Search

Gerald (Gerry) Rubin, pioneer in Drosophila genetics, is Founding Director of the HHMI-funded Janelia Research Campus. In this interview, Gerry recounts key events and collaborations that have shaped his unique approach to scientific exploration, decision-making, management and mentorship – an approach that forms the cornerstone of the model adopted at Janelia to tackle problems in interdisciplinary biomedical research. Gerry describes his remarkable journey from newcomer to internationally renowned leader in the fly field, highlighting his contributions to the tools and resources that have helped establish Drosophila as an important model in translational research. Describing himself as a ‘tool builder’, his current focus is on developing approaches for in-depth study of the fly nervous system, in order to understand key principles in neurobiology. Gerry was interviewed by Ross Cagan, Senior Editor of Disease Models & Mechanisms., Drosophila Collection: Ross Cagan spoke to Gerry Rubin about his career path and the story behind some of his key breakthroughs, as well as his unique and successful approach to management and mentorship.

The analysis of genetic mosaics, in which an animal carries populations of cells with differing genotypes, is a powerful tool for understanding developmental and cell biology. In 1990, we set out to improve the methods used to make genetic mosaics in Drosophila by taking advantage of recently developed approaches for genome engineering. These efforts led to the work described in our 1993 Development paper.

Differentiation in the Drosophila eye imaginal disc traverses the disc as a wave moving from posterior to anterior. The propagation of this wave is driven by hedgehog protein secreted by the differentiated cells in the posterior region of the disc. Hedgehog induces decapentaplegic expression at the front of differentiation, in the morphogenetic furrow. We have identified a gene, shortsighted, which is expressed in a hedgehog-dependent stripe in the undifferentiated cells just anterior to the furrow and which appears to be involved in the transmission of the differentiation-inducing signal; a reduction in shortsighted function leads to a delay in differentiation and to a loss of photoreceptors in the adult. shortsighted is also required for a morphogenetic movement in the brain that reorients the second optic lobe relative to the first. shortsighted encodes a cytoplasmic leucine zipper protein with homology to a mouse gene, TSC-22, which is transcriptionally induced in response to TGF-β.

Neural specification and differentiation in the Drosophila eye sweep across the unpatterned epithelial monolayer of the eye imaginal disc following a developmental wave termed the morphogenetic furrow. The furrow begins at the posterior margin of the eye imaginal disc and moves anteriorly as a linear front. Progression of the furrow requires the function of hedgehog, which encodes a secreted signaling protein. We characterize mutations in dachshund, a gene that encodes a novel nuclear protein required for normal cell-fate determination of imaginal disc cells. In the absence of dachshund function, cells at the posterior margin of the eye disc fail to follow a retinal differentiation pathway and appear to adopt a cuticle fate instead. These cells are therefore unable to respond to pattern propagation signals such as hedgehog and furrow initiation does not occur. In contrast, cells in more anterior portions of the eye disc are able to differentiate as retinal cells in the absence of dachshund activity and respond normally to patterning signals. These results suggest that posterior margin cells are distinct from other cells of the eye imaginal disc by early stages of development. dachshund is also necessary for proper differentiation of a subset of segments in the developing leg. Null mutations in dachshund result in flies with no eyes and shortened legs.

The glass gene encodes a DNA-binding zinc-finger protein required for the development of Drosophila photoreceptor cells and which appears to regulate a number of genes specifically expressed in photoreceptors. We have generated monoclonal antibodies to Glass and used them to examine Glass distribution during development. Glass is expressed in all cell types of the developing eye and in all other organs that contain photoreceptor cells in Drosophila, including a small number of cells in the brain. We altered the normal pattern of glass expression by placing the gene under the control of the hsp70 promoter. Our results suggest that nonphotoreceptor cells are restricted in their response to Glass expression. In an effort to discover the mechanism of this restriction, we examined the expression of a number of reporter gene constructs. Our results suggest that nonsensory cells are unable to express certain reporter constructs in response to Glass expression because another DNA-binding factor represses Glass activity in nonsensory cells.

During Drosophila ommatidial development, a single cell is selected within the ommatidial cluster to become the R7 photoreceptor neuron. The seven-up gene has been shown to play a role in this process by preventing four other photoreceptor precursors, R3/R4/R1/R6, from adopting the R7 cell fate. The seven-up gene encodes a steroid receptor-like molecule that is expressed only in those four cells that require seven-up function in the developing Drosophila ommatidium. We have examined the functional significance of the spatially restricted expression of seven-up by misexpressing seven-up isoforms. As expected from the function that seven-up performs in R3/R4/R1/R6, ubiquitous expression of seven-up causes transformation of the R7 cell to an R1-R6 cell fate. In addition, depending on the timing and spatial pattern of expression, various other phenotypes are produced including the loss of the R7 cell and the formation of extra R7 cells. Ubiquitous expression of seven-up close to the morphogenetic furrow interferes with R8 differentiation resulting in failure to express the boss protein, the ligand for the sevenless receptor tyrosine kinase, and the R7 cell is lost consequently. Extra R7 cells are formed by recruiting non-neuronal cone cells as photoreceptor neurons in a sevenless and bride of sevenless independent way. Thus, the spatiotemporal pattern of seven-up expression plays an essential role in controlling the number and cellular origin of the R7 neuron in the ommatidium. Our results also suggest that seven-up controls decisions not only between photoreceptor subtypes, but also between neuronal and non-neuronal fates.

In a screen for mutations affecting Drosophila eye devel-opment, we have identified a gene called fat facets (faf) which is required for cell interactions that prevent par-ticular cells in the developing eye from becoming pho-toreceptors. Analysis of eyes mosaic for faf+ and faf cells shows that faf is required in cells near to, but out-side, normal developing photoreceptors and also outside of the ectopic photoreceptors in mutant facets. faf is also essential during oogenesis, and we show that a faf-lacZ hybrid protein is localized via the first 392 amino acids of faf to the posterior pole of oocytes. Posterior local-ization of faf-lacZ depends on oskar. oskar encodes a key organizer of the pole plasm, a specialized cytoplasm at the posterior pole of embryos. The pole plasm is required for germ cell formation and contains the deter-minant of posterior polarity, encoded by nanos. Although other pole plasm components are required for localization of nanos RNA or for nanos protein func-tion, faf is not. We have cloned the faf gene, and have shown that it encodes two similar large (∼300 103Mr) proteins that are unique with respect to other known proteins.