A Method for Performing Islet Transplantation Using Tissue-Engineered Sheets of Islets and Mesenchymal Stem Cells

Abstract

Mesenchymal stem cells (MSCs) are known to have a protective effect on islet cells. Cell sheets developed using tissue engineering help maintain the function of the cells themselves. This study describes a tissue engineering approach using islets with MSC sheets to improve the therapeutic effect of islet transplantation. MSCs were obtained from Fischer 344 rats and engineered into cell sheets using temperature-responsive culture dishes. The islets obtained from Fischer 344 rats were seeded onto MSC sheets, and the islets with MSC sheets were harvested by low-temperature treatment after coculture. The functional activity of the islets with MSC sheets was confirmed by a histological examination, insulin secretion assay, and quantification of the levels of cytokines. The therapeutic effects of the islets with MSC sheets were investigated by transplanting the sheets at subcutaneous sites in severe combined immunodeficiency (SCID) mice with streptozotocin-induced diabetes. Improvement of islet function and viability was shown in situ on the MSC sheet, and the histological examination showed that the MSC sheet maintained adhesion factor on the surface. In the recipient mice, normoglycemia was maintained for at least 84 days after transplantation, and neovascularization was observed. These results demonstrated that islet transplantation in a subcutaneous site would be possible by using the MSC sheet as a scaffold for islets.

FIG. 1.

(A) A flow cytometric analysis. CD29 and CD90 such as mesenchymal markers were positive, and CD31 and CD34 such as hematopoietic markers were negative. (B) To confirm a capability of MSC differentiation into osteogenesis and adipogenesis, Alizarin Red S staining and Oil Red O staining were performed. MSCs, mesenchymal stem cells.

FIG. 2.

(A, B) The islets were seeded at a density of 50 islets/cm² in 35-mm diameter temperature-responsive dish. The islets were riding on the confluent MSCs in the temperature-responsive dish. (C) The islets+MSC sheets were harvested by low-temperature treatment after 72 h coculture. (D) The islet+MSC sheets were harvested while shrinking during low-temperature treatment. (E) The islets were stained with dithizone. (F) H&E staining showed that the sheets adhered to the islets in the shape of spheres. (G, H) Rat insulin and glucagon immunostaining of islets cocultured with MSC sheets. (I–M) Ultrastructures of the islets+MSC sheets were observed by electron microscopy. (I, J) ECM was partially detected between the islets and MSC sheets. (K) The islets and MSC sheets were connected through the formation of tight junctions. (L) The MSC sheets consisted of multiple layers. (M) Cell-to-cell connections were observed in the MSC sheets due to the formation of tight and gap junctions. ECM, extracellular matrix; H&E, hematoxylin and eosin; N, nucleus; TJ, tight junctions (arrow); GJ, gap junctions (arrowhead).

FIG. 3.

(A) The recovery rate was calculated after 24 and 72 h of incubation. (B) The viability of the islets was assessed using calcein-AM and PI. Viable cells were stained green and dead cells were stained red. Almost all MSCs and MSC sheets were viable. (C) Viability of the islets cultured alone and cocultured with MSCs and MSC sheets. (D) The insulin levels changed along with the change in the glucose concentration. The SI was calculated in the cultured-alone group and cocultured with MSCs and MSC sheet groups. (E–G) The secretions of VEGF, HGF, and TGFβ1 in the supernatants obtained from the islets alone, islets cocultured with MSCs, and MSC sheet groups. n=5 each. *p<0.05, **p<0.01 compared to the group of islets cultured alone. HGF, hepatocyte growth factor; PI, propidium iodide; SI, stimulation index; TGFβ1, transforming growth factor beta 1; VEGF, vascular endothelial growth factor.

FIG. 4.

(A) The islets+MSC sheets adhered to GP. The islets+MSC sheets were attached to the surrounding tissue. The implanted islets+MSC sheets are indicated by a dashed line. (B) The blood glucose levels of diabetic sham-operated (DM-sham) mice (n=5) and those of the recipient mice: 2000 islets alone (n=5), MSC sheet alone (n=5), 2000 islets with MSCs (n=5), two islets+MSC sheets (1000 islets) (n=5), and four islets+MSC sheets (2000 islets) (n=6). *p<0.05, **p<0.01 compared to the DM-sham group. (C) Body weight changes in the recipient mice treated with four islets+MSC sheets (2000 islets; black circles, n=6) and the DM-sham mice (white squares, n=5). *p<0.05, **p<0.01 compared to the DM-sham group. (D) The recipient mice were transplanted with four islets+MSC sheets (2000 islets; black circles, n=6). The graft tissue was surgically removed on day 84. ^†Graft removal. (E) The IPGTT was performed in the normal SCID mice (white circles, n=9) and recipient mice treated with four islets+MSC sheets (2000 islets; black circles, n=6) on day 56. DM, diabetes mellitus; GP, glass plates; IPGTT, intraperitoneal glucose tolerance tests; SCID, severe combined immunodeficiency.

FIG. 5.

(A) Histological, immunohistochemical, and immunofluorescence analyses on day 28 after the subcutaneous transplantation of the islets+MSC sheets. The expression of insulin and Pdx1 on islets was observed in the connective tissue. (B) Serum insulin level was investigated in the DM-sham (n=5), recipient SCID mice [2000 islets (n=5), 2000 islets with MSCs (n=5), two islets+MSC sheets (1000 islets) (n=5), four islets+MSC sheets (2000 islets) (n=6)], and normal SCID mice (n=7). **p<0.01 compared to the DM-sham group.

FIG. 6.

(A) The expression of vWF was observed in the subcutaneous site. (B) The degree of vascularization was evaluated according to the number of vessels for anti-vWF immunostaining. In the implanted subcutaneous tissue, the number of vessels per square millimeter was counted. At 28 days after treatment, specimens were obtained from DM-sham mice (n=5), recipient SCID mice treated with islets alone (n=5), 4 MSC sheets alone (n=5), 2000 islets with MSCs (n=5), and 4 islets+MSC sheet (n=6). *p<0.05, **p<0.01. vWF, von Willebrand factor.