ABSTRACT'25I-Labeled a-bungarotoxin has been used to determine the distribution of putative acetylcholine receptors in normal and chronically deafferented antennal lobes in the brain of the moth Manduca sexta. Toxin-binding sites are confined to synaptic regions in the neuropil of normal lobes and to rudiments of these regions in deafferented lobes. These findings suggest that receptors can develop in the insect central nervous system independently of normal synaptic influences. Many excitable cells require innervation for their normal development and for maintenance of their normally differentiated properties (for examples, see reviews 1-3). The most fully documented case of this role of innervation is the regulation of the distribution of acetylcholine (AcCho) receptors in vertebrate skeletal muscle fibers developing in vivo (4-7) and in vitro (8-13). AcCho receptors, which appear about the time of myotube formation, are initially distributed over the entire surface of the myotubes. As the developing muscle fiber becomes innervated and synaptic transmission begins, AcCho receptors are lost from the extrajunctional regions of the muscle fiber but are retained in the junctional membrane. If innervation is prevented, this redistribution of receptors does not occur, and also the immature muscle eventually dies. From these observations it appears that the motor innervation influences the muscle fiber to localize its AcCho receptors in the subsynaptic membrane.We know of no comparable study of development of receptors during de novo synapse formation in the central nervous system. Therefore, we have used the antennal lobes of the brain of the moth Manduca sexta to study the role of normal innervation in regulating the development and distribution of AcCho receptors on central neurons. The antennal lobes are an excellent preparation for such studies because they appear to have a high density of cholinergic synapses, a simple anatomical organization with clear separation of synaptic and nonsynaptic regions, and experimentally observable and manipulable development (14-16). To probe for AcCho receptors in the antennal lobes, we have used the snake venom protein a-bungarotoxin (a-Bgt), which is a powerful agent for identifying and quantifying nicotinic AcCho receptors at vertebrate neuromuscular junctions (17,18 Deafferentation involved amputation of antennal rudiments ("deantennation") on day 1 after the larval-pupal molt (16). This operation removes the epidermis in which the antennal sensory neurons ordinarily arise and is performed prior to the birthdays of these neurons.Histology. For dissection of the brain and subesophageal ganglion, a developing or pharate adult Manduca was anesthetized with CO2, and the brain was exposed through a window cut in the head cuticle (16). The brain was moistened with cold physiological salts solution during dissection (16). Tissues were fixed in modified Serra's fixative (25), dehydrated, and embedded in Paraplast Plus (Sherwood Medical Industries, St. Louis, MO). They were sectione...