Development 112, 747-753 (1991)Printed in Great Britain © The Company of Biologists Limited 1991

747

Localized and inducible expression of Xenopus-posterior {Xpo), a novel

gene active in early frog embryos, encoding a protein with a 'CCHC' finger

domain

SHERYL M. SATO1* and THOMAS D. SARGENT2

^Clinical Endocrinology Branch, NIDDK, and 2Laboratory of Molecular Genetics, NICHD, National Institutes of Health, Bethesda, MD20892, USA

* Corresponding author

Summary

Xenopus-posterior (Xpo) is a gene that is activated at orshortly after the midblastula transition (MBT). TheRNA accumulates to a relatively low level, whichremains constant until gastrulation, then rapidly andtransiently increases in posterior ectoderm and meso-derm. A single copy of a putative finger motif, of the'CCHC type, is located near the carboxyl terminus.One or two copies of similar sequence motifs are found inthe nucleocapsid protein of retroviruses where they areinvolved in protein-RNA interactions, and in cellularnucleic acid binding protein (CNBP), a protein that

binds to the sterol regulatory element. Xpo expression isinduced in ectodermal explants by treatment with basicnbroblast growth factor (bFGF) and with polypeptidegrowth factors found in medium conditioned by theXenopus XTC cell line (XTC-CM). Taken together,these properties suggest a possible role for Xpo in theorganization of the anteroposterior axis during develop-ment.

Key words: Xenopus, posterior (Xpo), gene activity,protein.

Introduction

An important problem in developmental biology is toestablish how the patterning of embryonic tissues isdetermined with respect to the dorsoventral andanteroposterior axes. One theory has suggested that thebody patterning is organized by concentration gradientsof morphogens which specify particular developmentalconsequences (Turing, 1952). The importance ofmorphogen gradients has been demonstrated elegantlyin Drosophila, where for example, the bicoid proteingradient has been directly implicated in the spatialorganization of body segmentation (Nusslein-Volhardet al. 1987; Struhl et al. 1989). In vertebrate develop-ment, the molecular mechanisms by which an organizedbody plan is achieved have been more elusive. Classicalembryological studies in amphibians have shown thatan important step in anteroposterior axis formation isthe polarization of prospective dorsal mesoderm, whichthen transfers this polarity by induction to overlyingectoderm during gastrulation (Spemann, 1938). Themolecular nature of the mesodermal polarity is notunderstood, but important clues are suggested byregional distribution of certain specific gene products.For example, at least two Xenopus homeoprotein

genes, Xhox3 and Xhox-36, display regional specificityin this tissue during gastrulation (Ruiz i Altaba andMelton, 1989a; Condie and Harland, 1987). Exper-imental over-expression of Xhox3 results in loss ofanterior structures (Ruiz i Altaba and Melton, 1989ft),supporting a primary role for this gene in mesodermalpatterning. Such studies have been quite valuable, butaxis formation is likely to involve a number ofadditional regulatory molecules.

Genes activated during late blastula or early gastrulaand expressed in an anterior or posterior-specificmanner represent potential candidates for such regu-lators. In this paper we describe such a gene, calledXenopus-posterior (Xpo). Xpo transcripts show aregion-specific distribution in posterior mesoderm andectoderm in gastrulating and neurulating embryos.Similar to Xhox3, Xpo expression is induced differen-tially by growth factors of the fibroblast-growth factor(FGF) and transforming growth factor ft (TGF£)families. Furthermore, the Xpo sequence includes aputative finger motif. Such elements have been impli-cated in interactions between regulatory proteins'andnucleic acids. These observations suggest a possible rolefor Xpo in regulating gene expression associated withanteroposterior axis formation in the Xenopus embryo.

748 S. M. Sato and T. D. Sargent

Materials and methods

EmbryosEggs were obtained from Xenopus laevis frogs (Xenopus I)primed with 1000 units of human chorionic gonadotropin(United States Biochemical). Eggs were fertilized artificiallyusing macerated testis, dejellied in 2% cysteine, pH7.8, for3-4min, and were placed in O.lxMMR (lxMMR is 0.1MNaCl, 2mM KC1, lmM MgSO4, 2mM CaCl2, 5mM Hepes,pH7.4, O.lmM EDTA) at 23°C. Embryos were stagedaccording to Nieuwkoop and Faber (1967).

Xpo clone isolation and sequencingA partial cDNA clone encoding the Xpo protein, designatedDG72, was isolated from a subtracted cDNA libraryrepresenting sequences that first appear during Xenopusgastrulation (Sargent and Dawid, 1983; Jamrich etal. 1985). Alonger cDNA clone of 4060 base pairs (bp) was subsequentlyobtained from a stage 17 cDNA library (Richter et al. 1988)and was sequenced completely on both strands by the chaintermination method using double-stranded DNA and aSequenase kit obtained from USB (Zagursky et al. 1985).

RNA blot analysisTotal RNA prepared from embryos was analyzed by RNA gelblotting as previously described (Sargent et al. 1986). A nick-translated, isolated restriction endonuclease fragment com-prising most of the Xpo open reading frame was used as aprobe.

In situ hybridizationEmbryos were fixed, embedded in paraffin, sectioned, andhybridized in situ with 35S RNA probes as previouslypublished (Jamrich and Sato, 1989). Exposures ranged from 1to 2 weeks.

Induction experimentsEctodermal explants dissected at stage 8 were cultured untilsibling, undissected embryos reached stage 11, 13 or 28.Culture was in 67 % Lebowitz's medium supplemented witheither bFGF (50ngmr]; Collaborative Research) or XTC-conditioned medium (XTC-CM, diluted 1:10; a gift of F.Rosa). Total RNA prepared from the explants was subjectedto RNA blot analysis as described above. Blots werehybridized with nick-translated, isolated inserts from cDNAclones encoding a-actin (Spac-9; Sargent et al. 1986) or Xpoprotein.

Results

Developmental expression of XpoXenopus-posterior (Xpo) was originally selected from asubtraction library containing genes differentially ex-pressed during gastrulation (DG72; Sargent andDawid, 1983). This gene was chosen for furthercharacterization because its transcript was found to behighly enriched in the posterior portion of early tailbudstage embryos (Fig. 1). The Xpo transcript was firstdetected by RNA gel blot analysis shortly after the mid-blastula transition (Fig. 2). This RNA persists at a lowlevel shortly after the onset of gastrulation, thenundergoes rapid accumulation, peaking in abundancesometime during neurulation. By late tailbud, Xpo

T3(00)

I28S

Fig. 1. Localization of Xpo transcripts in tailbud embryos.Stage 22-25 embryos were dissected into head, trunk, andtail regions. Total RNA (2 fig per lane) was analyzed byRNA gel blotting, using as a probe a P-labeled DNAfragment derived from the Xpo cDNA clone. RNAsamples were prepared in parallel and were equally intactas judged from the ethidium bromide staining pattern (notshown). The blot was exposed to X-ray film with anintensifier screen for 3 days at -70°C.

ft

8-9

ft

o

ft10

.5ft

T -

ft

12.5

ft

O)

(0

CMCM

ft

OCM

ft

35-3

6

(0

00CO

ft

CM U

(0 W

18S>

Fig. 2. Developmental expression of Xpo. Total RNA(2 fig per lane) from embryos of the indicated stages (st)was analyzed by RNA gel blotting as described in Fig. 1. Afaint band is visible in the stage 8-9 lane in the originalautoradiogram. Exposure to X-ray film was as for Fig. 1.

RNA has disappeared from the embryo. Thus, Xpoexpression is bimodal and transient. The rapid accumu-lation phase coincides with the period during gastru-lation in which the anteroposterior axis is beingelaborated.

Localization of Xpo by in situ hybridizationMore precise localization of Xpo transcripts wasachieved by in situ hybridization studies. At stages10-10.5, the transcript abundance is quite low, whichcomplicates in situ hybridization analysis. However,

Expression of Xenopus-posterior 749

Blastocoel

Anteriordorsal mesoderm

Posteriordorsal mesoderm

Yolk plug

B

Posteriormesoderm

Neural Plate

Anterior

Archenteron

Endoderm

Fig. 3. Localization of Xpo transcript during mid- to late gastrulation by in situ hybridization. Cross sections of embryos atstage 11 (A) or stage 12.5-13 (B) were hybridized to 35S-labeled antisense Xpo RNA probe. A schematic of the plane ofsection and hybridization pattern (cross-hatching) is shown to the right of the dark field image. In panel A the dorsal sideof the embryo is oriented to the right. Note that in gastrula endoderm, air bubbles create an apparent background in thedark field image; however, these bubbles can be distinguished clearly from silver grains under the microscope.

attempts to detect Xpo RNA by this method did notreveal any areas of noticeably higher transcript accumu-lation (data not shown), suggesting that the early phaseof Xpo expression is uniform throughout the embryo, atleast relative to later stages. By midgastrula (stage 11),Xpo RNA levels have increased substantially, and insitu hybridization at this stage revealed distinct concen-trations of Xpo RNA in the mesoderm and ectodermsurrounding the yolk plug, with the strongest signal inposterior dorsal mesoderm (Fig. 3). There may also bea small amount of Xpo RNA in endoderm, but thisapparent hybridization is not clearly distinguishablefrom nonspecific background.

At stage 13, the transcript remains predominantlylocalized in the posterior mesoderm and ectoderm ofthe embryo (Fig. 3B). In addition, there was a smallamount of transcript detected in the anterior ectodermin a region adjacent to the neural plate (pre-cementgland). At mid to late tailbud stages, near the end of theXpo expression period, in situ hybridization showedthis RNA was highly localized in mesoderm andectoderm derivatives in the extreme posterior of the tail(data not shown).

Xpo primary structureThe mRNA encoding the Xpo protein is approximately

4400 nucleotides (nt) in length, estimated from RNAgels. In order to obtain cDNA clones including acomplete open reading frame, a stage 17 Xenopusembryonic cDNA library in Agtll (Richter et al. 1988)was screened with a probe derived from the originalDG72 clone. This screen resulted in the isolation ofA72A, which has an insert size of 4060 bp (Fig. 4). DNAfrom this A clone was transferred to plasmid vectors forsequencing. Primer extension experiments (data notshown) indicated that A72A lacks approximately 200 ntof sequence from the Xpo mRNA 5' end. However, theentire Xpo protein sequence is probably represented inA72A by the open reading frame starting at position 291and ending at position 1956 (Fig. 4), since in the regionupstream from the AUG at position 291 there are atleast 22 stop codons, 10 of which are in phase with thelong open reading frame. There is at least one AUGcodon upstream from position 291, at position 84, butthis is not located in a favorable translation initiationmotif (Kozak, 1989). Thus Xpo has a predicted size of555 amino acids. Xpo mRNA has substantial stretchesof 5' and 3' untranslated sequence, the significance ofwhich is not known.

A search of the GenBank database with the Xposequence revealed only one region of substantialhomology to known proteins. Near the carboxylterminus (amino acid residues 414 to 429) is located a

750 5. M. Sato and T. D. Sargent

GAATTCCATTTTGAGAAGCCTGAGACTGACGCGTTAAAGGTCTGACCAGAAAAGAAAGCCTGGAGCCTAACACAGGTAAGTATATGTTTC 9 0

ITTTATCCAATTTTAACTGAAAGTTGTTGTATTAGATACGTTTCTGTTTTITTTTAGTGGAAAGAAGTTAAGGAACACTTTTTGTAAGGT 1 8 0

AGTAAGGTCAGTTATTGTGTAGAAGCTAAGGTATAGTTTGTTGTTGTTGCGAAGCnAGTGTTTTGTTTGTGTAAGTAATTGTTTGTAGT 2 7 0

GTGTATTGTTTGTTGTAAACATGGAATTTGTTGAAAGATATGTAAACAAATATGCCTCAGCTGATTACCATTCATGTGCAGATGAGTCTG 3 6 0M E F V E R Y V N K Y A S A D Y H S C A D E S V ( 2 4 )

T H O . F K S L L T Q C Q R Y N N K V K C N H L A K M V K K I ( 5 4 )

K M A N E I L A L L K M K V I A E H K A E Q W R N E K V Q F ( 8 4 )

R E D L E K F G E S L I V A A S T S E E H I S E L E E L R E ( 1 1 4 )

K V E L L A K Q N D L L E E K L K D C E E R C R L R E Q E V ( 1 4 4 )

I S L E S K A G Y E L N V P V S V C P V T E Q E I K D G E Q ( 1 7 4 )

G I R P Q T L P S P T L F S P Q S E C A R Q R I N N T Y V P ( 2 0 4 )

T D N N I H S N S A L K I Q E V I S L T Q I L G K F D T N L ( 2 3 4 )

S P I S L Y N K L E A V V K Q Y N L G N K D A C A L L R A W ( 2 6 4 )

L P Y Q L A A E L R P P V G K H I G T L S N I N E N W G S T ( 2 9 4 )

S E R L R E L Q R I L G G R D I R G T N A L E N A R Y R K G ( 3 2 4 )

D D P M L F C T D Y L S L Y K V V F N C P D M L P D E P N F ( 3 5 4 )

TCCTCTACTCAATGGCAAATAAATGrAATGTTGATTACAACACAAGAACTGCCCTTAGGTATGCCACCTCATACAACAACTTTATAAATA 1 4 4 0L Y S M A N K C N V D Y N T R T A L R Y A T S Y N N F I N T ( 3 8 4 )

AATGTTACAGTTGTGGTAAGTATGGTCATATTGCTCGATTTTGCAGAACTTCTGCCAATCAGCATGATACTTACCCAATCCATTCAATAC 1 6 2 0C Y S C G K Y G H I A R F C R T S A N Q H D T Y P I H S I Q . ( 4 4 4 )

R Q E G D A Q E N L N D Y L L D H S P P S E T E T V V S S D ( 4 7 4 )

H I D T G S N S A G D Q K E S Q N E P K R E F H T P P C K T ( 5 0 4 )

G K E G T T A P P F I F W V N I P L W L Y S S W S H I A M L ( 5 3 4 )

TAATGCTAATGGACAATATAGCACAGTTTGCCCAGGGTGTACCAAACACAAATATGCCTGTTTTCTAATGACAAGGAATGTGATGGAATT 1 9 8 0M L M D N I A Q F A Q G V P N T N M P V F * (555)

Fig. 4. Partial DNA sequence and deduced protein sequence of Xpo. The DNA sequence of the Xpo cDNA clone, A72Ais shown only for the existing 5' untranslated leader and the protein coding region. The remaining 2080 nt of 3'untranslated sequence has been included in the GenBank submission (Accession Number X58487). Stop codons in the 5'untranslated region are underlined, as is the 'CCHC motif. Nucleotide and amino acid (parentheses) residue numbers aregiven at right.

16-amino acid sequence that is identical at as many as 12 1989). This sequence, referred to as the 'CCHC motif,positions to a motif found in one or two copies in is similar to zinc-binding finger domains of manyretroviral nucleocapsid protein (Fig 5; Green and Berg, transcriptional regulatory proteins. However, while

XpoHIV

CopiaCNBP con

RKCYSCGKYGHIARFCRTVKCFNCGKEGHIARNCRAVKCHHCGREGHIKKDCFHXXCYXCGXXGHXAXXCXX

Fig. 5. Comparison of CCHC motifs from Xpo and otherproteins. Identical residues are shaded. References forsequences are as follows: HIV (Human ImmunodeficiencyVirus), Ratner et al. 1985; Copia, Mount and Rubin, 1985;CNBP con (consensus of 7 copies in Cellular Nucleic AcidBinding Protein), Rajavashisth et al. 1989.

CCHC peptides bind zinc in vitro (Green and Berg,1989), this element is essentially absent from retroviralparticles, suggesting that either the CCHC motif doesnot bind zinc in vivo, or binds zinc transiently (Katz andJentoft, 1989). Similar sequences have also been foundin transposable elements such as copia (Mount andRubin, 1985) and Txl and Tx2 in Xenopus (Garrett etal. 1989) and in CNBP, a protein that binds to the sterolregulatory element present in the HMG CoA reductaseand LDL receptor promoters (Rajavashisth etal. 1989).

Induction of Xpo by growth factorsSeveral laboratories have shown that exposure topolypeptide growth factors of the FGF and TGF/3families results in the conversion of competent ecto-derm to endoderm and mesoderm. FGF-type inducerstend to elicit tissues with ventral and posteriorcharacteristics while TGF/3-types induce more dorsaland anterior differentiation (Smith, 1987; Slack et al.1987; Kimelman and Kirschner, 1987; Rosa et al. 1988;Rosa, 1989; Green et al. 1990; Smith et al. 1990; Vanden Eijnden-Van Raaij et al. 1990; Asashima et al.1990). To examine the effects of these factors on Xpoexpression, ectodermal explants dissected at stage 8

Expression of Xenopus-posterior 751

were exposed to either bFGF or XTC-CM and thenmonitored for Xpo or cr-actin expression at stages 11,13and 28 (Fig. 6). In agreement with earlier work (Greenet al. 1990; Kimelman and Kirschner, 1987; Rosa et al.1988), a--actin RNA was not detectable at stage 11. Atstage 13, only XTC-CM-treated ectoderm containedmeasurable a--actin RNA, but by stage 28, both growthfactor treatments had resulted in accumulation ofsignificant amounts of this RNA. Xpo RNA accumu-lation was also increased by both types of growth factor,but this response contrasted with that of a'-actin inseveral ways. First, there is a small but significantbackground level of Xpo RNA accumulation inuntreated ectodermal explants, possibly correspondingto the local accumulation of this RNA in anteriorectoderm that occurs during normal development(Fig. 3B). Second, a significant increase in Xpoexpression was evident as early as stage 11, prior to theappearance of any a'-actin RNA. Third, FGF is at leastas effective as XTC-CM in the induction of Xpoexpression. In fact, when compared with the degree ofa'-actin RNA accumulation, FGF is significantly moreactive than XTC-CM in the induction of XPO. Finally,in contrast to a'-actin, which is mesoderm-specific, XpoRNA appears in both ectodermal and mesodermaltissues during normal development. Thus, the increasein Xpo expression elicited by growth factors presum-ably results from posteriorization as opposed tomesoderm induction per se.

Discussion

In amphibian development, axis formation commenceswith a cytoplasmic rotation occurring between fertiliz-ation and first cleavage (Gerhart et al. 1984). Bycompletely unknown mechanisms this rearrangement

a b e d a b e d a b e d

Stage 11 Stage 13 Stage 28

Fig. 6. Induction of Xpo by treatment of ectodermal explants with XTC-CM or bFGF. Explants dissected from stage 8blastulae were incubated in various conditions until sibling controls reached stages 11, 13 or 28 as indicated. Incubationconditions were as follows: lane b, no treatment; lane c, bFGF at SOngml"1; lane d, XTC-CM (1:10 dilution). Total RNAprepared from whole embryo siblings (lane a, 1 embryo equivalent) or explants (lanes b-d, 10 explants) was subjected togel blot analysis. The blot was hybridized to a mixture of nick-translated -"P-labelled Xpo and a'-actin probes. Arrowsindicate the position of the RNA for Xpo, cytoskeletal actin (CA), a'-actin (muscle actin, MA), and the 28S and 18Sribosomal RNAs.

752 5. M. Sato and T. D. Sargent

leads to the establishment of a dorsalizing center in thevegetal hemisphere, which in turn appears to induce theformation of the 'organizer' region of the dorsalmarginal zone (Dale and Slack, 1987; Gimlich andGerhart, 1984). During gastrulation, the organizertissue comprises dorsal mesoderm, predominantlyfuture notochord, and involutes extensively along theblastocoel roof. Classical dissection and transplantationexperiments revealed that this involuted dorsal meso-derm exhibits an anterior-posterior polarity, with themost extensively involuted tissue adopting the mostanterior character (Spemann, 1938). Furthermore,dorsal mesoderm can impose this polarity by inductionon adjacent gastrula ectoderm. This has been demon-strated recently by the use of molecular markersspecific to anterior and posterior neural plate deriva-tives such as cement gland (Sive et al. 1989), brain andspinal cord (Sharpe and Gurdon, 1990). Thus, amphib-ian axis formation can be broken down into five steps:(1) dorsoventral polarization of the fertilized egg, (2)formation of the organizer, (3) induction of dorsalmesoderm, (4) anteroposterior polarization of dorsalmesoderm, and (5) inductive transfer of this polarity toadjacent tissues.

The molecular mechanisms of these processes remainlargely a matter of speculation. However, it has becomereasonably clear that important roles are played inmesoderm induction by at least two different categoriesof polypeptide growth factors, the FGF and TGF/3families, which tend to induce posterior and anteriormesoderm, respectively. Recently, Ruiz i Altaba andMelton (1989c; 1990) have proposed that in the cleavingembryo, the graded distribution of TGF/3- or FGF-related polypeptides (or their cognate receptors),coupled to the differential expression of certainhomeoprotein genes, could be the molecular basis ofembryonic axes. In this model, homeoprotein genes areregulated by the growth factor gradients, leadingdirectly to tissue specification. The prototypes for suchan arrangement are the homeoprotein genes Xhox3 andXlHboxl, the former displays posterior-specific ex-pression in dorsal mesoderm, and is activated inectoderm explants to a significantly higher extent bybFGF than by growth factors of the TGF/3 family (Ruizi Altaba and Melton, 1989c; 1990), and the latter isexpressed anteriorly and exhibits a reciprocal responseto the two growth factor classes (Cho and De Robertis,1990).

There is a certain element of circularity to thisargument, i.e., one would anticipate that genesexpressed predominantly in posterior tissues would bepreferentially activated by bFGF, which elicits pos-terior morphology in ectodermal induction exper-iments. However, in the case of these two homeopro-tein genes, there is reasonably convincing evidence thatexpression of either gene is in itself sufficient to specifya posterior or anterior phenotype, as appropriate (Ruizi Altaba and Melton, 19896; Wright et al. 1989). Thissupports causative rather than passive roles for Xhox3and XlHboxl in axis formation.

At present, we cannot attribute such an active role to

Xpo. Nevertheless, we believe that Xpo has someinvolvement in the determination of the embryonic axesfor several reasons. First, Xpo expression begins veryearly in development, close to the midblastula tran-sition, and is strongly concentrated in the posteriorregion of dorsal mesoderm by early to mid-gastrula.Second, Xpo expression is quite transient, which seemsmore consistent with a regulatory as opposed to astructural function. Third, like Xhox3, Xpo expressionis significantly stimulated by bFGF and to a lesserextent by XTC-CM. Finally, the presence in the Xpopolypeptide sequence of a CCHC motif implies that thisprotein may interact with DNA or RNA. The CCHCmotif in nucleocapsid protein has been shown to beessential for normal retroviral assembly; if the cysteineor histidine residues are eliminated, the resulting virusis unable to specifically package viral RNA (Fu et al.1988; Gorelick et al. 1988, 1990; Meric and Goff, 1989).Although a direct role for the CCHC motifs found inCNBP has not been explicitly demonstrated, CNBPdoes interact with nucleic acids, and the CCHCdomains are likely to be important in this function aswell.

Elucidation of gene function in amphibian embryos isnot a trivial undertaking. However, it should bepossible to approach this problem in the case of Xpo bymaking use of the methods that have been successfullyemployed to investigate the developmental function ofother Xenopus proteins, such as introduction into theembryo of mRNA, specific antisera or constructionsproducing antisense RNA (e.g. Wright etal. 1989; Ruizi Altaba and Melton, 19896; Giebelhaus et al. 1988).

References

ASASHIMA, M., NAKANO, H. , SH1MADA, K., KlNOSHlTA, K., ISHII,K., SHIBAI, H. AND UENO, N. (1990). Mesodermal induction inearly amphibian embryos by Activin A (erythroid differentiationfactor). Roux's Arch, devl Biol. 198, 330-335.

CHO, K. W. Y. AND D E ROBERTIS, E. M. (1990). Differentialactivation of Xenopus homeo box genes by mesoderm-inducinggrowth factors and retinoic acid. Genes & Devi. 4, 1910-1916.

CONDIE, B. G. AND HARLAND, R. M. (1987). Posterior expressionof a homeobox gene in early Xenopus embryos. Development101, 93-105.

DALE, L. AND SLACK, J. M. W. (1987). Regional specificationwithin the mesoderm of early embryos of Xenopus laevis.Development 100, 279-295.

Fu, X. D., KATZ, R. A., SKALKA, A. M. AND LEIS, J. (1988). Site-directed mutagenesis of the avian retrovirus nucleocapsidprotein, pp 12. Mutation which affects RNA binding in vitroblocks viral replication. J. biol. Chem. 263, 2140-2145.

GARRETT, J. E., KNUTZON, D. S. AND CARROLL, D. (1989).Composite transposable elements in the Xenopus laevis genome.Molec. cell. Biol. 9, 3018-3027.

GERHART, J. C , VINCENT, J.-P., SCHARF, S. R., BLACX, S. D.,GIMLICH, R. L. AND DANILCHIK, M. (1984). Localization andinduction in early development of Xenopus. Phil. Trans. Roy.Soc. Lond. B 307, 319-330.

GIEBELHAUS, D. H., EIB, D. W. AND MOON, R. T. (1988).Antisense RNA inhibits expression of membrane skeletonprotein 4.1 during embryonic development of Xenopus. Cell 53,601-615.

GIMLICH, R. L. AND GERHART, J. C. (1984). Early cellularinteractions promote embryonic axis formation in Xenopuslaevis. Devi Biol. 104, 117-130.

Expression of Xenopus-posterior 753

GOREUCK, R. J., HENDERSON, L. E., HANSER, J. P. AND REIN, A.(1988). Point mutants of Moloney murine leukemia virus thatfail to package viral RNA: Evidence for specific RNArecognition by a 'zinc finger-like' protein sequence. Proc. natn.Acad. Sci. U.S.A. 85, 8420-8424.

GOREUCK, R. J., NIGIDA, S. M. JR. , BESS, J. W. JR, ARTHUR, L.

O., HENDERSON, L. E. AND REIN, A. (1990). Noninfectioushuman immunodeficiency virus type 1 mutants deficient ingenomic RNA. J. Virol. 64, 3207-3211.

GREEN, L. M. AND BERG, J. M. (1989) A retroviral Cys-Xaa2-Cys-Xaa4-His-Xaa4-Cys peptide binds metal ions: Spectroscopicstudies and a proposed three-dimensional structure. Proc. natn.Acad. Sci. U.S.A. 86, 4047-4051.

GREEN, J. B. A., HOWES, G., SYMES, K., COOKE, J. AND SMITH, J.

C. (1990). The biological effects of XTC-MIF: quantitativecomparison with Xenopus bFGF. Development 108, 173-183.

JAMRICH, M., SARGENT, T. D. AND DAWID, I. B. (1985). Altered

morphogenesis and its effects of gene activity in Xenopus laevisembryos. Cold Spring Harbor Symp. Quart. Biol. L 31-35.

JAMRICH, M. AND SATO, S. M. (1989). Differential gene expressionin the anterior neural plate during gastrulation of Xenopus laevisembryos. Development 105, 779-786.

KATZ, R. A. AND JENTOFT, J. E. (1989). What is the role of theCvs-His motif in retroviral nucleocapsid (NC) proteins?BioEssays 11, 176-181.

KIMELMAN, D. AND KIRSCHNER, M. (1987). Synergistic induction ofmesoderm by FGF and TGF/S and the identification of anmRNA coding for FGF in the early Xenopus embryo. Cell 51,869-877.

KOZAK, M. (1989). The scanning model for translation: an update./. Cell Biol. 108, 229-241.

MERIC, C. AND GOFF, S. P. (1989). Characterization of Moloneymurine leukemia virus mutants with single-amino-acidsubstitutions in the Cys-His box of the nucleocapsid protein. /.Virol. 63, 1558-1568.

MOUNT, S. M. AND RUBIN, G. M. (1985). Complete nucleotidesequence of the Drosophila transposable element copia:homology between copia and retroviral proteins. Molec. cell.Biol. 5, 1630-1638.

NIUEWKOOP, P. D. AND FABER, J. (1967). Normal Table ofXenopus laevis (Daudin). North-Holland, Amsterdam.

NUSSLEIN-VOLHARD, C , FROHNHOFER, H . G . AND LEHMANN, R.

(1987). Determination of Anteroposterior Polarity inDrosophila. Science 238, 1675-1681.

RAJAVASHISTH, T. B., TAYLOR, A. K., ANDAUBI, A., SVENSON, K.

L. AND Lusis, A. J. (1989). Identification of a zinc fingerprotein that binds to the sterol regulatory element. Science 245,640-643.

RATNER, L., HASELTINE, W., PATARDA, R., LIVAK, K. J., STARCICH,

B., JOSEPHS, S. J., DORAN, E. R., RAFALSKI, J. A., WHITEHORN,E. A., BAUMEISTER, K., IVANOFF, L., PETTEWAY, S. R. JR,

PEARSON, M. L., LAUTENBERGER, J. A., PAPAS, T. S., GHRAYEB,

J., CHANG, N. T., GALLO, R. C. AND WONG-STAAL, F. (1985).Complete nucleotide sequence of the AIDS virus, HTLV-IILNature 313, 277-284.

RICHTER, K., GRUNZ, H. AND DAWID, I. B. (1988). Gene

expression in the nervous system of Xenopus laevis. Proc. natn.Acad. Sci. U.S.A. 85, 8086-8090.

ROSA, F., ROBERTS, A. B., DANIELPOUR, D., DART, L. L., SPORN,

M. B. AND DAWID, I. B. (1988). Mesoderm induction in

amphibians: the role of TGF-b2-like factors. Science 239,783-785.

ROSA, F. M. (1989). Mix.l, a homeobox mRNA inducible bymesoderm inducers, is expressed mostly in the presumptiveendodermal cells of Xenopus embryos. Cell 57, 965-974.

Ruiz i ALTABA, A. AND MELTON, D. A. (1989a). Bimodal andgTaded expression of the Xenopus homeobox gene Xhox3 duringembryonic development. Development 106, 173-183.

Ruiz I ALTABA, A. AND MELTON, D. A. (19896). Involvement ofthe Xenopus homeobox gene Xhox3 in pattern formation alongthe anterior-posterior axis. Cell 57, 317-326.

Ruiz I ALTABA, A. AND MELTON, D. A. (1989c). Interactionbetween peptide growth factors and homeobox genes in theestablishment of antero-postenor polarity in frog embryos.Nature 341, 33-38.

Ruiz i ALTABA, A. AND MELTON, D. A. (1990). Axial patterningand the establishment of polarity in the frog embryo. Trends inGenetics 6, 57-64.

SARGENT, T. D., JAMRICH, M. AND DAWTD, I. B. (1986). Cell

interactions and the control of gene activity during earlydevelopment of Xenopus laevis. Devi Biol. 114, 238-246.

SARGENT, T. D. AND DAWID, I. B. (1983). Differential geneexpression in the gastrula of Xenopus laevis. Science 222,135-139.

SHARPE, C. R. AND GURDON, J. B. (1990). The induction ofanterior and posterior neural genes in Xenopus laevis.Development 109, 765-774.

SIVE, H. L., HATTORI, K. AND WEINTRAUB, H. (1989). Progressive

determination during formation of the anteroposterior axis inXenopus laevis. Cell 58, 171-180.

SLACK, J. M. W., DARLINGTON, B. G., HEATH, J. K. AND

GODSAVE, S. F. (1987). Mesoderm induction in early Xenopusembryos by heparin-binding growth factors. Nature 326,197-200.

SMITH, J. C , PRICE, B. M., VAN NMMEN, K. AND HUYLEBROECK,

D. (1990). Identification of a potent Xenopus mesoderm-inducing factor as a homologue of activin A. Nature 345, 729-731.

SMITH, J. C. (1987). A mesoderm-inducing factor is produced by aXenopus cell line. Development 99, 3-14.

SPEMANN, H. (1983). Embryonic Development and Induction. NewHaven: Yale University Press.

• STRUHL, G., STRUHL, K. AND MACDONALD, P. M. (1989). The

gradient morphogen bicoid is a concentration-dependenttranscriptional activator. Cell 57, 1259-1273.

TURING, A. (1952). The chemical basis of morphogenesis. Phil.Trans. Roy. Soc. Land. B 237, 37-72.

VAN DEN EUNDEN-VAN RAALT, A. J., VAN ZOELENT, E. J., VAN

NIMMEN, K., KOSTER, C. H., SNOEK, G. T., DURSTON, A. J. AND

HUYLEBROECK, D. (1990). Activin-hke factor from a Xenopuslaevis cell line responsible for mesoderm induction. Nature 345,732-734.

WRIGHT, C. V. E., CHO, K. W. Y., HARDWICKE, J., COLLINS, R.

H. AND D E ROBERTIS, E. M. (1989). Interference with functionof a homeobox gene in Xenopus embryos producesmalformations of the anterior spinal cord. Cell 59, 81-93.

ZAGURSKY, R. S., BAUMEISTER, K., LOMAX, N. AND BERMAN, M.

L. (1985). Rapid and easy sequencing of large linear double-stranded DNA and super-coiled plasmid DNA. Gene Anal.Technol. 2, 89-94.

{Accepted 4 March 1991)