Discovering what genes do the high-throughput way

Researchers at the Howard Hughes Medical Institute in Boston have developed a fast and systematic method that could make it easier to understand how cells from complex animals work. Their results, published this week in Journal of Biology, should inspire scientists to perform comprehensive screens of the fruit fly genome to find molecules that control a variety of cellular processes.

The research team, led by Norbert Perrimon, systematically inhibited the function of around 1,000 Drosophila genes that are predicted to affect diverse cellular processes. They observed that 16% of the inhibited genes altered the form or structure of the cells in some way.

Genes that caused the same changes in the cells when inhibited are likely to work together in a complex or pathway. Clustering genes by their effects allowed the researchers to assign functions to about 50 previously uncharacterised genes. Author Buzz Baum says, “The most exciting thing for me is that now you can take a step back and look at the bigger picture. You can find out which genes act together to do something, so you begin to build up a system-wide understanding of how cells work. Genes work in a community to do something, not on their own. With big-scale experiments you can start to see the internal logic of the cell.”

The screening method makes use of RNA interference (RNAi) – introducing double stranded RNA into cells, to interfere with the expression of specific genes. In order to scale up the procedure, which normally tests one gene at a time, the researchers plated out cells into 384-well dishes and then added double stranded RNA to each well. After three days, when the targeted gene should be inhibited, they stained the cells so that they could visualise both DNA and components of the cytoskeleton. They then photographed the cells using an automated microscope.

Two postdoctoral researchers, Baum and Amy Kiger, independently studied the thousands of photographs generated by the screen to characterize the effects of the RNAi treatment on the cells. They created a formal set of criteria to judge the cells consisting of seven classes of change that could have been induced. These included changes to cell number, shape, size and viability. Any changes were only considered significant if both scientists recorded them in replica experiments.

The researchers were also keen to find out if their method could be used to screen for genes that worked in, or inhibited specific molecular pathways. Screens in whole flies for genes that modify the effect of a particular genetic mutation have proved powerful, though time consuming. By adding two sets of double stranded RNA to each well, one that targeted the tumor suppressor gene pten and the other the gene to be tested, the researchers found that they were able to identify genes that modified the effects of inhibiting pten in cells. “These results demonstrate that modifier screens, as previously done in vivo, can be extended to RNAi screening methodology in cell culture”, write the researchers.

Drug companies are becoming interested in this technique, as the rate-limiting step in cell-based drug discovery is finding out which protein is inhibited by a particular drug. Using Perrimon’s method they can screen to see which gene, when inhibited, changes the cell in the same way as adding the drug.

“RNAi screens can complement classical Drosophila genetics to assign functions to both known and novel genes,” write the researchers. “The same technology can be easily adapted to a wide variety of cell-based studies and a greater genomic scale.”

Baum says, “The major difference between this and whole fly screens is that here we can be systematic. We can choose to look at any cell biological process and systematically test the set of genes that could be involved. In the future we will be able to screen the full genome in a few weeks, and look at any cell biological phenomenon.”

This press release is based on the following article:

A functional genomic analysis of cell morphology using RNA interference
A A Kiger, B Baum, S Jones, M R Jones, A Coulson, C Echeverri, N Perrimon
Journal of Biology 2:27
http://jbiol.com/content/2/3/27
Published 1 October 2003