SEARCH

SEARCH BY CITATION

Keywords:

  • cDNA array;
  • localized RNA;
  • maternal transcript;
  • Xenopus laevis

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References

In many organisms, proper embryo development depends on the asymmetrical distribution of mRNA in the cytoplasm of the egg. Here we report comprehensive screening of RNA localized in the animal or vegetal hemisphere of the Xenopus egg. Macroarrays including over 40 000 independent embryonic cDNA clones, representing at least 17 000 unigenes, were differentially hybridized with labeled probes synthesized from the mRNA of animal or vegetal blastomeres. After two rounds of screening, we identified 33 clones of transcripts that may be preferentially distributed in the vegetal region of the early stage embryo, but transcripts localized in the animal region were not found. To assess the array results, we performed northern blot and quantitative real-time reverse transcription–polymerase chain reaction analysis. As a result, 21 transcripts of the 33 were confirmed to be localized in the vegetal region of the early stage embryo. Whole-mount in situ hybridization analysis revealed that 11 transcripts, including 7 previously reported genes, were localized in the vegetal hemisphere of the egg. These 11 transcripts were categorized into three groups according to their expression patterns in the egg. The first group, which contained four transcripts, showed uniform expression in the vegetal hemisphere, similar to VegT. The second group, which contained three transcripts, showed gradual expression from the vegetal pole to the equator, similar to Vg1. The last group, which contained three transcripts, was expressed at the germ plasm, similar to Xdazl. One transcript, Xwnt11, showed both the second and the third expression patterns.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References

In many vertebrates and invertebrates, several RNA are known to be localized asymmetrically in the cytoplasm of the egg. This asymmetrical distribution of RNA in the egg is thought to be necessary for proper embryo development. In the Xenopus egg, vegetally localized VegT mRNA, which encodes a T-box transcription factor of the brachyury family, is essential to primary germ layer formation of both endoderm and mesoderm (Lustig et al. 1996; Stennard et al. 1996; Zhang & King 1996; Horb & Thomsen 1997; Zhang et al. 1998; Kloc et al. 2001). The transcripts for Vg1, Xotx1, xBic-C, and XNIF are also reported to be localized in the vegetal hemisphere of the Xenopus egg (Rebagliati et al. 1985; Melton 1987; Pannese et al. 2000; Wessely & De Robertis 2000; Kloc et al. 2001; Claussen et al. 2004). Another group of vegetally localized RNA such as Xcat-2, Xdazl, Xpat, DEADSouth, germes, and Xwnt11 are deposited in the vegetal pole region of the egg cytoplasm, which is believed to contain determinants of germ cell fate and is termed germ plasm (Ku & Melton 1993; Mosquera et al. 1993; Houston et al. 1998; Hudson & Woodland 1998; MacArthur et al. 2000; Kloc et al. 2001; Berekelya et al. 2003). For example, Xdazl encodes an RNA-binding protein that can act as a functional homologue of Drosophila boule. Xdazl is required for early primordial germ cell differentiation and is necessary for the migration of primordial germ cells through the endoderm (Houston & King 2000). In contrast, An1, An2, and An3 RNA are localized in the animal hemisphere (Rebagliati et al. 1985; Kloc et al. 2001). Comprehensive screening of asymmetrically localized mRNA and functional analysis of such genes is necessary to complete our understanding of early embryonic development and germ cell development.

Large-scale analysis using cDNA arrays is a powerful tool for elucidating the developmental biology of Xenopus (Altmann et al. 2001; Chung et al. 2004; Peiffer et al. 2005; Shin et al. 2005; Tazaki et al. 2005). In combination with the recently developed DNA array method and expressed sequence tag (EST) information (Lockhart & Winzeler 2000), we succeeded in identifying 33 candidate genes whose transcripts are localized in the vegetal hemisphere of the Xenopus egg.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References

Eggs and embryos

Wild-type and albino South African clawed frogs (Xenopus laevis) were purchased commercially and raised at 22°C until use. Embryos were obtained by in vitro fertilization, dejellied, and cultured in autoclaved tap water until the first cleavage. For the dissociation of blastomeres, embryos were transferred to 80% phosphate-buffered saline lacking calcium and magnesium (2.7 mm KCl, 0.14 m NaCl, 1.5 mm KH2PO4, 8.1 mm Na2HPO4) in agarose-coated plastic dishes at the 2-cell stage, and were cultured at 14°C. During the 8-cell stage, the embryos were manually demembranated and the blastomeres were dissociated. If the blastomeres did not dissociate spontaneously, they were dissociated by gentle shaking.

cDNA array

Description of the preparation and characterization of the Xenopus neurula (stage 15) and tailbud (stage 25) cDNA libraries can be found at the XDB3 website, http://xenopus.nibb.ac.jp (Kitayama, unpubl. data). Inserts of 42 240 independent cDNA clones were polymerase chain reaction (PCR)-amplified with vector pBluescript SK (Stratagene, La Jolla, CA, USA) primers (T3/T7) from bacterial stock and arrayed on a nylon membrane filter (Hybond-N+, Amersham Biosciences, Amersham, UK) with a gridding robot (microGrid TAS, Bio Robotics, Cambridge, UK) at high density (384 × 5 × 5 cDNA grids per 8 cm × 12 cm filter). The arrayed high-density filter, the macroarray, was denatured with 0.5 m NaOH, neutralized with 0.5 m Tris-HCl (pH 7.0), and baked at 80°C for 2 h. Recovery of cDNA clones and mini-array preparation were performed as previously described (Tazaki et al. 2005). DNA for the mini-array was amplified with purified plasmid, electrophoresed on agarose gel to confirm the amount, and arrayed as duplicated spots for each clone.

RNA preparation and differential array hybridization

Total RNA was extracted from animal or vegetal blastomeres of 8-cell-stage embryos with Trizol reagent (Invitrogen, San Diego, CA, USA), and was further purified by centrifugation on 5.7 m CsCl cushion containing a 2.5 mm EDTA. For probe labeling, 40 µg of purified total RNA was incubated in 40 µL of solution containing 8 µg of oligo-d(T), 1200 units of M-MLV reverse transcriptase (Invitrogen) and [α-32P]dCTP (Amersham Biosciences) at 37°C for 2 h. Probe was purified with a G-50 spin column (Amersham Biosciences). To mask the poly T sequence in the probe, labeled DNA was prehybridized with 4 µg of oligo-d(A)30 in 400 µL of solution containing 4× SSC at 65°C for 12 h. Duplicated macroarray filters for each probe were preincubated in hybridization solution containing 4× SSC, 5× Denhardt's solution without bovine serum albumin (0.1% polyvinylpyrrolidone/0.1% Ficoll), 0.5% SDS, and 100 µg/mL salmon sperm DNA at 65°C for 1 h, then incubated in the same solution containing the prehybridized probe at 65°C for 20 h. The hybridized filters were washed four times in 0.1× SSC containing 0.1% SDS at 60°C for 20 min, and exposed to a Fuji imaging plate for 24 h. The exposed imaging plate was scanned with an FLA3000 (Fuji BAS system, Fujifilm, Tokyo, Japan) at 50 µm resolution. The scanned BAS image was analyzed with ArrayGauge software (Fujifilm). The photo-stimulated luminescence (PSL) of each spot on the macroarray was normalized by the sum total PSL of all spots on the membranes between animal and vegetal blastomeres. The intensities of corresponding spots from the two duplicated filters were averaged, and were compared between animal and vegetal blastomeres using Microsoft Excel, as described (Mochii et al. 1999). For the mini-array analysis, elongation factor 1-alpha (EF1-α) was used as a positive control for normalization.

Northern blot

Total RNA was prepared from animal or vegetal blastomeres of 8-cell-stage embryos, as descried above. Ten micrograms of total RNA was glyoxylated, electrophoresed in a 1.2% agarose gel, and transferred to a nylon membrane (Hybond N, Amersham Biosciences; Sambrook et al. 1989). The membrane was incubated at 80°C for 2 h. To confirm that the RNA was transferred, the membrane was stained with 0.04% methylene blue containing 0.5 m sodium acetate (pH 5.2; Sambrook et al. 1989). The plasmid insert was amplified by PCR with the T3 and T7 primers, and was labeled using a BcaBEST labeling kit (Takara, Kusatsu, Japan) and [α-32P]dCTP (Amersham Biosciences). The probe was purified with a G-50 spin column (Amersham Biosciences). The nylon membrane was pre-incubated in hybridization solution (see differential array hybridization) at 65°C for 1 h, then incubated in the same solution containing the labeled probe at 65°C for 20 h. The hybridized membrane was washed four times in 0.1× SSC containing 0.1% SDS at 55°C for 20 min, and was exposed to the Fuji imaging plate for at least 24 h. Scanning of the images was described above (see differential array hybridization). The scanned BAS image was analyzed with Image Gauge software (Fujifilm). The intensities of hybridization signals were normalized by EF1-α between animal and vegetal. The single band of BAS images was compared between animal and vegetal blastomeres. If there were double or triple bands, the total intensities of all bands between animal and vegetal blastomeres were compared.

In situ hybridization

Whole-mount in situ hybridization was performed essentially as previously described (Sive et al. 2000). For easy identification of mRNA localized in animal or vegetal hemispheres, we used both pigmented (wild type) and unpigmented (albino) eggs. Wild-type and albino unfertilized eggs were fixed in 25% methanol and 20% formalin for 3 h at room temperature and stored in 100% methanol at −20°C. To detect a transcript inside the egg, some of fixed pigmented eggs were bisected along the animal–vegetal axis with a razor in 100% methanol. The plasmid inserts were PCR-amplified with T3 and T7 primers and purified with an S-400 spin column (Amersham Biosciences). Antisense riboprobes were synthesized from purified PCR products in the presence of digoxigenin-11-UTP (Roche Diagnostics, Mannheim, Germany) by T7 RNA polymerase (Invitrogen). The bisected and intact unfertilized eggs were hybridized with the probe at 65°C for 16 h. Hybridized probe was detected by alkaline phosphatase conjugated antidigoxigenin Fab fragments (Roche) and BMpurple (Roche) as a substrate for color development.

Quantitative real-time reverse transcription–polymerase chain reaction

Total RNA was prepared from animal or vegetal blastomeres of 8-cell stage embryos as described above. Complementary DNA was synthesized from 0.5 or 1 µg of total RNA using Ready-To-Go You-Prime First-Strand Beads and random hexamers (Amersham Biosciences), according to the manufacturer's instructions. The quantification of each gene was done using the PRISM 7000 Sequence Detection System (Applied Biosystems) and SYBR GREEN PCR Master Mix (Applied Biosystems). The specific primers used to quantify are described in Table 1. The amplification profile for these primer pairs was as follows: 95°C for 10 min, 40 cycles of 95°C for 30 s, 60°C for 30 s and 72°C for 30 s. Standard curves were generated for each gene by serial dilution of cDNA from whole 8-cell-stage embryos. Unknown quantities were calculated by comparison to the standard curve for each gene. Quantities of all samples were normalized to levels of the endogenous standard, EF1-α. Experiments were repeated at least twice on different embryo batches to ensure that the quantities described were reproducible from one experiment to the next.

Table 1.  Primers for QT RT–PCR Analysis
Clone IDSense primer (5′−3′)Antisense primer (5′−3′)
  • Did not yield reliable data using these primer sets in the QT RT–PCR analysis. QT RT–PCR, quantitative real-time reverse transcription–polymerase chain reaction.

EF1-αCAGGCCAGATTGGTGCTGGATATGCGCTCTCCACGCACATTGGCTTTCCT
XL002d04TGATTCGCATCTTAGCGTACACGTTTGTTGGCACCAGGTATAGACAAGGATGGCAGCATA
XL003i03GCCTCAAAGTAAATCTGACGCATATTAAGGATAAGGAGCTGGCCATTTTACCTAGGAAA
XL010a22GAACAATGTCGCTTGCTGGACGGTCTAGTGTTCCAGCGCATGTCAGATAAGGTCATTTG
XL011a11CCCTGGTGGACCTTTTTAACATGAATCTAAGGCCATCTTTTCTCAGTTGACGACATTACT
XL013k14TTACACGGGCTTTGGGTGTTAATAGTCCTTCACCGACGAGGAGCTTAGGGTCTTTAGA
XL014k14CTAACAATCATTGTTTCCCTTGGTGTGATCCACATCCTTGGAGTTGCCTTTGTATTTATG
XL017a13TCTCCTGGCATTCTAGTCCTATTGTGGAATAAAAGCTGACCCCTACACACTTATGCATGT
XL019p18CTGGAACTTGGGATCCTGCACTAACAACACCACAAAGGGCCGTTGTGGATAGAATCACTA
XL025a01TAAGAGATGTGGAAAAGGAACCAGTACTTTGTCATGGTATATTTACATGCACACAAGTTG
XL026p20GACAGGAGTGTCCAAGGGGTATGGCTTCATGTTCCATGCATGGTGGAATGGAGTAGGGTG
XL027p20TTCTCTGCCAACTCCAGTGAAGCCTGTATGTCCATTGCTTTCCTGGTATACAACCGCTTT
XL032c03ATATGCTCAAAAAGTTGGCCTGCTCATAAACAGTTAGAGGAGCATTGCTATTTTTGC
XL037i04AGCTGGATGTGCTGGAGTCCCTGTTCTACTCTGGACTCGGGCAGGTTGATCTTTAG
XL050g20AGAGGAAAATCTGCAAAGGACCAGCAGACAGTCTCTGGAAGAGGCAGAGGAGGATGTTGA
XL064m20CTCAGAAGGTCCAAAAGTGGCTCATTCTTGGGGTTCCTGCATCCTGATAGAAGCTATCAA
XL071o17AGACCCAACACCAGACTAAGTGGACTCCTTCAGCCTAAAGAACTAGCACACACACTGCAA
XL076e07CGCAGGCAGAATAAGGAATGAATCTCTTGGTCATATTCCTCTCCCAAAACATGGCAGTT
XL076l14GAAGTAAAGCATCCGTTCATTGTGGACCTTCAGGCTGTGTCCTCCATAAATATCCCTTCT
XL078d10CAGTGGATGGACACAACGAATTTTCAGCTATGGGTTTGTGTGGGAGATCTTGTAGGGTAA
XL081p07AAAGATGTTGGATGGCAAAACTGGGACAGCAAGGAAGAGGAATGTCTTCT
XL084d10CTCTTACTGCAAGCACTGGTTCCCCTGTTTGCGGGTATTCTTGATGGTTTGGTTTCATA
XL086i14TGGGAAGGGCAAAAGACCTCTGAAATGAGCAGCCCAGTTCAGATCCACTACTC
XL087b08CCAGAACTGCCTCCATTTTCCCCTAACATAAGGGTGAAAGCTGTTACGCAGGTGTTTCTT
XL088b08CGAGACCTTCCAGCATCCATACAACACCTTCAGTACACGCCATCTTCTTGCTCTGGAATG
XL096g22AGAGAACATCACTACAGGTGCCCAAATCCACGGACTGCAGCTCCACAGAACTTCAGAT
XL099d22TCTCTCTTATGGGCAACAGCAAGCAGTAGAACCATTACAGCATTTTGCTGTGGAGGATGT
XL103d24ACACTTACATTTTTGGTGTTACTGCTCCTTTTCTACATAACTAGCCATCAAAGCGACAT
XL104f05CTTTCTTTCTCCTGACGGCTTCCTTGACTCAAGGGATTTGTGTACTGTCAACACGAATG
XL105l17CAGTGCCGGATTCCGATCCAGATTCTTTTAGTGGGTGAGCTACTGCTCCTTGTGTGCTTAA
XL106g04CCCCTCGCTTTTTGCCCTTTTACTATGCCCACCATGCTAAGTTTTCTCCGACAA
XL106l17CTGGTCAGTTGGCTGTATCATGGCAGAGATGGAGCTTCTGGACAAAATCCTGAGTTGGT
XL107g04AGATCCCTTCTCCTGGGTTACTCCGTGTCTAGAGGTTCAATGCATAGTGCTGGCTTGCT
XL108g04GTATTATAGCTGTGGGCACCTGGGATGCTTCCAACATGGTGAGACTGCCTTCCA

Results and discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References

Differential hybridization analysis with arrayed cDNA

To identify maternal RNA preferentially localized in the animal or vegetal hemisphere of the unfertilized egg and blastomeres of the early cleavage embryo, we performed a differential hybridization of a cDNA-based macroarray. The macroarray contains 42 240 independent Xenopus cDNA clones, including at least 17 000 unigenes isolated from the neurula and tailbud normalized libraries (Shin et al. 2005). We used this array because a cDNA array prepared from oocytes was not available at that time. The EST sequence information at the XDB3 website, http://xenopus.nibb.ac.jp, indicated that many maternal transcripts were contained in this cDNA array, including VegT, Vg1, Xdazl, etc. It seemed useful as an initial step to screen for maternal transcripts with the array. 32P-labeled probes were therefore synthesized with total RNA prepared from animal and vegetal blastomeres of 8-cell-stage embryos, and were hybridized against the duplicated macroarrays for each probe. The animal and vegetal blastomeres were isolated in calcium- and magnesium-free medium, as described in Materials and methods (Fig. 1).

image

Figure 1. A dissociated embryo. An 8-cell-stage wild-type embryo was cultured in calcium-free, magnesium-free medium, and vitelline membrane was removed. Four animal blastomeres are pigmented on the left, and the vegetal blastomeres are on the right. Animal view. Bar, approximately 400 µm.

Download figure to PowerPoint

The radioactivity of each spot was quantified, and compared between animal and vegetal blastomeres. The screening resulted in 135 clones whose intensities differed by more than twofold between animal and vegetal blastomeres. Of the 135 clones, 103 had intensities higher than 10 PSL in at least one of the hybridized spots on duplicated filters and were chosen for further analysis, as intensities lower than 10 PSL were probably masked by the background signals (Mochii et al. 1999). We obtained 5′ and 3′ sequences of the selected clones from the XDB3 EST database. However, we determined the nucleotide sequences of several cDNA clones whose sequence information was absent from the database.

The macroarray includes redundant clones for several genes. The 103 cDNA clones were classified into 81 groups (animal, 41; vegetal, 40) representing independent genes using the XDB3 EST database information and the sequences determined by us. Inserts of 81 representative cDNA clones were PCR-amplified from the plasmid, and cDNA of the 64 amplified clones (animal, 26; vegetal, 38) was confirmed by agarose gel electrophoresis (17 clones were not amplified in this step).

Second differential hybridization analysis with mini-array

To confirm the reliability of the selected clones, the cDNA of the 64 clones was blotted as duplicated spots on a nylon membrane to create a mini-array. Each spot of the mini-array contained much more DNA than the first array (see Materials and methods) and exhibited clearer and stronger signals. The mini-arrays were hybridized with the labeled probes from the animal or vegetal blastomeres, as described above (Fig. 2). The second screening resulted in 33 positive clones whose intensities of hybridization signals in vegetal blastomeres were twofold or greater than in animal. However, none of the animally abundant clones was selected. Based on BLAST searches and the XDB3 database, we provide descriptions of the genes and the ratios of the screening in Table 2.

image

Figure 2. Hybridized mini-array filter images. The amplified cDNA was blotted as duplicated spots on a nylon membrane (EF1-α cDNA was blotted as a control at the four corners), and hybridized with the probe prepared from RNA of animal (A) or vegetal (V) blastomeres in the 8-cell-stage embryo, as represented in the upper two panels. The lower left two panels (a) and (v), are magnifications of the spots in the upper panels. For each clone, a schematic representation of the orientation of that spot paired within the image is provided at the lower right.

Download figure to PowerPoint

Table 2.  Summary of the screening in the 8-cell-stage embryo and in situ hybridization analysis in the egg
A. Representing already known vegetally localized genes
Representative cloneRedundant clonesDescriptionRatio Vegetal/Animal byGene expression pattern by in situ hybridization
ArrayMini arrayNorthern blotReal-time RT–PCR
XL010a22XL074k18, XL078n12Xwnt11 (Xenopus laevis)2.03.0>6830Germ plasm and Gradual in vegetal hemisphere
XL026p20 Xdazl (Xenopus laevis)3.39.5>63011Germ plasm
XL037i04 Xotx1 (Xenopus laevis)2.05.28.9 3.9Gradual in vegetal hemisphere
XL081p07 Vg1 (Xenopus laevis)2.29.725 6.4Gradual in vegetal hemisphere
XL087b08 XNIF (Xenopus laevis)2.55.74.4 5.6Gradual in vegetal hemisphere
XL096g22 Xpat (Xenopus laevis)2.03.4>66017Germ plasm
XL105l17 VegT (Xenopus laevis)2.69.112 6.8Uniform in vegetal hemisphere
B. Representing newly identified vegetally localized genes
Representative cloneRedundant clonesDescriptionRatio Vegetal/Animal byGene expression pattern by in situ hybridization
ArrayMini-arrayNorthern blotReal-time RT–PCR
  1.  Representative clones of the groups are shown in the left column and the redundant clones were also selected simultaneously by the macroarray. Ratio = photo stimulated luminescence (PSL) of vegetal blastomeres/PSL of animal blastomeres. The PSL of each spot of the macroarray was normalized by the sum total PSL of all spots on the membranes between animal and vegetal blastomeres. Ratio on the mini-array, the northern blots and the real-time reverse transcription–polymerase chain reaction (RT–PCR) was normalized by EF1-α between animal and vegetal. Ratio of the real-time RT–PCR was the averaged data from two independent experiments. UD, unreliable data; ND, not detected. The PSL of the animal blastomeres could not be detected. In this case, ratio = PSL of vegetal blastomeres/PSL of the background.

XL002d04XL050o15No significant similarity2.0 4.13314Uniform in vegetal hemisphere
XL003i03XL048n21, XL060i14Phosphotyrosine binding protein (Xenopus laevis)3.0 5.54.5 5.6Uniform in vegetal hemisphere
XL011a11 Tob protein (Xenopus laevis)2.1 3.3  6.0ND
XL013k14XL053n06No significant similarity7.012>45 6.0Germ plasm
XL014k14 Similar to neogenin (Gallus gallus)3.1 3.01.1 0.9ND
XL017a13 No significant similarity3.6 6.5 UDND
XL019p18 Similar to endocytic receptor Endo180 (Homo sapiens)2.2 3.8 UDND
XL025a01XL008g12, XL009g12, XL011l01, XL015a07, XL069j15, XL071o18, XL083d10, XL108i06Similar to long chain fatty acyl CoA synthetase protein(Mus musculus)5.2 8.027 7.8ND
XL027p20XL028p20Similar to phosphoribosylpyrophosphate synthetase-associated Protein 39 (Rattus norvegicus)2.2 2.4ND 1.0ND
XL032c03 No significant similarity2.0 9.2  9.2ND
XL050g20 No significant similarity2.3 2.2  0.8ND
XL064m20XL051j20, XL076e08Similar to PRTD-NY3 protein (Homo sapiens)2.4 4.9  4.1ND
XL071o17 No significant similarity2.0 2.9  0.7ND
XL076e07XL065j05Calcium/calmodulin-dependent protein kinase II Gamma L subunit (Xenopus laevis)2.5 5.1ND 7.7ND
XL076l14 Ribosomal protein S6 kinase (Xenopus laevis)2.2 6.2ND 1.1ND
XL078d10 Similar to matrix metalloproteinase 14 protein(Xenopus laevis)2.7 2.4  2.2ND
XL084d10 S-ADENOSYLMETHIONINE DECARBOXYLASE PROENZYME (Xenopus laevis)2.4 2.0  1.0ND
XL086i14 Similar to transcription factor E2F (Mus musculus)2.0 2.63.9 2.4ND
XL088b08 KREMEN, kringle-containing transmembrane protein(Xenopus laevis)2.2 2.1ND 1.2ND
XL099d22XL007a15, XL050i20, XL056c02, XL100d22Similar to ring finger protein38 (Homo sapiens)2.3 2.65.9 6.1Uniform in vegetal hemisphere
XL103d24 No significant similarity2.2 4.3 16ND
XL104f05 No significant similarity2.3 2.2>30 1.1ND
XL106g04 No significant similarity3.0 7.3ND12ND
XL106l17 Xp38gamma/SAPK3 protein kinase (Xenopus laevis)2.1 2.1  1.0ND
XL107g04 Similar to ubiquitin-specific protease isotype 40(Homo sapiens)2.8 4.1  3.6ND
XL108g04 Similar to putative galactose-binding protein (Danio rerio)2.0 2.3  0.1ND

Of the 33 vegetally abundant clones, 14 encode the proteins already reported in Xenopus. Ten clones had some sequence similarities to known genes or EST in Xenopus or other organisms, but the other nine clones had no significant sequence similarities. Vegetally localized genes identified by us include VegT, Vg1, Xotx1, XNIF, Xdazl, Xpat, and Xwnt11 (Table 2A). Transcripts of these seven genes have previously been reported to be localized vegetally in the egg (Rebagliati et al. 1985; Melton 1987; Ku & Melton 1993; Lustig et al. 1996; Stennard et al. 1996; Zhang & King 1996; Horb & Thomsen 1997; Houston et al. 1998; Hudson & Woodland 1998; Zhang et al. 1998; Pannese et al. 2000; Kloc et al. 2001; Claussen et al. 2004). This result confirms the credibility of the macroarray analysis. Our screening could not identify other reported genes of asymmetrically localized transcripts such as xBic-C, Xcat-2, DEADSouth, germes, An1, An2, or An3 (Rebagliati et al. 1985; Mosquera et al. 1993; MacArthur et al. 2000; Wessely & De Robertis 2000; Kloc et al. 2001). One reason for this screening failure was that the macroarray used in this study did not contain these cDNA. In fact, we could not find EST for xBic-C, Xcat-2, germes, or An1 in the database. Some clones showed intensity ratios lower than 2.0. For example, the intensity of the DEADSouth clone XL035o16 in vegetal blastomeres was approximately 1.6-fold that in animal blastomeres. The intensities of An2 (XL088n11) and An3 (XL039l13, XL070b20, XL075j11, XL090d02, and XL092o24) clones, however, did not differ significantly between the animal and vegetal blastomeres.

Further analysis using a cDNA array made with oocyte cDNA should make possible identification of many more RNA localized along the animal–vegetal axis. Nonetheless, we found in the present study the asymmetrical mRNA distribution of 26 candidate genes in vegetal blastomeres of the Xenopus early stage embryo in addition to 7 genes that had been previously reported. These 26 candidates contained 9 novel genes.

Reliability of the array analysis

To assess the reliability of our macroarray screen, we performed differential hybridization by northern blot analysis and quantitative real-time reverse transcription–polymerase chain reaction (QT RT–PCR). Northern blot analysis was expected to provide more reliable data than the array, and was performed with 20 selected genes out of 33 positive genes from the array analyses. In this selection, we chose 13 genes randomly in addition to 7 previously reported genes serving as a positive control. As a result, 14 transcripts were vegetally abundant in 8-cell-stage embryos, but the signals for one transcript, representative clone XL014k14, did not differ between animal and vegetal blastomeres (Fig. 3, Table 2A,B). The other five genes were not detected in this analysis (Table 2B).

image

Figure 3. Differential hybridization analysis by northern blots. To confirm the reliability of the array analysis, 20 selected genes out of 33 positive ones were subjected to the differential hybridization by northern blot. 10 µg of total RNA from animal or vegetal blastomeres was loaded at each membrane (animal, left lane; vegetal, right lane). Fourteen transcripts were abundant in the vegetal blastomeres of the 8-cell-stage embryo, while one, XL014k14, was not. Signals of five other genes were not detected in this analysis. The upper end of each image, except for EF1-α, coincides with the loading start level. The arrowheads represent 28S and 18S ribosomal RNA. XL087b08 and XL099d22 hybridized with transcripts of at least two different sizes. EF1-α was the internal control for these RNA.

Download figure to PowerPoint

We expected QT RT–PCR to be more sensitive than the hybridization methods, the array and northern blot, and it was performed with all 33 genes selected by the array analysis. As a result, 21 transcripts were confirmed to be over 2.0-fold abundant in vegetal blastomeres of 8-cell-stage embryos, but the value for 9 transcripts did not differ between animal and vegetal blastomeres (Table 2A,B). One transcript, clone XL108g04, was animally abundant on QT RT–PCR analysis. The other two clones XL017a13 and XL019p18 did not show reliable data (Table 2B). We believed that the amounts of these transcripts were too small in 8-cell-stage embryos to amplify by this method, because DNA fragments were amplified from the plasmid using the same primer sets.

Vegetal abundance of the 21 transcripts was confirmed by these three methods, but the array results of the 12 transcripts were not consistent with that of northern blot and/or QT RT–PCR. In some cases, the quantitative data from array analysis may contain a few doubts, as described previously (Taniguchi et al. 2001). Nonetheless, macroarray was a useful tool for the first screening of asymmetrical localized transcripts.

Expression analysis by in situ hybridization

To reveal the detailed distribution patterns of the transcripts of the selected clones by the array analysis, we performed whole-mount in situ hybridization (WISH) with antisense digoxigenin (DIG)-labeled riboprobes for the 33 genes on the unfertilized eggs. WISH revealed that mRNA for 11 genes was vegetally localized. The other 22 genes showed no obvious expression patterns, suggesting either low-level expression of these genes or that these transcripts might be deposited deep inside the egg. To identify a distribution pattern of the transcripts inside the egg, we performed in situ hybridization on the bisected eggs. However, hybridization signals for the 22 transcripts were not clearly detected even in the bisected egg.

The 11 representative clones showing signals in WISH on the unfertilized egg were categorized into three groups according to their staining patterns from outside view. The first group containing XL002d04, XL003i03, XL099d22, and XL105l17 (VegT) showed uniform staining in the vegetal hemisphere (Fig. 4A–D). These transcripts showed the clear border of the staining on the equatorial region (Fig. 4A–D and unpubl. data). On hybridization to bisected eggs, we could not detect clear staining of these transcripts inside the egg. One reason for this result is that these transcripts might diffuse inside the egg, as described previously (Stennard et al. 1996). The second group containing XL037i04 (Xotx1), XL087b08 (XNIF), and XL081p07 (Vg1) showed gradual staining from the vegetal pole to the equator (Fig. 4E–G). The gradual signal of this group was also confirmed on the bisected eggs (Fig. 4E′–G′). The last group contained XL013k14, XL026p20 (Xdazl) and XL096g22 (Xpat), whose expression signals were concentrated at the vegetal pole region, which was the germ plasm region (Fig. 4I–K). On the bisected eggs, these transcripts formed a granular appearance within the vegetal cortex (Fig. 4I′–K′). XL010a22 (Xwnt11) showed the gradual staining from vegetal pole to the equator and also the signals in the germ plasm. This gene was categorized into both the second and the last groups (Fig. 4H, H′ and arrow in the lowest panel). WISH using 8-cell-stage embryos showed that the transcripts of 11 genes remained in the vegetal blastomeres (data not shown). In further analysis to confirm the zygotic expression of these 11 genes in early development, we also performed WISH on gastrula-, neurula-, and tailbud-stage embryos. As a result, we detected the expression patterns of already reported genes, XL105l17 (VegT; Zhang & King 1996), XL037i04 (Xotx1; Kablar et al. 1996), XL010a22 (Xwnt11; Ku & Melton 1993), XL026p20 (Xdazl; Houston et al. 1998), and XL096g22 (Xpat; Hudson & Woodland 1998). The expression patterns of XL013k14 reminded us of Xdazl or Xpat (detailed analysis will be published elsewhere), but we could not detect obvious signals of the others.

image

Figure 4. In situ hybridization showing vegetally localized genes. Unfertilized eggs were obtained from albino (A–K) and wild-type (E′–K′) Xenopus laevis. XL002d04 (A), XL003i03 (B), XL099d22 (C), and XL105l17 (VegT) (D) were expressed uniformly in the vegetal hemisphere. XL037i04 (Xotx1) (E, E′), XL087b08 (XNIF) (F, F′), and XL081p07 (Vg1) (G, G′) were gradually expressed with the most condensed staining at the vegetal pole. XL13k14 (I, I′), XL026p20 (Xdazl) (J, J′), and XL096g22 (Xpat) (K, K′) were expressed in the germ plasm. XL010a22 (Xwnt11) (H, H′) was gradually expressed with the most condensed staining at the vegetal pole, and also expressed in the germ plasm (arrows). (A–G) Lateral view of unpigmented eggs with the animal pole on the top, and (H–K) Vegetal view. (E′–K′) Cut face of the egg, which was bisected along animal-vegetal axis, with the pigmented animal pole on the top. Lowest panels were higher magnification view of H′–K′. Scale bar in A–K, approximately 200 µm.

Download figure to PowerPoint

The three different expression patterns of these transcripts in the egg are thought to depend on the different RNA localization mechanisms in oogenesis (Kloc & Etkin 1995, 2005). Further in situ hybridization analysis in oogenesis is required to determine the relationship between RNA localization mechanisms and the expression pattern in the egg.

The last group of mRNA was expected to play a role in germ cell differentiation or migration, similar to the role of Xdazl (Houston & King 2000). The different distribution patterns of the first and second mRNA groups may reflect their different functions in development. The precise role of the asymmetrically localized mRNA, except for VegT, has not yet been determined.

Comprehensive analysis using a cDNA array is a powerful tool for elucidating the developmental biology of Xenopus. Using cDNA array and QT RT–PCR, we succeeded in identifying 21 vegetally localized transcripts in Xenopus egg, including 7 genes whose localization patterns have previously been reported. These maternal transcripts may play a role in endodermal and part of mesodermal development or germ cell development. Systematic functional analysis of these genes should bring about a profound understanding of the early development of Xenopus.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References

This work was supported by an award from the ‘Research for the Future Program’ of the Japanese Society for the Promotion of Science to N.U. and a Grant-in-Aid from the Ministry of Education, Science, Sports, and Culture, Japan to M.M. We thank Dr Orii, and other members of our laboratory for helpful discussions and maintenance of animals.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results and discussion
  6. Acknowledgements
  7. References