RNAi misses the target

signaling update home

SOURCE:

Off-target effects can dominate the results of genome-wide RNAi screens in Drosophila, but can be reduced by excluding certain sequences from dsRNAs.

It is well known that RNAi screens can produce spurious candidates through off-target effects. In Drosophila, the use of long dsRNA has been thought to minimize this problem, but Philip Beachy and colleagues now show that off-target effects can dominate the results of a genome-wide RNAi screen in this species.

The authors were looking for novel components of the Wingless (Wg) signalling pathway using more than 20,000 dsRNAs. They found seven dsRNAs that reduced expression of their reporter below a particular threshold. However, further analysis showed that the effects were specific to these dsRNAs and were not replicated by most other dsRNAs that targeted the same ORFs.

By systematically substituting bases in one of the original dsRNAs, the authors honed in on the core sequence that was required for the Wg-repressing effect. A homology search showed that this was a perfect match for part of the mRNA sequence of armadillo, a known Wg-pathway component. Equivalent homology was found in the other dsRNAs, and quantitative PCR showed that armadillo expression was decreased — in other words, all the effects were off-target.

Further analysis of this screen, an unpublished Hedgehog screen and a previously published Wg screen showed that off-target effects were prevalent. In some cases this was because of a particular trinucleotide repeat, CAN, that is found in many genes. Although these results might seem discouraging, the authors point out that excluding certain sequences from dsRNAs and always using multiple dsRNAs for each gene will significantly reduce the false-positive rate.

Patrick Goymer

References

Ma, Y. et al. Prevalence of off-target effects in Drosophila RNA interference screens. Nature 443, 359-363 (2006) | Article |