Molecular Endocrinology, doi:10.1210/me.2004-0243
Molecular Endocrinology 18 (11): 2765-2776
Copyright © 2004 by The Endocrine Society
Pancreatic Islet Progenitor Cells in Neurogenin 3-Yellow Fluorescent Protein Knock-Add-On Mice
Georg Mellitzer,
Mercè Martín,
Marjorie Sidhoum-Jenny,
Christophe Orvain,
Jochen Barths,
Philip A. Seymour,
Maike Sander and
Gérard Gradwohl
Institut National de la Santé et de la Recherche Médicale (INSERM) Unité 381 (G.M., M.M., M.S.-J., C.O., G.G.), 67200 Strasbourg, France; Institut de Génétique et de Biologie Moléculaire et Cellulaire (J.B.), Centre National de la Recherche Scientifique (CNRS)/INSERM/University Louis Pasteur, P.B. 10142, F-67404 Illkirch Cedex; and Department of Developmental & Cell Biology (P.A.S., M.S.), Irvine, California 92697-2300
Address all correspondence and requests for reprints to: Gérard Gradwohl, Institut National de la Santé et de la Recherche Médicale (INSERM) Unité 381, 3 avenue Molière, 67200, Strasbourg, France. E-mail: gerard.gradwohl{at}inserm.u-strasbg.fr.
 |
ABSTRACT
|
|---|
The basic helix-loop-helix transcription factor Neurogenin 3 (NGN3) controls endocrine cell fate specification in uncommitted pancreatic progenitor cells. Ngn3-deficient mice do not develop any islet cells and are diabetic. All the major islet cell types, including insulin-producing ß-cells, derive from Ngn3-positive endocrine progenitor cells. Therefore, the characterization of this population of immature cells is of particular interest for the development of novel strategies for cell replacement therapies in type 1 diabetes. To explore further the biology of islet progenitor cells we have generated a mouse in which Ngn3-expressing cells are labeled with the enhanced yellow fluorescent protein (EYFP) using a knock-add-on strategy. In this approach, the EYFP cDNA is introduced into the 3'-untranslated region of the proendocrine transcription factor, Neurogenin 3, without deleting any endogenous coding or regulatory sequences. In Ngn3EYFP/+ and Ngn3EYFP/EYFP mice, the EYFP protein is targeted to Ngn3-expressing progenitors in the developing pancreas, and islets develop normally. Islet progenitors can be purified from whole embryonic pancreas by fluorescence-activated cell sorting from Ngn3EYFP/+ mice and their development can be monitored in real time in pancreas explant cultures. These experiments showed that endocrine progenitors can form de novoand expand, in vitro, in the absence of signals from the surrounding mesenchyme, suggesting that endocrine commitment is a default pathway. The Ngn3EYFP mice represent a valuable tool to study islet cell development and neogenesis in normal and diabetic animals as well as for the determination of the conditions to generate ß-cells in vitro.
|
INTRODUCTION
|
|---|
GENE INACTIVATION IN the mouse has generated significant progress in our understanding of the hierarchy of transcription factors controlling the successive steps of islet cell differentiation (1, 2). This knowledge can now be translated to specifically label and characterize in depth islet progenitor cells at the molecular and cellular level. Such lineage labeling can be realized by genetically engineering mice to express various autofluorescent reporter proteins under the control of regulatory elements of transcription factors regulating key steps in islet cell development. Several studies have demonstrated the feasibility of monitoring simultaneously multiple spectrally distinct fluorescent proteins noninvasively within a single animal (3, 4, 5). Such a compound transgenic mouse or rainbow mouse will be instrumental in the purification and determination of the transcriptome and proteome of specific subtypes of islet progenitor cells. Moreover, multispectral visualization in live samples of pancreas cultures will also be possible, and the effect of extrinsic factors on the expansion and differentiation of various populations of tagged islet progenitors will be able to be tested. Finally, such tools will be of central importance for studies aimed at generating ß-cells in vitro starting from various sources of plastic stem/progenitor cells to monitor normal islet gene expression and function. We believe that in the perspective of the generation of such a rainbow mouse, one essential feature is the faithful recapitulation of the spatiotemporal expression of the tagged genes by the fluorescent reporter proteins. Due to the potential lack of important regulatory elements, as well as positional effects, this cannot always be achieved by traditional promoter approaches or BAC transgenesis. An additional problem is that the random insertion of the foreign DNA within the genome might interfere with the function of other genes. Both of these problems are potentially circumvented using a more sophisticated and long term knock-in approach in which the endogenous gene is replaced by a fluorescent reporter protein by homologous recombination in embryonic stem (ES) cells. In most cases, inactivation of one allele of islet-specific transcription factors does not impair endocrine differentiation. However, one cannot exclude that in a compound rainbow mouse, the multiple allelic combination of the different XFP variants targeted to the locus of genes controlling pancreatic endocrine cell development might generate an islet phenotype. Because of these considerations and with the perspective of the future generation of such a technicolor mouse, we decided to produce a mouse in which islet progenitor cells have been genetically labeled with the enhanced yellow fluorescent protein (EYFP) using a knock-add-on strategy. In this approach, the EYFP cDNA is introduced into the 3'-untranslated region (UTR) of the proendocrine transcription factor Neurogenin 3 (Ngn3) without deleting any endogenous coding or regulatory sequences.
The Neurogenin 3 transcription factor has emerged as a master gene controlling endocrine cell fate decision in pluripotent pancreatic progenitors during embryogenesis. Ngn3-deficient mice lack islets of Langerhans as well as islet precursor cells, and they develop diabetes and die postnatally (6). Lineage tracing experiments demonstrated that all the islet cell types, including the insulin-secreting ß-cells, derive from Ngn3-expressing cells (7). These data, together with other studies (8, 9, 10), thus clearly demonstrate that Ngn3 is the earliest marker of pancreatic epithelial cells committed toward an islet cell fate. Ngn3 has also been shown recently to control enteroendocrine commitment in the developing and adult intestinal epithelium (11), whereas gastric endocrine cell differentiation is only partially dependent on Ngn3 function (11, 12). For an in-depth characterization of the biology of immature islet progenitor cells, the issues of identification, visualization, and isolation of islet progenitor cells must be addressed. We thus decided to generate genetically labeled mice that would permit purification of pancreatic but also intestinal and gastric endocrine progenitor cells. Single transgenic mice, expressing the enhanced green fluorescent protein (EGFP) driven by Neurogenin3 regulatory sequences using promoter (7) and knock-in (12) approaches have been reported previously. We describe here the production and characterization of a mouse generated by a knock-add-on strategy in which Ngn3-expressing cells are tagged with EYFP. We demonstrate the usefulness of this approach for the future generation of multiple fluorescent mice as well as for the study, in real time, of the development of islet progenitor cells in pancreatic explant cultures.
|
RESULTS
|
|---|
Generation of Ngn3-EYFP Mice
We have generated a mouse line expressing the enhanced yellow fluorescent protein (EYFP) targeted to the Ngn3 locus using a knock-add-on approach. Briefly, a floxed internal ribosome entry site (IRES)-EYFP-Puromycine cassette has been introduced in the 3'-UTR of the mouse Ngn3 gene by homologous recombination in ES cells (Fig. 1
). This cassette insertion does not delete any sequence from the Ngn3 locus. In previous efforts, a 3'-flanking region from the mouse Ngn3 gene failed to target a LacZ reporter in Ngn3-expressing cells (Orvain, C., A. Dierich, and G. Gradwohl, unpublished observation). This observation suggested that the 3'-region lacks elements that regulate tissue-specific expression of the Ngn3 gene, indicating that the introduction of the EYFP gene should not interfere with the normal regulation of Ngn3 transcription. Two correctly targeted ES clones, as determined by Southern blot hybridization (Fig. 1C), have been injected into blastocysts. The resulting chimeric mice successfully transmitted the mutation to their progeny. The excision of the puromycin resistance gene (pac, puromycin N-acetyltransferase) has been performed by breeding the Ngn3EYFP-Pac/+ mouse with a transgenic mouse expressing the Cre recombinase under the control of the cytomegalovirus promoter (CMV-Cre) (13) (Fig. 1D). We observed no obvious difference either in the expression pattern or in the fluorescence intensity of the EYFP protein in the mice carrying or not carrying the pac gene. Using a strictly identical strategy, we also generated EGFP-tagged mice (data not shown).
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 1. Strategy for the Introduction of the EYFP Gene into the 3'-UTR of Ngn3 Locus by Homologous Recombination
Several independent correctly targeted ES cells clones could be established. A, Maps of the wild-type Ngn3 locus and the targeting construct. B, EYFP-targeted Ngn3 allele before and after the excision of the floxed puromycin resistance gene (pac) by the Cre recombinase. Stars in panel B indicate the position of the 5'- and 3'-external probes used for Southern blot analysis. Black box indicates the Ngn3 coding sequence, the IRES-EYFP sequence, and puromycin resistance gene (pac); the loxP sites are indicated as well. C, Genotype analysis by Southern blots of six independent recombinated ES cell clones by hybridization with a 5'- and 3'-external genomic probe. D, Representative genotyping by PCR of a litter from intercrosses between Ngn3EYFP/+ heterozygous and CMV-Cre mice with primers EYFP-pA-5' and Ngn3–3' as described in Material and Methods. When the pac gene has been successfully excised by the CRE recombinase, a 558-bp PCR band is amplified. Positions of primers are indicated by arrows. Wt, Wild type; mt, mutant; X, XbaI; Sp, SpeI; E, EcoRI; Pm, PmeI; As, AscI.
|
|
EYFP Is Targeted to Ngn3-Expressing Cells in the Pancreas
To determine whether the EYFP expression reflects the distribution of the Ngn3-expressing cells, we first examined directly the intrinsic EYFP fluorescence in developing embryonic pancreas. As shown in Fig. 2A, EYFP fluorescence was detected both in the dorsal and ventral embryonic pancreas in a pattern reminiscent of Ngn3-expressing cells, i.e. isolated or small clusters of cells located along the epithelial duct-like structures. To localize EYFP cells more precisely, we performed immunofluorescence experiments for EYFP on cryosections of pancreas at different developmental stages in Ngn3EYFP/+ embryos (Fig. 2B). The EYFP expression mirrors the dynamic expression of Ngn3, first detected at the bud stage: the number of EYFP+ cells peaks at embryonic d 15.5 (E15.5) dpc (days post coitus) and decreases thereafter. At E18.5, only rare EYFP-expressing cells are detected in the developing pancreas as was previously observed for Ngn3 protein (data not shown). To further characterize EYFP+ cells, we performed double immunofluorescence for EYFP and either Ngn3 or endocrine hormones. With an antiserum specific for Ngn3, we first compared the overlap between EYFP- and Ngn3-expressing cells. As expected and shown in Fig. 3A, the EYFP protein is targeted to Ngn3+ progenitor cells starting from the early bud stage at E10.5 dpc, where most of the Ngn3 expressing cells coexpress EYFP. At E13.5, 88% and at E15.5, 83% of the Ngn3-expressing cells are labeled with EYFP, whereas not all of the EYFP expressing cells colabel for Ngn3 at all stages analyzed. At E13.5 the EYFP reporter is exclusively expressed in immature cells that do not express endocrine hormones (Fig. 3B). However, at E10.5 and at E15.5 dpc, approximately 12% of the EYFP-positive cells costain for insulin or glucagon (Ins/Glu). In summary, these studies show that in the EYFP knock-add-on Ngn3 mice we generated, the EYFP transgene is, as expected, targeted to Ngn3 progenitors in the developing pancreas. A fraction of more mature cells is also labeled by the reporter protein, which is likely due to differences in the stability and/or turnover between Ngn3 and YFP proteins.
View larger version (61K):
[in this window]
[in a new window]
|
Fig. 2. EYFP Expression in the Pancreas from Ngn3EYFP/+ Mice at Different Developmental Stages
A, Pancreases were harvested from Ngn3EYFP/+ mice and intrinsic EYFP fluorescence observed in dissected dorsal and ventral pancreas at E13.5 and E15.5. Arrows indicate clear single EYFP-expressing cells. B, Immunofluorescence with an anti-EGFP antibody also recognizing EYFP showing the specific expression of EYFP in the pancreatic epithelium at different developmental stages. EYFP+ cells can be detected in a pattern reminiscent of Ngn3-expressing cells as soon as E10.5 in the dorsal pancreatic bud, showing its highest expression at E15.5, after which it gradually decreases. At E18.5 only a very few EYFP+ cells can still be detected. Arrows indicate clear EYFP+ cells. Magnification, x20.
|
|
View larger version (80K):
[in this window]
[in a new window]
|
Fig. 3. Labeling of Ngn3-Expressing Cells with EYFP in the Developing Pancreas
A, Double immunofluorescence for EYFP (green) and Ngn3 (red) on cryosections of pancreas from Ngn3EYFP/+ mice at different embryonic stages. EYFP is coexpressed in most of the Ngn3+ cells at E13.5 and E15.5. B, Double immunofluorescence for EYFP (green) and insulin/glucagon (red). EYFP is mostly expressed in undifferentiated islet progenitors cells that do not costain for hormones. EYFP is coexpressed with insulin and glucagon (Ins/Glu) in some cells at E10.5 and E15.5 (arrowheads), whereas at E13.5 its expression is exclusive. The fact that not all cells coexpress EYFP and Ngn3 might be due to a longer stability of the EYFP protein and/or the low sensitivity of the Ngn3 antibody. Arrows in panels A and B indicate examples of EYFP, Ngn3, or Ins/Glu single positive cells, whereas arrowheads indicate cells either coexpressing EYFP and Ngn3 or EYFP and Ins/Glu. Magnification, x40.
|
|
Labeling of Enteroendocrine and Neural Progenitor Cells in the Gastrointestinal Tract and Developing Central Nervous System (CNS)
Ngn3 is also expressed in isolated progenitors of the endocrine lineage in the embryonic and adult gastrointestinal tract. Intrinsic EYFP fluorescence of Ngn3-expressing cells can be observed in the embryonic intestine of Ngn3EYFP/+ mice at E15.5 dpc (data not shown). The weak intensity of EYFP fluorescence and the paucity of Ngn3-expressing cells in the adult digestive tract hampered the identification of individual EYFP+ cells in the adult intestinal epithelium and the glandular stomach. Immunodetection with an antibody against EYFP/EGFP, however, clearly revealed rare and isolated EYFP+ cells in the adult duodenal crypts and the gastric epithelium of newborn mice in a pattern reminiscent of Ngn3 expression in enteroendocrine progenitors (Fig. 4, A–C). As expected, we also successfully targeted the EYFP reporter protein to developing neural progenitors in the central nervous system (Fig. 4D) where Ngn3-expressing cells have also been described.
View larger version (46K):
[in this window]
[in a new window]
|
Fig. 4. EYFP Expression in Organs outside the Pancreas in Ngn3EYFP/+ Mice
Fluorescence immunohistochemistry showing that EYFP expression follows the normal expression pattern of Ngn3 in organs outside the pancreas. EYFP+ cells can easily be detected in enteroendocrine progenitors in the duodenum during embryogenesis (A) and in the adult (B); in the latter it is restricted, like Ngn3, to the crypts as well as in the postnatal stomach (C). D, EYFP expression in neural progenitors of the developing forebrain neuroepithelium.
|
|
Ngn3EYFP/EYFP Mice Do Not Develop Diabetes
Ngn3 mutant mice develop diabetes and very rapidly die at early postnatal stages. To determine whether, as expected, the insertion of the EYFP in the 3'-UTR of Ngn3 locus does not interfere with Ngn3 proendocrine function we generated Ngn3EYFP/EYFP mice. In contrast to Ngn3-deficient mice the Ngn3EYFP/EYFP animals develop normally and reach adulthood. Islets of Langerhans can clearly be observed (Fig. 5A), and we did not find any difference in the urine and blood glucose content between wild-type and transgenic mice. Moreover, ip glucose tolerance tests performed on 8- to 10-wk-old mice showed no significant difference in the response between wild-type, Ngn3EYFP/+, and Ngn3EYFP/EYFP mice (Fig. 5B). Thus, islet cells are correctly specified demonstrating that the introduction of the EYFP tag at the Ngn3 locus does not perturb the normal function of this gene during islet cell development.
View larger version (76K):
[in this window]
[in a new window]
|
Fig. 5. Normal Islet Development and Function in Ngn3EYFP/EYFP Mice
Hematoxylin-eosin staining of heterozygote (Ngn3EYFP/+) and homozygote (Ngn3EYFP/EYFP) adult pancreas showing normally developed Islets of Langerhans. B, ip glucose tolerance test showing a normal glucose response of heterozygote (Ngn3EYFP/+) and homozygote (Ngn3EYFP/EYFP) adult mice. Sex- and age-matched wild-type CD1 mice were used as control.
|
|
Fluorescence-Activated Cell Sorting (FACS) Analysis of EYFP-Labeled Endocrine Progenitor Cells
The opportunity to isolate and study the population of Ngn3-expressing islet and enteroendocrine progenitor cells is one of the main purposes of the Ngn3EYFP/+ mouse. To test whether endocrine progenitors can be purified, embryonic pancreases were dissociated to single cells by enzymatic digestion and subjected to FACS. The EYFP-labeled endocrine progenitor cells could readily be detected and purified by FACS (Fig. 6). Approximately 1300 EYFP+ cells can be purified from one E15.5 pancreas. Although intrinsic EYFP fluorescence was weaker in the intestine than in the pancreas, an average of 1700 EYFP+ enteroendocrine progenitors were successfully purified from each E15.5 small intestine (data not shown). This result demonstrates that a single-copy insertion of EYFP expressed under the control of Ngn3-regulatory sequences provides enough intrinsic fluorescence to detect and sort single EYFP-expressing cells.
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 6. Flow Cytometric Analysis of a Single-Cell Suspension Prepared from Whole Pancreas of E15.5 ngn3eYFP/+ Mutant Mouse Embryos
Fluorescence intensities are expressed in arbitrary units on a four-decade logarithmic scale. Of the pancreatic cells from the Ngn3eYFP/+ mouse at E15.5, 0.9% express EYFP (visible due to the shift of the emission spectrum to the green compared with the autofluorescence of the negative cells). x-Axis, log FL 1-A intensity (530/30); y-axis, log FL 2-A intensity (585/42).
|
|
Development of EYFP-Labeled Islet Progenitors Cells in Vitro in Real Time
The labeling of islet progenitor cells with noninvasive fluorescent markers provides a unique tool to follow, in vitro and in real time, the induction and expansion of endocrine progenitor cells. We developed explant cultures of pancreatic epithelium dissected from Ngn3EYFP/+ E12.5 (d 0) embryos and removed from the surrounding mesenchyme. The development of islet progenitor cells can thus be followed in the same explant over the time of the culture period by simply monitoring the EYFP fluorescence. As shown in Fig. 7, at d 1 only a few islet progenitor cells are observed; their number next increases, peaks at d3–d4, and decreases at later stages when the number of 
- and ß-cells increases (data not shown). Like in vivo, this expression pattern reflects the wave of Ngn3+ cells observed during the secondary transition phase in the developing pancreas. To gain a more dynamic insight into endocrine progenitor cell development, time-lapse image acquisition of EYFP-labeled living cells was performed on E12.5 dpc embryonic pancreas epithelium between d2 and d4 of culture (see video published as supplemental data on The Endocrine’s Society’s Journals Online web site at http://mend.endojournals.org). In this sequence the expansion of the EYFP+ endocrine progenitor cells can clearly be observed during this period, with cells turning on and then off the EYFP protein, illustrating the transient expression of the Ngn3 gene in the developing endocrine lineage. Importantly, the division of EYFP-labeled cells into two EYFP+ cells can clearly be observed in some cases, which indicates that NGN3-EYFP+ endocrine progenitors are formed not only de novo but also by division from existing Ngn3-EYFP+ cells giving rise to two equally Ngn3-EYFP+ daughter cells.
View larger version (81K):
[in this window]
[in a new window]
|
Fig. 7. Pancreas Explant Cultures of Ngn3EYFP/+ Embryos
E12.5 dorsal pancreas was dissected and, after removal of the mesenchyme, the epithelium was subsequently cultured for 8 d on collagen-coated Petri dishes. From d1 to d4 the number of EYFP-labeled cells (EYFP) progressively increases (exposure time, 1 sec) after which (d5 to d8) it rapidly decreases (exposure time, 6 sec). EYFP expression pattern mimics the in vivo wave of Ngn3-expressing islet progenitor cells during pancreas development. Bright-field images show the overall development of the explant. Magnification, x20.
|
|
Effects of the Mesenchyme on the Development of EYFP-Labeled Endocrine Progenitor Cells
Recent studies have shown that, in vitro, the mesenchyme may not be required for the differentiation of the endocrine pancreas but rather necessary for the development of the exocrine counterpart (14, 15). Moreover, it was suggested that signals from the mesenchyme stimulate epithelial cell proliferation but repress the development of the pancreatic epithelium into endocrine cells. To determine directly the effects of the mesenchyme on the development of endocrine progenitor cells, pancreatic rudiments dissected from Ngn3EYFP/+ E12.5 embryos were grown either with or without mesenchyme (Fig. 8A). The numbers of both immature (EYFP+) and differentiated endocrine cells (insulin or glucagon+) per explant were counted on d3 and d5 of culture (Fig. 8B). We found, as previously observed by others, that islet cell differentiation is repressed in the presence of mesenchyme. At d5, for example, we counted about 3 times more insulin- and glucagon-producing cells when the mesenchyme was removed. In contrast, at d3 no major differences were observed in the development of the EYFP+ islet progenitor cells in the pancreatic epithelia grown with or without the mesenchyme. Their number was initially high and decreased thereafter concomitantly with the increase of differentiated cells. At d5, however, we observed a slightly higher number of progenitor cells in the absence of mesenchyme. These results suggest that the generation of islet progenitor cells does not require signals from the mesenchyme and that endocrine fate commitment would be a default pathway. They also suggest that the mesenchyme does not massively repress the development of EYFP-progenitors in vitro in our experimental system.
View larger version (47K):
[in this window]
[in a new window]
|
Fig. 8. Effects of the Pancreatic Mesenchyme on the Development and Differentiation of EYFP-Labeled Islet Progenitor Cells in Vitro
Ngn3EYFP/+ pancreatic E12.5 dpc explants were cultivated in vitro with or without their surrounding mesenchyme for 3 or 5d, and the expression of EYFP (green) and insulin and glucacon (Ins/Glu, red) was analyzed by immunofluorescence on cryosections. B, Quantitative analysis of pancreatic development in vitro. The absolute number of EYFP and insulin/glucagon-expressing cells was analyzed in explants of Ngn3EYFP/+embryos surrounded by its mesenchyme (black bars) or depleted of it (hatched bars), after 3 and 5 d in culture. Each data point represents the mean ± SEM of three to five (n) rudiments. The repressive effect of the mesenchyme on the generation of - and ß-cells can clearly be seen, whereas the development of EYFP+ islet progenitor cells is almost unaffected at d 3.
|
|
|
DISCUSSION
|
|---|
An important step in our understanding of islet cell development has been the identification of the islet cell progenitor marker, the transcription factor Ngn3. Indeed, Ngn3-expressing cells can give rise to all the four islet cell types in vivo, and the inactivation of this gene in mice has proven its essential function in controlling islet fate in uncommitted pancreatic endodermal cells. The extensive study of this population of progenitor cells is thus of particular interest in the light of the important international effort to develop a ß-cell replacement therapy for type 1 diabetes. The identification of this marker allows the labeling of Ngn3-expressing progenitor cells, which is essential for their purification and subsequent characterization both in vivo and in vitro. Here we have generated a mouse line in which islet progenitor cells expressing Ngn3 transcription factor have been genetically labeled with EYFP using a knock-add-on approach. The genetically modified animals appear healthy, and functional pancreatic islets develop in mice heterozygous or homozygous for the EYFP transgene. This demonstrates that the expression of EYFP protein in the developing islet progenitor cells is not toxic and that insertion of the reporter gene in the 3'-UTR region of Ngn3 does not impair the proendocrine function of Ngn3. Thus the knock-add-on approach we have adopted here could potentially be applied for other genes to genetically label a specific cell population in the developing pancreas. This approach combines the advantages of ensuring the tissue-specific expression of the reporter gene without simultaneously inactivating the function of the targeted gene at the same time. Robust fluorescence was observed in the developing pancreas of Ngn3EYFP/+ mice after dissection or when grown in vitro, and the background fluorescence was low and clearly distinguishable from the true signal. This demonstrates that one copy of the EYFP gene expressed under the control of Ngn3-regulatory sequences is enough to monitor islet progenitor cell development using standard fluorescence microscopy. After standard fixation procedure of the tissue, intrinsic fluorescence of the EYFP could not be detected, but the EYFP protein can be visualized by immunofluorescence.
In Ngn3EYFP/+ mice, more then 80% of the pancreatic cells expressing the Ngn3 protein during the secondary transition phase, when most of the endocrine cells are generated, are successfully labeled with EYFP reporter protein. The remaining Ngn3+/EYFP-negative cells might still express EYFP protein at levels that are not detected by the anti-EGFP polyclonal antibody which also recognizes EYFP. Alternatively, these Ngn3+ cells are recently born and do not yet express the reporter due to the slower maturation of the EYFP protein (21). This might be the main reason why at E10.5, where Ngn3 expression has just recently started, a slightly higher percentage of Ngn3+ cells (20%) do not costain with EYFP compared with later stages. The proportion of Ngn3-/EYFP+ cells might reflect the lack of sensitivity of the Ngn3 antibody, which would not detect all the Ngn3-expressing cells. An additional possibility, which does not exclude the previous hypothesis, is that some more mature progenitor cells are labeled due to a greater stability of the EYFP protein maintained for a short period in cells that have transiently expressed Ngn3. Supporting this hypothesis a fraction of differentiated islet cells (EYFP+/GLU+; EYFP/INS+) coexpress the reporter protein. Similar observations have already been made in transgenic mice in which the expression of the ß-galactosidase gene is driven by a human (19) or a mouse (11) Ngn3 promoter fragment, also finding Ngn3 and ß-galactosidase single positive cells as well as occasionally ß-cells coexpressing insulin and ß-galactosidase. The latter problem could be bypassed by using destabilized fluorescent reporter proteins when designing future strategies to label specific stages of developing islets cells. Using time lapse video (see supplemental data), we could monitor in real time the development of EYFP+ cells in cultures of pancreas explants, mimicking the wave of generation of islet progenitor cells during E13.5 and E15.5. We could also observe that EYFP+ cells can give rise to two EYFP+ daughter cells, suggesting that Ngn3 endocrine progenitors can also divide into two equivalent cells. This demonstrates the usefulness of this experimental system to follow the effect of extrinsic factors on the induction and expansion of islet progenitor cells, simply by monitoring the fluorescence. We took advantage of our explant cultures to follow the EYFP-labeled endocrine progenitor cells in the presence or absence of the surrounding mesenchyme, a source of signaling molecules (15). In both conditions, islet progenitor cells developed in cultures to a similar level. In contrast, the differentiation of insulin- and glucagon-expressing cells was clearly repressed in the presence of the mesenchyme as previously reported (14). These results suggest that endocrine cell fate commitment is occurring independently of signals from the mesenchyme or that growth factors present in the fetal calf serum (FCS) are sufficient to trigger endocrine fate decisions in cultures of embryonic pancreases. Moreover, these data suggest that in our experimental system, the repressing effects of the mesenchymal signals on islet cell development act by inhibiting the differentiation of islet progenitor cells into hormone-producing cells rather than on the generation of these progenitors from uncommitted epithelial cells.
We believe that the Ngn3EYFP/+ mice will be a useful tool to isolate and characterize islet progenitor cells at the molecular and cellular levels as well as Ngn3+ progenitors from different origins such as the intestine, stomach, and neural epithelium. Indeed, we will be able to define the gene expression profiles of purified progenitors and their differentiation potential in vitro. Moreover, compound technicolor mice can be generated by crossing the Ngn3EYFP/+ mice with other genetically engineered mouse lines in which different fluorescent proteins have been tagged to key genes controlling the development of islet cells and/or mice in which ß-cells are labeled with GFP (16). Such rainbow mice will be instrumental for the purification of subpopulations of pancreatic epithelial cells and the characterization of the sequential maturation steps of islet cells by functional genomics, in particular to identify specific cell surface markers. Moreover, such rainbow mice can be used as a unique source of multipotent stem cells in experiments aimed at generating ß-cells in vitro. Indeed, the appropriate culture conditions that would recapitulate normal islet cell development could be easily monitored in real time because the activation of the proper gene cascade would also turn on the different fluorescent reporters.
|
MATERIALS AND METHODS
|
|---|
Targeted Introduction of EYFP into the Ngn3 Locus
Using ET (RecE/RecT) recombination in bacteria (17) an AscI and a PmeI recognition site was introduced into the 3' UTR of a 7.8-kb SpeI–SpeI genomic fragment from the Ngn3 locus using the primers 5'-TTC TGG CTT TCA CTA CTT GGA TCC CTA GCC CTC TCA CAG GGC TTA ACT AGG CCG TTT AAA CCG AAT AAA TAC CTG TGA CGG AAG and 3'-AGA GCA GCC CCC AAT GTC TGC TGT GCG CAG CAG CAA GGG TAC CGA TGA GAA GGC GCG CCT AAC GAC CCT GCC CTG AAC CGA CGA. An "IRES-EYFP-PGKpA-loxP-SV40 early-promoter-pac-HSVtkpA-LoxP" cassette containing the EYFP reporter and the Puromycin resistance gene pac was subcloned into the AscI and PmeI site. The IRES-EYFP-PGKpA-loxP-SV40 early-promoter-pac-HSVtkpA-LoxP cassette was generated by replacing the IRES-EGFP sequence from the pIGuX1 plasmid (generously provided by F. Stewart) with the IRES-EYFP sequence from pIRES-EYFP Clontech vector (CLONTECH, Palo Alto, CA). Homologous recombination in ES cells and generation of chimeric mice were performed at the ICS (Mouse Clinical Institute), Illkirch, France. The final targeting vector was linearized with an external unique NotI restriction site and electroporated into R1 ES cells (18), and puromycin selection was performed (1.5 µg/ml during 10 d). Homologous recombination events were detected in resistant colonies by Southern blotting using a 1.2-kb 3'-external probe generated from a KpnI–NotI Ngn3 genomic subclone. Independently targeted ES cell clones (n = 44) were obtained from 298 clones analyzed and further examined with a 0.48-kb 5'-external probe generated from a Ngn3 genomic clone by PCR using the primers 5'-CCC CTC TTC TCC CTT TGT TC and 3'-ACA CAT GGA TTT GGC ACT GA. Two correct targeted independent ES cell clones (K20/79, K20/119) were used to generate chimeras by standard procedures. Germline transmission was obtained by crossing the chimeras with C57BL/6J females, and the colony was amplified by crossing heterozygous animals with CD1 mice. Genotyping was performed by PCR of genomic DNA obtained from postnatal tails by using the primers Ngn3 5'and 3' as described by Gradwohl et al. (6) to detect the wild-type allele and EYFP 5'-CCT GAA GTT CAT CTG CAC CAC and EYFP 3'-TTG TAG TTG TAC TCC AGC TTG TGC to detect the mutant allele (300 bp). Heterozygous females were crossed with CMV-Cre males (13) to excise the puromycin resistance pac gene. When successful excision of the pac gene occurred, a 558-bp PCR product is amplified using the primer EYFP-pA 5'-CAT TTT GAA TGG AAG GAT TGG and Ngn3 3' from genomic tail DNA of the offspring.
Live Imaging
Pancreases were dissected in ice-cold 1x PBS from Ngn3EYFP/+ embryos in E13.5, E15.5, E18.5, and adult mice. The intrinsic EYFP expression from dissected pancreas or cultivated pancreatic explants was analyzed using a Leica MZFLIII stereo-dissecting microscope or a Leica DMIRE2 microscope (Leica Corp., Deerfield, IL), respectively, with a YFP filter, and pictures were taken with a CoolSnap HQ camera (Roper Scientific, Tucson, AZ) and the MetaView software (Universal Imaging Corp., Downingtown, PA).
Immunohistochemistry
Immunofluorescence experiments were performed on cryosections as described previously (20). The following primary antibodies were used: rabbit anti-GFP at 1:500 (Molecular Probes, Eugene, OR), mouse antiinsulin at 1:1000 (Sigma Chemical Co., St. Louis, MO), mouse antiglucagon at 1:2000 (Sigma). The guinea pig anti-Ngn3 was used at 1:1000 and was generated by the Sander laboratory against the same glutathione-S-transferase-Ngn3 fusion protein as described by Schwitzgebel et al. (10). Secondary antibodies used were: Alexa 488 antirabbit at 1:500 (Molecular Probes), Cy3 antiguinea pig and Alexa 568 antimouse at 1:500 (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA; and Molecular Probes, respectively). Images were taken on an Olympus AX60 fluorescence microscope equipped with a F-view Olympus camera and SIS Olympus software (Olympus Corp., Lake Success, NY).
Culture of Embryonic Pancreas
Dorsal pancreases were dissected from Ngn3EYFP/+ embryos on E12.5. The pancreatic epithelium was separated from its surrounding mesenchyme by incubating the pancreases in DMEM/F12 (1:1) (Life Technologies, Gaithersburg, MD) supplemented with 0.8 mg/ml of Collagenase IAS (Sigma) at 37 C for 25 min. The collagenase is then substituted by FCS:DMEM/F12 (1:1). The epithelium was then mechanically removed from the surrounding mesenchyme using needles on a gel in a Petri dish [0.25% Agar, 25% HBSS (Life Technologies), 75% F12/DMEM]. Pancreatic epithelia were cultivated on rat tail collagen I coated Petri dishes (6 µg collagen/1 cm2) in F12/DMEM medium supplemented with 10% FCS (Life Technologies), Pen/Strep (Life Technologies), 2 mM L-glutamine (Life Technologies), and 0.04 mg/ml gentamycin (Life Technologies). Cultures were maintained at 37 C in a humidified atmosphere of 95% air-5% Co2, and the medium was changed every 2 d. For the immunohistochemistry analysis, pancreatic explants were fixed in p-formaldehyde 4% on ice for 15 min, followed by three washes with ice-cold PBS for 5 min, and equilibration in 20% sucrose for 10 min. Finally, explants were embedded in OCT compound (Tissue-Tek, Sakura, Japan) and frozen on dry ice, and consecutive sections of 6 µm were collected on Superfrost Plus (Menzel Glaser) slides.
Quantitative Analysis
To determine the number of each cell type per pancreatic rudiment, serial 6-µm sections were cut and collected. To avoid analyzing and counting the same cell twice, one of two consecutive sections (i.e. sections separated by 12 µm) were collected on a parallel prepared glass slide. One of the two glass slides was then analyzed by immunocytochemistry for a given antigen. The results of each experimental point were obtained by quantifying the absolute number of each pancreatic cell type in at least three pancreatic rudiments.
Glucose Tolerance Test
Glucose tolerance was tested in overnight-fasted mice by ip injection with 2 g/kg body weight of 15% D-glucose. Circulating glucose was measured in tail blood at 0, 15, 30, 45, 60, 90, 120, and 180 min using a GlucoMen Sensor (A. Menarini Diagnostics, Firenze, Italy).
Preparation of Single-Cell Suspensions from Dissected Pancreases and Flow Cytometric Cell Sorting
Dissected pancreases from E15.5 ngn3EYFP/+ embryos were incubated in a solution of 0.05% (vol/vol) trypsin-EDTA (Life Technologies) in PBS at 37 C for 5min. The digestion was stopped by adding F10 medium containing 10% (vol/vol) FCS. The resulting suspension of single cells was washed twice with PBS and filtered through a 70-µm nylon mesh. Flow cytometric analysis and sorting of EYFP-labeled progenitors were carried out using a FACS Vantage SE DiVa high-speed cell sorter with DiVa software (Becton Dickinson Biosciences, San Jose, CA). EYFP-expressing cells were detected after excitation with the 488-nm line of an Argon Ion Laser (200 mW) using the signal area of the FL 1 (530/30 nm band pass) and FL 2 (585/42 nm band pass) channels, separated by a 560 short pass dichroic filter (corresponding to the standard fluorescein isothiocyanate and phycoerythrin channels of the sorter).
Sorting of the EYFP-positive cells was performed using a 100-µm flow cell at 40 psi and 52 kHz. The samples and sorted fractions were kept at 4 C. All cells within the gate, as shown in Fig. 6
, were sorted as positives.
Video Time-Lapse Analysis
Ngn3EYFP/+ pancreatic explants were prepared as described above. Video time-lapse analysis were started on d 2 of culture using a Leica DMIRB2 microscope with a YFP filter, and pictures were taken every 30 min with a CoolSnap HQ camera and the MetaMorph software (see supplemental data).
|
ACKNOWLEDGMENTS
|
|---|
We thank Michèle Kedinger for critical reading of the manuscript and for helpful discussions; Caroline Daurat, Viviane Hauer, and Mélanie Messmer for excellent technical assistance; Yollande Huss and the ICS (Illkirch, France) ES cell core for the generation of the recombined ES cell and the transgenic core of the ICS for the generation of the chimeric mice; Claudine Ebel and the Cell Sorting Core of the IGBMC Illkirch for assistance with FACS analysis; and Marcel Boeglin, Didier Hentsch, Jean Luc Vonesch, and the ICS Imaging Core for assistance with the time-lapse analysis.
|
FOOTNOTES
|
|---|
This work was supported by funds from Institut National de la Santé et de la Recherche Médicale (INSERM) (Avenir grant), National Institutes of Health (NIH Grant 1U19-DK611244-01), and the Juvenile Diabetes Research Fundation (JDRF; Grant 4-2001-434) to G.G. G.G. is a partner of the JDRF center for ß cell therapy in Europe. G.M. was supported by a postdoctoral fellowship from Human Frontier Science Program and M.M was supported by a postdoctoral fellowship from INSERM. M.J. is a recipient of a Ph.D. studentship from the Ministère de la Recherche et de la Technologie.
Abbreviations: CMV-Cre, Cre recombinase under the control of cytomegalovirus promoter; dpc, days post coitus; E15.5, embryonic d 15.5; EGFP, enhanced green fluorescent protein; ES cell, embryonic stem cell; EYFP, enhanced yellow fluorescent protein; FACS, fluorescence-activated cell sorting; FCS, fetal calf serum; IRES, internal ribosome entry site; Ngn3, neurogenin 3; UTR, untranslated region.
Received for publication June 15, 2004.
Accepted for publication July 27, 2004.
|
REFERENCES
|
|---|
- Jensen J 2004 Gene regulatory factors in pancreatic development. Dev Dyn 229:176–200[CrossRef][Medline]
- Wilson ME, Scheel D, German MS 2003 Gene expression cascades in pancreatic development. Mech Dev 120:65–80[CrossRef][Medline]
- Feng G, Mellor RH, Bernstein M, Keller-Peck C, Nguyen QT, Wallace M, Nerbonne JM, Lichtman JW, Sanes JR 2000 Imaging neuronal subsets in transgenic mice expressing multiple spectral variants of GFP. Neuron 28:41–51[CrossRef][Medline]
- Hadjantonakis AK, Macmaster S, Nagy A 2002 Embryonic stem cells and mice expressing different GFP variants for multiple non-invasive reporter usage within a single animal. BMC Biotechnol 2:11[CrossRef][Medline]
- Hadjantonakis AK, Dickinson ME, Fraser SE, Papaioannou VE 2003 Technicolour transgenics: imaging tools for functional genomics in the mouse. Nat Rev Genet 4:613–625[Medline]
- Gradwohl G, Dierich A, LeMeur M, Guillemot F 2000 neurogenin3 is required for the development of the four endocrine cell lineages of the pancreas. Proc Natl Acad Sci USA 97:1607–1611[Abstract/Free Full Text]
- Gu G, Dubauskaite J, Melton DA 2002 Direct evidence for the pancreatic lineage: NGN3+ cells are islet progenitors and are distinct from duct progenitors. Development 129:2447–2457
- Apelqvist A, Li H, Sommer L, Beatus P, Anderson DJ, Honjo T, Hrabe de Angelis M, Lendahl U, Edlund H 1999 Notch signalling controls pancreatic cell differentiation. Nature 400:877–881[CrossRef][Medline]
- Jensen J, Heller RS, Funder-Nielsen T, Pedersen EE, Lindsell C, Weinmaster G, Madsen OD, Serup P 2000 Independent development of pancreatic - and ß-cells from neurogenin3-expressing precursors: a role for the notch pathway in repression of premature differentiation. Diabetes 49:163–176[Abstract]
- Schwitzgebel VM, Scheel DW, Conners JR, Kalamaras J, Lee JE, Anderson DJ, Sussel L, Johnson JD, German MS 2000 Expression of neurogenin3 reveals an islet cell precursor population in the pancreas. Development 127:3533–3542[Abstract]
- Jenny M, Uhl C, Roche C, Duluc I, Guillermin V, Guillemot F, Jensen J, Kedinger M, Gradwohl G 2002 Neurogenin3 is differentially required for endocrine cell fate specification in the intestinal and gastric epithelium. EMBO J 21:6338–6347[CrossRef][Medline]
- Lee CS, Perreault N, Brestelli JE, Kaestner KH 2002 Neurogenin 3 is essential for the proper specification of gastric enteroendocrine cells and the maintenance of gastric epithelial cell identity. Genes Dev 16:1488–1497[Abstract/Free Full Text]
- Dupe V, Davenne M, Brocard J, Dolle P, Mark M, Dierich A, Chambon P, Rijli FM 1997 In vivo functional analysis of the Hoxa-1 3' retinoic acid response element (3'RARE). Development 124:399–410[Abstract]
- Miralles F, Czernichow P, Scharfmann R 1998 Follistatin regulates the relative proportions of endocrine versus exocrine tissue during pancreatic development. Development 125:1017–1024[Abstract]
- Scharfmann R 2000 Control of early development of the pancreas in rodents and humans: implications of signals from the mesenchyme. Diabetologia 43:1083–1092[CrossRef][Medline]
- Hara M, Wang X, Kawamura T, Bindokas VP, Dizon RF, Alcoser SY, Magnuson MA, Bell GI 2003 Transgenic mice with green fluorescent protein-labeled pancreatic ß-cells. Am J Physiol Endocrinol Metab 284:E177–E183
- Angrand PO, Daigle N, van der HF, Scholer HR, Stewart AF 1999 Simplified generation of targeting constructs using ET recombination. Nucleic Acids Res 27:e16
- Nagy A, Rossant J, Nagy R, Abramow-Newerly W, Roder JC 1993 Derivation of completely cell culture-derived mice from early-passage embryonic stem cells. Proc Natl Acad Sci USA 90:8424–8428[Abstract/Free Full Text]
- Lee JC, Smith SB, Watada H, Lin J, Scheel D, Wang J, Mirmira RG, German MS 2001 Regulation of the pancreatic pro-endocrine gene neurogenin 3. Diabetes 50:928–936[Abstract/Free Full Text]
- Cau E, Gradwohl G, Fode C, Guillemot F 1997 Mash1 activates a cascade of bHLH regulators in olfactory neuron progenitors. Development 124:1611–1621[Abstract]
- Nagai T, Ibata K, Park ES, Kubota M, Mikoshiba K, Miyawaki A 2002 A variant of yellow fluorescent protein with fast and efficient maturation for cell-biological applications. Nat Biotechnol 20:87–90[CrossRef][Medline]
This article has been cited by other articles:
|
|
|
|
 
T. Sugiyama, R. T. Rodriguez, G. W. McLean, and S. K. Kim
Conserved markers of fetal pancreatic epithelium permit prospective isolation of islet progenitor cells by FACS
PNAS,
January 2, 2007;
104(1):
175 - 180.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. Chamson-Reig, S. M Thyssen, E. Arany, and D. J Hill
Altered pancreatic morphology in the offspring of pregnant rats given reduced dietary protein is time and gender specific.
J. Endocrinol.,
October 1, 2006;
191(1):
83 - 92.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
S. Thyssen, E. Arany, and D. J. Hill
Ontogeny of Regeneration of {beta}-Cells in the Neonatal Rat after Treatment with Streptozotocin
Endocrinology,
May 1, 2006;
147(5):
2346 - 2356.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
