Chapter 4 Transcriptional Networks in the Early Development of Sensory–Motor Circuits

Abstract

The emergence of coordinated locomotor behaviors in vertebrates relies on the establishment of selective connections between discrete populations of neurons present in the spinal cord and peripheral nervous system. The assembly of the circuits necessary for movement presumably requires the generation of many unique cell types to accommodate the intricate connections between motor neurons, sensory neurons, interneurons, and muscle. The specification of diverse neuronal subtypes is mediated largely through networks of transcription factors that operate within progenitor and postmitotic cells. Selective patterns of transcription factor expression appear to define the cell‐type‐specific cellular programs that govern the axonal guidance decisions and synaptic specificities of neurons, and may lay the foundation through which innate motor behaviors are genetically predetermined. Recent studies on the developmental programs that specify two highly diverse neuronal classes—spinal motor neurons and proprioceptive sensory neurons—have provided important insights into the molecular strategies used in the earliest phases of locomotor circuit assembly. This chapter reviews progress toward elucidating the early transcriptional networks that define neuronal identity in the locomotor system, focusing on the pathways controlling the specific connections of motor neurons and sensory neurons in the formation of simple reflex circuits.

Introduction

Many organisms are born with a set of innate behaviors that have evolved so that they can confront the challenges imposed by their specific environments. The neural circuits controlling basic motor behaviors such as feeding, breathing, and walking are often functional at the time of birth, independent of any prior interaction with the external world. These genetically hard‐wired circuits can be essential for survival by imprinting behaviors such as the predator escape response—a system where sensory input must be tightly linked to motor output. The identification of the substrates for simple and complex innate behaviors has been a major challenge.

There is emerging evidence that stereotyped patterns of movement can be programmed through the actions of a few key regulatory genes, neurons, and microcircuits. In Drosophila gender‐specific courtship behaviors are specified by a transcription factor encoded by the fruitless gene, which is sufficient to interconvert a specific pattern of mating behavior between males and females (Demir and Dickson, 2005). In Caenorhabditis elegans, the neural substrates controlling innate patterns of motor behavior are comprised of a relatively small number of anatomically well‐defined groups of neurons (Hobert, 2003). Whether similar master gene regulators, or discrete neural assemblies, function to program the behavioral outputs of the vertebrate motor system remain to be determined. Presumably such a factor, or group of factors, would need to function in distinct classes of interconnected neuronal subtypes that by other criteria might be considered dissimilar.

The problem of defining behaviorally relevant circuits in vertebrate nervous systems is confounded by the shear volume of neurons, and the relative complexity and number of synaptic connections. The spinal cord and hindbrain have provided tractable model systems for defining the neural circuits necessary for basic motor functions such as breathing and walking, and contain the sensory feedback systems that are required for reflex responses and locomotor adaptation (reviewed in Goulding and Pfaff, 2005, Kiehn and Butt, 2003). As a system for studying locomotor behaviors, the spinal cord has an advantage in that the anatomy of the system is relatively well defined and the sensory inputs and motor outputs are accessible and quantifiable. One successful approach to the study of the assembly of locomotor circuits has been to define the embryonic programs that contribute to the identity and connectivity of the cells within the circuit, to try to link control of synaptic specificity with the emergence of a defined behavior.

In this chapter, I review our current understanding of the genetic programs which control the specification of motor neurons and sensory neurons in the vertebrate spinal cord and peripheral nervous system. Emphasis will be placed upon the transcriptional networks which dictate the early identity of these two neuronal classes, and on recent advances that have enriched our understanding of the general principles of circuit assembly. Activity‐dependent steps in the wiring of locomotor circuits will not be addressed as this aspect has been the subject of recent reviews (Hanson et al., 2008, Ladle et al., 2007). The potential mechanisms that may contribute to the assembly of sensory–motor circuits will be explored, with a focus on the formation of monosynaptic stretch‐reflex circuits, a collection of neural circuits that are critical for coordinated movement.