A cDNA clone coding for a sea urchin embryonic protein was isolated from a prehatching blastula λgtll library. The predicted translation product is a secreted 64 × 103 Mr enzyme designated as BP10. The protein contains several domains: a signal peptide, a putative propeptide, a catalytic domain with an active center typical of a Zn2+-metalloprotease, an EGF-like domain and two internal repeats similar to repeated domains found in the Cis and Clr serine proteases of the complement cascade. The BP10 protease is constructed with the same domains as the human bone morphogenetic protein BMP-1, a protease described as a factor involved in bone formation, and as the recently characterized product of the tolloid gene which is required for correct dorsal-ventral patterning of the Drosophila embryo. The transcription of the BP10 gene is transiently activated around the 16-to 32-cell stage and the accumulation of BP10 transcripts is limited to a short period at the blastula stage. By in situ hybridization with digoxygenin-labelled RNA probes, the BP10 transcripts were only detected in a limited area of the blastula, showing that the transcription of the BP10 gene is also spatially controlled. Antibodies directed against a fusion protein were used to detect the BP10 protein in embryonic extracts. The protein is first detected in early blastula stages, its level peaks in late cleavage, declines abruptly before ingression of primary mesenchyme cells and remains constant in late development. The distribution of the BP10 protein during its synthesis and secretion was analysed by immunostaining blastula-stage embryos. The intracellular localization of the BP10 staining varies with time. The protein is first detected in a perinuclear region, then in an apical and submembra-nous position just before its secretion into the perivitel-line space. The protein is synthesized in a sharply delimited continuous territory spanning about 70% of the blastula. Comparison of the size and orientation of the labelled territory in the late blastula with the fate map of the blastula stage embryo shows that the domain in which the BP10 gene is expressed corresponds to the presumptive ectoderm.

Developing embryos treated with purified antibodies against the BP10 protein and with synthetic peptides derived from the EGF-like domain displayed pertur-bations in morphogenesis and were radialized to various degrees. These results are consistent with a role for BP10 in the differentiation of ectodermal lineages and sub-sequent patterning of the embryo. On the basis of these results, we speculate that the role of BP10 in the sea urchin embryo might be similar to that of tolloid in Drosophila. We discuss the idea that the processes of spatial regulation of gene expression along the animal-vegetal in sea urchin and dorsal-ventral axes in Drosophila might have some similarities and might use common elements.

A major objective in developmental biology is to understand the mechanisms underlying early determination of cell lineages and pattern formation. The characterization of mutations affecting development in Drosophila has allowed the isolation of genes whose expression precedes and controls pattern formation along the anterior-posterior and dorsal-ventral axes of the egg and early embryo. In other organisms in which isolation of mutants is not feasible practically, the mechanisms responsible for determination of cell fate and patterning remain less well understood. Neverthe-less, several genes showing a structural homology with genes controlling anterior-posterior pattern formation in Drosophila have been characterized in vertebrates, revealing that some molecular mechanisms of pattern formation are conserved in evolution (Duboule and Dollé, 1989). As far as the dorsal-ventral axis is concerned, four vertebrates genes (or gene families) homologous to Drosophila patterning genes have been isolated. Genes encoding mammalian transcription factors of the rel family have remarkable homologies with the maternal effect gene dorsal (Steward, 1987; Kieran et al., 1990; Ghosh et al., 1990). The Xenopus xtwist (Hopwood et al., 1989) and xsnail (Sargent and Bennett, 1990) genes have sequence similarities with the zygotic genes twist and snail which are positively regulated by the morphogen dorsal and finally, genes coding for polypeptides of the TGF-μ family are structurally related to the product of the decapentaple-gic gene, a zygotic gene negatively regulated by dorsal (Padgett et al., 1987).

For many years, the sea urchin embryo has been an attractive material to study cell determination and patterning in deuterostomes. The egg possesses a primordial organization in only one axis, the animal-vegetal axis (Boveri, 1901; reviewed in Davidson, 1986). It is well known, from the classical embryology experiments, that the development of animal and vegetal merogones follows widely divergent directions. The animal halves develop into spherical ciliated embryoids, which do not gastrulate, while the vegetal halves develop into small pluteus larvae with an almost normal morphology except that the oral lobe is underdeveloped (Boveri, 1901; Horstadius, 1973; Mar-uyama et al., 1985). The localization of factors responsible for determination in the sea urchin egg is still a debated question. Egg bisection experiments strongly argue for the asymmetric distribution of morphogenetic factors in the egg. Determinants re-sponsible for mesoderm and endoderm formation would be localized in the vegetal hemisphere while those controlling ectoderm formation would be restric-ted to the animal hemisphere. On the other hand, isolation and transplantation experiments carried out by Horstadius (reviewed in Horstadius, 1973) have clearly demonstrated that up to the 64-cell stage the potencies of most of the blastomeres are much greater than their normal fates. With the exception of the micromeres, all of the other blastomeres can be reprogrammed by putting them in contact with cells from other parts of the early embryo. In addition, treatments of cleaving embryos with various chemicals perturb determination in two opposite directions, animalization and vegetalization. These experiments led Runnstrom and Horstadius to propose a model of early determination in the sea urchin embryo based on the existence of two maternal morphogenetic gradients. Decreasing in intensity from the animal to the vegetal pole, the animal gradient would be responsible for the differentiation of ectodermal structures. The vegetal gradient, running in the opposite direction would mediate the differentiation of mesodermal (skeleton) and endodermal tissue. Recently, two authors have called into question this model and critically reexa-mined the interpretations of classical embryology in the light of recent information on cell lineage and lineage-specific gene expression (Wilt, 1987; Davidson, 1989). A molecular mechanism of cell lineage specification in this embryo has been proposed (Davidson, 1989). The key idea of this model is that specification would be mediated by a series of inductive interactions between cell layers. The first signal would originate from the vegetal pole blastomeres and be transmitted, stepwise, by ligand-receptor interactions to the overlying cells.

Until now, despite many experimental and theoreti-cal efforts to understand determination and morpho-genesis in the sea urchin embryo, no gene or gene product, either maternal or zygotic, controlling deter-mination or pattern formation along the animal-vegetal axis has been isolated, essentially because very early expressed and spatially restricted genes were not yet characterized. We report here, the cloning, characteriz-ation and preliminary functional analysis of such a marker gene that might also play a role in pattern formation. The spatiotemporal pattern of expression of this gene, together with the homologies of the corre-sponding protein with a Drosophila protein recently characterized, lead us to speculate that the processes of determination and positional signalling along the animal-vegetal axis in the sea urchin may share some key features with the process of dorsal-ventral axis determination in Drosophila.

Gametes and embryos

Paracentrotus lividus adults were collected along the Côte d’Azur. Spawning, fertilization and embryo culturing were carried out in Millipore filtered sea water (MFSW) or artificial sea water (ASW) as described previously (Lepage and Gaché, 1989, 1990). When required, the fertilization envelopes were removed using an adaptation of the procedure of Showman and Foerder (1979).

RNA extraction

Embryos were collected at various stages by low-speed centrifugation, washed twice with ASW, and treated immedi-ately or frozen in liquid nitrogen and stored at — 70°C. Total cellular RNA was prepared by the method of Cathala et al. (1983).

cDNA library construction and screening

Total RNA from prehatching blastulas (about 300 cells) was prepared as indicated. Poly (A)+-RNAs were purified by two cycles of chromatography on oligo(dT)-cellulose. A cDNA library was constructed in Agtll as described (Lepage and Gaché, 1990). Immunological screening was done according to Mierendorf and Pfeffer (1987).

DNA sequencing

The sequencing strategy involved the construction of nested deletions of the pBPIO plasmid in both directions and the subcloning of various restriction fragments. In all cases, direct sequencing of denatured plasmid DNA was done by the dideoxynucleotide chain termination method of Sanger et al. (1977).

DNA probes

’We used as probe a 0.4 kb EcoRV restriction fragment corresponding to positions 1887 to 2292 in the cDNA. The plasmid pHB 70 containing the early H2A and H3 histones genes has been described (Spinelli et al., 1979). When needed, DNA probes were labelled to a specific activity of about 109 cts min−1 μg−1 by random priming according to Feinberg and Vogelstein (1983).

Northern blots

Samples of total RNA were fractionated by electrophoresis on 1% agarose gels containing 1.2 M formaldehyde and transferred to nitrocellulose membranes by standard methods (Maniatis et al., 1982). Blots were prehybridized for one hour at 68°C in 5 ×SSPE, 5 × Denhardt’s solution, 0.5% SDS, 10% dextran sulfate and 150 μg/ml denatured salmon sperm DNA. Hybridization was carried out overnight at 68°C in the same solution with the addition of the 32 P-labelled DNA probe at 4 × 106 cts min−1 ml−1. After hybridization the filters were washed at 68°C with decreasing salt concentrations. The most-stringent wash was in 0.1×SSPE, 0.1% SDS at68°Cfor 15 min.

Nuclei preparation and run-on transcription by isolated nuclei

At the desired stage of development, the cultures were chilled on ice and the embryos collected by low-speed centrifugation. All the subsequent steps were done at 0–4°C. The nuclei were prepared essentially according to Morris and Marzluff (1983) as detailed by Lepage and Gaché (1990). Transcription reaction with the isolated nuclei, monitoring of the reaction, extraction and purification of the labelled RNAs, hybridiz-ation and autoradiography were carried out as described (Lepage and Gaché, 1990).

Whole-mount in situ hybridization

The procedure was adapted from Cox et al. (1984), Tautz and Pfeifle (1989), Hemmati-Brivanlou et al. (1990) and various manufacturer protocols (Promega, Boehringer).

The plasmid pBPIO, which contains a full-length BP10 cDNA, was linearized by the appropriate restriction enzyme and used as template to synthesize single-strand RNA probes labelled with digoxygenin-ll-UTP (DIG-RNA probes). The embryos were collected as described above, fixed with 1% glutaraldehyde, dehydrated in an ethanol series to 70% ethanol and stored at — 20°C (Angerer et al., 1987). The fixed embryos were rehydrated, treated with proteinase K and refixed with formaldehyde. Hybridization was carried out overnight at 46°C. The nonspecifically bound probe was digested with RNAses and the embryos were washed at high stringency. RNA hybrids were detected using an alkaline-phosphatase-conjugated anti-digoxygenin antibody and the chromogenic substrates NBT and BCIP. The nuclei were stained with Hoechst 33342. The embryos were mounted in 50% glycerol, examined under bright-field illumination and fluorescence with a Zeiss Axiophot microscope and photo-graphed using a colour reversal film Ektachrome EPY 64 T. The detailed procedure will be described elsewhere.

Production and purification of polyclonal antibodies

A BP10-cro-lacZ in frame gene was constructed by inserting a 1.2 kb EcoRV fragment derived from the cDNA into a Smal- digested pEx2 vector (Stanley and Luzio, 1984). This fragment codes for amino acids 177–575 of the BP10 protein.

Host bacteria carrying the construct were induced for 3 h. Bacteria were harvested by centrifugation at 7000 g and inclusion bodies purified by standard methods (Harlow and Lane, 1988).

To generate polyclonal antibodies, rabbits were first injected with 5 mg of inclusion bodies mixed with Freund’s complete adjuvant at 10 subcutaneous and 4 intramuscular sites. They were boosted 4 times at one month intervals by injection of 2 mg of inclusion bodies in Freund’s incomplete adjuvant at the same sites. Bleeding was done 15 days after each boost and the sera were tested by immunoblotting. Antibodies directed against the fusion protein were purified by affinity chromatography. 40 mg of inclusion bodies were dissolved in 6 M urea, 0.1 M EPPS pH 8.0 and coupled to 4 ml of Affigel 10 according to the manufacturer’s protocol. 5 ml of antiserum were diluted 1:10 with 10 mM Tris-HCl pH 8.0 and passed 3 times through the column. The column was washed sequentially with 20 bed volumes of 10 mM Tris-HCl pH 8.0 and 20 bed volumes of 500 mM NaCl in the same buffer. Specific antibodies were eluted in 3 steps with 8 bed volumes of the following solutions: 0.1 M glycine pH 2.5, 0.1 M triethylamine pH 11.5, and 3 M MgClj. Fractions containing proteins were neutralized, pooled and the antibody solution was concentrated either on a small protein-A sepharose column or by ultrafiltration. Protein concentrations were determined either by the method of Bradford (1976) or by spectrophotometry. For in vivo experiments, the affinity-purified IgGs were dialysed against ASW and stored at 4°C. For immunostaining of fixed embryos, BSA was added to 1% and IgGs were dialysed against TBS (150 mM NaCl, 50 mM Tris-HCl pH 8.0) containing 0.01% sodium azide and extensively preadsorbed against fixed late gastarlas, which do not express the BP10 mRNA.

Western blot analysis

At the desired stage of development, embryos were har-vested, washed twice with MFSW and the embryos pellets were frozen in liquid nitrogen. Each pellet was solubilized by adding 10 volumes of sample buffer containing 6 M urea at 95°C. Complete solubilization was achieved by vortexing, sonication and incubation at 95°C. Insoluble material was removed by centrifugation at 10000 g and the protein concentration of the sample was measured by the method of Bradford (1976). Proteins samples were separated by SDS-gel electrophoresis and electrophoretically transferred to nitro-cellulose sheets using a semi-dry transfer cell of BioRad and following instructions of the manufacturer. The replicate was blocked for 2 h in 5% non-fat dry milk in TBST and incubated overnight at 4°C in the same solution with the affinity-purified anti-BP10 antibodies. After 5 washes of 10 min each, the filter was reacted for 1 h with an alkaline-phosphatase-conjugated goat anti-rabbit IgG secondary antibody diluted 1/7500 in TBST and washed as before. The immune complexes were revealed by using the chromogenic substrates NBT and BCIP.

Immunostaining of embryos

Embryos at the desired stage were collected by low-speed centrifugation, washed once with MFSW and the pellet resuspended in 10 volumes of MFSW. The embryos were fixed with 4% paraformaldehyde, 0.1% glutaraldehyde, 10 mM EPPS pH 8.0 in MFSW. The fixative was made fresh as a twofold concentrated solution and added to an equal volume of resuspended embryos. Fixation was done for 10-15 min at room temperature under gentle agitation, then the fixative was removed and the embryos washed twice in MFSW. Free reactive groups were blocked by incubating the embryos for 5 min in 1 M ethanolamine pH 8.0. After an additional wash, the embryos were quickly transferred to methanol at -20°C and stored at —20°C. All subsequent steps were done at room temperature in 1.5 ml Eppendorf tubes on a rotating wheel. The embryos were rehydrated in 50% MeOH, 50% TBST and rinsed several times in TBST. Embryos were blocked 20 min in 5% non-fat dry milk in TBST and incubated for 1 h in the same buffer with the affinity-purified anti-BP10 antibodies at 0.5–1 μg/ml. After 6 washes of 5 min each in TBST, embryos were incubated with an alkaline-phosphatase-conjugated goat anti-rabbit IgG antibody diluted 1/7500, which had been preadsorbed against fixed embryos. After washing as above, the embryos were rinsed briefly with alkaline phosphatase buffer (100 mM Tris-HCl pH 9.5, 100 mM NaCl, 5 mM MgCl2) and cell staining was developed in the same buffer after addition of the chromogenic substrates NBT and BCIP. The embryos were handled and observed as indicated above for in situ hybridization experiments. In control experiments, the first antibody was omitted or replaced with IgGs purified from preimmune sera.

Peptides

Two derivatives of the 25 mer synthetic peptide DDIEDCA-GANECLNGGYHDTECNCV corresponding to positions 284-309 in the BP10 protein sequence were synthesized by the Protein and Nucleic Acid Chemistry Center of the Woods Hole Oceanographic Institution. Peptide PI is a partially refolded intermediate obtained by oxidation of cys 6 and cys 22 residues. Peptide PII was obtained by oxidation of the remaining cys and cys 24 residues of peptide PI. Both peptides were purified by HPLC (90% purity). In control experiments, we used a 13-mer synthetic bovine neurotensin.

In vivo effects of antibodies and synthetic peptides

Embryos whose fertilization envelopes had been removed were cultured in ASW up to the 32- to 64-cell stage and then samples of 100–200 embryos were transferred to the wells (200 μl) of a microtitration plate. Anti-BP10 antibodies and synthetic BP10 peptides (or the control IgGs and control peptide) were added at various concentrations and the cultures incubated in a humid chamber at 20°C. Development was monitored by microscopy on a Zeiss Axiophot. After two days, the embryos were mounted with and without coverslips and examined using bright field and polarization.

Computer sequence analysis

Sequence handling and analysis, and homology searches in protein data banks were performed with the softwares from Intelligenetics and CITI 2.

Chemicals

cDNA synthesis and non-radioactive detection kits were from Boehringer, λgtll cloning kit and DNA labelling kit from Amersham, Bluescript plasmid and nested deletions kit from Stratagene, Sequenase sequencing kit from US Biochemicals. All these kits were used following instructions from the manufacturers. Restriction endonucleases were from Boehr-inger or Appligene and oligo(dT)-cellulose from Boehringer. Affigel 10 and Bradford reagent were from BioRad. NACS columns were from BRL, reagents for in vitro transcription and goat anti-rabbit IgG antibodies coupled to alkaline phosphatase from Promega. Labelled nucleotides were from Amersham. Artificial sea water (ASW) was 510 mM NaCl, 10 mM KC1, 11 mM CaCl2, 34 mM MgC32, 22 mM MgSO4, 10 mM Tris-HCl pH 8.2 and 50 μg/ml penicillin and strepto-mycin.

Screening of a prehatching blastula library and cloning of BP 10

The BP10 protein was identified during a search for the hatching enzyme cDNA, a secreted protease specific to the blastula stage. In order to clone the hatching enzyme cDNA, a prehatching blastula expression library was constructed in Âgtll and screened with a polyclonal antibody directed against the purified protease. We isolated 29 independent clones which were arranged in 3 groups on the basis of cross-hybridization and partial restriction analyses. The 26 clones that formed the largest group have been shown to code for the hatching enzyme (Lepage and Gaché, 1990).

Two other clones isolated using the same antibody did not hybridize with clones of the first group, had a clearly different restriction map and thus appeared to code for a protein other than the hatching enzyme. One of them, the λBP10 clone, appeared to bear a nearly full-length cDNA of 2.7 kb which was subcloned in Bluescript as the plasmid pBPlO.

Nucleotide sequence of the BP 10 clone

The nucleotide sequence of the BP10 clone was determined as indicated in Materials and Methods. In addition, the sequence was confirmed by sequencing the coding region of the BP10 gene, which is identical to the cDNA sequence, except for a few polymorphic changes (data not shown). The complete cDNA sequence is shown in Fig. 1. It contains a long open reading frame of 1791 bp, which begins with the ATG at position 157. The sequence adjacent to this codon AAAAATG matches the initiation consensus sequence C/A AAA/CATG described by Cavener (1987) for Drosophila and is very similar to the initiation sequence of the sea urchin hatching enzyme mRNA (TAAAATG) (Lepage and Gaché, 1990). The open reading frame is flanked by a 156 bp noncoding leader sequence which contains several in frame stop codons and by a 753 bp 3’ noncoding sequence with a canonical polyadenylation signal AATAAA, 19 bp upstream of the poly(A) tail.

Fig. 1.

Nucleotide sequence and deduced amino acid sequence of the BP10 clone. The nucleotide sequence is numbered starting from the first nucleotide of the putative initiation codon. This Met codon is the origin of the amino acid sequence numbering. The vertical arrowheads indicate 2 probable cleavage sites of the signal sequence. Several features discussed in the text are underlined; from the N terminus to the C terminus: a basic amino acid stretch (89–93), the active site (187–200), an EGF-like domain (294–329) and a series of threonine clusters (452–476). The double bar underlines the polyadenylation signal. The nucleotide sequence data reported will appear in the EMBL, GenBank and DDJB Nucleotide Sequence Databases under the accession number X56224.

Fig. 1.

Nucleotide sequence and deduced amino acid sequence of the BP10 clone. The nucleotide sequence is numbered starting from the first nucleotide of the putative initiation codon. This Met codon is the origin of the amino acid sequence numbering. The vertical arrowheads indicate 2 probable cleavage sites of the signal sequence. Several features discussed in the text are underlined; from the N terminus to the C terminus: a basic amino acid stretch (89–93), the active site (187–200), an EGF-like domain (294–329) and a series of threonine clusters (452–476). The double bar underlines the polyadenylation signal. The nucleotide sequence data reported will appear in the EMBL, GenBank and DDJB Nucleotide Sequence Databases under the accession number X56224.

Predicted amino acid sequence of the BP10 protein

The predicted protein coded for by the BP10 cDNA contains 597 amino acids and has a calculated Mr of 66×103. Its sequence is shown in Fig. 1. The protein begins with its most hydrophobic series of amino acids which has the characteristics of a signal sequence. Analysis of the N-terminal sequence by the statistical weight-matrix method of von Heijne (1986) indicates 2 plausible cleavage sites with high scores, between residues Cys 16 and Thr 17 and between Ala 19 and Ala 20. The first site is clearly favored over the second one (score 9.18 versus 7.67) and thus Thr 17 will be taken provisionally as the first residue of the secreted protein.

The secreted protein should comprise 581 amino acids for a Mr of 64 × 103; it is a hydrophilic protein with an excess of acidic (81 Asp + Glu) over basic amino acids (63 Arg + Lys + His) and unusually high numbers clusters of 2-3 residues. A threonine-rich region is also of methionine (13) and cysteine residues (18). Most of present in the hatching enzyme but its structure is the cysteine residues (15) are found in the C-terminal different (Lepage and Gaché, 1990). These threonine half of the molecule and 8 of them are very close to each clusters could be the site of O-glycosylations (Yama-other in a small cysteine-rich region between residues moto et al., 1984) and might be a common epitope to 293 and 339. No potential N-glycosylation site was the hatching enzyme and BP10 proteins. The threonine-found but we noted at positions 451-475 a series of 25 rich region splits the terminal half of the protein into 2 amino acids containing 14 threonines grouped in five regions which have almost exactly the same length, each containing 4 cysteine residues at the same positions and sharing 35% identity. The C-terminal end of the BP10 protein is thus made up of 2 homologous repeats of the same domain, arranged in tandem. Finally, we found at positions 88-93, the basic stretch RKKRKA whose sequence is reminiscent of those present in some precursors recognized by intracellular processing enzymes with a trypsin-like specificity (Bentley et al., 1986).

Homologies between the protein encoded by the BP10 cDNA and other proteins

Comparison of the BP10 protein and the hatching enzyme did not reveal any homology except for the short threonine-rich sequences. However, we noted at position 190 the HEXXH motif already found in the hatching enzyme and considered to be part of the zinc binding site of several metalloproteases (Whitham et al., 1986; Van-Wart and Birkedal-Hansen, 1990). The surrounding sequence satisfies the requirements for the “zinc signature” sequence defined by Jongeneel et al. (1989); it also shares a number of key amino acids with the active center consensus of the collagenases (San-chez-Lopez et al., 1988; Alexander and Werb, 1989), a metalloproteinase family to which the hatching enzyme belongs, but is different enough from this consensus to indicate that BP10 is probably not an extracellular matrix-degrading metalloprotease.

The BP10 amino acid sequence was then compared to the protein sequences of the NBRF and Swiss-Prot databases. The survey revealed striking homologies with proteins from different groups.

A 200 amino acid domain of the BP10 protein, beginning around position 90 and extending to about residue 295, is homologous to a whole small digestive protease of 22 × 103, isolated from the crayfish {Astacus), which is a classical zinc-dependent metal-loenzyme (Titani et al., 1987; Stricker et al., 1988). The alignment between the entire Astacus protease se-quence and the homologous domain in the protein encoded by the BP10 clone is presented in Fig. 2A. The overall homology score is 40% in a 200 amino acid overlap. A sequence homologous to this domain was also found in the human protein BMP-1. As shown in Fig. 2A, 42% of the 200 amino acid residues of the BP10 and BMP-1 domains are identical. The BMP-1 protein has been isolated from bones and is thought to play an important role in cartilage differentiation and bone formation. Its proteolytic activity was not demonstrated directly but was deduced from sequence comparisons and its substrate is not known. The triple alignment in Fig. 2A shows that the central region of this domain, which includes the zinc binding site, is particularly well conserved (over 60% homology). The positions of the four cysteines of the Astacus protease are conserved in the BP10 and BMP-1 proteins, suggesting that these domains are similarly folded. The N-terminal of the crayfish protease, determined for the purified active enzyme, aligns with position 94 in the BP10 sequence and position 121 in the BMP-1 sequence, just down-stream of the basic peptides RKKRK (BP10) and RSRSRR (BMP-1) which resemble many cleavage sites by intracellular processing enzymes. This suggests that, like most of the proteolytic enzymes, these proteases might be synthesized as precursors and later processed by proteolytic cleavage.

Fig. 2.

Homologies between the BP10 protein and other proteins. (A) Alignment between the BP10 protein, a protease from the crayfish (ASTPR) and the human protein BMP-1. The conserved cysteine residues are indicated by an asterisk. (=) indicates amino acid identity; (—) indicates conservative substitution between amino acids of the same homology group defined as (H), (KR), (LIVM), (AG), (ST), (YFW), (DE), (QN), (P), (C). (B) Homology between a BP10 domain and EGF-like domains of Delta (417–451), (Vaessin et al., 1987) and Notch (repeat 31), (Wharton et al., 1985) from Drosophila’, lin-12 (284– 319), (Greenwald, 1985) from Caenorhabditis’, factor IX (repeat 1), (Jaye et al., 1983), EGF-like protein (77–107), (Ciccodicola et al., 1989) and urokinase-type human plasminogen activator (26–63), (Riccio et al., 1985); uEGF-1 (repeat 7), (Hursh et al., 1987) from sea urchin. Part of the sequence of the mouse EGF is added for comparison (positions 977–1019 of the precursor) (Gray et al., 1983). Highly conserved residues are boxed. The cysteines are designated Cl to C6. Alignment at cysteine 3 was done by hand. (C) Alignment of the internal repeats of BP10, BMP-1, Cis and Clr. The sequences aligned are domain III of Cis (160–272), (Mackinnon et al., 1987); domain III of Clr (193–302), (Leytus et al., 1986); repeat R1 (339–449) and R2 (484–595) from BP10; repeat R1 (322–434), R2 (435–547) and R3 (591–703) from BMP-1. Identical residues in 4 sequences or more are boxed and shaded. Similar residues defined as above (A) are in open boxes. The cysteines are indicated by an asterisk.

Fig. 2.

Homologies between the BP10 protein and other proteins. (A) Alignment between the BP10 protein, a protease from the crayfish (ASTPR) and the human protein BMP-1. The conserved cysteine residues are indicated by an asterisk. (=) indicates amino acid identity; (—) indicates conservative substitution between amino acids of the same homology group defined as (H), (KR), (LIVM), (AG), (ST), (YFW), (DE), (QN), (P), (C). (B) Homology between a BP10 domain and EGF-like domains of Delta (417–451), (Vaessin et al., 1987) and Notch (repeat 31), (Wharton et al., 1985) from Drosophila’, lin-12 (284– 319), (Greenwald, 1985) from Caenorhabditis’, factor IX (repeat 1), (Jaye et al., 1983), EGF-like protein (77–107), (Ciccodicola et al., 1989) and urokinase-type human plasminogen activator (26–63), (Riccio et al., 1985); uEGF-1 (repeat 7), (Hursh et al., 1987) from sea urchin. Part of the sequence of the mouse EGF is added for comparison (positions 977–1019 of the precursor) (Gray et al., 1983). Highly conserved residues are boxed. The cysteines are designated Cl to C6. Alignment at cysteine 3 was done by hand. (C) Alignment of the internal repeats of BP10, BMP-1, Cis and Clr. The sequences aligned are domain III of Cis (160–272), (Mackinnon et al., 1987); domain III of Clr (193–302), (Leytus et al., 1986); repeat R1 (339–449) and R2 (484–595) from BP10; repeat R1 (322–434), R2 (435–547) and R3 (591–703) from BMP-1. Identical residues in 4 sequences or more are boxed and shaded. Similar residues defined as above (A) are in open boxes. The cysteines are indicated by an asterisk.

At the exact position where the Astacus protease homologous domain stops, there begins a ≈41) amino acid sequence which contains 6 cysteines and 14 polar residues that are either Asp/Glu or Asn/Gln. The search for homology revealed that this short region is significantly homologous to the EGF-like domain of several proteins from vertebrates and invertebrates. In Fig. 2B are shown the most notable homologies with: Delta and Notch, two proteins from Drosophila involved in neurogenesis (Vaessin et al., 1987; Wharton et al., 1985); lin-12, a protein involved in cell fate determination in Caenorhabditis, (Greenwald, 1985); two human enzymes, factor IX (Jaye et al., 1983) and urokinase (Riccio et al., 1985); and a human EGF-like protein (Ciccodicola et al., 1989). To these computer-selected sequences, we have added an EGF-like repeat from the sea urchin protein uEGF-1 (Hursh et al., 1987) and a mouse EGF sequence (Gray et al., 1983). The characteristic features of EGF-like domains are a specific pattern of 6 cysteines (cys 1 to cys 6) and the presence of conserved glycine, proline and aromatic residues (Greenwald, 1985; Cooke et al., 1987). The aligments presented in Fig. 2B are anchored to the CXNGG motif at cys 2 and to the consensus sequence CXC(X)5G(X)2CE at the end of the domain. The CXNGG sequence has been identified as a critical flanking sequence of the receptor binding site in EGF and urokinase (Komoriya et al., 1984; Appella et al.,1987) and the C-terminal side of the domain is particulary well conserved among many proteins (Engel, 1989). The relative positions of the cys residues of BP10 conform to the EGF consensus except for cys 3.

The region between cys 2 and cys 3 in BP10 has 9 amino acids instead of 5 found in most EGF-like sequences and the interval between cys 3 and cys 4 is unusually short. Such a cysteine spacing is also present in the EGF-like domain of the human complement C9 (Stanley et al., 1985). Thus, this short central region of the BP10 protein is clearly an EGF-like domain.

The proteins listed in Fig. 2B are a few cases among a large number of proteins which contain EGF-like domains (Engel, 1989). The function of these domains is still unclear, however, since they are found in proteins known to be involved in protein-protein interactions, it has been suggested that they are structural determinants necessary for these interactions (Davis et al., 1987; Appella et al., 1988; Furie and Furie, 1988). This view was supported by several studies showing a correlation between alteration of the EGF-like domains and loss of function (Davis et al., 1987; Rees et al., 1988). Many of the proteins containing EGF-like structures are enzymes. All are proteases of the serine protease family, with the notable exception of the BMP-1 metalloprotease from human bone. The EGF domain of BMP-1 is not strongly homologous to the EGF domain of BP10 and is located in the C-terminal half of the molecule as opposed to a central position in BP10. In addition, BP10 and BMP1 contain homologous internal repeats: the 2 C-terminal similar domains of BP10 have the same length and the same cysteine spacing as the 3 repeats of BMP-1 and several positions are occupied by identical or similar residues in the 5 repeats. A search for homologous repeats in other proteins of the NBRF and Swiss-Prot databases re-vealed significant similarities only with Clr and Cis, two serine proteases of the complement system (Leytus et al., 1986; MacKinnon et al., 1987). These proteases contain two repeats of this domain, I and in, flanking an EGF domain which have been shown to be involved in protein-protein interaction between components of the Cl complex (Tosi et al., 1987; Busby and Ingham, 1990). The sequences of the internal repeats of the two metalloproteases BP10 and BMP1 are aligned with the homologous domains of the two serine proteases Clr and Cis in Fig. 2C.

The domain organization of the BP10 and BMP-1 molecules is shown in Fig. 3. A similar structural analysis has been done recently (Shimell et al., 1991; O’Connor, personal communication) for the product of the tolloid gene of Drosophila which plays a role in the process of dorsal-ventral patterning of the embryo. The homologies with the crayfish protease, EGF, and Clr/Cls domains described above have also been described for the tolloid protein together with its structural similarities with the human BMP-1 protein.

Fig. 3.

Domain organization of the BP10 (top), BMP-1 (midle) and tollo id (bottom) proteins. Characteristic regions are shown as marked areas and designated as discussed in the text. Homologous domains are labelled in the same way except for dashed boxes which designate non-homologous domains. Rl, R2, R3, R4 and R5 designate similar repeats of the same domain.

Fig. 3.

Domain organization of the BP10 (top), BMP-1 (midle) and tollo id (bottom) proteins. Characteristic regions are shown as marked areas and designated as discussed in the text. Homologous domains are labelled in the same way except for dashed boxes which designate non-homologous domains. Rl, R2, R3, R4 and R5 designate similar repeats of the same domain.

The domain organization of the tolloid protein commu-nicated by Shimell et al. (1991) can be compared to the domain organization of BP10 in Fig. 3.

Expression of the BP10 gene during development

The expression of the BP10 gene during embryogenesis was first approached by northern blot analysis of total RNA extracted from staged embryos between the unfertilized egg and the pluteus. The results are shown in Fig. 4B, together with those obtained previously for the hatching enzyme gene (Fig. 4A). The BP10 probe recognizes a single transcript of ≈2.9 kb, about 200 bp longer than the cDNA, not including the poly (A)+ tail. The abundance of the BP10 transcript varies sharply during development. This transcript was not detected in unfertilized eggs or in embryos during the first cleavage stages. At the early blastula stage (about 128 cells), the BP10 transcript begins to accumulate, reaches its highest level at the prehatching blastula stage then decreases to an undetectable level when the blastula is hatched. As shown in Fig. 4, the patterns of the BP10 and hatching enzyme mRNAs accumulation are very similar. Both mRNAs are early embryonic messengers and they are both only transiently present during the blastula stages. The BP10 transcript reaches its highest prevalence about an hour after the hatching enzyme transcripts peak.

Fig. 4.

Accumulation of the BP10 and hatching enzyme transcripts during development. Northern blot analysis of total RNA prepared from embryos at the indicated stages and fractionated on a 1% formaldehyde agarose gel (10 μg per lane). The blot was probed with (A) a 0.5 kb EcoRI restriction fragment from the 3’ non-coding region of the hatching enzyme cDNA (data taken from Lepage and Gaché, 1990); (B) a 0.4 kb EcoRV restriction fragment from the BP10 cDNA. E, unfertilized egg; 2, 2-cell stage, 16, 16-cell stage; Bl to B6 blastula stages one hour apart beginning with the 128-cell stage; SB, swimming blastula; MB, mesenchyme blastula; EG, early gastrula; G, gastrula; D, prism; P, pluteus.

Fig. 4.

Accumulation of the BP10 and hatching enzyme transcripts during development. Northern blot analysis of total RNA prepared from embryos at the indicated stages and fractionated on a 1% formaldehyde agarose gel (10 μg per lane). The blot was probed with (A) a 0.5 kb EcoRI restriction fragment from the 3’ non-coding region of the hatching enzyme cDNA (data taken from Lepage and Gaché, 1990); (B) a 0.4 kb EcoRV restriction fragment from the BP10 cDNA. E, unfertilized egg; 2, 2-cell stage, 16, 16-cell stage; Bl to B6 blastula stages one hour apart beginning with the 128-cell stage; SB, swimming blastula; MB, mesenchyme blastula; EG, early gastrula; G, gastrula; D, prism; P, pluteus.

The transcriptional activity of the BP10 gene during early development has been measured by run-on assays using nuclei isolated from embryos at different stages. The results are shown in Fig. 5, together with an internal control provided by the early histones genes. Run-on transcripts of the BP10 gene are hardly detectable at the 16-cell stage, their prevalence in-creases rapidly until the mid-blastula stage (8 hours, about 200 cells per embryo), then falls off rapidly to become undetectable after hatching. We have pre-viously discussed the significance of this type of results. Comparative run-on assays at constant input of newly synthesized mRNA measure the relative transcriptional activity of the probed gene. If during early embryogen-esis of the sea urchin the total transcriptional activity increases several times (Davidson, 1986), an increase in relative activity corresponds to an even greater increase in absolute activity (Lepage and Gaché, 1990). This transient activity clearly parallels the variation of the transcript abundance, showing that during embryogen-esis the BP10 gene is mainly controlled at the transcriptional level. The variation of the relative transcriptional activities of the BP10 and hatching enzyme genes as a function of time are compared in Fig. 6. Also shown are the time course of hatching and the increase in cell number during the period analysed. The activity of the two genes follows the same type of kinetics, the BP10 gene being maximally activated 1.5 hours after the hatching enzyme gene.

Fig. 5.

Run-on assays of the transcriptional activity of the BP 10 gene during development. 5 ng of the pBP10 plasmid DNA were loaded per slot and hybridized with 6 × 106 cts min−1 of RNA synthesized by nuclei isolated from embryos at the indicated stages (BP10). As an internal control the labelled RNA was also hybridized to a plasmid carrying the H2A and H3 early histone genes (HIST). These genes have been shown to be transcribed early in development and transcriptionally repressed at the late blastula stages (Davidson, 1986). Stages are designated as indicated in Fig. 4.

Fig. 5.

Run-on assays of the transcriptional activity of the BP 10 gene during development. 5 ng of the pBP10 plasmid DNA were loaded per slot and hybridized with 6 × 106 cts min−1 of RNA synthesized by nuclei isolated from embryos at the indicated stages (BP10). As an internal control the labelled RNA was also hybridized to a plasmid carrying the H2A and H3 early histone genes (HIST). These genes have been shown to be transcribed early in development and transcriptionally repressed at the late blastula stages (Davidson, 1986). Stages are designated as indicated in Fig. 4.

Fig. 6.

Comparison between the transcriptional activities of the hatching enzyme (HE6) and BP10 genes. Relative transcriptional activity as percent of the maximal value: (▵) for the BP10 gene (densitométrie scanning of the run-on in Fig. 6); (▪) for the HE6 gene (data taken from Lepage and Gaché, 1990). Fertilization is at time zero. The rightmost curve is a typical hatching curve (in %) for a population of embryos raised at 18°C in the conditions described in Materials and Methods. The left curve describes the number of cell per embryo as a function of time.

Fig. 6.

Comparison between the transcriptional activities of the hatching enzyme (HE6) and BP10 genes. Relative transcriptional activity as percent of the maximal value: (▵) for the BP10 gene (densitométrie scanning of the run-on in Fig. 6); (▪) for the HE6 gene (data taken from Lepage and Gaché, 1990). Fertilization is at time zero. The rightmost curve is a typical hatching curve (in %) for a population of embryos raised at 18°C in the conditions described in Materials and Methods. The left curve describes the number of cell per embryo as a function of time.

Localization of the BP10 transcripts by in situ hybridization

The spatial distribution of the BP10 transcript in the blastula was determined by in situ hybridization to whole-mount embryos. The probe used in these experiments was a single-strand BP10 antisense RNA labelled with digoxigenin-ll-UTP. This probe recog-nizes exclusively the BP10 mRNA on a northern blot of total RNA prepared at different developmental stages (not shown). The results are shown in Fig. 7. Consistent with the temporal accumulation profile of the mRNA, the BP10 transcripts are not detectable in the 16-cell stage or late gastrula embryos (Fig. 7A). The BP10 mRNA can be detected only in blastula stages embryos, after the BP10 gene has reached its maximal activity. Surface views of a blastula comprising about 300 cells hybridized with an antisense BP10 RNA probe together with a control embryo hybridized with a sense probe are shown in Fig. 7 B, C. At this stage, the distribution of the transcripts in the blastula appears to be clearly asymmetric. In all the embryos observed, the labelling was restricted to a broad domain of contiguous cells spanning over 50% of the embryo. This result shows that the BP10 gene expression is spatially regulated. The differential activity of the BP10 gene allows recognition of two distinct domains in the early blastula.

Fig. 7.

Distribution of the BP10 transcripts in sea urchin embryos. Whole embryos were hybridized with single-strand RNA probes labelled with digoxygenin. (A) Embryos fixed at the 16-cell (4 h), blastula (8 h), and gastrula (24 h) mixed and hybridized with an antisense probe (optical section viewing). (B) Surface view of a blastula hybridized with a sense probe. (C) Surface view of a blastula hybridized with an antisense probe.

Fig. 7.

Distribution of the BP10 transcripts in sea urchin embryos. Whole embryos were hybridized with single-strand RNA probes labelled with digoxygenin. (A) Embryos fixed at the 16-cell (4 h), blastula (8 h), and gastrula (24 h) mixed and hybridized with an antisense probe (optical section viewing). (B) Surface view of a blastula hybridized with a sense probe. (C) Surface view of a blastula hybridized with an antisense probe.

Identification of the BP10 protein in embryonic extracts

To characterize the BP10 protein, we generated polyclonal antibodies against a BP10-μ-gal fusion protein that contains most of the metalloprotease homologous domain, the EGF-like region and the two C-terminal repeats. The antibodies were purified by affinity chromatography.

A western blot analysis of proteins extracted from unfertilized eggs or embryos at various stages of development is shown in Fig. 8. The antibodies recognize a single protein of relative molecular mass 55 × 103 in protein extracts (nonreducing conditions). This value is significantly smaller than the 64 × 103 value predicted from the cDNA sequence for the mature protein. This suggests that the BP10 protein has an abnormal migration on SDS-PAGE. Actually, this high relative mobility was expected because the protein is predicted to contain 18 cysteine residues and is thus probably tightly folded by several disulfide bridges. Preliminary experiments showed that the relative molecular mass of the protein recognized on western blots shifts from 55 × 103 to about 66 × 103 upon reduction.

Fig. 8.

Western blot analysis of proteins extracted from unfertilized eggs and staged embryos. Proteins (150 μg/ lane) were separated by electrophoresis on a 10% polyacrylamide gel and transferred to a nitrocellulose filter,’rhe blot was incubated with purified anti-BPIO antibodies and processed with an alkaline-phosphatase-conjugated secondary antibody. E, unfertilized egg; 16, 16-cell stage; EB, early blastula; PHB, prehatching blastula; SB, swimming blastula; EMB, early mesenchyme blastula; MB, mesenchyme blastula; EG, early gastrula; G, gastrula; D, prism stage.

Fig. 8.

Western blot analysis of proteins extracted from unfertilized eggs and staged embryos. Proteins (150 μg/ lane) were separated by electrophoresis on a 10% polyacrylamide gel and transferred to a nitrocellulose filter,’rhe blot was incubated with purified anti-BPIO antibodies and processed with an alkaline-phosphatase-conjugated secondary antibody. E, unfertilized egg; 16, 16-cell stage; EB, early blastula; PHB, prehatching blastula; SB, swimming blastula; EMB, early mesenchyme blastula; MB, mesenchyme blastula; EG, early gastrula; G, gastrula; D, prism stage.

As shown in Fig. 8, the abundance of the BP10 protein varies significantly during development from the egg to the late gastrula stage. The BP10 protein is not detectable either in the unfertilized egg, which indicates that there is no maternal BP10 protein, or in the very early cleavage stages. It begins to be detected, albeit weakly, at the early blastula stage (about 250 cells), its abundance peaks at the prehatching and swimming blastula stages then decreases rapidly and remains constant throughout development until the prism stage which was the last analysed.

Immunolocalization of the BP10 protein in embryos

The purified anti-BPIO antibodies were used to localize the protein during its synthesis and secretion by immunostaining whole embryos. As shown by the northern blot analysis, the BP10 transcripts accumulate transiently during a short period at the blastula stage. We thus tried to localize the protein at three different blastula stages: early, hatching and late blastula. The results are presented in Fig. 9. We first detected the BP10 protein at the early blastula stage, at about the same time as the transcripts accumulate in the embryo. At this time, the protein is present in a diffuse pattern in the perinuclear region suggesting that it is synthesized in the endoplasmic reticulum surrounding the nuclei. In all the labelled embryos, the BP10 staining is restricted to a domain covering about half the embryo, roughly the same size as the territory labelled by in situ hybridization (Fig. 9A). Thus, the localization of the BP10 protein clearly correlates with the distribution of its mRNA.

Fig. 9.

Immunolocalization of the BP10 protein in whole embryos. Embryos at the early, hatching and late blastula stages were stained with anti-BPIO and alkaline-phosphatase-conjugated secondary antibodies. (A) Embryo at the early blastula stage (about 250 cells); (B and C) Hatching blastula stage embryo (about 450 cells) upper and lower halves (polar views). (D,E,F) Embryo at the hatching blastula stage (side views): (D) Hoechst 33342 fluorescence and bright-field images superimposed; (E and F) bright-field image of both sides of the same embryo; (G) and (H) surface and optical section views of a late blastula; (I) late blastula at the beginning of ingression of primary mesenchyme cells.

Fig. 9.

Immunolocalization of the BP10 protein in whole embryos. Embryos at the early, hatching and late blastula stages were stained with anti-BPIO and alkaline-phosphatase-conjugated secondary antibodies. (A) Embryo at the early blastula stage (about 250 cells); (B and C) Hatching blastula stage embryo (about 450 cells) upper and lower halves (polar views). (D,E,F) Embryo at the hatching blastula stage (side views): (D) Hoechst 33342 fluorescence and bright-field images superimposed; (E and F) bright-field image of both sides of the same embryo; (G) and (H) surface and optical section views of a late blastula; (I) late blastula at the beginning of ingression of primary mesenchyme cells.

Close to the time of hatching, the blastulas are quite intensely labelled by the antibodies. In each labelled cell, the protein is detected as a brightly stained spot located between the nucleus and the apical surface indicating that the protein is concentrated in a subcellu-lar compartment (Fig. 9D). The strong signals allowed us to examine the staining on both sides of whole blastulas (Fig. 9B,C,E,F). We found a percentage of labelled cells ranging from 60 to 73% when the embryos are composed of 400–500 cells. This variation in the percentage of labelled cells was observed even among embryos with the same number of cells. The border between labelled and unlabelled domains was generally rather sharply delineated but in a significant number of cases a graded transition over a few cells was clearly observed. In the same way, the boundary was in most cases straight but sometimes rather irregular.

About 1 h 30 min after hatching, the embryo is a late blastula whose shape is no longer spherical. The embryo is lengthened along the animal-vegetal axis, the cell layer is flattened at the vegetal pole and thicker than the meridional sides. By localizing the protein at this stage, it was thus possible to determine the orientation of the territory expressing the BP10 gene by reference to the animal-vegetal axis of the embryo. Most embryos at the late blastula stage were weakly stained indicating that the synthesis and secretion of the protein was almost terminated. However, a few percent of the embryos still displayed an intense labelling. As shown in Fig. 9H, this labelling is located just under the apical membrane of the blastomeres. Both surface (Fig. 9G) and optical section viewing (Fig. 9H) of the same labelled embryo show that no staining is visible in the thickened vegetal plate. This result shows that BP10 is synthesized by the cells of a territory covering the whole animal hemisphere and extending into the vegetal hemisphere up to the boundary of the vegetal plate. The orientation of the domain expressing the BP10 gene was also confirmed by using as polarity marker the subequatorial pigment ring of Paracentrotus lividus embryos (results not shown), as described for the hatching enzyme localization (Lepage et al., results to be published). Finally, at the beginning of ingression of the primary mesenchyme, the embryos did not show any labelling above background (Fig. 91). These results show that BP10 is a protein specific to the blastula stage, synthesized in a distinct territory in the embryo and probably secreted into the perivitelline space before the beginning of gastrulation.

Effects of antibodies and peptides on hatching and morphogenesis

Comparisons of the BP10 protein sequence with other proteins suggested that BP10 is a secreted protease containing structural domains important for function probably because they are implicated in protein-protein interactions. To evaluate the role of BP10 during development, we used two complementary approaches. First, we attempted to block its activity with a polyclonal antibody, and second, we tried to interfere with the protein-protein interactions using a synthetic peptide derived from the EGF-like domain of interaction. Assuming that the BP10 protein is secreted into the perivitelline space, these reagents could be effective if simply added to the culture medium of early embryos whose fertilization envelopes had been removed. These approaches, however, might provide evidence for a protein function, but cannot provide proof like mutations or gene “knock-outs” do in organisms where genetic tools are available.

Effects of antibodies on hatching

The BP10 gene and the hatching enzyme gene have a very similar pattern of expression and both code for secreted metalloproteases. Thus we should ask if BP10 is also involved in hatching through processes which involve protein-protein interactions. BP10 might be a second hatching enzyme acting synergistically with the collagenase-like hatching enzyme to digest the fertiliz-ation envelope (Lepage and Gaché, 1989, 1990). However, all the metalloproteases involved in extra-cellular matrix degradation characterized so far belong to the collagenase family, suggesting that only this class of enzymes is responsible for extracellular matrix degradation. Furthermore, we never detected any protein with hatching activity other than the collagen-ase-like hatching enzyme. Alternatively, the BP10 protease might activate the zymogen of the hatching enzyme, and thus would be similar to the proteases of the activation cascades both by its structural organiz-ation and by its function, i.e., instead of a serine protease activating a serine protease zymogen, a metalloprotease would activate a collagenase.

To test these two hypotheses, embryos devoid of their own fertilization envelopes were raised to the blastula stage. Shortly before the time of hatching, isolated fertilization envelopes were added to the embryos together with anti-BPIO antibodies at various concentrations. The fertilization envelopes were digested at the same rate, whether the antibodies were present or not. Under the same conditions, an antibody directed against the hatching enzyme blocks the degradation of the fertilization envelopes.

(b) Effects of antibodies on morphogenesis

Purified anti-BPIO antibodies or purified IgGs from the preimmune animals were added at concentrations ranging from 40 to 200 μ μ g/ml. Purified IgGs from preimmune animals did not affect morphogenesis but the embryos developed slightly slower than the controls (Fig. 10A). Morphogenesis was also normal when embryos were raised in the presence of 40 μg/ml of purified anti-BPIO antibodies. In the presence of 100 μg/ml of antibodies, about 30% of the embryos developed into pluteus-like larvae with an abnormal shape and at 200 μg/ml, almost all the embryos developed into larvae radialized to various degrees. In the less severely affected embryos the oral lobe (their most animal part) and the anal arms were often underdeveloped or projected laterally. These embryos frequently had supernumerary spicules and defects in bilateral symmetry. Fully radialized embryos were characterized by their ovoid shape, the axial position of the archenteron and the numerous (4 to 8) and frequently ramified spicules which formed an entangle-ment at the base of the archenteron (Fig. 10B). In all these experiments, the antibodies were added at the 64-cell stage. When the antibodies were added after hatching, embryo development was normal even with the highest concentrations of antibodies.

Fig. 10.

Inhibition of morphogenesis by anti-BPIO antibodies and synthetic peptides derived from the EGF-like domain. Antibodies or peptides were added to 200 μl cultures of 150–200 embryos without fertilization envelopes at the 16- to 32-cell stage. The cultures were incubated at 20°C in a humid chamber and the embryos were examined 48 hours after fertilization. Embryos were photographed with bright-field and polarized light. (A) Untreated pluteus larva raised in ASW. Normal plutei were also obtained in control experiments with 200 μg/ml of IgGs purified from preimmune sera or with 400 μ g/ml of synthetic neurotensin. (B) Radialized larva obtained by treatment with purified anti-BPIO antibodies (200 μ g/ml). (C and D) Larvae obtained by treatment with peptide PII (250 μ μ ml): (C) pluteus with slight morphological defects; (D) fully radialized larva.

Fig. 10.

Inhibition of morphogenesis by anti-BPIO antibodies and synthetic peptides derived from the EGF-like domain. Antibodies or peptides were added to 200 μl cultures of 150–200 embryos without fertilization envelopes at the 16- to 32-cell stage. The cultures were incubated at 20°C in a humid chamber and the embryos were examined 48 hours after fertilization. Embryos were photographed with bright-field and polarized light. (A) Untreated pluteus larva raised in ASW. Normal plutei were also obtained in control experiments with 200 μg/ml of IgGs purified from preimmune sera or with 400 μ g/ml of synthetic neurotensin. (B) Radialized larva obtained by treatment with purified anti-BPIO antibodies (200 μ g/ml). (C and D) Larvae obtained by treatment with peptide PII (250 μ μ ml): (C) pluteus with slight morphological defects; (D) fully radialized larva.

(c) Effects of the EG F domain peptides on morphogenesis

Two synthetic peptides corresponding to the N-terminal part of the EGF-like domain of BP10 were synthesized. The disulfide pairing of this domain is not known but is likely to be the same as that of mammalian EGF (Cooke et al., 1987). Therefore, the peptides were either partially (PI) or fully (PII) refolded by selective oxidation of the cysteine residues according to the known structure of EGF.

In the presence of 400 μg/ml of a control peptide, the embryos developed into normal plutei (Fig. 10A). In the presence of 50 μg/ml of peptide PII, most of the larvae were normal but a few percent displayed some aberrations, such as the axial position of the oral lobe, the presence of supernumerary spicules, and bilateral asymmetries. In the presence of 250 μg/ml of peptide PII, 70-80% of the embryos developed into small plutei with defects in the position of the oral lobe or anal arms (Fig. 10C). The size and shape of the spicules were frequently abnormal. These larvae resembled the partially radialized larvae obtained by treatment of the embryos with 8-chloroxanthine as described by Horsta-dius (1973). The remaining 30% of the embryos developed into ovoid larvae very similar to the fully radialized larvae obtained with the antibodies (Fig. 10D). A tripartite gut was always present but in most cases the mouth was not formed. Peptide PI elicited the same kind of perturbations but at twice the concen-tration of peptide PII. Thus both peptides had less pronounced effects on morphogenesis than antibodies but elicited the same kind of aberrations.

The BP 10 gene is expressed transiently in a spatially restricted pattern during cleavage

We have described the cloning and characterization of a sea urchin mRNA encoding an embryonic protein that we call BP10 (Blastula Protease 10). The expression pattern of the corresponding gene was analysed at different levels. The results of northern blot exper-iments showed that the BP10 transcripts are undetect-able in the unfertilized egg, accumulate rapidly during cleavage stages then disappear at hatching. The BP10 mRNA is present transiently during a short period during cleavage and is thus, with the hatching enzyme mRNA, among the most precocious transcripts specific to the early blastula stage characterized so far. By run-on experiments, we have demonstrated that this transient appearance of the mRNA results from a transcriptional regulation of the corresponding gene. Therefore, the BP10 mRNA is a zygotic gene product synthesized in the very first hours following fertilization and activation of the zygotic genome. The results of in situ hybridization demonstrate that the BP10 gene has a spatially restricted expression in the early embryo. This gene is only activated in a clearly delimited territory covering about 2μ of the embryo. Thus, the BP 10 gene should be a useful marker to study the control of transcriptional activity during early development and the spatial regulation of gene expression in the sea urchin embryo.

BP 10 is synthesized by the cells constituting the presumptive ectoderm and probably secreted in the perivitelline space

The BP10 protein begins to be detected on western blots at the early blastula stage, its abundance reaches its highest level late in cleavage and decreases signifi-cantly just before gastrulation. The protein seems to persist at a low level late in embryogenesis. Whether this residual protein is functional or not is not known. This persistence might related to a low susceptibility to proteolytic degradation of a tightly folded protein.

The protein was localized during its synthesis and secretion by immunolabelling whole embryos. The results show that only blastula stage embryos are strongly and specifically labelled. The BP10 protein could not be detected in early cleavage or gastrula stage embryos. In addition, at all stages where the protein has been detected it was present in a restricted area of the embryo covering about 2μ of its surface. Thus, the temporal and spatial distribution of the protein is consistent with the accumulation profile of the mRNAs and their spatial localization.

To determine the orientation of the domain express-ing the BP10 gene, we localized the protein at the late blastula stage where it is still present in a significant amount and where the morphology can easily be related to the embryonic polarity. The results of this experiment demonstrate that BP10 is synthesized in a territory covering the whole animal hemisphere and extending to the edge of the vegetal plate. A comparison of the position of the domains expressing or not the BP10 protein with the fate map of the late blastula stage shows that the limit separating the two domains corresponds to a sharp border of cell fates. The vegetal plate will differentiate into skeletogenic mesenchyme (mesoderm), gut (endoderm) and secondary mesenchyme cells, while the upper territory of the blastula will only contribute to the ectoderm of the larva. Thus, the BP10 protein and its mRNA are precocious molecular markers of the presumptive ectoderm territory.

Finally, BP10 is predicted to be a secreted protein but it was not ascertained whether the protein was exported to the blastocoel or to the perivitelline space. Localization of the BP10 protein in whole embryos (and in individual cells from dissociated embryos, unpublished results) clearly showed that in hatching blastulas, the BP10 protein is transported towards the apical face of the epithelial cells, and can be seen just beneath the membrane in older blastulas. As the predicted protein sequence does not contain any putative hydrophobic transmembrane spanning region, it is likely that BP10 is secreted into the perivitelline space. Thus, the BP10 protein is synthesized and secreted by ectoderm cells and probably acts on the surface of the embryo.

BP10 is a regulatory metalloprotease

The developmental regulation of the BP10 gene and its pattern of spatial expression are indicative of both an early and important role in embryogenesis but do not provide clues to the function of the corresponding protein. The analysis of the BP10 protein structure and the homologies found do yield some information. From the N terminus to the end of the molecule, we identified 7 domains: a signal peptide, a putative activation peptide, a catalytic domain, an EGF-like module and two repeats about 110 amino acid long, separated by a threonine-rich region.

A portion of the molecule (40%) is homologous to a small digestive enzyme known to be a Zn2+-dependent protease. The critical residues at the active site are well conserved. Thus BP10 appears to be a secreted metalloprotease. The same protease homologous domain has been found in the human protein BMP-1 and in the Drosophila protein tolloid. This domain functions isolated as a digestive protease and consti-tutes the catalytic domain of larger proteins probably evolved to fullfill more complex tasks. The largest part of BP10 protein (60%) is made of non-catalytic modules, the EGF-like domain and 2 repeats. It is commonly believed that this complex modular organiz-ation distinguishes regulatory proteases from other proteases involved in simpler digestive processes (Neur-ath, 1989; Patthy, 1985). The cardinal examples of those complex enzymes are the vertebrate serine proteases of blood coagulation, fibrinolysis and complement sys-tems. In addition to the catalytic domain, most of these proteases contain a propeptide, one to five autonomous structural and folding units referred to as kringles (3 disulfide bridges) or fingers (2 disulfide bridges) and one or several EGF-like domains (Furie and Furie, 1988; Patthy, 1985). All the proteases containing an EGF-like sequence characterized to date are involved in proteolytic cascades of activation of zymogens or precursor molecules. Several lines of evidence suggest that EGF-like domains and repeated domains are necessary for the binding of these enzymes to other macromolecules, cellular receptors, zymogens, or cofactors (Patthy et al., 1984; Appella et al., 1988). Most of the enzymes containing the domains described above are serine proteases. The presence of an EGF-like domain and of Cis homologous repeats in a Zn2+-metalloprotease is a rare characteristic feature of the BP10 protein, only found so far in BMP-1, a human protein which has been shown to play a role in bone morphogenesis (Wozney et al., 1988) and in the recently characterized product of the tolloid gene necessary for dorsal-ventral patterning in Drosophila (O’Connor, personal communication). Actually, BP10, BMP-1 and tolloid are constructed with the same building blocks: the metalloprotease domain, the EGF-like sequence and the large C-terminal repeats. The three metalloproteases BP10, BMP-1 and tolloid might thus be the first members of a family of widely conserved enzymes involved in developmental or differentiation processes.

Functional implications of the homology between BP10 and the morphogenetic proteins BMP-1 and tolloid

The BP10 protein is globally homologous to two proteins from distantly related organisms, the BMP-1 protein from human and the tolloid protein from Drosophila.

The BMP-1 protein has been characterized during a search for factors involved in bone morphogenesis. In mammals, it has been shown that bone extracts implanted at ectopic sites induce the formation of bone by a process that mimics normal bone development (Urist, 1965; Reddi and Huggins, 1972). A fraction that can induce the bone morphogenetic response has been extracted with denaturing solvents from demineralized bone and highly purified (Wang et al., 1988). The preparation contained 3 tightly associated proteins called BMP-1, BMP-2, and BMP-3. Sequence analysis of their cDNAs revealed that two of them, BMP-2 and BMP-3, are peptide growth factors of the TGF-μ family. The third is the metalloprotease BMP-1 to which BP10 is homologous. The recombinant BMP-1 protein, like the two others, can induce the formation of cartilage (Wozney et al., 1988; Wang et al., 1990). Its precise role is not known, but its close association with growth factors of the TGF-β family suggests that BMP-1 regulates directly or indirectly the activity of these factors (Wozney et al., 1988).

The tolloid locus was identified during a screening for embryonic lethal mutations in Drosophila (Jürgens et al., 1984). Mutations in the tolloid gene affect the dorsal-ventral patterning of the larva. Strong alleles of tolloid exhibit phenotypes resembling the ventralized phenotype of the dominant alleles of Toll, that is, the larva lacks the amnioserosa, which derives from the most dorsal cells of the blastoderm, and the ventrola-teral neurogenic ectoderm is expanded dorsally. Gen-etic analysis suggests that tolloid is involved in a cascade of interactions between zygotic gene products necessary for cell fate specification in the dorsal region of the embryo. The decapentaplegic gene (dpp), which is expressed in the dorsal region, plays a crucial role in pattern formation of the dorsal half of the blastoderm stage embryo (St. Johnston and Gelbart, 1987; Irish and Gelbart, 1987; Ferguson and Anderson, 1991). On the basis of genetic and molecular evidence, it has been suggested that the function of tolloid is to potentiate the activity of dpp and that this regulation is mediated by a direct interaction between the two proteins (Shimell et al., 1991; Ferguson and Anderson, 1991). The dpp gene codes for a polypeptide of the TGF-β family (Padgett et al., 1987). It is to date the only TGF-β-like growth factor characterized in an invertebrate.

Thus, two homologous proteases, BMP-1 and tolloid, appear to be closely associated to growth factors of the TGF-β family. Strikingly, the mature peptide sequences of BMP-2 and dpp have 75% amino acid identity. Because of this exceptionally high conservation be-tween proteins from so distantly related organisms, it has been suggested that BMP-2 might be the vertebrate homologue of the Drosophila dpp protein (Gelbart, 1989; Wozney, 1989). In both cases of BMP-1 and tolloid, the associated growth factors play a pivotal role in a morphogenetic process. This provides compelling evidence of the strong structural and functional hom-ology between two pairs of interacting proteins, BMP-1 and BMP-2 on one hand, and tolloid and dpp on the other. As the BP10 protein is homologous to BMP-1 and tolloid, it is possible that the sea urchin metalloprotease is involved in a regulatory interaction with a polypeptide of the TGF-β family, which might be the homologue of dpp in Drosophila and BMP-2 in mammals.

The involvement of BP10 in a process analogous to the process of bone formation and morphogenesis in which the BMP-1 protein takes part seems unlikely. In contrast, the data that we present on the BP10 gene expression compare strikingly to the data on the spatio-temporal expression pattern of the tolloid gene (O’Con-nor, personal communication). Both genes are zygotic, transcribed early in embryogenesis and expressed transiently before germ layer formation. The spatial patterns of expression of both genes share several important features. At the cellular blastoderm stage, the tolloid transcripts are present in a territory spanning about 50% of the embryo circumference. At the blastula stage, the BP10 mRNAs are detected in a territorry spanning over 50% of the embryo. Most importantly, the boundaries of these territories corre-spond to borders of cell fates and in both cases these boundaries are perpendicular to a primordial maternal axis, the dorsal-ventral axis in Drosophila and the animal-vegetal axis in the sea urchin embryo. Thus, the BP10 gene might be the functional homologue in sea urchin of the tolloid gene in Drosophila. In this case, BP10, together with a TGF-μ homologous to dpp, could mediate short range cell-cell interactions playing an important role in pattern formation along the animal-vegetal axis.

Is BP10 involved in pattern formation along the animal-vegetal axis?

In order to test the role of BP10 in development of the embryo, we tried to interfere with the activity of the protein by using anti-BPIO antibodies and synthetic peptides. These reagents had dramatic effects on morphogenesis. Actually, the in vivo effects of the antibodies and peptides can be described using the same words as those employed by Hdrstadius to describe the development of vegetal fragments of early sea urchin embryos obtained by surgical removal of the animal cap cells: “⃛ovoid larva with a straight digestive tract⃛”, “⃛some irregular spicules under a cap of thick epithelium⃛”, “⃛dwarf plutei with an oral lobe underdevelopped⃛” (Hdrstadius, 1973). Thus, the results are consistent with a role for BP10 in differen-tiation and morphogenesis of the structures derived from the animal region of the egg. We cannot completely rule out the possibility that some of the effects observed result from a non-specific inhibition unrelated to BP10. However, the action of the antibodies is stage specific and occurs at moderate concentrations, and furthermore, two different approaches gave essentially the same results. Thus, the perturbations of morphogenesis observed apparently result from an inhibition of the BP10 protein function.

Possible implications in the processes of pattern formation along the animal-vegetal axis

From a comparative point of view, sea urchins and flies have adopted very dissimilar strategies for oogenesis and early development. A current idea regarding the various forms of animal development is that the complex genetic systems involved in generation of positional information along the two axes of the Drosophila egg and embryo are very atypical and unique to this embryo (Davidson, 1990). The anterior-posterior axis of this embryo is defined by a set of maternal coordinate genes (Nüsslein-Volhard et al., 1987). The action of these genes results in the appearance of morphogenetic activities at the anterior and posterior poles of the egg with long-range influence through the syncitial embryo. This is likely a special process closely associated with a meroistic oogenesis, the structure of this large egg, the syncitial nature of cleavage and the segmented body plan the Drosophila embryo.

This might be far less true for the genetic system involved in pattern formation along the dorsal-ventral axis. In Drosophila the determinants for dorsal-ventral polarity are maternal. They belong to a defined set of gene products including surface receptors and their ligands, extracellular serine proteases, transducing molecules and transcription factors, which interact during oogenesis and after fertilization through a complex pathway (Anderson, 1987; Govind and Stew-ard, 1991). The end result of this regulatory cascade is the formation of a ventral-dorsal concentration gradient of a maternal morphogen, the protein dorsal, in the blastoderm nuclei (Roth et al., 1989; Rushlow et al., 1989; Steward, 1989). The asymmetric distribution of the dorsal protein controls in turn the selective activation or repression of the transcription of zygotic regulatory genes such as twist, snail, dpp, zen and probably tolloid, which ultimately specify cell fates and refine pattern along the dorsal-ventral axis (Rushlow et al., 1987; Thisse et al., 1987, 1991; Ip et al., 1991). The homology between the tolloid and BP10 proteins and the strong similarities between the expression pattern of the tolloid and BP10 genes suggest that the mechanisms of spatial regulation of gene expression and possibly of cell fate specification in the early sea urchin embryo may share common elements with the mechanisms generating the dorsal-ventral polarity of the Drosophila embryo.

It will be interesting to determine if the spatially restricted expression of BP10 results from a negative regulation by a maternal transcription factor related to dorsal. If such a transcription factor does exist in the sea urchin embryo, is it asymmetrically distributed during cleavage and does it control cell fate specification by regulating the spatial expression of a set of zygotic regulatory genes just as in Drosophila? By using the available molecular probes, it should now be possible to determine which genes among the maternal and zygotic dorsal-ventral polarity genes of Drosophila are con-served in sea urchins, to describe their pattern of expression and eventually to study their function. A clearer picture of the components involved in determi-nation along the animal-vegetal axis of the sea urchin embryo might then emerge.

We wish to thank M.B. O’Connor (University of Califor-nia, Irvine) for communicating the sequence of the tolloid protein and exchanging results before publication. We are very grateful to the F. Cuzin and J. Pouyssegur teams (Centre de Biochimie, Nice) for their constant interest. We thank G. Leblanc (C.E.A., Villefranche) for many generous gifts, M. Busslinger (IMP, Vienna) for his gift of the histone clone, M. Piovant (Université de Marseille-Luminy) for fruitful dis-cussions and all our colleagues for their help, especially C. Sardet for advice and P. Chang for stimulating discussions and carefully reading the manuscript. This work was supported in part by grants from the Centre National de la Recherche Scientifique, the Association pour la Recherche sur le Cancer and the Fondation pour la Recherche Medicale.

Alexander
,
C. M.
and
Werb
,
Z.
(
1989
).
Proteinases and extracellular matrix remodeling
.
Curr. Opinion Cell Biol
.
1
,
974
982
.
Anderson
,
K. V.
(
1987
).
Dorsal-ventral embryonic pattern genes of Drosophila
.
Trends Genet
.
3
,
91
97
.
Angerer
,
L. M.
,
Cox
,
K. H.
and
Angerer
,
R. C.
(
1987
).
in Guide to Molecular Cloning Techniques
,
Methods in Enzymology
152
,
649
661
.
Appella
,
E.
,
Robinson
,
E. A.
,
Ulrich
,
S. J.
,
Stoppdli
,
M. P.
,
Corti
,
A.
,
Cassani
,
G.
and
Blasi
,
F.
(
1987
).
The receptor binding sequence of urokinase. A biological function for the growth factor module of proteases
.
J. biol. Chem
.
262
,
4437
4440
.
Appella
,
E.
,
Weber
,
I.
and
Blasi
,
F.
(
1988
).
Structure and function of epidermal growth factor-like regions in proteins
.
FEBS Letters
231
,
1
4
.
Bentley
,
A. K.
,
Rees
,
D. J. G.
,
Rlzza
,
C.
and
Brownlee
,
G. G.
(
1986
).
Defective propeptide processing of blood clotting factor IX caused by mutation of arginine to glutamine at position 4
.
Cell
45
,
343
348
.
Boveri
,
T.
(
1901
).
Uber die polaritàt des Seeigel-Eies
.
Verh. Phys, med. Ges., Wuzburg
,
34
,
145
175
.
Bradford
,
M.
(
1976
).
A rapid and sensitive method for the quantitation of micrograms quantities of protein utilizing the principle of protein-dye binding
.
Anal. Biochem
.
72
,
248
254
.
Busby
,
T. F.
and
Ingham
,
K. C.
(
1990
).
NH2-terminal calcium binding domain of human complement Cis mediates the interaction of Clr with Clq
.
Biochemistry
29
,
4613
4618
.
Cathala
,
G.
,
Savouret
,
J. F.
,
Mendez
,
B.
,
West
,
B. L.
,
Karin
,
M.
,
Martial
,
J. A.
and
Baxter
,
J. D.
(
1983
).
A method for isolation of intact, translationally active ribonucleic acid
.
DNA
2
,
327
333
.
Cavener
,
D. R.
(
1987
).
Comparison of the consensus sequence flanking translational start sites in Drosophila and vertebrates
.
Nucleic Acids Res
.
15
,
1353
1361
.
Ciccodicola
,
A.
,
Dono
,
R.
,
Obici
,
S.
,
Simeone
,
A.
,
Zoilo
,
M.
and
Pérsico
,
M. G.
(
1989
).
Molecular characterization of a gene of the “EGF family” expressed in undifferentiated human NTERA2 teratocarcinoma cells
.
EMBO J
.
8
,
1987
1991
.
Cooke
,
R. M.
,
Wilkinson
,
A. J.
,
Baron
,
M.
,
Pastore
,
A.
,
Tappin
,
M. J.
,
Campbell
,
I. D.
,
Gregory
,
H.
and
Sheard
,
B.
(
1987
).
The solution structure of human epidermal growth factor
.
Nature
327
,
339
341
.
Cox
,
K. H.
,
Deleon
,
D. V.
,
Angerer
,
L. M.
and
Angerer
,
R. C.
(
1984
).
Detection of mRNAs in sea urchin embryos by in situ hybridization using asymmetric RNA probes
.
Dev. Biol
.
101
,
485
502
.
Davidson
,
E. H.
(
1986
).
Gene Activity in Early Development. 3rd edition. Orlando, Florida: Academic Press
.
Davidson
,
E. H.
(
1989
).
Lineage-specific gene expression and the regulative capacities of the sea urchin embryo: a proposed mechanism
.
Development
105
,
421
445
.
Davidson
,
E. H.
(
1990
).
How embryos work: a comparative view of diverse modes of cell fate specification
.
Development
108
,
365
389
.
Davis
,
C. G.
,
Goldstein
,
J. L.
,
Südhof
,
T. C.
,
Anderson
,
R. G. W.
,
Russell
,
D. W.
and
Brown
,
M. S.
(
1987
).
Acid-dependent ligand dissociation and recycling of the LDL receptor mediated by growth factor homology region
.
Nature
326
,
760
765
.
Duboule
,
D.
and
Dollé
,
P.
(
1989
).
The structural and functional organization of the murine hox family resembles that of Drosophila homeotic genes
.
EMBO J
.
8
,
1497
1505
.
Engel
,
J.
(
1989
).
EGF-like domains in extracellular matrix proteins: localized signals for growth and differentiation?
FEBS Lett
.
251
,
1
7
.
Feinberg
,
A. P.
and
Vogelstein
,
B.
(
1983
).
A technique for radiolabelling DNA restriction fragments to high specific activity
.
Anal. Biochem
.
132
,
6
13
.
Ferguson
,
E. L.
and
Anderson
,
K. V.
(
1991
).
Dorsal-ventral pattern formation in the Drosophila embryo: the role of zygotically active genes
.
Curr. Top. Dev. Biol, (in press)
.
Furie
,
B.
and
Furie
,
B.
(
1988
).
The basis of blood coagulation
.
Cell
53
,
505
518
.
Gelbart
,
W. M.
(
1989
).
The decapentaplegic gene: a TGF-β homologue controlling pattern formation in Drosophila
.
Development
Supplement
,
65
74
.
Ghosh
,
S.
,
Gifford
,
A. M.
,
Riviere
,
L. R.
,
Tempst
,
P.
,
Nollan
,
G. P.
and
Baltimore
,
D.
(
1990
).
Cloning of the p50 DNA binding subunit of NF-xB: homology to rel and dorsal
.
Cell
62
,
1019
1029
.
Govind
,
R.
and
Steward
,
R.
(
1991
).
Dorsoventral pattern formation in Drosophila
.
Trends Genet
.
7
,
119
125
.
Gray
,
A.
,
Dull
,
T. J.
and
Ullrich
,
A.
(
1983
).
Nucleotide sequence of epidermal growth factor cDNA predicts a 128,000 molecular weight protein precursor
.
Nature
303
,
722
725
.
Greenwald
,
I.
(
1985
).
lin-12, a nematode homeotic gene, is homologous to a set of mammalian proteins that includes epidermal growth factors
.
Cell
43
,
583
590
.
Harlow
,
E.
and
Lane
,
D.
(
1988
).
Antibodies, a Laboratory Manual. Cold Spring Harbor Laboratory
.
Hemmati-Brivanlou
,
A.
,
Frank
,
D.
,
Boice
,
M. E.
,
Brown
,
B. D.
,
Sive
,
H. L.
and
Harland
,
R. M.
(
1990
).
Localization of specific mRNAs in Xenopus embryos by whole-mount in situ hybridization
.
Development
110
,
325
330
.
Hopwood
,
N. D.
,
Pluck
,
A.
and
Gurdon
,
J. B.
(
1989
).
A Xenopus mRNA related to Drosophila twist is expressed in response to induction in the mesoderm and the neural crest
.
Cell
59
,
893
903
.
Horstadlus
,
S.
(
1973
).
Experimental Embryology of Echinoderms. Oxford University Press
.
Hursh
,
D. A.
,
Andrews
,
M. E.
and
Raff
,
R. A.
(
1987
).
A sea urchin gene encodes a polypeptide homologous to epidermal growth factor
.
Science
237
,
1487
1490
.
Ip
,
Y. T.
,
Kraut
,
R.
,
Levine
,
M.
and
Rushlow
,
C. A.
(
1991
).
The dorsal morphogen is a sequence specific DNA-binding protein that interacts with a long-range repression element in Drosophila
.
Cell
,
64
,
439
446
.
Irish
,
V. F.
and
Gelbart
,
W. M.
(
1987
).
The decapentaplegic gene is required for dorsal-ventral patterning of the Drosophila embryo
.
Genes Dev
.
1
,
868
879
.
Jaye
,
M.
,
De La Salle
,
H.
,
Schamber
,
F.
,
Balland
,
A.
,
Kohli
,
V.
,
Findeli
,
A.
,
Tolstoshev
,
P.
and
Lecocq
,
J. P.
(
1983
).
Isolation of a human anti-haemophilic factor IX cDNA clone using a 52-base synthetic oligonucleotide probe deduced from the amino acid sequence of bovine factor IX
.
Nucleic Acids Res
.
11
,
2325
2335
.
Jongeneel
,
C. V.
,
Bouvier
,
J.
and
Baairoch
,
A.
(
1989
).
A unique signature identifies a family of zinc-dependent metallopeptidases
.
FEBS Lett
.
242
,
211
214
.
Jürgens
,
G.
,
Wieschauss
,
E.
,
Nüsslein-volhard
,
C.
and
Kludlng
,
H.
(
1984
).
Mutations affecting the pattern of the larval cuticle in Drosophila melanogaster
.
Roux’ Arch. Dev. Biol
.
193
,
283
295
.
Kieran
,
M.
,
Blank
,
V.
,
Logeât
,
F.
,
Vanderkerckhove
,
J.
,
Lottspeicb
,
F.
,
Le Bail
,
O.
,
Urban
,
M.
,
Kourilsky
,
P.
,
Baeurle
,
P.
and
Israel
,
A.
(
1990
).
The DNA binding subunit of NF-xB is identical to factor KBF1 and homologous to the rel oncogene product
.
Cell
62
,
1007
1018
.
Komoriya
,
A.
,
Hortsch
,
M.
,
Meyers
,
C.
,
Smith
,
M.
,
Kanety
,
H.
and
Schlessinger
,
J.
(
1984
).
Biologically active synthetic of epidermal growth factor: localization of a major receptor binding region
.
Proc, natn Acad. Sci. USA
.
81
,
1351
1355
.
Lepage
,
T.
and
Gacbe
,
C.
(
1989
).
Purification and characterization of the sea urchin embryo hatching enzyme
.
J. biol. Chem
.
264
,
4787
4793
.
Lepage
,
T.
and
Gaché
,
C.
(
1990
).
Early expression of a collagenase-like hatching enzyme gene in the sea urchin embryo
.
EMBO J
.
9
,
3003
3012
.
Leytus
,
S. P.
,
Kurachi
,
K.
,
Sakariassen
,
K. S.
and
Davie
,
E. W.
(
1986
).
Nucleotide sequence of the cDNA coding for human complement Clr
.
Biochemistry
25
,
4855
4863
.
Mackinnon
,
C.
,
Carter
,
P. E.
,
Smyth
,
J.
,
Dunbar
,
B.
and
Fothergill
,
J. E.
(
1987
).
Molecular cloning of cDNA for human complement component Cis
.
Eur. J. Biochem
.
169
,
547
553
.
Manlatis
,
T.
,
Fritsch
,
E. F.
and
Sambrook
,
J.
(
1982
).
Molecular Cloning: a Laboratory Manual. Cold Spring Harbor Laboratory
.
Maruyama
,
Y. K.
,
Nakaseko
,
Y.
and Y
Agi
,
S.
(
1985
).
Localization of cytoplasmic determinants responsible for primary mesenchyme formation and gastrulation in the unfertilized egg of the sea urchin Hemicentrotus pulcherrimus
.
J. exp. Zool
.
236
,
155
163
.
Mierendorf
,
R. C.
and
Pfeffer
,
D.
(
1987
).
Direct sequencing of denatured plasmid DNA
.
Methods Enzymol
.
152
,
556
562
.
Morris
,
G. F.
and
Marzluff
,
W. F.
(
1983
).
A factor in sea urchin eggs inhibits transcription in isolated nuclei by sea urchin RNA polymerase III
.
Biochemistry
22
,
645
653
.
Neurath
,
H.
(
1989
).
Proteolytic processing and physiological regulation
.
TIBS
14
,
268
271
.
Nüsslein-volhard
,
C.
,
Frohnhôfer
,
H. G.
and
Lehmann
,
R.
(
1987
).
Determinants of anteroposterior polarity in Drosophila
.
Science
238
,
1675
1681
.
Padgett
,
R. W.
,
St Johnston
,
R. D.
and
Gelbart
,
W. M.
(
1987
).
A transcript from a Drosophila pattern gene predicts a protein homologous to the TGF-μ family
.
Nature
325
,
81
84
.
Patthy
,
L.
(
1985
).
Evolution of the proteases of the blood coagulation and fibrinolysis by assembly from modules
.
Cell
41
,
657
663
.
Patty
,
L.
,
Trexler
,
M.
,
Vali
,
Z.
,
Banyai
,
L.
and Varad!, A
. (
1984
).
Kringles: modules specialized for protein binding
.
FEBS Lett
.
171
,
131
136
.
Reddi
,
A. H
, and
Huggins
,
C. B.
(
1972
).
Biochemical sequences in the transformation of normal fibroblasts in adolescent rats
.
Proc, natn Acad. Sci. USA
.
69
,
1601
1605
.
Rees
,
D. J. G.
,
Jones
,
I. M.
,
Handford
,
P. A.
,
Walter
,
S. J.
,
Esnouf
,
M. P.
,
Smith
,
K. J.
and
Brownlee
,
G. G.
(
1988
).
The role of j3-hydroxyaspartate and adjacent carboxylate residues in the first EGF domain of human factor IX
.
EMBO J
.
1
,
2053
2061
.
Riccio
,
A.
,
Grimaldi
,
G.
,
Verde
,
P.
,
Sebastio
,
G.
,
Boast
,
S.
and
Blasi
,
F.
(
1985
).
The human urokinase-plasminogen activator and its promoter
.
Nucleic Acids Res
.
13
,
2759
2771
.
Roth
,
S.
,
Stein
,
D.
and Nüsslein-Volhard
(
1989
).
A gradient of nuclear localization of the dorsal protein determines dorsoventral pattern in the Drosophila embryo
.
Cell
59
,
1189
1202
.
Rushlow
,
C.
,
Frasch
,
M.
,
Doyle
,
H.
and
Levine
,
M.
(
1987
).
Maternal regulation of zerkniillt: a homeobox gene controlling differentiation of dorsal tissues in Drosophila
.
Nature
330
,
583
587
.
Rushlow
,
C. A.
,
Han
,
K.
,
Manley
,
J.
and
Levine
,
M.
(
1989
).
The graded distribution of the dorsal morphogen is initiated by selective nuclear transport in Drosophila
.
Cell
59
,
1165
1177
.
Sanchez-Lopez
,
R.
,
Nicholson
,
R.
,
Gesnel
,
M. C.
,
Matrisian
,
L. M.
and
Breathnach
,
R.
(
1988
).
Structure-function relationships in the collagenase family member transin
.
J. biol. Chem
.
263
,
11892
11899
.
Sanger
,
F.
,
Nicklen
,
S.
and
Coulson
,
A. R.
(
1977
).
DNA sequencing with chain-terminating inhibitors
.
Proc. natn. Acad. Sci. USA
74
,
5463
5467
.
Sargent
,
M. G.
and
Bennett
,
M. F.
(
1990
).
Identification in Xenopus of a structural homologue of the Drosophila gene snad
.
Development
109
,
967
973
.
Shimell
,
M. J.
,
Ferguson
,
E. L.
,
Childs
,
S. R.
and
O’connor
,
M. B.
(
1991
).
The Drosophila dorsal-ventral patterning gene tolloid is homologous to human Bone Morphogenetic Protein-1
.
Cell (in press)
.
Showman
,
R. M.
and
Foerder
,
C. A.
(
1979
).
Removal of the fertilization membrane of sea urchin embryos employing aminotriazole
.
Expl. Cell Res
.
120
,
253
255
.
Spinelli
,
G.
,
Gianguzza
,
F.
,
Casano
,
C.
,
Ademo
,
P.
and
Burckhardt
,
J.
(
1979
).
Evidence of two different sets of histones genes active during embryogenesis of the sea urchin Paracentrotus lividus
.
Nucleic Acids Res
.
6
,
545
560
.
Stanley
,
K.
,
Kocher
,
H. P.
,
Luzio
,
J. P.
,
Jackson
,
P.
,
Tschopp
,
J.
and
Dickson
,
J.
(
1985
).
The sequence and topology of human complement component C9
.
EMBO J
.
4
,
375
382
.
Stanley
,
K. K.
and
Luzio
,
J. P.
(
1984
).
Construction of a new family of high efficiency bacterial expression vectors: identification of cDNA clones coding for human liver proteins
.
EMBO J
.
3
,
1429
.
Steward
,
R.
(
1987
).
Dorsal, an embryonic polarity gene in Drosophila, is homologous to the vertebrate proto-oncogene, c-rel
.
Science
238
,
692
694
.
Steward
,
R.
(
1989
).
Relocalization of the dorsal protein from the cytoplasm to the nucleus correlates with its function
.
Cell 59,1179-1188
.
St Johnston
,
R. D.
and
Gelbart
,
W. M.
(
1987
).
Decapentaplegic transcripts are localized along the dorsal-ventral axis of the Drosophila embryo
.
EMBO J
.
6
,
2785
2791
.
Stacker
,
W.
,
Woltz
,
R. L.
,
Zwilllng
,
R.
,
Strydom
,
D. J.
and
Auld
,
D. S.
(
1988
).
Astacus protease, a zinc metalloenzyme
.
Biochemistry
27
,
5026
5032
.
Tautz
,
D.
and
Pfeifle
,
C.
(
1989
).
A non-radioactive in situ hybridization method for the localization of specific RNAs in Drosophila embryos reveals translational control of the segmentation gene hunchback
.
Chromosoma
98
,
81
85
.
Thisse
,
B.
,
Stoetzel
,
C.
,
Messal
,
M. E.
and
Perrin-Schmitt
,
F.
(
1987
).
Genes of the Drosophila maternal dorsal group control the specific expression of the zygotic gene twist in the presumptive mesodermal cells
.
Genes Dev
.
1
,
709
715
.
Thisse
,
C.
,
Perrin-Schmitt
,
F.
,
Stoetzel
,
C.
and
Thisse
,
B.
(
1991
).
Sequence specific transactivation of the Drosophila twist gene by the dorsal gene product
.
Cell
65
,
1191
1201
.
Titanl
,
K.
,
Torff
,
H. J.
,
Hormel
,
S.
,
Kumar
,
S.
,
Walsh
,
K.
,
Roedel
,
J.
,
Neurath
,
H.
and
Zwilling
,
R.
(
1987
).
Aminoacid sequence of a unique protease from the crayfish Astacus fluviatihs
.
Biochemistry
26
,
22
226
.
Tosí
,
M.
,
Duponchel
,
C.
,
Meo
,
T.
and
Juher
,
C.
(
1987
).
Complete cDNA sequence of human complement Cis and close physical linkage of the homologous genes Cis and Clr
.
Biochemistry
26
,
8516
8524
.
Urist
,
M. R.
(
1965
).
Bone: formation by autoinduction
.
Science
150
,
893
899
.
Vaessin
,
H.
,
Bremer
,
K. A.
,
Knust
,
E.
and
Campos-Ortega
,
J. A.
(
1987
).
The neurogenic gene Delta of Drosophila melanogaster is expressed in neurogenic territories and encodes a putative transmenbrane protein with EGF-like repeats
.
EMBO J
.
6
,
3431
3440
.
Van Wart
,
H. E.
and
Birdekal-Hansen
,
H.
(
1990
).
The cysteine switch: a principle of regulation of metalloproteinase activity with potential applicability to the entire matrix metalloproteinase gene family
.
Proc. natn. Acad. Sci. USA
87
,
5578
5582
.
Von Hejne
,
G.
(
1986
).
A new method for predicting signal cleavage sites
.
Nucleic Acids Res
.
14
,
4683
4690
.
Wang
,
E. A.
,
Rosen
,
V
,,
Cordes
,
P.
,
Hewick
,
R. M.
,
Kritz
,
M. J.
,
Luxenberg
,
D. P.
,
Sibley
,
B. S
, and
Wozney
,
J. M.
(
1988
).
Purification and characterization of other distinct bone-inducing factors
.
Proc. natn. Acad. Sci. USA
.
85
,
9484
9488
.
Wang
,
E. A.
,
Rosen
,
V.
,
D’alessandro
,
J. S.
,
Bauduy
,
M.
,
Cordes
,
P.
,
Harada
,
T.
,
Israel
,
D.
,
Hewick
,
R. M.
,
Kerns
,
K.
,
Lapan
,
P.
,
Luxenberg
,
D. P.
,
Mcquaid
,
D.
,
Moutsatsos
,
I.
,
Nove
,
J.
,
Sibley
,
B. S.
and
Wozney
,
J. M.
(
1990
).
Recombinant bone morphogenetic protein induces bone formation
.
Proc. natn. Acad. Sci. USA
.
87
,
2220
2224
.
Wharton
,
K. A.
,
Johansen
,
K. M.
,
Xu
,
T.
and
Artavanis-Tsakonas
,
S.
(
1985
).
Nucleotide sequence from the neurogenic locus Notch implies a gene product that shares homology with proteins containing EGF-like repeats
.
Cell
43
,
567
581
.
Whitham
,
S. E.
,
Murphy
,
G.
,
Angel
,
P.
,
Rahmsdorf
,
H. J.
,
Smith
,
B. J.
,
Lyons
,
A.
,
Harris
,
T. J. R.
,
Reynolds
,
J. J.
,
Herrlich
,
P.
and
Docherty
,
A. J. P.
(
1986
).
Comparison of human stromelysin and collagenase by cloning and sequence analysis
.
Biochem. J. 240,913-916
.
Wilt
,
F. H.
(
1987
).
Determination and morphogenesis in the sea urchin embryo
.
Development
100
,
559
575
.
Wozney
,
J. M.
(
1989
).
Bone morphogenetic proteins
.
Progress in Growth Factor Res
.
1
,
267
280
.
Wozney
,
J. M.
,
Rosen
,
V.
,
Celeste
,
A. J.
,
Mitsock
,
L. M.
,
Whitters
,
M. J.
,
Kritz
,
R. W.
,
Hewick
,
R. M.
and
Wang
,
E. A.
(
1988
).
Novel regulators of bone formation: molecular clones and activities
.
Science
242
,
1528
1534
.
Yamamoto
,
T.
,
Davis
,
G.
,
Brown
,
M.
,
Schneider
,
W. J.
,
Casey
,
M. L.
,
Goldstein
,
J. L.
and
Russel
,
D. W.
(
1984
).
The human LDL receptor: A cysteine rich protein with multiple Aiu sequences in its mRNA
.
Cell
39
,
27
38
.