Fibroblast growth factors (FGFs) are thought to specify distal fates along the proximal–distal axis in the mouse limb.
It has been known for 50 years that the apical ectodermal ridge (AER) is essential for the development of the tetrapod limb along its proximal–distal (P–D) axis. Fifteen years ago it was shown that members of the fibroblast growth factor (FGF) family are the vital signals produced by the AER. But what do these FGFs do? A series of genetic studies shows that, contrary to the long-standing view, these molecules are not merely permissive but are instructive for limb P–D patterning in the mouse.
Four FGF genes are expressed in the AER: Fgf4, Fgf8, Fgf9 and Fgf17. To assess how these factors contribute to P–D patterning, the phenotype of the forelimb was examined in several FGF gene mutant combinations. The limb skeletal pattern of Fgf4;9,17 triple-knockout mutants was indistinguishable from wild type — an unexpected finding that revealed that Fgf8 is sufficient for normal limb development. Does this mean that the other three FGF genes are functionally redundant? Knocking out Fgf8 plus one or more of the other three FGF alleles generated a series of phenotypes with increasing severity — a finding that suggested that all four FGFs produced by the AER can contribute in the same way to limb patterning, but each to a different degree. By analysing both the skeletal phenotypes and the strength of FGF signalling, the relative contribution made by each FGF gene to limb development could be assessed — with Fgf8 contributing the most, followed by Fgf4, Fgf9, and then Fgf17.
The conclusion that FGFs are instructive for distal cell fates was reached by following the expression of Meis1, a homeobox transcription factor that is expressed in the proximal portion of the limb bud and that has been proposed to be involved in proximal specification. Because there is a stage at which mutant limb buds with progressively less FGF signalling are of the same size, the effect of reducing FGF signalling on limb P–D patterning could be determined. As FGF signalling was reduced, the Meis1-positive proximal domain occupied a greater proportion of the limb bud, whereas the Meis1-negative distal domain occupied a smaller proportion. Thus, the authors conclude that FGF signalling 'instructs' the P–D pattern of gene expression during limb development.
How can this information be incorporated into a model of P–D patterning? In the prevailing models, proximal-to-distal fates would be acquired either gradually or all in one go in the early limb bud. A third model, presented here, postulates that proximal and distal domains are specified by two separate signals — one (which might be retinoic acid) that specifies proximal fates and one (contributed by FGFs) that specifies distal fates — and that the interaction between the two domains or two signals specifies a third, central domain. Crucial tests of this model will rely on following markers of different limb segments, which have yet to be discovered.
Tanita Casci
References
Mariani, F. V. et al. Genetic evidence that FGFs have an instructive role in limb proximal–distal patterning. Nature453, 401–405 (2008) | Article | PubMed |
