TBP is not universally required for zygotic RNA polymerase II transcription in zebrafish

282 Brief Communication

TBP is not universally required for zygotic RNA polymerase II transcription in zebrafish Ferenc Mu¨ller†§, Lo`ra`nt Lakatos*§, Jean-Christophe Dantonel*‡, Uwe Stra¨hle† and La`szlo` Tora* General transcription factors TFIIA, B, D, E, F, H, and RNA polymerase II (Pol II) are required for accurate initiation of Pol II transcription. The TATA binding protein (TBP), a subunit of TFIID, is responsible for recognition of the TATA box, a core element shared by a category of class II promoters [1]. Recently, novel TBP-like factors (TLFs) have been described in metazoan organisms [2]. In spite of the numerous in vitro studies describing the general role of TBP in RNA polymerase II (Pol lI) transcription initiation, the precise function of TBP and the newly described TLF is poorly understood in vivo. We inhibited TBP and TLF function in zebrafish embryos to study the role of these factors during zygotic transcription. A dominant-negative variant of TLF mRNA and a TBP morpholino antisense oligo was used to block either TLF or TBP function. Both TBPor TLF-blocked embryos developed normally until the midblastula stage; however, they then failed to gastrulate. Several zygotic regulatory genes were downregulated by a block in either TBP or TLF function, while others were differentially affected. These results suggest that TBP is not universally required for Pol II transcription in vertebrates and that there is a differential requirement for TBP and TLF during early embryogenesis. Address: * Department of Transcriptional and Posttranscriptional Control of Gene Regulation and †Department of Developmental Biology, Institut de Ge´ne´tique et de Biologie Mole´culaire et Cellulaire, CNRS/INSERM/ULP BP 163, F-67404 Illkirch Cedex, C.U. de Strasbourg, France. ‡ Present

address: CNRS UMR 5535, IGM, 34293 Montpellier Cedex 5, France. Correspondence: La`szlo` Tora E-mail: [email protected] § F. M. and L. L. contributed equally to this work. Received: 27 November 2000 Revised: 21 December 2000 Accepted: 12 January 2001 Published: 20 February 2001 Current Biology 2001, 11:282–287 0960-9822/01/$ – see front matter  2001 Elsevier Science Ltd. All rights reserved.

Results and discussion The early embryonic development of many invertebrate and vertebrate species is characterized by an initial period

during which the zygotic genome is transcriptionally silent. TATA binding protein (TBP), a key player in initiation of transcription by Pol II, is a component of several class II transcription factors, i.e., different TFIID complexes [3] and the recently identified TAC complex [4]. From experiments in Xenopus embryos, it has been suggested that either the translation of TBP and/or its ability to interact with promoter sequences due to a closed chromatin configuration may be critical parameters that control the onset of zygotic transcription at midblastula transition (MBT) [5–7]. The genomes of several metazoan species contain a gene that encodes, in addition to TBP, a related factor called TBP-like factor (TLF) or TRF-2 (reviewed in [2]). While the structural organization of TLFs is very similar to that of TBPs [2], the structural model predicts important differences between the backbone of the DNA binding surface of TBPs and vertebrate TLFs, but also between nematode TLFs and that of vertebrate TLFs [2]. The differences in the predicted structures among TBP-type factors (including TBPs and TLFs) indicate potentially distinct functions for the vertebrate TLF homologs. TBP has been assumed to play a predominant function in transcription initiation of class II genes, and thus questions have arisen regarding the function of TLFs. The suppression of Caenorhabditis elegans (ce) TLF by RNA interference (or RNAi) showed that ceTLF is required for correct zygotic transcription during early embryogenesis [8, 9]. To investigate the role of TLF in vertebrates, mutations were generated in TLF that may affect Pol II transcription and tested in zebrafish (z) embryos. We generated point mutations in TLF residues (E131A; E133A) predicted to affect binding surfaces for TFIIB (TLFmutB) [10]. Injection of synthetic mRNA corresponding to wild-type TLF into early cleavage stage zebrafish embryos resulted in no or weak malformations in 6 hr old early gastrula stage embryos. Embryos were slightly retarded in the progression of epiboly, the cell movement by which the cells of the blastoderm envelope the yolk cell [11] (Figure 1a,b and Table 1). Thus, overexpressed vertebrate TLF may act as a repressor to prevent zygotic transcription in very early embryos, similar to what has been observed with ceTLF [8]. In contrast, a strong phenotype was obtained when TLFmutB RNA was injected at the same concentration as wild-type TLF (200 ng/␮l); 100% of the embryos failed to initiate epiboly or arrested before dome stage [11] (Figure 1c, Table 1). TLFmutB was still very

Brief Communication 283

Figure 1 Inhibition of TLF by expression of a dominant negative mutant (TLFmutB) interferes with epiboly and transcription in the zebrafish embryo. (a–d) Expression of TLFmutB results in severe inhibition of the epibolic spread of the blastoderm over the yolk cell. (a) Noninjected control at 50% epiboly stage. Expression of wild-type TLF (b) did not cause significant inhibition of epiboly.(c) Expression of the dominant negative TLFmutB had similar blocking effect on epiboly as injection of embryos with ␣-amanitin (d). (e,f) Zygotic expression of green fluorescent protein (gfp) from the reporter construct CMV::gfp is inhibited by coinjection of dominant-negative TLFmutB (f) but not by wild-type TLF (e). (g–i): Expression of notail (dark ring in [g]) is inhibited by expression of TLFmutB (h) or

injection with ␣-amanitin (i). (j–l) The correct phosphorylation of the CTD of Pol II on serine 2 is impaired in TLFmutB-injected embryos. Immunohistochemical staining of noninjected embryo (j), TLFmutB-injected

potent in blocking epiboly when injected at a 10-fold lower concentration (20 ng/␮l; Table 1). Interestingly, the majority of TLFmutB-expressing embryos arrested development at a similar stage to embryos that were injected with ␣-amanitin, an inhibitor of Pol II transcription (Figure 1, compare [c] and [d]) [12, 13]. This finding suggests that expression of TLFmutB results in inhibition of Pol II transcription in the embryos and that the phenotype is the result of a failure to initiate efficient zygotic transcription. In agreement with this hypothesis, cytomegalovirus (CMV) promoter–driven expression of Table 1 Injection of mRNA encoding variants of TLF causes retardation or complete block of epiboly and loss of ntl expression in zebrafish embryos. Retardation of epiboly (% embryos) Injected material TLF (200 ng/␮l) TLFmutB (200 ng/␮l) TLFmutB (20 ng/␮l) ␣-amanitin (200 ng/␮l)

Wild type

Weak

98 19 0 13 2

2 70 0 22 0

Strong Embryos (n) 0 11 100 65 98

82 61 70 88 43

Expression of notail (% embryos) Injected material Uninjected TLFmutB (100 ng/␮l) TLFmutB (100 ng/␮l) ⫹ TLF (100 ng/␮l) TLF (100 ng/␮l) ␣-amanitin (200 ng/␮l)

Wild type Partial

Lost

Embryos (n)

100 6 32

0 34 58

0 60 10

51 35 36

74 0

26 0

0 100

31 40

Strong retardation represents embryos showing a dome stage phenotype or complete loss of epiboly. Weak retardation represents embryos reaching up to 30%–40% epiboly. ntl expression in injected embryos was scored for loss or reduction of expression as compared to noninjected embryos at 6 hr gastrulation stage. Partial: embryos showing an asymmetrical expression pattern around the margin distinctive of partial loss of ntl expression. Lost: embryos with complete loss of ntl expression.

embryo (k), or ␣-amanitin-injected embryo (l) at 50% epiboly stage using the H5 antibody. All embryos are shown at 50% epiboly stage in lateral views, animal pole up (a–d), or in dorsal views of the animal pole (e–l).

green fluorescent protein (gfp) is inhibited by coinjected TLFmutB (Figure 1e,f). Moreover, expression of endogenous notail transcription, which is characterized by a ringlike expression pattern at early gastrula stage embryos (Figure 1g), was abolished in both TLFmutB- and ␣-amanitin-injected embryos (Figure 1h,i and Table 1). Importantly, the TLFmutB phenotype was rescued by coinjection of wild-type TLF mRNA into the embryos. Coinjected embryos progressed through epiboly (data not shown) and also reexpressed notail (ntl) mRNA (Table 1), indicating that the effects seen with TLFmutB were specific. These results show that TLFmutB acts as a dominant-negative inhibitor of TLF function, and suggest an important role for a TLF-like activity in early zygotic transcription.

The seemingly broad inhibition of Pol II transcription by TLFmutB in the early embryo (see also Figure 3) was surprising in view of the previously assumed important role of TBP in these process. To examine the role of TBP during early embryonic development, a recently described antisense morpholino oligonucleotide (MO) technique [14] was used to specifically eliminate zTBP expression. TBP-antisense MO (TBP-MO)-injected embryos, even when injected only at very low concentration (0.3 mM), severely impaired epiboly, while embryos injected with 5 mM control MO oligos continued to develop without any defects in gastrulation (Figures 2a–d). Epibolic movements were completely blocked by injection of 5 mM TBP-MO, resulting in developmental arrest similar to that caused by TLFmutB or ␣⫺amanitin (compare Figure 2d to Figure 1c,d). The efficiency of inhibition of TBP mRNA translation by the MO antisense oligonucleotide was analyzed by Western blot analysis exploiting the cross-reactivity of the 3G3 monoclonal antibody (mAb) directed against human TBP [15]. In protein extracts prepared from TBP-MO-injected embryos at 6 hr, early gas-

284 Current Biology Vol 11 No 4

Figure 2 Inhibition of TBP expression by antisense morpholino oligonucleotides (MOs) impairs epiboly and blocks correct phosphorylation of the CTD of Pol II. (a–d) A TBP-MO overlaping the first AUG codon of zTBP mRNA blocks epiboly in a dose-dependent manner. In contrast to embryos injected with control MOs (a), increasing doses of TBP-MO (0.3 mM, 1 mM, 5 mM in [b–d], respectively) inhibit epiboly. Embryos are oriented with the animal pole up. (e) TBP expression in extracts from MO-injected embryos. Injection of the TBPMO but not of the control MO (c-MO) completely eliminated zebrafish TBP (zTBP) expression at 50% epiboly stage. Furthermore, expression of human TBP (hTBP) from a coinjected plasmid (CMV:hTBP) was clearly detectable at 75% epiboly in the embryos; however, the endogenous TBP expression at this stage is significantly lower than at 6 hr gastrula stage. As a loading control (Pol II), an antibody raised against the CTD of Pol II was used (10% SDS-PAGE). (f–h) Phosphorylation of the CTD of Pol II on serine 2 is abolished in TBP-MO-injected zebrafish embryos. Immunohistochemical staining of control-MO injected embryo (f) and zTBP-MO injected embryo (g) at 50% epiboly stage using the H5 antibody. (h) Western blot analysis of protein extracts from 50% epiboly

stage embryos using the H5 or the 7G5 mAbs. The hyperphosphorylated (Pol IIO) and the hypophosphorylated (Pol IIA) forms of the

trula stage (60% epiboly) zTBP was not detected on the immunoblot, whereas in noninjected and control MOinjected embryos, zTBP protein levels were clearly detectable (Figure 2e; data not shown). Interestingly, the severe phenotype of the TBP-MO-injected embryos was partially rescued by the coinjection of purified recombinant human TBP protein (Table 2). Importantly, however, the ntl expression blocked by TLFmutB RNA injection was not affected in the TBP-MO-injected embryos at 6 hr early gastrula stage (Figure 3e,f), further demonstrating the specificity of the methods used. These data suggest that despite the dramatic phenotype observed, Table 2 Injection of morpholino antisense oligonucleotides (MO) designed against TBP results in the block of epiboly movements in zebrafish embryos. Retardation of epiboly (% embryos) Injected material Uninjected control MO (5 mM) TBP-MNO (0.3 mM) TBP-MO (1 mM) TBP-MO (5 mM) TBP-MO (1 mM) ⫹ hTBP Control MO (5 mM) ⫹ hTBP

Wild type Weak Strong Embryos (n) 100 98 4 10 2 26 100

0 2 86 33 0 59 0

0 0 10 57 98 15 0

65 56 45 67 50 210 65

Categories of the degree of retardation in epiboly were scored as in Table 1.

large subunit of Pol II are labeled (8% SDS PAGE). ns: non-specific loading control.

TBP is not universally required for all zygotic Pol II transcription in the zebrafish embryo. The similarity in the phenotypes caused by either blockage of TBP expression or TLF activity suggests that zygotic transcription and progression of development into gastrulation requires both proteins. One explanation for these findings is that both factors act on the same genes. To investigate this possibility, the expression from seven endogenous early patterning genes, initially expressed at late blastula (4.5 hr; ntl, forkhead4 [fkd4], spadetail [spt], goosecoid [gsc]) or early gastrula (5.5 hr; even-skipped1 [eve1], sonic hedgehog [shh], T-box6 [tbx6]) was compared in embryos in which TLF or TBP was blocked (Figure 3). TLFmutB abolished expression of six of the seven genes completely. Curiously, eve1, which encodes a transcriptional repressor [16], lost its specific pattern of expression in the germ ring and instead became expressed ubiquitously in TLFmutB-injected embryos. Lack of TBP caused a different pattern of effects. While expression of eve1, shh, spt, tbx6, and gsc were abolished, expression of ntl and fkd4 were not affected or were only weakly affected by injection of zTBP-MO. Although we may not look at direct effects of TBP or TLF on the genes investigated here, nevertheless, we did not find a direct correlation between the timing of the onset of gene activity and effects of blocking TBP-type factors. However, our results show that TBP and TLF can have distinct functions in

Brief Communication 285

Figure 3

the zebrafish embryo. Interestingly, expression of certain genes depends on both activities either directly or indirectly. The absence of TLF in C. elegans embryos prevented the correct soma-specific phosphorylation of the C-terminal domain (CTD) of the large subunit of Pol II and resulted in a delay in the onset of transcription [8, 9]. To analyze the phosphorylation state of Pol II in zebrafish embryos, we have carried out immunohistochemical staining and Western blot analysis. Monoclonal antibodies H5, which recognizes the phosphoserine in position 2 on the CTD [17], and 7G5, which recognizes both the hyper- and hypophosphorylated forms of Pol II [18], were used. In control embryos at 50% epiboly stage H5 mAb immunoreactivity was detected in a restricted number of scattered cells (Figure 1j and Figure 2f). Compared to the control embryos, TLFmutB-injected embryos showed a significantly rearranged spatial distribution of H5 staining (Figure 1k), while block of Pol II transcription by ␣-amanitin abolished H5 staining (Figure 1l). Moreover, TBP-MO injection resulted in a loss of H5 immunoreactivity (Figure 2g). Western blot analysis using the H5 mAb confirmed the immunohistochemical H5 staining, and using the 7G5 mAbs showed that, potentially, other phosphorylated forms of the CTD of Pol II are also missing from TBPMO-injected embryos (Figure 2h). Thus, blocking TBP or TLF function may differentially influence CTD phosphorylation and, consequently, transcription.

Conclusions

TLF and TBP differentially regulate early zygotic transcription. Embryos were injected with either the dominant-negative TLFmutB or with the antisense TBP-MO. Embryos were split when controls had reached the 40%–50% epiboly stage and subjected to in situ hybridization with evenskipped (eve1) (a–c), notail (ntl) (d–f), forkhead4 (fkh4) (g–i), sonic hedgehog (shh) (j–l), spadetail (spt) (m–o), t-box6 (tbx6) (p–r), and goosecoid (gsc) (s–u) probes. Expression was abolished by inhibition of either TLF or TBP in case of shh (87.5%, n ⫽ 23; 95.0%, n ⫽ 20, respectively), spt (88.9%, n ⫽ 18; 95.2%, n ⫽ 21), tbx6 (85.0%, n ⫽ 20; 100.0%, n ⫽ 22), and gsc (90.0%, n ⫽ 20; 100.0%, n ⫽ 17). In contrast, ntl and fkh4 were downregulated only by inhibition of TLF ([e,h] 94.3%, n ⫽ 35, and 87.9%, n ⫽ 33, respectively) while blocking of TBP had no or little effect on these genes (ntl, 0.0%, n ⫽ 42; fkd4, 23.8%, n ⫽ 21). Expression of eve1 was abolished by inhibition of TBP ([c] 100.0%, n ⫽ 18); however, TLF-blocked embryos showed no effect or an expansion of eve1 mRNA expression into the animal pole area ([b] 72.8% normal or ectopic expression, n ⫽ 22). View onto animal pole with dorsal up.

Taken together, our results suggest that TBP is required for epiboly but not universally for early zygotic transcription of specific genes in zebrafish. Furthermore, TBP and TLF can have distinct functions as exemplified by the differential effects on eve1, fkd4, and ntl expression. Thus, TBP and possibly also TLF regulate subsets of Pol II genes specifically. However, both proteins appear to be required for the transcriptional activation of the zygotic genome at MBT and for the subsequent epibolic movements. In the case of the early zygotic genes spt and gsc, both TBP- and TLF-like function are required for expression. It is feasible that TBP and TLF act directly together in the control of transcription initiation of these genes. As transcription initiation and reinitiation differ mechanistically, TLF might function to organize embryonically transcribed genes into an active state that is then further transcribed by TBP. This hypothesis is in good agreement with the C. elegans tlf-1(RNAi) results, suggesting that the earliest stages of embryogenesis may critically depend on TLF activity [8, 9]. It is conceivable that TLF has a higher affinity or accessibility for promoters than TBP before the MBT, because the state of the nuclei and that of the chromatin structure seems to be inhibitory for TBP function prior to MBT [6, 19]. Our observations together suggest that Pol II transcription does not univer-

286 Current Biology Vol 11 No 4

sally require TBP in metazoans and that distinct members of the TBP-type factors are differentially required to initiate zygotic transcription during embryogenesis.

Materials and methods Plasmid constructions TLFmutB [10] was generated by PCR using the Xenopus TLF sequence as a template. Xenopus TLF was used, as at present the zebrafish TLF has not yet been identified from the available databanks. TLF variants were cloned into the pCS2 vector [20] and verified by sequencing.

Maintenance of zebrafish, embryo production, and microinjection Fish were kept and embryos were produced and staged as described [21]. Zebrafish embryonic stages were identified as described in [11]. Synthesis of capped mRNA and microinjection into the zebrafish zygote and early cleavage stages were carried out as described [22]. pCMV::gfp (gift of A. Ungar and J. Miller) and pXJ40-hTBP expression vectors have been described [23]. Morpholinos were purchased from GeneTools and dissolved as described by the supplier. Sequences of morpholino oligonucleotides were zTBP-MO, 5⬘-GAGGTAGGCTGTTGTTCTGCT CCAT-3⬘; standard control-MO, 5⬘-CCTCTTACCTCAGTTACAATT TATA-3⬘; and specific TBP control-MO containing seven mismatches (underlined), 5⬘-AAGGTGGGCTATTGTCATGTCCCAT-3⬘. MO oligos were diluted in 1⫻ Danieau solution [14] prior to microinjection.

In situ hybridization and immunohistochemistry In situ hybridization on whole-mount zebrafish embryos was carried out as described in [24]. Digoxygenin-labeled antisense probes were synthesized in vitro as described: eve1 [25], shh [26], spt [27], Ttbx6 [28], gsc [29], ntl [29], and fkd4 [30]. Immunohistochemical staining was carried out using a rhodamine-conjugated anti-mouse IgM antibody as described previously [31].

Western blot analysis Yolk cells were removed by trituration, and animal caps were then collected and washed twice in 1⫻ PBS. Protein extracts were then made by freezing and thawing the cells three times. Aliquots containing 5–10 ␮g proteins were boiled in SDS sample buffer, separated on SDSPAGE, transferred to nitrocellulose membrane, and probed with the indicated primary antibodies. Chemilluminescence detection was performed according to manufacturer’s instructions. The anti-hTBP mAb 3G3 has been described previously [15].

Note added in proof While this manuscript has been processed for publication, Veenstra et al. [32] also reported that TBP and TLF function differentially during early embryonic transcription in Xenopus.

Acknowledgements We are grateful to A. Karmin for technical assistance; P. Ingham, S. SchulteMerker, S. Joly, T. Dickmeis, and A. Ungar for reagents; S. Rastegar for advice; M. Leid for a critical reading of the manuscript; and A. Wolffe and J.G. Veenstra for communicating unpublished data. F. M. was supported by fellowships from EMBO and AFM. The present work was supported by grants from the CNRS, the INSERM, the Hoˆpital Universitaire de Strasbourg, the Ministe`re de la Recherche et Technologie, FRM, ARC, Ligue National Contre le Cancer, Association of International Cancer Research, AFM, and HFSP.

References 1. Orphanides G, Lagrange T, Reinberg D: The general transcription factors of RNA polymerase II. Genes Dev 1996, 10:26572683. 2. Dantonel JC, Wurtz JM, Poch O, Moras D, Tora L: The TBP-like factor: an alternative transcription factor in metazoa? Trends Biochem Sci 1999, 24:335-339. 3. Bell B, Tora L: Regulation of gene expression by multiple forms of TFIID and other novel TAFII-containing complexes. Exp Cell Res 1999, 246:11-19.

4. Mitsiou DJ, Stunnenberg HG: TAC, a TBP-sans-TAFs complex containing the unprocessed TFIIAalphabeta precursor and the TFIIAgamma subunit. Mol Cell 2000, 6:527-537. 5. Prioleau MN, Huet J, Sentenac A, Mechali M: Competition between chromatin and transcription complex assembly regulates gene expression during early development. Cell 1994, 77:439449. 6. Prioleau MN, Buckle RS, Mechali M: Programming of a repressed but committed chromatin structure during early development. EMBO J 1995, 14:5073-5084. 7. Veenstra GJ, Destree OH, Wolffe AP: Translation of maternal TATA-binding protein mRNA potentiates basal but not activated transcription in Xenopus embryos at the midblastula transition. Mol Cell Biol 1999, 19:7972-7982. 8. Dantonel JC, Quintin S, Lakatos L, Labouesse M, Tora L: TBP-like factor is required for embryonic RNA polymerase II transcription in C. elegans. Mol Cell 2000, 6:715-722. 9. Kaltenbach L, Horner MA, Rothman JH, Mango SE: The TBP-like factor CeTLF is required to activate RNA polymerase II transcription during C. elegans embryogenesis. Mol Cell 2000, 6:705-713. 10. Moore PA, Ozer J, Salunek M, Jan G, Zerby D, Campbell S, et al.: A human TATA binding protein-related protein with altered DNA binding specificity inhibits transcription from multiple promoters and activators. Mol Cell Biol 1999, 19:7610-7620. 11. Kimmel CB, Ballard WW, Kimmel SR, Ullmann B, Schilling TF: Stages of embryonic development of the zebrafish. Dev Dyn 1995, 203:253-310. 12. Kedinger C, Gniazdowski M, Mandel JL Jr, Gissinger F, Chambon P: Alpha-amanitin: a specific inhibitor of one of two DNApendent RNA polymerase activities from calf thymus. Biochem Biophys Res Commun 1970, 38:165-171. 13. Kane DA, Hammerschmidt M, Mullins MC, Maischein HM, Brand M, van Eeden FJ, et al.: The zebrafish epiboly mutants. Development 1996, 123:47-55. 14. Nasevicius A, Ekker SC: Effective targeted gene ‘knockdown’ in zebrafish. Nat Genet 2000, 26:216-220. 15. Lescure A, Lutz Y, Eberhard D, Jacq X, Krol A, Grummt I, et al.: The N-terminal domain of human TATA-binding protein plays a role in transcription from TATA-containing RNA polymerase II and III promoters. EMBO J 1994, 13:1166-1175. 16. Li C, Manley JL: Even-skipped represses transcription by binding TATA binding protein and blocking the TFIID-TATA box interaction. Mol Cell Biol 1998, 18:3771-3781. 17. Patturajan M, Schulte RJ, Sefton BM, Berezney R, Vincent M, Bensaude O, et al.: Growth-related changes in phosphorylation of yeast RNA polymerase II. J Biol Chem 1998, 273:4689-4694. 18. Besse S, Vigneron M, Pichard E, Puvion-Dutilleul F: Synthesis and maturation of viral transcripts in herpes simplex virus type 1 infected HeLa cells: the role of interchromatin granules. Gene Expr 1995, 4:143-161. 19. Kikyo N, Wade PA, Guschin D, Ge H, Wolffe AP: Active remodeling of somatic nuclei in egg cytoplasm by the nucleosomal ATPase ISWI. Science 2000, 289:2360-2363. 20. Turner DL, Weintraub H: Expression of achaete-scute homolog 3 in Xenopus embryos converts ectodermal cells to a neural fate. Genes Dev 1994, 8:1434-1447. 21. Westerfield M: The Zebrafish Book. Eugene: University of Oregon Press; 1993. 22. Muller F, Albert S, Blader P, Fischer N, Hallonet M, Strahle U: Direct action of the nodal-related signal cyclops in induction of sonic hedgehog in the ventral midline of the CNS. Development 2000, 127:3889-3897. 23. Veschambre P, Simard P, Jalinot P: Evidence for functional interaction between the HIV-1 Tat transactivator and the TATA box binding protein in vivo. J Mol Biol 1995, 250:169180. 24. Oxtoby E, Jowett T: Cloning of the zebrafish krox-20 gene (krx20) and its expression during hindbrain development. Nucleic Acids Res 1993, 21:1087-1095. 25. Joly JS, Joly C, Schultemerker S, Boulekbache H, Condamine H: The ventral and posterior expression of the zebrafish homeobox gene eve1 is perturbed in dorsalized and mutant embryos. Development 1993, 119:1261-1275. 26. Krauss S, Concordet JP, Ingham PW: A functionally conserved homolog of the Drosophila segment polarity gene hh is expressed in tissues with polarizing activity in zebrafish embryos. Cell 1993, 75:1431-1444.

Brief Communication 287

27. Griffin KJ, Amacher SL, Kimmel CB, Kimelman D: Molecular identification of spadetail: regulation of zebrafish trunk and tail mesoderm formation by T-box genes. Development 1998, 125:3379-3388. 28. Hug B, Walter V, Grunwald DJ: tbx6, a Brachyury-related gene expressed by ventral mesendodermal precursors in the zebrafish embryo. Dev. Biol. 1997, 183:61-73. 29. Schulte-Merker S, Hammerschmidt M, Beuchle D, Cho KW, De Robertis EM, Nusslein-Volhard C: Expression of zebrafish goosecoid and no tail gene products in wild-type and mutant no tail embryos. Development 1994, 120:843-852. 30. Odenthal J, Nusslein-Volhard C: fork head domain genes in zebrafish. Dev Genes Evol 1998, 208:245-258. 31. Strahle U, Blader P, Henrique D, Ingham PW: Axial, a zebrafish gene expressed along the developing body axis, shows altered expression in cyclops mutant embryos. Genes Dev 1993, 7:1436-1446. 32. Veenstra GJ, Weeks DL, Wolffe AP: Distinct roles for TBP and TBP-like factor in early embryonic gene transcription in Xenopus. Science 2001:2312–2315.