Journal Logo

Basic and Experimental Research

Trophic Molecules Derived From Human Mesenchymal Stem Cells Enhance Survival, Function, and Angiogenesis of Isolated Islets After Transplantation

Park, Ki-Soo1,2; Kim, Young-Seok1,3; Kim, Jae-Hyeon3; Choi, Bongkum1,2; Kim, Sa-Hyun2; Tan, Alice Hyun-Kyung3; Lee, Myung-Shik3; Lee, Moon-Kyu3; Kwon, Choon-Hyuck4; Joh, Jae-Won4; Kim, Sung-Joo2,4,5; Kim, Kwang-Won3,5

Author Information
doi: 10.1097/TP.0b013e3181c7dc99



In the March 2010 issue of Transplantation, the article by Park et al (Trophic molecules derived from human mesenchymal stem cells enhance survival, function, and angiogenesis of isolated islets after transplantation. Transplantation 2010;89(5):509–517), was published with an incorrect DOI. The correct DOI is: 10.1097/

Transplantation. 101(8):e290, August 2017.

Islet death, including β-cell loss during the culture period and after transplantation, remains a significant obstacle in successful islet transplantation (ITx). The primary reasons for the observed cell death include apoptosis and reduced β-cell function as a result of hypoxic stress and lack of nutrients during the isolation process and the culture period (1–5). Therefore, many strategies for improving islet culture conditions as a means of preventing islet death and enhancing the function of isolated islets have been tested (6–10). However, most of these studies have not evaluated the effect on islets in vivo. Furthermore, the signaling cascade of molecules and appropriate conditions for using feeder cells in the generation of a trophic effect remain poorly understood (9, 10).

Mesenchymal stem cells (MSCs) have been recognized as a useful source of cells for tissue or organ regeneration and immune modulation (11, 12). Recently, the trophic effects related to MSC-secreted bioactive molecules were elucidated (13). Moreover, Yarmush and coworkers (14) propose that the molecules produced by MSCs can directly modulate hepatocyte death and regeneration in vitro and in vivo. In various disease models, including radiation burn injuries, myocardial infarction, and Parkinson's disease, the recovery of damaged tissue or protective properties of MSC-secreted molecules has also been reported (15–17). Although the potential trophic effect of MSCs on transplanted islets is not entirely clear, several MSC-secreted factors, including interleukin (IL)-6, vascular endothelial growth factor (VEGF), and hepatocyte growth factor (HGF), have demonstrated enhanced islet engraftment after transplantation (18–20).

A recent report by Urbán et al. (21) describes MSCs promote the regeneration of recipient-derived pancreatic insulin-secreting cells by tail vein injection in streptozotocin (STZ)-induced diabetic mice. The study also suggests that the MSC-induced reversible effect is caused by a paracrine effect originating from the MSCs. These results suggest that pancreatic islet quality may be modulated by trophic factors from MSCs. We tested the hypothesis that molecules secreted by MSCs induce a trophic effect on isolated islets during in vitro culture and that MSCs regulate an islet intracellular signaling cascade. We developed an islet culture system with MSCs and MSC-conditioned medium (CM) to evaluate the potential for MSC-derived molecules to enhance the isolated mouse cell viability and function in vitro and to investigate the effects on apoptosis and angiogenesis-related signal molecules, which may be linked to the MSC trophic effects. We also examined whether islet quality after transplantation was enhanced through angiogenesis and assessed the improvement of islet function after culture with MSC-CM.


Pancreatic Islet Isolation

Male inbred syngeneic C57BL/6 mice, aged 9 to 10 weeks, were purchased from Orient-Bio Laboratory (Seongnam, Korea). Islets were isolated with an intraductal injection of 0.8 mg/mL collagenase P (Roche Applied Science, Indianapolis, IN) in Hank's-buffered saline solution purified by Bicoll (Biochrom; Berlin, Germany) gradient. To control for the trophic effect of MSCs, isolated islets were cultured while free floating in minimal islet culture medium (M199, Cellgro Mediatech Inc., Herndon, VA) with 10% fetal bovine serum (Gibco, Grand Island, NY).

Generation of MSCs From Human Cord Blood

Our umbilical cord blood collection protocol was approved by the Institutional Review Board of Samsung Medical Center. Written informed consent was obtained from healthy pregnant women who delivered without complications. MSCs were isolated according to methods described by Bieback et al. (22). Surface phenotypes were analyzed with a FACS Vantage cell sorter (Becton Dickinson, Mountain View, CA).

Islet-MSC Co-Culture and Islet Culture Using MSC-CM

MSCs (cultured seven to nine passages for most experiments and not exceeding 12 passages) were seeded at 3×104 cells/well in 24-well culture plates in growth medium (Dulbecco's minimum essential medium-low glucose with the addition of 10% fetal calf serum with 2 mmol/L l-glutamine and 1% penicillin-streptomycin [Gibco]) and incubated in a 5%-CO2 humidified atmosphere at 37°C. After overnight incubation, these cells were washed twice with phosphate-buffered saline. The medium was then changed to islet culture medium. Isolated mouse islets were inoculated into 12.0-μm pore culture-insert (Millipore, Bedford, MA) and placed in MSC-preseeded 24-well culture plates. In the control group, islets were also inoculated into the culture-insert; however, MSCs were not preseeded (Fig. 1A). MSC-CM for islet culture was obtained by the following procedure. MSCs were seeded at 5×105 cells per T-75 cm2 culture plates in growth medium. After overnight incubation, the medium was changed to islet culture medium (Fig. 1B). Then, MSC-CM was collected at 24, 48, and 72 hr after medium change (referred to MSC-CM 1, 2, and 3, respectively).

Experimental design for islet culture conditions and in vivo islet transplantation. (A) In the mesenchymal stem cell (MSC)-islet co-culture system, 3×104 MSCs were seeded in 24-well plates with MSC growth medium overnight, and the isolated islets were then co- cultured with MSCs using a culture insert (to inhibit cell-to-cell contact) in islet culture medium for 48 hr. (B) For the MSC-conditioned medium (CM) islet culture, 5×105 MSCs were placed in culture plates containing MSC growth medium. After overnight incubation, cells were washed twice with phosphate-buffered saline, and the medium was switched to M199 with the addition of 10% fetal calf serum and 1% penicillin-streptomycin. MSC-CM was prepared by harvest and filtration after 48 hr of medium change. (C) Islets cultured with MSC-CM for 48 hr were collected and transplanted to the renal capsule of streptozotocin-induced diabetic mice.

In Vitro Assessments of the Viability, ADP/ATP Ratio, and Function of Islets

To assess the islet viability, acridine orange (AO; 0.67 μmol/L) and propidium iodide (PI; 75 μmol/L) (AO/PI) staining was used to simultaneously observe living and nonviable islet. To confirm islet viability after the co-culture with MSCs, or MSC-CM culture, a DNA fragmentation-based ELISA kit (Roche Applied Science) measuring the amount of fragmented and solubilized nucleosomal DNA was used to evaluate apoptosis (23). Each sample was analyzed with 100-islet equivalent (IEq) hand-picked islets (n=6).

To assess the insulin secretory function of islets after co-culture with MSCs or MSC-CM culture, the 100-IEq hand-picked islets (n=5) were washed twice and incubated at 37°C for 1 hr in Krebs-Ringer bicarbonate buffer solution containing low (60 mg/dL, basal) and high (300 mg/dL, stimulated) glucose concentrations. The supernatants were then collected and analyzed by measuring the amount of secreted insulin with a Rat/Mouse Insulin ELISA kit (Millipore).

The ADP/ATP ratio of islets was assessed using a commercially manufactured kit (Lonza, Rockland, ME). One hundred-IEq hand-picked islets (n=5) were washed twice and mixed with 100 μL of nucleotide-releasing reagent for 10 min at room temperature. Thereafter, 20 μL of nucleotide-monitoring reagent was added, and the ATP levels were measured using a luminometer (Thunder Designs, Sunnyvale, CA) and expressed as the number of relative light units. After 10 min, the ATP was converted to ADP by adding 20 μL ADP converting reagent and then measured. Subsequently, the ADP/ATP ratio of the islet was calculated.

Western Blot Analysis

Mouse islet pellets after co-culture were lysed in a buffer containing 60 μmol/L Tris-HCl (pH 6.8), 1% sodium dodecyl sulfate, 10% glycerol, 0.5% β-mercaptoethanol, 0.05% NP-40, and a protease inhibitor mixture (1:100 dilutions). Equivalent amounts of protein from islets of each group were run on 6% to 12% sodium dodecyl sulfate polyacrylamide gels. Proteins were electrically transferred to nitrocellulose filters and incubated with primary antibodies. Mouse anti-X-linked inhibitor of apoptosis protein (XIAP; 1:500) and mouse anti-heat shock protein (HSP)-32 (1:2000) antibodies were purchased from BD Transduction Laboratories (San Diego, CA), rabbit anti-Ki-67 (1:1000) antibodies were purchased from Abcam (Cambridge, UK), and goat anti-pancreatic and duodenal homeobox-1 (PDX-1) (1:500) antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Rabbit antibodies for Bcl-xL (1:1000), anti-Bcl-2 (1:1000), anti-Akt (1:1000), anti-Phospho-Akt (Ser473) (1:1000), anti-p44/42 (1:1000), anti-phospho-p44/42 (Thr202/Tyr204) (1:1000), anti-phospholipase c (PLC)-γ (1:1000), anti-phospho-PLC-γ (Ser1248; 1:1000), anti-focal adhesion kinase (FAK; 1:1000), anti-Phospho-FAK (Tyr397) (1:1000), anti-phospho-Tie-2 (Tyr992) (1:1000), anti-β-actin (1:2000), and mouse anti-Tie-2 (1:1000) antibodies were purchased from Cell Signaling Technology (Beverly, MA). Secondary antibodies were incubated with anti-mouse, anti-rabbit, or anti-goat IgG conjugated with horseradish peroxidase (1:2500).

Reverse Transcription Polymerase Chain Reaction

Total RNA was extracted from mouse islets co-cultured with MSCs and from mouse colon adenocarcinoma cell lines (MC-38) or mouse bone marrow-derived MSCs as a positive control using an RNeasy Mini Kit (QIAGEN; Hilden, Germany). Primers were designed according to a GenBank search synthesized by Bioneer Corporation (Cheongwon, Korea). Reverse transcription polymerase chain reaction was performed according to the manufacturer's instructions (Invitrogen, Carlsbad, CA). Each pair of primer sequences, annealing temperatures, and amplification cycles are described in a supplemental Table (see Supplemental Digital Content 1,

Islet Transplantation

All procedures were conducted at the Samsung Biomedical Research Institute, Laboratory Animal Research Center (Seoul, Korea) in an Association for Assessment and Accreditation of Laboratory Animal Care-approved facility (No. 001003) following an Institutional Animal Care and Use Committee-approved protocol. Male syngeneic C57BL/6 mice, aged 9 to 10 weeks, were used as donors and transplantation recipients. Diabetes (defined by a blood glucose >400 mg/dL) was produced by a single intraperitoneal injection of STZ (180 mg/kg; Sigma, St. Louis, MO) 2 days before transplantation. Recipient mice were classified into two groups: ITx with standard-cultured islets (control; n=6) and ITx with MSC-CM2–cultured islets (MSCs; n=6). Islets were cultured with MSC-CM2 for 48 hr before ITx. Two hundred IEq islets, an intentionally marginal number of islets (approximately half the amount needed for full glycemic control), were hand picked for transplantation. Animals were anesthetized with an intraperitoneal injection of ketamine hydrochloride (100 mg/kg) and Rompun xylazine (20 mg/kg). Islets were implanted under the renal capsule of the left kidney in recipient mice. Body weight and blood glucose levels were monitored to assess islet graft function.

Intraperitoneal Glucose Tolerance Test

The glucose tolerance of transplanted islets was evaluated 2 weeks after ITx. Animals were fasted for approximately 14 hr and intraperitoneally injected with a 20% glucose solution (1 mg/kg). Blood glucose levels were then monitored at 0, 30, 60, 90, and 120-min intervals by tail snipping.


We analyzed the MSC-cultured medium to identify the soluble factors secreted by MSCs. MSCs were seeded according to the earlier described co-culture conditions. After overnight incubation, the growth medium was changed to M199. After a 48-hr incubation period, supernatants were collected (n=3). The levels of soluble factors (insulin, insulin growth factor [IGF]-1, IL-6, VEGF-A, HGF, transforming growth factor [TGF]-β, fibroblast growth factor [FGF]-2, stem- cell factor [SCF], tumor necrosis factor [TNF]-α, and interferon [IFN]-γ) were analyzed using ELISA (R&D Systems; Minneapolis, MN). Background levels for all factors in the normal growth medium were determined and subtracted from MSC-cultured medium samples.

Immunohistochemical Staining

Two weeks after transplantation, islet grafts from the kidney and residual pancreata were retrieved, fixed, and then paraffin embedded. Consecutive sections (6- and 4-μm thick) of islet grafts and pancreas were subjected to a heat-based antigen retrieval process with citric buffer (pH 6.0) by microwave (three times at approximately 90°C for 5 min). Retrieved sections were washed with tris-buffered saline Twen-20 buffer (pH 7.6) and blocked with 10% normal goat serum in tris-buffered saline buffer for 1 hr. Subsequently, these sections were incubated with guinea pig anti-insulin (1:500) at 4°C for overnight. Signal amplification and visualization were accomplished by incubation with an avidin-biotin peroxidase system (DAKO Corporation, Carpinteria, CA). Blood vessels of retrieved islet grafts were stained with lectin I (1:2000, Fluorescein Griffonia Simplicifolia Lectin I; Vector Laboratories, Burlingame, CA) for 2 hr at room temperature (24). Images of five randomly selected slides were captured from each tissue. To evaluate remnant islet in residual pancreata and vessel formation in transplanted islet grafts, images of the insulin-stained-to-total pancreas area and lectin I-stained-to-total graft area were acquired by fluorescence microscope (Axiophot, Zeiss, Oberkochen, Germany) and analyzed by the 50 or 200× magnification using Image-Pro Plus version 5.0 (Media Cybernetics; Bethesda, MD).

Statistical Analysis

Values are expressed as mean±SD, with statistical significance calculated using the Student's t test. A P value of less than 0.05 was considered statistically significant.


Characterization of MSCs From Umbilical Cord Blood

MSCs were purified as the cells adhered to plastic culture flasks and then characterized for surface protein expression. Approximately seven to eight passages of CB-MSCs were analyzed to define each phenotype. MSCs expressed CD13, CD29, CD44, CD73, CD90, CD105, and HLA-ABC, but they were negative for CD14, CD34, CD38, CD45, CD117, and HLA-DR (data not shown).

Quality Assessment of Islets Co-Cultured With MSCs

To identify the effects of islet-MSC co-culture, isolated mouse islets were co-cultured with human MSCs and then assessed for islet viability, ADP/ATP ratio, and insulin secretory function. The viability of islets co-cultured with MSCs was investigated during the 7 days of co-culture. Control islets cultured in the standard culture medium demonstrated a higher ratio of nonviable cells that were evenly distributed within the islets, with no apparent localization (e.g., β cell), as noted by AO/PI staining (Fig. 2A, upper; see Figure, Supplemental Digital Content 2, However, islets co-cultured with MSCs demonstrated fewer nonviable cells (Fig. 2A, lower), reduced DNA fragmentation (0.52±0.1-fold change MSCs vs. control, P<0.05, Fig. 2B), and lower ADP/ATP ratio than control islets after 48 hr co-culture (MSCs vs. control, 0.04±0.01 vs. 0.10±0.01, P<0.05, Fig. 2C). β-cell function was assessed by the glucose stimulated insulin secretion test, which demonstrated increases in glucose-stimulated insulin secretion (MSCs vs. control, 43.45±4.30 ng/mL vs. 24.24±4.26 at 300 mg/dL glucose, P<0.05, Fig. 2D) and higher secretion indexes (MSCs vs. control, 4.08±0.48 vs. 2.57±0.36) for islets co-cultured with MSCs for 48 hr.

Viability, ADP/ATP ratio and insulin secretory function of co-cultured islets. (A) The viability of mouse islet cells was assessed with acridine orange and propidium iodide staining. Green indicates living cells, whereas red indicates dead cells. Isolated islets were cultured with standard islet culture medium (M199) and co-cultured with mesenchymal stem cells (MSCs) for 7 days. Co-cultured islets demonstrated a lower cell death rate than control islets (×200). (B) After 48 hr, co-cultured islets also demonstrated a lower DNA fragmentation than the control (0.52±0.1-fold change vs. control, *P<0.05, n=6). (C) In addition, islets co-cultured with MSCs for 48 hr showed a lower ADP/ATP ratio than control islets (*P<0.05 vs. control, n=5). (D) The glucose stimulated insulin secretion of co-cultured islets also increased when compared with control islets after 48 hr (*P<0.05 vs. control, n=5).

Alteration of Cell Survival, Apoptosis, and Angiogenesis-Related Signaling

Islet-MSC co-culture enhanced islet viability and β-cell insulin secretory function while reducing ADP/ATP ratios. To assess the islet quality improvement as a result of MSC co-culture, the control of survival, function, and angiogenesis- related signal proteins were analyzed by western blot after a 48-hr co-culture. PDX-1 production increased in islets co-cultured with MSCs. Proliferation marker Ki-67, however, exhibited no difference between islets co-cultured with MSCs and control. In contrast, XIAP, Bcl-xL, and Bcl-2, known as anti-apoptotic signal molecules, and HSP-32, a cell survival-related protein, exhibited greater expression in MSC–co- cultured islets than in control (Fig. 3A). The activation of Akt and ERK1/2 were also analyzed. The phosphorylation state of Akt was altered in MSC–co-cultured islets, but the phosphorylation of ERK1/2 demonstrated no changes (Fig. 3B). In addition, the mRNA expression of VEGF families and their receptors (VEGFs, VEGFRs, and Tie-2) were screened to investigate the induction of angiogenesis-related signals in islets co-cultured with MSCs. Our analysis demonstrated increases in gene expression levels of VEGF receptor 2 (VEGFR2) and Tie-2 in MSC–co-cultured islets (Fig. 3C). At the protein level, PLC-γ phosphorylation, which is a known upstream signal molecule related to the VEGF-A-mediated proliferation of endothelial cells, was not observed. However, FAK and Tie-2 phosphorylation, which are known upstream signaling molecules related to the VEGF-A mediated migration of endothelial cells, were significantly induced by islets co-cultured with MSCs (Fig. 3D).

Western blot analysis to investigate the control of survival, proliferation, and angiogenesis related- signal induction of co-cultured islet. (A) Mesenchymal stem cell (MSC)–co-cultured islets altered expression and phosphorylation of various signal molecules. Alterations in Ki-67 levels were not evident in any group, however, pancreatic and duodenal homeobox-1 (PDX-1) increased in co- cultured islets. The expression of anti-apoptosis and cell survival-related proteins, heat shock protein (HSP)-32, XIAP, Bcl-xL, and Bcl-2 increased in MSC–co-cultured islets. (B) The phosphorylation of Akt increased in MSC–co-cultured islets. (C) As demonstrated by amplification of genes involved in the angiogenesis of MSC–co-cultured islets, the expression of VEGFR2 and Tie-2 mRNA were increased. (D) Tie-2 and FAK phosphorylation were stimulated in the MSC co-culture. In contrast, the induction of phospholipase c-γ phosphorylation, a known upstream signal molecule related to endothelial cell proliferation, was not observed.

Quality Assessment of Islets Cultured in MSC-CM

To optimize the MSC-CM conditions and confirm whether the effects of islet-MSC co-culture are caused by trophic factors derived from MSCs, we prepared a series of MSC-CMs of different MSC-culture durations. MSC-cultured supernatants were collected at 24, 48, and 72 hr after plating (referred to as MSC-CM1, MSC-CM2, and MSC-CM3, respectively). All groups cultured for 48 hr in various MSC-CM demonstrated lower ADP/ATP ratios than islets cultured in the standard medium (control: 0.14±0.01, MSC-CM1: 0.11±0.01, MSC-CM2: 0.06±0.01, and MSC-CM3: 0.88±0.01). Of the three types of MSC-CM tested, MSC-CM2 was the most effective in reducing ADP/ATP ratios (control vs. MSC-CM 2, P<0.05, Fig. 4A). Therefore, the optimized MSC-CM2 was chosen for subsequent assays.

Evaluation of ADP/ATP ratio, insulin secretion, and DNA fragmentation in mesenchymal stem-cell (MSC)-conditioned medium (CM)–cultured islets. (A) A series of MSC-CMs were prepared according to MSC culture duration. MSC-cultured supernatants were collected at 24, 48, and 72 hr after medium change (referred to as MSC-CM1, MSC-CM2, and MSC-CM3, respectively). Among these preconditioned medium, the MSC-CM2 culture group demonstrated the lowest ADP/ATP ratio (*P<0.05 vs. control, n=5) after 48 hr. (B) During the 5-day culture, the ADP/ATP ratio of mouse islets cultured with MSC-CM2 demonstrated a lower ratio than that of the control group over the entire culture period (*P<0.05 or *P<0.01 vs. control at the same time point, n=5). (C) The glucose-stimulated insulin secretion of MSC-CM2–cultured mouse islets also increased compared with control islets (*P<0.05 vs. control, n=5) after 48 hr. (D) In addition, MSC-CM2–cultured islets demonstrated lower DNA fragmentation than the control (*P<0.05 vs. control, n=6).

Mouse islets were cultured with MSC-CM2 for 120 hr, with analysis of ADP/ATP ratios taking place at 24, 48, 72, and 120 hr. Although the ADP/ATP ratios of all islets increased after the 48-hr culture, the ADP/ATP ratios of islets cultured in MSC-CM2 were consistently lower than that of the control (control vs. MSC-CM2 at 48 hr: 0.1±0.01 vs. 0.05±0.01, at 72 hr: 0.16±0.01 vs. 0.08±0.01, at 120 hr: 0.23±0.01 vs. 0.12±0.01, P<0.05 at 48 and 72 hr, P<0.01 at 120 hr, Fig. 4B). Moreover, islets cultured with MSC-CM demonstrated increases in glucose-stimulated insulin secretion (control vs. MSC-CM2 at 300 mg/dL glucose, 25.42±3.967 vs. 42.06±3.36 ng/mL, P<0.05, Fig. 4C) and an increased secretion index (control: 2.85±0.53, MSC-CM 2: 3.87±0.44) after 48 hr. Lower amounts of DNA fragmentation occurred in the MSC-CM islets as compared with control (0.43±0.03-fold change in MSC-CM vs. control, P<0.05, Fig. 4D). Both MSC-CM–cultured and co-cultured islets demonstrated similar positive effects on islet function. These results suggest that the improved islet quality, including survival and function, in islets co-cultured with MSCs was caused by MSC-derived factors.

In Vivo Function of Islets Cultured in MSC-CM

Diabetic mouse models were produced by an STZ-injection, and mice with sustained hyperglycemia of more than 400 mg/dL were used as islet graft recipients. Enhanced islet quality related to culture in optimized MSC-CM (MSC-CM2) was investigated with a transplantation model using a marginal number (200 IEq) of islets. Diabetic mice transplanted with islets cultured in MSC-CM demonstrated enhanced glycemic control and increased body weight gain when compared with the control. During the intraperitoneal glucose tolerance test, a 14-hr fast resulted in similar blood glucose levels for both groups. After glucose administration, however, the group that was transplanted with MSC-CM–cultured islets demonstrated significantly lower levels of glucose at 90 and 120 min than the ITx group transplanted with control-cultured islets (*P<0.05 vs. control at the same time point, Fig. 5A–C).

Function of mesenchymal stem cell (MSC)-conditioned medium (CM)–cultured islets after in vivo islet transplantation. (A–C) Compared with the control group (dashed line), islets cultured with MSC-CM (solid line) for 48 hr demonstrated enhanced glycemic control and glucose tolerance ability in random blood glucose levels and intraperitoneal glucose tolerance test and increase in body weight (*P<0.05 vs. control at the same time point, n=6). (D–F) After euthanization of the animals, the paraffin-embedded grafts from each group were stained with lectin/insulin (brown, insulin; green, blood vessel) to analyze β-cell function and angiogenesis. The insulin staining intensity was increased in islet grafts cultured with MSC-CM. The relative blood vessel area of islets cultured with MSC-CM also increased (*P<0.05 vs. control, n=6).

During histologic analysis, no residual or regenerative islets were identified in either STZ-treated group. Moreover, statistical analysis revealed no differences in the insulin positive areas of destructive islet remnants in the MSC-CM group and the control (see Figure, Supplemental Digital Content 3, Transplanted grafts using islets cultured in MSC-CM revealed higher insulin content than those using control islets (Fig. 5D and E). The blood vessel structure of grafts using MSC-CM–cultured cells demonstrated improved blood vessel formation and enhanced blood vessel area compared with the control (control vs. MSC-CM: 3.89±0.36 vs. 5.07±0.28, P<0.05, Fig. 5F).

Identification of MSC-Secreted Molecules

The islet-MSC co-culture improved islet quality and induced related cell signals, despite the lack of any direct contact between islets and MSCs. This implies that improved islet quality is linked to MSC-derived molecules such as cytokines and growth factors. Therefore, we performed an ELISA to identify MSC-secreted soluble factors: insulin, IGF-1, IL-6, VEGF, HGF, TGF-β, SCF, FGF-2, TNF-α, and IFN-γ. Among these factors, HGF, IL-6, TGF-β, and VEGF-A were detected at significant concentrations in supernatants form MSCs culture (476.50±22.20, 4404.63±700.86, 1118±95.41, and 6824±441.8 pg/mL, respectively). In particular, a high concentration of VEGF-A was detected (>6 ng/mL; Fig. 6). SCF and FGF-2 were also detected, but at very low concentrations. The proinflammatory cytokines (TNF-α and IFN-γ), insulin, and IGF-1 were not observed at significant concentrations in MSC-CM (data not shown). A small amount of TGF-β was observed in the normal growth medium, whose level was excluded from obtained level of MSC-CM.

To identify the secreted soluble factors, mesenchymal stem cells were cultured for 48 hr, after which the medium supernatant was harvested to determine the identity and amount of soluble factors using ELISA. Interleukin (IL)-6, vascular endothelial growth factor (VEGF)-A, hepatocyte growth factor (HGF), and transforming growth factor (TGF)-β were detected at significant levels (n=3).


Successful ITx is dependent not only on the number or mass of the islets but also on the stable engraftment and prevention of immune-mediated rejection of transplanted islets. Furthermore, high islet quality, as defined by viability, metabolic activity, and function, is necessary (25, 26). Apart from immune-mediated mechanisms, various factors, including a deficiency in trophic agents, have been proposed to account for the low survival rate for transplanted islet grafts (3). MSCs have the potential to repair and regenerate nearby tissue through the secretion of active or trophic agents (15–17). Thus, we hypothesized that appropriate trophic molecules may effectively inhibit islet quality degradation during islet culture and that MSCs can secreted active agents that are beneficial in maintaining islet viability and function.

To evaluate this hypothesis, we established a co-culture system. Without direct cell-to-cell contact, islets co-cultured with MSCs successfully increased mouse islet qualities, including viability and insulin secretory function, and reduced ADP/ATP ratios when compared with islets cultured in the standard culture medium (Fig. 2). The lower ADP/ATP ratios imply enhanced metabolic activity or increased cytoplasmic ATP content, which are both reported to be related to insulin secreting function (27). In this study, we used primary isolated MSCs that displayed some variation based on feeder cells or conditioned medim (CM). MSCs isolation from cord blood occasionally failed due to factors such as donor age or delivery time from blood collection. However, once established, the majority of MSCs (seven to nine passages) showed similar trophic effects as feeder cells. In addition, even within a single MSC line, some variation in trophic effect is present due to the quality of the islet itself, which is also a primary isolated cell. Islet quality is highly variable due to varying isolation conditions such as digestion time and pancreas harvest time lapse. Therefore, we performed islet quality assays (ADP/ATP ratio and glucose stimulated insulin secretion) using five different MSC lines (see Table, Supplemental Digital Content 4, In co-culture, we evaluated the stability of MSCs, in which excessive variation was not seen as feeder cells for trophic effect (see Table, Supplemental Digital Content 4, On the other hand, to demonstrate the MSC-specific or MSC-enhanced effect in co-culture condition, we evaluated the mouse embryonic fibroblast cell line to evaluate the effect on the islet quality (The same coculture conditions as those used for MSCs). Islet cocultured with MSCs showed a lower ADP/ATP ratio than the negative control and the NIH3T3 cocultured group. Moreover, compared to control, islets cocultured with NIH3T3s showed a higher ADP/ATP ratio. This unexpected tendency was also observed with the GSIR test used for evaluation of beta cell function, which demonstrated an increase in glucose-stimulated insulin secretion as well as higher secretion indexes (for islets cocultured with MSCs for 48 hr., in conclusion, NIH3T3 cells showed pixilated potency as feeder cells for islet culture under our experimental conditions (see Figure, Supplemental Digital Content 5, Subsequently, we used the most stable MSC line in terms of cell proliferation, cell morphology, and doubling time for CM evaluation.

The expressions and phosphorylation states of various signaling proteins were evident in MSC–co-cultured mouse islets (Fig. 3). PDX-1 expression, which is known to increase the regeneration and proliferation of β cells (28, 29), was increased in MSC–co-cultured islets, but Ki-67 levels did not change (Fig. 3A). This suggests that the MSC co-culture may influence islet insulin secretion, but not proliferation. HSP-32, which is known to have a cytoprotective effect on islets, functions by suppressing inflammatory reactions and oxidative stress (30, 31). In our study, HSP-32 expression increased in the co-cultured group (Fig. 3A). In addition, XIAP, Bcl-2, and Bcl-xL, which are known as anti-apoptotic signaling molecules in β cells (32, 33), showed increased expression (Fig. 3A). The up-regulation of anti-apoptotic, β-cell repair, and survival-related proteins implies that islet-MSC co-culture may support islet survival, which is supported by our in vitro assay analyzing survival, ATP/ADP ratios, and insulin secretory functions. Akt is known to play a crucial role in the survival of pancreatic β cells (34, 35). The phosphorylation of Akt signaling increased in co-cultured mouse islets, although ERK1/2 phosphorylation did not experience any changes (Fig. 3B). Therefore, the up-regulation of HSP-32, XIAP, and Bcl-families could result from Akt pathway activation and not from ERK1/2 signaling.

At the level of mRNA, we observed changes in the expression of angiogenesis/revascularization-related genes, including VEGFR2 and Tie-2, in islet cells co-cultured with MSCs (Fig. 3C). VEGF-A has been shown to induce the revascularization, angiogenesis, and proliferation of endothelial cells in islet grafts after transplantation. Reports also indicate potential increases in β-cell function and mass in isolated islets (19, 36, 37). VEGFR2 is a well-known receptor of VEGF-A (38). Therefore, the up-regulated gene expression of VEGFR2 implies an increase in VEGF-A-induced signaling. In addition, Tie-2 signaling triggered by Ang-2 is involved in the formation and revascularization of blood vessel networks in various cells (39), a signal that could be involved with VEGF-induced signaling (40, 41). Furthermore, we identified the induction of FAK signaling, but an alteration in the PLC-γ signal (Fig. 3C) as a VEGF down-regulated signal was not observed (42). The expression of Tie-2 was not altered in any group at the protein level, even though Tie-2 mRNA levels generally increased compared with control group. However, increased levels of phosphorylated Tie-2 were observed in MSC–co-cultured islets (Fig. 3D). This implies that stimulated Tie-2 signaling may depend on phosphorylation triggered by a signal cascade related to MSC- secreted ligands or VEGF-down-regulated signaling.

With the use of MSC-CM, islets demonstrated improved ADP/ATP ratio, glucose-stimulated insulin secretion, secretion index, and prevention of islet cell death that was similar to that observed with the islet-MSC co-culture (Fig. 4). These results confirm that islet quality enhancement is caused by an MSC-derived trophic factor. Islets cultured in MSC-CM facilitated the regulation of blood glucose at normoglycemic levels, with the recipient mice demonstrating superior weight gain and glucose tolerance capacities. In addition, residual pancreata of recipient mice were rarely stained by insulin. Therefore, these effects were not related to an endogenous pancreatic regenerative effect.

Histologic analysis revealed an increase in the intensity of insulin staining and significantly enhanced blood vessel formation in islets cultured with MSC-CM (Fig. 5). Reports indicate that a well-developed blood vessel structure can improve insulin secretion through interaction between β and endothelial cells (43). It is possible that the difference in histologic results between the two groups could be due to the effects of glucose toxicity related to hyperglycemia in the control group. However, the in vitro evidence demonstrating differences in VEGF and angiogenesis-related signaling and β-cell function enhancement support the hypothesis that increased blood vessel formation of MSC-CM–cultured islet grafts is caused by a trophic factor related to MSCs, which induces signals involved in β-cell survival and intra-islet endothelial revascularization.

Finally, the MSC-CM was screened for alterations in level of several cytokines and growth factors using ELISA to determine the specific factors that modulate intracellular signaling related to increased islet survival and function in vitro and in vivo. Among these factors, IL-6, VEGF-A, HGF, and TGF-β were secreted at significant levels. VEGF-A was detected at higher concentrations (>6 ng/mL) and demonstrated a critical correlation with islet quality (Fig. 6). TGF-β reportedly stimulates the enhancement of HSP-32 and XIAP in human epithelial cells (44), whereas IL-6 induces Bcl-2 and Bcl-xL expression (18, 45). IL-6 could, therefore, be related to decreased islet cell death. Furthermore, HGF and TGF-β are known to trigger the Akt pathway (46, 47). Consequently, HGF and TGF-β may trigger the PI-3K-PKB/Akt pathway to regulate insulin secretion and cell survival signaling in the islet-MSC coculture.

In conclusion, MSC-secreted trophic agents seem to prevent islet cell death through the induction of anti-apoptotic and survival signaling, enhancement of islet function, and stimulation of migration and revascularization of intra-islet endothelial cells. Islets co-cultured with MSCs result in enhanced islet quality and thereby improve in vivo glycemic control. These results strongly support the hypothesis that MSC-originated factors have a positive effect on isolated islets. Further studies are needed to investigate the specific signaling mechanism involved, and to identify the critical trophic factors and to determine their role in enhanced islet quality to link their biological activities to islet function or viability. Evaluation of human islets involving the ITx model used for immune-deficient mice should also be performed. Our study contributes to the current body of research on mesenchymal trophic factors as facilitators of islet intracellular signaling and the optimization of human islet culture conditions.


The authors thank the Laboratory Animal Research Center of the Samsung Biomedical Research Institute for animal care and technical support.


1.Goto M, Yoshikawa Y, Matsuo K, et al. Optimization of a prominent oxygen-permeable device for pancreatic islets. Transplant Proc 2008; 40: 411.
2.Ricordi C, Fraker C, Szust J, et al. Improved human islet isolation outcome from marginal donors following addition of oxygenated perfluorocarbon to the cold-storage solution. Transplantation 2003; 75: 1524.
3.Ilieva A, Yuan S, Wang RN, et al. Pancreatic islet cell survival following islet isolation: The role of cellular interactions in the pancreas. J Endocrinol 1999; 161: 357.
4.Keymeulen B, Gillard P, Mathieu C, et al. Correlation between beta cell mass and glycemic control in type 1 diabetic recipients of islet cell graft. Proc Natl Acad Sci USA 2006; 103: 17444.
5.Ichii H, Sakuma Y, Pileggi A, et al. Shipment of human islets for transplantation. Am J Transplant 2007; 7: 1010.
6.Ling Z, Hannaert JC, Pipeleers D. Effect of nutrients, hormones and serum on survival of rat islet beta cells in culture. Diabetologia 1994; 37: 15.
7.Woods EJ, Walsh CM, Sidner RA, et al. Improved in vitro function of islets using small intestinal submucosa. Transplant Proc 2004; 36: 1175.
8.Rutzky LP, Bilinski S, Kloc M, et al. Microgravity culture condition reduces immunogenicity and improves function of pancreatic islets1. Transplantation 2002; 74: 13.
9.Miki A, Narushima M, Okitsu T, et al. Maintenance of mouse, rat, and pig pancreatic islet functions by coculture with human islet-derived fibroblasts. Cell Transplant 2006; 15: 325.
10.Luo L, Badiavas E, Luo JZ, et al. Allogeneic bone marrow supports human islet beta cell survival and function over six months. Biochem Biophys Res Commun 2007; 361: 859.
11.Granero-Molto F, Weis JA, Longobardi L, et al. Role of mesenchymal stem cells in regenerative medicine: Application to bone and cartilage repair. Expert Opin Biol Ther 2008; 8: 255.
12.Le Blanc K, Ringden O. Immunobiology of human mesenchymal stem cells and future use in hematopoietic stem cell transplantation. Biol Blood Marrow Transplant 2005; 11: 321.
13.Caplan AI, Dennis JE. Mesenchymal stem cells as trophic mediators. J Cell Biochem 2006; 98: 1076.
14.van Poll D, Parekkadan B, Cho CH, et al. Mesenchymal stem cell-derived molecules directly modulate hepatocellular death and regeneration in vitro and in vivo. Hepatology 2008; 47: 1634.
15.Lataillade JJ, Doucet C, Bey E, et al. New approach to radiation burn treatment by dosimetry-guided surgery combined with autologous mesenchymal stem cell therapy. Regen Med 2007; 2: 785.
16.Zhang M, Mal N, Kiedrowski M, et al. SDF-1 expression by mesenchymal stem cells results in trophic support of cardiac myocytes after myocardial infarction. FASEB J 2007; 21: 3197.
17.Park HJ, Lee PH, Bang OY, et al. Mesenchymal stem cells therapy exerts neuroprotection in a progressive animal model of Parkinson's disease. J Neurochem 2008; 107: 141.
18.Choi SE, Choi KM, Yoon IH, et al. IL-6 protects pancreatic islet beta cells from pro-inflammatory cytokines-induced cell death and functional impairment in vitro and in vivo. Transpl Immunol 2004; 13: 43.
19.Cheng Y, Liu YF, Zhang JL, et al. Elevation of vascular endothelial growth factor production and its effect on revascularization and function of graft islets in diabetic rats. World J Gastroenterol 2007; 13: 2862.
20.Fiaschi-Taesch NM, Berman DM, Sicari BM, et al. Hepatocyte growth factor enhances engraftment and function of nonhuman primate islets. Diabetes 2008; 57: 2745.
21.Urbán VS, Kiss J, Kovacs J, et al. Mesenchymal stem cells cooperate with bone marrow cells in therapy of diabetes. Stem Cells 2008; 26: 244.
22.Bieback K, Kern S, Kluter H, et al. Critical parameters for the isolation of mesenchymal stem cells from umbilical cord blood. Stem Cells 2004; 22: 625.
23.Gozal E, Ortiz LA, Zou X, et al. Silica-induced apoptosis in murine macrophage: Involvement of tumor necrosis factor-alpha and nuclear factor-kappaB activation. Am J Respir Cell Mol Biol 2002; 27: 91.
24.Invernici G, Emanueli C, Madeddu P, et al. Human fetal aorta contains vascular progenitor cells capable of inducing vasculogenesis, angiogenesis, and myogenesis in vitro and in a murine model of peripheral ischemia. Am J Pathol 2007; 170: 1879.
25.Boker A, Rothenberg L, Hernandez C, et al. Human islet transplantation: Update. World J Surg 2001; 25: 481.
26.Robertson RP. Islet transplantation as a treatment for diabetes—A work in progress. N Engl J Med 2004; 350: 694.
27.Miwa I, Taniguchi S. Acceleration by fructose of the ATP-sensitive K(+) channel-independent pathway of glucose-induced insulin secretion. Horm Metab Res 2002; 34: 450.
28.McKinnon CM, Docherty K. Pancreatic duodenal homeobox-1, PDX-1, a major regulator of beta cell identity and function. Diabetologia 2001; 44: 1203.
29.Taniguchi H, Yamato E, Tashiro F, et al. beta-cell neogenesis induced by adenovirus-mediated gene delivery of transcription factor pdx-1 into mouse pancreas. Gene Ther 2003; 10: 15.
30.Pileggi A, Molano RD, Berney T, et al. Heme oxygenase-1 induction in islet cells results in protection from apoptosis and improved in vivo function after transplantation. Diabetes 2001; 50: 1983.
31.Lee DY, Lee S, Nam JH, et al. Minimization of immunosuppressive therapy after islet transplantation: Combined action of heme oxygenase-1 and PEGylation to islet. Am J Transplant 2006; 6: 1820.
32.Emamaullee JA, Rajotte RV, Liston P, et al. XIAP overexpression in human islets prevents early posttransplant apoptosis and reduces the islet mass needed to treat diabetes. Diabetes 2005; 54: 2541.
33.Plesner A, Liston P, Tan R, et al. The X-linked inhibitor of apoptosis protein enhances survival of murine islet allografts. Diabetes 2005; 54: 2533.
34.Aikin R, Hanley S, Maysinger D, et al. Autocrine insulin action activates Akt and increases survival of isolated human islets. Diabetologia 2006; 49: 2900.
35.Meng ZX, Sun JX, Ling JJ, et al. Prostaglandin E2 regulates Foxo activity via the Akt pathway: Implications for pancreatic islet beta cell dysfunction. Diabetologia 2006; 49: 2959.
36.Cross SE, Richards SK, Clark A, et al. Vascular endothelial growth factor as a survival factor for human islets: Effect of immunosuppressive drugs. Diabetologia 2007; 50: 1423.
37.Chae HY, Lee BW, Oh SH, et al. Effective glycemic control achieved by transplanting non-viral cationic liposome-mediated VEGF-transfected islets in streptozotocin-induced diabetic mice. Exp Mol Med 2005; 37: 513.
38.Shibuya M, Claesson-Welsh L. Signal transduction by VEGF receptors in regulation of angiogenesis and lymphangiogenesis. Exp Cell Res 2006; 312: 549.
39.Suzuki Y, Komi Y, Ashino H, et al. Retinoic acid controls blood vessel formation by modulating endothelial and mural cell interaction via suppression of Tie2 signaling in vascular progenitor cells. Blood 2004; 104: 166.
40.Chong AY, Caine GJ, Lip GY. Angiopoietin/tie-2 as mediators of angiogenesis: A role in congestive heart failure? Eur J Clin Invest 2004; 34: 9.
41.Holash J, Wiegand SJ, Yancopoulos GD. New model of tumor angiogenesis: Dynamic balance between vessel regression and growth mediated by angiopoietins and VEGF. Oncogene 1999; 18: 5356.
42.Matsumoto T, Mugishima H. Signal transduction via vascular endothelial growth factor (VEGF) receptors and their roles in atherogenesis. J Atheroscler Thromb 2006; 13: 130.
43.Johansson M, Mattsson G, Andersson A, et al. Islet endothelial cells and pancreatic beta-cell proliferation: Studies in vitro and during pregnancy in adult rats. Endocrinology 2006; 147: 2315.
44.Kutty RK, Nagineni CN, Kutty G, et al. Increased expression of heme oxygenase-1 in human retinal pigment epithelial cells by transforming growth factor-beta. J Cell Physiol 1994; 159: 371.
45.Ogawa M, Nishiura T, Oritani K, et al. Cytokines prevent dexamethasone-induced apoptosis via the activation of mitogen-activated protein kinase and phosphatidylinositol 3-kinase pathways in a new multiple myeloma cell line. Cancer Res 2000; 60: 4262.
46.Garcia-Ocana A, Vasavada RC, Cebrian A, et al. Transgenic overexpression of hepatocyte growth factor in the beta-cell markedly improves islet function and islet transplant outcomes in mice. Diabetes 2001; 50: 2752.
47.Lin CC, Chiang LL, Lin CH, et al. Transforming growth factor-beta1 stimulates heme oxygenase-1 expression via the PI3K/Akt and NF-kappaB pathways in human lung epithelial cells. Eur J Pharmacol 2007; 560: 101.

Islet culture; Islet transplantation; Mesenchymal stem cells; Trophic effect

Supplemental Digital Content

© 2010 Lippincott Williams & Wilkins, Inc.