Develop. Growth Differ. (2003) 45, 85–93
Novel pattern of AtXlox gene expression in starfish Archaster typicus embryos Sheng-Ping L. Hwang, Jen-Yi Wu, Chaolun A. Chen, Cho-Fat Hui and Chang-Po Chen* Institute of Zoology, Academia Sinica, Nankang, Taipei 11529, Taiwan.
An Xlox homologue gene (AtXlox) was identified in the starfish Archaster typicus. The gene consists of two exons, and encodes a polypeptide containing 228 amino acids. Although AtXlox shared 54.6 and 50.3% global amino acid sequence similarity with sea urchin SpXlox and Xenopus XlHhox8, respectively, the homeodomain of AtXlox was highly conserved. Amino acid sequence identity as high as 85 to 100% was identified between the AtXlox homeodomain and its homologues from various vertebrate and invertebrate organisms. In addition, a conserved histidine residue located at position 44 of the homeodomain of all known Xlox homologues was also identified. Results of a phylogenetic analysis based on the 60 amino acid sequence of the homeodomain indicated that AtXlox was closely related to sea urchin SpXlox. Temporal developmental mRNA expression pattern analyzed by reverse transcription (RT)–polymerase chain reaction (PCR) showed that AtXlox mRNA was mainly expressed in the early gastrula stage embryos. Whole-mount in situ hybridization revealed a ubiquitous mRNA expression pattern in archenterons as well as in ectodermal cells near the vegetal region of early and mid-gastrula stage embryos. This spatial expression pattern is very different from those of Xlox homologues in the leech, amphioxus, and in various vertebrate organisms with spatially restricted mRNA expression patterns in endodermal cells. Key words: archenteron, homeodomain, spatial distribution, starfish, Xlox.
Introduction Transcription factors such as homeobox genes have been shown to modulate developmental processes in a variety of organisms ranging from yeast to mammals (Manak & Scott 1994). All homeobox proteins possess a homeodomain that contains a helix-turn-helix DNA binding motif located in the C-terminal region (Gehring 1994). Among homeobox genes, Hox genes have been shown to confer regional identity to the antero-posterior axis of developing vertebrate and invertebrate embryos (Krumlauf 1994; Gellon & McGinnis 1998), while other classes of homeobox genes contain evolutionarily conserved functions in the development of organogenesis (Manak & Scott 1994). For example, Drosophila engrailed (Patel et al. 1989) as well as mouse EN-1 and EN-2 (Joyner & Martin 1987) have been implicated in neural development. Xlox family homeobox genes have been suggested to be involved in endodermal differentiation because of their unique spatial distribution pattern
*Author to whom all correspondence should be addressed. Email:
[email protected] Received 3 September 2002; revised 8 October 2002; accepted 26 November 2002.
(Burglin 1994). In vertebrates, Xenopus XlHbox8 is expressed in the nucleus of endodermal cells of the duodenum and pancreas in embryos, and expression within the pancreatic epithelium continues into the adult stage (Wright et al. 1988). In invertebrate organisms such as amphioxus, AmphiXlox is expressed in the presumptive gut region from the neurula stage through to the larval stage (Brooke et al. 1998). The development of echinoderms such as sea urchins and starfish begins with bilateral embryos containing archenterons in the blastocoele cavity. After metamorphosis, an adult starfish with radial symmetry develops from the left coelomic pouch that is derived from archenterons (Wray 1997). Thus the development of archenterons in embryos is not only necessary for the ingestion of food during the larval stage but also essential for the differentiation of the adult rudiment. Because Xlox homeobox genes have been implicated in the function of endodermal differentiation in both vertebrate and invertebrate organisms, we speculate that Xlox homologues in starfish embryos may play important roles in their development. In this report, we have cloned the Xlox homologue gene (named AtXlox) from the starfish Archaster typicus, and characterized its temporal and spatial mRNA expression pattern during embryonic development.
86
S-P. L. Hwang et al.
Materials and Methods Materials and embryo culture Starfish (Archaster typicus) were collected from the Penghu Islands in the Taiwan Strait. All chemicals were purchased from Merck (Frankfurt, Germany) unless otherwise specified. Embryonic cultures of Archaster typicus were performed according to Chen & Run (1991).
DNA isolation and PCR Testes were used for DNA extraction following the procedures described in Sambrook et al. (1989). A pair of conserved degenerate primers located in the homeodomain designed by Wedeen et al. (1990) was used in the polymerase chain reaction (PCR). The PCR was conducted in a 50 µL reaction mixture containing DNA (0.5 µg), a forward degenerate primer (5-GAA TTC GAA/G C/TTC GAA/G AAA/G GAA/G TTC/T CAT-3; 100 pmol), a reverse degenerate primer (5-TTA/G TTC/T TGA/G AAC CAG ATC/T TTA/G/T AT-3; 100 pmol), dNTP (0.2 mM), Taq buffer (1), and super Taq polymerase (1.5 U) in a thermal cycler (Biometra Trio-Thermoblock; Biometra, Goettingen, Germany). PCR cycles were 94C for 2 min for 1 cycle; 94C for 1 min, 40C for 1 min and 72C for 1 min for 36 cycles; and 72C for 7 min for 1 cycle.
Construction and screening of a starfish genomic DNA library Genomic DNA was partially digested with Sau3AI, sized to 9–23 kb on a sucrose gradient and ligated into the BamHI site of the lambda DASHII vector (Stratagene, La Jolla, CA, USA). The genomic DNA library was screened using a 119 bp digoxygenin (DIG)-labeled PCR product according to Sambrook et al. (1989). One positive plaque was obtained, and lambda phage DNA was isolated as described in Donovan et al. (1993). After BamHI digestion, an approximately 5 kb DNA fragment was ligated into a pBluescript II vector digested with the same restriction enzyme.
DNA sequencing and sequence analysis DNA sequencing was performed on an ABI Prism 377 automatic DNA sequencer (Applied Biosystems, Foster City, CA, USA). The nucleotide sequence has been deposited in GenBank with the accession no.: AF439973. GENESCAN (http://genes.mit.edu/genescan.html) was used for exon prediction. The GAP program from the Wisconsin Sequence Analysis
Package of the Genetics Computer Group (GCG) was used for various sequence alignments. Phylogenetic relationships were constructed based on 60 amino acid sequences of homeodomains of Xlox homologues from various vertebrate and invertebrate organisms. Phylogenetic tree analyses were performed using PHYLIP 3.6 (Felsenstein 2000). NEIGHBOR option after genetic distances were calculated based on the Dayhoff PAM model (Dayhoff et al. 1978) in the PROTDIST option. The robustness of the NJ phylogenies was assessed by 1000 bootstrap replicates using the SEQBOOT and CONSENSE options.
Preparation of digoxygenin-labeled probes We used two DIG-labeled probes of different sizes for various experiments. A 119 bp DIG-labeled DNA probe corresponding to amino acid numbers 99–138 was used for genomic DNA library screening and for the construction of a temporal mRNA expression profile using reverse transcription (RT)–PCR and Southern blot analysis. A 385 bp DIG-labeled DNA probe corresponding to amino acid numbers 78–206 was used in the Southern blot analysis of RT–PCR for the confirmation of exon location and the length of coding region predicted. These two DNA probes were generated by PCR using dNTP containing DIG-11-dUTP following the manufacturer’s protocol (Roche, Mannheim, Germany). The plasmid containing the 385 bp DNA, a segment that encodes the complete homeodomain plus 60 amino acids 3 of the homeodomain, was also used to synthesize both antisense and sense RNA probes for wholemount in situ hybridization. RNA probes were synthesized using T7 or T3 RNA polymerase according to the protocols provided by the manufacturer (Roche).
RNA isolation, RT–PCR and Southern blot analysis Total RNA from different developmental stages of starfish embryos was isolated using TRI REAGENT (Molecular Research Center, Cincinnati, OH, USA). To confirm exon location and the length of the coding region predicted, poly(A)+ mRNA of the gastrula stage was isolated using PolyATract mRNA Isolation Systems (Promega, Madison, WI, USA). The RT reaction was conducted in a 30 µL reaction mixture containing Poly(A)+ mRNA (approximately 1 µg), random hexamers (1 µg), MgCl2 (5 mM), an RNase inhibitor (40 U), dNTP (1 mM) and Improm-II reverse transcriptase (Promega) at 25C for 10 min, 55C for
Spatial distribution of starfish AtXlox
90 min and 70C for 15 min, with a pause at 4C. The PCR was performed in a 50 µL reaction mixture containing cDNA (5 µL), a forward primer (5-AGA TTG ATT GAG CGA CAC TGG-3; 20 pmol), a reverse primer (5-ACT TGA CGA TGA GAT GTT CTG-3; 20 pmol), dNTP (0.2 mM), MgCl2 (2 mM), Ex Taq buffer (1), and Ex Taq polymerase (Takara, Shiga, Japan) in a thermal cycler (Biometra Trio-Thermoblock). PCR cycles were 94C for 2 min for 1 cycle; 94C for 1 min, 56C for 1 min and 72C for 1 min for 35 cycles; and 72C for 10 min. RT–PCR products were fractionated on a 2% 0.5 Tris borate ethylenediaminetetraacetic acid (EDTA) (TBE) agarose gel, and DNA was transferred onto a nylon membrane following standard methods (Sambrook et al. 1989). After 2 h of prehybridization, the membrane was hybridized in prehybridization buffer (5 sodium citrate chloride (SSC), 50% formamide, 2% blocking solution, 0.1% sodium lauryl sarcosine and 0.02% sodium dodecylsulfate (SDS)) containing 25 ng/mL of the 385 bp DIGlabeled DNA probe at 42C overnight. After hybridization, the membrane was washed with 2 SSC and 0.1% SDS for 5 min twice at room temperature, followed by two 0.1 SSC and 0.1% SDS washes for 15 min at 55C. CDPStar chemiluminescent detection was carried out following instructions from the manufacturer (Roche). The developmental profile of AtXlox mRNA expression was further analyzed by RT–PCR. The RT reaction was conducted in a 10 µL reaction mixture containing total RNA (1 µg) from different developmental stages, random hexamers (2.5 µM), MgCl2 (5 mM), dNTP (1 mM), an RNase inhibitor (40 U), and MMLV reverse transcriptase (Applied Biosystems) at 25C for 10 min, 42C for 60 min and 99C for 5 min. The PCR was conducted in a 50 µL reaction mixture containing cDNA (10 µL), a forward primer (5-AAG TAC ATC TCC CGA CCC-3; 10 pmol), a reverse primer (5-CTC TGT CAG GTT CAG CAT-3; 10 pmol), dNTP (0.2 mM), MgCl2 (1.5 mM), 1 Replitherm reaction buffer, and Replitherm DNA polymerase (Epicentre Technologies, Madison, WI, USA) in a thermal cycler. PCR cycles were 94C for 2 min for 1 cycle; 94C for 1 min, 53C for 1 min and 72C for 45 s for 35 cycles; and 72C for 7 min. In addition, control RT reactions without the addition of reverse transcriptase were conducted for each developmental stage and subsequent PCR reactions were performed at the same conditions as described above. RT–PCR products were fractionated on a 12% 0.5 TBE native polyacrylamide gel, and DNA was transferred onto a nylon membrane. Subsequently, Southern blot hybridization was conducted as described above except that the 119 bp DIGlabeled DNA probe was used.
87
Whole-mount in situ hybridization Embryos from different developmental stages were fixed in 50 mM phosphate buffer (pH 7.5) containing 1% glutaraldehyde and 0.426 M NaCl on ice for a total of 1 h with one change of fixative. Subsequent washing and dehydration were performed according to Angerer et al. (1987). Whole-mount in situ hybridization was conducted based on procedures described in Shih et al. (2002) with some modifications. Basically, embryos were rehydrated and treated with proteinase K (8 µg/mL) for 15 min at room temperature. PBST (0.1% Tween-20 in phosphatebuffered saline (PBS)) washes were conducted to stop the proteinase K reaction, and embryos were refixed with 4% paraformaldehyde in PBS for 30 min. Following two washes with PBST, embryos were treated with acetic anhydride (2.5 µL in 1 mL 0.1 M triethanolamine) for 1 h to remove endogenous phosphatase. Embryos were gradually introduced to pH 5.0 hybridization buffer containing 50% formamide, 5 SSC, 500 µg/mL yeast tRNA, 50 µg/mL heparin and 0.1% Tween-20 after two 10 min PBST washes. Embryos were prehybridized for 2 h at 53C and hybridized overnight with 100 ng of the 385 bp DIG-labeled RNA probe. Excess probe was removed by two 30 min washes with an equal mixture of 50% formamide, 5 SSC and 50% 2 SSC, and 0.1% Tween-20; one 15 min wash with 2 SSC and 0.1% Tween-20; and two 30 min washes with 0.2 SSC and 0.1% Tween-20 at 53C. After three 15 min washes with PBST, embryos were blocked in PBST buffer containing 5% sheep serum and 2 mg/mL bovine serum albumin (BSA) for 2 h before incubating with 1:1000 preabsorbed alkaline phosphatase-conjugated anti-DIG antibody for 1 h at room temperature. Excess antibody was removed by eight 15 min washes with blocking solution and the embryos were stored in PBST at 4C overnight. Following one PBST wash and two rinses with pH 9.5 solution containing 0.1 M Tris-HCl, 50 mM MgCl2, 100 mM NaCl, 0.1% Tween-20, and 3 mM levamisol, embryos were stained with the same pH 9.5 solution containing 3.4 µL/mL nitro blue tetrazolium (NBT; 100 mg/mL stock) and 3.5 µL/mL 5-bromo-4-chloro-3-indolyl phosphate (BCIP; 50 mg/mL stock) for 3–4 h at room temperature. The staining reaction was terminated by several PBST washes. Embryos were refixed with 4% paraformaldehyde for 20 min followed by several PBST washes, and then they were mounted using 50% glycerol. Embryos were analyzed using an Olympus BX60 microscope (Olympus, Tokyo, Japan) equipped with Nomarski lenses. The images were recorded
88
S-P. L. Hwang et al.
using a SPOT digital camera (Diagnostic, Sterling Heights, MI, USA).
Results Using genomic DNA and a pair of conserved degenerate primers, a 119 bp AtXlox gene fragment corresponding to amino acid numbers 99–138 was obtained by PCR. Subsequently, we obtained the full-length
AtXlox gene by screening a lambda phage starfish genomic DNA library using the 119 bp PCR product as probe. The AtXlox gene contains two exons and one intron, as predicted by the GENESCAN program. It encodes a 228 amino acid long open reading frame. The 60 amino acid homeodomain is located in the second exon. In addition, a conserved histidine residue located at position 44 of the homeodomain that is present in all Xlox homeodomain proteins was
Fig. 1. Nucleic acid and deduced amino acid sequences of the AtXlox gene. The exon nucleotide sequence is shown in uppercase and the intron nucleotide sequence is shown in lowercase. The amino acid sequence of the homeodomain is underlined. The unique histidine residue of the Xlox family of homeobox proteins is located at position 44 of the homeodomain (shown with a bold italicized letter). Locations of forward and reverse degenerate primers used in the initial polymerase chain reaction (PCR) are shaded. One pair of specific forward and reverse primers used in the reverse transcription (RT)–PCR to analyze developmental mRNA expression are boxed, while another pair of specific forward and reverse primers used in the RT–PCR to confirm exon location and the length of the predicted coding region are shown by double underlining.
Spatial distribution of starfish AtXlox
also identified (Fig. 1). To confirm exon location and the length of the coding region predicted by the GENESCAN program, RT–PCR using a primer pair located at the beginning of exon 1 (amino acid numbers 4–10) and the end of exon 2 (amino acid numbers 220–226) was conducted. Although no prominent amplified DNA fragment in the expected size range could be viewed in the agarose gel, an approximately 680 bp DNA band was detected by Southern blot analysis using the 385 bp DNA probe (Fig. 2). A second DNA band with lower molecular weight on the Southern blot was probably a RNA degradation product. Full length amino acid sequence comparison showed that AtXlox shares 54.6 and 50.3% sequence similarity with sea urchin SpXlox and Xenopus XlHbox8, respectively (Table 1). However, high amino acid sequence identity ranging from 85 to 100% was shared between the 60 amino acid homeodomain of AtXlox and those of Xlox homologues from various vertebrate and invertebrate organisms (Fig. 3). Together, these data indicate that AtXlox is an Xlox homologue gene in starfish. To further understand the relationship between AtXlox and other Xlox homologues, phylogenetic analyses were performed. The full length coding region of Xlox homologues has been reported in only a few species, including human, mouse, rat, Xenopus, zebrafish and sea urchin. In contrast, partial amino acid sequences of the Xlox homeodomain region were described for various invertebrate organisms (Fig. 3). Thus, a phylogenetic tree was constructed based on an alignment of 60 amino acids of the Xlox homeodomain from various vertebrate and invertebrate organisms (Fig. 4). The Neighbor-Joining (NJ) tree showed that the Xlox homeodomains of all vertebrates (subphylum Vertebrata) were clustered as a monophyletic group. In contrast, Xlox homeodomains of the amphioxus, ascidian (subphylum Cephalochordata), and other invertebrates were divided into three major groups. The starfish AtXlox formed a clade with sea urchin SpXlox, and clustered with a group
89
composed of a starfish (AmXlox), two sea urchin homologues (PjXlox and HpXlox), an amphioxus AmphiXlox, and a polychaete CtsXlox. The second group contained the ophiuroid SsXlox, crinoid OjXlox and polychaete CHv-Hb. The third group formed a paraphyletic relationship to the rest of the invertebrates, and was composed of the sipunculan PsXlox, ascidian CionaXlox, and HtrA2 and Lox3c from leeches (Fig. 4). A temporal developmental mRNA expression pattern was examined by RT–PCR using a primer
Fig. 2. Confirmation of exon location and the predicted length of the AtXlox coding region by reverse transcription (RT)– polymerase chain reaction (PCR) and Southern blot analyses. Poly(A)+ mRNA isolated from the gastrula stage was used in RT– PCR. The products were fractionated on a 2% 0.5 TBE agarose gel and viewed by ethidium bromide staining (A). Subsequently, DNA was transferred onto a nylon membrane and hybridized with the 385 bp DNA probe (B). The arrow indicates the expected 680 bp DNA band. Lane 1 shows the PCR reaction from RT mixture containing poly(A)+ mRNA, while lane 2 shows PCR reaction from RT mixture containing dH2O instead of poly(A)+ mRNA as a control. Molecular weight markers (M) are shown in bases.
Table 1. Amino acid sequence comparison among AtXlox and vertebrate Xlox homologues†
Human PDX-1 Mouse IPX-1 Rat IDX-1 Xenopus XlHhox8 Zebrafish PDX-1 Starfish AtXlox Sea urchin SpXlox
Human PDX-1
Mouse IPX-1
Rat IDX-1
Xenopus XlHhox8
Zebrafish PDX-1
Starfish AtXlox
Sea urchin SpXlox
100
92.6 100.0
93.6 93.6 100.0
66.5 69.2 66.2 100.0
63.3 63.8 63.1 63.9 100.0
43.0 44.0 46.9 50.3 48.2 100.0
51.1 50.5 53.6 55.2 59.2 54.6 100.0
† The GAP program from the Wisconsin Sequence Analysis Package of the Genetics Computer Group (GCG) was used for the comparison. The percentage similarity is shown.
90
S-P. L. Hwang et al.
Fig. 3. Amino acid sequence alignment of the homeodomain of AtXlox and Xlox homologues from various vertebrate and invertebrate organisms. Identical amino acid sequences are indicated by dots. Sequences are human PDX-1 (Peshavaria et al. 1997), rat IDX-1 (Miller et al. 1994), mouse IPF-1 (Ohlsson et al. 1993), Xenopus XlHbox8 (Wright et al. 1988), zebrafish PDX-1 (Milewski et al. 1998), amphioxus AmphiXlox (Brooke et al. 1998), ascidian (Ciona intestinalis) CiXlox (Ferrier & Holland 2002), sea urchin (Strongylocentrotus purpuratus) SpXlox (AF541970), sea urchin (Holopneustes purpurescens) HpXlox (Morris et al. 1997), sea urchin (Peronella japonica) PjXlox (Hano et al. 2001), starfish (Asterina minor) AmXlox (Mito & Endo 1997), crinoid (Oxycomanthus japonicus) OjXlox (Mito & Endo 2000), ophiuroid (Stegophiura sladeni) SsXlox (Mito & Endo 2000); polychaete (Ctenodrilus serratus) CTs-Xlox (Dick & Buss 1994), polychaete (Chaetopterus variopedatus) CHv-Hb7 (Irvine et al. 1997), leech (Helobdella triserialis) Htr-A2 (Wedeen et al. 1990), leech (Hirudo medicinalis) Lox3C (Wysocka-Diller et al. 1995) and sipunculan (Phascolion stromus) PsXlox (Ferrier & Holland 2001). The percent identity of amino acids between the homeodomain of AtXlox and those of other Xlox homologues is shown. The GAP program from the Wisconsin Sequence Analysis Package of the Genetics Computer Group (GCG) was used for the comparison.
Fig. 4. Phylogenetic analysis of AtXlox and Xlox homologues from various vertebrate and invertebrate organisms. The phylogenetic tree was constructed using a 60 amino acid sequence of the homeodomain of AtXlox as well as those of 18 Xlox homologues from various vertebrate and invertebrate organisms. Both Drosophila Zen1 and Zen 2 (Rushlow et al. 1987) were used as outgroups for the phylogenetic tree construction. The bar represents a distance of 10.
pair located within the homeodomain (Figs 1,5). We observed the expected 58 bp RT–PCR products in both the 22 h postfertilization (h.p.f.) early gastrula and the 36 h.p.f. late gastrula stages (Fig. 5A; lanes 6,7). The identity of the RT–PCR products was confirmed by Southern blot hybridization (Fig. 5B; lanes 6,7). A prominent DNA band was detected in the early gastrula stage and a faint DNA band was detected in the late gastrula stage. In the control experiment, no DNA band was observed on both the agarose gel and the Southern blot when reverse transcriptase was omitted in RT reactions (data not shown). Thus, AtXlox was mainly expressed in the early gastrula stage. Furthermore, the spatial distribution pattern of AtXlox was analyzed by whole-mount in situ hybridization. Transcripts of Atxlox were localized in the archenterons and ectodermal cells near the vegetal region of early gastrula embryos (Fig. 6A). A slightly decreased mRNA level was observed during extension of the archenterons and migration of mesenchyme cells (Fig. 6C). However, no expression of AtXlox could be observed during development of coelomic pouches in the late gastrula stage (Fig. 6E). No staining could be observed in control embryos treated with sense RNA probes (Fig. 6B,D,F).
Spatial distribution of starfish AtXlox
Discussion We obtained the AtXlox gene from the starfish Archaster typicus. The gene encodes a polypeptide containing 228 amino acids. Based on the full length amino acid sequence, AtXlox shared the highest similarity (54.6%) with sea urchin SpXlox. High (85–100%) amino acid sequence identity was observed between the homeodomain of AtXlox and those from various vertebrate and invertebrate Xlox homologues. In addition, the conserved histidine residue at amino acid 44
Fig. 5. Expression of AtXlox mRNA during starfish embryonic development. Total RNA isolated from different developmental stages was used in reverse transcription (RT)–polymerase chain reaction (PCR) reaction. The amplified RT–PCR products were fractionated on a 12% 0.5 TBE native polyacrylamide gel and viewed by ethidium bromide staining (A). After transferring DNA onto a nylon membrane, the membrane was hybridized with the 119 bp DNA probe (B). The arrow indicates the expected 58 bp DNA band. Different developmental stages included are unfertilized egg (lane 1), fertilized egg (lane 2), 2.5 h postfertilization (h.p.f.) cleavage (lane 3), 7 h.p.f. blastula (lane 4), 14 h.p.f. prehatching (lane 5), 22 h.p.f. early gastrula (lane 6), 36 h.p.f. late gastrula (lane 7), 50 h.p.f. early bipinnaria (lane 8), 72 h.p.f. bipinnaria (lane 9), 4 days postfertilization (d.p.f.) late bipinnaria (lane 9), 5 d.p.f. early brachiolaria (lane 10), 6 d.p.f. brachiolaria (lane 11) and 7 d.p.f. brachiolaria (lane 12). Lane 13 represents PCR reaction from RT mixture containing dH2O instead of total RNA as a control. Molecular weight markers (lane M) are shown in bases.
91
of the homeodomain found in all Xlox homologues was also present. The unique histidine is located in the ‘recognition helix’ that determines DNA-binding specificity (Desplan et al. 1988; Wright et al. 1988). Overall, these results indicate that starfish AtXlox belongs to the Xlox family of homeobox genes. Although amino acid sequences within the homeodomain are identical and phylogenetic analysis indicated a closer relationship between AtXlox and sea urchin SpXlox, this NJ tree should be interpreted with caution. First, either Xlox of the annelids or those of echinoderms did not form a phylum-based monophyly accordingly. Second, most of the bootstrap values on the branches were relatively low, suggesting low statistical support for grouping of taxa.
Fig. 6. Spatial distribution of AtXlox mRNA in early, mid and late gastrula embryos. Both the 385 bp sense and antisense RNA probes were used in whole-mount in situ hybridization experiments. Embryos in (A), (C) and (E) were treated with antisense RNA probes, while embryos in (B), (D) and (F) were treated with sense RNA probes. Transcripts are expressed in the archenterons and ectodermal cells near the vegetal region of early and mid-gastrula embryos (A,C). No dark purple staining can be observed in a late gastrula stage embryo (E). Ar, archenteron; l.c.v., left coelomic vesicle. Bar, 100 µm.
92
S-P. L. Hwang et al.
These results may imply an early divergence of Xlox homeodomain in metazoan evolution; however, this is unlikely to be the case, as other paradox studies (e.g. Mox) showed a concordance to the organismal phylogeny such that AmphiMox diverged immediately outside two branches formed by vertebrate Mox1 and Mox2 (Minguillon & Garcia-Fernandez 2002). In contrast, we argue that these inconsistencies are probably due to the conservation of Xlox homeodomain and incomplete amino acid sequences of Xlox homeodomains in eight of the 19 taxa: they only contain 40% of amino acid sequence within the homeodomain. These drawbacks significantly decreased the information for phylogenetic reconstruction of the Xlox homeodomain. When those taxa with incomplete amino acid sequences were excluded from the phylogenetic analysis, AtXlox was still grouped with SpXlox, and bootstrap values were greatly improved not only locally but also globally (data not shown). The positions of other Xlox in the phylogenetic tree needs to be further confirmed when the complete 60 amino acid homeodomain is available. Our results indicated that AtXlox developmental expression was restricted to the early gastrula stage, but its spatial expression in gastrula embryos was quite broad and ubiquitous. Such an expression pattern vastly differs from those reported for other vertebrate and invertebrate organisms. For instance, mouse IPF-1 expression is restricted to the dorsal and ventral walls of the primitive foregut where the pancreas will later form in the embryo (Ohlsson et al. 1993). In addition, IPF-1 expression was also observed in -cells of the adult mouse pancreas (Ohlsson et al. 1993). In invertebrate organisms, amphioxus AmphiXlox is expressed in a stripe of the archenteron wall, the presumptive gut region, from the neurula stage through to the larval stage (Brooke et al. 1998). In the leech, early Lox3 expression is restricted to the midgut primordium, with mRNA expression being observed from embryonic day (E)11 embryos to adults (Wysocka-Diller et al. 1995). All these results indicate a restricted spatial expression pattern in endoderms and a broad temporal expression period from embryos to adult stages. Recently, ParaHox genes were found in amphioxus and a sipunculan (Brooke et al. 1998; Ferrier & Holland 2001). Three genes, Gsx, Xlox and Cdx, are included in the ParaHox genes. In amphioxus, AmphiGsx expression was detected in the anterior cerebral vesicle, while AmphiCdx was expressed posteriorly (Brooke et al. 1998). It was suggested that ParaHox genes are an ancient paralogue of the Hox gene cluster and arose by duplication of ProtoHox
genes. The ParaHox genes have also been suggested to exist in mammals (Pollard & Holland 2000). However, not all members of the ParaHox genes exist in the two protostome invertebrates Drosophila and Caenorhabditis. Only the Gsx homologue (intermediate neuroblasts defective) and Cdx homologue (caudal) have been identified in Drosophila (Weiss et al. 1998; Macdonald & Struhl 1986), while the Cdx homologue (pal-1), but not Gsx or Xlox, exists in Caenorhabditis elegans (Ruvkun & Hobert 1998). Similarly, we speculate that not all members of the ParaHox genes exist in starfish. In this case, AtXlox may be forced to conduct multiple functions, leading to its ubiquitous expression in the archenteron. Additional expression of AtXlox in ectodermal cells near the vegetal region possibly reflects its multiple roles during starfish embryonic development.
Acknowledgements We would like to thank Ms C. C. Lin and Ms D. N. Lin for their help with DNA sequencing using an ABI Prism 377 automatic DNA sequencer (Applied Biosystems). Thanks also go to Dr David E. K. Ferrier and Dr Peter Holland for providing us with references of some echinoderm Xlox sequences. This work was supported by grants issued by the National Science Council of the Republic of China (Grant no.s NSC 88-2311-B-001-075 and NSC 892611-B-001-002) to Chang-Po Chen.
References Angerer, L. M., Cox, K. H. & Angerer, R. C. 1987. Demonstration of tissue-specific gene expression by in situ hybridization. Meth. Enzymol. 152, 649–661. Brooke, N. M., Garcia-Fernandez, J. & Holland, P. W. H. 1998. The ParaHox gene cluster is an evolutionary sister of the Hox gene cluster. Nature 392, 920–922. Burglin, T. R. 1994. A comprehensive classification of homeobox genes. In Guidebook to the Homeobox Genes (ed. D. Duboule), pp. 27–71. Oxford University Press, New York. Chen, C. P. & Run, J. Q. 1991. Induction of larval metamorphoses of Archaster typicus (Echinodermata: Asteroidea). Bull. Inst. Zool. Acad. Sinica 30, 257–260. Dayhoff, M. O., Schwartz, R. E. & Hung, L. T. 1978. A model of evolutionary changes in proteins. In Atlas of Protein Sequence and Structure (ed. M. O. Dayhoff), pp. 345–352. National Biomedical Research Foundation, Silver Spring, MD. Desplan, C., Theis, J. & O’Farrell, P. H. 1988. The sequence specificity of homeodomain–DNA interaction. Cell 54, 1081–1090. Dick, M. H. & Buss, L. W. 1994. A PCR-based survey of homeobox genes in Ctenodrilus serratus (Annelida: Polychaeta). Mol. Phylogenet. Evol. 3, 146–158. Donovan, J., Lu, X. & Nagarajan, L. 1993. Rapid purification of bacteriophage DNA. Biotechniques 15, 602–603.
Spatial distribution of starfish AtXlox
Felsenstein, J. 2000. phylip 3.6: Phylogeny Inference Package. Available from URL: http://evolution.genetics.washington. edu/phylip/ Ferrier, D. E. K. & Holland, P. W. H. 2001. Sipunculan ParaHox genes. Evol. Dev. 3, 263–270. Ferrier, D. E. K. & Holland, P. W. H. 2002. Ciona intestinalis ParaHox genes: evolution of Hox/ParaHox cluster integrity, developmental mode, and temporal colinearity. Mol. Phyogenet. Evol. 24, 412–417. Gehring, W. J. 1994. A history of the homeobox. In Guidebook to the Homeobox Genes (ed. D. Duboule), pp. 1–10. Oxford University Press, New York. Gellon, G. & McGinnis, W. 1998. Shaping animal body plans in development and evolution by modulation of Hox expression patterns. Bioessays 20, 116–125. Hano, Y., Hayashi, A., Yamaguchi, S. & Yamaguchi, M. 2001. Hox genes of the direct-type developing sea urchin Peronella japonica. Zool. Sci. 18, 353–359. Irvine, S. Q., Warinner, S. A., Hunter, J. D. & Martindale, M. Q. 1997. A survey of homeobox genes in Chaetopterus variopedatus and analysis of polychaete homeodomains. Mol. Phylogenet. Evol. 7, 331–345. Joyner, A. L. & Martin, G. R. 1987. En-1 and En-2, two mouse genes with sequence homology to the Drosophila engrailed gene: expression during embryogenesis. Genes Dev. 1, 29–38. Krumlauf, R. 1994. Hox genes in vertebrate development. Cell 78, 191–201. Macdonald, P. M. & Struhl, G. 1986. A molecular gradient in early Drosophila embryos and its role in specifying body pattern. Nature 324, 537–545. Manak, J. R. & Scott, M. P. 1994. A class act: conservation of homeodomain protein functions. Development 1994 (Suppl.), 61–71. Milewski, W. M., Duguay, S. J., Chan, S. J. & Steiner, D. F. 1998. Conservation of PDX-1 structure, function, and expression in zebrafish. Endocrinology 139, 1440–1449. Miller, C. P., McGehee, R. E. Jr & Habener, J. F. 1994. IDX-1: a new homeodomain transcription factor expressed in rat pancreatic islets and duodenum that transactivates the somaostatin gene. EMBO J. 13, 1145–1156. Minguillon, C. & Garcia-Fernandez, J. 2002. The single amphioxus Mox gene: insights into the functional evolution of Mox genes, somites, and the asymmetry of amphioxus somitogenesis. Dev. Biol. 246, 455–465. Mito, T. & Endo, K. 1997. A PCR survey of Hox genes in the sea star, Asterina minor. Mol. Phylogenet. Evol. 8, 218– 224. Mito, T. & Endo, K. 2000. PCR survey of Hox genes in the crinoid and ophiuroid: evidence for anterior conservation and
93
posterior expansion in the echinoderm Hox gene cluster. Mol. Phylogenet. Evol. 14, 375–388. Morris, V. B., Brammall, J., Byrne, M. & Frommer, M. 1997. Hoxtype and non-Hox homeobox gene sequences in genomic DNA of the sea urchin Holopneustes purpurescens. Gene 201, 107–110. Ohlsson, H., Karlsson, K. & Edlund, T. 1993. IPF1, homeodomain-containing transactivator of the insulin gene. EMBO J. 12, 4251–4259. Patel, N. H., Martin-Blanco, E., Coleman, K. G. et al. 1989. Expression of engrailed proteins in arthropods, annelids, and chordates. Cell 58, 955–968. Peshavaria, M., Henderson, E., Sharma, A., Wright, C. V. E. & Stein, R. 1997. Functional characterization of the transactivation properties of the PDX-1 homeodomain protein. Mol. Cell Biol. 17, 3987–3996. Pollard, S. L. & Holland, P. W. H. 2000. Evidence for 14 homeobox gene clusters in human genome ancestry. Curr. Biol. 10, 1059–1062. Rushlow, C., Doyle, H., Hoey, T. & Levine, M. 1987. Molecular characterization of the zerknullt region of the Antennapedia gene complex in Drosophila. Genes Dev. 1, 1268–1279. Ruvkun, G. & Hobert, O. 1998. The taxonomy of developmental control in Caenorhabdtis elegans. Science 282, 2033–2041. Sambrook, J., Fritsch, E. F. & Maniatis, T. 1989. Molecular Cloning: A Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Shih, L. J., Chen, C. A., Chen, C. P. & Hwang, S. P. L. 2002. Identification and characterization of bone morphogenetic protein 2/4 gene from the starfish Archaster typicus. Compar. Biochem. Physiol. Part B 131, 143–151. Wedeen, C. J., Kostriken, R. G., Matsumura, I. & Weisblat, D. A. 1990. Evidence for a new family of evolutionarily conserved homeobox genes. Nucl. Acids Res. 18, 1908. Weiss, J. B., Von Ohlen, T., Mellerick, D. M., Dressler, G., Doe, C. Q. & Scott, M. P. 1998. Dorsoventral patterning in the Drosophila central nervous system: the intermediate neuroblasts defective homeobox gene specifies intermediate column identity. Genes Dev. 12, 3591–3602. Wray, G. A. 1997. Echinoderms. In Embryology: Constructing the Organism (eds S. F. Gilbert & A. M. Raunio), pp. 309– 329. Sinauer Associates, Sunderland, MA. Wright, C. V. E., Schnegelsberg, P. & DeRobertis, E. M. 1988. Xlhbox 8: a novel Xenopus homeo protein restricted to a narrow band of endoderm. Development 104, 787–794. Wysocka-Diller, J., Aisemberg, G. O. & Macagno, E. R. 1995. A novel homeobox cluster expressed in repeated structures of the midgut. Dev. Biol. 171, 439–447.
