HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Shiraki, N. Articles by Kume, S. Search for Related Content PubMed PubMed Citation Articles by Shiraki, N. Articles by Kume, S.

Nobuaki Shiraki^1,, Cheng-Jung Lai^2,, Yosuke Hishikari¹ and Shoen Kume^1,*

¹ Institute of Molecular Embryology and Genetics, Kumamoto University, Honjo 2-2-1, Kumamoto 860-0811, Japan
² Howard Hughes Medical Institute, Department of Molecular and Cellular Biology, Harvard University, Cambridge, MA 02138, USA

	Abstract

Top Abstract Introduction Results Discussion Experimental procedures References

 
Embryonic stem (ES) cells have the capacity to differentiate to every cell type that constitutes fetal or adult tissues. To trace and quantitatively assess the differentiation of ES cells into gut endodermal cells, we used an ES cell line with the lacZ gene inserted into the pdx-1 locus. Targeted mutations of pdx-1 in mice demonstrate that pdx-1 is required for pancreatic and rostral duodenal development; therefore, pdx-1 serves as an excellent early gut regional specific marker. When these ES cells were differentiated by removal of leukemia inhibitory factor (LIF), only fractional cells turned into lacZ positive, which indicates pancreatic-duodenal differentiation. Co-cultivation of ES cells with pancreatic rudiments induced a significant increase in the proportion of lacZ positive cell numbers and this increase was further enhanced by forced expression of a chick putative endoderm inducer gene, cmix. Transforming growth factor (TGF)-ß2 mimicked the effects of pancreatic rudiments and this effect was enhanced by cmix expression. Expression analysis showed over-expression of cmix induced endodermal marker genes. These data indicate that one can make use of this knowledge on molecular events of embryonic development to drive ES cells to differentiate into pdx-1 expressing endodermal cells in vitro.

	Introduction

Top Abstract Introduction Results Discussion Experimental procedures References

 
Because of their pluripotency, ES cells are a useful model system to use for studies of gene function during development and may provide a source of transplantable cells for medical application. Use of ES cells for cell-replacement therapy in diseases such as diabetes mellitus requires controlled differentiation and this remains a great challenge. While remarkable progress has been made in understanding the differentiation of ES cells to neural, hematopoietic, and cardiac tissue (Yamashita et al. 2000; Ying et al. 2003), limited knowledge has been acquired for their differentiation into the tissues of the definitive endoderm lineage.

In 2001, it was reported that ES cells form pancreatic insulin-containing cells in vitro when cultivating them, first in a condition under which nestin-positive neuronal cell populations are enriched, and then with specific supplements (Lumelsky et al. 2001). But the insulin-positive cells do not express pdx-1, a marker for both pancreatic and duodenal endoderm and functional pancreatic ß cells (Lumelsky et al. 2001). They did not express insulin I transcripts, either, but only a trace amount of insulin II transcripts. Moreover, the cells, which are positively stained for insulin, did not stain with an antibody against C-peptide, a byproduct of de novo insulin synthesis (Rajagopal et al. 2003). These findings indicated that the insulin staining resulted from insulin uptake from the medium and the cells do not synthesize insulin for themselves, casting doubt as to whether or not nestin-positive cells generate insulin producing ß cells (Rajagopal et al. 2003). However, using similar conditions selective for nestin-positive cells, other researchers reported generation of insulin-secreting cells from ES cells, thus leaving this issue elusive (Hori et al. 2002; Blyszczuk et al. 2003; Moritoh et al. 2003). Recently, it is reported that nestin-positive progenitors give rise to a population of cells that release insulin when glucose is added to the media, but, notably that, C-peptide release is never detected (Hansson et al. 2004), thus suggesting that they are not functional insulin secreting ß cells. It remains as a challenging goal for engineering ES cells into pancreatic endocrine ß cells.

In vertebrates, endodermal induction is mediated by activation of TGF-ß signaling. In loss-of-function studies, over-expression of dominant-negative constructs of TGF-ß receptors that block TGF-ß signals converted the prospective endodermal cells to mesodermal and ectodermal cells (Henry et al. 1996). In gain-of-function studies, over-expression of an activated form of zebrafish TGF-ß type I receptor kinase, TARAM-A, induced mesendodermal markers and transfated early blastomeres into endoderm. This implicates that activation of TGF-ß type I receptor kinase functions in the induction of mesoderm and endoderm (David & Rosa 2001). Studies in amphibians, mice and zebrafish have shown that Nodal, a ligand of TGF-ß family receptor, is a potent inducer of mesodermal and endodermal cell fates (Conlon et al. 1994; Tremblay et al. 2000). Smads function at the downstream of TGF-ß receptor signaling, and the Smad2 knockout mice exhibit defects in the formation of the primitive streak, mesoderm and endoderm, which are similar to Nodal mutants (Tremblay et al. 2000). Higher levels of Nodal signaling are required for endoderm formation whereas lower levels induce mesoderm formation (Schier 2003). Upon activation of Nodal signaling, the receptor-activated Smad2 forms a complex with Smad4 and translocates to the nucleus. The Smads then form a complex with other DNA-binding proteins, such as Foxh1, or Mix family protein, and regulate Nodal target gene expression (Derynck et al. 1998; Germain et al. 2000). Ectopic expressions of Mix.1, Milk and Mixer lead to the expression of mesodermal and endodermal genes in Xenopus animal caps (Ecochard et al. 1998; Henry & Melton 1998; Lemaire et al. 1998). Loss-of-function analysis of zebrafish Mix gene, bonnie and clyde (Bon), revealed a reduction in endodermal precursors (Kikuchi et al. 2000).

By the end of gastrulation when all three germ layers are identifiable, gut folding takes place and early gut endoderm is patterned along the anterior-posterior axis prior to organogenesis. Derivatives of the respiratory and digestive system give rise from specific regions of the endoderm, as a result of interaction of endoderm with adjacent ectoderm or mesoderm. For example, signals from the notochord and the dorsal aorta have been shown to be sending permissive signals required for the induction of the dorsal pancreas (Kim et al. 1997; Lammert et al. 2001).

Although the process of gut endoderm formation has been observed morphologically, little is known about the roles of secreted molecules involved in organogenesis of the endodermal tissues. In mouse embryos, pdx-1 expression is by far the first sign of differentiation detected at embryonic 8.5 days post coitum (dpc) in the dorsal endoderm of the gut. At 9.5 dpc, pdx-1 expression marks the dorsal and ventral pancreatic buds, and the duodenal endoderm. In the adult, pdx-1 expression is maintained in the duodenal epithelium and in the insulin-secreting islet ß-cells, where it plays a critical role in the regulation of insulin gene transcription (Offield et al. 1996). Targeted mutations of pdx-1 in mice demonstrate that pdx-1 is required for pancreatic and rostral duodenal development (Ahlgren et al. 1996; Offield et al. 1996). Therefore, pdx-1 is an essential molecule for the pancreatic development and also serves as an early marker for pancreatic precursors and other endodermal derivatives such as stomach, duodenum and bile duct.

In an attempt at engineering ES cells into gut endoderm, we use an ES cell line, which bears lacZ reporter in the pdx-1 locus. We developed protocols for efficient generation of pdx-1 positive cells and over-expressed in ES cells a putative endoderm inducer chick mix (cmix) gene, which has been reported to be a chick homolog of Mix family gene (Peale et al. 1998; Stein et al. 1998). We show here that ES cells are efficiently induced to a pdx-1 expressing gut endodermal fate when TGF-ß signaling pathways are activated.

	Results

Top Abstract Introduction Results Discussion Experimental procedures References

 
Signals from pancreatic rudiments induce ES cells to differentiate into pdx-1 expressing gut endoderm

Previous reports demonstrated that pdx-1 is required for pancreatic and rostral duodenal development, as well as for maintenance of a functional ß-cell phenotype (Ahlgren et al. 1996, 1998; Offield et al. 1996). To quantitatively evaluate pancreatic development of ES cells, we used an ES cell line bearing the lacZ gene integrated into the pdx-1 locus (Offield et al. 1996). Differentiation of ES cells was initiated by withdrawal of LIF. Embryoid bodies (EBs) were formed by cultivating in suspension. EBs differentiated for 21 days expressed both pdx-1 and lacZ transcripts revealed by RT-PCR analysis (Fig. 1B). We then asked if we could quantitatively evaluate the pdx-1/lacZ expressing cells by in situ X-gal staining. We dissociated EBs on day 2 and replated the cells in tissue culture treated 24 well plates to allow them to grow in monolayers. We found that differentiation by withdrawal of LIF yielded 0–10 pdx-1/ß-gal positive cells out of 10⁶ cells in a single well, which corresponded to 0–0.001% of total cells (Fig. 1C,D). We tried to find a source of inducer that enhanced pdx-1/ß-gal expression. We tested the effects of recombining differentiated ES cells with embryonic tissues. Figure 1A shows a schematic drawing of the differentiation experiments. ES cells after differentiation of 2 days were recombined with embryonic tissues and grown until day 14 of differentiation, then assayed for ß-gal activity by X-gal staining. We tested pancreatic rudiments, liver, lung, intestine buds dissected from 10.5 dpc embryos for the source of inducer. We found that ES cells cocultivated with embryonic lung and intestine buds showed an approximately two-fold increase in the number of lacZ-positive cells. ES cells cocultivated with embryonic liver buds showed no significant change compared to findings in control ES cells cultivated alone. When ES cells were recombined with pancreatic rudiments or mesenchyme, a 10-fold increase of pdx-1 expressing cells at approximately 500 out of every 10⁶ cells (0.05%) was observed (Fig. 1D,E). Pancreatic rudiments isolated from embryos at 9.5 dpc through 16.5 dpc are effective. Pancreatic mesenchyme isolated from 11.5 dpc embryos (Fig. 1E) showed a similar extent of activities in inducing the differentiation of ES/EB cells into pdx-1/ß-gal positive cells.

View larger version (60K): [in this window] [in a new window]   Figure 1  Specific potentiation of differentiation of ES cells to pdx-1 expressing cells by cocultivation of ES cells with pancreatic rudiments. (A) Schematic drawing of the differentiation procedure of ES cells. Differentiation of ES cells is initiated by withdrawal of LIF from the medium. Day 2 EB is dissociated, cultivated alone or recombined with pancreatic rudiments (PAN) or other tissues. Recombined cells were plated and cultivated on 24-well culture dishes for an additional 12 days, which is equivalent to day 14 after LIF removal, and were fixed and assayed for ß-gal activities. (B) RT-PCR analysis of lacZ and pdx-1 expressions in undifferentiated ES cells (lane, ES) and EBs differentiated for 21 days in suspension (lane, EB21). PCR reactions were run for 35 cycles. (C) X-gal staining for detection of pdx-1/ß-gal positive cells. ES cells cultivated alone (ES), recombined with pancreatic rudiments (PAN), pancreatic mesenchyme (PM), or other embryonic tissues: lung bud (LUNG), intestine buds (INT), or liver buds (LIVER) are shown. ES cells differentiated as described in (A) are fixed and assayed for X-gal staining on day 14. Embryonic tissues from embryos at 10.5 dpc were used. (D) Total number (left Y-axis) or percentage (right Y-axis) of lacZ-positive cells in C are shown. Co-cultivation of ES cells with pancreas rudiments showed increase in number and proportion of pdx-1/ß-gal positive cells. (E) Total number and percentage of lacZ-positive cells in ES cells cultured alone (ES, ), cocultivated with 10.5 dpc or 12.5 dpc whole pancreatic rudiments (PAN), or with embryonic 11.5 dpc pancreatic mesenchyme (PM) isolated at embryonic 11.5 dpc. Similar numbers of pdx-1/ß-gal positive cells were yielded between ES cells cocultivated with pancreatic mesenchyme and with whole pancreatic rudiments (). (F) To optimize the cultivation system, we use EB cells that were differentiated for 0, 2 or 8 days. No significant differences in variable groups of EBs were observed. For D–F, ES cells cultivated alone (), or in cocultivation with embryonic tissues (). For D–F, values are expressed as means ± s.e.m. Numbers of experiments are shown above the bars.

Addition of TGF-ß2 induced differentiation of ES cells into pdx-1 expressing cells

The finding that pancreatic mesenchyme potentiate ES cells to differentiate into pdx-1 expressing cells suggests that a soluble factor secreted from the pancreatic mesenchyme may play a role. To identify signals that specifically enhance differentiation of ES cells into pdx-1/ß-gal positive cells, we tested the effects of growth factors added to the dissociated day 2 EBs.

The growth factors tested were: Activin, TGF-, TGF-ß1, TGF-ß2, FGF-2, FGF-4, FGF-5, FGF-6, FGF-7, FGF-8b, FGF-9, IGF-1, IGF-2, EGF, HGF or VEGF. Of the growth factors tested, the addition of TGF-ß2 reproducibly yielded a higher proportion of pdx-1/ß-gal positive cells (Fig. 2A,B), compared to findings in controls. However, the proportion of the pdx-1/ß-gal positive cells obtained by the addition of TGF-ß2 was always lower than the ES cells cocultivated with pancreatic rudiments (Fig. 2B). Moreover, we observed no synergistic or additive effects between pancreatic rudiments and TGF-ß2 (Fig. 2B). These results suggest that TGF-ß2 is one of the candidate factors secreted from pancreatic rudiments and which induces differentiation of ES cells into pdx-1 expressing cells.

View larger version (47K): [in this window] [in a new window]   Figure 2  The activity of pancreatic rudiments in potentiating differentiation of ES cells into pdx-1 expressing cells is mimicked by TGF-ß2. (A) X-gal staining of ES cells differentiated in the presence of purified growth factors such as TGF-ß2, TGF-ß1, TGF-, IGF-2, FGF-4 or VEGF are shown. ES cells added with TGF-ß2 showed increased numbers of pdx-1/ß-gal positive cells than control ES cells cultivated alone. (B) ES cells were cultivated alone (ES) (open bar) or in cocultivation with pancreatic rudiments (PAN) in the presence (PAN Tß2) or absence of TGF-ß2 (Tß2). Addition of TGF-ß2 to the cultivation partially mimicked action of pancreatic rudiments and yields an increased number/proportion of pdx-1/ß-gal positive cells.

TGF-ß signaling induce endodermal development through functional Mix/Bix family protein. It has been shown that ectopic expression of endodermal inducing Mix family genes led to activation of the mesendodermal pathway. We thus asked if over-expression of the endodermal inducing gene would potentiate differentiation of ES cells into pdx-1/ß-gal positive cells. To date, one murine Mix-like gene, Mixl1 (Mml, Mix) (Pearce & Evans 1999; Robb et al. 2000) has been identified. Mixl1 mutants exhibited complex defects in axial mesodermal and endodermal structures, hence a possible role for Mixl1 in the regulation of morphorgenetic movements associated with gastrulation. However, we found no effects of over-expression of Mixl1 gene in ES cells (see Fig. 4G,H). We then examined the effect of a putative endoderm inducer chick mix (cmix) gene, which has been reported to be a chick homolog of Mix family gene (Peale et al. 1998; Stein et al. 1998).

To define the function of cmix as an endodermal inducing gene, we injected cmix RNA into Xenopus early embryos. Ventral injection of cmix RNA did not lead to significant effects. Dorsal injection of low doses (160 pg/embryo) of cmix RNA resulted in retarded involution and malformation of the head. Injection of higher doses (320–640 pg/embryo) resulted in a loss of axial structures (Fig. 3A). A similar loss of axis was seen in the embryo injected dorsally with Mixer RNA (160 pg/embryo), which suggests that the cmix gene functions similarly to the Mixer gene in endodermal induction (Fig. 3A). Histological analysis of embryos given Mixer RNA or cmix RNA injection revealed that the embryos consisted of epidermal and endodermal like structures and that no axial structures were observed (Fig. 3B). Cmix blocked mesodermal cell involution, when ectopically expressed in the dorsal marginal zone, and this effect resembles that of Xenopus endodermal inducing gene milk (Ecochard et al. 1998). We next did animal cap assays to examine if endodermal molecular markers were induced by ectopically expressed cmix RNA.

View larger version (51K): [in this window] [in a new window]   Figure 3  Endodermal induction activity of Cmix RNA in Xenopus embryos. (A) Effects of cmix RNA injection into Xenopus early embryos. Pictures of the followings are shown: Xenopus embryos given dorsal injections at a dose of 160, 320 or 640 pg/embryo cmix RNA; dorsal injection of 160 pg/embryo Mixer RNA; ventral injection of 320 or 640 pg/embryo cmix RNA; control uninjected embryos. Injections are given into marginal zones. Embryos are shown at stage 27 equivalent. (B) Histological analysis of embryos dorsally injected with 160 pg/embryo Mixer RNA or 320 pg/embryo cmix RNA at stage 27. No axial structures are observed and the embryos consisted of epidermal and yolky endodermal-like structures. (C) Expressions of molecular markers in Xenopus animal caps injected with a graded dose of cmix RNA or Mixer RNA. RNA was injected at 160, 320 or 640 pg/embryo. RT-PCR analyses of mid- gastrula stage (stage 11.5) (upper panel), or tailbud stage (stage 25) (lower panel) animal caps are shown. Uninjected, RNA from uninjected animal caps; WE, RNA from whole embryos; -RT, negative control without addition of RT products in PCR reaction.

Over-expression of cmix potentiates ES cells to respond to pancreatic rudiments and TGF-ß2 and to differentiate into pdx-1 expressing cells

We then examined the effect of over-expression of cmix gene in ES cells. In order to detect protein expression, we constructed a cmix expression vector with a V5-tag fused to its C-terminus under the control of the CMV enhancer/chick ß-actin promoter (pCAGGScmixV5) (Niwa et al. 1991). A band with a molecular mass of 23.7 kDa corresponding to cmixV5 protein specifically appeared in COS cells transiently transfected with pCAGGScmixV in our Western blot analysis (data not shown). A band of the same size was detected in ES cells transfected with pCAGGScmixV5. We isolated several stable ES cell transformants over-expressing cmixV5. In these transformants, distinct levels of cmixV5 protein were detected (Fig. 4A). We then examined whether or not the expression of cmixV5 protein would be maintained during differentiation of these cmix transformants. Figure 4B shows a typical temporal change of the cmixV5 expression during differentiation of the #59/cmix transformant (Fig. 4B). CmixV5 protein is expressed throughout in vitro differentiation of ES cells at least up to 14 days of differentiation. Similar sustained expression patterns of cmixV5 protein are also observed in other cmix transformants (data not shown).

View larger version (35K): [in this window] [in a new window]   Figure 4  Potentiation of differentiation of ES cells into pdx-1/ß-gal positive cells by over-expression of cmixV5. (A) Western blot analysis of cmix stable transformants and control ES cell. Clones #30, #35, #47, #41 and #59/cmix express cmixV5 protein with a molecular mass of 23.7 kDa. (B) Temporal expression pattern of cmixV5 protein during in vitro differentiation of ES cells. Differentiation was triggered by LIF withdrawal. EBs were formed in suspension and recovered at days indicated. The expression of cmixV5 was retained until day 14 of differentiation. (C) The cmix ES cell transformants were differentiated alone or cocultivated with pancreatic rudiments. ES cells differentiated alone. ES cells cocultivated with pancreatic rudiments. N, numbers of samples studied. Percentages of lacZ-positive cells are shown. Studies are done by comparing cmixV5 transformants with control ES cells. Differences are observed in clones #30, #35, #47, #41, and #59 (paired Student's t-test, *P < 0.1, and **P < 0.05). Values are expressed as means +/– s.e.m. Almost no lacZ-positive pdx-1 expressing cells were observed when these ES cells were cultured alone. (D, E) Temporal appearance of lacZ-positive cells in control and #59/cmix ES cells during differentiation. (D) Absolute numbers of the lacZ-positive cells in a single well; (E) percentages of the lacZ-positive cells are shown. ES cells cultured alone. Note that no lacZ-positive cells are observed in ES cells cultivated alone. ES cells cocultivated with pancreatic rudiments. LacZ-positive cells appeared on day 9. Values are expressed as means ± s.e.m. (F) The effect of TGF-ß2 on the percentage of pancreatic differentiation of #59/cmix ES cells. ES cells alone; gray bars, with the addition of TGF-ß2; cocultivated with pancreas rudiments. Values are expressed as means ± s.e.m. N, numbers of experiments done. (G) Western blot analysis of stable ES cell transformants over-expressing Mixl1V5. Clone #79/Mixl1 expressed Mixl1V5 protein of 26.5 kDa. (H) #79/Mixl1 ES cell transformants did not show increased lacZ-positive cells either differentiated alone or cocultivated with pancreatic rudiments, compared to control ES cells.

Analysis revealed that the pdx-1/ß-gal positive cells appeared as early as day 7 of differentiation and a substantial number of pdx-1/ß-gal positive cells were observed on day 9. The absolute numbers of pdx-1/ß-gal positive cells increased after day 9 and reached a plateau on day 12, remaining unchanged thereafter until day 14 of differentiation (Fig. 4D). However, there was an intense increase in the total cell number, thereby resulting in an apparent decrease in the proportion of pdx-1/ß-gal positive cells after day 9 of differentiation (Fig. 4E). The decrease in the proportion of pdx-1/ß-gal positive cells after a prolonged cultivation is also observed in control cmix untransformed ES cells, meaning that this is probably not cmix dependent. Since day 12 reproducibly gives the highest number of pdx-1/ß-gal positive cells, we assayed for pdx-1/ß-gal positive cells on day 12 of differentiation, in the following experiments.

Next we studied the effects of addition of TGF-ß2 on cmix transformants. Upon removal of LIF, #59/cmix ES cells responded to TGF-ß2 and showed an increase in the proportion of pdx-1/ß-gal positive cells compared to findings in control ES cells (Fig. 4F). This finding suggested that TGF-ß2 potentiates differentiation of ES cells into pdx-1/ß-gal positive cells, an event further enhanced by cmix gene over-expression. Therefore, we are of the view that signals from pancreatic rudiments activate differentiation pathways to an endodermal fate and this is further enhanced by over-expression of cmix in ES.

Unlike cmix, Hepatocyte nuclear factor (HNF)-3ß led to no significant difference. It was reported that exogenously expressed Mixl1 RNA induced a truncated axial axis in Xenopus early embryos. The amount of Mixl1 RNA needed is higher (Mohn et al. 2003) compared to cmix in order to induce a truncated axis (present study). As described above, we initially attempted to study the effect of over-expression of Mixl1 gene in ES cells, which turned out to be not potent (Fig. 4G,H). We constructed an expression vector for expression of Mixl1V5 protein (pCAGGSMixl1V5). Stable ES cell transformants over-expressing Mixl1V5 (26.5 kDa) were isolated and the expression of Mixl1V5 protein in #79 was confirmed in our Western blot analysis (Fig. 4G). However, #79/Mixl1 transformant did not show a significant increase in the proportion of pdx-1/ß-gal positive cells, when cocultivated with pancreatic rudiments (Fig. 4H). This result suggests that Mixl1 is not potent in inducing endodermal differentiation in ES cells.

Over-expression of cmix induced endodermal differentiation of ES cells

We examined expression of several germ layer specific molecular markers during #59/cmix and control ES cells differentiation (Fig. 5). Differentiation of ES cells is triggered by withdrawal of LIF, followed by formation of EBs. RNAs were extracted from EBs at indicated days after the withdrawal of LIF and were proceeded for semiquantitative RT-PCR analysis. The mesodermal marker Brachyury (T), and neuroectoderm marker Otx2 were not affected. Several genes are up-regulated specifically in #59/cmix ES cells after 7 days of differentiation. These genes include the visceral endodermal (VE) marker Ihh; -Fetoprotein (AFP) and Transthyretin (TTR), which are liver markers as well as the VE marker, and genes associated with endodermal induction, such as Sox-17, HNF-3, HNF-4, GATA-4, GATA-6. A gene which associates with anterior visceral endoderm differentiation, Hesx1, and Bone morphorgenetic protein (BMP)-2, expressed in VE, are also up-regulated. BMP-4, which is found in the inner cell mass (ICM) and embryonic ectoderm of the embryo, is gradually up-regulated during ES cell differentiation in #59/cmix ES cell transformants. FGF-4, which is implicated in trophoblast cell differentiation, and FGF-5, a marker of the primitive ectoderm in early development, is little affected. Nodal, which is implicated in mesodermal induction in addition to endodermal induction, is up-regulated in control ES cells after differentiation, and in undifferentiated #59/cmix ES cells.

View larger version (86K): [in this window] [in a new window]   Figure 5  RT-PCR analysis of early endodermal markers or pancreatic endocrine molecular marker expressions in differentiated ES cells.Differentiation of control or #59/cmix ES cells is triggered by removal of LIF and EBs formed in suspension. Aliquots of differentiated EBs were sampled on days 4,7,9 of differentiation, RNAs were extracted and examined for expression of molecular markers. ES, undifferentiated ES cells. RT (–), negative controls without RT reaction. All PCR reactions are run for 20–28 cycles (see Table 2) and confirmed empirically that the detections are of linear ranges. Detections were done by staining with SYBRGreen I.

HNF-3

Limited up-regulation of regional markers of the gut endoderm was observed by RT-PCR analysis (Fig. 5). Hex, which is implicated in definitive endoderm specification, and NeuroD, which is implicated in differentiation of the pancreatic endocrine ß cells, are up-regulated in undifferentiated #59/cmix ES cells. Nkx2.2, which is implicated in pancreatic ß cell differentiation, and somatostatin, a marker for endocrine cells, and pdx-1 are up-regulated after 4–9 days of differentiation in #59/cmix EBs.

Overall, the expression pattern of early and late molecular markers show that cmix over-expression in ES cells potentiates endodermal differentiation and that some extent of regional specification of gut endoderm is potentiated by cmix over-expression. However, endocrine markers such as insulin or glucagon were not detected by RT-PCR. Moreover, we are not able to detect Nkx2.2, somatostatin or pancreatic peptide expression using immunohistochemistry, probably due to low level of protein expression or low proportion of the positive cells (SK and NS unpublished observation).

	Discussion

Top Abstract Introduction Results Discussion Experimental procedures References

 
Here, we used an ES line bearing the lacZ gene integrated into the pdx-1 locus, and found that cocultivation of mouse ES cells with pancreatic rudiments or pancreatic mesenchyme differentiated toward a gut endodermal fate with a 10-fold increase compared to findings in control ES cells. Forced expression of cmix, a gene with endodermal inducing activities, in ES cells in combination with the cocultivation with pancreatic mesenchyme further potentiated the ES cells to differentiate into more pdx-1 expressing cells. TGF-ß2 partially mimicked pancreatic rudiments and worked synergistically with cmix over-expression. Normal pathways of endodermal induction are driven by activation of the TGF-ß signaling through binding of Nodal to its receptor. Our findings suggest that pancreatic rudiments secrete the TGF-ß ligand that binds to the receptor of the TGF-ß family on ES cells, which activates TGF-ß signaling components and the activation is further relayed by binding of cmix to its partner and switch on downstream targets. Actually, high level early expression of the TGF-ß2 as well as its receptor TßRII, has been observed in embryonic pancreas epithelium and mesenchyme (Crisera et al. 1999; Dichmann et al. 2003) and ActRIIA was seen to be expressed in undifferentiated ES cells in an Affymetrix gene chip analysis (Ramalho-Santos et al. 2002). Our data suggests that TGF-ß2 is one of the endodermal inducing signaling molecules secreted from the pancreatic rudiments. No other secreted growth factors we tested showed endodermal inducing effects. We interpret this result to mean that under conditions where serum is present, the endodermal inducing activity might be masked by yet unknown inhibitory factors in the serum. In supporting this, it is reported that EB cultured in serum free medium for a certain period expressed endodermal gene, and that addition of activin A potentiated endodermal differentiation under condition without serum (Kubo et al. 2004).

In the chick, a single Mix-like gene (cmix), which has approximately 70% amino acid identity within the homeodomain with Mix.1 has been cloned. Cmix is expressed at the position of endoderm progenitors in the streak. It ceases to be expressed when the streak loses the potentials to generate endoderm, and it is not expressed in definitive endoderm (Peale et al. 1998; Stein et al. 1998). The expression pattern of cmix suggests that endodermal precursors express cmix but turn it off upon differentiation into definitive endoderm.

Mixl1 RNA was first detected in the visceral endoderm of the pregastrula embryo. By 6.5 dpc, expression was seen in the nascent primitive streak and emerging mesoderm, being maintained in the emerging mesoderm in mid- to late-streak embryos then becoming restricted to the posterior primitive streak by the head-fold stage. The cells expressing Mixl1 do not overlap with the previously described endoderm precursor in the epiblast (Lawson et al. 1991). Analysis of Mixl1 mutant mice (Hart et al. 2002) revealed that the definitive endoderm is specified but gut morphogenesis is defective in Mixl1 mutants, which suggests a role for Mixl1 in the regulation of morphorgenetic cell movements rather than for endodermal differentiation. Unlike other members of the mix family, Mixl1 lacked the PPNK-containing Smad2 interaction motif (Germain et al. 2000; Randall et al. 2002). Therefore, Mixl1 might function differently from other Mix family proteins. The expression pattern of Mixl1 suggest that Mixl1 do not act cell autonomously to induce endoderm, whereas cmix seems likely to function in a cell autonomous manner. This discrepancy between cmix and Mixl1 turns out to be critical in ES cells. Mixl1 is the only Mix gene being identified to date in mouse. Another gene(s) might substitute the function of Mix in mouse. However, considering that the GC content of both cmix (73%) and Mixl1 (64%) genes are high, it is possible that a yet to be identified mix-like family protein that contains the Smad2 interaction motif might exist.

Use of the ES cells with the lacZ gene targetedly inserted into the pdx-1 locus allows spatial analysis and rapid detection of lacZ-positive pdx-1 expressing cells by X-gal staining. The target insertion overcomes the disadvantages of random integration of the promoter driving reporter gene in terms of fidelity of the promoter. Therefore, this ES cell system provides a reliable tool for dissecting signal pathways for differentiation of ES cells into regional specific gut endoderm.

Our RT-PCR results show up-regulation of pdx-1 expression in undifferentiated cmix transformant ES cells and which did not stained positively by X-gal staining. We interpret this result to mean that the RT-PCR is a much more sensitive way to detect the average RNA expression. In contrast, in situ X-gal staining detects expression over a threshold, and gives spatial expression information of the positive cells. Expression of cmix gene is observed before and during formation of the chick primitive streak, and its expression is switched off when gastrulation is completed and endoderm layer is formed. Expression of cpdx-1 gene is detectable by RT-PCR at the 9- to 10-somite stage (Kumar et al. 2003), when regional specification of gut endoderm starts. The time lag between cmix and cpdx-1 expression suggest that pdx-1 is not directly activated by cmix.

In Fig. 1, we showed that there is no difference between ES cells differentiated for 0, 2 or 8 days in induction of Pdx1-positive cells by cocultivation with pancreatic rudiments. We routinely yield pdx-1/ß-gal positive cells without EB formation. This suggests that EB formation is not required for endodermal differentiation. Mouse ES cells were reported to have differentiated into various cell types, including the VE, through formation of EB (Abe et al. 1996). However, there is evidence that EB formation is not a necessary step for induction of cells of ectodermal and mesodermal origin (Yamashita et al. 2000; Ying et al. 2003). Considering that endodermal cells and mesodermal cells possibly arise from mesendoderm (Rodaway & Patient 2001), which is the common precursor for mesoderm and endoderm, it is possible that EB formation is not required for endodermal differentiation.

Here we found that we could induce ES cells to an endodermal fate by activating TGF-ß pathways, a first step of differentiating ES cells into specific endodermal derived organs. The ES cells we yielded in our culture are pdx-1 positive, insulin, glucagon negative cells, which do not express other duct markers. These cells express transcripts of regional markers of gut endoderm such as Nkx2.2, NeuroD, somatostatin and pancreatic peptide. Further insights into the mechanism of how the endoderm is induced, how pdx-1 expression is turned on in the endodermal field, and how insulin-secreting ß-cells are generated from pancreatic precursor cells, will be useful for generating insulin producing pancreatic ß-cells from ES cells which will be suitable for transplantation of diabetic patients.

	Experimental procedures

Top Abstract Introduction Results Discussion Experimental procedures References

 
ES cell culture and differentiation

The 129/Sv-derived ES cell line R1 with integration of lacZ gene into the pdx-1 locus was provided by Dr Christopher Wright (Vanderbilt University). ES cells were maintained in the presence of mitomycin-C-treated mouse embryonic fibroblast (MEF) feeders in Dulbecco's modified Eagle medium (DMEM) supplemented with LIF (1000 units/mL of ESGRO, Chemicon), 15% fetal bovine serum, 100 µM non-essential amino acids and 2 mM L-glutamine, for a maximum of 12 passages.

Undifferentiated ES cells were passaged on gelatin coated plates once and used for differentiation. For the differentiation studies, ES cells were dissociated with trypsin and cultured in suspension in the absence of LIF in untreated culture dishes (bacteria grade) to allow them to form embryoid bodies (EBs). Except where stated, EBs differentiated for 2 days were trypsinized, 5 x 10⁴ dissociated cells were aliquoted, and used for further differentiation experiments. Aliquots of dissociated EB cells were added with or without embryonic tissues, centrifuged at 270 g for 3 min to achieve close contact of the cells, followed by incubation at 37 °C for up to 1 h, then plated on to 24-well cell culture dishes. The replated EB cells were cultured until day 14, with daily change of medium. On day 14 of differentiation, X-gal stainings were done. Embryonic tissues were prepared by dissecting out the embryonic tissues in phosphate buffered saline (PBS) and torn into pieces mechanically using forceps. For pancreatic mesenchyme isolation, 11.5 dpc embryos were used. In some of the experiments, transgenic mice introduced with a construct with pdx-1 promoter driving eGFP expression (Gu et al. 2004) were used. No pancreas epithelium was contaminated as confirmed by lack of the GFP expression in isolated pancreatic mesenchyme tissues (data not shown).

Growth factors

Recombinant soluble growth factors purchased from R & D System Inc. were used at the concentrations shown below. Recombinant human FGF-2 (10 or 50 ng/mL), FGF-4 (1 ng/mL), FGF-5 (0.1 µg/mL), FGF-6 (1.25 ng/mL), FGF-7 (40 ng/mL), FGF-9 (2.5 ng/mL), TGF- (5 ng/mL), TGF-ß1 (0.2 ng/mL), TGF-ß2 (0.8 ng/mL), epidermal growth factor (EGF) 10 ng/mL, insulin like growth factor (IGF)-1 (5 ng/mL), IGF-2 (10 ng/mL), HGF (50 ng/mL), recombinant mouse FGF-8b (5 ng/mL) and VEGF (5 ng/mL). Activin-ßB (5 U/mL) was provided by Dr K. Symes (Boston University). Growth factors were added to dissociated EB cells after being replated on to 24 well plates on day 2 of differentiation then were cultivated until day 14 of differentiation when X-gal stainings were done.

X-gal staining for detection of ß-galactosidase activity

ES cell cultures were fixed in 4% paraformaldehyde in PBS for 20 min at room temperature, rinsed several times with PBS then stained for ß-galactosidase activity. For X-gal staining, cells were incubated with 1 mg/mL X-gal in buffer containing 20 mM K₃Fe(CN)₆, 20 mM K₄Fe(CN)₆-3H₂O, 2 mM MgCl₂, 0.02% Nonidet-P40, 0.01% deoxycholate in PBS, overnight at 37 °C. The numbers of lacZ-positive cells in each well were counted under a microscope. After the X-gal staining, DNA were extracted from the cells and the quantities determined. Total cell numbers of each well were determined from the DNA standards and the proportion of X-gal positive cells in each well were calculated. Total cell number after differentiation for 12 days ranged from 3 x 10⁵ to 1 x 10⁶ cells. DNA extraction was done according to a standard protocol. The fixed cells were washed once with PBS, lyzed with 200 µL of lysis buffer (0.1 M NaCl, 10 mM Tris-HCl (pH 8.0), 5 mM EDTA, 0.2% SDS, 200 µg/mL Proteinase K), transferred to eppendorf tubes, incubated for 3 h at 55 °C, followed by ethanol precipitation. DNA precipitates were dissolved in TE (10 mM Tris-HCl (pH 8.0), 1 mM EDTA), incubated with 5 µg/mL RNase for 30 min at 37 °C, followed by phenol extraction, ethanol precipitation. DNA was dissolved in TE and OD at 260 nm was measured. DNA prepared from a series of known numbers of differentiated ES cells served as standards which were prepared each time with the samples.

Xenopus embryo microinjection

Xenopus embryo manipulations were done as described (Kume et al. 1997). Adult Xenopus were purchased commercially (Hamamatsu-Seibutsu). Xenopus embryos were obtained by injection with 800 units of human chorionic gonadotropin (Sigma) into both male and female Xenopus. The embryos were transferred to 2% Ficoll in 1 x Steinberg's solution (SS) [SS: 60 mM NaCl, 0.67 mM KCl, 0.34 mM Ca(NO₃)₂, 0.83 mM MgSO₄, 10 mM HEPES pH 7.4], injected with capped RNA, then cultivated in 0.1xSS. Staging was according to Nieuwkoop & Faber (1967). In vitro transcription of RNA was done as described (Kume et al. 1997). Plasmids were linearized and capped mRNAs were synthesized in vitro using mMESSAGE mMACHINE (Ambion), according to manufacturer's instructions.

Embryos were injected in four blastomeres at the 4-cell stage at the animal pole with a dose of 160, 320 or 640 pg cmix RNA per embryo. Animal caps were dissected from stage 8-9 embryos with a tungsten needle in 1 x modified Barth's solution (MBS) [MBS; 88 mM NaCl, 1 mM KCl, 2.4 mM NaHCO₃, 10 mM HEPES (pH 7.5), 0.82 mM MgSO₄, 0.33 mM Ca(NO₃)₂, 0.41 mM CaCl₂] on 1% agarose lined dishes. The explants were cultivated in 1xMBS until siblings reached desired stages. Animal caps were explanted at late blastula stages of 8–9 in 1xMBS and cultured on 1% agarose lined dish until collection.

Reverse transcription-polymerase chain reaction (RT-PCR) analysis

RNA was extracted from ten Xenopus animal cap explants (Kume et al. 1997), 0.8 µg RNA was used for each reverse transcription (RT) reaction. As for extraction of RNA from ES/EB cells, TRIZOL reagent (Gibco BRL) was used, followed by DNase (Sigma) treatment. Four µg of RNA from ES/EB cells was used for RT reaction. These reactions for both Xenopus and mouse were carried out using SuperscriptII reverse transcriptase (Gibco BRL) with oligo dT primers (Gibco BRL), one µL of 1/5 diluted cDNA (1/100 equivalent of the RT product) was used, with the exception for EF1, for PCR analyses.

For each set of primers, cycle numbers, cDNA required for linear ranges in the RT-PCR were determined empirically. The cycle numbers used for each primer set are shown in Tables 1 and 2. With the exception of detection of markers in Xenopus whole embryos as a positive control: 30 cycles were used for detecting XBra, Xgsc, Xlhbox8, IFABP, and for 24 cycles were used for detection of Sox17ß and Edd. The PCR condition is as follows: denaturing at 95 °C for 30 s, annealing at 60 °C for 30 s and extention at 72 °C for 45 s. One fifth of the RT-PCR products was loaded on to 5% non-denaturing PAGE and the gels were overlaid with DNA sensitive dye, SYBRGreenI (Molecular Probes), and analyzed using FluoroImager (Molecular Dynamics).

Expression vectors were constructed through standard protocols. Short cDNAs of chick mix was cloned by PCR and the full-length transcript was isolated from a homemade Hamburger & Hamilton XPS-PEX 1961 stage HH3-5 cDNA library in pSport. In order to construct an expression vector tagged with V5, we used primers designed so that a V5-tag was fused to the C-terminus of cmix. The cmixV5 fragment was subcloned into either pCS2+ (a gift from Dr David Turner) for RNA in vitro transcription or subcloned downstream of the CMV enhancer/chick ß-actin promoter in pCAGGS (a gift from Dr J. Miyazaki). A drug resistant cassette of phosphoglycerokinase (PGK) promoter driving hygromycin gene was also inserted. Undifferentiated ES cells were trypsinized and electroporated following standard protocols with linearized plasmids prepared using Midiprep Kits (Qiagen). One or two days after electroporation, cultures were switched to selective medium (150 µg/mL of hygromycin) and grown further for 1 week. Drug-resistant colonies were picked and expanded. Stable transformants were further confirmed by Western blot analysis or by PCR analysis with specific primers. Mouse Mixl1 cDNA was cloned using PCR. Expression vector for mouse Mixl1 cDNA was constructed similarly to the above description.

Western blot analysis

ES cells or EBs were lyzed in 2xSDS-PAGE sample buffer then boiled for 5 min. Samples were loaded at approximately 5 x 10⁴ cells/lane equivalent on to 15% SDS-PAGE and were transferred to PVDF membrane (Millipore). To detect cmixV5 or Mixl1V5 protein, the membranes were probed with an affinity-purified rabbit anti-V5-horseradish peroxidase (HRP) antibody (Invitrogen) 1: 2000 dilution and visualized using an ECL plus system (Amersham).

	Acknowledgements

 
This work was initiated when SK and CL were in Dr Douglas A. Melton's laboratory at Harvard University. The authors would like to thank Dr Melton for his kind support. We also thank Drs Christopher Wright and Maureen Gannon for pdx-1/LacZ ES cells, Dr Anne Grapin-Botton and Dr Douglas A. Melton for cmix and Mixer cDNAs, Dr Junichi Miyazaki for pCAGGS vector, and Dr David Turner for pCS2+ construct. Special thanks to Drs Anne Grapin-Botton, Jay Rajagopal, Quoqiang Gu, Kazuhiko Kume and Mariko Ohara for critical discussion of this manuscript, and Masako Yamazaki and Aki Hamaji for excellent technical assistance. This work was also supported by grants from the Mitsubishi Foundation to SK, and for 21st Century COE Research ‘Cell Fate Regulation Research and Education Unit,’ and by a Grant-in-Aid for Scientific Research on priority Areas to SK (15039226), from Ministry of Education, Culture, Sports, Science and Technology of Japan.

	Footnotes

 
These authors contributed equally to this work.

Communicated by: Shinichi Aizawa*Correspondence: E-mail: skume{at}kaiju.medic.kumamoto-u.ac.jp

	References

Top Abstract Introduction Results Discussion Experimental procedures References

 
Abe, K., Niwa, H., Iwase, K., et al. (1996) Endoderm-specific gene expression in embryonic stem cells differentiated to embryoid bodies. Exp. Cell Res. 229, 27–34.[CrossRef][Medline]

Ahlgren, U., Jonsson, J. & Edlund, H. (1996) The morphogenesis of the pancreatic mesenchyme is uncoupled from that of the pancreatic epithelium in IPF1/PDX1-deficient mice. Development 122, 1409–1416.[Abstract]

Ahlgren, U., Jonsson, J., Jonsson, L., Simu, K. & Edlund, H. (1998) Beta-cell-specific inactivation of the mouse Ipf1/Pdx1 gene results in loss of the beta-cell phenotype and maturity onset diabetes. Genes Dev. 12, 1763–1768.[Abstract/Free Full Text]

Blyszczuk, P., Czyz, J., Kania, G., et al. (2003) Expression of Pax4 in embryonic stem cells promotes differentiation of nestin-positive progenitor and insulin-producing cells. Proc. Natl. Acad. Sci. U S A 100, 998–1003.[Abstract/Free Full Text]

Conlon, F.L., Lyons, K.M., Takaesu, N., et al. (1994) A primary requirement for nodal in the formation and maintenance of the primitive streak in the mouse. Development 120, 1919–1928.[Abstract]

Crisera, C.A., Rose, M.I., Connelly, P.R., et al. (1999) The ontogeny of TGF-beta1-beta2-beta3, and TGF-beta receptor-II expression in the pancreas: implications for regulation of growth and differentiation. J. Pediatr. Surg. 34, 689–694.[CrossRef][Medline]

David, N.B. & Rosa, F.M. (2001) Cell autonomous commitment to an endodermal fate and behaviour by activation of Nodal signalling. Development 128, 3937–3947.

Derynck, R., Zhang, Y. & Feng, X.H. (1998) Smads: transcriptional activators of TGF-beta responses. Cell 95, 737–740.[CrossRef][Medline]

Dichmann, D.S., Miller, C.P., Jensen, J., Scott Heller, R. & Serup, P. (2003) Expression and misexpression of members of the FGF and TGFbeta families of growth factors in the developing mouse pancreas. Dev. Dyn. 226, 663–674.[CrossRef][Medline]

Ecochard, V., Cayrol, C., Rey, S., et al. (1998) A novel Xenopus mix-like gene milk involved in the control of the endomesodermal fates. Development 125, 2577–2585.[Abstract]

Germain, S., Howell, M., Esslemont, G.M. & Hill, C.S. (2000) Homeodomain and winged-helix transcription factors recruit activated Smads to distinct promoter elements via a common Smad interaction motif. Genes Dev. 14, 435–451.[Abstract/Free Full Text]

Gu, G., Wells, J.M., Dombkowski, D., Preffer, F., Aronow, B. & Melton, D.A. (2004) Global expression analysis of gnen regulatory pathways during endocrine pancreatic development. Development 131, 165–179.[Abstract/Free Full Text]

Hansson, M., Tonning, A., Frandsen, U., et al. (2004) Artifactual insulin release from differentiated embryonic stem cells. Diabetes 53, 2603–2609.[Abstract/Free Full Text]

Hart, A.H., Hartley, L., Sourris, K., et al. (2002) Mixl1 is required for axial mesendoderm morphogenesis and patterning in the murine embryo. Development 129, 3597–3608.[Abstract/Free Full Text]

Henry, G.L., Brivanlou, I.H., Kessler, D.S., Hemmati-Brivanlou, A. & Melton, D.A. (1996) TGF-beta signals and a pattern in Xenopus laevis endodermal development. Development 122, 1007–1015.[Abstract]

Henry, G.L. & Melton, D.A. (1998) Mixer, a homeobox gene required for endoderm development. Science 281, 91–96.[Abstract/Free Full Text]

Hori, Y., Rulifson, I.C., Tsai, B.C., Heit, J.J., Cahoy, J.D. & Kim, S.K. (2002) Growth inhibitors promote differentiation of insulin-producing tissue from embryonic stem cells. Proc. Natl. Acad. Sci. U S A 99, 16105–16110.[Abstract/Free Full Text]

Kikuchi, Y., Trinh, L.A., Reiter, J.F., Alexander, J., Yelon, D. & Stainier, D.Y. (2000) The zebrafish bonnie and clyde gene encodes a Mix family homeodomain protein that regulates the generation of endodermal precursors. Genes Dev. 14, 1279–1289.[Abstract/Free Full Text]

Kim, S.K., Hebrok, M. & Melton, D.A. (1997) Notochord to endoderm signaling is required for pancreas development. Development 124, 4243–4252.[Abstract]

Kubo, A., Shinozaki, K., Shannon, J.M., et al. (2004) Development of definitive endoderm from embryonic stem cells in culture. Development 131, 1651–1662.[Abstract/Free Full Text]

Kumar, M., Jordan, N., Melton, D. & Grapin-Botton, A. (2003) Signals from lateral plate mesoderm instruct endoderm toward a pancreatic fate. Dev. Biol. 259, 109–122.[CrossRef][Medline]

Kume, S., Muto, A., Inoue, T., Suga, K., Okano, H. & Mikoshiba, K. (1997) Role of inositol 1,4,5-trisphosphate receptor in ventral signaling in Xenopus embryos. Science 278, 1940–1943.[Abstract/Free Full Text]

Lammert, E., Cleaver, O. & Melton, D. (2001) Induction of pancreatic differentiation by signals from blood vessels. Science 294, 564–567.[Abstract/Free Full Text]

Lawson, K.A., Meneses, J.J. & Pedersen, R.A. (1991) Clonal analysis of epiblast fate during germ layer formation in the mouse embryo. Development 113, 891–911.[Abstract]

Lemaire, P., Darras, S., Caillol, D. & Kodjabachian, L. (1998) A role for the vegetally expressed Xenopus gene Mix.1 in endoderm formation and in the restriction of mesoderm to the marginal zone. Development 125, 2371–2380.[Abstract]

Lumelsky, N., Blondel, O., Laeng, P., Velasco, I., Ravin, R. & McKay, R. (2001) Differentiation of embryonic stem cells to insulin-secreting structures similar to pancreatic islets. Science 292, 1389–1394.[Abstract/Free Full Text]

Mohn, D., Chen, S.W., Dias, D.C., et al. (2003) Mouse Mix gene is activated early during differentiation of ES and F9 stem cells and induces endoderm in frog embryos. Dev. Dyn. 226, 446–459.[CrossRef][Medline]

Moritoh, Y., Yamato, E., Yasui, Y., Miyazaki, S. & Miyazaki, J. (2003) Analysis of insulin-producing cells during in vitro differentiation from feeder-free embryonic stem cells. Diabetes 52, 1163–1168.[Abstract/Free Full Text]

Nieuwkoop, P.D. & Faber, J. (1967) Normal Table of Xenopus Laevis (Daudin). Amsterdam: North-Holland Publishing Co.

Niwa, H., Yamamura, K. & Miyazaki, J. (1991) Efficient selection for high-expression transfectants with a novel eukaryotic vector. Gene 108, 193–199.[CrossRef][Medline]

Offield, M.F., Jetton, T.L., Labosky, P.A., et al. (1996) PDX-1 is required for pancreatic outgrowth and differentiation of the rostral duodenum. Development 122, 983–995.[Abstract]

Peale, F.V. Jr, Sugden, L. & Bothwell, M. (1998) Characterization of CMIX, a chicken homeobox gene related to the Xenopus gene mix.1. Mech. Dev. 75, 167–170.[CrossRef][Medline]

Pearce, J.J. & Evans, M.J. (1999) Mml, a mouse Mix-like gene expressed in the primitive streak. Mech. Dev. 87, 189–192.[CrossRef][Medline]

Rajagopal, J., Anderson, W.J., Kume, S., Martinez, O.I. & Melton, D.A. (2003) Insulin staining of ES cell progeny from insulin uptake. Science 299, 363.[Free Full Text]

Ramalho-Santos, M., Yoon, S., Matsuzaki, Y., Mulligan, R.C. & Melton, D.A. (2002) ‘Stemness’: transcriptional profiling of embryonic and adult stem cells. Science 298, 597–600.[Abstract/Free Full Text]

Randall, R.A., Germain, S., Inman, G.J., Bates, P.A. & Hill, C.S. (2002) Different Smad2 partners bind a common hydrophobic pocket in Smad2 via a defined proline-rich motif. EMBO J. 21, 145–156.[CrossRef][Medline]

Robb, L., Hartley, L., Begley, C.G., et al. (2000) Cloning, expression analysis, and chromosomal localization of murine and human homologues of a Xenopus mix gene. Dev. Dyn. 219, 497–504.[CrossRef][Medline]

Rodaway, A. & Patient, R. (2001) Mesendoderm. an ancient germ layer? Cell 105, 169–172.[CrossRef][Medline]

Schier, A.F. (2003) Nodal signaling in vertebrate development. Annu. Rev. Cell Dev. Biol. 19, 589–621.[CrossRef][Medline]

Stein, S., Roeser, T. & Kessel, M. (1998) CMIX, a paired-type homeobox gene expressed before and during formation of the avian primitive streak. Mech. Dev. 75, 163–165.[CrossRef][Medline]

Tremblay, K.D., Hoodless, P.A., Bikoff, E.K. & Robertson, E.J. (2000) Formation of the definitive endoderm in mouse is a Smad2-dependent process. Development 127, 3079–3090.[Abstract]

Yamashita, J., Itoh, H., Hirashima, M., et al. (2000) Flk1-positive cells derived from embryonic stem cells serve as vascular progenitors. Nature 408, 92–96.[CrossRef][Medline]

Ying, Q.L., Stavridis, M., Griffiths, D., Li, M. & Smith, A. (2003) Conversion of embryonic stem cells into neuroectodermal precursors in adherent monoculture. Nature Biotechnol. 21, 183–186.[CrossRef][Medline]

Received: 6 January 2005
Accepted: 22 February 2005

This article has been cited by other articles:

				T. Yoshida, N. Shiraki, H. Baba, M. Goto, S. Fujiwara, K. Kume, and S. Kume Expression patterns of epiplakin1 in pancreas, pancreatic cancer and regenerating pancreas. Genes Cells, July 1, 2008; 13(7): 667 - 678. [Abstract] [Full Text] [PDF]

				N. Shiraki, T. Yoshida, K. Araki, A. Umezawa, Y. Higuchi, H. Goto, K. Kume, and S. Kume Guided Differentiation of Embryonic Stem Cells into Pdx1-Expressing Regional-Specific Definitive Endoderm Stem Cells, April 1, 2008; 26(4): 874 - 885. [Abstract] [Full Text] [PDF]

				M. Puceat TGF{beta} in the differentiation of embryonic stem cells Cardiovasc Res, May 1, 2007; 74(2): 256 - 261. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Shiraki, N. Articles by Kume, S. Search for Related Content PubMed