A Drosophila Axin homolog

Like in many other signaling pathways, most components in Wnt signal transduction are highly conserved in evolution. In organisms ranging from C. elegans to mice, a similar hierarchy of signaling molecules transmits the Wnt signal to the nucleus. In current models of this pathway, the secreted Wnt protein is received by Frizzled-like cell surface receptors. In the cytoplasm, Dishevelled (Dsh) relays the signal to a complex of several proteins, including the protein kinase glycogen synthase kinase-3β (GSK-3β) and β-catenin (see URL: http://www.stanford.edu/~rnusse/wntwindow.html and reviewed by Cadigan and Nusse, 1997). In this complex, the β-catenin protein becomes targeted for degradation after being phosphorylated by GSK-3β (Aberle et al., 1997; Yost et al., 1996). After Wnt signaling and the resulting down-regulation of GSK-3β kinase activity (Cook et al., 1996), β-catenin escapes from degradation. β-catenin then can form a complex with the nuclear DNA binding protein TCF (Behrens et al., 1996; Molenaar et al., 1996; Morin et al., 1997) and participate in transcriptional activation of Wnt target genes (Korinek et al., 1998). Most components of the Wnt signaling pathway were identified in genetic screens in Drosophila, where the Wnt gene wingless (wg) patterns the embryo through the β-catenin homolog armadillo (arm). However, other molecules that regulate β-catenin levels have emerged from mammalian systems. APC, for example, which binds β-catenin to facilitate its degradation, was discovered as a human tumor suppressor gene (reviewed by Polakis, 1997). Subsequent cloning of the Drosophila homolog of APC showed that it is involved in regulating Arm in photoreceptor cells in the eye (Ahmed et al., 1998). Wnt signaling in the C. elegans embryo is also regulated by an APC related protein, Apr-1 (Rocheleau et al., 1997). Another component in β-catenin regulation is the vertebrate Axin protein, which binds to β-catenin, GSK-3β and APC (Behrens et al., 1998; Hart et al., 1998; Ikeda et al., 1998; Itoh et al., 1998; Kishida et al., 1998; Nakamura et al., 1998; Sakanaka et al., 1998; Yamamoto et al., 1998). Axin is the product of a mouse gene, fused, that affects embryogenesis (Zeng et al., 1997). An Axin-related vertebrate protein, Conductin/Axil, has been identified as a binding partner for βcatenin (Behrens et al., 1998; Yamamoto et al., 1998). Axin and Conductin overexpression phenotypes in Xenopus indicate that these proteins are negative regulators of Wnt signaling (Behrens et al., 1998; Itoh et al., 1998; Zeng et al., 1997), in agreement with the biochemical interactions between Axin and β-catenin. The consequences of loss of function of Axin are not entirely clear; there are several alleles of the mouse fused gene but it is not established whether these are nulls (Vasicek et al., 1997). Other Axin-like genes in Drosophila or C. elegans would provide important information on their genetic requirements in Wnt signaling, but it is only recently that they have been identified. The worm genome actually may not contain an Axin homolog (Ruvkun and Hobert, 1998). We therefore were interested in identifying Axin in the fly. By screening the Drosophila EST database, we have found a fly Axin homolog, Daxin. Using double-stranded RNA interference methods, we generate a loss-of-function phenotype and show that loss of Daxin in the embryo mimics overexpression of wg and of arm. Conversely, overexpression of Daxin is similar to loss of wg. In addition to interfering with the Wg signal, we show that Daxin overexpression can block the DWnt-2 signal. We have also examined the Daxin protein and show that in embryo lysates it interacts with Arm and 4165 Development 126, 4165-4173 (1999) Printed in Great Britain © The Company of Biologists Limited 1999 DEV8604