Abstract
Mammalian locomotion is a complex task in which hundreds of muscles work together in a coordinated fashion. Neural networks in the spinal cord, known as central pattern generators (CPGs), carry all the components necessary to produce the cyclical pattern of muscle activity needed for locomotion. The fact that the locomotor CPG is innate and highly localized makes it outstanding as a subject to study how a complex, but concrete behavior, is produced by a neuronal network.
Two fundamental aspects of the CPG are rhythm generation, and flexor-extensor coordination. These two properties have sometimes been linked together, such as in the half-center model, in which the alternating activity between flexors and extensors are the cause of the rhythm. Other models for rhythm generation have also been postulated, consequently no consensus exists regarding the overall structure of the CPG for locomotion.
Pharmacological investigations have indicated that glutamatergic neurons as essential for locomotion. To further elucidate the function of these neurons, the work presented in this thesis has made use of a set of new tools to target glutamatergic neurons, and elucidate their specific contribution to locomotion.
A mouse was produced that expressed the optically gated ion channel Channelrhodopsin-2, making it possible for the first time to selectively activate a genetically specific sub-population of neurons in the spinal cord. The experiments using this mouse show that glutamatergic neuron activation is sufficient to produce locomotor-like activity, both in the spinal cord, and in the hindbrain.
With the use of another set of recently produced transgenic mice, it was possible to probe deeper into the structure of the CPG, and illuminate several key aspects of the organization of the network. Several proposed network models could be refuted and one in particular was promoted. The results show that the CPG network is build up from intrinsically rhythmic modules.
Furthermore, a mouse without glutamatergic neurotransmission was examined. What was found was that the locomotion deficient mouse could produce locomotor-like activity under special conditions, and this activity depended solely on inhibitory interneurons, specifically, reciprocally connected Ia interneurons.
Overall, glutamatergic neurons are shown to form intrinsically rhythmic modules that are indispensable for rhythm generation, and network function. The use of genetics and electrophysiology is a powerful combination that will continue to provide conclusions about how neural networks produce and control complex motor behavior.
