Groucho corepressor proteins regulate otic vesicle outgrowth

DEVELOPMENTAL DYNAMICS 233:760 –771, 2005

RESEARCH ARTICLE

Groucho Corepressor Proteins Regulate Otic Vesicle Outgrowth Baubak Bajoghli,† Narges Aghaallaei,† and Thomas Czerny*

The Groucho/Tle family of corepressor proteins is known to regulate multiple developmental pathways. Applying the dominant-negative effect of the short member Aes, we demonstrate here a critical role of this gene family also for ear development. Misexpression of Aes in medaka embryos resulted in reduced size or loss of otic vesicles, whereas overexpression of the full-length Groucho protein Tle4 gave the opposite phenotype. These results are in close agreement with phenotypes observed for eye formation, suggesting a similar role for Groucho/Tle proteins in the developmental pathways of both sensory organs. Furthermore, by using the heat-inducible HSE promoter, we observed reversible branching of the embryonic axis upon Aes misexpression, indicating a transient duplication of the organizer. Groucho proteins, therefore, are critical for organizer maintenance. Developmental Dynamics 233:760 –771, 2005. © 2005 Wiley-Liss, Inc. Key words: Medaka; Groucho; Aes; heat-inducible misexpression; otic vesicle; eye development; lens placode; axis duplication Received 3 October 2004; Revised 21 January 2005; Accepted 22 January 2005

INTRODUCTION Groucho or Transducin-like Enhancer of Split (Tle) proteins serve as nonDNA binding corepressors for specific subsets of DNA binding transcription factors. Upon interaction, they can switch the activating potential of its binding partner to a repressing function (Valentine et al., 1998; Eberhard et al., 2000; Cai et al., 2003). Members of the Groucho/Tle family have been identified in vertebrates such as humans (Stifani et al., 1992), mice (Miyasaka et al., 1993), Xenopus (Choudhury et al., 1997), zebrafish (Wu¨lbeck and CamposOrtega, 1997), and medaka (Lopez-Rios et al., 2003). Groucho/Tle members are widely expressed both during development and in the adult, in contrast to the more

limited expression pattern of their DNA binding partners (Hartley et al., 1988; Stifani et al., 1992; Miyasaka et al., 1993; Schmidt and Sladek, 1993; Choudhury et al., 1997; Jimenez et al., 1997; Pflugrad et al., 1997; Sharief et al., 1997). The expression of Groucho/ Tle members in diverse sensory organs indicates a role of this gene family during development of these organs. For example, it has been shown that Tle1 and Tle4 are expressed in the developing eye of medaka fish. In addition, Tle3 has been found in the lens placodes of mice (Leon and Lobe, 1997) and medaka (Lopez-Rios et al., 2003). In Xenopus, Grg4 and Grg5 are expressed in otic vesicles (Molenaar et al., 2000). An interesting member of Groucho/ Tle family is Amino Enhancer of Split

(Aes) or Grg5, which exhibits strong similarity to the amino-terminal domains of other family members but lacks the C-terminal WD-40 repeats (Miyasaka et al., 1993). Aes/Grg5 genes have been identified from various vertebrates (Schmidt and Sladek, 1993; Choudhury et al., 1997; Molenaar et al., 2000; Lopez-Rios et al., 2003). During early embryonic development, it is widely expressed, switching to a tissue-specific pattern in adults (Choudhury et al., 1997; Molenaar et al., 2000). The amino-terminal part of Aes/Grg5 harbors the Qdomain, a multimerization motif, facilitating interactions with other Groucho family members. Lack of binding motifs for histone deacetylases (HDAC; Brantjes et al., 2001),

Institute of Animal Breeding and Genetics, University of Veterinary Medicine, Vienna, Austria Grant sponsor: Austrian Science Fund (FWF); Grant number: P15185. † B. Bajoghli and N. Aghaallaei contributed equally to this work. *Correspondence to: Thomas Czerny, Institute of Animal Breeding and Genetics, University of Veterinary Medicine, Veterina¨rplatz 1, A-1210 Vienna, Austria. E-mail: [email protected] DOI 10.1002/dvdy.20398 Published online 28 April 2005 in Wiley InterScience (www.interscience.wiley.com).

© 2005 Wiley-Liss, Inc.

Groucho REGULATES OTIC VESICLE OUTGROWTH 761

TABLE 1. Phenotypic Effects of Aes Misexpression in Medaka Embryos DNA concentration Number of induced embryos Dead embryos Surviving embryos GFP positive after inductionb Malformations (%)c Otic vesicle defects Eye defects Mid-hindbrain defects Heart position

a

10 ng/␮l

30 ng/␮l

50 ng/␮l

134 20 114 105 (92%)

180 26 154 150 (97%)

284 97 187 170 (91%)

9 (8.5%) 8 (7.6%) 7 (6.6%) 4 (4%)

34 (23%) 29 (19%) 15 (10%) 23 (15%)

54 (32%) 57 (33%) 45 (26%) 36 (21%)

a

Embryos were injected at the one-cell stage and GFP negative embryos were selected for heat treatment (39°C/2h) at the early gastrula stage. b Percentages were calculated from the surviving embryos. c Percentages were calculated from the GFP positive embryos. TABLE 2. Eye and Otic Vesicle Phenotypes Observed after Aes Misexpression

Stage of activation

one-cella

MidBlastulab

EarlyGastrulac,d

MidGastrulac

LateGastrulac

2-Somitec

Totally induced embryos Dead embryos Surviving embryos GFP positive after induction Malformations (%)e Reduced otic vesicle Loss of otic vesicle Reduced eye size Loss of eyes

145 84 61

150 43 107

180 26 154 150

137 26 111 91

67 3 64 60

84 15 69 67

17% 0% 13% 0%

19% 0% 16% 0%

-

13% 23% 18% 21%

-

15% 6% 11% 5%

21% 2% 14% 5%

18% 2% 15% 5%

a,b

The concentrations for mRNA (a200 ng/␮l) and DNA (bCMV:Aes; 200 ng/␮l) were adjusted to result in similar percentages of reduzed size phenotypes. c The GFP:HSE:Aes construct was injected at 30ng/␮l. GFP negative embryos were selected for heat treatment (2h/39°C). d These data correspond to those presented in Table 1 and are included for comparison. e Percentages were calculated from surviving embryos with GFP expression after induction.

nevertheless, interferes with binding of this major repressive component to Groucho multiprotein complexes. The derepression effect of Aes/Grg5 is further enhanced by the absence of the WD-40 repeats, responsible for most protein–protein interactions (Chen and Courey, 2000). This dominant-negative function has successfully been applied to various experimental systems (Roose et al., 1998; Ren et al., 1999). Otic induction in vertebrates starts at mid-gastrula stage, initiated by factors from hindbrain and subjacent mesoderm (reviewed in Whitfield et al., 2002; Riley and Phillips, 2003). This process occurs in a series of steps that are likely to be regulated independently (Groves and Bronner-Fraser, 2000; Liu et al., 2003). The best can-

didates for otic-inducing factors are members of the fibroblast growth factor (Fgf) family of peptide ligands. In particular, Fgf8 and Fgf3 are sufficient to ectopically induce otic vesicles (Lombardo et al., 1998; Vendrell et al., 2000; Bajoghli et al., 2004; Phillips et al., 2004). In zebrafish acerebellar mutants, which contain a disrupted Fgf8 gene, the otic vesicles show a reduced size (Whitfield et al., 1996; Reifers et al., 1998). The same phenotype has been observed in morpholino knockdown experiments for Fgf3 (Phillips et al., 2001; Leger and Brand, 2002; Maroon et al., 2002). Another candidate for an otic-inducing factor is Wnt8. In chick, an ortolog of the Wnt8 gene (Wnt8c) is expressed in the hindbrain between the two prospective otic anla-

gen and human Wnt8 can induce a variety of otic markers in chick embryos (Ladher et al., 2000). In zebrafish, the indirect function of Wnt8 in this process has been revealed; disruption of the gene leads to a delay in preotic expression of Fgf8 and Fgf3, thus, indirectly causing otic vesicles of reduced size (Phillips et al., 2004). The Sox gene family also plays an important role in the development of the vertebrate ear (de Martino et al., 2000; Groves and Bronner-Fraser, 2000; Koster et al., 2000; Chiang et al., 2001; Dutton et al., 2001; SaintGermain et al., 2004). For example, overexpression of Sox3 in medaka leads to the formation of ectopic otic vesicle-like structures (Koster et al., 2000), whereas inactivation of Sox10

762 BAJOGHLI ET AL.

Fig. 1.

Fig. 3.

Fig. 2.

Groucho REGULATES OTIC VESICLE OUTGROWTH 763

TABLE 3. Effects of Aes and Tle4 Misexpression on Endogenous Eya1, Six1, and Sox3 Aes misexpression Sox3

Tle4 misexpression

Six1 a

Eya1 a

Eya1 b

c

Ectopic

Repression

Ectopic

Repression

Ectopic

Repression

Ectopic

Repression

0/31

10/31

0/31

7/31

0/26

17/26

18/43

0/43

a

Embryos with Sox3 or Six1 repression in the otic vesicle region. Embryos with Eya1 repression in the somites. c Embryos with ectopic Eya1 in rhombencephalic ectoderm. b

in colourless mutants causes defects in patterning of the otic vesicle and small ears (Whitfield et al., 1996). Pax8 a member of Pax2/5/8 family is one of the earliest markers of preotic development, first detectable at late gastrulation in zebrafish (Pfeffer et al., 1998) and Xenopus (Heller and Brandli, 1999). Loss of both Fgf8 and Fgf3 function prevents Pax8 expression in the ear primordium (Phillips et al., 2001; Leger and Brand, 2002). Although inactivation of Pax8 in mice has no effect on ear development (Mansouri et al., 1998), knock-down of Pax8 in zebrafish delays development of the otic placode (Hans et al., 2004). Nothing is known about the role of Groucho/Tle proteins during ear formation. In this report, therefore, we investigated the function of this gene family in early otic development. For

this purpose, we applied the dominantnegative effect of Aes/Grg5 on other family members. Using the recently developed heat-inducible HSE-system for misexpression of Aes (Bajoghli et al., 2004), we observed hypoplasia of otic vesicles and eyes. These results were complemented by Tle4 overexpression experiments giving opposite phenotypes. In addition, we report that ectopic Aes is able to induce partial axis duplications in medaka embryos.

RESULTS Blocking Groucho Corepressor Function in Otic Vesicles by Misexpression of Aes The equilibrium between activating and repressive functions is of major

importance for developmental decisions and Groucho proteins are key factors in such processes. We tested the effects of these corepressors on otic vesicle formation. As a tool, we used Aes, a truncated Groucho protein, known for its antagonistic functions on other members of the vertebrate Groucho family (Roose et al., 1998; Ren et al., 1999; Lopez-Rios et al., 2003). For misexpression, we chose the heat-inducible HSE system (Bajoghli et al., 2004), allowing ectopic activation of the gene-of-interest in a stage-dependent manner. The embryos were first induced during early gastrulation, and the effects of Aes misexpression were observed 48 hours later. Between 91 and 97% of the surviving embryos developed green fluorescent protein (GFP) fluorescence (Table 1). GFP is used as a marker

Fig. 1. Misexpression of Aes leads to size reduction or loss of otic vesicles. The heat-inducible Aes construct was coinjected with meganuclease enzyme into one- or two-cell stage embryos. All embryos were heat treated at early gastrula (stage 13) for 2 hr at 39°C and analyzed for phenotypes 2 days after induction. A: An embryo with a typical Aes-induced phenotype displays reduced size of the eye (arrowhead) and the otic vesicle (arrow). The midbrain is dramatically reduced, and the heart tube position is shifted by 90° to the affected side. B,Bⴕ: Embryos with smaller otic vesicles (B) exhibited green fluorescent protein (GFP) activity only at the affected side (B⬘; the dotted line demarcates the normal otic vesicle that is GFP-negative). C: An otic vesicle that started splitting. Cⴕ: GFP activity at the indentation indicates the involvement of Aes in this process. D,Dⴕ: An embryo with loss of an otic vesicle (D) in agreement with high levels of GFP activity (D⬘). The heart tube (h) of this embryo also pointed to the side of high misexpression levels. In all embryos, anterior is to the top (dorsal views). h, heart; mb, midbrain; ov, otic vesicle. Scale bar in A ⫽ 100 ␮m for A, 40 ␮m for B,B⬘, 35 ␮m for C,C⬘, 60 ␮m for D,D⬘. Fig. 2. Expression of otic vesicle marker genes after Aes activation. A–K: Dorsal (A–F) and lateral (G–K) views of embryos at stage 25 (19 somites) injected with HSE:Aes and induced at early gastrula stage (B,C,E,F,H,I,K). Anterior is to the top (A–F) or to the left (G–K). A,D: Wild-type expression of Pax2 (A) and Eya1 (D) is shown. B,E: The expression of Pax2 (B) and Eya1 (E) was unaffected in embryos with reduced otic vesicle size. The brackets indicate the different size of the otic vesicle territory in the affected and the normal half of the embryo. C,F: Loss of otic vesicles was associated with lack of Pax2 (C, arrowhead) and Eya1 (F, arrowhead) expression in this region. G: Sox3 is strongly expressed in the rhombencephalic ectoderm of wild-type embryos. H: Misexpression of Aes leads to repression of Sox3 in this region (arrowhead). I: Same embryo as in H, but a different optical plane and an increase of background light makes visible the reduced size of the otic vesicle (marked by a red dotted line). J,K: Overexpression of Aes (K) represses the expression of Six1 in the otic vesicle region. MHB, midbrain– hindbrain boundary. Scale bars ⫽ 50 ␮m in A; 40 ␮m in B–F,J,K; 20 ␮m in G–I. Fig. 3. Misexpression of Tle4 leads to enlargement of otic vesicles. A–F: Dorsal views of embryos at stage 25 (A,B; 19 somites), stage 24 (C,E,F; 16 somites) and stage 22 (D; 9 somites) with Tle4 induced at early gastrula stage; anterior is to the top (A–E) or to the left (F). A,B: Misexpression of Tle4 resulted in embryos with enlarged otic vesicles (A), which correlated with the green fluorescent protein activity in the affected region (B; the dotted line demarcates the normal otic vesicle). C,D: Expression of Eya1 (C) and Pax2 (D) was not affected by the enlargement of the otic vesicle. E: Ectopic otic vesicle-like structures were observed after Eya1 marker gene analysis (arrow; inset, lateral view, anterior to the left, dorsal to the top). F: The lens placode was enlarged (arrow; the dotted line demarcates the eye field), or in rare cases, the eye became lost (arrowhead). Scale bar in A ⫽ 50 ␮m for A,B, 35 ␮m for C, 25 ␮m for D; 40 ␮m for E, 65 ␮m for inset in E, 100 ␮m for F.

764 BAJOGHLI ET AL.

gene in the HSE construct, indicating the efficiency of misexpression in the embryos. As hypothesized, we found dramatic effects on otic development for Aes misexpression. The total number of otic vesicle defects depended on the Aes expression level. Increasing the DNA concentration from 10 to 50 ng/␮l concomitantly raised the appearance of phenotypes from 9 to 32% (Table 1). The most prominent malformation was a size reduction of the otic vesicles; typically, the affected organs were half the size of those of wild-type embryos. In rare cases, they became split or were lost completely. Aes, therefore, negatively affects otic vesicle outgrowth. Consistent with the idea that interference with Groucho protein function affects multiple developmental pathways, we observed additional malformations throughout the embryo, most obvious in the eyes,

Fig. 4.

Fig. 5.

Groucho REGULATES OTIC VESICLE OUTGROWTH 765

the midbrain–hindbrain region, and the heart. The size of the eye was reduced (Fig. 1A, arrowhead), in extreme cases leading to complete absence of the eye. This strong effect on eye development has been described previously for Aes mRNA injection experiments (Lopez-Rios et al., 2003) and, therefore, served as a reference for Aes activity. In the midbrain– hindbrain region, we saw a dramatic reduction of midbrain size (Fig. 1A); in some cases, a quite variable phenotype was observed, which turned out to be caused by a partial axis duplication (see section Induction of Axis Duplications by Aes in the Results section). Furthermore, in Aesinjected embryos, the position of the heart tube appeared altered, pointing straight away from the body axis, either to the right side (Fig. 1A) or the left side of the embryo (Fig. 1D). All described phenotypes showed a strict dependence on Aes dose and appeared at similar ratios (Table 1). We often observed a coincidence of the malformations on the same side of the embryos, either right or left. Typical embryos developed, on the affected side, small eyes and small otic vesicles, a largely reduced midbrain,

and a heart tube pointing to the same side (Fig. 1A). Such embryos appeared at a frequency of 15% for 50 ng/␮l. A comparison of the affected side with the wild-type reference side of the embryo was of particular importance for qualifying size differences of the eyes and the otic vesicles. We thus could unequivocally conclude a size reduction for both sensory organs as a result of Aes misexpression. Further evidence came from analysis of GFP expression patterns. GFP is expressed in a bidirectional manner from the HSE promoter, thus marking Aesmisexpressing cells (Bajoghli et al., 2004). Tracing GFP fluorescence indicated a strict overlap of the described phenotypes with high misexpression levels. Examples for hypoplasia or complete loss of otic vesicles are shown in Figure 1B and 1D, respectively. Splitting of the otic vesicle was observed rarely but also was correlated with GFP expression in an area where the splitting was initiated (Fig. 1C). Later, a complete separation into two vesicles took place (data not shown). The heart tube strictly pointed to the side of high GFP expression (Fig. 1D⬘). Thus, interference with Groucho function generates a reproducible phenotype in the embryo.

Time Dependence of the AesInduced Phenotypes Deviations between the results obtained for the HSE-inducible system, compared with mRNA injections (Lopez-Rios et al., 2003; and J. Lo´pezRı´os, personal communication) suggested a time dependence of the effects of ectopic Aes activity in the embryos. As a major advantage, inducible systems can be activated during different time points, thus covering different windows of competence for developmental decisions. We, therefore, performed a time-course experiment by systematically varying induction from early gastrula to the two-somite stage (Table 2). To qualify earlier stages, we included mRNA (translated immediately) and a DNA expression construct (transcribed during mid-blastula). A reduction of otic vesicle size was observed for all time windows of Aes activation at comparable rates, whereas complete loss was a phenotype preferentially seen for early induction (23% for mRNA injection). Almost identical results, both for size reduction and loss were observed for eye vesicles.

Misexpression of Aes Results in Repression of Six1 and Sox3 in Otic Vesicles Fig. 4. Opposing effects of ectopic Tle/Aes on Eya1 and Sox3 expression. A–I: Lateral views (A,D,G) or dorsal views (B,C,E,F,H,I) of embryos at stage 27 (A,D; 24 somites), stage 24 (B,C,E,F,G,H,I; 16 somites). In B,E, and H, anterior is to the top. Embryos were injected either with the Aes- or with the Tle4-inducible construct. All injected embryos were heat induced at the same time during early gastrula stage (for 39°C/2 hr). D,E,GH: Ectopic Aes leads to reduction of Eya1 in somites (D,E; arrowheads), but overexpression of Tle4 induces ectopic Eya1 in ectoderm (G,H; arrows). F,I: The expression of Sox3 in the lens placode was reduced in Aes-misexpressed embryos (F, arrowhead) in contrast to Tle4-injected embryos (I). ey, eye; lp, lens placode; nl, nasal placode; ov, otic vesicle. Scale bar in A ⫽ 25 ␮m for A,D, 50 ␮m for C,F,H,I, 70 ␮m for B,E, 20 ␮m for G. Fig. 5. A–N: Induction of Aes at early gastrula stage leads to partial axis duplication. A,C–F,L–N: Dorsal views of embryos at stage 26 (A,C–F; 22 somites), stage 16 (L,M; late gastrulation), and stage 15 (N; mid-gastrulation) injected with the Aes-inducible construct at the one-cell stage. All embryos were induced at early gastrula (stage 13) at 39°C for 2 hr. Anterior (except F,I,J, left) is to the top. C,D,L,M: The brackets indicate the length of the duplication. A: An embryo with partial axis duplication in the midbrain–hindbrain region. C–F,H–N: This phenotype was analysed for Pax2 (C–F) and Sox3 (H–N) expression. C–E: Embryos with partial axis duplication display a duplicated midbrain–hindbrain boundary (MHB) and spinal cord neurons. The position and length of axis duplication differed between embryos (compare C and D). C–E,H,I: The axis duplication was observed in the anterior part (C–E,I) as well as the posterior part of the embryos (F,H). F: Embryos with four rows of spinal cord neurons in the tail region (arrow indicating the beginning of the duplication). E: In some cases, otic vesicles were observed between the two neural tubes (arrow). J: We observed embryos with sequential branching and fusion of the neural tube (the bracket and numbers mark the duplicated regions). L–N: Confirming the phenotype at later stages, analysis of embryos at late gastrula also revealed axis duplications. Strong green fluorescent protein–positive embryos were selected for whole-mount in situ hybridization against Sox3. L–N: Variations in the extent of the duplicated region became visible, but head ectoderm was never affected. ey; eye; fb, forebrain; hec, head ectoderm; MHB, midbrain–hindbrain boundary; ov; otic vesicle; wt, wild-type; y, yolk. Scale bar in A ⫽ 100 ␮m for A,B,J, 80 ␮m for C–E, 60 ␮m for F–H, 120 ␮m for I,K–N.

To learn more about the molecular mechanisms, we investigated the effect of ectopic Aes on otic placode marker genes. We analyzed the expression of Pax2, Eyes Absent 1 (Eya1), Six1, and Sox3. In all vertebrates, Pax2 is expressed in preotic cells during early somitogenesis (Pfeffer et al., 1998; Heller and Brandli, 1999) until organogenesis (Whitfield et al., 2002). Eya1 and Six1 are expressed early on in the preplacodal domain and later show distinct expression patterns in the otic placode (Sahly et al., 1999; Koster et al., 2000; Bessarab et al., 2004). In medaka, Sox3 expression is found in all sensory placodes (Koster et al., 2000). These four marker genes, therefore, can be used to analyze formation of the otic placode. Aes-injected embryos were first analyzed for Pax2 expression by wholemount in situ hybridization (Fig.

766 BAJOGHLI ET AL.

2B,C). In general, the Pax2 expression pattern appeared normal in otic vesicles with reduced size (Fig. 2B), although some early induced embryos developed weak repression of this gene (data not shown). On the contrary, all endogenous Pax2 in this region disappeared for embryos, which displayed complete loss of an otic vesicle (Fig. 2C, arrowhead). Eya1 expression showed the same behavior upon size reduction or loss of otic vesicles (Fig. 2E,F). Of interest, in 65% of the embryos Eya1 expression in the somites was reduced (17 of 26; Table 3; Fig. 4D,E, arrowheads). On the contrary, Six1 and Sox3 expression was strongly repressed in otic vesicles of reduced size at a frequency of 23% (7 of 31) and 32% (10 of 31), respectively (see Table 3). In Figure 2K,H, examples of this repression are shown. In some embryos, Sox3 expression in the lens placodes also was reduced (4 of 31; Fig. 4F). Taken together, hypoplasia of otic vesicles induced by Aes is accompanied by normal expression for Eya1, whereas Sox3 and Six1 expression is reduced.

Overexpression of Tle4 Leads to an Enlargement of the Otic Vesicle and Ectopic Activation of Eya1 Our strategy was based on the interference of Aes with Groucho protein function. A clear prediction was, therefore, an opposite effect upon overexpression of a full-length Groucho family member. We selected Tle4 for this experiment, based on the expression of this gene in Xenopus otic vesicles (Molenaar et al., 2000). Using the HSE system, we induced Tle4 misexpression during early gastrulation. Upon low concentrations (10 ng/ ␮l), 7% of the embryos (6 of 81) showed enlarged otic vesicles. Higher Tle4 concentrations increased the percentage of this phenotype to 13% (5 of 38) for 20 ng/␮l and to 15% (17 of 111) for 60 ng/␮l. Regions of high Tle4 misexpression correlated with this phenotype, as concluded from GFP marker gene expression (Fig. 3A,B). Most importantly, reduction of otic vesicle size was not detected in a single case, whereas complete loss of the eye appeared as a rare phenotype (Fig. 3F,

arrowhead). In addition, we observed in few embryos an enlargement of the lens (2 of 43; Fig. 3F, arrow), but at a low frequency compared with the alterations of the otic vesicles. When coinjecting 20 ng/␮l Tle4 together with 20 ng/␮l Aes (HSE-constructs induced during early gastrulation), no enlarged otic vesicles were detected any more (n ⫽ 70) and, similarly, the reduced otic vesicle phenotype of Aes was largely reduced due to the presence of the agonist (data not shown). These data further strengthen the opposing effects of both proteins on otic vesicle outgrowth. Tle1 overexpression experiments resulted in enlargement of the eye field, in addition Tle1 was shown to induce patches of ectopic Six3 and Pax6 expression, indicating the potential of Groucho proteins to assist in the formation of ectopic eye structures (Lopez-Rios et al., 2003). Using the HSE induction system, we could detect the formation of additional otic vesicles upon Tle4 overexpression. For identification, we used Eya1 as a marker (Fig. 3E, arrow). Contrary to a splitting of the vesicles, which we observed for Aes misexpression (Fig. 1C; adjacent vesicles), the position of these ectopic sensory organs seen for Tle4 overexpression was quite distant from the endogenous vesicles (Fig. 3E, inset; ectopic vesicle at the dorsal side of hindbrain). As expected, Pax2 and Eya1 expression within the enlarged otic vesicles was normal (Fig. 3C,D). Contrary to repression of Eya1 induced by Aes, Tle4 overexpression led to ectopic Eya1 in rhombencephalic ectoderm at a frequency of 42% (18 of 43; Fig. 4G,H). Taken together, these data strongly support a positive effect of full-length Groucho proteins for otic vesicle outgrowth in contrast to Aes and confirm the size determining effect of this corepressor in development of the otic vesicle.

Induction of Axis Duplications by Aes In several cases, Aes-injected embryos developed a variable phenotype in neural tube-derived structures. In severely affected embryos, a duplicated midbrain– hindbrain region became visible (Fig. 5A). To get a clearer pic-

ture, we analyzed the expression of the midbrain– hindbrain boundary marker Pax2. In situ hybridization data for this gene clearly indicated a duplication of the anterior neural tube (Fig. 5C–F). The two neural tubes were completely separated, but otherwise showed a normal expression pattern for Pax2, a marker for differentiating neurons along the neural tube. Of interest, the duplication was partial, resulting in a fusion of the neural tube posteriorly. We did not see forebrain structures affected, but otherwise observed branching at different positions along the axis (Fig. 5F). The length of the duplicated region differed between the embryos (Fig. 5C,D). In extreme cases, where the two neural tubes were clearly separated from each other, we even observed otic vesicles at the expected position between the two neural tubes (Fig. 5E, arrow). Because misexpression with the HSE promoter generates a peak of Aes misexpression, slowly declining with time, we reasoned that position of the partial duplication of the axis might depend on the time of induction. We, therefore, induced embryos at different time points during gastrulation and analyzed for axis duplication by whole-mount in situ hybridization against Pax2. We did not detect any time dependence of this phenotype between early- and mid-gastrula, but embryos induced during late gastrula did not display partial axis duplications any more (data not shown). mRNA injections of Aes resulted in axis duplications at a very low frequency (2 of 61; data not shown). To back up the results for Pax2, we analyzed the embryos with Sox3, a marker for neuroectoderm along the entire body axis (Koster et al., 2000). Labeling the neural tube with Sox3 even better visualized the partial axis duplication (Fig. 5H–J). Applying the higher resolution of this marker, we observed embryos where splitting and fusion of the neural tube appeared several times along the axis (Fig. 5J). Of interest, we did not detect a single case of a complete axis duplication, suggesting that after the decline of Aes misexpression, axis induction continues normally, resulting in a fusion of the duplicated structures. To examine earlier stages of axis

Groucho REGULATES OTIC VESICLE OUTGROWTH 767

induction, we analyzed Sox3 expression during late gastrulation (Aes induced at early gastrula). The duplication could clearly be detected at this early stage (Fig. 5L–N) and appeared in GFP-expressing embryos at a frequency of 36% (17 of 47). The extent and distance of the duplicated neuroectodermal structures differed dramatically (compare Fig. 5L–N). Therefore, Aes transiently interferes with normal axis induction.

DISCUSSION The study of mutants is the key method for analysis of gene function. Unfortunately, several gene families exhibit a high degree of functional overlap. A typical example is the Groucho/Tle family. The amino acid sequences of these proteins are highly conserved between different Groucho family members, in particular in the C-terminal WD-40 repeats, the major protein–protein interaction domain (Choudhury et al., 1997; Fisher and Caudy, 1998; Li, 2000). This finding explains why diverse Groucho/Tle members often interact with the same transcription factors (Choi et al., 1999; Ren et al., 1999; Eberhard et al., 2000). Therefore, the analysis of mutants or morpholino knockdown experiments will only cover subsets of the complete Groucho corepressor function. The Groucho-related protein Aes/Grg5 lacks the WD-40 repeats, thus blocking many protein–protein interactions. Due to deviations in the amino acid sequence, this protein does not interact with HDAC1 (Brantjes et al., 2001), a major source of repressive function conferred by Groucho proteins (Chen et al., 1999; Brantjes et al., 2001). With its N-terminal Q-domain, Aes/Grg5 nevertheless can multimerize with other Groucho proteins, thus explaining the dominant-negative effect seen for this short member of the Groucho gene family (Roose et al., 1998; Ren et al., 1999). By overexpressing Aes in medaka embryos, we applied the suppressive effect on all Groucho family members at the same time. This strategy resulted in a typical phenotype showing eyes, midbrain, and otic vesicles of reduced size. A critical control for this dominant-negative approach was the application of full-length Tle proteins.

Shifting the equilibrium toward the repressive function, overexpression of Tle4 indeed resulted in opposite phenotypes.

Groucho/Tle Proteins Strongly Affect Otic Vesicle Development The role of Groucho proteins in otic vesicle development has not been investigated to date. Expression analysis has shown that in Xenopus, Grg4, and Grg5 are active during early otic vesicle development (Molenaar et al., 2000). In zebrafish Gro1 (a Tle3 homologue) transcripts were also detected in this region (Wu¨lbeck and CamposOrtega, 1997). In medaka, neither Tle1, Tle3, or Tle4 are expressed in early otic vesicles (Lopez-Rios et al., 2003), but we have data that two new full-length Groucho family members are activated in this region (N. Aghaallaei, B. Bajoghli, I. Walter, and T. Czerny, submitted). When interfering with Groucho protein function, we observed two different phenotypes in ear development, namely absence or reduced size of the otic vesicle. Loss of this structure in principle could be the extreme version of a reduced size phenotype, but differences in time dependence indicate an involvement of Groucho protein function in two separate developmental pathways. Misexpression of Aes by mRNA injection generated embryos with loss of otic vesicles at a high frequency (23%). The same phenotype was rarely observed when Aes was induced at mid-gastrula (2%) but was not detected for Aes activation at later stages. On the contrary, size reduction of the otic vesicle was observed at roughly the same frequency (between 13 and 21%) for all Aes activation times tested (up to the two-somite stage). In addition, unspecific perturbation effects of surrounding tissue, preventing proper induction, cannot be excluded for loss of otic vesicles. Take together, an early involvement of Groucho protein function is suggested for preplacodal stages (initiation phase), whereas a continuously acting system determining otic vesicle size is sensitive to Aes during all stages (maintenance phase). To characterize the molecular basis of these phenotypes, we analyzed the ex-

pression of otic vesicle marker genes. In embryos lacking an otic vesicle due to Aes misexpression, transcripts for Pax2 were lost in this region, whereas otherwise Pax2 expression was almost normal in these embryos. Pax2 is expressed early in otic placodes (three-somite stage) and has an important regulator function during subsequent otic development (for review, see Riley and Phillips, 2003). Nevertheless, disruption of mouse Pax2 does not alter otic induction (Torres et al., 1996) and does not affect Eya1 and Six1 expression (Xu et al., 1999; Zheng et al., 2003; Burton et al., 2004). Eya1 is expressed earlier (end of gastrulation) than Pax2 and affects the early inductive signalling events involved in otic vesicle formation, as demonstrated by Eya1-deficient mice (Xu et al., 1999). Although not affected in formed otic vesicles, Aes misexpression resulted in a dramatic downregulation of Eya1 in the somites. Furthermore, ectopic expression was detected in rhombencephalic ectoderm upon Tel4 overexpression, thus clearly indicating the sensitivity of this gene to Groucho activity, which similarly might play a role in early induction of otic vesicles. Even more striking was the reduction of Six1 expression for vesicles of reduced size. A highly conserved network of interaction has been described for Pax-Six-Eya-Dach genes acting during organogenesis (Bonini et al., 1997; Heanue et al., 1999; Li et al., 2003). Coexpression of these genes during otic development suggests functional interactions, strongly supported by gene inactivation experiments. Tle/Aes misexpression, therefore, might critically affect correct formation of this gene network during otic vesicle formation. Furthermore, Sox3 expression in this region appeared strongly reduced. Sox3 has been implicated previously in sensory organ development (Koster et al., 2000). Misexpression of Sox3 in medaka embryos leads to the formation of ectopic otic vesicle structures, which express the markers Pax2 and Eya1. Overexpression of a dominant-negative form of Sox3, which encodes the DNA-binding HMG-domain without transactivation domain, leads to dysgenesis of endogenous sensory organs (Koster et al., 2000). Thus, Sox3 is also a good candidate to control both initiation and size regulation of these organs.

768 BAJOGHLI ET AL.

Groucho/Tle proteins exhibit their function by binding to other transcription factors. The question, therefore, arises, which up-stream regulators could confer the Aes effect on Six1, Eya1, and Sox3. Among the genes expressed during otic vesicle development, members of the Pax2/5/8 subfamily have been identified as interaction partners for Groucho/Tle proteins (Eberhard et al., 2000; Cai et al., 2003). As discussed before, Pax2 inactivation does not affect Eya1 or Six1 expression and Pax5 is expressed later during inner ear development. On the contrary, Pax8 can be detected in preotic cells during late gastrulation in all vertebrates, representing one of the earliest markers in this region (Pfeffer et al., 1998; Heller and Brandli, 1999). Loss of Pax8 function in the mouse does not exhibit an otic phenotype (Mansouri et al., 1998), but knockdown of Pax8 in zebrafish delays otic placode development (Hans et al., 2004). This variability of phenotypes could be explained by the high degree of functional overlap between members of the Pax2/5/8 subfamily. Pax8 and Pax2 are sequentially expressed, and combined inactivation of both genes in zebrafish results in complete loss of otic vesicles (Hans et al., 2004). Interaction with Groucho proteins has been described also for members of the Six-family (Kobayashi et al., 2001; Zhu et al., 2002; Lopez-Rios et al., 2003). Thus, Six3 could be identified to confer the effect of Aes/Tle misexpression on eye development, resulting in similar phenotypes compared with ear formation (see below). Here, autoregulation of Six3 could be made responsible for the effect on its own expression (Zhu et al., 2002) and might similarly affect Six1. Therefore, currently Six1 is the best candidate for a critical target of Groucho protein function during otic vesicle formation. Mechanistically, the effect of Groucho proteins on the formation of otic vesicles and eyes is quite unexpected. Down-regulation of the repressing activity by Aes misexpression leads to reduced size or complete loss, whereas enhanced repression results in hyperplasia of these sense organs. In a linear genetic pathway, the effect of Groucho proteins, therefore, could be transmitted only indirectly by another factor, which itself would suppress

sensory organ development. In principle, Tcf/Lef transcription factors would be good candidates for transmission of a Groucho effect. They are known to interact with Groucho family members (Brantjes et al., 2001), making the canonical wnt-signalling pathway dependent on the Tle/Aes equilibrium. In zebrafish, wnt signalling has been proposed to play a role in otic placode induction, by regulating expression of Fgf3 and Fgf8 in the hindbrain. Furthermore, injection of Wnt antagonists (GSK3b; Saint-Germain et al., 2004) and Wnt8 morpholino oligos (Phillips et al., 2004) resulted in otic vesicles of reduced size. Therefore, Tcf/Lef proteins play a positive role in the proposed pathway, the phenotypic consequences for Tle/Aes misexpression nevertheless would be contradictory to our results. Provided that Pax2/5/8 and Six genes also positively affect otic vesicle outgrowth, the same criterion would argue against these gene families. Interestingly Six genes are regulated by a feedback loop (Zhu et al., 2002) and quite surprisingly Tlecorepressors binding to Six proteins enhance gene expression. It will be revealing, therefore, to study in detail the mechanism of Groucho protein function during sensory organ development.

Parallel Pathways of Otic Vesicle and Eye Development Groucho/Tle family members have been identified previously to participate in eye development (Kobayashi et al., 2001; Zhu et al., 2002; LopezRios et al., 2003). Gain-of-function analysis of Tle1 resulted in ectopic expression of the eye retina marker Rx2 (Lopez-Rios et al., 2003). Similarly, we observed an enlargement of the eye field upon Tle4 overexpression in our experiments. Blocking Groucho/Tle function by overexpression of the dominant-negative Aes led to reduction or loss of the eye (Lopez-Rios et al., 2003). A detailed analysis of the time dependence in our experiments indicates that loss of the eye is a phenotype for early Aes activation (most prominent for mRNA injection), whereas a size reduction is seen at

similar frequencies up to the twosomite stage. We found extensive similarities between otic vesicle and eye phenotypes in our experiments. Typical embryos with unilateral Aes misexpression showed size reduction of both eye and otic vesicles, similarly Tle4 overexpression resulted in the opposite phenotypes for both organs. Unexpected were the highly corresponding statistics for these malformations, when analyzed for different induction times. Similar ratios, both for complete loss as well as for size reduction were seen for all time points tested. These data suggest equal pathways for early development of both organs. After initiation of eye development, a complicated network of genetic interactions starts. Six3 and Pax6 are key players in this process (Hill et al., 1991; Ton et al., 1991; Quiring et al., 1994; Oliver et al., 1995; Loosli et al., 1999), and Six genes have been identified to confer the Tle corepressor effect (Kobayashi et al., 2001; Zhu et al., 2002; Lopez-Rios et al., 2003). Our experiments suggest that the basic architecture of the developmental pathways for both sensory organs have to be similar, strongly suggesting an important function for Six family members in otic development. The exact targets still have to be identified, but our results support the prominent role of Groucho corepressor proteins during sensory organ formation.

Induction of Aes During Gastrulation Leads to Partial Axis Duplication The genetic network controlling the induction of the anterior/posterior axis in vertebrates is well understood (for reviews, see De Robertis et al., 2000; Schier, 2001). Gain-of-function analysis has identified multiple components of the Wnt-signalling pathway with the potential to initiate this process (Moon and Kimelman, 1998). Downstream effectors of this pathway are Tcf/Lef transcription factors, which interact with ␤-catenin in the nucleus (Behrens et al., 1996; Molenaar et al., 1996; van de Wetering et al., 1997). In the absence of ␤-catenin activation, Tcf/Lef transcription factors are transformed into transcriptional repressors through interaction

Groucho REGULATES OTIC VESICLE OUTGROWTH 769

with Groucho/Tle proteins (Roose et al., 1998; Brantjes et al., 2001). Thus, Grg4 inhibits axis formation, whereas Aes/Grg5 enhances the axis-inducing potential of misexpressed effectors of wnt signalling (Roose et al., 1998). Dorsal injection of Aes/Grg5 mRNA alone into Xenopus embryos did not result in significant phenotypes (Roose et al., 1998). We rarely observed a duplication of the endogenous axis after mRNA injection (3.2%) but at a frequency of 36% upon Aes activation during early gastrulation. The duplications mainly affected the midbrain and the hindbrain but, in some cases, also more posterior regions of the spinal chord. We never saw duplicated heads. In all cases, the branched neural tubes fused again at a more posterior position, consistent with a transient effect of Aes after heat-shock induction. Taken together, misexpression of Aes/Grg5 through application of a heat-inducible promoter has shown phenotypes in sensory organ development and organizer formation. During otic vesicle and eye formation, Groucho proteins seem to control both initiation as well as size regulation. The next step will be to identify the transcription factors conferring the effects of these corepressors during otic vesicle development.

EXPERIMENTAL PROCEDURES Fish Strains and Maintenance Medaka embryos and adults of the Cab inbred strain were used for all experiments (Loosli et al., 2000). Adult fish were kept under a reproduction regimen (14 hr light/10 hr dark) at 26°C. Embryos were collected daily immediately after spawning. Embryonic stages were determined according to Iwamatsu (2004).

DNA, RNA Injection, and Heat-Shock Treatment Medaka Aes (Lopez-Rios et al., 2003) and mouse Tle4 (Eberhard et al., 2000) cDNAs were cloned into the HSE expression construct (Bajoghli et al., 2004). DNA was coinjected with the I-SceI meganuclease enzyme as

described (Thermes et al., 2002) into single blastomeres at the one- to twocell stage. After injection, the embryos were incubated at 28°C. Before heatshock treatment, the embryos were selected for background activity under the fluorescence microscope. Heat treatment was performed for 2 hr at 39°C as described previously (Bajoghli et al., 2004). After heat shock, the embryos were incubated at 28°C. In all transient experiments, background GFP expression was seen in 0 –10% of the injected embryos, which were eliminated from further analysis. As a control group, 100 embryos were injected with GFP:HSE:Aes at 30 ng/␮l. Twenty-four hours after injection, five embryos exhibited background GFP expression. The remaining embryos (n ⫽ 95) were incubated at 28°C until hatching. All embryos were GFP-negative and did not show any phenotypes (data not shown). For DNA injections (CMV:Aes), a pCS2 expression construct containing the medaka Aes cDNA was used. Messenger RNA of Aes was in vitro transcribed using the SP6 Message Machine Kit (Ambion). The CMV:Aes DNA and the mRNA were injected in 1⫻ Yamamoto buffer.

Whole-Mount In Situ Hybridization Embryos were fixed in 4% paraformaldehyde/2PTW (2 ⫻ PBS at pH 7.5, 0.1% Tween-0). The chorion of embryos beyond gastrula stages was removed. Whole-mount in situ hybridization was performed at 65°C as described (Quiring et al., 2004) using DIG-labeled probes for Sox3, Eya1, Six1, and Pax2 (Koster et al., 2000).

ACKNOWLEDGMENTS We thank Viktoriya Titova for cloning of DNA constructs, Jochen Wittbrodt and Felix Loosli for providing in situ probes (Sox3, Six1, Pax2, Eya1), Javiar Lo´pez-Rı´os for medaka Aes cDNA and Dirk Eberhard for mouse Tle4 cDNA. The work was supported by the Austrian Genome Project GEN-AU.

REFERENCES Bajoghli B, Aghaallaei N, Heimbucher T, Czerny T. 2004. An artificial promoter construct for heat-inducible misexpres-

sion during fish embryogenesis. Dev Biol 271:416 –430. Behrens J, von Kries JP, Kuhl M, Bruhn L, Wedlich D, Grosschedl R, Birchmeier W. 1996. Functional interaction of betacatenin with the transcription factor LEF-1. Nature 382:638 –642. Bessarab DA, Chong SW, Korzh V. 2004. Expression of zebrafish six1 during sensory organ development and myogenesis. Dev Dyn 230:781–786. Bonini NM, Bui QT, Gray-Board GL, Warrick JM. 1997. The Drosophila eyes absent gene directs ectopic eye formation in a pathway conserved between flies and vertebrates. Development 124:4819 – 4826. Brantjes H, Roose J, van De Wetering M, Clevers H. 2001. All Tcf HMG box transcription factors interact with Grouchorelated co-repressors. Nucleic Acids Res 29:1410 –1419. Burton Q, Cole LK, Mulheisen M, Chang W, Wu DK. 2004. The role of Pax2 in mouse inner ear development. Dev Biol 272:161–175. Cai Y, Brophy PD, Levitan I, Stifani S, Dressler GR. 2003. Groucho suppresses Pax2 transactivation by inhibition of JNK-mediated phosphorylation. Embo J 22:5522–5529. Chen G, Courey AJ. 2000. Groucho/TLE family proteins and transcriptional repression. Gene 249:1–16. Chen G, Fernandez J, Mische S, Courey AJ. 1999. A functional interaction between the histone deacetylase Rpd3 and the corepressor groucho in Drosophila development. Genes Dev 13:2218 –2230. Chiang EF, Pai CI, Wyatt M, Yan YL, Postlethwait J, Chung B. 2001. Two sox9 genes on duplicated zebrafish chromosomes: expression of similar transcription activators in distinct sites. Dev Biol 231:149 –163. Choi CY, Kim YH, Kwon HJ, Kim Y. 1999. The homeodomain protein NK-3 recruits Groucho and a histone deacetylase complex to repress transcription. J Biol Chem 274:33194 –33197. Choudhury BK, Kim J, Kung HF, Li SS. 1997. Cloning and developmental expression of Xenopus cDNAs encoding the Enhancer of split groucho and related proteins. Gene 195:41–48. de Martino S, Yan YL, Jowett T, Postlethwait JH, Varga ZM, Ashworth A, Austin CA. 2000. Expression of sox11 gene duplicates in zebrafish suggests the reciprocal loss of ancestral gene expression patterns in development. Dev Dyn 217: 279 –292. De Robertis EM, Larrain J, Oelgeschlager M, Wessely O. 2000. The establishment of Spemann’s organizer and patterning of the vertebrate embryo. Nat Rev Genet 1:171–181. Dutton KA, Pauliny A, Lopes SS, Elworthy S, Carney TJ, Rauch J, Geisler R, Haffter P, Kelsh RN. 2001. Zebrafish colourless encodes sox10 and specifies non-ectomesenchymal neural crest fates. Development 128:4113–4125.

770 BAJOGHLI ET AL.

Eberhard D, Jimenez G, Heavey B, Busslinger M. 2000. Transcriptional repression by Pax5 (BSAP) through interaction with corepressors of the Groucho family. EMBO J 19:2292–2303. Fisher AL, Caudy M. 1998. Groucho proteins: transcriptional corepressors for specific subsets of DNA-binding transcription factors in vertebrates and invertebrates. Genes Dev 12:1931–1940. Groves AK, Bronner-Fraser M. 2000. Competence, specification and commitment in otic placode induction. Development 127:3489 –3499. Hans S, Liu D, Westerfield M. 2004. Pax8 and Pax2a function synergistically in otic specification, downstream of the Foxi1 and Dlx3b transcription factors. Development 131:5091–5102. Hartley DA, Preiss A, Artavanis-Tsakonas S. 1988. A deduced gene product from the Drosophila neurogenic locus, enhancer of split, shows homology to mammalian G-protein beta subunit. Cell 55: 785–795. Heanue TA, Reshef R, Davis RJ, Mardon G, Oliver G, Tomarev S, Lassar AB, Tabin CJ. 1999. Synergistic regulation of vertebrate muscle development by Dach2, Eya2, and Six1, homologs of genes required for Drosophila eye formation. Genes Dev 13:3231–3243. Heller N, Brandli AW. 1999. Xenopus Pax2/5/8 orthologues: novel insights into Pax gene evolution and identification of Pax-8 as the earliest marker for otic and pronephric cell lineages. Dev Genet 24: 208 –219. Hill RE, Favor J, Hogan BL, Ton CC, Saunders GF, Hanson IM, Prosser J, Jordan T, Hastie ND, van Heyningen V. 1991. Mouse small eye results from mutations in a paired-like homeobox-containing gene. Nature 354:522–525. Iwamatsu T. 2004. Stages of normal development in the medaka Oryzias latipes. Mech Dev 121:605–618. Jimenez G, Paroush Z, Ish-Horowicz D. 1997. Groucho acts as a corepressor for a subset of negative regulators, including Hairy and Engrailed. Genes Dev 11:3072– 3082. Kobayashi M, Nishikawa K, Suzuki T, Yamamoto M. 2001. The homeobox protein Six3 interacts with the Groucho corepressor and acts as a transcriptional repressor in eye and forebrain formation. Dev Biol 232:315–326. Koster RW, Kuhnlein RP, Wittbrodt J. 2000. Ectopic Sox3 activity elicits sensory placode formation. Mech Dev 95:175– 187. Ladher RK, Anakwe KU, Gurney AL, Schoenwolf GC, Francis-West PH. 2000. Identification of synergistic signals initiating inner ear development. Science 290: 1965–1967. Leger S, Brand M. 2002. Fgf8 and Fgf3 are required for zebrafish ear placode induction, maintenance and inner ear patterning. Mech Dev 119:91–108. Leon C, Lobe CG. 1997. Grg3, a murine Groucho-related gene, is expressed in the developing nervous system and in

mesenchyme-induced epithelial structures. Dev Dyn 208:11–24. Li SS. 2000. Structure and function of the Groucho gene family and encoded transcriptional corepressor proteins from human, mouse, rat, Xenopus, Drosophila and nematode. Proc Natl Sci Counc Repub China B 24:47–55. Li X, Oghi KA, Zhang J, Krones A, Bush KT, Glass CK, Nigam SK, Aggarwal AK, Maas R, Rose DW, Rosenfeld MG. 2003. Eya protein phosphatase activity regulates Six1-Dach-Eya transcriptional effects in mammalian organogenesis. Nature 426:247–254. Liu D, Chu H, Maves L, Yan YL, Morcos PA, Postlethwait JH, Westerfield M. 2003. Fgf3 and Fgf8 dependent and independent transcription factors are required for otic placode specification. Development 130:2213–2224. Lombardo A, Isaacs HV, Slack JM. 1998. Expression and functions of FGF-3 in Xenopus development. Int J Dev Biol 42: 1101–1107. Loosli F, Winkler S, Wittbrodt J. 1999. Six3 overexpression initiates the formation of ectopic retina. Genes Dev 13:649 – 654. Loosli F, Koster RW, Carl M, Kuhnlein R, Henrich T, Mucke M, Krone A, Wittbrodt J. 2000. A genetic screen for mutations affecting embryonic development in medaka fish (Oryzias latipes). Mech Dev 97:133–139. Lopez-Rios J, Tessmar K, Loosli F, Wittbrodt J, Bovolenta P. 2003. Six3 and Six6 activity is modulated by members of the groucho family. Development 130:185– 195. Mansouri A, Chowdhury K, Gruss P. 1998. Follicular cells of the thyroid gland require Pax8 gene function. Nat Genet 19: 87–90. Maroon H, Walshe J, Mahmood R, Kiefer P, Dickson C, Mason I. 2002. Fgf3 and Fgf8 are required together for formation of the otic placode and vesicle. Development 129:2099 –2108. Miyasaka H, Choudhury BK, Hou EW, Li SS. 1993. Molecular cloning and expression of mouse and human cDNA encoding AES and ESG proteins with strong similarity to Drosophila enhancer of split groucho protein. Eur J Biochem 216:343– 352. Molenaar M, van de Wetering M, Oosterwegel M, Peterson-Maduro J, Godsave S, Korinek V, Roose J, Destree O, Clevers H. 1996. XTcf-3 transcription factor mediates beta-catenin-induced axis formation in Xenopus embryos. Cell 86:391– 399. Molenaar M, Brian E, Roose J, Clevers H, Destree O. 2000. Differential expression of the Groucho-related genes 4 and 5 during early development of Xenopus laevis. Mech Dev 91:311–315. Moon RT, Kimelman D. 1998. From cortical rotation to organizer gene expression: toward a molecular explanation of axis specification in Xenopus. Bioessays 20: 536 –545.

Oliver G, Mailhos A, Wehr R, Copeland NG, Jenkins NA, Gruss P. 1995. Six3, a murine homologue of the sine oculis gene, demarcates the most anterior border of the developing neural plate and is expressed during eye development. Development 121:4045–4055. Pfeffer PL, Gerster T, Lun K, Brand M, Busslinger M. 1998. Characterization of three novel members of the zebrafish Pax2/5/8 family: dependency of Pax5 and Pax8 expression on the Pax2.1 (noi) function. Development 125:3063–3074. Pflugrad A, Meir JY, Barnes TM, Miller DM, 3rd. 1997. The Groucho-like transcription factor UNC-37 functions with the neural specificity gene unc-4 to govern motor neuron identity in C. elegans. Development 124:1699 –1709. Phillips BT, Bolding K, Riley BB. 2001. Zebrafish fgf3 and fgf8 encode redundant functions required for otic placode induction. Dev Biol 235:351–365. Phillips BT, Storch EM, Lekven AC, Riley BB. 2004. A direct role for Fgf but not Wnt in otic placode induction. Development 131:923–931. Quiring R, Walldorf U, Kloter U, Gehring WJ. 1994. Homology of the eyeless gene of Drosophila to the Small eye gene in mice and Aniridia in humans. Science 265:785–789. Quiring R, Wittbrodt B, Henrich T, Ramialison M, Burgtorf C, Lehrach H, Wittbrodt J. 2004. Large-scale expression screening by automated whole-mount in situ hybridization. Mech Dev 121:971– 976. Reifers F, Bohli H, Walsh EC, Crossley PH, Stainier DY, Brand M. 1998. Fgf8 is mutated in zebrafish acerebellar (ace) mutants and is required for maintenance of midbrain-hindbrain boundary development and somitogenesis. Development 125: 2381–2395. Ren B, Chee KJ, Kim TH, Maniatis T. 1999. PRDI-BF1/Blimp-1 repression is mediated by corepressors of the Groucho family of proteins. Genes Dev 13:125– 137. Riley BB, Phillips BT. 2003. Ringing in the new ear: resolution of cell interactions in otic development. Dev Biol 261:289 –312. Roose J, Molenaar M, Peterson J, Hurenkamp J, Brantjes H, Moerer P, van de Wetering M, Destree O, Clevers H. 1998. The Xenopus Wnt effector XTcf-3 interacts with Groucho-related transcriptional repressors. Nature 395:608 –612. Sahly I, Andermann P, Petit C. 1999. The zebrafish eya1 gene and its expression pattern during embryogenesis. Dev Genes Evol 209:399 –410. Saint-Germain N, Lee YH, Zhang Y, Sargent TD, Saint-Jeannet JP. 2004. Specification of the otic placode depends on Sox9 function in Xenopus. Development 131:1755–1763. Schier AF. 2001. Axis formation and patterning in zebrafish. Curr Opin Genet Dev 11:393–404. Schmidt CJ, Sladek TE. 1993. A rat homolog of the Drosophila enhancer of split

Groucho REGULATES OTIC VESICLE OUTGROWTH 771

(groucho) locus lacking WD-40 repeats. J Biol Chem 268:25681–25686. Sharief FS, Tsoi SC, Li SS. 1997. cDNA cloning and genomic organization of enhancer of split groucho gene from nematode Caenorhabditis elegans. Biochem Mol Biol Int 43:327–337. Stifani S, Blaumueller CM, Redhead NJ, Hill RE, Artavanis-Tsakonas S. 1992. Human homologs of a Drosophila Enhancer of split gene product define a novel family of nuclear proteins. Nat Genet 2:119 –127. Thermes V, Grabher C, Ristoratore F, Bourrat F, Choulika A, Wittbrodt J, Joly JS. 2002. I-SceI meganuclease mediates highly efficient transgenesis in fish. Mech Dev 118:91–98. Ton CC, Hirvonen H, Miwa H, Weil MM, Monaghan P, Jordan T, van Heyningen V, Hastie ND, Meijers-Heijboer H, Drechsler M, et al. 1991. Positional cloning and characterization of a paired boxand homeobox-containing gene from the aniridia region. Cell 67:1059 –1074.

Torres M, Gomez Pardo E, Gruss P. 1996. Pax2 contributes to inner ear patterning and optic nerve trajectory. Development 122:3381–3391. Valentine SA, Chen G, Shandala T, Fernandez J, Mische S, Saint R, Courey AJ. 1998. Dorsal-mediated repression requires the formation of a multiprotein repression complex at the ventral silencer. Mol Cell Biol 18:6584 –6594. van de Wetering M, Cavallo R, Dooijes D, van Beest M, van Es J, Loureiro J, Ypma A, Hursh D, Jones T, Bejsovec A, Peifer M, Mortin M, Clevers H. 1997. Armadillo coactivates transcription driven by the product of the Drosophila segment polarity gene dTCF. Cell 88:789 –799. Vendrell V, Carnicero E, Giraldez F, Alonso MT, Schimmang T. 2000. Induction of inner ear fate by FGF3. Development 127:2011–2019. Whitfield TT, Granato M, van Eeden FJ, Schach U, Brand M, Furutani-Seiki M, Haffter P, Hammerschmidt M, Heisenberg CP, Jiang YJ, Kane DA, Kelsh RN, Mullins MC, Odenthal J, Nusslein-Vol-

hard C. 1996. Mutations affecting development of the zebrafish inner ear and lateral line. Development 123:241–254. Whitfield TT, Riley BB, Chiang MY, Phillips B. 2002. Development of the zebrafish inner ear. Dev Dyn 223:427–458. Wu¨lbeck C, Campos-Ortega JA. 1997. Two zebrafish homolugues of the Drosophila neurogenic gene groucho and their pattern of transcription during early embryogenesis. Dev Genes Evol 207:156 –166. Xu PX, Adams J, Peters H, Brown MC, Heaney S, Maas R. 1999. Eya1-deficient mice lack ears and kidneys and show abnormal apoptosis of organ primordia. Nat Genet 23:113–117. Zheng W, Huang L, Wei ZB, Silvius D, Tang B, Xu PX. 2003. The role of Six1 in mammalian auditory system development. Development 130:3989 –4000. Zhu CC, Dyer MA, Uchikawa M, Kondoh H, Lagutin OV, Oliver G. 2002. Six3-mediated auto repression and eye development requires its interaction with members of the Groucho-related family of co-repressors. Development 129:2835–2849.