he paper at #8 offers researchers interested in gene function a powerful new tool to study the role of particular genes. With it, they will be able to switch off specific sequences selectively. Ultimately, there is hope for a new therapy to tackle genetic disorders.

Thomas Tuschl and his colleagues at the Max Planck Institute for Biophysical Chemistry in Gottingen, Germany, have extended the phenomenon called RNA interference (RNAi) to mammalian cells. RNAi, discovered in the late 1990s, has rapidly become indispensable. It works like this: Once a gene sequence is known, a piece of double stranded RNA (dsRNA) is made that matches part of the gene. Inserted into the cell, the dsRNA is broken down into short pieces that somehow—the mechanisms are not yet well understood—act to destroy the messenger RNA of the target sequence. By studying the effects of interfering with a particular gene, researchers can deduce its function.

While RNAi has proved very useful in the study of invertebrates and plants (and is therefore a very ancient pathway), in mammals the presence of dsRNA in the cell triggers a non-specific shutdown of all protein synthesis. This is somewhat surprising because specific RNAi does occur in mouse egg cells and early embryos. Tuschl’s group had previously shown that the specific shutdown can be triggered by the short interfering RNA (siRNA) cut from the larger dsRNA. And they knew that the non-specific shutdown needs dsRNA more than about 30 base pairs long. So they created 21 and 22 bp siRNAs (the same length as natural siRNAs) and put them directly into mammalian cells to see whether that would bypass the non-specific shutdown. It did.

The group’s first targets were various luciferase genes that had been inserted into four different kinds of mammalian cell line. In every case, the luminescence of the cells was reliably and specifically reduced. It was not always entirely suppressed, as it is in Drosophila, but this may be because expression of luciferase is much higher in mammalian cells. Larger dsRNA corresponding to the luciferase reporter genes also reduced expression, but entirely non-specifically, as expected. Superimposed on the general suppression, however, was a specific component, probably triggered by siRNAs created from the long dsRNA. This suggests that the specific and non-specific interference pathways are at least partially independent.

Having shown that siRNAs work specifically on inserted reporter genes, Tuschl and his colleagues turned their attention to endogenous genes. Again, the siRNAs interfered specifically with their targets but not with any other genes. One target gene, vimentin, was not silenced as expected. Tuschl’s group suggests that this could be because vimentin is very abundant--several percent of the total cell mass--or because the siRNA sequence was somehow not optimal for interfering with the vimentin mRNA.

This suggests that the use of siRNAs to study gene function in mammalian cells will not be plain sailing. It will depend both on the cell type and the gene under study. In that respect it does not differ from the main method in use before, antisense RNA. This involves creating an RNA strand complementary to the mRNA of interest, which binds to and thus silences the mRNA. But Tuschl and his group have already shown that siRNAs can effectively silence genes at concentrations several times lower than those used for antisense experiments.

The non-specific shutdown caused by long pieces of dsRNA is part of the cellular response that leads to synthesis of interferon. The dsRNA activates two enzymes, one of which blocks the translation of DNA into RNA while the other triggers a ribonuclease that destroys existing mRNA. Some of the citation popularity of the Tuschl paper reflects a search for the corresponding mechanisms underlying specific RNAi. Far more, however, represents RNAi being put to work directly to discover how particular genes work and what they do. And, as ever, researchers are already dreaming of ways in which RNAi might be employed to silence aberrant genes of medical importance.