SEARCH

SEARCH BY CITATION

Keywords:

  • immunohistochemistry;
  • islet;
  • morphogenesis;
  • pancreas;
  • proliferation;
  • mantle–core arrangement;
  • three-dimensional reconstruction

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The organogenesis of islets in rat pancreas was studied by three-dimensional reconstructions from serial section micrographs. On embryonic day (E) 12, an endocrine cluster consisting mainly of glucagon-expressing cells maintained connection with the pancreatic endoderm at several regions. On E15–E17, the cluster enlarged by fusion of newly formed buds. Although the proportion of insulin-expressing cells increased, they were located in the periphery of the cluster. On the day of birth, insulin-expressing cell clusters enlarged and fused to form several cores within the islet. The glucagon-expressing cell mass expanded to form a thin mantle covering the cores. During islet organogenesis, proliferation activity was high in the exocrine duct system. Moreover, the endocrine cell clusters maintained contact with the duct epithelium throughout. We conclude that the pancreatic islet is generated by the unification of multiple endocrine clusters originated from separate regions of the duct system. The mechanism of mantle–core formation is discussed. Developmental Dynamics 236:3451–3458, 2007. © 2007 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The pancreas is an integrated organ involved in both exocrine and endocrine functions. The exocrine portion occupies more than 95% of the pancreas, is composed of acinar and ductal cells, and secretes digestive enzymes that promote nutrient digestion. In contrast, the endocrine portion is composed of many cell clusters called pancreatic islets, or islets of Langerhans. The islets contain four principal endocrine cell types: insulin-producing β cells, glucagon-producing α cells, somatostatin-producing δ cells, and pancreatic polypeptide-producing PP cells (Kim and Hebrok,2001). The arrangement of these endocrine cells within the islet is described as mantle–core, in which a core of β cells is surrounded by a mantle of non-β cells of one to three cells in thickness (Bonner-Weir and Smith,1994). There is quite a sharp segregation of β and non-β cells within the islet in rats, although this segregation is less clear or partial in humans and mice.

In embryonic rat, glucagon-expressing cells appear individually or as small cell clusters distinct from other endodermal cells until about embryonic day (E) 11. This finding is followed by the appearance of a large number of insulin-expressing cells generated from the duct epithelium at around E16. However, the typical mantle–core arrangement becomes evident around E17–E21 (Kim and Hebrok,2001; Jensen,2004). Complex combinations of transcription factors are required for the development of individual cell type (Dohrmann et al.,2000). Both descriptive (Hard,1944; Pictet and Rutter,1972) and experimental (Kim and Hebrok,2001; Miralles et al.,1998) studies have attempted to reveal the mechanisms of the formation of the mantle–core structure; however, its dynamic process remains largely unknown.

The pancreatic islet is a complex structure. In fact, the islet is not a cell mass isolated from the surrounding exocrine portion, but is connected directly to exocrine duct cells (Bouwens et al.,1994; Bertelli et al.,2001). Moreover, the islet is too large for structural study by conventional two-dimensional (2-D) histology. In the present study, we elucidated the mechanisms underlying the formation of pancreatic islet by examining islet organogenesis by computer-aided 3-D reconstruction of embryonic rat pancreas in combination with immunohistochemistry of several islet cell hormones. As a result, we suggest that the rapidly increasing insulin-expressing cell clusters fuse to form central cores within the endocrine portion, whereas the slowly growing glucagon-expressing cell mass spreads to form the thin peripheral mantle layer.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Detection of Endocrine Cells in the Developing Pancreas

As a preliminary study, a set of three serial sections from pancreas of E12, E15, E17, and newborn rats were prepared. Hematoxylin and eosin (H&E) staining, double immunohistochemistry for glucagon and insulin, and double immunofluorescence for glucagon and insulin were conducted. In H&E stained sections from E15 pancreas, two populations of cells could be distinguished: solitary or clusters of cells with abundant eosinophilic cytoplasm and small immature darkly stained exocrine duct cells (Fig. 1A). Comparison of serial sections stained by H&E, immunohistochemistry (Fig. 1B) and immunofluorescence (Fig. 1C) indicated that the large eosinophilic cells corresponded to the cells expressing either glucagon or insulin. Similar findings were obtained in E12 pancreas (data not shown). Therefore, in the case of E12 and E15 pancreas, contours of the areas occupied by large eosinophilic cells in H&E-stained serial sections were fed into a computer for 3-D reconstruction of endocrine portion. As development advanced, however, it became difficult to discriminate endocrine from exocrine cells only by H&E staining. In a preliminary study, we confirmed that the majority of the cells expressing secretogranin III (Sakai et al.,2004) also expressed either insulin or glucagon in the dorsal pancreas (data not shown). Thus, 3-D reconstructions of endocrine portions from E17 were performed using serial sections stained immunohistochemically for secretogranin III.

thumbnail image

Figure 1. Three serial sections of embryonic day (E) 15 pancreas stained by three different methods. A: Hematoxylin and eosin (H&E) staining. B: Double immunohistochemical staining for insulin (brown) and glucagon (blue). C: Double immunofluorescent staining for insulin (green) and glucagon (red). A group of cells with abundant eosinophilic cytoplasm stained by H&E coincides exactly with a group of endocrine cells demonstrated both by immunohistochemistry (B) and immunofluorescence (C). Scale bar = 50 μm.

Download figure to PowerPoint

Dorsal Pancreatic Endoderm and Its Endocrine Cells in E12 and E15 Rats

On E12, the dorsal pancreatic rudiment consisted of columnar or cuboidal stratified epithelial cells. They surrounded a definite lumen that continued to the duodenal lumen. There was no morphological indication of cytodifferentiation into acinar and ductal cells. Instead, it was possible to distinguish clusters of 10 to 20 cells with more abundant eosinophilic cytoplasm than other undifferentiated epithelial cells by H&E staining (Fig. 2A–C, arrowheads). Double immunohistochemistry for glucagon and insulin performed on the neighboring section revealed that the large eosinophilic cell cluster was composed mainly of glucagon-expressing cells (Fig. 2D, arrowhead), with a few randomly distributed insulin-expressing cells. Some endocrine cells were also seen scattered in the undifferentiated epithelium, but they appeared solitary or in small aggregates of two to three cells. To examine the structural relationship between undifferentiated epithelium and endocrine cell clusters, we further studied the morphology of dorsal pancreatic rudiment by computer-aided 3-D reconstruction of the rudiment (Fig. 2E,F). Three sets of serial sections were reconstructed and analyzed. Surprisingly, each dorsal pancreas was always found to contain one large dorsal endocrine cell cluster and several small endocrine cell clusters scattering throughout the endoderm. The clusters were always connected to the undifferentiated endoderm at several regions as shown in Figure 2D.

thumbnail image

Figure 2. A–C: A dorsal pancreas of an embryonic day (E) 12 embryo, as shown in three sections, each separated by 15 μm (hematoxylin and eosin [H&E] staining). An eosinophilic endocrine cluster (arrowhead) and solitary or small aggregate of endocrine cells (double arrows) are seen in all three sections. The large cluster appears separated from the pancreatic endoderm by a thin layer of mesenchyme in B; but connections with the undifferentiated endoderm are evident at three independent regions (A and C, single arrow). D: Double immunohistochemical staining of dorsal pancreatic bud for insulin (brown) and glucagon (blue). Glucagon-expressing cells are the most abundant in the large endocrine cluster (arrowhead). Solitary or small aggregates of glucagon-expressing cells are also observed. Insulin-expressing cells are only seen as single cells within the large endocrine cluster. E: An example of computer-aided three-dimensional (3-D) reconstruction image of the dorsal pancreatic bud reconstructed from a set of 21 sections. Undifferentiated endoderm and endocrine cells are colored light green and red, respectively. The rudiment contains one large endocrine part (arrowhead) and several smaller aggregates or solitary endocrine cells. F: A 3-D image of the endocrine portions (red) and areas of contact with undifferentiated endoderm (yellow). The image was reconstructed from E by removing the exocrine portion using computer graphics. Note that the endocrine portions are in contact with the pancreatic endoderm at several points. Scale bars = 50 μm.

Download figure to PowerPoint

In E15 rats, polarized ductal cells surrounding a definite lumen became more evident in the rudimental pancreas. In addition, there were clusters of 10 to 20 undifferentiated eosinophilic cells (Fig. 3A–C, arrowheads). Indeed, double immunohistochemistry for glucagon and insulin performed on serial sections revealed that these clusters consisted mainly of glucagon-expressing cells (Fig. 3D, arrowhead). We reconstructed 3-D images from three sets of serial sections and found that there was only one large endocrine portion within each dorsal pancreas (Fig. 3E,F). The continuities between endodermal duct and large endocrine cell cluster were also confirmed in 3-D reconstructive images (Fig. 3F). These results suggest that the endocrine cluster is enlarged by fusion of several small endocrine clusters that originated from different regions of pancreatic endoderm.

thumbnail image

Figure 3. A–C: A large endocrine cluster of embryonic day (E) 15 dorsal pancreas, as shown in three sections, each separated by 15 μm (hematoxylin and eosin [H&E] staining). A large cluster of eosinophilic cells (arrowhead) is surrounded by developing exocrine tissue. Connections of the cluster with endodermal duct are shown by arrows in A and C. D: Double immunohistochemistry for insulin (brown) and glucagon (blue) showing that eosinophilic cell clusters consist mainly of glucagon-expressing cells. Insulin-expressing cells are seen scattering within the cluster. Solitary or small aggregates of glucagon-expressing cells are also observed. E: A computer-aided three-dimensional (3-D) reconstruction image of a set of 29 serial sections. The undifferentiated endoderm and endocrine cells are colored light green and red, respectively. F: A 3-D image of the endocrine portions (red) and areas of contact with undifferentiated endoderm (yellow) reconstructed from E by removing the exocrine portion. Note that the endocrine clusters are connected with each other to form a complex endocrine islet. There is an increase in number of connections between endocrine and exocrine portions compared with E12. Scale bars = 50 μm.

Download figure to PowerPoint

Morphological Changes of Endocrine Cluster and Increasing Number of Insulin-Expressing Cells on E17

On E17, the dorsal pancreas grew rapidly to form a highly branched acino-tubular structure with polarized exocrine cells. At the same time, the number of endocrine cell clusters increased, although it became difficult to fully discriminate endocrine cells from exocrine portion by H&E staining alone. Hence, endocrine cells were carefully identified by immunohistochemistry for secretogranin III, a ubiquitous endocrine cell marker (Sakai et al.,2004). The individual endocrine cluster became branched but maintained good contact with the adjacent duct systems (Fig. 4A, arrowhead). Three-dimensional reconstruction analysis showed that the endocrine clusters were connected with each other in the central part of the dorsal pancreas to form a large islet (Fig. 4B,C). As shown in Figure 4C, an immature islet was in contact with ducts at more than 20 sites. We then examined the distribution of glucagon-expressing and insulin-expressing cells within the immature islet. Although the number of insulin-expressing cells increased significantly at this stage of development, there was no clear mantle–core arrangement of cells. In fact, glucagon-expressing cells assembled to form branched clusters, and most of them were found in the center of immature islets (Fig. 4D–F). However, some glucagon-expressing cell clusters elongated and made contact with the ductal epithelium. In contrast, the majority of insulin-expressing cells were located in the periphery of the branched endocrine cell clusters (Fig. 4D–F). A small number of solitary insulin-expressing cells were also recognized within the ductal epithelium.

thumbnail image

Figure 4. A: A section of dorsal pancreas from an embryonic day (E) 17 embryo immunohistochemically stained for secretogranin III, a ubiquitous marker for pancreatic endocrine cells. The endocrine clusters (arrowhead) branch and connect with several exocrine ducts. Some epithelial cells located in the duct wall are also positive for secretogranin III. B: Computer-aided three-dimensional (3-D) reconstructed image of a part of dorsal pancreas. The developing exocrine system and endocrine cells are colored light green and red, respectively. This image was reconstructed from a set of 34 serial sections. Endocrine clusters that appear separated in individual sections are shown to connect with each other to form a large endocrine part. C: A 3-D image of the endocrine portions (red) and areas of contact with the exocrine system (yellow) reconstructed from B by removing the exocrine portion. More than 20 connected regions are shown in C. D–F: Three sections of the same region of dorsal pancreas, each separated at 20-μm intervals, double immunostained for insulin (brown) and glucagon (blue). Four independent ducts, each surrounding a lumen independently, are marked d1 to d4. Glucagon-expressing cells are most abundant and appear to have increased in number forming aggregates, often located adjacent to the duct epithelium (arrowhead). They are connected distally to large glucagon-expressing cell clusters (arrow). Both glucagon- and insulin-expressing cells are also present singly or form small aggregates within the ductal endoderm. Scale bars = 50 μm.

Download figure to PowerPoint

Newly Formed Pancreatic Islet Maintaining Contact With the Duct System in Neonatal Rat

Differentiation of duct and acinar cells was evident in the neonatal rat pancreas. Double immunohistochemistry showed that, in several mature islets, insulin-expressing cells aggregated to form an inner core, while glucagon-expressing cells formed a mantle surrounding the core (Fig. 5A). A capillary network penetrated the islet (Fig. 5B, arrows). Three-dimensional reconstruction analysis clearly showed that large islets were connected with each other and formed one large endocrine portion. As shown in Figure 5C, insulin-expressing cells forming the inner cores arranged beads-like along the duct, while glucagon-expressing cells formed a thin belt-like mantle that partially surrounded the cores. Direct contact of ductal epithelium with both the glucagon-expressing cell mantle and insulin-expressing cell inner cores remained evident, but the number of contacts with the latter was reduced.

thumbnail image

Figure 5. Neonatal pancreatic islet double-stained for insulin (brown) and glucagon (blue). A: A section showing the distribution of glucagon- and insulin-expressing cells in an islet. A typical mantle–core arrangement is seen, in which insulin-expressing cells occupy the central core and are surrounded by glucagon-expressing cells. A mass of glucagon-expressing cells connects two separate insulin-expressing cell clusters (arrows). Mesenchymal tissue rich in capillaries is seen to penetrate the glucagon-expressing cell region connecting two insulin-expressing cells clusters. B: The islet is penetrated by a complex capillary network as marked by arrows. Note the close association of the exocrine duct (d) to an islet. C: A computer-aided three-dimensional (3-D) reconstruction image of an endocrine portion and duct system. Glucagon-expressing cell clusters, insulin-expressing cell clusters, and duct are colored light blue, red, and brown, respectively. Four large endocrine clusters (a–d), which might be identified as four independent islets in a section, are connected together to form one large endocrine portion. Partial but definite connecting regions are marked by white arrows. A close association of the endocrine portion with the branched duct system is also evident. This image was reconstructed from a set of 34 serial sections. Scale bars = 100 μm in A, 50 μm in B.

Download figure to PowerPoint

Cell Proliferation Analysis

To address the mechanisms underlying the dynamic changes in size and arrangement of the developing pancreatic endocrine portion, cell type-specific proliferation rates were examined by simultaneous bromodeoxyuridine (BrdU) -uptake analysis and immunohistochemistry for endocrine or duct cells markers using E15 and E17 pancreas. In E17 pancreas, a large proportion of ductal epithelial cells, which expressed duct-specific cytokeratin (Means et al.,2005), showed reactivity for BrdU (Fig. 6A,D). In contrast, BrdU-positive nuclei were seldom seen in glucagon-expressing cells (Fig. 6B,E) and insulin-expressing cells (Fig. 6C,F). Cells with BrdU-incorporated cells but lacking cytoplasmic expression of cytokeratin, glucagon, or insulin were mesenchymal cells or acinar cells. Similar results were seen in E15 rats (data not shown). We next counted the frequencies of BrdU-incorporated cells in cytokeratin-, insulin-, and glucagon-expressing cells. The results are summarized in Figure 7. A large proportion of proliferating cells were found in exocrine ductal epithelial cells but not in endocrine cells.

thumbnail image

Figure 6. Bromodeoxyuridine (BrdU) incorporation assay of embryonic day (E) 17 dorsal pancreas. A rat at gestational day 17 was injected intraperitoneally with BrdU. A–F: The embryos were removed, and distributions of BrdU-incorporated nuclei (arrowheads) in cells expressing cytokeratin (A,D), glucagon (B,E), and insulin (C,F) were examined by double labeling immunohistochemistry. All nuclei were counterstained with 4′,6-diamidine-2-phenylidole-dihydrochloride (DAPI; A,D) or hematoxylin (B,C,E,F). Portions surrounded by rectangular frame on A, B, and C are illustrated by higher magnification in D, E, and F, respectively. In A and D, many BrdU-incorporated nuclei (marked by green) are seen in the cells expressing cytokeratin (red). In B and E, and C and F, no BrdU-incorporated nuclei (marked by dark brown) are found in the cells expressing glucagon or insulin (blue). Scale bars = 50 μm in A–C, 100 μm in D–F.

Download figure to PowerPoint

thumbnail image

Figure 7. Quantification of bromodeoxyuridine (BrdU) -incorporated nuclei in insulin-, glucagon-, and cytokeratin-expressing cells in embryonic day (E) 15 and E17 dorsal pancreas. The vertical axis indicates the frequency of BrdU-incorporated nuclei (%). The proportion (%) of BrdU-incorporated nuclei among all cells expressing each marker was calculated and plotted. The open, shaded, and solid bars represent the labeling indexes in insulin-, glucagon-, and cytokeratin-expressing cells, respectively. Proliferating cells are recognized at higher rate in cytokeratin-expressing exocrine cells but not in insulin- or glucagon-expressing cells both on E15 and E17 pancreas. Data are means ± SE. *P < 0.005, **P < 0.01.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Several lines of evidence indicate that the early glucagon- and insulin-expressing cells are not necessary for generation of adult endocrine cells, and in fact do not contribute to the formation of definitive islet (Herrera,2002; Collombat et al.,2006). Previous studies also demonstrated that these early hormone-expressing cells often coexpress PP-like immunoreactivity (Herrera et al.,1991; Teitelman et al.,1993; Mulder et al.,1998). In fact, Jackerott et al. (1996) showed that the earlier PP-like immunoreactivity was attributed to pancreatic YY (PYY). According to their observation, the early glucagon-expressing cells, which also express PYY, remain until late fetal development and are gradually replaced by adult glucagon-expressing cells (mature α cells); and the adult insulin-expressing cells (mature β cells), which are negative for PYY, increase dramatically in embryonic day-14 to -18 rat pancreas.

Based on the regulated appearances of early and mature pancreatic hormone-expressing cells mentioned above, the pancreatic islet morphogenesis in the present study can be described as follows (Fig. 8). (1) Endocrine cell cluster consisting mainly of early glucagon-expressing cells (shown in blue) is generated in the central part of dorsal pancreas by fusion of several independent early glucagon-expressing cell bulges derived from different regions of the undifferentiated pancreatic duct system (Fig. 8A,B), with a few insulin-expressing cells scattered in the cluster. (2) The endocrine cell cluster enlarges and becomes more complex (Fig. 8C). At the same time, the proportion of mature β cells increases dramatically (Fig. 8D). These β cells are generated from the ductal epithelium and have been present in the islet periphery from the beginning. (3) The numbers of newly formed mature α and β cells increase. Due to a relatively rapid increase of mature β cells, their clusters fuse together to form central cores within the endocrine portion. The mature α cell mass spreads to form a thin mantle layer partially covering the mature β cells cores. Then, this large endocrine portion appears to be separated into several smaller islets by mesenchymal tissue (Fig. 8E). (4) The early glucagon-expressing cells, which were located distally to the newly generated β cells, might be abolished at this stage, although the elimination mechanism is not determined. At the same time, the mature α cells emerge to form a thin mantle layer partially covering the cores of mature β cells. Then, this large endocrine portion appears to be separated into several smaller islets by mesenchymal tissues. During this process, the area of contact between endocrine cells and the duct system decreases gradually.

thumbnail image

Figure 8. Schematic representation of islet morphogenesis of rat pancreas. A: Small early glucagon-expressing cell buds (shown in blue) originated from different regions of undifferentiated pancreatic duct epithelia (brown) are formed before embryonic day (E) 12. B: These endocrine buds, which consist mainly of early glucagon-expressing cells, fuse to form a large cluster. C: As the pancreatic endoderm grows to form highly branched structure, early glucagon-expressing cell buds increase in number. The endocrine cluster becomes more complex due to fusion of the newly formed endocrine buds with the pre-existing cluster, and is connected to the duct at multiple regions. D: The proportion of mature β cells (red) within the endocrine clusters increases dramatically after E17. At the same time, the endocrine cluster enlarges and becomes more complex. These mature β cells are generated from the duct epithelium and are located in the islet periphery from the beginning. E: Newly formed mature α and β cells increase in number rapidly in the perinatal stage. Due to a relatively rapid increase in mature β cells, these cell clusters fuse to form central cores within the endocrine portion. A mass of mature α cells spread to form a thin mantle layer to cover the mature β cell cores partially. This large endocrine element is then separated into several islets of smaller sizes by invasion of the mesenchyme. During organogenesis of pancreatic islet, endocrine cells are supplied by proliferation of ductal epithelial cells.

Download figure to PowerPoint

The present study showed that the endocrine portion maintained connection with the duct or undifferentiated endoderm at multiple points throughout development, confirming a report by Pictet and Rutter (1972). Because the BrdU uptake study revealed that cell proliferation occurred mostly in ductal cells, we conclude that the majority of endocrine cells originate from various segments of the ductal region and not from preexisting endocrine cells. It should be noted that the connecting points increase in number as the islets enlarge, suggesting that fusion of newly generated endocrine clusters with a pre-existing endocrine portion may account for the enlargement of the endocrine part, as suggested by Hard (1944). This explanation is in good agreement with the previous study using chimeric mice by Deltour et al. (1991) that demonstrated the polyclonal origin of islets β cells. Because a glucagon-expressing cell cluster is also connected to the duct system at multiple points, polyclonal origin of islet glucagon-expressing cells is also very likely.

Based on the finding of abnormal arrangements of islet cells in transgenic mice expressing dominant-negative forms of N-CAM (Esni et al.,1999) and cadherin (Dahl et al.,1996) molecules, a schematic model has been proposed, which explains that the mantle–core arrangement of the islet is achieved by endocrine cell dissociation from the endoderm and cell type-specific sorting during reaggregation (Kim and Hebrok,2001). However, we failed to observe any indication of dissociation of endocrine glucagon- and insulin-expressing cells within the clusters during islet morphogenesis. Instead, both clusters maintained good cell–cell contact and direct contact with the duct epithelium during this period. Therefore, dissociation and re-aggregation are less likely to be responsible. On this account, we cannot fully exclude the possibility that we might have missed very minute disconnection of islet cells less than 5 μm, because 3-D reconstruction was performed by serial sections with 5 μm thickness.

If dissociation and re-aggregation are not responsible for the mantle–core arrangement of islet glucagon- and insulin-expressing cells, how were preexisting glucagon-expressing cells eliminated from the center of immature islets; and how do peripherally located insulin-expressing cells penetrate to form the core of islet? The first question may be explained by gradual removal of early glucagon-expressing cells at the late embryonic stage (Jackerott et al.,1996). It is likely that early glucagon-expressing cells are eliminated by apoptosis (Collombat et al.,2006). The second question may be explained by fusion of rapidly expanding mature β cell clusters at the center of the endocrine portion. In this study, we examined the process of mantle–core formation from static observations. However, active cell movement should have taken place during the formation of special structure of the islet. Also cell adhesion molecules are most probably involved in endocrine cell arrangement. The dynamic and molecular mechanisms of islet morphogenesis in vivo require further studies.

Endocrine islets are distributed throughout the pancreas. In addition to the large endocrine cell cluster in the dorsal pancreatic rudiment, which has been addressed in the present study, endocrine cells also appeared in various dorsal and ventral regions (data not shown). However, we observed no clustering of endocrine cells in the E12 ventral pancreas, suggesting that islet morphogenesis varies in different parts of the pancreas. Further studies are required to investigate the morphogenesis of islets in ventral pancreas.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Animal and Tissue Preparations

Dated pregnant Wistar rats were purchased from Oriental Yeast Co. (Tokyo, Japan). They were kept in an experimental animal facility controlled at 23°C room temperature with a 12-hr light and dark cycle, and with free access to standard chow and water. The morning when the vaginal plug was detected was designated as embryonic day 0 (E0). Under ether anesthesia, embryos at E12, E15, or E17 were dissected from the dated pregnant rats. Whole bodies of embryos were immediately fixed in Bouin's fixative. For the newborn, the whole digestive tract including the pancreas was dissected under ether anesthesia. The specimen was fixed in Bouin's fixative at 4°C overnight. Twenty-four pregnant rats and 19 neonatal rats were used in this study. At least 7 pregnant rats at each stage were killed to obtain embryos at E12, E15, and E17. The animal experiment was conducted according to the guidelines for animal care issued by The Animal Experiment and Ethic Committee, Kitasato University, School of Medicine.

Histology and Immunohistochemistry

The fixed tissues were dehydrated with a graded ethanol series, cleared with xylene, and embedded in paraffin. For general morphology, serial horizontal sections (5 μm in thickness) were cut, deparaffinized, and stained with hematoxylin and eosin. The dorsal pancreas was identified as the organ in the dorsal mesogastrium connecting the duodenum and spleen, located dorsal to the stomach. For immunohistochemistry, sections were blocked with 0.3% hydrogen peroxidase in methanol (20 min) to inhibit endogenous peroxidase activity, treated with 1% bovine serum albumin (BSA) in phosphate buffered saline (PBS, 30 min) to block nonspecific antibody binding, and processed as described previously (Hayashi et al.,2003). The following primary antibodies were diluted with PBS containing 1% BSA and applied to the sections: rabbit polyclonal anti-glucagon antibody (1:500; Zymed Labs, Co., San Francisco, CA), guinea pig polyclonal anti-insulin antibody (1:1,000; DAKO, Carpinteria, CA), rabbit polyclonal anti-cytokeratin wide spectrum screening antibody (1:100; DAKO), and goat polyclonal anti-secretogranin III antibody (1:500; Santa Cruz Biotechnology, Inc., Santa Cruz, CA). For enzymatic detection of the bound primary antibodies, alkaline phosphatase (ALP) -conjugated anti-rabbit IgG, horseradish peroxidase (HRP) -conjugated anti–guinea pig IgG (both from Chemicon, Temecula, CA), and biotinylated anti-rabbit IgG and biotinylated anti-goat IgG (both from Vector Laboratories, Burlingame, CA) were used at 1:200 dilution (PBS with 1% BSA). For biotinylated antibodies, the sections were further incubated with HRP-conjugated streptavidin (1:500, Vector Laboratories). Enzyme reaction was detected using 3,3′-diaminobenzidine tetrahydrochloride (DAB) for HRP, and Vector blue ALP substrate (Vector Laboratories) for ALP. For immunofluorescence, Alexa Fluor 568–conjugated anti-rabbit IgG, and Alexa Fluor 488–conjugated anti-mouse IgG (both from Molecular Probes, Eugene, OR) were used at 1:200 dilution (PBS with 1% BSA). Sections were observed under a microscope (AX70, OLYMPUS, Tokyo, Japan). For double immunostaining for glucagon and insulin, both antibodies were simultaneously applied on sections. For negative control, sections were incubated with 1% BSA in PBS instead of the primary antibody and treated similarly. No significant reactivity other than nonspecific autofluorescence of erythrocytes was observed in all the negative controls.

Detection of Proliferating Cells

Proliferating cells were detected as previously described (Hayashi et al.,2003). In brief, rats at the indicated gestational days were injected intraperitoneally with BrdU at a dose of 3 mg/100 g body weight. After 4 hr, the rats were killed and the embryos were removed, fixed, and then processed for immunohistochemistry. Sections were double immunostained for incorporated BrdU (according to the manufacturer's instruction, Amersham Biosciences Co., Piscataway, NJ) and for specific cell markers as described above. As an indication of proliferation activity of insulin-, glucagon-, and cytokeratin-positive cells, we determined the labeling indices of BrdU-incorporated nuclei for insulin-, glucagon-, and cytokeratin-positive cells per specimen. Five specimens were analyzed at E15 and E17 separately. Statistical significance was determined using Student's t-test. All data are presented in mean ± SE.

Computer-Aided 3-D Reconstruction of Serial Sections

The 3-D images of the developing endocrine portions were reconstructed from 20 to 60 horizontal serial sections from each of three embryos at E12, E15, E17, and two neonatal rats. Sets of images from serial sections were acquired under light microscope equipped with a CCD camera (Olympus, Tokyo, Japan) and enlarged individual images (×400) were printed out. Using a digitizing tablet, the contours of the whole endocrine portion, the endocrine cell clusters, the pancreatic endoderm, the pancreatic ducts, and the contact areas of endocrine portions with ducts were fed into the computer as independent data (Yamashina et al.,1999), and reconstruction was performed with TRI 3-D reconstruction software (Ratoc System Engineering Co., Ltd., Tokyo). Representative images are illustrated in Figures 2 to 5.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

The authors thank Aya Nakagawa for secretarial assistance.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES
  • Bertelli E, Regoli M, Orazioli D, Bendayan M. 2001. Association between islets of Langerhans and pancreatic ductal system in adult rat. Where endocrine and exocrine meet together? Diabetologia 44: 575584.
  • Bonner-Weir S, Smith FE. 1994. Islets of Langerhans: morphology and its implications. In: KahnCR, WeirGC, eds. Joslin's diabetes mellitus. 13th ed. Philadelphia: Lea and Febiger. p 1525.
  • Bouwens L, Wang R, Blay ED, Pipeleers DG, Kloppel G. 1994. Cytokeratins as markers of ductal cell differentiation and islet neogenesis in the neonatal rat pancreas. Diabetes 43: 12791283.
  • Collombat P, Hecksher-Sorensen J, Serup P, Mansouri A. 2006. Specifying pancreatic endocrine cell fates. Mech Dev 123: 501512.
  • Dahl U, Sjodin A, Semb H. 1996. Cadherins regulate aggregation of pancreatic β-cells in vivo. Development 122: 28952902.
  • Deltour L, Leduque P, Paldi A, Ripoche MA, Dubois P, Jami J. 1991. Polyclonal origin of pancreatic islets in aggregation mouse chimera. Development 112: 11151121.
  • Dohrmann C, Gruss P, Lemaire L. 2000. Pax genes and the differentiation of hormone-producing endocrine cells in the pancreas. Mech Dev 92: 4754.
  • Esni F, Taljedal I, Perl A, Cremer H, Christofori G, Semb H. 1999. Neural cell adhesion molecule (N-CAM) is required for cell type segregation and normal ultrastructure in pancreatic islets. J Cell Biol 144: 325337.
  • Hard WL. 1944. The origin and differentiation of the alpha and beta cells in the pancreatic islets of the rat. Am J Anat 75: 369403.
  • Herrera PL. 2002. Defining the cell lineages of the islets of Langerhans using transgenic mice. Int J Dev Biol 46: 97103.
  • Herrera PL, Huarte J, Sanvito F, Meda P, Orci L, Vassalli JD. 1991. Embryogenesis of the murine endocrine pancreas; early expression of pancreatic polypeptide gene. Development 113: 12571265.
  • Hayashi K, Tamaki H, Handa K, Takahashi T, Kakita A, Yamashina S. 2003. Differentiation and proliferation of endocrine cells in the regenerating rat pancreas after 90% pancreatectomy. Arch Histol Cytol 66: 163174.
  • Jackerott M, Oster A, Larsson LI. 1996. PYY in developing murine islet cells: comparisons to development of islet hormones, NPY, and BrdU incorporation. J Histochem Cytochem 44: 809817.
  • Jensen J. 2004. Gene regulatory factors in pancreatic development. Dev Dyn 229: 176200.
  • Kim SK, Hebrok M. 2001. Intercellular signals regulating pancreas development and function. Genes Dev 15: 111127.
  • Means AL, Meszoely IM, Suzuki K, Miyamoto Y, Rustgi AK, Coffey RJ, Wright CVE, Stoffers DA, Leach SD. 2005. Pancreatic epithelial plasticity mediated by acinar cell transdifferentiation and generation of nestin-positive intermediates. Development 132: 37673776.
  • Miralles F, Battelino T, Czernichow P, Scharfmann R. 1998. TGF-β plays a key role in morphogenesis of the pancreatic Islets of Langerhans by controlling the activity of the matrix metalloproteinase MMP-2. J Cell Biol 143: 827836.
  • Mulder H, Myrsen-Axcrona U, Gebre-Medhin S, Ekblad E, Sundler F. 1998. Expression of non-classical islet hormone-like peptides during the embryonic development of the pancreas. Microsc Res Tech 43: 313321.
  • Pictet R, Rutter WJ. 1972. Development of the embryonic endocrine pancreas. In: RoyOG, EdwinBA, DonaldFS, NorbertF, StephenRG, editors. Handbook of physiology. Washington: American Physiological Society. p 2566.
  • Sakai Y, Hosaka M, Yoshinaga A, Hira Y, Harumi T, Watanabe T. 2004. Immunocytochemical localization of secretogranin III in the endocrine pancreas of male rats. Arch Histol Cytol 67: 5764.
  • Teitlelman G, Alpert S, Polak JM, Martinez A, Hanahan D. 1993. Precursor cells of mouse endocrine pancreas coexpress insulin, glucagon and the neural proteins tyrosine hydroxylase and neuropeptide Y, but not pancreatic polypeptide. Development 118: 10311039.
  • Yamashina S, Tamaki H, Katsumata O. 1999. The serous demilune of rat sublingual gland is an artificial structure produced by conventional fixation. Arch Histol Cytol 62: 347354.