Embryonic stem cells can develop into almost any type of cell. What cues prompt this remarkable transformation? This was a question that Ihor Lemischka, a molecular biologist at Princeton University in New Jersey, was keen to answer. He wanted to find out more about the genes that determine when and how mouse stem cells differentiate and renew, hoping that this information would one day be useful for human stem-cell-based therapeutics.

Lemischka's group opted to use RNA interference (RNAi), an increasingly popular technique for selectively reducing the function of individual genes. “We wanted to study whole ensembles of gene products at the same time and not be limited by the complications of traditional genetics approaches,” he says.

Lemischka's group, led by postdoc Natalia Ivanova, began by using microarrays to identify gene products required by embryonic stem cells to remain as stem cells. They looked for genes that are rapidly turned off after the cells begin to diversify, reasoning that these genes were potentially crucial to this process. The screen yielded 900 candidate genes. The researchers focused on the 65 genes that encoded transcription factors, which are known for their role in development, and five other genes of interest from previous work.

As reported on page 533, they used RNAi to knock out the function of each gene, one at a time. Next, they monitored how well the cells continued to divide. Ten genes seemed to have an important role, seven of which were particularly active.

The group found seven genes that are required by mouse embryonic stem cells for efficient self-renewal in vitro. In particular, they were interested in how other genes would be affected when these transcriptional regulators were inactive.

Ihor Lemischka (right) with Natalia Ivanova.

More than three years ago, Lemischka became intrigued by the potential for systems-biology approaches. He hopes, one day, to identify the heart of the stem cell's 'self-renewal machine', and then to tweak the system by modulating levels of gene products. “We're really at the very, very beginning of understanding how this works,” he says. As part of a collaboration with researchers at Cold Spring Harbor Laboratory in New York, he is expanding his search to find proteins that give mouse embryonic stem cells the ability to develop into so many cell types.

The next step will be to move into human embryonic stem-cell work. Lemischka says the federally approved US embryonic stem-cell lines will probably be sufficient at this stage; he would prefer to avoid the extra costs and effort involved in setting up a separate lab to study privately funded cell lines. Although his methodology will transfer to human systems, the differences that might exist between mouse and human embryonic stem cells remain unclear. But comparing stem-cell differentiation in the two species should provide some interesting insight into developmental biology — and perhaps help to translate that work to clinical treatments. “It's an absolute thrill to think that something we're doing with mouse stem cells might actually impact on clinical medicine in the future,” Lemischka says.

He adds that he has found the lab work so exciting that he's considering spending more time at the bench. “It's a scary thought for people in the lab,” he jokes.