The maturation of locomotor networks

Abstract

In both vertebrates and invertebrates, the elaboration of locomotion, and its neural control by the central nervous system, are extremely flexible. This is due not only to the network properties of relevant sets of central neurons, but also to the active participation of mutually co-operative central and peripheral loops of neural projections and activity. In this chapter, we describe experiments in which the above concepts have been advanced by comparing locomotor properties in the adult vs. neonatal rat preparation. Data obtained from the in vivo vs. in vitro preparation, and swimming vs. walking behavior, suggest that the locomotor pattern progressively exhibited after birth corresponds to successive steps in the maturation of locomotor networks. Our work emphasises that during the late pre- and early postnatal period, three distinct neural entities—segmental sensory input, descending pathways, and motoneurons—play a key role in the maturation of locomotion and its neural control. We propose that the neonatal rat preparation is an excellent model for studying the conversion from immature to adult locomotion. Some neural controls are more clearly demonstrable in the developing animal preparation than in the adult because the latter exhibits an array of complex and redundant adaptive mechanisms.

Introduction

The neural control of locomotion has been extensively studied these last four decades and it continues to be a pace-setting topic for movement neuroscientists. It is a complex, coordinated behavior whose most basic feature is its pattern of muscle contractions. Kinematically, locomotion consists of an assemblage of motor sequences with several levels of coordination (intramuscular, intrajoint, interjoint, inter-bodypart, etc.). The extensive literature available on locomotor patterns is due to the extreme diversity of both the species studied (from mollusc to human) and the preparations used (from in vitro tissue to freely moving animals; Orlovsky et al., 1999). In some extremely valuable, albeit still exceptional cases, it is now possible to “… bridge the gap from ion channels to networks and behavior” (Grillner, 1999).

At first glance, locomotor organization and control by the CNS are remarkably similar from one species to another. It is generally accepted that the fundamental rhythm of locomotion originates within relatively low-order central circuits which produce alternation between functionally antagonistic (usually flexor and extensor) muscles. This program (often termed a central pattern generator or CPG) is triggered into activity by an intrinsic command and/or an extrinsic sensory input. It is regulated on a moment-to-moment basis by central and peripheral feedback loops which are necessary for the organism's adaptation to its external environment. The central neural patterns are extremely flexible as is the behavior, itself. Pearson (2000) has emphasized distinctions between rapid adaptations, wherein motor programs are able to switch immediately from one pattern to another, and much longer adaptations brought on by anatomical and morphological changes due to training or traumatic injuries. During maturation, both adaptive effects are present, with the organism able to both adapt quickly to relevant surrounding events and integrate ontogenetic processes on a much slower timescale.

In recent years, the in vitro isolated spinal cord of a neonatal rat in a manipulable bathing medium has become a widely used preparation (Cazalets et al., 1992, Kiehn and Kjaerulff, 1996). Its locomotor pattern can be studied for several hours with intracellular and extracellular recording techniques before central myelination has occurred (postnatal days 0–12; P_0–12). In parallel, ethologists have studied a variety of motor behaviors over the same period (Westerga and Gramsbergen, 1990, Jamon and Clarac, 1998). The close phylogenetic association between the rat and mouse, and the great potential of transgenic models of the latter, has increased even further the general interest of movement neuroscientists in rodents.

In this chapter, we proceed from the flexibility of locomotor behavior in adult species to comparison of locomotor patterns in adult versus neonatal rat preparations and finally, to the maturation of the fundamental locomotor pattern in the rat embryo (see also Kudo et al., this volume). There is an emphasis on comparing data obtained from in vivo versus in vitro preparations (cf. Ezure, Chapter 7 of this volume) and swimming versus walking behavior. All in all, the results to be presented suggest that the locomotor patterns progressively exhibited after birth correspond to the successive steps in the maturation of the locomotor networks in the embryo.