The many lives of SHH in limb development and evolution

G Model

ARTICLE IN PRESS

YSCDB-1906; No. of Pages 9

Seminars in Cell & Developmental Biology xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Seminars in Cell & Developmental Biology journal homepage: www.elsevier.com/locate/semcdb

Review

The many lives of SHH in limb development and evolution Javier Lopez-Rios Development and Evolution, Department of Biomedicine, University of Basel, 4058 Basel, Switzerland

a r t i c l e

i n f o

Article history: Received 22 October 2015 Received in revised form 21 December 2015 Accepted 23 December 2015 Available online xxx Keywords: Limb bud Digit SHH GLI HOX Morphological evolution

a b s t r a c t The SHH signaling pathway is essential for proper formation of the limb skeleton, as is required for the survival and expansion of distal chondrogenic progenitor cells. At the same time, SHH is important to specify digit identities along the anterior–posterior axis. Upon gain or loss of activity of the SHH pathway, bones are gained, lost or malformed, and such deregulation underlies the aetiology of various human congenital limb defects. Likewise, accumulating evidence suggests that evolutionary tampering with SHH signaling underlies the morphological diversification of the tetrapod appendicular skeleton. This review summarizes the roles of the SHH pathway in the context of limb development and evolution and incorporates recent evidence into a mechanistic view of how the positioning of digit condensations is integrated with the specification of distinct bone morphologies. © 2016 Elsevier Ltd. All rights reserved.

Contents 1. 2. 3. 4. 5. 6. 7. 8.

Basic concepts in mouse limb development. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .00 Early AP polarization of the limb bud upstream of Shh expression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 Shaping and sensing the SHH gradient . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 Signal interpretation at the transcriptional level: GLI activators, GLI repressors and GLI motifs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 SHH pathway and growth control during limb development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 SHH and the establishment of AP polarity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 SHH and the morphological evolution of the limb . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 Concluding remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

“On the other hand, you have different fingers” Steven Wright

1. Basic concepts in mouse limb development Forelimb primordia emerge from the flank around mouse embryonic day E9.0, while hindlimb development is delayed for about half a day. The early limb bud, composed of undifferentiated mesenchyme encapsulated by ectoderm, grows out and elongates along the proximal–distal (PD) axis. Around embryonic day E10.75, the distal part of the forelimb bud starts to expand along the anterior–posterior (AP) axis, forming the handplate,

E-mail address: [email protected]

which contains the progenitors that will give rise to the digits. All pre-chondrogenic condensations that prefigure the definitive limb skeleton are already distinguishable in E12.5 forelimbs. The proximal segment of the forelimb, the stylopod, will give rise to the humerus (femur in the hindlimb), the middle segmentor zeugopod will form the ulna and radius (fibula and tibia), while the distal part, the autopod, contains the progenitors for the carpal bones of the wrist (tarsal bones of the ankle), the metacarpal bones of the palm (metatarsals in the feet) and the phalanges of the digits (Fig. 1A). Growth and patterning of the limb skeleton are coordinated along the three anatomical axes through the concerted interaction of two signaling centers or organizers (reviewed in [1,2]). The AP organizer is called the polarizing region or zone of polarizing activity (ZPA), which is a small group of SHH-producing cells located in the posterior limb bud mesenchyme (Fig. 1B). The PD organizer

http://dx.doi.org/10.1016/j.semcdb.2015.12.018 1084-9521/© 2016 Elsevier Ltd. All rights reserved.

Please cite this article in press as: J. Lopez-Rios, The many lives of SHH in limb development and evolution, Semin Cell Dev Biol (2015), http://dx.doi.org/10.1016/j.semcdb.2015.12.018

G Model YSCDB-1906; No. of Pages 9 2

ARTICLE IN PRESS J. Lopez-Rios / Seminars in Cell & Developmental Biology xxx (2015) xxx–xxx

Fig. 1. Mouse limb development and the SHH pathway. (A) Mouse forelimb skeletal elements. Digits are numbered 1–5 along the anterior–posterior axis. (B, C) Expression of Shh and Fgf8 in mouse forelimbs. (D, E) Forelimb skeletal phenotypes of Shh and Gli3-deficient embryos. (F) Gli1 expression is a sensitive transcriptional readout of SHH signaling. The arrowhead points to the domain that becomes insensitive to SHH signaling. (G) GLI3R is cleared from the posterior limb bud mesenchyme under the influence of SHH signaling. Shh-expressing cells are labelled in green, while GLI3R (in red) is detected with antibodies that primarily recognize the repressor isoform on tissue sections [8]. Stages shown correspond to embryonic days E16.5 (A, D, E), E11.0 (B, C, F) and E10.25 (G).

is the apical ectodermal ridge (AER), a specialized ectoderm that runs along the limb bud margin and produces several factors, most importantly FGFs (AER–FGFs; Fig. 1C). Both signaling centers maintain each other’s activity via a series of interlinked feedback loops that converge on the regulation of the BMP-antagonist GREM1 [1]. Decades of research have uncovered the essential roles of SHH signaling in the growth and AP patterning of the limb, as evidenced by the phenotypes resulting from loss or gain of function of the pathway (Fig. 1D and E) [3,4]. This review summarizes the current knowledge on the functions and mechanisms of SHH signaling during mouse limb development and puts them in the context of human congenital defects and morphological evolution of the tetrapod limb. For more general accounts on limb development or on mechanistic aspects of Hedgehog signal transduction that are out of the scope of this article, the reader is referred to other recent reviews [1,2,5–7]. 2. Early AP polarization of the limb bud upstream of Shh expression Recent studies have provided strong evidence for a mechanism that polarizes the early limb bud mesenchyme along the anteriorposterior axis prior to the onset of Shh expression [8,9]. In the limb field, Hox9 paralogous genes in the forelimb and Isl1/Sall4 in the hindlimb contribute to position Gli3 and Hand2 domains in the anterior and posterior limb bud, respectively [10–12]. In turn, GLI3 and HAND2 proteins control a gene regulatory network of transcription factors that specifies abutting anterior and posterior mesenchymal compartments and enables Shh activation in the polarizing region [8,12–14]. Several transcriptional activators and repressors such as HAND2, HOX, PBX, ETS, TWIST1, ALX4, PLZF or GATA6 proteins are required to delimit the Shh domain [15–26]. Many of them do so by interacting with a distant conserved enhancer referred to as the ZRS/MFCS1, whose genetic inactivation phenocopies the limb defects observed in Shh-deficient embryos [27,28]. In contrast, point mutations affecting the ZRS are associated to polydactyly in humans and other species due to

ectopic Shh expression in the anterior limb bud [27,29]. In addition, FGF and WNT signals produced by the AER and surface ectoderm are required for Shh expression, while the BMP pathway negatively regulates it [30–37]. Moreover, a self-regulatory system operates in the polarizing region, as Shh is downregulated after increasing SHH pathway activity and vice versa [38,39]. Nevertheless, it remains to be mechanistically defined how the downstream effectors of all these morphogenetic pathways impact, directly or indirectly, on Shh expression. 3. Shaping and sensing the SHH gradient The release of SHH by the polarizing region results in graded distribution of the ligand in the posterior half of the distal limb bud [40]. This pattern fits well with the expression domains of the transcriptional targets Gli1 and Ptch1, which are sensitive readouts of pathway activation (Fig. 1F) [41,42]. PTCH1 is the main SHH receptor and constitutively represses the pathway by preventing the accumulation of another multi-pass membrane protein, Smoothened (SMO), in the primary cilium [5,6,43–45]. This ligandindependent antagonism ensures that the pathway is silenced in cells not exposed to SHH. In addition, Ptch1 is a transcriptional target of the pathway, which leads to high levels of PTCH1 receptor being produced close to the source of SHH, which results in ligand sequestration [43,46,47]. This negative feedback loop is referred to as ligand-dependent antagonism, and is essential to limit the range of SHH signaling. Additional mechanisms are important to shape the SHH gradient. The active SHH ligand is modified by the covalent addition of palmitoyl and cholesterol groups, which are required to target the protein to lipid rafts and for the assembly of soluble multimeric complexes [48–50]. Lack of cholesterol modification in SHH proteins induces the formation of ectopic digits in the anterior margin, which indicates that the cholesterol moiety is required to restrict the ligand to the posterior half [51]. In contrast, mice producing SHH proteins missing the palmitic acid group display loss and fusions of central digits, suggesting that the palmitoyl group is important for

Please cite this article in press as: J. Lopez-Rios, The many lives of SHH in limb development and evolution, Semin Cell Dev Biol (2015), http://dx.doi.org/10.1016/j.semcdb.2015.12.018

G Model YSCDB-1906; No. of Pages 9

ARTICLE IN PRESS J. Lopez-Rios / Seminars in Cell & Developmental Biology xxx (2015) xxx–xxx

long-range signaling [52]. Recently, it was shown that SHH particles are transported away from the polarizing region by long, actin-based filopodia similar to Drosophila cytonemes [53,54]. SHHcarrying filopodia contact in responding cells similar processes that contain the SHH co-receptors BOC and CDO that, together with GAS1, are known to facilitate long-range SHH signaling by forming multimolecular receptor complexes with PTCH [55–60]. In fact, their combined genetic inactivation results in loss and fusions of digits, similar to alterations in SHH palmitoylation [52,56,59]. An emerging picture for long-range signaling in the limb is therefore arising, and it will be important to challenge it genetically and to define the mechanisms that connect the cytonemal-delivery of SHH with signal transduction at the level of the primary cilium. Ultimately, SHH binding to PTCH results in the accumulation and activation of ciliary SMO. Oxysterols are natural cholesterol derivatives that have been found to bind and activate SMO, raising the possibility that PTCH (which contains a sterol-binding domain) regulates the availability of a still uncharacterized ligand of SMO [5,61]. Recent studies have also revealed that, upon pathway activation, SMO becomes preferentially localized to the base of the primary cilium through the interaction which EVC/EVC2 complexes, which are themselves anchored to the proximal cilium by EFCAB7–IQCE complexes [62–64]. Pathway activation also critically depends on intraflagellar transport (IFT) complexes trafficking PTCH, SMO, SUFU, GLI2 and GLI3 and other proteins in and out of the cilia. Extensive genetic and biochemical evidence indicates that primary cilia are essential both for activating the pathway in response to the SHH ligand and for keeping it inactive in its absence [5,6]. Consequently, mutations in ciliary components show limb defects consistent with pathway dysregulation, e.g.: polydactylies [65]. Recent publications have reported that NOTCH signaling modulates the response to SHH in the spinal cord by facilitating the trafficking of SMO to the primary cilium [66,67]. Interestingly, the NOTCH ligand JAG1 is itself a transcriptional target of SHH signaling in the mouse limb bud [68–70], which suggests the existence of SHH-NOTCH positive feedback loop operating in the limb mesenchyme, although this hypothesis remains to be genetically tested.

4. Signal interpretation at the transcriptional level: GLI activators, GLI repressors and GLI motifs Ultimately, HH pathway activation converges on the regulation of the GLI transcription factors at the level of the primary cilia [5,6,71]. In the absence of the signal, full-length GLI3 is constitutively processed to a GLI3R isoform that translocates to the nucleus and negatively regulates SHH target genes [72]. This process is controlled by phosphorylation of full length GLI3 by PKA, CKI and GSK3␤, which targets it for ubiquitination and partial processing by the proteasome to render GLI3R [5,72]. Conversely, HH pathway activation inhibits the production of GLI3R isoforms and promotes the formation of GLI activators (GLI2A and GLI3A), which upregulate SHH transcriptional targets. SUFU performs essential roles in these processes, as it negatively regulates HH pathway activation by promoting the formation of GLI repressor isoforms and preventing the production of GLI activator proteins. Upon HH signaling, dissociation of SUFU and GLI2/3 complexes is required for the formation of GLI2A/GLI3A activator isoforms, a process also regulated by phosphorylation [5,73]. Genetic analysis revealed that embryos lacking Gli1 and/or Gli2 develop normally patterned limbs, indicating that GLI3A and GLI3R suffice in controlling SHH transcriptional output during limb development [74]. However, genetic removal of Gli2 in a Gli3deficient background revealed that Gli2 also participates in limb patterning, mostly of the posterior digits [75,76]. In contrast, Gli1 is a downstream target of vertebrate Hedgehog signaling and encodes

3

a constitutive activator that amplifies the transcriptional response [74]. Another general characteristic of SHH signaling is that cells exposed to continuous high levels of SHH become desensitized and silence the pathway (arrowhead in Fig. 1F) through a mechanism that likely involves down-regulation of Gli2 [77–79]. In the limb bud, the polarized expression of Shh in the posterior margin leads to graded distributions of GLI isoforms, with highest levels of GLI3R in the anterior mesenchyme and of GLI2A and GLI3A in the posterior margin (Fig. 1G) [8,72]. This unequal distribution of repressor and activator isoforms ensures that the pathway is activated in the posterior half of the limb, while being silenced in the anterior. Quite strikingly, the simultaneous inactivation of Gli3 and Shh produces a polydactylous phenotype similar to that of the Gli3 deficiency alone [80,81]. This indicates that a main function of SHH is to deplete the posterior mesenchyme of GLI3R proteins in order to activate the pathway (Fig. 1G). In the absence of Gli3 (and hence of GLI3R), SHH is “dispensable” as the pathway is constitutively derepressed in the posterior margin. In agreement, recent analyses indicate that SHH target genes in the limb bud are mostly regulated by GLI-mediated repression [82]. Several studies have genetically addressed the relative contributions of GLI3A and GLI3R in controlling digit number and AP polarity by the use of Gli3 mutant alleles expressing either truncated GLI3R-like isoforms or GLI3 proteins that could not be processed into GLI3R [9,76,83–88]. Some of these studies raised some controversy due to the hypomorphic nature of some of these alleles and did not take into account the contribution of GLI2A to posterior patterning [75,76]. Altogether, the current genetic evidence suggests that proper AP patterning of the limb and digit number are both controlled by the asymmetric distribution of both GLI3R and GLI2A/GLI3A proteins. Genetic and expression analysis of limb buds deficient in different pathway components have identified a wealth of putative direct SHH downstream targets during mouse limb bud development [69,70,89] (see also Sections 5 and 6). GLI transcriptional activator and repressor forms bind directly to the genomic regions through the interaction with GLI-binding sites (GBS) [90]. All GLI1-3 proteins show similar binding affinities to the GLI motif, a sequence with optimal core consensus TGGGTGGTC [91,92]. GLI factors may also be potentially recruited to DNA through the interaction with other transcriptional regulators. Indeed, GLI proteins have been shown to interact in vitro with SMADs, ␤−Catenin, ZIC and SKI factors, although the functional relevance of these interactions for limb development is unclear [93–97]. In addition, GLI3R interacts with HOXD12 in the limb bud to render complexes with trans-activating capacities [98]. Analysis of GLI-ChIP datasets and bioinformatic predictions also uncovered that SOXB1 proteins coregulate SHH target genes in the spinal cord [92,99]. Analogous factors conferring limb bud specificity to SHH responses have not been identified so far, but 5 HOXD and, in particular, HAND2 proteins are good candidates, as several cis-regulatory modules (CRM) are targeted both by GLI3 and HAND2 in limb buds [8]. These functional genomics analyses also evidenced that the number and quality of the GBS motif can vary substantially between CRMs. The presence of high and low affinity GBSs is of particular importance to translate the SHH gradient into a robust transcriptional response and define activation thresholds. A general principle arising, at least in neural tissues, is that genes induced close to the source of HH are frequently regulated by CRMs with high-affinity GBS. Conversely, genes induced at low SHH levels present more complex regulatory regions with several low binding affinity GLI motifs [92,99–101]. This apparent paradox may reflect a general mechanism for translating the morphogenetic SHH signaling gradient into a diversified transcriptional response affecting hundreds of genes. Genes induced by high SHH may require loss of GLIR and active GLIA input mediated by high-affinity GBS; in contrast, genes that are induced at low levels of the ligand may require

Please cite this article in press as: J. Lopez-Rios, The many lives of SHH in limb development and evolution, Semin Cell Dev Biol (2015), http://dx.doi.org/10.1016/j.semcdb.2015.12.018

G Model YSCDB-1906; No. of Pages 9 4

ARTICLE IN PRESS J. Lopez-Rios / Seminars in Cell & Developmental Biology xxx (2015) xxx–xxx

poor quality GBS to facilitate their activation just by loss of GLI3R (without GLIA production). Indeed, the activation of many relevant SHH targets in the limb just requires GLI3R clearance from the posterior margin [80,81]. Finally, these long-range SHH targets are also predicted to require additional regulation from other pathways to precisely define their expression domains [92,99,102].

5. SHH pathway and growth control during limb development Shh-deficient limb buds form a single bone in the zeugopod, while the autopod is essentially absent in the forelimb (Fig. 1D) and represented by a single digit in the hindlimb [3,103]. This severe phenotype illustrates the critical functions of SHH in the survival and proliferation of the distal mesenchyme. The scapula and humerus are normal in Shh-deficient embryos, which indicates that the specification and AP patterning of proximal limb elements depends on upstream mechanisms such as the Gli3-Hand2 system [8,15]. Indeed, genetic lineage analyses showed that neither SHH-expressing nor SHH-responding cells contribute to proximal skeletal elements [77,104]. The distal outgrowth of the limb bud requires the establishment of a mesenchymal-epithelial feedback loop between the ZPA and the AER (Fig. 1B, C and Fig. 2A) [31,32,105,106]. Both limb organizers maintain each other’s activity so that inactivation of AER-Fgf genes leads to loss of Shh expression and vice versa [30,105]. Therefore, the loss of distal bones in Shh-deficient limbs can be explained by the collapse of the AER and the elimination of distal progenitors by apoptosis starting around E10.5 [3]. Conversely, experimental or genetic manipulations that enhance AER-FGF feedback signaling result in increased mesenchymal cell number and polydactyly (e.g [35,107,108]. SHH sustains AER activity through the upregulation of the BMP-inhibitor GREM1, which is essential to protect the AER from excess BMP signaling [70,105,109,110]. Grem1 expression becomes independent of SHH signaling as it expands anteriorly, which allows for the maintenance of the AER in the anterior-most mesenchyme [68]. In comparison to Shh deficiency, Grem1 inactivation leads to a slightly delayed disruption of the AER-ZPA feedback loop, allowing the formation of up to three digits in the autopod [109]. This SHH-GREM1-FGF system of interlinked feedback loops confers limb bud growth with robustness and auto-regulatory properties [1,106]. During early limb bud development, growth is most evident along the PD axis. Around E10.75, the distal limb bud starts to expand dramatically along the AP axis to form the handplate, and balanced control of this proliferation is essential to restrict the limb to five digits. This notable expansion of digit progenitors is controlled by the SHH pathway, not only by sustaining AER-FGF expression, but also through the direct transcriptional regulation of cell cycle genes (Fig. 2A) [111,112]. In the posterior margin, SHH upregulates the expression of genes involved in the G1 -S transition of the cell cycle such as Ccnd1, Cdk6 and Mycn [111,112]. In agreement, conditional inactivation of Shh at progressively earlier time points leads to a step-wise decrease in digit number and G1 cell cycle arrest [113]. In the anterior mesenchyme, the SHH pathway—including cell cycle targets- is silenced by the predominance of GLI3R-mediated repression [80,81,112]. This allows for the preferential expansion of posterior progenitors under the influence of SHH. Indeed, SHH-responding cells contribute to digits 2–5, albeit at different proportions [77]. In the absence of Gli3—and hence of GLI3R-, Ccnd1 and Cdk6 are ectopically upregulated in the anterior margin and digit progenitors proliferate faster as the result of enhanced G1 -S transition. In addition, due to AP patterning defects and Grem1 expansion, AER-FGF signaling is also increased. Overall, these alterations produce a surplus of cells that

is translated into the formation of several supernumerary digits in Gli3-deficient embryos (Fig. 1E) [112]. Conditional gene inactivation experiments revealed that Gli3 is also required to time the cell cycle exit of progenitors towards BMP-driven chondrogenic differentiation by abolishing Grem1 expression. This delay in differentiation produces an excess of progenitors that also contributes to preaxial polydactylies in Gli3-deficient limbs [69,112]. Furthermore, this mechanism may underlie the etiology of human polydactylies caused by heterozygous mutations in GLI3 that are not associated to alterations in digit identity [114]. Finally, SHH also signals directly to the ectoderm and is required to limit AER length in the posterior margin, so that additional postaxial condensations are formed when SHH signaling is inhibited in the AER [115]. This phenotype is indeed reminiscent of the genetic reduction in BMP pathway activity in the mesenchyme, as BMP signaling has an inhibitory effect on the maintenance of the AER at these developmental stages [35,36]. Lastly, the disruption of the ZPAAER feedback loop marks the end of the proliferative expansion phase. Different non-exclusive mechanisms for self-termination of growth have been proposed, but all require the extinction of Grem1 expression [104,116]. In addition, it has recently been proposed in the chicken that ZPA cells are able to measure the time of Shh expression via an intrinsic clock linked to the cell cycle that would represent a cell-autonomous mechanism for termination of SHH signaling and growth [117].

6. SHH and the establishment of AP polarity The ZPA was identified by Saunders and Gasseling as a region with organizing properties [118]. (see also [119] for a historical account). When dissected and transplanted to the anterior margin of another wing bud, these posterior grafts were able to induce mirror-image digit duplications. These experiments led to the proposal that the specification of digits could be controlled by a morphogen, a diffusible signal that would provide positional information in a dose dependent manner [120]. The morphogen secreted by the polarizing region was later identified to be SHH and further shown to indeed be capable of evoking different responses in a concentration dependent manner [121,122]. In contrast to the spinal cord, SHH signaling in the limb is not involved in the specification of different cellular identities, but rather in controlling, in addition to cell number, the pattern and morphology of cartilage condensations. The proximal skeleton forms in a SHH-independent manner [3,77,104] (see also Section 5). Instead, the early AP polarization of the limb bud and specification of stylopod elements (humerus/femur) is mediated by the interactions between the GLI3 and HAND2 transcription factors, likely in combination with Hox9 and Hox10 paralogous genes [8,13,15,123–125]. In the zeugopod, in contrast, Shh-deficient embryos develop a single bone, reflecting the shortage of progenitors due to proliferation/survival defects [3]. Genetic lineage tracing in the mouse revealed that cells exposed to SHH only contribute to the posterior bone (the ulna/fibula), suggesting that SHH may also control its AP specification [77,104]. However, the radius and ulna appear normal in the absence of both Gli3 and Shh (loss of GLIR and GLIA activities), which suggests that AP polarity in the zeugopod may not be specified simply by graded SHH pathway activation [80,81]. A plausible interpretation of these observations is that the specification of zeugopod AP identities is also mediated by the Gli3-Hand2 system that polarizes the limb upstream of SHH, in combination with Hox10 and Hox11 paralogous genes and Plzf [8,13–15,123,125–128]. In support of this hypothesis, Gli3/ ;Hand2/ mutant limbs typically form symmetrical zeugopodal bones, in contrast to the unperturbed patterning of the Gli3/ ;Shh/ zeugopod [15,80,81]. Interestingly, only ZPA grafts

Please cite this article in press as: J. Lopez-Rios, The many lives of SHH in limb development and evolution, Semin Cell Dev Biol (2015), http://dx.doi.org/10.1016/j.semcdb.2015.12.018

G Model YSCDB-1906; No. of Pages 9

ARTICLE IN PRESS J. Lopez-Rios / Seminars in Cell & Developmental Biology xxx (2015) xxx–xxx

5

Fig. 2. Overlapping mechanisms for the positioning of digit condensations and the specification of digit identities. (A) SHH drives the proliferative expansion of digit progenitor cells along the anterior–posterior (AP) axis through the ZPA-AER feedback loop and the direct transcriptional control of the cell cycle. (B) Simultaneously, a BMP-SOX9-WNT (BSW) Turing-type patterning mechanism controls the localized onset of chondrogenic differentiation of digit precursor cells. Distal Hox genes and AER–FGFs modulate the spacing of digit condensations. (C) Digit morphologies (=identities) are likely specified through AP asymmetric 5 Hoxd expression, which is regulated by SHH signaling through yet to be defined cis-regulatory mechanisms. A, anterior; P, posterior.

in very early chicken embryos—before the limb bud is evident- can duplicate the ulna, probably as a consequence of disturbing normal GLI3 functions upstream of Shh [129]. Genetic analysis has led to different models for digit specification in the mouse (reviewed in [130]). Genetic lineage tracing revealed that cells that express Shh (ZPA-descendants) contribute to the two posterior digits and, to a lesser extent, to d3 [104]. A similar genetic strategy mapped the location of cells that had been exposed to the ligand, which indicated that they contribute to the same skeletal structures as SHH-descendants and, in addition, to digit 2. These studies led to a “temporal expansion” model by which SHH would specify digit identities by integrating the concentration and time of exposure to the ligand [39,104,111,130]. Hence, digits 5 and 4 would be both specified by high levels of SHH but would differ in the duration of exposure, as digit 4 progenitors cease to express Shh earlier. Digit 3 progenitors would be a mixture of cells receiving autocrine and paracrine SHH, while digit 2 would be specified by low paracrine signaling only. In contrast, digit 1 forms in a SHH-independent manner [39,104]. A second model was proposed after the temporal requirements for SHH signaling were assayed by Shh conditional inactivation [113,131]. Shh ablation at progressively earlier time points led to sequential digit loss but in the order 3–5–2–4, which is the reverse order of condensation of digit primordia. In other words, the first digit to be lost upon compromising Shh function was the last to condense, and so on and so forth. These findings led the authors to propose a “biphasic” model by which digit AP identities are first specified by SHH morphogenetic signaling, followed by a proliferative expansion phase that would elaborate this early patterning into digits. Hence, digit loss would just reflect the absence of sufficient progenitors of a defined identity to produce a condensation. The consequences upon genetic alterations of the SHH pathway can be interpreted in the light of both models depending on the identity assigned to the digit(s) lost, which is difficult in the mouse

[130,132]. To reconcile data from both models, Towers and Tickle proposed the “alternating specification” model that postulates that a mechanism of progressive digit specification driven by proliferation operates in the anterior to specify digits 2 and 3. In the posterior, digits 4 and 5 would arise through a temporal expansion system. This model would integrate current models from digit specification in the chicken wing, in which the ZPA does not contribute to any of the digits, and provide an explanation for the alternating loss of digits upon the temporal deletion of Shh [113,130,133]. Adding to the complexity of the process, it was recently described that the periodic digit-interdigit pattern is controlled by a selfregulatory BMP-SOX9-WNT (BSW) Turing system that positions the digits through the localized onset of chondrogenic differentiation (Fig. 2B) [134]. Moreover, genetic evidence indicated that the wavelength of this BSW Turing patterning mechanism is controlled by HOXD and HOXA factors in a genetically redundant manner, as well as by AER–FGFs [135,136]. None of these models specifically addresses how different skeletal morphologies arise. Most importantly, it remains unclear how SHH-driven growth is integrated with the iterative process of digit formation controlled by such Turing-type patterning system [136]. The best effector candidates to mediate both of these aspects are the Hoxd10-d13 genes. In the handplate, these 5 Hoxd genes are expressed in nested domains, and this pattern was long ago suggested to control digit morphologies [126,137–139]. While these genes are essential to activate Shh in the posterior margin, SHH signaling quantitatively controls the transcription of 5 Hoxd genes during this late phase of expression [17,18,140,141]. In addition, GLI3R is required at early stages to posteriorly restrict 5 Hoxd expression. In fact, constitutive inactivation of Gli3 leads to the loss of anterior digit identity, which correlates with the precocious and ectopic expansion of 5’Hoxd gene expression domains in the anterior mesenchyme [112,142]. This pattern is an intrinsic property of the regulatory landscape that drives this late phase of 5 Hoxd

Please cite this article in press as: J. Lopez-Rios, The many lives of SHH in limb development and evolution, Semin Cell Dev Biol (2015), http://dx.doi.org/10.1016/j.semcdb.2015.12.018

G Model YSCDB-1906; No. of Pages 9 6

ARTICLE IN PRESS J. Lopez-Rios / Seminars in Cell & Developmental Biology xxx (2015) xxx–xxx

expression and, as a consequence, a “code” of graded HOX activity is established along the AP axis of the autopod in response to SHH influence (Fig. 2C). Interestingly, 5 Hoxd genes show differential temporal dependence on SHH signaling, and the establishment of their respective autopod domains progressively becomes independent of SHH pathway activity [68,112]. It is not known how SHH/GLI signaling impacts on the regulation of 5 Hoxd genes in the presumptive digit domain, but GLI-containing complexes have been shown to bind to the Hoxd regulatory landscape [70,143]. Taking all the previous and recent evidence together, a formal possibility for the elaboration of digit patterning is schematized in Fig. 2. This proposal relies on the assumption that growth and patterning of the autopod are disconnected molecular circuits controlled by SHH and that the specification of digit identity is a late process. In fact, previous evidence in the chicken indicates that digit identity remains plastic but is determined late in development by a BMP-related mechanism involving the interdigital mesenchyme and the tips of the digits (the phalanx forming region, which has also been identified in the mouse) [144–147] Under this view, growth driven by SHH through the feedback loop and cell cycle modulation would induce AP expansion of distal progenitors (Fig. 2A). Simultaneously, SHH signaling would impose graded HOX activity along the AP axis of the handplate. This HOX code would then, in turn, confer digit identity (i.e. specific bone morphologies) and at the same time, modulate the periodicity of the digit-interdigit pattern as part of a Turing-like system that controls the positioning of the digit condensations. When more or less cells are produced due to proliferative alterations, the available progenitors “read” the HOX code available at each location along the AP axis, while the BSW system defines the periodic pattern according to the size of the field. Chondrogenesis and the BSW network need not to be regulated directly by SHH, but in its absence no autopod is formed as the distal progenitors die. In contrast, ectopic activation of the pathway will result in enhanced proliferation and extra digits. The identity of these digits will then depend on the extent of the disturbance of HOX expression. Most telling, this set of overlapping systems can be genetically uncoupled. For example, removal of Gli3 and almost total genetic ablation of the HOX code leads to an enlarged handplate similar in size to that of the Gli3-deficiency alone (growth control by the SHH pathway removed) but with multiple thin “digits” with no identity (HOX code disrupted). In these mutants, the Turing-system of localized chondrogenesis is still working, but the wavelength is diminished due to lower HOX activity, leading to as many “digits” as the width of the handplate can accommodate given the intrinsic properties of the BSW gene regulatory network [135]. As this overlapping set of patterning mechanisms act simultaneously in the handplate, they are all directly or indirectly under the genetic control of the SHH pathway.

is reminiscent of the temporal series of Shh inactivation in the mouse or the experimental decrease in cell numbers using mitotic inhibitors in amphibians [113,150,151]. Birds constitute another classic case of evolutionary digit loss, although it is controversial which are the digits lost and the identity of the ones remaining (reviewed in [152]). Interestingly, the chicken ZPA does not contribute to any digit in the wing (with three digits) and only to one in the leg (with four digits), which suggests that SHH-descendants do not expand or that they die [133]. Analysis of condensation patterns in different bird limb buds with variable digit number also revealed that posterior condensations are more refractory to evolutionary disappearance, probably reflecting strong developmental constrains associated to SHH-regulated networks [153]. Finally, a recent study on cattle limb development illustrates the power of manipulating SHH signaling for morphological evolution [154]. Bovines form two symmetric digits, of which only the distal phalanges contact the ground. This so-called unguligrade posture is an adaptation for running, as is the evolutionary loss of lateral digits, which has independently occurred in several artiodactyl lineages [155]. In contrast to the mouse, Ptch1 is not upregulated in the posterior bovine limb bud mesenchyme and remains instead expressed at low levels. Consistent with these observations, a CRM that mediates Ptch1 expression in the distal limb bud was found to have functionally degenerated in the bovine lineage. Given that a main function of PTCH1 is to spatially restrict ligand distribution (see Section 3), SHH proteins extend further anterior in bovine than in mouse limb buds. This alteration in SHH graded signaling results in the progressive loss of anterior–posterior polarity in the distal limb bud, as revealed by the symmetrical and distalized domains of expression of Hoxd13, Grem1 and Fgf8. Hence, the molecular loss of polarity prefigures the anatomical symmetry of the bovine handplate skeleton. Interestingly, lateral digit condensations do form in the bovine handplate, but are not elaborated into adult digits as they fall out of AER–FGF range due to the loss of AP polarity and distal shift of the AER [154]. As in cattle, Ptch1 is not upregulated in pig limb buds, but it is still expressed in the camel limb bud mesenchyme. In contrast to cattle and pig, lateral digit primordia in the camel limb bud are eliminated by apoptosis [154,156,157]. These results suggest that different mechanisms have contributed to the progressive loss of asymmetry and digits in different artiodactyl lineages. Likewise, additional alterations in the gene regulatory networks that control limb patterning and growth must have bolstered up alterations in Ptch1 regulation to facilitate loss of AP polarity in the handplate and the acquisition of unguligrade postures in artiodactyls.

8. Concluding remarks 7. SHH and the morphological evolution of the limb Given the essential roles of SHH in the growth and patterning of the extremities, it is not surprising that this pathway has been tweaked for evolutionary adaptations affecting the limb. For example, the Shh ZRS enhancer sequences have degenerated in limbless reptiles (colubrid snakes) and amphibians (caecilians) [148]. However, it is not known if these evolutionary changes are causative or secondary to other alterations, such as e.g. AER degeneration, as it occurs in python hindlimb buds [148,149]. Another frequent adaptation of the limb skeleton is the loss and/or reduction of digits, which has occurred in all major groups of tetrapods. For example, closely related lizard species of the genus Hemiergis display a serially reduced digit pattern with 5, 4, 3 or 2 complete digits. Shapiro and collaborators showed that the degree of digit loss inversely correlates with the time digit progenitors are exposed to SHH, which

The limb bud constitutes a fascinating system to study morphogenesis. While classic models have proven invaluable to conceptualize such a complex process, the advent of functional genomics is providing mechanistic insight into these processes at the molecular level. Specifically concerning the growth and patterning of the limb, it will be essential to define the catalog of regulatory interactions mediated by GLIR and GLIA isoforms with spatio-temporal resolution, as well as those of HOXD and HOXA proteins. At the same time, it will remain a challenge to understand how SHH signaling is integrated with that of other developmental pathways at the transcriptional level during limb development. Unravelling the mechanisms behind regulatory systems may identify genomic regions that harbour mutations in patients with congenital malformations of the extremities or that have been functionally altered in one way or the other during tetrapod evolution.

Please cite this article in press as: J. Lopez-Rios, The many lives of SHH in limb development and evolution, Semin Cell Dev Biol (2015), http://dx.doi.org/10.1016/j.semcdb.2015.12.018

G Model YSCDB-1906; No. of Pages 9

ARTICLE IN PRESS J. Lopez-Rios / Seminars in Cell & Developmental Biology xxx (2015) xxx–xxx

Acknowledgements I wish to apologize to all colleagues whose work could not be discussed due to space constraints. I thank Marco Osterwalder, Dario Speziale and Virginie Tissières for comments on the manuscript and pictures and Barbara Widmer for help in the preparation of the text. Research in my lab is supported by the Olga Mayenfisch Foundation and the University of Basel.

References [1] R. Zeller, J. Lopez-Rios, A. Zuniga, Vertebrate limb bud development: moving towards integrative analysis of organogenesis, Nat. Rev. Genet. 10 (2009) 845–858. [2] C.J. Sheeba, R.P. Andrade, I. Palmeirim, Getting a handle on embryo limb development: molecular interactions driving limb outgrowth and patterning, Semin. Cell Dev. Biol. (2015), http://dx.doi.org/10.1016/j. semcdb.2015.01.007 (in press). [3] C. Chiang, Y. Litingtung, M.P. Harris, B.K. Simandl, Y. Li, P.A. Beachy, et al., Manifestation of the limb prepattern: limb development in the absence of sonic hedgehog function, Dev. Biol. 236 (2001) 421–435. [4] C.C. Hui, A.L. Joyner, A mouse model of greig cephalopolysyndactyly syndrome: the extra-toesJ mutation contains an intragenic deletion of the Gli3 gene, Nat. Genet. 3 (1993) 241–246. [5] J. Briscoe, P.P. Therond, The mechanisms of Hedgehog signalling and its roles in development and disease, Nat. Rev. Mol. Cell Biol. 14 (2013) 416–429. [6] Y.I. Nozawa, C. Lin, P.T. Chuang, Hedgehog signaling from the primary cilium to the nucleus: an emerging picture of ciliary localization, trafficking and transduction, Curr. Opin. Genet. Dev. 23 (2013) 429–437. [7] A. Zuniga, Next generation limb development and evolution: old questions, new perspectives, Development 142 (2015) 3810–3820. [8] M. Osterwalder, D. Speziale, M. Shoukry, R. Mohan, R. Ivanek, M. Kohler, et al., HAND2 targets define a network of transcriptional regulators that compartmentalize the early limb bud mesenchyme, Dev. Cell 31 (2014) 345–357. [9] O. Zhulyn, D. Li, S. Deimling, N.A. Vakili, R. Mo, V. Puviindran, et al., A switch from low to high Shh activity regulates establishment of limb progenitors and signaling centers, Dev. Cell 29 (2014) 241–249. [10] B. Xu, D.M. Wellik, Axial Hox9 activity establishes the posterior field in the developing forelimb, Proc. Natl. Acad. Sci. U. S. A. 108 (2011) 4888–4891. [11] J. Itou, H. Kawakami, T. Quach, M. Osterwalder, S.M. Evans, R. Zeller, et al., Islet1 regulates establishment of the posterior hindlimb field upstream of the Hand2-Shh morphoregulatory gene network in mouse embryos, Development 139 (2012) 1620–1629. [12] R. Akiyama, H. Kawakami, J. Wong, I. Oishi, R. Nishinakamura, Y. Kawakami, Sall4–Gli3 system in early limb progenitors is essential for the development of limb skeletal elements, Proc. Natl. Acad. Sci. U. S. A. 112 (2015) 5075–5080. [13] P. te Welscher, M. Fernandez-Teran, M.A. Ros, R. Zeller, Mutual genetic antagonism involving GLI3 and dHAND prepatterns the vertebrate limb bud mesenchyme prior to SHH signaling, Genes Dev. 16 (2002) 421–426. [14] D. Li, R. Sakuma, N.A. Vakili, R. Mo, V. Puviindran, S. Deimling, et al., Formation of proximal and anterior limb skeleton requires early function of Irx3 and Irx5 and is negatively regulated by Shh signaling, Dev. Cell 29 (2014) 233–240. [15] A. Galli, D. Robay, M. Osterwalder, X. Bao, J.D. Benazet, M. Tariq, et al., Distinct roles of Hand2 in initiating polarity and posterior Shh expression during the onset of mouse limb bud development, PLoS Genet. 6 (2010) e1000901. [16] T.D. Capellini, G. Di Giacomo, V. Salsi, A. Brendolan, E. Ferretti, D. Srivastava, et al., Pbx1/Pbx2 requirement for distal limb patterning is mediated by the hierarchical control of Hox gene spatial distribution and Shh expression, Development 133 (2006) 2263–2273. [17] M. Kmita, B. Tarchini, J. Zakany, M. Logan, C.J. Tabin, D. Duboule, Early developmental arrest of mammalian limbs lacking HoxA/HoxD gene function, Nature 435 (2005) 1113–1116. [18] B. Tarchini, D. Duboule, M. Kmita, Regulatory constraints in the evolution of the tetrapod limb anterior-posterior polarity, Nature 443 (2006) 985–988. [19] L.A. Lettice, I. Williamson, J.H. Wiltshire, S. Peluso, P.S. Devenney, A.E. Hill, et al., Opposing functions of the ETS factor family define Shh spatial expression in limb buds and underlie polydactyly, Dev. Cell 22 (2012) 459–467. [20] Z. Zhang, J.M. Verheyden, J.A. Hassell, X. Sun, FGF-regulated Etv genes are essential for repressing Shh expression in mouse limb buds, Dev. Cell 16 (2009) 607–613. [21] J. Mao, E. McGlinn, P. Huang, C.J. Tabin, A.P. McMahon, Fgf-dependent Etv4/5 activity is required for posterior restriction of Sonic Hedgehog and promoting outgrowth of the vertebrate limb, Dev. Cell 16 (2009) 600–606. [22] E. Kozhemyakina, A. Ionescu, A.B. Lassar, GATA6 is a crucial regulator of Shh in the limb bud, PLoS Genet. 10 (2014) e1004072.

7

[23] S. Qu, S.C. Tucker, J.S. Ehrlich, J.M. Levorse, L.A. Flaherty, R. Wisdom, et al., Mutations in mouse Aristaless-like4 cause Strong’s luxoid polydactyly, Development 125 (1998) 2711–2721. [24] D. Krawchuk, S.J. Weiner, Y.T. Chen, B.C. Lu, F. Costantini, R.R. Behringer, et al., Twist1 activity thresholds define multiple functions in limb development, Dev. Biol. 347 (2010) 133–146. [25] B. Xu, S.M. Hrycaj, D.C. McIntyre, N.C. Baker, J.K. Takeuchi, L. Jeannotte, et al., Hox5 interacts with Plzf to restrict Shh expression in the developing forelimb, Proc. Natl. Acad. Sci. U. S. A. 110 (2013) 19438–19443. [26] E. Anderson, R.E. Hill, Long range regulation of the sonic hedgehog gene, Curr. Opin. Genet. Dev. 27 (2014) 54–59. [27] L.A. Lettice, S.J. Heaney, L.A. Purdie, L. Li, P. de Beer, B.A. Oostra, et al., A long-range Shh enhancer regulates expression in the developing limb and fin and is associated with preaxial polydactyly, Hum. Mol. Genet. 12 (2003) 1725–1735. [28] T. Sagai, M. Hosoya, Y. Mizushina, M. Tamura, T. Shiroishi, Elimination of a long-range cis-regulatory module causes complete loss of limb-specific Shh expression and truncation of the mouse limb, Development 132 (2005) 797–803. [29] E. Anderson, S. Peluso, L.A. Lettice, R.E. Hill, Human limb abnormalities caused by disruption of hedgehog signaling, Trends Genet. 28 (2012) 364–373. [30] X. Sun, F.V. Mariani, G.R. Martin, Functions of FGF signalling from the apical ectodermal ridge in limb development, Nature 418 (2002) 501–508. [31] E. Laufer, C.E. Nelson, R.L. Johnson, B.A. Morgan, C. Tabin, Sonic hedgehog and Fgf-4 act through a signaling cascade and feedback loop to integrate growth and patterning of the developing limb bud, Cell 79 (1994) 993–1003. [32] L. Niswander, S. Jeffrey, G.R. Martin, C. Tickle, A positive feedback loop coordinates growth and patterning in the vertebrate limb, Nature 371 (1994) 609–612. [33] Y. Yang, L. Niswander, Interaction between the signaling molecules WNT7a and SHH during vertebrate limb development: dorsal signals regulate anteroposterior patterning, Cell 80 (1995) 939–947. [34] B.A. Parr, A.P. McMahon, Dorsalizing signal Wnt-7a required for normal polarity of D-V and A-P axes of mouse limb, Nature 374 (1995) 350–353. [35] J. Selever, W. Liu, M.F. Lu, R.R. Behringer, J.F. Martin, Bmp4 in limb bud mesoderm regulates digit pattern by controlling AER development, Dev. Biol. 276 (2004) 268–279. [36] A. Bandyopadhyay, K. Tsuji, K. Cox, B.D. Harfe, V. Rosen, C.J. Tabin, Genetic analysis of the roles of BMP2, BMP4, and BMP7 in limb patterning and skeletogenesis, PLoS Genet. 2 (2006) e216. [37] M.F. Bastida, R. Sheth, M.A. Ros, A BMP-Shh negative-feedback loop restricts Shh expression during limb development, Development 136 (2009) 3779–3789. [38] J.J. Sanz-Ezquerro, C. Tickle, Autoregulation of Shh expression and Shh induction of cell death suggest a mechanism for modulating polarising activity during chick limb development, Development 127 (2000) 4811–4823. [39] P.J. Scherz, E. McGlinn, S. Nissim, C.J. Tabin, Extended exposure to Sonic hedgehog is required for patterning the posterior digits of the vertebrate limb, Dev. Biol. 308 (2007) 343–354. [40] A. Gritli-Linde, P. Lewis, A.P. McMahon, A. Linde, The whereabouts of a morphogen: direct evidence for short- and graded long-range activity of hedgehog signaling peptides, Dev. Biol. 236 (2001) 364–386. [41] V. Marigo, R.L. Johnson, A. Vortkamp, C.J. Tabin, Sonic hedgehog differentially regulates expression of GLI and GLI3 during limb development, Dev. Biol. 180 (1996) 273–283. [42] V. Marigo, M.P. Scott, R.L. Johnson, L.V. Goodrich, C.J. Tabin, Conservation in hedgehog signaling: induction of a chicken patched homolog by Sonic hedgehog in the developing limb, Development 122 (1996) 1225–1233. [43] Y. Chen, G. Struhl, Dual roles for patched in sequestering and transducing hedgehog, Cell 87 (1996) 553–563. [44] N.C. Butterfield, V. Metzis, E. McGlinn, S.J. Bruce, B.J. Wainwright, C. Wicking, Patched 1 is a crucial determinant of asymmetry and digit number in the vertebrate limb, Development 136 (2009) 3515–3524. [45] V. Marigo, R.A. Davey, Y. Zuo, J.M. Cunningham, C.J. Tabin, Biochemical evidence that patched is the Hedgehog receptor, Nature 384 (1996) 176–179. [46] J. Briscoe, Y. Chen, T.M. Jessell, G. Struhl, A hedgehog-insensitive form of patched provides evidence for direct long-range morphogen activity of sonic hedgehog in the neural tube, Mol. Cell 7 (2001) 1279–1291. [47] J. Jeong, A.P. McMahon, Growth and pattern of the mammalian neural tube are governed by partially overlapping feedback activities of the hedgehog antagonists patched 1 and Hhip1, Development 132 (2005) 143–154. [48] J.A. Porter, K.E. Young, P.A. Beachy, Cholesterol modification of hedgehog signaling proteins in animal development, Science 274 (1996) 255–259. [49] R.B. Pepinsky, C. Zeng, D. Wen, P. Rayhorn, D.P. Baker, K.P. Williams, et al., Identification of a palmitic acid-modified form of human Sonic hedgehog, J. Biol. Chem. 273 (1998) 14037–14045. [50] A. Rietveld, S. Neutz, K. Simons, S. Eaton, Association of sterol- and glycosylphosphatidylinositol-linked proteins with Drosophila raft lipid microdomains, J. Biol. Chem. 274 (1999) 12049–12054. [51] Y. Li, H. Zhang, Y. Litingtung, C. Chiang, Cholesterol modification restricts the spread of Shh gradient in the limb bud, Proc. Natl. Acad. Sci. U. S. A. 103 (2006) 6548–6553.

Please cite this article in press as: J. Lopez-Rios, The many lives of SHH in limb development and evolution, Semin Cell Dev Biol (2015), http://dx.doi.org/10.1016/j.semcdb.2015.12.018

G Model YSCDB-1906; No. of Pages 9 8

ARTICLE IN PRESS J. Lopez-Rios / Seminars in Cell & Developmental Biology xxx (2015) xxx–xxx

[52] M.H. Chen, Y.J. Li, T. Kawakami, S.M. Xu, P.T. Chuang, Palmitoylation is required for the production of a soluble multimeric Hedgehog protein complex and long-range signaling in vertebrates, Genes Dev. 18 (2004) 641–659. [53] T.A. Sanders, E. Llagostera, M. Barna, Specialized filopodia direct long-range transport of SHH during vertebrate tissue patterning, Nature 497 (2013) 628–632. [54] F.-A. Ramírez-Weber, T.B. Kornberg, Cytonemes: cellular processes that project to the principal signaling center in Drosohila imaginal discs, Cell 97 (1999) 599–607. [55] B.L. Allen, T. Tenzen, A.P. McMahon, The Hedgehog-binding proteins Gas1 and Cdo cooperate to positively regulate Shh signaling during mouse development, Genes Dev. 21 (2007) 1244–1257. [56] B.L. Allen, J.Y. Song, L. Izzi, I.W. Althaus, J.S. Kang, F. Charron, et al., Overlapping roles and collective requirement for the coreceptors GAS1, CDO, and BOC in SHH pathway function, Dev. Cell 20 (2011) 775–787. [57] T. Tenzen, B.L. Allen, F. Cole, J.S. Kang, R.S. Krauss, A.P. McMahon, The cell surface membrane proteins Cdo and Boc are components and targets of the Hedgehog signaling pathway and feedback network in mice, Dev. Cell 10 (2006) 647–656. [58] D.C. Martinelli, C.M. Fan, Gas1 extends the range of Hedgehog action by facilitating its signaling, Genes Dev. 21 (2007) 1231–1243. [59] Y. Liu, C. Liu, Y. Yamada, C.M. Fan, Growth arrest specific gene 1 acts as a region-specific mediator of the Fgf10/Fgf8 regulatory loop in the limb, Development 129 (2002) 5289–5300. [60] W. Zhang, J.S. Kang, F. Cole, M.J. Yi, R.S. Krauss, Cdo functions at multiple points in the Sonic Hedgehog pathway, and Cdo-deficient mice accurately model human holoprosencephaly, Dev. Cell 10 (2006) 657–665. [61] S. Nachtergaele, L.K. Mydock, K. Krishnan, J. Rammohan, P.H. Schlesinger, D.F. Covey, et al., Oxysterols are allosteric activators of the oncoprotein Smoothened, Nat. Chem. Biol. 8 (2012) 211–220. [62] K.V. Dorn, C.E. Hughes, R. Rohatgi, A Smoothened-Evc2 complex transduces the Hedgehog signal at primary cilia, Dev. Cell 23 (2012) 823–835. [63] C. Yang, W. Chen, Y. Chen, J. Jiang, Smoothened transduces Hedgehog signal by forming a complex with Evc/Evc2, Cell Res. 22 (2012) 1593–1604. [64] G.V. Pusapati, C.E. Hughes, K.V. Dorn, D. Zhang, P. Sugianto, L. Aravind, et al., EFCAB7 and IQCE regulate hedgehog signaling by tethering the EVC–EVC2 complex to the base of primary cilia, Dev. Cell 28 (2014) 483–496. [65] F. Hildebrandt, T. Benzing, N. Katsanis, Ciliopathies, N. Engl. J. Med. 364 (2011) 1533–1543. [66] J.H. Kong, L. Yang, E. Dessaud, K. Chuang, D.M. Moore, R. Rohatgi, et al., Notch activity modulates the responsiveness of neural progenitors to sonic hedgehog signaling, Dev. Cell 33 (2015) 373–387. [67] M. Stasiulewicz, S.D. Gray, I. Mastromina, J.C. Silva, M. Bjorklund, P.A. Seymour, et al., A conserved role for Notch signaling in priming the cellular response to Shh through ciliary localisation of the key Shh transducer Smo, Development 142 (2015) 2291–2303. [68] L. Panman, A. Galli, N. Lagarde, O. Michos, G. Soete, A. Zuniga, et al., Differential regulation of gene expression in the digit forming area of the mouse limb bud by SHH and gremlin 1/FGF-mediated epithelial–mesenchymal signalling, Development 133 (2006) 3419–3428. [69] E. McGlinn, K.L. van Bueren, S. Fiorenza, R. Mo, A.M. Poh, A. Forrest, et al., Pax9 and Jagged1 act downstream of Gli3 in vertebrate limb development, Mech. Dev. 122 (2005) 1218–1233. [70] S.A. Vokes, H. Ji, W.H. Wong, A.P. McMahon, A genome-scale analysis of the cis-regulatory circuitry underlying sonic hedgehog-mediated patterning of the mammalian limb, Genes Dev. 22 (2008) 2651–2663. [71] K.N. Falkenstein, S.A. Vokes, Transcriptional regulation of graded Hedgehog signaling, Semin. Cell Dev. Biol. 33 (2014) 73–80. [72] B. Wang, J.F. Fallon, P.A. Beachy, Hedgehog-regulated processing of Gli3 produces an anterior/posterior repressor gradient in the developing vertebrate limb, Cell 100 (2000) 423–434. [73] E.W. Humke, K.V. Dorn, L. Milenkovic, M.P. Scott, R. Rohatgi, The output of Hedgehog signaling is controlled by the dynamic association between Suppressor of Fused and the Gli proteins, Genes Dev. 24 (2010) 670–682. [74] H.L. Park, C. Bai, K.A. Platt, M.P. Matise, A. Beeghly, C.C. Hui, et al., Mouse Gli1 mutants are viable but have defects in SHH signaling in combination with a Gli2 mutation, Development 127 (2000) 1593–1605. [75] M. Bowers, L. Eng, Z. Lao, R.K. Turnbull, X. Bao, E. Riedel, et al., Limb anterior–posterior polarity integrates activator and repressor functions of GLI2 as well as GLI3, Dev. Biol. 370 (2012) 110–124. [76] O. Zhulyn, C.C. Hui, Sufu and Kif7 in limb patterning and development, Dev. Dyn. 244 (2015) 468–478. [77] S. Ahn, A.L. Joyner, Dynamic changes in the response of cells to positive hedgehog signaling during mouse limb patterning, Cell 118 (2004) 505–516. [78] M. Cohen, A. Kicheva, A. Ribeiro, R. Blassberg, K.M. Page, C.P. Barnes, et al., Ptch1 and Gli regulate Shh signalling dynamics via multiple mechanisms, Nat. Commun. 6 (2015) 6709. [79] J.P. Junker, K.A. Peterson, Y. Nishi, J. Mao, A.P. McMahon, A. van Oudenaarden, A predictive model of bifunctional transcription factor signaling during embryonic tissue patterning, Dev. Cell 31 (2014) 448–460. [80] P. te Welscher, A. Zuniga, S. Kuijper, T. Drenth, H.J. Goedemans, F. Meijlink, et al., Progression of vertebrate limb development through SHH-mediated counteraction of GLI3, Science 298 (2002) 827–830.

[81] Y. Litingtung, R.D. Dahn, Y. Li, J.F. Fallon, C. Chiang, Shh and Gli3 are dispensable for limb skeleton formation but regulate digit number and identity, Nature 418 (2002) 979–983. [82] J.P. Lewandowski, F. Du, S. Zhang, M.B. Powell, K.N. Falkenstein, H. Ji, et al., Spatiotemporal regulation of GLI target genes in the mammalian limb bud, Dev. Biol. 406 (2015) 92–103. [83] J. Bose, L. Grotewold, U. Ruther, Pallister-hall syndrome phenotype in mice mutant for Gli3, Hum. Mol. Genet. 11 (2002) 1129–1135. [84] P. Hill, K. Gotz, U. Ruther, A SHH-independent regulation of Gli3 is a significant determinant of anteroposterior patterning of the limb bud, Dev. Biol. 328 (2009) 506–516. [85] P. Hill, B. Wang, U. Ruther, The molecular basis of Pallister Hall associated polydactyly, Hum. Mol. Genet. 16 (2007) 2089–2096. [86] C. Wang, U. Ruther, B. Wang, The Shh-independent activator function of the full-length Gli3 protein and its role in vertebrate limb digit patterning, Dev. Biol. 305 (2007) 460–469. [87] C. Wang, Y. Pan, B. Wang, A hypermorphic mouse Gli3 allele results in a polydactylous limb phenotype, Dev. Dyn. 236 (2007) 769–776. [88] T. Cao, C. Wang, M. Yang, C. Wu, B. Wang, Mouse limbs expressing only the Gli3 repressor resemble those of Sonic hedgehog mutants, Dev. Biol. 379 (2013) 221–228. [89] S. Probst, C. Kraemer, P. Demougin, R. Sheth, G.R. Martin, H. Shiratori, et al., SHH propagates distal limb bud development by enhancing CYP26B1-mediated retinoic acid clearance via AER-FGF signalling, Development 138 (2011) 1913–1923. [90] K.W. Kinzler, B. Vogelstein, The GLI gene encodes a nuclear protein which binds specific sequences in the human genome, Mol. Cell. Biol. 10 (1990) 634–642. [91] O. Hallikas, K. Palin, N. Sinjushina, R. Rautiainen, J. Partanen, E. Ukkonen, et al., Genome-wide prediction of mammalian enhancers based on analysis of transcription-factor binding affinity, Cell 124 (2006) 47–59. [92] K.A. Peterson, Y. Nishi, W. Ma, A. Vedenko, L. Shokri, X. Zhang, et al., Neural-specific Sox2 input and differential Gli-binding affinity provide context and positional information in Shh-directed neural patterning, Genes Dev. 26 (2012) 2802–2816. [93] F. Liu, J. Massague, A. Ruiz i Altaba, Carboxy-terminally truncated Gli3 proteins associate with Smads, Nat. Genet. 20 (1998) 325–326. [94] Y. Koyabu, K. Nakata, K. Mizugishi, J. Aruga, K. Mikoshiba, Physical and functional interactions between Zic and Gli proteins, J. Biol. Chem. 276 (2001) 6889–6892. [95] P. Dai, T. Shinagawa, T. Nomura, J. Harada, S.C. Kaul, R. Wadhwa, et al., Ski is involved in transcriptional regulation by the repressor and full-length forms of Gli3, Genes Dev. 16 (2002) 2843–2848. [96] F. Ulloa, N. Itasaki, J. Briscoe, Inhibitory Gli3 activity negatively regulates Wnt/beta-catenin signaling, Curr. Biol. 17 (2007) 545–550. [97] M.D. Nye, L.L. Almada, M.G. Fernandez-Barrena, D.L. Marks, S.F. Elsawa, A. Vrabel, et al., The transcription factor GLI1 interacts with SMAD proteins to modulate transforming growth factor beta-induced gene expression in a p300/CREB-binding protein-associated factor (PCAF)-dependent manner, J. Biol. Chem. 289 (2014) 15495–15506. [98] Y. Chen, V. Knezevic, V. Ervin, R. Hutson, Y. Ward, S. Mackem, Direct interaction with Hoxd proteins reverses Gli3-repressor function to promote digit formation downstream of Shh, Development 131 (2004) 2339–2347. [99] T. Oosterveen, S. Kurdija, Z. Alekseenko, C.W. Uhde, M. Bergsland, M. Sandberg, et al., Mechanistic differences in the transcriptional interpretation of local and long-range Shh morphogen signaling, Dev. Cell 23 (2012) 1006–1019. [100] D.S. Parker, M.A. White, A.I. Ramos, B.A. Cohen, S. Barolo, The cis-regulatory logic of Hedgehog gradient responses: key roles for gli binding affinity, competition, and cooperativity, Sci. Signal 4 (2011) ra38. [101] A.I. Ramos, S. Barolo, Low-affinity transcription factor binding sites shape morphogen responses and enhancer evolution, Philos. Trans. R. Soc. Lond. B Biol. Sci. 368 (2013) 20130018. [102] M. Cohen, J. Briscoe, R. Blassberg, Morphogen interpretation: the transcriptional logic of neural tube patterning, Curr. Opin. Genet. Dev. 23 (2013) 423–428. [103] P. Kraus, D. Fraidenraich, C.A. Loomis, Some distal limb structures develop in mice lacking Sonic hedgehog signaling, Mech. Dev. 100 (2001) 45–58. [104] B.D. Harfe, P.J. Scherz, S. Nissim, H. Tian, A.P. McMahon, C.J. Tabin, Evidence for an expansion-based temporal Shh gradient in specifying vertebrate digit identities, Cell 118 (2004) 517–528. [105] A. Zuniga, A.P. Haramis, A.P. McMahon, R. Zeller, Signal relay by BMP antagonism controls the SHH/FGF4 feedback loop in vertebrate limb buds, Nature 401 (1999) 598–602. [106] J.D. Benazet, M. Bischofberger, E. Tiecke, A. Goncalves, J.F. Martin, A. Zuniga, et al., A self-regulatory system of interlinked signaling feedback loops controls mouse limb patterning, Science 323 (2009) 1050–1053. [107] P. Lu, G. Minowada, G.R. Martin, Increasing Fgf4 expression in the mouse limb bud causes polysyndactyly and rescues the skeletal defects that result from loss of Fgf8 function, Development 133 (2006) 33–42. [108] K. Handschuh, J. Feenstra, M. Koss, E. Ferretti, M. Risolino, R. Zewdu, et al., ESCRT-II/Vps25 constrains digit number by endosome-mediated selective modulation of FGF-SHH signaling, Cell. Rep. 9 (2014) 674–687. [109] O. Michos, L. Panman, K. Vintersten, K. Beier, R. Zeller, A. Zuniga, Gremlin-mediated BMP antagonism induces the epithelial–mesenchymal

Please cite this article in press as: J. Lopez-Rios, The many lives of SHH in limb development and evolution, Semin Cell Dev Biol (2015), http://dx.doi.org/10.1016/j.semcdb.2015.12.018

G Model YSCDB-1906; No. of Pages 9

ARTICLE IN PRESS J. Lopez-Rios / Seminars in Cell & Developmental Biology xxx (2015) xxx–xxx

[110]

[111] [112]

[113]

[114] [115]

[116] [117]

[118]

[119] [120] [121] [122]

[123] [124]

[125] [126] [127]

[128] [129] [130] [131]

[132] [133]

[134]

feedback signaling controlling metanephric kidney and limb organogenesis, Development 131 (2004) 3401–3410. A. Zuniga, F. Laurent, J. Lopez-Rios, C. Klasen, N. Matt, R. Zeller, Conserved cis-regulatory regions in a large genomic landscape control SHH and BMP-regulated Gremlin1 expression in mouse limb buds, BMC Dev. Biol. 12 (2012) 23. M. Towers, R. Mahood, Y. Yin, C. Tickle, Integration of growth and specification in chick wing digit-patterning, Nature 452 (2008) 882–886. J. Lopez-Rios, D. Speziale, D. Robay, M. Scotti, M. Osterwalder, G. Nusspaumer, et al., GLI3 constrains digit number by controlling both progenitor proliferation and BMP-dependent exit to chondrogenesis, Dev. Cell 22 (2012) 837–848. J. Zhu, E. Nakamura, M.T. Nguyen, X. Bao, H. Akiyama, S. Mackem, Uncoupling Sonic hedgehog control of pattern and expansion of the developing limb bud, Dev. Cell 14 (2008) 624–632. L.G. Biesecker, What you can learn from one gene: GLI3, J. Med. Genet. 43 (2006) 465–469. C.M. Bouldin, A. Gritli-Linde, S. Ahn, B.D. Harfe, Shh pathway activation is present and required within the vertebrate limb bud apical ectodermal ridge for normal autopod patterning, Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 5489–5494. J.M. Verheyden, X. Sun, An Fgf/Gremlin inhibitory feedback loop triggers termination of limb bud outgrowth, Nature 454 (2008) 638–641. K. Chinnaiya, C. Tickle, M. Towers, Sonic hedgehog-expressing cells in the developing limb measure time by an intrinsic cell cycle clock, Nat. Commun. 5 (4230) (2014). J.W. Saunders Jr., M.T. Gasseling, Ectodermal-mesenchymal interactions in the origin of limb symmetry, in: R. Fleichmeyer, R.E. Billingham (Eds.), Epithelial–Mesenchymal Interactions, Williams and Wilkin, Baltimore, 1968, pp. 78–97. C. Tickle, The early history of the polarizing region: from classical embryology to molecular biology, Int. J. Dev. Biol. 46 (2002) 847–852. L. Wolpert, Positional information and the spatial pattern of cellular differentiation, J. Theor. Biol. 25 (1969) 1–47. R.D. Riddle, R.L. Johnson, E. Laufer, C. Tabin, Sonic hedgehog mediates the polarizing activity of the ZPA, Cell 75 (1993) 1401–1416. Y. Yang, G. Drossopoulou, P.T. Chuang, D. Duprez, E. Marti, D. Bumcrot, et al., Relationship between does, distance and time in Sonic Hedgehog-mediated regulation of anteroposterior polarity in the chick limb, Development 124 (1997) 4393–4404. J. Zakany, D. Duboule, The role of Hox genes during vertebrate limb development, Curr. Opin. Genet. Dev. 17 (2007) 359–366. C. Fromental-Ramain, X. Warot, S. Lakkaraju, B. Favier, H. Haack, C. Birling, et al., Specific and redundant functions of the paralogous Hoxa-9 and Hoxd-9 genes in forelimb and axial skeleton patterning, Development 122 (1996) 461–472. D.M. Wellik, M.R. Capecchi, Hox10 and Hox11 genes are required to globally pattern the mammalian skeleton, Science 301 (2003) 363–367. J. Zakany, M. Kmita, D. Duboule, A dual role for Hox genes in limb anterior-posterior asymmetry, Science 304 (2004) 1669–1672. M. Barna, P.P. Pandolfi, L. Niswander, Gli3 and Plzf cooperate in proximal limb patterning at early stages of limb development, Nature 436 (2005) 277–281. A.P. Davis, D.P. Witte, H.M. Hsieh-Li, S. Potter, M. Capecchi, Absence of radius and ulna in mice lacking hoxa-11 and hoxd-11, Nature 375 (1995) 791–795. L. Wolpert, A. Hornbruch, Positional signalling and the development of the humerus in the chick limb bud, Development 100 (1987) 333–338. M. Towers, C. Tickle, Growing models of vertebrate limb development, Development 136 (2009) 179–190. J. Zhu, S. Mackem, Analysis of mutants with altered shh activity and posterior digit loss supports a biphasic model for shh function as a morphogen and mitogen, Dev. Dyn. 240 (2011) 1303–1310. C.J. Tabin, A.P. McMahon, Developmental biology grasping limb patterning, Science 321 (2008) 350–352. M. Towers, J. Signolet, A. Sherman, H. Sang, C. Tickle, Insights into bird wing evolution and digit specification from polarizing region fate maps, Nat. Commun. 2 (2011) 426. J. Raspopovic, L. Marcon, L. Russo, J. Sharpe, Digit patterning is controlled by a Bmp-Sox9-Wnt turing network modulated by morphogen gradients, Science 345 (2014) 566–570.

9

[135] R. Sheth, L. Marcon, M.F. Bastida, M. Junco, L. Quintana, R. Dahn, et al., Hox genes regulate digit patterning by controlling the wavelength of a turing-type mechanism, Science 338 (2012) 1476–1480. [136] J.B. Green, J. Sharpe, Positional information and reaction-diffusion: two big ideas in developmental biology combine, Development 142 (2015) 1203–1211. [137] P. Dollé, J.C. Izpisua-Belmonte, H. Falkenstein, A. Renucci, D. Duboule, Coordinate expression of the murine Hox-5 complex homeobox-containing genes during limb pattern formation, Nature 342 (1989) 767–772. [138] C.J. Tabin, Why we have (only) five fingers per hand: hox genes and the evolution of paired limbs, Development 116 (1992) 289–296. [139] B.A. Morgan, J.C. Izpisua Belmonte, D. Duboule, C.J. Tabin, Targeted misexpression of Hox-4.6 in the avian limb bud causes apparent homeotic transformations [see comments], Nature 358 (1992) 236–239. [140] T. Montavon, J.F. Le Garrec, M. Kerszberg, D. Duboule, Modeling Hox gene regulation in digits: reverse collinearity and the molecular origin of thumbness, Genes Dev. 22 (2008) 346–359. [141] J.M. Woltering, D. Duboule, The origin of digits: expression patterns versus regulatory mechanisms, Dev. Cell 18 (2010) 526–532. [142] D. Buscher, B. Bosse, J. Heymer, U. Ruther, Evidence for genetic control of Sonic hedgehog by Gli3 in mouse limb development, Mech. Dev. 62 (1997) 175–182. [143] T. Montavon, N. Soshnikova, B. Mascrez, E. Joye, L. Thevenet, E. Splinter, et al., A regulatory archipelago controls Hox genes transcription in digits, Cell 147 (2011) 1132–1145. [144] R.D. Dahn, J.F. Fallon, Interdigital regulation of digit identity and homeotic transformation by modulated BMP signaling, Science 289 (2000) 438–441. [145] T. Suzuki, S.M. Hasso, J.F. Fallon, Unique SMAD1/5/8 activity at the phalanx-forming region determines digit identity, Proc. Natl. Acad. Sci. U. S. A. 105 (2008) 4185–4190. [146] F. Witte, D. Chan, A.N. Economides, S. Mundlos, S. Stricker, Receptor tyrosine kinase-like orphan receptor 2 (ROR2) and Indian hedgehog regulate digit outgrowth mediated by the phalanx-forming region, Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 14211–14216. [147] J.A. Montero, C.I. Lorda-Diez, Y. Ganan, D. Macias, J.M. Hurle, Activin/TGFbeta and BMP crosstalk determines digit chondrogenesis, Dev. Biol. 321 (2008) 343–356. [148] T. Sagai, H. Masuya, M. Tamura, K. Shimizu, Y. Yada, S. Wakana, et al., Phylogenetic conservation of a limb-specific, cis-acting regulator of Sonic hedgehog (Shh), Mamm. Genome 15 (2004) 23–34. [149] M.J. Cohn, C. Tickle, Developmental basis of limblessness and axial patterning in snakes, Nature 399 (1999) 474–479. [150] M.D. Shapiro, J. Hanken, N. Rosenthal, Developmental basis of evolutionary digit loss in the Australian lizard Hemiergis, J. Exp. Zool. B Mol. Dev. Evol. 297 (2003) 48–56. [151] P. Alberch, E.A. Gale, Size dependence during the development of the amphibian foot: colchicine-induced digital loss and reduction, J. Embryol. Exp. Morphol. 76 (1983) 177–197. [152] X. Xu, S. Mackem, Tracing the evolution of avian wing digits, Curr. Biol. 23 (2013) R538–R544. [153] M.A. de Bakker, D.A. Fowler, K. den Oude, E.M. Dondorp, M.C. Navas, J.O. Horbanczuk, et al., Digit loss in archosaur evolution and the interplay between selection and constraints, Nature 500 (2013) 445–448. [154] J. Lopez-Rios, A. Duchesne, D. Speziale, G. Andrey, K.A. Peterson, P. Germann, et al., Attenuated sensing of SHH by Ptch1 underlies evolution of bovine limbs, Nature 511 (2014) 46–51. [155] A.B. Clifford, The evolution of the unguligrade manus in artiodactyls, J. Vert. Paleontol. 30 (2010) 1827–1839. [156] K.E. Sears, A.K. Bormet, A. Rockwell, L.E. Powers, L. Noelle Cooper, M.B. Wheeler, Developmental basis of mammalian digit reduction: a case study in pigs, Evol. Dev. 13 (2011) 533–541. [157] K.L. Cooper, K.E. Sears, A. Uygur, J. Maier, K.S. Baczkowski, M. Brosnahan, et al., Patterning and post-patterning modes of evolutionary digit loss in mammals, Nature 511 (2014) 41–45.

Please cite this article in press as: J. Lopez-Rios, The many lives of SHH in limb development and evolution, Semin Cell Dev Biol (2015), http://dx.doi.org/10.1016/j.semcdb.2015.12.018