Fibroblast growth factor (FGF) signalling has been implicated in neurulation and in the induction of mesodermal cell types. How can a single signalling pathway bring about such dissimilar effects? Two recent reports in Cell dissect out the downstream signals that mediate the dual responses to FGF in two different species, shedding light on this fundamental embryological problem.

In the first study, Sheng et al. carried out a genetic screen in chick embryos, looking for genes that were differentially expressed five hours after the initiation of neural induction by FGF. This interval is needed to sensitize cells to the action of bone morphogenetic protein (BMP) antagonists, which allow ectodermal cells to adopt a neural fate by interfering with BMP signalling. The authors identified the gene Churchill (Chch), which encodes a transcription factor that is expressed with a slow time course by neural tissue in response to FGF. The functional characterization of Chch disclosed that it is required for the expression of the transcriptional repressor Sip1 (Smad-interacting protein 1), which inhibits the expression of the mesodermal gene brachyury. In addition, Chch expression arrests the movement of epiblast cells that would otherwise migrate through the primitive streak to form mesoderm. These results indicate that the decision to adopt a neural or mesodermal fate in response to FGF is regulated in time and space by Chch. As this transcription factor appears rather late after the tissue is exposed to FGF, it will be of interest to identify the factors that act upstream of Chch to control its expression.

Bertrand et al., the authors of the second paper, charted the FGF signalling pathway in the ascidian Ciona intestinalis, and discovered a different way in which FGF exerts its dual action. In the embryo of this organism, FGF induces animal cells to form neural tissue and vegetal cells to form mesoderm. Trying to understand the mechanism that governs this choice, the authors first mapped the regulatory elements of the Otx gene, the earliest known marker of ascidian neural tissue, and subsequently identified the molecules that interact with such sequences to control Otx expression. The authors found that expression of the endogenous gene Fgf9/16/20 activates an enhancer of Otx expression through the action of the transcription factors Ets1/2 and GATAa, which can directly bind this regulatory sequence. Overexpression and knockdown experiments led Bertrand et al. to conclude that, whereas Ets1/2 activity regulates the expression of Otx in both animal and vegetal cells, the regulatory role of GATAa is limited to animal cells. So, GATAa seems to be the key factor that restricts the manifestation of the neural programme in a specific cell population. But how is GATAa targeted to the right cells in the first place? Further experiments will be required to answer this intriguing question.