Long before synaptic networks are fully established, electrical activity present in developing neurons regulates neuronal differentiation (1-4). In particular, electrical activity mediated by the NMDA 1 class of glutamate receptors is required for normal neuronal development. Loss of NMDA receptor function during development increases neuronal cell death (5), prevents the formation of precise neural circuits (6, 7), diminishes respiration and feeding (8 -10), and has been implicated in fetal alcohol syndrome (11) and schizophrenia (12). NMDA receptor function can also regulate neuronal proliferation (13) and migration (14). Despite the importance of NMDA receptors for normal development and adult brain function, knowledge of molecular mechanisms regulated by NMDA receptors in developing neurons is rudimentary.Evidence from development as well as adult models of learning and memory indicate that regulation of gene expression is an important strategy that can be used to mediate changes in neuronal structure and function. Developmental studies suggest that mRNA abundance is rate-limiting for the accumulation of functional neurotransmitter receptors on neurons during development in vivo (15). Changes in gene expression also accompany activity regulated synaptic plasticity events in the developing visual system (16), as well as at the neuromuscular junction (17). Transcription factors have recently been shown to play a critical role for the development of synaptic connectivity (18 -20). Similarly, transcription may be required for activity dependent long term changes in synaptic strength in mature nervous systems (21-23). Finally, the electrical activity of neurons, including NMDA receptor function, has been shown to regulate gene expression in the adult hippocampus (24 -27), a structure critical for the establishment and maintenance of memories (28 -30).To test the hypothesis that NMDA receptor-dependent regulation of gene expression is required for, and directs, molecular and cellular mechanisms for neuronal development, we have designed a screen based on the neural circuit that connects highly specialized whiskers (mystacial vibrissae) found on the snout of the mouse to their synaptic targets in the brain stem. These whiskers are richly innervated by pseudounipolar sensory neurons of the trigeminal ganglion that project centrally to the brain stem trigeminal complex. Here synaptic inputs from the trigeminal ganglion that correspond to the mystacial vibrissae are organized in a pattern that matches their topographic organization on the face. This pattern of whisker representations, called barrelettes at the level of the brain stem, is relayed and reiterated in the thalamus and finally the cerebral cortex, and requires NMDA receptor function for normal development (8,10,31,32). The whisker representations are comparable to other sensory maps, for example, for touch, hearing, and vision, that are found in humans, and serve as a powerful mammalian model for highly patterned synaptic development and organization in viv...
