But because each microRNA may target hundreds of messenger RNAs-and it hasn't been clear how effectively it targets each of those messenger RNAs-scientists have not been able to assess the ultimate impact of microRNAs on protein production.

Now, in the first large-scale study of its kind, scientists at Whitehead Institute and Harvard Medical School have looked more comprehensively at the protein output of genes targeted by specific microRNAs. The results are reported in the July 30, 2008 online issue of Nature.

"This is the first time a large-scale data set has been used to observe the effect of microRNAs on the production of thousands of proteins," says Whitehead Member and Howard Hughes Medical Institute Investigator David Bartel, who is co-senior author on the study. "Previous technology enabled us to only look at how microRNAs affect messenger RNAs. That provided us with crucial information about their involvement in the first stage of protein translation but not their ultimate effect on gene expression."

The study confirms that microRNAs typically fine-tune protein production, providing much less of a repressive effect than has been seen for some transcription factors, proteins that can also act as gene regulators. The study also shows that microRNAs mainly operate by degrading target messenger RNAs rather than simply throwing a wrench into protein translation.

Additionally, the researchers showed that microRNAs produce their greatest impact when they interact with a specific part of the messenger RNA molecule that doesn't code for proteins.

Working in collaboration with the lab of co-senior-author Steven Gygi, associate professor of cell biology at Harvard Medical School, the Bartel lab introduced individual microRNAs into human cells and then used quantitative mass spectrometry to measure the resulting changes in protein levels.

In addition, working with Whitehead Fellow Fernando Camargo, researchers used a knockout mouse model lacking a single microRNA (microRNA-223) known to influence the production and function of white blood cells called neutrophils. Researchers then measured and compared the protein levels of neutrophils from normal mice and mice lacking microRNA-223.

"By measuring protein levels, we could illustrate that microRNAs fine-tune gene expression rather than simply working as a switch to turn a gene on or off," says Daehyun Baek, co-lead author on the paper and a postdoctoral fellow in Bartel's lab. "The knockout mouse model enabled us to more accurately gauge the interactions of microRNAs with their regulatory targets, and to confirm what we found earlier in human cells."

"Surprisingly, the protein levels of cells missing microRNA-223 nearly matched those of normal cells," adds Gygi. "This microRNA changed the expression level of hundreds of proteins, but only by a small amount."

Identifying the proteins influenced by the presence of miR-223 also enabled the researchers to evaluate different predictions that scientists had made regarding which genes are regulated by each miRNA.

"We still cannot predict perfectly which genes are targeted by a particular microRNA, but at least now we know which of the proposed prediction methods are most useful," says Bartel.

By comparing protein levels and measuring RNA levels, the researchers demonstrated that microRNAs are more likely to change gene expression by destabilizing messenger RNAs rather than simply fouling up the protein translation process.

"There had been a suspicion that the majority of microRNA regulation would happen in a manner that doesn't really change the amount of messenger RNA," says Bartel. "We found that this wasn't true, at least for this particular microRNA."

Cristin Carr.

wi.mit