Cell Transplantation, Vol. 25, pp. 97–108, 2016 Printed in the USA. All rights reserved. Copyright Ó 2016 Cognizant, LLC.

0963-6897/16 $90.00 + .00 DOI: http://dx.doi.org/10.3727/096368915X687732 E-ISSN 1555-3892 www.cognizantcommunication.com

Islet Endothelial Cells Derived From Mouse Embryonic Stem Cells Neha Jain and Eun Jung Lee New Jersey Institute of Technology, Department of Biomedical Engineering, Newark, NJ, USA

The islet endothelium comprises a specialized population of islet endothelial cells (IECs) expressing unique markers such as nephrin and a-1 antitrypsin (AAT) that are not found in endothelial cells in surrounding tissues. However, due to difficulties in isolating and maintaining a pure population of these cells, the information on these islet-specific cells is currently very limited. Interestingly, we have identified a large subpopulation of endothelial cells exhibiting IEC phenotype, while deriving insulin-producing cells from mouse embryonic stem cells (mESCs). These cells were identified by the uptake of low-density lipoprotein (LDL) and were successfully isolated and subsequently expanded in endothelial cell culture medium. Further analysis demonstrated that the mouse embryonic stem cell-derived endothelial cells (mESC-ECs) not only express classical endothelial markers, such as platelet endothelial cell adhesion molecule (PECAM1), thrombomodulin, intercellular adhesion molecule-1 (ICAM-1), and endothelial nitric oxide synthase (eNOS) but also IEC-specific markers such as nephrin and AAT. Moreover, mESC-ECs secrete basement membrane proteins such as collagen type IV, laminin, and fibronectin in culture and form tubular networks on a layer of Matrigel, demonstrating angiogenic activity. Further, mESC-ECs not only express eNOS, but also its eNOS expression is glucose dependent, which is another characteristic phenotype of IECs. With the ability to obtain highly purified IECs derived from pluripotent stem cells, it is possible to closely examine the function of these cells and their interaction with pancreatic b-cells during development and maturation in vitro. Further characterization of tissue-specific endothelial cell properties may enhance our ability to formulate new therapeutic angiogenic approaches for diabetes. Key words: Islet endothelial cells (IECs); Embryonic stem cells (ESCs); Differentiation

INTRODUCTION Several efforts have recently been made to differentiate pluripotent stem cells into endothelial progenitor cells or endothelial cells using various strategies (2,10,17,38,64). These pluripotent stem cell-derived endothelial cells can provide an abundant cell source for tissue engineering applications as well as more personalized medical treatments. Levenberg et al. reported the derivation of endothelial cells from human embryonic stem cells (hESCs) (38), and Blancas et al. demonstrated a chemically defined protocol to differentiate mouse embryonic stem cells (mESCs) into endothelial cells (9,10). Other groups have shown that fluid shear stress can promote an endotheliallike phenotype from early embryonic stem cell differentiation (31,63), and the role of matrix stiffness has also been explored for endothelial differentiation (16,69). Recently, Nolan et al. established a library containing molecular signatures of tissue-specific microvascular endothelial cells. They demonstrated that when murine embryonic stem cell-derived endothelial cells were transplanted into animal models, the generic endothelial cells acquired specific

characteristics of the tissues in which they were transplanted (45). Notably, endothelial cells are recognized as a heterogeneous cell population in their structure and function based on the location in the body (3). However, protocols for in vitro stem cell differentiation into organspecific endothelial lineages are not yet available. Endothelial cells found in pancreatic islets, known as islet endothelial cells (IECs), are being increasingly appreciated as an important contributing factor in pancreatic islet development and maturation (28,65), as pancreatic islets have a very dense capillary network in vivo. IECs are known to exhibit unique phenotypes that distinguish them from neighboring endothelial cell populations. It has previously been reported that isolated human IECs have more fenestrations than endothelial cells found within the exocrine part of the pancreas (20,46,66,67). In addition to classical endothelial cell makers such as von Willebrand factor, platelet endothelial cell adhesion molecule (PECAM1), and uptake of acetylated lowdensity lipoprotein (LDL), IECs are shown to express unique markers including nephrin and a-1 antitrypsin (AAT)

Received June 17, 2014; final acceptance March 3, 2015. Online prepub date: March 6, 2015. Address correspondence to Eun Jung Lee, Ph.D., New Jersey Institute of Technology, Department of Biomedical Engineering, 323 Dr. MLK Blvd, Fenster 615, Newark, NJ 07102, USA. Tel: +1-973-596-8471; E-mail: [email protected]

97

98

(42,67). Other studies have shown that rat IECs constitutively express endothelial nitric oxide synthase (eNOS), and the expression of eNOS is glucose dependent, unlike other endothelial cells (55,66,67). These unique properties of IECs contribute to normal development of b-cells in the pancreas. IECs secrete various molecules such as collagen type IV (29), laminin (26,44,59), connective tissue growth factor (CTGF) (14), and hepatocyte growth factor (HGF) (27) that are important for insulin secretion, b-cell proliferation, differentiation, and endocrine lineage specification. Recent studies have also highlighted the role of IECs in the revascularization and stabilization of transplanted islets in the host tissue (11). Mattsson et al. showed that the expression of angiogenic and inhibitory factors by IECs vary at different time points posttransplantation (43). However, due to difficulty in isolating a pure population of IECs and expanding in in vitro culture, the information on IECs is still relatively limited. In the current study, we report the first demonstration of endothelial cells exhibiting IEC markers differentiated from mESCs. This population of IECs appeared as a side product in the differentiation culture of mESCs into pancreatic b-cells. A pure population of endothelial cells was successfully isolated from the vicinity of insulinproducing cells and expanded in in vitro culture for further characterization. These highly purified endothelial cells with islet-specific characteristics will further enhance our understanding of tissue-specific endothelial cell function. Moreover, studies investigating the interaction between neighboring insulin-producing b-cells can lead to a significant step in formulating new therapeutic angiogenic approaches for diabetes. MATERIALS AND METHODS mESC Culture and Differentiation Into Pancreatic β-Cells mESCs (a generous gift from Dr. Yibing Qyang) were cultured on a mouse embryonic fibroblast (MEF; Globalstem, Rockville, MD, USA) layer in a DMEM medium supplemented with 10% fetal bovine serum (FBS), 1% penicillin–streptomycin, 1% L-glutamine, 1% sodium pyruvate, and 1% nonessential amino acids. All reagents were purchased from Life Technologies (Carlsbad, CA, USA). To derive insulin-producing cells from mESCs, a previously established protocol was used (50). Briefly, embryoid bodies (EBs) were formed by a hanging drop method and were collected after 2 days. EBs were cultured for an additional 3 days in a suspension culture and were then plated on a 0.1% gelatin-coated dish for 9 days to derive endodermal lineage in an IMDM medium supplemented with 20% FBS, 2 mM L-glutamine, 2 mM nonessential amino acids, 450 µM monothioglycerol (Sigma-Aldrich, St. Louis, MO, USA), 0.05 mg/ml streptomycin, and

JAIN AND LEE

0.03 mg/ml penicillin. EBs were then digested and plated on a polyornithine and laminin (Sigma-Aldrich)coated dish. Pancreatic differentiation medium containing DMEM/F12 medium supplemented with 10% FBS, 20 nM progesterone, 100 µM putrescine, 1 µg/ml laminin, 10 mM nicotinamide, 25 µg/ml insulin, 30 nM sodium selenite, 50 µg/ml transferrin, B27 (Life Technologies), 0.05 mg/ml streptomycin, and 0.03 mg/ml penicillin was added to the cells. All the reagents, unless stated otherwise, were purchased from Sigma-Aldrich. After 24 h, the culture medium was switched to a serum-free differentiation medium without 10% FBS. Cells remained in culture for an additional 18 days until completion of the differentiation. Insulin-producing cells were detected after 33 days of differentiation by a dithizone (DTZ) staining as previously described (51). To further confirm b-cell identity, an immunohistochemistry analysis was performed. Fixed cells were incubated with a monoclonal mouse anti-insulin antibody (1:1,000; Abcam, Cambridge, MA, USA) overnight and then with secondary goat anti-mouse (1:200; Santa Cruz, Dallas, TX, USA) for 2 h. Identification and Isolation of mESC-Derived Endothelial Cells (mESC-ECs) During various stages of mESC differentiation into pancreatic b-cells, acetylated LDL (Biomedical Technologies Inc., Stoughton, MA, USA) was used to identify a mature endothelial cell population in the culture. Cells were incubated with LDL diluted 1:10 in culture medium for 4 h at 37°C, and LDL-positive cells were detected under a florescence microscope (IX81 DSU, Olympus, Somerset, NJ, USA). At the end of the differentiation process, that is, day 33, LDL-positive cells in culture were selectively isolated from the rest of the population using sterile cloning discs (Capitol Scientific, Austin, TX, USA). Isolated endothelial cells were placed in a new dish coated with 0.1% gelatin and were initially cultured in the pancreatic differentiation medium supplemented with 10% FBS. After a few passages, culture medium was replaced with MCDB complete endothelial growth media containing 10% FBS, 1% penicillin–streptomycin (Sigma-Aldrich), and Endogro, an endothelial cell growth supplement from VEC Technologies (Rensselaer, NY, USA). Cells were passaged at 70–80% confluence, and the cell culture medium was exchanged every 2–3 days. Characterization of mESC-ECs Fluorescence-activated flow cytometry (FACS) was used to confirm the phenotype and the purity of the isolated cell population. For FACS analysis, cells were incubated with a polyclonal rabbit anti-PECAM1 (1:20; Santa Cruz) in FACS buffer (PBS with 1% BSA and 2 mM EDTA) for 30 min at 4°C. Cells were then washed three times with cold PBS, and secondary donkey anti-rabbit

ISLET ENDOTHELIAL CELLS FROM mESCs

APC (1:20; Fisher Scientific, Pittsburgh, PA, USA) was added. Samples were analyzed using a FC500 Flow cytometer (Beckman Coulter, Pasadena, CA, USA) and FlowJo software. Further characterization of cell phenotype was performed using Western blotting analysis. Cell and tissue lysates were prepared in RIPA buffer with 1% Triton X-100 (Boston Bioproducts, Ashland, MA, USA). Cell lysates were diluted (1:1) in Laemili buffer (Bio-Rad, Hercules, CA, USA) containing 5% mercaptoethanol and 2% sodium dodecyl sulfate (SDS). For immunoblotting, primary antibodies used were a polyclonal rabbit antiPECAM1 (1:200; Santa Cruz), a monoclonal mouse antithrombomodulin (1:200; Abcam), a monoclonal mouse anti-ICAM-1 (1:500; Abcam), a polyclonal rabbit antieNOS (1:200; Abcam), a polyclonal rabbit anti-EphB2 (1:200; Santa Cruz), a polyclonal goat anti-EphB4 (1:200; Santa Cruz), a polyclonal rabbit anti-Notch-1 (1:100; Santa Cruz), a polyclonal rabbit anti-nephrin (1:250; Abcam), and a monoclonal mouse anti-b-actin (1:2,000; SigmaAldrich). After multiple washes with TBS-Tween buffer, blots were incubated with a secondary goat anti-rabbit IgG HRP, goat anti-mouse IgG HRP antibodies, or a goat antidonkey IgG HRP antibody (1:2,000; Santa Cruz) for 1 h. The blots were developed using a Supersignal chemiluminescent substrate (Thermo Fisher Scientific, Rockford, IL, USA). Fresh rat aortic tissue homogenates and low passage of cultured rat aortic endothelial cell (RAECs) lysates were used as controls. mESC-EC lysates from passages 6–8 were used and compared to that of passage 16. To determine whether mESC-ECs deposit basement membrane proteins in culture, cells were cultured on glass slides for 5–7 days for immunohistological evaluation. Cells were fixed for 30 min at room temperature using 4% para-formaldehyde (Boston Bioproducts) and washed three times with PBS. Cells were then blocked with 10% horse serum in PBS with 0.1% Tween-20 for 30 min followed by an overnight incubation with primary antibodies at 4°C. Primary antibodies used were a polyclonal goat anti-laminin (1:200; Santa Cruz), a polyclonal rabbit anti-collagen type IV (1:200; Santa Cruz), and a polyclonal rabbit anti-fibronectin (1:200; Santa Cruz). Cells were then incubated with secondary donkey antigoat (1:200; Santa Cruz) and goat anti-rabbit (1:200; Santa Cruz) for 30 min. Cells were counterstained with hematoxylin (Boston Bioproducts) and mounted using Permount (Fisher Scientific, Hampton, NH, USA). Cells were imaged using an inverted microscope with a color camera (Nikon Eclipse Ti-S, Tokyo, Japan). To determine the expression of FLK1 by mESC-ECs, immunofluorescence analysis was performed. Cells were fixed for 30 min at room temperature using 4% paraformaldehyde (Boston Bioproducts) and washed three times with PBS. Cells were then blocked with 10% goat

99

serum in PBS with 0.1% Tween-20 for 1 h followed by an overnight incubation with a polyclonal rabbit anti-Flk-1 (1:1,000; Abcam) at 4°C and with secondary goat antimouse (1:2,000; Abcam) for 1 h. Cells were counterstained with DAPI to visualize the nuclei and were imaged using a confocal microscope (IX81 DSU; Olympus). Endothelial Nitric Oxide Synthase (eNOS) Expression by mESC-ECs eNOS enzyme-linked immunosorbent assay (ELISA) kit (R&D Systems, Minneapolis, MN, USA) was performed to quantify eNOS expression of mESC-ECs in their culture medium following the manufacturer’s instruction. Culture medium was collected from the culture of cells seeded at a density of 0.3 × 106 cells. Cells cultured in MCDB complete endothelial growth media containing 17.5 mM of glucose served as a control. The culture medium was then collected from cultures after 24 h of exposure in either 25 mM or 35 mM of glucose concentrations to investigate whether eNOS expression by mESC-ECs is dependent on glucose concentrations. The ELISA results were read using an Emax microplate reader (Molecular Devices, Sunnyvale, CA, USA) at the wavelength of 570 nm. Vascular Endothelial Growth Factor (VEGF) Expression by Cells During Pancreatic Differentiation Process VEGF concentration in the culture medium during the pancreatic differentiation process was quantified using a VEGF ELISA kit, following the manufacturer’s instruction (R&D Systems). Culture media were collected from pancreatic differentiation culture on days 20, 26, and 32. Initial cell plating density was approximately 65 × 105 cells/cm2. To determine whether cultured mESC-ECs also express VEGF, media was collected after 72 h of culture. Fresh differentiation medium was used as a control. Matrigel Assay A 0.8-mm-thick layer of Matrigel (BD Biosciences, San Jose, CA, USA) was prepared by adding Matrigel diluted in ice-cold DMEM (1:1) into a well of a 12-well plate, and 1.5 × 104 mESC-ECs were seeded onto the layer of Matrigel and cultured for up to 48 h. Cord formation by mESC-ECs was imaged using a microscope with a color camera (Nikon Eclipse Ti-S, Tokyo, Japan). Preparation of Sandwiched Collagen Gels Using mESC-ECs 3D collagen gels were created by a sandwich method, which contains two layers of cells prepared in between three layers of collagen. Rat tail collagen type I (BD Biosciences), 10× DMEM (Sigma-Aldrich), and 10× reconstitution buffer. The reconstitution buffer had the following composition: 0.05 N NaOH (Sigma-Aldrich) with 0.16 M HEPES (Sigma-Aldrich) and 0.25 M NaHCO3

100

(J. T. Baker, Center Valley, PA, USA) were mixed to form a collagen solution as described previously (37). To form 3D collagen gels, collagen solution was first poured into tissue culture plates and incubated at 37°C for 30 min to allow polymerization before mESC-ECs suspended in MCDB complete medium were added on top. After 3 h, the medium was carefully removed, and another layer of cold collagen gel solution was added on top of the cell layer. This process was repeated to obtain in total two layers of cells sandwiched between three layers of collagen gel. The medium was changed every other day. After 3–5 days in culture, collagen gels were fixed in 3.7% formalin. Fixed samples were paraffin embedded, and histological sections (10 µm thick) were prepared for immunofluorescent analysis. Once the sections were incubated with an antigen retrieval buffer (10 mM citric acid, pH = 6; Sigma Aldrich) at 95°C for 25 min, they were incubated with 10% goat serum in PBS for 1 h followed by incubation with a polyclonal rabbit anti-PECAM1 (1:200; Santa Cruz) at 4°C overnight. Sections were then incubated with a secondary donkey anti-rabbit (1:200; Santa Cruz) for 1 h at room temperature. Images were acquired using a confocal microscope (IX81 DSU; Olympus). Effect of VEGF Inhibitor on mESC-EC Yield To determine whether the presence of VEGF is one of the critical factors contributing to the appearance of mESC-ECs in pancreatic differentiation culture, VEGF

JAIN AND LEE

inhibitor was used. Thalidomide (Sigma-Aldrich), which is known as a potent VEGF inhibitor (33), was added to the differentiation culture on day 15 at a final concentration of either 10 µM or 20 µM in the culture medium and continued until the end of the differentiation (i.e., day 33). At the end of the differentiation, cells were incubated with LDL as previously described to detect endothelial cells and to determine the yield. Statistical Analysis Results are presented as mean ± SD. Statistical differences between groups were determined by single-factor analysis of variance (ANOVA) using SPSS software, and post hoc comparisons were based on Tukey’s test. Statistical significance was accepted for p < 0.05. RESULTS Identification of mESC-ECs Upon following a previously established protocol to derive insulin-producing cells from mESCs (50), we have successfully obtained insulin-producing cells as shown in Figure 1A. Insulin-producing cell clusters were identified after 33 days of differentiation by staining with DTZ, which is known to selectively stain insulin-producing cells (51). Insulin production by these cells was further confirmed by immunohistochemical analysis using an insulin antibody as shown in Figure 1B. Interestingly, in the vicinity of DTZ-positive cell clusters, we detected a population

Figure 1. Identification of mESC-ECs. (A) A brightfield image of an insulin-producing cell cluster stained with DTZ is shown. (B) An immunohistochemical image showing the expression of insulin on fixed DTZ-positive cluster at day 33 of differentiation. Scale bar: 100 µm. (C) Cells neighboring the DTZ-positive cell cluster (in the region indicated by a box inset in A) are LDL positive. Scale bar: 100 µm. (D) A timeline with important milestone marks for differentiation of mESCs into pancreatic b-cells. EBs formed by a hanging drop method were plated on day 6. A definitive endoderm marker, FoxA2, was first detected on day 15. LDL-positive cells started to appear on day 20. The mESC-ECs were isolated on day 33, which corresponds to the end time point of pancreatic differentiation.

ISLET ENDOTHELIAL CELLS FROM mESCs

of cells that uptake LDL (Fig. 1C). By examining the differentiation culture at various time points, we found that the LDL-positive cells start to appear as early as day 20 of differentiation. It was also found that the number of LDL-positive cells progressively increases, reaching approximately 6.3 ± 0.9% of the total number of cells on the final day of differentiation, day 33. Characterization of mESC-ECs Isolated mESC-ECs positive for LDL were expanded and cultured for up to 16 passages. Cultured mESCECs exhibited a cobblestone morphology and expressed FLk-1, a VEGF receptor as shown in Figure 2A. To determine whether these cells express an endothelial cell adhesion marker, PECAM1, a Western blot analysis was performed on cells at various stages of differentiation. It was found that while undifferentiated mESCs and cells differentiated for up to 15 days do not express PECAM1, only isolated and cultured mESC-ECs express PECAM1 (Fig. 2B). In order to determine the purity of the cultured cell population, a FACS analysis using PECAM1 was performed. Figure 2C shows that 99% of mESC-ECs express PECAM1, confirming that the isolated cell population is a pure endothelial cell

101

population. Moreover, immunohistochemistry results revealed that these cells deposit basement membrane proteins such as collagen type IV, laminin, and fibronectin in culture (Fig. 2D, E, and F, respectively), although a difference in the protein expression level at various culture times was not observed. Figure 3A demonstrates that cultured mESC-ECs express classical endothelial markers such as thrombomodulin, eNOS, and ICAM-1 in addition to PECAM1 and EphB2. Unlike RAECs that only express an arterial marker, Notch-1, mESC-ECs express both venous and arterial markers, EphB4 and Notch-1, respectively. More importantly, mESC-ECs express nephrin, which is known as one of the islet endothelial cell-specific markers. Consistent with mESC-ECs at passage 7–8, cells at passage 16 also express nephrin and another islet-specific marker, AAT, in 3D culture conditions, indicating that the cells do not undergo dedifferentiation even at passage numbers (Fig. 3B). To determine whether eNOS expression by mESCECs is glucose dependent, cells were subjected to different glucose concentrations. A negligible amount of eNOS was detected in the culture medium, which contains 17.5 mM glucose (Fig. 4A). However, when the cells were

Figure 2. Characterization of mESC-ECs. (A) A representative phase contrast image showing mESC-ECs (passage 7) exhibiting cobblestone morphology in culture (scale bar: 100 µm). Insets show immunofluorescent images displaying the expression of FLk-1 marker by mESC-ECs (left) and nuclei counterstained with DAPI (right) (scale bar: 20 µm). (B) Western blot analysis showing PECAM1 expression only by isolated mESC-ECs and not by undifferentiated mESCs or mESCs at earlier time points of differentiation, that is, before day 15. Rat aortic tissue lysate and primary RAECs served as positive controls, and b-actin served as a loading control. (C) A FACS histogram demonstrating a complete shift of peak for PECAM1-positive mESC-ECs (right) compared to isotype control (left). (D) Representative immunohistochemical micrographs demonstrating the deposition of collagen type IV, (E) fibronectin, and (F) laminin by cultured mESCECs. mESC-ECs were counterstained with hematoxylin for visualization of nuclei (scale bar: 50 µm).

102

JAIN AND LEE

Figure 3. Immunoblot analysis. (A) Classical endothelial markers such as PECAM1, thrombomodulin, ICAM-1, eNOS, EphB2, and both venous and arterial makers, EphB4 and Notch-1, respectively, as well as islet endothelial-specific marker, nephrin, was examined. RAEC lysate was used for comparison. b-Actin served as a loading control. (B) The expression of nephrin by mESC-ECs is maintained even at a higher passage number (P16). AAT is also expressed by mESC-ECs (P16) statically cultured in a 3D collagen gel for 3 days (SD3).

Figure 4. (A) Expression of eNOS expression by mESC-ECs exposed to different glucose levels (25 mM and 35 mM) in the medium. A significant increase in eNOS expression was observed when mESC-ECs were exposed to 35 mM glucose concentration compared to the control DMEM/F12 culture medium containing 17.5 mM glucose, demonstrating glucose-dependent eNOS expression (n = 5, *p < 0.05). (B) VEGF concentrations in culture medium measured at various time points (days 20, 26, and 32) during pancreatic differentiation. The amount of VEGF in culture was significantly higher on days 26 and 32 compared to that of fresh culture medium (n = 3, *p < 0.05 vs. culture medium). The VEGF concentration increased significantly with more days in differentiation culture was observed (#p < 0.05 vs. day 20, **p < 0.05 vs. day 26). No significant amount of VEGF was produced by cultured mESC-ECs.

ISLET ENDOTHELIAL CELLS FROM mESCs

exposed to the glucose concentration of 35 mM, eNOS expression was significantly upregulated compared to the control (n = 5, p < 0.05). Although there appeared to be an increasing trend of eNOS expression with increased glucose levels, the eNOS values at 25 mM and 35 mM were not statistically significant. As the number of mESC-ECs progressively increased with the time in differentiation culture, VEGF ELISA was performed to examine the presence of VEGF in the culture and to determine whether VEGF plays a role in the appearance of the endothelial cell. Culture medium collected at various time points during differentiation were examined, and fresh medium was used as a control. The VEGF ELISA results showed a significant increase in VEGF concentrations at days 26 and 32 of differentiation compared to that of fresh culture medium (p < 0.05 vs. culture medium) (Fig. 4B). It was also observed that the VEGF concentration is significantly higher in culture medium at days 26 and 32 compared to that of day 20 of differentiation (p < 0.05 vs. day 20). VEGF concentration continues to increase by day 32 compared to that of day 26 (p < 0.05 vs. day 26). The increase in VEGF expression in culture correlates to the increasing number of endothelial cells as the differentiation progresses. With the treatment of 10 µM or 20 µM of thalidomide, a VEGF inhibitor in the culture for 18 days, a significantly less number of LDL-positive endothelial cells at the end of the differentiation were detected as shown in Figure 5. While thalidomide did not completely inhibit endothelial cell appearance in the culture, it is evident that VEGF plays an important role in EC differentiation to form stem cells in our pancreatic differentiation culture. Matrigel Assay and 3D Sandwich Gels Using mESC-ECs When mESC-ECs were cultured on a layer of Matrigel, they were able to spontaneously organize into cord-like

103

structures that were maintained for up to 48 h as shown in Figure 6A. This in vitro Matrigel angiogenesis assay demonstrates the angiogenic capacity of mESC-ECs, consistent with known characteristics of endothelial cells (40). Furthermore, to examine mESC-EC behavior in a 3D environment, collagen gels were constructed using a sandwich method as shown in Figure 6B. Cells within each layer were spread out and formed networks with the neighboring cells after 5 days in culture, as demonstrated by F-actin staining (Fig. 6C, D). A representative immunofluorescence image further confirmed that the lumens were lined with PECAM1-positive mESC-ECs (Fig. 6E, F). Although connections between cells across two separate layers were not observed, cells within the layer formed lumen-like structures in the 3D collagen gels. This was apparent in cross-sections of gels stained with hematoxylin and eosin (H&E; Sigma-Aldrich) (Fig. 6G). Interestingly, it was found that while mESC-ECs in a traditional 2D culture do not express AAT, they express AAT when cultured in a 3D collagen gel environment (Fig. 6H). DISCUSSION In this study, we describe the identification, isolation, and characterization of a pure endothelial cell population exhibiting islet microvascular endothelial cell phenotype derived from mESCs. These cells appeared as a side product in mESC differentiation into pancreatic b-cell culture following a previously established protocol (50). This is a unique finding, as differentiation was not achieved through a conventional direct differentiation method, which uses successive maturation steps to derive endothelial cells. While endothelial cells are recognized as a highly heterogeneous population exhibiting different phenotypes and functions depending on the location of the endothelium within the body (4), tissue-specific endothelial cell populations, especially derived from pluripotent stem cells in vitro, have not been demonstrated. Thus, to

Figure 5. Effect of VEGF inhibitor on mESC-EC yield. Differentiation culture exposed to 20 µM thalidomide for 18 days resulted in a significantly less number of LDL-positive endothelial cells at the end of the differentiation, compared to the control samples without thalidomide (scale bar: 100 µm).

104

JAIN AND LEE

Figure 6. Angiogenic capacity of mESC-ECs. (A) mESC-ECs plated on Matrigel forming tube-like structures 12 h after plating (scale bar: 500 µm). (B) A schematic of a 3D sandwiched collagen gel depicting two layers of cells between three layers of collagen. (C) An immunofluorescence image of F-actin demonstrating formation of lumen-like structures in a sandwiched 3D collagen gel (scale bar: 100 µm). (D) DAPI counterstaining was performed for visualization of nuclei. (E) An immunofluorescence image confirming the presence of lumens lined by PECAM1-positive mESC-ECs, and (F) counterstained with DAPI for visualization of nuclei (scale bar: 50 µm). (G) A cross-sectional H&E micrograph demonstrating lumen-like structures in both cell layers (scale bar: 100 µm). (H) Western blot analysis showing no expression of AAT by mESC-ECs cultured in a 2D condition but only in 3D collagen gels after 3 and 7 days of static culture. b-Actin served as a loading control.

ISLET ENDOTHELIAL CELLS FROM mESCs

the best of our knowledge, this is the first report on stem cell-derived islet-specific endothelial cells. The isolated mESC-EC population incorporates Dio-ac-LDL, which is a hallmark of mature endothelial cells. In addition, they exhibit classical endothelialspecific markers such as PECAM1, eNOS, and thromobomodulin, which are essential to endothelial adhesion and vascular network formation even at high passage numbers. mESC-ECs also express EphB2, which is known to play an important role in sprouting angiogenesis (1,60). Unlike a previous study that showed expression of PECAM1 by mESCs (40), our results demonstrated that neither our mESCs nor cells during early differentiation periods express PECAM1. Only cells selectively isolated from the differentiation culture exhibited expression of PECAM1, indicating a pure population of isolated endothelial cells. Endothelial cells are generally considered to exhibit cobblestone morphology in culture; however, these cells can vary in thickness, aspect ratio, and orientation in vivo. In the present study, mESC-ECs initially had more of an elongated morphology when first detected near insulin-producing cells, which is consistent with previous reports on primary human and rat IECs (67). Upon culturing in endothelial cell medium, however, they progressed more toward a cobblestonelike morphology. As information on mouse IECs is currently lacking due to difficulty in deriving purified IECs, whether the morphological changes induced any functional changes needs further studies. However, detailed characterization of these cells was performed on cultured mESC-ECs with cobblestone morphology, and as the cells possessed properties of islet-specific endothelial cells, significant functional difference due to morphological changes was not expected. Pancreatic b-cells are unable to form a basement membrane in the absence of IECs (44). Instead, endothelial cells secrete proteins that are known to promote insulin-producing cell proliferation and insulin regulation (29,34,44). IEC-secreted collagen type IV is shown to potentiate insulin secretion via interaction with integrin a1b1 on b-cells (30). Likewise, laminin has been shown to upregulate insulin gene expression and secretion and induce b-cell differentiation and proliferation (26,44,59). Through immunohistochemical analysis, it was found that mESC-ECs in close proximity to insulin-positive cells deposit collagen type IV and laminin during differentiation culture (data not shown). Isolated mESC-ECs in culture also secrete ECM proteins such as fibronectin, laminin, and collagen type IV. This further confirms their endothelial cell phenotype and provides additional evidence for possible interactions of mESC-ECs with the pancreatic b-cell populations. It has previously been shown that VEGF plays an important role in recruitment of endothelial cells and formation

105

of IEC fenestrations (12,23,24,35). Our results demonstrate the presence of VEGF in the culture medium during differentiation, which progressively increases with more days in culture. This increase in concentration of VEGF coincided with the increased number of mESC-ECs present in the culture. As our populations of endothelial cells were detected in the vicinity of insulin-producing cells, it is plausible that the signals originating from endodermal pancreatic cells induced formation and growth of endothelial cell population in the differentiation culture. Since it was found that mESC-ECs do not produce significant amounts of VEGF in culture and pancreatic b-cells are known to constitutively secrete VEGF to recruit endothelial cells (44), it is likely that VEGF is mostly originated from insulin-producing b-cells in our culture. Moreover, as IECs are known to constitutively express VEGF-R1 (FLk-1) (13,18,58), enabling them to bind to VEGF, the expression of FLk-1 by our isolated mESC-ECs further supports the hypothesis that pancreatic b-cells play an important role in endothelial cell growth, although the exact mechanisms involved needs to be further elucidated. Furthermore, the addition of thalidomide substantially suppressed the formation of mESC-ECs in our culture (33), suggesting that VEGF is one of the key factors mediating the formation of mESC-ECs in the differentiation culture. In addition to classical endothelial cell phenotype, mESC-ECs exhibit distinct functional characteristics of IECs that are manifested by unique markers that can aid in their identification. mESC-ECs express nephrin, which is considered one of the specific IEC markers as shown by Zanone et al. (67). While nephrin is expressed by other cell types, including podocytes in kidneys (57), some parts of the central nervous system in mice (8,47), and by the Sertoli cells where it forms the blood–testis barrier (41), other tissue-specific endothelial cells have not been shown to express nephrin. The precise function of nephrin on IECs, however, remains unknown. Studies have also shown that AAT is another IEC-specific marker that is not expressed by endothelial cells in surrounding tissues (42). A study by Zhang et al. showed that AAT significantly reduces cytokine- and streptozotocin-induced pancreatic b-cell apoptosis (68). Additionally, the administration of clinical-grade AAT after islet transplantation has been shown to improve the graft survival in mice (39,54). While our mESC-ECs cultured in traditional 2D condition do not express AAT, it is important to note that mESC-ECs cultured in a 3D collagen gel environment express AAT. As endothelial cells may lose some of their phenotypical characteristics when maintained in a 2D culture condition, it is possible that mESC-ECs cultured in a 3D condition regained their AAT expression, which was originally downregulated from long-term 2D culture. This indicates the need for a physiological microenvironment to maintain phenotype and function of cells in vitro.

106

JAIN AND LEE

Moreover, IECs are known to produce a host of vasodilators including nitric oxide (NO) (25). NO is generated by NO synthase (NOS), and eNOS is the major NOS isoform predominantly expressed in endothelium. Both constitutive and cytokine-induced eNOS are present in pancreatic islets, although the role of constitutive eNOS in b-cell physiology has been controversial (55). Studies using islet cells and b-cell lines have reported that the constitutive eNOS either stimulates (15,36,49,52,53,62) or inhibits (5–7,19,21,22,48) the insulin release. Unlike other endothelium, constitutive and cytokine-inducible eNOS in islets is specifically upregulated depending on glucose levels (55). Thus, the fact that our mESC-ECs express eNOS, which is regulated by a higher glucose level in the culture medium, further demonstrates their behavior as IECs. However, eNOS expression was not significantly upregulated in culture medium containing 17.5 mM of glucose, which is yet much higher than physiological glucose levels of 4–6 mM for endothelial cells (32,56,61). Since mESC-ECs were cultured in medium containing 17.5 mM of glucose, it is plausible that cells became insensitive to that glucose level from being exposed for prolonged time. Thus, additional stimuli such as cytokines may be required to induce cells to produce eNOS at that level. Given the increasing interest in tissue-specific endothelial cell populations, the availability of an unlimited source of IECs derived from pluripotent stem cells provide a promising source of cells for both research and clinical use. Although primary endothelial cells are widely used in in vitro studies, stem cell-derived endothelial cells, especially organ-specific endothelial cells, provide the opportunity to further study their specific functions in in vitro conditions. Better understanding of mESC-EC function and their interaction with their neighboring pancreatic b-cells can lead to significant advancements in the development of therapeutic strategies for diabetes treatment. ACKNOWLEDGMENTS: We acknowledge the support of Dr. Rameshwar and Dr. Singh with FACS analysis and Dr. Cho for his fruitful suggestions. The authors declare no conflicts of interest.

REFERENCES 1. Adams, R. H.; Wilkinson, G. A.; Weiss, C.; Diella, F.; Gale, N. W.; Deutsch, U.; Risau, W.; Klein, R. Roles of Ephrin ligands and Eph receptors in cardiovascular development: Demarcation of arterial/venous domains, vascular morphogenesis, and sprouting angiogenesis. Genes Dev. 13(3):295–306; 1999. 2. Ahsan, T.; Nerem, R. M. Fluid shear stress promotes an endothelial-like phenotype during the early differentiation of embryonic stem cells. Tissue Eng. Part A 16(11):3547– 3553; 2010. 3. Aird, W. C. Phenotypic heterogeneity of the endothelium: I. Structure, function, and mechanisms. Circ. Res. 100(2):158–173; 2007.

4. Aird, W. C. Phenotypic heterogeneity of the endothelium: Ii. Representative vascular beds. Circ. Res. 100(2):174– 190; 2007. 5. Akesson, B.; Henningsson, R.; Salehi, A.; Lundquist, I. Islet constitutive nitric oxide synthase and glucose regulation of insulin release in mice. J. Endocrinol. 163(1):39–48; 1999. 6. Åkesson, B.; Lundquist, I. Influence of nitric oxide modulators on cholinergically stimulated hormone release from mouse islets. J. Physiol. 515(2):463–473; 1999. 7. Antoine, M.-H.; Ouedraogo, R.; Sergooris, J.; Hermann, M.; Herchuelz, A.; Lebrun, P. Hydroxylamine, a nitric oxide donor, inhibits insulin release and activates K+Atp channels. Eur. J. Pharmacol. 313(3):229–235; 1996. 8. Beltcheva, O.; Kontusaari, S.; Fetissov, S.; Putaala, H.; Kilpelainen, P.; Hokfelt, T.; Tryggvason, K. Alternatively used promoters and distinct elements direct tissue-specific expression of nephrin. J. Am. Soc. Nephrol. 14(2):352–358; 2003. 9. Blancas, A. A.; Lauer, N. E.; McCloskey, K. E. Endothelial differentiation of embryonic stem cells. Curr. Protoc. Stem Cell Biol. Chapter 1: Unit 1F 5; 2008. 10. Blancas, A. A.; Shih, A. J.; Lauer, N. E.; McCloskey, K. E. Endothelial cells from embryonic stem cells in a chemically defined medium. Stem Cells Dev. 20(12):2153–2161; 2011. 11. Brissova, M.; Fowler, M.; Wiebe, P.; Shostak, A.; Shiota, M.; Radhika, A.; Lin, P. C.; Gannon, M.; Powers, A. C. Intraislet endothelial cells contribute to revascularization of transplanted pancreatic islets. Diabetes 53(5):1318–1325; 2004. 12. Brissova, M.; Shostak, A.; Shiota, M.; Wiebe, P. O.; Poffenberger, G.; Kantz, J.; Chen, Z.; Carr, C.; Jerome, W. G.; Chen, J.; Baldwin, H. S.; Nicholson, W.; Bader, D. M.; Jetton, T.; Gannon, M.; Powers, A. C. Pancreatic islet production of vascular endothelial growth factor-a is essential for islet vascularization, revascularization, and function. Diabetes 55(11):2974–2985; 2006. 13. Christofori, G.; Naik, P.; Hanahan, D. Vascular endothelial growth factor and its receptors, Flt-1 and Flk-1, are expressed in normal pancreatic islets and throughout islet cell tumorigenesis. Mol. Endocrinol. 9(12):1760–1770; 1995. 14. Crawford, L. A.; Guney, M. A.; Oh, Y. A.; Deyoung, R. A.; Valenzuela, D. M.; Murphy, A. J.; Yancopoulos, G. D.; Lyons, K. M.; Brigstock, D. R.; Economides, A.; Gannon, M. Connective tissue growth factor (CTGF) inactivation leads to defects in islet cell lineage allocation and betacell proliferation during embryogenesis. Mol. Endocrinol. 23(3):324–336; 2009. 15. Ding, Y.; Rana, R. S. Nitric oxide does not initiate but potentiates glucose-induced insulin secretion in pancreatic beta-cells. Biochem. Biophys. Res. Commun. 251(3):699– 703; 1998. 16. Fioretta, E. S.; Fledderus, J. O.; Baaijens, F. P.; Bouten, C. V. Influence of substrate stiffness on circulating progenitor cell fate. J. Biomech. 45(5):736–744; 2012. 17. Glaser, D. E.; Gower, R. M.; Lauer, N. E.; Tam, K.; Blancas, A. A.; Shih, A. J.; Simon, S. I.; McCloskey, K. E. Functional characterization of embryonic stem cell-derived endothelial cells. J. Vasc. Res. 48(5):415–428; 2011. 18. Gorden, D. L.; Mandriota, S. J.; Montesano, R.; Orci, L.; Pepper, M. S. Vascular endothelial growth factor is increased in devascularized rat islets of langerhans in vitro. Transplantation 63(3):436–443; 1997. 19. Gross, R.; Roye, M.; Manteghetti, M.; Hillaire-Buys, D.; Ribes, G. Alterations of insulin response to different

ISLET ENDOTHELIAL CELLS FROM mESCs

20. 21.

22.

23.

24.

25. 26. 27. 28.

29.

30.

31.

32. 33.

34.

beta cell secretagogues and pancreatic vascular resistance induced by N omega-nitro-L-arginine methyl ester. Br. J. Pharmacol. 116(3):1965–1972; 1995. Henderson, J. R.; Moss, M. C. A morphometric study of the endocrine and exocrine capillaries of the pancreas. Q. J. Exp. Physiol. 70(3):347–356; 1985. Henningsson, R.; Alm, P.; Ekström, P.; Lundquist, I. Hemeoxygenase and carbon monoxide: Regulatory roles in islet hormone release: A biochemical, immunohistochemical, and confocal microscopic study. Diabetes 48(1):66–76; 1999. Henningsson, R.; Alm, P.; Lindström, E.; Lundquist, I. Chronic blockade of no synthase paradoxically increases islet no production and modulates islet hormone release. Am. J. Physiol. Endocrinol. Metab. 279(1):E95–E107; 2000. Iwashita, N.; Uchida, T.; Choi, J. B.; Azuma, K.; Ogihara, T.; Ferrara, N.; Gerber, H.; Kawamori, R.; Inoue, M.; Watada, H. Impaired insulin secretion in vivo but enhanced insulin secretion from isolated islets in pancreatic beta cellspecific vascular endothelial growth factor-a knock-out mice. Diabetologia 50(2):380–389; 2007. Jabs, N.; Franklin, I.; Brenner, M. B.; Gromada, J.; Ferrara, N.; Wollheim, C. B.; Lammert, E. Reduced insulin secretion and content in VEGF-a deficient mouse pancreatic islets. Exp. Clin. Endocrinol. Diabetes 116 Suppl. 1:S46–49; 2008. Jansson, L.; Carlsson, P. O. Graft vascular function after transplantation of pancreatic islets. Diabetologia 45(6):749– 763; 2002. Jiang, F. X.; Cram, D. S.; DeAizpurua, H. J.; Harrison, L. C. Laminin-1 promotes differentiation of fetal mouse pancreatic beta-cells. Diabetes 48(4):722–730; 1999. Johansson, M.; Andersson, A.; Carlsson, P.-O.; Jansson, L. Perinatal development of the pancreatic islet microvasculature in rats. J. Anat. 208(2):191–196; 2006. Johansson, M.; Mattsson, G.; Andersson, A.; Jansson, L.; Carlsson, P. O. Islet endothelial cells and pancreatic betacell proliferation: Studies in vitro and during pregnancy in adult rats. Endocrinology 147(5):2315–2324; 2006. Kaido, T.; Yebra, M.; Cirulli, V.; Montgomery, A. M. Regulation of human beta-cell adhesion, motility, and insulin secretion by collagen IV and its receptor alpha1beta1. J. Biol. Chem. 279(51):53762–53769; 2004. Kaido, T.; Yebra, M.; Cirulli, V.; Rhodes, C.; Diaferia, G.; Montgomery, A. M. Impact of defined matrix interactions on insulin production by cultured human beta-cells: Effect on insulin content, secretion, and gene transcription. Diabetes 55(10):2723–2729; 2006. Kattman, S. J.; Huber, T. L.; Keller, G. M. Multipotent Flk-1+ cardiovascular progenitor cells give rise to the cardiomyocyte, endothelial, and vascular smooth muscle lineages. Dev. Cell 11(5):723–732; 2006. Kim, J. A.; Berliner, J. A.; Natarajan, R. D.; Nadler, J. L. Evidence that glucose increases monocyte binding to human aortic endothelial cells. Diabetes 43(9):1103–1107; 1994. Komorowski, J.; Jerczyńska, H.; Siejka, A.; Barańska, P.; Ławnicka, H.; Pawłowska, Z.; Stępień, H. Effect of thalidomide affecting VEGFsecretion, cell migration, adhesion and capillary tube formation of human endothelial Ea.Hy 926 cells. Life Sci. 78(22):2558–2563; 2006. Konstantinova, I.; Nikolova, G.; Ohara-Imaizumi, M.; Meda, P.; Kucera, T.; Zarbalis, K.; Wurst, W.; Nagamatsu, S.; Lammert, E. Epha-ephrin-a-mediated beta cell communication regulates insulin secretion from pancreatic islets. Cell 129(2):359–370; 2007.

107

35. Lammert, E.; Gu, G.; McLaughlin, M.; Brown, D.; Brekken, R.; Murtaugh, L. C.; Gerber, H. P.; Ferrara, N.; Melton, D. A. Role of VEGF-a in vascularization of pancreatic islets. Curr. Biol. 13(12):1070–1074; 2003. 36. Laychock, S. G.; Modica, M. E.; Cavanaugh, C. T. L-arginine stimulates cyclic guanosine 3¢,5¢-monophosphate formation in rat islets of langerhans and Rinm5f insulinoma cells: Evidence for L-arginine:nitric oxide synthase. Endocrinology 129(6):3043–3052; 1991. 37. Lee, E. J.; Niklason, L. E. A novel flow bioreactor for in vitro microvascularization. Tissue Eng. Part C Methods 16(5):1191–1200; 2010. 38. Levenberg, S.; Golub, J. S.; Amit, M.; Itskovitz-Eldor, J.; Langer, R. Endothelial cells derived from human embryonic stem cells. Proc. Natl. Acad. Sci. USA 99(7):4391–4396; 2002. 39. Lewis, E. C.; Mizrahi, M.; Toledano, M.; DeFelice, N.; Wright, J. L.; Churg, A.; Shapiro, L.; Dinarello, C. A. A1-antitrypsin monotherapy induces immune tolerance during islet allograft transplantation in mice. Proc. Natl. Acad. Sci.USA 105(42):16236–16241; 2008. 40. Li, Z.; Wu, J. C.; Sheikh, A. Y.; Kraft, D.; Cao, F.; Xie, X.; Patel, M.; Gambhir, S. S.; Robbins, R. C.; Cooke, J. P.; Wu, J. C. Differentiation, survival, and function of embryonic stem cell–derived endothelial cells for ischemic heart disease. Circulation 116(11 Suppl):I-46–I-54; 2007. 41. Liu, L.; Aya, K.; Tanaka, H.; Shimizu, J.; Ito, S.; Seino, Y. Nephrinis an important component of the barrier system in the testis. Acta Med. Okayama 55(3):161–165; 2001. 42. Lou, J.; Triponez, F.; Oberholzer, J.; Wang, H.; Yu, D.; Buhler, L.; Cretin, N.; Mentha, G.; Wollheim, C. B.; Morel, P. Expression of alpha-1 proteinase inhibitor in human islet microvascular endothelial cells. Diabetes 48(9):1773–1778; 1999. 43. Mattsson, G. The endothelial cells in islets of Langerhans. Ups. J. Med. Sci. 110(1):1–15; 2005. 44. Nikolova, G.; Jabs, N.; Konstantinova, I.; Domogatskaya, A.; Tryggvason, K.; Sorokin, L.; Fassler, R.; Gu, G.; Gerber, H. P.; Ferrara, N.; Melton, D. A.; Lammert, E. The vascular basement membrane: A niche for insulin gene expression and beta cell proliferation. Dev. Cell 10(3):397–405; 2006. 45. Nolan, D. J.; Ginsberg, M.; Israely, E.; Palikuqi, B.; Poulos, M. G.; James, D.; Ding, B.-S.; Schachterle, W.; Liu, Y.; Rosenwaks, Z.; Butler, J. M.; Xiang, J.; Rafii, A.; Shido, K.; Rabbany, S. Y.; Elemento, O.; Rafii, S. Molecular signatures of tissue-specific microvascular endothelial cell heterogeneity in organ maintenance and regeneration. Dev. Cell 26(2):204–219; 2013. 46. Olsson, R.; Carlsson, P. O. The pancreatic islet endothelial cell: Emerging roles in islet function and disease. Int. J. Biochem. Cell Biol. 38(5–6):710–714; 2006. 47. Putaala, H.; Soininen, R.; Kilpelainen, P.; Wartiovaara, J.; Tryggvason, K. The murine nephrin gene is specifically expressed in kidney, brain and pancreas: Inactivation of the gene leads to massive proteinuria and neonatal death. Hum. Mol. Genet. 10(1):1–8; 2001. 48. Salehi, A.; Parandeh, F.; Lundquist, I. Signal transduction in islet hormone release: Interaction of nitric oxide with basal and nutrient-induced hormone responses. Cell. Signal. 10(9):645–651; 1998. 49. Schmidt, H. H.; Warner, T. D.; Ishii, K.; Sheng, H.; Murad, F. Insulin secretion from pancreatic B cells caused by L-argininederived nitrogen oxides. Science 255(5045):721–723; 1992.

108

50. Schroeder, I. S.; Rolletschek, A.; Blyszczuk, P.; Kania, G.; Wobus, A. M. Differentiation of mouse embryonic stem cells to insulin-producing cells. Nat. Protoc. 1(2):495–507; 2006. 51. Shiroi, A.; Yoshikawa, M.; Yokota, H.; Fukui, H.; Ishizaka, S.; Tatsumi, K.; Takahashi, Y. Identification of insulinproducing cells derived from embryonic stem cells by zincchelating dithizone. Stem Cells 20(4):284–292; 2002. 52. Spinas, G.; Laffranchi, R.; Francoys, I.; David, I.; Richter, C.; Reinecke, M. The early phase of glucose-stimulated insulin secretion requires nitric oxide. Diabetologia 41(3):292–299; 1998. 53. Spinas, G. A. The dual role of nitric oxide in islet b-cells. Physiology 14(2):49–54; 1999. 54. Strom, T. B. Saving islets from allograft rejection. Proc. Natl. Acad. Sci. USA 102(36):12651–12652; 2005. 55. Suschek, C.; Fehsel, K.; Kroncke, K. D.; Sommer, A.; KolbBachofen, V. Primary cultures of rat islet capillary endothelial cells. Constitutive and cytokine-inducible macrophagelike nitric oxide synthases are expressed and activities regulated by glucose concentration. Am. J. Pathol. 145(3):685–695; 1994. 56. Talavera-Adame, D.; Wu, G.; He, Y.; Ng, T. T.; Gupta, A.; Kurtovic, S.; Hwang, J. Y.; Farkas, D. L.; Dafoe, D. C. Endothelial cells in co-culture enhance embryonic stem cell differentiation to pancreatic progenitors and insulinproducing cells through Bmp signaling. Stem Cell Rev. 7(3):532–543; 2011. 57. Tryggvason, K.; Wartiovaara, J. Molecular basis of glomerular permselectivity. Curr. Opin. Nephrol. Hypertens. 10(4):543–549; 2001. 58. Vasir, B.; Aiello, L. P.; Yoon, K. H.; Quickel, R. R.; BonnerWeir, S.; Weir, G. C. Hypoxia induces vascular endothelial growth factor gene and protein expression in cultured rat islet cells. Diabetes 47(12):1894–1903; 1998. 59. Virtanen, I.; Banerjee, M.; Palgi, J.; Korsgren, O.; Lukinius, A.; Thornell, L. E.; Kikkawa, Y.; Sekiguchi, K.; Hukkanen, M.; Konttinen, Y. T.; Otonkoski, T. Blood vessels of human islets of langerhans are surrounded by a double basement membrane. Diabetologia 51(7):1181–1191; 2008. 60. Wang, H. U.; Chen, Z. F.; Anderson, D. J. Molecular distinction and angiogenic interaction between embryonic

JAIN AND LEE

61. 62.

63.

64.

65.

66. 67.

68.

69.

arteries and veins revealed by Ephrin-B2 and its receptor Eph-B4. Cell 93(5):741–753; 1998. Wang, X. L.; Fu, A.; Spiro, C.; Lee, H. C. Proteomic analysis of vascular endothelial cells-effects of laminar shear stress and high glucose. J. Proteomics Bioinform. 2:445; 2009. Willmott, N. J.; Galione, A.; Smith, P. A. Nitric oxide induces intracellular Ca 2+mobilization and increases secretion of incorporated 5-hydroxytryptamine in rat pancreatic beta cells. FEBS. Lett. 371(2):99–104; 1995. Yamamoto, K.; Sokabe, T.; Watabe, T.; Miyazono, K.; Yamashita, J. K.; Obi, S.; Ohura, N.; Matsushita, A.; Kamiya, A.; Ando, J. Fluid shear stress induces differentiation of Flk-1-positive embryonic stem cells into vascular endothelial cells in vitro. Am. J. Physiol. Heart Circ. Physiol. 288(4):H1915–1924; 2005. Yang, L.; Soonpaa, M. H.; Adler, E. D.; Roepke, T. K.; Kattman, S. J.; Kennedy, M.; Henckaerts, E.; Bonham, K.; Abbott, G. W.; Linden, R. M.; Field, L. J.; Keller, G. M. Human Cardiovascular progenitor cells develop from a Kdr+ embryonic-stem-cell-derived population. Nature 453(7194):524–528; 2008. Yoshitomi, H.; Zaret, K. S. Endothelial cell interactions initiate dorsal pancreas development by selectively inducing the transcription factor Ptf1a. Development 131(4):807– 817; 2004. Zanone, M. M.; Favaro, E.; Camussi, G. From endothelial to beta cells: Insights into pancreatic islet microendothelium. Curr. Diabetes Rev. 4(1):1–9; 2008. Zanone, M. M.; Favaro, E.; Doublier, S.; LozanoskaOchser, B.; Deregibus, M. C.; Greening, J.; Huang, G. C.; Klein, N.; CavalloPerin, P.; Peakman, M.; Camussi, G. Expression of nephrin by human pancreatic islet endothelial cells. Diabetologia 48(9):1789–1797; 2005. Zhang, B.; Lu, Y.; Campbell-Thompson, M.; Spencer, T.; Wasserfall, C.; Atkinson, M.; Song, S. A1-antitrypsin protects b-cells from apoptosis. Diabetes 56(5):1316–1323; 2007. Zhang, S.; Sun, A.; Ma, H.; Yao, K.; Zhou, N.; Shen, L.; Zhang, C.; Zou, Y.; Ge, J. Infarcted myocardium-like stiffness contributes to endothelial progenitor lineage commitment of bone marrow mononuclear cells. J. Cell. Mol. Med. 15(10):2245–2261; 2011.