The animated movie Antz follows the misadventures of a neurotic ant. In the laboratory, unusual insects are usually the result of genetic engineering rather than anthropomorphism as researchers generate Drosophila models of human genetic diseases. However, it is difficult to study the phenotype of a genetic mutation without knowledge of the gene's normal function. Now Chintapalli et al. report FlyAtlas, a free resource that identifies sites of expression and enrichment for each gene in the Drosophila genome in a recent article in Nature Genetics.
Until the publication of its genome in 2000, only 20% of Drosophila melanogaster genes had been identified, many of which were primarily involved in development. Currently, most high-throughput genomics studies of Drosophila use mRNA from the whole animal, likely underestimating local expression of rare genes.
The authors analyzed 9 tissues from adult Drosophila (brain, head, crop, midgut, Malpighian tubules, hindgut, testis, ovary and male accessory glands) with 1000 Drosophila Genome 2 microarray chips from the UK Drosophila Affymetrix Microarray Facility. Each chip contained 18,800 probes for 18,500 unique transcripts, and the authors ran 4 replicates of each microarray. Including data from Drosophila larvae and previously published data from whole Drosophila, the FlyAtlas dataset contains approximately 900,000 gene expression values.
Tissue-specific microarrays were more sensitive than whole organism microarrays, Each tissue expressed approximately 50% of all Drosophila genes. Together, the tissue-specific datasets were positive for approximately 85%, whereas whole animal samples were positive for only 67% of Drosophila genes. The authors detected 665 features never before reported in adult Drosophila in their tissue-specific screen. More than 25% of Drosophila genes showed predominant expression and nearly 10% showed unique expression in one tissue relative to all others. With each tissue contributing only a small percentage of organismal gene expression, the authors estimated that tissue-specific gene expression would have to change by 1000% to cause a 50% increase in a gene's expression in the whole animal.
The authors found surprising expression patterns. For example, several genes encoding odorant-binding proteins did not show head-specific expression. Obp56d was enriched in the hindgut as well as the head, and Obp56f and Obp22a were uniquely expressed in the accessory gland. In addition, fasciclin2, which is important in axon guidance and is implicated in memory, was enriched in the Malpighian tubule, which constitutes the Drosophila excretory system, and ventral nerve system defective (vnd), which is important in neural development, was enriched in the adult hindgut.
Several disease-associated genes showed similar expression patterns in humans and Drosophila. For example homologs for the GABAA receptor gamma 2 subunit, RAPSYN and SLITRK1, which are associated with myoclonic epilepsy, myasthenia gravis and Tourette's syndrome, respectively, were all enriched in the Drosophila brain. These data support the creation of new Drosophila models of human neurological diseases.