Stem Cells
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

First published online March 23, 2006
Stem Cells Vol. 24 No. 7 July 2006, pp. 1738 -1749
doi:10.1634/stemcells.2005-0367; www.StemCells.com
© 2006 AlphaMed Press

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Figures
Right arrow All Versions of this Article:
2005-0367v1
24/7/1738    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Reprints/Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ta, M.
Right arrow Articles by Lumelsky, N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ta, M.
Right arrow Articles by Lumelsky, N.

TISSUE-SPECIFIC STEM CELLS

The Defined Combination of Growth Factors Controls Generation of Long-Term-Replicating Islet Progenitor-Like Cells from Cultures of Adult Mouse Pancreas

Malancha Ta, Yong Choi, Fouad Atouf, Cheol Hong Park, Nadya Lumelsky

Islet and Autoimmunity Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Department of Health and Human Services, Bethesda, Maryland, USA

Key Words. Adult stem cells • Bone morphogenetic protein-4 • Progenitors • Diabetes • Pancreatic islets • Proliferation • Differentiation

Correspondence: Nadya Lumelsky, Ph.D.,National Institute of Dental and Craniofacial Research/NIH, 45 Center Drive, Building 45, Room 4N 24J, Bethesda, Maryland 20892-6402, USA.Telephone: 301-594-7703;Fax: 301-480-8318;email: nadyal{at}nidcr.nih.gov

Received August 8, 2005; accepted for publication March 20, 2006.
First published online in STEM CELLS EXPRESS   March 23, 2006.


   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 Acknowledgments
 References
 
Application of pancreatic islet transplantation to treatment of diabetes is severely hampered by the inadequate islet supply. This problem could in principle be overcome by generating islet cells from adult pancreas in vitro. Although it is possible to obtain replicating cells from cultures of adult pancreas, these cells, when significantly expanded in vitro, progressively lose pancreatic-specific gene expression, including that of a "master" homeobox transcription factor Pdx1. Here we show for the first time that long-term proliferating islet progenitor-like cells (IPLCs) stably expressing high levels of Pdx1 and other genes that control early pancreatic development can be derived from cultures of adult mouse pancreas under serum-free defined culture conditions. Moreover, we show that cells derived thus can be maintained in continuous culture for at least 6 months without any substantial loss of early pancreatic phenotype. Upon growth factor withdrawal, the IPLCs organize into cell clusters and undergo endocrine differentiation of various degrees in a line-dependent manner. We propose that our experimental strategy will provide a framework for developing efficient approaches for ex vivo expansion of islet cell mass.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 Acknowledgments
 References
 
Adult pancreatic islet cells, and in particular insulin-producing ß-cells, can replicate to a limited extent in vivo [1], and ß-cells are capable of partial regeneration following damage [2]. Although these properties indicate that adult pancreas can provide a source for new ß-cells, efficient ex vivo protocols for generating new ß-cells have not yet been developed. In fact, it is presently unclear which adult pancreatic cell type (mature ß-cells, pancreatic ductal cells, acinar cells, or still unidentified pancreatic stem cells) is best suited for generation of new ß-cells. This issue remains highly controversial [3–6]. Regardless of the mechanism, however, it is evident that obtaining clinically relevant ß-cell mass from ex vivo cultures of adult pancreas will require expansion of ß-cells or stem/progenitor cells followed by their differentiation. It is known that during development, all pancreatic cells arise from endodermal progenitors [7]. If such cells could be obtained and expanded in vitro, they would provide a potential source for generation of new ß-cells.

In this work, we show that long-term-replicating islet progenitor-like cells (IPLCs) uniformly expressing high levels of endodermal and pancreatic genes, and Notch pathway-associated genes can be obtained from cultures of adult mouse pancreas. Derivation of IPLCs is strictly dependent on a combination of basic fibroblast growth factor (bFGF), leukemia inhibitory factor (LIF), and bone morphogenetic protein-4 (BMP-4). The IPLCs can be maintained in culture for at least 6 months without a substantial loss of endodermal/pancreatic progenitor phenotype and can also be induced to undergo partial endocrine differentiation.


   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 Acknowledgments
 References
 
Isolation of Islet-Enriched Fractions and Derivation of Islet Progenitor-Like Cells
Mouse islet-enriched fractions (IEFs) (FVB/NJ3 and CD1/C57BL) were isolated from pancreas perfused with collagenase V (Sigma-Aldrich, St. Louis, http://www.sigmaaldrich.com) using described protocols. Depending on the preparation, the ß-cell content of the IEFs ranged from 50%–60% (determined by staining with the Zn2+ binding agent, diphenyl thiocarbazone [dithizone]; Sigma-Aldrich). The cells were cultured as described in ref. 8, with some modifications (Fig. 1). Specifically, the IEFs (500 cell clusters per 60-mm dish) were plated in Dulbecco’s modified Eagle’s medium (25 mM glucose) with 10% fetal bovine serum (both from Invitrogen, Carlsbad, CA, http://www.invitrogen.com) in tissue culture grade plates precoated overnight with bovine fibronectin (2 µg/ml; Sigma-Aldrich) and concanavalin A (ConA, 1 µg/ml; Sigma-Aldrich). After 48 hours in serum-containing medium, 50%–80% of the IEF cell clusters attached to tissue culture plates. At this time, the serum-containing medium was replaced with serum-free ITSFn medium (26 mM glucose) [9] containing 20 ng/ml bFGF (R&D Systems Inc., Minneapolis, http://www.rndsystems.com) and 1,400 U/ml LIF (Chemicon, Temecula, CA, http://www.chemicon.com). A serum-free medium was used throughout the remaining culture period. BMP-4 (R&D Systems) was added on days 5–7 of culture at a final concentration 10 ng/ml. At the time of BMP-4 addition, the ITSFn medium [9] was replaced with serum-free N2 medium (26 mM glucose) [9] containing B27 supplement (Invitrogen) and 1,400 U/ml LIF. During the course of these studies, we tested BMP-4 concentrations ranging from 2–50 ng/ml, and we found 10 ng/ml to be optimal for IPLC derivation. On days 10–12 of culture, the cells were detached from culture dishes by mild trypsin digestion (5 minutes, 37°C, 0.05% trypsin/0.53 mM EDTA; Invitrogen). The viable cell count was determined using trypan blue (Bio Whittaker, Walkersville, MD), and the cells were replated at a density of 1 x 104 cells per cm2 in N2 medium on Fb/ConA plates as above, but without bFGF. During IPLC derivation (Fig. 1, stage I) and IPLC expansion (Fig. 1, stage II), the cells were passaged every 6–7 days at constant cell density (1 x 104 per cm2). For IPLC differentiation (Fig. 1, stage III), the IPLCs were first allowed to expand under stage II conditions in the presence of BMP-4 and LIF until they reached confluency. This period was followed by incubation for an additional 1–2 weeks with no growth factors in a N2/B27 medium that contained nicotinamide (Sigma-Aldrich) at a final concentration 10 mM.


Figure 1
View larger version (39K):
[in this window]
[in a new window]
  		Figure 1. Scheme of optimized culturing protocol for generation of IPLCs. Details of the experimental protocol are given in Materials and Methods. Abbreviations: bFGF, basic fibroblast growth factor; BMP, bone morphogenetic protein; FN, fibronectin; IEF, islet-enriched fraction; IPLC, islet progenitor-like cell; LIF, leukemia inhibitory factor.

 
In the experiments with {gamma}-secretase inhibitor L-685458 (Calbiochem, La Jolla, CA, http://www.emdbiosciences.com), the inhibitor was dissolved in dimethyl sulfoxide (DMSO) and added to the cells at a final concentration 4–8 µM. The DMSO vehicle was added to control samples.

In the experiments with BMP-4 inhibitor noggin (R&D Systems), noggin was added at 500 ng/ml (50-fold excess over BMP-4 concentration). Depending on the experiment, noggin was added at stage I or to established IPLCs at stage II.

RNA Isolation, cDNA Synthesis, and Reverse Transcription-Polymerase Chain Reaction Analysis
Total cellular RNA was isolated using an RNeasy kit (Qiagen, Valencia, CA, http://www1.qiagen.com). To eliminate traces of genomic DNA, the RNA samples were treated with DNA-free DNase according to manufacturer instructions (Ambion, Austin, TX, http://www.ambion.com). The RNA was reverse-transcribed into cDNA using Superscript II reverse transcriptase (Invitrogen) and random hexamer primers (Invitrogen), according to the manufacturer’s instructions.

Semiquantitative Polymerase Chain Reaction (PCR).   PCRs were performed with 3 ng of the input cDNA in 25-µl reactions containing Taq DNA polymerase, primers, dNTP mix, and a buffer (all from Invitrogen). To ensure semiquantitative reverse transcription (RT)-PCR analysis, we chose the input amount of cDNA by first normalizing the cDNA concentration in different cDNA samples, using the relative expression of an internal control, 18S rRNA. To determine the optimal cycle number in the linear range of PCR amplification, for each set of the primers, the initial PCRs were first carried out at several different cycle numbers. The expected fragment sizes, annealing temperatures, and optimal PCR cycle numbers are shown in Table 1. After the initial denaturation step (94°C, 2 minutes), the PCRs were carried out as follows: denaturation, 94°C for 30 seconds; primer annealing, 1 minute; extension, 72°C, for 1 minute; final extension, 72°C for 7 minutes. The PCR products were resolved on 1.5% or 2% agarose gels, depending on their size.


View this table:
[in this window]
[in a new window]
  		Table 1. Parameters of semiquantitative RT-PCR

 

Quantitative PCR.   Three nanograms of cDNA were used as a template for TaqMan amplifications. The amplification reactions (25 µl) consisting of cDNA templates, Universal PCR Master Mix, and primers/probes (PE Applied Biosystems, Branchburg, NJ) were carried out in the Prism 7900HT sequence detection system (Applied Biosystems). Mouse 18S rRNA was used as an internal control in all assays. The fold change in expression level for each set of time points was calculated using the formula 2–({Delta}{Delta}CT), where {Delta}{Delta}CT is the cycle threshold difference between the time points corrected for 18S rRNA. All PCRs were performed in duplicates. The results are presented as means of the duplicate measurements.

Immunocytochemisty
Depending on the primary antibody, the cells were fixed with 4% paraformaldehyde in phosphate-buffered saline (PBS) for 20 minutes at room temperature, treated with 100% ice-cold methanol for 10 minutes following paraformaldehyde fixation, or treated with 100% ice-cold methanol for 5 minutes without paraformaldehyde fixation. If not treated with methanol, the cells were permeabilized in 0.3% Triton X-100 and blocked in PBS/10% normal goat or donkey serum for 1 hour at room temperature; if treated with methanol, the cells were blocked with PBS/10% normal goat or donkey serum, omitting Triton X-100. The following primary antibodies were used at the indicated dilutions: C-peptide rabbit polyclonal, 1:100 (Linco Research, St. Charles, MO, http://www.lincoresearch.com); PDX1, (a) rabbit polyclonal, 1:2,000 (gifts from Joel Habener, Harvard University; and Chris Wright, Vanderbilt University) and (b) mouse monoclonal, 1:2,000 (contributed by ß-Cell Biology Consortium); HNF-4{alpha} mouse monoclonal, 1:500 (R&D Systems); Foxa2 rabbit polyclonal, 1:5,000 (a gift from Jeffrey Whitsett, Cincinnati Children’s Hospital Medical Center); E-cadherin mouse monoclonal, 1:200 (BD Pharmingen, San Diego, http://www.bdbiosciences.com/pharmingen); and {alpha}-smooth muscle actin ({alpha}-SMA) mouse monoclonal, 1:400 (Sigma-Aldrich). For detection of primary antibodies, Alexa Fluor 488-conjugated (green; Molecular Probes Inc., Eugene, OR, http://probes.invitrogen.com) and CY-3-conjugated (red; Jackson Immunoresearch Laboratories, West Grove, PA, http://www.jacksonimmuno.com) secondary antibodies were used according to the methods recommended by the manufacturers. The negative controls were performed by omitting the primary antibodies from the reactions.

Immunoblotting
Whole cell lysates were prepared in lysis buffer containing 10 mM Tris-HCl (pH 7.5), 150 mM NaCl, 0.1% SDS, 1% Triton X-100, 1% sodium deoxycholate, and a cocktail of protease inhibitors (Roche Applied Sciences, Indianapolis). Twenty to 25 µg of whole cell lysate protein was fractionated on 12% (hairy and enhancer of split 1 [Hes1], Pdx1, and SMA) or 8% (E-cadherin) SDS-polyacrylamide gel electrophoresis, and immunoblotting was carried out using standard protocols. The following primary antibodies were used at the indicated dilutions: goat anti-Hes1 polyclonal, 1:200 (gift from Yuh Nung Jan, University of California San Francisco); rabbit anti-PDX1 polyclonal, 1:1,000 (contributed by ß-Cell Biology Consortium); mouse anti-{alpha}-SMA monoclonal, 1:1,000 (Sigma-Aldrich); mouse anti-E cadherin monoclonal, 1:1,000 (BD Biosciences Pharmingen). The immunoblots were reacted with the appropriate horseradish peroxidase-conjugated secondary antibodies (Santa Cruz Biotechnology Inc., Santa Cruz, CA, http://www.scbt.com) and developed using ECL Western Blotting Detection System (Amersham Biosciences, Piscataway, NJ, http://www.amersham.com).

Derivation of Clonal IPLC Lines
We initiated clonal cultures at the time of appearance of first morphologically identifiable IPLCs at the end of Stage I (day in vitro 25). The cells were trypsinized, diluted to five cells per ml in 50% conditioned medium from the nonclonal cultures/50% fresh N2 medium, containing BMP-4/LIF (final concentration of BMP-4, 10 ng/ml; LIF, 1,400 U/ml), and plated in 96-well plates (Falcon, BD Labware, NJ) precoated with Fn/ConA. The wells containing single cells were marked, and these wells were followed to detect the emerging colonies. The culture medium was replaced every other day. After approximately 2 weeks, individual colonies were dispersed by trypsin and transferred into a Fb/ConA precoated 24-well plate (Falcon) for further expansion.

Quantitative Chromatin Immunoprecipitation (ChiP) Assays with Anti-Acetyl-Histone H3 Antibody
The ChiP assays were performed essentially as described [10, 11]. To generate stage III IPLCs (Fig. 1), the cells were cultured at high cell density without the growth factors for 10 days. To generate stage II IPLCs, the cells were cultured at approximately 30% confluency for 10 days with the growth factors in the medium. The confluent monolayers of NIH3T3 fibroblasts and 70% confluent insulin producing MIN6 insulinoma cells (both from American Type Culture Collection, Manassas, VA, http://www.atcc.org) were used as negative and positive controls, respectively. Approximately 1 x 107 cells of each type were used per assay. The ChiP co-immunoprecipitation was carried out with the rabbit polyclonal anti-acetyl-histone H3 antibody (Upstate Biotechnology, Waltham, MA) or control rabbit IgG (Santa Cruz Biotechnology). Two ng of a plasmid DNA containing ß-galactosidase gene was added at the final elution step to control for recovery of co-immunoprecipitated promoter fragments during washing steps.

The precipitated promoter fragments were quantified by real-time PCR (ABI Prism; ABI Biosystems) using continuous SYBR Green I (Molecular Probes) monitoring [11]. The PCR primers were as follows: ß-galactosidase sense, 5'-TCC AGA TAA CTG CCG TCA CTC CAA C-3'; ß-galactosidase antisense, 5'-TCA ATC CGC CGT TTG TTC CCA C-3'; mouse insulin promoter 1 sense, 5'-TAC CTT GCT GCC TGA GTT CTG C-3'; mouse insulin promoter 1 antisense, 5'-GCA TTT TCC ACA TCA TTC CCC-3'. The PCR-amplified mouse insulin promoter 1 fragment is located –463 to –299 base pairs relative to the transcriptional start site. The PCRs were carried out using JumpStart Taq polymerase (Sigma-Aldrich), 0.3x SYBR green I dye, and 200 nM each primer, using standard protocols. All PCRs were performed in triplicate. The PCR threshold cycle values were first normalized by ß-galactosidase. Then the relative H3-acetylation (R) values were calculated using the equation R = 2(input – H3)/ 2(input – IgG), where input is the normalized threshold cycle value for the control samples subjected to cross-linking and reverse cross-linking, but not to immunoprecipitation, and H3 and IgG are normalized threshold values for anti-acetyl-histone H3 antibody and rabbit IgG, respectively. The results are presented as means ± standard deviations of triplicate measurements. Similar results were obtained in at least three independent experiments. The p values were obtained using t tests comparing the cycle threshold values in the corresponding samples.

Transplantation of IPLCs into NOD/SCID Mice
Nonobese diabetic/severe combined immunodeficient (NOD/SCID) immunodeficient mice were purchased from Jackson Laboratory (Bar Harbor, ME, http://www.jax.org). Cell transplantation was carried out as described previously [12]. Briefly, an incision was made to expose the kidney. IPLCs suspended in the culture medium (1 x 106 to 2 x 106) were injected beneath the renal capsule using a 27-gauge butterfly needle. After injection, the kidney was returned to its original location, and the incision was closed using a double closure technique.


   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 Acknowledgments
 References
 
IPLCs Express a Wide Range of Pancreatic and Endodermal Genes
We have previously found that a wide range of stem cell-specific mRNAs are induced when adult mouse islet-enriched fractions are cultured for several weeks in a serum-free medium in the presence of bFGF and LIF [8]. Although these results suggested to us that developmentally primitive progenitor-like cells might be generated in IEF cultures, we were unable to expand these primitive cells; after several passages, the cultures became quiescent. In an attempt to obtain replicating progenitor-like cells, we decided to incorporate BMP-4 into our original culturing protocol [8]. We chose BMP-4 because BMP-4 and other members of the transforming growth factor-ß (TGF-ß) family are thought to play a functional role in pancreatic endoderm determination during development [13–15].

We found that all three growth factors (bFGF, LIF, and BMP-4) were required for the outgrowth of long-term-replicating IPLCs (Figs. 1, 2A, 2E; supplemental online Fig. 2). Moreover, obtaining continuously replicating cells required addition of BMP-4 several days after addition of bFGF and a timely removal of bFGF. When noggin, an extracellular inhibitor of BMP-4 [16], was added to the culture (Materials and Methods), no long-term cell expansion was observed. This implies that BMP-4 is specifically required for derivation of the continuously replicating cells. The early stage I was characterized by a drastic downregulation of the mRNAs coding for endodermal/early pancreatic transcription factors Pdx1, Foxa2, Hnf1{alpha}, Hnf1ß, and Hnf4{alpha} [17] (Fig. 2B, 2C). Insulin and epithelial marker E-cadherin mRNAs [18] were also downregulated. Conversely, mesenchymal mRNAs ({alpha}-SMA [19] and snail [20]) were strongly upregulated (Fig. 2D). Also upregulated was neural progenitor marker nestin mRNA [21] (Fig. 2D). This upregulation of nestin is consistent with our previous work where we found that pancreatic nestin-expressing cells coexpress additional neural markers [8]. Other investigators have found that pancreatic mesenchyme of apparently non-neural origin also expresses nestin [22, 23]. It is thus likely that nestin-expressing cells in the present work might be composed of both neural and non-neural cell types.


Figure 2
View larger version (68K):
[in this window]
[in a new window]
  		Figure 2. (A): Typical growth kinetics of the IEF culture. Emergence of rapidly proliferating cells marks the stage I to stage II transition. Details of the culturing protocol are given in Materials and Methods and Figure 1. (B, C): At the time of stage I to stage II transition, the islet progenitor-like cells (IPLCs) coordinately activate endodermal, pancreatic, and epithelial genes. Results of semiquantitative reverse transcription-polymerase chain reaction (RT-PCR) (B) and real-time RT-PCR (C) were normalized by expression of 18S rRNA. Results in (C) for all the examined markers are presented as fold increase in the level of expression of a given marker compared with that in day 0 IEFs. The expression level of each marker on day 0 is arbitrarily set at 1. All real-time PCRs were performed in duplicates. The results are presented as means of the duplicate measurements. (D): At the time of stage I to stage II transition, the IPLCs coordinately repress expression of mesenchymal genes. Results of real-time RT-PCR were normalized by expression of 18S rRNA. Results are presented as in (C). Fold increase in expression levels for SMA and Snail are shown on the left vertical axis, and that for nestin is shown on the right vertical axis. (E): Cells arising in the absence and in the presence of LIF have drastically different morphology. Shown are phase contrast images (Zeiss Axiovert 200 microscope) of stage II cells emerging in the presence of LIF (top panel) and in the absence of LIF (bottom panel). Both cultures were of the same age and were cultured under identical conditions, except for the presence or absence of LIF. (F): Pdx1-expressing IPLCs become predominant species in stage II cultures. Compare the left (late stage I) and right (stage II) panels. Images were obtained with Zeiss Axiovert 2 Plus microscope. Details of immunocytochemistry are given in Materials and Methods. Abbreviations: D, day; LIF, leukemia inhibitory factor; P, passage; SMA, smooth muscle actin.

 
At the end of stage I, we observed the emergence of morphologically identifiable rapidly proliferating cells. These small cells, 10–20 µm in diameter, expressed Pdx1, Foxa2, Hnf1{alpha}, Hnf1ß, and Hnf4{alpha} (Figs. 2, 3). The appearance of these rapidly proliferating cells coincided with the inhibition of mesenchymal genes (Fig. 2D). As the IPLCs became the predominant species in culture, the overall level of endodermal/pancreatic genes continued to increase and that of mesenchymal genes continued to fall (Fig. 2B–2D, 2F; in Fig. 2F compare the left and the right panels). We found that BMP-4 inhibitor noggin decreased cell proliferation and Pdx1 gene expression in stage II IPLCs (supplemental online Fig. 1). It is noteworthy that with increasing passage number, the IPLCs gradually became nearly independent of BMP-4 for proliferation and pancreatic phenotype maintenance (unpublished results; also, see below). We observed this phenomenon after the IPLC lines were passaged four to six times; the exact passage number was cell line-specific.


Figure 3
View larger version (38K):
[in this window]
[in a new window]
  		Figure 3. Immunocytochemical analysis of pancreatic/endodermal transcription factors in islet progenitor-like cell (IPLC) cultures. Each horizontal panel shows a typical image of the same split microscopic field. To reveal nuclear localization of the transcription factors, nuclear stain DAPI is shown on the left and is omitted on the right. Cell fixation protocols differed for different antibody combinations (additional details in Results and Materials and Methods). The cells were fixed as follows: (A): paraformaldehyde/methanol; (B): methanol; (C): paraformaldehyde/methanol; and (D): methanol. The images were obtained with Zeiss Axiovert 2 Plus microscope. Details of immunocytochemistry are given in Materials and Methods. Abbreviation: DAPI, 4,6-diamidino-2-phenylindole.

 
In stage II IPLCs, the transcription factors Pdx1, Foxa2, Hnf1{alpha}, Hnf1ß, and Hnf4{alpha} were all expressed at levels at least as high or significantly higher than in time 0 IEFs (Fig. 2B, 2C; Table 2). Whereas Pdx1, Foxa2, Hnf1{alpha}, Hnf1ß, and Hnf4{alpha} were all stably expressed over multiple passages for at least 6 months of culture, Hnf6, a transcription factor normally expressed during pancreatic development but not in mature ß-cells [24], appeared only transiently during the course of derivation of four out of five IPLC lines examined. These results suggest that development of IPLCs in vitro may recapitulate at least some of the steps of pancreatic development in vivo. On the other hand, Ptf1a, another transcription factor associated with normal pancreatic development [25], was not expressed in any of the IPLC lines. Despite their pancreatic phenotype, the IPLCs expressed only low levels of insulin: 0.01%–0.02% of normal ß-cell level.


View this table:
[in this window]
[in a new window]
  		Table 2. Pdx-1 and Foxa2 mRNA expression in 11 independently derived stage 2 IPLC lines

 
To examine whether individual IPLCs coexpressed pancreatic genes, we performed dual immunostaining with Pdx1-, Foxa2-, and Hnf4{alpha}-specific antibodies (Fig. 3). Our results demonstrate a close overlap between Foxa2 and Pdx1 gene expression. Moreover, both transcription factors were appropriately localized to the nuclei of the cells (Fig. 3A). A predominantly nuclear pattern was also observed for HNF4{alpha} (Fig. 3B). Notably, we encountered certain technical difficulties when performing HNF4{alpha}/Foxa2 and HNF4{alpha}/Pdx1 double immunostaining (Fig. 3C, 3D). We found, in particular, that the Hnf4{alpha} antibody reaction was sensitive to the cell fixation protocol—it required methanol fixation without paraformaldehyde (details in Materials and Methods; compare Fig. 3B and 3D with Fig. 3C). Conversely, Pdx1 and Foxa2 antibodies performed significantly better with the paraformaldehyde fixation step included. This made it difficult to obtain high-quality images for HNF4{alpha}/Foxa2 and HNF4{alpha}/Pdx1 combinations. Note, for example, the diffused Pdx1 pattern of staining with methanol fixation (Fig. 3D). On the other hand, Pdx1 appears to be nuclear with cells fixed by paraformaldehyde/methanol (Fig. 3A). Also, note the diffused Hnf4{alpha} with paraformaldehyde/methanol fixation (Fig. 4C) and nuclear Hnf4{alpha} with cells fixed by methanol (Fig. 3B). Despite these technical difficulties, our results strongly suggest that individual IPLCs coexpress several pancreatic/endodermal transcription factors.


Figure 4
View larger version (42K):
[in this window]
[in a new window]
  		Figure 4. Activation of Notch pathway in IPLC cultures. (A–C): A broad range of Notch pathway-associated genes is induced in the IEF cultures. (A): Results of semiquantitative reverse transcription-polymerase chain reaction (RT-PCR); 18S rRNA was used as an internal control. (B): Results of real-time RT-PCR were normalized by expression of 18S rRNA. Results are presented as fold increase in the level of Pdx1 and Hes1 gene expression compared with that in day 0 IEFs. The expression levels of Pdx1 and Hes1 on day 0 are arbitrarily set at 1. All the PCRs were performed in duplicates. The results are presented as means of the duplicate measurements. (C): Results of Western blot analysis of Pdx1 and Hes1 in IPLC cultures (passage 4 after derivation) were quantified by densitometry (bottom panel) and were plotted relative to day 0 IEFs. Details of the Western blot protocol are given in Materials and Methods. (D): BMP-4 induces Hes1 gene expression. Scheme of the experiment is shown on the left side of the panel. The results of real-time RT-PCR analysis of Hes1 gene expression are shown on the right. The results were normalized by expression of 18S rRNA and are shown relative to Hes1 expression in Stage II IPLCs at the start of the experiment, which is arbitrarily set at 1. All the PCRs were performed in duplicate. The results are presented as means of the duplicate measurements. The concentrations of {gamma}-secretase inhibitor L-685458 are displayed above the corresponding plots. The identity of the plotted samples is specified by the numbered bars; these numbers correspond to the numbers on the diagram on the left. Abbreviations: BMP, bone morphogenetic protein; D, day; IPLC, islet progenitor-like cell; P, passage.

 
In all the IPLC lines we observed, a reciprocal shift from the mesenchymal to the epithelial pattern of gene expression during transition from stage I to stage II. The extent of this shift, however, varied between the individual lines. For example, in some lines, the level of E-cadherin in stage II IPLCs was below that in day 0 IEFs. Accordingly, in "low E-cadherin" lines, snail, SMA, and nestin were expressed at higher levels than in 0 IEFs. We found that the morphology (Fig. 2E) and the mRNA expression pattern (supplemental online Fig. 2A) of the cells generated in the presence and in the absence of LIF were different. Namely, in the absence of LIF, stage I cultures ceased proliferation after four to six passages, mesenchymal to epithelial change in gene expression pattern was not observed, and IPLCs were not generated. Interestingly, although LIF was required for IPLC derivation, it was not necessary for the maintenance of pancreatic phenotype of stage II IPLCs (supplemental online Fig. 2B).

Multiple Notch Pathway-Associated Genes Are Induced in IPLC Cultures
It is known that Notch signaling controls progenitor cell proliferation and maintenance during pancreatic development [26, 27]. Notch activation results in the induction of a basic helix-loop-helix (bHLH) transcriptional repressor Hes1. Hes1 inhibits two key pro-endocrine transcriptional activators, neurogenin3 (ngn3) and NeuroD, thus preventing endocrine differentiation [17]. The initiation of endocrine differentiation results, at least in part, from repression of Hes1. This repression relieves inhibition of ngn3 and NeuroD, leading to activation of insulin and other islet-specific genes. Consistent with this model, it has recently been found that constitutive activation of Notch pathway in developing mouse pancreas "traps" progenitors in an undifferentiated proliferating state [28, 29]. It was thus proposed that expansion of pancreatic progenitor cells could be achieved via a controlled manipulation of Notch pathway [29]. Based on this information, we decided to examine the status of Notch-specific gene expression in the IEF cultures.

Our results show that Notch receptors (Notch1, Notch2, and Notch3), as well as Notch ligands (Jagged1 and Jagged2) were all coordinately and strongly activated during the course of the culture (Fig. 4A). Furthermore, Hes1 mRNA was induced 40–100-fold (Fig. 4B) and Hes1 protein at least 8–10-fold over their respective levels in day 0 IEFs (Fig. 4C). Moreover, the activation of Notch pathway temporally preceded the induction of Pdx1 (Fig. 4B). These results suggest that (as in embryonic pancreas, where activation of Notch pathway is required for expansion of the early progenitor cell pool [27, 30]), Notch might also be required for generation of the IPLCs. Note that although stage II IPLCs expressed a 2–4-fold higher level of Pdx1 mRNA than day 0 IEFs, they expressed approximately the same level of Pdx1 protein (compare Fig. 4B, top panel, with Fig. 4C). These results suggest that post-transcriptional regulation might contribute to Pdx1 gene expression in the IPLC cultures. It is important to emphasize that expression of Pdx1 and other pancreatic transcription factor genes was stably maintained at or above the ß-cell levels throughout multiple passages for at least 6 months in culture.

BMP-4 Induces Hes1 Gene Expression
Whereas the functional role of Notch pathway in proliferation and differentiation of pancreatic progenitors is well-documented [30], the contribution of BMP and TGF-ß pathways to these processes is only beginning to be explored [31, 32]. Our findings that BMP-4 is required for generation of IPLCs and that Notch pathway is activated during this process suggest that in the pancreas, BMP-4 might exert its action through Notch pathway. Indeed, interactions between BMP and Notch signaling have previously been documented in the muscle, in the central nervous system, and in the endothelial cells [33–35]. These works provide evidence for a direct cross-talk between Notch and BMP pathways, suggesting that Notch signaling may be playing a functional role in some of the biological effects of BMP.

To explore whether, as in the other cell types, the BMP-4 and Notch pathways might also interact in the IPLCs, we examined the effect of BMP-4 on Hes1 in the presence of different concentrations of {gamma}-secretase inhibitor, L-685458. This inhibitor blocks Notch-mediated signaling by interfering with the proteolytic cleavage of Notch receptor, and preventing generation of active intracellular form of Notch [36]. The scheme of the experiment is shown on the left of Figure 4D. When we cultured stage II IPLCs without BMP-4 for 48 hours, the level of Hes1 mRNA decreased approximately 40% (compare samples 1 and 2, Fig. 4D, right side of the panel). After an additional 48 hours without BMP-4, a further 20% decrease in Hes1 gene expression was observed (compare samples 2 and 3). Consistent with the known role of Notch receptor signaling in controlling Hes1 gene expression [30], exposure of the IPLC cultures to L-685458 in the absence of BMP-4 resulted in an additional 10%–20% reduction in Hes1 mRNA (compare samples 5 and 7). Importantly, the addition of BMP-4 back into the cultures for the last 48 hours of the experiment resulted in an increase in Hes1 mRNA in the absence and in the presence of L-685458 (compare samples 5 and 6 with samples 7 and 8).

To further investigate the effect of BMP-4 and Notch pathway on Hes1 gene expression, we performed four additional experiments similar to those above, but with the IPLC lines of different passage numbers. We used L-685458 at 4 µm in these experiments. The results shown in Figure 4D and supplemental online Figure 3A–3C were obtained with the same IPLC line at passages 3 to 8 after derivation. Passage numbers were in increasing order as follows: Figure 4D, supplemental online Figure 3A, 3B, 3C. The results in supplemental online Figure 3D were obtained with a different IPLC line at passage 6 after derivation. Collectively, these results indicate that BMP-4 becomes less efficient at inducing Hes1 in older IPLCs (compare samples 5 and 6 in Fig. 4D; supplemental online Fig. 3A–3C), thus suggesting that IPLCs undergo phenotypic changes in long-term cultures. This agrees with our finding of the blunted effect of BMP-4 on IPLC proliferation and Pdx1 gene expression in long-term cultures (see above). Interestingly, the BMP-4-mediated induction of Hes1 was partially restored in the presence of Notch inhibitor (compare samples 5 and 6 with samples 7 and 8 in supplemental online Fig. 3A, 3B, and 3C).

High Cell Density and BMP-4 Withdrawal Promote Endocrine Differentiation of IPLCs
Given that activation of Notch pathway and Hes1 during pancreatic development is known to inhibit endocrine differentiation [28–30], it is perhaps not surprising that IPLCs express only low levels of insulin. It is reasonable to expect that endocrine differentiation of IPLCs should be augmented by downregulation of Notch signaling and Hes1 gene expression. When BMP-4 was removed from stage II IPLCs even in the presence of Notch inhibitor, the level of Hes1 remained at least 10–15-fold higher than in day 0 IEFs (compare Fig. 4B, bottom panel, with Fig. 4D). Accordingly, no substantial upregulation of endocrine genes was observed under these conditions: we tested expression of insulin1, insulin 2, NeuroD, Nkx2.2, Nkx6.1, Pax4, ngn3, pax6, and Glut2 by semiquantitative RT-PCR (unpublished results). However, when the IPLCs were allowed to organize into high-density cell clusters in the absence of BMP-4 (stage III, Fig. 1; Materials and Methods section), several pro-endocrine transcription factors, such as Nkx6.1 and Pax6 [37], as well as E-cadherin, insulin/C-peptide, and glut2, were induced (Fig. 5A and 5C; also see below). We found that not all of the IPLC lines induced expression of all the pro-endocrine genes tested (Fig. 5A, table on the right side of the panel). We also found that with time in culture, the IPLCs gradually became resistant to endocrine differentiation.


Figure 5
View larger version (54K):
[in this window]
[in a new window]
  		Figure 5. Analysis of differentiating IPLC cultures. (A–D): High cell density and bone morphogenetic protein-4 (BMP-4) withdrawal induce differentiation of islet progenitor-like cells (IPLCs) and partial chromatin unfolding at the site of insulin promoter. (A): Results of differentiation were examined by semiquantitative reverse transcription-polymerase chain reaction (RT-PCR). On the left side of the panel, 18S rRNA was used as internal control. On the right side of the panel the table summarizes results of differentiation of five independently derived IPLC lines. The numbers on the left denote the number of IPLC lines (out of five analyzed) in which the expression of a given marker was increased as a result of differentiation. (B): Chromatin immunoprecipitation (ChiP) assay of mouse insulin 1 promoter with anti-acetylated histone H3 antibody. Real-time PCR was used as a readout for the ChiP assay. The results of PCR were normalized by the PCR signal from exogenously spiked ß-galactosidase gene-containing plasmid. Min6 and NIH3T3 cells were used as positive and negative controls, respectively. Shown are the results obtained with clonal IPLC line 3 (E). These results were reproduced in at least three independent experiments with different IPLC lines. Details of ChiP experimental protocol and calculation of relative H3 acetylation values are given in Materials and Methods. Statistical significance: *, p < .0001; **, p < .02. (C): Left panel, stage III IPLCs organize into cell clusters. Shown is a phase-contrast image (Zeiss Axiovert 200 microscope) of stage III IPLCs. (C): Right panel, cells within clusters express C-peptide. Note that the cells surrounding the cluster (identified by DAPI nuclear staining) are C-peptide-negative. (D): High cell density promotes epithelial organization of IPLCs. In the IPLCs cultured at low cell density (top panel), the epithelial marker E-cadherin is not localized to the plasma membrane. Cell clustering induces plasma membrane-associated pattern of E-cadherin expression (bottom panel). Note that brightly stained Pdx1-expressing cells are primarily localized in the E-cadherin-positive cell clusters. Each horizontal panel shows a typical image of the same split microscopic field. DAPI nuclear staining is not shown on the right of each panel to reveal nuclear localization of Pdx1. The images were obtained with Zeiss Axiovert 2 Plus microscope. (E, F): Clonal IPLC lines undergo mesenchymal to epithelial transition during differentiation. (E): Reciprocal changes in epithelial and mesenchymal mRNA expression in clonally derived IPLCs between stage II and stage III. Results of real-time RT-PCR were normalized by expression of 18S rRNA and are presented as fold change in the level of gene expression between stage II and stage III (increase, for E-cadherin; decrease, for snail and SMA). All the PCRs were performed in duplicates. The results are presented as means of the duplicate measurements. Shown are the results obtained with three independently derived IPLC lines (1, 2, and 3). (F): Reciprocal changes in epithelial and mesenchymal protein expression in IPLCs between stage II and stage III. Results of the immunoblot analysis of E-cadherin and SMA (top panel) were quantified by densitometry (bottom panel). Details of the immunoblotting protocol are given in Materials and Methods. Abbreviations: D0, day 0 islet-enriched fractions; DAPI, 4,6-diamidino-2-phenylindole; SMA, smooth muscle actin.

 
The recent work of Chakrabarti et al. suggests that lysine acetylation of histone H3 is required for establishment of an open chromatin structure at the site of insulin 1 gene promoter in pancreatic ß-cells [10]. To determine whether stage III culture conditions might induce chromatin opening at the site of insulin promoter in the IPLC cultures, we took advantage of the Chakrabarti et al. protocol of quantitative ChiP with anti-acetylated histone H3 antibody [10]. As those investigators examined histone H3 acetylation status of insulin 1 promoter, to be able to compare our results with theirs, we also examined the insulin 1 gene promoter. Insulin-expressing MIN6 insulinoma and NIH3T3 fibroblasts were used as positive and negative controls, respectively. We found that the level of histone H3 acetylation at the site of insulin 1 gene promoter was significantly higher in stage III than stage II cultures (Fig. 5B), suggesting that the stage III culture conditions promote partial unfolding of chromatin at insulin gene promoter. We reproduced these results in at least three independent experiments with independently derived IPLC lines.

High Cell Density and BMP-4 Withdrawal Promote Epithelial Organization of Clonally Derived IPLCs
As described above, a partial mesenchymal to epithelial phenotypic shift takes place at the end of stage I (Fig. 2C, 2D). Although as a result of this shift the IPLCs upregulate E-cadherin mRNA (Fig. 2C), its level is still below that found in day 0 IEFs. Also, the majority of stage II IPLCs do not exhibit plasma membrane-associated E-cadherin typical of normal epithelial cells (Fig. 5D, top panel). This contrasts with clear plasma membrane-associated pattern of E-cadherin expression in high-density stage III IPLCs (Fig. 5D, bottom panel). Moreover, the majority of cells with plasma membrane-associated E-cadherin express strong nuclear Pdx1 (Fig. 5D, bottom panel). Note that E-cadherin protein level was increased, and SMA protein level was reciprocally decreased in stage III cells (Fig. 5F).

To determine whether individual IPLCs can be induced to upregulate their epithelial phenotype and downregulate mesenchymal phenotype, we examined the dynamics of mesenchymal and epithelial gene expression in clonal IPLC lines derived from single cells. Whereas early stage I cells are not amenable to single cell cloning, it is feasible to clone late-stage I IPLCs. Single cell-derived IPLC lines were expanded in the presence of BMP-4 and were then allowed to differentiate, as described above. The results with three independently derived clonal lines (Fig. 5E) demonstrate that stage III IPLCs coordinately upregulate E-cadherin and downregulate snail and SMA genes during differentiation. These results strongly suggest that the shift from mesenchymal to epithelial phenotype occurs on a single cell level, thus implying that mesenchymal-epithelial transition (MET) [38] might take place in this system. Since epithelial architecture is known to play a functional role in normal islet development [18, 39], we hypothesize that enhancing epithelial organization of IPLCs should further augment their endocrine differentiation.


   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 Acknowledgments
 References
 
In this work, we demonstrate for the first time that adult mouse pancreas can generate long-term-replicating IPLCs that robustly and stably express a constellation of genes characteristic of embryonic endodermal/pancreatic progenitors, including Notch pathway-associated genes. As of today, we have derived more than 15 independent IPLC lines, all stably expressing a high level of endodermal/pancreatic transcription factors through many passages in culture. Despite this prominent pancreatic progenitor phenotype, the IPLCs express only a low level of insulin. When transplanted under the kidney capsule of immunodeficient NOD/SCID mice, the IPLCs did not generate detectable tumors when examined 6 weeks after transplantation. More work will be required to fully access possible tumorigenic potential of the IPLCs.

Given the known role of Notch pathway in inhibition of endocrine differentiation in vivo [28, 29], low insulin expression was not entirely unexpected. Since downregulation of Notch pathway is required for endocrine differentiation during embryogenesis [27], we hypothesized that endocrine differentiation of IPLCs would also be controlled by the similar mechanisms. Indeed, we found that downregulation of Hes1 induced by high cell density and growth factor withdrawal resulted in an induction of several pro-endocrine transcription factors and insulin. These events were accompanied by chromatin unfolding at the insulin 1 promoter region. We failed, however, to reproducibly detect ngn3 gene expression in differentiating IPLC cultures. This suggests that that ngn3 might not be required for generation of new ß-cells in adult pancreas, as is also indicated by recent results of other investigators, [40]. Collectively, our results suggest that IPLCs may be primed to undergo further endocrine differentiation.

In light of previous reports that point to functional role of bFGF, BMP-4, and Notch pathways in pancreatic endoderm patterning and early pancreatic development [13–15], the activity of these pathways in IPLC cultures is intriguing. Our experiments with different concentrations of Notch pathway inhibitor suggest that BMP-4 can at least to some extent induce Hes1 independently of Notch (Fig. 4D). Nevertheless, in IPLCs of high passage number, inhibition of Notch pathway augmented BMP-4-mediated induction of Hes1, suggesting a possible cooperation between the BMP-4 and Notch pathways [33–35]. The interaction between these two pathways in the pancreas merits further investigation.

Although not all pancreatic markers characteristic of normal pancreatic development were expressed by the IPLCs (for instance, the IPLCs did not express Ptf1a), our results point to many similarities between IPLCs and the developmentally primitive pancreatic progenitors present transiently during pancreatic development before initiation of endocrine differentiation. Since large quantities of IPLCs can readily be obtained in vitro, these cells provide a useful experimental tool for addressing mechanistic questions about early pancreatic development that are difficult to address in vivo. For example, as has been found in other tissues [33–35], our results suggest that BMP-mediated signaling may be involved in the activation of Notch pathway in the pancreas. It is tempting to speculate that in pancreatic development, BMP-4-mediated signaling may function upstream of the Notch pathway. In addition, interactions between LIF and BMP-4 pathways, which are likely to occur in the IPLC cultures, might have implication for pancreatic development. In this regard, it is noteworthy that functional LIF and BMP-4 pathway interactions have been documented in the development of astrocytes [41, 42].

The mechanism of generation of IPLCs in our cultures is still to be determined. In one scenario, the IPLCs could arise through replication of a small pool of pre-existing pancreatic stem/progenitor cells. Another possibility is that IPLCs are generated via lineage reprogramming or dedifferentiation of mature cells. Pancreatic epithelial or nonpancreatic cell types residing in the pancreas could be targets for such dedifferentiation, which might involve epithelial-mesenchymal transition (EMT) and the reverse process, MET [38]. In fact, our results with clonal IPLC lines suggest that partial endocrine differentiation of IPLCs during stage III is accompanied by MET. It has recently been proposed that EMT/MET-based mechanisms operate in cultures of adult human islets and during mouse islet regeneration in vivo [4, 43, 44]. We are currently using a genetic lineage tracing analysis to directly address the functional significance of EMT and MET for generation of IPLCs.

Production of clinically relevant quantities of functional islets from adult pancreas in vitro will require overcoming a number of formidable obstacles. Two strategies appear to hold promise. First, direct replication of mature islet cells could be induced. Although this approach has had limited success so far, recent results of Dor et al. in the mouse [1] suggest that this could be a viable option. It is still unknown, however, whether significant replication of islet cells can be achieved in human islet cultures. A second strategy is the expansion of differentiation-competent progenitors followed by endocrine differentiation. Although the IPLCs become partially independent of BMP-4 during long-term culture, they retain their strong endodermal/pancreatic progenitor phenotype for many months. We propose that by virtue of their progenitor phenotype, the IPLCs will offer a useful experimental tool for investigating the mechanisms of generation of new ß-cells in adult pancreas and will provide an abundant source of differentiation-competent cells.


   DISCLOSURES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
Acknowledgments
 References
 
The authors indicate no potential conflicts of interest.


   ACKNOWLEDGMENTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 Acknowledgments
References
 
We are grateful to Rocky Tuan for comments on the manuscript prior to submission. We thank the staff of the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) mouse core facility, in particular Oksana Gavrilova, for help with IEF isolations. Our thanks go also to the members of Raghu Mirmira laboratory for help with quantitative ChiP assays; to Joel Habener and Chris Wright for the gifts of Pdx1 antibodies; to Yuh Nung Jan for Hes1 antibody; and to Jeff Whitsett for Foxa2 antibody. This research was supported by the Intramural Research Program of the NIH, NIDDK. Y.C. and F.A. contributed equally to the work. Y.C. is currently affiliated with the National Institute of Neurological Disorders and Stroke/NIH; F.A. is currently affiliated with the Department of Standards Development, U.S. Pharmacopeia, Rockville, MD; C.H.P. is currently affiliated with the Department of Biochemistry and Molecular Biology, Mayo Clinic College of Medicine, Rochester, MN; and N.L. is currently affiliated with the National Institute of Dental and Craniofacial Research/NIH.


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 Acknowledgments
 References
 

  1. Dor Y, Brown J, Martinez OIet al. Adult pancreatic beta-cells are formed by self-duplication rather than stem-cell differentiation. Nature 2004; 429: 41–46.[CrossRef][Medline]

  2. Butler AE, Janson J, Soeller WCet al. Increased beta-cell apoptosis prevents adaptive increase in beta-cell mass in mouse model of type 2 diabetes: Evidence for role of islet amyloid formation rather than direct action of amyloid. Diabetes 2003; 52: 2304–2314.[Abstract/Free Full Text]

  3. Bonner-Weir S, Toschi E, Inada Aet al. The pancreatic ductal epithelium serves as a potential pool of progenitor cells. Pediatr Diabetes 2004; 5suppl 2, : 16–22.[Medline]

  4. Habener JF. A perspective on pancreatic stem/progenitor cells. Pediatr Diabetes 2004; 5suppl 2, : 29–37.[Medline]

  5. Hayek A. In search of endocrine progenitor/stem cells in the human pancreas. Pediatr Diabetes 2004; 5suppl 2, : 70–74.[Medline]

  6. Means AL, Meszoely IM, Suzuki Ket al. Pancreatic epithelial plasticity mediated by acinar cell transdifferentiation and generation of nestin-positive intermediates. Development 2005; 132: 3767–3776.[Abstract/Free Full Text]

  7. Slack JM. Developmental biology of the pancreas. Development 1995; 121: 1569–1580.[Abstract]

  8. Choi Y, Ta M, Atouf Fet al. Adult pancreas generates multipotent stem cells and pancreatic and nonpancreatic progeny. STEM CELLS 2004; 22: 1070–1084.[Abstract/Free Full Text]

  9. Lumelsky N, Blondel O, Laeng Pet al. Differentiation of embryonic stem cells to insulin-secreting structures similar to pancreatic islets. Science 2001; 292: 1389–1394.[Abstract/Free Full Text]

  10. Chakrabarti SK, Francis J, Ziesmann SMet al. Covalent histone modifications underlie the developmental regulation of insulin gene transcription in pancreatic beta cells. J Biol Chem 2003; 278: 23617–23623.[Abstract/Free Full Text]

  11. Chakrabarti SK, James JC, Mirmira RG. Quantitative assessment of gene targeting in vitro and in vivo by the pancreatic transcription factor, Pdx1. Importance of chromatin structure in directing promoter binding. J Biol Chem 2002; 277: 13286–13293.[Abstract/Free Full Text]

  12. Brissova M, Fowler M, Wiebe Pet al. Intraislet endothelial cells contribute to revascularization of transplanted pancreatic islets. Diabetes 2004; 53: 1318–1325.[Abstract/Free Full Text]

  13. Hebrok M, Kim SK, St Jacques Bet al. Regulation of pancreas development by hedgehog signaling. Development 2000; 127: 4905–4913.[Abstract]

  14. Tiso N, Filippi A, Pauls Set al. BMP signalling regulates anteroposterior endoderm patterning in zebrafish. Mech Dev 2002; 118: 29–37.[CrossRef][Medline]

  15. Kumar M, Melton D. Pancreas specification: A budding question. Curr Opin Genet Dev 2003; 13: 401–407.[CrossRef][Medline]

  16. Balemans W, Van Hul W. Extracellular regulation of BMP signaling in vertebrates: A cocktail of modulators. Dev Biol 2002; 250: 231–250.[CrossRef][Medline]

  17. Wilson ME, Scheel D, German MS. Gene expression cascades in pancreatic development. Mech Dev 2003; 120: 65–80.[CrossRef][Medline]

  18. Dahl U, Sjodin A, Semb H. Cadherins regulate aggregation of pancreatic beta-cells in vivo. Development 1996; 122: 2895–2902.[Abstract]

  19. Street CN, Lakey JR, Seeberger Ket al. Heterogenous expression of nestin in human pancreatic tissue precludes its use as an islet precursor marker. J Endocrinol 2004; 180: 213–225.[Abstract]

  20. Nieto MA. The snail superfamily of zinc-finger transcription factors. Nat Rev Mol Cell Biol 2002; 3: 155–166.[CrossRef][Medline]

  21. Lendahl U, Zimmerman LB, McKay RD. CNS stem cells express a new class of intermediate filament protein. Cell 1990; 60: 585–595.[CrossRef][Medline]

  22. Selander L, Edlund H. Nestin is expressed in mesenchymal and not epithelial cells of the developing mouse pancreas. Mech Dev 2002; 113: 189–192.[CrossRef][Medline]

  23. Lardon J, Rooman I, Bouwens L. Nestin expression in pancreatic stellate cells and angiogenic endothelial cells. Histochem Cell Biol 2002; 117: 535–540.[CrossRef][Medline]

  24. Wilding L, Gannon M. The role of pdx1 and HNF6 in proliferation and differentiation of endocrine precursors. Diabetes Metab Res Rev 2004; 20: 114–123.[CrossRef][Medline]

  25. Kawaguchi Y, Cooper B, Gannon Met al. The role of the transcriptional regulator Ptf1a in converting intestinal to pancreatic progenitors. Nat Genet 2002; 32: 128–134.[CrossRef][Medline]

  26. Kim SK, MacDonald RJ. Signaling and transcriptional control of pancreatic organogenesis. Curr Opin Genet Dev 2002; 12: 540–547.[CrossRef][Medline]

  27. Jensen J. Gene regulatory factors in pancreatic development. Dev Dyn 2004; 229: 176–200.[CrossRef][Medline]

  28. Hald J, Hjorth JP, German MSet al. Activated Notch1 prevents differentiation of pancreatic acinar cells and attenuate endocrine development. Dev Biol 2003; 260: 426–437.[CrossRef][Medline]

  29. Murtaugh LC, Stanger BZ, Kwan KMet al. Notch signaling controls multiple steps of pancreatic differentiation. Proc Natl Acad Sci U S A 2003; 100: 14920–14925.[Abstract/Free Full Text]

  30. Jensen J, Pedersen EE, Galante Pet al. Control of endodermal endocrine development by Hes-1. Nat Genet 2000; 24: 36–44.[CrossRef][Medline]

  31. Harmon EB, Apelqvist AA, Smart NGet al. GDF11 modulates NGN3+ islet progenitor cell number and promotes beta-cell differentiation in pancreas development. Development 2004; 131: 6163–6174.[Abstract/Free Full Text]

  32. Zaret KS. Hepatocyte differentiation: From the endoderm and beyond. Curr Opin Genet Dev 2001; 11: 568–574.[CrossRef][Medline]

  33. Dahlqvist C, Blokzijl A, Chapman Get al. Functional Notch signaling is required for BMP4-induced inhibition of myogenic differentiation. Development 2003; 130: 6089–6099.[Abstract/Free Full Text]

  34. Takizawa T, Ochiai W, Nakashima Ket al. Enhanced gene activation by Notch and BMP signaling cross-talk. Nucleic Acids Res 2003; 31: 5723–5731.[Abstract/Free Full Text]

  35. Itoh F, Itoh S, Goumans MJet al. Synergy and antagonism between Notch and BMP receptor signaling pathways in endothelial cells. EMBO J 2004; 23: 541–551.[CrossRef][Medline]

  36. Karlstrom H, Bergman A, Lendahl Uet al. A sensitive and quantitative assay for measuring cleavage of presenilin substrates. J Biol Chem 2002; 277: 6763–6766.[Abstract/Free Full Text]

  37. Schwitzgebel VM, Scheel DW, Conners JRet al. Expression of neurogenin3 reveals an islet cell precursor population in the pancreas. Development 2000; 127: 3533–3542.[Abstract]

  38. Savagner P. Leaving the neighborhood: Molecular mechanisms involved during epithelial-mesenchymal transition. Bioessays 2001; 23: 912–923.[CrossRef][Medline]

  39. Shih DQ, Heimesaat M, Kuwajima Set al. Profound defects in pancreatic beta-cell function in mice with combined heterozygous mutations in Pdx-1, Hnf-1alpha, and Hnf-3beta. Proc Natl Acad Sci U S A 2002; 99: 3818–3823.[Abstract/Free Full Text]

  40. Lee CS, De Leon DD, Kaestner KHet al. Regeneration of pancreatic islets after partial pancreatectomy in mice does not involve the reactivation of neurogenin-3. Diabetes 2006; 55: 269–272.[Abstract/Free Full Text]

  41. Nakashima K, Yanagisawa M, Arakawa Het al. Synergistic signaling in fetal brain by STAT3-Smad1 complex bridged by p300. Science 1999; 284: 479–482.[Abstract/Free Full Text]

  42. Rajan P, Panchision DM, Newell LFet al. BMPs signal alternately through a SMAD or FRAP-STAT pathway to regulate fate choice in CNS stem cells. J Cell Biol 2003; 161: 911–921.[Abstract/Free Full Text]

  43. Gershengorn MC, Hardikar AA, Hardikar Aet al. Epithelial-to-mesenchymal transition generates proliferative human islet precursor cells. Science 2004; 306: 2261–2264.[Abstract/Free Full Text]

  44. Kulkarni RN, Jhala US, Winnay JNet al. PDX-1 haploinsufficiency limits the compensatory islet hyperplasia that occurs in response to insulin resistance. J Clin Invest 2004; 114: 828–836.[CrossRef][Medline]




This article has been cited by other articles:


Home page
 Stem CellsHome page
V. S. Urban, J. Kiss, J. Kovacs, E. Gocza, V. Vas, E. Monostori, and F. Uher
Mesenchymal Stem Cells Cooperate with Bone Marrow Cells in Therapy of Diabetes
Stem Cells, January 1, 2008; 26(1): 244 - 253.
[Abstract] [Full Text] [PDF]

Home page
 DiabetesHome page
F. Atouf, C. H. Park, K. Pechhold, M. Ta, Y. Choi, and N. L. Lumelsky
No Evidence for Mouse Pancreatic {beta}-Cell Epithelial-Mesenchymal Transition In Vitro
Diabetes, March 1, 2007; 56(3): 699 - 702.
[Abstract] [Full Text] [PDF]

Home page
 Stem CellsHome page
T. Tayaramma, B. Ma, M. Rohde, and H. Mayer
Chromatin-Remodeling Factors Allow Differentiation of Bone Marrow Cells into Insulin-Producing Cells
Stem Cells, December 1, 2006; 24(12): 2858 - 2867.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Figures
Right arrow All Versions of this Article:
2005-0367v1
24/7/1738    most recent
Right arrow Alert me when this article is cited
Right arrow