Toll signaling in dorsoventral patterning of Xenopus embryos

Mechanisms of Development 71 (1998) 99–105

Conserved Spa¨tzle/Toll signaling in dorsoventral patterning of Xenopus embryos Neil J. Armstrong, Herbert Steinbeisser, Christian Prothmann, Robert DeLotto, Ralph A.W. Rupp* Friedrich Miescher Laboratorium, MPG, Spemannstrasse 37–39, 72076 Tu¨bingen, Germany Received 11 December 1997; revised version received 18 December 1997; accepted 19 December 1997

Abstract The Spa¨tzle/Toll signaling pathway controls ventral axis formation in Drosophila by generating a gradient of nuclear Dorsal protein. Dorsal controls the downstream regulators dpp and sog, whose patterning functions are conserved between insects and vertebrates. Although there is no experimental evidence that the upstream events are conserved as well, we set out to ask if a vertebrate embryo can respond to maternal components of the fly Dorsal pathway. Here we demonstrate a dorsalizing activity for the heterologous Easter, Spa¨tzle and Toll proteins in UV-ventralized Xenopus embryos, which is inhibited by a co-injected dominant Cactus variant. We conclude that the Dorsal signaling pathway is a component of the conserved dorsoventral (d/v) patterning system in bilateria.  1998 Elsevier Science Ireland Ltd. Keywords: Evolutionary conserved mechanism of dorsoventral patterning; Axis induction; Xenopus; Drosophila; Easter; Spa¨tzle; Toll; Cactus; Dorsal; Rel

1. Introduction Recent findings that the structurally-related proteins Dpp/ Bmp-4 and Sog/Chd carry out conserved functions in Drosophila and Xenopus argue for a common plan for dorsoventral (d/v) patterning in bilateria (De Robertis and Sasai, 1996; Ferguson, 1996). Given the strikingly-different strategies of how the d/v axis forms in flies and in vertebrates, it has not been addressed whether conservation of d/v-patterning is restricted to these functions or includes additional components. In Drosophila zygotic expression of the dpp and sog genes is controlled by the morphogen Dorsal, which becomes locally activated through the signal-dependent degradation of its inhibitor Cactus on the ventral side of the embryo (reviewed in Morisato and Anderson, 1995). Three maternal proteins are directly involved in the generation of this ventralizing signal. These are Toll (Tl), a transmembrane receptor; its presumptive ligand Spa¨tzle (Spz), which is secreted as an inactive precursor into the perivitelline space and the extracellular serine protease Easter (Ea), * Corresponding author. Tel.: +49 7071 601830; fax: +49 7071 601455; e-mail: [email protected]

0925-4773/98/$19.00  1998 Elsevier Science Ireland Ltd. All rights reserved PII S0925-4773 (98 )0 0003-3

which by cleavage releases the mature Spa¨tzle ligand. So far, there is no evidence that vertebrates contain homologues of these proteins with conserved patterning functions. Model systems for dorsoventral axis formation in vertebrates are amphibian embryos. The development of dorsoanterior structures in Xenopus requires the translocation of maternal determinants from the vegetal pole of the egg to the prospective dorsal side of the embryo, a process which is achieved by a rotation of the cortex relative to the inner core of the egg. This cortical rotation involves the local activation of a wnt-signaling pathway and the subsequent nuclear accumulation of b-catenin. While gain- and loss-of-function experiments in the frog suggest that activation of b-catenin is both required and sufficient for complete axis formation, it is not known to what extent other pathways might contribute to this patterning process (reviewed in Miller and Moon, 1996). To test whether key functions of the Spa¨tzle/Toll signaling pathway have been conserved in vertebrates, we used the sensitive assay of mRNA microinjection into UV-ventralized Xenopus embryos (Smith and Harland, 1991; Sokol et al., 1991). Such embryos lack dorsal mesodermal and neural derivatives, because UV-irradiation of the vegetal

100

N.J. Armstrong et al. / Mechanisms of Development 71 (1998) 99–105

pole of the zygote during the first cell cycle blocks cortical rotation (Scharf and Gerhart, 1980). In this assay, the Drosophila proteins would be expected to specify the dorsal, rather than ventral, side of the Xenopus embryo, because the d/v axis appears to have been inverted between arthropods and vertebrates (Ferguson, 1996). Phenotypically, this should result in a rescue of dorsal structures.

2. Results and discussion 2.1. Drosophila Spa¨tzle rescues axis structures in UVventralized Xenopus embryos We started our investigation by injecting in vitro-transcribed Spz8.19 mRNA (Fig. 1), which encodes the fulllength Spa¨tzle protein (Morisato and Anderson, 1994). The injections were targeted subequatorially into two opposing sites of UV-irradiated embryos at the 4-cell stage, i.e. into the region with highest competence to respond to endogenous dorsal determinants (Kageura, 1997). Based on external morphological criteria, their degree of ventralization can be quantitatively assessed by the dorso anterior index (DAI, (Kao and Elinson, 1988)). Under our conditions, typically about 70% of the UV-irradiated embryos were completely ventralized, lacking all axial structures such as heads, tails and somites (Fig. 1B,I). The residual 30% show single axis rudiments (Fig. 1I). This heterogeneity is related to the experimental procedure and reflects differences in the actual UV-dose received at the vegetal pole of individual embryos. In contrast, more than 90% of untreated control embryos developed a normal morphology (Fig. 1A,I). At a dose of 500 pg Spa¨tzle (Spz) mRNA per site, over 90% of the injected embryos developed dorsal structures such as somites, tail mesenchyme and hindbrain (DAI ≥ 1; Table 1), which in the majority of the cases were confined to a single axis rudiment (Fig. 1C). More significantly, over one third of these embryos (i.e. 116 of 315 embryos; Tables 1 and 2) contained twinned axes (Fig. 1D,H), consistent with the two injection sites acting as independent dorsalizing centers. In contrast, twinning was never observed among over 6800 control- and UV-irradiated embryos, nor was it triggered by mRNA injection per se (see below and Table 1). The rescue of dorsal structures with 500 pg Spz mRNA per site was observed in 32 out of 33 embryo populations. No rescue was observed at a threefold lower dose, while higher doses did not increase the frequency or the extent of dorsal structures (data not shown). Most spz-dependent axes contained muscle and neural tissue, as well as neural crest derivatives like melanophores (Fig. 1C,D,F). Overall, induced axes exhibited quite normal trunk and tail regions, but lacked head structures anterior to the otic vesicle (Fig. 1F). In three independent experiments rare cases (n = 5/116 (4%)) were observed, in which twinned axes were almost

complete, with well-developed eye cups and lenses, cement glands and forebrain structures (Fig. 1E,H). They also contained notochords, which otherwise we have not found in cross sections (n = 15; data not shown) of the typical twinned axes (Fig. 1D). We also tested a splicing variant of Spz, spz8.24, whose open reading frame contains most of the prodomain, but terminates upstream of the presumptive ligand domain (R.DL., unpublished data). This variant caused neither the formation of twinned embryos nor a significant shift in the average DAI (Fig. 1G,I; Table 1). Together, these results unambiguously demonstrate a dorsalizing activity for Spa¨tzle in UV-ventralized Xenopus embryos, which requires the presumptive ligand domain of the Spa¨tzle protein. 2.2. The Easter protease and the Toll transmembrane receptor dorsalize UV-ventralized Xenopus embryos This dorsalizing effect of Spa¨tzle in Xenopus could be mediated by an as-yet unidentified Toll receptor orthologue, or represent the artefactual activation of an unrelated receptor(s). One way to discriminate between these two possibilities is to test whether other components of the Drosophila dorsal group also induce d/v polarity. Applying the same experimental approach described for Spz, we microinjected Ea and Tl mRNAs as two other components directly involved in generation of the ventralizing signal in Drosophila. As shown in Fig. 2A, injection of an mRNA coding for a dominant active form of Easter, EaDN (Chasan et al., 1992), also partially rescued dorsal structures and generated twinned axes at the same RNA-dose as Spz. With respect to the frequency and the average axis rescue, EaDN activity was indistinguishable from that of Spz8.19 (see DDAI in Table 1 and Table 2). In contrast to Spz, eyecups and lenses were not observed with EaDN-rescued axes. The full-length Easter protease also had some, albeit weaker, activity in this assay (DDAI = 1.03; n = 30; data not shown). In contrast, a protease-dead point mutant (EaDN(S → A), see Chasan et al., 1992) was completely inactive (Fig. 2D, Table 1), suggesting that the enzymatic activity of the Easter protease is required for the induction of dorsal tissues. Injection of synthetic Toll receptor mRNA resulted in a partial axis rescue in UV-ventralized embryos (Fig. 2B). However, the rescue by Toll was less frequent than with Ea and Spz (Table 1), and did not produce retinal pigment (data not shown). Several gain-of-function mutations of Toll have been characterized, and Drosophila embryos carrying the Tl10B allele are strongly ventralized (Schneider et al., 1991). Surprisingly, when injected, RNA encoding the Tl10B variant was slightly less active than wildtype Tl (Fig. 2C; Table 1). We currently do not know whether this reflects biochemical differences between Tl and Tl10B proteins in the heterologous environment, or differences in Toll signaling between Drosophila and Xenopus. We finally tested a truncated variant (TlDC), which lacks the intracellular domain. This truncation is expected to cause a loss of function, based

N.J. Armstrong et al. / Mechanisms of Development 71 (1998) 99–105

101

Fig. 1. Spz rescues dorsal structures in UV-embryos. For quantities of mRNA injected see Table 1A. (A) Untreated control embryos (DAI = 5). (B) Ventralized UV-siblings (DAI = 0). Typical phenotypes of spz8.19 mRNA-injected UV-ventralized embryos include: (C) single axis, DAI = 2; (D) twinned axes labeled #1 and #2, each with a DAI of 2; (E) rare, almost complete twinned axes (DAI = 4/4); (F) cross-section through a typical twinned embryo (DAI = 2/2). nt, neural tissue; m, muscle; ov, otic vesicle. Cut arranged so that one axis (#1) produces a transverse section and the second axis (#2) produces an oblique section. (G) UV-ventralized embryos injected with inactive spz8.24 mRNA (DAI = 0). (H) Histogram showing the spread of twinned axes generated by spz8.19 (axes were scored individually). (I) Distribution of the DAI scores in control and mRNA-injected embryo populations with single axes.

102

N.J. Armstrong et al. / Mechanisms of Development 71 (1998) 99–105

Table 1 Induction of single/twinned axes in UV-treated embryos by Spz, Ea and Tl RNAs injected (ng/injection site) Untreated control (N/A) UV-irradiated control (N/A) Spz8.19 (0.5) Spz8.24 (0.5) EaDN (0.5) EaDN(S → A) (1.0) Tl (1.0) Tl10B(1.0) TlDC (1.0)

Exp.

29 29 17 7 21 4 4 7 5

n

3009 2408 241 157 231 160 187 198 129

Single axes

Twinned axes

%

DAI

DDAI

%

DAI

DDAI

100 31 61 64 57 62 51 64 67

4.91 0.46 1.55 0.64 2.01 0.76 1.67 1.56 0.98

N/A N/A 1.24 0.32 1.60 0.45 1.15 0.84 0.06

0 0 36 0 37 0 16 13 5

N/A N/A 1.91 N/A 1.83 N/A 1.54 1.32 1.00

N/A N/A 1.60 N/A 1.42 N/A 1.02 0.60 0.08

exp., number of independent experiments; n, total number of embryo scores; %, percentage of total embryos with axes of DAI ≥ 1; DAI, mean DAI over all populations; DDAI, mean DAI increase of injected embryos over uninjected UV-ventralized siblings; N/A, not applicable.

on indirect evidence from the functional analysis of the structurally related domain of the human IL-1 receptor, and from the fact that similar mutations have rendered several transmembrane receptors deficient in signal transduction (Amaya et al., 1991; Heguy et al., 1992; HemmatiBrivanlou and Melton, 1992). TlDC-injected embryos were indistinguishable from uninjected UV-siblings (compare Fig. 1B with Fig. 2E; Table 1), indicating that Toll’s intracellular domain is required for the partial rescue of UVventralized embryos. 2.3. A dominant variant of the cytoplasmic Cactus protein blocks Easter and Spa¨tzle activity in the frog The ventralizing signal in Drosophila triggers the nuclear accumulation of Dorsal, which belongs to the Rel/NF-kB protein family. Although several members of this family have been isolated in Xenopus, the presently available data do not suggest that any have functions homologous to Dorsal (for details see Richardson et al., 1995; Suzuki et al., 1995; Tannahill and Wardle, 1995; Kao and Lock-

wood, 1996). Indirect evidence for the involvement of Rel proteins, however, can be acquired by asking whether the d/ v axis-rescuing activity of Drosophila proteins can be inhibited by Cactus. Recently, N- and C-terminal domains have been identified in Cactus, which are required for both its signal-dependent and constitutive degradation (Bergmann et al., 1996), but which do not overlap with a centrallylocated ankyrin repeat domain required for the interaction with Dorsal (Geisler et al., 1992; Kidd, 1992). The deletion of these sequences creates a dominant gain-of-function allele (DN100cactusDPEST), which in vivo stably retains Dorsal in the cytoplasm (Bergmann et al., 1996). The dorsalizing activity of both Spz and EaDN was strongly diminished when DN100cactusDPEST mRNA was co-injected (compare Fig. 3B,D with 3A,C; Table 2). This inhibition was a specific effect, since neither co-injection of the same amount of wildtype Cactus nor of control b-galactosidase mRNA prevented the formation of twinned axes (Table 2). Injection of DN100cactusDPEST mRNA into dorsal blastomeres of wildtype embryos did not ventralize the primary axis (Fig. 3F; Table 2), indicating that the activated

Table 2 Suppression of EaDN- and Spz8.19-dependent axis induction in UV-ventralized embryos by co-injection of DN100CactDPEST RNAs injected (ng/injected site) Untreated control (N/A) UV-irradiated control (N/A) EaDN/nlacZ (0.5/1.0) EaDN/DN100CactDPEST (0.5/1.0) Spz8.19/nlacZ (0.5/1.0) Spz8.19/DN100CactDPEST (0.5/1.0) Spz8.19/Cactus (0.5/1.0) DN100CactDPEST (1.0) DN100CactDPESTa (1.0)

Exp.

4 4 4 4 4 4 3 3 4

n

857 617 59 126 74 147 105 107 162

Single axes

Twinned axes

%

DAI

DDAI

%

DAI

DDAI

100 32 49 59 54 77 37 75 100

4.97 0.49 1.91 0.96 1.67 0.97 1.73 1.07 4.66

N/A N/A 1.42 0.48 1.34 0.58 0.94 0.68 N/A

0 0 46 10 39 1 36 0 0

N/A N/A 1.72 1.0 1.85 1.0 1.63 N/A N/A

N/A N/A 1.20 0.49 1.46 0.61 0.84 N/A N/A

exp., number of independent experiments; n, total number of embryo scores; %, percentage of total embryos with axes of DAI ≥ 1; DAI, mean DAI over all populations; DDAI, mean DAI increase of injected embryos over uninjected UV-ventralized siblings; N/A, not applicable; a, injected into both dorsal blastomeres of wildtype embryos at the 4-cell stage.

N.J. Armstrong et al. / Mechanisms of Development 71 (1998) 99–105

103

Fig. 2. Injection of Easter and Toll mRNAs into UV-ventralized embryos. For quantities of mRNA injected see Table 1. (A) partial dorsalization of UVembryos by the activated allele EaDN (DAI = 2/2). Similarly, injection of Toll mRNA was able to rescue axes up to a DAI of 2 (B), as was Tl10B (C). Comparatively there was an absence of UV-rescue in embryos injected with EaDN(S → A), (DAI = 0) (D) or TlDC (DAI = 0) (E) mRNAs. Fig. 3. Inhibition of Spz/Ea-dependent axis rescue by dominant Cactus. For quantities of mRNA injected see Table 2. (A) Group of UV-irradiated embryos injected with spz8.19. (B) Sibling embryos co-injected with spz8.19 and DN100cactDPEST mRNAs. (C) EaDN-injected UV-irradiated embryos. (D) Siblings co-injected with EaDN and DN100cactDPEST mRNAs. (E) UV-ventralized embryos injected with DN100cactDPEST alone. (F) Wildtype embryos injected dorsally with DN100cactDPEST mRNA.

Cactus variant is not generally inhibitory to the formation of dorsoanterior structures. Since members of the Cactus/ I-kB protein family are thought to bind exclusively Rellike proteins, these experiments strongly suggest that the axis rescue by Ea and Spz proteins is mediated through a common signaling pathway, which, as in Drosophila activates a Dorsal-related transcription factor. Injection of DN100cactusDPEST mRNA alone into UV-ventralized embryos caused a slight, but reproducible DAI-increase (Table 2). The reason for this is unknown.

3. Conclusions In this study, we have demonstrated a significant (P ≤ 0.005 in standard t-tests) dorsalizing activity for the maternal Drosophila proteins Easter, Spa¨tzle and Toll in UVventralized Xenopus embryos (summarized in Fig. 4). For

a number of reasons, this activity cannot be simply explained by non-specific interference with some endogenous process(es). First, consistent results were obtained with injections of eight different mRNAs at moderate dose. Second, the Drosophila proteins are structurally unrelated to each other (serine protease, cysteine-knot type ligand, transmembrane receptor) and are active in different compartments (extra/intracellular). Third, specific mutations which interfere with the activity of these proteins in Drosophila, also inactivate axis rescue in Xenopus. Finally, the epistatic relationships between upstream and downstream components of the Drosophila d/v pathway are maintained in the frog, as is evident from the inhibition of Spz and Easter activity by the dominant Cactus mutation. We note that the activity of Cactus is limited, however, since it blocked neither primary axis formation in wildtype embryos (Fig. 3F; Table 2), nor axis induction by b-Catenin in UV-ventralized embryos (data not shown). One possible explana-

104

N.J. Armstrong et al. / Mechanisms of Development 71 (1998) 99–105

Fig. 4. Overview over the cellular localization and epistatic relationships (indicated by arrows) between the four tested components of the Drosophila Dorsal signaling pathway, and their respective dorsalizing activity in UV-ventralized Xenopus embryos. Co-injection of dominant Cactus inhibits the dorsalizing activities of Easter and Spa¨tzle.

tion for this is that the relative impact of the Spz/Tl signaling pathway on d/v patterning is different between flies and frogs. Alternatively, redundancy of maternal and/ or zygotic regulatory functions may mask the Cactus phenotype in Xenopus. Collectively, these results strongly support a functional conservation between maternal components involved in d/v patterning of Drosophila and Xenopus. It will be highly informative to delineate further the conserved and diverged elements of this process by identifying both immediate target genes and homologous components of the Dorsal signaling pathway in Xenopus.

divided by the total number of embryos (see Kao and Elinson, 1988). Injected embryos were always compared with uninjected siblings which had been ventralized in the same tray. To allow for a direct comparison of the activities of different proteins we calculated the average axis rescuing activity (i.e. ‘DDAI’) by subtracting the mean DAI of control siblings from the mean DAI of injected UV-ventralized embryos. Eleven from a total of 143 UV-ventralized embryo populations were not used to calculate results, either because the uninjected controls were not satisfactorily ventralized (i.e. they had a mean DAI ≥ 1.0), or because more than 5% of control embryos showed developmental abnormalities.

4. Experimental procedures

4.2. RNA synthesis and cDNAs

4.1. Embryo manipulations and DAI-scoring

The in vitro synthesis of capped mRNAs has been described (Rupp et al., 1994). The following cDNA templates are described: pG4spz8.19, the open reading frame of Spa¨tzle (large protein splice variant (Morisato and Anderson, 1994)), was cloned into pGem4 by PCR as described (Smith et al., 1994) using the oligonucleotides 5′-GACTCTGCAGGATCCATGATGACGCCCATGTGGATA-3′ and 5′-GATCAAGCTTGGTA CCTCACCCAGTCTTCAACGCGCACT-3′; pG4spz8.24 (truncated splice variant, R.DL. unpublished data); pEa (Easter; Smith et al., 1995); pEaDN and pEADN(S → A) (Chasan et al., 1992); pToll and Toll10B (Schneider et al., 1991); TollDC, truncated at DNA-sequence position 3482 (Hashimoto et al., 1988), was generated by cleaving pToll with StuI and SmaI and religating; Cactus and DN100cactusDPEST (Bergmann et al., 1996) were PCR-subcloned into the pCS2 + plasmid (Rupp et al., 1994). All plasmid templates have

Embryos were dejellied in cysteine approximately 20 min after fertilization and placed in quartz glass-bottomed trays containing 5% Ficoll400 in 0.1× MBS (Rupp et al., 1994). The trays were suspended over the tubes of an upturned stratalinker (Stratagene). Where the embryos had not rotated or were stuck to the glass they were manually orientated so that the vegetal poles faced downwards. Embryos were irradiated with 90–95 mJoules of UV irradiation approximately 35–40 min post fertilization. The trays were not moved until the embryos reached the early 4-cell stage. UV-treated embryos were injected sub-equatorially into two opposing cells and allowed to develop until NF 37/38. Embryos were scored according to the DAI, with the mean DAI being the sum of the products of the frequency of each DAI grade and the respective grade numbers

N.J. Armstrong et al. / Mechanisms of Development 71 (1998) 99–105

been tested by in vitro translation, new PCR-subclones were verified by sequencing (N.J.A., C.P., unpublished data). Acknowledgements We thank Dr. Herbert Ja¨ckle for the gift of plasmids and Drs. Steve Cohen and Siegfried Roth for critical comments on the manuscript. References Amaya, E., Musci, T.J., Kirschner, M.W., 1991. Expression of a dominant negative mutant of the FGF receptor disrupts mesoderm formation in Xenopus embryos. Cell 66, 257–270. Bergmann, A., Stein, D., Geisler, R., Hagenmaier, S., Schmid, B., Fernandez, N., Schnell, B., Nu¨sslein-Volhard, C., 1996. A gradient of cytoplasmic Cactus degradation establishes the nuclear localization gradient of the dorsal morphogen in Drosophila. Mech. Dev. 60, 109–123. Chasan, R., Jin, Y., Anderson, K.V., 1992. Activation of the easter zymogen is regulated by five other genes to define dorsal-ventral polarity in the Drosophila embryo. Development 115, 607–616. De Robertis, E.M., Sasai, Y., 1996. A common plan for dorsoventral patterning in Bilateria. Nature 380, 37–40. Ferguson, E.L., 1996. Conservation of dorsal-ventral patterning in arthropods and chordates. Curr. Opin. Gen. Dev. 6, 424–431. Geisler, R., Bergmann, A., Hiromi, Y., Nu¨sslein-Volhard, C., 1992. cactus, a gene involved in dorsoventral pattern formation of Drosophila, is related to the IKB gene family of vertebrates. Cell 71, 613–621. Hashimoto, C., Hudson, K.L., Anderson, K.V., 1988. The Toll gene of Drosophila, required for dorsal-ventral embryonic polarity, appears to encode a transmembrane protein. Cell 52, 269–279. Heguy, A., Baldari, C.T., Macchia, G., Telford, J.L., Melli, M., 1992. Amino acids conserved in interleukin-1 receptors (IL-1Rs) and the Drosophila Toll protein are essential for IL-1R signal transduction. J. Biol. Chem. 267, 2605–2609. Hemmati-Brivanlou, A., Melton, D.A., 1992. A truncated activin receptor inhibits mesoderm induction and formation of axial structures in Xenopus embryos. Nature 359, 609–614. Kageura, H., 1997. Activation of dorsal development by contact between the cortical dorsal determinant and the equatorial core cytoplasm in eggs of Xenopus laevis. Development 124, 1543–1551. Kao, K.R., Elinson, R.P., 1988. The entire mesodermal mantle behaves as

105

Spemann’s organizer in dorsoanterior enhanced Xenopus laevis embryos. Dev. Biol. 127, 64–77. Kao, K.R., Lockwood, A., 1996. Negative regulation of dorsal patterning in early embryos by overexpression of XrelA, a Xenopus homologue of NF-kB. Mech. Dev. 58, 129–139. Kidd, S., 1992. Characterization of the Drosophila cactus locus and analysis of interactions between cactus and dorsal proteins. Cell 71, 623– 635. Miller, J.R., Moon, R.T., 1996. Signal transduction through b-catenin and specification of cell fate during embryogenesis. Genes Dev. 10, 2527– 2539. Morisato, D., Anderson, K.V., 1995. Signaling pathways that establish the dorsal-ventral pattern of the Drosophila embryo. Ann. Rev. Genet. 29, 371–399. Morisato, D., Anderson, K.V., 1994. The spa¨tzle gene encodes a component of the extracellular signaling pathway establishing the dorso-ventral pattern of the Drosophila embryo. Cell 76, 677–688. Richardson, J.C., Gatherer, D., Woodland, H.R., 1995. Developmental effects of over-expression of normal and mutated forms of a Xenopus NF-kB homologue. Mech. Dev. 52, 165–177. Rupp, R.A.W., Snider, L., Weintraub, H., 1994. Xenopus embryos regulate the nuclear localization of XMyoD. Genes Dev. 8, 1311–1323. Scharf, S., Gerhart, J.C., 1980. Determination of the dorsoventral axis in eggs of Xenopus laevis: complete rescue of UV-impaired eggs by oblique orientation before first cleavage. Dev. Biol. 79, 181–198. Schneider, D.S., Hudson, K.L., Lin, T.-Y., Anderson, K.V., 1991. Dominant and recessive mutations define functional domains of Toll, a transmembrane protein required for dorsal-ventral polarity in the Drosophila embryo. Genes Dev. 5, 797–807. Smith, C., Giordano, H., DeLotto, R., 1994. Mutational analysis of the Drosophila snake protease: an essential role for domains within the proenzyme polypeptide. Genetics 136, 1355–1365. Smith, C.L., Giordano, H., Schwartz, M., DeLotto, R., 1995. Spatial regulation of Drosophila snake protease activity in the generation of dorsalventral polarity. Development 121, 4127–4135. Smith, W.C., Harland, R.M., 1991. Injected Xwnt-8 RNA acts early in Xenopus embryos to promote formation of a vegetal dorsalizing center. Cell 67, 753–765. Sokol, S., Christian, J.L., Moon, R.T., Melton, D.A., 1991. Injected Wnt RNA induces a complete body axis in Xenopus embryos. Cell 67, 741– 752. Suzuki, K., Yamamoto, T., Inoue, J.-I., 1995. Molecular cloning of cDNA encoding the Xenopus homolog of mammalian RelB. Nucleic Acids Res. 23, 4664–4669. Tannahill, D., Wardle, F.C., 1995. Control of axis formation in Xenopus by the NF-KB-IKB system. Int. J. Dev. Biol. 39, 549–558.