br Conflict of interest statement br Acknowledgments
Conflict of interest statement
Introduction Hedgehog (Hh) signaling controls key steps of development in most tissues and organs of invertebrates and vertebrates (Briscoe and Therond, 2013, Ingham et al., 2011, Wilson and Chuang, 2010). The unique cellular composition and morphological movement in individual tissues require distinct modes of Hh signaling. For example, in the mammalian neural tube and limb, Hh acamprosate calcium from a localized source, such as the notochord/floor plate and the zone of polarizing activity (ZPA), is known to exert dose-dependent long-range signaling effects on tissue patterning. By contrast, in several branching organs such as the lung, epithelial Hh signaling to the mesenchyme mediates critical aspects of epithelial-mesenchymal interactions that drive lung branching morphogenesis. Hh signaling thus generates different outputs in diverse tissues, which underlie cellular changes during tissue patterning. Uncovering the whole complement of Hh targets and how they control cellular changes in each tissue is required for understanding the development of a given tissue. This knowledge will also contribute to our mechanistic understanding of tissue regeneration and repair and cancer development, in which Hh signaling is frequently activated (Barakat et al., 2010, Bijlsma and Roelink, 2010, Scales and de Sauvage, 2009). The Hh pathway has been extensively studied for two decades, culminating in a basic framework of mammalian Hh signal transduction that depends on Gli transcription factors (Gli1-3) to mediate Hh responses (Beachy et al., 2010, Chen and Jiang, 2013, Eggenschwiler and Anderson, 2007, Farzan et al., 2008, Hui and Angers, 2011, Rabinowitz and Vokes, 2012, Robbins et al., 2012, Ryan and Chiang, 2012, Wang et al., 2007). Gli3 (and to some extent Gli2) undergoes limited proteolysis in the absence of the Hh ligand to produce a transcriptional repressor (Pan et al., 2006, Wang et al., 2000). Hh signaling not only inhibits proteolysis of Gli proteins but also promotes the conversion of Gli proteins (primarily Gli2) into transcriptional activators. Gli1, like Ptch1 and Hhip, is a transcriptional target of Hh signaling and Gli1 induction is believed to amplify Hh responses. The combinatorial effects of Gli activators and repressors likely mediate graded Hh responses in diverse tissues. In this regard, a large gap remains in our ability to correlate Hh signaling outputs with phenotypic outcomes since it is difficult to delineate the contributions of individual Gli protein or its processed form. This is further complicated by the differential expression and requirement of Gli proteins (Bai et al., 2004, Bowers et al., 2012, Cao et al., 2013, Ding et al., 1998, Matise et al., 1998) and their complex interactions in diverse tissues (Bowers et al., 2012, Liu et al., 2012). One of the critical events in mammalian Hh signaling involves regulation of Gli by Suppressor of fused (Sufu), a major negative regulator. Studies of Sufu thus provide a unique opportunity to uncover the molecular mechanisms by which Gli proteins control Hh signaling. Sufu can sequester Gli proteins (Barnfield et al., 2005, Ding et al., 1999, Kogerman et al., 1999, Murone et al., 2000), regulate Gli2/3 protein levels (Chen et al., 2009, Jia et al., 2009, Wang et al., 2010), facilitate the production of Gli repressor and inhibit the generation of Gli activators (Humke et al., 2010, Tukachinsky et al., 2010). Perhaps all of these actions ensure the production of appropriate amounts of Gli activators and repressors as well as a pertinent Gli activator/repressor ratio necessary for tissue development and homeostasis. The relative contribution of multiple effects of Sufu to Gli protein functions has not been clearly delineated. A key aspect of Hh signaling is to turn on Hh target genes through Gli activators. In this report, we investigate how Sufu controls Gli1 protein levels both in vitro and in vivo (lungs). These studies not only revealed a conserved mechanism by which Sufu controls Gli protein levels but also led to the unexpected finding that Hh targets can exhibit different responses when the Hh pathway is activated. We speculate that this is because different combinations of Gli proteins are present in a particular tissue for activating Hh targets. Thus some Gli targets are upregulated while others are downregulated, depending on the availability of Gli proteins that control their expression in a specific tissue. This result reveals the complexity of Hh responses in diverse tissues and increases our understanding of how the Sufu/Gli circuitry controls Hh pathway activation.