A genome-wide transgenic RNAi library for conditional gene inactivation in Drosophila.

Forward genetic screens in model organisms have provided important insights into numerous aspects of development, physiology and pathology. With the availability of complete genome sequences and the introduction of RNA-mediated gene interference (RNAi), systematic reverse genetic screens are now also possible. Until now, such genome-wide RNAi screens have mostly been restricted to cultured cells and ubiquitous gene inactivation in Caenorhabditis elegans. This powerful approach has not yet been applied in a tissue-specific manner. Here we report the generation and validation of a genome-wide library of Drosophila melanogaster RNAi transgenes, enabling the conditional inactivation of gene function in specific tissues of the intact organism. Our RNAi transgenes consist of short gene fragments cloned as inverted repeats and expressed using the binary GAL4/UAS system. We generated 22,270 transgenic lines, covering 88% of the predicted protein-coding genes in the Drosophila genome. Molecular and phenotypic assays indicate that the majority of these transgenes are functional. Our transgenic RNAi library thus opens up the prospect of systematically analysing gene functions in any tissue and at any stage of the Drosophila lifespan.

The tumor suppressors Brat and Numb regulate transit-amplifying neuroblast lineages in Drosophila.

In both vertebrates and insects, neurons typically arise from neural stem cells or terminally dividing intermediate progenitors. Here, we describe another mode of neurogenesis where neural stem cells generate secondary precursors that undergo multiple rounds of self-renewing transit-amplifying divisions. We identify the Posterior Asense-Negative (PAN) neuroblasts, which do not express the transcription factors Asense or Prospero. PAN neuroblasts rely on the segregating determinants Numb and Brat to generate smaller, secondary neuroblasts that in turn give rise to ganglion mother cells (GMCs) and neurons throughout larval development. In brat or numb mutants, misspecified secondary neuroblasts are unable to produce differentiated progeny and initiate tumor-like overgrowth. In prospero mutants, however, tumors arise from GMCs while secondary neuroblasts are correctly specified. Our data describe a transit-amplifying lineage in the Drosophila nervous system and suggest that different vulnerabilities in intermediate cell types can affect the outcome of tumor suppressor loss in stem cell lineages.