Two systematic RNAi screens in worms provide the first large-scale reverse genetic analyses of a multicellular organism.
By William Wells(firstname.lastname@example.org) | November 20, 2000
In the 16 November Nature, Fraser et al. and Gönczy et al. present the first large-scale reverse genetic analyses of a multicellular organism (Nature 2000, 408:325-330; Nature 2000, 408:331-336). Fraser et al. use RNA-mediated interference (RNAi to target 2,416 predicted genes on chromosome I of the worm Caenorhabditis elegans by feeding the worms with bacteria expressing double-stranded RNA. Of the analyzed genes, 13.9% show a phenotype, increasing the number of sequenced chromosome I genes with a known phenotype from 70 to 378. The identified genes include 90% of known embryonic lethal genes from chromosome I, but only 45% of genes with known post-embryonic phenotypes, with genes involved in nerve and sperm cell function apparently resistant to RNAi. The majority (60%) of the phenotypes were embryonic lethal, including many genes involved in basic metabolism. The largest class of post-embryonic phenotypes are in the uncoordinated (Unc) class, which generally relate to neuromuscular function. Extrapolating from this screen, Fraser et al. estimate that the worm requires about 5,400 genes to live under standard laboratory conditions.
Gönczy et al. target 2232 genes from chromosome III using double-stranded RNA injected into worm gonads. They use time-lapse microscopy to look for any phenotype affecting cell division in the early worm embryo. There are 133 genes (around 6%) that show a definite phenotype, suggesting that a total of over 1,000 worm genes are essential for the first two cleavage divisions. From the microscopic observations, the genes are grouped into classes involved in processes such as nuclear migration, cytokinesis and spindle positioning. Worm genes that have orthologs in both flies and yeast represent only 12.9% of genes tested, but they comprise 47.3% of those associated with a cell division phenotype.
Despite the best of intentions, sometimes a Western blot goes bad. When that happens, you can cry into your blocking buffer (not recommended), or you can interpret the signs your Western is sending and address them! Can you read between the bands and determine where these blots went bad?