Hedgehog signaling in animal development: paradigms and principles

Since their isolation in the early 1990s, members of the Hedgehog family of intercellular signaling proteins have come to be recognized as key mediators of many fundamental processes in embryonic development. Their activities are central to the growth, patterning, and morphogenesis of many different regions within the body plans of vertebrates and insects, and most likely other invertebrates. In some contexts, Hedgehog signals act as morphogens in the dose-dependent induction of distinct cell fates within a target field, in others as mitogens regulating cell proliferation or as inducing factors controlling the form of a developing organ. These diverse functions of Hedgehog proteins raise many intriguing questions about their mode of operation. How do these proteins move between or across fields of cells? How are their activities modulated and transduced? What are their intracellular targets? In this article we review some well-established paradigms of Hedgehog function inDrosophila and vertebrate development and survey the current understanding of the synthesis, modification, and transduction of Hedgehog proteins. Embryological studies over much of the last century that relied primarily on the physical manipulation of cells within the developing embryo or fragments of the embryo in culture, provided many compelling examples for the primacy of cell–cell interactions in regulating invertebrate and vertebrate development. The subsequent identification of many of the signaling factors that mediate cellular communication has led to two general conclusions. First, although there are many important signals, most of these fall into a few large families of secreted peptide factors: theWnt (Wodarz and Nusse 1998), fibroblast growth factor (Szebenyi and Fallon 1999), TGFsuperfamily (Massague and Chen 2000), plateletderived growth factor (Betsholtz et al. 2001), ephrin (Bruckner and Klein 1998), and Hedgehog families. Second, parallel studies in invertebrate and vertebrate systems have shown that although the final outcome might look quite different (e.g., a fly vs. a mouse), there is a striking conservation in the deployment of members of the same signaling families to regulate development of these seemingly quite different organisms. This review focuses on one of the most intriguing examples of this phenomenon, that of the Hedgehog family. As with many of the advances in our understanding of the genetic regulation of animal development, hedgehog (hh) genes owe their discovery to the pioneering work of Nusslein-Volhard and Wieschaus (1980). In their screen for mutations that disrupt the Drosophila larval body plan, these authors identified several that cause the duplication of denticles (spiky cuticular processes that decorate the anterior half of each body segment) and an accompanying loss of naked cuticle, characteristic of the posterior half of each segment (see Fig. 1). The ensuing appearance of a continuous lawn of denticles projecting from the larval cuticle evidently suggested the spines of a hedgehog to the discoverers, hence the origin of the name of one of these genes. Other loci identified by mutants with this phenotype included armadillo, gooseberry, and wingless (wg). In contrast, animals mutant for the aptly named naked gene showed the converse phenotype, with denticle belts replaced by naked cuticle in every segment. On the basis of these mutant phenotypes, Nusslein-Volhard and Wieschaus (1980) proposed that these so-called segment-polarity genes regulate pattern within each of the segments of the larval body, individual genes acting within distinct subregions of the emerging segmental pattern. The first important breakthrough in unraveling how segment-polarity genes act came in the mid-1980s with the cloning of two members of the class, wingless and engrailed (en). Wg was shown to be the ortholog of the vertebrate proto-oncogene int1 (subsequently renamed Wnt1 and the founder member of the Wnt family of secreted peptide factors; Rijsewijk et al. 1987), whereas the sequence of en revealed that it encodes a homeodomaincontaining transcription factor (Fjose et al. 1985; Poole et al. 1985). Intriguingly, the two genes were found to be expressed in adjacent narrow stripes of cells in each segment (Martinez Arias et al. 1988). A close spatial relationship between Wnt1 and En expression domains was also reported in the primordial midbrain and hindbrain of the vertebrate embryo (McMahon et al. 1992). AnalyWe dedicate this review to the memory of our dear friend and colleague Rosa Beddington, whose encouragement led to our initial collaboration. 3Corresponding authors. E-MAIL p.w.ingham@sheffield.ac.uk; FAX 0114-222-288. E-MAIL amcmahon@biosun.harvard.edu; FAX (617) 496-3763. Article and publication are at http://www.genesdev.org/cgi/doi/10.1101/ gad.938601.