Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Dev Cell. Author manuscript; available in PMC 2011 August 30.
Published in final edited form as:
PMCID: PMC3164484

Fgf-dependent Etv4/5 activity is required for posterior restriction of Sonic hedgehog and promoting outgrowth of the vertebrate limb


Cross-talk between the fibroblast growth factor (FGF) and Sonic Hedgehog (Shh) pathways is critical for proper patterning and growth of the developing limb bud. Here, we show that FGF dependent activation of the ETS transcription factors Etv4 and Etv5 contributes to proximal-distal limb outgrowth. Surprisingly, blockage of Etv activity in early distal mesenchyme also resulted in ectopic, anterior expansion of Shh leading to a polydactylous phenotype. These data indicate an unexpected function for an FGF/Etv pathway in anterior-posterior patterning. In this, FGF activity in the limb is not only responsible for maintaining posterior-specific Shh expression, but also acts via Etvs to prevent inappropriate anterior expansion of Shh. This study identifies another level of genetic interaction between the orthogonal axes during limb development.


Patterning along the anterior-posterior axis (A-P; thumb-little finger) is instructed by a discrete region of cells located at the posterior margin known as the zone of polarizing activity (ZPA), identifiable by the expression of Shh, a secreted molecule known to recapitulate all function of the ZPA (McGlinn and Tabin, 2006; McMahon et al., 2003). In the absence of a Shh input, all but the most anterior digit fail to form (Chiang et al., 2001; Chiang et al., 1996; Lewis et al., 2001). Conversely, ectopic Shh expression in more anterior positions results in extra digits (Riddle et al., 1993).

A number of factors contribute, positively or negatively, to specifying the domain of Shh expression (Charite et al., 2000; Fernandez-Teran et al., 2000; Masuya et al., 1995; Nissim et al., 2006). In particular, broad posterior expression of Hand2 generates a zone of mesenchyme that is competent to express Shh (Charite et al., 2000). Marginal restriction may be conferred by a requirement for signals emanating from the non-AER dorsal-ventral border ectoderm (Nissim et al., 2006). Gli3 repressor function also helps prevent ectopic Shh expression in the anterior limb bud (Hui and Joyner, 1993). However, the active mechanisms that tightly restrict normal Shh expression to distal posterior limb mesenchyme are not fully defined.

Outgrowth along the proximal-distal (P-D; shoulder-finger tip) axis requires signals emanating from a strip of columnar epithelium at the distal tip of the limb bud know as the apical ectodermal ridge (AER). Fibroblast growth factor (Fgf) signaling is sufficient to recapitulate all AER functions (Fallon et al., 1994; Niswander and Martin, 1993); AER-derived Fgf8 and Fgf4 are the major determinants of P-D patterning (Boulet et al., 2004; Mariani et al., 2008; Sun et al., 2002). Importantly, co-ordinated limb growth requires cross-talk between the orthogonal axes (Vogel and Tickle, 1993). In this, Fgf signaling is required for both the initiation and maintenance of Shh expression. Reciprocally, Shh acts indirectly to antagonize BMP-mediated inhibition of Fgf expression and AER morphology (Capdevila et al., 1999; Laufer et al., 1994; Niswander et al., 1994; Zuniga et al., 1999). Thus, there is a tight interdependency of each axis for proper limb development.

Three members of the Etv4/Pea3 subfamily of ETS transcription factors (Etv4/Pea3, Etv5/Erm and Etv1/Er81) are transcriptionally regulated by Fgf-signals in many embryonic contexts (Brent and Tabin, 2004; Firnberg and Neubuser, 2002; Raible and Brand, 2001; Roehl and Nusslein-Volhard, 2001). Two of these, Etv4 and Etv5, share similar overlapping domains of expression at many embryonic sites where Fgf signaling is critical including the vertebrate limb (Chotteau-Lelievre et al., 2001). Here, we identify roles for Etvs in P-D outgrowth and, unexpectedly, in restricting mesenchymal Shh expression from the anterior, thus contributing to the local definition of the ZPA. These findings highlight an unexpected role for FGF signaling in establishing the appropriate posterior restriction of the ZPA signal.


FGF dependent Etv4/5 expression in the limb mesenchyme

Previous reports have documented expression of the Etv4 family of ETS transcriptional factors at select stages of embryonic development (Chotteau-Lelievre et al., 2001; Kawakami et al., 2003) We re-examined the expression pattern of three members of this family, Etv1, Etv4 and Etv5. Etv1 expression was not detected at high levels in the mouse limb bud and was not examined further; however distal limb mesenchymal expression has been reported in chick (Kawakami et al., 2003). Etv4 and Etv5 share a similar pattern of expression in mouse and chick (Figure 1 and Figure S1). Prior to limb bud outgrowth, at the earliest stages of mouse limb bud development, Etv4 and Etv5 are broadly expressed in the limb mesenchyme (E8.75; Figure 1). As the limb bud extends, the expression domains of Etv4 and Etv5 become more distally restricted, underlying the FGF expressing AER (highlighted by Fgf8 expression, Figure 1). Distal-restricted mesenchymal expression of both genes is maintained until at least E12.5. This general distribution of Etv factors is conserved in the chick limb; moreover, their expression is inhibited by SU5402-treatment (Figure S2A) consistent with a role for an FGF receptor-like kinase activity maintaining distal Etv4/5 expression.

Figure 1
Etv4/5 and Fgf8 exhibit dynamic expression in the developing mouse limb bud

Repression of Etv targets in the limb mesenchyme results in P-D and A-P pattern defects

Etv5 null embryos die at E9.5 prior to limb development (Liu et al., 2003). Etv4 null mutants are viable with no apparent limb phenotype (Laing et al., 2000). However, their similar molecular structure and tightly overlapping expression make it highly likely that Etv4 and Etv5 have redundant activities. We therefore undertook a constitutive repressor approach to counter their normal activator functions. In this, the Engrailed repressor domain was fused immediately downstream of the ETS DNA-binding domain of Etv4. In vitro studies have previously shown that this construct represses transcriptional targets of all three members of this sub-family (Shepherd et al., 2001); consequently, we term this inhibitory form EtvEnR. A conditional allele of EtvEnR, R26-EtvEnR, was generated by targeting a cDNA fragment encoding the Etv4EnR fusion protein and an IRES-nuclear GFP tag reporter, into the ubiquitously expressed Rosa26 locus (R26). The presence of a LoxP-flanked polyadenylation stop sequence cassette (Figure S3A) blocks EtvEnR transcription at the R26 locus until Cre-mediated excision of the cassette. We then used a Prx1-Cre transgenic driver strain, in which the promoter is broadly active in the fore-limb mesenchyme prior to limb outgrowth (Logan et al., 2002), to examine effects on limb development (Figure 2A). Zeugopod elements (radius and ulna/tibia and fibular) were shortened; consistent with the hypothesis that Etv activity normally mediates, at least in part, Fgf-dependent limb growth. We failed to detect any significant change in cell death (using TdT-mediated dUTP nick-end labeling) or proliferation (using an anti-phospho-histone H3 antibody) in E10.5 Prx1-Cre;R26-EtvEnR forelimb buds (data not shown) reflecting the subtle phenotype observed. However, although heterozygosity for Fgfr1 does not cause a phenotype on its own, the R26-EtvEnR zeugopod phenotype was markedly enhanced on reduction of FgfR1 dosage in FgfR1n/+ embryos (Figure 2C).

Figure 2
Repressing Etv target genes results in an FgfR1 dosage sensitive shortening of limb elements and Shh dependent polydactyly

Mostly strikingly, Prx1-Cre;R26-EtvEnR limbs were polydactylous with 6-7 digits. Importantly, the most anterior digit was triphalangeal not biphalangeal as is the norm, suggesting a loss of digit 1 identity and the adoption of a more posterior digit pattern as all other digits have a triphalangeal organization. Less well-formed ectopic cartilaginous elements were also observed in the anterior limb. In a small number of cases, ectopic digits were located more centrally consisting of 3 phalangeal elements without an additional metacarpal element. Taken together, our data indicates a dual role for Etvs in development of both the P-D and A-P axes.

Repression of Etv targets leads to ectopic Shh expression in the distal mesenchyme

An increased number of digits, and posteriorization of the most anterior digit, would be predicted to result from ectopic Shh activities. Expression of Shh in wild type embryos is restricted to the posterior mesenchyme of the ZPA, but is markedly extended anteriorly in E9.5 limbs of Prx1-Cre;R26-EtvEnR embryos (Figure 3A). By E10.5, ectopic Shh expression was evident across the entire distal limb bud to the anterior border (Figure 3A). Ectopic Shh expression faded along the distal A-P extent of the autopod by E11.5, whereas expression remained strong at both the posterior and anterior margins, suggesting the superposition of a distinct mode of regulation (Figure 3A; see also Figure 4 below). Strikingly, ectopic Shh was only observed in the distal-most mesenchyme, despite Cre-mediated activation of EtvEnR throughout the entire limb mesenchyme from early stages. While these observations are surprising in the context of previously described epistatic relationship between FGF and Shh, they are consistent with, and likely explain, the polydactylous phenotype.

Figure 3
Ectopic activation of Shh and Shh targets results from an inhibition of Etv-dependent regulation
Figure 4
Etv4/5 function in early distal mesenchyme to restrict Shh expression

The expanded distal domain of Shh expression was accompanied by up-regulation and anterior expansion of general Shh target genes such as Gli1 and Patched1, as well as downstream limb specific regulators, including Hand2, Hoxd12, Hoxd13, and Gremlin (Figure 3B and data not shown). Gremlin expression was also observed within the very distal-most mesenchyme at E10.5, a time when cumulative AER-Fgf signaling normally silences Gremlin transcription in this zone (Figure 3B; Verheyden and Sun, 2008). Conversely, expression of the anterior specific regulator, Alx4, was reduced and more proximally restricted (Figure 3B). Together the data support a simple model: the loss of Etv target gene activity results in an anterior expansion of Shh expression resulting in ectopic Shh signaling in anterior limb regions. Gli3 and Hand2 were expressed normally on activation of EtvEnR prior to Shh expression (E9.25); thus, an alternative model were prepatterning of the limb is perturbed seems unlikely (Figure S3B). As Etv4 and Etv5 are themselves dependent on Fgf signals, we conclude that Fgf signaling normally restricts Shh expression to the posterior mesenchyme of the ZPA.

EtvEnR activity likely reflects endogenous negative regulation of Shh by Etv4/5

The conclusion-that Etvs normally act downstream of Fgf signaling to negatively regulate Shh expression is surprising given the body of evidence demonstrating that Fgfs positively regulate Shh expression in the posterior limb bud. Could the phenotype reflect a non-specific action of the EtvEnR regulator outside of the Fgf-pathway? Arguing against this possibility, similar constitutive repressor constructs have been used to address the function of this subfamily of ETS transcription factors in the developing mouse lung (Liu et al., 2003) and somite (Brent and Tabin, 2004) as well as in tumorigenesis (Shepherd and Hassell, 2001); in all cases the results are consistent with their acting in a highly specific manner. Nonetheless, given the unexpected nature of our results, we sought to verify that the Etv4/5 subfamily negatively regulates Shh expression using an independent approach knocking-down Etv4/5 activity in the zebrafish.

Formation of the zebrafish pectoral fin is highly conserved at the molecular level to that of the tetrapod limb (Mercader, 2007), this includes Fgf-dependent transcriptional responses in the fin/limb bud (Kawakami et al., 2003; Mercader, 2007). The zebrafish contains three Etv4/5 orthologues; pea3 is orthologous to mEtv4 while erm and etv5 appear to be most closely related to mEtv5 (Roussigne and Blader, 2006). All three genes exhibit early expression in the developing pectoral fin ((Mercader et al., 2006); Figure S4); erm expression here has been shown to be FGF-dependent (Mercader et al., 2006). Expression of all three Etv4/5 relatives in the zebrafish fin bud was inhibited by treatment with SU5402, consistent with their positive regulation through SU5402-sensitive receptor tyrosine kinases that include the FGF receptor family (Figure S4). Knockdown of pea3, erm and etv5 was achieved by a combined morpholino oligonucleotide (MO) injection (Figure S5); Shh expression was then assessed 48 hours post fertilization (for comparison, a detailed wild-type series of Shh expression is presented in Figure S6). A clear expansion of the Shh expression domain was observed in pea3;erm;etv5 MO fin buds when compared to either uninjected or mismatch controlled MO injected embryos (Figure 3C and Figure S6). Of 52 morphant embryos analyzed, 20 exhibited a slight reduction in size of the fin and an expanded domain of Shh encompassing at least one quarter of the distal fin bud (Figure 3C). In 20 additional embryos, comparison was complicated by significant developmental delays associated with the systemic nature of morpholino action. Despite this, the domain of Shh in mutant fins exceeds that observed for stage-matched wild-type embryos (Figure S6). These results suggest that Etv4/5 orthologues restrict Shh expression in the zebrafish fin, highlighting a fin to limb conservation in this regulatory pathway.

Fgf inhibits polarizing activity in cultured anterior mesenchyme

Our findings that Etv4/5 activity restricts Shh from the anterior limb bud, and that Etv4/5 expression is lost following AER removal, beg the question: why ectopic Shh expression is not observed after AER extirpation? A potential insight comes from early work demonstrating that anterior mesenchymal cells removed from the chick limb bud, and cultured in vitro without the influence of the AER, gain polarizing activity when assayed by grafting to a host chick wing bud (Anderson et al., 1994) . Importantly, this polarizing potential was suppressed when cultured cells were treated with Fgf2. We readdressed this work from the perspective of Shh signaling. Following micro-dissociation and high density culturing of anterior E10.0 limb cells as described in the earlier study (Anderson et al., 1994) , we observe a dramatic increase in the expression of Gli1, a sensitive and specific read-out of hedgehog activity (Figure 3D). The induction of Gli1 was completely abrogated by the addition of Fgf8 to cultures (Figure 3D) supporting the conclusion that Fgf signaling blocks hedgehog signaling in the anterior mesenchyme. The lengthy time for hedgehog signaling to initiate in the absence of Fgf activity, which may in part reflect turnover of existing ligand, could lead to this outcome being obscured in vivo by the wave of cell death that occurs following this surgical manipulation.

Polydactyly in Prx1-Cre; R26-EtvEnR mice is Shh dependent

Ectopic induction of Shh has been identified in a number of other polydactylous mouse lines, including Gli3 and Alx4 null mutants, and in both mouse and chick limb buds when Hand2 is broadly overexpressed (Chan et al., 1995; Fernandez-Teran et al., 2000; Masuya et al., 1995; Qu et al., 1997). In these cases, the ectopic domain is generally confined to the anterior margin. However, in at least one of these, the Gli3 loss of function mouse Gli3XtJ/XtJ (Hui and Joyner, 1993), a similar polydactyly is observed in the presence or absence of Shh (Litingtung et al., 2002; te Welscher et al., 2002). To test whether ectopic Shh signaling contributes to the Prx1-Cre; R26-EtvEnR limb patterning phenotype, we removed the Shh gene from mesenchymal cells in the mouse limb by crossing Prx1-Cre; Shhn/+ driver males to R26-EtvEnR; Shhc/+ females that were heterozygous for a conditional Shh (Shhc) allele (Dassule et al., 2000). Prx1-Cre; R26-EtvEnR; Shhn/c and Prx1-Cre; Shhn/c limbs exhibited an identical number of digits (Figure 2B and Lewis et al., 2001). Thus, the polydactlylous phenotype in Prx1-Cre; R26-EtvEnR mutants is indeed dependent on ectopic Shh activity. Interestingly, digits of Prx1-Cre; R26-EtvEnR; Shhn/c embryos appear posteriorized when comparing relative digit length and phalanx morphology to those of Prx1-Cre; Shhn/c embryos (Figure 2B). This likely reflects an expansion of the domain of Shh production (relative to that in Prx1-Cre; Shhn/c limbs) during the short time window prior to complete Prx1-Cre mediated recombination (Lewis et al., 2001).

Etvs restrict Shh expression in early limb mesenchyme

At E9.5-9.75, when robust Shh mRNA is first observed in wild-type embryos, a significant expansion of Shh is already observed into anterior limb mesenchyme in Prx1-Cre; R26-EtvEnR embryos (Figure 3A). This suggests Etvs act either at the time of Shh initiation or very soon after. While this expansion is maintained and strengthened up until at least E11.5 (Figure 3A), this analysis does not address whether blocking Etv activity induces ectopic Shh after limb outgrowth is initiated? Preliminary experiments using a Shh-Cre driver line (Harfe et al., 2004) indicated that introduction of EtvEnR after the onset of Shh expression did not strongly affect limb development (Figure S7). These data argue against a role for Etvs in later FGF-dependent maintenance of Shh expression suggesting they only act early in refining Shh expression to the ZPA (Laufer et al., 1994; Niswander et al., 1994).

To examine the temporal requirement in more detail, we crossed R26-EtvEnR mice carrying a Rosa26 lacZ reporter allele (R26R ; (Soriano, 1999)), to a mouse line carrying a tamoxifen (TM)-dependent Cre transgene (CAGGS-CreER; (Hayashi and McMahon, 2002)). This transgene enables widespread TM-dependent activation of both the EtvEnR and lacZ throughout the embryo 12 hours after intraperitoneal injection of TM into dams (Hayashi and McMahon, 2002). TM was injected at E8.5 (6mg/40g body weight), leading to recombination from E9.0 onwards. Examination of embryos at E10.5 demonstrated high levels of mesenchymal recombination in the limb (Figure 4A) and accompanying this, ectopic distal anterior Shh expression (Figure 4A). In contrast, when TM was administered at E9.5 (recombination initiating around E10), Shh expression was unaltered despite similar high levels of recombination of the reporter allele evident at E11.5 (Figure 4A). Thus, only the early limb exhibits Etv-mediated regulation of Shh expression.

EtvEnR induces ectopic Shh in the absence of either an AER or ZPA

In the experiments described thus far, it is interesting to note that the ectopic anterior Shh domain that resulted from EtvEnR expression was always distally restricted despite Cre expression throughout the entire limb mesechyme (Figures 3A and and4A).4A). To determine whether ectopic Shh expression requires an additional positive input from the overlying AER we performed surgical manipulations on the chick limb in ovo.

To recapitulate the Prx1-Cre; R26-EtvEnR phenotype in chick, we expressed an EtvEnR construct (Brent and Tabin, 2004) using the RCAS system of retroviral gene delivery (Morgan and Fekete, 1996). Injection of RCAS-EtvEnR vrus into the lateral plate mesoderm of stage 10 chick results in shortening of the limb and anterior expansion of Shh as expected (Figures S8A-C). In control uninfected animals, removal of the AER at stage 19/20 resulted in rapid down-regulation of endogenous Shh ((Laufer et al., 1994; Niswander et al., 1994); (Figure 4B; n=5/5). However, in RCAS-EtvEnR infected limbs, a distal anterior expansion of Shh expression was observed even after AER removal (Figure 4B; n=2/3). These results suggest that while blocking Etv activity allows induction of Shh in an AER-independent fashion, additional factor(s) must account for the distal restriction of ectopic Shh despite widespread EtvEnR infection. To our knowledge, this is the first example whereby persistent Shh expression in the limb bud is observed in the absence of either an AER or exogenous FGF protein source.

We also performed ZPA removal experiments. In an uninfected limb, removal of the posterior quarter of the limb at Stage 17/18, just prior to expression of Shh, resulted in a failure to establish the ZPA and halted all further limb development (5/5 embryos, Figure 4C; (Pagan et al., 1996). When we infected the limb with RCAS-EtvEnR at Stage 10, then performed ZPA removal at Stage 17/18 allowing embryos to develop for an additional 24 hours, ectopic Shh was induced in a distal anterior domain (5/8 embryos; Figure 4C). Thus, ectopic Shh activation in the anterior limb mesenchyme is independent of Shh activity from the normal ZPA.


Our data support a role for an FGF-Etv pathway in repressing Shh expression in the anterior mesenchyme thereby contributing to the posterior restriction of a Shh-producing organizing center. Based on the temporal pattern of ectopic Shh expression seen in the limb bud following EtvEnR misexpression, we speculate that endogenous Etv4/5 acts to repress Shh in two distinct domains, one extending from the ZPA across the distal limb bud, and a second, later domain in the anterior of the limb bud. The later timing of ectopic anterior Shh expression is consistent with the fact that in other mutants exhibiting ectopic anterior Shh activity, the anterior domain is observed at a later time in limb development than the normal domain at the posterior margin (Chan et al., 1995; Fernandez-Teran et al., 2000; Masuya et al., 1995; Qu et al., 1997). Both of these functions of Etv4/5, in the distal and anterior limb, must occur before ~E10.5, based on the data presented in Figure 4A. Thus, there is an early time window, prior to Shh activation, when all distal mesenchyme is competent to activate Shh but broad activation is blocked by Fgf-Etv4/5 action. This broad competence is then lost shortly thereafter. Whether the loss of competence reflects a change in the intrinsic properties of mesenchyme, or a redundant mechanism for restricting Shh expression to the posterior margin, is unclear.

The model that Etv4/5 regulates repression of Shh is further supported by recent work from Xin Sun and colleagues (co-submitted, personal communication), In this, a T-Cre driver was used to remove both Etv4 and Etv5 activity from the limb bud resulting in ectopic anterior expression of Shh in the limb bud. Their approach did not reveal a role for Etvs in the zeugopod; it is unclear whether the differences observed are attributable to differential recombination efficiency, persistence of protein, or residual functions of Etv1. Regardless, their results provide an independent validation that ectopic Shh expression induced by EtvEnR reflects an endogenous requirement for Etv4/5 in restricting Shh expression within the developing vertebrate limb.

While we can not exclude the possibility that endogenous Etv4/5 act directly to repress Shh transcription in the distal and anterior mesenchyme, the vast majority of reports indicate these factors act to positively regulate transcription (de Launoit et al., 2006). Thus, Etv4/5 may induce transcription of an intermediary molecule whose normal function is to suppress Shh transcription. However, as Etv4 and Etv5 expression domains encompass the ZPA, assuming their messages are translated within this region, other mechanisms must block their inhibitory actions specifically within the ZPA. Given the recent finding that signals from the non-AER dorsal-ventral border ectoderm are sufficient to induce Shh expression in competent posterior mesenchyme (Nissim et al., 2006), a delicate balance may be created between a positively-acting, posterior, Shh-inducing pathway and an FGF-Etv4/5-dependent, anterior inhibitory input that ultimately refines the precise, posterior Shh expression domain, and thereby enables appropriate A-P axis development in the vertebrate limb.


In situ hybridization, skeletal preparations, β-galactosidase analysis and immunohistochemistry

Whole-mount digoxigenin in situ hybridization, skeleton preparation and β-gal staining were carried out as previously described (Lewis et al., 2001). For immunohistochemistry, embryos were collected and fixed for 2 hours in 4% paraformaldehyde. TUNEL analysis was performed according to manufacturer's instructions (Roche). Primary antibodies for immunodetection included Anti-phospho Histone H3 (Upstate; diluted 1:200, and AMV-3C2 (DSHB; diluted 1:5). Cy5-conjugated secondary antibodies (Jackson ImmunoResearch; diluted 1:500).

Chick limb cultures, manipulation and retroviral infection

Chick embryos were staged according to Hamburger and Hamilton (Hamburger and Hamilton, 1951). Stage 20 limb buds were removed and cultured on nucleoprotein filters in DMEM supplemented with 10% fetal calf serum, 1% pen-strap and 1% glutamine. SU5402 (Caliches) was made up in DMSO and added to media at a final concentration of 30μM. Control cultures included DMSO alone.

Stage 10 lateral plate mesoderm infection of the RCAS construct was performed as previously described (Logan and Tabin, 1998). AER removals were performed essentially as described in Dudley et al., 2002 (Dudley et al., 2002). In brief, following removal of the vitteline membrane, 0.1% nile blue was placed over the limb to allow visualization and the AER removed using a tungsten needle. ZPA removal experiments were preformed as previously described (Pagan et al., 1996).

Morpholino Injection

Antisense morpholino oligonucleotides against pea3 (5’- TTAAAAGTCTAATGTTTACCTCCTC-3’) erm (5’- GCTTCTATAACATACTGACCTCCTC-3’) and etv5 (5’- ATACATTAGGGAGTACCTGTAGCTG-3’) were purchased from Gene Tools. Morpholinos were diluted in distilled water at a concentration of 0.25 mM. The triple morpholino mix was the then injected into one-cell stage embryos at the volume of 1.5 nL (0.3 pmol per morpholino; 0.9 pmol total concentration). Mismatch controls directed against erm (5’- GGTTCTTTAAGATACTCACCTGCTC-3’) and pea3 (5’- TTATAAGTGTAATCTTTAGCTCGTC-3’) were diluted to maintain the total concentration injected for triple morpholino (0.9 pmol).

Micro-dissociation cultures and real-time PCR

Micro-dissociation cultures were performed as detailed in (Anderson et al., 1994) with the following modifications: eighty anterior ¼ forelimb bud fragments from E10.0 mouse embryos were collected and dissociated in 20 μL 0.05% Trypsin; 0.53mM EDTA for 5 minutes at 37°C. Trypsin-EDTA was removed, cells washed once in PBS and resuspended in 160 μL culture media (see below). 10 μL drops of cell suspension were cultured in the presence or absence of recombinant Fgf8 (R&D Systems; final concentration 100ng/ml) for 1.5 hours to allow cells to adhere. At this time, 0.5 ml of culture media was added maintaining similar Fgf8 levels in the Fgf8-treated samples and micro-dissociation cultures incubated for a further 22.5 hours. Culture media consisted of DMEM supplemented with 10% heparin-treated FBS (FBS was pre-incubated with 30 μL bed-volume heparin-coated acrylic beads (Sigma) for 6h at 4°C and beads removed by centrifugation at 3000rpm for 3 minutes). RNA extraction was performed using RNeasy (Qiagen). 250 ng total RNA was used for reverse transcription (Transcriptor; Roche) and quantitative real-time PCR performed (Lightcycler 2000; Roche) using primers for Shh (Mariani et al., 2008), Cycolphilin(Mariani et al., 2008) and Gli1 (GCCATCTCTCCATTGGTACCATG and CGAGTAGAGTCATGTGGTACAC).

Supplementary Material


Supplementary Figure 1. Etv4/5 exhibit dynamic expression in the developing chick limb bud. Both Etv4 and Etv5 are expressed in the limb mesenchyme from the earliest stages of limb development and become progressively restricted to the distal mesenchyme during limb outgrowth. The expression of these transcripts is highly conserved across species, when compared to their orthologous genes in mouse (see Figure 1).

Supplementary Figure 2. Small molecule mediated inhibition of FGF receptor-like kinase activities supports a role for Fgf signaling in the regulation of Etv4 and Etv5 expression. (A) Limb buds were dissected from Stage20/21 chick embryos and cultured in the presence of 30μM SU5402 dissolved in DMSO. Control limb buds were cultured in the presence of DMSO alone; In addition to the distal Etv4/5 expression characteristic of this developmental stage, some diffuse proximal expression was observed, an artifact of culturing conditions. Following 24h treatment, complete abolition of Etv4 and Etv5 expression (both proximal and distal) was observed in SU5402-treated limbs but not in control limbs, consistent with the requirement for an FGF receptor-like tyrosine kinase activity in Etv4/5 regulation. The maintenance of Tbx2 expression following SU5402 treatment suggests the loss of Etv4 and Etv5 is not a consequence of global loss of gene expression.

Supplementary Figure 3 (A) Schematic representation of the EtvEnR Rosa26 targeted allele, R26-EtvEnR. A cDNA fragment encoding the Etv4EnR fusion protein with a IRES-nuclear GFP tag was targeted into the ubiquitously expressed Rosa26 locus (R26) 3’ to a LoxP-flanked polyadenylation stop sequence cassette. (B) Constitutive repression of Etv4/5 does not perturb limb prepatterning. Whole mount in situ hybridization analysis of Gli3 and Hand2 in E9.25 wild-type and Prx1-Cre;R26-EtvEnR mouse embryos reveals no alterations in their expression domains.

Supplementary Figure 4. Shh expression in the zebrafish pectoral fin is dependent on a FGF receptor-like activity. Embryos were cultured from 12 hours post-fertilization in the presence of 30μM SU5402 dissolved in DMSO. Control embryos were cultured in the presence of DMSO alone. A complete abolition of erm, pea3 and etv5 expression was observed in SU5402-treated fin buds 12 hours after treatment commenced while Shh was robustly expressed in control embryos (fin buds marked with arrow), suggesting their transcription depends on Fgf signaling.

Supplementary Figure 5. Morpholino knockdown of erm, pea3 and etv5 in zebrafish. For each gene, the primary transcript is schematized; morpholino target sites are marked above with a red bar and the position of primers used to determine the resulting splicing pattern is indicated below. The DNA binding domain of erm, pea3 and etv5 is located in the 3’ region of each transcript and is spread across 3 exons as indicated in green. The predicted protein following MO injection is presented below: a non-specific amino acid fusion is indicated in red. erm morpholino was directed against the splice acceptor site upstream of exon 11 and results in a 79bp insertion of the entire intron 10. A stop codon is observed 3 amino acids into the predicted fusion protein, terminating the protein prior to critical elements required for DNA binding based on homology with other Ets-DNA binding transcription factors. pea3 morpholino was directed against the splice acceptor site upstream of exon 10 and results in a 173bp deletion, removing exon 9. This deletion puts the resultant protein out of frame creating an in-frame stop codon 15 amino acids into the predicted fusion protein. etv5 morpholino was directed against the splice acceptor site upstream of exon 7 and results in a 94bp insertion of the intron 6 resulting in translational termination at a stop codon 35 amino acids into the predicted fusion transcript.

Supplementary Figure 6. erm;pea3;etv5 morpholino knockdown results in expansion of Shh expression within the fin bud. (A) Wild-type series of Shh expression in the zebrafish embryo, from 29 hours post fertilization (hpf) to 48 hpf. A higher magnification image of the fin is shown below; scale bar represents 0.02mm. (B) Embryos were injected with triple MO mix or mismatch MO mix at the 1-2 cell stage and Shh expression assessed 48 hours post-fertilization. Shh expression in mismatch MO control embryos (n=40) appeared identical to wild-type embryos (n=20). In 20/52 experimental MO injected embryos (pea3;erm;etv5 MO_a), an increase in the intensity and domain of Shh expression was observed in the distal fin bud. For embryos where fin bud development was significantly delayed (pea3;erm;etv5 MO_b; n=20/52), an expanded domain of Shh was still observed relative to age-matched controls, the ectopic anterior Shh expression domain is arrowed. No change in Shh expression was observed in 9/52 embryos, while a decrease/loss of Shh was recorded in 3/52 embryos that also exhibited severe developmental defects.

Supplementary Figure 7. Limb development in Shh-Cre;R26-EtvEnR embryos. (A) Shh expression in the limb buds of Shh-Cre;R26-EtvEnR and R26-EtvEnR embryos at E10.5. (B) Skeletal analysis of forelimbs and hindlimbs development in R26-EtvEnR and Shh-Cre;R26-EtvEnR embryos at E16.5.

Supplementary Figure 8. Expression of EtvEnR in the chick limb bud results in anterior expansion of Shh and altered skeletal pattern. (A) EtvEnR was expressed throughout the chick limb bud using the RCAS system of gene delivery. Embryos were injected at Stage 10 into the lateral plate mesoderm and assayed for expression of Shh (blue) and Fgf8 (brown) at various stages of development. Anterior expansion of Shh was observed in EtvEnR infected limbs when compared to uninfected age-matched control embryos at early (Stage 21) and later (Stage25) of development, reflecting the pattern of ectopic Shh observed following expression of EtvEnR in mouse limb bud (compare with Figure 2B and Figure 3A). A slight reduction in the robustness of ectopic Shh expression in some embryos likely reflects technical limitations of infection. (B) DAPI-stained cross-section through a developing chick embryo, infected with RCAS-EtvEnR at Stage 10 in the right lateral plate mesoderm. The degree of limb truncation can be appreciated when comparing infected limb (white arrow) with the contralateral control. (C) RCAS-EtvEnR infection in chick results in a range of limb abnormalities shown here by alcian blue staining of limb cartilage at Day 5 or 6 of development. Distal limb truncation was observed in a small number of embryos. More commonly, a reduction of the zeugopod with an expanded autopod with or without ectopic digits was observed.

Supplemental experimental procedures

Generation of mouse strains. To generate the R26-EtvEnR allele, a cDNA encoding the Etv4EnR fusion protein (Shepherd et al., 2001) was cloned together a C-terminal IRES-nuclear GFP tag into the BigT vector (Srinivas et al., 2001). The resulting plasmid was digested to release the floxed Etv4EnR-IRES-GFP cassette, which was then inserted into pROSA26PAS (Mao et al., 2005). This targeting vector was linearized and electroporated into 3-1 embryonic stem cells (Mao et al., 2005). After G418 selection, recombinants at the ROSA26 locus were identified by visualization of loss of YFP expression, and the chimeras were then generated by blastocyst injection.

Prx1-Cre;Shhn/c;R26-EtvEnR mutant mice were obtained from the crosses of Prx1-Cre;Shhn/+ males and Shhc/+;R26-EtvEnR females, while Prx1-Cre;Fgfr1n/+;R26-EtvEnR mice were generated from the cross of Prx1-Cre;Fgfr1n/+ males and R26-EtvEnR females. To generate CAGGS-CreER; R26-EtvEnR; R26R embryos, CAGGS-CreER;R26R males were crossed to R26-EtvEnR females. Tamoxifen (Sigma) was administered at a single dose of 6mg/40g body weight by intraperitoneal injection into the pregnant dam at E8.5 and E9.5 and embryos collected at E10.5 and E11.5, respectively.


Mao, J., Barrow, J., McMahon, J., Vaughan, J., and McMahon, A. P. (2005). An ES cell system for rapid, spatial and temporal analysis of gene function in vitro and in vivo. Nucleic Acids Res 33, e155.

Shepherd, T. G., Kockeritz, L., Szrajber, M. R., Muller, W. J., and Hassell, J. A. (2001). The pea3 subfamily ets genes are required for HER2/Neu-mediated mammary oncogenesis. Curr Biol 11, 1739-1748.

Srinivas, S., Watanabe, T., Lin, C. S., William, C. M., Tanabe, Y., Jessell, T. M., and Costantini, F. (2001). Cre reporter strains produced by targeted insertion of EYFP and ECFP into the ROSA26 locus. BMC Dev Biol 1, 4.

50 word summary for DC-D-08-00516

The FGF and Hedgehog pathways interlink to drive outgrowth and patterning of the vertebrate limb. We present evidence that an FGF/Etv pathway confines Sonic hedgehog expression to a posterior zone of polarizing activity. Our studies provide new insights into the signaling and transcriptional networks that govern vertebrate limb development.


We thank Dr. John Hassell for providing the Etv4EnR cDNA, H. Steinbeisser for pea3 and erm plasmids, Dr. Juha Partanen for the Fgfr1 null allele and Dr. Alex Schier for discussions and assistance with the zebrafish experiments. We are grateful to Jill McMahon for blastocyst injection and members of McMahon and Tabin labs for helpful discussion. J.M. is a fellow of Charles H. Hood Foundation. This work was supported by grants from NIH (NS033642 to APM and R37 HD32443 to CJT).


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Further discussions of the mouse methods used in this paper are available in the online supplement.


  • Anderson R, Landry M, Reginelli A, Taylor G, Achkar C, Gudas L, Muneoka K. Conversion of anterior limb bud cells to ZPA signaling cells in vitro and in vivo. Dev Biol. 1994;164:241–257. [PubMed]
  • Boulet AM, Moon AM, Arenkiel BR, Capecchi MR. The roles of Fgf4 and Fgf8 in limb bud initiation and outgrowth. Dev Biol. 2004;273:361–372. [PubMed]
  • Brent AE, Tabin CJ. FGF acts directly on the somitic tendon progenitors through the Ets transcription factors Pea3 and Erm to regulate scleraxis expression. Development. 2004;131:3885–3896. [PubMed]
  • Capdevila J, Tsukui T, Rodriquez Esteban C, Zappavigna V, Izpisua Belmonte JC. Control of vertebrate limb outgrowth by the proximal factor Meis2 and distal antagonism of BMPs by Gremlin. Mol Cell. 1999;4:839–849. [PubMed]
  • Chan DC, Laufer E, Tabin C, Leder P. Polydactylous limbs in Strong's Luxoid mice result from ectopic polarizing activity. Development. 1995;121:1971–1978. [PubMed]
  • Charite J, McFadden DG, Olson EN. The bHLH transcription factor dHAND controls Sonic hedgehog expression and establishment of the zone of polarizing activity during limb development. Development. 2000;127:2461–2470. [PubMed]
  • Chiang C, Litingtung Y, Harris MP, Simandl BK, Li Y, Beachy PA, Fallon JF. Manifestation of the limb prepattern: limb development in the absence of sonic hedgehog function. Dev Biol. 2001;236:421–435. [PubMed]
  • Chiang C, Litingtung Y, Lee E, Young KE, Corden JL, Westphal H, Beachy PA. Cyclopia and defective axial patterning in mice lacking Sonic hedgehog gene function. Nature. 1996;383:407–413. [PubMed]
  • Chotteau-Lelievre A, Dolle P, Peronne V, Coutte L, de Launoit Y, Desbiens X. Expression patterns of the Ets transcription factors from the PEA3 group during early stages of mouse development. Mech Dev. 2001;108:191–195. [PubMed]
  • Dassule HR, Lewis P, Bei M, Maas R, McMahon AP. Sonic hedgehog regulates growth and morphogenesis of the tooth. Development. 2000;127:4775–4785. [PubMed]
  • de Launoit Y, Baert JL, Chotteau-Lelievre A, Monte D, Coutte L, Mauen S, Firlej V, Degerny C, Verreman K. The Ets transcription factors of the PEA3 group: transcriptional regulators in metastasis. Biochim Biophys Acta. 2006;1766:79–87. [PubMed]
  • Dudley AT, Ros MA, Tabin CJ. A re-examination of proximodistal patterning during vertebrate limb development. Nature. 2002;418:539–544. [PubMed]
  • Fallon JF, Lopez A, Ros MA, Savage MP, Olwin BB, Simandl BK. FGF-2: apical ectodermal ridge growth signal for chick limb development. Science. 1994;264:104–107. [PubMed]
  • Fernandez-Teran M, Piedra ME, Kathiriya IS, Srivastava D, Rodriguez-Rey JC, Ros MA. Role of dHAND in the anterior-posterior polarization of the limb bud: implications for the Sonic hedgehog pathway. Development. 2000;127:2133–2142. [PubMed]
  • Firnberg N, Neubuser A. FGF signaling regulates expression of Tbx2, Erm, Pea3, and Pax3 in the early nasal region. Dev Biol. 2002;247:237–250. [PubMed]
  • Hamburger H, Hamilton HL. A series of normal stages in the development of the chick embryo. J Exp Morphol. 1951;88:49–92. [PubMed]
  • Harfe BD, Scherz PJ, Nissim S, Tian H, McMahon AP, Tabin CJ. Evidence for an expansion-based temporal Shh gradient in specifying vertebrate digit identities. Cell. 2004;118:517–528. [PubMed]
  • Hayashi S, McMahon AP. Efficient recombination in diverse tissues by a tamoxifen-inducible form of Cre: a tool for temporally regulated gene activation/inactivation in the mouse. Dev Biol. 2002;244:305–318. [PubMed]
  • Hui CC, Joyner AL. A mouse model of greig cephalopolysyndactyly syndrome: the extra-toesJ mutation contains an intragenic deletion of the Gli3 gene. Nat Genet. 1993;3:241–246. [PubMed]
  • Kawakami Y, Rodriguez-Leon J, Koth CM, Buscher D, Itoh T, Raya A, Ng JK, Esteban CR, Takahashi S, Henrique D, et al. MKP3 mediates the cellular response to FGF8 signalling in the vertebrate limb. Nat Cell Biol. 2003;5:513–519. [PubMed]
  • Laing MA, Coonrod S, Hinton BT, Downie JW, Tozer R, Rudnicki MA, Hassell JA. Male sexual dysfunction in mice bearing targeted mutant alleles of the PEA3 ets gene. Mol Cell Biol. 2000;20:9337–9345. [PMC free article] [PubMed]
  • Laufer E, Nelson CE, Johnson RL, Morgan BA, Tabin C. Sonic hedgehog and Fgf-4 act through a signaling cascade and feedback loop to integrate growth and patterning of the developing limb bud. Cell. 1994;79:993–1003. [PubMed]
  • Lewis PM, Dunn MP, McMahon JA, Logan M, Martin JF, St-Jacques B, McMahon AP. Cholesterol modification of sonic hedgehog is required for long-range signaling activity and effective modulation of signaling by Ptc1. Cell. 2001;105:599–612. [PubMed]
  • Litingtung Y, Dahn RD, Li Y, Fallon JF, Chiang C. Shh and Gli3 are dispensable for limb skeleton formation but regulate digit number and identity. Nature. 2002;418:979–983. [PubMed]
  • Liu Y, Jiang H, Crawford HC, Hogan BL. Role for ETS domain transcription factors Pea3/Erm in mouse lung development. Dev Biol. 2003;261:10–24. [PubMed]
  • Logan M, Martin JF, Nagy A, Lobe C, Olson EN, Tabin CJ. Expression of Cre Recombinase in the developing mouse limb bud driven by a Prxl enhancer. Genesis. 2002;33:77–80. [PubMed]
  • Logan M, Tabin C. Targeted gene misexpression in chick limb buds using avian replication-competent retroviruses. Methods. 1998;14:407–420. [PubMed]
  • Mariani FV, Ahn CP, Martin GR. Genetic evidence that FGFs have an instructive role in limb proximal-distal patterning. Nature. 2008;453:401–405. [PMC free article] [PubMed]
  • Masuya H, Sagai T, Wakana S, Moriwaki K, Shiroishi T. A duplicated zone of polarizing activity in polydactylous mouse mutants. Genes Dev. 1995;9:1645–1653. [PubMed]
  • McGlinn E, Tabin CJ. Mechanistic insight into how Shh patterns the vertebrate limb. Curr Opin Genet Dev. 2006;16:426–432. [PubMed]
  • McMahon AP, Ingham PW, Tabin CJ. Developmental roles and clinical significance of hedgehog signaling. Curr Top Dev Biol. 2003;53:1–114. [PubMed]
  • Mercader N. Early steps of paired fin development in zebrafish compared with tetrapod limb development. Dev Growth Differ. 2007;49:421–437. [PubMed]
  • Mercader N, Fischer S, Neumann CJ. Prdm1 acts downstream of a sequential RA, Wnt and Fgf signaling cascade during zebrafish forelimb induction. Development. 2006;133:2805–2815. [PubMed]
  • Morgan BA, Fekete DM. Manipulating gene expression with replicationcompetent retroviruses. Methods Cell Biol. 1996;51:185–218. [PubMed]
  • Nissim S, Allard P, Bandyopadhyay A, Harfe BD, Tabin CJ. Characterization of a novel ectodermal signaling center regulating Tbx2 and Shh in the vertebrate limb. Dev Biol. 2006 [PMC free article] [PubMed]
  • Niswander L, Jeffrey S, Martin GR, Tickle C. A positive feedback loop coordinates growth and patterning in the vertebrate limb. Nature. 1994;371:609–612. [PubMed]
  • Niswander L, Martin GR. FGF-4 regulates expression of Evx-1 in the developing mouse limb. Development. 1993;119:287–294. [PubMed]
  • Pagan SM, Ros MA, Tabin C, Fallon JF. Surgical removal of limb bud Sonic hedgehog results in posterior skeletal defects. Dev Biol. 1996;180:35–40. [PubMed]
  • Qu S, Niswender KD, Ji Q, van der Meer R, Keeney D, Magnuson MA, Wisdom R. Polydactyly and ectopic ZPA formation in Alx-4 mutant mice. Development. 1997;124:3999–4008. [PubMed]
  • Raible F, Brand M. Tight transcriptional control of the ETS domain factors Erm and Pea3 by Fgf signaling during early zebrafish development. Mech Dev. 2001;107:105–117. [PubMed]
  • Riddle RD, Johnson RL, Laufer E, Tabin C. Sonic hedgehog mediates the polarizing activity of the ZPA. Cell. 1993;75:1401–1416. [PubMed]
  • Roehl H, Nusslein-Volhard C. Zebrafish pea3 and erm are general targets of FGF8 signaling. Curr Biol. 2001;11:503–507. [PubMed]
  • Roussigne M, Blader P. Divergence in regulation of the PEA3 family of ETS transcription factors. Gene Expr Patterns. 2006;6:777–782. [PubMed]
  • Shepherd T, Hassell JA. Role of Ets transcription factors in mammary gland development and oncogenesis. J Mammary Gland Biol Neoplasia. 2001;6:129–140. [PubMed]
  • Shepherd TG, Kockeritz L, Szrajber MR, Muller WJ, Hassell JA. The pea3 subfamily ets genes are required for HER2/Neu-mediated mammary oncogenesis. Curr Biol. 2001;11:1739–1748. [PubMed]
  • Soriano P. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat Genet. 1999;21:70–71. [PubMed]
  • Sun X, Mariani FV, Martin GR. Functions of FGF signalling from the apical ectodermal ridge in limb development. Nature. 2002;418:501–508. [PubMed]
  • te Welscher P, Zuniga A, Kuijper S, Drenth T, Goedemans HJ, Meijlink F, Zeller R. Progression of vertebrate limb development through SHH-mediated counteraction of GLI3. Science. 2002;298:827–830. [PubMed]
  • Vogel A, Tickle C. FGF-4 maintains polarizing activity of posterior limb bud cells in vivo and in vitro. Development. 1993;119:199–206. [PubMed]
  • Zuniga A, Haramis AP, McMahon AP, Zeller R. Signal relay by BMP antagonism controls the SHH/FGF4 feedback loop in vertebrate limb buds. Nature. 1999;401:598–602. [PubMed]