Extracellular matrix gel is necessary for in vitro cultivation of insulin producing cells from human umbilical cord blood derived mesenchymal stem cells

Pancreatic islet cell transplantation is an effective approach to treat type 1 diabetes instead of insulin injection.^1,2 But this therapy is not widely used because of the severe shortage of transplantable donor islets. Recently many attempts have been made to generate new insulin producing cells in vitro from non-β cells.^3-11 As an excellent source for cell based therapy, mesenchymal stem cells (MSCs) are multipotent and have high ex vivo expansive potential.^12-14 It has been demonstrated that murine bone marrow derived MSCs differentiated into insulin expressing cells.^15,16 Using MSCs to treat diabetes seems promising. Currently, bone marrow (BM) represents the main source of MSCs for both experimental and clinical studies,¹⁷ but the source of BM is limited and the amount of bone marrow derived MSCs and their capacity for differentiation declines with age.¹⁸ In addition, obtaining a BM sample requires a painful invasive procedure. These have led investigators to search for other BM substitutes as sources of MSCs.

Umbilical cord blood (UCB) is abundantly available and can be routinely harvested without risk to the donor and infectious agents such as cytomegalovirus are rare exceptions.¹⁹ These characteristics make UCB a very promising source of MSCs. However, a key problem that remained is whether human UCB derived MSCs can differentiate into insulin producing cells. In this study, we demonstrate that UCB derived MSCs can be induced to differentiate into pancreatic endocrine cells in vitro in the presence of extracellular matrix (ECM) gel for maturation and formation of three dimensional structures. This study provides support for continuing efforts utilizing adult stem cells as a steady and renewable source of insulin producing cells for transplantation into patients with type 1 diabetes.

Collection of human UCB
Human UCB (n=42, 36–40 weeks) was obtained from Department of Obstetrics and Gynaecology, the Second Hospital of Harbin Medical University, China. UCB was collected from the umbilical cord vein with informed consent of the mother. A bag system containing 17 ml of anticoagulant (citrate, phosphate, dextrose) was used. All UCB units were processed within 3 hours after delivery.

Culture of MSCs from UCB
To isolate mononuclear cells (MNCs), each UCB unit was diluted 1:1 with phosphate buffered solution (PBS) and carefully loaded onto Ficoll Hypaque solution (1.077 g/ml, Sigma-Aldrich Co, USA). After density gradient centrifugation at 800 g for 16 minutes at room temperature, MNCs were removed from the interphase, washed twice with PBS and resuspended in low glucose Dulbecco's modified Eagle's medium (L-DMEM, 5.5 mmol/L glucose, Invitrogen Corporation, USA) supplemented with 30% foetal bovine serum (FBS, Invitrogen). After counting, cell suspension was seeded in uncoated T25 culture flasks (Orange Scientific, Belgium) at a concentration of 1×10⁶ cells/ml. Cultures were maintained at 37°C in a humidified atmosphere containing 5% CO₂ and the medium changed 8 days later. When fibroblast like cells at the base of the flask reached confluence, they were harvested with 0.25% trypsin EDTA (Sigma-Aldrich) and passaged at 1:3 dilution as passage one. Then half of the medium was changed with L-DMEM containing 20% FBS every other day to make 30% FBS concentration decrease to 20%.

In vitro differentiation cultures
For differentiation into osteogenic cells, the UCB-MSCs at third passage were plated at 1×10⁴ cells/well in 24-well plates. At 70% confluency, the cells were cultured for 14–21 days in L-DMEM supplemented with 10% FBS, 10^–7 mol/L dexamethasone (Sigma-Aldrich), 50 µmol/L ascorbic acid-2 phosphate (Sigma-Aldrich) and 10 mmol/L β-glycerol phosphate (Sigma-Aldrich).²⁰Osteogenic differentiation was confirmed by Von Kossa staining.

For differentiation into adipogenic cells, the cells at third passage were plated at 1×10⁴ cells/well in 24-well plates. At 70% confluency, the cells were cultured for 14–21 days in L-DMEM supplemented with 0.5 mmol/L 3-isobutyl-1-methylxanthine (Sigma-Aldrich), 1 µmol/L dexamethasone, 0.1 mmol/L indomethacin (Sigma-Aldrich) and 10% FBS.²¹ Adipogenic differentiation was evaluated by the cellular accumulation of neutral lipid vacuoles that were stained with oil red O.

For pancreatic endocrine differentiation, expanded MSCs from passage three were allowed to reach 80%–90% confluence and induced to differentiate into insulin secreting cells by an adjusted 3-step protocol.²² In step 1, the cell monolayer was treated for 24 hours with high glucose DMEM (H-DMEM, 25 mmol/L glucose) supplemented with 10% FBS and 10^–6 mol/L retinoic acid (RA, Sigma-Aldrich ), then the medium was changed to H-DMEM with only 10% FBS for 2 days. In step 2, the cells were detached with 0.25% trypsin EDTA and seeded in 12-well plates with or without ECM gel (Sigma- Aldrich) coating. The medium was changed to L-DMEM, supplemented with 10% FBS, 10 mmol/L nicotinamide (Sigma-Aldrich) and 20 ng/ml epidermal growth factor (EGF, Peprotech, UK) for 6 days. In step 3, to mature the insulin producing cells, the low glucose medium was supplemented with 10% FBS and 10 nmol/L exendin-4 (Sigma-Aldrich)) for 6 days. Cellular differentiation was monitored by observation of three dimensional formation of islet like cell cluster, the expression of genes related to pancreatic endocrine cell development and insulin production. As a control group, cells were cultured in L-DMEM containing only 10% FBS.

Electron microscopic analysis
Pre-induced cells and differentiated pancreatic endocrine cells were fixed in 5% pentan-1, 5 diol for 2 hours at 4°C, washed in PBS, transferred to 1% osmic acid for 2 hours at 4°C, then washed in PBS again, dehydrated in ethanoic acid and embedded. Ultra thin sections were counterstained with uranyl ethanoate and lead citrate and viewed by electron microscope (JEM-1220, JEOL, Japan).

Immunofluorescence
Induced cells with ECM gel coating were released by 0.25% trypsin and washed three times with PBS. Then cytospin slides were made for insulin, C-peptide and glucagon expression. The cells were fixed with 4% methanal for 30 minutes at room temperature and incubated overnight at 4°C with primary antibodies: mouse antihuman insulin 1:100 (Sigma-Aldrich), mouse antihuman C-peptide 1:100 (Abcam Co., Cambridge, UK) and goat antihuman glucagon 1:100 (Santa Cruz Biotechnology, USA). Subsequently, the cells were washed with PBS three times and incubated at 37°C for 1 hour with fluorescent labelled second antibody: fluorescein isothiocyanate (FITC) labelled goat antimouse IgG 1:50, rhodamine labelled goat antimouse IgG 1:50 and rhodamine labelled rabbit antigoat IgG 1:50 (Zhongshan Goldenbridge, China). After being washed with PBS, cells were mounted with propan-1, 2, 3 triol PBS (9:1). The cells were visualized and photographed through a laser scanning, confocal microscope TE2000-U (Nikon, Japan). Pre-induced cells and induced cells without ECM coating, which were cultured on coverslips in plastic 12-well plates, were treated as above.

Reverse transcription polymerase chain reaction (RT-PCR) analysis
Total RNA was extracted by TRIzol (Invitrogen) according to the manufacturer's instructions and quantified by UV spectroscopy. To prepare RNA for PCR analysis, 2 µg total RNA was converted to the cDNA using SuperScript II reverse transcriptase (Invitrogen) with oligo (dT) (Promega, USA) and random hexamer primers (Promega). PCR was performed using Taq DNA polymerase (Invitrogen). All PCR experiments were performed using a PCR system TC-XP-G (Bioer, China). Products were analysed by polyacrylamide gel electro- oresis. The name and sequences of the primers, the sizes of PCR products, cycles and annealing temperature for each pair are listed in Table. Gene β-actin was used as an internal control.

Flow cytometric analysis
Immunophenotyping of human UCB-MSCs was performed with antibodies against human antigens CD19, CD29, CD34, CD44, CD45 (BD Biosciences, USA) and CD105 (Caltag Laboratories, USA). The MSCs at third passage were released by 0.25% trypsin- EDTA. Cells were resuspended, 1×10⁶ in 200 µl PBS and incubated with fluorescein isothiocyanate or phycoerythrin conjugated antibodies for 30 minutes at room temperature. After being washed, the cells were analyzed with a flow cytometer.

To determine the amount of insulin positive cells, pre-induced cells and differentiated pancreatic endocrine cells were released by trypsinization and analyzed by flow cytometry. Briefly, cells were fixed with cold 1% methanal, incubated with 0.01% Triton X-100, washed twice with PBS, then stained with antiinsulin antibodies and rhodamine labelled second antibodies. After being washed twice with PBS, cells were analyzed with a FACSCalibur (BD Biosciences).

Insulin detection assay
To test whether the insulin release of induced endocrine cells was glucose dependent, two glucose concentrations (5.5 mmol/L and 28 mmol/L) were used. The differentiated cells on ECM gel, which were plated in 12-well dishes, were washed twice with PBS and incubated in 0.5 ml serum free DMEM containing 5.5 mmol/L glucose for 4 hours at 37°C. The media were collected and measured for basal insulin secretion and fresh media with either 5.5 mmol/L or 28 mmol/L glucose were added. At the end of this second 4 hours of incubation, the media were collected and stored at –70°C until they were assayed for insulin content. To estimate total cellular insulin levels, the total cell protein content was tested with the BCA Protein Assay Kit (Beyotime, China). Measurement of cellular and secreted insulin was performed with chemiluminescence immunoassay system ADVIA Centaur (Bayer, USA). Pre-induced MSCs treated as above were used as a control group.

Statistical analysis
Data were presented as mean ± standard deviation (SD). Results were analyzed by independent samples t test and one way analysis of variance. Statistical significance was set at P < 0.05.

Isolation and characterization of human UCB-MSCs
In this study, we isolated 16 MSCs from 42 term UCB units (16/42, 38%). The onset of fibroblast like cell colony formation could be observed during 6 to 12 days after first seeding and the attachment of osteoclast like cells to culture flasks was rare. Specific markers for MSCs; CD29, CD44 and CD105 were detected by flow cytometry, but haematopoietic antigens CD34, CD45 and endothelial antigen CD19 could not be detected. Osteogenic differentiation and adipogenic differentiation could be induced in those cells defined by von Kossa and oil-red O staining.

In vitro differentiation of UCB-MSCs into insulin producing cells
UCB-MSCs were differentiated into insulin secreting cells by a three-step protocol with ECM gel coated. At step 1, changes in cell morphology could not be observed. During further culturing, the rate of cell proliferation became slower and these spindle like cells became short and changed into round epithelial like cells by the end of step 2: about 9 days after differentiation. Meanwhile, some new islet like clusters started to appear, ranging from 100 to 150 µm in diameter. At step 3, more islet like clusters were formed and there were (3.0±1.4) clusters per cm² (Figure 1). Ultrastructural analysis showed that the induced cells on ECM gel contained small secretory granules: a characteristic of pancreatic endocrine cells (Figure 2).

view in a new window

Figure 1. Morphological changes in endocrine differentiation. A: The cells were spindle shaped and fibroblast like before inducement (original magnification ×200); B: In step 2, the induced cells on ECM gel turned shorter and changed into round epithelial like (original magnification×200); C and D: Islet like cell clusters appeared on ECM gel at end of step 2 and more islet like clusters were formed in step 3 (C: original magnification ×100; D: original magnification ×400); E: There were no cell clusters formed without ECM gel (original magnification×200). ECM: extracellular matrix.

To investigate whether cluster formation depended on ECM gel, cells were seeded on dishes without ECM gel in step 2. Under this condition, cell clusters did not form. Moreover, the control of MSCs cultured on ECM gel coated dishes did not produce clusters either.

RT-PCR analysis of gene expression
To determine whether UCB-MSCs had differentiated into pancreatic endocrine cells, gene expression profiles for pancreatic β-cell differentiation markers and hormones were assessed by RT-PCR. As illustrated in Figure 3, endocrine cells differentiation related genes were not expressed in pre-induced cells. After differentiation, many characteristic pancreatic endocrine cell marker genes were expressed by the end of our protocol: PDX-1, Pax4, Ngn3, insulin, glucagon and Glut-2. When induced by the three-step protocol without ECM gel coating, the differentiated cells only weakly expressed PDX-1, Pax4 and Ngn3 and small cluster formation could not be observed. This result suggested that ECM gel might be necessary for endocrine cells' differentiation and maturation in vitro.

Immunofluorescence analysis of insulin, C-peptide and glucagon
Figure 4 shows that the differentiated cells induced with ECM gel, stained for insulin and C-peptide, but the induced cells without ECM gel did not stain. Though the anti–C-peptide antibody recognized both C-peptide and proinsulin, the staining for C-peptide indicated that the differentiated cells were likely to synthesize and process insulin. Furthermore, we found some cells stained for glucagon. We performed double immunofluorescent staining for insulin and glucagon and found that the insulin positive cells did not stain for glucagon. However, considering the insulin content and response to glucose stimulation, we suspected the induced cells were most likely to represent immature pancreatic endocrine cells, which expressed glucagon. Therefore, the result needs further testing.

Flow cytometric analysis of insulin positive cells
It was found that (1.11±0.15) % of pre-induced cells and (1.30±0.31)% of induced cells without ECM coating were insulin positive (n=6, t=1.33, P >0.05). However, the rate of the cells with ECM gel coating was (25.2±3.36) %, which was higher than the other two groups (n=6, F=302, P <0.001) (Figure 5A).

Insulin content and release in response to glucose stimulation
Analysis indicated that pre-induced cells showed no significant release of insulin in the presence or absence of glucose challenge (Figure 5B). After differentiation with ECM gel, these islet like cells produced much more insulin and secreted it into extracellular medium. However, these differentiated cells were not very responsive to glucose challenge. The data indicated that the islet like cells might represent immature β like cells and further induction might be required to reach a high degree of differentiation and maturation.

view in a new window

Figure 5. Frequencies of insulin positive cells and insulin secretion of pre- and induced human UCB-MSCs. A: Flow cytometric analysis of insulin positive cells; B: Basal insulin secretion of pre-induced MSCs was low, but that of induced cells was dramatically increased. ECM: extracellular matrix.

As an additional assessment, we evaluated the content of total cellular insulin in both pre-induced ((0.14±0.02) ng/mg protein) and induced cells ((12.28±4.59) ng/mg protein) and found that induced cells showed a significant increase (n=7, t=6.99, P <0.01). The proportion of insulin release to insulin content in differentiated cells was approximately 9.2%.

The capacity of MSCs to differentiate may be related to tissue source. Kern et al²³ indicated that human UCB-MSCs do not differentiate into adipogenic cells, in contrast to bone marrow and adipose tissue MSCs. Several studies have demonstrated that MSCs derived from mouse bone marrow can be differentiated into insulin secreting cells in vitro. A key problem remains unsolved: whether human UCB-MSCs can differentiate into insulin secreting cells in vitro. In the study reported here, we have obtained insulin producing cells by a three step protocol with ECM gel. These induced cells can express pancreatic β-cell markers, synthesize and secrete functional Islet proteins. It indicates the possibility of deriving functional beta-cells from human UCB-MSCs and warrants further investigation.

Recently, several laboratories have isolated MSCs from UCB,^23-25 but some researchers do not think UCB is an adequate source of MSCs given the low proportion of MSCs from UCB cells.²⁶ In this study, we have obtained cells which show some characteristic features of MSCs from term UCB with an efficiency of greater than 38%, using L-DMEM containing 30% FBS. The immunophenotype of these MSCs like cells, which appeared similar to that of MSCs reported by Kern et al,²³was negative for CD45, CD34 and CD19, but positive for CD29, CD44 and CD105. In addition, these MSCs like cells have the capability to differentiate into osteoblasts and adipocytes. These findings are consistent with the characterization of MSCs, which has been reported.²⁷ Therefore, we believe that the cells we have obtained most likely represent mesenchymal stem cells derived from UCB. Furthermore, the rate of MSCs appearance in this study suggests that it is possible to use UCB-MSCs in clinical therapeutic approaches.

During embryonic development, a cascade of transcription factors is activated to initiate the development of pancreas. A key substance in this system is the transcription factor PDX-1, which is expressed in all pancreatic progenitor cells.²⁸ In this study, the combination of retinoic acid (RA) and high glucose specifically activated pancreatic endocrine cell differentiation from UCB-MSCs by inducing the cells to express the key factors PDX-1 and pancreatic endocrine cell marker Ngn3. The result is consistent with reports that RA and high glucose are essential for the development of the early embryonic pancreas and adult stem cell differentiation into insulin producing cells.^15,29,30 Nicotinamide, a polyADP-ribose synthetase inhibitor, can induce formation of islets from pancreatic progenitor cells, transdifferentiation and maturation of liver stem cells into insulin producing cells.³¹ Epidermal growth factor (EGF) can increase the number of undifferentiated endocrine precursor cells. Upon removal of EGF, a large number of β-cells are differentiated.³²

In the present study, obvious morphological changes of induced cells and the expression of Pax4 were observed after nicotinamide and EGF were added. Nicotinamide and EGF accelerated the differentiation of pancreatic precursor cells into endocrine cells. Exendin-4 is a potent glucagon like peptide 1 agonist that has been shown to stimulate both β-cell replication and neogenesis from ductal progenitor cells. In step 3 of the protocol, the induced cells matured quickly upon the withdrawal of EGF and addition of exendin-4. More islet like clusters and mRNA expression of insulin, glucagon, Glut-2 are observed in this stage. The functional islet proteins were detected too. However, the insulin content of insulin secreting cells we acquired was low and glucose induced, insulin secretion and its proportion (secretion expressed as percent of content) were somewhat lower than in native islets cells. In addition, the differentiated cells also expressed early pancreatic genes, such as Pdx1, Ngn3, and Pax4, suggesting that the in vitro generated insulin secreting cells are immature and some unknown inducing factors are necessary to complete differentiation.

Stem cells differentiation is influenced by multiple factors, such as interactions with adjacent cells, cytokines and extracellular matrix (ECM). Previous study has demonstrated that microenvironment plays an important role in differentiation of stem cells³³ and ECM is essential for cell differentiation through rearrangement of the cytoskeletal network.³⁴ Matrigel, a type of ECM, is necessary for migration of pancreatic progenitor cell, formation of three dimensional cystic structures and protrusion of islet bud.⁶ When embryonic stem cells are cultured on Matrigel, they can form insulin positive, islet like clusters.¹¹ In this study, we found the morphology of differentiated cells seems more β-like on ECM than without it. When UCB-MSCs were induced without ECM gel, the differentiation stopped at an early stage. Islet like cell clusters formation could not be observed by 15 days and secretion of functional islet proteins could not be detected.

There is a very close relationship between the formation of three dimensional structures and pancreatic endocrine cell maturation when UCB-MSCs are induced in vitro. The probable reason is that cell clusters put more cells in contact beneficial for interactions between adjacent cells to accelerate the differentiation process and promote maturation of induced cells. In this study, the collagen and laminin rich ECM gel provide the matrix for formation of three dimensional structures. Though some studies obtained insulin secreting cells from murine MSCs without ECM gel, the induction strategies took about 2 to 4 months¹⁵ or cluster formation and islet protein secretion were not observed.¹⁶ Therefore, matrix and growth factors in ECM gel are critical for maturation of pancreatic endocrine cell derived from UCB-MSCs.

It has been reported that murine bone marrow MSCs can transdifferentiate into insulin producing cells within 2−4 months, which are responsive to glucose stimulation under culturing conditions containing high concentrations of glucose.¹⁵ But, we found that human UCB-MSCs were susceptible to high glucose induced apoptosis (unpublished data). The cells aged early (within 10 days) when cultured in high glucose condition. This indicated that some biological characteristics might be different between murine bone MSCs and human UCB-MSCs, or other factors were involved in differentiation of human UCB-MSCs into competent insulin producing cells.

In conclusion, this study indicates that human UCB-MSCs are capable of differentiating into insulin secreting cells in vitro by a three-step protocol including ECM gel. However, these cells are immature in terms of insulin production and glucose responsiveness. Obviously, more research is needed to make possible the use of human UCB-MSCs in therapeutic approaches to diabetes.

1. Ryan EA, Lakey JRT, Rajotte RV, Korbutt GS, Kin T, Imes S, et al. Clinical outcomes and insulin secretion after islet transplantation with the Edmonton protocol. Diabetes 2001; 50: 710-719.

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

3. Zulewski H, Abraham EJ, Gerlach MJ, Daniel PB, Moritz W, Muller B, et al. Multipotential nestin-positive stem cells isolated from adult pancreatic islets differentiate ex vivo into pancreatic endocrine, exocrine, and hepatic phenotypes. Diabetes 2001; 50: 521-533.

4. Bonner-Weir S, Taneja M, Weir GC, Tatarkiewicz K, Song KH, Sharma A, et al. In vitro cultivation of human islets from expanded ductal tissue. Proc Natl Acad Sci USA 2000; 97: 7999-8004.

5. Petropavlovskaia M, Rosenberg L. Identification and characterization of small cells in the adult pancreas: potential progenitor cells? Cell Tissue Res 2002; 310: 51-58.

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. 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.

8. Leon-Quinto T, Jones J, Skoudy A, Burcin M, Soria B. In vitro directed differentiation of mouse embryonic stem cells into insulin producing cells. Diabetologia 2004; 47: 1442-1451.

9. Jahr H, Bretzel B. Insulin positive cells in vitro generated from rat bone marrow stromal cells. Transplant Proc 2003; 35: 2140-2141.

10. Ianus A, Holz GG, Theise ND, Hussain MA. In vivo derivation of glucose-competent pancreatic endocrine cells from bone marrow without evidence of cell fusion. J Clin Invest 2003; 111: 843-850.

11. Shi Y, Hou L, Tang F, Jiang W, Wang P, Ding M, et al. Inducing embryonic stem cells to differentiate into pancreatic ß cells by a novel three-step approach with activin A and all-trans retinoic acid. Stem Cells 2005; 23: 656-662.

12. Oswald J, Boxberger S, Jorgensen B, Feldmann S, Ehninger G, Bornhauser M, et al. Mesenchymal stem cells can be differentiated into endothelial cells in vitro. Stem Cells 2004; 22: 377-384.

13. Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, Mosca JD, et al. Multilineage potential of adult human mesenchymal stem cells. Science 1999; 284: 143-147.

14. Kang XQ, Zang WJ, Bao LJ, Li DL, Song TS, Xu XL, et al. Fibroblast growth factor-4 and hepatocyte growth factor induce differentiation of human umbilical cord blood derived, mesenchymal stem cells into hepatocytes. World J Gastroenterol 2005; 11: 7461-7465.

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

16. Chen LB, Jiang XB, Yang L. Differentiation of rat marrow mesenchymal stem cells into pancreatic islet beta-cells. World J Gastroenterol 2004; 10: 3016-3020.

17. Koc ON, Gerson SL, Cooper BW, Dyhouse SM, Haynesworth SE, Caplan AI, et al. Rapid hematopoietic recovery after coinfusion of autologous-blood stem cells and culture-expanded marrow mesenchymal stem cells in advanced breast cancer patients receiving high-dose chemotherapy. J Clin Oncol 2000; 18: 307-316.

18. Mueller SM, Glowacki J. Age-related decline in the osteogenic potential of human bone marrow cells cultured in three-dimensional collagen sponges. J Cell Biochem 2001; 82: 583-590.

19. Rubinstein P, Rosenfield RE, Adamson JW, Stevens CE. Stored placental blood for unrelated bone marrow reconstitution. Blood 1993; 8: 1679-1690.

20. Okamoto T, Aoyama T, Nakayama T, Nakamata T, Hosaka T, Nishijo K, et al. Clonal heterogeneity in differentiation potential of immortalized human mesenchymal stem cells. Biochem Biophys Res Commun 2002; 295: 354-361.

21. Lee OK, Kuo TK, Chen WM, Lee KD, Hsieh SL, Chen TH. Isolation of multipotent mesenchymal stem cells from umbilical cord blood. Blood 2004; 103: 1669-1675.

22. Seeberger KL, Dufour JM, Shapiro AM, Lakey JR, Rajotte RV, Korbutt GS. Expansion of mesenchymal stem cells from human pancreatic ductal epithelium. Lab Invest 2006; 86: 141-153.

23. Kern S, Eichler H, Stoeve J, Kluter H, Bieback K. Comparative analysis of mesenchymal stem cells from bone marrow, umbilical cord blood, or adipose tissue. Stem Cells 2006; 24: 1294-1301.

24. Erices A, Conget P, Minguell JJ. Mesenchymal progenitor cells in human umbilical cord blood. Br J Haematol 2000; 109: 235-242.

25. Lu FZ, Fujino M, Kitazawa Y, Uyama T, Hara Y, Funeshima N, et al. Characterization and gene transfer in mesenchymal stem cells derived from human umbilical-cord blood. J Lab Clin Med 2005; 146: 271-278.

26. Wexler SA, Donaldson C, Denning-Kendall P, Rice C, Bradley B, Hows JM. Adult bone marrow is a rich source of human mesenchymal stem cells but umbilical cord and mobilized adult blood are not. Br J Haematol 2003; 121: 368-374.

27. Jackson L, Jones DR, Scotting P, Sottile V. Adult mesenchymal stem cells: Differentiation potential and therapeutic applications. J Postgraduate Med 2004; 53: 121-127.

28. Edlund H. Pancreatic organogenesis-developmental mechanisms and implications for therapy. Nat Rev Genet 2002; 3: 524-532.

29. Micallef SJ, Janes ME, Knezevic K, Davis RP, Elefanty AG, Stanley EG. Retinoic acid induces pdx1-positive endoderm in differentiating mouse embryonic stem cells. Diabetes 2005; 54:301-305.

30. Kojima H, Fujimiya M, Matsumura K, Nakahara T, Hara M, Chan L. Extrapancreatic insulin producing cells in multiple organs in diabetes. Proc Natl Acad Sci U S A 2004; 101: 2458-2463.

31. Yang L, Li S, Hatch H, Ahrens K, Cornelius JG, Petersen BE, et al. 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.

32. Cras-Meneur C, Elghazi L, Czernichow P, Scharfmann R. Epidermal growth factor increases undifferentiated pancreatic embryonic cells in vitro: a balance between proliferation and differentiation. Diabetes 2001; 50: 1571-1579.

33. Watt FM, Hogan B. Out of eden: stem cells and their niches. Science 2000; 287: 1427-1430.

34. Levenberg S, Golub JS, Amit M, Itskovitz-Eldor J, Langer R. Endothelial cells derived from human embryonic stem cells. Proc Natl Acad Sci U S A 2002; 99: 4391-4396.