| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Laboratory of Molecular Endocrinology, Massachusetts General Hospital, Harvard Medical School, Howard Hughes Medical Institute, Boston, Massachusetts 02114
Address all correspondence and requests for reprints to: Joel F. Habener, M.D., Laboratory of Molecular Endocrinology, Massachusetts General Hospital, 55 Fruit Street, Their 320, Boston, Massachusetts 02114. E-mail: jhabener{at}partners.org.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In many instances, the dedifferentiation of cells is accompanied by drastic changes in phenotype in which their morphology changes from that of epithelial cells containing extensive cell-cell junctions and cytokeratin filament networks, to cells with a fibroblast or mesenchymal appearance, fewer cell-cell attachments, and a vimentin filament network (5, 6, 7). This process of dedifferentiation is known as an epithelial to mesenchymal transition (EMT) and is believed to be mediated in large part by the induction of the zinc finger transcription factor Snail (8, 9). Snail functions as a powerful transcriptional repressor of both epithelial-specific and lineage-specific genes. Upon binding to the promoters of these genes, Snail down-regulates gene expression and promotes a change in cell identity from a differentiated to an undifferentiated cell, or in some cases from an epithelial cell to a mesenchymal cell.
Recently, interest in EMT has increased in the area of stem cell biology and tissue regeneration. For example, Slug, a Snail family member, has been implicated in the re-epithelialization of cutaneous wounds, as well as in the regeneration of damaged skeletal muscle after damage (10, 11). Most notably in the area of pancreatic stem cells, Gershengorn and co-workers (12) demonstrated that cultured primary ß-cells are capable of undergoing EMT and dedifferentiating into a fibroblastoid-like cell type. This process can be reversed, and these fibroblastoid cells can then be redifferentiated into pancreatic endocrine cells by a so-called mesenchymal to epithelial transition.
These circumstances prompted us to examine several immortalized pancreatic endocrine cell lines to determine whether EMT is functioning in their growth and differentiation. In this report, we examine the MIN6, INS1, and RIN1027-B2 pancreatic endocrine cell lines and find that Snail is highly and dynamically expressed within a subfraction of non-hormone-expressing cells where it represses proinsulin and proglucagon gene expression. As the density of the culture increases and the cells become more confluent and contacted, levels of Snail decrease and levels of hormone mRNAs rise. Transient transfection assays indicate that Snail functions directly on hormone gene expression and exerts its repressive function within the proximal 410 bp of the insulin promoter. Reduction of Snail levels by treatment of cells with small interfering RNA (siRNA) increases gene expression from this promoter. We propose that Snail functions as a modulator of differentiation in cultured cell lines by temporarily repressing the expression of differentiation-specific genes.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Northern hybridizations
PCR-amplified cDNA fragments were internally labeled with 
-32P-deoxy-CTP using Klenow and random hexamer primers and purified over a G50 Sepharose column. Five micrograms of total RNA were resolved on a 1.2% formaldehyde agarose gel and transferred to Hybond-N+ (Amersham Biotech, Pittsburgh, PA) by capilliary action. Membranes were prehybridized for 1 h in hybridization buffer [50% formamide, 5x saline sodium phosphate EDTA (SSPE), 2x Denhardt’s, 0.1% sodium dodecyl sulfate (SDS)] at 42 C. Hybridization was performed overnight with 1 x 106 cpm probe per ml of hybridization buffer at 42 C. The next day, the blots were sequentially washed with 2x saline sodium citrate (SSC)/0.1% SDS, 1x SSC/0.1% SDS, and 0.1x SSC/0.1% SDS (all washes at room temperature).
Cloning
The plasmid pcDNA3-myc was generated by annealing the oligonucleotides (AGCTTGCCGCCATGGAGCAGAAACTCATCTCTGAAGAGGATCTGGGTAC; CCAGATCCTCTTCAGAGATGAGTTTCTGCTCCATGGCGGCA) and cloning them into the HindIII/KpnI sites of pcDNA3 (Invitrogen). The coding sequence for mouse Snail was then amplified using the following primers (TAGGATCCATGCCGCGCTCCTTCCTG; TAGAGCTCTCAGCGAGGGCCTCCGGA) and cloned into the BamHI/EcoRV sites of pcDNA3-myc, generating a fusion of the myc-epitope onto the N terminus of the mouse Snail protein. The Snail S8A clone (pCMV-Tag2B-Snail S8A) was a generous gift from Dr. Mien-Chie Hung (13).
Cell culture and transfections
INS1, MIN6, PANC1, and RIN1027-B2 cells were grown under standard conditions. For the generation of Myc-Snail stable cell lines, plasmids were linearized by digestion with PvuI, and transfected using Lipofectamine 2000 according to the manufacturer’s recommendations (Invitrogen). Stable cells were selected by growth in media containing 400 µg/ml Geneticin. Individual clones were isolated using cloning cylinders and assayed for Myc expression by immunocytochemistry and Western immunoblot.
Reporter gene assays were performed by transfecting 2 x 105 cells per well of a 24-well plate with 0.5 µg DNA using Lipofectamine 2000 according to the manufacturer’s recommendations (Invitrogen). Luciferase activity was read using Luciferase Assay Reagent (Promega, Madison, WI) in a Wallac Victor2 Multilabel Counter (PerkinElmer, Wellesley, MA) and values normalized to total protein content. The rat Insulin I –410 promoter-luciferase vector was a kind gift of Dr. M. German and the rat Proglucagon –361 promoter-luciferase vector was described previously (14).
Protein isolation and Western blots
Total cellular extracts were isolated in RIPA buffer [50 mM Tris (pH 7.4), 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS] supplemented with Complete protease inhibitor cocktail (Roche, Indianapolis, IN). Extracts were quantitated using bicinchoninic acid protein assay reagent (Pierce, Rockford, IL). For Western immunoblots, 20 µg of total protein were resolved on 4–12% NuPage Novex Bis-Tris gels (Invitrogen) and transferred to polyvinylidene fluoride membranes. Primary antibodies were detected using the Roche BM Chemiluminescent Western blotting kit.
Immunocytochemistry
Cells were grown on glass coverslips and processed for immunocytochemistry by standard methods. Cells were fixed with 4% paraformaldehyde in PBS, permeablized for 5 min with 0.1% Triton X-100 in PBS, blocked for 1 h in 10% serum, and incubated for 1 h at room temperature with the primary antibody. Primary antibodies used were guinea pig antiinsulin (Linco, St. Charles, MO), guinea pig antiglucagon (Linco), mouse anti-myc (Clone 9E10; Santa Cruz Biotechnology, Inc., Santa Cruz, CA), goat anti-Ki67 (Santa Cruz), mouse anti-Cytokeratin20 (Dako, Carpinteria, CA) and rabbit antiactin (Santa Cruz). Polyclonal antibodies recognizing Snail were generated against synthetic peptides corresponding to the C terminus of mouse Snail (amino acids 249–264) (Covance Research Products, Denver, PA). Cy2 and Cy3 secondary antibodies were purchased from Jackson ImmunoResearch (West Grove, PA). Images were pseudo-colored for clarity and consistency among figures.
siRNA
Stealth siRNA oligos were purchased from Invitrogen with the following RNA sequences: Scramble (AGACCCACUCGGAUGUGAAGAGAUA); Snail 683 (AGAACUCGCAGGUGUAAAGGCCAUA); Snail 697 (GUGAAGAGAUACCAGUGUCAGGCCU). Cells were transfected using Lipofectamine 2000 according to the manufacturer’s recommendations (Invitrogen).
Real-time PCR
Real-time PCR was performed on a Cepheid Smart-Cycler Thermocycler. PCRs were carried out using the ONE-STEP RT-PCR kit from Invitrogen using 100 ng RNA per 30-µl reaction. Products were amplified using the following probes: glyceraldehyde-3-phosphate dehydrogenase (GAPDH) forward (TGGTCTACATGTTCCAGTATGACT), GAPDH reverse (CCATTTGATGTTAGCGG GATCTC), GAPDH probe (CCACGGCAAGTTCAACGGCACAGT), Ins1 forward (AACAGCACCTTTGTGG TCCTC), Ins1 reverse (CTCCACCCAGCTCCAGTTGT), Ins1 Probe (TGCGGGTCCTCCACTTCACGACGG), Snail forward (CACCAAGAGTCTGGCTGCTC), Snail reverse (GGACCAAGGCTGGAAGGAGT), Snail probe (CCCTCGCTGACCCTGCTACCTCCC). Insulin and Snail mRNA levels were normalized by standard methods using GAPDH as a control.
Partial pancreatectomy
Partial (70%) pancreatectomy was performed on 6-month-old C57BL/6 mice by previously described methods (15). All experimentation was conducted in accord with accepted standards of humane animal care.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). These cell lines also expressed varying amounts of insulin, depending on the characteristics of the cell type. The MIN6, INS1, and ß-TC6 cells are insulinoma cells lines, whereas the RIN1027-B2 cell line is a multihormonal line. The majority of the cells in most, if not all of these clonal cell lines are undifferentiated, and lack hormone expression. This cellular heterogeneity of hormone expression appears to be a general property of clonal cell lines derived from pancreatic endocrine cells (16, 17). The RIN1027-B2 cells can undergo spontaneous differentiation and express any of the pancreatic endocrine hormones (17). During the course of our studies, the RIN1027-B2 cells primarily expressed glucagon and, to a lesser extent, insulin.
|
).
Nuclear Snail was also detected by immunocytochemistry in each of the cell lines; however, not every cell within the culture expressed Snail (Fig. 1C). The percentage of cells expressing Snail varied depending on the cell line examined as well as the density of each culture. Cells that were noncontacted (less dense) or were present at the edges of cell clusters expressed high amounts of nuclear Snail, whereas those cells present in the center of the clusters contained little or no detectable Snail.
Snail expression inversely correlated to differentiation
Because cell aggregation is often a prerequisite for cell differentiation, we postulated that Snail expression might be related to hormone expression. Indeed, when cells were costained for Snail and antibody combinations for insulin and/or glucagon, the cells expressing the highest amounts of Snail did not contain any hormone, whereas cells expressing little to no nuclear Snail protein contained readily detectable amounts of hormone (Fig. 2A). This finding was consistent among all of the cell lines we examined, although in the case of MIN6 and INS1 cells, this relationship was most clearly observed when the cells were cultured for extended periods at low density to maintain elevated Snail expression levels. This wash-out period is necessary to allow for the secretion of the stored insulin granules as well as for the degradation of the previously transcribed insulin mRNA present in the cytoplasm, allowing for a more accurate estimation of the cellular differentiation state.
|
). However, when these cells were allowed to cluster and differentiate, levels of Snail protein fell dramatically, suggesting a relationship between Snail expression and cell differentiation. We quantitated these results by counting fields of cells and scoring for expression of Snail or hormone to get a more accurate view of the dynamics of individual cell fate decisions within the culture. Notably, at low densities, Snail was expressed in virtually every cell; however, there still remained a small percentage of cells containing hormone. We suggest that these are cells in a transition between a hormone-positive and hormone-negative state. However, due to the long half-life of the proinsulin mRNA [T1/2 = approximately 30 h (19)], and the amount of time it takes for stored hormone to be secreted from the cell, it is difficult to estimate the percentage of cells undergoing this transition.
Snail-expressing cells proliferate but do not undergo EMT
A common feature of the endocrine pancreas-derived cell lines is that they appear to shift between a hormone-positive and hormone-negative state. We reasoned that cells that are less differentiated, and therefore contain fewer specialized structures (such as secretion granules, etc.) would be able to undergo cytokinesis more readily than cells with extensive cell-cell contacts and polarized subcellular systems. Because Snail expression has been implicated in promoting cell proliferation in some systems (20, 21), we hypothesized that Snail, proliferation, and dedifferentiation may be linked. We examined the expression of the proliferation marker Ki67 which is expressed within all non-G0 stages of the cell division cycle. Although Snail expression appears to be inversely correlated with hormone expression, it appears to coincide with cell proliferation (Fig. 2C). Immunostaining for Snail and Ki67 in RIN1027-B2 cells indicates that Snail is almost invariably expressed in actively cycling cells; as the cells cluster, both Snail and Ki67 become extinguished and the cells differentiate. Similar to the results seen with Snail and hormone expression, a small number of Ki67-positive cells also coexpress hormone. A similar correlation was seen between Snail and Ki67 in INS1 cells and MIN6 cells.
|
). Similarly, E-Cadherin levels remained constant (data not shown). One plausible explanation for our observations is that these cell lines may have been selected for because of their capacity to maintain an epithelial phenotype; unknown compensatory changes must have occurred to prevent Snail from inducing a mesenchymal phenotype.
Overexpression of Snail inhibits cell differentiation
To test the hypothesis that Snail is acting as an inhibitor of endocrine cell differentiation, we constitutively overexpressed Snail in RIN1027-B2 cells to maintain high levels of Snail during cell clustering and differentiation. Stable cell lines were generated with a myc-epitope-tagged version of Snail driven from a cytomegalovirus promoter and individual clones were isolated. High levels of Snail expression appeared to be deleterious to cell survival because the majority of clones either did not survive growing with superphysiological levels of Snail, or down-regulated expression of the transgene over time. Of the clones that were able to maintain Snail levels below a cytotoxic level, clone no. 1 was the most useful and maintained elevated levels of nuclear Snail in most of the cells within the culture during continuous passage, whereas clone nos. 11 and 16 expressed Snail in approximately 50% of the cells. Immunohistochemical staining of these cell lines revealed that overexpression of Snail is sufficient to prevent hormone expression in these clones, regardless of cell density (Fig. 3A). Virtually no hormone expression was seen in clone no. 1 by immunohistochemistry or RT-PCR, even after extended culture at high density (Fig. 3B). Hormone expression was detected in a subpopulation of cells in clone nos. 11 and 16; however, these differentiated cells were devoid of myc-Snail expression, suggesting that differentiation required the absence of Snail expression. Comparisons of growth curves of the parental line to the myc-Snail no. 1 line revealed no significant difference in cell cycle time or saturation density (data not shown), suggesting that Snail expression does not regulate cell growth directly.
|
); however, similar proportions of cells expressed the PDX-1 protein (Fig. 3C). This suggests that all aspects of endocrine cell identity were not altered by Snail, but instead Snail may function to inhibit differentiation into hormone-producing cells. Similarly, Snail overexpression also caused a decrease in the expression of a number of cell adhesion molecules known to be under direct regulation by Snail (Fig. 3B) (24, 25). Expression of desmoplakin was affected most strongly, with levels of occludin, E-Cadherin, and Zona Occludins 1 (ZO1) showing a moderate decrease.
We next determined whether the effects of Snail on cellular differentiation are reversible and whether undifferentiated cells could be redifferentiated by reducing the levels of Snail. The RIN1027-B2 myc-Snail no. 1 cell line was treated for 72 h with two different Snail siRNA oligonucleotides, which resulted in a significant reduction in Snail (Fig. 3D). No effect on Snail was observed in cells treated with a scrambled version of the Snail siRNA. This reduction in Snail protein corresponded to a greater than 2-fold increase in insulin mRNA expression in the myc-Snail no. 1 cell line (Fig. 3E), suggesting that Snail may function as a temporary repressor of hormone expression. This increase in hormone expression was also evident in immunohistochemical staining of the siRNA-treated samples in which a small number of insulin-expressing cells were detected (Fig. 3E). Although some recovery of differentiation was observed by treatment with Snail siRNA, the proportion of differentiation was not nearly that of the original cell line. We speculate that this is due to a combination of the limited percentage of cells that were transfected with the siRNA, the low level of differentiation normally seen in the parental cell line (
40%), and the relatively short time course of the experiment. Experiments are currently ongoing to knock-down Snail expression in a larger proportion of cells using a lentiviral-based RNA interference approach.
Snail represses insulin promoter-driven luciferase activity
To further investigate how Snail affects hormone expression and the state of cellular differentiation, RIN1027-B2 cells were transfected with luciferase reporter constructs driven by either the rat insulin I promoter or the rat proglucagon promoter (Fig. 4A). Transfection of a Snail expression plasmid with these reporter constructs resulted in an approximately 60% decrease in reporter gene activity, suggesting that Snail is able to directly affect hormone gene transcription.
|
). Cotransfection with a stabilized form of the Snail protein (Snail S8A) (13) that can no longer be efficiently exported from the nucleus and degraded is even more effective in repressing insulin reporter activity. However, a deletion construct of Snail that lacks the N-terminal repressor domain (SNAG domain) does not affect insulin reporter expression, suggesting that Snail represses insulin gene expression by means of an active recruitment of corepressors rather than by competitive displacement of activator proteins residing on the insulin promoter. It is worth noting that the closely related Snail family member Slug has no effect on the insulin promoter, despite evidence in other systems that these two proteins bind to similar DNA sites and perform partially overlapping functions. Studies have revealed that Snail and Slug have differing DNA binding affinities that may affect their relative abilities to repress transcription (26). We speculate that this may be a potential cause for the observed differences between Snail and Slug on insulin gene transcription.
Because Snail is normally expressed in the pancreatic endocrine cell lines, we postulated that reducing the levels of endogenous Snail would increase insulin gene expression. When INS1 cells were transfected with the combination of the insulin reporter gene construct and Snail siRNA oligonucleotides, insulin reporter gene activity was increased approximately 3-fold over basal levels (Fig. 4C), suggesting that Snail may be an endogenous modulator of insulin gene expression in the INS1 cell line.
Snail is expressed during pancreatic development and regeneration
Snail expression is detected in numerous pancreatic cell lines; however, it may be a result of the transformed nature of the cell lines. To address whether or not Snail is expressed during pancreas development, total RNA was isolated from newborn mice [postnatal d 1 (PN1)] and 2-month-old mice. RNA was also isolated from 6-month-old mice that had undergone either a partial pancreatectomy to induce pancreatic regeneration, or a sham control operation. Samples were isolated at 1 and 3 d after pancreatectomy and Snail mRNA assayed by real-time PCR (Fig. 5). The highest expression of Snail mRNA was detected in the PN1 pancreas, with levels declining as the organ matures into adulthood. Snail mRNA was reinduced during the regenerative response after partial pancreatectomy, increasing by 3 d after procedure to levels similar to that seen in the neonatal pancreas. Experiments are ongoing to identify the cells within the pancreas that express Snail and to further characterize the role of Snail during pancreatic regeneration.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We examined the role of Snail, an effector of EMT, during the growth and differentiation of cultured endocrine pancreatic cell lines. We found Snail expressed in subpopulations of cells in all of the lines examined, and its expression correlated negatively with the level of cellular differentiation. Cells with high levels of Snail expression are undifferentiated, but as the level of Snail decreases, the extent of differentiation into insulin or glucagon producing cells increases. Forced overexpression of Snail in these cells is sufficient to prevent this induction of hormone expression.
We tentatively conclude from our studies that Snail is capable of both preventing cellular differentiation as well as inducing cells to dedifferentiate, but not necessarily by converting epithelial cells into mesenchymal cells. It appears that Snail may play additional roles in modulating differentiation that are independent of the regulation of some of the broader epithelial-specific genes such as E-Cadherin. Minimal changes in E-Cadherin or Cytokeratin 20 mRNA or protein levels were seen with the conditions tested in the RIN1027-B2 cells, even in the cell lines with ectopic overexpression of Snail. This circumstance suggests that Snail may function more subtly than previously appreciated, allowing cells to acquire motility and undifferentiated characteristics while still retaining some markers of an epithelial morphology.
Additionally, this Snail-mediated dedifferentiation is a reversible process. When RIN1027-B2 cells are maintained for extended periods in noncontacted conditions, all of the cells eventually lose their differentiated characteristics. However, these cells undergo differentiation again if allowed to aggregate and undergo a concurrent decrease in nuclear Snail levels. These differentiated cells can then be replated at a low density and induced to dedifferentiate again. It is of interest to note that, as these cells undergo repeated differentiation and dedifferentiation, they remain committed to a pancreatic endocrine phenotype. These observations suggest that although the cells are undergoing EMT, they retain a memory from their differentiated state, allowing them to resume hormone expression after reduction of Snail expression and relief of transcriptional repression.
The capability of the endocrine pancreatic cell lines to stably cycle between differentiated and undifferentiated states may be a result of both the transformed nature of these cells along with the fact that the cell lines were originally selected based on their ability to both proliferate and maintain differentiation. Nevertheless, this cell model demonstrates that the potential exists for Snail to play a role in the modulation of endocrine cell differentiation in vivo. Reports from numerous groups indicate that EMT plays a role in the development and regeneration of organ systems such as the kidney, hair follicle, epidermis, and skeletal muscle, and that the induction of EMT is central to the regenerative process (10, 11, 30, 31, 32). Preliminary evidence from our laboratory indicates that Snail is expressed during pancreatic development and is also induced during the in vivo regeneration of the pancreas, providing the possibility that Snail may be a target for modulation to promote the regeneration of lost or damaged tissue within the pancreas. Finally, it is worth mentioning the cellular heterogeneity previously reported (16, 17) and the capacity for these established cell lines to undergo dedifferentiation and differentiation is culture should be considered in future studies of these widely distributed cell lines.
|
Footnotes |
|---|
J.M.R., M.U., M.V.J. and J.F.H. have nothing to declare.
First Published Online March 23, 2006
Abbreviations: EMT, Epithelial to mesenchymal transition; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; PDX-1, pancreas duodenum homeobox-1; SDS, sodium dodecyl sulfate; siRNA, small interfering RNA.
Received November 2, 2005.
Accepted for publication March 10, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-cell-specific expression of the proglucagon gene through interaction with a novel proximal promoter G1 element. Mol Cell Biol 17:7186–7194[Abstract]This article has been cited by other articles:
 |
|
 |
|
  Z. Liu and J. F. Habener Glucagon-like Peptide-1 Activation of TCF7L2-dependent Wnt Signaling Enhances Pancreatic Beta Cell Proliferation J. Biol. Chem., March 28, 2008; 283(13): 8723 - 8735. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
