RNA interference (RNAi) has emerged as a powerful tool to systematically dissect gene function by depleting transcripts through the introduction of homologous short or long double-stranded RNAs. Discovered in C. elegans and plants, RNAi is now also widely employed for gene-silencing studies in many model systems, including Drosophila and human cells. With the availability of and the development of genome-wide libraries of RNAi reagents, combining it with high-throughput screening techniques it can be used to query genome for a variety of cell or organism-based phenotypes.
Screening in Drosophila cells
RNAi in cultured Drosophila cells can be triggered by adding in vitro generated double-stranded RNAs to the cell culture medium. Because of the relative ease and the availability of libraries, a large number of Drosophila RNAi screens have been done using cell-based assays. Screening experiments can be performed for a small number of genes, for example those in specific functional groups, to genome-wide applications of libraries. Compared to screens in mammalian cells, Drosophila has the key advantages of relatively low genetic redundancy and high efficacy of RNAi reagents.
Cell-based assays can measure a diverse set of phenotypes, ranging from homogenous cell viability readouts to alterations in reporter gene expression to high-content assays using automated microscopy. Luciferase reporters are especially amenable for screening and have been used to identify genes involved in signalling pathways, including JAK/STAT, Wnt and IMD and others.
Screening in human cells
Because long dsRNAs activate interferon responses leading to apoptosis, RNAi screens in human cells use siRNAs of 21-23nt long that evade the interferon response. Genome-wide siRNA (shRNA) libraries that cover almost the complete human genome with several independent siRNAs are available from several commercial companies and academic groups. Since the knock-down efficiencies of siRNAs vary, multiple independent siRNAs per gene are required. These can be screened individually or in pools. Pooling of siRNA can reduce costs of screening but increases the likelihood that the phenotype is caused by an off-target effect of a single siRNA. Validation experiments are usually performed by retesting positives with multiple independent siRNAs. Although human siRNA libraries are expensive compared to libraries generated in-house for Drosophila and C. elegans, a small aliquot is usually sufficient for a large number of screens in a high-throughput format (such as 384-well plates).
1. Boutros M. and Ahringer, J. (2008). The art and design of genetic screens: RNA interference. Nature Rev. Genet., 9:554-566