A POU protein regulates mesodermal competence to FGF in Xenopus

Mechanisms of Development 71 (1998) 131–142

A POU protein regulates mesodermal competence to FGF in Xenopus Clara Henig, Sarah Elias, Dale Frank* Department of Biochemistry, The Rappaport Family Institute for Research in the Medical Sciences, Faculty of Medicine, Technion-Israel Institute of Technology, Haifa 31096, Israel Received 10 November 1997; revised version received 6 December 1997; accepted 22 December 1997

Abstract XLPOU91, a POU-homeobox gene is expressed in a narrow window during early Xenopus development. We show that ectopic expression of XLPOU91 RNA causes severe posterior truncations in embryos without inhibiting the formation of Spemann’s organizer. Ectopic XLPOU91 expression also inhibits mesoderm induction by fibroblast growth factor (FGF) and activin in animal cap explants. Using antisense RNA, we depleted endogenous XLPOU91 protein in animal caps. Gastrula-stage animal caps expressing XLPOU91 antisense RNA do not lose competence to FGF, unlike controls, these animal caps express XBra after FGF treatment. Endogenous XLPOU91 levels are peaking when FGF mesoderm-inducing competence is lost in animal caps. Thus XLPOU91 protein may act as a competence switch during early development, as XLPOU91 levels increase in the embryo, the mesoderm response to FGF is lost.  1998 Elsevier Science Ireland Ltd. Keywords: XLPOU91; Xenopus laevis; Basic fibroblast growth factor (bFGF); Mesoderm induction; Competence; Animal caps; XBra expression

1. Introduction The development of vertebrate embryos requires several inductive interactions to form the three-dimensional body plan. In Xenopus laevis embryos, mesoderm forms by a multi-stepped inductive process, which eventually forms such tissues as notochord, muscle, heart, mesenchyme and blood (Kimelman et al., 1992; Sive, 1993; Kessler and Melton, 1994). Mesoderm is induced in the equatorial region (marginal zone) by underlying vegetal pole cells (Nieuwkoop, 1969; Sudwarti and Nieuwkoop, 1971). The most dorsal mesoderm (Spemann organizer) is induced in the equatorial region by underlying dorsal vegetal cells (Nieuwkoop center). Ventral mesoderm (blood, mesenchyme) is induced in the ventral marginal zone by underlying ventral vegetal cells. There are several candidate molecules for endogenous mesoderm inducers (Kimelman and Kirschner, 1987; Green et al., 1990; Sokol et al., 1990; Dale et al., 1992; Jones et al., 1992; Isaacs et al., 1994) including mem-

* Corresponding author. Tel.: +972 48 295286; fax: +972 48 535773; email: [email protected]

0925-4773/98/$19.00  1998 Elsevier Science Ireland Ltd. All rights reserved PII S0925-4773 (98 )0 0006-9

bers of the FGF (basic fibroblast growth factor, bFGF, and eFGF), and TGF-b families (activin/Vg1, BMP-4). The use of dominant-negative receptors for FGF, activin and BMP-4 suggests that these molecules are essential for proper mesoderm patterning (Amaya et al., 1991; Hemmati-Brivanlou and Melton, 1992; Graff et al., 1994). These inducing molecules modulate cell fate throughout the activation of regionspecific transcription factors. FGF activates expression of the XBra transcription factor, which is necessary for the formation of posterior mesoderm such as notochord and muscle (Smith et al., 1991; Cunliffe and Smith, 1992; Isaacs et al., 1994; Conlon et al., 1996). BMP-4 probably regulates ventral mesoderm formation by activating the Xvent-1, Xvent-2, and Mix.1 transcription factors (Gawantka et al., 1995; Mead et al., 1996; Onichtchouk et al., 1996; Papalopulu and Kintner, 1996; Schmidt et al., 1996). Activin/Vg1 may regulate formation of dorsal mesoderm by activating organizer-specific transcription factors such as gsc, Xlim-1 and XFKH-1 (Cho et al., 1991; Dirksen and Jamrich, 1992; Green et al., 1992; Taira et al., 1992). We focused our studies on understanding the role of POU-domain genes during early vertebrate development. POU genes belong to a large family of transcription factors

132

C. Henig et al. / Mechanisms of Development 71 (1998) 131–142

which contain a diverged-homeobox and a highly-conserved POU-specific box of 76–78 amino acids. These genes regulate cell-fate decisions in many invertebrate and vertebrate systems. POU genes have been implicated in regulating differentiation of cells in the central nervous system, brain, pituitary gland and hemapoetic systems (revewied by Ryan and Rosenfeld, 1997). We previously cloned the Xenopus POU-gene, XLPOU91 from a gastrula stage cDNA library (Frank and Harland, 1992). This gene is activated in a cell-autonomous manner in all embryonic cells at the mid-blastula transition. Transcript levels reach a transient peak at late gastrula stages and the gene undergoes an inductive-dependent transcriptional shutdown in the dorsoanterior regions of the embryo. By late neurula stages, XLPOU91 expression is reduced to the posterior embryo regions fated for tailbud. This gene is most similar to the mammalian class V POU-domain gene, Oct3/4 (74% in the POU-domain, 58% in the homeo domain). Like XLPOU91, Oct 3/4 is expressed in early embryonic cells and its expression is downregulated in the embryo at about the time of gastrulation (Okamoto et al., 1990; Rosner et al., 1990). While the direct biological role of Oct 3/4 is still unclear, it has been suggested that this gene may prevent differentiation, and maintain cells in a totipotential state during early development (Okamoto et al., 1990; Rosner et al., 1990; Scholer et al., 1990). We examined the role of XLPOU91 protein during early development. Ectopic expression of XLPOU91 caused severe posterior truncations in whole embryos. These embryos expressed reduced levels of the pan-mesodermal XBra gene, but normal levels of the Spemann organizerspecific gsc and Xlim-1 genes. The similarity of this phenotype to embryos injected with the dominant-negative FGF receptor (Amaya et al., 1991, 1993; Isaacs et al., 1994), suggested that XLPOU91 modulates endogenous FGF signaling within the embryo. Therefore, we examined the effects of ectopic XLPOU91 expression on FGF mesoderm induction in animal cap explants. Animal cap explants failed to differentiate as mesoderm and to express the XBra transcription factor when overexpressing XLPOU91 protein. FGF induces mesoderm through activation of the MAP kinase (MAPK) cascade which in turn activates XBra expression (Graves et al., 1994; Hartley et al., 1994; LaBonne and Whitman, 1994; Gotoh et al., 1995; LaBonne et al., 1995; Umbhauer et al., 1995). XLPOU91 did not inhibit FGF activation-phosphorylation of MAPK in animal caps, suggesting it acts downstream of this cascade. Using XLPOU91 antisense RNA, we examined the role of endogenous XLPOU91 expression in the animal cap induction assay. Stage 10.5 animal caps depleted of XLPOU91 protein respond to bFGF, in contrast to uninjected controls. These results suggest that XLPOU91 may act as a mesodermal competence switch during development. As XLPOU91 levels transiently increase during gastrulation, cells lose their responsiveness to bFGF, which may be important for regulating the timing of mesoderm induction.

2. Results 2.1. Ectopic XLPOU91 expression perturbs posterior axis formation We injected 2 ng of in vitro-synthesized XLPOU91 RNA into 1-cell fertilized Xenopus embryos. Embryos were scored at tailbud to tadpole stages for phenotypes. Overexpression of XLPOU91 caused posterior truncations in 80% of the injected embryos; about 75% of these perturbed embryos have severe posterior truncations (Fig. 1A; Table 1). In addition, we also injected RNA encoding a mutant form of XLPOU91, called XLPOU91*. XLPOU91* has an altered homeobox region (see Section 4), which binds DNA about three-fold less efficiently than wild type protein (Fig. 2). Ectopic XLPOU91* expression caused a less extreme gain-of-function phenotype in comparison with wild type XLPOU91. While 80% of the embryos still have posterior truncations, only 25% are severely truncated, while 75% of the embryos have moderate posterior truncations (Fig. 1B; Table 1). We could routinely monitor DNA binding of XLPOU91 and XLPOU91* in a DNA electrophoretic mobility shift assay (EMSA) to an octamer motif (OCT) oligonucleotide (Fig. 2). Protein was isolated for EMSA from injected and control embryos at stages 11–12. Ectopic XLPOU91 expression routinely increased the EMSA about 7–15 fold in comparison with uninjected embryos (Fig. 2). Ectopic XLPOU91* expression stimulated the EMSA by only 3–5 fold (Fig. 2). As negative controls, embryos were injected with a truncated version of XLPOU91, DNsi-XLPOU91, which lacks 164 amino acids spanning most of the POUhomeobox region (see Section 4). DNsi-XLPOU91-injected embryos had normal phenotypes (Fig. 1C). The DNsiXLPOU91 protein did not bind the OCT sequence and thus did not alter the EMSA in extracts from injected embryos (not shown). Therefore, the penetrance of the posterior truncation phenotype is dependent on the DNA binding activity of the XLPOU91 protein. To further characterize these embryos, gastrula stage mesoderm markers were examined by Northern analysis. Ectopic expression of XLPOU91 reduced transcription of the pan-mesodermal marker XBra (Smith et al., 1991) by 3– 5 fold versus control embryos (Fig. 3A). Expression of the ventral marker, XPO (Amaya et al., 1993) was also reduced in gastrula stage embryos (not shown). In contrast, the Spemann organizer markers gsc and Xlim-1 (Cho et al., 1991; Taira et al., 1992) were expressed normally in perturbed embryos (Fig. 3A). Moreover, expression of muscle and notochord markers such as muscle-specific actin and collagen type II (Amaya et al., 1993) were also reduced in posterior-truncated late neurula stage embryos (Fig. 3B). Expression of the neural marker, XIF-3 (Sharpe et al., 1989) and the anterior marker XAG-1 (Sive et al., 1989) were unaffected (Fig. 3B). In addition, blastopore closure in XLPOU91-injected late gastrula embryos is retarded (Fig.

C. Henig et al. / Mechanisms of Development 71 (1998) 131–142

133

Fig. 1. Ectopic expression of XLPOU91 RNA causes posterior truncations in Xenopus embryos. (A) Embryos at the 1-cell stage were injected with 2 ng of in vitro synthesized Xenopus XLPOU91 RNA. Embryos developed until stages 39–41; embryos have normal heads, but a deleted trunk-tail axis. This is not a cell-toxicity effect, since uninjected control embryos and posterior truncated embryos have the same number of cells as determined by DNA and RNA content (not shown). (B) Embryos at the 1-cell stage were injected with 2 ng of in vitro synthesized Xenopus XLPOU91* RNA. Embryos developed until stages 39–41; embryos have normal heads, but a shortened trunk-tail axis. (C) Embryos at the 1-cell stage were injected with either 2 ng of in vitro synthesized Xenopus DNsi-XLPOU91 RNA or XLPOU91 RNA. Embryos developed until stages 39–41; DNsi-XLPOU91-injected embryos (top) were normal, like uninjected control embryos (middle) in comparison with posterior-truncated XLPOU91-injected embryos (bottom). (D) Embryos at the 1-cell stage were injected with 2 ng of in-vitro synthesized XLPOU91 RNA. Embryos developed until late gastrula stage 12.5; XLPOU91-injected embryos (right panel) have retarded blastopore closure in comparison with control embryos (left panel).

134

C. Henig et al. / Mechanisms of Development 71 (1998) 131–142

Table 1 Phenotypes of embryos overexpressing XLPOU91 RNA Injection

Normal/total

Intermediate posterior truncations/total

Severe posterior truncations/total

% Posterior truncations

XLPOU91 XLPOU91* DNsi-XLPOU91

9/53 10/43 40/44

10/53 25/43 4/44

34/53 8/43 –

83 77 0.9

2 ng RNA were injected at the 1-cell stage. Embryos were scored for posterior truncation at stages 39–41. One representative experiment is shown, experiments were performed at least seven times. DNsi-XLPOU91-injected embryos served as negative controls and resembled uninjected embryos in frequency of posterior truncation.

1D), and it appears that mesoderm tissue spreads laterally around the blastopore. Blastopore closure retardation, posterior truncations, and loss of XBra expression without a loss of Spemann’s organizer are all phenotypes observed in embryos injected with the FGF dominant-negative receptor (Amaya et al., 1991, 1993; Isaacs et al., 1994). This striking phenotypic similar-

Fig. 2. DNA electrophoretic mobility shift assay (EMSA): binding of the XLPOU91 protein to the OCT-consensus sequence. Embryos at the 1-cell stage were injected with 2 ng of XLPOU91 or XLPOU91* RNAs. Embryos developed until stage 11.5 and protein was extracted. After incubation with the radioactive OCT motif oligonucleotide, products were resolved on an 8% polyacrylamide gel. The arrows mark the mobility shift bands and the free octamer DNA. XLPOU91 EMSAs are specifically inhibited by an antibody (anti-91) which recognizes native XLPOU91 protein (right well in each group, Hinkley and Perry, 1992). This shift is OCT sequence-specific, being inhibited by ten-fold excess of non-radioactive OCT oligonucleotide; but not inhibited by a ten-fold molar excess of the SP1 or CREB consensus oligonucleotides (not shown). The upper band in each well is the Oct-1 mobility shift, which was used as an internal standard to normalize between different extracts. b, blank, reaction with no extract.

ity suggested that ectopic XLPOU91 expression may inhibit endogenous FGF signaling within the embryo. 2.2. Ectopic XLPOU91 expression inhibits mesoderm induction by bFGF To address the potential role of XLPOU91 in regulating mesoderm induction by FGF, we used the animal cap mesoderm induction assay. Blastula stage animal cap explants, normally fated to differentiate as epidermal skin, differentiate to mesoderm after treatment with soluble mesoderminducing factors such as bFGF and activin (Green et al., 1990; Sokol et al., 1990). Using this assay, we could directly address the role of XLPOU91 in mesoderm induction. Embryos were injected with 2–4 ng of XLPOU91 RNA at the 1-cell stage and animal caps were removed at the blastula stage for treatment with Xenopus bFGF (XbFGF). Ani-

Fig. 3. Ectopic expression of XLPOU91 inhibits expression of posterior mesoderm markers but not Spemann organizer and neural markers. (A) Embryos at the 1-cell stage were injected with 4 ng of XLPOU91 RNA. Total RNA was isolated from pools of ten gastrula (stage 11) embryos. One embryo-equivalent of RNA was loaded per well for Northern analysis. The filters were sequentially hybridized with Xenopus cDNA probes for XBra, Xlim-1, goosecoid (gsc) and SRC. SRC is used as a positive control to compare RNA levels loaded per well. (B) Embryos at the 1-cell stage were injected with 1 ng of XLPOU91 RNA. Total RNA was isolated from pools of ten neurula (stage 18) embryos. One embryo-equivalent of RNA was loaded per well for Northern analysis. The filters were sequentially hybridized with Xenopus cDNA probes for muscle actin (m-actin), notochord-specific type II collagen (C. Type II), XIF-3, XAG-1 and EF1a. EF1a is used as a positive control to compare RNA levels loaded per well.

C. Henig et al. / Mechanisms of Development 71 (1998) 131–142

mal caps were grown to gastrula stages and RNA was isolated for Northern analysis to examine mesoderm marker

135

gene expression. Animal caps ectopically expressing XLPOU91 and treated with XbFGF do not express genes

Fig. 4. Ectopic expression of XLPOU91 inhibits mesoderm-inducing activity of bFGF and activin in animal cap explants. (A) Embryos at the 1-cell stage were injected with 4 ng of XLPOU91 or DNsi-XLPOU91 RNA. Animal cap explants were removed at stages 8–9 and treated with 150 ng/ml XbFGF until stage 11.5. Total RNA was isolated from pools of 15 gastrula and neurula stage animal cap explants from each treatment. All animal cap RNA was loaded per well for Northern analysis. The filters were sequentially hybridized with Xenopus cDNA probes for the XBra, 1A11 and SRC. These results were observed in five independent experiments; over 70 animal cap explants were compared in each treatment group. (B) Animal cap explants from XLPOU91-injected (2 ng) and uninjected embryos were treated with 100 ng/ml XbFGF. At the neurula stage, explants were scored for elongation, characteristic of bFGF induction. Left panel: animal caps from uninjected embryos +XbFGF, 5/6 elongated; right panel: XLPOU91-injected +XbFGF, 0/6 elongated like untreated control animal caps (not shown). (C) Embryos at the 1-cell stage were injected with 2 ng of XLPOU91 RNA. Animal cap explants were removed at stages 8–9 and treated with PIF/Activin until stage 11.5. Total RNA was isolated from pools of 15 gastrula stage animal cap explants from each treatment. All animal cap RNA was loaded per well for Northern analysis. The filters were sequentially hybridized with Xenopus cDNA probes for the XBra, XPO, gsc, XFKH (Xenopus Forkhead-1) and SRC.

136

C. Henig et al. / Mechanisms of Development 71 (1998) 131–142

normally induced by bFGF (Smith et al., 1991; Greene et al., 1993; Pownall et al., 1996), such as XBra, 1A11 (Fig. 4A) and XhoxA7 (not shown). Ectopic expression of the truncated DNsi-XLPOU91 protein did not inhibit XBra induction by FGF (Fig. 4A); these animal caps behaved identically to those taken from uninjected control embryos (Fig. 4A). Animal caps converted to mesoderm by bFGF characteristically elongate in response to bFGF treatment (Symes and Smith, 1987), but XLPOU91-injected caps did not undergo elongation movements after XbFGF treatment (Fig. 4B). Mesoderm induction by activin is FGFdependent (LaBonne and Whitman, 1994; Cornell et al., 1995). We found that activin failed to induce mesoderm in animal caps over expressing XLPOU91. Ventroposterior-expressed genes (XBra, XPO) were inhibited more than dorsoanterior-expressed genes (gsc, XFKH-1) in this assay (Fig. 4C); a similar observation was seen in activintreated animal caps injected with the FGF dominant-negative receptor (LaBonne and Whitman, 1994). These results suggest that XLPOU91 can directly interfere with the FGF signaling pathways during mesoderm induction in animal cap explants. 2.3. Overexpression of XLPOU91 does not inhibit activation of the MAPK cascade We wanted to determine what aspect of the FGF signaling pathway is inhibited by XLPOU91. FGF mediates mesoderm induction by activating the MAPK cascade. FGF induces mesoderm through its tyrosine kinase receptor which activates the Ras/Raf pathway. Ras/Raf activate the MAPK kinase (MEK) enzyme which in turn phosphorylates and activates the MAPK enzyme. Microinjection of RNA encoding a constuitively-activated form of MAPK suffices to induce mesoderm by turning on XBra expression in the absence of exogenous bFGF signaling (LaBonne et al., 1995; Umbhauer et al., 1995). Thus, MAPK presumably induces mesoderm by activating XBra gene expression. We examined the effect of ectopic XLPOU91 in regulating FGF mediated MAPK activation. Embryos were injected with XLPOU91 RNA and blastula stage caps were removed for XbFGF treatment. Half of the animal cap explants were scored for XBra expression by Northern blot analysis at mid-gastrula stages; while the other half were assayed for MAPK activity using a Western blot system (PhosphoPlus p44/p42 MAPK kit; New England Biolabs) which specifically recognizes the phosphorylated, and thus activated, MAPK protein. After XbFGF treatment, no decrease in phosphorylated MAPK protein was observed in XLPOU91-injected animal caps (Fig. 5A) while a parallel decrease in XBra expression was always observed (Fig. 4A). In a second system, we co-injected XLPOU91 RNA along with RNA encoding an activated MEK enzyme. MEK phosphorylates MAPK which then activates XBra expression (Gotoh et al., 1995; LaBonne et al., 1995; Umbhauer et al., 1995). Ectopic XLPOU91 expression inhibits MEK-

Fig. 5. Overexpression of XLPOU91 inhibits mesoderm induction by MAPK and MEK in animal explants. (A) Embryos at the 1-cell stage were injected with 4 ng of XLPOU91 RNA. Animal cap explants were removed at stages 8–9 and treated for 20 min with XbFGF (150 ng/ml). Protein was isolated from each treatment group for Western blot analysis using the New England Biolabs PhosphoPlus p44/p42 MAPK Antibody kit. This antibody specifically recognizes the phosphorylated-activated MAP kinase enzyme. For examination of the interaction between MEK and XLPOU91, embryos at the 1-cell stage were injected with 4 ng of XLPOU91 RNA and 2.5 ng of MEK* RNA (encoding activated MEK protein). Protein was isolated from each treatment group immediately after microdissection (see Section 4). Western analysis was performed as described. (B) Embryos at the 1-cell stage were injected with either 4 ng of XLPOU91 RNA and 2.5 ng of MEK* RNA or 5 ng of MAPK* RNA (encoding activated MAPK protein). Total RNA was isolated from pools of 15 gastrula stage animal cap explants from each treatment. All animal cap RNA was loaded per well for Northern analysis. The filter was sequentially hybridized with Xenopus cDNA probes for XBra and SRC.

mediated XBra expression (Fig. 5B). However, MAPK is activated to identical levels in both MEK and MEK/ XLPOU91 co-injected animal cap explants (Fig. 5A). XLPOU91 RNA was also co-injected along with RNA encoding an activated MAPK enzyme; ectopic XLPOU91 expression inhibited transcriptional activation of XBra by MAPK (Fig. 5B). These results demonstrate that ectopic XLPOU91 expression inhibits the animal cap mesoderm response to FGF without inhibiting phosphorylation of MAPK. Therefore, XLPOU91 probably inhibits XBra expression by acting downstream of the MAPK cascade. 2.4. Knockout of endogenous XLPOU91 activity The best way to determine the function of a protein is to eliminate its activity from the embryo. We used an antisense RNA approach to address the role of XLPOU91 in early Xenopus development. At the beginning of gastrulation (stage 10), animal caps lose competence to respond to

C. Henig et al. / Mechanisms of Development 71 (1998) 131–142

bFGF (Green et al., 1990). This is the developmental window, when endogenous XLPOU91 transcript levels are increasing to peak levels in all embryonic cells (Frank and Harland, 1992; Hinkley et al., 1992). Since ectopic expression of XLPOU91 inhibits FGF signaling (Figs. 4 and 5), we thought that XLPOU91 protein could regulate the timing of mesodermal competence to FGF. Animal cap explants from XLPOU91 antisense-injected embryos were removed at stages 8 (blastula) and 10.5 (gastrula) and treated with XbFGF; expression of the mesoderm marker XBra was determined at stage 11.5 (midgastrula). As controls, embryos were injected with comparable levels of antisense RNA of the human Kv1.3 potassium channel (Fig. 6A). We found that mesoderm competence to FGF was

137

extended in animal caps injected with antisense XLPOU91 RNA. Antisense-injected animal caps removed at stage 10.5 and treated with XbFGF expressed high levels of XBra at stage 11.5 (Fig. 6A) and high levels of 1A11 and XhoxA7 at neurula stages (Fig. 6B) in comparison with controls. We also found that XBra expression was stimulated in antisense-injected animal caps treated with XbFGF at stage 8, in comparison with the control XbFGF-treated animal caps (Fig. 6A). Interestingly, MAPK activity was stimulated identically by XbFGF in both the antisense and control groups of animal caps (not shown), again supporting our observation that XLPOU91 acts downstream of MAPK activation. Animal caps do not normally express genes induced by mesoderm inducing factors, yet antisense-injected animal caps (removed at st. 8 or 10.5) also express XBra and 1A11 mRNAs (Fig. 6A,B) and elongate in the absence of bFGF (not shown). Thus, depletion of endogenous

Fig. 6. Overexpression of XLPOU91 antisense RNA extends mesoderm competence in animal cap explants. (A) Embryos at the 1-cell stage were injected with 5 ng of full-length antisense XLPOU91 RNA. Animal cap explants were removed at either stage 8 or stage 10.5 and treated with XbFGF (150 ng/ml). Total RNA was isolated from pools of seven embryos and 15 stage 11.5 animal cap explants from each treatment. One embryo equivalent of RNA was loaded per well for Northern analysis; all animal cap RNA was loaded. The filter was sequentially hybridized with Xenopus cDNA probes for XBra and SRC. As controls for antisense RNA injection, embryos were injected with antisense RNA of the human Kv1.3 potassium channel. These results were observed in eight independent experiments; over 120 animal cap explants were compared in each treatment group. (B) Embryos at the 1-cell stage were injected with 5 ng of antisense XLPOU91 RNA. Animal cap explants were removed at stage 10.5 and treated with XbFGF (150 ng/ml) until stage 11.5. Total RNA was isolated from pools of 15 stage 15 animal cap explants from each treatment. All animal cap RNA was loaded per well for Northern analysis. The filter was sequentially hybridized with Xenopus cDNA probes for 1A11, Xenopus HoxA7 (XhoxA7) and SRC. As controls for antisense RNA injection, embryos were injected with antisense RNA of the human Kv1.3 potassium channel. (C) Embryos at the 1-cell stage were injected with 5 ng of antisense XLPOU91 RNA. Embryos developed until stage 12 and protein was extracted. After incubation with the radioactive OCT consensus oligonucleotide, products were resolved on a 8% polyacrylamide gel. The antisense RNA eliminates the XLPOU91 shift present in uninjected embryos. The Oct-1 shift is unaffected in antisense-injected embryos serving as an internal standard to normalize between the different extracts. By not inhibiting the Oct-1 EMSA, this demonstrates the XLPOU91 antisense RNA specifically inhibits the XLPOU91 protein EMSA. The arrows mark the mobility shift bands and free octamer DNA. In addition, EMSA was also performed using an SP1-specific oligonucleotide (Promega) and no difference in the SP1 mobility shift was observed between controls and antisense-injected embryos (not shown). (D) Embryos at the 1-cell stage were co-injected with 4–5 ng of antisense XLPOU91 RNA and 1.5–2 ng of sense XLPOU91 RNA. Embryos developed until stage 11 and protein was extracted. After incubation with the radioactive OCT consensus oligonucleotide, products were resolved on a 8% polyacrylamide gel. The antisense RNA reduces the stimulation of the XLPOU91 EMSA in coinjected embryos. As negative controls, embryos were injected with 4–5 ng of Kv1.3 potassium channel antisense RNA. This RNA does not alter the XLPOU91 mobility shift. XLPOU91 EMSAs are specifically inhibited by an antibody (anti-91) which recognizes native XLPOU91 protein (right well in each group; Hinkley and Perry, 1992).

138

C. Henig et al. / Mechanisms of Development 71 (1998) 131–142

XLPOU91 protein permits expression of XBra in animal cap explants. We have measured the changes in the levels of endogenous and ectopic XLPOU91 protein after antisense injection by EMSA. EMSAs of endogenous (Fig. 6C) and ectopic (Fig. 6D) XLPOU91 protein are consistently inhibited 2–5 fold by XLPOU91 antisense RNA injection. These results suggest that antisense XLPOU91 RNA injection depletes XLPOU91 protein levels by inhibiting translation of its mRNA.

3. Discussion We have shown that overexpression of XLPOU91 and XLPOU91* RNAs caused posterior truncations in 80% of the injected embryos (Table 1). Expression of the wild type XLPOU91 protein caused severe posterior perturbations in 75% of the truncated embryos (Fig. 1A; Table 1). In contrast, expression of the XLPOU91* protein with reduced DNA binding activity caused severe posterior truncations in only 25% of the perturbed embryos; 75% of these perturbed embryos had moderate posterior truncations (Fig. 1B; Table 1). Injection of RNA encoding a non DNA-binding protein, DNsi-XLPOU91, caused no phenotype (Fig. 1C; Table 1). Thus the posterior truncation phenotype is dependent on the DNA-binding characteristics of the XLPOU91 protein. XLPOU91-injected embryos expressed reduced levels of the pan-mesodermal XBra marker at gastrula stages; the muscle-specific actin and notochord-specific collagen markers were also reduced in injected neurula stage embryos (Fig. 3). In contrast, gastrula stage marker genes of the Spemann organizer such as gsc and Xlim-1 are expressed at normal levels (Fig. 3) in injected embryos; the anterior/neural markers XAG-1 and XIF-3 are also expressed normally in posterior-truncated embryos (Fig. 3). This phenotype is reminiscent of embryos injected with RNA encoding the FGF dominant-negative receptor or the XBra dominant-negative protein (Amaya et al., 1991, 1993; Isaacs et al., 1994; Conlon et al., 1996). Embryos lose expression of XBra and have similar posterior truncations in which both muscle and notochord tissues are lost (Figs. 1 and 3). XBra appears to be the major target of FGF signaling, since XBra autonomously induces mesoderm in animal caps and XBra protein is necessary for formation of posterior mesoderm in whole embryos (Cunliffe and Smith, 1992; Conlon et al., 1996). On the basis of these similarities, we investigated the possibility that ectopic XLPOU91 expression perturbed endogenous FGF signaling within the embryos. Using animal caps as a model system for mesoderm induction, we found that ectopic XLPOU91 expression in animal caps inhibited bFGF activation of XBra, 1A11 and HoxA7 expression and explant elongation movements associated with mesoderm induction (Fig. 4). FGF induces XBra expression through the activation of MAPK. We therefore investigated the interaction of XLPOU91 with members of the MAPK activation cascade. Ectopic

XLPOU91 expression inhibited induction of XBra mRNA by either bFGF or injection of RNA encoding an activated MEK enzyme (Figs. 4 and 5). However, XLPOU91 did not inhibit their activation of the MAPK protein (Fig. 5). XLPOU91 also inhibited XBra expression induced by a constuitively-activated MAPK enzyme (Fig. 5). It is not known how MAPK activates XBra transcription. However, these results suggest that XLPOU91 acts downstream of the MAPK cascade, and that it does not suppress FGF signal transduction by inhibiting MAPK activation. The role of endogenous XLPOU91 activity in regulating FGF signaling was examined in animal caps. XLPOU91 protein was depleted in animal caps by injection of antisense RNA. Two striking phenomena were observed in these animal caps. First, XLPOU91 antisense-injected animal caps removed at the blastula stage, expressed high levels of XBra in the absence of FGF in contrast to control animal caps (Fig. 6). Secondly, animal caps injected with XLPOU91 antisense RNA had an extended window of FGF-competence response in comparison with control animal caps (Fig. 6). The former observation suggests that XLPOU91 could act to prevent endogenous FGF mesoderm induction in animal cap ectoderm. Functional FGF signaling is necessary in animal caps for mesoderm induction by activin (LaBonne and Whitman, 1994; Cornell et al., 1995) or neural induction by noggin (Launay et al., 1996). This residual FGF activity could be maintained at low levels by endogenous XLPOU91 in animal cap cells. Too much of this endogenous FGF signaling could possibly activate XBra expression and cause auto-induction of animal cap ectoderm into mesoderm. XLPOU91 appears required to prevent spontaneous XBra expression in animal cap ectoderm since XLPOU91 antisense expressing caps express XBra (Fig. 6) and elongate as mesoderm (not shown). Animal caps removed at blastula stages are more sensitive to the loss of XLPOU91 activity than caps removed at gastrula stages, since XLPOU91 antisense-injected gastrula stage animal caps express less XBra than blastula animal caps. Perhaps latent FGF signaling within the animal pole is higher in explants removed at blastula versus gastrula stages. In an analogous manner, vegetal pole cells also have an intrinsic capacity to respond to FGF but this potential is repressed (Cornell et al., 1995). These authors suggest that this intrinsic inhibition of FGF signaling may maintain vegetal cells in their proper non-mesodermal fate (Cornell et al., 1995). In previous studies, we found that XLPOU91 transcript levels were quite high in the vegetal pole (Frank and Harland, 1992). Therefore, animal cap and vegetal pole cells may share a similar capacity to suppress residual FGF signaling, which could be important for maintaining the unique ectodermal and endodermal lineages of these cells. These results also suggest that XLPOU91 is an endogenous competence regulator. Classic embryological experiments have shown that ectoderm loses its ability to respond to mesoderm induction during gastrulation (Jones

C. Henig et al. / Mechanisms of Development 71 (1998) 131–142

and Woodland, 1987). This occurs through a built-in timing mechanism in a cell-autonomous manner (Grainger and Gurdon, 1989). This loss of competence probably limits the extent of cell types in which ectoderm can be induced. Increasing or decreasing XLPOU91 levels may determine whether a cell can respond to FGF and make mesoderm at a given time during development. This is important, because there appears to be no strict spatial regulation of either FGF ligand or receptor expression during development; yet FGF is known to be involved in regulating specific spatial aspects of mesoderm and neural induction. Interestingly, when mesoderm competence to FGF is lost in animal caps at gastrula stages, FGF still stimulates MAPK activation (Graves et al., 1994). Thus at later gastrula stages, MAPK does not suffice to activate XBra expression and mesoderm induction. This observation suggests that pathways downstream of MAPK, such as XLPOU91, may be required for directing cells to their correct fates in the proper temporal manner. Thus, XLPOU91, which increases to peak levels at gastrula stages, could act as a switch to modify FGF mesoderm-response in the embryo. This is the first observation of a transcription factor which can modify the timing of competence during Xenopus development. Little is known about the molecular regulation of competence. Xenopus notch protein, and most recently the Xenopus somatic histone H1 protein, have been shown to be involved in regulating mesoderm competence for activin in animal caps (Coffman et al., 1993; Steinbach et al., 1997). However, in these studies it was not reported whether these molecules regulate competence to FGF. XLPOU91’s expression pattern and function fit a number of criteria to suggest a role in cellular competence. First, XLPOU91 expression is activated a cell-autonomous manner throughout all cells in the embryo (Frank and Harland, 1992). Second, gene expression peaks at the time when mesoderm induction competence is lost. XLPOU91 could be viewed as a built-in timer which shuts down FGF responses over a certain threshold. In Xenopus, posterior mesoderm formation is dependent on FGF induction of XBra expression through the MAPK cascade. It is unclear how MAPK turns on XBra transcription the nucleus. It is also unknown how the mesoderm response to FGF is lost during gastrulation in animal caps, despite their intrinsic capacity to still activate MAPK. These observations suggest that mechanisms exist to inhibit XBra expression in the presence of activated MAPK. In preliminary studies, we have observed that XLPOU91 represses transcription from OCT-dependent promoters co-injected into Xenopus embryos (not shown). In a number of other systems, POU proteins have been shown to act as transcriptional repressors in addition to acting as differentiation inhibitors (Lillycrop et al., 1994; Weinstein et al., 1995; Dawson et al., 1996; Wu et al., 1997). Either directly or indirectly, XLPOU91 protein could repress XBra promoter transcription. Thus, by expressing a repressor of XBra transcription, it may be possible to eliminate the mesoderm

139

response to FGF in the animal cap, while perhaps maintaining the MAPK activation by FGF for other non-mesodermal functions. FGF has been implicated as a potential caudalizer in the developing nervous system (Cox and Hemmati-Brivanlou, 1995; Doniach, 1995; Kenkgaku and Okamoto, 1995; Lamb and Harland, 1995). It is intriguing to think that proteins such as XLPOU91 could potentiate a role in the switch from mesodermal to neural competence by modulating FGF response in the embryo.

4. Experimental Procedures 4.1. Xenopus embryos, explants and inducing factors Ovulation, in vitro fertilization, embryo culture and dissections were carried out as described (Re’em-Kalma et al., 1995). Embryos were staged according to Nieuwkoop and Faber (1967). XbFGF treated (100–150 ng/ml) animal cap explants were routinely grown in 1/2× modified Ringer’s (MR) with 0.1 mg/ml gentamycin and 0.2 mg/ml BSA from stages 8–9 until stage 11.5. PIF-activin was prepared from the P338D1 cell line (Sokol et al., 1990), and activintreated animal caps were grown in the same conditions as XbFGF treated explants. 4.2. RNA injections, mutagenesis and plasmids Different concentrations of capped sense in vitro transcribed full-length XLPOU91 RNA (1.0–4.0 ng in a volume of 10 nl) were injected into embryos at the 1-cell stage. Capped antisense XLPOU91 RNA (5.0 ng) was also injected at the 1-cell stage. In antisense experiments, 5 ng of capped human potassium channel Kv1.3 antisense RNA was injected as a negative control (Marom et al., 1993). Using in vitro site directed mutagenesis (Kunkel, 1985), the XLPOU91* mutant was generated by deleting two basic amino acids (a histidine and a lysine residue) in the highly-conserved basic motif adjacent to the first homeobox helix domain of the XLPOU91 protein (Frank and Harland, 1992). This region is critical for DNA-binding by POU proteins (Ingraham et al., 1990; Treacy et al., 1991), and some studies have reported that this deletion causes dominant-negative like phenotypes in Drosophila (Treacy et al., 1991, 1992). Equal molar amounts of in vitrotranslated (reticulocyte lysates) wild type XLPOU91 and XLPOU91* proteins were used in EMSAs. We found that the XLPOU91* protein bound DNA about three-fold less efficiently than wild type XLPOU91 protein. After injection of equal amounts of these RNAs into embryos, and extraction of proteins at mid-gastrula stages, the wild type protein stimulates the EMSA about 15-fold versus uninjected embryos. XLPOU91* stimulates EMSA about five-fold versus uninjected embryos, about three-fold lower than the wild type protein. Thus, XLPOU91* causes a weaker gain-of-function phenotype and not a dominant-negative

140

C. Henig et al. / Mechanisms of Development 71 (1998) 131–142

phenotype in comparison with the wild type protein. Another deletion of the XLPOU91 protein, DNsiXLPOU91 was generated by deleting an internal PstI–Nsi fragment of 492 bp (164 amino acids). The XLPOU91 POU-homeo domain spans amino acids 202–354, and the DNsi mutation deletes amino acids 240–403, thus removing nearly all of the POU-domain region. This mRNA encodes a protein of 265 amino acids instead of the wild type protein of 429 amino acids. DNsi-XLPOU91 protein does not bind DNA as determined by EMSAs and ectopic expression does not cause posterior truncations or inhibit mesoderm induction by XbFGF in animal caps. Plasmids containing cDNAs encoding the constuitively active MAPK (Xp42D324N) and the constuitively active MEK (MEK1S217E/S221E) proteins were transcribed in vitro as described (Umbhauer et al., 1995).

and approximately 12 animal cap equivalents (approximately 6 mg of protein extract) were loaded per well for electrophoresis.

Acknowledgements We thank Drs. R. Harland, J. Smith, M. Whitman, I. Dawid, E. Amaya, A. Fainsod, M. Frank, M. Perry, M. Jamrich and H. Isaacs for plasmids. We thank M. Perry for the XLPOU91 antiserum. Xenopus bFGF (XbFGF) was generously provided by Dr. Gera Neufeld, Technion, Israel. This research was supported by a grant from the Israel Science Foundation (Israel Academy of Sciences and Humanities) and by Bruce Rappaport and Technion research funds.

4.3. Northern blot analysis RNA was extracted from embryos and prepared for Northern blot analysis as described (Re’em-Kalma et al., 1995). Electrophoresis, probe preparation, filter hybridization and exposure was performed as described (Re’emKalma et al., 1995). Quantitation was performed using a Fuji phosphor-imaging system. EF1a or SRC was used as a positive standard for comparing amounts of RNA loaded per well at any given stage. 4.4. DNA electrophoretic mobility shift assay (EMSA) Embryos (stages 11–12) were lysed at 10 ml/embryo and extracts prepared according to Taylor et al. (1991). Lysis buffer contains 50mM Tris-HCl, pH 7.9; 25% glycerol; 50 mM KCl; 0.1 mM EDTA; 0.5 mM PMSF; 25 mg/ml Aprotinin; 25 mg/ml Leupeptin and 10 mg/ml Bestatin. EMSAs were performed with an octamer motif (OCT) oligonucleotide (Promega) in 25 ml total volume (8 ml extract/reaction) and resolved by electrophoresis on 8% polyacrylamide gels (Hinkley and Perry, 1992; Hinkley et al., 1992). Gels were dried and quantitation was performed using a Fuji phosphorimaging system. 4.5. MAP kinase Western blot analysis Western blot analysis was performed using the PhosphoPlus p44/p42 MAPK antibody kit (New England Biolabs). For XbFGF treated animal caps, blastula stage explants were immersed in XbFGF for 20 minutes and lysed. Lysis buffer contains 62.5 mM Tris-HCl, pH 6.8; 2% SDS; 10% glycerol; 50 mM DTT and 0.1% bromophenol blue. For animal caps injected with MEK RNA, explants were lysed immediately after surgical removal. Filters were routinely incubated first with the phospho-specific MAPK antibody, stripped and then re-incubated with the control MAPK antibody which recognizes both the active and inactive forms of MAPK. Usually, 12–18 animal caps were lysed per sample

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. Amaya, E., Stein, P.A., Musci, T.J., Kirschner, M.W., 1993. FGF signaling in the early specification of mesoderm in Xenopus. Development 118, 477–487. Cho, K.W.Y., Blumberg, B., Steinbeisser, H., De Robertis, E.M., 1991. Molecular nature of Spemann’s organizer: the role of the Xenopus homeobox gene goosecoid. Cell 67, 1111–1120. Coffman, C.R., Skoglund, P., Harris, W.A., Kintner, C.R., 1993. Expression of an extracellular deletion of Xotch diverts cell fate in Xenopus embryos. Cell 73, 659–671. Conlon, F.L., Sedgwick, L.G., Weston, K.M., Smith, J.C., 1996. Inhibition of XBra transcription activation causes defects in mesodermal patterning and reveals autoregulation of XBra in dorsal mesoderm. Development 122, 2427–2435. Cornell, R.A., Musci, T.J., Kimelman, D., 1995. FGF is a prospective competence factor for early activin-type signals in Xenopus mesoderm induction. Development 121, 2429–2437. Cox, W.G., Hemmati-Brivanlou, A., 1995. Caudalization of neural fate by tissue recombination and bFGF. Development 121, 4349–4358. Cunliffe, V., Smith, J., 1992. Ectopic mesoderm formation in Xenopus caused by widespread expression of Brachyury homologue. Nature 358, 427–430. Dale, L., Howes, G., Price, M.J., Smith, J.C., 1992. Bone morphogenetic protein-4: a ventralizing factor in early Xenopus development. Development 115, 573–585. Dawson, S.J., Morris, P.J., Latchman, D.S., 1996. A single amino acid change converts an inhibitory transcription factor into an activator. J. Biol. Chem. 271, 11631–11633. Dirksen, M.L., Jamrich, M., 1992. A novel, activin inducible, blastopore lip-specific gene of Xenopus laevis contains a fork head DNA-binding domain. Genes Dev. 6, 599–608. Doniach, T.B., 1995. Basic FGF as an inducer of anteroposterior neural pattern. Cell 83, 1067–1070. Frank, D., Harland, R., 1992. Localized expression of a Xenopus POU gene depends on cell-autonomous transcriptional activation and induction-dependent inactivation. Development 115, 439–448. Gawantka, V., Delius, H., Hirschfeld, K., Blumenfeld, C., Niehrs, C., 1995. Antagonizing the Spemann organizer: role of the homeobox gene Xvent-1. EMBO J. 14, 6268–6279. Gotoh, Y., Masuyama, N., Suzuki, A., Ueno, N., Nishida, E., 1995. Invol-

C. Henig et al. / Mechanisms of Development 71 (1998) 131–142 vement of the MAP kinase cascade in Xenopus mesoderm induction. EMBO J. 14, 2491–2498. Graff, J.M., Thies, R.S., Song, J.J., Celeste, A.J., Melton, D.A., 1994. Studies of a Xenopus BMP receptor suggest that ventral mesoderminducing signals override dorsal signals in vivo. Cell 79, 169–179. Grainger, R.M., Gurdon, J.B., 1989. Loss of competence in amphibian induction can take place in single nondividing cells. Proc. Natl. Acad. Sci. USA 86, 1900–1904. Graves, L.M., Northrop, J., Potts, B.C., Krebs, E.G., Kimelman, D., 1994. Fibroblast growth factor but not activin, is a potent activator of mitogenactivated protein kinase in Xenopus explants. Proc. Natl. Acad. Sci. USA 91, 1662–1666 Green, J.B.A., Howes, G., Symes, K., Cooke, J., Smith, J.C., 1990. The biological effects of XTC-MIF, quantitative comparison with Xenopus bFGF. Development 108, 173–183. Green, J.B.A., New, H.V., Smith, J.C., 1992. Responses of embryonic Xenopus cells to activin and FGF are separated by multiple dose thresholds and correspond to distinct axes of the mesoderm. Cell 71, 731–739. Greene, J.M., Otani, H., Good, P.J., Dawid, I.B., 1993. A novel family of retrotransposon-elements in Xenopus laevis with a transcript inducible by two growth factors. Nucleic Acids Res. 21, 2375–2381. Hartley, R.S., Lewellyn, A.L., Maller, J.L., 1994. MAP kinase is activated during mesoderm induction in Xenopus laevis. Dev. Biol. 163, 521–524. Hemmati-Brivanlou, A., Melton, D., 1992. A truncated activin receptor dominantly inhibits mesoderm induction and formation of axial structures in Xenopus embryos. Nature 359, 609–614. Hinkley, C.S., Martin, J.F., Leibham, D., Perry, M., 1992. Sequential expression of multiple POU proteins during amphibian early development. Mol. Cell. Biol. 12, 638–649. Hinkley, C., Perry, M., 1992. Histone H2B gene transcription during Xenopus early development requires functional cooperation between proteins bound to the CCAAT and octamer motifs. Mol. Cell. Biol. 12, 4400– 4411. Ingraham, H.A., Flynn, S.E., Voss, J.W., Albert, V.R., Kapiloff, M.S., Wilson, L., Rosenfeld, M.G., 1990. The POU-specific domain of Pit-1 is essential for sequence-specific, high affinity DNA binding and DNAdependent Pit-1–Pit-1 interactions. Cell 61, 1021–1033. Isaacs, H.V., Pownall, M.E., Slack, J.M.W., 1994. eFGF regulates XBra expression during Xenopus gastrulation. EMBO J. 13, 4469–4481. Jones, C.M., Lyons, K.M., Lapan, P.M., Wright, C.V.E., Hogan, B.L.M., 1992. DVR-5 (Bone Morphogenetic Protein-4) as a posterior-ventralizing factor in Xenopus mesoderm induction. Development 115, 639–647. Jones, E.A., Woodland, H.R., 1987. The development of animal cap cells in Xenopus: a measure of animal cap competence to form mesoderm. Development 101, 557–563. Kenkgaku, M., Okamoto, H., 1995. bFGF as a possible morphogen for the anteroposterior axis of the central nervous system in Xenopus. Development 121, 3121–3130. Kessler, D.S., Melton, D.A., 1994. Vertebrate embryonic induction: mesodermal and neural patterning. Science 266, 596–604. Kimelman, D., Kirschner, M., 1987. Synergistic induction of mesoderm by FGF and TGF-b and the identification of an mRNA coding for FGF in the early Xenopus embryo. Cell 51, 869–877. Kimelman, D., Christian, J.L., Moon, R.T., 1992. Synergistic principles of development: overlapping patterning systems in Xenopus mesoderm induction. Development 116, 1–9. Kunkel, T.A., 1985. Rapid and efficient site-specific mutagenesis without phenotypic selection. Proc. Natl. Acad. Sci. USA 82, 488–492. LaBonne, C., Burke, B., Whitman, M., 1995. Role of MAP kinase in mesoderm induction and axial patterning during Xenopus development. Development 121, 1475–1486. LaBonne, C., Whitman, M., 1994. Mesoderm induction by activin requires FGF-mediated intracellular signals. Development 120, 463–472. Lamb, T.M., Harland, R.M., 1995. Fibroblast growth factor is a direct neural inducer, which combined with noggin generates anterior-posterior neural pattern. Development 121, 3627–3636.

141

Launay, C., Fromentoux, V., De-Li, S., Boucat, J.C., 1996. A truncated FGF receptor blocks neural induction by endogenous Xenopus inducers. Development 122, 869–880. Lillycrop, K.A., Dawson, S.J., Estridge, J.K., Gerster, T., Matthias, P., Latchman, D.S., 1994. Repression of the herpes simplex virus immediate-early promoter by the Oct-2 transcription factor is dependent on an inhibitory region at the N terminus of the protein. Mol. Cell. Biol. 14, 7633–7642. Marom, S., Goldstein, S.A.N., Kupper, J., Levitan, I.B., 1993. Mechanism and modulation of inactivation of the Kv3 potassium channel. Recept. Channels 1, 81–88. Mead, P.E., Brivanlou, I.H., Kelley, C.M., Zon, L.I., 1990. BMP-4 responsiveness regulation of dorsal-ventral patterning by the homeobox protein Mix.1. Nature 382, 357–360. Nieuwkoop, P.D., 1969. The formation of mesoderm in urodelan amphibians. I. The induction by the endoderm. Roux’s Arch. Dev. Biol. 162, 341–373. Nieuwkoop, P., Faber, J., 1967. Normal Table of Xenopus laevis (Daudin). North-Holland, Amsterdam. Okamoto, K., Okazawa, H., Okuda, A., Sakai, M., Muramatsu, M., Hamada, H., 1990. A novel octamer binding transcription factor is differentially expressed in mouse embryonic cells. Cell 60, 461–472. Onichtchouk, D., Gawantka, V., Dosch, R., Delius, H., Hirschfeld, K., Blumenstock, C., Niehrs, C., 1996. The Xvent-2 homeobox gene is part of the BMP-4 signalling pathway controlling the dorsoventral patterning of Xenopus mesoderm. Development 122, 3045–3053. Papalopulu, N., Kintner, C., 1996. A Xenopus gene, Xbr-1, defines a novel class of homeobox genes and is expressed in the dorsal ciliary margin of the eye. Dev. Biol. 174, 104–114. Pownall, M.E., Tucker, A.S., Slack, J.M.W., Isaacs, H.V., 1996. eFGF, Xcad3, and Hox genes form a molecular pathway that establishes the anteroposterior axis in Xenopus. Development 122, 3881–3892. Re’em-Kalma, Y., Lamb, T., Frank, D., 1995. Competition between noggin and bone morphogenetic protein 4 activities may regulate dorsalization during Xenopus development. Proc. Natl. Acad. Sci. USA 92, 12141–12145. Rosner, M.H., Vigano, M.A., Ozato, K., Timmons, P.M., Poirier, F., Rigby, P.W.J., Staudt, L.M., 1990. A POU-domain transcriptional factor in early stem cells and germ cells of the mammalian embryo. Nature 345, 686–692. Ryan, A.K., Rosenfeld, M.G., 1997. Pou domain family values: flexibility, partnerships, and developmental codes. Genes Dev. 11, 1207–1225. Schmidt, J.E., von Dassow, G., Kimelman, D., 1996. Regulation of dorsalventral patterning: the ventralizing effects of the novel Xenopus homeobox gene Vox. Development 122, 1711–1721. Scholer, H.R., Ruppert, S., Suzuki, N., Chowdhury, K., Gruss, P., 1990. New type of POU domain in germ line-specific protein Oct-4. Nature 344, 435–439. Sharpe, C.R., Pluck, A., Gurdon, J.B., 1989. XIF3, a Xenopus peripherin gene, requires an inductive signal for enhanced expression in anterior neural tissue. Development 107, 701–714. Sive, H., 1993. The frog prince-cess: a molecular formula for dorsoventral patterning in Xenopus. Genes Dev. 7, 1–12. Sive, H.L., Hattori, K., Weintraub, H., 1989. Progressive determination during formation of the anteroposterior axis in Xenopus laevis. Cell 58, 171–180. Smith, J.C., Price, B.M.J., Green, J.B.A., Weigel, D., Herrmann, B.G., 1991. Expression of a Xenopus homolog of Brachyury (T) is an immediate-early response to mesoderm induction. Cell 67, 79–87. Sokol, S., Wong, G.G., Melton, D.A., 1990. A mouse macrophage factor induces head structures and organizes a body plan in Xenopus embryos. Science 249, 561–564. Steinbach, O.C., Wolffe, A., Rupp, R.A.W., 1997. Somatic linker histones cause a loss of mesodermal competence in Xenopus. Nature 389, 395– 399. Sudwarti, S., Nieuwkoop, P.D., 1971. Mesoderm formation in the anuran Xenopus laevis (Daudin). Roux’s Arch. Dev. Biol. 166, 189–204.

142

C. Henig et al. / Mechanisms of Development 71 (1998) 131–142

Symes, K., Smith, J.C., 1987. Gastrulation movements provide an early marker of mesoderm induction in Xenopus laevis. Development 101, 339–349. Taira, M., Jamrich, M., Good, P.J., Dawid, I.B., 1992. The LIM domaincontaining homeo box gene Xlim-1 is expressed specifically in the organizer region of Xenopus gastrula embryos. Genes Dev. 6, 356–366. Taylor, M.V., Gurdon, J.B., Hopwood, N.D., Towers, N., Mohun, T.J., 1991. Xenopus embryos contain a somitic-specific, MyoD-like protein that binds to a promoter site required for muscle actin expression. Genes Dev. 5, 1149–1160. Treacy, M.N., He, X., Rosenfeld, M.G., 1991. I-POU: a POU-domain protein that inhibits neuron-specific gene activation. Nature 350, 577– 584. Treacy, M.N., Neilson, L.I., Turner, E.E., He, X., Rosenfeld, M.G., 1992.

Twin of I-POU: a two amino acid difference in the I-POU homeodomain distinguishes an activator from an inhibitor of transcription. Cell 68, 491–505. Umbhauer, M., Mardhall, C.J., Mason, C.S., Old, R.W., Smith, J.C., 1995. Mesoderm induction in Xenopus caused by activation of MAP kinase. Nature 376, 58–62. Weinstein, D.E., Burrola, P.G., Lemke, G., 1995. Premature Schwann cell differentiation and hypermyelination in mice expressing a targeted antagonist of the POU transcription factor SCIP. Mol. Cell. Neurosci. 6, 212–229. Wu, G.D., Lai, E.J., Huang, N., Wen, X., 1997. Oct-1 and CCAAT/enhancer -binding protein (C/EBP) bind to overlapping elements within the interleukin-8 promoter. The role of Oct-1 as a transcriptional repressor. J. Biol. Chem. 272, 2396–2403.