• neural induction;
  • BMP;
  • nervous system;
  • Xenopus;
  • microarray


  1. Top of page
  2. Abstract
  7. Acknowledgements

To isolate novel genes regulating neural induction, we used a DNA microarray approach. As neural induction is thought to occur by means of the inhibition of bone morphogenetic protein (BMP) signaling, BMP signaling was inhibited in ectodermal cells by overexpression of a dominant-negative receptor. RNAs were isolated from control animal cap explants and from dominant-negative BMP receptor expressing animal caps and subjected to a microarray experiment using newly generated high-density Xenopus DNA microarray chips representing over 17,000 unigenes. We have identified 77 genes that are induced in animal caps after inhibition of BMP signaling, and all of these genes were subjected to whole-mount in situ hybridization analysis. Thirty-two genes showed specific expression in neural tissues. Of the 32, 14 genes have never been linked to neural induction. Two genes that are highly induced by BMP inhibition are inhibitors of Wnt signaling, suggesting that a key step in neural induction is to produce Wnt antagonists to promote anterior neural plate development. Our current analysis also proves that a microarray approach is useful in identifying novel candidate factors involved in neural induction and patterning. Developmental Dynamics 232:432–444, 2005. © 2004 Wiley-Liss, Inc.


  1. Top of page
  2. Abstract
  7. Acknowledgements

Formation of the vertebrate nervous system is initiated at early gastrula stages, when signals from the organizer trigger neural development in the ectoderm. In recent years, it became clear that a blockade of bone morphogenetic protein (BMP) signaling during gastrula stages is sufficient to initiate neural induction of the ectoderm in Xenopus (Hawley et al., 1995; Sasai et al., 1995; Wilson and Hemmati-Brivanlou, 1995; Hemmati-Brivanlou and Melton, 1997), whereas in the presence of endogenous BMP signaling, ectodermal specification is maintained. Consistent with this notion, the organizer secretes several BMP inhibitors such as noggin, chordin, Xnr3, follistatin, and cerberus, and promotes the prospective ectodermal cells to adopt a neural fate (Lamb et al., 1993; Hemmati-Brivanlou et al., 1994; Sasai et al., 1995; Bouwmeester et al., 1996; Hansen et al., 1997). Notably, inhibition of BMP signaling in ectoderm can only lead to induction of anterior neural marker genes, suggesting that additional factors are necessary for posterior neural patterning (Hawley et al., 1995; Wilson and Hemmati-Brivanlou, 1995). Unknown at this time is how neural induction proceeds after BMP signaling inhibition or the identity of target genes that affect the neural induction process.

We have used a genome-wide DNA microarray approach in combination with a Xenopus animal cap assay to identify genes involved in the early event of neural induction. High-density DNA microarray slides consisting of over 38,300 Xenopus cDNA spots were generated to identify genes involved in neural induction (expressed sequence tag [EST] sequence data are available from RNA was isolated from control uninjected animal cap explants and from animal cap explants expressing a dominant-negative BMP receptor (Suzuki et al., 1994). DNA microarray analysis using these RNAs led to the identification of 77 genes that are up-regulated upon inhibition of BMP signaling. These 77 genes were subjected to a systematic whole-mount in situ analysis, and here we report on the spatiotemporal expression patterns of these genes. We anticipate that some of these target genes will serve as useful markers in studying neural induction in the future.


  1. Top of page
  2. Abstract
  7. Acknowledgements

Generation of Xenopus cDNA Microarrays

To produce high-density microarray slides for the analysis of Xenopus gene activity, normalized neurula and tailbud (stage 15 and stage 25) cDNA libraries were generated. Normalization was used to simultaneously reduce the frequency of abundant genes and to increase the frequency of rare cDNAs within the library. A total of 42,240 cDNA clones representing 19,200 neurula and 23,040 tailbud stage Xenopus clones were arrayed into 384-well plates (Table 1). Arrayed cDNAs were subjected to EST sequencing and annotated by N. Ueno's group at NIBB (National Institute for Basic Biology, Japan; Kitayama et al., manuscript in preparation). The EST sequence information is available at the XDB Web site, From the 3′ sequence and contig analyses, we estimate that 42,240 cDNAs represent 17,000 unigenes. To this collection, we have included additional 326 cDNAs that were absent in the arrayed libraries. These cDNAs represent replicates of 93 genes listed in Table 2, which includes lambda phage DNA fragments and a cDNA encoding green fluorescence protein (GFP). All 42,566 clones (42,240 plus 326 cDNAs) were subjected to polymerase chain reaction (PCR) amplification. The overall success rate of 42,566 PCR reactions was 90% and an average insert size was 2 kb. Amplified DNAs were purified, transferred to 384-well plates, and loaded into the robotic arrayer to generate high-density cDNA microarray slides. More than 38,300 PCR amplicons were spotted separately onto two slides (set 1 and set 2, Table 1); each set containing approximately 19,000 PCR products.

Table 1. Clones Used for Xenopus Microarrays
SlideSourceNumber of clones
Set1Mochii Xenopus neurula library19,200
 Mochii Xenopus tailbud library1,920
 Collection of clones in the lab144
Set2Mochii Xenopus tailbud library21,120
 Collection of clones in the lab182
Table 2. List of Control Genes
  • a

    These genes were printed on both slide sets (set1 and set2).

Known genes (62)     
 Activinαe cateninBMP-1BMP-7Cardiac actinCG1
 Cytochrome C oxidase I αaE.hairyElongation factor 1αaEngrailed 2
 Xhairy 2BXlhbox4Xlhbox-8Xlim-1Xmox2Xnot
Clones that derived from Axeldb (18)    
 lambda phage DNA fragments (13)    

Potential Neural Genes Identified by Microarray Analysis

We have used our high-density microarray slides to identify genes that are induced upon inhibition of BMP signaling. Two-cell stage embryos were microinjected with 4 ng of dominant negative BMP receptor I (DNBR) mRNA. Animal cap explants were isolated from blastula stage (stage 8–8.5) embryos injected with DNBR mRNA and control uninjected embryos and then allowed to develop until early gastrula stage (stage 10.5). RNAs isolated from the explants were subjected to microarray experiments.

The following criteria were used to identify genes up-regulated in response to inhibition of BMP signaling. First, microarray experiments were repeated twice, and bad spots were flagged out using GeneData Expressionist (GeneData). The surviving spots were then analyzed by using Bayesian statistical analysis as used in the Cyber-T program (Long et al., 2001) to determine confidence levels for each spot value in the replicates. P values of less than 0.01 were considered to be statistically reliable, and only these spots were subjected to further analysis. Among the qualified spots, spots exhibiting differential expression above an arbitrary cutoff were scored as genes “induced” or “suppressed” in response to BMP signaling inhibition. Among 38,300 spots examined, 30,034 spots survived the initial analysis, representing 11,918 and 18,116 clones from set 1 and set 2 slide, respectively. From those clones, we selected the clones whose induction level was more than 1.5-fold in both of the duplicated experiments, and average induction folds of two experiments were calculated. Among these, 42 clones representing 17 unigenes were induced more than three-fold, 101 clones representing 29 unigenes were induced more than two-fold, and 188 clones representing 77 unigenes were induced more than 1.7-fold (Table 3). Although a cutoff value of two-fold is often used to identify genes that are induced after an array experimental treatment, we decided to systematically characterize the 77 genes that survived the 1.7-fold cutoff. By lowering cutoff values, we may be able to identify novel neural genes that have not yet been characterized previously, although the likelihood of picking false-positive clones certainly increases.

Table 3. Summary of Genes Induced After BMP Signaling Inhibition and Their Expression Patternsa
IDDescriptionRT-PCRFoldExpression pattern at stageExpression pattern referred in
  • a

    Genes were organized into categories based on their expression patterns at the neural and gastrula stages and sorted by average fold induction (Fold) in microarray experiments. PNE, prospective neural ectoderm in gastrula stage embryos; ANP, anterior neural plate; MNP, mid-neural plate; PNP, posterior neural plate; NC, neural crest of neurula stage embryos. Specific expression was scored as −, no specific expression; +/−, broad but somewhat specific expression pattern; +, specific expression. Twenty three genes in category C were subjected to RT-PCR analyses. Induction after inhibition of BMP signaling was scored as +, induced; +/−, no change. RT-PCR, reverse transcriptase-polymerase chain reaction; BMP, bone morphogenetic protein; N/D, not determined.

A. Strong expression in both prospective neuroectoderm in gastrula and in neural tissues at neurula
 XL056i07Otx2 [Xenopus laevis] 13.68++Blitz and Cho, 1995
 XL007b09Retinoic acid hydrolase XCYP26 [Xenopus laevis] 5.04+++Hollemann et al., 1998
 XL099i04Zic-related-1 protein [Xenopus laevis] 4.43+++Mizuseki et al., 1998
 XL0c2d20Xanf-1 [Xenopus laevis] 4.01++Zaraisky et al., 1995
 XL006c04Secreted frizzled-related protein 2-related [Xenopus laevis] 3.94+++This paper
 XL098p21KIAA0806 hypothetical protein [Homo sapiens] ortholog 3.78+++/−+/−This paper
 XL063p10XLS13A protein [Xenopus laevis] 3.47+++++This paper
 XL092o07Sox2 [Xenopus laevis] 3.27++++Mizuseki et al., 1998
 XL032a12Geminin H [Xenopus laevis] 3.27+++Kroll et al., 1998
 XL060c23Secreted frizzled-related protein 2 [Xenopus laevis] 3.16+++Pera and De Robertis, 2000
 XL061n17Zic5 [Xenopus laevis] 3.01+++Nakata et al., 2000
 XL105b21Frizzled 7 [Xenopus laevis] 2.92+++Sumanas et al., 2000
 XL037i04Transcription factor Otx1 [Xenopus laevis] 2.68++Kablar et al., 1996
 XL066a18Histone H3 [Xenopus laevis] 2.58++++This paper
 XL025p08Sox3 protein [Xenopus laevis] 2.21++++Penzel et al., 1997
 XL002o08No homology 2.16+++This paper
 XL091p04HES-related IB [Xenopus laevis] 2.10+++Shinga et al., 2001
 XL094m13Pyruvate dehydrogenase kinase 2 [Phodopus sungorus] ortholog 1.93+++++This paper
 XL022a18Cyclin D1 [Xenopus laevis] 1.92++Vernon and Philpott, 2003
 XL058k21Prothymosin alpha [Homo sapiens] ortholog 1.89++++Munoz-Sanjuan et al., 2002
 XL052j22Transcription factor XTCF-3c [Xenopus laevis] 1.75+++Molenaar et al., 1998
B. Strong expression in prospective neuroectoderm at gastrula and weak expression in neural tissues at neurula
 XL102a14Serine dehydratase [Mus musculus] ortholog 1.93++/−+/−+/−This paper
 XL105m16No homology 1.80++/−+/−+/−This paper
C. Strong expression in neural tissues at neurula and weak/no expression in prospective neuroectoderm at gastrula
 XL061f17RTP801 [Rattus norvegicus] ortholog+2.14+This paper
 XL096a13No homology+2.09++This paper
 XL040i10Glycine amidinotransferase [Xenopus laevis]+2.05+++Zhao et al., 2001
 XL102p15Patched-2 [Xenopus laevis]+/−2.01+++Takabatake et al., 2000
 XL019d13Repulsive guidance molecule [Gallus gallus] ortholog+2.00+++This paper
 XL084i22Histone H2A.ZI [Xenopus laevis]+1.95+++This paper
 XL093119No homology+/−1.94++This paper
 XL103j13Tetraspan TSPAN-1 [Homo sapiens] ortholog+/−1.93+/−+This paper
 XL090115No homology+/−1.87+/−+This paper
 XL079a10Claudin-4L1 [Xenopus laevis]+/−1.82+++Fujita et al., 2002
 XL042i05Marginal coil Xmc [Xenopus laevis]+/−1.80+Frazzetto et al., 2002
 XL098c04No homology+1.79+++This paper
 XL008d01Ferritin H [Xenopus laevis]+/−1.79+++This paper
 XL010j21Claudin-4L2 [Xenopus laevis]+/−1.78+++Fujita et al., 2002
 XL068m21DM-GRASP precursor [Gallus gallus] homolog+1.77+++This paper
 XL098123Sp1-like zinc-finger protein XSPR-2 [Xenopus laevis]+/−1.76+++Ossipova et al., 2002
 XL013j09No homology+/−1.75+++This paper
 XL098j24[X76487] hypothetical protein [Homo sapiens] homolog+1.73+++This paper
 XL032k22LIM protein prickle b [Xenopus laevis]+1.73+Takeuchi et al., 2003
 XL104e24Zinc finger protein Fez [Xenopus laevis] homolog+1.73+Matsuo-Takasaki et al., 2000
 XL045a10Histone-binding protein N1/N2 [Xenopus laevis]+1.72++This paper
 XL102a23hnRNP D-like protein JKTBP [Mus musculus] ortholog+/−1.72+++This paper
 XL078a09Novel G protein-coupled receptor [Xenopus laevis]+/−1.70+This paper
D. Weak expression in neural tissues at neurula but no expression in prospective neuroectoderm at gastrula
 XL063a03No homology 2.67+/−+/−+/−This paper
 XL032101Transcription elongation factor type xTFHS.1 [Xenopus laevis] 2.04+/−+/−+/−Taira et al., 2000
 XL065h02No homology 2.03+/−+/−+/−This paper
 XL098c19Diamine N-acetyltransferase [Mus musculus] ortholog 1.98+/−+/−+/−This paper
 XL041d07Cellular nucleic acid binding protein [Mus musculus] ortholog 1.95+/−+/−+/−Flink et al., 1998
 XL043b17[AF047601] gi:4689090 SMCD [Mus musculus] ortholog 1.86+/−+/−+/−This paper
 XL092p03[AX136273] gi:14272680 Sequence 195 [Homo sapiens] ortholog 1.84+/−+/−+/−This paper
 XL026i17Ribonucleoside-diphosphate reductase M2 chain [Mus musculus] ortholog 1.84+/−+/−+/−This paper
 XL095f16G protein-coupled receptor RDC1 [Mus musculus] ortholog 1.83+/−+/−+/−This paper
 XL031c18No homology 1.82+/−+/−+/−This paper
 XL063i10No homology 1.82+/−+/−+/−This paper
 XL089a16MCT-1 gene product [Homo sapiens] ortholog 1.80+/−+/−+/−This paper
 XL102a04Thymosin beta 4 peptide [Xenopus laevis] 1.80+/−+/−+/−This paper
 XL036j02[AK015619] gi:12854023 hypothetical protein [Mus musculus] ortholog 1.79+/−+/−+/−This paper
 XL089a09Ubiquitin carboxyl-terminal hydrolase 7 [Homo sapiens] ortholog 1.76+/−+/−+/−This paper
 XL014i22CBP/p300-interacting transactivator CITED3 [Gallus gallus] ortholog 1.75+/−+/−+/−This paper
 XL074i03D7 protein (AA 1–278) [Xenopus laevis] 1.72+/−+/−+/−This paper
 XL079a15High mobility group HMG-14 [Xenopus laevis] 1.71+/−+/−+/−This paper
E. No expression in both prospective neuroectoderm at gastrula and neural tissues at neurula
 XL061o05No homology 2.45This paper
 XL092f11Muscle-cadherin precursor [Mus musculus] ortholog 2.31This paper
 XL084d01No homology 2.26This paper
 XL102h22No homology 2.04This paper
 XL070i21Myosin light chain kinase [Homo sapiens] ortholog 1.92This paper
 XL088a09Nuclear matrix protein NMP200 [Homo sapiens] ortholog 1.78This paper
 XL091m15CCCH zinc finger protein C3H-2 [Xenopus laevis] 1.78This paper
 XL093b05[AK003263] gi:12833824 hypothetical protein [Mus musculus] ortholog 1.75This paper
 XL092a22[BC018407] gi:17390960 Unknown protein [Mus musculus] ortholog 1.72This paper
 XL076a10Claudin-7L1 [Xenopus laevis] 1.71Fujita et al., 2002
 XL098c24Cyclin G2 [Homo sapiens] ortholog 1.71This paper
 XL042a03Embryonic serine protease-1 [Xenopus laevis] homolog 1.70This paper
F. Not determined
 XL006j22Calnexin [Rana rugosa] ortholog 1.71  N/D  N/D

As shown in Table 3, 65 of the 77 Xenopus genes appear to have orthologs or homologs in other organisms such as mouse, human, and chick, of which 55 genes were previously identified; the remaining 10 genes encode conserved hypothetical proteins of unknown function. The remaining 12 genes show no homology to any known genes. At present, it is not clear whether the lack of a homolog is due to the lack of full-length DNA sequence information or due to the Xenopus-specific nature of the genes. Through an extensive literature search, we identified the expression patterns of 23 genes in both gastrula and neurula stage embryos, and three additional genes only in neurula stage embryos. To obtain more complete expression data and to examine whether the candidate genes identified by this microarray approach are indeed expressed in neural ectoderm of the gastrula and/or neural stage embryo, we subjected the remaining 54 genes to whole-mount in situ hybridization analysis. With the exception of a calnexin ortholog that was not recovered from the original plate, all 53 clones were analyzed. As shown in Table 3, all of the genes that are induced strongly (higher than threefold, representing 11 genes) by inhibition of BMP signaling are indeed expressed in the prospective neuroectoderm of the gastrula embryo. Whereas most of these genes previously were shown to be expressed in neuroectoderm, we also identified three additional genes, XLS13A, secreted frizzled related protein 2-related, and Xenopus ortholog of KIAA0806 hypothetical protein (discussed later), to be expressed in neuroectoderm of the early gastrula embryo. Among the moderately induced (two- to three-fold, representing 18 genes) and weakly induced genes (1.7- to 2.0-fold, representing 47 genes), only 33% (6 genes) and 15% (6 genes) of these genes, respectively, were expressed in the prospective neuroectoderm region of the gastrula embryo (Table 3). Among a total of 23 genes that showed specific expression patterns in the prospective neuroectoderm of gastrula embryos, with the exception of serine dehydratase and XL105m16, the remainder were also expressed in the neural tissues at the neurula stage.

We find that, among the remaining 53 genes (of 77 genes) that were detected by microarrays, 23 genes were expressed in the neural tissues of late neurula stage embryos (Table 3), while their expression in the prospective neuroectoderm at the gastrula stage was not confirmed by whole-mount in situ hybridization. To better determine whether these 23 genes are expressed in the ectoderm of the gastrula stage embryo, a more sensitive reverse transcriptase-PCR (RT-PCR) analysis was performed. Eleven genes were confirmed to be significantly expressed after inhibition of BMP signaling (Table 3). These results suggest that microarray data at low induction levels should be interpreted with caution and that other means of verification such as RT-PCR or Northern blot are required. In summary, among the 77 genes characterized, 34 genes were expressed in the prospective neuroectoderm of the gastrula stage embryos. However, of the 34 genes, two genes were only expressed in the neuroectoderm of the gastrula embryos and not expressed in the neural tissues at the late neurula stage. Thus, in addition to the 18 genes previously reported to be expressed in the neural tissues, we identified 14 additional neural genes (Table 3; Fig. 1).

thumbnail image

Figure 1. Expression patterns of novel neural genes. Whole-mount in situ hybridization was performed to confirm localization of transcripts in whole embryos. Gene expression patterns shown are at gastrula stage (stage 10.5), neurula stage (stage 18), tailbud stage (stage 30). Average fold induction from microarray experiments are presented as “Fold.” a, anterior; ba, branchial arches; c, cranial ganglia; cg, cement gland; d, dorsal ectoderm; nc, neural crest; np, neural plate; ov, otic vesicle; p, posterior; *, dorsal blastopore lip. Induction of these genes in animal cap ectoderm after inhibition of bone morphogenetic protein (BMP) signaling was confirmed by reverse transcriptase-polymerase chain reaction analysis.

Download figure to PowerPoint

Transcription Factors

Among the 32 neural genes, 12 genes are transcription factors, most of which are already known. This list includes FeZ, Geminin H, HES-related 1B, Otx1, Otx2, Sox2, Sox3, Xanf-1, XLS13A, XTCF-3, Zic1, and Zic5. With the exception of XLS13A, these transcription factors were all implicated previously in neural induction. Based on the spatiotemporal expression patterns of XSL13A, which is expressed in the neuroectoderm of the gastrula embryo as well as in the entire neural plate and neural crest cells of the neurula embryo (Fig. 1A), it is tempting to speculate about its potential role as a global regulator of neural induction.

Wnt Signaling Modifiers

Regulation of Wnt signaling is thought to be crucial for the proper patterning of neural tissues (McGrew et al., 1992; Augustine et al., 1993; Wolda et al., 1993). Simultaneous inhibition of Wnt and BMP signaling pathways is necessary to form the proper head structures in Xenopus (Glinka et al., 1997, 1998). Furthermore, compound mouse mutations lacking both BMP inhibitor (noggin) and Wnt inhibitor (dickkopf1) cause more severe head defects (del Barco Barrantes et al., 2003) than either single mutation alone (McMahon et al., 1998; Mukhopadhyay et al., 2001). Our microarray analysis suggests that one of the major consequences of inhibiting BMP signaling is to promote the induction of Wnt antagonists, perhaps to ensure the anterior head specification. Both secreted frizzled-related protein 2 (SFRP2) and its closely related clone, SFRP2-related (SFRP2-r) are induced strongly (3.2- and 4.0-fold, respectively) after inhibition of BMP signaling. SFRP2 is a member of the frizzled related protein that has been shown to function as a Wnt antagonist (Pera and De Robertis, 2000). At present, we are not certain whether SFRP2 and SFRP2-r represent two different genes or allelic variants of the same gene. Nonetheless, the spatiotemporal expressed patterns of these genes are identical during early Xenopus embryogenesis. As shown in Figure 1A, these genes are expressed at high levels in the entire prospective neuroectoderm and in the organizer mesoderm of the gastrula stage embryo, and by mid-neurula stage, they are expressed preferentially in the anterior and posterior neural plate.

In addition to these Wnt antagonists, we also find that a Wnt receptor, frizzled 7, is strongly expressed in the middle region of the neural plate where SFRP2 and SFRP2-r is expressed weakly (Sumanas et al., 2000; our unpublished data). The expression of frizzled 7 also appears to coincide with the expression of engrailed-2. This expression is interesting, as it is known that expression of engrailed-2 requires Wnt signaling (McGrew et al., 1999). High expression levels of frizzled 7 and concomitant weak expression of SFRP2 and SFRP2-r may create high levels of Wnt signaling in the region. In summary, we find up-regulation of Wnt modifiers in the animal cap ectoderm after inhibition of BMP signaling, suggesting a close link between Wnt and BMP signaling pathways in neural induction.

Chromatin Remodeling Proteins

Three molecules, histone H3, histone H2A.ZI, and histone-binding protein N1/N2, are induced by inhibiting BMP signaling. Strong expression of histone H3 is detected both in the organizer and in the prospective neuroectoderm of the gastrula embryo (Fig. 1C) and subsequently in the neural plate. As active cell proliferation occurs in neural plate, increased expression of histone H3 may be consistent with the behavior of cells undergoing extensive cell proliferation and morphogenetic movements. Histone H2A.ZI is also expressed strongly in the entire neural plate at neural stages, but at gastrula stages, the expression is detected in the entire animal hemisphere. Histone-binding protein N1/N2 is expressed in the entire animal hemisphere of the gastrula embryo and in the neural plate of the neurula embryo. Later, the expression becomes more confined in the anterior neural plate (Fig. 1C). It is interesting to note that the expression pattern of H3 only partially overlaps with those of histone binding protein N1/N2, despite that they have been implicated to interact in Xenopus oocyte (Kleinschmidt et al., 1986). This finding suggests that histone binding protein N1/N2 has a histone H3-independent function.

Membrane Associated Proteins

Several interesting membrane associated proteins were shown to be neural genes. An ortholog of the hypothetical protein KIAA0806 encoding a putative transmembrane protein is expressed in the organizer and prospective neuroectoderm of the gastrula embryo (Fig. 1D). DM-GRASP precursor is a cell surface glycoprotein, known to be localized in axons during early development in chick and supports neurite extension from chick sensory neurons (Burns et al., 1991). At tailbud stage, strong expression is detected in cranial ganglia, nerves, and spinal cord (Fig. 1D). Repulsive axon guidance molecule is a membrane-associated glycoprotein having repulsive and axon-specific guiding activity (Monnier et al., 2002). The transcripts are detected in the forebrain, midbrain, hindbrain, and spinal cord at tailbud stage (Fig. 1D).

House Keeping Genes

Pyruvate dehydrogenase 2 (PDK2) is expressed in the neuroectoderm of the gastrula embryo and subsequently in the entire neural plate and neural crest region of the neurula embryo (Fig. 1E). RTP801 is expressed strongly in neural crest cells of the neurula embryos (Fig. 1E). Whereas RT-PCR analysis confirms the expression in the ectoderm at the gastrula stage, the presence cannot be detected by whole-mount in situ hybridization.

Novel Genes of Unknown Function

XL002o08 is expressed in the neuroectoderm of the gastrula embryo (Fig. 1F). At the neurula stage, high levels of XL002o08 expression are detected in the anterior half of the neural plate, and at tailbud stage, the expression is particularly strong in the eye and branchial arches (Fig. 1F). Two genes, XL098c04 and XL098j24, are expressed in the entire neural plate, and XL096a13 is expressed in the anterior neural plate (Fig. 1F). These genes are weakly expressed in the neuroectoderm of the gastrula embryo.


  1. Top of page
  2. Abstract
  7. Acknowledgements

We have successfully used Xenopus DNA microarrays to identify genes that are expressed in neural tissues after inhibition of BMP signaling. Although many of these genes have been shown previously to be expressed in neural tissues, we also identified new genes that were never linked to neural tissue development. Further analysis is required to define the roles these genes play in neural induction.

From our microarray, RT-PCR, and whole-mount in situ hybridization analyses, the following general conclusions can be drawn regarding amphibian neural induction. First, transcription factors are by far the most common genes induced upon BMP signaling inhibition. Perhaps this finding reflects that a battery of critical genes needs to be transcribed in the early phase of neural induction to control the cell specification at the transcriptional level. Second, whereas many genes are induced after inhibition of BMP signaling, we identified the genes that are down-regulated by the process (data not shown). These down-regulated genes include bona-fide BMP targets such as Xvent-2 and Vox-1. As these genes are normally expressed in the animal cap ectoderm of Xenopus blastula embryos where BMP signaling allows epidermal development to proceed, expression of a dominant-negative BMP receptor is expected to interfere with the expression of these BMP targets. Third, some Wnt signaling inhibitors are significantly induced after the inhibition of BMP signaling. This finding is interesting, as it has been suggested that the induction of anterior neural tissue requires inhibition of both BMP and Wnt signaling pathways. By inducing Wnt signaling inhibitors, inhibition of BMP signaling is likely to ensure the specification of anterior neural ectoderm to proceed. In this regard, it should be noted that inhibition of BMP signaling at early gastrula stage induces preferentially anterior neural genes such as Cyclin D1, Geminin H, HES-related 1B, Otx1, Otx2, Sox3, Xanf-1, XTcf-3, Zic1, Zic5 including a novel gene XL002o08, whereas expression of the most posterior neural marker genes (e.g., engrailed-2 and krox20) was unaffected. This finding is consistent with the notion that neural induction initially specifies the anterior state, subsequently other signaling molecules such as FGF, wnt, and retinoic acid control anteroposterior patterning of the neuroectoderm during gastrulation and throughout neurulation (Blumberg et al., 1997; Xu et al., 1997; McGrew et al., 1999). In addition to the Wnt signaling inhibitors, we also find the induction of a Wnt receptor frizzled 7, indicating a close link between BMP and Wnt signaling.

One of the advantages of the microarray system is the ease in identifying the clones of interest and making educated guesses as to the function of cDNAs so that researchers can quickly identify potentially important genes. Being able to easily access cDNAs to perform whole-mount in situ hybridization analysis, thus, is a major advantage of these cDNA arrays over Affymetrix or long oligo-based chip analyses. Despite the advantage of the cDNA array system, our current level of annotation is still not satisfactory as many genes are not assigned to other orthologs. To effectively use this array resource, further efforts should be made to better annotate the cDNAs, based on the future Xenopus tropicalis and full-length cDNA sequencing projects. We also believe that the sensitivity of cDNA arrays could limit the use of this potentially very powerful technology in the Xenopus system. Many false-positive clones are detected around the two-fold induction, forcing us to independently evaluate the array data based on other methods such as RT-PCR and whole-mount in situ hybridization analyses. Perhaps, the fidelity of our array analysis could improve if array experiments are repeated more than twice. Additionally, development of better algorithms to flag out false-positive signals at low signal ranges will greatly facilitate the use of a microarray approach. In this regard, we previously have reported an approach to minimize false-positive spots by measuring the mean and median ratio (Tran et al., 2002). When a similar analysis was performed on 48 genes that are induced between 1.7- and 2-fold, six of six genes (100%) shown to be expressed in the neural tissues by whole-mount in situ hybridization survived the screen. On the other hand, only 21 of 41 genes (51%) found not to be expressed in neuroectoderm survived the flagging. With further refinement of the algorithm, we may be able to increase the accuracy of microarray data at the lower end signals.


  1. Top of page
  2. Abstract
  7. Acknowledgements

Microarray Slides

Bacterial clones selected for the array were grown to confluence overnight in flat-bottom 384-well plates with 55 μl of ampicillin containing LB medium per well. Two microliters of bacteria culture were transferred into 18 μl of TE (Tris-HCl, 100 mM, pH 8.0, ethylenediaminetetraacetic acid 0.1 mM) and boiled for 3 min at 95°C. PCR amplification was performed using standard 384-well plates in a 26-μl reaction volume (28 cycles, 94°C for 30 sec, 57.5°C for 30 sec, 72°C for 2 min; Hegde et al., 2000). The size and amount of each PCR product were verified by running 1% agarose gel. The 384-well PCR plates showing over 90% percent successful PCR amplification rate were subjected to ethanol precipitation, followed by a 70% ethanol wash. DNA was resuspended in a buffer containing 50% dimethylsulfoxide and spotted with a home-made robotic microarray spotter onto Corning CMT-GAP slides (Tran et al., 2002). After printing, spots were briefly rehydrated and ultraviolet crosslinked using a Stratalinker (Stratagene).

Probe Generation, Array Hybridization, and Slide Scanning

Xenopus embryos were obtained by in vitro fertilization of eggs with testes homogenates. Embryos were dejellied in 2% cysteine and staged according to Nieuwkoop and Faber (Faber et al., 1996). Synthetic mRNA for dominant-negative BMP receptor (DNBR) was injected into both blastomeres of two-cell stage embryos. Animal caps were dissected at stage 8–8.5 and cultured in 1× Barth's solution. Total RNA was isolated from animal cap ectodermal explants using TRIzol Reagent (Gibco), followed by a lithium chloride precipitation method.

Fluorescent labeled probes were generated according to Tran et al. (2002) with the following modifications. Fifty micrograms of total RNA was used in reverse transcription reactions with oligo(dT)20 using Superscript II enzyme (Gibco) in the presence of Cy3- and Cy5-dUTP. RNA from animal caps injected with DNBR mRNA was labeled with Cy5-dUTP, and control RNA was labeled with Cy3-dUPT. The samples were treated with 1 μl of 1 M NaOH to degrade the RNA. Cy3 and Cy5 fluorescent-labeled probes were combined, purified first by using a PCR purification kit (Qiagen) and then by using Microcon YM-30 spin column filters (Amicon) to remove unincorporated dyes and free nucleotides. The eluate volume was adjusted to 17 μl and mixed with an equal volume of 2× hybridization buffer (0.2% sodium dodecyl sulfate, 6.8× standard saline citrate and yeast tRNA at 1 μg/μl). Hybridization was performed at 65°C for 20 hr, and the slides were washed as described (Tran et al., 2002). Hybridized slides were scanned with a GenePix 4000B scanner (Axon Instruments). The laser power was set to 100%m and PMT voltage (normally 700–900 volts) was adjusted depending upon levels of signal saturation. For array data analysis, we used a GenePix Pro image acquisition program, a Cyber-T program ( and GeneData Expressionist (GeneData).

Whole-Mount In Situ Hybridization Analysis

In situ hybridization was performed essentially as described previously (Harland, 1991), except that BM purple (Boehringer-Mannheim) was used as the chromogenic substrate. As cDNAs were directionally cloned into pBSIISK, inserts were PCR-amplified by using T3 and T7 primers. Antisense RNA probes were generated by using a T7 RNA polymerase (New England Biolabs)

RT-PCR Analysis

The protocol for RT-PCR is previously described (Blitz and Cho, 1995), except that the reaction products were visualized on ethidium bromide–stained agarose gels. Unless otherwise indicated, the basic PCR protocol was 94°C, 5 min; 1 cycle (94°C, 30 sec; 55°C, 30 sec, 72°C, 30 sec); 72°C, 10 min. Specific PCR cycles, annealing temperature, and primer pair sequences for genes are as follows: XL061f17 (RTP801): 25 cycles, 5′-TTGTGGACTTAGGGGAGCAC-3′ and 5′-CTGGCTTGGTGCTAAAGAGG-3′; XL096a13 (no homology): 26 cycles, 5′-GGCATTGTGGGAATCTGAAT-3′ and 5′-ATGATTGCCCTTGTTTGGAG-3′; XL040i10 (Glycine amidinotransferase): 30 cycles, 5′-TCAGGAAAGGCAGCTTCAAT-3′ and 5′-CCAGCTTTTG-GAAAGACAGC-3′; XL102p15 (Patched-2): 30 cycles, 5′-AATATGCCCCTGTGT-TGCTC-3′ and 5′-GGGCTTTGGGGTA-ACAAAT-3′; XL019D13 (repulsive guidance molecule): 26 cycles, 5′-CCGTCAGGTGGGTCACTACT-3′ and 5′-GAGGAGGTCAAAGACGCAAG-3′; XL084i22 (Histone H2A.ZI); 24 cycles, 5′-CTCTATCACCCGATCCTCCA-3′ and 5′-TTCCAGCCAATTCAAGAACC-3′; XL093l19 (No homology): 32 cycles, 5′-CACCTAGACGTGACCAGCAA-3′ and 5′-TCCAAACCGACTACCAGA-GG-3′; XL103j13 (Tetraspan TSPAN-1): 24 cycles, 5′-GTGGAAGAAGGCCCATGTTA-3′ and 5′-GGGGAGCTGT-TGAAGATGAA-3′; XL090l15 (No homology): 23 cycles, 5′-TGATGTTCTGGGGAAGGAAG-3′ and 5′-ACGTTGGGTCCTAACACAGC-3′; XL079a10 (Claudin-4L1): 24 cycles, 5′-TCCTCTGCTCTGATGCTCCT-3′ and 5′-CCAG-CGAGTTATGCCCTTAG-3′; XL042105 (Xmc): 23 cycles 5′-CTGGTGTTACAGACCAAGGGG-3′ and 5′-ACCTGTGCTTTTGCCACTC-3′; XL098c04 (No homology): 32 cycles, annealing temperature 65°C, 5′-CCGGTGTATGAGAA-GGGAGA-3′ and 5′-ATGAGGAGGTCAACCCACTG-3′; XL008d01 (Ferritin H): 28 cycles, 5′-AGGAGCAGAGTCACGAGGAA-3′ and 5′-CCTTCACCTGTTCCTCCAAG-3′; XL010j21 (Claudin-4L2): 24 cycles, 5′-GTGGTGGAGGCT-CAGAAGAG-3′ and 5′-CTGGGAGG-CAGCAGTGTATT-3′; XL068m21 (DM-GRASP precursor): 30 cycles, 5′-CATTGTGGGAATTGTCGTTG-3′ and 5′-GGCATCGGACTTGTGGTTAT-3′; XL098l23 (XSPR-2): 23 cycles, 5′-CAAACTGTTGCCTCTCATGAG-3′ and 5′-CACTTACACCTCCGGCAGC-GC-3′; XL013j09 (No homology): 32 cycles, 5′-CCTTGTTTTCCCAGCACTTT-3′ and 5′-ACTCCTTCCCATTCTGAGCA-3′; XL098j24 ([X76487] hypothetical protein): 26 cycles, 5′-CTGGTAGTGGAA-AGGCCAAA-3′ and 5′-GAGTTTGCTG-GGGTTGAAAA-3′; XL032k22 (LIM protein prickle b): 24 cycles, 5′-TGCGCTCAGGAGGTCTAAAT-3′ and 5′-TCCAA-AGAGGTCAGCGTTCT-3′; XL104e24 (FeZ): 32 cycles, annealing temperature 64°C, 5′-TCAACTCCTCGTCCATACCC-3′ and 5′-AGCAGCTTGTTCCTCTCAGC-3′; XL045a10 (Histone- binding protein N1/N2): 25 cycles, 5′-GCCACTGCTCCTTCAACTTC-3′ and 5′-CTCGCCATACTTCTGTGCAA- 3′; XL102a23 (JKTBP): 26 cycles, 5′-TTTTGGAGCCTTTGGAGAGA-3′ and 5′-TAATCATAGCCGCCATAGCC-3′;XL078a09 (Novel G protein-coupled receptor): 25 cycles, 5′-TTTACACCGCAAAGGAGGAC-3′ and 5′-TCACACAAATCACAGCAGCA-3′. Histone H4 (23 cycles) primers were used for the control as described previously (Blitz et al., 2003).


  1. Top of page
  2. Abstract
  7. Acknowledgements

We thank Dr. Christof Niehrs for providing cDNAs from Axeldb and Mrs. Cherryl Nugas for helpful technical assistance. K.W.Y.C. was funded by the NIH.


  1. Top of page
  2. Abstract
  7. Acknowledgements
  • Augustine K, Liu ET, Sadler TW. 1993. Antisense attenuation of Wnt-1 and Wnt-3a expression in whole embryo culture reveals roles for these genes in craniofacial, spinal cord, and cardiac morphogenesis. Dev Genet 14: 500520.
  • Blitz IL, Cho KW. 1995. Anterior neurectoderm is progressively induced during gastrulation: the role of the Xenopus homeobox gene orthodenticle. Development 121: 9931004.
  • Blitz IL, Cho KW, Chang C. 2003. Twisted gastrulation loss-of-function analyses support its role as a BMP inhibitor during early Xenopus embryogenesis. Development 130: 49754988.
  • Blumberg B, Bolado J Jr, Moreno TA, Kintner C, Evans RM, Papalopulu N. 1997. An essential role for retinoid signaling in anteroposterior neural patterning. Development 124: 373379.
  • Bouwmeester T, Kim S, Sasai Y, Lu B, De Robertis EM. 1996. Cerberus is a head-inducing secreted factor expressed in the anterior endoderm of Spemann's organizer. Nature 382: 595601.
  • Burns FR, von Kannen S, Guy L, Raper JA, Kamholz J, Chang S. 1991. DM-GRASP, a novel immunoglobulin superfamily axonal surface protein that supports neurite extension. Neuron 7: 209220.
  • del Barco Barrantes I, Davidson G, Grone HJ, Westphal H, Niehrs C. 2003. Dkk1 and noggin cooperate in mammalian head induction. Genes Dev 17: 22392244.
  • Faber KN, Westra S, Waterham HR, Keizer-Gunnink I, Harder W, Veenhuis GA. 1996. Foreign gene expression in Hansenula polymorpha. A system for the synthesis of small functional peptides. Appl Microbiol Biotechnol 45: 7279.
  • Flink IL, Blitz I, Morkin E. 1998. Characterization of cellular nucleic acid binding protein from Xenopus laevis: expression in all three germ layers during early development. Dev Dyn 211: 123130.
  • Frazzetto G, Klingbeil P, Bouwmeester T. 2002. Xenopus marginal coil (Xmc), a novel FGF inducible cytosolic coiled-coil protein regulating gastrulation movements. Mech Dev 113: 314.
  • Fujita M, Itoh M, Shibata M, Taira S, Taira M. 2002. Gene expression pattern analysis of the tight junction protein, Claudin, in the early morphogenesis of Xenopus embryos. Gene Expr Patterns 2: 2326.
  • Glinka A, Wu W, Onichtchouk D, Blumenstock C, Niehrs C. 1997. Head induction by simultaneous repression of Bmp and Wnt signalling in Xenopus. Nature 389: 517519.
  • Glinka A, Wu W, Delius H, Monaghan AP, Blumenstock C, Niehrs C. 1998. Dickkopf-1 is a member of a new family of secreted proteins and functions in head induction. Nature 391: 357362.
  • Hansen CS, Marion CD, Steele K, George S, Smith WC. 1997. Direct neural induction and selective inhibition of mesoderm and epidermis inducers by Xnr3. Development 124: 483492.
  • Harland RM. 1991. In situ hybridization: an improved whole-mount method for Xenopus embryos. Methods Cell Biol 36: 685695.
  • Hawley SH, Wunnenberg-Stapleton K, Hashimoto C, Laurent MN, Watabe T, Blumberg BW, Cho KW. 1995. Disruption of BMP signals in embryonic Xenopus ectoderm leads to direct neural induction. Genes Dev 9: 29232935.
  • Hegde P, Qi R, Abernathy K, Gay C, Dharap S, Gaspard R, Hughes JE, Snesrud E, Lee N, Quackenbush J. 2000. A concise guide to cDNA microarray analysis. Biotechniques 29: 548550, 552–544, 556 passim.
  • Hemmati-Brivanlou A, Melton D. 1997. Vertebrate neural induction. Annu Rev Neurosci 20: 4360.
  • Hemmati-Brivanlou A, Kelly OG, Melton DA. 1994. Follistatin, an antagonist of activin, is expressed in the Spemann organizer and displays direct neuralizing activity. Cell 77: 283295.
  • Hollemann T, Chen Y, Grunz H, Pieler T. 1998. Regionalized metabolic activity establishes boundaries of retinoic acid signalling. EMBO J 17: 73617372.
  • Kablar B, Vignali R, Menotti L, Pannese M, Andreazzoli M, Polo C, Giribaldi MG, Boncinelli E, Barsacchi G. 1996. Xotx genes in the developing brain of Xenopus laevis. Mech Dev 55: 145158.
  • Kleinschmidt JA, Dingwall C, Maier G, Franke WW. 1986. Molecular characterization of a karyophilic, histone-binding protein: cDNA cloning, amino acid sequence and expression of nuclear protein N1/N2 of Xenopus laevis. EMBO J 5: 35473552.
  • Kroll KL, Salic AN, Evans LM, Kirschner MW. 1998. Geminin, a neuralizing molecule that demarcates the future neural plate at the onset of gastrulation. Development 125: 32473258.
  • Lamb TM, Knecht AK, Smith WC, Stachel SE, Economides AN, Stahl N, Yancopolous GD, Harland RM. 1993. Neural induction by the secreted polypeptide noggin. Science 262: 713718.
  • Long AD, Mangalam HJ, Chan BY, Tolleri L, Hatfield GW, Baldi P. 2001. Improved statistical inference from DNA microarray data using analysis of variance and a Bayesian statistical framework. Analysis of global gene expression in Escherichia coli K12. J Biol Chem 276: 1993719944.
  • Matsuo-Takasaki M, Lim JH, Beanan MJ, Sato SM, Sargent TD. 2000. Cloning and expression of a novel zinc finger gene, Fez, transcribed in the forebrain of Xenopus and mouse embryos. Mech Dev 93: 201204.
  • McGrew LL, Otte AP, Moon RT. 1992. Analysis of Xwnt-4 in embryos of Xenopus laevis: a Wnt family member expressed in the brain and floor plate. Development 115: 463473.
  • McGrew LL, Takemaru K, Bates R, Moon RT. 1999. Direct regulation of the Xenopus engrailed-2 promoter by the Wnt signaling pathway, and a molecular screen for Wnt-responsive genes, confirm a role for Wnt signaling during neural patterning in Xenopus. Mech Dev 87: 2132.
  • McMahon JA, Takada S, Zimmerman LB, Fan CM, Harland RM, McMahon AP. 1998. Noggin-mediated antagonism of BMP signaling is required for growth and patterning of the neural tube and somite. Genes Dev 12: 14381452.
  • Mizuseki K, Kishi M, Matsui M, Nakanishi S, Sasai Y. 1998. Xenopus Zic-related-1 and Sox-2, two factors induced by chordin, have distinct activities in the initiation of neural induction. Development 125: 579587.
  • Molenaar M, Roose J, Peterson J, Venanzi S, Clevers H, Destree O. 1998. Differential expression of the HMG box transcription factors XTcf-3 and XLef-1 during early xenopus development. Mech Dev 75: 151154.
  • Monnier PP, Sierra A, Macchi P, Deitinghoff L, Andersen JS, Mann M, Flad M, Hornberger MR, Stahl B, Bonhoeffer F, Mueller BK. 2002. RGM is a repulsive guidance molecule for retinal axons. Nature 419: 392395.
  • Mukhopadhyay M, Shtrom S, Rodriguez-Esteban C, Chen L, Tsukui T, Gomer L, Dorward DW, Glinka A, Grinberg A, Huang SP, Niehrs C, Belmonte JC, Westphal H. 2001. Dickkopf1 is required for embryonic head induction and limb morphogenesis in the mouse. Dev Cell 1: 423434.
  • Munoz-Sanjuan I, Bell E, Altmann CR, Vonica A, Brivanlou AH. 2002. Gene profiling during neural induction in Xenopus laevis: regulation of BMP signaling by post-transcriptional mechanisms and TAB3, a novel TAK1-binding protein. Development 129: 55295540.
  • Nakata K, Koyabu Y, Aruga J, Mikoshiba K. 2000. A novel member of the Xenopus Zic family, Zic5, mediates neural crest development. Mech Dev 99: 8391.
  • Ossipova O, Stick R, Pieler T. 2002. XSPR-1 and XSPR-2, novel Sp1 related zinc finger containing genes, are dynamically expressed during Xenopus embryogenesis. Mech Dev 115: 117122.
  • Penzel R, Oschwald R, Chen Y, Tacke L, Grunz H. 1997. Characterization and early embryonic expression of a neural specific transcription factor xSOX3 in Xenopus laevis. Int J Dev Biol 41: 667677.
  • Pera EM, De Robertis EM. 2000. A direct screen for secreted proteins in Xenopus embryos identifies distinct activities for the Wnt antagonists Crescent and Frzb-1. Mech Dev 96: 183195.
  • Sasai Y, Lu B, Steinbeisser H, De Robertis EM. 1995. Regulation of neural induction by the Chd and Bmp-4 antagonistic patterning signals in Xenopus. Nature 376: 333336.
  • Shinga J, Itoh M, Shiokawa K, Taira S, Taira M. 2001. Early patterning of the prospective midbrain-hindbrain boundary by the HES-related gene XHR1 in Xenopus embryos. Mech Dev 109: 225239.
  • Sumanas S, Strege P, Heasman J, Ekker SC. 2000. The putative wnt receptor Xenopus frizzled-7 functions upstream of beta-catenin in vertebrate dorsoventral mesoderm patterning. Development 127: 19811990.
  • Suzuki A, Thies RS, Yamaji N, Song JJ, Wozney JM, Murakami K, Ueno N. 1994. A truncated bone morphogenetic protein receptor affects dorsal-ventral patterning in the early Xenopus embryo. Proc Natl Acad Sci U S A 91: 1025510259.
  • Taira Y, Kubo T, Natori S. 2000. Participation of transcription elongation factor XSII-K1 in mesoderm-derived tissue development in Xenopus laevis. J Biol Chem 275: 3201132015.
  • Takabatake T, Takahashi TC, Takabatake Y, Yamada K, Ogawa M, Takeshima K. 2000. Distinct expression of two types of Xenopus Patched genes during early embryogenesis and hindlimb development. Mech Dev 98: 99104.
  • Takeuchi M, Nakabayashi J, Sakaguchi T, Yamamoto TS, Takahashi H, Takeda H, Ueno N. 2003. The prickle-related gene in vertebrates is essential for gastrulation cell movements. Curr Biol 13: 674679.
  • Tran PH, Peiffer DA, Shin Y, Meek LM, Brody JP, Cho KW. 2002. Microarray optimizations: increasing spot accuracy and automated identification of true microarray signals. Nucleic Acids Res 30: e54.
  • Vernon AE, Philpott A. 2003. The developmental expression of cell cycle regulators in Xenopus laevis. Gene Expr Patterns 3: 179192.
  • Wilson PA, Hemmati-Brivanlou A. 1995. Induction of epidermis and inhibition of neural fate by Bmp-4. Nature 376: 331333.
  • Wolda SL, Moody CJ, Moon RT. 1993. Overlapping expression of Xwnt-3A and Xwnt-1 in neural tissue of Xenopus laevis embryos. Dev Biol 155: 4657.
  • Xu RH, Kim J, Taira M, Sredni D, Kung H. 1997. Studies on the role of fibroblast growth factor signaling in neurogenesis using conjugated/aged animal caps and dorsal ectoderm-grafted embryos. J Neurosci 17: 68926898.
  • Zaraisky AG, Ecochard V, Kazanskaya OV, Lukyanov SA, Fesenko IV, Duprat AM. 1995. The homeobox-containing gene XANF-1 may control development of the Spemann organizer. Development 121: 38393847.
  • Zhao H, Cao Y, Grunz H. 2001. Expression of Xenopus L-arginine:glycine amidinotransferase (XAT) during early embryonic development. Dev Genes Evol 211: 358360.