Function of FGF signaling in the developmental process of the median fin fold in zebrafish

Developmental Biology 304 (2007) 355 – 366 www.elsevier.com/locate/ydbio

Function of FGF signaling in the developmental process of the median fin fold in zebrafish Gembu Abe, Hiroyuki Ide, Koji Tamura ⁎ Department of Developmental Biology and Neurosciences, Graduate School of Life Sciences, Tohoku University, Aobayama, Aoba-ku, Sendai 980-8578, Japan Received for publication 11 September 2006; revised 18 December 2006; accepted 19 December 2006 Available online 23 December 2006

Abstract Median fins, unpaired appendages in fish, are fundamental locomotory organs that are believed to have evolved before paired lateral appendages in vertebrates. However, the early process of median fin development remains largely unknown. We investigated the early development of the median fin fold, a rudiment of median fins, and report here the process in zebrafish embryos and the function of FGF signaling in the process. Using expressions of three genes, dlx5a, sp9 and fgf24, as markers of different phases of fold development, our findings suggest that the early process of median fin fold development can be divided into two steps, specification of the median fin fold territory and construction of the fold structure. Both loss-of-function and gain-of-function assays revealed that FGF signaling plays roles in each step, suggesting a common mechanism for the development of median appendages and paired lateral appendages. © 2007 Elsevier Inc. All rights reserved.

Introduction Most fish have two kinds of fins: paired fins located bilaterally at the ventral–lateral body wall and median (unpaired) fins on the trunk midline along the anteroposterior body axis. In bony fish, both paired and median fins consist of the same elements: endoskeletons with skeletal muscles for the proximal part and dermal finrays (lepidotrichia), a non-muscularized fin lobe, for the distal part (Hall, 1999; van Eeden et al., 1996). It has also been shown that they share a similar embryonic structure as an epithelial rudiment of fins, apical fold/median fin fold (MFF), which becomes a fin lobe (Dane and Tucker, 1985; Grandel and Schulte-Merker, 1998). During vertebrate evolution, median fins are believed to have evolved before paired fins (Coates, 1994; Freitas et al., 2006; Mabee et al., 2002; Zhang and Hou, 2004). In extant vertebrates, median fins can be seen in agnathans (lampreys and hagfish) as well as gnathostomes, while paired fins/limbs are regarded as the synapomorphy that defines the gnathostomes (Coates, 1994; Donoghue et al., 2000). It has also been shown that there is a group of ancestral craniates in fossil records that have no paired appendages equipped with a ⁎ Corresponding author. Fax: +81 22 795 3489. E-mail address: [email protected] (K. Tamura). 0012-1606/$ - see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.ydbio.2006.12.040

continuous median fin (Coates, 1994; Zhang and Hou, 2004). Thus, the ancestral mechanism of median fin development may have been co-opted for the development of paired appendages (Freitas et al., 2006; Mabee et al., 2002). Despite the evolutionary significance, many issues about the mechanisms underlying median fin development, especially the mechanism by which development of the MFF in early embryonic stages is initiated, remain unknown. In fact, there are several reports on mutants with an abnormal MFF in zebrafish and medaka (Amsterdam et al., 1999; Fritz et al., 1996; Golling et al., 2002; Ishikawa, 2000; Loosli et al., 2000; van Eeden et al., 1996) that provide interesting insights into molecules involved in fin development but without sufficient discussion of these molecules because of a lack of information on the early process of MFF development. On the other hand, because of the major homology between paired fins and tetrapod limbs, paired fin development has been extensively investigated (Ahn et al., 2002; Grandel and SchulteMerker, 1998; Kawakami et al., 2004b; Neumann et al., 1999; Ng et al., 2002; Norton et al., 2005), and remarkable similarities of developmental and genetic mechanisms between their rudiments, fin buds and limb buds have been found. The formation and function of the apical structure in these buds are good examples of such similarities: the formation and function

356

G. Abe et al. / Developmental Biology 304 (2007) 355–366

of the apical structure in these buds. The apical ectodermal ridge (AER), an essential structure for the outgrowth and proximodistal patterning in limb bud development, is an ectodermal thickening that is located at the distal margin and runs along the anteroposterior axis of the limb bud (reviewed by Capdevila and Izpisua Belmonte, 2001; Martin, 1998; Niswander, 2003; Tickle, 2002). A morphologically/functionally equivalent structure is seen in the fin bud (Grandel et al., 2000; Grandel and SchulteMerker, 1998). The AER-like structure in fishes, which is a pleat-like epithelial sheet at the distal apex of the fin bud, is called the apical fold because the structure itself keeps elongating in the distal direction, resulting in an extended fold at later developmental stages. The apical fold and the AER share a common developmental mechanism. It has been shown that the AER in the limb bud is induced and maintained by mesenchymederived FGF10 (Min et al., 1998; Ohuchi et al., 1997; Sekine et al., 1999; Yonei-Tamura et al., 1999). Then FGF8, another member of the FGF family, which is expressed throughout the entire AER, maintains FGF10 expression in the limb bud mesenchyme, thereby completing a positive feedback loop as an epithelial–mesenchymal interaction (reviewed by Xu et al., 1999). A comparable event is also known in fin bud development of the zebrafish embryo. Targeted knockdown of the fgf10 with morpholinos or a mutant of fgf10 locus (dae mutant) shows deficiency of pectoral fin buds resulting from a disruption of the above-described crosstalk (Ng et al., 2002; Norton et al., 2005). In the dae mutant, for example, initiation of fin bud can be seen, but induction of apical fold formation does not occur (Norton et al., 2005). This abnormality prevents later fin bud development, resulting in the formation of only a shoulder girdle, and, moreover, the mutant can be rescued by application of FGF4, whose transcripts are expressed in the apical fold (Grandel et al., 2000; Norton et al., 2005). Sp9 and Sp8 encode buttonhead (btd)-like zinc (Zn) finger transcription factors that are expressed in the AER and the apical fold including pre-AER cells (Kawakami et al., 2004b). Knockdown analysis with morpholinos in zebrafish has revealed that these transcription factors are involved in mediating the actions of FGF signaling and are essential for function of the apical fold and fin bud development (Kawakami et al., 2004b). Since Sp9 and Sp8 have similar functions in tetrapod limb development, these results suggest that the apical fold is homologous to the AER and that these two structures share conserved functions and molecular pathways. Adult median fins in major living fishes are positioned within the initially continuous MFF (Mabee et al., 2002; Suzuki et al., 2003; van Eeden et al., 1996). In zebrafish, a continuous epidermal fin fold fully develops on the dorsal and ventral midline during embryo and larval stages, and this structure expresses a set of molecules (sp genes, dlxs, msxs, fgfs, etc.), similar to the set of molecules expressed in the apical fold in paired fins (Akimenko et al., 1994, 1995; Draper et al., 2003; Fischer et al., 2003; Kawakami et al., 2004b; Nomura et al., 2006; Reifers et al., 1998). In the chick embryo, moreover, additional application of FGF7/10 to the dorsal medial region induces the formation of an AER-like structure in the dorsal midline, reminiscent of the MFF (Tamura et al., 2001; Yonei-Tamura

et al., 1999). Taken together, these findings suggest that the dorsal and lateral ectoderms in tetrapods and fishes possess a common competence and mechanism for establishing the special structures, the AER and the apical fold/MFF, for each appendage development. We report here the developmental process of the MFF, focusing on early embryonic stages. Our findings suggest that early MFF development includes two steps, specification of a presumptive territory where the fin fold arises and construction of the actual fin fold structure, and FGF is likely to play roles in both steps. Materials and methods Fish maintenance Zebrafish, Danio rerio, were maintained at 27 °C on a 14 h light/10 h dark cycle. Embryos were raised at 28.5 °C until the appropriate stages. Embryos obtained from natural crosses were staged according to Kimmel et al. (1995).

Histological analysis and whole-mount in situ hybridization For histological analysis, embryos were fixed in Bouin's fixative for 1–3 h at room temperature, dehydrated, embedded in paraffin wax, and cut at 4–6 μm. Slides were then stained with hematoxylin and eosin. Whole-mount in situ hybridization reactions were essentially performed as described previously (Schulte-Merker et al., 1992; Tamura et al., 1999) except that embryos were treated with proteinase K (0.5–2 μg/ml) and hybridized at 60 °C. Antisense RNA probes for dlx5a (a kind gift from Dr. Atsushi Kawakami), sp9 (a kind gift from Dr. Yasuhiko Kawakami), and fgf24 (a kind gift from Dr. Kyo Yamasu) were described previously (Draper et al., 2003; Kawakami et al., 2004a,b). Several stained embryos were embedded in OCT compound (Miles) and sectioned to a thickness of 8 μm.

SU5402 treatment Inhibition of signaling through FGF receptors was performed with the lipophilic reagent SU5402 (CalBiochem) (Mohammadi et al., 1997). Embryos were incubated in the dark at 28.5 °C in 20 μM SU5402 containing aquarium water, prepared from 4 mM SU5402 stock solution in DMSO (Jackman et al., 2004; Roehl and Nusslein-Volhard, 2001). Control embryos were incubated in the dark in aquarium water with the corresponding amount of DMSO.

Implantation of beads Implantation of beads was performed as described previously (Hirate and Okamoto, 2006). Heparin acrylic beads (Sigma) soaked in 0.5 mg/ml of each FGF as described below or PBS control bead were implanted into the midline of the 2, 3-somite level of 18 hpf (hours post fertilization) embryos, and the embryos were incubated at 28.5 °C for the indicated periods. FGF4 (recombinant human, PeproTech EC), FGF8b (recombinant human, R&D System), FGF7 (recombinant human, PeproTech EC), and FGF10 (recombinant human, PeproTech EC) were used for this experiment.

Results Morphological observation of the early process of median fin fold development The MFF is an epithelial structure only during a short period of time as mesenchymal cells invade the MFF very rapidly starting during embryogenesis. This structure originates at the anal/cloaca area on the ventral midline (arrow in Fig. 1A), and it

G. Abe et al. / Developmental Biology 304 (2007) 355–366

357

Fig. 1. (A, B) Overview of the median fin fold in a 42 hpf zebrafish embryo from the lateral view (A) and higher magnification of anterior trunk region (B). The continuous midline fin fold (mff, median fin fold (MFF)) surrounds the tail and trunk region from the cloaca (arrow in panel A) to the rostral edge at the 8th somite level (arrow in panel B). paff, pre-anal fin fold; y, yolk; ye, yolk extension. Numbers indicate the somite number. (C–H, a–l) Histological analysis of median fin fold. (C–H) 15–24 hpf zebrafish embryos from the lateral view. (a–l) HE staining of transverse sections at the level of each bar in panels C–H. At 15 hpf (C, a) and 16 hpf (D, b), the presumptive fin fold epidermis (indicated by the black broken lines in a and b that are under the epidermal cell layer) was still connected to the neural ectoderm in the tail bud. (E, c) Epidermis covered the midline of the tail and trunk at 17 hpf (indicated by the black broken lines in c). ep, epidermis; nc, notochord; nt, neural tube; pe, peridermis; pnt, prospective neural tube; so, somite. (F) At 18 hpf, wedge-shaped epidermal cells were distinguishable at the midline of the tail bud (e, f; indicated by brackets), but these cells were not observed at a more rostral region (d). (G, H) Then the area in which wedge-shaped cells were seen (h, i for 21 hpf, and k, l for 25 hpf; indicated by brackets) expanded rostrally and caudally as the tail elongated. The dots indicate the final position of the anterior boundary of the fin fold (8th somite level). (j′, f ′) High magnifications of j and f. Typical cuboidal-shaped epidermal cells and wedge-shaped epidermal cells indicated by the blue (in j′) and red (in f ′) broken lines, respectively. Scale bars in panels A–C, F, and (a) are 500 μm, 100 μm, 500 μm, 200 μm, and 50 μm, respectively. The panel width of (d–l) is 100 μm.

358

G. Abe et al. / Developmental Biology 304 (2007) 355–366

encircles the tail and extends up to the 8th somite level on the dorsal midline (Fig. 1B). Some previous studies (Dane and Tucker, 1985; Kimmel et al., 1995) showed that morphogenesis of the MFF starts at around 18 hpf. The early MFF consists of two epidermal layers, the epidermis and overlying peridermis (Dane and Tucker, 1985). To further elucidate the early step of MFF morphogenesis, focusing on MFF formation along the anteroposterior axis of the main body, we observed the detailed process of morphological change of the epidermis histologically. To identify MFF cells in sections, cell shape was observed because epidermal cells at the dorsal midline are known to change in shape from cuboidal (non-fin fold cells; Fig. 1F-d′) to wedge (fin fold cells; Fig. 1F-f ′) in the early phase of MFF formation (Dane and Tucker, 1985). Transformation from a neural keel to a neural tube is still proceeding from anterior to posterior at 15 and 16 hpf, and the neural keel is not covered with the dorsal epidermis at the tail region at these stages (Figs. 1C-a and D-b) (Kimmel et al., 1995). At these stages, we did not observe wedge-shaped MFF cells anywhere in the dorsal midline region (Figs. 1C-a and D-b; not shown). Neural tube formation appeared to be completed by 17 hpf, and the entire trunk and tail midline was covered with epidermis at this stage (Fig. 1E-c). However, midline epidermal cells along the anteroposterior axis were still cuboidal in shape (Fig. 1E-c). Distinguishable wedge-shaped epidermal cells were first detected in a small area of the tail bud at 18 hpf (Figs. 1F-e, F-f, and F-f′), and a slightly more anterior region included no wedge-shaped cells (Figs. 1F-d and F-d′). Since embryos at this stage have about 18 somites and the future anterior–dorsal border of the MFF is at the 7/8th somite level (Fig. 1B), formation of the MFF structure is initiated not from the anterior boundary but in a more caudal region. As development proceeds, the region in which wedge-shaped cells were observed was extended anteriorly from the tail bud of the 18 hpf embryo (Figs. 1G and H). At 21 hpf, the area extended approximately from the 12th somite level to the tail (Figs. 1G-h and G-i), and wedge-shaped cells were still not seen in a more anterior region (Fig. 1G-g). The gap between the anterior border of the wedge-shaped cell region and the prospective edge of the MFF structure was smaller at 21 hpf. Then at 25 hpf, the MFF structure was observed in the entire region (Figs. 1H-k and H-l). These observations indicate that the median fin fold starts to develop from the tail bud region and extends toward the head as the tail elongates (summarized in Fig. 3A). Different gene expression patterns in the MFF We next examined the expression of several genes because the dynamic temporal changes of their expression in the MFF have not previously been fully elucidated. Among the genes that we examined, we describe here the expression patterns of three genes, dlx5a (Akimenko et al., 1994), sp9 (Kawakami et al., 2004b) and fgf24 (Draper et al., 2003), that had interesting and distinct expression patterns in the MFF from 16 hpf to 24 hpf. Dlx genes, which encode Distal-less-related transcription factor, are known to be expressed and function in the AER/apical fold during vertebrate limb/fin bud development (Panganiban

and Rubenstein, 2002; Quint et al., 2000). Some members of the dlx gene family are also expressed in the MFF of the zebrafish embryo (Akimenko et al., 1994; Ellies et al., 1997). Among them, we chose dlx5a as an accurate marker for the early development of MFF. At 16 hpf, dlx5a was expressed in midline epidermal cells as a long single stripe (Fig. 2B), and this expression domain was bifurcated at the tail bud region, juxtaposing along the neural keel (Fig. 2A) (Akimenko et al., 1994). At this stage, the anterior border of the dlx5a expression domain on the back was located at approximately the 8th somite level (Fig. 2B, arrowhead and dot), corresponding to the prospective anterior border of the MFF (Fig. 1B). This midline expression of dlx5a was maintained throughout early stages of MFF development (from 16 hpf to 24 hpf), and the anterior border of the dlx5a expression domain was always at around the 8th somite (Figs. 2A–E, summarized in Fig. 3B), suggesting that dlx5a continues to be expressed in the presumptive MFF territory throughout early MFF morphogenesis. It was not possible to precisely determine the anterior border of dlx5a expression because the border varies within a half of one somite anteriorly or posteriorly. This variety seems to correspond to the variation of the anterior limit of the MFF structure (Fig. 1B). dlx5a expression was also detected in the ventral midline of the yolk extension, where the pre-anal fin fold is formed. sp9, which encodes buttonhead (btd)-like zinc (Zn) finger transcription factor, is expressed in the AER/apical fold. It functions as a positive regulator of the AER/apical fold formation in limb/fin bud development (Kawakami et al., 2004b). We detected sp9 expression in the epidermis of the tail bud at 16 hpf (Figs. 2F and G). The sp9 expression domain in the non-neural ectoderm of the tail bud was wider than the dlx5a expression domain (Fig. 2F, compare with A), but the domain was restricted to the caudal end (arrowhead in Fig. 2G). At 18 hpf, sp9 expression was detected in the midline epidermis of the tail bud (arrowhead in Fig. 2H) as well as in the neural tube (bracketed in Fig. 2H). This sp9-expressing domain in the dorsal epidermis approximately corresponds to the region that contains wedge-shaped cells. The sp9-expressing domain gradually extended toward the head (Figs. 2G–J), and the anterior extension of the sp9-expressing domain ended up at the 8th somite level by 24 hpf (Fig. 2J). The expression change in sp9 appears to be consistent (or maybe preceding) with the anterior extension of MFF structure formation (compare Fig. 3C with A), although it is not clear whether the anterior borders of the two areas always agree. Fgf24 is a member of the Fgf8/17/18 subfamily in Fgf ligands, and its transcripts are detected in the apical fold together with Fgf8, which has a redundant function with fgf24 in apical fold formation (Draper et al., 2003; Fischer et al., 2003). fgf24 has been reported to be expressed in the caudal fin ectoderm (Draper et al., 2003). fgf24 expression in the epidermis was not detected in the tail bud region at 16 hpf (Figs. 2K and L). After the dorsal midline of the tail bud had been closed and covered with non-neural ectoderm, fgf24 started to be expressed in the epidermal cells of the tail bud at 18 hpf (Fig. 2M). fgf24 expression remained restricted to the tail tip of the MFF at least until 24 hpf (Figs. 2N and O), and fgf24

G. Abe et al. / Developmental Biology 304 (2007) 355–366

359

Fig. 2. Expression patterns of some marker genes during development of the median fin fold. Expression patterns of dlx5a (A–E), sp9 (F–J), and fgf24 (K–O): lateral view (B–E, G–J, and L–O) with head toward the left, and tail view (A, F, and K) with dorsal midline toward the top. Insets in panels M–O show higher magnifications focused on the dorsal half of the tail bud corresponding with the region in panels M–O, respectively; lateral view with rostral to the left. Arrowheads indicate the rostral border of the expression domain. Dots indicate the 8th somite level. (A) dlx5a was expressed around the tail bud and in the epidermal cells juxtaposed with the neural keel at 16 hpf. (B) At this stage, dlx5a expression was seen on the dorsal midline at the 8th somite level and caudally. (C–E) The rostral border of the dlx5a expression domain remained around the 8th somite level at 18 hpf (C), 21 hpf (D), and 24 hpf (E). (F, G) At 16 hpf, sp9 expression was observed in neural tissue (indicated by the bracket) and non-neural epidermis around the tail bud. At this stage, sp9 expression was restricted to the tail bud, and the more anterior dorsal midline was sp9negative. (H–J) The rostral border of the sp9 expression domain expanded toward the head as development proceeded and the tail bud elongated. Compare the position of the arrowhead with the position of the 8th somite indicated by a dot. (K, L) fgf24 expression was seen in the mesenchyme of the tail bud and intermediate mesoderm at 16 hpf. (M–O and insets) fgf24 expression in the epidermal cells was always restricted to the midline of the tail bud, and there was no expansion of the expression domain of fgf24. Scale bars are 250 μm in panels B, C, and E and 125 μm in panel D.

expression was never detected in other regions of the MFF, indicating that fgf24 expression exclusively marks a caudal part of the MFF. Expression profiles of these genes during MFF development are summarized in Figs. 3B–D. Inhibition of Fgfr function affects MFF formation To further investigate the early phase of MFF development, we next focused on FGF signaling because FGF plays essential roles in paired fin development, including apical fold formation (Fischer et al., 2003; Grandel et al., 2000; Ng et al., 2002; Nomura et al., 2006; Norton et al., 2005). We examined FGF function by a pharmacological assay with SU5402, a widely

used inhibitor of Fgf receptor (Fgfr) activation (Mohammadi et al., 1997). SU5402 is used in a wide range of developmental systems in zebrafish to specifically block Fgfr signaling (Jackman et al., 2004; Maroon et al., 2002; Shinya et al., 2001). SU5402 is useful for assessing requirements for FGF signaling in the later stage of development of the zebrafish embryo because it can be applied in late developmental events such as organogenesis, leaving early FGF-dependent processes unaffected. Moreover, SU5402 treatment potentially uncovers FGF requirements that might not be revealed by knocking down specific FGF ligands or receptors owing to redundancy (Jackman et al., 2004). We exposed developing embryos to SU5402 at several time points, and we found that the MFF structure formation was

360

G. Abe et al. / Developmental Biology 304 (2007) 355–366

Fig. 3. Schematic representation of the initial stage of MFF development. (A) Morphological architecture of the median fin fold that is composed of wedge-shape cells starts to develop from the tail bud at 18 hpf, and this structure expands toward the head as the tail elongates. (B) Expression domain of dlx5a represents a prospective “fin fold-forming region” at the 8th somite level and caudal midline before the fin fold structure is constructed. (C) sp9-expressing region that expands rostrally corresponds to the architecture of the median fin fold. Compare panel C with panel A. (D) fgf24 expression is always restricted around the tail bud.

affected in embryos exposed to 20 μM SU5402 for 3 h starting at 18 hpf. After SU5402 treatment, wedge-shaped epidermal cells were not detected on the dorsal midline at the level of middle yolk extension (Fig. 4B-c), while wedge-shaped cells were observed around the tail bud region (Fig. 4B-d). The MFF structure formation normally reached the trunk level in control embryos (exposed to 0.17% DMSO) (Figs. 4A-a and A-b). These results indicate that the signaling pathway mediated by Fgfr is essential for anterior extension of the MFF structure formation. We also examined the effects of SU5402 treatment on the expression of the genes described above. dlx5a expression was detected in SU5402-treated embryos (Fig. 4D) as in control embryos (Fig. 4C). sp9 expression in the MFF became undetectable at the dorsal midline of SU5402-treated embryos, and the expression was restricted posteriorly to the caudal end of yolk extension (Fig. 4F). The intensity of signal for sp9 in SU5402-treated embryos was less than that in the control embryos (compare Fig. 4F with E). fgf24 expression was suppressed in the MFF around the tail bud in SU5402-

treated embryos (Figs. 4G and H). These observations showing that expression of sp9 and fgf24, whose normal patterns of expression are related to the process of MFF structure formation, is dependent on FGF signaling suggest that sp9 and fgf24 play roles in MFF structure formation. While we found that FGF signaling is required for anterior progress of MFF structure formation, SU5402 employed at that time period (for 3 h starting at 18 hpf) was not sufficient for complete suppression of MFF structure formation. Thus, we further examined the effect of SU5402 at an earlier time point (for 3 h starting at 15 hpf). Embryos treated with SU5402 for 3 h from 15 hpf showed no MFF structure even in the tail bud region, and the posterior midline of those embryos was not covered with epidermis (Fig. 5B-b). Both sp9 expression and fgf24 expression in the MFF disappeared with this SU5402 treatment (Figs. 5G–J). In those embryos, however, dlx5a expression was observed in epidermal cells (Figs. 5D and F), the posterior part of which was juxtaposed with the neural keel and showed bilateral stripes (Fig. 5F).

G. Abe et al. / Developmental Biology 304 (2007) 355–366

361

mostly had the anterior border of the MFF at the normal position (Figs. 6A and F). FGF7 and FGF10 induced expansion of the MFF-like structure (Fig. 6F), and in most cases this extended structure was continuously elongated from the normal MFF structure, reaching over the 1st somite level in some cases (Figs.

Fig. 4. SU5402 inhibits extension of the MFF structure formation. Embryos were exposed to 0.17% DMSO for control (A, C, E, G) or 20 μM SU5402 (B, D, F, H) from 18 hpf for 3 h. (A, B) Lateral view of embryos, anterior to the left. (a, b, c, d) HE staining of transverse sections at the level of each bar in panels A and B. While the dorsal midline (at the level of yolk extension) of control embryos consisted of wedge-shaped cells (a; bracket), SU5402-treated embryos did not have such cells at the same level (c). At the tail bud, the fin fold was seen in SU5402-treated embryos (d; bracket) but was immature compared with that in control embryos (b; bracket). (C–H) Expression patterns of dlx5a (C, D), sp9 (E, F), and fgf24 (G, H). (D) dlx5a expression was unaffected in SU5402-treated embryos. (F) Expression of sp9 was suppressed in the median fin fold after SU5402 treatment, and the expression domain was limited caudally to the cloaca level (arrowhead). Expression of fgf24 was not detectable in the tail tip of SU5402-treated embryos (H). Arrowheads in panels C–F indicate rostral border of the expression domain. Scale bars in panels A and C are 200 μm. The panel width of (a–d) is 100 μm.

FGF induces the formation of an additional MFF Lastly, we performed a gain-of-function assay for FGF by implantation of FGF-soaked beads. Beads soaked with FGFs (FGF4, 7, 8, or 10) were implanted at 18 hpf into the dorsal midline at the 2nd or 3rd somite level, where the MFF is not formed in normal development (see Fig. 1B). The effects of FGFs were evaluated by observation of the embryo morphology at the implanted site after 24 h. Embryos with control beads

Fig. 5. SU5402 did not affect dlx5a expression in the presumptive fin fold epidermis. Embryos were exposed to 0.17% DMSO for control (A, C, E, G, I) or 20 μM SU5402 (B, D, F, H, J) from 15 hpf for 3 h. Lateral view of embryos, anterior to the left. (a, b) HE staining of transverse sections at the level of each bar in panels A and B. Bracket in (a) indicates the area in which wedge-shaped cells were seen. The dorsal midline in the SU5402-treated embryo (b) was not covered with epidermis (arrowheads), and wedge-shaped cells for the fin fold were not formed. (C–F) In SU5402-treated embryos, dlx5a expression was retained as bilateral stripes in epidermal cells juxtaposed with the neural keel (D, F). (G–J) Expression patterns of sp9 (G, I) and fgf24 (H, J). Neither sp9 (H) nor fgf24 (J) was detected in the dorsal midline of SU5402-treated embryos. Bracket in (G, I) indicates sp9 expression observed in neural tissue. Scale bars in panels A and C are 200 μm. The panel width of (a, b) is 100 μm.

362

G. Abe et al. / Developmental Biology 304 (2007) 355–366

FGF application is an additional MFF. The onset of the ectopic expression was different, and dlx5a expression was detected at 5 h (Fig. 7G) after FGF7 application, at which time sp9 had not yet been induced (Fig. 7I). It took about 8 h for sp9 expression to be induced in the additional structure (Fig. 7J), suggesting that time schedule of induction of gene expression in the additional MFF is consistent with that during normal MFF development. Interestingly, in FGF7-applied embryos, fgf24 expression, the domain of which is normally restricted to the tail region, was also seen ectopically in a part of the additional MFF on the bead (Fig. 7F). Taken together with the fact that the MFF

Fig. 6. Ectopic median fin fold-like structure formation was induced by FGF application to the dorsal midline. (A–E) Lateral view of the anterior trunk region in 42 hpf or 24 h post implantation embryos. FGF-soaked beads were implanted into the dorsal midline at around the second and third somite level of 18 hpf embryos. Arrows indicate the anterior edge of the median fin fold and ectopic median fin fold-like structure. (A) Control embryos usually had the anterior end of the median fin fold at the 8th somite level. (B–E) Ectopic median fin fold-like structure could be seen rostrally to the intrinsic median fin fold after implantation of FGF beads (B; FGF4, C; FGF7, D; FGF8, E; FGF10). Arrowheads in panel B indicate ectopic hypertrophy that was disconnected from the intrinsic fin fold. (F) The ratio of the position of the anterior end of the median fin fold-like structures in the bead-implanted embryos. The colors of bars in the graph indicate the position of the anterior end in the median fin fold and the median fin fold-like structure: yellow is somite 7–8 level, orange is somite 5–6 level, and maroon is somite 4< region. The colors of bars correspond to the colors of arrows in panels A–E.

6C and E). FGF4 and FGF8 had much less effect (Fig. 6F), and the additional structure induced by FGF4 sometimes displayed an ectopic hypertrophy around the implanted bead, which was disconnected from the normal MFF (Fig. 6B). We also analyzed the expression pattern of dlx5a/sp9/fgf24 in FGF-treated embryos and found that all of these genes were induced in the ectopic structures (Fig. 7). dlx5a and sp9 transcripts were detected in the additional structure induced by FGF7/10, continuously from their normal domain located at and posterior to the 8th somite (Figs. 7B and D, compare with A and C), supporting the idea that the structure ectopically induced by

Fig. 7. Expression of marker genes for the median fin fold in ectopic structure formation induced by FGF7/10. Expression of genes in control bead-implanted embryos (A, C, E) and FGF7/10 bead-implanted embryos (B, D, F) at 12 h after implantation of beads. (A, B) dlx5a, (C, D) sp9, dlx5a (B), and sp9 (D) are expressed in the ectopic structure, continuously from the normal domain. (E, F) fgf24, which is normally expressed only in the tail bud, is (F) also seen ectopically but restricted to the epidermis around the bead. (G–J) Onset of the expression of dlx5a and sp9; dlx5a expression was detected from 5 h after implantation of beads (G, H), while sp9 expression was not detected at 5 h after (I) but was detected at 8 h after implantation of beads (J). Red arrowheads indicate an additional domain of gene expression. Numbers indicate the somite number. Scale bars in panels A and G are 100 μm.

G. Abe et al. / Developmental Biology 304 (2007) 355–366

structure is normally formed first at the tail bud (Fig. 3A), fgf24 may contribute to MFF structure formation. Our findings from experiments on FGF signaling suggest that the FGF signaling pathway, particularly signaling mediated by FGF7/10, is sufficient and necessary for MFF formation. Discussion Early process of MFF structure formation Investigations of MFF development (this study) (Dane and Tucker, 1985; Kimmel et al., 1995) showed that wedge-shaped epidermal cells characteristic of the MFF appear at the dorsal midline at around 18 hpf. We also found that these morphologically distinct MFF cells emerge from the tail bud region and that the area containing these MFF cells extends anteriorly toward the head as the tail elongates. Anterior extension in the dorsal midline reached the final border of the MFF at the 8th somite level at 25 hpf, and the entire MFF structure continuously elongated distally and radially. In this process, the caudal-torostral direction of MFF structure formation is a distinctive feature because many events of embryogenesis such as somitogenesis usually progress in the rostral-to-caudal direction in zebrafish embryos as well as other vertebrate embryos. In agreement with this morphological change, we found that sp9 expression in the dorsal midline, which starts from the tail bud region, extended anteriorly toward the 8th somite level (Figs. 2 and 3). Thus, it is thought that there is a distinct developmental mechanism underlying the extension of MFF formation from caudal to rostral. Although the molecular mechanism for this extension remains unknown, all genes that we analyzed in the present study, dlx5a, sp9 and fgf24, presumably contribute to this phenomenon. In addition, the tail bud may have an important role in this process because the fin fold structure is first seen in the midline epidermis around the tail bud, and FGF signaling must be involved as described below. Some mutants in zebrafish have a complete absence of ventral MFF (Connors et al., 1999; Dick et al., 2000; Kishimoto et al., 1997; Kramer et al., 2002; Mintzer et al., 2001; Mullins et al., 1996; van Eeden et al., 1996), and these mutants have deficient function of the posterior mesoderm in the tail bud (Connors et al., 1999; Pyati et al., 2005, 2006). It is assumed that dorsal MFF formation may also have a relationship with function of the tail bud, but, unfortunately, we have not found any reported mutants in zebrafish that display a complete lack of the dorsal MFF (Amsterdam et al., 1999; Fritz et al., 1996; Golling et al., 2002; Ishikawa, 2000; Loosli et al., 2000; van Eeden et al., 1996). FGF signaling functions in MFF development Judging from fragmented information in previous reports, the role of FGF signaling in MFF development appears to be complicated, and functional redundancy of the signaling should be assumed. No clear phenotype in the MFF has been reported in any mutants of FGF ligands, receptors, and related molecules in zebrafish (Draper et al., 2003; Fischer et al., 2003; Grandel et al., 2000; Herzog et al., 2004; Lee et al., 2004; Norton et al.,

363

2005; Reifers et al., 1998; van Eeden et al., 1996; Whitehead et al., 2005). This is partially because of the difficulty in analysis of the MFF. For example, fgf24 and fgf8, both of which belong to the same subgroup of the FGF family, show redundant function in posterior body formation (Draper et al., 2003). Both single gene disruption mutants only display mild or no phenotype of posterior body formation, and normal MFF could be seen in these single mutants (Draper et al., 2003; Fischer et al., 2003; Reifers et al., 1998). fgf24/fgf8 double mutant shows complete loss of the posterior body (Draper et al., 2003), and MFF development therefore cannot be investigated in the double mutant. In support of redundancy in signaling for formation of the MFF structure, it is noteworthy that mutant fish that completely lacked the dorsal MFF were not isolated from mutagenesis screens, although fin reduction was observed in some mutants (Fritz et al., 1996; Golling et al., 2002; van Eeden et al., 1996). We showed via SU5402 inhibition that FGF signaling is required for progression of the anterior extension of MFF structure formation. SU5402 is a chemical inhibitor of Fgfr, which has been reported to specifically inhibit the kinase activity of nearly all types of Fgfr (Furthauer et al., 2001; Jackman et al., 2004; Mandler and Neubuser, 2001), and it can potentially reveal requirements of FGF signaling that may not be revealed by knocking down specific FGF ligands or receptors because of redundancy. A short period (3 h) of SU5402 administration was sufficient for inhibition of the anterior progress of MFF structure formation and reduction in sp9 expression in the MFF, suggesting that SU5402 directly affects the MFF cells. However, our loss-of-function study did not result in complete disruption of the MFF, and the posterior part of the dorsal midline around the tail bud still had wedgeshaped cells. These results suggest that maintenance of the MFF structure involves other signaling mechanisms. The reduction of sp9 expression in the dorsal midline after SU5402 treatment suggests involvement of sp9 in MFF formation. Embryological analyses have revealed that sp9 expressed in the apical fold of the pectoral fin buds plays an essential role in the maintenance of the apical fold and pectoral fin outgrowth (Kawakami et al., 2004b). sp9 plays a role in pectoral fin formation together with FGF signaling, suggesting its function in MFF formation mediated by FGF signaling. Further investigations with spatially and temporally specific disruption of FGF signaling and sp9 genes will uncover the role of each gene in MFF structure formation. Our gain-of-function analysis of FGF confirmed that FGF is sufficient to induce an additional MFF, strongly suggesting that FGF signals have an important role in MFF structure formation. Taken together with the fact that FGF signaling is crucial for induction of the AER and the apical fold in tetrapod limbs and fish pectoral fins, respectively, our findings suggest that these three epidermal structures that are essential for each appendage formation share the same mechanism mediated by FGF for the initiation. The FGFs we used in the present study showed different activities for the additional MFF induction, and FGF7 and FGF10 had much higher activity levels. These two FGFs belong to the same FGF subfamily and share the same receptor,

364

G. Abe et al. / Developmental Biology 304 (2007) 355–366

Fgfr2b (reviewed by Itoh and Ornitz, 2004; Powers et al., 2000). They also show similar activity in many systems, and the main target of their activity is epidermal cells and epithelial tissues (reviewed by Bates, 2006; Ware and Matthay, 2002). In the chick, moreover, it has been demonstrated that FGF7 and FGF10 have direct effects on epidermal cells for AER induction (Yonei-Tamura et al., 1999). Also in the case of MFF structure formation, it is possible that the FGF7/10 activity directly targets epidermal cells. Some FGF family genes have been shown to be expressed in several regions of the trunk and tail (Cao et al., 2004; Draper et al., 2003; Ng et al., 2002; Nomura et al., 2006; Reifers et al., 1998; Shimizu et al., 2005). However, no significant role of these FGFs in MFF formation has not been reported, and, moreover, as described above, it is possible that multiple FGF ligands contribute to the formation in a redundant manner. Thus, which FGF ligand(s) is responsible for MFF structure formation remains unknown, but we speculate the idea that the FGF function is exerted at the tail bud. FGF7/10 subfamily members, fgf3 and fgf10, are expressed in the tail bud mesenchyme (Furthauer et al., 2001; Ng et al., 2002; Shimizu et al., 2005), suggesting their function there. In this study, FGF7/10 proteins induced ectopic expression of fgf24 in the extended MFF formed at the anterior–dorsal epidermis. Since fgf24 is intrinsically expressed in the MFF only around the tail bud, it is thought that the induction of the additional MFF formation involves molecular mechanisms exerted in the tail bud. Our finding that MFF structure is formed from the tail bud and extends toward the head also supports the idea that the signaling center to organize the MFF is at the tail bud.

(Da), suggested that somites have an important role in this process (Ohtsuka et al., 2004). In the Da mutant, the MFF shows a phenotype in which the dorsal portion of the MFF shifts toward the head. Causative genes for the Da mutation are zic1 and/or zic4, and these genes are not expressed in the MFF itself but in the somite derivatives and dorsal spinal cord. Moreover, zic1/4 expression in somites is only affected in the Da mutant, suggesting that alteration of zic gene expression in somite derivatives gives rise to the Da phenotype, including the MFF deformity. Our findings suggest that the early stage of MFF development consists of two fundamental steps: specification of the MFF territory and construction of the MFF structure. The latter step has an interesting feature, that is, construction proceeding from caudal to rostral. It is likely that FGF signaling contributes in different ways to both steps, and the contribution of FGF signaling to the process of MFF formation suggests a conserved mechanism of development between the apical fold/AER in paired appendages and the MFF in median fins. Acknowledgments We thank Dr. Yasuhiko Kawakami (The Salk institute, USA), Dr. Atsushi Kawakami (Tokyo Institute of Technology, Japan), and Dr. Kyo Yamasu (Saitama University, Japan) for kind gifts of the zebrafish sp9, dlx5a, and fgf24 cDNA plasmid, respectively. We also thank Dr. Hitoshi Okamoto and Dr. Yoshikazu Hirate (Riken BSI, Japan) for technical advice on bead implantation into the zebrafish embryo. This work was supported by research grants from the Ministry of Education, Science, Sports and Culture of Japan.

Specification of the MFF-forming field dlx5a started to be expressed in the non-neural dorsal epidermis adjacent to the neural keel at 16 hpf before MFF structure formation begins (18 hpf) (Figs. 2 and 3). The anterior boundary of the dlx5a-expressing region was at the 8th somite level, being consistent with the anterior edge of the MFF, and this agreement continued to later stages. These results suggest that this dlx5a-expressing region corresponds to the prospective MFF territory. If this is the case, it is thought that the territory for the MFF on the dorsal midline has already been specified by 16 hpf, before the MFF structure formation. FGF7/10 application was sufficient to induce additional dlx5a expression, suggesting that FGF signaling is involved in the specification of the MFF territory. It is likely, however, that FGF signaling is not essential for this process because inhibition of FGF signaling with SU5402 did not alter dlx5a expression, and it appears that other signaling plays a role in the maintenance of dlx5a expression. Although upstream mechanisms for the region specificity of dlx5a expression remain to be determined, one reasonable assumption is that specification of the MFF territory is regulated by the Hox code of the axial mesoderm. Supporting this idea, recent findings for median fin development in the catshark suggest that expression of Hoxd genes in the axial mesoderm specifies the positions of median fins (Freitas et al., 2006). Previous studies on a medaka mutant, Double anal fin

References Ahn, D.G., Kourakis, M.J., Rohde, L.A., Silver, L.M., Ho, R.K., 2002. T-box gene tbx5 is essential for formation of the pectoral limb bud. Nature 417, 754–758. Akimenko, M.A., Ekker, M., Wegner, J., Lin, W., Westerfield, M., 1994. Combinatorial expression of three zebrafish genes related to distal-less: part of a homeobox gene code for the head. J. Neurosci. 14, 3475–3486. Akimenko, M.A., Johnson, S.L., Westerfield, M., Ekker, M., 1995. Differential induction of four msx homeobox genes during fin development and regeneration in zebrafish. Development 121, 347–357. Amsterdam, A., Burgess, S., Golling, G., Chen, W., Sun, Z., Townsend, K., Farrington, S., Haldi, M., Hopkins, N., 1999. A large-scale insertional mutagenesis screen in zebrafish. Genes Dev. 13, 2713–2724. Bates, C.M., 2006. Role of fibroblast growth factor receptor signaling in kidney development. Pediatr. Nephrol. (Published online: 25 Aug 2006, doi:10.1007/s00467-006-0239-7). Cao, Y., Zhao, J., Sun, Z., Zhao, Z., Postlethwait, J., Meng, A., 2004. fgf17b, a novel member of Fgf family, helps patterning zebrafish embryos. Dev. Biol. 271, 130–143. Capdevila, J., Izpisua Belmonte, J.C., 2001. Patterning mechanisms controlling vertebrate limb development. Annu. Rev. Cell Dev. Biol. 17, 87–132. Coates, M.I., 1994. The origin of vertebrate limbs. Dev. 169–180. Connors, S.A., Trout, J., Ekker, M., Mullins, M.C., 1999. The role of tolloid/ mini fin in dorsoventral pattern formation of the zebrafish embryo. Development 126, 3119–3130. Dane, P.J., Tucker, J.B., 1985. Modulation of epidermal cell shaping and extracellular matrix during caudal fin morphogenesis in the zebra fish Brachydanio rerio. J. Embryol. Exp. Morphol. 87, 145–161.

G. Abe et al. / Developmental Biology 304 (2007) 355–366 Dick, A., Hild, M., Bauer, H., Imai, Y., Maifeld, H., Schier, A.F., Talbot, W.S., Bouwmeester, T., Hammerschmidt, M., 2000. Essential role of Bmp7 (snailhouse) and its prodomain in dorsoventral patterning of the zebrafish embryo. Development 127, 343–354. Donoghue, P.C., Forey, P.L., Aldridge, R.J., 2000. Conodont affinity and chordate phylogeny. Biol. Rev. Cambridge Philos. Soc. 75, 191–251. Draper, B.W., Stock, D.W., Kimmel, C.B., 2003. Zebrafish fgf24 functions with fgf8 to promote posterior mesodermal development. Development 130, 4639–4654. Ellies, D.L., Stock, D.W., Hatch, G., Grioux, G., Weiss, K.M., Ekker, M., 1997. Relationship between the genomic organization and the overlapping embryonic expression patterns of the zebrafish dlx genes. Genomics 45, 580–590. Fischer, S., Draper, B.W., Neumann, C.J., 2003. The zebrafish fgf24 mutant identifies an additional level of Fgf signaling involved in vertebrate forelimb initiation. Development 130, 3515–3524. Freitas, R., Zhang, G., Cohn, M.J., 2006. Evidence that mechanisms of fin development evolved in the midline of early vertebrates. Nature 442, 1033–1037. Fritz, A., Rozowski, M., Walker, C., Westerfield, M., 1996. Identification of selected gamma-ray induced deficiencies in zebrafish using multiplex polymerase chain reaction. Genetics 144, 1735–1745. Furthauer, M., Reifers, F., Brand, M., Thisse, B., Thisse, C., 2001. sprouty4 acts in vivo as a feedback-induced antagonist of FGF signaling in zebrafish. Development 128, 2175–2186. Golling, G., Amsterdam, A., Sun, Z.X., Antonelli, M., Maldonado, E., Chen, W.B., Burgess, S., Haldi, M., Artzt, K., Farrington, S., Lin, S.Y., Nissen, R.M., Hopkins, N., 2002. Insertional mutagenesis in zebrafish rapidly identifies genes essential for early vertebrate development. Nat. Genet. 31, 135–140. Grandel, H., Schulte-Merker, S., 1998. The development of the paired fins in the zebrafish (Danio rerio). Mech. Dev. 79, 99–120. Grandel, H., Draper, B.W., Schulte-Merker, S., 2000. dackel acts in the ectoderm of the zebrafish pectoral fin bud to maintain AER signaling. Development 127, 4169–4178. Hall, B.K., 1999. The Neural Crest in Development and Evolution. SpringerVerlag, New York. Herzog, W., Sonntag, C., von der Hardt, S., Roehl, H.H., Varga, Z.M., Hammerschmidt, M., 2004. Fgf3 signaling from the ventral diencephalon is required for early specification and subsequent survival of the zebrafish adenohypophysis. Development 131, 3681–3692. Hirate, Y., Okamoto, H., 2006. Canopy1, a novel regulator of FGF signaling around the midbrain–hindbrain boundary in zebrafish. Curr. Biol. 16, 421–427. Ishikawa, Y., 2000. Medakafish as a model system for vertebrate developmental genetics. BioEssays 22, 487–495. Itoh, N., Ornitz, D.M., 2004. Evolution of the Fgf and Fgfr gene families. Trends Genet. 20, 563–569. Jackman, W.R., Draper, B.W., Stock, D.W., 2004. Fgf signaling is required for zebrafish tooth development. Dev. Biol. 274, 139–157. Kawakami, A., Fukazawa, T., Takeda, H., 2004a. Early fin primordia of zebrafish larvae regenerate by a similar growth control mechanism with adult regeneration. Dev. Dyn. 231, 693–699. Kawakami, Y., Esteban, C.R., Matsui, T., Rodriguez-Leon, J., Kato, S., Belmonte, J.C., 2004b. Sp8 and Sp9, two closely related buttonhead-like transcription factors, regulate Fgf8 expression and limb outgrowth in vertebrate embryos. Development 131, 4763–4774. Kimmel, C.B., Ballard, W.W., Kimmel, S.R., Ullmann, B., Schilling, T.F., 1995. Stages of embryonic development of the zebrafish. Dev. Dyn. 203, 253–310. Kishimoto, Y., Lee, K.H., Zon, L., Hammerschmidt, M., Schulte-Merker, S., 1997. The molecular nature of zebrafish swirl: BMP2 function is essential during early dorsoventral patterning. Development 124, 4457–4466. Kramer, C., Mayr, T., Nowak, M., Schumacher, J., Runke, G., Bauer, H., Wagner, D.S., Schmid, B., Imai, Y., Talbot, W.S., Mullins, M.C., Hammerschmidt, M., 2002. Maternally supplied Smad5 is required for ventral specification in zebrafish embryos prior to zygotic Bmp signaling. Dev. Biol. 250, 263–279. Lee, J.S., von der Hardt, S., Rusch, M.A., Stringer, S.E., Stickney, H.L., Talbot, W.S., Geisler, R., Nusslein-Volhard, C., Selleck, S.B., Chien, C.B., Roehl,

365

H., 2004. Axon sorting in the optic tract requires HSPG synthesis by ext2 (dackel) and extl3 (boxer). Neuron 44, 947–960. Loosli, F., Koster, R.W., Carl, M., Kuhnlein, R., Henrich, T., Mucke, M., Krone, A., Wittbrodt, J., 2000. A genetic screen for mutations affecting embryonic development in medaka fish (Oryzias latipes). Mech. Dev. 97, 133–139. Mabee, P.M., Crotwell, P.L., Bird, N.C., Burke, A.C., 2002. Evolution of median fin modules in the axial skeleton of fishes. J. Exp. Zool. 294, 77–90. Mandler, M., Neubuser, A., 2001. FGF signaling is necessary for the specification of the odontogenic mesenchyme. Dev. Biol. 240, 548–559. Maroon, H., Walshe, J., Mahmood, R., Kiefer, P., Dickson, C., Mason, I., 2002. Fgf3 and Fgf8 are required together for formation of the otic placode and vesicle. Development 129, 2099–2108. Martin, G.R., 1998. The roles of FGFs in the early development of vertebrate limbs. Genes Dev. 12, 1571–1586. Min, H., Danilenko, D.M., Scully, S.A., Bolon, B., Ring, B.D., Tarpley, J.E., DeRose, M., Simonet, W.S., 1998. Fgf-10 is required for both limb and lung development and exhibits striking functional similarity to Drosophila branchless. Genes Dev. 12, 3156–3161. Mintzer, K.A., Lee, M.A., Runke, G., Trout, J., Whitman, M., Mullins, M.C., 2001. Lost-a-fin encodes a type I BMP receptor, Alk8, acting maternally and zygotically in dorsoventral pattern formation. Development 128, 859–869. Mohammadi, M., McMahon, G., Sun, L., Tang, C., Hirth, P., Yeh, B.K., Hubbard, S.R., Schlessinger, J., 1997. Structures of the tyrosine kinase domain of fibroblast growth factor receptor in complex with inhibitors. Science 276, 955–960. Mullins, M.C., Hammerschmidt, M., Kane, D.A., Odenthal, J., Brand, M., van Eeden, F.J., Furutani-Seiki, M., Granato, M., Haffter, P., Heisenberg, C.P., Jiang, Y.J., Kelsh, R.N., Nusslein-Volhard, C., 1996. Genes establishing dorsoventral pattern formation in the zebrafish embryo: the ventral specifying genes. Development 123, 81–93. Neumann, C.J., Grandel, H., Gaffield, W., Schulte-Merker, S., NussleinVolhard, C., 1999. Transient establishment of anteroposterior polarity in the zebrafish pectoral fin bud in the absence of sonic hedgehog activity. Development 126, 4817–4826. Ng, J.K., Kawakami, Y., Buscher, D., Raya, A., Itoh, T., Koth, C.M., Rodriguez Esteban, C., Rodriguez-Leon, J., Garrity, D.M., Fishman, M.C., Izpisua Belmonte, J.C., 2002. The limb identity gene Tbx5 promotes limb initiation by interacting with Wnt2b and Fgf10. Development 129, 5161–5170. Niswander, L., 2003. Pattern formation: old models out on a limb. Nat. Rev., Genet. 4, 133–143. Nomura, R., Kamei, E., Hotta, Y., Konishi, M., Miyake, A., Itoh, N., 2006. Fgf16 is essential for pectoral fin bud formation in zebrafish. Biochem. Biophys. Res. Commun. 347, 340–346. Norton, W.H., Ledin, J., Grandel, H., Neumann, C.J., 2005. HSPG synthesis by zebrafish Ext2 and Extl3 is required for Fgf10 signalling during limb development. Development 132, 4963–4973. Ohtsuka, M., Kikuchi, N., Yokoi, H., Kinoshita, M., Wakamatsu, Y., Ozato, K., Takeda, H., Inoko, H., Kimura, M., 2004. Possible roles of zic1 and zic4, identified within the medaka Double anal fin (Da) locus, in dorsoventral patterning of the trunk–tail region (related to phenotypes of the Da mutant). Mech. Dev. 121, 873–882. Ohuchi, H., Nakagawa, T., Yamamoto, A., Araga, A., Ohata, T., Ishimaru, Y., Yoshioka, H., Kuwana, T., Nohno, T., Yamasaki, M., Itoh, N., Noji, S., 1997. The mesenchymal factor, FGF10, initiates and maintains the outgrowth of the chick limb bud through interaction with FGF8, an apical ectodermal factor. Development 124, 2235–2244. Panganiban, G., Rubenstein, J.L.R., 2002. Developmental functions of the Distal-less/Dlx homeobox genes. Development 129, 4371–4386. Powers, C.J., McLeskey, S.W., Wellstein, A., 2000. Fibroblast growth factors, their receptors and signaling. Endocr. Relat. Cancer 7, 165–197. Pyati, U.J., Webb, A.E., Kimelman, D., 2005. Transgenic zebrafish reveal stagespecific roles for Bmp signaling in ventral and posterior mesoderm development. Development 132, 2333–2343. Pyati, U.J., Cooper, M.S., Davidson, A.J., Nechiporuk, A., Kimelman, D., 2006. Sustained Bmp signaling is essential for cloaca development in zebrafish. Development 133, 2275–2284. Quint, E., Zerucha, T., Ekker, M., 2000. Differential expression of orthologous

366

G. Abe et al. / Developmental Biology 304 (2007) 355–366

Dlx genes in zebrafish and mice: implications for the evolution of the Dlx homeobox gene family. J. Exp. Zool. Mol. Dev. Evol. 288, 235–241. Reifers, F., Bohli, H., Walsh, E.C., Crossley, P.H., Stainier, D.Y., Brand, M., 1998. Fgf8 is mutated in zebrafish acerebellar (ace) mutants and is required for maintenance of midbrain–hindbrain boundary development and somitogenesis. Development 125, 2381–2395. Roehl, H., Nusslein-Volhard, C., 2001. Zebrafish pea3 and erm are general targets of FGF8 signaling. Curr. Biol. 11, 503–507. Schulte-Merker, S., Ho, R.K., Herrmann, B.G., Nusslein-Volhard, C., 1992. The protein product of the zebrafish homologue of the mouse T gene is expressed in nuclei of the germ ring and the notochord of the early embryo. Development 116, 1021–1032. Sekine, K., Ohuchi, H., Fujiwara, M., Yamasaki, M., Yoshizawa, T., Sato, T., Yagishita, N., Matsui, D., Koga, Y., Itoh, N., Kato, S., 1999. Fgf10 is essential for limb and lung formation. Nat. Genet. 21, 138–141. Shimizu, T., Bae, Y.K., Muraoka, O., Hibi, M., 2005. Interaction of Wnt and caudal-related genes in zebrafish posterior body formation. Dev. Biol. 279, 125–141. Shinya, M., Koshida, S., Sawada, A., Kuroiwa, A., Takeda, H., 2001. Fgf signalling through MAPK cascade is required for development of the subpallial telencephalon in zebrafish embryos. Development 128, 4153–4164. Suzuki, T., Haga, Y., Takeuchi, T., Uji, S., Hashimoto, H., Kurokawa, T., 2003. Differentiation of chondrocytes and scleroblasts during dorsal fin skeletogenesis in flounder larvae. Dev. Growth Differ. 45, 435–448.

Tamura, K., Yonei-Tamura, S., Belmonte, J.C., 1999. Differential expression of Tbx4 and Tbx5 in zebrafish fin buds. Mech. Dev. 87, 181–184. Tamura, K., Kuraishi, R., Saito, D., Masaki, H., Ide, H., Yonei-Tamura, S., 2001. Evolutionary aspects of positioning and identification of vertebrate limbs. J. Anat. 199, 195–204. Tickle, C., 2002. Molecular basis of vertebrate limb patterning. Am. J. Med. Genet. 112, 250–255. van Eeden, F.J., Granato, M., Schach, U., Brand, M., Furutani-Seiki, M., Haffter, P., Hammerschmidt, M., Heisenberg, C.P., Jiang, Y.J., Kane, D.A., Kelsh, R.N., Mullins, M.C., Odenthal, J., Warga, R.M., Nusslein-Volhard, C., 1996. Genetic analysis of fin formation in the zebrafish Danio rerio. Development 123, 255–262. Ware, L.B., Matthay, M.A., 2002. Keratinocyte and hepatocyte growth factors in the lung: roles in lung development, inflammation, and repair. Am. J. Physiol.: Lung. Cell Mol. Physiol. 282, L924–L940. Whitehead, G.G., Makino, S., Lien, C.L., Keating, M.T., 2005. fgf20 is essential for initiating zebrafish fin regeneration. Science 310, 1957–1960. Xu, X., Weinstein, M., Li, C., Deng, C., 1999. Fibroblast growth factor receptors (FGFRs) and their roles in limb development. Cell Tissue Res. 296, 33–43. Yonei-Tamura, S., Endo, T., Yajima, H., Ohuchi, H., Ide, H., Tamura, K., 1999. FGF7 and FGF10 directly induce the apical ectodermal ridge in chick embryos. Dev. Biol. 211, 133–143. Zhang, X.G., Hou, X.G., 2004. Evidence for a single median fin-fold and tail in the Lower Cambrian vertebrate Haikouichthys ercaicunensis. J. Evol. Biol. 17, 1162–1166.