The Westerfield Lab Institute of Neuroscience 1254 University of Oregon Eugene, OR 97403-1254 USA

Inductive interactions between cells regulate many aspects of vertebrate development, including formation and patterning of the nervous system and body. To understand how inductive signals regulate the specification and differentiation of different types of skeletal muscle cells, we have studied muscle precursors in zebrafish embryos. We have identified precursor cells that give rise to fast muscles, slow muscles and muscle pioneers, a subset of the slow muscles, and we have characterized the inductive role of the notochord in establishment of the muscle pioneer cell fate. Our results indicate that the fates of fast and slow muscle precursor cells are regulated by competing influences of two key signaling molecules, Hedgehog and TGF§.

Signaling by members of the Hedgehog family has been shown to function in many developmental systems. The signaling molecule, Sonic hedgehog (Shh), was originally discovered in Drosophila and homologues were subsequently found in vertebrates. Several lines of evidence suggest that Hedgehog signaling is transduced intracellularly by two transmembrane proteins, Patched and Smoothened, that form a receptor complex (Hammerschmidt et al., 1996; Ingham, 1998). Patched is thought to be the receptor that binds Hedgehog and induces conformational changes in itself and Smoothened. Genetic analysis in Drosophila suggests that Patched suppresses Smoothened in the absence of Hedgehog signaling. Smoothened is believed to activate an intracellular signaling cascade constitutively, but is repressed by Patched when the receptor complex forms. The biochemical properties of Hedgehog, Patched and Smoothened proteins have been studied in vitro. Mutations in PATCHED lead to nevoid basal cell carcinoma in humans and mice mutant for Patched have severe developmental defects. Homozygous mutant mice defective for Shh are severely cyclopic and lack floorplate and motoneurons (Chiang et al. 1996). Recently, several mutations have been isolated in zebrafish that affect the Hedgehog signaling pathway. Paradoxically, mutation of the zebrafish shh gene in sonic-you (syu) mutants leaves the floorplate apparently unaffected and syu mutants develop both primary and secondary motoneurons, although these cells form abnormal axonal projections. Thus, in contrast to mice, Shh does not appear to be required for floorplate or motoneuron induction in zebrafish. Alternatively, because zebrafish have multiple hedgehog genes, another member of this family in addition to or other than Shh may be sufficient to signal floorplate and motoneuron induction.

To distinguish among these possibilities, we analyzed new mutations that affect Hedgehog signaling in zebrafish. We analyzed two alleles of smooth muscle undeveloped, smu(b577) and smu (b641). The smu mutants have reduced numbers of primary and secondary motoneurons and have a very poorly formed floorplate, in addition to cyclopia and other defects in systems that express members of the hedgehog gene family during early development. The smu mutations map to the same genetic map location as the smoothened gene. Wild-type smoothened cDNA rescues both the motoneuron and floorplate defects in smu mutants. Thus, we conclude that the smu mutation affects the smoothened gene. We suggest that Smoothened acts as a common point in the signaling pathways of various members of the Hedgehog family. These results may thus reconcile the apparent discrepancy between mouse and zebrafish. In both systems, Hedgehog signaling induces motoneurons and floorplate, but in zebrafish one or more Hedgehog family members other than Shh are sufficient.