β-catenin signaling pathway

Mechanisms of Development 89 (1999) 93±102 www.elsevier.com/locate/modo

Relationship of vegetal cortical dorsal factors in the Xenopus egg with the Wnt/b -catenin signaling pathway Yusuke Marikawa, Richard P. Elinson* Department of Zoology, University of Toronto, 25 Harbord Street, Toronto, ON M5S 3G5, Canada Received 16 July 1999; received in revised form 16 August 1999; accepted 16 August 1999

Abstract In Xenopus, the dorsal factor in the vegetal cortical cytoplasm (VCC) of the egg is responsible for axis formation of the embryo. Previous studies have shown that VCC dorsal factor has properties similar to activators of the Wnt/b -catenin-signaling pathway. In this study, we examined the relationship of the VCC dorsal factor with components of the pathway. First, we tested whether b -catenin protein, which is known to be localized on the dorsal side of early embryos, accounts for the dorsal axis activity of VCC. Reduction of b -catenin mRNA and protein in oocytes did not diminish the activity of VCC to induce a secondary axis in recipient embryos. The amount of b -catenin protein was not enriched in VCC compared to animal cortical cytoplasm, which has no dorsal axis activity. These results indicate that b -catenin is unlikely to be the VCC dorsal axis factor. Secondly, we examined the effects of four Wnt-pathway-interfering constructs (dominant-negative Xdsh, XGSK3, Axin, and dominant-negative XTcf3) on the ability of VCC to induce expression of the early Wnt target genes, Siamois and Xnr3. The activity of VCC was inhibited by Axin and dominant negative XTcf3 but not by dominant negative Xdsh or XGSK3. We also showed that VCC decreased neither the amount nor the activity of exogenous XGSK3, suggesting that the VCC dorsal factor does not act by affecting XGSK3 directly. Finally, we tested six Wnt-pathway activating constructs (Xwnt8, Xdsh, dominant negative XGSK3, dominant negative Axin, XAPC and b -catenin) for their responses to the four Wnt-pathway-interfering constructs. We found that only XAPC exhibited the same responses as VCC; it was inhibited by Axin and dominant negative XTcf3 but not by dominant negative Xdsh or XGSK3. Although the connection between XAPC and the VCC dorsal factor is not yet clear, the fact that APC binds Axin suggests that the VCC dorsal factor could act on Axin rather than XGSK3. q 1999 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Dorsal axis factor; Xenopus; Vegetal cortical cytoplasm; Wnt/b -catenin signaling pathway; XGSK3; Axin; b -catenin

1. Introduction Cytoplasmic factors localized in eggs are responsible for the establishment of embryonic body patterns in various animals. Cortical cytoplasm around the vegetal pole of the Xenopus egg contains a factor which plays a critical role in the formation of the dorsal axis of the embryo. Transplantation of the vegetal cortical cytoplasm (VCC) of the egg into a ventral blastomere of a normal embryo results in formation of a secondary dorsal axis (Fujisue et al., 1993; Holowacz and Elinson, 1993; Kageura, 1997), and deletion of the vegetal pole region of the egg impairs development of dorsal structures (Kikkawa et al., 1996; Sakai, 1996). The dorsal factor in the VCC moves towards the future dorsal side before ®rst cleavage by a cytoplasmic movement called the cortical rotation. A parallel array of microtubules * Corresponding author. Tel.: 11-416-978-4445; fax: 11-416-9788532. E-mail address: [email protected] (R.P. Elinson)

formed in the vegetal cortex is responsible for the cortical rotation, and disturbance of the microtubule array results in ventralization of the embryo (Elinson and Rowning, 1988). The Wnt/b -catenin pathway is an evolutionary conserved signal transduction cascade, responsible for various developmental processes, such as cell proliferation and pattern formation (reviewed in Cadigan and Nusse, 1997). The secreted ligand wnt acts through the cytoplasmic protein dsh to inhibit the activity of GSK3. GSK3 with the aid of Axin phosphorylates b -catenin. Phosphorylation of b -catenin leads to its ubiquitination, followed by the proteosomemediated degradation (reviewed in Maniatis, 1999). As a consequence, wnt activation results in the stabilization of b -catenin. b -catenin binds to transcriptional enhancers of the LEF/Tcf family in the nucleus and activates the target genes. A homeobox gene Siamois and a TGFb family member Xnr3 have been identi®ed as direct targets of Wnt-activation in Xenopus early embryos (Brannon et al., 1997; McKendry et al., 1997; Fan et al., 1998). A number of studies have demonstrated that the Wnt/b -

0925-4773/99/$ - see front matter q 1999 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0925-477 3(99)00210-5

94

Y. Marikawa, R.P. Elinson / Mechanisms of Development 89 (1999) 93±102

Neither the molecular identity of VCC dorsal factor nor its relationship with components of the Wnt/b -catenin pathway are clear. However, previous studies have shown that b -catenin is necessary and suf®cient for dorsal axis development. Depletion of b -catenin mRNA in oocytes by an antisense oligo leads to development of ventralized embryos, indicating that b -catenin is necessary for the dorsal axis formation (Heasman et al., 1994). Overexpression of b -catenin is suf®cient to induce a secondary dorsal axis, the expression of Siamois, and the nuclear accumulation of b -catenin protein (Funayama et al., 1995; Brannon and Kimelman, 1996; Carnac et al., 1996). b -catenin protein is enriched on the dorsal side of early embryos (Larabell et al., 1997), and co-localizes with the cortical microtubule array on the dorsal side before the ®rst cleavage (Rowning et al., 1997). In our study, we ®rst tested the possibility that b -catenin itself is the dorsal factor in the VCC. Since our result suggested that this was not the case, we examined the relationship between the VCC dorsal factor and other components of the Wnt/b -catenin-signaling pathway.

2. Results 2.1. Depletion of b -catenin in oocytes does not diminish dorsal axis activity of the VCC Fig. 1. Transplantation of VCC from b -catenin-depleted eggs. (A) A schematic diagram showing the experimental procedure. Defolliculated full-grown oocytes were injected with antisense or sense oligo, incubated overnight, matured with progesterone and activated by pricking with a glass needle. Twenty to 30 min after the activation, VCC was transplanted into a ventral vegetal blastomere of a recipient normal embryo to test its dorsal axis activity. (B) Northern blot of RNAs from matured oocytes which were injected with antisense (4 ng and 2 ng) and sense oligos (4 ng). Histone H4 is used as a loading control. The antisense oligo depleted b -catenin mRNA from the oocytes. (C) Western blot of proteins from matured oocytes which were injected with antisense (4 ng and 2 ng) and sense oligos (4 ng). b tubulin is used as a loading control. The antisense oligo depleted b -catenin protein from the oocytes. (D) Western blot of proteins from the same volumes of VCC and ACC, collected from a single batch of eggs. In this blot, the relative amount of b -catenin protein in the VCC was 97% of that in the ACC.

To test the possibility that b -catenin itself is the dorsal factor in the VCC, we examined the dorsal axis activity of VCC from eggs which were depleted of b -catenin (Fig. 1A). Oocytes were injected with an antisense oligo of b catenin to destroy b -catenin mRNA (Heasman et al., 1994). The antisense oligo eliminated most of the b -catenin mRNA, while the control sense oligo did not (Fig. 1B). The antisense oligo also reduced the amount of b -catenin protein, while the sense oligo did not (Fig. 1C). Both the antisense and the sense oligo-injected oocytes were matured with progesterone, activated, and used as donors of VCC. VCC taken from antisense oligo-injected eggs induced a secondary axis in recipient embryos as effectively as VCC taken from sense oligo-injected eggs (Table 1). Thus, the

catenin-signaling pathway is involved in dorsal axis formation of Xenopus embryos. Reagents which activate the Wntsignaling cascade, such as Xwnt8, dsh, inhibitors of GSK3, a dominant negative Axin, and b -catenin, are able to induce a secondary axis when overexpressed in the ventral side of the embryo. Inhibition of the signaling activities of b -catenin and LEF/Tcf diminishes the formation of the dorsal axis (reviewed in Moon and Kimelman, 1998). Moreover, ectopically translocated VCC can induce the nuclear accumulation of b -catenin protein and the expression of Siamois and Xnr3, indicating that VCC dorsal factor has properties similar to activators of the Wnt/b -catenin-signaling pathway (Darras et al., 1997; Marikawa et al., 1997).

Table 1 Dorsal axis activity of the vegetal cortical cytoplasm of b -catenin-depleted eggs Expt. no.

Oligo injected a

Secondary axis (%)

1

Sense Antisense Sense Antisense Sense Antisense Sense Antisense

9/16 (56) 7/13 (54) 7/13 (54) 11/14 (79) 12/16 (75) 5/12 (42) 4/6 (67) 4/5 (80)

2 3 4 a

Each oocyte is injected with 4 ng of oligo.

Y. Marikawa, R.P. Elinson / Mechanisms of Development 89 (1999) 93±102

95

2.3. The activity of VCC is inhibited by Axin and D N-XTcf3 but not by Xdd1 or XGSK3

Fig. 2. Effects of Wnt-pathway-interfering constructs on the action of VCC. (A) A schematic diagram showing the experimental procedure. Animal poles of two blastomeres were injected at the 2-cell stage with mRNA and transplanted with VCC at the 8-cell stage. Animal caps were excised at stage 8±9, cultured until stage 10 and used for RT±PCR of Siamois, Xnr3 and EF1a . Each experiment using a single batch of embryos was comprised of two types of treatment: one was the injection of mRNA for a Wnt-pathway-interfering construct followed by transplantation with VCC, and the other was the injection of b -gal mRNA followed by transplantation with VCC. The expression levels of Siamois and Xnr3 were compared between these two. (B) RT±PCR showing the inhibition of the VCC activity by the Wnt-pathway-interfering constructs. The results are summarized in Table 3. The amounts of mRNAs injected per embryo are 1.0 ng for Xdd1, XGSK3 and Axin, and 0.5 ng for D N-XTcf3. EF1a is used as a loading control.

dorsal axis activity of VCC was not diminished by the reduction of b -catenin in the oocytes. 2.2. b -catenin protein is not enriched in the VCC We compared the amount of b -catenin protein in VCC with that in animal cortical cytoplasm (ACC), which has no dorsal axis activity (Holowacz and Elinson, 1993). Equal volumes of VCC and ACC were taken from the same batch of eggs using a calibrated microcapillary needle, and the amounts of b -catenin protein in the VCC and ACC were compared by Western blots (Fig. 1D). Three independent experiments showed that the relative amounts of b -catenin protein in the VCC to that in the ACC were 97%, 83%, and 73%, suggesting that b -catenin protein was not enriched in the VCC. This result together with the result in the previous section suggest that b -catenin is not the dorsal factor in the VCC.

To clarify the functional relationship between the VCC dorsal factor and components of the Wnt/b -catenin-signaling pathway, we examined whether the activity of VCC is suppressed by constructs that interfere with the Wnt-signaling cascade at speci®c levels. The constructs used were a dominant negative Xdsh (Xdd1; Sokol, 1996), XGSK3 (Dominguez et al., 1995; He et al., 1995; Pierce and Kimelman, 1995), Axin (Zeng et al., 1997) and a dominant negative XTcf3 (D N-XTcf3; Molenaar et al., 1996). The experimental procedure is shown in Fig. 2A. Animal poles of two blastomeres were injected at the 2-cell stage with mRNA for one of the Wnt-pathway-interfering constructs or a control b -galactosidase (b -gal), and then transplanted at the 8-cell stage with VCC which was taken from fertilized eggs. Animal caps were excised from the injected embryos when they became blastula (stage 8±9), and assayed for the expression of Siamois and Xnr3 by RT-PCR when the sibling embryos started gastrulation (stage 10). The results are shown in Fig. 2B. The expression levels of both Siamois and Xnr3 in animal caps injected with Xdd1 mRNA plus VCC appeared the same as those injected with b -gal mRNA plus VCC, indicating that Xdd1 did not interfere with the action of VCC. Similarly, pre-injection of XGSK3 mRNA did not reduce the expression levels of Siamois and Xnr3 induced by VCC. In contrast, Axin and D N-XTcf3 mRNAs signi®cantly suppressed the expression levels of Siamois and Xnr3 induced by VCC. Thus, the action of VCC was inhibited by Axin and D N-XTcf3 but not by Xdd1 or XGSK3. To con®rm that Xdd1 and XGSK3 mRNAs are working properly, we tested their inhibitory actions on Xwnt8 and Xdsh mRNAs. Siamois and Xnr3 induction by Xwnt8 and Xdsh mRNAs were signi®cantly suppressed by pre-injection of Xdd1 and XGSK3 mRNAs (Fig. 3). The suppression of Xwnt8 and Xdsh activity but not VCC activity may have been due to a higher level of activity of VCC in our test. To control for this possibility, we compared the activity of Xwnt8, Xdsh, and VCC in inducing a secondary axis, when injected into the ventral marginal zone (Table 2). In this assay, Xwnt8 and Xdsh mRNAs had stronger axis-inducing activities than VCC, indicating that the amounts of Xdd1 and XGSK3 should have been suf®cient to suppress upstream activity of the Wnt/b -catenin pathway. 2.4. VCC did not decrease the amount or the activity of XGSK3 protein Overexpression of XGSK3 was suf®cient to diminish the activities of Xwnt8 and Xdsh mRNAs to induce Siamois and Xnr3, in agreement with the idea that XGSK3 is a negative regulator of Wnt signaling and acts downstream of Xwnt8 and Xdsh (reviewed in Moon and Kimelman, 1998). However, overexpression of XGSK3 did not suppress the activity of VCC (Fig. 2). A possible explanation for this

96

Y. Marikawa, R.P. Elinson / Mechanisms of Development 89 (1999) 93±102

Fig. 3. Effects of Wnt-pathway-interfering constructs on the actions of Wnt-pathway-activating constructs. Animal poles of two blastomeres were injected at the 2-cell stage with mRNA (18) for one of the Wnt-pathway-interfering constructs (Xdd1, XGSK3, Axin and D N-XTcf3) or a control b -gal, and then injected at the 4-cell stage with mRNA (28) for one of the Wnt-pathway-activating constructs (Xwnt8, Xdsh, DN-XGSK3, D RGS-Axin, XAPC and b -catenin). Animal caps were excised from the injected embryos at stage 8±9 and assayed for the expression of Siamois and Xnr3 by RT-PCR at stage 10. Each experiment using a single batch of embryos was comprised of two types of treatment: one was the injection of mRNAs for a Wnt-pathway-interfering construct and an activating construct, and the other was the injection of mRNAs for b -gal and the same activating construct. The expression levels of Siamois and Xnr3 were compared between these two. The results are summarized in Table 3. The amounts of mRNAs injected per embryo are 1.0 ng for Xdd1, XGSK3 and Axin, 0.5 ng for D NXTcf3, 10 pg for Xwnt8, 1.0 ng for Xdsh, 0.5 ng for DN-XGSK3 and D RGS-Axin, 2.0 ng for XAPC, and 0.3 ng for b -catenin.

lack of suppression was that little XGSK3 protein was made from the exogenous RNA before the VCC was injected into the embryo. To test this, we injected XGSK3 mRNA at the 2-cell stage and measured by Western blot the amount of XGSK3 protein produced by 8-cell, 64-cell, or blastula stage. We con®rmed that a high amount of XGSK3 protein was translated from the injected mRNA (Fig. 4A). We tested further whether the injection of XGSK3 mRNA led to a high level of XGSK3 kinase activity, and this was the case (Fig. 4B). Therefore, XGSK3 failed to antagonize VCC Table 2 Dorsal axis activities of VCC and Wnt-pathway-activating constructs Activators a (amount)

VCC Xwnt8 (10 pg) Xdsh (1 ng) DN-XGSK3 (0.5 ng) DRGS-Axin (0.5 ng) XAPC (2 ng) b -catenin (0.3 ng) a

n

36 18 32 23 33 39 27

Secondary axis (%) DAI ˆ 1±2

DAI ˆ 3±4

DAI ˆ 5

53 17 38 22 36 51 22

19 28 28 43 30 13 22

6 44 16 30 15 0 41

The activators were injected into a ventral vegetal blastomere of the 16cell embryo. Dorsoanterior index (DAI) is based on Kao and Elinson (1988).

activity, even though there was a high level of both XGSK3 protein and kinase activity from the injected XGSK3 mRNA. We then tested whether the VCC acted on XGSK3 to decrease either the amount or the activity of its protein, and in that way prevented inhibition by XGSK3 mRNA. Animal poles of two blastomeres were injected at the 2cell stage with XGSK3 mRNA, and then transplanted at the 8-cell stage with VCC. Animal caps were excised from the injected embryos before MBT (stage 7±8), and assayed for the amount and activity of XGSK3 protein. Compared to the animal caps injected with XGSK3 mRNA alone, there was no signi®cant reduction in the amount of XGSK3 protein by VCC transplantation (Fig. 4C). Similarly, XGSK3 kinase activity was not reduced by VCC transplantation (Fig. 4D). These results show that the VCC dorsal factor does not act by decreasing either the amount or the activity of XGSK3. 2.5. Comparison of VCC with constructs that activate the Wnt/b -catenin pathway To compare the VCC dorsal factor with other Wnt-pathway components, we tested the activities of six Wnt-pathway activating constructs: Xwnt8 (Smith and Harland, 1991; Sokol et al., 1991), Xdsh (Sokol et al., 1995), a dominant

Y. Marikawa, R.P. Elinson / Mechanisms of Development 89 (1999) 93±102

negative XGSK3 (DN-XGSK3, Dominguez et al., 1995; He et al., 1995; Pierce and Kimelman, 1995), a dominant negative Axin (D RGS-Axin, Zeng et al., 1997), XAPC (Vleminckx et al., 1997) and b -catenin (Funayama et al., 1995) in response to the four Wnt-pathway interfering constructs in animal caps. As shown in Fig. 3 and Table 3, responses of mRNAs for Xwnt8, Xdsh, DN-XGSK3, D RGS-Axin and b -catenin to the four Wnt-pathway-interfering constructs were different from that of VCC. For example, while VCC was inhibited by Axin but not by Xdd1 or XGSK3, Xwnt8 and Xdsh were inhibited by Xdd1 and XGSK3. DN-XGSK3 and D RGS-Axin were inhibited by XGSK3, and b -catenin was not inhibited by Axin. Note that in all cases, the axis inducing activity of the activating mRNAs was greater than that of VCC (Table 2). These results suggest that the mode of action of the VCC dorsal factor is different from that of Xwnt8, Xdsh, DN-XGSK3, D RGS-Axin and b -catenin mRNAs. In contrast, XAPC mRNA was inhibited by Axin and D N-XTcf3 but not by Xdd1 or XGSK3, which is the same response to the four Wnt-pathway-interfering constructs as VCC. Although this result implies that the VCC dorsal factor may have a mode of action similar to XAPC mRNA, XAPC mRNA was not as potent as VCC or other Wnt-pathway activating constructs in inducing a secondary axis (Table 2). 3. Discussion In this study, we examined the relationship between the

97

VCC dorsal factor and components of the Wnt/b -cateninsignaling pathway. First, we showed that antisense depletion of b -catenin in oocytes did not diminish the dorsal axis activity of VCC, and that the amount of b -catenin protein was not higher in VCC than in ACC. These results suggest that b -catenin itself is not the VCC dorsal factor. Secondly, we showed that the activity of VCC was inhibited by Axin and dominant negative XTcf3 but not by dominant negative Xdsh or XGSK3, and that VCC decreased neither the amount nor the activity of XGSK3. These results suggest that VCC dorsal factor does not act on XGSK3 directly. Finally, we showed that among six Wnt-pathway activating constructs (Xwnt8, Xdsh, dominant negative XGSK3, dominant negative Axin, XAPC and b -catenin), only XAPC exhibited the same properties as VCC. b -catenin is both necessary and suf®cient for dorsal axis formation in Xenopus embryos. Moreover, b -catenin protein is localized on the dorsal side along the cortical microtubule before the ®rst cleavage (Rowning et al., 1997). One interpretation of these observations is that b catenin itself may be the dorsal axis factor localized in VCC. However, our present data suggests that this is not the case. Instead, it is likely that the VCC factor is translocated along the cortical microtubule array, and causes the accumulation of b -catenin protein on the dorsal side. Considering the timing of the dorsal accumulation of b catenin protein, VCC may start to function before ®rst cleavage. If this is the case, injection of mRNAs for inhibitory constructs into the dorsal blastomeres at the 4-cell stage may

Fig. 4. Effects of VCC on the amount and the kinase activity of XGSK3 protein. (A) Western blot for XGSK3 proteins. Embryos were injected with 1.0 ng of XGSK3 mRNA into animal poles at the 2-cell stage. Whole embryo lysates were prepared when the injected embryos reached the 8-cell stage or the 64-cell stage, and animal cap lysates (AC) were prepared at stage 7. The exogenous XGSK3 protein (arrowhead), translated from the injected mRNA, has a higher molecular weight than the endogenous XGSK3 protein (arrow) due to the myc-tag (Dominguez et al., 1995). The XGSK3 injected embryos (1) had a large amount of exogenous XGSK3, absent in the control embryos (2). (B) Kinase assay for XGSK3. Embryos were injected with 1.0 ng XGSK3 mRNA into animal poles at the 2-cell stage, and whole embryo lysates were prepared at the 8-cell stage. XGSK3 was immunoprecipitated and assayed for kinase activity using myelin basic protein as a substrate. The injected (1) embryos had more activity than the uninjected embryos (2). (C) Western blot for XGSK3 proteins. Animal poles of embryos were injected with 1.0 ng XGSK3 mRNA at the 2-cell stage and transplanted with VCC at the 8-cell stage. Animal caps were isolated at stage 7±8, and lysates were prepared. The VCC did not decrease the amount of XGSK3 protein. (D) Kinase assay for XGSK3. The same samples, used in C, were assayed for XGSK3 kinase activity. The left lane (no I.P.) stands for a mock control where the kinase reaction is carried out without the XGSK3 immunoprecipitate. The VCC did not decrease the activity of XGSK3.

98

Y. Marikawa, R.P. Elinson / Mechanisms of Development 89 (1999) 93±102

Table 3 Effects of Wnt-pathway-interfering constructs on VCC and Wnt-pathway-activating constructs a Inhibitors

Xdd1 XGSK3 Axin DN-XTcf3

Activators VCC

Xwnt8

Xdsh

DN-XGSK3

DRGS-Axin

XAPC

b -catenin

W W £ £

£ £ n.d. n.d.

£ £ n.d. n.d.

W £ £ £

W £ £ £

W W £ £

W W W £

a £ , The Siamois/Xnr3 expression level was diminished by the inhibitor; W, the Siamois/Xnr3 expression level was unaffected by the inhibitor. n.d., the experiment was not done.

not be able to antagonize the action of VCC. To avoid this problem, we performed an animal cap assay to test the effect of the inhibitory constructs on VCC. In this assay, embryos are injected with mRNAs at the 2-cell stage, followed at the 8-cell stage by transplantation of VCC taken from fertilized eggs. This procedure allows the injected mRNA to be translated into functional protein before the VCC dorsal factor is introduced. In fact, we showed that injection of XGSK3 mRNA at the 2-cell stage led to the accumulation of a high amount of active XGSK3 protein by the 8-cell stage. The current model of the Wnt/b -catenin-signaling cascade depicts Wnt activation leading to inhibition of GSK3, leading in turn to the accumulation of b -catenin (reviewed in Moon and Kimelman, 1998). In accordance with this model, the actions of Xwnt8 and Xdsh are suppressed by overexpression of XGSK3 (Dominguez et al., 1995; He et al., 1995; Pierce and Kimelman, 1995). However, we showed that overexpression of XGSK3 was not suf®cient to inhibit the action of VCC, suggesting that the VCC dorsal factor does not act by inhibiting XGSK3 activity. Several results indicate that this is the case. First, the same amount of XGSK3 mRNA was enough to suppress the actions of Xwnt8 and Xdsh mRNAs, which had higher activities of secondary axis induction than VCC. From this result, we exclude the possibility that the XGSK3 mRNA we used was non-functional or insuf®cient. Secondly, since Axin and D N-XTcf3 mRNAs suppressed the action of VCC, we exclude the possibility that exogenous mRNAs cannot act on the transplanted VCC. The inhibition by Axin mRNA is particularly important for our interpretation as GSK3, Axin, b -catenin, and other proteins form a complex (Behrens et al., 1998; Ikeda et al., 1998; Itoh et al., 1998; Nakamura et al., 1998; Yamamoto et al., 1998; Fagotto et al., 1999; Hedgepeth et al., 1999; Kishida et al., 1999). If the VCC dorsal factor that we transfer is already part of this complex, access to inhibitors may be restricted. The lack of restriction of new Axin protein to such a complex, in contrast to GSK3 or dsh, would then require an explanation. Thirdly, a high amount of active XGSK3 protein accumulated by the 8-cell stage, excluding the possibility that the injected XGSK3 mRNA was not translated enough before VCC transplantation. Finally, since VCC transplantation reduced neither the amount nor the activity

of XGSK3 protein, we exclude the possibility that the VCC dorsal factor destroyed or inactivated the overexpressed XGSK3 protein. In contrast to our results, overexpression of XGSK3 or rat GSK3 in dorsal blastomeres caused some ventralization as indicated by a reduction in head development and of goosecoid expression in Xenopus embryos (He et al., 1995; Pierce and Kimelman, 1995). These observations suggest that overexpression of GSK3 can partially suppress the action of the VCC dorsal factor, although a dorsal axis and Xnot expression were still present. An alternative interpretation of these results is that overexpression of GSK3 mRNA has effects other than inhibition of pre-MBT Wnt/b -catenin signaling, and these effects lead to reductions in head development. Consistent with this interpretation are the ®ndings that XGSK3 over-expression affects epidermal patterning (Itoh et al., 1995; Pierce and Kimelman, 1996), and injection of XGSK3 mRNA into a dorsal animal blastomere at the 8-cell stage leads to loss of an eye (Itoh et al., 1995). The activity of VCC in the presence of high XGSK3 activity tends to rule out several strong candidates for the VCC dorsal factor. The induction of Siamois and Xnr3 by Xwnt8 and Xdsh was inhibited by overexpression of XGSK3, suggesting that wnt ligands or Xdsh itself is not the VCC dorsal factor. Similar conclusions were reached by other researchers. Injection of a dominant negative Xwnt8 (Hoppler et al., 1996) and Xdd1 (Sokol, 1996) into dorsal blastomeres of 4-cell embryos does not interfere with formation of dorsoanterior structures or expression of dorsal markers. Our study using the animal cap assay supports these results. GBP has recently been discovered as a GSK3 binding protein (Yost et al., 1998). Overexpression of GBP on the ventral side can induce a secondary axis, and antisense oligo-mediated depletion of maternal GBP mRNA results in de®ciency in axis formation. These observations raise the possibility that GBP is the VCC dorsal factor. However, the VCC dorsal factor does not behave like GBP as the action of GBP can be inhibited by overexpression of XGSK3 (Yost et al., 1998). Since GBP is required for dorsal axis formation, it is possible that GBP is necessary for the function of the VCC dorsal factor. We and others demonstrated that b -TrCP, an F-box/

Y. Marikawa, R.P. Elinson / Mechanisms of Development 89 (1999) 93±102

WD40 repeat family member, is a negative regulator of the Wnt/b -catenin signaling pathway and of dorsal axis formation in Xenopus embryos (Marikawa and Elinson, 1998; Lagna et al., 1999; Liu et al., 1999). An F-box deletion construct of b -TrCP (D F) acts as a dominant negative form and leads to dorsal axis formation. Since distinct forms of b -TrCP transcripts are localized at the vegetal cortex of the oocyte (Hudson et al., 1996), we speculated that these transcripts might encode D F-like molecules and act as a dorsal axis factor. Although the identity of the vegetal cortical b -TrCP transcripts has not been clari®ed, this speculation appears not to be the case. The induction of Siamois and Xnr3 by D F in animal caps is inhibited by overexpression of XGSK3 (Marikawa and Elinson, 1998), while the action of VCC was not inhibited by XGSK3. We showed that XAPC mRNA behaved similarly to the VCC dorsal factor with regard to the response to the four Wnt-interfering constructs. Although we used XAPC mRNA as a Wnt-pathway activating construct, the role of endogenous XAPC in the Wnt/b -catenin signaling pathway is unclear (Vleminckx et al., 1997). Loss of function of APC is a cause of many colon cancers, in which the amount of b catenin protein is maintained at a high level. Introduction of full-length APC into such cancer cells decreases the level of b -catenin protein, so that APC appears to act as a negative regulator of the Wnt/b -catenin signaling pathway (reviewed in Polakis, 1999). Conversely, in Xenopus embryos, introduction of mRNA that encodes full length XAPC or human APC can induce a secondary axis, suggesting that APC acts as a positive regulator of the pathway (Vleminckx et al., 1997). This apparent contradiction may be due to the difference between the experimental systems, so that the APC protein may act either negatively or positively on the pathway depending on the cellular context. Alternatively, as suggested by Moon and Kimelman (1998), injected XAPC mRNA produces full-length as well as truncated XAPC proteins. The latter may act as a dominant negative form and interfere with the function of endogenous XAPC protein. A clear understanding of the role of XAPC in Xenopus embryos awaits further analysis, and we are reluctant to identify the VCC dorsal factor with XAPC at this point. Nonetheless, the recent ®ndings that APC binds directly to Axin raise the possibility that the mode of action of the VCC dorsal factor is similar to that of XAPC. APC binds to the RGS domain of Axin (Kishida et al., 1998; Hart et al., 1998; Nakamura et al., 1998). In Xenopus embryos, the RGS domain is critical for the function of Axin. While the full length of Axin acts as a negative regulator of the Wnt/b catenin signaling pathway, the RGS domain deletion construct of Axin (D RGS-Axin) acts as a positive regulator (Zeng et al., 1997). APC may control the action of Axin by binding to the RGS domain. If so, the action of XAPC would be inhibited by overexpression of Axin, as we found. In contrast, direct association between APC and GSK3 has not been reported, which may be the reason why the action of XAPC was not inhibited by overexpression of XGSK3. A

99

hypothesis to explain the similarity of behaviour of XAPC and the VCC dorsal factor would be that the latter also binds the RGS domain of Axin. 4. Materials and methods 4.1. Oocytes and embryos Xenopus laevis oocytes and embryos were obtained as described in Elinson and Pasceri (1989); Marikawa et al. (1997), respectively. The procedure for the antisense depletion of b -catenin mRNA and the sequences of the antisense and sense oligos were as in Heasman et al. (1994). For VCC transplantation, the oligo-treated oocytes were matured with progesterone, prick-activated and used as donors for VCC as in Holowacz and Elinson (1995). Embryonic stages were determined according to Nieuwkoop and Faber (1967). 4.2. Northern blot analysis For each experiment, total RNA was extracted from ®ve matured oocytes by the Proteinase K/SDS method (Marikawa et al., 1997). Electrophoresis, blotting and hybridization were performed as described in Sambrook et al. (1989) using one oocyte equivalent total RNA per lane. Radioactive probes were synthesized by random priming of the excised inserts of b -catenin (XbaI/XhoI; Funayama et al., 1995) and Histone H4 (BamHI; Turner and Woodland, 1982). 4.3. Western blot analysis For b -catenin detection, protein extracts were prepared by lysing oocytes in NP-40 buffer (0.5% NP-40, 20% glycerol, 150 mM NaCl, 2 mM EDTA, 1 mM DTT and 50 mM Tris, pH 7.5). Homogenates were cleared by centrifugation and the supernatants were combined with an equal volume of 2 £ SDS buffer (4% SDS, 20% glycerol, 10% b mercaptoethanol and 120 mM Tris, pH 6.8). For each experiment, VCC and ACC were collected from 20 fertilized eggs of a single batch using a calibrated microcapillary needle, transferred into NP-40 buffer, and mixed with an equal volume of 2 £ SDS buffer. One oocyte or ten cortical cytoplasm equivalents was loaded per lane. For GSK3 detection, protein extracts were prepared by lysing embryos and animal caps in 0.5% NP-40, 150 mM NaCl, 2 mM EDTA, 0.6 mM PMSF and 50 mM Tris, pH 7.5. Lysates were cleared by centrifugation, and the supernatants were combined with an equal volume of the 2£ SDS buffer. A quarter embryo or one animal cap equivalent was loaded per lane. After separation in a 10% SDS-polyacrylamide gel, samples were transferred onto PVDF membrane (Amersham) at 25 V for 1.5 h. The membrane was blocked with 5% skimmed milk for 1 h and incubated overnight with antib -catenin (Transduction lab), anti-GSK3 (Transduction lab)

100

Y. Marikawa, R.P. Elinson / Mechanisms of Development 89 (1999) 93±102

or anti-b -tubulin (N357, Amersham) monoclonal antibody. After washing with Tris-buffered saline plus 0.1% Tween 20, the membrane was incubated with HRP-conjugated secondary antibody (Jackson Immuno Research Labs) for 1 h. HRP activity was visualized using the ECL detection kit (Amersham) and quanti®ed using a scanner (ScanMaker III, Microtek) linked to a Macintosh computer running the NIH Image software. Each experiment was repeated at least three times. 4.4. Synthetic mRNAs Capped mRNAs were synthesized by in vitro transcription and injected into embryos as described in Moon and Christian (1989). The following mRNAs were prepared: Xwnt8 (Sokol et al., 1991), Xdsh (Sokol et al., 1995), Xdd1 (Sokol, 1996), XGSK3, DN-XGSK3 (Dominguez et al., 1995), Axin, D RGS-Axin (Zeng et al., 1997), XAPC (Vleminckx et al., 1997), b -catenin (Funayama et al., 1995), D N-XTcf3 (Molenaar et al., 1996) and b -gal (Smith and Harland, 1991). Synthesized mRNAs were treated with RNase-free DNase, extracted with phenol-chloroform, precipitated twice in 2.5 M LiCl and 25 mM EDTA, extensively washed in 70% ethanol, and dissolved in DEPCwater at desired concentrations. Microinjection was done with the aid of Nanoject automatic oocyte injector (Drummond). 4.5. RT±PCR Reverse transcription±polymerase chain reaction (RT± PCR) was performed as described previously (Marikawa and Elinson, 1998). For each sample, total RNA was extracted from eight to ten animal caps. Oligo dT-primed ®rst strand cDNA was prepared from 1 mg of total RNA with an M-MLV RT (Gibco BRL) in a 30-ml reaction. 20 ml PCR reaction was carried out using Taq DNA polymerase (1 unit) and PCR buffer (Gibco BRL) with 1.5 mM MgCl2, 200 mM dNTPs, 2 mCi [a - 32P]dCTP, 40 ng of each primer and 2 ml of the ®rst strand cDNA. The PCR conditions were initial denaturation at 948C for 5 min, 23 cycles of 948C for 50 s (denaturation), 558C for 50 s (annealing) and 728C for 50 s (extension), and the ®nal extension at 728C for 10 min. Under these conditions, the signals obtained for Siamois, Xnr3 and EF1a were linear to the amount of the targets as checked by serial dilution of total RNAs and cDNAs. Control reactions in which M-MLV RT was omitted yielded no signals. Each experiment was repeated at least three times using different batches of embryos. Observations were made at different exposure times to ensure that we were not comparing saturated signals. The primer sequences were as described: Siamois (Brannon and Kimelman, 1996), Xnr3 (Yang-Snyder et al., 1996) and EF1a (HemmatiBrivanlou and Melton, 1994).

4.6. XGSK3 kinase assay The XGSK3 kinase assay was carried out according to Itoh et al. (1998) with slight modi®cations. Five embryos or ten animal caps were lysed in 250 ml or 100 ml lysis buffer (0.5% NP40, 50 mM NaCl, 1 mM EDTA, 0.6 mM PMSF, 10 mM Tris, pH 7.5), respectively. Lysates were cleared by centrifugation at 13 000 £ g for 8 min. AntiGSK3 antibody (0.25 mg, Transduction lab) was incubated with 30 ml cleared lysate overnight at room temperature. Protein G-agarose (Sigma) was added and incubated for an additional 3 h. The immunoprecipitates were washed twice in the lysis buffer and twice in 10 mM MgCl2 and 30 mM Tris (pH 7.5). Half of the immunoprecipitate was incubated in 10 mM MgCl2, 1 mM DTT, 5% glycerol, 7.5 mM ATP, 5 mCi [g - 32P]ATP, 30 mM Tris, pH 7.5, and 0.5 mg/ml myelin basic protein (Sigma) as a GSK3 substrate. After incubation for 30 min at 308C, the sample was mixed with an equal volume of 2£ SDS buffer and separated in a 15% polyacrylamide gel. The gel was dried and exposed to an X-ray ®lm. Each experiment was repeated at least three times.

Acknowledgements We thank F. Costantini for the Axin and D RGS-Axin plasmids, O. DestreÂe for D N-XTcf3 plasmid, R.T. Moon for the Xwnt8 plasmid, B.M. Gumbiner for the b -catenin and XAPC plasmids, and S.Y. Sokol for the XGSK3, DNXGSK3, Xdsh and Xdd1 plasmids. We also thank K. Itoh for advice on preparation of synthetic mRNAs and procedure of GSK3 kinase assay. Y. Li helped us with b -catenin Northern and Western blottings. We are grateful to P. Klein, Y. Masui and members of the Elinson lab for helpful comments. The procedures used for experimentation with Xenopus laevis were approved by the Animal Care Committee of the University of Toronto. Y.M. is supported by a postdoctoral fellowship from the Japan Society for the Promotion of Science. This work is funded by a grant to R.P. E. from Medical Research Council of Canada.

References Behrens, J., Jerchow, B.A., Wurtele, M., Grimm, J., Asbrand, C., Wirtz, R., KuÈhl, M., Wedlich, D., Birchmeier, W., 1998. Functional interaction of an axin homolog, conductin, with b -catenin, APC, and GSK3b . Science 280, 596±599. Brannon, M., Kimelman, D., 1996. Activation of Siamois by the Wnt pathway. Dev. Biol. 180, 344±347. Brannon, M., Gomperts, M., Sumoy, L., Moon, R.T., Kimelman, D., 1997. A b -catenin/XTcf-3 complex binds to the siamois promoter to regulate dorsal axis speci®cation in Xenopus. Genes Dev. 11, 2359±2370. Cadigan, K.M., Nusse, R., 1997. Wnt signaling: a common theme in animal development. Genes Dev. 11, 3286±3305. Carnac, G., Kodjabachian, L., Gurdon, J.B., Lemaire, P., 1996. The homeobox gene Siamois is a target of the Wnt dorsalisation pathway and

Y. Marikawa, R.P. Elinson / Mechanisms of Development 89 (1999) 93±102 triggers organiser activity in the absence of mesoderm. Development 122, 3055±3065. Darras, S., Marikawa, Y., Elinson, R.P., Lemaire, P., 1997. Animal and vegetal pole cells of early Xenopus embryos respond differently to maternal dorsal determinants: implication for the patterning of the organiser. Development 124, 4275±4286. Dominguez, I., Itoh, K., Sokol, S.Y., 1995. Role of glycogen synthase kinase 3b as a negative regulator of dorsoventral axis formation in Xenopus embryos. Proc. Natl. Acad. Sci. USA 92, 8498±8502. Elinson, R.P., Rowning, B., 1988. A transient array of parallel microtubules in frog eggs: potential tracks for a cytoplasmic rotation that speci®es the dorso-ventral axis. Dev. Biol. 128, 185±197. Elinson, R.P., Pasceri, P., 1989. Two UV-sensitive targets in dorsoanterior speci®cation of frog embryos. Development 106, 511±518. Fagotto, F., Jho, E.-H., Zeng, L., Kurth, T., Joos, T., Kaufmann, C., Costantini, F., 1999. Domains of axin involved in protein±protein interactions, Wnt pathway inhibition, and intracellular localization. J. Cell Biol. 145, 741±756. Fan, M.J., Gruning, W., Walz, G., Sokol, S.Y., 1998. Wnt signaling and transcriptional control of Siamois in Xenopus embryos. Proc. Natl. Acad. Sci. USA 95, 5626±5631. Fujisue, M., Kobayakawa, Y., Yamana, K., 1993. Occurrence of dorsal axis-inducing activity around the vegetal pole of an uncleaved Xenopus egg and displacement to the equatorial region by cortical rotation. Development 118, 163±170. Funayama, N., Fagotto, F., McCrea, P., Gumbiner, B.M., 1995. Embryonic axis induction by the armadillo repeat domain of b -catenin: evidence for intracellular signaling. J. Cell Biol. 128, 959±968. Hart, M.J., de los Santos, R., Albert, I.N., Rubinfeld, B., Polakis, P., 1998. Downregulation of b -catenin by human Axin and its association with the APC tumor suppressor, b -catenin and GSK3b . Curr. Biol. 8, 573± 581. He, X., Saint-Jeannet, J.P., Woodgett, J.R., Varmus, H.E., Dawid, I.B., 1995. Glycogen synthase kinase-3 and dorsoventral patterning in Xenopus embryos. Nature 374, 617±622. Heasman, J., Craeford, A., Goldstone, K., Garner-Hamrick, P., Gumbiner, B., McCrea, P., Kintner, C., Noro, C.Y., Wylie, C., 1994. Overexpression of cadherin and underexpression of b -catenin inhibit dorsal mesoderm induction in early Xenopus embryos. Cell 79, 791±803. Hedgepeth, C.M., Deardorff, M.A., Klein, P.S., 1999. Xenopus axin interacts with glycogen synthase kinase-3 b and is expressed in the anterior midbrain. Mech. Dev. 80, 147±151. Hemmati-Brivanlou, A., Melton, D.A., 1994. Inhibition of activin receptor signaling promotes neuralization in Xenopus. Cell 77, 273±281. Holowacz, T., Elinson, R.P., 1993. Cortical cytoplasm, which induces dorsal axis formation in Xenopus, is inactivated by UV irradiation of the oocyte. Development 119, 277±285. Holowacz, T., Elinson, R.P., 1995. Properties of the dorsal activity found in the vegetal cortical cytoplasm of Xenopus eggs. Development 121, 2789±2798. Hoppler, S., Brown, J.D., Moon, R.T., 1996. Expression of a dominantnegative Wnt blocks induction of MyoD in Xenopus embryos. Genes Dev. 10, 2805±2817. Hudson, J.W., AlarcoÂn, V.B., Elinson, R.P., 1996. Identi®cation of new localized RNAs in the Xenopus oocyte by differential display PCR. Dev. Genet. 19, 190±198. Ikeda, S., Kishida, S., Yamamoto, H., Murai, H., Koyama, S., Kikuchi, A., 1998. Axin, a negative regulator of the Wnt signaling pathway, forms a complex with GSK-3b and b -catenin and promotes GSK-3b -dependent phosphorylation of b -catenin. EMBO J. 17, 1371±1384. Itoh, K., Tang, T.L., Neel, B.G., Sokol, S.Y., 1995. Speci®c modulation of ectodermal cell fates in Xenopus embryos by glycogen synthase kinase. Development 121, 3979±3988. Itoh, K., Krupnik, V.E., Sokol, S.Y., 1998. Axis determination in Xenopus involves biochemical interactions of axin, glycogen kinase 3 and b catenin. Curr. Biol. 8, 591±594. Kageura, H., 1997. Activation of dorsal development by contact between

101

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 Spemann's organizer in dorsoanterior enhanced Xenopus laevis embryos. Dev. Biol. 127, 64±77. Kikkawa, M., Takano, K., Shinagawa, A., 1996. Location and behavior of dorsal determinants during ®rst cell cycle in Xenopus eggs. Development 122, 3687±3696. Kishida, S., Yamamoto, H., Ikeda, S., Kishida, M., Sakamoto, I., Koyama, S., Kikuchi, A., 1998. Axin, a negative regulator of the wnt signaling pathway, directly interacts with adenomatous polyposis coli and regulates the stabilization of b -catenin. J. Biol. Chem. 273, 10823±10826. Kishida, S., Yamamoto, H., Hino, S., Ikeda, S., Kishida, M., Kikuchi, A., 1999. DIX domains of Dvl and axin are necessary for protein interactions and their ability to regulate b -catenin stability. Mol. Cell. Biol. 19, 4414±4422. Lagna, G., Carnevali, F., Marchiani, M., Hemmati-Brivanlou, A., 1999. Negative regulation of axis formation and Wnt signaling in Xenopus embryos by the F-box/WD40 protein b -TrCP. Mech. Dev. 80, 101±106. Larabell, C.A., Torres, M., Rowning, B.A., Tost, C., Miller, J.R., Wu, M., Kimelman, D., Moon, R.T., 1997. Establishment of the dorso-ventral axis in Xenopus embryos is presaged by early asymmetries in b -catenin that are modulated by the Wnt signaling pathway. . J. Cell Biol. 136, 1123±1136. Liu, C., Kato, Y., Zhang, Z., Do, V.M., Yankner, B.A., He, X., 1999. b Trcp couples b -catenin phosphorylation-degradation and regulates Xenopus axis formation. Proc. Natl. Acad. Sci. USA 96, 6273±6278. Maniatis, T., 1999. A ubiquitin ligase complex essential for the NF-k B, Wnt/Wingless, and Hedgehog signaling pathways. Genes Dev. 13, 505± 510. Marikawa, Y., Elinson, R.P., 1998. b -TrCP is a negative regulator of Wnt/ b -catenin signaling pathway and dorsal axis formation in Xenopus embryos. Mech. Dev. 77, 75±80. Marikawa, Y., Li, Y., Elinson, R.P., 1997. Dorsal determinants in the Xenopus egg are ®rmly associated with the vegetal cortex and behave like activators of the Wnt pathway. Dev. Biol. 191, 69±79. McKendry, R., Hsu, S.C., Harland, R.M., Grosschedl, R., 1997. LEF-1/TCF proteins mediate wnt-inducible transcription from the Xenopus nodalrelated 3 promoter. Dev. Biol. 192, 420±431. Molenaar, M., van de Wetering, M., Oosterwegel, M., Peterson-Maduro, J., Godsave, S., Korinek, V., Roose, J., DestreÂe, O., Clevers, H., 1996. XTcf-3 transcription factor mediates b -catenin-induced axis formation in Xenopus embryos. Cell 86, 391±399. Moon, R.T., Christian, J.L., 1989. Microinjection and expression of synthetic mRNAs in Xenopus embryos. Technique 1, 76±89. Moon, R.T., Kimelman, D., 1998. From cortical rotation to organizer gene expression: toward a molecular explanation of axis speci®cation in Xenopus. Bioessays 20, 536±545. Nakamura, T., Hamada, F., Ishidate, T., Anai, K., Kawahara, K., Toyoshima, K., Akiyama, T., 1998. Axin, an inhibitor of the Wnt signalling pathway, interacts with b -catenin, GSK-3b and APC and reduces the b -catenin level. Genes Cells 6, 395±403. Nieuwkoop, P.D., Faber, J., 1967. Normal Table of Xenopus laevis (Daudin), North Holland, Amsterdam. Pierce, S.B., Kimelman, D., 1995. Regulation of Spemann organizer formation by the intracellular kinase Xgsk-3. Development 121, 755±765. Pierce, S.B., Kimelman, D., 1996. Overexpression of Xgsk-3 disrupts anterior ectodermal patterning in Xenopus. Dev. Biol. 175, 256±264. Polakis, P., 1999. The oncogenic activation of b -catenin. Curr. Opin. Genet. Dev. 9, 15±21. Rowning, B.A., Wells, J., Wu, M., Gerhart, J.C., Moon, R.T., Larabell, C.A., 1997. Microtubule-mediated transport of organelles and localization of b -catenin to the future dorsal side of Xenopus eggs. Proc. Natl. Acad. Sci. USA 94, 1124±1129. Sakai, M., 1996. The vegetal determinants required for the Spemann organizer move equatorially during the ®rst cell cycle. Development 122, 2207±2214.

102

Y. Marikawa, R.P. Elinson / Mechanisms of Development 89 (1999) 93±102

Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning: A Laboratory Manual, 2. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. 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, 829±840. 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. Sokol, S.Y., Klingensmith, J., Perrimon, N., Itoh, K., 1995. Dorsalizing and neuralizing properties of Xdsh, a maternally expressed Xenopus homolog of dishevelled. Development 121, 1637±1647. Sokol, S.Y., 1996. Analysis of Dishevelled signalling pathway during Xenopus development. Curr. Biol. 6, 1456±1467. Turner, P.C., Woodland, H.R., 1982. H3 and H4 histone cDNA sequences from Xenopus: a sequence comparison of H4 genes. Nucleic Acids Res. 10, 3769±3780. Vleminckx, K., Wong, E., Guger, K., Rubinfeld, B., Polakis, P., Gumbiner,

B.M., 1997. Adenomatous polyposis coli tumor suppressor protein has signaling activity in Xenopus laevis embryos resulting in the induction of an ectopic dorsoanterior axis. J. Cell Biol. 136, 411±420. Yamamoto, H., Kishida, S., Uochi, T., Ikeda, S., Koyama, S., Asashima, M., Kikuchi, A., 1998. Axil, a member of the Axin family, interacts with both glycogen synthase kinase 3b and b -catenin and inhibits axis formation of Xenopus embryos. Mol. Cell. Biol. 18, 2867±2875. Yang-Snyder, J., Miller, J.R., Brown, J.D., Lai, C.J., Moon, R.T., 1996. A frizzled homolog functions in a vertebrate Wnt signaling pathway. Curr. Biol. 6, 1302±1306. Yost, C., Farr III, G.H., Pierce, S.B., Ferkey, D.M., Chen, M.M., Kimelman, D., 1998. GBP, an inhibitor of GSK-3, is implicated in Xenopus development and oncogenesis. Cell 93, 1031±1041. Zeng, L., Fagotto, F., Zhang, T., Hsu, W., Vasicek, T.J., Perry III, W.L., Lee, J.J., Tilghman, S.M., Gumbiner, B.M., Costantini, F., 1997. The mouse Fused locus encodes Axin, an inhibitor of the Wnt signaling pathway that regulates embryonic axis formation. Cell 90, 181± 192.