The present study has investigated the rhythmic properties of spinal networks in the neonatal
rat spinal cord in vitro, by means of intracellular recordings from single motoneurons and
extracellular recordings from ventral roots.
The occurrence and basic characteristics of an alternating locomotor-like pattern, triggered by
stimulating afferent dorsal root fibers, was described, suggesting that sensory inputs from the
periphery can activate the spinal locomotor network. It was proposed that activation of the
locomotor CPG could occur via an increase in extracellular K+ and network neuron
depolarization.
Furthermore, it was demonstrated that spinal networks could generate rhythms exhibiting a
variety of different properties in terms of phase, frequency, and duration of patterns. Due to
the importance of inhibitory transmission in shaping such rhythmic patterns, and the complex
role that glycine and GABA seem to have in the developing central nervous system, the
nature and function of er mediated synaptic transmission was investigated on motoneurons.
To this aim, the glycinergic and GABAergic recurrent postsysnaptic potential (PSP) mediated
by Renshaw cells was used to assess its impact on excitatory synaptic inputs from dorsal
afferent fibers. Despite its depolarizing nature, the recurrent PSP consistently inhibited
synaptic excitation of lumbar motoneurons.
Different patterns of rhythmic activity were obtained by activating certain classes of
metabotropic receptors present in the spinal cord. In particular, tachykinin NK3 receptors
could trigger neuronal bursting, which outlasted the stimulus and appeared predominantly
with alternation at segmental level and synchronous coupling between ipsilateral motor pools.
Such bursting was accompanied by depression of GABAergic dorsal root potentials evoked
by dorsal root stimulation and of the recurrent inhibitory PSP recorded from motoneurons,
indicating the possibility that fully alternating pattern generation by the CPG was partly
impaired due to a decrease in the efficacy of inhibition. Nevertheless, NK3 receptor activation
could facilitate fictive locomotor patterns, since they could operate synergistically with
NMDA and 5-HT to trigger fully alternating locomotor-like rhythms. Furthermore, NK3
receptor antagonism disrupted NMDA and 5-HT induced fictive locomotion. Activation of glutamate metabotropic receptors elicited a wide range of effects. Group I
receptors mediated depolarization and onset of oscillatory activity (via the subclasses
mGluRl and 5, respectively), which usually appeared synchronously at homosegmental and
homolateral level. This type of rhythm represented another example of pattern that spinal
cord networks could generate in addition to previously reported ones. While the role of group
I metabotropic glutamate receptors in fictive locomotion was limited, they seemed to
participate in regulating disinhibited rhythm, i.e. bursting induced by block of GABA and
glycine receptors. In addition to an excitatory effect, group I receptors also reduced reflex
responses, at least partly via facilitation of endogenous glycine transmission. Group II and III
metabotropic glutamate receptors, on the other hand, were always inhibitory on spinal
neurons and their bursting behavior. Experiments performed with selective agonists and
antagonists indicated that group II and III receptors could modulate, and were involved in
controlling, the duration and frequency of disinhibited bursts.
Thus, the present study broadened our current understanding of the rhythmic patterns
generated by spinal networks and identified certain forms of oscillatory behavior as induced
by activation of selective classes of transmitter receptors.
