Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Apr 6.
Published in final edited form as: Cell Stem Cell. 2012 Apr 6;10(4):371–384. doi: 10.1016/j.stem.2012.02.024

Self-Renewing Endodermal Progenitor Lines Generated from Human Pluripotent Stem Cells

Xin Cheng 1,2, Lei Ying 1,2, Lin Lu 1,2, Aline M Galvão 1,2, Jason A Mills 1,2, Henry C Lin 4, Darrell N Kotton 5, Steven S Shen 6, M Cristina Nostro 7, John Kim Choi 1, Mitchell J Weiss 3, Deborah L French 1,2, Paul Gadue 1,2,*
PMCID: PMC3580854  NIHMSID: NIHMS364580  PMID: 22482503

SUMMARY

The use of human pluripotent stem cells for laboratory studies and cell-based therapies is hampered by their tumor-forming potential and limited ability to generate pure populations of differentiated cell types in vitro. To address these issues, we established endodermal progenitor (EP) cell lines from human embryonic and induced pluripotent stem cells. Optimized growth conditions were established that allow near unlimited (>1016) EP cell self-renewal in which they display a morphology and gene expression pattern characteristic of definitive endoderm. Upon manipulation of their culture conditions in vitro or transplantation into mice, clonally derived EP cells differentiate into numerous endodermal lineages, including monohormonal glucose-responsive pancreatic β-cells, hepatocytes, and intestinal epithelia. Importantly, EP cells are nontumorigenic in vivo. Thus, EP cells represent a powerful tool to study endoderm specification and offer a potentially safe source of endodermal-derived tissues for transplantation therapies.

INTRODUCTION

Human pluripotent stem cells (PSCs), including embryonic stem cells (ESCs) and induced pluripotent stem cells (iPSCs), hold tremendous promise for both basic biology and cell-based therapies due to their unlimited in vitro proliferation capacity and their potential to generate all tissue types (Murry and Keller, 2008; Stadtfeld and Hochedlinger, 2010). Upon in vitro differentiation, these stem cell populations recapitulate early embryonic development, giving rise to a spectrum of mature cell types (Murry and Keller, 2008).

During embryogenesis, the blastocyst inner cell mass gives rise to an epithelial population known as the epiblast. These cells traverse the primitive streak during gastrulation, giving rise to mesoderm and definitive endoderm (DE) (Lu et al., 2001). The epithelial sheet of nascent DE then folds to form the primitive gut tube consisting of three major domains along the anterior-posterior axis: the foregut, midgut, and hindgut. These domains are further refined into specific regions from which the rudiments of various endodermal organs bud (Zorn and Wells, 2009). The foregut eventually gives rise to esophagus, trachea, lungs, thyroid, parathyroid, thymus, stomach, liver, biliary system, and pancreas, while the midgut and hindgut form the small intestine and colon.

Endodermal-derived tissues, including liver and pancreas, are potentially useful for cell replacement therapies. It is possible to generate DE and its derivative lineages from PSCs in vitro through sequential exposure to cytokines that mimics that in embryonic morphogenesis. In this fashion, hepatic, intestinal, and pancreatic cells can be produced from ESCs and iPSCs (D’Amour et al., 2006; Gouon-Evans et al., 2006; Basma et al., 2009; Spence et al., 2011a). While these studies highlight the promise of PSC-derived endodermal tissues for transplantation therapies, several obstacles remain. Endodermal cells generated from PSCs tend to display immature phenotypes and in many instances are not fully functional. For example, most pancreatic β-cells currently generated in vitro from human ESCs are polyhormonal and not glucose responsive (D’Amour et al., 2006; Nostro et al., 2011). In addition, the pluripotent nature of ESCs and iPSCs results in production of multiple cells types from different germ layers in most differentiation protocols. Thus, it is difficult to produce pure monolineage cultures of a desired cell type from PSCs (Murry and Keller, 2008). Finally, undifferentiated ESCs and iPSCs are tumorigenic and therefore must be completely removed from their derivative tissues to be used for transplantation (Hentze et al., 2009).

To address these issues, we generated self-renewing DE progenitor lines from both human ESCs and iPSCs. These cells, termed EP cells for endodermal progenitor cells, display a proliferative capacity similar to ESCs yet lack teratoma-forming ability. In addition, EP cell lines generate endodermal tissues representing liver, pancreas, and intestine, both in vitro and in vivo. Thus, EP cell lines provide a powerful reagent to study how different gut tissues are specified from a common multipotent endodermal progenitor and to optimize monolineage differentiation. Moreover, creation of EP cells from ESCs/iPSCs may represent a strategy to optimize the production of pure, nontumorigenic cells for tissue replacement therapies.

RESULTS

Identification of a Definitive Endoderm Progenitor Population from Human PSCs

We adapted a step-wise differentiation protocol in serum free media that was previously shown to induce DE and its derivative hepatic lineages from mouse and human ESCs (Gadue et al., 2006; Gouon-Evans et al., 2006). This protocol uses Activin A to induce DE and then BMP4 and basic FGF (bFGF) among other factors to specify a hepatic fate from DE in a process that mimics embryogenesis. Using a variation of this protocol (see Experimental Procedures), we could generate hepatic cells after 2 weeks of induction (data not shown). Interestingly, after 3–4 weeks, these cultures contained two cell populations with distinct morphology (Figure 1A). One population resembled immature hepatocytes, being large and highly vacuolated (Figure 1A, “Differentiated Cells”), while the other population resembled undifferentiated ESC colonies (Figure 1A, “Progenitor Colonies”). These mixed cultures maintained both colony types after passaging for over 9 months (data not shown). We used a pipette to manually isolate colonies of a given morphology and performed gene expression analysis using QRT-PCR (Figure 1B). The ESC markers NANOG and OCT4 were not expressed in either cell type, suggesting that these colonies were not an outgrowth of contaminating ESCs in the culture. The early hepatocyte marker AFP was enriched in the differentiated colonies while the early panendoderm marker SOX17 was enriched in the putative progenitor colonies. The early endoderm/liver/pancreas marker HNF4A (Duncan et al., 1997) was found in both. Based on these findings, we hypothesized that the undifferentiated colonies might represent EP cells undergoing both self-renewal and differentiation into hepatocytes in culture.

Figure 1. A Putative Definitive Endoderm Progenitor Population Is Identified in Differentiation Cultures from Human ESCs.

Figure 1

Open in a new tab

The human ESC line H9 was differentiated toward the hepatic lineage.

(A) Phase contrast image of a culture maintained for 4 weeks shows two distinct types of populations (Differentiated versus Progenitor). The scale bar represents 100 μm. Image was captured with a 20× objective.

(B) Colony types indicated in (A) were manually picked and pooled into groups, and gene expression was analyzed by quantitative real-time RT-PCR, comparing to undifferentiated ESCs. ES, embryonic stem cells; Diff, differentiated cells; Prog, progenitor colonies. The expression of indicated genes is shown normalized to the housekeeping gene PPIG. Values represent the mean of three individual pools of each population. Error bars represent the standard errors.

(C) Cultures were assayed for CD117 (KIT) versus CXCR4, SOX17 versus FOXA1, and SOX17 versus CXCR4 by flow cytometry in three separate stains. See also Table S4.

The bulk differentiation cultures were then examined by flow cytometry for expression of KIT (CD117) and CXCR4, which are expressed on early endoderm (Gouon-Evans et al., 2006; D’Amour et al., 2005). A subpopulation of CXCR4+CD117+ cells (3%–60%) was consistently present in the cultures (Figure 1C and data not shown). These double-positive cells also express the panendoderm marker FOXA1 (Ang et al., 1993) and SOX17, which is expressed transiently in immature endoderm in vivo (Kanai-Azuma et al., 2002).

The CXCR4+CD117+FOXA1+SOX17+ population likely marks putative EP cells from the undifferentiated colonies, which we showed to be enriched for SOX17 mRNA (Figure 1B). To test this, we used FACS to isolate CXCR4+ and CXCR4− cells for further study. The early hepatocyte marker AFP was enriched in the CXCR4− cells while the CXCR4+ population expressed the immature endoderm marker SOX17 and was negative for AFP expression (Figure 2A). Purified CXCR4+ cells could be expanded in culture, but within three passages, reverted to a mixed population containing both SOX17+ and SOX17− cells, indicative of differentiation in culture (Figure 2A, right panel). In contrast, the CXCR4− population had little proliferative capacity, consistent with a more mature phenotype of differentiated hepatocytes in vitro (data not shown).

Figure 2. Purification and Maintenance of Endodermal Progenitor Cells.

Figure 2

Open in a new tab

(A) Cultures containing endoderm progenitors as in Figure 1 were sorted based upon CXCR4 expression. Gene expression was analyzed by QRT-PCR. CXCR4+ cells were cultured as in Figure 1 and after 1 or 3 passages, cultures were assayed for SOX17 versus FOXA1 expression by intracellular flow cytometry.

(B) EP cell lines are established by sorting definitive endoderm (CXCR4+CD117high) cells from day 5 of human ESC differentiation cultures induced with high levels of Activin A, and culturing the cells in optimized conditions (see text). Images show typical morphology of EP cell colonies (H9-EP), captured using 10× and 20× objectives.

(C) EP cell lines derived from H9 ESCs or iPSCs were cultured and cell growth was analyzed by cell count at each passage. Left panel: graph shows relative cell growth over time (Square: iPSC2-derived EP cell line; Circle: H9-derived EP cell line). Right panel: SOX17 versus FOXA1 expression by intracellular flow cytometry after either 24 (H9-EP) or 20 (iPSC2-EP) passages.

See also Figures S1 and S2 and Tables S1–S4.

Our data indicate that self-renewing EP cells can be produced through in vitro manipulation of human ESCs. Differentiation of CXCR4+ EP cells into AFP-expressing hepatocytes was expected in our initial experiments (Figure 1) because the culture conditions were initially formulated to drive hepatocyte differentiation (Gouon-Evans et al., 2006). To better expand the EP cell population while maintaining the FOXA1+SOX17+ phenotype of these cells, we identified culture conditions, which consist of a serum free media containing BMP4, bFGF, EGF, and VEGF plated on matrigel and MEF feeders (see Experimental Procedures). Utilizing matrigel was critical because suspension cultures were unable to maintain EP cells (data not shown).

Using these optimized culture conditions, we developed a simplified protocol for EP cell production. Human ESCs or iPSCs were induced to differentiate in the presence of high Activin A, which promotes endoderm formation (D’Amour et al., 2005). Definitive endoderm cells (CXCR4+CD117+) were purified by cell sorting and cultured directly in EP cell medium (Figure 2B, left panel). When cultured in the optimized conditions, the majority of cells form epithelial structures reminiscent of gut epithelium and contain what appears to be a central lumen, especially at higher cell density (Figure 2B, right panel). Both ESC-derived and iPSC-derived EP cell lines proliferated extensively and displayed a homogenous EP cell phenotype (FOXA1+SOX17+FOXA2+) (Figure 2C and data not shown). The optimized culture conditions could also maintain EP lines generated from 3–4 week liver differentiation cultures as described in Figure 1 (liver-derived EP cell). These cells displayed characteristics similar to those of EP cells derived directly from definitive endoderm (data not shown).

We were able to derive EP cell lines from the human ESC lines H9 and CHB8 and from two human iPSC lines (H9-EP, CHB8-EP, iPS1-EP and iPS2-EP; see Supplemental Experimental Procedures). Endoderm progenitor populations generated from all of these PSC lines have been maintained for more than 20 passages with an expansion of >1016 while maintaining the progenitor phenotype (Figure 2C and data not shown). Thus, self-renewing EP cell lines can be generated reproducibly from human PSCs. Expansion of EP cells is not associated with a “crisis” where cells undergo a high rate of senescence as indicated by a transient flattening of the growth curve (Figure 2C). Moreover, karyotype analyses were performed on EP cells derived from H9 ESCs and one iPSC line and were found to be normal (Figure S1 available online). These data indicate that the high rate of EP cell proliferation is not due to acquired genetic instability or chromosomal rearrangements. Cryopreserved EP cells can be thawed with high viability (data not shown).

Among the cytokines used, BMP4 and FGF signaling were the most critical for EP cell maintenance. Withdrawal of BMP4 led to an almost complete loss of SOX17 expression within 1 week and upregulation of PDX1, indicating pancreatic differentiation (Figure S2A). Withdrawal of bFGF or disruption of the FGF/Ras/ MEK pathway with the small molecule inhibitors PD173074 and PD0325901 reduced proliferation dramatically (Figure S2B), consistent with previous reports that FGF signaling is essential for DE expansion (Morrison et al., 2008). Withdrawal of VEGF or EGF also led to a significant decrease in proliferation (Figure S2B). TGF-β signaling was also found to be essential for maintaining the EP cells because treatment with a small molecule inhibitor of TGF-β signaling, SB431542, completely abolished SOX17 expression in these cells and drastically suppressed proliferation (Figures S2C and S2D). The inhibition of TGF-β signaling in EP cells also inhibited their differentiation into pancreatic β-cells, possibly due to spontaneous differentiation prior to pancreatic induction (Figure S2E).

Nine subclones from the H9-EP line were established by depositing single CXCR4+CD117+ cells into 96-well plates to expand. These clones displayed the same proliferative capacity and marker expression as the parental line (data not shown). Thus, we have established clonal populations of EP cell lines that exhibit extensive proliferative capacity and express markers of early multipotent endoderm.

Characterization of EP Cells

We used gene expression microarrays to compare the transcriptomes of purified H9 ESCs, Activin A-induced day 5 transient endoderm, and H9-EP (Figure 2A) (see Supplemental Experimental Procedures). Cluster analysis of the microarray data revealed that EP cells are more similar to day 5 transient endoderm than they are to ESCs (Figure 3A). Principal component analysis revealed that while EP cells are more similar to transient endoderm than they are to ESCs, they are also distinct from transient endoderm (Figure 3B). As expected, high level expression of the pluripotency markers NANOG, OCT4, SOX2, and DNMT3B was limited to ESCs (Figures 3C and 3D). Interestingly, the proto-oncogene MYCN was expressed in both ESCs and EP cells. MYCN is required for pluripotency and self-renewal of mouse ESCs (Varlakhanova et al., 2010) and may have a similar role in EP cells. Genes expressed in the primitive streak or mesendoderm, including CER1, FGF17, FGF8, GSC, and MIXL1, are expressed in day 5 transient endoderm, but not EP cells, while a subset of genes including EOMES and LHX1 are maintained in both cell types (Figures 3C and 3D). As expected, expression of most endodermal genes including FOXA2, FOXA3, GATA4, and SOX17 was limited to day 5 endoderm and EP cells (Figures 3C and 3D).

Figure 3. Gene Expression Microarray Analyses of EP Cells Compared to ESCs and Transient Definitive Endoderm.

Figure 3

Open in a new tab

(A–C) Purified ESCs (sorted SSEA3high SSEA4high), transient endoderm (day 5 high Activin-treated, sorted CXCR4high CD117high), and EP cells (sorted CXCR4high CD117high) were analyzed using gene expression microarrays. T0, ESCs; T5, transient endoderm; EPC, EP cells. (A) ESC, transient endodermal cell, and EP cell global expression profile clustered by dendrogram. (B) Principal component analysis of populations in (A). (C) Heat map expression levels of genes related to pluripotency, primitive streak/mesendoderm, definitive endoderm, and cell surface markers.

(D) Quantitative RT-PCR confirmation of a subset of genes examined in (C). Expression levels were normalized to 1 in ESCs for pluripotency genes, transient endoderm for primitive streak genes, and EP cells for definitive endoderm genes, respectively. Error bars represent the standard error of three individual pools of each population.

(E) Flow cytometry analyses distinguish EP cells from ESCs and transient endoderm cells by cell surface markers ALCAM, LGR5, and CXCR4. (ESCs: gray filled histograms; transient endoderm: red histogram; EP cells: yellow histograms.)

See also Figure S3 and Tables S1–S4.

The microarray data predicted surface markers that may be used to distinguish EP cells from ESCs and day 5 transient endoderm (Figure 3C). For example, the ESC marker CD9 was depleted in EP cells. Messenger RNAs encoding endodermal markers CXCR4 and CD117 were expressed in both EP cells and transient endoderm (Figure 3C), confirming our flow cytometry studies (Figure 1 and Figure 2). In contrast, activated leukocyte adhesion molecule ALCAM/CD166 (Bowen et al., 1995) was expressed only by EP cells but not day 5 transient endoderm, while CD177/NB1, a glycoprotein expressed by neutrophils (Stroncek, 2007), was reduced in EP cells compared to day 5 endoderm (Figure 3C). The stem cell/progenitor markers CD133, CD34, and LGR5 (Mizrak et al., 2008; Barker et al., 2010) were also expressed at relatively high levels in EP cells. Differential expression of a subset of the cell surface markers predicted by the microarray studies was confirmed by flow cytometry on ESCs, EP cells, and day 5 endoderm, demonstrating that ALCAM and LGR5 are enriched on EP cells (Figure 3E).

Our data indicate that EP cells closely resemble nascent endoderm, yet exhibit key differences in gene expression. The microarray data were analyzed for transcriptional regulators that are enriched in EP cells and therefore might confer unique biological properties to this population. The top 13 hits are shown in Table 1, along with the fold increase in expression over day 5 transient endoderm. GATA3, a GATA family member enriched in mouse DE (Sherwood et al., 2007), ranks first in the list. HEY2, a target of Notch signaling, is enriched in EP cells, suggesting a possible role of Notch signaling in EP cell maintenance. Interestingly, transcription factors known to be important regulators of early liver (TBX3) (Lüdtke et al., 2009), pancreas (ISL1 and RFX6) (Ahlgren et al., 1997; Smith et al., 2010), lung (FOXP2) (Shu et al., 2001), and intestine (ISX1) (Choi et al., 2006) were enriched in EP cells, suggesting the capacity to form these multiple endodermal lineages.

Table 1.

Transcriptional Regulators Enriched in EP Cells

Gene Fold Expression (EP Cell / Transient Endoderm)
GATA3 >10×
TBX3 7.8×
ID2 8.4×
MSX2 8.2×
RFX6 8×
MEIS2 6×
ID1 5.5×
SALL1 4.3×
HEY2 3.4×
FOXP2 3.4×
ISL1 3.1×
LMO7 3.0×
ISX 2.7×
Open in a new tab

Gene expression was analyzed by microarray as described in Figure 3. The top 13 transcriptional regulators that were enriched in EP cells over transient definitive endoderm are shown.

EP Cells Exhibit Multilineage Endodermal Differentiation In Vitro

We next tested whether expanded EP cells retain the ability for multipotent endodermal differentiation upon manipulation of culture conditions. EP cell lines were differentiated under conditions known to promote hepatocyte, pancreatic, or intestinal specification in vitro (Gouon-Evans et al., 2006; Nostro et al., 2011; Spence et al., 2011a). For all induction protocols, EP cells were assumed to be at the DE stage and were stimulated as such. Representative data are shown for a polyclonal H9-EP line and two single-cell-derived subclones (Figure 4, Figure 5, and Figure S4–S6). In addition, hepatocyte and pancreatic differentiation from two iPSC-derived EP cell lines (iPS1-EP and iPS2-EP) were examined with similar results (Figures S4 and S5).

Figure 4. Hepatic Differentiation of EP Cells In Vitro.

Figure 4

Open in a new tab

(A–C) Cells from the H9-EP line and from one single-cell-derived subclone (clone 2) were differentiated into hepatocytes as indicated, and were harvested for analyses at days 14, 20, and 27 (T14, T20, and T27). (A) Gene expression was analyzed by QRT-PCR on samples harvested at three time points from H9-EP and clone 2 lines, comparing to undifferentiated EP cells (T0). The expression of indicated genes is normalized to the housekeeping gene PPIG. Values represent the mean of three independent differentiation experiments. Error bars represent the standard error. (B) Flow cytometry analyses of AFP versus FOXA1, AAT versus FOXA1, or ASGPR1 versus CXCR4 at day 20 of differentiation, comparing to undifferentiated EP cells (T0 EP cell). Percentage of cells within each quadrant is indicated. Shown are the representative data from H9-EP (passage 10) and clone 2 EP (passage 6) cells. (C) To assess CYP3A4 activity of EP cell-derived hepatocytes, H9-EP cells (passage 15) were differentiated into hepatocytes for 24 days and then cultured in the presence (Rifampicin) or absence (DMSO) of Rifampicin for 3 days, and compared to HepG2 cells. Values represent means of measured luminescence units normalized to 6 × 104 cells, and error bars represent standard error (n = 3). See also Figure S4 and Tables S1–S4.

Figure 5. Pancreatic Differentiation of EP Cells In Vitro.

Figure 5

Open in a new tab

(A–D) Cells from the H9-EP line (passages 6–12) and from one single-cell-derived subclone (clone 2) (passages 5–11) were differentiated into pancreatic cells as indicated in the text. Cells were harvested for analyses at day 12 (T12) to day 18. (A) Immunofluorescence staining of human islets (Islet) and day 14 EP cell-derived pancreatic cultures (EP) for c-peptide (red) and GCG (green) or PDX1 (green). Scale bars represent 100 μm. (B) Intracellular flow cytometric analyses of c-peptide versus GCG and c-peptide versus SST at day 12 of differentiation. (C) Gene expression was analyzed by QRT-PCR on samples harvested at day 18 of differentiation and compared to adult islets. The expression of indicated genes is normalized to 1 in islets. Values represent the mean of three independent differentiation experiments or two batches of adult islets. Error bars represent the standard error. (D) The percentage of c-peptide+ cells that coexpress either SST or GCG was quantified. Data are shown for endocrine cells generated from EP cells (H9-EP, n = 12; and iPS2-EP, n = 5) and directly from H9 ESCs (ESC, n = 4).

(E) Intracellular flow cytometric analyses of c-peptide expression, gated on c-peptide+ cells from day 12 EP cell-derived pancreatic cultures (blue histogram) and from human islets (red filled histogram). The values shown are mean fluorescence intensity of c-peptide in H9-EP-derived cells (blue, n = 6) and human islets (red, n = 2).

(F) The H9 EP cells were differentiated into β-cells and stimulated with D-glucose at 2 mM (Basal) or 20 mM (glucose) at day 14, and were compared to adult islets. C-peptide release was measured at various time points (5, 10, 15, 20, and 40 min) (H9-EP, n = 3; islets, n = 2; error bars represent the standard error). Asterisk indicates statistical significance as determined by t test. p = 0.022.

See also Figures S5 and Tables S1–S4.

To induce hepatocyte specification, EP cells were stimulated using a step-wise protocol adapted from previously published studies (see Experimental Procedures and Supplemental Experimental Procedures) (Gouon-Evans et al., 2006; Basma et al., 2009). Hepatocyte specification was monitored at days 14, 20, and 27 of induction by QRT-PCR and at day 20 by flow cytometry. By day 20 most cells exhibited features of hepatocyte morphology including polygonal shape and multinuclearity (Figure S4A). QRT-PCR analysis revealed upregulation of the hepatic markers alpha-1-antitrypsin (AAT), alpha-fetoprotein (AFP), Albumin (ALB), CYP3A4, CYP3A7, and glucose-6-phosphatase (G6PC) (Figures 4A and S4B). Intracellular flow cytometry at day 20 of differentiation showed that more than 90% of cells were AFP+ and more than 80% of cells expressed AAT, indicating commitment to the hepatocyte fate (Figures 4B and S4C). The mature hepatocyte marker ASGPR1 (Basma et al., 2009) was expressed on about 20% of the cells while the EP cell marker CXCR4 was absent (Figure 4B). EP cells generated from two iPSC lines also differentiated into hepatocytes (Figure S4B). When treated with Rifampicin, EP cell-derived hepatocytes exhibited inducible CYP3A4 activities comparable to HepG2 cells (Figures 4C and S4D). Thus, EP cells consistently exhibit hepatocyte potential.

Endoderm components of liver and pancreas are thought to arise from the putative common foregut precursor (Spence et al., 2009). To investigate the pancreatic potential of EP cells, we modified the protocol reported by Nostro et al. (2011) (see Supplemental Experimental Procedures). Immunofluorescence analysis of differentiated EP cultures showed colonies that contain c-peptide+ but no GLUCAGON+ (GCG+) cells as compared to primary human islets that express both of these hormones (Figure 5A). Colonies generated from EP cells were also PDX1+, indicative of pancreatic specification (Figure 5A). PDX1 expression was confirmed by flow cytometric analyses at days 10–11 of differentiation in the H9-EP line (not shown), single-cell derived subclones, and two iPSC-derived EP cell lines (Figure S5A). In addition, approximately 5%–30% of the differentiated EP cells (H9 and iPSC derived) expressed c-peptide by day 14 of pancreatic induction (Figures 5B and S5A). The efficiency of c-peptide+ cell generation was variable depending on the genetic background of the cell line (Figures 5B and S5A). However, c-peptide+ cells generated from all EP cell lines, regardless of genetic background, were negative/low for GCG and SOMATOSTATIN (SST) expression (Figures 5A, 5B, and S5A). We also examined the differentiation cultures by QRT-PCR. Consistent with pancreatic differentiation, expression of PDX1, NKX6-1, and NEUROD1 were strongly induced (Figures 5C and S5C), while the expression of the EP cell marker SOX17 declined (Figure S5C). Transient expression of NEUROG3 (NGN3) in the EP cell differentiation cultures is indicative of endocrine specification (Figure S5C). Insulin (INS) RNA was robustly induced with levels approximately 20% of that found in adult islets. When these data are corrected for the percentages of c-peptide+ cells in the samples (islet ~60%, H9-EP 15% ± 3%, H9-EP clone 2 22% ± 4%) (Figure 5C), the levels of INS RNA in EP cell-derived β-cells are estimated to be ~70% of that found in primary β-cells. This is consistent with protein levels of c-peptide in primary β-cells and EP cell-derived pancreatic cells as determined by intracellular flow cytometry (Figure 5E). Notably, MAFA, a critical transcription factor expressed in mature β-cells, was upregulated in the EP cell differentiation cultures (Figure 5C) to 30% of the level of RNA found in adult islets. If the same correction for INS expression is used, then both MAFA and NKX6-1 may be expressed in EP cell-derived c-peptide+ cells at comparable levels to those in adult β-cells.

Importantly, c-peptide+ cells derived from EP cells exhibit minimal expression of GCG and relatively low expression of SST (Figures 5B and 5C). This contrasts with the majority of reports describing the generation of c-peptide+ cells from human PSCs, in which most pancreatic endocrine progeny display a polyhormonal phenotype (D’Amour et al., 2006; Nostro et al., 2011; Rezania et al., 2011). Polyhormonal pancreatic cells are found during mouse development but do not contribute to the β-cell population in the adult mouse (Herrera, 2000). In agreement with this, H9 ESCs exposed to the same pancreatic differentiation protocol produced a greater proportion of polyhormonal cells compared to pancreatic endocrine cells derived from EP cells (Figure 5D). EP cells generated a c-peptide+ population with less than 2% of the cells coexpressing GCG and ~5% of the cells coexpressing SST. In contrast, c-peptide+ cells generated directly from ESCs were 38% ± 9% GCG+ and >20% ± 1% SST+ (Figure 5D). Thus, c-peptide+ cells generated from multiple EP cell lines are more similar to mature β-cells than populations generated directly from PSCs.

Previous reports have shown that polyhormonal cells derived from ESCs are not fully functional and cannot efficiently respond to glucose stimulation by releasing insulin (D’Amour et al., 2006). A time course analysis of glucose-induced c-peptide release was performed, comparing EP cell (H9-EP and iPS2-EP)-derived pancreatic differentiation cultures to primary human islets (Figures 5F and S5D). Data were collected over 40 min of stimulation with 20 mM glucose and was normalized, setting the 5 min basal stimulation time point for each culture as 1. The kinetics of c-peptide release was similar between the EP cell-derived cultures and primary human islets, with a consistent 3-fold stimulation index in the high glucose group over basal (low glucose) conditions. EP cell-derived cultures were statistically indistinguishable from primary islets, with the exception of the 10 min time point. When the cells were challenged again after 3 or 18 hr, they remained responsive to glucose stimulation in a similar manner as in the first challenge (data not shown), indicating that these cells maintain glucose-responsive function. Assuming that all of the proinsulin is processed into c-peptide and insulin is at a 1:1 molar ratio, the c-peptide release assay revealed an estimated secretion of 9 ± 1.7 ng insulin per 105 EP cell-derived c-peptide+ cells (n = 3), which is ~20% of the secretion of adult islet β-cells (estimated at 37 ng/105 cells, n = 2; and Lukowiak et al., 2001). These findings demonstrate that EP cell-derived β-cells are functionally responsive to the physiologic stimulus (D-glucose) for insulin release in a manner similar to that of adult islets in vitro.

To compare the differentiation capacities of the EP cells derived from day 5 transient endoderm and the EP cells derived from liver differentiation cultures of ESCs, liver-derived EP cells were allowed to differentiate in the aforementioned liver or pancreas conditions. These cells generated hepatocyte progenitors that expressed AFP and AAT (Figure S4C) and pancreatic cells that expressed PDX1 and c-peptide, but not GCG (Figure S5A, bottom panel), with efficiencies similar to the EP cells derived from transient endoderm. These data suggest that EP cells are a robust stem cell population that can be generated from multiple stages of differentiation while maintaining multipotency.

Liver and pancreas are of foregut origin. We next determined whether EP cells could produce intestinal epithelia, which arises from mid/hindgut regions. Using a protocol similar to that of Spence et al. (2011a), we induced EP cells to undergo intestinal differentiation (see Supplemental Experimental Procedures). Typical “organoids” indicative of intestinal differentiation were identified upon day 30 of induction (Figure S6A). These cultures expressed CDX2 and KLF5, transcription factors enriched in mid/hindgut lineages (Spence et al., 2011a), and the Paneth cell marker Lysozyme (LYZ) (Figure S6B). In contrast, the intestinal stem cell marker LGR5, which is also expressed on EP cells, was downregulated after 30 days of induction to levels similar to those in adult intestine. Intracellular flow cytometric analysis revealed that 85%–90% of cells expressed CDX2 at day 30 of differentiation, similar to the intestinal tumor cell line CACO-2 (Figure S6C). These data suggest that EP cells have intestinal potential.

To determine if EP cell developmental potential was restricted to endodermal lineages, cultures were induced with conditions established to drive ESCs toward either neuroectoderm (Greber et al., 2011) or mesoderm (J.M., L.Y., and X.C., unpublished data) (see Supplemental Experimental Procedures). The upregulation of neuroectoderm markers ZIC1, SOX1, or PAX4 (Figure S3A), and mesoderm markers MIXL1, T, or PECAM (Figure S3B), was not observed, suggesting that EP cells are committed to the endoderm germ layer.

EP Cells Lack Tumorigenicity and Form Endodermal Tissues In Vivo

The propensity for ESCs and iPSCs to form teratomas represents a major impediment to transplantation studies. To determine if EP cells retain tumor-forming potential, 0.5 million H9 ESCs or H9-EPs were transplanted intramuscularly into immune-compromised mice. After 4–6 weeks, all mice injected with H9 ESCs developed tumors, while mice injected with an equal number of EP cells did not (Figure 6A). In fact, it was difficult to detect any injected EP cells under these conditions. Therefore, larger numbers (8 to 10×106) of EP cells were transplanted in concentrated matrigel containing growth factors to promote cell survival and recruit host blood vessels (see Experimental Procedures). Under these conditions, no EP cell-derived tumors developed in 35 transplantations (Figure 6B) over a period of 3–60 weeks. In contrast, when the cells from day 5 transient endoderm differentiation cultures of ESCs were transplanted in this system, all animals developed teratomas. Surprisingly, even purified CXCR4+CD117+ DE cells isolated by cell sorting from day 5 ESC differentiation cultures still generated tumors within 6 weeks of transplantation (Figure 6B). It is possible that CXCR4+CD117+ DE cells may not be fully committed to endoderm and still retain teratoma-forming ability, or that a small number of contaminating ESCs present even after cell sorting led to tumor formation. In addition, when the cells from day 20 liver differentiation cultures of ESCs were transplanted, four out of five animals still developed teratomas (Figure 6B). These data indicate that EP cells are less tumorigenic than ESCs, iPSCs, transient endoderm, and ESC-derived hepatic cultures.

Figure 6. Tumor Potential and Spontaneous Differentiation of EP cells In Vivo.

Figure 6

Open in a new tab

(A) To determine if EP cells retain teratoma forming potential, 0.5 million H9 ESCs or H9 EP cells were transplanted intramuscularly into immune compromised mice. Images show teratoma formation in a mouse injected with ESCs (ES, left panel) and lack of tumor formation in a mouse injected with EP cells (EP, right panel).

(B) A summary of teratoma-forming ability of pluripotent stem cells (ES/iPSC), day 5 (T5) transient endoderm, ESC-derived day 20 hepatocyte cultures (ES-derived hepatocytes), and EP cells (EPC) and EP cell-derived day 9 hepatocyte cultures (EP cell-derived hepatocytes). For transient endoderm, both bulk cultures and CXCR4+CD117high sorted cells were used for transplantation.

(C–F) Analyses of EP cell transplants. Eight to ten million EP cells mixed with matrigel were injected subcutaneously into immunodeficient mice, and the resultant matrigel plugs were isolated 3–12 weeks posttransplantation. Scale bars represent 100 μm. (C) Sections of matrigel plugs were stained with hematoxylin and eosin. Shown are endoderm-like structures at lower (10×, left panel) and higher magnifications (20×, right panel). (D) Immunohistochemistry using an anti-human antibody reveals human-cell-derived endodermal structures surrounded by nonhuman mesenchyme and adipocytes. (E) Immunohistochemistry for intestinal markers reveals gut/intestinal structures in the matrigel plug. (F) Immunohistochemistry for hepatocyte markers in the matrigel plug.

See also Figure S6 and Tables S1–S3.

Next, we examined transplanted matrigel-EP cell plugs for evidence of cellular differentiation. By 3–8 weeks after transplantation, cells formed various endodermal tissues, ranging from epithelium-forming cysts and tubes that appear similar to early gut tube endoderm (Figure 6C, top panels) to larger, more complex structures containing differentiated cells (Figure 6C, bottom panels; and data not shown). These included intestinallike structures with epithelial cells surrounding a central lumen (Figure 6C, bottom panels) and hepatoblast-like structures (data not shown). Staining with a pantissue human-specific antibody demonstrated that these endodermal structures arose from the transplanted EP cells, in contrast to surrounding fat and mesenchyme, which was murine host derived (Figure 6D).

Immunohistochemistry with human-specific antibodies was used to further characterize the developmental potential of transplanted EP cells. All EP cell-derived morphologically endodermal structures expressed FOXA1 and FOXA2, consistent with an endodermal origin (Figure 6E and data not shown). The structures resembling intestinal epithelia expressed the transcription factors HNF4A and CDX2, key regulators of intestinal homeostasis (Spence et al., 2011b). These cells also stained for the intestinal epithelial markers human fatty acid binding protein (IFABP), VILLIN1, and MUCIN2 (Figure 6E). In addition, immunohistochemistry revealed that the hepatoblast/hepatocyte-like structures harbor cells positive for AFP and AAT (Figure 6F). These data clearly demonstrate that EP cells are able to differentiate into gut epithelia and liver in vivo.

DISCUSSION

The ability to generate stem cell lines that self-renew in vitro and differentiate into various mature tissues after manipulation of culture conditions or transplantation into animals has revolutionized biology and offers great promise for medical applications. In this regard, it is likely that different stem cell lines with distinct developmental potentials will be exploited for unique applications. For example, ESCs, trophectoderm stem cells (TSCs), and extraembryonic endoderm stem cells (XENCs), generated from three distinct cell types of the mouse blastocyst, display developmental potentials similar to those of their respective progenitors (Tanaka et al., 1998; Kunath et al., 2005). Here we describe the establishment of continuously replicating, clonal endoderm-committed stem cell lines from human ESCs and iPSCs. These cells, termed EP cells, can self-renew and specifically generate endodermal lineages from both the foregut and mid/hindgut.

EP cells exhibit several unique features in comparison to other endodermal progenitors previously reported. Hepatocyte stem cell populations can be derived from adult liver (Schmelzer et al., 2006). In addition, a possible hepatic stem cell population was generated from human ESCs (Zhao et al., 2009). These two cell types differentiate into liver lineages only, in contrast to the more broad developmental potential exhibited by EP cells. Another report indicates that SOX17 overexpression in human ESCs generates a progenitor population that expresses genes indicative of both ESCs and endoderm and can differentiate into endoderm derivatives (Séguin et al., 2008). These cells express ESC markers such as OCT4 and NANOG and are not endoderm-committed because they generate mesoderm-containing teratomas. A cell population similar to EP cells but derived from mouse ESCs has been reported (Morrison et al., 2008). Perhaps the most distinguishing feature of human EP cells described here is their extensive proliferative capacity. Mouse endoderm progenitors were only reported to expand ~2,000-fold in culture (Morrison et al., 2008), whereas human EP cells exhibit virtually unlimited self-renewal (>1016) (Figure 2 and data not shown). In contrast to the current study, mouse EP cells were not analyzed at the clonal level and were tested only for hepatic and pancreatic development and may therefore be restricted in developmental potential. In addition, EP cells can be maintained as a homogenous SOX17+FOXA1+ undifferentiated cell population. This is a critical point because partially differentiated EP cell cultures (see Figure 1 and Figure 2A) cannot be reproducibly differentiated into either hepatocytes or β-cells at the efficiencies reported here, starting with pure undifferentiated EP cultures (data not shown and Figure 5 and Figure 6). This is a well-known phenomenon also seen with directed differentiation of ESCs and may be due to the inability of partially differentiated cells to be respecified down a different developmental path. It will be important to systematically compare the proliferative and developmental capacities of EP cells to various other endoderm progenitor populations in order to better define their biological features and utility for downstream applications.

The ability of EP cells to generate glucose-responsive monohormonal insulin-expressing cells in vitro is promising (Figure 5). Most attempts to generate functional β-cells from human PSCs in culture have failed to generate glucose-responsive cells (D’Amour et al., 2006; Nostro et al., 2011). Those studies that have reported glucose responsiveness have low levels of c-peptide+ cells, have not carefully examined the percentage of cells in a population with the abnormal polyhormonal phenotype (Jiang et al., 2007; Zhang et al., 2009; Thatava et al., 2011), or have some combination thereof. It will be important in the future to compare β-cells generated using different methodologies to determine which generates the most robust functional cell type. In agreement with these prior studies, we also found that human ESCs stimulated to undergo pancreatic differentiation produce polyhormonal β-cells (data not shown and Figure 5D). It is possible that production of functionally superior β-cells from EP cells reflects biological features related to the developmental timing of pancreatic endoderm production in vivo. During normal human embryogenesis, β-cells are not generated until ~10 weeks after endoderm specification (Spence and Wells, 2007), while in ESC differentiation cultures this typically occurs in ~2 weeks (D’Amour et al., 2006). It is possible that this accelerated timeframe in ESC differentiation cultures precludes the establishment of essential transcriptional networks and/or epigenetic modifications required for proper β-cell formation. Further analysis and comparison of ESC endoderm and EP cell differentiation cultures may reveal specific genes and epigenetic modifiers that regulate optimal formation of functional β-cells.

Our experiments indicate that EP cells lack intrinsic mesoderm and ectodermal potential. Directed differentiation protocols developed for mesoderm or ectoderm were unable to induce gene expression profiles specific to these lineages when applied to EP cells, suggesting an endoderm-restricted program (Figure S3). Moreover, in vivo transplanted EP cells exclusively generate endodermal structures that stained with a human-specific antibody (Figure 6D). In addition, pathological analysis of 24 independent transplants did not detect any morphologically ectodermal tissues (data not shown). This lineage restriction makes EP cells a powerful platform to dissect the signals necessary for the specification of endodermal cell types.

Flow cytometry and microarray analyses reveal a unique gene expression pattern in human EP cells, distinct from undifferentiated ESCs or CXCR4+CD117+ transient endoderm. LHX1 and EOMES, while not typically defined as mature gut tube markers, are maintained in EP cells (Figure 3). This raises the possibility that these transcription factors may play a role in EP cell fate or self-renewal. These data along with the expression of markers typical of cells undergoing specification of various endodermal lineages (Table 1) suggest that EP cells are not gut tube endoderm “frozen in time,” but are a distinct in vitro stem cell population. Investigation of these differences could identify key regulators of endoderm maintenance and development.

In summary, we describe a simple culture procedure to reproducibly and efficiently generate endoderm stem cell lines from human PSCs. Resultant EP cells self-renew rapidly and can be stimulated to form hepatic, pancreatic, and intestinal tissues. Moreover, EP cells are nontumorigenic, reflecting their potential utility for tissue replacement therapies. Our work challenges the notion that ESCs/iPSCs must be used as a starting point for directed tissue differentiation studies. Rather, EP cells serve as an intermediate between iPSCs/ESCs and mature endodermal derivatives. The ability to generate functional monohormonal β-cells from iPSCs will enable in vitro modeling of diseases of the β-cell, including multiple genetic forms of diabetes and hyperinsulinemias. These lines will provide innovative experimental platforms to investigate mechanisms of endodermal differentiation and a safer, more efficient starting point for tissue replacement therapies aimed at common human disorders including liver failure and diabetes.

EXPERIMENTAL PROCEDURES

Human PSC Culture and Differentiation

Human ESC lines H9 and CHB8 were obtained from the National Stem Cell Bank and Massachusetts Human Stem Cell Bank, respectively. The human iPSC lines iPSC1 (CHOP_WT1.2) and iPSC2 (CHOP_WT2.2) were derived from wild-type human fibroblasts (J.M., L.Y., and X.C., unpublished data, and see Supplemental Experimental Procedures). For generating day 5 transient endoderm cells, human PSCs were differentiated in the serum free differentiation (SFD) media (Gadue et al., 2006) as either embryoid bodies (EBs) (Gouon-Evans et al., 2006) or monolayer cultures (D’Amour et al., 2006; Nostro et al., 2011). Further details are described in the Supplemental Experimental Procedures.

Establishment and Maintenance of EP Cell Lines

EP cells were first purified from day 27 hepatocyte differentiation cultures of H9 ESCs by sorting for CXCR4+/CD117high cells and culturing them in hepatic induction media on matrigel (BD Biosciences) with mouse embryonic feeder (MEF) cells. The optimal culture conditions for maintaining EP cells were established by culturing cells on growth-factor-reduced matrigel and MEF cells ([0.5 × 106]/58 cm2) and eliminating cytokines that promote hepatocyte differentiation. This medium is composed of SFD-based media supplemented with BMP4 (50 ng/ml), bFGF (10 ng/ml), VEGF (10 ng/ml), and EGF (10 ng/ml) (see more details in Supplemental Experimental Procedures). EP cells from other PCS lines were established from day 5 transient endoderm by sorting for CXCR4+/CD117high cells and maintaining them as described above in a 5% CO2, 5% O2, 90% N2 environment.

Differentiation of EP Cells

The hepatic, pancreatic, and intestinal in vitro differentiation protocols were adapted from previously published reports (Gouon-Evans et al., 2006; Nostro et al., 2011; Spence et al., 2011a). Detailed protocols can be found in Supplemental Experimental Procedures.

Teratoma/Transplantation Assay

For the intramuscular injection, 0.5 × 106 H9 ESCs or EP cells were dissociated, mixed with 250 μl matrigel, and transplanted intramuscularly into the leg of SCID/Beige immunodeficient mice. Teratoma formation was monitored over a period of 4–8 weeks. In the following experiments, 5 to 10 × 106 cells were embedded in 300 μl high-concentration matrigel (BD Biosciences) that was supplemented with a combination of cytokines (see details in Supplemental Experimental Procedures), and they were injected subcutaneously into the neck of SCID/Beige mice. Teratoma formation was monitored over a period of 3–60 weeks. Resultant matrigel plugs/transplants were surgically removed from the mice, fixed in 10% formaldehyde, embedded in paraffin, and analyzed using hematoxylin and eosin stains for morphology or immunohistochemistry for specific antigens as described in Supplemental Experimental Procedures. All animal work was approved by the institutional IACUC committee.

Supplementary Material

01

ACKNOWLEDGMENTS

We thank Dr. Gordon Keller for sharing his pancreatic differentiation protocol ahead of publication. We also thank members of the Gadue and French groups for critical discussion of the manuscript. X.C. was supported by a Ruth L. Kirschstein Institutional Training Grant (T32).

Footnotes

ACCESSION NUMBERS The microarray data for this experiment can be accessed at NCBI Gene Expression Omnibus with accession number GSE35461.

SUPPLEMENTAL INFORMATION Supplemental Information for this article includes six figures, four tables, and Supplemental Experimental Procedures and can be found with this article online at doi:10.1016/j.stem.2012.02.024.

REFERENCES

  1. Ahlgren U, Pfaff SL, Jessell TM, Edlund T, Edlund H. Independent requirement for ISL1 in formation of pancreatic mesenchyme and islet cells. Nature. 1997;385:257–260. doi: 10.1038/385257a0. [DOI] [PubMed] [Google Scholar]
  2. Ang SL, Wierda A, Wong D, Stevens KA, Cascio S, Rossant J, Zaret KS. The formation and maintenance of the definitive endoderm lineage in the mouse: involvement of HNF3/forkhead proteins. Development. 1993;119:1301–1315. doi: 10.1242/dev.119.4.1301. [DOI] [PubMed] [Google Scholar]
  3. Barker N, Huch M, Kujala P, van de Wetering M, Snippert HJ, van Es JH, Sato T, Stange DE, Begthel H, van den Born M, et al. Lgr5(+ve) stem cells drive self-renewal in the stomach and build long-lived gastric units in vitro. Cell Stem Cell. 2010;6:25–36. doi: 10.1016/j.stem.2009.11.013. [DOI] [PubMed] [Google Scholar]
  4. Basma H, Soto-Gutiérrez A, Yannam GR, Liu L, Ito R, Yamamoto T, Ellis E, Carson SD, Sato S, Chen Y, et al. Differentiation and transplantation of human embryonic stem cell-derived hepatocytes. Gastroenterology. 2009;136:990–999. doi: 10.1053/j.gastro.2008.10.047. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bowen MA, Patel DD, Li X, Modrell B, Malacko AR, Wang WC, Marquardt H, Neubauer M, Pesando JM, Francke U, et al. Cloning, mapping, and characterization of activated leukocyte-cell adhesion molecule (ALCAM), a CD6 ligand. J. Exp. Med. 1995;181:2213–2220. doi: 10.1084/jem.181.6.2213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Choi MY, Romer AI, Hu M, Lepourcelet M, Mechoor A, Yesilaltay A, Krieger M, Gray PA, Shivdasani RA. A dynamic expression survey identifies transcription factors relevant in mouse digestive tract development. Development. 2006;133:4119–4129. doi: 10.1242/dev.02537. [DOI] [PubMed] [Google Scholar]
  7. D’Amour KA, Agulnick AD, Eliazer S, Kelly OG, Kroon E, Baetge EE. Efficient differentiation of human embryonic stem cells to definitive endoderm. Nat. Biotechnol. 2005;23:1534–1541. doi: 10.1038/nbt1163. [DOI] [PubMed] [Google Scholar]
  8. D’Amour KA, Bang AG, Eliazer S, Kelly OG, Agulnick AD, Smart NG, Moorman MA, Kroon E, Carpenter MK, Baetge EE. Production of pancreatic hormone-expressing endocrine cells from human embryonic stem cells. Nat. Biotechnol. 2006;24:1392–1401. doi: 10.1038/nbt1259. [DOI] [PubMed] [Google Scholar]
  9. Duncan SA, Nagy A, Chan W. Murine gastrulation requires HNF-4 regulated gene expression in the visceral endoderm: tetraploid rescue of Hnf-4(−/−) embryos. Development. 1997;124:279–287. doi: 10.1242/dev.124.2.279. [DOI] [PubMed] [Google Scholar]
  10. Gadue P, Huber TL, Paddison PJ, Keller GM. Wnt and TGF-beta signaling are required for the induction of an in vitro model of primitive streak formation using embryonic stem cells. Proc. Natl. Acad. Sci. USA. 2006;103:16806–16811. doi: 10.1073/pnas.0603916103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Gouon-Evans V, Boussemart L, Gadue P, Nierhoff D, Koehler CI, Kubo A, Shafritz DA, Keller G. BMP-4 is required for hepatic specification of mouse embryonic stem cell-derived definitive endoderm. Nat. Biotechnol. 2006;24:1402–1411. doi: 10.1038/nbt1258. [DOI] [PubMed] [Google Scholar]
  12. Greber B, Coulon P, Zhang M, Moritz S, Frank S, Müller-Molina AJ, Araúzo-Bravo MJ, Han DW, Pape HC, Schöler HR. FGF signalling inhibits neural induction in human embryonic stem cells. EMBO J. 2011;30:4874–4884. doi: 10.1038/emboj.2011.407. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Hentze H, Soong PL, Wang ST, Phillips BW, Putti TC, Dunn NR. Teratoma formation by human embryonic stem cells: evaluation of essential parameters for future safety studies. Stem Cell Res. (Amst.) 2009;2:198–210. doi: 10.1016/j.scr.2009.02.002. [DOI] [PubMed] [Google Scholar]
  14. Herrera PL. Adult insulin- and glucagon-producing cells differentiate from two independent cell lineages. Development. 2000;127:2317–2322. doi: 10.1242/dev.127.11.2317. [DOI] [PubMed] [Google Scholar]
  15. Jiang J, Au M, Lu K, Eshpeter A, Korbutt G, Fisk G, Majumdar AS. Generation of insulin-producing islet-like clusters from human embryonic stem cells. Stem Cells. 2007;25:1940–1953. doi: 10.1634/stemcells.2006-0761. [DOI] [PubMed] [Google Scholar]
  16. Kanai-Azuma M, Kanai Y, Gad JM, Tajima Y, Taya C, Kurohmaru M, Sanai Y, Yonekawa H, Yazaki K, Tam PP, Hayashi Y. Depletion of definitive gut endoderm in Sox17-null mutant mice. Development. 2002;129:2367–2379. doi: 10.1242/dev.129.10.2367. [DOI] [PubMed] [Google Scholar]
  17. Kunath T, Arnaud D, Uy GD, Okamoto I, Chureau C, Yamanaka Y, Heard E, Gardner RL, Avner P, Rossant J. Imprinted X-inactivation in extra-embryonic endoderm cell lines from mouse blastocysts. Development. 2005;132:1649–1661. doi: 10.1242/dev.01715. [DOI] [PubMed] [Google Scholar]
  18. Lu CC, Brennan J, Robertson EJ. From fertilization to gastrulation: axis formation in the mouse embryo. Curr. Opin. Genet. Dev. 2001;11:384–392. doi: 10.1016/s0959-437x(00)00208-2. [DOI] [PubMed] [Google Scholar]
  19. Lüdtke TH, Christoffels VM, Petry M, Kispert A. Tbx3 promotes liver bud expansion during mouse development by suppression of cholangiocyte differentiation. Hepatology. 2009;49:969–978. doi: 10.1002/hep.22700. [DOI] [PubMed] [Google Scholar]
  20. Lukowiak B, Vandewalle B, Riachy R, Kerr-Conte J, Gmyr V, Belaich S, Lefebvre J, Pattou F. Identification and purification of functional human beta-cells by a new specific zinc-fluorescent probe. J. Histochem. Cytochem. 2001;49:519–528. doi: 10.1177/002215540104900412. [DOI] [PubMed] [Google Scholar]
  21. Mizrak D, Brittan M, Alison MR. CD133: molecule of the moment. J. Pathol. 2008;214:3–9. doi: 10.1002/path.2283. [DOI] [PubMed] [Google Scholar]
  22. Morrison GM, Oikonomopoulou I, Migueles RP, Soneji S, Livigni A, Enver T, Brickman JM. Anterior definitive endoderm from ESCs reveals a role for FGF signaling. Cell Stem Cell. 2008;3:402–415. doi: 10.1016/j.stem.2008.07.021. [DOI] [PubMed] [Google Scholar]
  23. Murry CE, Keller G. Differentiation of embryonic stem cells to clinically relevant populations: lessons from embryonic development. Cell. 2008;132:661–680. doi: 10.1016/j.cell.2008.02.008. [DOI] [PubMed] [Google Scholar]
  24. Nostro MC, Sarangi F, Ogawa S, Holtzinger A, Corneo B, Li X, Micallef SJ, Park IH, Basford C, Wheeler MB, et al. Stage-specific signaling through TGFβ family members and WNT regulates patterning and pancreatic specification of human pluripotent stem cells. Development. 2011;138:861–871. doi: 10.1242/dev.055236. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Rezania A, Riedel MJ, Wideman RD, Karanu F, Ao Z, Warnock GL, Kieffer TJ. Production of functional glucagon-secreting α-cells from human embryonic stem cells. Diabetes. 2011;60:239–247. doi: 10.2337/db10-0573. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Schmelzer E, Wauthier E, Reid LM. The phenotypes of pluripotent human hepatic progenitors. Stem Cells. 2006;24:1852–1858. doi: 10.1634/stemcells.2006-0036. [DOI] [PubMed] [Google Scholar]
  27. Séguin CA, Draper JS, Nagy A, Rossant J. Establishment of endoderm progenitors by SOX transcription factor expression in human embryonic stem cells. Cell Stem Cell. 2008;3:182–195. doi: 10.1016/j.stem.2008.06.018. [DOI] [PubMed] [Google Scholar]
  28. Sherwood RI, Jitianu C, Cleaver O, Shaywitz DA, Lamenzo JO, Chen AE, Golub TR, Melton DA. Prospective isolation and global gene expression analysis of definitive and visceral endoderm. Dev. Biol. 2007;304:541–555. doi: 10.1016/j.ydbio.2007.01.011. [DOI] [PubMed] [Google Scholar]
  29. Shu W, Yang H, Zhang L, Lu MM, Morrisey EE. Characterization of a new subfamily of winged-helix/forkhead (Fox) genes that are expressed in the lung and act as transcriptional repressors. J. Biol. Chem. 2001;276:27488–27497. doi: 10.1074/jbc.M100636200. [DOI] [PubMed] [Google Scholar]
  30. Smith SB, Qu HQ, Taleb N, Kishimoto NY, Scheel DW, Lu Y, Patch AM, Grabs R, Wang J, Lynn FC, et al. Rfx6 directs islet formation and insulin production in mice and humans. Nature. 2010;463:775–780. doi: 10.1038/nature08748. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Spence JR, Wells JM. Translational embryology: using embryonic principles to generate pancreatic endocrine cells from embryonic stem cells. Dev. Dyn. 2007;236:3218–3227. doi: 10.1002/dvdy.21366. [DOI] [PubMed] [Google Scholar]
  32. Spence JR, Lange AW, Lin SC, Kaestner KH, Lowy AM, Kim I, Whitsett JA, Wells JM. Sox17 regulates organ lineage segregation of ventral foregut progenitor cells. Dev. Cell. 2009;17:62–74. doi: 10.1016/j.devcel.2009.05.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Spence JR, Mayhew CN, Rankin SA, Kuhar MF, Vallance JE, Tolle K, Hoskins EE, Kalinichenko VV, Wells SI, Zorn AM, et al. Directed differentiation of human pluripotent stem cells into intestinal tissue in vitro. Nature. 2011a;470:105–109. doi: 10.1038/nature09691. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Spence JR, Lauf R, Shroyer NF. Vertebrate intestinal endoderm development. Dev. Dyn. 2011b;240:501–520. doi: 10.1002/dvdy.22540. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Stadtfeld M, Hochedlinger K. Induced pluripotency: history, mechanisms, and applications. Genes Dev. 2010;24:2239–2263. doi: 10.1101/gad.1963910. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Stroncek DF. Neutrophil-specific antigen HNA-2a, NB1 glycoprotein, and CD177. Curr. Opin. Hematol. 2007;14:688–693. doi: 10.1097/MOH.0b013e3282efed9e. [DOI] [PubMed] [Google Scholar]
  37. Tanaka S, Kunath T, Hadjantonakis AK, Nagy A, Rossant J. Promotion of trophoblast stem cell proliferation by FGF4. Science. 1998;282:2072–2075. doi: 10.1126/science.282.5396.2072. [DOI] [PubMed] [Google Scholar]
  38. Thatava T, Nelson TJ, Edukulla R, Sakuma T, Ohmine S, Tonne JM, Yamada S, Kudva Y, Terzic A, Ikeda Y. Indolactam V/GLP-1-mediated differentiation of human iPS cells into glucose-responsive insulin-secreting progeny. Gene Ther. 2011;18:283–293. doi: 10.1038/gt.2010.145. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Varlakhanova NV, Cotterman RF, deVries WN, Morgan J, Donahue LR, Murray S, Knowles BB, Knoepfler PS. myc maintains embryonic stem cell pluripotency and self-renewal. Differentiation. 2010;80:9–19. doi: 10.1016/j.diff.2010.05.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Zhang D, Jiang W, Liu M, Sui X, Yin X, Chen S, Shi Y, Deng H. Highly efficient differentiation of human ES cells and iPS cells into mature pancreatic insulin-producing cells. Cell Res. 2009;19:429–438. doi: 10.1038/cr.2009.28. [DOI] [PubMed] [Google Scholar]
  41. Zhao D, Chen S, Cai J, Guo Y, Song Z, Che J, Liu C, Wu C, Ding M, Deng H. Derivation and characterization of hepatic progenitor cells from human embryonic stem cells. PLoS ONE. 2009;4:e6468. doi: 10.1371/journal.pone.0006468. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Zorn AM, Wells JM. Vertebrate endoderm development and organ formation. Annu. Rev. Cell Dev. Biol. 2009;25:221–251. doi: 10.1146/annurev.cellbio.042308.113344. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01
NIHMS364580-supplement-01.doc (32.5MB, doc)

RESOURCES