Drosophila melanogaster has been a favored tool for genetic studies for over 90 years and is an excellent model system to identify genes involved in developmental and cellular processes. The contribution made by studies in Drosophila are numerous, and many important discoveries were made first in this organism. In particular, Drosophila is a model of choice for developmental studies, signal transduction, cell biology and immunity. The Drosophila genome is about 180 Mb in size and a third of which is centric heterochromatin. Completion of the sequence of the Drosophila euchromatic regions provides us with an unprecedented resource, as we can now fully evaluate the degree of conservation of this organism with others (Adams, M. et al. 2000. The genome sequence of Drosophila melanogaster. Science 287: 2185-2195). The relevance of Drosophila to humans is perhaps best illustrated by the realization that more than 60% of the genes identified in human diseases (177 out of 289) have counterparts in Drosophila (Rubin G.M. et al. 2000. Comparative genomics of the Eukaryotes. Science 287: 2204-2215).

The initial analysis of the Drosophila genome by the Berkeley Drosophila Genome Project (BDGP) and Celera has led to the annotation of 13,600 genes. Interestingly, the current literature discusses only approximately 20% of these genes, and only half of these have been characterized genetically. Thus, it is clear that a wealth of information remains to be mined from this model organism. Although conventional genetic approaches will clearly continue to provide valuable information, new powerful methods that can systematically and quickly analyze the functions of all ~14,000 predicted genes in specific assays are needed.

The recent advent of RNA-mediated interference (RNAi) in Drosophila cell cultures provides a direct and powerful method (Clemens et al., 2000. Use of double-stranded RNA interference in Drosophila cell lines to dissect signal transduction pathways.  Proc. Natl. Acad. Sci. USA 12: 6499-6503.) to test the effects of specific gene disruption, yielding efficient phenocopy of loss-of-function mutations. The availability of gene-specific sequences from the complete genome in combination with RNAi application in cell cultures allows the remarkable opportunity to design a functional genomic approach to many cell biological and signal transduction processes. Towards this goal we have generated a set of 21,000 dsRNAs that cover every annotated gene in the Drosophila genome (in collaboration with Dr. Renato Paro’s group), and have developed high-throughput RNAi screens using Drosophila cells cultured in 384-well plates. The use of model organisms to elucidate gene functions and molecular pathways has relied on genetic epistasis tests and modifier screens. Our current work demonstrates that the powerful classical genetic tools, such as loss of function analyses and genetic modifier screens, can be conducted in cell culture now as previously done in vivo.

Method

The screens consist of cell-based assays conducted in high density, 384-well tissue culture plates. The methodology consists of three major steps: (1) array gene-specific dsRNAs into 384-well assay plates, (2) add cells to assay plates, then (3) incubate and assay the cells (Figure 1).

(1) First, gene-specific dsRNAs are transcribed and then aliquoted into unique wells of multiple 384-well assay plates using robotics. A full-genome screen involves fifty six 384-well assay plates. Full-genome sets of pre-aliquoted dsRNA assay plates are stored at –80˚, readily available for future screens. Given the robustness of RNAi and the small volume that can be assayed in the high-density wells (~10-30 ul), only small amounts of the total transcribed dsRNA are needed for any one screen.  We determined, consistent with what has been published, that dsRNA concentrations of 25-75 nM (~0.2 ug of 500 nt length) are sufficient to diminish or deplete endogenous mRNA levels, concomitant with the appearance of penetrant phenotypes.

(2) Once a robust cell-based assay is developed, cells are plated directly into the dsRNA-containing assay plate wells.  The cells can be modified as needed for the assay. Cells are uniformly and rapidly dispensed into the 384-well plates using a MultiDrop liquid dispenser at around 10⁴ cells per well, with some variation depending on the cell line and assay timecourse.  For efficient RNAi, the cells are first exposed to the dsRNAs in serum-free medium.  After one-half an hour in the serum-free medium, the MultiDrop is used again to add additional culture medium with serum for the remainder of the incubation time.

(3) After the appropriate incubation time, as determined per assay, cells can be either first exposed to a specific treatment/induction or directly analyzed for the assay read-out. Depending on the method for data acquisition, the read-out can be based on any of a wide-range of assays that rely on the detection of either fluorescence or luciferase activity. There are two methods for data acquisition from the 384-well plates, using either a plate reader or automated microscopy.  The plate reader rapidly quantifies relative luciferase or fluorescence levels to a single numerical read-out representing the entire cell population per well. Alternatively, an inverted fluorescent microscope with automated acquisition software (Metamorph, Universal Imaging), or “autoscope”, can be used to automatically track, focus, and capture fluorescent images of the cells within each well across an entire plate.  The image data provides information on fluorescence levels and/or localization for each cell within the cell population by either manual visualization or automated analysis.