Locomotor pattern generation and descending control: a historical perspective.

The ability to generate and control locomotor movements depends on complex interactions between many areas of the nervous system, the musculoskeletal system, and the environment. How the nervous system manages to accomplish this task has been the subject of investigation for more than a century. In vertebrates, locomotion is generated by neural networks located in the spinal cord referred to as central pattern generators. Descending inputs from the brain stem initiate, maintain, and stop locomotion as well as control speed and direction. Sensory inputs adapt locomotor programs to the environmental conditions. This review presents a comparative and historical overview of some of the neural mechanisms underlying the control of locomotion in vertebrates. We have put an emphasis on spinal mechanisms and descending control.

The serotonin reuptake blocker citalopram destabilizes fictive locomotor activity in salamander axial circuits through 5-HT1A receptors.

Serotoninergic (5-HT) neurons are powerful modulators of spinal locomotor circuits. Most studies on 5-HT modulation focused on the effect of exogenous 5-HT and these studies provided key information about the cellular mechanisms involved. Less is known about the effects of increased release of endogenous 5-HT with selective serotonin reuptake inhibitors. In mammals, such molecules were shown to destabilize the fictive locomotor output of spinal limb networks through 5-HT1A receptors. However, in tetrapods little is known about the effects of increased 5-HT release on the locomotor output of axial networks, which are coordinated with limb circuits during locomotion from basal vertebrates to mammals. Here, we examined the effect of citalopram on fictive locomotion generated in axial segments of isolated spinal cords in salamanders, a tetrapod where raphe 5-HT reticulospinal neurons and intraspinal 5-HT neurons are present as in other vertebrates. Using electrophysiological recordings of ventral roots, we show that fictive locomotion generated by bath-applied glutamatergic agonists is destabilized by citalopram. Citalopram-induced destabilization was prevented by a 5-HT1A receptor antagonist, whereas a 5-HT1A receptor agonist destabilized fictive locomotion. Using immunofluorescence experiments, we found 5-HT-positive fibers and varicosities in proximity with motoneurons and glutamatergic interneurons that are likely involved in rhythmogenesis. Our results show that increasing 5-HT release has a deleterious effect on axial locomotor activity through 5-HT1A receptors. This is consistent with studies in limb networks of turtle and mouse, suggesting that this part of the complex 5-HT modulation of spinal locomotor circuits is common to limb and axial networks in limbed vertebrates.NEW & NOTEWORTHY Little is known about the modulation exerted by endogenous serotonin on axial locomotor circuits in tetrapods. Using axial ventral root recordings in salamanders, we found that a serotonin reuptake blocker destabilized fictive locomotor activity through 5-HT1A receptors. Our anatomical results suggest that serotonin is released on motoneurons and glutamatergic interneurons possibly involved in rhythmogenesis. Our study suggests that common serotoninergic mechanisms modulate axial motor circuits in amphibians and limb motor circuits in reptiles and mammals.

Reproducing Five Motor Behaviors in a Salamander Robot With Virtual Muscles and a Distributed CPG Controller Regulated by Drive Signals and Proprioceptive Feedback.

Diverse locomotor behaviors emerge from the interactions between the spinal central pattern generator (CPG), descending brain signals and sensory feedback. Salamander motor behaviors include swimming, struggling, forward underwater stepping, and forward and backward terrestrial stepping. Electromyographic and kinematic recordings of the trunk show that each of these five behaviors is characterized by specific patterns of muscle activation and body curvature. Electrophysiological recordings in isolated spinal cords show even more diverse patterns of activity. Using numerical modeling and robotics, we explored the mechanisms through which descending brain signals and proprioceptive feedback could take advantage of the flexibility of the spinal CPG to generate different motor patterns. Adapting a previous CPG model based on abstract oscillators, we propose a model that reproduces the features of spinal cord recordings: the diversity of motor patterns, the correlation between phase lags and cycle frequencies, and the spontaneous switches between slow and fast rhythms. The five salamander behaviors were reproduced by connecting the CPG model to a mechanical simulation of the salamander with virtual muscles and local proprioceptive feedback. The main results were validated on a robot. A distributed controller was used to obtain the fast control loops necessary for implementing the virtual muscles. The distributed control is demonstrated in an experiment where the robot splits into multiple functional parts. The five salamander behaviors were emulated by regulating the CPG with two descending drives. Reproducing the kinematics of backward stepping and struggling however required stronger muscle contractions. The passive oscillations observed in the salamander's tail during forward underwater stepping could be reproduced using a third descending drive of zero to the tail oscillators. This reduced the drag on the body in our hydrodynamic simulation. We explored the effect of local proprioceptive feedback during swimming and forward terrestrial stepping. We found that feedback could replace or reduce the need for different drives in both cases. It also reduced the variability of intersegmental phase lags toward values appropriate for locomotion. Our work suggests that different motor behaviors do not require different CPG circuits: a single circuit can produce various behaviors when modulated by descending drive and sensory feedback.

The mesencephalic locomotor region sends a bilateral glutamatergic drive to hindbrain reticulospinal neurons in a tetrapod.

In vertebrates, stimulation of the mesencephalic locomotor region (MLR) on one side evokes symmetrical locomotor movements on both sides. How this occurs was previously examined in detail in a swimmer using body undulations (lamprey), but in tetrapods the downstream projections from the MLR to brainstem neurons are not fully understood. Here we examined the brainstem circuits from the MLR to identified reticulospinal neurons in the salamander Notophthalmus viridescens. Using neural tracing, we show that the MLR sends bilateral projections to the middle reticular nucleus (mRN, rostral hindbrain) and the inferior reticular nucleus (iRN, caudal hindbrain). Ca(2+) imaging coupled to electrophysiology in in vitro isolated brains revealed very similar responses in reticulospinal neurons on both sides to a unilateral MLR stimulation. As the strength of MLR stimulation was increased, the responses increased in size in reticulospinal neurons of the mRN and iRN, but the responses in the iRN were smaller. Bath-application or local microinjections of glutamatergic antagonists markedly reduced reticulospinal neuron responses, indicating that the MLR sends glutamatergic inputs to reticulospinal neurons. In addition, reticulospinal cells responded to glutamate microinjections and the size of the responses paralleled the amount of glutamate microinjected. Immunofluorescence coupled with anatomical tracing confirmed the presence of glutamatergic projections from the MLR to reticulospinal neurons. Overall, we show that the brainstem circuits activated by the MLR in the salamander are organized similarly to those previously described in lampreys, indicating that the anatomo-physiological features of the locomotor drive are well conserved in vertebrates. J. Comp. Neurol. 524:1361-1383, 2016. © 2015 The Authors The Journal of Comparative Neurology Published by Wiley Periodicals, Inc.

Neonatal testosterone suppresses a neuroendocrine pulse generator required for reproduction.

The pituitary gland releases hormones in a pulsatile fashion guaranteeing signalling efficiency. The determinants of pulsatility are poorly circumscribed. Here we show in magnocellular hypothalamo-neurohypophyseal oxytocin (OT) neurons that the bursting activity underlying the neurohormonal pulses necessary for parturition and the milk-ejection reflex is entirely driven by a female-specific central pattern generator (CPG). Surprisingly, this CPG is active in both male and female neonates, but is inactivated in males after the first week of life. CPG activity can be restored in males by orchidectomy or silenced in females by exogenous testosterone. This steroid effect is aromatase and caspase dependent, and is mediated via oestrogen receptor-α. This indicates the apoptosis of the CPG network during hypothalamic sexual differentiation, explaining why OT neurons do not burst in adult males. This supports the view that stereotypic neuroendocrine pulsatility is governed by CPGs, some of which are subjected to gender-specific perinatal programming.

Modular functional organisation of the axial locomotor system in salamanders.

Most investigations on tetrapod locomotion have been concerned with limb movements. However, there is compelling evidence that the axial musculoskeletal system contributes to important functions during locomotion. Adult salamanders offer a remarkable opportunity to examine these functions because these amphibians use axial undulations to propel themselves in both aquatic and terrestrial environments. In this article, we review the currently available biological data on axial functions during various locomotor modes in salamanders. We also present data showing the modular organisation of the neural networks that generate axial synergies during locomotion. The functional implication of this modular organisation is discussed.

Anatomical and electrophysiological plasticity of locomotor networks following spinal transection in the salamander.

Recovery of locomotor behavior following spinal cord injury can occur spontaneously in some vertebrates, such as fish, urodele amphibians, and certain reptiles. This review provides an overview of the current status of our knowledge on the anatomical and electrophysiological changes occurring within the spinal cord that lead to, or are associated with the re-expression of locomotion in spinally-transected salamanders. A better understanding of these processes will help to devise strategies for restoring locomotor function in mammals, including humans.

A salamander's flexible spinal network for locomotion, modeled at two levels of abstraction.

Animals have to coordinate a large number of muscles in different ways to efficiently move at various speeds and in different and complex environments. This coordination is in large part based on central pattern generators (CPGs). These neural networks are capable of producing complex rhythmic patterns when activated and modulated by relatively simple control signals. Although the generation of particular gaits by CPGs has been successfully modeled at many levels of abstraction, the principles underlying the generation and selection of a diversity of patterns of coordination in a single neural network are still not well understood. The present work specifically addresses the flexibility of the spinal locomotor networks in salamanders. We compare an abstract oscillator model and a CPG network composed of integrate-and-fire neurons, according to their ability to account for different axial patterns of coordination, and in particular the transition in gait between swimming and stepping modes. The topology of the network is inspired by models of the lamprey CPG, complemented by additions based on experimental data from isolated spinal cords of salamanders. Oscillatory centers of the limbs are included in a way that preserves the flexibility of the axial network. Similarly to the selection of forward and backward swimming in lamprey models via different excitation to the first axial segment, we can account for the modification of the axial coordination pattern between swimming and forward stepping on land in the salamander model, via different uncoupled frequencies in limb versus axial oscillators (for the same level of excitation). These results transfer partially to a more realistic model based on formal spiking neurons, and we discuss the difference between the abstract oscillator model and the model built with formal spiking neurons.

From lamprey to salamander: an exploratory modeling study on the architecture of the spinal locomotor networks in the salamander.

The evolutionary transition from water to land required new locomotor modes and corresponding adjustments of the spinal "central pattern generators" for locomotion. Salamanders resemble the first terrestrial tetrapods and represent a key animal for the study of these changes. Based on recent physiological data from salamanders, and previous work on the swimming, limbless lamprey, we present a model of the basic oscillatory network in the salamander spinal cord, the spinal segment. Model neurons are of the Hodgkin-Huxley type. Spinal hemisegments contain sparsely connected excitatory and inhibitory neuron populations, and are coupled to a contralateral hemisegment. The model yields a large range of experimental findings, especially the NMDA-induced oscillations observed in isolated axial hemisegments and segments of the salamander Pleurodeles waltlii. The model reproduces most of the effects of the blockade of AMPA synapses, glycinergic synapses, calcium-activated potassium current, persistent sodium current, and [Formula: see text]-current. Driving segments with a population of brainstem neurons yields fast oscillations in the in vivo swimming frequency range. A minimal modification to the conductances involved in burst-termination yields the slower stepping frequency range. Slow oscillators can impose their frequency on fast oscillators, as is likely the case during gait transitions from swimming to stepping. Our study shows that a lamprey-like network can potentially serve as a building block of axial and limb oscillators for swimming and stepping in salamanders.

Decoding the mechanisms of gait generation in salamanders by combining neurobiology, modeling and robotics.

Vertebrate animals exhibit impressive locomotor skills. These locomotor skills are due to the complex interactions between the environment, the musculo-skeletal system and the central nervous system, in particular the spinal locomotor circuits. We are interested in decoding these interactions in the salamander, a key animal from an evolutionary point of view. It exhibits both swimming and stepping gaits and is faced with the problem of producing efficient propulsive forces using the same musculo-skeletal system in two environments with significant physical differences in density, viscosity and gravitational load. Yet its nervous system remains comparatively simple. Our approach is based on a combination of neurophysiological experiments, numerical modeling at different levels of abstraction, and robotic validation using an amphibious salamander-like robot. This article reviews the current state of our knowledge on salamander locomotion control, and presents how our approach has allowed us to obtain a first conceptual model of the salamander spinal locomotor networks. The model suggests that the salamander locomotor circuit can be seen as a lamprey-like circuit controlling axial movements of the trunk and tail, extended by specialized oscillatory centers controlling limb movements. The interplay between the two types of circuits determines the mode of locomotion under the influence of sensory feedback and descending drive, with stepping gaits at low drive, and swimming at high drive.

Where are we in understanding salamander locomotion: biological and robotic perspectives on kinematics.

Salamanders have captured the interest of biologists and roboticists for decades because of their ability to locomote in different environments and their resemblance to early representatives of tetrapods. In this article, we review biological and robotic studies on the kinematics (i.e., angular profiles of joints) of salamander locomotion aiming at three main goals: (i) to give a clear view of the kinematics, currently available, for each body part of the salamander while moving in different environments (i.e., terrestrial stepping, aquatic stepping, and swimming), (ii) to examine what is the status of our current knowledge and what remains unclear, and (iii) to discuss how much robotics and modeling have already contributed and will potentially contribute in the future to such studies.

Sensory feedback plays a significant role in generating walking gait and in gait transition in salamanders: a simulation study.

Here, we investigate the role of sensory feedback in gait generation and transition by using a three-dimensional, neuro-musculo-mechanical model of a salamander with realistic physical parameters. Activation of limb and axial muscles were driven by neural output patterns obtained from a central pattern generator (CPG) which is composed of simulated spiking neurons with adaptation. The CPG consists of a body-CPG and four limb-CPGs that are interconnected via synapses both ipsilaterally and contralaterally. We use the model both with and without sensory modulation and four different combinations of ipsilateral and contralateral coupling between the limb-CPGs. We found that the proprioceptive sensory inputs are essential in obtaining a coordinated lateral sequence walking gait (walking). The sensory feedback includes the signals coming from the stretch receptor like intraspinal neurons located in the girdle regions and the limb stretch receptors residing in the hip and scapula regions of the salamander. On the other hand, walking trot gait (trotting) is more under central (CPG) influence compared to that of the peripheral or sensory feedback. We found that the gait transition from walking to trotting can be induced by increased activity of the descending drive coming from the mesencephalic locomotor region and is helped by the sensory inputs at the hip and scapula regions detecting the late stance phase. More neurophysiological experiments are required to identify the precise type of mechanoreceptors in the salamander and the neural mechanisms mediating the sensory modulation.

Axial dynamics during locomotion in vertebrates lesson from the salamander.

Much of what we know about the flexibility of the locomotor networks in vertebrates is derived from studies examining the adaptation of limb movements during stepping in various conditions. However, the body movements play important roles during locomotion: they produce the thrust during undulatory locomotion and they help to increase the stride length during legged locomotion. In this chapter, we review our current knowledge about the flexibility in the neuronal circuits controlling the body musculature during locomotion. We focus especially on salamander because, as an amphibian, this animal is able to display a rich repertoire of aquatic and terrestrial locomotor modes.

Rhythmogenesis in axial locomotor networks: an interspecies comparison.

During locomotion, specialized neural networks referred to as "central pattern generators" ensure precise temporal relations between the axial segments, both in limbed and limbless vertebrates. These neural networks are intrinsically capable of generating coordinated patterns of rhythmic activity in the absence of sensory feedback or descending command from higher brain centers. Rhythmogenesis in these neural circuits lies on several mechanisms, both at the cellular and the network levels. In this chapter, we compare the anatomical organization of the axial networks, the role of identified spinal neurons, and their interactions in rhythmogenesis in four species: lamprey, zebrafish, Xenopus tadpole, and salamander. The comparison suggests that several principles in axial network design are phylogenetically conserved among vertebrates.

Segmental oscillators in axial motor circuits of the salamander: distribution and bursting mechanisms.

The rhythmic and coordinated activation of axial muscles that underlie trunk movements during locomotion are generated by specialized networks in the spinal cord. The operation of these networks has been extensively investigated in limbless swimming vertebrates. But little is known about the architecture and functioning of the axial locomotor networks in limbed vertebrates. We investigated the rhythm-generating capacity of the axial segmental networks in the salamander (Pleurodeles waltlii). We recorded ventral root activity from hemisegments and segments that were surgically isolated from the mid-trunk cord and chemically activated with bath-applied N-methyl-d-aspartate (NMDA). We provide evidence that the rhythmogenic capacity of the axial network is distributed along the mid-trunk spinal cord without an excitability gradient. We demonstrate that the burst generation in a hemisegment depends on glutamatergic excitatory interactions. Reciprocal glycinergic inhibition between opposite hemisegments ensures left-right alternation and lowers the rhythm frequency in segments. Our results further suggest that persistent sodium current contributes to the rhythmic regenerating process both in hemisegments and segments. Burst termination in hemisegments is not achieved through the activation of apamine-sensitive Ca(2+)-activated K(+) channels and burst termination in segments relies on crossed glycinergic inhibition. Together our results indicate that the basic design of the salamander axial network is similar to most of axial networks investigated in other vertebrates, albeit with some significant differences in the cellular mechanism that underlies segmental bursting. This finding supports the view of a phylogenetic conservation of basic building blocks of the axial locomotor network among the vertebrates.

A 3D Musculo-Mechanical Model of the Salamander for the Study of Different Gaits and Modes of Locomotion.

Computer simulation has been used to investigate several aspects of locomotion in salamanders. Here we introduce a three-dimensional forward dynamics mechanical model of a salamander, with physically realistic weight and size parameters. Movements of the four limbs and of the trunk and tail are generated by sets of linearly modeled skeletal muscles. In this study, activation of these muscles were driven by prescribed neural output patterns. The model was successfully used to mimic locomotion on level ground and in water. We compare the walking gait where a wave of activity in the axial muscles travels between the girdles, with the trotting gait in simulations using the musculo-mechanical model. In a separate experiment, the model is used to compare different strategies for turning while stepping; either by bending the trunk or by using side-stepping in the front legs. We found that for turning, the use of side-stepping alone or in combination with trunk bending, was more effective than the use of trunk bending alone. We conclude that the musculo-mechanical model described here together with a proper neural controller is useful for neuro-physiological experiments in silico.

Muscarinic control of the excitability of hindlimb motoneurons in chronic spinal-transected salamanders.

The excitability of spinal motoneurons (MNs) is regulated by acetylcholine via the activation of muscarinic receptors. The objective of the present study was to determine whether this cholinergic modulation of MN excitability is altered following a chronic spinal cord transection. Juvenile salamanders (Pleurodeles waltlii) were spinally transected at the mid-trunk level, and patch-clamp recordings from hindlimb MNs in spinal cord slices were performed 9-30 days after transection, with and without bath application of muscarine (20 mum). Our results showed that the input-output relationship was larger in MNs recorded 2 weeks after spinal transection than in MNs recorded 3-4 weeks after spinal transection. They further revealed that muscarine increased both the gain of MNs and the proportion of MNs that could exhibit plateau potentials and afterdischarges, whereas it decreased the amplitude of the medium afterhypolarizing potential. Moreover, muscarine had no effect on the hyperpolarization-activated cation current (I(h)), whereas it increased the inward rectifying K(+) current (I(Kir)) in MNs recorded > or = 2 weeks after spinal transection. We conclude that following chronic spinal cord injury, the muscarinic modulation of some intrinsic properties of MNs previously reported in acute spinal-transected animals [S. Chevallier et al. (2006)The Journal of Physiology, 570, 525-540] was preserved, whereas that of other intrinsic properties of MNs was suppressed, either transiently (I(Kir)) or definitively (I(h)). These alterations in muscarinic modulation of MN excitability may contribute to the spontaneous recovery of locomotion displayed in long-term chronic spinal-transected salamanders.

Fibroblast growth factor-2 mRNA expression in the brainstem and spinal cord of normal and chronic spinally transected urodeles.

Descending pathways in the spinal cord of adult urodele amphibians show a high regenerative ability after body spinal cord transection; regenerated axons regrow into the transected spinal cord, and hindlimb locomotor recovery occurs spontaneously. Little is currently known about the molecular basis of spinal cord regeneration in urodeles, but it is believed that fibroblast growth factor-2 (FGF2) may play an important role by inducing proliferation of neural progenitor cells. The aim of our study, using in situ hybridization in adult Pleurodeles waltlii, was twofold: 1) to document FGF2 mRNA expression pattern along the brainstem-spinal cord of intact salamanders and 2) to investigate the changes in this pattern in animals unable to display hindlimb locomotor movements and in animals having fully recovered hindlimb locomotor activity after body spinal cord transection. This design establishes a firm basis for further studies on the role of FGF2 in functional recovery of hindlimb locomotion. Our results revealed a decreasing rostrocaudal gradient in FGF2 mRNA expression along the brainstem-spinal cord in intact animals. They further demonstrated a long-lasting up-regulation of FGF2 mRNA expression in response to spinal transection at the midtrunk level, both in brainstem and in the spinal cord below the injury. Finally, double immunolabeling showed that FGF2 was up-regulated in neuroglial, presumably undifferentiated, cells. Therefore, we propose that FGF2 may be involved in cell proliferation and/or neuronal differentiation after body spinal cord transection in salamander and could thus play an important role in functional recovery of locomotion after spinal lesion.

Organisation of the spinal central pattern generators for locomotion in the salamander: biology and modelling.

Among living tetrapods, salamanders are regarded as most closely resembling the first terrestrial vertebrates, and are therefore an interesting group in which the evolutionary changes in the locomotor behaviour from aquatic to terrestrial habitats can be inferred. Salamanders exhibit two locomotor modes: swimming and terrestrial stepping. The swimming is anguilliform and resembles closely that of the lamprey. On the ground, the salamander switches to a stepping gait with axial undulations that is also observed in many reptiles. The salamander is therefore ideally suited for examining the neural mechanisms for the generation of these two locomotor modes, as well as the neural mechanisms of gait transition. In the present paper, we describe the kinematics and patterns of activation of axial and limb muscles during stepping and swimming in adult salamanders. We then review the current neurobiological data about the organisation of the spinal networks underlying swimming and stepping, and the mechanisms of gait transition. Finally we report modelling studies aimed at understanding the organisation and operation of the salamander locomotor circuits. Altogether, the neurobiological and the modelling data support the hypothesis of a phylogenetic conservatism from agnathians to amphibians of the spinal locomotor networks generating axial motor patterns.

From swimming to walking with a salamander robot driven by a spinal cord model.

The transition from aquatic to terrestrial locomotion was a key development in vertebrate evolution. We present a spinal cord model and its implementation in an amphibious salamander robot that demonstrates how a primitive neural circuit for swimming can be extended by phylogenetically more recent limb oscillatory centers to explain the ability of salamanders to switch between swimming and walking. The model suggests neural mechanisms for modulation of velocity, direction, and type of gait that are relevant for all tetrapods. It predicts that limb oscillatory centers have lower intrinsic frequencies than body oscillatory centers, and we present biological data supporting this.