| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Cell Biology and Genetics, College of Life Sciences, Peking University, Beijing, China;
b Institute of Biological Science and Technology, College of Science, Beijing Jiaotong University, Beijing, China
Key Words. Embryonic stem cells • Pancreatic ß cells • Differentiation • All-trans retinoic acid • Activin A
Correspondence: Hongkui Deng, Ph.D., Department of Cell Biology and Genetics, College of Life Sciences, Peking University, Beijing, 100871, P. R. China. Telephone: 86-10-6275-6954; Fax: 86-10-6275-6954; e-mail: hongkui_deng{at}pku.edu.cn; and Mingxiao Ding, M.S., Department of Cell Biology and Genetics, College of Life Sciences, Peking University, Beijing, 100871, P. R. China. Telephone: 86-10-6275-6954; Fax: 86-10-6275-6954; e-mail: dingmx01{at}pku.edu.cn
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
One attractive approach is the generation of functional ß cells from embryonic stem cells (ESCs). ESCs have been shown to be able to differentiate into pancreatic islet–like clusters, especially pancreatic ß cells [2– 4]. First, Soria et al. [2] successfully induced ESCs to differentiate into pancreatic ß cells by a cell-trapping system. However, this is a complicated process with genetic manipulation. Then Lumelsky et al. [3] designed a five-stage protocol to induce ESCs to differentiate into insulin-producing islet-like structures without genetic modification. Hori et al. [5] and Blyszczuk et al. [6] improved Lumelsky’s five-stage induction protocol by adding growth inhibitor LY294002 or overexpression of the pax4 gene. However, all these induction approaches are somewhat complicated and take a long period of time. Later Hansson et al. [7] used the five-stage induction protocol and found that ESC-derived insulin-positive cells absorbed insulin from the culture medium instead of producing it themselves. Therefore, it is necessary to find the specific induction factors that can induce ESCs to differentiate into pancreatic ß cells more simply and rapidly.
It has been suggested that some factors are able to promote definitive endoderm differentiation. Activin A, a member of the transforming growth factor–beta (TGF-ß) superfamily, is critical for mesoderm and endoderm formation during gastrulation. When used at a high concentration, it primarily induces endoderm formation [8, 9]. All-trans retinoic acid (RA) is a well-characterized signaling molecule that acts in anteroposterior patterning of neuroectoderm and mesoderm in vertebrates [10]. Recent evidence indicates that RA is also involved in the regulation of the embryonic endoderm differentiation pattern, especially the early pancreas bud formation, and it is able to improve insulin expression in pancreatic ß cells and the INS-1 cell line [11, 12]. It has been demonstrated that the combination of activin A and RA was able to induce the Xenopus presumptive ectoderm region of the blastula to differentiate into pancreatic insulin-positive cells [13].
We report here a novel three-step approach based on combination of activin A, RA, and other factors that mature pancreatic ß cells. This three-step protocol can induce ESCs to differentiate into insulin-producing cells within only 2 weeks. These insulin-positive cells express characteristic pancreatic ß-cell marker genes such as insulinI, pdx1, glut2, hnf3ß, and is11. Moreover, we use insulin promoter–enhanced green fluorescent protein (EGFP)–marked ESCs to further demonstrate that this strategy can indeed induce ESCs to differentiate into insulin-producing cells instead of uptaking insulin. Finally, we provide evidence that insulin-producing cells are able to fully rescue the streptozocin (STZ)–induced diabetic mice when they are transplanted under their renal capsules.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Step 1: To induce embryonic body (EB) formation, ESCs were dissociated with trypsin and suspended into Petri dishes (Alpha Medical Instrument Corp., Qingdao, China) with 10% FBS/DMEM without LIF. After 24–48 hours, EBs were collected and replated into 10% FBS/DMEM without LIF in 1% Matrigel-coated dishes (BD Biosciences, Bedford, MA, http://www.bdbioscience.com). Two hours later, EBs began to spread onto the dishes and were cultured in 10% FBS/DMEM with 100 ng/ml activin A (Sigma Chemical Corp., St. Louis, http://www.sigma-aldrich.com) for 24 hours. Then EBs were switched to 10% FBS/DMEM for 6–8 hours as an interval. After this interval, the differentiated EBs were cultured in 10% FBS/DMEM with 10–6 M RA (Sigma) for another 24 hours.
Step 2: To expand insulin-producing precursors, the differentiated EB cells were cultured in 10% FBS/DMEM with 10 ng/ml basic fibroblast growth factor (bFGF; Sigma) for 3–5 days.
Step 3: To mature the insulin-producing cells, the expanded cells in step 2 were switched to DMEM/F12 with N2 supplement, B27 supplement (all from Gibco-BRL), 1 µg/ml laminin (Sigma), 10 ng/ml bFGF, and 10 mM nicotinamide (Sigma) and cultured for 3–5 days. The morphology of cells was observed under a Nikon phase-contrast microscope (TS-100).
Reverse Transcription Polymerase Chain Reaction (RT-PCR) Analysis of Gene Expression in Differentiated Cells
Total RNA was extracted using the Catrimox-14 kit (Takara Bio, Otsu, Shiga, Japan, http://www.takara-bio.co.jp/english/index.htm) from the cells in step 2 or step 3, which were induced with and without both activin A and RA or with only one of them. Then RNA was reverse-transcribed into cDNA by Moloney murine leukemia virus reverse transcriptase (Promega Corp., Madison, WI, http://promega.com). PCR was performed with Ex Taq polymerase (Takara Bio) in PCR buffer. Cycle conditions were as follows: 94°C for 5 minutes; followed by 35 cycles of 94°C denaturation for 50 seconds, 56°C to 58°C annealing for 30 seconds, and 72°C elongation for 40 seconds; with a final incubation at 72°C for 4 minutes. PCR primers of pdx1, glut2, is11, and ß-actin were designed with Prime Premier 5.0. InsulinI and hnf3ß primers were as cited from [3]. The sequences for the PCR primers (sense and antisense, respectively) are:
insulinI: TAGTGACCAGCTATAATCAGAG and ACGC-CAAGGTCTGAAGGTCC (288 bp)
hnf3ß: ACCTGAGTCCGAGTCTGACC and GGCACCTT-GAGAAAGCAGTC (345 bp)
pdx1: CTTAGCGTGTCGCCACAGCCCTCCA and TCCAA-CAGCCGCCTTTCGTTATTCT (472 bp)
glut2: GGATAAATTCGCCTGGATGA and TTCCTTTG-GTTTCTGGAACT (299 bp)
is11: ATTTGCCACCTAGCCACAGCACC and CGCATTT-GATCCCGTACAACCTG (335 bp)
ß-actin: CCTGAACCCTAAGGCCAACCGTGAA and ATACCCAAGAAGGAAGGCTGGAAAA (480 bp)
hnf4
: ACACGTCCCCATCTGAAG and CTTCCTTCTTCAT-GCCAG (269 bp)
EGFP Reporter Construction and ESC Transfection
A human-insulin promoter-EGFP vector was constructed. A 0.4-kb XbaI and HindIII fragment containing human-insulin promoter from pFOXCAT-362 hIns (kindly provided by Dr. M.S. German, Hormone Research Institute, University of California, San Francisco) was ligated to the pCMV-EGFPN3 (BD Biosciences) whose cytomegalovirus promoter was removed by AseI-HindIII digestion. This vector with neomycin resistance was transfected into the ESC R1 line by electroporation, and positive ESC clones were selected with 250 µg/ml G418 (Gibco-BRL). Selected cells were induced by the three-step method described above.
Immunohistochemistry Assay
Induced cells in step 3 were fixed in 4% paraformaldehyde and washed three times by phosphate-buffered solution (PBS), then incubated with 10% normal goat serum for 20 minutes at room temperature. Goat serum was removed, and the cells were incubated with primary antibody to insulin (rabbit polyclonal immunoglobulin G [IgG], 1:200; Santa Cruz Biotechnology, Santa Cruz, CA, http://www.scbt.com) or to C-peptide (Linco Research, St.Charles, MO, http://www.lincoresearch.com) overnight at 4 °C, and further incubated with secondary antibody rhodamine-conjugated goat anti-rabbit IgG, fluorescein isothiocyanate–conjugated rabbit anti-goat IgG (1:200; Santa Cruz), or biotin-conjugated IgG (work solution; Santa Cruz). DAB (3,3'-diaminobenzidine) was used as the reaction substrate when biotin-conjugated IgG was the secondary antibody. Images were captured with an Olympus phase contrast microscope (Olympus IX-71).
DTZ (dithizone) stock solution (Merck & Co., Whitehouse Station, NY, http://www.merck.com) was prepared with 50 mg of DTZ in 5 ml of dimethyl sulfoxide. To prepare a DTZ working solution, DTZ stock solution was diluted at 1:100 with PBS. Induced cells were washed with PBS three times and stained with DTZ working solution at 37°C for 30 minutes.
Insulin Release Test by Enzyme-Linked Immunosorbent Assay (ELISA)
To further test whether the insulin release of induced cells was glucose-dependent, two glucose concentrations (5.5 mM and 27.7 mM) were used. The cells were preincubated with Krebs-Ringer buffer containing 2.5 mM glucose at 37°C for 90 minutes. To induce insulin release, the cells were incubated with Krebs-Ringer buffer containing 27.7 mM glucose for 15 minutes at 37°C. The control was incubated with 5.5 mM glucose for 15 minutes. Then the conditioned medium was collected. The two medium samples were tested for insulin release content using the Rat/Mouse Insulin ELISA Kit (Crystal Chem. Inc., Downers Grove, IL, http://www.crystalchem.com). The total cell protein content was tested using the BCA [Bicinchoninic Acid] Protein Assay Kit (Pierce Biotechnology, Rockford, IL, http://www.piercenet.com).
Transplantation of Insulin-Producing Cells Under Renal Subcapsule
The Institutional Animal Care and Use Committee of Peking University approved all animal procedures. To induce experimental diabetes before cell or PBS transplantation, STZ was injected intraperitoneally for 5 days at 50 mg/kg into 129 male mice, 6–8 weeks old. When the blood glucose of STZ-treated mice was above 13.9 mM, 1 x 106 insulin-producing cells were transplanted into the left renal capsule. PBS treatment was the control. Blood glucose was measured by GlucoTREND2 (Roche, Indianapolis, http://www.roche-applied-science.com) from snipped tails. Cryostat sections of the operated kidneys were prepared, and insulin expression of transplanted cells in the renal capsule was tested by immunohistochemistry.
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). First, ESCs (Fig. 1B) were suspended for EB formation. Then EBs were placed into 1% Matrigel-coated dishes and induced by activin A and RA sequentially. Second, the cells were switched to a medium containing 10% FBS/DMEM and bFGF, then cultured for another 3 days, which improved the proliferation of pancreatic progenitor cells derived from EBs. During this stage, most of the remaining cells were epithelial-like (Fig. 1C). Some new clusters appeared from large, spreading EBs. Third, the cells were cultured in DMEM/F12 medium with N2 supplement, B27 supplement, laminin, bFGF, and nicotinamide, which appeared to promote pancreatic ß cells to maturation [3]. At this stage, many small cell clusters appeared (Fig. 1D, E).
|
). In the second step, the differentiated epithelial-like cells expressed pancreatic progenitor markers such as hnf3ß, hnf4, and pdx1 (Fig. 2B). If induced by the three-step protocol with both activin A and RA, the differentiated cells strongly expressed pdx1, hnf3ß, insulinI, glut2, and is11 (Fig. 2C, lane 3). These cells did not express markers of other islet endocrine cells (non-ß cell) such as glucagon or somatostatin (data not shown). This result suggests that TGF-ß and RA signalings may be specific for pancreatic ß-cell development and maturation. We found this induction was strictly dependent on activin A and RA, as absence of both of them resulted in almost no small cluster formation and the induced cells did not express or only weakly expressed pdx1, hnf3ß, insulinI, glut2, and is11 (Fig. 2C, lane 2). Moreover, we found that if the dishes were coated with laminin or gelatin, instead of Matrigel, no cell clusters formed in the third step.
|
). If cultured without activin A and RA, the cells did not express insulin (Fig. 3C, D). In addition, we also used DTZ to detect the induced cells [15] and found that some clusters in the third step appeared crimson red after DTZ staining (data not shown). This proved that, within those clusters, there were cells that had the same characteristics as pancreatic ß cells.
|
), and their expression pattern matched that of insulin detected by immunohistochemistry (Fig. 3F, H). These data indicated that activin A and RA, combining with other mature factors, could indeed induce ESCs to differentiate into insulin-producing cells.
To analyze whether insulin secretion from those differentiated ESCs could be regulated by glucose, we treated about 106 induced cells with Krebs-Ringer buffer containing either low (5.5 mM) or high (27.7 mM) concentration of glucose and then analyzed the insulin release level in the culture medium by ELISA. We found that the insulin release in the high-glucose medium was nearly six times higher than that in the low-glucose medium (Fig. 3I). This result suggested that the ESC-derived insulin-positive cells secreted insulin in a glucose-dependent manner, as do normal pancreatic ß cells.
Transplantation of ESC-Derived Insulin-Positive Cells
To investigate whether the ESC-derived insulin-producing cells could rescue mice with diabetes, we transplanted the induced cells into the left renal capsules of STZ-treated diabetic mice (n = 9). After transplantation, the survival probability of cell-transplanted mice was about three times higher than that of PBS sham-operated control mice (n = 12) (Fig. 4A). The survival mice continued to gain weight. Two weeks later, the blood glucose of cell-treated mice reduced to a normal level (less than 13.9 mM). In contrast, the blood glucose of the PBS sham-operated control mice was still above 16 mM (Fig. 4B). In addition, after removal of the left kidney transplanted with ESC-derived cells, the blood glucose level of STZ-treated mice reversed back to over 13.9 mM within 3 days, and these mice could survive for nearly 1 week. The recrudescent diabetes and weak body may be the main reasons for their death. Cryostat sections of the operated kidneys were prepared, and insulin-positive cells could be visualized only in the differentiated cell–treated kidneys by immunohistochemistry (Fig. 5A, B). These results showed that the ESC-derived insulin-producing cells, when transplanted into the renal capsule, could improve glucose control and functionally rescue the STZ-treated mice with diabetes.
|
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Activin A and RA have important roles in our experimental induction of pancreatic ß cells. Activin A is important for early definitive endoderm development, and it is a disulfide-stabilized protein that belongs to the TGF-ß superfamily. Its binding to the cell surface receptor can induce the expression of many genes, including mix11 and goosecoid, which is important for early endoderm development [8, 9]. Kubo et al. [16] reported that activin A could induce ESCs to differentiate into definitive endoderm cells. In addition, activin A can improve insulin secretion in cultured human pancreatic islets [17] and regulate differentiation of pancreatic ß cells during development and regeneration of ß cells in diabetic neonatal rats [18]. A combination of these functions of activin A may explain how the activin A works in our induction system.
We then used RA to further promote pancreatic lineage differentiation. It has recently been demonstrated that RA is an important signaling molecule in the development of the early embryonic pancreas besides functions on induction of ectoderm and mesoderm development [10]. During zebrafish development, increased RA signaling can induce remarkable anterior expansion of the pancreas and liver endoderm. Conversely, inhibition of RA signaling by BMS493 inhibits early pancreas differentiation from embryonic endoderm [11]. Micallef et al. [19] reported that RA could induce pdx1+ endoderm formation when added at the fourth day of mouse ESC differentiation. Therefore, it is clear that RA can facilitate the development of pancreatic precursor cells. The unique feature of our induction strategy is to use the combination of activin A and RA to activate specifically pancreatic ß differentiation from ESCs. We discovered that after activin A and RA induction, differentiated ESCs expressed pancreatic progenitor markers such as pdx1, hnf3ß, and hnf4. We also found that induction with both activin A and RA was more efficient than with activin A or RA alone.
Matrigel also has an important role in our induction strategy. It has been demonstrated that Matrigel, a mixture of extracellular matrix, including laminin and growth factors such as fibroblast growth factor and TGF-ß, is essential for pancreatic progenitor cell migration, the three-dimensional cystic structures formation and protrusion of islet bud [20]. If human pancreatic ductal epithelial cells were cultured on the Matrigel, they could form insulin-positive islet-like clusters [21]. It has also been reported that laminin, one of the major components of Matrigel, was involved in mouse embryonic pancreatic duct lineage selection [22, 23]. In our study, we found that Matrigel was necessary for small islet-like cluster formation during induction. If ESCs were cultured on the laminin or gelatin matrix alone, cell clusters could not form well. And the differentiated cells seemed to grow better on Matrigel than on laminin or gelatin. Because laminin is a component of Matrigel, our data suggested that other matrix and growth factors in Matrigel may be also useful for pancreatic endocrine cell maturation.
Transplantation of ESC-derived insulin-producing cells was sufficient to normalize the blood glucose levels of diabetic mice in our experiment. However, we observed tumor formation in the kidney of some mice transplanted with induced cells. In fact, until now, most reported that insulin-producing cells derived from ESCs developed tumors when transplanted into recipient mice [5, 6]. Because of tumorigenesis, our strategy may not be sufficient to induce all treated ESCs into terminal-differentiated pancreatic ß cells. In our future work, we will test many other maturation factors, such as GLP-1 [24], to further improve pancreatic ß-cell terminal differentiation. Alternatively, more effort should be made to overcome ESC tumorigenesis problems. For example, it may be useful to block PI3K signaling [5] or to reduce Eras gene expression [25] to prevent ESC tumorigenesis.
In summary, we have demonstrated that a novel three-step protocol with a combination of activin A, RA, and other mature factors can induce ESCs to differentiate into functional insulin-producing cells in a short period. The results indicate that TGF-ß and RA signals are critical for pancreatic ß–cell development and maturation. The induction system described here offers a new in vitro induction model for studying the mechanism of pancreatic ß–cell formation and differentiation, and it provides a potential source of insulin-producing cells for transplantation therapy of type I diabetes mellitus.
|
ACKNOWLEDGMENTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We thank Dr. M.S. German for kindly providing the insulin promoter vector. We also thank Dr. Tung-Tien Sun for critical reading of the manuscript. In addition, we acknowledge other colleagues in our lab for advice during experiments. Y.S., L.H., and F.T. contributed equally to this article.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. Bertolino, R. Holmberg, E. Reissmann, O. Andersson, P.-O. Berggren, and C. F. Ibanez Activin B receptor ALK7 is a negative regulator of pancreatic {beta}-cell function PNAS, May 20, 2008; 105(20): 7246 - 7251. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. C. Robson, D. J. Phillips, T. McAlpine, A. Shin, S. Svobodova, T. Toy, V. Pillay, N. Kirkpatrick, D. Zanker, K. Wilson, et al. Activin-A: a novel dendritic cell-derived cytokine that potently attenuates CD40 ligand-specific cytokine and chemokine production Blood, March 1, 2008; 111(5): 2733 - 2743. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Qin, T. Yu, T. Qing, Y. Liu, Y. Zhao, J. Cai, J. Li, Z. Song, X. Qu, P. Zhou, et al. Regulation of Apoptosis and Differentiation by p53 in Human Embryonic Stem Cells J. Biol. Chem., February 23, 2007; 282(8): 5842 - 5852. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-G. Chen, Q. Wang, S.-L. Lin, C. D. Chang, J. Chung, and S.-Y. Ying Activin Signaling and Its Role in Regulation of Cell Proliferation, Apoptosis, and Carcinogenesis. Experimental Biology and Medicine, May 1, 2006; 231(5): 534 - 544. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. J. Rippon, J. M. Polak, M. Qin, and A. E. Bishop Derivation of Distal Lung Epithelial Progenitors from Murine Embryonic Stem Cells Using a Novel Three-Step Differentiation Protocol Stem Cells, May 1, 2006; 24(5): 1389 - 1398. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Stafford, R. J. White, M. D. Kinkel, A. Linville, T. F. Schilling, and V. E. Prince Retinoids signal directly to zebrafish endoderm to specify insulin-expressing {beta}-cells Development, March 1, 2006; 133(5): 949 - 956. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. P. Raikwar, T. Mueller, and N. Zavazava Strategies for Developing Therapeutic Application of Human Embryonic Stem Cells Physiology, February 1, 2006; 21(1): 19 - 28. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| STEM CELLS | THE ONCOLOGIST | CME | ALPHAMED PRESS JOURNALS |
