Role for cGATA-5 in Transcriptional Regulation of the Embryonic Chicken Pepsinogen Gene by Epithelial–Mesenchymal Interactions in the Developing Chicken Stomach

Developmental Biology 223, 103–113 (2000) doi:10.1006/dbio.2000.9731, available online at http://www.idealibrary.com on

Role for cGATA-5 in Transcriptional Regulation of the Embryonic Chicken Pepsinogen Gene by Epithelial–Mesenchymal Interactions in the Developing Chicken Stomach Nobuyuki Sakamoto, 1,2 Kimiko Fukuda, 1 Kumiko Watanuki, Daisuke Sakai, Teruya Komano, Paul J. Scotting,* and Sadao Yasugi 3 Department of Biological Sciences, The Graduate School of Science, Tokyo Metropolitan University, 1-1 Minamiohsawa, Hachioji, Tokyo 192-0397, Japan; and *Institute of Genetics, University of Nottingham Medical School, Queen’s Medical Centre, Nottingham NG7 2UH, United Kingdom

A gene encoding embryonic chicken pepsinogen (ECPg), a zymogen of the digestive enzyme pepsin, is expressed specifically in epithelial cells of glands of embryonic stage proventriculus (glandular stomach) under the influence of mesenchyme. We found four GATA and one Sox binding motifs in 1.1 kb of the 5ⴕ flanking region of the ECPg gene which are essential to the organ-specific expression of the gene. The expression of cGATA-5 and cSox2 in the proventriculus from day 6 to day 12 of incubation was therefore analyzed. cGATA-5 was more strongly expressed in glandular epithelial cells than in luminal epithelial cells, while cSox2 gene expression was weaker in glandular epithelial cells. Using heterologous recombination explants we also discovered that the expression of cGATA-5 and cSox2 in epithelial cells was affected by mesenchyme when the latter induced ECPg gene expression in epithelial cells. Introduction of expression constructs into epithelial cells by electroporation demonstrated that cGATA-5 upregulated transcription of a reporter luciferase gene via a cis element in the 5ⴕ flanking region of the ECPg gene. The gel mobility shift assay revealed that the cGATA-5 protein specifically binds to the GATA binding sites. cSox2 downregulated the activity of luciferase but it was not through the Sox binding motif. These results suggest that cGATA-5 positively regulates transcription of the ECPg gene and is involved in spatial regulation of the pepsinogen gene during development. © 2000 Academic Press Key Words: GATA; pepsinogen gene; gene expression; epithelial–mesenchymal interaction.

INTRODUCTION Epithelial–mesenchymal interactions play important roles in the development of many organ systems including the gut. In the early development of the chicken embryo, the digestive tract is a simple double-layered tube consisting of epithelium derived from endoderm and mesenchyme derived from mesoderm. The epithelium gradually acquires organ-specific functions and morphological features. Thus on day 6 of incubation, proventricular (glandular stomach) 1

These two authors made an equal contribution to the paper. Present address: The Biohistory Research Hall, 1-1 Murasakicho, Takatsuki, Osaka 569-1125, Japan. 3 To whom correspondence should be addressed. Fax: ⫹81 (426) 77-2559. E-mail: [email protected]. 2

0012-1606/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.

epithelial cells invaginate into the mesenchyme and begin to form glands and, in 9-day embryos, glandular epithelial cells begin to express prochymosin-type pepsinogen, ECPg (embryonic chicken pepsinogen; Hayashi et al., 1988a,c). The epithelium that does not invaginate is called luminal epithelium. Studies on epithelial–mesenchymal interactions in gut development have revealed that the formation of glands and expression of ECPg are controlled by the surrounding mesenchyme (for review, see Yasugi 1994, 1995; Yasugi and Fukuda, 2000): ECPg expression can be induced by the proventricular mesenchyme in the gizzard (muscular stomach) and esophageal epithelia which never express ECPg in normal development (Takiguchi et al., 1986; Urase et al., 1996). Interestingly, pulmonary mesenchyme exhibits even stronger gland and ECPg-inducing

103

104

Sakamoto et al.

activity than was found in proventricular mesenchyme (Urase et al., 1996). On the other hand, gizzard and esophageal mesenchymes exhibit a strong inhibitory influence on gland formation and ECPg expression in the epithelium and have no inductive activity (Urase and Yasugi, 1993). Recent studies identified genes expressed in epithelial cells such as chicken spasmolytic polypeptide (cSP; a marker of luminal epithelium; Tabata and Yasugi, 1998), CdxA (Ishii et al., 1997), cSox2 (Ishii et al., 1998), and Sonic hedgehog (Shh) (Narita et al., 1998) and demonstrated that their expression patterns are also affected by the underlying mesenchyme. Elucidation of the mechanisms of transcriptional regulation of these genes will lead to understanding of the molecular nature of epithelial–mesenchymal interactions during development. Studies on the mechanisms of transcriptional regulation of vertebrate pepsinogen genes have shown that hypomethylation of pepsinogen genes occurs concurrently with the activation of their transcription (Ichinose et al., 1988a,b; Fukuda et al., 1994a; Pals et al., 1995). cDNAs for various vertebrate pepsinogen mRNAs have been isolated (Sogawa et al., 1981; Ichihara et al., 1986; Evers et al., 1988; Hayashi et al., 1988a; Kageyama et al., 1990; Sakamoto et al., 1998), and genomic clones of pepsinogens have also been cloned, mainly in mammals (Sogawa et al., 1983; Hayano et al., 1988; Ishihara et al., 1989) and in chicken (Hayashi et al., 1988b; Sakamoto et al., 1998). Although it has been revealed that 1.1 kb of the 5⬘ flanking region of the ECPg gene is sufficient to drive reporter gene expression in the same patterns as in the epithelial–mesenchymal recombination experiments (Fukuda et al., 1994b), a transcription factor(s) involved in the regulation of vertebrate pepsinogen genes, including ECPg, is still unknown. Recently, information about transcription factors expressed in the gut has increased. GATA-4/5/6 are expressed in the gut of several vertebrate species. In the rat, GATA-4 and GATA-6 (named GATA-GT2 and GATA-GT1, respectively) activate the transcription of a reporter gene downstream of the control region of H ⫹/K ⫹-ATPase ␣ or ␤ subunit genes (Mushiake et al., 1994; Nishi et al., 1997). Gene disruption experiments in the mouse revealed that GATA-4 is required for ventral morphogenesis such as foregut formation (Kuo et al., 1997; Molkentin et al., 1997; Narita et al., 1997). Xenopus GATA-4/5/6 transactivate the promoter of the xIFABP gene (Gao et al., 1998). In Caenorhabditis elegans, it has been suggested that gut-specific expression of the ges-1 gene is regulated by the GATA transcription factor (Egan et al., 1995). Therefore, GATA factors regulate gut-specific genes in both invertebrates and vertebrates. In addition, the GATA family is necessary to determine and/or maintain the endodermal lineage in the worm (Zhu et al., 1997) and mouse (Soudais et al., 1995). Chicken cSox2 is a member of the transcription factor family containing an Sry box as their DNA binding domain and is expressed in anterior gut endoderm from early stages of development. Its expression is downregulated in regions where epithelial morphogenesis occurs, such as proven-

tricular glands, and was influenced by the effect of proventricular mesenchyme (Ishii et al., 1998). cSox2 is also expressed in the lens where its target appears to be a crystallin gene. cSox2 can upregulate transcription of a reporter gene ligated downstream of the promoter of the ␦1-crystallin gene (Kamachi et al., 1995, 1998). In the gut, however, its targets are still unclear. In this paper, we observed the expression patterns of cGATA-5 and cSox2, which are candidates for transcriptional regulators of the ECPg gene, using double-color fluorescence in situ hybridization and revealed a correlation between expressions of these transcription factors and the ECPg gene. Next, to elucidate the mechanisms of transcriptional regulation of the ECPg gene, we established a gene transfection method using electroporation and showed that cGATA-5 can regulate ECPg transcription. Moreover, a gel mobility shift assay using recombinant cGATA-5 protein clearly demonstrated that the protein can bind to GATA binding sites in the ECPg promoter. Although cSox2 was able to inhibit ECPg expression, this was independent of the Sox binding motif found upstream of the ECPg gene and it is not clear therefore whether this is direct transcriptional regulation.

MATERIALS AND METHODS Animals Embryos of White Leghorn chicken (Gallus gallus domesticus) were used throughout the experiments.

Cloning and Sequencing of cDNAs and Genes To obtain cGATA-5 cDNA, a 14-day posthatching chicken proventricular cDNA library was screened with PCR product from RT-PCR using mRNA from 7-day posthatching chicken proventriculus. The 5⬘ and 3⬘ untranslated regions of cGATA-5 cDNA were subcloned into pBluescript KS(⫹) (Stratagene) as templates for transcription of RNA probe.

Chromogenic and Double-Color Fluorecence in Situ Hybridization Clones with inserts of 5⬘ and 3⬘ untranslated regions of the cGATA-5 cDNA were used to prepare digoxigenin-labeled or fluorescein-labeled antisense riboprobes. For detection of cSox2 and ECPg, cDNA clones described by Ishii et al. (1998) and Hayashi et al. (1988a), respectively, were used. As a specific marker of luminal epithelial cells, we used the cSP cDNA reported earlier (Tabata and Yasugi, 1998). Chromogenic in situ hybridization on frozen sections and double-color fluorescence in situ hybridization were carried out according to the method described (Ishii et al., 1998) with modifications. Tissues were fixed in 4% paraformaldehyde in phosphatebuffered saline (PBS) overnight at 4°C and embedded in OCT compound. Frozen sections of 10 or 12 ␮m were cut in a cryostat and immediately dried. After rehydration in PBS containing 0.1% Tween 20 (PBT), the sections were digested with 1 mg/ml proteinase K in PBT at 37°C for 8 min and postfixed in 4% paraformalde-

Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved.

105

cGATA-5 Regulates Pepsinogen Gene Expression

hyde in PBS at room temperature for 20 min. Hybridization was performed overnight at 70°C in hybridization solution (50% formamide, 5⫻ SSC (pH 4.5), 1% sodium dodecyl sulfate (SDS), 50 mg/ml yeast RNA, 50 mg/ml heparin) containing digoxigeninlabeled or fluorescein-labeled riboprobes. The slides were washed twice with 50% formamide, 5⫻ SSC (pH 4.5), 1% SDS for 30 min at 65°C and three times with 50% formamide, 2⫻ SSC (pH 4.5) for 30 min at 65°C. After three washes with Tris-buffered saline containing 0.1% Tween 20 (TBST) at room temperature for 5 min, the slides were incubated in blocking solution (0.5% blocking reagent (Boehringer Mannheim) in TBST) at room temperature for 60 min and then overlaid with 375 U/ml of anti-digoxigenin Fab–peroxidase-conjugated antibody (Boehringer Mannheim) or anti-fluorescein Fab–alkaline phosphatase-conjugated antibody (Boehringer Mannheim) in blocking solution and kept at 4°C overnight. After three washes with TBST containing 2 mM levamisole for 20 min and two with 0.1 M Tris–HCl (pH 7.5), 0.15 M NaCl, 0.05% Tween 20 (TNT) for 5 min with agitation, transcripts of the first gene were detected by treatment with 100 ml of BT solution (biotynyl tyramid (NEL Life Science Product, Inc.) diluted to 1/50 with amplification diluent) for 10 –30 min at room temperature. After three washes with TNT, the slides were overlaid with 100 ␮l of streptavidin–fluorescein (SA-Flu) solution diluted to 1/200 with 0.1 M Tris–HCl (pH 7.5), 0.15 M NaCl, 0.5% blocking reagent (NEL Life Science Product, Inc.) and placed at room temperature for 30 min. To detect transcripts of the second gene, the slides were washed three times with TNT at room temperature for 5 min and then overlaid with 100 ml of HNPP/Fast red mixture (0.1 M Tris–HCl (pH 8.0), 0.1 M NaCl, 10 mM MgCl 2 containing 250 mg/ml Fast red, 100 mg/ml HNPP (Boehringer Mannheim)) and incubated at room temperature for 30 min. This reaction was repeated until signal became strong enough. To stop the reaction, slides were washed three times with TNT at room temperature for 5 min. The signals of SA-Flu conjugate and HNPP/Fast red were detected by fluorescence microscopy using filter sets for fluorescein and rhodamine, respectively.

Recombination and Culture of Esophageal Epithelium and Lung Mesenchyme The esophagus and lung were dissected from 5.5- to 6.0-day chicken embryos in Tyrode’s solution. To separate epithelium from mesenchyme, tissues were incubated in Tyrode’s solution containing 0.03% collagenase at 37°C for 60 min. After digestion, tissues were washed twice with Tyrode’s solution containing 10% fetal bovine serum and epithelium was removed from mesenchyme by using fine forceps. Two lung mesenchymes were laid together on a Nuclepore filter (pore size 0.8 ␮m) and several fragments of esophageal epithelium were placed onto them. The filter was rested on a stainless grid, put into a 24-well culture dish, and cultured at the medium– gas interphase for 6 days in 5% CO 2 and 95% air at 37°C. The medium was 199 with Earle’s salt containing an equal volume of heat-treated embryo extract (Koike and Yasugi, 1999; Takiguchi et al., 1988). The medium was changed every other day. We analyzed 12 explants.

Construction of Reporter Plasmids and Effector Plasmids A genomic fragment from the start codon to a PstI site about 1.1 kb upstream of the transcription start site of the ECPg gene was inserted into a KpnI- and NcoI-cut PGV-P2 luciferase expression

vector (Toyo Inki) and named Pm1k. The deleted plasmids Hm800, Am400, and Km120 were constructed by removing the sequences from PstI to HinfI, to ApaI, and to KpnI, respectively, from Pm1k (Fig. 6). -SS230 is a reporter plasmid with nearly 230 bp deleted between two SmaI sites from Pm1k, and Hm-SS230 is deleted from PstI to HinfI of -SS230. The coding region of cGATA-5 or cSox2 was inserted in pEGFP-N1 and pEGFP-C1 vectors, respectively. When PCR was used during construction, the fragments amplified by PCR were always confirmed by sequencing. pEGFP-N1 (Clontech Laboratories, Inc.) was used as a control GFP-expressing plasmid. The ␤-galactosidase gene from pCH110 (Pharmacia LKB Biotechnology) was subcloned into pME18S (gift from Dr. Maruyama) and used as an internal control.

Transfection of Plasmids into Gut Epithelium by Electroporation The apparatus used for electroporation of plasmids into epithelium was set up as follows (Fig. 5A). Platinum electrodes were fixed on a glass dish. A chamber (7 mm in height, 8 mm in width, and 5 mm in length) was formed with electrodes and resin walls. A vessel made of 1% agarose/PBS gel was put into an electrode chamber and filled with about 10 ␮l of PBS containing DNA. Outside of the gel vessel was filled with PBS. The proventriculus and gizzard were dissected from 5.5- to 6.0-day chicken embryos in Tyrode’s solution. They were cut open and put into the vessel with their epithelial side toward the cathode. The vessel was filled with PBS containing the luciferase reporter plasmid and the ␤-galactosidase reference reporter plasmid or, in addition to this, cGATA-5, cSox2, or GFP expression vector DNA. The concentration of plasmid DNA was 0.2 nmol/ml. For the maximum efficiency of transfection, five 50-ms pulses of 30 V were generated by electroporator T820 (BTX), and tissues were immediately washed with Tyrode’s solution containing 10% fetal bovine serum. Two transfected proventriculi or gizzards were laid on a Nuclepore filter (pore size 0.8 ␮m) and cultured for 4 days as described for tissue recombination explants.

Measurement of Luciferase Activity Explants were washed twice with PBS, frozen in 20 ␮l of 1⫻ Passive Lysis Buffer (Promega) at ⫺80°C, and immediately sonicated several times until tissues were completely dissociated. The lysate was centrifuged at 13,000 rpm for 1 min. The supernatant was stored at ⫺80°C. The samples were diluted 1/50 with 1⫻ Passive Lysis Buffer and assayed for their luciferase and ␤-galactosidase activities. For the luciferase assay, 5 ␮l of sample was added to 50 ␮l of luminescence substrate solution (Toyo Inki) and luminescence was measured with a scintillation counter. For the ␤-galactosidase assay, 5 ␮l of sample was added to 50 ␮l of 50 mM sodium phosphate buffer (pH 7.2), 1.5 mM MgCl 2, 1 mg/ml p-nitrophenyl ␤-D-galactopyranoside and incubated at 37°C for 90 min. Twenty-five microliters of 0.5 M Na 2CO 3 was added to the sample and immediately mixed to stop the reaction. The activities were measured by absorption at 405 nm with a spectrophotometer. The luciferase activity of each sample was normalized against ␤-galactosidase activity.

Protein Preparation and Gel Mobility Shift Assay The coding region of cGATA-5 cDNA was cloned into the pET expression plasmid containing a His-tag sequence (pET15b; Nova-

Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved.

106

Sakamoto et al.

FIG. 1. Diagram of cis elements in 1.1 kb of 5⬘ upstream sequences of the ECPg gene (Sakamoto et al., 1998). Putative transcription factor-binding sites in 5⬘ upstream sequences are indicated as follows: GATA, filled ellipse; Sox, open ellipse. P, PstI; H, HinfI; A, ApaI; S, SmaI; K, KpnI. gs1 to gs4 indicate probes used in the gel mobility shift assay.

gen). This construct was introduced into Escherichia coli and expression was induced by the addition of IPTG. His-tagged cGATA-5 protein was purified from the soluble fraction of cells using Ni 2⫹-charged resin. DNA fragments containing one of four consensus GATA binding sites (gs) in the 5⬘-upstream region of the ECPg gene were isolated using PCR (gs1, -794 to -693; gs2, -716 to -613; gs3, -637 to -493; gs4, -517 to -383; Fig. 1) and end-labeled. Each probe (0.2 ␮g) was incubated with recombinant cGATA-5 protein (3 ng) in a total volume of 25 ␮l containing 10 mM Tris–HCl (pH 7.4), 5 mM MgCl 2, 1 mM EDTA, 6.25% glycerol, 1 mM DTT for 30 min at room temperature. Unlabeled DNA (competitor) or salmon sperm DNA (noncompetitor) was added at 400-fold excess to the mixture to test for specificity of DNA binding. The DNA–protein complex was separated by electrophoresis on 5% native polyacrylamide gel in TBE and an autoradiogram was made.

RESULTS Expression Patterns of Genes during Proventricular Gland Formation GATA and Sox transcription factors are candidates for transcriptional regulators of the ECPg gene because of the existence of putative binding sequences in the 5⬘ flanking region of the ECPg gene (Fig. 1; Sakamoto et al., 1998). Previous studies have shown that cSox2, a member of the Sox transcription factor family, is expressed in the anterior gut epithelium of the chicken embryo, and the expression of this gene is complementary to that of the ECPg gene in the proventricular epithelium (Ishii et al., 1998). It is also known that cGATA-5, a member of the GATA transcription factor family, is expressed in chicken gut and heart (Morrisey et al., 1997). However, the detailed expression pattern of cGATA-5 during gland formation in the chicken embryonic proventriculus has not been studied. Accordingly, to compare the expression patterns of ECPg, cSP, cGATA-5, and cSox2 genes, proventriculi from 6- to 10-day chicken embryos were examined in serial sections by in situ hybridization (Fig. 2). Expression of the ECPg gene begins in 9-day embryos, 2 or 3 days after the onset of gland formation (Fig. 2D). As described above, this expression is restricted to glandular epithelial cells of the proventriculus

during development (Figs. 2D and 2E). Expression of the cSP, cGATA-5, and cSox2 genes differed between glandular epithelium and luminal epithelium from the beginning of gland formation. On day 7 of incubation, when gland formation starts, cSP is already restricted to luminal epithelium (Fig. 2G), indicating that epithelial cells have already differentiated into luminal and glandular epithelial cells. The cSP gene was expressed complementary to the ECPg gene as described previously (Figs. 2E and 2J; Tabata and Yasugi, 1998). In 6-day embryos, cGATA-5 was expressed in the entire proventricular epithelium and mesenchyme (Fig. 2K). It was expressed stronger in glandular epithelial cells than in luminal epithelial cells at the onset of gland formation (Fig. 2L). The cSox2 gene was expressed in all epithelial cells in 6-day embryos. However, in contrast to cGATA-5, when gland formation started, cSox2 was downregulated in glandular epithelial cells (Fig. 2Q, arrowhead), but was maintained in luminal epithelial cells. In the proventriculus of 10-day embryos in which the ECPg gene was extensively expressed, these transcription factor genes were expressed in patterns similar to those of 7-day embryos (Figs. 2L, 2O, 2Q, and 2T). To examine expression of these genes more closely, we established a method for double-color fluorescence in situ hybridization which allowed us to observe simultaneously the expression of two genes in a single cell. Using this technique, expression of the ECPg and cGATA-5 or cSox2 genes was compared in sections of proventriculus of 12-day embryos (Fig. 3). The expression of cGATA-5 was stronger in ECPg-positive cells than in ECPg-negative cells (Figs. 3A and 3C). In contrast, the expression of cSox2 in ECPgpositive cells was weaker than in ECPg-negative cells (Figs. 3B and 3D). The expression patterns of these genes at a cellular level suggested the hypothesis that transcription of the ECPg gene is upregulated by cGATA-5 and/or downregulated by cSox2.

Expression of cGATA-5 and cSox2 in Esophageal Epithelium under the Influence of Pulmonary Mesenchyme The ECPg gene is expressed in the proventriculus but not in the esophagus and gizzard during normal development. However, esophageal and gizzard epithelia can express the ECPg gene when they are cultured on proventricular or pulmonary mesenchyme (Urase et al., 1996). Thus, expression of ECPg in the epithelium is controlled by underlying mesenchyme. The expression patterns of cGATA-5 and cSox2 during gland formation suggest that their expression is controlled by mesenchyme as is that of the ECPg gene. Indeed, expression of cSox2 is induced in small-intestinal epithelium by proventricular mesenchyme (Ishii et al., 1998). If cGATA-5 and cSox2 regulate the transcription of the ECPg gene, the expression patterns of cGATA-5 and cSox2 in esophageal and gizzard epithelia will be altered by proventricular or pulmonary mesenchyme. In the esophageal epithelium of 6-day embryos, expression of cGATA-5

Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved.

107

cGATA-5 Regulates Pepsinogen Gene Expression

FIG. 2. Expression patterns of ECPg (A, B, C, D, E), cSP (F, G, H, I, J), cGATA-5 (K, L, M, N, O), and cSox2 (P, Q, R, S, T) in proventriculus during development. 6- (A, F, K, P), 7- (B, G, L, Q), 8- (C, H, M, R), 9- (D, I, N, S), and 10-day (E, J, O, T) proventriculi were analyzed by in situ hybridization on serial sections. ECPg begins to be expressed in glandular cells of 9-day proventriculus (D, arrowheads). cSP is expressed in luminal epithelial cells in 7-day proventriculus, but not expressed in glandular epithelial cells (G, arrowhead). cGATA-5 and cSox2 are expressed throughout the epithelium in 6-day proventriculus. In 7-day proventriculus, cGATA-5 is upregulated (L, arrowhead), while cSox2 (Q, arrowhead) is downregulated in invaginating epithelial cells. These patterns of cGATA-5 and cSox2 expression are also evident in 10-day proventriculus (O and T). Bar, 100 ␮m (A–T).

and cSox2 is uniform (Figs. 4A and 4B). In esophageal epithelium–pulmonary mesenchyme recombination explants, expression of cGATA-5 is stronger in ECPg-positive cells than in ECPg-negative cells (Figs. 4C and 4E) and expression of cSox2 in ECPg-positive cells is clearly weaker than in ECPg-negative cells (Figs. 4D and 4F). Thus, the change of expression patterns of cGATA-5 and cSox2 in esophageal epithelium was caused by the pulmonary mesenchyme. These results also show that ectopic expression of the ECPg gene is accompanied by a change of expression of these genes as in normal development of the proventriculus. Based on these results, we hypothesized that cGATA-5 upregulates and cSox2 downregulates transcription of the ECPg gene in glandular epithelial cells under the influences of the mesenchyme.

Transcriptional Regulation of the ECPg Gene by cGATA-5 and cSox2 We have shown the existence of one Sox binding motif and four GATA binding motifs in 1.1 kb of the 5⬘ flanking region of the ECPg gene (Fig. 1; Sakamoto et al., 1998). To

test whether these motifs are involved in cell-type-specific regulation of transcription of the ECPg gene, the reporter luciferase gene connected to this region was introduced into epithelial cells of proventricular or gizzard explants (Fig. 5A). After 4 days of cultivation, the explants were examined for luciferase activity. Luciferase activity was definitely detected in the proventriculus, but none was detected in the gizzard explants (Fig. 5B). This means that 1.1 kb of 5⬘ flanking region of the ECPg gene is sufficient for organspecific expression of the reporter gene, as described previously (Fukuda et al., 1994b). Next, luciferase-expressing cells were detected on sections by in situ hybridization. In proventricular explants, the reporter gene was clearly expressed only in glandular cells (Fig. 5C). Thus, 1.1 kb of 5⬘ flanking region of the ECPg drives expression of the reporter gene in an organ-specific as well as a cell-type-specific fashion. To analyze further the regulatory activity of cis elements in the 1.1 kb of the 5⬘ flanking region of the ECPg gene, this region was partially deleted and connected with a luciferase gene. This deletion series was introduced into the proventricular epithelium and the luciferase activity was mea-

Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved.

108

Sakamoto et al.

sured after 4 days cultivation (Fig. 6A). Explants transfected with Pm1k, Hm800, -SS230, or Hm-SS230 showed luciferase activity, whereas explants transfected with plasmids lacking about 430 bp including four GATA motifs (Am400 and Km120) could not support luciferase activity, suggesting that a GATA transcription factor is definitely needed for expression of the reporter gene in proventricular epithelium. Hm800, -SS230, and Hm-SS230 showed higher activity than Pm1k, but only the activity of Hm-SS230 was statistically significant. The effects of cGATA-5 and cSox2 on expression of the reporter gene were examined by cotransfecting the reporter plasmid mentioned above and cGATA-5 or cSox2 expression constructs. In explants transfected with Pm1k and cGATA-5, luciferase activity was 1.7 times higher than in controls. On the other hand, no luciferase activity was detected in explants transfected with Am400 and cGATA-5 (Fig. 6B). Thus, cGATA-5 has an enhancing effect on Pm1k but no effect on Am400, indicating that cGATA-5 acts on GATA motifs in the 430-bp region of the 5⬘ flanking region of the ECPg gene. In explants transfected with Pm1k, Hm800, or Hm-SS230 together with cSox2, luciferase activity decreased to 51–101 of controls (Fig. 6B). This decrease was also shown in a reporter plasmid which lacked the Sox motif (Hm800 and Hm-SS230), suggesting that cSox2 may indirectly repress transcription of the reporter gene or that the Sox binding motif -816 bp from the transcription start site of the ECPg gene is not involved in the repression by cSox2.

Gel Mobility Shift Assay To determine whether cGATA-5 protein can bind to the putative GATA binding sites, we performed a gel mobility shift assay using recombinant protein. As shown in Fig. 7, cGATA-5 bound to distal fragments (gs1, 2, and 3) and the binding was competed by the cognate DNA fragment, but not by noncompetitor DNA. In contrast, cGATA-5 showed no binding to a proximal element (gs4). These results suggest that cGATA-5 enhances the transcription of the ECPg gene by binding directly to its tissue-specific enhancer region.

FIG. 3. Expression of cGATA-5 (A), cSox2 (B), and ECPg (C, D) examined in 12-day proventriculus by double in situ hybridization method. In ECPg-expressing glandular cells (C, gl), cGATA-5 is upregulated (A, gl). In contrast, in ECPg-expressing glandular cells (D, gl), cSox2 is downregulated (B, gl). Note that the expression boundary of ECPg gene (D, arrowhead) coincides with the boundary of expression intensity of cSox2 (B, arrowhead). E and F are superimposed photographs of A and C and B and D, respectively. Bar, 50 ␮m (A–F). FIG. 4. Effect of pulmonary mesenchyme on expression of cGATA-5, cSox2, and ECPg in esophageal epithelium examined by double in situ hybridization method. In 6-day esophagus, cGATA-5

(A) and cSox2 (B) are uniformly expressed in the epithelium. Expression of ECPg (E, F), cGATA-5 (C), and cSox2 (D) in esophageal epithelium recombined and cultivated with pulmonary mesenchyme is shown. In ECPg-expressing glandular cells (E, arrowhead), cGATA-5 is upregulated (C, arrowhead). In contrast, in ECPg-expressing glandular cells (F), cSox2 is downregulated (D). Note that the expression boundary of the ECPg gene (F, arrowhead) coincides with the boundary of intensive expression of cSox2 (D, arrowhead). G and H are superimposed photographs of C and E and D and F, respectively. epi, epithelium. Bar, 50 ␮m (A–H).

Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved.

109

cGATA-5 Regulates Pepsinogen Gene Expression

FIG. 5. Transfection of Pm1k (see Fig. 6A) reporter plasmid into proventricular or gizzard epithelium. (A) Diagram of apparatus of electroporation to transfect DNA into proventricular epithelium (see Materials and Methods). (B) Luciferase activity is detected in proventricular (PV) explants, but not in gizzard (GZ) explants. Data are normalized against ␤-galactosidase activity and expressed relative to the activity of proventricular explants. (C) A proventricular explant analyzed by in situ hybridization for expression of luciferase mRNA. The signal was detected almost exclusively in glandular cells (arrowheads). epi, epithelium; mes, mesenchyme.

DISCUSSION Spatial Regulation of Transcription of Chicken Pepsinogen Genes by cGATA-5 In this study, we showed that cGATA-5 can alter ECPg gene transcription, by reporter analysis using 1.1 kb of 5⬘ flanking region of ECPg gene. Moreover, cGATA-5 protein was shown to bind to the putative GATA binding sites in that region. However, cGATA-5 is expressed throughout the gut (Figs. 2K and 4A, and data not shown), so it is not likely that it controls the organ specificity of ECPg expression. The organ-specific expression of ECPg may be brought about by some organ-specific signals coming from, for example, the mesenchyme. This idea is consistent with experimental results of Takiguchi et al. (1986) and Urase et al. (1996). cGATA-5 seems to regulate the cell-type-specific expression of ECPg. cGATA-5 is more strongly expressed in glandular epithelium than in luminal epithelium. The boundary of regions with high and low expression of the cGATA-5 gene coincides with the boundary of ECPg expression (Fig. 3). This was also confirmed when ECPg gene expression was experimentally induced in esophageal epithelium by pulmonary mesenchyme (Fig. 4). These observations strongly suggest that cGATA-5 regulates cell-typespecific expression of the ECPg genes.

Putative binding motifs for the GATA transcription factor family exist in the 5⬘ flanking regions of all three chicken pepsinogen genes (Sakamoto et al., 1998). We can therefore imagine that cGATA-5 also acts as a regulator of the spatial transcription patterns of chicken pepsinogen A and C genes. There are two GATA binding motifs in the promoter region of the human PgC gene (Hayano et al., 1988). However, it is still unknown whether there are GATA motifs in the promoter region of other pepsinogen genes, because the sequences of the 5⬘ flanking regions of these genes have not been investigated sufficiently. In the rat, GATA-4 and GATA-6 are known to control the transcription of the H ⫹/K ⫹-ATPase ␣ and ␤ subunit genes expressed in acid-producing parietal cells of the stomach (Mushiake et al., 1994). Parietal cells and pepsinogenexpressing chief cells in the stomach of mammals are different cell types, but in the chicken, acid and pepsinogens are produced by the same cells (Toner, 1963). Although the expression and functions of cGATA-4/6 in the proventriculus are obscure at present, it is possible that they regulate the transcription of the chicken H ⫹/K ⫹-ATPase genes and/or pepsinogen genes. Xenopus GATA-4/5/6 can activate a reporter gene through the promoter region of the xIFABP gene with different efficiencies (Gao et al., 1998). In chicken proventriculus, cGATA-4/5/6 may also have differ-

Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved.

110

Sakamoto et al.

study showed that Sox2 antagonizes transactivation by Oct-4 in transcription of the Osteopontin gene (Botquin et al., 1998). However, it is not clear at present that cSox2 works through the 5⬘ sequence of the ECPg gene, because deletion of the Sox binding site could not cancel the effect of cSox2 (Fig. 6). This suggests that, for the suppressive action of cSox2, other factors are necessary. Alternatively, cSox2 may bind some motifs within the 5⬘ flanking region of the ECPg gene. These problems may be clarified in the future using gel mobility shift assay and foot printing techniques.

An Epithelial–Mesenchymal Interaction Involved in Transcription of the ECPg Gene

FIG. 6. (A) Transfection of a deletion series of reporter plasmid into the proventricular epithelium. (B) Cotransfection of a deletion series of reporter plasmid with GFP, cGATA-5 (G5), or cSox2 (S2). Luciferase activities are normalized against ␤-galactosidase activity and expressed relative to the activity of explants transfected with Pm1k and GFP expression plasmid. Stars and circles mean statistically significant differences. A diagram of the promoter region of each plasmid is shown in the middle column. GATA motifs are shown as filled ellipses, Sox motifs as open ellipses. P, PstI; H, HinfI; A, ApaI; S, SmaI; K, KpnI.

ent transcriptional activities to regulate pepsinogen genes or other stomach-specific genes.

cSox2 as a Possible Transcriptional Repressor Members of the Sox transcription factor family are generally known as transcriptional activators. For example, chicken cSox2 activates the chicken lens ␦-1 crystalline gene (Karachi et al., 1995) and upregulates transcription of the FGF-4 gene in cooperation with Oct-3 in F9 cells (Ambrosetti et al., 1997). In the present study, however, cSox2 was shown to act as a possible repressor of ECPg transcription, since cotransfection of a full-length (Pm1k) or deleted construct with luciferase and a cSox2 construct resulted in a drastic decrease in luciferase activity (Figs. 5 and 6). Moreover, expression of cSox2 is downregulated in the proventricular gland epithelial cells (Fig. 2). A recent

The nature of inductive signals from the mesenchyme to the epithelium is one of the most important problems in developmental biology. An elaborate experimental system revealed that transcription of the ECPg gene is induced by a diffusible factor(s) from the proventricular mesenchyme (Koike and Yasugi, 1999). One of these factors is BMP-2, a member of the TGF-␤ superfamily, since it is expressed only in the proventricular mesenchyme at the time of gland formation, and ectopic expression of BMP-2 in the mesenchyme can induce much higher ECPg expression in proventricular epithelium. Moreover, ectopic expression of noggin, an antagonist for BMPs, in the epithelium or in the mesenchyme leads to the downregulation of ECPg expression (Narita et al., 2000). A similar relationship has been reported in ventralization during early Xenopus development, in which BMP-4 induces ventral fates such as blood cell differentiation and represses dorsal structures such as neurons (Maeno et al., 1994). Ectopic BMP signaling downregulates expression of Sox2, a panneural marker (Mizuseki et al., 1998), and upregulates hematopoietic GATA in the animal cap (Zhang and Evans, 1996) and in ventral mesoderm (Maeno et al., 1996). In the chicken, ectopic application of BMP-2 protein can elicit cardiac mesoderm specification, including cGATA-4 expression (Schultheiss et al., 1997; Andree et al., 1998). These cascades may also be working in the chicken proventriculus. Thus, BMP signals from proventricular mesenchyme may induce cGATA-5 and repress cSox2 and bring about the expression of the ECPg gene. The alteration of expression patterns of cGATA-5 and cSox2 occurs 2 days before the onset of ECPg transcription (Fig. 2). We assume the molecular events during this time lag to be as follows: The expression patterns of cGATA-5 and cSox2 are under the influence of mesenchyme. At the same time, Shh, a mediator of cell– cell interactions which is expressed throughout the gut epithelium, is downregulated in cells in the course of gland formation (Narita et al., 1998). Since Shh is known to exert its influence on the differentiation of the adjacent mesenchyme in the gut (Roberts et al., 1995, 1998; Sukegawa et al., 2000), the decrease of Shh signaling may cause a new signal from mesenchyme surrounding the glandular epithelium. This

Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved.

111

cGATA-5 Regulates Pepsinogen Gene Expression

FIG. 7. Binding of cGATA-5 protein to the GATA motifs in the ECPg gene promoter region. 32P-labeled DNA fragments encompassing one of four GATA motifs (see Fig. 1; gs1, A–D; gs2, E–H; gs3, I–L; gs4, M–P) were used in gel mobility shift assays with recombinant cGATA-5 protein translated in E. coli. The DNA–protein complexes shown in B, F, and J (arrowheads) were competed specifically by excess unlabeled probe (lanes C, G, and K), but not by noncompetitor DNA (lanes D, H, and L). DNA–protein complexes were not observed in lanes without cGATA-5 protein (A, E, I, and M). There was no shifted band when gs4 was used as a probe (M–P).

signal induces another transcription factor(s) essential to ECPg expression. It is also possible that the downregulation of Shh may be controlled by cGATA-5 and/or cSox2. Further experiments to stimulate or block Shh signaling will elucidate the relationship between Shh signaling and ECPg expression. Such experiments are currently being carried out in our laboratory using cyclopamine, a specific inhibitor of sonic hedgehog signaling.

ACKNOWLEDGMENTS We thank Dr. Kazuo Maruyama of Tokyo Medical and Dental University for the generous gift of the plasmid pME18S. We are also grateful to Dr. Hidetoshi Saiga for helpful discussion. This study was supported in part by the Sasakawa Scientific Research Grant from The Japan Science Society to N.S. and by Grants-in-Aid for Scientific Research on Priority Area and for International Scientific Research (Joint Research) from the Ministry of Education, Science, and Culture of Japan to S.Y.

REFERENCES Ambrosetti, D. C., Basilico, C., and Dailey, L. (1997). Synergistic activation of the fibroblast growth factor 4 enhancer by Sox2 and Oct-3 depends on protein–protein interactions facilitated by a specific spatial arrangement of factor binding sites. Mol. Cell. Biol. 17, 6321– 6329. Andree, B., Duprez, D., Vorbusch, B., Arnold, H. H., and Brand, T. (1998). BMP 2 induces ectopic expression of cardiac lineage markers and interferes with somite formation in chicken embryos. Mech. Dev. 70, 119 –131.

Botquin, V., Hess, H., Fuhrmann, G., Anastassiadis, C., Gross, M. K., Vriend, G., and Scho¨ler, H. R. (1998). New POU dimer configuration mediates antagonistic control of an osteopontin preimplantation enhancer by Oct-4 and Sox-2. Genes Dev. 12, 2073–2090. Egan, C. R., Chung, M. A., Allen, F. L., Heschl, M. F. P., Van Buskirk, C. L., and McGhee, J. D. (1995). A gut-to-pharynx/tail switch in embryonic expression of the Caenorhabditis elegans ges-1 gene centers on two GATA sequences. Dev. Biol. 170, 397– 419. Evers, M. P. J., Zelle, B., Bebelman, J. P., Pronk, J. C., Mager, W. H., Planta, R. J., Eriksson, A. W., and Frants, R. R. (1988). Cloning and sequencing of rhesus monkey pepsinogen A cDNA. Gene 65, 179 –185. Fukuda, K., Ichinose, M., Saiga, H., Shiokawa, K., and Yasugi, S. (1994a). Developmental changes of DNA methylation pattern of embryonic chick pepsinogen gene. J. Biochem. 115, 502–506. Fukuda, K., Ishii, Y., Saiga, H., Shiokawa, K., and Yasugi, S. (1994b). Mesenchymal regulation of epithelial gene expression in developing avian stomach: 5⬘-flanking region of pepsinogen gene can mediate mesenchymal influence on its expression. Development 120, 3487–3495. Gao, X., Sedgwick, T., Shi, Y.-B., and Evans, T. (1998). Distinct functions are implicated for the GATA-4, -5, and -6 transcription factors in the regulation of intestine epithelial cell differentiation. Mol. Cell. Biol. 18, 2901–2911. Hayano, T., Sogawa, K., Ichihara, Y., Fijii-Kuriyama, Y., and Takahashi, K. (1988). Primary structure of human pepsinogen C gene. J. Biol. Chem. 263, 1382–1385. Hayashi, K., Agata, K., Mochii, M., Yasugi, S., Eguchi, G., and Mizuno, T. (1988a). Molecular cloning and the nucleotide sequence of cDNA for embryonic chicken pepsinogen: Phylogenetic relationship with prochymosin. J. Biochem. 103, 290 –296.

Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved.

112

Sakamoto et al.

Hayashi, K., Yasugi, S., and Mizuno, T. (1988b). Isolation and structural analysis of embryonic chicken pepsinogen gene: Avian homologue of prochymosin gene. Biochem. Biophys. Res. Commun. 152, 776 –782. Hayashi, K., Yasugi, S., and Mizuno, T. (1988c). Pepsinogen gene transcription induced in heterologous epithelial–mesenchymal recombinations of chicken endoderms and glandular stomach mesenchyme. Development 103, 725–731. Ichihara, Y., Sogawa, K., Morohashi, K., Fujii-Kuriyama, Y., and Takahashi, K. (1986). Nucleotide sequence of a nearly full-length cDNA coding for pepsinogen of rat gastric mucosa. Eur. J. Biochem. 161, 7–12. Ichinose, M., Miki, K., Furihata, C., Tatematsu, M., Ichihara, Y., Ishihara, T., Katsura, I., Sogawa, K., Fujii-Kuriyama, Y., Tanji, M., Oka, H., Matsushima, T., and Takahashi, K. (1988a). DNA methylation and expression of the rat pepsinogen gene in embryonic, adult, and neoplastic tissues. Cancer Res. 48, 1603–1609. Ichinose, M., Miki, K., Tatematsu, M., Furihata, C., Nobuhara, M., Ichihara, Y., Tanji, M., Sogawa, K., Fujii-Kuriyama, Y., Oka, H., Takahashi, T., Kageyama, T., and Takahashi, K. (1988b). Hypomethylation and expression of pepsinogen A genes in the fundic mucosa of human stomach. Biochem. Biophys. Res. Commun. 151, 275–282. Ishihara, T., Ichihara, Y., Hayano, T., Katsura, I., Sogawa, K., Fujii-Kuriyama, Y., and Takahashi, K. (1989). Primary structure and transcriptional regulation of rat pepsinogen C gene. J. Biol. Chem. 264, 10193–10199. Ishii, Y., Fukuda, K., Saiga, H., Matsushita, S., and Yasugi, S. (1997). Early specification of intestinal epithelium in the chicken embryo: A study on the localization and regulation of CdxA expression. Dev. Growth Differ. 39, 643– 653. Ishii, Y., Rex, M., Scotting, P. J., and Yasugi, S. (1998). Regionspecific expression of cSox2 in the developing gut and lung epithelium: Regulation by epithelial–mesenchymal interactions. Dev. Dyn. 213, 464 – 475. Kageyama, T., Tanabe, K., and Koiwai, O. (1990). Structure and development of rabbit pepsinogens: Stage-specific zymogens, nucleotide sequences of cDNAs, molecular evolution, and gene expression during development. J. Biol. Chem. 265, 17031– 17038. Kamachi, Y., Sockanathan, S., Liu, Q., Breitman, M., Lovell-Badge, R., and Kondoh, H. (1995). Involvement of SOX proteins in lens-specific activation of crystallin genes. EMBO J. 14, 3510 – 3519. Kamachi, Y., Uchikawa, M., Collignon, J., Lovell-Badge, R., and Kondoh, H. (1998). Involvement of Sox1, 2 and 3 in the early and subsequent molecular events of lens induction. Development 125, 2521–2532. Koike, T., and Yasugi, S. (1999). In vitro analysis of mesenchymal influences on the differentiation of stomach epithelial cells of the chicken embryo. Differentiation 65, 13–25. Kuo, C. T., Morrisey, E. E., Anandappa, R., Sigrist, K., Lu, M. M., Parmacek, M. S., Soudais, C., and Leiden, J. M. (1997). GATA4 transcription factor is required for ventral morphogenesis and heart tube formation. Genes Dev. 11, 1048 –1060. Maeno, M., Mead, P. E., Kelley, C., Xu, R. H., Kung, H. F., Suzuki, A., Ueno, N., and Zon, L. I. (1996). The role of BMP-4 and GATA-2 in the induction and differentiation of hematopoietic mesoderm in Xenopus laevis. Blood 88, 1965–1972. Maeno, M., Ong, R. C., Suzuki, A., Ueno, N., and Kung, H. F. (1994). A truncate bone morphogenetic protein 4 receptor alters the fate of ventral mesoderm to dorsal mesoderm: Roles of

animal pole tissue in the development of ventral mesoderm. Proc. Natl. Acad. Sci. USA 91, 10260 –10264. Mizuseki, K., Kishi, M., Matsui, M., Nakanishi, S., and 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, 579 –587. Molkentin, J. D., Lin, Q., Duncan, S. A., and Olson, E. N. (1997). Requirement of the transcription factor GATA4 for heart tube formation and ventral morphogenesis. Genes Dev. 11, 1061– 1072. Morrisey, E. E., Ip, H. S., Tang, Z., Lu, M. M., and Parmacek, M. S. (1997). GATA-5: A transcriptional activator expressed in a novel temporally and spatially restricted pattern during embryonic development. Dev. Biol. 183, 21–36. Mushiake, S., Etani, Y., Shimada, S., Tohyama, M., Hasebe, M., Futai, M., and Maeda, M. (1994). Genes for members of the GATA-binding protein family (GATA-GT1 and GATA-GT2) together with H ⫹/K ⫹-ATPase are specifically transcribed in gastric parietal cells. FEBS Lett. 340, 117–120. Narita, N., Bielinska, M., and Wilson, D. B. (1997). Wild-type endoderm abrogates the ventral developmental defects associated with GATA-4 deficiency in the mouse. Dev. Biol. 189, 270 –274. Narita, T., Ishii, Y., Nohno, T., Noji, S., and Yasugi, S. (1998). Sonic hedgehog expression in developing chicken digestive organs is regulated by epithelial–mesenchymal interactions. Dev. Growth Differ. 40, 67–74. Narita, T., Saitoh, K., Kameda, T., Kuroiwa, A., Mizutani, M., Koike, C., Iba, H., and Yasugi, S. (2000). BMPs are necessary for stomach gland formation in the chicken embryo: A study using virally induced BMP-2 and Noggin expression. Development 127, 981–988. Nishi, T., Kubo, K., Hasebe, M., Maeda, M., and Futai, M. (1997). Transcription activation of H ⫹/K ⫹-ATPase genes by gastric GATA binding proteins. J. Biochem. (Tokyo) 121, 922–929. Pals, G., Meijerink, P. H. S., Defize, J., Bebelman, J. P., Strunk, M., Arwert, F., Timmerman, A., and Mager, W. H. (1995). Transcription regulation of human and porcine pepsinogen A. Roberts, D. J., Johnson, R. L., Burke, A. C., Nelson, C. E., Morgan, B. A., and Tabin, C. (1995). Sonic hedgehog is an endodermal signal inducing Bmp-4 and Hox genes during induction and regionalization of the chick hindgut. Development 121, 3163– 3174. Roberts, D. J., Smith, D. M., Goff, D. J., and Tabin, C. J. (1998). Epithelial–mesenchymal signaling during the regionalization of the chick gut. Development 125, 2791–2801. Sakamoto, N., Saiga, H., and Yasugi, S. (1998). Analysis of temporal expression pattern and cis-regulatory sequences of chicken pepsinogen A and C. Biochem. Biophys. Res. Commun. 250, 420 – 424. Schultheiss, T. M., Burch, J. B., and Lassar, A. B. (1997). A role for bone morphogenetic proteins in the induction of cardiac myogenesis. Genes Dev. 11, 451– 462. Sogawa, K., Fujii-Kuriyama, Y., Mizukami, Y., Ichihara, Y., and Takahashi, K. (1983). Primary structure of human pepsinogen gene. J. Biol. Chem. 258, 5306 –5311. Sogawa, K., Ichihara, Y., and Takahashi, K. (1981). Molecular cloning of complementary DNA to swine pepsinogen mRNA. J. Biol. Chem. 256, 12561–12565. Soudais, C., Bielinska, M., Heikinheimo, M., MacArthur, C. A., Narita, N., Saffiz, J. E., Simon, M. C., Leiden, J. M., and Wilson, D. B. (1995). Targeted mutagenesis of the transcription factor

Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved.

113

cGATA-5 Regulates Pepsinogen Gene Expression

GATA-4 gene in mouse embryonic stem cells disrupts visceral endoderm differentiation in vitro. Development 121, 3877–3888. Sukegawa, A., Narita, T., Kameda, T., Saitoh, K., Nohno, T., Iba, H., Yasugi, S., and Fukuda, K. (2000). The concentric structure of the developing gut is regulated by sonic hedgehog derived from endodermal epithelium. Development 127, 1971–1980. Tabata, H., and Yasugi, S. (1998). Tissue interaction regulates expression of a spasmolytic polypeptide gene in chicken stomach epithelium. Dev. Growth Differ. 40, 519 –526. Takiguchi, K., Yasugi, S., and Mizuno, T. (1986). Gizzard epithelium of chick embryos can express embryonic pepsinogen antigen, a marker protein of proventriculus. Roux’s Arch. Dev. Biol. 195, 475– 483. Takiguchi, K., Yasugi, S., and Mizuno, T. (1988). Pepsinogen induction in chick stomach epithelia by reaggregated proventricular mesenchyme cells in vitro. Dev. Growth Differ. 30, 241–250. Toner, P. G. (1963). The fine structure of resting and active cells in the submucosal glands of the fowl proventriculus. J. Anat. London 97, 575–583. Urase, K., Fukuda, K., Ishii, Y., Sakamoto, N., and Yasugi, S. (1996). Analysis of mesenchymal influence on the pepsinogen gene expression in the epithelium of chicken embryonic digestive tract. Roux’s Arch. Dev. Biol. 205, 382–390.

Urase, K., and Yasugi, S. (1993). Induction and inhibition of epithelial differentiation by the mixed cell aggregates of the mesenchymes from the chicken embryonic digestive tract. Dev. Growth Differ. 35, 33– 40. Yasugi, S. (1994). Regulation of pepsinogen gene expression in epithelial cells of vertebrate stomach during development. Int. J. Dev. Biol. 38, 273–279. Yasugi, S. (1995). Differentiation of avian digestive tract epithelium in vitro. Tissue Cult. Res. Commun. 14, 177–184. Yasugi, S., and Fukuda, K. (2000). The mesenchymal factors regulating epithelial morphogenesis and differentiation of the chicken stomach. Zool. Sci. 17, 1–9. Zhang, C., and Evans, T. (1996). BMP-like signals are required after the midblastula transition for blood cell development. Dev. Genet. 18, 267–278. Zhu, J., Hill, R. J., Heid, P. J., Fukuyama, M., Sugimoto, A., Priess, J. R., and Rothmann, J. H. (1997). end-1 encodes an apparent GATA factor that specifies the endoderm precursor in Caenorhabditis elegans embryos. Genes Dev. 11, 2883–2896.

Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved.

Received January 6, 2000 Revised March 10, 2000 Accepted March 28, 2000