Qi-Long Ying, University of Southern California

Robust, well characterized embryonic stem (ES) cell lines exist only for monkeys, humans and mice. ES cell lines from other species could be a huge boon for all sorts of applications, from creating transgenic livestock to using larger animals as models to study potential stem-cell therapies. The biggest barrier to obtaining stem cell lines from more species has been the lack of appropriate culture conditions: no one knows what the cells require to sustain self-renewal. A paper published this month in Nature1 might change that.

A team led by Austin Smith reports that mouse ES cells can be maintained without adding any of the growth factors, feeder cells or other conditions generally thought necessary to stimulate self-renewal. Indeed, the researchers conclude that instead of finding collections of species-specific 'on' switches for self-renewal, it just might be possible to flip universal 'off' switches for differentiation. If that's the case, the derivation of ES cells from additional species might happen soon.

The work began when Qi-Long Ying, the paper's first author and at the time a postdoc in Smith's laboratory at the University of Edinburgh, UK, decided he wanted to find a way to grow mouse ES cells without using the cytokine LIF (leukemia inhibitory factor). Smith had identified the cytokine in the 1987 as one of the most important ingredients secreted by the layer of feeder cells on which mouse ES cells are typically grown. LIF activates the intracellular JAK-STAT3 signalling pathway, which is now known to promote self-renewal. “The more experiments I did, the more I realized how important LIF was for mouse ES cells,” recalls Ying. “I spent two or three years without progress.”

Then Ying came upon a compound with contradictory roles. Non-specific inhibitors of glycogen synthase kinase 3 (GSK-3) are sometimes added to ES cell cultures to promote growth. Acting on advice from co-author Philip Cohen at the University of Dundee, UK the researchers decided to try a more specific inhibitor, the aminopyrimidine CHIR99021. When Ying added this compound to ES cells, he noticed that this inhibitor promoted both self-renewal and differentiation. Low concentrations of the compound could, even in the absence of LIF, temporarily sustain self-renewal of up to 5% of ES cells. High concentrations in addition to LIF enhanced cell survival but also enhanced differentiation.

Ying realized that he could stop the differentiation by adding a second chemical, one that inhibits an important differentiation pathway known as mitogen-activated protein kinase (MAPK) signaling. “I tested it the next day and found it really worked,” says Ying. But there was a problem: “differentiation is blocked [initially,] but after two weeks the cells undergo differentiation and start dying.” So Ying added a third chemical that inhibited differentiation, and also began optimizing the concentrations of the three inhibitors, to get the right balance to maintain ES cells over the long term.

Eventually, Ying and colleagues found a recipe using the GSK-3 inhibitor CHIR99021, which, among other effects, stimulates the WNT/beta-catenin pathway, plus the small molecules PD184352 and SU5402, which inhibit pathways that activatedifferentiation. These pathways involve, respectively, the Erk cascade and autoinductive fibroblast growth factor-4 receptor tyrosine kinase, both of which feed into MAPK signalling. If these three small molecules were added to ES cell cultures, the cells grew and proliferated normally, and expressed protein markers characteristic of ES cells.

But that still was not sufficient evidence that these cells were truly independent of LIF, because the mouse ES cells could actually be promoting their own self-renewal by secreting LIF themselves. To be sure, the research team derived ES cells from mouse embryos in which the STAT3 gene was knocked out ES cells, making them incapable of responding to LIF. These cells could not grow in traditional culture conditions, with serum and additional LIF, but they could grow in the three-inhibitor combination without LIF or serum. Ying and colleagues also performed additional controls, such as newly deriving mouse ES cell lines to make sure the cells had not somehow pre-adapted to culture conditions.

The next step is to show whether the strategy of stopping differentiation, rather than promoting self-renewal, will ease growing conditions for ES cells from other animals. Ying is convinced that he and Smith have already derived rat ES cell lines. In 2006, Ying established his own lab at the Keck School of Medicine at the University of Southern California in Los Angeles, and now they are waiting for the key evidence that these new cell lines are truly ES cells. For this, they need to show that these cells, when mixed into a normal rat embryo, can contribute to all tissues, including the gametes. The chimeric rats that have grown up so far have not demonstrated this, but Ying hopes the rat that will has already been born.