progressively restores first inhibitory post-synaptic potentials and then excitatory post-synaptic potentials. (5) The loss of synaptic transmission can be prevented by in vivo chronic sensory nerve stimulation. (6) Using simulations based on the geometric arrangements of synapses of the monosynaptic excitatory transmission and disynaptic inhibitory pathways, we show that antidromic activity in the CBCO nerve could play a role in the maintenance of synaptic function of inhibitory pathways to MNs, but not monosynaptic excitatory transmission to MNs. Our study confirms the deep changes in glial nuclei observed in axons deprived of their nucleus. We further show that the machinery for spike conduction and synaptic release persists for several months, even if there is no longer any activity. Indeed, we were able to restore, with electrical activity, spike conduction and synaptic function after long silent periods (>6 months).

Neonatal Mice Spinal Cord Interneurons Send Axons through the Dorsal Roots

Spontaneous interneuron activity plays a critical role in developing neuronal networks. Discharges conducted antidromically along the dorsal root (DR) precede those from the ventral root’s (VR) motoneurons. This work studied whether spinal interneurons project axons into the neonate’s dorsal roots. Experiments were carried out in postnatal Swiss-Webster mice. We utilized a staining technique and found that interneurons in the spinal cord’s dorsal horn send axons through the dorsal roots. In vitro electrophysiological recordings showed antidromic action potentials (dorsal root reflex; DRR) produced by depolarizing the primary afferent terminals. These reflexes appeared by stimulating the adjacent dorsal roots. We found that bicuculline reduced the DRR evoked by L5 dorsal root stimulation when recording from the L4 dorsal root. Simultaneously, the monosynaptic reflex (MR) in the L5 ventral root was not affected; nevertheless, a long-lasting after-discharge appeared. The addition of 2-amino-5 phosphonovaleric acid (AP5), an NMDA receptor antagonist, abolished the MR without changing the after-discharge. The absence of DRR and MR facilitated single action potentials in the dorsal and ventral roots that persisted even in low Ca2+ concentrations. The results suggest that firing interneurons could send their axons through the dorsal roots. These interneurons could activate motoneurons producing individual spikes recorded in the ventral roots. Identifying these interneurons and the persistence of their neuronal connectivity in adulthood remains to be established.

Antidromic Spike Propagation and Dissimilar Expression of P2X, 5-HT, and TRPV1 Channels in Peripheral vs. Central Sensory Axons in Meninges

Background: The terminal branches of the trigeminal nerve in meninges are supposed to be the origin site of migraine pain. The main function of these peripheral sensory axons is the initiation and propagation of spikes in the orthodromic direction to the second order neurons in the brainstem. The stimulation of the trigeminal ganglion induces the release of the neuropeptide CGRP in meninges suggesting the antidromic propagation of excitation in these fibers. However, the direct evidence on antidromic spike traveling in meningeal afferents is missing. Methods: By recording of spikes from peripheral or central parts of the trigeminal nerve in rat meninges, we explored their functional activity and tested the expression of ATP-, serotonin-, and capsaicin-gated receptors in the distal vs. proximal parts of these nerves. Results: We show the significant antidromic propagation of spontaneous spikes in meningeal nerves which was, however, less intense than the orthodromic nociceptive traffic due to higher number of active fibers in the latter. Application of ATP, serotonin and capsaicin induced a high frequency nociceptive firing in peripheral processes while, in central parts, only ATP and capsaicin were effective. Disconnection of nerve from trigeminal ganglion dramatically reduced the tonic antidromic activity and attenuated the excitatory action of ATP. Conclusion: Our data indicate the bidirectional nociceptive traffic and dissimilar expression of P2X, 5-HT and TRPV1 receptors in proximal vs. distal parts of meningeal afferents, which is important for understanding the peripheral mechanisms of migraine pain.

The M-current works in tandem with the persistent sodium current to set the speed of locomotion

The central pattern generator (CPG) for locomotion is a set of pacemaker neurons endowed with inherent bursting driven by the persistent sodium current (I_NaP). How they proceed to regulate the locomotor rhythm remained unknown. Here, in neonatal rodents, we identified a persistent potassium current critical in regulating pacemakers and locomotion speed. This current recapitulates features of the M-current (I_M): a subthreshold noninactivating outward current blocked by 10,10-bis(4-pyridinylmethyl)-9(10H)-anthracenone dihydrochloride (XE991) and enhanced by N-(2-chloro-5-pyrimidinyl)-3,4-difluorobenzamide (ICA73). Immunostaining and mutant mice highlight an important role of Kv7.2-containing channels in mediating I_M. Pharmacological modulation of I_M regulates the emergence and the frequency regime of both pacemaker and CPG activities and controls the speed of locomotion. Computational models captured these results and showed how an interplay between I_M and I_NaP endows the locomotor CPG with rhythmogenic properties. Overall, this study provides fundamental insights into how I_M and I_NaP work in tandem to set the speed of locomotion.

Afferent Input Induced by Rhythmic Limb Movement Modulates Spinal Neuronal Circuits in an Innovative Robotic In Vitro Preparation

Locomotor patterns are mainly modulated by afferent feedback, but its actual contribution to spinal network activity during continuous passive limb training is still unexplored. To unveil this issue, we devised a robotic in vitro setup (Bipedal Induced Kinetic Exercise, BIKE) to induce passive pedaling, while simultaneously recording low-noise ventral and dorsal root (VR and DR) potentials in isolated neonatal rat spinal cords with hindlimbs attached. As a result, BIKE evoked rhythmic afferent volleys from DRs, reminiscent of pedaling speed. During BIKE, spontaneous VR activity remained unchanged, while a DR rhythmic component paired the pedaling pace. Moreover, BIKE onset rarely elicited brief episodes of fictive locomotion (FL) and, when trains of electrical pulses were simultaneously applied to a DR, it increased the amplitude, but not the number, of FL cycles. When BIKE was switched off after a 30-min training, the number of electrically induced FL oscillations was transitorily facilitated, without affecting VR reflexes or DR potentials. However, 90 min of BIKE no longer facilitated FL, but strongly depressed area of VR reflexes and stably increased antidromic DR discharges. Patch clamp recordings from single motoneurons after 90-min sessions indicated an increased frequency of both fast- and slow-decaying synaptic input to motoneurons. In conclusion, hindlimb rhythmic and alternated pedaling for different durations affects distinct dorsal and ventral spinal networks by modulating excitatory and inhibitory input to motoneurons. These results suggest defining new parameters for effective neurorehabilitation that better exploits spinal circuit activity.

Origins of antidromic activity in sensory afferent fibers and neurogenic inflammation

Neurogenic inflammation results from the release of biologically active agents from the peripheral primary afferent terminal. This release reflects the presence of releasable pools of active product and depolarization-exocytotic coupling mechanisms in the distal afferent terminal and serves to alter the physiologic function of innervated organ systems ranging from the skin and meninges to muscle, bone, and viscera. Aside from direct stimulation, this biologically important release from the peripheral afferent terminal can be initiated by antidromic activity arising from five anatomically distinct points of origin: (i) afferent collaterals at the peripheral-target organ level, (ii) afferent collaterals arising proximal to the target organ, (iii) from mid-axon where afferents lacking myelin sheaths (C fibers and others following demyelinating injuries) may display crosstalk and respond to local irritation, (iv) the dorsal root ganglion itself, and (v) the central terminals of the afferent in the dorsal horn where local circuits and bulbospinal projections can initiate the so-called dorsal root reflexes, i.e., antidromic traffic in the sensory afferent.

Prochlorperazine Increases KCC2 Function and Reduces Spasticity after Spinal Cord Injury

In mature neurons, low intracellular chloride level required for inhibition is maintained by the potassium-chloride cotransporter, KCC2. Impairment of Cl- extrusion after KCC2 dysfunction has been involved in many central nervous system disorders, such as seizures, neuropathic pain, or spasticity, after a spinal cord injury (SCI). This makes KCC2 an appealing drug target for restoring Cl- homeostasis and inhibition in pathological conditions. In the present study, we screen the Prestwick Chemical Library® and identify conventional antipsychotics phenothiazine derivatives as enhancers of KCC2 activity. Among them, prochlorperazine hyperpolarizes the Cl- equilibrium potential in motoneurons of neonatal rats and restores the reciprocal inhibition post-SCI. The compound alleviates spasticity in chronic adult SCI rats with an efficacy equivalent to the antispastic agent, baclofen, and rescues the SCI-induced downregulation of KCC2 in motoneurons below the lesion. These pre-clinical data support prochlorperazine for a new therapeutic indication in the treatment of spasticity post-SCI and neurological disorders involving a KCC2 dysfunction.

Glucose is an adequate energy substrate for the depolarizing action of GABA and glycine in the neonatal rat spinal cord in vitro

In vitro studies have repeatedly demonstrated that the neurotransmitters γ-aminobutyric acid (GABA) and glycine depolarize immature neurons in many areas of the CNS, including the spinal cord. This widely accepted phenomenon was recently challenged by experiments showing that the depolarizing action of GABA on neonatal hippocampus and neocortex in vitro was prevented by adding energy substrates (ES), such as the ketone body metabolite dl-β-hydroxybutyric acid (DL-BHB), lactate, or pyruvate to the artificial cerebrospinal fluid (ACSF). It was suggested that GABA-induced depolarizations in vitro might be an artifact due to inadequate energy supply when glucose is the sole energy source, consistent with the energy metabolism of neonatal rat brain being largely dependent on ESs other than glucose. Here we examined the effects of these ESs (DL-BHB, lactate, pyruvate) on inhibitory postsynaptic potentials (IPSPs) recorded from neonatal rat lumbar spinal cord motoneurons (MNs), in vitro. We report that supplementing the ACSF with physiologic concentrations of DL-BHB, lactate, or pyruvate does not alter the reversal potential of IPSPs (E(IPSP)). Only high concentrations of pyruvate hyperpolarized E(IPSP). In addition, the depolarizing action of GABA on primary afferent terminals was not affected by supplementing the ACSF with ES at physiologic concentrations. We conclude that depolarizing IPSPs in immature MNs and the primary afferent depolarizations are not caused by inadequate energy supply. Glucose at its standard concentration appears to be an adequate ES for the neonatal spinal cord in vitro.

Chapter 1--importance of chloride homeostasis in the operation of rhythmic motor networks

GABA and glycine are classically called "inhibitory" amino acids, despite the fact that their action can rapidly switch from inhibition to excitation and vice versa. The postsynaptic action depends on the intracellular concentration of chloride ions ([Cl(-)](i)), which is regulated by proteins in the plasma membrane: the K(+)-Cl(-) cotransporter KCC2 and the Na(+)-K(+)-Cl(-) cotransporter NKCC1, which extrude and intrude Cl(-) ions, respectively. A high [Cl(-)](i) leads to a depolarizing (excitatory) action of GABA and glycine, as observed in mature dorsal root ganglion neurons and in motoneurons both early during development and in several pathological conditions, such as following spinal cord injury. Here, we review some recent data regarding chloride homeostasis in the spinal cord and its contribution to network operation involved in locomotion.

Presynaptic inhibition of primary afferents by depolarization: observations supporting nontraditional mechanisms

Primary afferent neurotransmission is the fundamental first step in the central processing of sensory stimuli and is controlled by pre- and postsynaptic inhibitory mechanisms. Presynaptic inhibition (PSI) is probably the more powerful form of inhibitory control in all primary afferent fibers. A major mechanism producing afferent PSI is via a channel-mediated depolarization of their intraspinal terminals, which can be recorded extracellularly as a dorsal root potential (DRP). Based on measures of DRP latency it has been inferred that this primary afferent depolarization (PAD) of low-threshold afferents is mediated by minimally trisynaptic pathways with pharmacologically identified GABAergic interneurons forming last-order axo-axonic synapses onto afferent terminals. There is still no "squeaky clean" evidence of this organization. This paper describes recent and historical work that supports the existence of PAD occurring by more direct pathways and with a complex pharmacology that questions the proprietary role of GABA and GABA(A) receptors in this process. Cholinergic transmission in particular may contribute significantly to PAD, including via direct release from primary afferents.

GABAergic and glycinergic interneuron expression during spinal cord development: dynamic interplay between inhibition and excitation in the control of ventral network outputs

A key objective of neuroscience research is to understand the processes leading to mature neural circuitries in the central nervous system (CNS) that enable the control of different behaviours. During development, network-constitutive neurons undergo dramatic rearrangements, involving their intrinsic properties, such as the blend of ion channels governing their firing activity, and their synaptic interactions. The spinal cord is no exception to this rule; in fact, in the ventral horn the maturation of motor networks into functional circuits is a complex process where several mechanisms cooperate to achieve the development of motor control. Elucidating such a process is crucial in identifying neurons more vulnerable to degenerative or traumatic diseases or in developing new strategies aimed at rebuilding damaged tissue. The focus of this review is on recent advances in understanding the spatio-temporal expression of the glycinergic/GABAergic system and on the contribution of this system to early network function and to motor pattern transformation along with spinal maturation. During antenatal development, the operation of mammalian spinal networks strongly depends on the activity of glycinergic/GABAergic neurons, whose action is often excitatory until shortly before birth when locomotor networks acquire the ability to generate alternating motor commands between flexor and extensor motor neurons. At this late stage of prenatal development, GABA-mediated excitation is replaced by synaptic inhibition mediated by glycine and/or GABA. At this stage of spinal maturation, the large majority of GABAergic neurons are located in the dorsal horn. We propose that elucidating the role of inhibitory systems in development will improve our knowledge on the processes regulating spinal cord maturation.

In Vivo Analysis of Proprioceptive Coding and Its Antidromic Modulation in the Freely Behaving Crayfish

Although sensory nerves in vitro are known to convey both orthodromic (sensory) and antidromic (putatively modulating) action potentials, in most cases very little is known about their bidirectional characteristics in intact animals. Here, we have investigated both the sensory coding properties and antidromic discharges that occur during real walking in the freely behaving crayfish. The activity of the sensory nerve innervating the proprioceptor CBCO, a chordotonal organ that monitors both angular movement and position of the coxo-basipodite (CB) joint, which is implicated in vertical leg movements, was recorded chronically along with the electromyographic activity of the muscles that control CB joint movements. Two wire electrodes placed on the sensory nerve were used to discriminate orthodromic from antidromic action potentials and thus allowed for analysis of both sensory coding and antidromic discharges. A distinction is proposed between 3 main classes of sensory neuron, according to their firing in relation to levator muscle activity during free walking. In parallel, we describe 2 types of antidromic activity: one produced exclusively during motor activity and a second produced both during and in the absence of motor activity. A negative correlation was found between the activity of sensory neurons in each of the 3 classes and identified antidromic discharges during walking. Finally, a state-dependent plasticity of CBCO nerve activity has been found by which the distribution of sensory orthodromic and antidromic activity changes with the physiological state of the biomechanical apparatus.

Characteristics of the electrical oscillations evoked by 4-aminopyridine on dorsal root fibers and their relation to fictive locomotor patterns in the rat spinal cord in vitro

4-Aminopyridine (4-AP) is suggested to improve symptomatology of spinal injury patients because it may facilitate neuromuscular transmission, spinal impulse flow and the operation of the locomotor central pattern generator (CPG). Since 4-AP can also induce repetitive discharges from dorsal root afferents, this phenomenon might interfere with sensory signals necessary to modulate CPG activity. Using electrophysiological recording from dorsal and ventral roots of the rat isolated spinal cord, we investigated 4-AP-evoked discharges and their relation with fictive locomotor patterns. On dorsal roots 4-AP (5-10 microM) induced sustained synchronous oscillations (3.3+/-0.8 s period) smaller than electrically evoked synaptic potentials, persistent after sectioning off the ventral region and preserved in an isolated dorsal quadrant, indicating their dorsal horn origin. 4-AP oscillations were blocked by tetrodotoxin, or 6-cyano-7-nitroquinoxaline-2,3-dione and d-amino-phosphonovalerate, or strychnine and bicuculline, suggesting they were network mediated via glutamatergic, glycinergic and GABAergic transmission. Isolated ventral horn areas could not generated 4-AP oscillations, although their intrinsic disinhibited bursting was accelerated by 4-AP. Thus, ventral horn areas contained 4-AP sensitive sites, yet lacked the network for 4-AP induced oscillations. Activation of fictive locomotion by either application of N-methyl-D-aspartate and serotonin or stimulus trains to a single dorsal root reversibly suppressed dorsal root oscillations induced by 4-AP. This suppression was due to depression of dorsal network activity rather than simple block of root discharges. Since dorsal root oscillations evoked by 4-AP were turned off when the fictive locomotor program was initiated, these discharges are unlikely to interfere with proprioceptive signals during locomotor training in spinal patients.

Development of posture and locomotion: an interplay of endogenously generated activities and neurotrophic actions by descending pathways

The adult pattern of locomotion is observed at the end of the second postnatal week in the rat. The in vitro spinal cord isolated from immature rats has served as a valuable preparation to study the mechanisms underlying the development of locomotion. Although the rat is unable to walk at birth, because of an immature posture, its spinal cord networks can generate at least two kinds of motor patterns in vitro. One activity is called 'fictive locomotion' because it shares several common features with locomotion observed in vivo. This fictive locomotor pattern is rarely observed spontaneously and its release requires either pharmacological or electrical stimulation of the spinal cord. A second endogenously generated activity observed in this preparation occurs spontaneously and exhibits phase relationships between motor outputs that are quite different from the fictive locomotor pattern. Here we review some of the developmental functions this spontaneous activity may subserve. It is likely a major trigger for the maturation of lumbar networks in the fetus, at a stage when inputs from both the periphery and supraspinal structures are weak. Pathways descending from the brainstem arrive in the lumbar enlargement during the last week in utero and the first two postnatal weeks. These pathways, through the neurotransmitters they contain, especially monoamines, are essential for the expression of some neuronal properties and may regulate several ongoing developmental processes.

Picrotoxin and bicuculline have different effects on lumbar spinal networks and motoneurons in the neonatal rat

Bicuculline is the most commonly used GABA(A) receptor antagonist to investigate the contribution of these receptors in motor control. However, this compound has been shown recently to potentiate the burst firing of neurons in various brain regions by blocking a calcium-activated potassium current underlying the spike after-hyperpolarization (AHP). This effect may distort our understanding of the role of GABA(A) receptors at the network level. In vitro brainstem-spinal cord preparations isolated from neonatal rats were used to compare the effects of bicuculline methiodide (bicuculline-M) and picrotoxin (PTX), another GABA(A) receptor antagonist, on the AHP of lumbar motoneurons as well as on spontaneous and locomotor-like motor activities. Intracellular recordings of lumbar motoneurons showed that bicuculline-M (20 microM) reduced the AHP to 57% of control whereas PTX (20-60 microM) had no significant effect. Bath-application of increasing concentrations of PTX caused an increase in spontaneous ventral root activity, which further increased significantly when bicuculline-M was added. The effects of both antagonists were tested on fictive locomotion. The left-right alternation was disrupted in the presence of bicuculline-M. A slow synchronous bursting activity of large amplitude also appeared in the presence of PTX. This slow rhythm was superimposed on a faster rhythm which still exhibited some degree of left-right alternation. These data demonstrate that bicuculline-M may not reveal accurately the contribution of GABA(A) receptors in motor control and the intrinsic properties of disinhibited networks.

Perinatal development of lumbar motoneurons and their inputs in the rat

The rat is quite immature at birth and a rapid maturation of motor behavior takes place during the first 2 postnatal weeks. Lumbar motoneurons undergo a rapid development during this period. The last week before birth represents the initial stages of motoneuron differentiation, including regulation of the number of cells and the arrival of segmental and first supraspinal afferents. At birth, motoneurons are electrically coupled and receive both appropriate and inappropriate connections from the periphery; the control from supraspinal structures is weak and exerted mainly through polysynaptic connections. During the 1st postnatal week, inappropriate sensori-motor contacts and electrical coupling disappear, the supraspinal control increases gradually and myelin formation is responsible for an increased conduction velocity in both descending and motor axons. Both N-methyl-D-aspartate (NMDA) and non-NMDA receptors are transiently overexpressed in the neonatal spinal cord. The contribution of non-NMDA receptors to excitatory amino acid transmission increases with age. Activation of gamma-aminobutyric acid(A) and glycine receptors leads to membrane depolarization in embryonic motoneurons but to hyperpolarization in older motoneurons. The firing properties of motoneurons change with development: they are capable of more repetitive firing at the end of the 1st postnatal week than before birth. However, maturation does not proceed simultaneously in the motor pools innervating antagonistic muscles; for instance, the development of repetitive firing of ankle extensor motoneurons lags behind that of flexor motoneurons. The spontaneous embryonic and neonatal network-driven activity, detected at the levels of motoneurons and primary afferent terminals, may play a role in neuronal maturation and in the formation and refinement of sensorimotor connections.

Formation of the central pattern generator for locomotion in the rat and mouse

It is well known that in the neonatal rat spinal cord preparation, alternating rhythmic bursts in the left and right ventral roots in a given lumbar segment can be induced by bath-application of N-methyl-D-aspartate or 5-hydroxytryptamine. Alternation between L2 and L5 ventral roots on the same side, representing the activity of flexor and extensor muscles, respectively, can be observed as well. In the prenatal period in the rat, alternation between the left and right ventral roots is established between embryonic day (E) 16.5 and E18.5. The alternation between the L2 and L5 ventral roots emerges at E20.5. Recent findings show that locomotor-like rhythmic activity with similar characteristics can be induced in the neonatal mouse preparation. In the lumbar spinal cord in the neonatal mouse, it is likely that the rhythm-generating network is distributed throughout the lumbar region with a rostro-caudal gradient, a situation similar to that in the neonatal and fetal rat spinal cord. With this review we hope to highlight the dramatic changes that neuronal networks generating locomotor-like activity undergo during the prenatal development of the rat. Moreover, the distribution of the neuronal network generating the locomotor rhythm in the neonatal rat and mouse spinal cord is compared.

Ubiquity of motor networks in the spinal cord of vertebrates

In a recent paper, we found that it is possible to record motor activity in sacral segments in the in vitro neonatal rat spinal cord preparation. This motor activity recorded in segments that are not innervating hindlimbs is driven by the lumbar locomotor network. Indeed, compartimentalizations of the cord with Vaseline walls or section experiments, reveals that the sacral segments possess their own rhythmogenic capabilities but that in an intact spinal cord they are driven by the lumbar locomotor network. In this review, these recent findings are placed in the context of spinal motor network interactions. As previously suspected, the motor networks do not operate in isolation but interact with each other according to behavioural needs. These interactions provide some insight into the discrepancies observed in several studies dealing with the localization of the lumbar locomotor network in the neonatal rat spinal cord. In conclusion, the spinal cord of quadrupeds appears as an heterogeneous structure where it is possible to identify neuronal networks that are crucial for the genesis of locomotor-related activities.