Xenopus Brachyury regulates mesodermal expression of Zic3, a gene controlling left–right asymmetry


  • Tetsuya Kitaguchi,

    1. Laboratory for Developmental Neurobiology, Riken Brain Science Institute, Wako-shi, Saitama 351-0198 and
    2. The Division of Molecular Neurobiology, The Department of Basic Medical Science, The Institute of Medical Science, The University of Tokyo, Minato-ku, Tokyo 108-8639, Japan.
    Search for more papers by this author
  • Kiyomi Mizugishi,

    1. Laboratory for Developmental Neurobiology, Riken Brain Science Institute, Wako-shi, Saitama 351-0198 and
    Search for more papers by this author
  • Minoru Hatayama,

    1. Laboratory for Developmental Neurobiology, Riken Brain Science Institute, Wako-shi, Saitama 351-0198 and
    Search for more papers by this author
  • Jun Aruga,

    Corresponding author
    1. Laboratory for Developmental Neurobiology, Riken Brain Science Institute, Wako-shi, Saitama 351-0198 and
    Search for more papers by this author
  • Katsuhiko Mikoshiba

    1. Laboratory for Developmental Neurobiology, Riken Brain Science Institute, Wako-shi, Saitama 351-0198 and
    2. The Division of Molecular Neurobiology, The Department of Basic Medical Science, The Institute of Medical Science, The University of Tokyo, Minato-ku, Tokyo 108-8639, Japan.
    Search for more papers by this author

*Author to whom all correspondence should be addressed. Correspondence address: Laboratory for Developmental Neurobiology, Brain Science Institute, Riken, 2-1 Hirosawa, Wako-shi, Saitama 351-0198, Japan


The Brachyury gene has a critical role in the formation of posterior mesoderm and notochord in vertebrate development. A recent study showed that Brachyury is also responsible for the formation of the left–right (L–R) axis in mouse and zebrafish. However, the role of Brachyury in L–R axis specification is still elusive. Here, it is demonstrated that Brachyury is involved in L–R specification of the Xenopus laevis embryo and regulates expression of Zic3, which controls the L–R specification process. Overexpression of Xenopus Brachyury (Xbra) and dominant-negative type Xbra (Xbra-EnR) altered the orientation of heart and gut looping, concomitant with disturbed laterality of nodal-related 1 (Xnr1) and Pitx2 expression, both of which are normally expressed in the left lateral plate mesoderm. Furthermore, activation of inducible type Xbra (Xbra-GR) induces Zic3 expression within 20 min. These results suggest that a role of Brachyury in L–R specification may be the direct regulation of Zic3 expression.


Left–right (L–R) asymmetry of the internal organs is a consistent feature in vertebrates (Harvey 1998; Ramsdell & Yost 1998; Burdine & Schier 2000; Capdevila et al. 2000). The molecular pathway that establishes the L–R asymmetry has been studied in several species. Although a number of genes have been shown to control the L–R axis, the initial phase of L–R axial patterning is not fully understood in terms of its molecular mechanisms.

In the Xenopus laevis embryo, one group of the candidates for the initial L–R coordination is the TGF-β superfamily. This includes activinβB/Vg1 and BMP-4/BMP-2 (Hyatt et al. 1996; Hyatt & Yost 1998; Ramsdell & Yost 1999). Misexpression of Vg1/activinβB in the right blastomere can disturb the L–R pattern of the visceral organs and heart, whereas BMP-2/BMP-4 influences the L–R axis only if misexpressed in the left side. These results suggest the involvement of Vg1/activinβB and BMP-2/BMP-4 in the left- and right-sided signaling pathways, respectively. Although these factors are expressed before gastrulation, it is still unclear whether the effects of unilateral misexpression reflect their role at this stage, because the injection of activin protein into the right side of the neurula can still disturb the L–R asymmetry (Lohr et al. 1998; Toyoizumi et al. 2000).

In addition to the above hypothesis, the role of midline structures in the initial L–R patterning has been noted in multiple species (Danos & Yost 1995, 1996; Lohr et al. 1997; Pagan-Westphal & Tabin 1998). Surgical removal of the notochord in Xenopus embryos and node transplantation in chick embryos result in L–R axis disturbance. In addition, genetically defined notochord-affected mice (Brachyury, Shh, HNF3β, No turning, Ednrb and SIL mutants) and zebrafish (no tail (Brachyury), floating head and momo mutants) also showed anomalies related to L–R axis disturbance (Chen et al. 1997; Bisgrove et al. 2000; Burdine & Schier 2000; Welsh & O’Brien 2000).

Brachyury is a sequence-specific, DNA-binding, transcriptional activator required for differentiation of the posterior mesoderm and notochord (Herrmann et al. 1990; Smith 1997). Mouse and zebrafish Brachyury mutants show degeneration of the notochord and have an incidence of heart loop inversion of more than 50%. This suggests that Brachyury is involved in L–R axis establishment (Danos & Yost 1996; Chen et al. 1997; King et al. 1998; Schilling et al. 1999). However, the relationship between Brachyury and other L–R controlling genes has not been elucidated.

In a previous study, we showed that another transcription factor, Zic3, is involved in the L–R specification of the Xenopus embryo (Kitaguchi et al. 2000). Unilateral overexpression experiments suggest that Zic3 carries the left-sided signal from the initial activin-like signal to asymmetric genes such as nodal-related 1 (Xnr1) and Pitx2, which are expressed in the left lateral plate mesoderm (LPM) and function in the left specification process (Sampath et al. 1997; Ryan et al. 1998; Campione et al. 1999). Furthermore, in the early gastrulating embryo, Zic3 is expressed not only in the neuroectoderm but also in the involuting mesoderm, overlapping with Xenopus Brachyury (Xbra) expression (Kitaguchi et al. 2000).

In this study, we applied a unilateral overexpression assay to Xbra and examined its relationship to other L–R controlling genes. The results indicate that Xbra can affect L–R signaling genes irrespective of the overexpressed sides. In particular, we show that Xbra immediately induces Zic3. This regulatory cascade could provide a basis for the establishment of the L–R axis.

Materials and Methods

Whole-mount in situ hybridization and LacZ staining

Whole-mount in situ hybridization was done as described by Shain and Zuber (1996) with slight modifications. Briefly, embryos were cut before hybridization to detect mesodermal expression of Zic3 more efficiently. In addition, proteinase K treatment (10 μg/mL) was extended to 30 min to enhance the signal in the LPM. Digoxigenin-labeled probes were synthesized for Xenopus Zic3 (Nakata et al. 1997), Xnr1 (Jones et al. 1995), and Xenopus Pitx2 (Ryan et al. 1998). LacZ staining was carried out as described by Vize et al. (1991).

Embryo manipulation

Xenopus laevis were purchased from Hamamatsu Seibutsu Kyozai (Shizuoka, Japan). Embryos were obtained from human chorionic gonadotrophin (hCG)-injected adult pigmented females by in vitro fertilization. The jelly coats were removed by immersing the embryos in 1% sodium mercaptoacetate (pH 9.0) for a few minutes. Embryos were cultured in 0.1×Steinberg’s solution and staged according to Nieuwkoop and Faber (1967). Microinjection was carried out as previously described (Moon & Christian 1989). mRNA for injection was synthesized by in vitro transcription.

mRNA from Xbra, Xbra-EnR (Conlon et al. 1996) or activinβB (Sokol et al. 1991) was injected into the left or right blastomere of 4-cell stage embryos. The embryos were cultured until stage 47, and scored for heart and gut orientation. Xnr1 or Pitx2 expression was also examined at stage 25 in injected embryos. Misexpression of Xnr1 and Pitx2 was scored after the injected side was determined by LacZ staining.

For conditional activation of the Xbra protein, a hormone-inducible construct of Xbra (Xbra-GR) was injected into the dorsal blastomere of the 4-cell stage embryo (Tada et al. 1997). Injected embryos were first cultured without dexamethasone. At various stages the medium was replaced by fresh medium containing dexamethasone, in which the embryos were kept until stage 25. Misexpression of Pitx2 was scored in these embryos. Dexamethasone was given at a final concentration of 1 μM in 0.1× Steinberg’s solution.

For animal cap assays, mRNA from Xbra, Xbra-GR, Xbra-EnR or Xenopus Zic3 was injected into the animal side of both blastomeres at the 2-cell stage. Embryos were grown until stage 9 when the animal cap region was excised. The explants were cultured in 0.5× MMR (50 mM NaCl, 1 mM KCl, 1 mM CaCl2, 0.5 mM MgCl2, 2.5 mM HEPES–NaOH, pH 7.4) until the sibling embryo reached stage 25. Dexamethasone (1 μM) was added at stage 10.5 for Xbra-GR-injected explants, and the explants were cultured for 0, 20, 40 or 60 min in this medium.

RNA isolation and RT–PCR assay

Preparation of total RNA and reverse transcription (RT)– polymerase chain reaction (PCR) assays were carried out as previously described (Suzuki et al. 1994). EF1α was used to monitor RNA recovery. The sibling untreated embryo served as a positive control. PCR was also done with RNA that had not been reverse transcribed, to check for DNA contamination. Primer sequences were obtained from The Xenopus Molecular Marker Resource on the Internet (http:// vize222.zo.utexas.edu/) and from a previous paper (Kitaguchi et al. 2000). In addition, we designed the following primers for use in this study: Bix1 5′-TGGACACTCTACACTATGGC-3′, 5′-TATTGGCGGTGATGGTCTTG-3′; and Wnt11 5′-TGGAATGAGAGCGAACACTG-3′, 5′-CCAAAGAGTCCATCAGTACC-3′.


Xbra or Xbra-EnR overexpression alters heart and gut looping

To assess the involvement of Xbra in L–R axis establishment, we first examined the effect of unilateral Xbra or Xbra-EnR overexpression on heart and gut laterality. Xbra-EnR is a dominant-negative construct of Xbra, which was generated by replacing the activation domain of Xbra with a repressor domain from the Drosophila engrailed protein (Fig. 1A; Conlon et al. 1996). Xbra or Xbra-EnR mRNA was injected into the left or right side of 4-cell stage embryos. The embryos were cultured until stage 47 when the L–R laterality of the heart and gut was clearly observed. In normal Xenopus tadpoles, the heart ventricle is situated on the left side with the outflow trace looping to the right side, and the gut coils counterclockwise (Fig. 1B). Following injections of Xbra or Xbra-EnR mRNA into the left or right side of the embryo, a significant number of embryos showed situs abnormalities in the heart and gut (Fig. 1C; Table 1). However, effects on laterality by left or right side injection were not distinct, indicating that Xbra may be involved in the establishment of both left and right laterality.

Figure 1.

. Reversal of heart and gut looping in Xbra-injected embryos. (A) Xbra constructs used in this study. EnR and GR indicate the repressor derived from the Drosophila engrailed protein and the hormone-binding domain of human glucocorticoid receptor protein, respectively. (B) Wild-type Xenopus embryo (stage 47; control) with a rightward-looping heart and a counterclockwise-coiled gut, and an embryo that was injected with Xbra mRNA showing a leftward-looping heart and clockwise coiled-gut (Xbra). (C) The frequency of the reversed organs in left-sided (L) or right-sided (R) injection of Xbra (2000 pg), Xbra-EnR (1000 pg), activinβB (0.5 pg) mRNA, and uninjected (control) embryos (see also Table 1).

Table 1.  . Xbra injection induces reversed heart and gut laterality
 Xbra injectionXbra-EnR injectionActivin
  1. †The numbers of abnormal morphology of gut looping are given in parentheses.

Normal67 (7/67)100 (21/100)39 (7/39)51 (17/51)73 (9/73)84 (3/84)
Reversed heart12 (10/12)13 (10/13)3 (1/3)2 (2/2)13 (6/13)0
Reversed gut000020
Reversed heart and gut431180
Reversed organs (%)191496240

Xbra regulates the asymmetric expression of Xnr1 and Pitx2

To determine the molecular basis of the above result, we examined the expression of Xnr1 and Pitx2 in tailbud stage embryos in which Xbra or Xbra-EnR had been overexpressed. Expression patterns were determined by in situ hybridization. In most control embryos, Xnr1 and Pitx2 were expressed in the left LPM (Fig. 2A). When Xbra or Xbra-EnR was overexpressed on the right or left side, ectopic expression of both markers was observed in the right LPM (Fig. 2A). The frequency of the laterality change did not significantly differ between embryos injected on the left or the right side (Fig. 2B,C; Table 2). Thus, the change in laterality of Xnr1/Pitx2 expression correlated well with that of the heart and gut. These results showed that Xbra acts upstream of Xnr1 and Pitx2 in the establishment of the L–R axis.

Figure 2.

. Asymmetric expression of Xnr1 and Pitx2 is disturbed by Xbra or Xbra-EnR. (A) In situ hybridization of Xnr1 and Pitx2. Xnr1 and Pitx2 expression patterns were observed in the left lateral plate mesoderm (LPM) of the uninjected control embryos (control). Xbra-injected embryos show bilateral or right side expression (Xbra). (B,C) The frequency of Xnr1 and Pitx2 expression disturbed by Xbra (2000 pg), Xbra-EnR (1000 pg), activinβB (0.5 pg), and uninjected (control) embryos (see also Table 2).

Table 2.  . Xbra injection disturbs the expression of Xnr1 and Pitx2
 Xbra injectionXbra-EnR injectionActivin injection
Xnr1 expression
Altered expression (%)475235310484
Pitx2 expression
Altered expression (%)16211390340

Xbra is involved in the establishment of L–R asymmetry at early gastrula stage

Our next step was to determine at what stage Xbra acts in the regulation of L–R axis formation. To address this question, we used the Xbra-GR construct, in which the hormone-binding domain of the human glucocorticoid receptor is fused to the Xbra protein (Fig. 1A; Tada et al. 1997). In the absence of the hormone the fusion protein is inactive, while addition of dexamethasone results in rapid activation.

After injection of Xbra-GR mRNA into 4-cell stage embryos, dexamethasone was added at various stages to activate the Xbra protein. Dexamethasone-induced Xbra-GR disturbed the expression of Pitx2 on the left side at late blastula or early gastrula stage (stages 9 and 10.5), but not at late gastrula stage or later (stages 12 and 14)(Fig. 3; Table 3), indicating that Xbra specifies the L–R laterality at the early gastrula stage.

Figure 3.

. Xbra specifies left–right (L–R) laterality at the early gastrula stage. Disturbed expression of Pitx2 in Xbra-GR (400 pg)-injected embryos was observed when dexamethasone was added at several stages. The frequency of Pitx2 expression disturbed by Xbra-GR injection (see also Table 3). DEX, dexamethasone.

Table 3.  . Xbra determines the laterality of Pitx2 expression at the early gastrula stage
 Xbra-GRXbra-GRXbra-GRXbra-GRDexamethasone only
Altered expression (%)2945690

Xbra induces Zic3 expression in the embryos

We next examined the relationship between Xbra and Zic3. The expression of both genes in the mesodermal tissues is very similar at the early gastrula stage, and both Xbra and Zic3 are involved in the L–R specification at this stage of Xenopus embryo development (Kitaguchi et al. 2000).

We first injected Xbra mRNA into one blastomere of the 2-cell stage embryo and examined Zic3 expression by in situ hybridization at stage 10.5. When Xbra was overexpressed unilaterally, mesodermal expression of Zic3 was enhanced in the injected side (Fig. 4A). Conversely, Xbra-EnR overexpression resulted in suppression of Zic3 expression. The results of the animal cap explant assay were consistent with this finding. When Xbra was overexpressed in animal cap explants, Zic3 was markedly induced (Fig. 4B). The Zic3 induction by Xbra was abolished by coinjecting Xbra-EnR, which did not induce Zic3 by itself in the explants. Overexpression of Zic3 induced Xnr1 and Pitx2 but did not induce Xbra expression (Fig. 4C).

Figure 4.

. Xbra can induce Zic3 in the mesoderm. (A) Zic3 expression (purple) is shown in the cross-section of the gastrula. Animal pole is top. The embryos were cut along the broken line shown. Zic3 is expressed symmetrically in wild-type embryos (control). Zic3 expression was enhanced in the mesoderm at stage 10.5 when Xbra (2000 pg) was injected into the lateral side of one blastomere at the 2-cell stage (Xbra). Xbra-EnR suppressed the Zic3 expression at the injected side (Xbra-EnR). Co-injected LacZ expression is indicated in blue. inj, injected side; D, dorsal side; V, ventral side; bp, blastopore. (B) Reverse transcription (RT)–polymerase chain reaction (PCR) analysis of animal cap explants from embryos injected with Xbra (2000 pg) and/or Xbra-EnR (1000 pg). Zic3 was induced by Xbra whereas the expression was reduced in the presence of Xbra-EnR. (C) RT-PCR analysis of Zic3 (100 pg) overexpressed animal cap explants. Xnr1 and Pitx2 were induced by Zic3, but Xbra was not. (D) Zic3 was induced by Xbra-GR (400 pg) as early as 20 min after adding dexamethasone. This temporal profile is comparable to that of Bix1 and Wnt11 which have been shown to be regulated directly by Xbra. RT-PCR analyses were also carried out with RNA that had not been reverse transcribed (RT–), RNA from sibling embryos (Embryo) and uninjected animal cap explants (uninj).

To determine the time course of this induction, Zic3 induction by Xbra-GR was examined at multiple time points in comparison to other Xbra-inducible genes (Bix1, Wnt11; Tada et al. 1998; Tada & Smith 2000). When animal cap explants expressing Xbra-GR were cultured in medium with dexamethasone, Zic3 was induced in the Xbra-GR injected animal cap explants as early as 20 min after dexamethasone addition (Fig. 4D). This temporal expression profile is comparable to that of Bix1 and Wnt11, which have been shown to be regulated directly by Xbra (Tada et al. 1998; Tada & Smith 2000). The immediate Zic3 induction by Xbra indicates that Xbra could directly induce Zic3 in mesodermal tissues to specify the L–R laterality.


In this study, we showed that right-sided and left-sided overexpression of Xbra and Xbra-EnR equally caused a L–R pattern disruption. This disruption was different from that caused by other Xenopus L–R controlling genes, such as Vg1/activinβB (Hyatt et al. 1996; Hyatt & Yost 1998), BMP2/4 (Ramsdell & Yost 1999), antivin (lefty; Cheng et al. 2000), Zic3 (Kitaguchi et al. 2000), Xnr1 (Sampath et al. 1997) and Pitx2 (Ryan et al. 1998; Campione et al. 1999). This result indicates that Xbra does not have an exclusive role in either left- or right- specific developmental regulation.

How do the Brachyury proteins participate in the early phase of L–R axis establishment? Previous studies provide a basis for the hypothesis that disturbance of the midline structures due to the absence of Brachyury proteins affects presumptive L–R instructive signals (Danos & Yost 1996; King et al. 1998). The midline structures mentioned here include primitive streak, node, mesodermal tissue lateral to the node, and notochord, which are known to develop abnormally in the mouse and/or zebrafish mutants. In Xenopus embryos, overexpression of Xbra with Pintallavis (HNF3β) in animal cap explants causes ectopic notochord formation (O’Reilly et al. 1995), and overexpression of Xbra-EnR causes a lack of notochord (Conlon et al. 1996).

The equivalency of the left- and right-sided Xbra or Xbra-EnR injection can be rationalized if we assume that both left- and right-sided injection of Xbra or Xbra-EnR equally affected the midline structure which has a critical role in the L–R specification process. Although we did not detect apparant histological abnormalities in the notochord of the Xbra-overexpressed embryos (data not shown), Xbra probably contributes to L–R axis establishment through changing functional properties of the midline structures.

Concerning the role of the midline structures in L–R specification, the notochord may function as a physical barrier to prevent the spread of asymmetric signal (Burdine & Schier 2000; Capdevila et al. 2000). Mouse and zebrafish notochord-affected mutants have randomized heart looping and express the nodal gene symmetrically, suggesting that the role of the notochord is to maintain asymmetry in the embryo by preventing the right lateral plate from acquiring a left-sided identity. The midline barrier model may explain the L–R defects found in the Brachyury deficient mouse/ zebrafish and by Xbra injection in this study. It is also possible that the overexpressed Xbra affects the L–R axis by disturbing induction of the organizer, which determines L–R asymmetry (Nascone & Mercola 1997). The organizer may have an instructive role in L–R specification similar to that of the chick node (Pagan-Westphal & Tabin 1998).

In addition to the role of Brachyury in the development of the midline structures, this study revealed its additional role: the positive regulation of Zic3 expression. Because Xbra is known as a transcriptional activator (Conlon et al. 1996), Xbra could directly regulate Zic3 expression. This idea is also supported by the fact that Zic3 is immediately induced after Xbra-GR activation and that Zic3 and Xbra are expressed in a spatially overlapping fashion in early gastrula (Kitaguchi et al. 2000). The regulation of Zic3 expression by Xbra may have a role in L–R specification at the early gastrula stage, because activated Xbra-GR can disturb the L–R axis as late as the early gastrula stage when Zic3-GR is also effective (Kitaguchi et al. 2000).

In frog L–R patterning, Zic3 is involved in the left-sided signaling pathway (Kitaguchi et al. 2000). However, distinct results were not found between the right- and left-sided injection of Xbra. Therefore, Zic3 induction may not be the only mechanism by which Xbra controls the L–R axis. It is necessary to link Xbra/Zic3 to other L–R controlling factors for a more comprehensive understanding of the whole process.


We thank Dr C. V. E. Wright for the Xnr1 plasmid; Dr J. C. Izpisua-Belemonte for the Pitx2 plasmid; Dr D. A. Melton for the activinβB plasmid; Dr J. C. Smith and Dr M. Tada for the Xbra, Xbra-EnR and Xbra-GR plasmids; and Dr D. L. Turner for the pCS2+ plasmid.

This work was supported by Special Coordination Funds for Promoting Science and Technology, and grants from the Japanese Ministry of Education, Science, Sports and Culture to J. A. and K. M., and the Japan Society for Promotion of Science to T. K.