---

Original Paper	Pdf file on Synergy OPEN omments
Acta Biochim Biophys Sin 2008, 40: 830-839
doi:10.1111/j.1745-7270.2008.00459.x

Human amnion epithelial cells can be induced to differentiate into functional insulin�-producing cells

Yanan Hou^1,2, Qin Huang^1,2, Tianjin Liu^1,3, and Lihe Guo^1,3*

1 Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai 200031, China

2 Graduate School of the Chinese Academy of Sciences, Beijing 100049, China

3 Cellstar Biotechnologies Company, Shanghai 201210, China

Received: April 30, 2008������

Accepted: May 28, 2008

This work was supported by a grant from the Science and Technology Department of Shanghai Research Fund (No. 05DZ19329)

*Corresponding author: Tel/Fax, 86-21-54921392; E-mail, [email protected]

Pancreatic islet transplantation has demonstrated that long-term� insulin independence may be achieved in patients suffering� from diabetes mellitus type 1. However, limited availability� of islet tissue means that new sources of insulin-producing cells that are responsive to glucose are required. Here, we show that human amnion epithelial cells (HAEC) can be induced to differentiate into functional insulin-producing� cells in vitro. After induction of differentiation, HAEC expressed multiple pancreatic b-cell genes, including insulin, pancreas duodenum homeobox-1, paired box gene 6, NK2 transcription� factor-related locus 2, Islet 1, glucokinase, and glucose transporter-2, and released C-peptide in a glucose�-regulated manner in response to other extracellular� stimulations. The transplantation of induced HAEC into strepto�zotocin-induced diabetic C57 mice reversed hyper��glycemia, restored body weight, and maintained euglycemia for 30 d. These findings indicated that HAEC may be a new source for cell replacement therapy in type 1 diabetes.

Keywords������� amnion epithelial cell; diabetes; differentiation; transplantation

Type 1 diabetes is characterized by the autoimmune destruction� of pancreatic b-cells. The resulting lack of insulin production leads to hyperglycemia and serious long-term complications. Clinical trials have proven that allogeneic islet transplantation is an alternative treatment for patients with type 1 diabetes [1]. However, the shortage� of islet donors limits the large-scale use of this therapy. Recent studies have demonstrated that embryonic stem cells and adult stem/progenitor cells isolated from the pancreas� [2-6], liver [7,8], salivary gland [9], adipose [10], or nerve system are capable of differentiation into insulin-producing cells [11]. Stem cells hold great promise for supplying sufficient donor cells for transplantation.

Amniotic epithelial cells are generated from amnioblasts on the eighth day after fertilization and constitute the inner layer of the amnion [12]. Recent studies have shown that human amniotic epithelial cells (HAEC) are endowed with stem cell characteristics and have the potential to differentiate� into cells of all three germ layers, including endodermal pancreatic cells [13]. It has also been shown that after induction of differentiation, the transplantation of HAEC into streptozotocin (STZ)-induced diabetic mice can normalize blood glucose [14]. However, it is unclear whether HAEC could be induced to differentiate into functional� insulin-producing cells and whether, with regard� to its immune privilege property [15], HAEC could normalize� blood glucose levels after transplantation into diabetic C57 mice.

In the present study, we showed that, after induction of differentiation, human amniotic epithelial cells exhibit b-cell-like characteristics, including the expression of genes related to b-cell development and function as well as C-peptide production and release response to glucose and other extracellular stimulations. Transplantation of induced� HAEC into STZ-induced diabetic C57/B1 mice resulted in perfect control of blood glucose for 30 d and allowed body weight to be restored.

Materials and Methods

Isolation of HAEC

The study and use of human amnion were approved by the Patients and Ethics Committee of the Shanghai Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences (Shanghai, China). Human amnion membranes were mechanically� peeled from the chorion of a placenta obtained� from healthy mothers undergoing cesarean sections. The membrane was extensively scraped out to remove the underlying� tissues (i.e. the spongy and fibroblast layers) to obtain a pure epithelial layer with an intact basement membrane. The tissue was minced and incubated at 37 �C with 0.25% trypsin (Difco Laboratories, Detroit, USA) containing 0.53 mM EDTA for 30 min. Afterwards, the digested tissue was passed through a 200 mm filter and then a 74 mm filter to remove larger fibrous tissue remnants. The amniotic epithelial cells were collected by centrifugation� and suspended in medium. Standard culture� medium is Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (PAA, Pasching, Austria), 2 mM L-glutamine, 1% non-essential amino acid, 55 mM 2-mercaptoethanol, 1 mM sodium pyruvate, and 1% antibiotic-antimycotic (all from Gibco, Grand Island, USA). Incubation was carried out in a 5% CO2 atmosphere at 37 �C.

Induction of differentiation

To induce differentiation into insulin-producing cells, HAEC (1107) were seeded in a 10-cm culture dish and cultured in serum-free DMEM containing 25 mM D-glucose, N2 supplement (Gibco), 1% antibiotic-antimycotic, and 10 mM nicotinamide (Sigma, St. Louis, USA) for 1-3 weeks. Medium was replaced twice a week. Cell culture supernatants were collected at 0, 7, 14, and 21 d, centrifuged� briefly to remove cell debris, and then tested for C-peptide using human C-peptide radioimmunoassay� (RIA) kit (China Diagnostics Medical Corporation, Beijing, China).

Reverse transcription-polymerase chain reaction (RT-PCR)

Total RNA was prepared from HAEC at different times of induction (0, 7, 14 and 21 d) using TRIzol reagent (Invitrogen, Carlsbad, USA). Prior to RT, RNA samples were digested with DNase I (1 U/mg RNA; Fermentas, Glen Burnie, USA) for 30 min at 37 �C to eliminate genomic� DNA contamination. Total RNA (2 mg) underwent reverse transcription-polymerase chain reaction using� Moloney murine� leukemia virus reverse transcriptase (Promega, Madison, USA) in a 25 ml volume containing� 0.5 mg oligo dT, 400 mM deoxynucleotide triphosphate, and buffers� supplied by the manufacturer. As described in Table 1, 1 ml cDNA was subjected to PCR amplification using human primer pairs.

C-peptide release

After induction for 7 d, cells (2106 each well) were seeded in 6-well plates and grown in DMEM containing 25 mM glucose, 10 mM nicotinamide and 10% fetal bovine serum for 2 d. The cells were then washed twice in DMEM containing� 0.5 mM glucose and cultured in this medium for 1 h. The medium was then replaced by 2 ml DMEM containing 0.5, 5, or 25 mM glucose, respectively, for each well. As a control, the non-induced HAEC (at day 0) were washed twice in DMEM containing 0.5 mM glucose and cultured in DMEM containing 25 mM glucose. The culture supernatants were collected after 12 h of incubation, centrifuged briefly to remove cell debris, and then tested for C-peptide using human C-peptide RIA kit. For KCl and tolbutamide stimulation experiments, the cells were washed twice in DMEM containing 2 mM glucose and cultured in this medium for 1 h. The medium was then replaced by 2 ml DMEM containing 20 mM glucose and KCl (0, 10, 20, 30, and 40 mM, respectively, for each well) or tolbutamide (0, 50 and 100 mM, respectively, for each well; Sigma) for 1 h. The culture supernatants were collected after 12 h of incubation, centrifuged to remove debris, and then tested for C-peptide. The cells were harvested� for quantitation of the cell number.

Immunofluorescence

Following isolation from amniotic membrane, HAEC were seeded on glass cover slides in 6-well culture plates. After induction of differentiation, induced and non-induced HAEC were rinsed in phosphate-buffered saline (PBS) after� 14 d of induction, fixed with 4% polyformaldehyde in PBS at room temperature for 15 min, and permeabilized with 0.1% Triton X-100 in 0.1% sodium citrate at room temperature� for 15 min. Preparations were blocked by 5% goat serum in PBS at room temperature for 30 min, and then incubated overnight at 4 �C with primary guinea pig anti-insulin polyclonal antibody (1:200; Dako, Carpinteria, USA) and rabbit anti-Pdx-1 polyclonal antibody� (1:200; Chemicon, Temecula, USA). For insulin staining, the cells were incubated in Texas Red-labeled anti-guinea pig secondary� antibody (1:500; Santa Cruz Biotechnology� Inc., Santa Cruz, USA) at room temperature� for 1 h. For Pdx-1 staining, the cells were incubated in fluorescein-isothio�cyanate�-labeled goat-anti-rabbit secondary� antibody (1:500; Jackson ImmunoResearch Laboratories, West Grove, USA) at room temperature for 1 h. Nuclei were stained with 4,6-diamidino-2-phenylindole dihydro�chloride (DAPI). Goat serum was used as a negative control.

Kidneys were removed from HAEC-transplanted mice and sham-operated mice at 30 d after transplantation and then fixed with 4% polyformaldehyde in PBS, rinsed with PBS, and cryoprotected overnight in phosphate� buffer containing 30% sucrose. Tissues were embedded in optimal� cutting temperature compound (Tissue-Tek, Sakura-Finetek, Torrance, USA), frozen, sectioned at 6 mm and collected on gelatin pre-coated slides. Immunostaining was preformed. After immunofluorescence analysis, the sections� were stained with hematoxylin, dehydrated, and mounted for light microscopic examination. Three mice from each group were analyzed.

Transfection of HAEC with green fluorescence protein� (GFP)

Lentiviral vectors (pWPT-GFP, pMDlg/pRRE, pMD2.G and pRSV-REV) were kindly provided by Dr. Didier Trono (University of Geneva, Geneva, Switzerland) [16]. The pseudotyped viral particles were produced by four-plasmid� transient cotransfection into 293T cells with the calcium phosphate transfection system, harvested, and concentrated� by ultracentrifugation (72,000 g, 120 min, 4 �C). Concentrated supernatants were titrated with serial dilutions of vector stocks on 1�105 HeLa cells, followed by fluorescence-activated cytometric analysis (FACSCalibur flow cytometer; BD Biosciences, San Jose, USA). Primary cultured HAEC were transduced by 24 h exposure (three rounds of infection) to lentivirus-containing� supernatant at 30 multiplicity of infection in the presence of polybrene (8 mg/ml), followed by induction� of differentiation. After induction for 7 d, the cells were trypsinized for transplantation.

Animal and cell transplantation ���������������������

All animal experiments were performed in compliance with the standard institutional guidelines for animal care (Shanghai Institutes for Biological Sciences, Chinese Academy� of Sciences). Six- to eight-week-old male C57/B1 mice were made hyperglycemic by two intraperitoneal� injections of STZ (Sigma) of first 180 mg/kg, and then 60 mg/kg of body weight at 3 d interval. Animals with blood glucose level16.8 mM for 1 week were either transplanted� with induced HAEC (2-3�106 each) or injected� with saline (sham operation) into the left subrenal capsule. Blood glucose� levels were monitored twice a week in samples obtained from the tail vein of mice using the GlucoTrend glucose detector (Roche Diagnostics, Indianapolis, USA). The body weight of mice was also monitored twice a week. For intraperitoneal glucose tolerance test (IPGTT), at d 30 after transplantation, normal� non-diabetic mice (n=5), diabetic mice with normalized� glucose levels following the HAEC trans�plantation (n=5), and diabetic mice (n=5) were fasted for 8 h and then given an intraperitoneal injection of glucose (1 g/kg of body weight). Blood glucose was monitored� at 0, 15, 30, 60, 90, and 120 min after injection. Serum samples were collected at 30 d after transplantation� from HAEC-transplanted mice with euglycemia (n=5), sham-operated mice (n=5) and normal control mice (n=5) for human C-peptide levels analysis. The mice were fasted for 8 h and given an intraperitoneal injection of glucose (1 g/kg of body weight) before blood samples were collected� from the orbital plexus. The ultrasensitive human� C-peptide ELISA kit (Mercodia, Uppsala, Sweden) was used according to the manufacturer's instruction.

Statistical analysis

Data were expressed as mean�SD. Statistical differences between groups were assessed by Student's t-test. P<0.05 was considered statistically significant.

Results

Gene expression of HAEC after induction of differentiation

To determine whether the HAEC had undergone pancreatic� differentiation, gene expression profiles for pancreatic b-cells differentiation markers and hormones were assessed using RT-PCR. Insulin and other pancreatic b-cell related genes, such as pancreas duodenum homeobox-1 (Pdx-1), paired box gene 6 (Pax-6), NK2 transcription factor-related locus 2 (Nkx-2.2), Islet 1 (Isl-1), glucokinase (GCK), and glucose transporter-2 (Glut-2), were expressed at 7, 14 and 21 d of induction. The expression of GCK and Glut-2 indicated that the HAEC might have glucose-sensing ability after induction of differentiation. As shown in Fig. 1(A), the expression of Octamer-4 (Oct-4), a pluripotent cell marker, was detected in non-induced HAEC, but not in induced HAEC at 7 d of induction and thereafter. Immunostaining for insulin and Pdx-1 were also observed at 14 d of induction, but not in non-induced HAEC [Fig. 1(B)]. These results indicated that HAEC could be induced to differentiate into b-cell-like cells in vitro.

C-peptide release

Because of the controversy surrounding insulin uptake into cells from media supplements [17,18], we measured C-peptide in the cell culture supernatants to further investigate� the function of insulin synthesis in the induced HAEC. C-peptide is the byproduct of de novo insulin synthesis, co-secreted from the pancreas with and in equimolar amounts (i.e., an equal number of each molecule) to insulin [19]. Therefore, the presence of C-peptide in HAEC supernatants at 7 (111.43�0.7 pM per 106 cells), 14 (110.83�0.86 pM per 106 cells), 21 d (108.6�2.48 pM per 106 cells) of induction indicated that proinsulin is synthesized� and processed in these cells. In addition, there was no detectable amount of C-peptide in HAEC supernatants� at day 0 of induction [Fig. 2(A)].

To determine whether HAEC were responsive to different� concentrations of glucose, KCl and tolbutamide after induction of differentiation, the release of C-peptide into the culture medium was measured using human C-peptide RIA kit. The results showed that the induced HAEC could secret C-peptide in a glucose-regulated manner. No detectable C-peptide release was observed in the non-induced� HAEC culture medium, even in the presence of glucose stimulation [Fig. 2(B)]. Direct depolarization of the induced HAEC by adding sequential concentrations of KCl consistently resulted in notable increases in secreted C-peptide in a concentration-dependent manner during 1 h incubations [Fig. 2(C)]. We inferred the presence of functional sensitive potassium (KATP) channels in the cells from increases in C-peptide release over basal levels upon the addition of tolbutamide, an inhibitor of KATP-channels [Fig. 2(D)] [20]. These results indicated that HAEC could be induced to differentiate into functional insulin-producing� cells, exhibiting secretory characteristics similar to that of pancreatic b-cell.

Reversal of hyperglycemia in STZ-induced diabetic mice

To determine whether the induced HAEC possessed the capacity to correct hyperglycemia in diabetic mice, C57/B1 mice were either induced to become diabetic with STZ before transplantation with induced HAEC (2-3�106 per mouse) or treated with sham operation. As shown in Fig. 3(A), glucose levels in the HAEC-implanted mice decreased� and were below 13.9 mM within 30 d following� transplantation. In contrast, blood glucose levels in the diabetic control mice and in mice receiving sham operation� remained elevated (P<0.01). The body weights of the HAEC-implanted mice were restored gradually, whereas the mice in the other two groups persistently lost weight [Fig. 3(B)]. These results suggested that induced HAEC are functional in vivo and capable of reversing hyper�glycemia in diabetic mice.

To further evaluate the function of the implanted HAEC, we performed an IPGTT on induced HAEC-implanted mice (n=5), sham-operated mice (n=3) and non-diabetic control mice (n=3) after 30 d of normalized glucose levels following the transplantation. As illustrated in Fig. 3(C), blood glucose levels in normal control mice rose rapidly, with peak values obtained at 15 min, followed by a return to the normal range between 60 and 120 min. Blood glucose� levels in the HAEC-implanted mice were generally higher, but likewise displayed a peak at 15 min and returned to the normal range at 120 min. Blood glucose levels in the sham-operated mice kept higher than 20 mM during the test. These results indicated that the implanted HAEC were indeed responsive to a glucose challenge in vivo.

To confirm the presence and function of HAEC in vivo, HAEC were infected with lentivirus coding GFP before induction of differentiation. The C-peptide-secreting characteristics� and hyperglycemia-reversing ability of induced� HAEC were not changed by the lentivirus infection� (data not shown). The mice were killed, and the transplanted� cells in the subrenal capsule were examined for insulin expression by anti-insulin antibody. Transplanted HAEC were positive for both insulin and GFP expression at 30 d after transplantation [Fig. 4(A)]. As in the sham-operated controls, no diffuse interstitial expansion, tubular� atrophy and massive leukocytic infiltration were observed at the site of transplantation in the HAEC-transplanted mouse kidneys [Fig. 4(B)].

To confirm glucose-responsive insulin secretion in HAEC-transplanted mice, human C-peptide levels in the serum of the mice after glucose injection were measured using the ELISA kit. After injection, a significant increase of human C-peptide in HAEC-transplanted mice serum samples was detected [Fig. 4(C)].

�

Discussion

Although recent studies have demonstrated the feasibility of generating insulin-producing cells from stem/progenitor� cells of various cellular sources or genetically engineered somatic cell lines [2-10,21,22], some obstacles, such as immune rejection and autoimmunity against newly formed insulin-producing cells derived from pancreatic stem cells, still remain. Despite their promising potential, it may also prove difficult to obtain enough autologous adult stem cells from these organs.

To overcome these limitations, we explored the possibility� of using HAEC as sources for differentiation into insulin-producing cells under specific in vitro culture conditions. HAEC are a monolayer of epithelial cells endowed� with unique secretory properties that line the human� amniochorion [23]. HAEC develop from the epiblast� by 8 d after fertilization and before gastrulation, suggesting� that they might maintain the plasticity of pregastrulation embryo cells. Because of the lack of major� histocom�patibility complex class II antigens and mild expression� of major histocompatibility complex class I antigens [15], HAEC have been used in researching the treatment of neural degenerative diseases [24]. Previous studies have proven that HAEC can be induced to differentiate� into endodermal pancreatic cells and can normalize� blood glucose levels after transplantation into STZ-induced diabetic nude mice [13,14]. In the present study, we further detected the production and release of C-peptide of differentiated HAEC induced by nicotinamide. The secretion process can be regulated. By transplanting HAEC into diabetic C57 mice, the cells can normalize the blood glucose of the mice and maintain euglycemia for 30 d.

Previous studies have shown that nicotinamide can induce� differentiation and maturation of human fetal pancreatic� islet cells [25], and that duct tissue from human� tissue treated with nicotinamide can be directed to differentiate� into glucose responsive islet tissue in vitro [5]. The nicotinamide-induced differentiation activated a number of genes related to pancreatic b-cell development and function in HAEC, such as insulin, Pdx-1, Pax-6, Nkx-2.2, Isl-1, and the expression of GCK and glut-2, two important components in "glucose-sensing apparatus" of pancreatic -cells [26], indicating that these cells might have glucose-sensing ability. Unlike previous reports [13,14], we have not detected glucagon expression during the period of induction, and the expression of Pdx-1 could not be detected before induction of differentiation. Additionally, the expression of Nkx 6.1 and Pax-4 were not detected during differentiation (data not shown). Because� there was no obvious change in the expression of the detected transcription factors Pdx-1, Pax-6, Nkx-2.2 and Isl-1 at different times during induction, we speculated� that the nicotinamide-induced differentiation process did not recapitulate a normal pancreatic development pathway in vivo. As demonstrated by immuno�fluorescence analysis, the expression of insulin and Pdx-1 after 14 d of induction further proved that HAEC can exhibit b-cell-like characteristics after induction of differentiation� with nicotinamide for a longer period in vitro. Although the proportions of immunopositive cells are small, the possibility that the cells further differentiated� into b-cell-like cells in vivo after transplantation can not be ruled out [27,28].

The secretion of human C-peptide in HAEC cell culture detected by RIA kit indicated that insulin de novo synthesis� and processing may be present in induced HAEC. Although the secretion of human C-peptide did not exhibit a time-dependent manner after induction of differentiation, the cells immunopositive for insulin and Pdx-1 could not be detected until 14 d of induction in vitro. These results indicated that cell differentiation proceeded during the induction� process.

Glucose-stimulated insulin secretion from pancreatic b-cells is regulated by a series of electrogenic events leading� to exocytosis of insulin-containing granules [29]. The induced HAEC exhibited glucose responsiveness in C-peptide release [Fig. 2(B)], indicating that the induced HAEC may possess the major function of b-cell, namely, insulin release in response to changes in extracellular glucose� concentration. The induced HAEC were also responsive� to KCl [Fig. 2(C)], suggesting the presence of insulin-containing secretory granules. Insulin secretion from pancreatic b-cells is acutely regulated by a complex interplay of metabolic and electrogenic events. The electrogenic� mechanism regulating insulin secretion from b-cells is commonly referred to as the KATP channel-dependent� pathway [30]. Depolarization of b-cell by closing� the KATP channel results in the opening of voltage-dependent Ca2+ channels, and influx of Ca2+ is the main trigger for insulin secretion [31]. Our results showed that the C-peptide release of induced HAEC was responded to tolbutamide, a KATP channel inhibitor, suggesting the presence� of functional KATP channels. The C-peptide secreting� characteristics of induced HAEC indicated that these cells could function similarly to normal pancreatic b-cells in vitro.

Induced HAEC have the ability to replace b-cell function� in vivo. After implantation of induced HAEC, blood glucose� levels of diabetic C57 mice decreased and remained normal� within 30 d, while body weights continued to increase.

One problem inherent in our study is that normoglycemia reached after transplantation might be due to regeneration of b-cell mass. However, there are reasons that argue against this possibility. None of the sham-operated diabetic� mice, which remained hyperglycemic throughout the study, survived more than 8 weeks after surgery, and untreated diabetic mice all remained hyperglycemic and died within 10 weeks. HAEC-implanted mice reached normoglycemic within 1 week after implantation, even though it was reported that pancreas regeneration in mice takes approximately 6-8 weeks in the case of 60% of pancreatectomized animals [32]. About 60% of normoglycemic HAEC-implanted mice developed hyper�glycemia within 7-8 weeks after the transplantation, indicating� that this glucose decrease process is reversible; reversibility likely depends on the half-life of the implanted cell cluster, suggesting the absence of contribution of b-cell mass regeneration to glycemia normalization.

As tested by IPGTT, the blood glucose levels of transplanted� diabetic mice exhibited a similar kinetics to that of normal control mice, representing the recovery of insulin secretion ability of the HAEC-transplanted mice. The immunofluorescence analysis of kidney sections showed that induced GFP-labeled HAEC survived and presented� insulin-positive staining 30 d after transplantation. It was demonstrated that mild mononuclear cell infiltration� existed in kidneys receiving surgical operations. However, no intensive immunorejection was observed in HAEC-transplanted kidneys, indicating that the immune response was caused by surgery and that the transplanted cells did not cause severe immunorejection in vivo. Although the mouse serum samples caused high background signal in the ELISA assay, a statistically significant increase of human� C-peptide in HAEC-transplanted mouse serum could be detected. Taken together, these results suggest that these induced cells have similar function to b-cells in vivo and could be used as a b-cell replacement.

In summary, our findings present evidence that HAEC could be induced to differentiate into functional insulin-producing cells, which may provide a source of b-cells for the treatment of type 1 diabetes by transplantation.

Acknowledgements

We thank Dr. Xuejun Zhang and Dr. Bao Zhang for their technical assistance and helpful discussion. This study received technical support from Cellstar Biotechnologies Company (Shanghai, China).

References

 1�� Shapiro AM, Lakey JR, Ryan EA, Korbutt GS, Toth E, Warnock GL, Kneteman NM et al. Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. N Engl J Med 2000, 343: 230-238

 2�� Soria B, Roche E, Berna G, Leon-Quinto T, Reig JA, Martin F. Insulin-secreting cells derived from embryonic stem cells normalize� glycemia in streptozotocin-induced diabetic mice. Diabetes 2000, 49: 157-162

 3�� Assady S, Maor G, Amit M, Itskovitz-Eldor J, Skorecki KL, Tzukerman M. Insulin production by human embryonic stem cells. Diabetes 2001, 50: 1691-1697

 4�� Lumelsky N, Blondel O, Laeng P, Velasco I, Ravin R, McKay R. Differentiation of embryonic stem cells to insulin-secreting structures� similar to pancreatic islets. Science 2001, 292: 1389-1394

 5�� Bonner-Weir S, Taneja M, Weir GC, Tatarkiewicz K, Song KH, Sharma A, O�Neil JJ. In vitro cultivation of human islets from expanded ductal tissue. Proc Natl Acad Sci U S A 2000, 97: 7999-8004

 6�� Gao R, Ustinov J, Pulkkinen MA, Lundin K, Korsgren O, Otonkoski T. Characterization of endocrine progenitor cells and critical factors� for their differentiation in human adult pancreatic cell culture. Diabetes 2003, 52: 2007-2015

 7�� Yang L, Li S, Hatch H, Ahrens K, Cornelius JG, Petersen BE, Peck AB. In vitro trans-differentiation of adult hepatic stem cells into pancreatic endocrine hormone-producing cells. Proc Natl Acad Sci U S A 2002, 99: 8078-8083

 8�� Zalzman M, Anker-Kitai L, Efrat S. Differentiation of human liver-derived, insulin-producing cells toward the b-cell phenotype. Diabetes� 2005, 54: 2568-2575

 9�� Okumura K, Nakamura K, Hisatomi Y, Nagano K, Tanaka Y, Terada K, Sugiyama T, et al. Salivary gland progenitor cells induced� by duct ligation differentiate into hepatic and pancreatic lineages. Hepatology 2003, 38: 104-113

10� Timper K, Seboek D, Eberhardt M, Linscheid P, Christ-Crain M, Keller U, M�ller B et al. Human adipose tissue-derived mesenchymal� stem cells differentiate into insulin, somatostatin, and glucagon expressing cells. Biochem Biophys Res Commun 2006, 341: 1135-1140

11� Hori Y, Gu X, Xie X, Kim SK. Differentiation of insulin-producing cells from human neural progenitor cells. PLoS Med 2005, 2: e103

12� Sandler TW. In: Coryell P eds. First Week of Development: Ovulation� to Implantation Langman�s Medical Embryology. 7th edn. Baltimore, MD: William and Wilkins 1995

13� Miki T, Lehmann T, Cai H, Stolz DB, Strom SC. Stem cell characteristics� of amniotic epithelial cells. Stem Cells 2005, 23: 1549-1559

14 Wei JP, Zhang TS, Kawa S, Aizawa T, Ota M, Akaike T, Kato K et al. Human amnion-isolated cells normalize blood glucose in streptozotocin-induced diabetic mice. Cell Transplant 2003, 12: 545-552

15� Adinolfi M, Akle CA, McColl I, Fensom AH, Tansley L, Connolly P, Hsi BL et al. Expression of HLA antigens, b 2-microglobulin and enzymes by human amniotic epithelial cells. Nature 1982, 295: 325-327

16� Zufferey R, Dull T, Mandel RJ, Bukovsky A, Quiroz D, Naldini L, Trono D. Self-inactivating lentivirus vector for safe and efficient in vivo gene delivery. J Virol 1998, 72: 9873-9880

17� Hansson M, Tonning A, Frandsen U, Petri A, Rajagopal J, Englund MC, Heller RS et al. Artifactual insulin release from differentiated embryonic stem cells. Diabetes 2004, 53: 2603-2609

18� Paek HJ, Morgan JR, Lysaght MJ. Sequestration and synthesis: the source of insulin in cell clusters differentiated from murine embryonic� stem cells. Stem Cells 2005, 23: 862-867

19� Sherry NA, Tsai EB, Herold KC. Natural history of b-cell function� in type 1 diabetes. Diabetes 2005, 54 Suppl 2: S32-S39

20� Sturgess NC, Ashford ML, Cook DL, Hales CN. The sulphonylurea receptor may be an ATP-sensitive potassium channel. Lancet 1985, 2: 474-475

21� Qin XY, Shen KT, Song LJ, Zhang X, Han ZG. Regulated production� of mature insulin in rat hepatoma cells: insulin production� is up-regulated by dexamethasone and down-regulated by insulin. Acta Biochim Biophys Sin 2006, 38: 89-94

22� Zhang Y, Yao L, Shen K, Xu M, Zhou P, Yang W, Liu X et al. Genetically engineered K cells provide sufficient insulin to correct� hyperglycemia in a nude murine model. Acta Biochim Biophys Sin 2008, 40: 149-157

23� Matthews RN, Faulk WP, Bennett JP. A review of the role of amniotic membranes in surgical practice. Obstet Gynecol Annu 1982, 11: 31-58

24� Kakishita K, Nakao N, Sakuragawa N, Itakura T. Implantation of human amniotic epithelial cells prevents the degeneration of nigral dopamine neurons in rats with 6-hydroxydopamine lesions. Brain Res 2003, 980: 48-56

25� Otonkoski T, Beattie GM, Mally MI, Ricordi C, Hayek A. Nicotinamide� is a potent inducer of endocrine differentiation in cultured human fetal pancreatic cells. J Clin Invest 1993, 92: 1459-1466

26� Newgard CB, McGarry JD. Metabolic coupling factors in pancreatic� beta-cell signal transduction. Annu Rev Biochem 1995, 64: 689-719

27� Davani B, Ikonomou L, Raaka BM, Geras-Raaka E, Morton RA, Marcus-Samuels B, Gershengorn MC. Human islet-derived precursor� cells are mesenchymal stromal cells that differentiate and mature to hormone-expressing cells in vivo. Stem Cells 2007, 25: 3215-3222

28� Tang DQ, Cao LZ, Burkhardt BR, Xia CQ, Litherland SA, Atkinson MA, Yang LJ. In vivo and in vitro characterization of insulin-producing� cells obtained from murine bone marrow. Diabetes 2004, 53: 1721-1732

29� Easom RA. Beta-granule transport and exocytosis. Semin Cell Dev Biol 2000, 11: 253-266

30� Straub SG, Sharp GW. Glucose-stimulated signaling pathways in biphasic insulin secretion. Diabetes Metab Res Rev 2002, 18: 451-463

31� MacDonald PE, Wheeler MB. Voltage-dependent K(+) channels in pancreatic beta cells: role, regulation and potential as therapeutic� targets. Diabetologia 2003, 46: 1046-1062

32� Martin F, Andreu E, Rovira JM, Pertusa JA, Raurell M, Ripoll C, Sanchez-Andr�s JV et al. Mechanisms of glucose hypersensitivity in beta-cells from normoglycemic, partially pancreatectomized mice. Diabetes 1999, 48: 1954-1961