The FGFR Pathway Is Required for the Trunk-Inducing Functions of Spemann's Organizer

Developmental Biology 237, 295–305 (2001) doi:10.1006/dbio.2001.0385, available online at http://www.idealibrary.com on

The FGFR Pathway Is Required for the TrunkInducing Functions of Spemann’s Organizer Tracy S. Mitchell and Michael D. Sheets 1 Department of Biomolecular Chemistry, University of Wisconsin, Madison, Wisconsin 53706

Xenopus laevis embryogenesis is controlled by the inducing activities of Spemann’s organizer. These inducing activities are separated into two distinct suborganizers: a trunk organizer and a head organizer. The trunk organizer induces the formation of posterior structures by emitting signals and directing morphogenesis. Here, we report that the fibroblast growth factor receptor (FGFR) signaling pathway, also known to regulate posterior development, performs critical functions within the cells of Spemann’s organizer. Specifically, the FGFR pathway was required in the organizer cells in order for those cells to induce the formation of somitic muscle and the pronephros. Since the organizer influences the differentiation of these tissues by emitting signals that pattern the mesodermal germ layer, our data indicate that the FGFR regulates the production of these signals. In addition, the FGFR pathway was required for the expression of chordin, an organizer-specific protein required for the trunk-inducing activities of Spemann’s organizer. Significantly, the FGFR pathway had a minimal effect on the function of the head organizer. We propose that the FGFR pathway is a defining molecular component that distinguishes the trunk organizer from the head organizer by controlling the expression of organizer-specific genes required to induce the formation of posterior structures and somitic muscle in neighboring cells. The implications of our findings for the evolutionarily conserved role of the FGFR pathway in the functions of Spemann’s organizer and other vertebrate-signaling centers are discussed. © 2001 Academic Press Key Words: Spemann’s organizer; trunk organizer; Xenopus embryogenesis; FGFR pathway.

INTRODUCTION Spemann’s organizer is segregated into subdomains that each exhibit distinct inducing activities. Transplantation experiments in Xenopus indicate that the organizer cells furthest from the early blastopore lip act as a trunk organizer and induce the formation of trunk/tail structures that contain ectopic muscle (Zoltewicz and Gerhart, 1997). The trunk organizer functions nonautonomously to regulate posterior development by emitting inducing signals and directing morphogenesis. In contrast, the organizer cells closest to the early blastopore lip act as a head organizer and induce the formation of head structures. These results demonstrate that there is a clear functional distinction between the trunk and head suborganizers that regulate posterior and anterior development. The fibroblast growth factor receptor (FGFR) signal transduction pathway also regulates the posterior development of Xenopus embryos. The definitive evidence for this con1

To whom correspondence should be addressed. Fax: 608-2625253. E-mail: [email protected]. 0012-1606/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.

clusion was provided by experiments in which a dominant negative form of the FGFR (DN-FGFR or XFD) was expressed in developing Xenopus embryos (Amaya et al., 1991, 1993; Isaacs et al., 1994). The DN-FGFR blocks activation of the endogenous FGFR and therefore allows the identification of developmental processes that require the FGFR pathway. Expression of the DN-FGFR in Xenopus embryos causes severe posterior truncations, but the formation of head structures in these embryos is only minimally perturbed. These experiments clearly demonstrate that the FGFR pathway regulates posterior development, but the mechanisms underlying this regulation are unclear. An important aspect of posterior development regulated by the FGFR pathway is the formation of specific mesodermal cell types such as somitic muscle. Inhibiting the FGFR pathway in embryos disrupts the formation of both somitic muscle and notochord, and disrupts the expression of the mesodermal regulatory factor Xbra (Amaya et al., 1991, 1993; Isaacs et al., 1994). In addition, several in vitro studies have demonstrated that purified FGF ligands can cause naive embryonic cells to form mesoderm (Kimelman and Kirschner, 1987; Slack et al., 1987). These and other results

295

296

Mitchell and Sheets

clearly demonstrated the importance of the FGFR pathway for the induction and/or maintenance of mesoderm in the embryo and contributed to focusing subsequent research primarily on the autonomous functions of this pathway (Kimelman and Griffin, 1998). Although most studies have addressed how the autonomous functions of the FGFR pathway affect mesoderm formation, several observations point to a nonautonomous role of the FGFR pathway in mesoderm formation as well. For example, the most effective strategy for inhibiting posterior development controlled by the FGFR pathway was to bias DN-FGFR expression to the cells that give rise to Spemann’s organizer (Amaya et al., 1993; Isaacs et al., 1994). Since Spemann’s organizer controls mesoderm formation nonautonomously by producing specific signals and directing morphogenesis (Harland and Gerhart, 1997), these observations raised the possibility that the FGFR pathway might regulate the nonautonomous mesoderm-patterning functions of the organizer. However, the potential importance of the FGFR pathway for regulating organizer signaling and mesodermal patterning has not been directly investigated. This report provides evidence that the FGFR pathway regulates the functions of the trunk organizer in the Xenopus embryo. The FGFR pathway was required in the organizer cells in order for those cells to induce the formation of structures such as the somitic muscle and pronephros. Consistent with previous studies, blocking the FGFR pathway had minimal effects on anterior development, providing evidence that the FGFR pathway molecularly defines the trunk organizer and distinguishes it from the head organizer. In addition, the FGFR pathway was required for the expression of chordin, a protein secreted from Spemann’s organizer that induces posterior structures and patterns the mesoderm. Thus, one mechanism by which the FGFR pathway helps define the trunk/tail subdomain of Spemann’s organizer is to regulate the expression of trunkinducing proteins. These findings provide new insights into the molecular nature of Spemann’s organizer and contribute to a growing body of evidence that the FGFR pathway is a general molecular component of developmental signaling centers that regulate vertebrate embryogenesis.

MATERIALS AND METHODS Embryos and Injections Embryos were obtained by standard methods and staged according to Nieuwkoop and Faber (1994). Only embryos exhibiting clear pigmentation differences and symmetrical cleavage furrows were chosen for mRNA injection. mRNAs encoding the dominant negative FGF receptor (DN-FGFR, also called XFD), the nonfunctional FGF receptor HAVØ and ␤-galactosidase were generated as previously described (Amaya et al., 1991, 1993). All injections were performed in 0.25⫻ MMR containing 4% ficoll. For organizerdirected injections, either both AB1 blastomeres, both CD1 blastomeres, or one AB1 and one CD1 blastomere of 16-cell-stage embryos were injected at the marginal zone close to the medial

cleavage plane with 300 pg of FGFR mRNA. The AB1 and CD1 blastomeres are the cells of 16-cell-stage embryos that will give rise to the A1, B1, C1, and D1 cells of 32-cell-stage embryos. In some experiments, mRNA encoding ␤-galactosidase (100 pg) was coinjected to identify progeny of the injected cells. For surgical experiments and gene expression analysis, the marginal zone of the two anterior (lighter pigmented, dorsal) blastomeres of 4-cell embryos was injected near the cleavage planes with 100 pg of ␤-galactosidase mRNA and 300 pg of mRNA encoding either DN-FGFR or HAVØFGFR. Injected embryos were incubated overnight at 13–14°C in 0.25⫻ MMR/4% ficoll containing 10 ␮g/ml gentamicin.

Organizer Transplantations Gastrula-stage embryos that had been injected with mRNAs encoding the DN-FGFR and ␤-galactosidase or the HAVØ-FGFR and ␤-galactosidase were transferred to agarose-coated dishes containing 0.25⫻ MMR. Organizer tissue was excised from injected stage-10⫹ embryos with a hairknife and an eyepiece reticule engraved with a protractor. Two small incisions were made into the marginal zone ⫾ 45° to each side of the center of the early blastopore lip. Organizer tissue was then removed by cutting from the animal to vegetal pole through each incision, thus removing a 90° wedge of tissue (Lane and Sheets, 2000). Presumptive posterior tissue was removed from uninjected recipient embryos in a similar manner; however, in this case, the two small incisions were made into the marginal zone ⫾ 135° to each side of the center of the early blastopore lip. For each transplant, organizer tissue was inserted into the cavity generated by the removal of the presumptive posterior tissue. Each embryo receiving a transplant was transferred to an agarose well of a dish containing DFA/BSA solution and antimycotic/antibiotic (Sater et al., 1993). The remaining pieces of tissue from injected and control embryos were cultured in separate wells to monitor the accuracy of the microsurgery. All transplants and tissues were cultured overnight at room temperature. After 18 h, the embryos containing grafts were transferred to agarose wells in a new dish containing (1:1) DFA/BSA:0.25⫻ MMR and antimycotic/antibiotic. Embryos receiving transplants were fixed when sibling embryos reached stage 28 –32 and examined for the presence of secondary axes containing both an ectopic head and trunk. Ectopic heads were judged by the presence of the cement gland and at least one eye. Ectopic trunk formation was assessed morphologically by the presence of a dorsal fin and significant tissue separating the ectopic head from the trunk and head of the primary axis. Ectopic somitic muscle formation by the grafted organizer tissue was analyzed by either whole-mount immunocytochemistry with the muscle-specific antibody 12/101 (Kintner and Brockes, 1984) or whole-mount in situ hybridization with a muscle actin probe to detect muscle (Harland, 1991) combined with ␤-galactosidase staining to detect the transplanted organizer cells. The embryos shown in Figs. 2A and 2B were fixed at stage 39 to clearly show the morphology of the head structures. ␤-Galactosidase activity detection. The detection of ␤-galactosidase activity was adapted from existing methods (Houzelstein et al., 1997). Specifically, embryos coinjected with ␤-galactosidase mRNA were fixed at stage 28 –35 in MEMPFAT (1⫻ MEM, 4% paraformaldehyde, 0.1% Tween 20) for 1 h at room temperature. Following fixation, embryos were washed three times in PBS. ␤-galactosidase activity was detected for no more than 2 h at 37°C in 0.5 ml of the following solution: 5 mM K 3Fe(CN) 6, 5 mM K 4Fe(CN) 6, 1 mg/ml Rose ␤-D-gal (Biotium), 2 mM MgCl 2, 0.1% Tween 20, 0.2% paraformaldehyde in PBS. Following detection,

Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.

297

FGFR Pathway Regulates Organizer Function

embryos were washed three times in PBS and fixed in MEMPFA (1⫻ MEM, 4% paraformaldehyde) for either 90 min at room temperature or overnight at 4°C. Embryos were stored in methanol at ⫺20°C. Immunocytochemistry. Injected embryos were scored for morphology and fixed in MEMFA for 2 h at room temperature or overnight at 4°C at the stages indicated below. Somitic muscle was detected in stage 28 –35 embryos by using the muscle-specific 12/101 antibody (Kintner and Brockes, 1984) and an IgG HRPcoupled secondary antibody with DAB substrate. An APconjugated secondary antibody and BCIP (Boehringer Manheim) substrate were used for embryos that had been assayed for ␤-galactosidase activity. Embryos that lacked or exhibited a ⬎50% reduction in 12/101 staining compared to controls were considered defective for muscle formation. Muscle defects were visually assessed by comparing both the somite file organization and length to control embryos. The somite files of embryos exhibiting a ⬎50% reduction in muscle were less than half the length and width of control embryos and lacked the segmented chevron pattern indicative of trunk somites. Pronephric duct formation was detected in stage 38 – 40 embryos by using the duct-specific antibody 4A6 (Vize et al., 1995) and an IgG AP-coupled secondary antibody with BM purple substrate (Boehringer Mannheim). Reduction in duct formation was assessed by visual comparison of duct mass to control embryos. Pronephric duct formation was considered defective if the duct was either entirely absent or only a few 4A6 staining cells were detected. In situ hybridization. Embryos were fixed for 90 min at room temperature in RNase-free MEMFA and stored at ⫺20°C in methanol until analysis. In situ hybridization was performed as described by using the substrate BCIP (Harland, 1991). Northern blot analysis. Injected, stage-10.25 embryos were frozen on dry ice and stored at ⫺80°C until analysis. Sibling embryos were allowed to develop until stage 28 –35, fixed, and analyzed for somitic mesoderm using immunocytochemistry and the 12/101 antibody. Only experiments in which ⬎80% of the DN-FGFR-injected embryos exhibited significant muscle reduction (⬎50% reduction of both somite files) were analyzed by RNA blot hybridization. Total RNA was isolated from embryos by using Trizol reagent (Life Technologies) according to the manufacturer’s instructions. RNA was denatured with glyoxal, fractionated by agarose gel electrophoresis, and transferred to a nylon membrane (Pall Biodyne A) overnight by capillary action. The RNA was fixed to the membrane by UV treatment followed by baking under vacuum at 80°C for 2 h. Membranes were boiled in 20 mM Tris, pH 8.0, for 3 min, prehybidized, hybridized, washed, and exposed to X-ray film by standard protocols (Sheets et al., 1994). Radiolabeled, single-stranded antisense DNA probes synthesized by asymmetric PCR were used for hybridization.

RESULTS The FGFR Pathway Functions in the Cells of Spemann’s Organizer to Control Trunk/Tail Development To test whether the FGFR pathway regulates organizer function, we inhibited activation of this pathway in the organizer precursor cells and analyzed the effects on embryonic development. Fate maps indicate that the AB1 and CD1 cells give rise to the organizer and the anterior

structures of the embryo, but these cells make only minimal contributions to posterior structures of embryo (Figs. 1A and 1F) (Dale and Slack, 1987a; Moody, 1987; Lane and Sheets, 2000). To inhibit activation of the FGFR pathway, mRNA encoding the dominant negative FGF receptor (DNFGFR, also called XFD; Amaya et al., 1991) was injected into AB1 and CD1 cells of 16-cell-stage frog embryos. At the early tadpole stage, injected embryos were analyzed for the reduced trunk and tail structures characteristic of blocked FGFR signal transduction. Trunk and tail development were severely disrupted in the embryos injected with the DN-FGFR mRNA (Fig. 1D, Table 1), while head formation in these embryos was only minimally affected. In contrast, the control embryos injected with mRNA encoding the nonfunctional FGFR, HAVØ, exhibited no developmental abnormalities (Fig. 1C, Table 1). To further test whether the FGFR pathway regulates organizer function, the formation of somitic muscle of the trunk and tail was analyzed. Embryos injected at the 16-cell stage with mRNA encoding either the DN-FGFR or the HAVØFGFR were analyzed at the early tadpole stage for somitic muscle differentiation using whole-mount immunocytochemistry and the muscle-specific antibody 12/101 (Kintner and Brockes, 1985). In some experiments, embryos were coinjected with mRNA encoding ␤-galactosidase to follow the fate of the injected cells. Embryos in which FGFR pathway function was blocked in the organizer cells (DN-FGFR injected) exhibited significant reductions in somitic muscle formation (Fig. 1G, the muscle is stained blue, and Table 2), while the controls (HAVØ-FGFR-expressing embryos) contained normal amounts of muscle patterned into somites (Fig. 1F, Table 2). Staining for ␤-galactosidase activity confirmed that the AB1 and CD1 cells made only minor contributions to the trunk and somites of control embryos, as predicted from fate maps (Fig. 1F, cells expressing ␤-galactosidase are stained red) (Dale and Slack, 1987a; Moody, 1987; Lane and Sheets, 2000). The results of these two experiments indicated that blocking the FGFR pathway in cells that give rise to the organizer and the anterior structures of the embryo had a dramatic effect on the development of posterior structures, specifically the trunk and somitic muscle. The organizer functions as a nonautonomous regulator of trunk and somite development (Dale and Slack, 1987b; Harland and Gerhart, 1997). Therefore, these data provide compelling evidence that the FGFR pathway is necessary for the nonautonomous functions of the organizer. Furthermore, because blocking the FGFR pathway had only minimal effects on anterior development, these results suggest that the FGFR pathway is specifically necessary for the trunk-inducing functions of the organizer.

The Function of the FGFR Pathway Is Required for the Muscle- and Trunk-Inducing Activities of the Organizer If the FGFR pathway regulates the muscle- and trunkinducing functions of the organizer, then an organizer in

Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.

298

Mitchell and Sheets

FIG. 1. Inhibiting the FGFR pathway in the organizer cells blocked trunk/tail development and somitic muscle formation. (A) Blastomere map of a 16-cell-stage Xenopus embryo. AB1 blastomeres will give rise to A1 and B1 daughter cells and CD1 blastomeres give rise to the C1 and D1 daughter cells of the 32-cell-stage embryo. The A1, B1, C1, and D1 cells all contribute to the organizer as well as giving rise to the anterior structures of the embryo. The anterior (A) and posterior (P), formerly dorsal and ventral, origins of embryonic structures are indicated (Lane and Sheets, 2000; Lane and Smith, 1999). (B–D) Both CD1 blastomeres of 16-cell-stage Xenopus embryos were injected with the mRNAs indicated. Embryos were scored at the early tadpole stage for defects in trunk and tail development. The morphology of uninjected (B), HAVØ-FGFR-expressing (C), and DN-FGFR-expressing (D) embryos at stage 31–35. Embryos expressing the DN-FGFR exhibit severe posterior reductions and develop with an open blastopore. Head development in these embryos was relatively normal, the cement gland is marked with an asterisk (*). (E–G) One AB1 and one CD1 blastomere of 16-cell-stage embryos were injected with the mRNAs indicated. Muscle formation in uninjected (E), HAVØ-FGFR-expressing (F), and DN-FGFR-expressing (G) embryos at stage 31–35. All embryos were analyzed for somitic muscle using wholemount immunocytochemistry and the 12/101 antibody, shown in blue. Embryos were coinjected with the mRNA encoding ␤-galactosidase; ␤-galactosidase activity is stained red.

which the FGFR pathway was inhibited should fail to induce ectopic trunks and somitic muscle upon transplantation to a host embryo. To test this model, embryos were injected with mRNA encoding ␤-galactosidase together with mRNA encoding either the DN-FGFR or the HAVØFGFR. When the injected embryos reached the early gastrula stage, stage 10⫹, the organizer tissues were removed by microsurgery and transplanted to the presumptive posterior (ventral) region of stage 10⫹ recipient embryos. When the embryos receiving transplants reached the early tadpole stage, they were stained for ␤-galactosidase activity and assayed for somitic muscle using immunocytochemistry. The embryos receiving organizer transplants were scored for the formation of ectopic trunks and heads as well as for the induction of ectopic muscle directed by the transplant. Organizer cells in which the FGFR pathway was inhibited were defective in their ability to induce host cells to form ectopic trunks and somitic muscle (Figs. 2B and 2D, Table 3). Similar results were obtained by combining organizer tissues with presumptive posterior tissues to assay muscle induction (data not shown). Importantly, not all organizer function was eliminated by the expression of the DNFGFR. These transplanted organizers still directed the formation of ectopic head structures (Figs. 2B and 2D, Table 3). As expected, the control organizers, expressing HAVØFGFR, efficiently formed ectopic heads and induced ectopic trunks that contained somitic muscle (Figs. 2A and 2C). Notably, a significant amount of the somitic muscle tissue that formed in the control transplants arose from cells surrounding the transplanted organizer and not the organizer cells themselves (see Fig. 2C, Table 3, and note the substantial amount of muscle not labeled with ␤-galactosidase). This finding indicates that this muscle formed as a result of induction by the organizer cells. There are two important conclusions that result from these experiments. First, the trunk- and muscle-inducing functions of the organizer require a functional FGFR pathway. Second, the inducing functions of the two organizer subdomains can be distinguished by their dependence upon the FGFR pathway; the trunk organizer requires the FGFR pathway while the head organizer does not.

The FGFR Pathway Is Required for the Pronephric Duct-Inducing Activities of the Organizer The trunk contains multiple tissues, such as the pronephros, whose formation depends upon signals from the organizer (Harland and Gerhart, 1997; Seufert et al., 1999). To test whether the FGFR pathway regulates

The embryos expressing the DN-FGFR in the organizer cells exhibited significant defects in muscle formation, ranging from no muscle detected (the embryo on the left) to a ⬎50% reduction in muscle (embryo on the right).

Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.

299

FGFR Pathway Regulates Organizer Function

FIG. 2. Trunk and muscle induction by the organizer required the FGFR pathway. Organizer tissues from embryos expressing ␤-galactosidase and either the DN-FGFR or the HAVØ-FGFR were transplanted to the presumptive posterior region (ventral) of unmanipulated host embryos. (A, B) Embryos receiving transplants were examined at the early tadpole stage for the presence of secondary axes containing both an ectopic head and trunk. Ectopic heads were judged by the presence of at least one eye (red arrows) and a cement gland (indicated by the asterisk). Ectopic trunk formation was assessed morphologically by the presence of a dorsal fin and significant tissue separating the ectopic head from the trunk and head of the 1° axis. (A) Organizer grafts expressing HAVØ-FGFR induced complete secondary axes containing trunks and heads with eyes (red arrows) and cement glands (asterisk). (B) DN-FGFR-expressing organizer grafts only induced ectopic heads with cement glands and eyes; these tissues did not induce ectopic trunks. (C, D) Embryos receiving organizer grafts were analyzed for somitic muscle induced by the grafted tissue using either immunocytochemistry (shown in blue) or by in situ hybridization with a muscle actin probe (data not shown). Embryos were stained for ␤-galactosidase activity (shown in red) to detect the transplanted organizer cells. (C) Organizer grafts expressing HAVØ-FGFR induced ectopic trunks containing muscle. Note the absence of ␤-galactosidase-expressing cells in the ectopic muscle induced by the HAVØ-FGFR-expressing organizers (magnified image in Fig. 2C). (D) DN-FGFR-expressing organizer grafts did not induce ectopic trunks containing muscle. There was no muscle surrounding the ␤-galactosidase and DN-FGFR-expressing cells (magnified image in Fig. 2D). The results shown are representative of four separate experiments.

organizer signals required for the formation of mesodermal tissues other than somitic muscle, embryos were injected at the 16-cell-stage with mRNA encoding either the DN-FGFR or the HAVØ-FGFR. The injected embryos

were analyzed when they reached stage 38 – 40 for pronephric duct formation using whole-mount immunocytochemistry and the pronephric duct-specific antibody 4A6 (Vize et al., 1995). Embryos in which FGFR pathway

Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.

300

Mitchell and Sheets

TABLE 1 Trunk/Tail Development in Embryos Where the FGFR Pathway Was Inhibited in the Organizer Cells Morphology Cells injected CD1 (both)

AB1 (both)

AB1/CD1 (1 side)

mRNA injected

No. of embryos

% Embryos with trunk and tail defects

% Embryos with normal morphology

DN-FGFR HAVØ-FGFR Uninjected DN-FGFR HAVØ-FGFR Uninjected DN-FGFR HAVØ-FGFR Uninjected

492 234 931 86 55 95 104 78 260

78 4 2 95 0 1 98 1 1

21 90 95 1 100 99 2 96 99

Note. Specific cells of 16-cell-stage Xenopus embryos were injected with the mRNAs indicated and scored at the tadpole stage for defects in trunk and tail development. Embryos scored as defective had relatively normal heads, open blastopores directly behind the head, and reduced trunks and tails. Representative embryos from this experiment are shown in Figs. 1B–1D. The data in this table were compiled from 17 separate experiments.

function was blocked in the organizer cells (DN-FGFRinjected) exhibited significant reductions in pronephric duct formation (Fig. 3C, and Table 4), while the controls (uninjected, HAVØ-FGFR-expressing embryos and the contralateral side of DN-FGFR-expressing embryos) contained normal ducts (Figs. 3A, 3B and 3D, the duct is stained blue, Table 4). Therefore, the pronephric ductinducing functions of the organizer require a functional FGFR pathway. Moreover, these results provide additional evidence that the FGFR pathway is necessary for the nonautonomous functions of the organizer that are required for trunk formation and patterning the mesodermal germ layer into specific tissues.

TABLE 2 Somitic Muscle Formation in Embryos Where the FGFR Pathway Was Inhibited in the Organizer Cells Somitic muscle formation mRNA injected

No. of embryos

% Embryos with muscle defects

DN-FGFR HAVØ-FGFR Uninjected

81 61 159

67 0 0

Note. Both CD1 cells in 16-cell-stage Xenopus embryos were injected with the mRNAs indicated. At the early tadpole stage, these embryos were assayed for somitic muscle using whole-mount immunocytochemistry. Embryos were scored as defective for muscle formation if they exhibited a ⬎50% reduction in muscle compared to controls. Representative embryos from this experiment are shown in Figs. 1E–1G. The data in this table were compiled from four separate experiments.

The FGFR Pathway Regulates the Expression of the Chordin Gene in the Organizer Cells The inducing functions of the organizer have been attributed to secreted proteins, such as chordin and noggin that are synthesized from mRNAs transcribed specifically by the organizer cells (Smith and Harland, 1992; Sasai et al., 1994; Harland and Gerhart, 1997). Our results indicate that at least some of the inducing functions of the organizer are regulated by the FGFR pathway (Figs. 1 and 2). One potential mechanism to explain this regulation is that the FGFR pathway controls the expression of mRNAs by the organizer cells. The FGFR pathway could regulate the synthesis of mRNAs that encode inducing proteins or mRNAs that encode transcription factors that in turn control the synthesis of the inducing proteins. If this hypothesis is correct, then the expression of specific organizer genes should be reduced when the FGFR pathway is blocked. Total RNA was harvested from stage-10.25 embryos that expressed the DN-FGFR or HAVØ- FGFR. The expression of individual organizer-specific mRNAs was analyzed by denaturing agarose gel electrophoresis and RNA blot hybridization (Sheets et al., 1994). Significantly, the level of chordin mRNA was dramatically reduced in embryos expressing the DN-FGFR compared to uninjected control embryos and embryos expressing HAVØ-FGFR (Fig. 4). This effect was highly specific; expression of the mRNAs encoding the goosecoid, Xnot, noggin, and cytoskeletal actin proteins were relatively unaffected by the DN-FGFR, while Xlim expression was reduced to a small extent by DNFGFR (Blumberg et al., 1991; Smith and Harland, 1992; Taira et al., 1992; von Dassow et al., 1993). Thus, the FGFR pathway regulates the expression of the gene encoding the trunk-inducing protein chordin.

Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.

301

FGFR Pathway Regulates Organizer Function

TABLE 3 Trunk and Muscle Induction by Organizer Cells in Which the FGFR Pathway Was Inhibited Muscle induced by grafted organizer

Structures induced by grafted organizer Grafted organizer expressing

No. of embryos

% Induction of ectopic trunks and heads

% Induction of ectopic heads only

No. of embryos

% Induction of ectopic muscle

DN-FGFR HAVØ-FGFR

123 124

13 75

84 21

45 51

4 94

Note. Representative embryos from this experiment are shown in Fig. 2. The morphological data in this table were compiled from nine separate experiments and the muscle induction data were compiled from four separate experiments.

DISCUSSION The FGFR Pathway Is a Molecular Component of the Trunk Organizer Spemann’s organizer is a developmental signaling center that regulates the formation of both posterior and anterior structures in vertebrate embryos. Significantly, the inducing functions of Spemann’s organizer are regionalized into two distinct suborganizers: a trunk/tail organizer, which controls formation of most somitic muscle and posterior structures, and a head organizer, which controls the formation of head and other anterior structures (Zoltewicz and Gerhart, 1997). However, the molecular pathways that distinguish the functions of these suborganizers are unclear. In this report, we provide compelling evidence that the FGFR pathway is a molecular component of the trunk subdomain of Spemann’s organizer. The function of the FGFR pathway was required within the organizer cells for the organizer to induce posterior structures. Since the organizer functions by generating inducing signals that cause neighboring cells to differentiate, our findings indicate that the FGFR pathway regulates these nonautonomous functions of the organizer during early Xenopus embryogenesis. Significantly, the FGFR pathway was not required for the ability of the organizer to regulate the formation of anterior structures. Thus, the FGFR signaling pathway is a defining molecular component that distinguishes the two distinct subdomains of Spemann’s organizer.

The FGFR Pathway Regulates Posterior Development by Controlling the Nonautonomous Functions of the Trunk Organizer that Pattern the Mesoderm The FGFR pathway regulates organizer signaling that patterns the mesoderm. The diverse array of specific mesodermal cell types form during embryonic development as the result of two processes, the induction of the mesodermal germ layer and the patterning of a subset of these cells into particular differentiated cell types. Induction of the

mesoderm during early embryogenesis depends on specific molecular pathways, such as FGFR and TGF␤ growth factor pathways and the activation of specific transcription factors such as vegT (Kimelman and Griffin, 1998; Kofron et al., 1999). Several studies have indicated that the FGFR pathway is necessary for the formation of the mesodermal germ layer. Naı¨ve embryonic cells differentiate as muscle when exposed to FGF ligands, suggesting that FGF ligands may function as mesoderm inducers in the embryo (Kimelman and Kirschner, 1987; Slack et al., 1987). The most definitive evidence has come from inhibiting the function of the endogenous FGFR pathway within Xenopus through the expression of a dominant negative form of the FGFR (DNFGFR) (Amaya et al., 1991, 1993; Isaacs et al., 1994). The formation of many mesodermal cell types, such as muscle and notochord, are disrupted by the DN-FGFR. These and other results indicate that FGFR signaling is important for the induction and/or maintenance of the mesoderm in Xenopus embryos. Our results indicate that the FGFR pathway also controls the formation of mesodermal cell types by regulating the signals required for mesodermal patterning. The organizer emits signals that affect the differentiation of the mesodermal germ layer into specific tissues such as the somitic muscle, the pronephros, and the heart (Dale and Slack, 1987b; Sater and Jacobson, 1990; Nascone and Mercola, 1995; Seufert et al., 1999). These tissues do not form when organizer signaling is disrupted either by removal of the organizer cells or by eliminating organizer formation with UV irradiation. Our results demonstrate that somitic muscle and pronephros formation are also disrupted when organizer signaling is impaired by blocking the FGFR pathway in the organizer cells. Therefore, our results define a new function for the FGFR pathway in the embryo. Specifically, we propose that the FGFR pathway is a regulator of organizer signals required for patterning the mesoderm. Examination of the literature stimulated by our discovery provides additional support for the conclusions presented here. For example, the erk1/2 MAPK protein is activated by the FGFR pathway, and MAPK activation initially occurs in the cells of Spemann’s organizer (Christen and Slack, 1999;

Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.

302

Mitchell and Sheets

gest that the FGFR pathway is active preferentially in cells of the organizer, providing additional evidence consistent with our model for the FGFR pathway in Xenopus embryogenesis (Isaacs et al., 1994; Northrop et al., 1995). Significantly, the ability of the FGFR pathway to distinguish between the trunk/tail and head organizers is unique among the signaling pathways studied to date. For example, blocking either the activin pathway with a dominant negative activin receptor (Hemmati-Brivanlou and Melton, 1992) or the wnt pathway with an inhibitory siamois protein (Fan and Sokol, 1997; Kessler, 1997) eliminates all organizer-specific functions and gene expression. Therefore, the DN-FGFR is a powerful tool for functionally dissecting the trunk- and headinducing activities of Spemann’s organizer.

The Function of the FGFR Pathway in Spemann’s Organizer: An Attenuator of BMP Signal Transduction FIG. 3. Inhibiting the FGFR pathway in the organizer cells blocked formation of the pronephros. One AB1 and one CD1 blastomere of 16-cell-stage Xenopus embryos were injected with the mRNAs indicated. All embryos were analyzed at stage 38 – 40 for the presence of pronephric ducts using whole-mount immunocytochemistry and the 4A6 antibody, shown in blue (red arrow). Pronephric duct formation in uninjected (A) and HAVØ-FGFRexpressing stage 38 – 40 embryos (B). Pronephric duct formation was inhibited on the injected side of embryos expressing DN-FGFR (C) but was normal on the contralateral side of the same embryo (D). The cement gland is marked with an asterisk (*). The results shown are representative of two separate experiments.

Patterning of the early Xenopus embryo is controlled by reciprocal interactions between the presumptive posterior tissues and the cells of Spemann’s organizer (Harland and Gerhart, 1997). The cells fated to become posterior struc-

Curran and Grainger, 2000; LaBonne and Whitman, 1997). In addition, the ectopic overexpression of FGF8 protein in zebrafish mimics the transplantation of trunk organizer cells and directs the formation of ectopic trunk structures (Furthauer et al., 1997). These and other observations sug-

TABLE 4 Pronephric Duct Formation in Embryos Where the FGFR Pathway Was Inhibited in the Organizer Cells Pronephric duct formation

mRNA injected

No. of embryos

% Embryos with pronephric duct defects

DN-FGFR HAVØ-FGFR Uninjected

45 45 50

96 0 0

Note. The AB1 and CD1 cells on one side of 16-cell-stage Xenopus embryos were injected with the mRNAs indicated. At the tadpole stage, these embryos were assayed for presence of the pronephric duct using whole-mount immunocytochemistry. Representative embryos from this experiment are shown in Fig. 3. The data in this table were compiled from two separate experiments.

FIG. 4. The FGFR pathway regulated expression of the chordin gene. Total RNA was isolated from stage-10.25 embryos injected with mRNA encoding the DN-FGFR or the HAVØ-FGFR. The RNA was fractionated by denaturing agarose gel electrophoresis and analyzed by blot hybridization using probes to mRNAs expressed in the organizer cells. The results shown are representative of two separate experiments. Both results for the analysis of noggin expression are shown. The arrowhead marks the position of the noggin mRNA. The asterisk (*) indicates the position of the injected DN-FGFR or the HAVØ-FGFR mRNAs that hybridize to the noggin probe.

Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.

303

FGFR Pathway Regulates Organizer Function

tures produce secreted growth factors such as BMPs and wnts that antagonize anterior development and the actions of Spemann’s organizer (Christian and Moon, 1993; Dale et al., 1992). In turn, the organizer cells produce proteins such as chordin and noggin that antagonize the activity of BMPs, and proteins such as frzb that antagonize the activity of wnts (Leyns et al., 1997; Sasai et al., 1994; Smith and Harland, 1992). The expression of these secreted BMP and wnt antagonists by the organizer cells promotes cells to develop into anterior and dorsal structures of the embryo. Thus, normal embryonic development results from a balance between these antagonistic mechanisms that affect cell fate. Our results indicate that an important role of the FGFR pathway is to control the activities of Spemann’s organizer that are required for the development of the trunk and somitic muscle. The FGFR pathway may exert its effects on organizer function by regulating mechanisms that attenuate BMP signaling. One obvious mechanism by which the FGFR pathway may antagonize BMP signaling is by controlling the expression of proteins that function as BMP antagonists. Our results demonstrate that expression of the chordin protein, a protein that antagonizes BMP signaling and is necessary for the trunk/tail-inducing functions of Spemann’s organizer, requires the FGFR pathway. In addition, the FGFR pathway may also inhibit BMP signaling by a more direct mechanism. There are several examples that demonstrate that activation of tyrosine kinase receptors, such as the FGF receptor, can inhibit BMP signal transduction (Kretzschmar et al., 1997; de Caestecker et al., 1998; Kretzschmar et al., 1999; Ulloa et al., 1999). For instance, the Smad1 transcription factor is activated in response to BMP signaling, but Smad1 function is inhibited by activation of the MAPK proteins, known signaling components of the FGFR pathway (Kretzschmar et al., 1997; Szebenyi and Fallon, 1999). Thus, the FGFR and BMP signaling pathways can compete directly for alternate modes of regulation of common downstream signaling components. Therefore, a critical role of the FGFR pathway in Spemann’s organizer may be to attenuate BMP signaling both indirectly by controlling chordin gene expression and directly by regulating common downstream signaling components such as Smad1.

The FGFR Pathway as a Fundamental Regulator of Vertebrate Signaling Centers A number of studies indicate that the FGFR pathway and signaling centers act in concert to control various aspects of vertebrate development. Limb formation is governed by the activity of the AER signaling center (Johnson and Tabin, 1997; Ng et al., 1999), while tooth development is controlled by the activities of the primary enamel knot signaling center (Jernvall and Thesleff, 2000). Significantly, both limb formation and tooth development also require a functional FGFR pathway, but it has been difficult to assess directly whether the FGFR pathway is actually required within these signaling centers themselves (Jernvall et al.,

1994; Martin, 1998; Niswander et al., 1993). However, these studies do provide evidence that the FGFR pathway antagonizes BMP signaling associated with these signaling centers (Neubuser et al., 1997; Niswander and Martin, 1993; Pizette and Niswander, 1999). We have now shown that Spemann’s organizer, an evolutionary conserved and fundamental signaling center that controls the formation of the vertebrate body plan and associated tissues, requires the FGFR pathway for normal function. Our results, combined with observations associated with other signaling centers, lead us to propose that a balance between the FGFR and BMP signaling pathways is a fundamental mechanistic feature controlling the formation and function of all vertebrate signaling centers. Given the evolutionary conservation and relevance of the FGFR pathway to vertebrate signaling centers in general, a rigorous dissection of the FGFR pathway’s role in Spemann’s organizer should continue to provide insights into the fundamental mechanisms that regulate vertebrate development.

ACKNOWLEDGMENTS We thank Connie Lane for her invaluable intellectual and technical insights. We thank Catherine Fox and members of the Sheets and Fox labs for their inputs and discussions. Enrique Amaya generously supplied the FGF receptor constructs. We thank Peter Vize and Elizabeth Jones who kindly provided the antipronephros antibodies. Richard Harland, Igor Dawid, and David Kimelman generously provided plasmids. This work was supported by grants from the Beckman Foundation and the Pew Scholars in Biomedical Sciences Program. T.M. was supported by the Molecular Biosciences Training Grant to the University of WisconsinMadison.

REFERENCES Amaya, E., Musci, T. J., and Kirschner, M. W. (1991). Expression of a dominant negative mutant of the FGF receptor disrupts mesoderm formation in Xenopus embryos. Cell 66, 257–270. Amaya, E., Stein, P. A., Musci, T. J., and Kirschner, M. W. (1993). FGF signalling in the early specification of mesoderm in Xenopus. Development 118, 477– 487. Blumberg, B., Wright, C. V., De Robertis, E. M., and Cho, K. W. (1991). Organizer-specific homeobox genes in Xenopus laevis embryos. Science 253, 194 –196. Christen, B., and Slack, J. M. (1999). Spatial response to fibroblast growth factor signalling in Xenopus embryos. Development 126, 119 –125. Christian, J. L., and Moon, R. T. (1993). Interactions between Xwnt-8 and Spemann organizer signaling pathways generate dorsoventral pattern in the embryonic mesoderm of Xenopus. Genes Dev. 7, 13–28. Curran, K. L., and Grainger, R. M. (2000). Expression of activated MAP kinase in Xenopus laevis embryos: Evaluating the roles of FGF and other signaling pathways in early induction and patterning. Dev. Biol. 228, 41–56. Dale, L., Howes, G., Price, B. M., and Smith, J. C. (1992). Bone morphogenetic protein 4: a ventralizing factor in early Xenopus development. Development 115, 573–585.

Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.

304

Mitchell and Sheets

Dale, L., and Slack, J. M. (1987a). Fate map for the 32-cell stage of Xenopus laevis. Development 99, 527–551. Dale, L., and Slack, J. M. (1987b). Regional specification within the mesoderm of early embryos of Xenopus laevis. Development 100, 279 –295. de Caestecker, M. P., Parks, W. T., Frank, C. J., Castagnino, P., Bottaro, D. P., Roberts, A. B., and Lechleider, R. J. (1998). Smad2 transduces common signals from receptor serine-threonine and tyrosine kinases. Genes Dev. 12, 1587–1592. Fan, M. J., and Sokol, S. Y. (1997). A role for Siamois in Spemann organizer formation. Development 124, 2581–2589. Furthauer, M., Thisse, C., and Thisse, B. (1997). A role for FGF-8 in the dorsoventral patterning of the zebrafish gastrula. Development 124, 4253– 4264. Harland, R., and Gerhart, J. (1997). Formation and function of Spemann’s organizer [Review]. Annu. Rev. Cell Dev. Biol. 13, 611– 667. Harland, R. M. (1991). In situ hybridization: An improved wholemount method for Xenopus embryos. Methods Cell Biol. 36, 685– 695. Hemmati-Brivanlou, A., and Melton, D. A. (1992). A truncated activin receptor inhibits mesoderm induction and formation of axial structures in Xenopus embryos [see comments]. Nature 359, 609 – 614. Houzelstein, D., Cohen, A., Buckingham, M. E., and Robert, B. (1997). Insertional mutation of the mouse Msx1 homeobox gene by an nlacZ reporter gene. Mech. Dev. 65, 123–133. Isaacs, H. V., Pownall, M. E., and Slack, J. M. (1994). eFGF regulates Xbra expression during Xenopus gastrulation. EMBO J. 13, 4469 – 4481. Jernvall, J., Kettunen, P., Karavanova, I., Martin, L. B., and Thesleff, I. (1994). Evidence for the role of the enamel knot as a control center in mammalian tooth cusp formation: Non-dividing cells express growth stimulating Fgf-4 gene. Int. J. Dev. Biol. 38, 463– 469. Jernvall, J., and Thesleff, I. (2000). Reiterative signaling and patterning during mammalian tooth morphogenesis. Mech. Dev. 92, 19 –29. Johnson, R. L., and Tabin, C. J. (1997). Molecular models for vertebrate limb development. Cell 90, 979 –990. Kessler, D. S. (1997). Siamois is required for formation of Spemann’s organizer. Proc. Natl. Acad. Sci. USA 94, 13017–13022. Kimelman, D., and Griffin, K. J. (1998). Mesoderm induction: A postmodern view [comment]. Cell 94, 419 – 421. Kimelman, D., and Kirschner, M. (1987). Synergistic induction of mesoderm by FGF and TGF-beta and the identification of an mRNA coding for FGF in the early Xenopus embryo. Cell 51, 869 – 877. Kintner, C. R., and Brockes, J. P. (1984). Monoclonal antibodies identify blastemal cells derived from dedifferentiating limb regeneration. Nature 308, 67– 69. Kintner, C. R., and Brockes, J. P. (1985). Monoclonal antibodies to the cells of a regenerating limb. J. Embryol. Exp. Morphol. 89, 37–55. Kofron, M., Demel, T., Xanthos, J., Lohr, J., Sun, B., Sive, H., Osada, S., Wright, C., Wylie, C., and Heasman, J. (1999). Mesoderm induction in Xenopus is a zygotic event regulated by maternal VegT via TGFbeta growth factors. Development 126, 5759 –5770. Kretzschmar, M., Doody, J., and Massague, J. (1997). Opposing BMP and EGF signalling pathways converge on the TGF-beta family mediator Smad1. Nature 389, 618 – 622.

Kretzschmar, M., Doody, J., Timokhina, I., and Massague, J. (1999). A mechanism of repression of TGFbeta/Smad signaling by oncogenic Ras. Genes Dev. 13, 804 – 816. LaBonne, C., and Whitman, M. (1997). Localization of MAP kinase activity in early Xenopus embryos: Implications for endogenous FGF signaling. Dev. Biol. 183, 9 –20. Lane, M. C., and Sheets, M. D. (2000). Designation of the anterior/ posterior axis in pregastrula Xenopus laevis. Dev. Biol. 225, 37–58. Lane, M. C., and Smith, W. C. (1999). The origins of primitive blood in Xenopus: Implications for axial patterning. Development 126, 423– 434. Leyns, L., Bouwmeester, T., Kim, S. H., Piccolo, S., and Derobertis, E. M. (1997). Frzb-1 is a secreted antagonist of Wnt signaling expressed in the Spemann organizer. Cell 88, 747–756. Martin, G. R. (1998). The roles of FGFs in the early development of vertebrate limbs. Genes Dev. 12, 1571–1586. Moody, S. A. (1987). Fates of the blastomeres of the 32-cell-stage Xenopus embryo. Dev. Biol. 122, 300 –319. Nascone, N., and Mercola, M. (1995). An inductive role for the endoderm in Xenopus cardiogenesis. Development 121, 515–523. Neubuser, A., Peters, H., Balling, R., and Martin, G. R. (1997). Antagonistic interactions between FGF and BMP signaling pathways: A mechanism for positioning the sites of tooth formation. Cell 90, 247–255. Ng, J. K., Tamura, K., Buscher, D., and Izpisua-Belmonte, J. C. (1999). Molecular and cellular basis of pattern formation during vertebrate limb development. Curr. Top. Dev. Biol. 41, 37– 66. Nieuwkoop, P. D., and Faber, J. (1994). “Normal Table of Xenopus laevis.” Garland, New York. Niswander, L., and Martin, G. R. (1993). FGF-4 and BMP-2 have opposite effects on limb growth. Nature 361, 68 –71. Niswander, L., Tickle, C., Vogel, A., Booth, I., and Martin, G. R. (1993). FGF-4 replaces the apical ectodermal ridge and directs outgrowth and patterning of the limb. Cell 75, 579 –587. Northrop, J., Woods, A., Seger, R., Suzuki, A., Ueno, N., Krebs, E., and Kimelman, D. (1995). BMP-4 regulates the dorsal-ventral differences in FGF/MAPKK-mediated mesoderm induction in Xenopus. Dev. Biol. 172, 242–252. Pizette, S., and Niswander, L. (1999). BMPs negatively regulate structure and function of the limb apical ectodermal ridge. Development 126, 883– 894. Sasai, Y., Lu, B., Steinbeisser, H., Geissert, D., Gont, L. K., and De Robertis, E. M. (1994). Xenopus chordin: A novel dorsalizing factor activated by organizer-specific homeobox genes. Cell 79, 779 –790. Sater, A. K., and Jacobson, A. G. (1990). The role of the dorsal lip in the induction of heart mesoderm in Xenopus laevis. Development 108, 461– 470. Sater, A. K., Steinhardt, R. A., and Keller, R. (1993). Induction of neuronal differentiation by planar signals in Xenopus embryos. Dev. Dyn. 197, 268 –280. Seufert, D. W., Brennan, H. C., DeGuire, J., Jones, E. A., and Vize, P. D. (1999). Developmental basis of pronephric defects in Xenopus body plan phenotypes. Dev. Biol. 215, 233–242. Sheets, M. D., Fox, C. A., Hunt, T., Vande Woude, G., and Wickens, M. (1994). The 3⬘-untranslated regions of c-mos and cyclin mRNAs stimulate translation by regulating cytoplasmic polyadenylation. Genes Dev. 8, 926 –938. Slack, J. M., Darlington, B. G., Heath, J. K., and Godsave, S. F. (1987). Mesoderm induction in early Xenopus embryos by heparin-binding growth factors. Nature 326, 197–200.

Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.

305

FGFR Pathway Regulates Organizer Function

Smith, W. C., and Harland, R. M. (1992). Expression cloning of noggin, a new dorsalizing factor localized to the Spemann organizer in Xenopus embryos. Cell 70, 829 – 840. Szebenyi, G., and Fallon, J. F. (1999). Fibroblast growth factors as multifunctional signaling factors. Int. Rev. Cytol. 185, 45–106. Taira, M., Jamrich, M., Good, P. J., and Dawid, I. B. (1992). The LIM domain-containing homeo box gene Xlim-1 is expressed specifically in the organizer region of Xenopus gastrula embryos. Genes Dev. 6, 356 –366. Ulloa, L., Doody, J., and Massague, J. (1999). Inhibition of transforming growth factor-beta/SMAD signalling by the interferongamma/STAT pathway. Nature 397, 710 –713. Vize, P. D., Jones, E. A., and Pfister, R. (1995). Development of the Xenopus pronephric system. Dev. Biol. 171, 531–540.

von Dassow, G., Schmidt, J. E., and Kimelman, D. (1993). Induction of the Xenopus organizer: Expression and regulation of Xnot, a novel FGF and activin-regulated homeo box gene. Genes Dev. 7, 355–366. Zoltewicz, J. S., and Gerhart, J. C. (1997). The Spemann organizer of Xenopus is patterned along its anteroposterior axis at the earliest gastrula stage. Dev. Biol. 192, 482– 491.

Received for publication February 26, Revised June 20, Accepted June 22, Published online August 9,

Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved.

2001 2001 2001 2001