Activity‐Dependent Regulation of Transcription During Development of Synapses

Publisher Summary

This chapter discusses the role of signaling and transcription during long-term neural change. During the establishment of long-term changes, new proteins are synthesized in a highly context-dependent fashion. This is brought about both by the translation of preexisting messenger ribonucleic acids (mRNAs) and by the activation of specific transcription factors. The Drosophila larval neuromuscular junction (NMJ) has provided a unique paradigm in which the role of transcription in neural development and plasticity has been assessed with remarkable resolution. At this motor synapse, the effect of modulating signaling and transcription factor activation on synaptic strength and size can be determined easily. The availability of classical and modern genetic and genomic methods further enhances the utility of this system. The results indicate the presence of conserved activity-dependent signaling networks that trigger particular transcription factors during long-term synaptic plasticity. Furthermore, as the NMJ grows through development, communication between the pre- and postsynaptic compartments provides signals and constraints to maintain parity. Taken together, the regulation of transcription and protein synthesis during developmental plasticity of the NMJ offers valuable insights into conserved mechanisms of learning and memory across species.

Introduction

Perhaps one of the most intriguing physiological processes in biology is the ability of the brain to acquire, process, store, and retrieve information. Encountering new experiences, making associations, and recalling them at will underscore the utility of learning and memory. Most animals, however primitive, have a semblance of a “neural circuitry” that subserves this function. Instances in which this function of the nervous system is compromised offer one of the best (and the most unfortunate) opportunities for their study. Nowhere is the ability to form new memories more obvious than in cases in which it is lacking, for example, anterograde amnesia (Sacks, 1998). Efforts to understand memory and describe its underpinnings have been remarkably productive. As in most of experimental biology, a top‐down approach has been complemented continually by bottom‐up, reductionist ones. Thus, while an incredible amount of knowledge exists now on information processing by various vertebrate brains, equally substantial progress has been made in understanding the cellular and molecular correlates that operate in neurons.

When an organism learns, there are discrete changes in relevant neural circuits. These changes occur simultaneously at several levels and fundamentally alter the way in which neurons connect to one another. Consequently, both the physical connectivity and firing properties of these neurons change (Fregnac, 1996). These modifications, a necessity for long‐term plasticity, are brought about by cellular processes that direct synthesis of new proteins. In several animal models, blocking protein synthesis precludes long‐term plasticity (Castellucci 1986, Frey 1988, Stanton 1984). Based on comparable experiments designed to discriminate between discrete steps of plasticity, the following largely conserved sequential steps can be identified (Fig. 1). On initial patterned neural activity in a given circuit (such as in response to encountering novel stimuli), short‐term changes are quickly set in motion. These changes typically do not require protein synthesis and instead utilize modifications of preexisting proteins in the relevant neurons (Abraham 1991, Bailey 2000). Membrane properties are altered as a consequence of protein modifications and signaling mechanisms are activated. However, these changes are transient and, in turn, are further consolidated by continued neural activity and long‐lasting activation of intracellular signaling (such as those that might be triggered by repetitive trials; Mauelshagen 1996, Mauelshagen 1998). This late phase, invariably involves the synthesis of new proteins, resulting in structural modifications and changes in membrane and synaptic properties that are more permanent. This later phase is known as late long‐term plasticity (L‐LTP) of synapses.

This sequence during the establishment of long‐term changes has been consistently observed in several systems, both in vivo and in “reduced” preparations (Sharma 2003, Waddell 2001). Thus, formation of long‐term memory requires protein synthesis either through translation of preexisting RNA (Klann 2004, Martin 2004, Richter 2001, Steward 2001, Sutton 2005) or through transcription driven primarily by transcription factors that are activated following instructive neural activity and intracellular signaling (Hevroni 1998, Lonze 2002). In this chapter, we focus on the role of signaling and transcription during long‐term neural change. At the outset, it is important to bear in mind that although protein synthesis is a must for L‐LTP, it is usually not the identity of the proteins but rather the “place” where this happens that encodes the “type” of memory. Protein synthesis thus forms the necessary permissive step that underlies synaptic modifications. It does not dictate what kind of memory will be formed and therefore all long‐term changes are expected to share a large majority of new proteins being made.