RESEARCH

Physiological and molecular organization of neuronal networks controlling walking
Brain functions are generated by activity in dedicated neuronal networks. A major challenge to neuroscientists is to understand the functional mode of operation of such circuits in the mammalian brain. For locomotor behaviors, like walking, motor circuits in the spinal cord itself generate the actual timing and coordination of the rhythmic muscle activity. A great part of the research in our lab is carried out in an attempt to understand the molecular and functional organization of the spinal motor circuits underlying walking in mammals, including the activation of these circuits from the brain and sensory control. In these studies, we use the in vitro rodent spinal cord, which is an excellent preparation for experimental manipulation of rhythmic locomotor activity in mammals and is amenable to molecular genetic manipulations of neuronal networks.

Early work from our lab has revealed aspects of the overall organization of the locomotor network, the implication of cellular properties for rhythmicity, and the nature of neuronal spike coding. We have developed numerous techniques to analyze network activity and introduced the technique of whole cell recording from interneurons in the intact mammalian spinal cord.

More recent work focuses on the network organization. The key features that characterize limbed locomotion in mammals are: 1) the rhythm generation itself, 2) the coordination of flexors and extensors across the same or different joints in a limb or between limbs, and 3) the left/right coordination. Our work has provided a detailed anatomical and electrophysiological characterization of the locomotor network in mammals that controls left-right coordination. This network organization may serve as a vantage point for characterization of the mammalian locomotor system. Using molecular genetics as direct tools in the network analysis, we have started to identify excitatory network components in the mammalian spinal cord and brainstem locomotor networks. We have initially targeted two molecularly defined classes of ipsilaterally projecting excitatory glutamatergic interneurons: V2a neurons - expressing the transcription factor Chx10 - and neurons expressing the axon guidance molecule, EphA4.

Targeted deletions of EphA4 lead to an abnormal motor behavior where null mice display a hopping, rabbit-like gait. In electrophysiological studies and anatomical studies we showed that the hopping gait is explained by a reconfiguration of the spinal locomotor network where EphA4 positive neurons aberrantly cross the midline whereas they normally are uncrossed. A defined a subset of EphA4 positive spinal neurons are excitatory locomotor related neurons.

Studies of the V2a neurons show that these interneurons play little or no role in rhythm-generation but drive commissural interneurons to insure left-right alternation during locomotion. The V2a network is organized in a modular fashion along the cord. Ongoing work aims at defining how V2a interneurons, together with other excitatory neurons, control left-right alternation

In a collaborative effort, we have determined the functional role of mouse V1 neurons, a major class of spinal inhibitory interneurons that selectively expresses the transcription factor Engrailed1. These experiments outlined a surprising role for inhibition in regulating the frequency of the locomotor rhythm.

We have implemented BAC-technology to produce transgenic mice lines with expression of the Cre-lox system in excitatory neurons in the spinal cord and brainstem. Using a transgenic mouse line that selectively drives light-sensitive channels - channelrhodopsin 2 - in glutamatergic excitatory neurons in the spinal cord and brainstem, we show that such neurons in the cord are directly responsible for rhythm-generation and that glutamatergic neurons in the lower hindbrain serve as command neurons for initiating locomotor activity. In future experiments, we will use optogenetic techniques to probe the function of specific classes of neurons in the network. These tools will provide a basis for the sophisticated functional and network studies needed to understand the principal mode of operation of a large-scale mammalian motor circuit.

Plateau potentials: a cellular mechanism underlying spasticity after spinal cord injury
Severe abnormal motor function like spasticity develops as a consequence of spinal cord injury or damage to motor pathways from the brain. We previously made the hypothesis and provided evidence that the pathophysiology of spasticity is positively related to chronic expression of plateau properties in motor neurons after spinal cord injury. Plateau potentials in vertebrate motor neurons are mediated by activation of prolonged sodium/calcium currents and the expression is conditional upon activation of noradrenergic and/or serotoninergic receptors. The normal function of motor neuron plateau potentials seems to be to maintain persistent motor output and amplify synaptic inputs during rhythmic motor activity. To find possible targets for regulation of the chronic expression of plateau potentials after spinal cord injury we have used global gene-expression analysis from isolated rat motor neurons before and after injury to the cord using Micro-array analysis. These studies have shown that the chronic expression of plateaux may be related to changes in genes coding for the regulatory units for persistent sodium and calcium channels as well as a number of intracellular pathways that may regulate plateau expression. Moreover, we have identified transcription factors that seem to regulate clusters of genes with the same dynamic change in gene expression after injury. In ongoing experiments we aim at performing targeted manipulations of these pathways with the hope of defining new therapies for symptoms associated with spinal cord injury.