The mechanisms behind the action of Wingless (Wg) and its vertebrate homologue Wnt have slowly been unravelled over the past few years. But there are still surprises. New work published in Nature and Nature Cell Biology shows that Dickkopf (Dkk), an antagonist of the Wnt pathway, inhibits signal transduction through a completely unsuspected mechanism.

We know that Wg/Wnt signals through the seven-transmembrane receptor Frizzled (Fz) to inhibit phosphorylation of Armadillo (Arm)/β-catenin. Without activation of the Wg/Wnt signals, phosphorylation of Arm/β-catenin is maintained by glycogen synthase kinase-3β (GSK-3β). When Wg/Wnt is coupled to Fz, GSK-3β is inhibited by a large cytoplasmic complex, which includes Dishevelled and adenomatous polyposis coli (APC). This inhibition of GSK-3β allows unphosphorylated Arm/β-catenin to shuttle into the nucleus and affect the transcription of downstream target genes through the TCF/LEF transcription factor.

Cerberus, Wnt inhibitory factor (WIF), Frizzled-related proteins (FRPs) and Dickkopf (Dkk, which means 'fat head' in German) are all known antagonists of the Wnt pathway, but the mechanism by which Dkk inhibits Wnt has remained unknown. Previous work has shown that Cerberus, WIF and the FRPs all act by binding and sequestering Wnt, preventing the signal from interacting with its receptor. Mao et al. and Bafico et al. now show that Dkk does not bind to either Wnt or Fz.

As both groups had seen that Dkk is part of a large cellular complex, they identified the components of this complex that interact with Dkk. The two groups showed that LRP6 and LRP5 — which encode low-density lipoprotein-related co-receptors and act as positive regulators of the Wg/Wnt pathway — are specific, high-affinity receptors for the two Xenopus laevis Dkk homologues, Dkk1 and Dkk2. The domains necessary for this interaction were identified and are distinct from the regions of LRP6 that interact with either Wnt or Fz. Bafico et al. went on to show that overexpression of LRP6 markedly interferes with Dkk inhibition of Wnt signalling — if excess LRP6 is present, the concentration of Dkk needed to prevent Wnt signalling is increased 100-fold.

The Wg/Wnt pathway is vital during the development of many organisms including Drosophila melanogaster and Xenopus, and APC and β-catenin have been implicated in human cancer. So it is imperative to understand how antagonists of the Wg/Wnt pathway work. It seems that Dkk works by binding to the LRP6 and LRP5 co-receptors, which increase the interaction between Wnt and Fz and might also interact directly with the β-catenin degradation complex to inhibit Wnt signalling in an indirect manner. As this is the first inhibitor of the Wg/Wnt pathway found to act in this way, this work should expand our understanding of this increasingly important signalling pathway.