Recently, an extremely large multigene family likely to encode odorant receptors has been identified, which may consist of as many as several hundred divergent genes. The structural diversity of this gene family raises the question of how the brain encodes information about which receptors are occupied by odorous ligands. The solution to this problem appears to be that sensory neurons in the olfactory epithelium exhibit a high degree of diversity with respect to receptor gene expression. In catfish, individual receptor genes are expressed in only about 1% of the population of sensory neurons, and estimates suggest that each neuron may express less than five receptor genes and perhaps as few as one. The vast diversity of olfactory sensory neurons with respect to receptor gene expression immediately poses the interesting problem of how individual neurons regulate expression of one or a few receptor genes, while excluding expression of as many as 1000 other genes of the same family. Do transcription factors specify the active genes, or, alternatively, are transcriptionally silent genes translocated into active loci by genomic rearrangement? Efforts to study expression of a single receptor gene in the intact epithelium are complicated by the fact that each gene is active in only. about 0.1% of the epithelial cells. To enrich for neurons which express the same receptor genes, clonal lines of olfactory sensory neurons will be generated and induced to differentiate. If a population of cells expressing a known set of receptors is isolated, then Southern blotting will be used to test whether receptor gene expression is associated with genomic rearrangement. If rearrangement occurs, it would be extremely exciting because it would be the first example of directed genomic recombination outside of the immune system. Alternatively, if the choice of which receptor gene to express does not involve recombination, then it may be regulated by transcription factors. To test this possibility, several kilobases of sequence upstream from both active and inactive receptor genes will be linked to a reporter construct and transfected into the cell lines to test whether the cells can limit expression to only a subset of genes. If cells regulate transfected receptor genes in a specific fashion, then regulatory elements will be mapped. Whatever the mechanism turns out to be, this problem will provide an exciting opportunity to probe neural gene regulation in an extremely complex system.
