Long-term glycemic control using polymer-encapsulated human stem cell-derived beta cells in immune-competent mice

Abstract

The transplantation of glucose-responsive, insulin-producing cells offers the potential for restoring glycemic control in individuals with diabetes. Pancreas transplantation and the infusion of cadaveric islets are currently implemented clinically, but these approaches are limited by the adverse effects of immunosuppressive therapy over the lifetime of the recipient and the limited supply of donor tissue. The latter concern may be addressed by recently described glucose-responsive mature beta cells that are derived from human embryonic stem cells (referred to as SC-β cells), which may represent an unlimited source of human cells for pancreas replacement therapy. Strategies to address the immunosuppression concerns include immunoisolation of insulin-producing cells with porous biomaterials that function as an immune barrier. However, clinical implementation has been challenging because of host immune responses to the implant materials. Here we report the first long-term glycemic correction of a diabetic, immunocompetent animal model using human SC-β cells. SC-β cells were encapsulated with alginate derivatives capable of mitigating foreign-body responses in vivo and implanted into the intraperitoneal space of C57BL/6J mice treated with streptozotocin, which is an animal model for chemically induced type 1 diabetes. These implants induced glycemic correction without any immunosuppression until their removal at 174 d after implantation. Human C-peptide concentrations and in vivo glucose responsiveness demonstrated therapeutically relevant glycemic control. Implants retrieved after 174 d contained viable insulin-producing cells.

Figure 1

SC-β cells encapsulated with TMTD alginate sustain normoglycemia in STZ-treated immune competent C57BL/6J mice. (a) SC-β cells were generated using the differentiation protocol described. FACS analysis shows surface markers on cells at indicated differentiation stages. Data is representative of 10 separate differentiations from the HUES8 stem cell line. (Editor: Stage 1–3 is previously described and not relevant to this manuscript) (b) Brightfield images of encapsulated SC-β cells.. Scale bar = 400 µm, N = 15. (c–e) SC-β cells encapsulated as shown in (b) were transplanted into the intraperitoneal space of STZ-treated C57BL/6 mice, and blood glucose concentrations were measured at indicated times. (c) 500 µm SLG20 alginate microcapsules; (d) 1.5 mm SLG20 alginate microspheres; (e) 1.5 mm TMTD alginate spheres. Three different doses of cell clusters (100, 250, and 1000 cluster per mouse) were implanted under each encapsulation condition. The red dashed line indicates the blood glucose cutoff for normoglycemia in mice. For reference 250 clusters equates to approximately 1 million cells. Error bars, mean ± s.e.m. Quantitative data shown is the average of N = 5 mice per treatment. All experiments were repeated three times for a total of N = 15 mice per treatment.

Figure 2

SC-β cells encapsulated with TMTD alginate elicit weaker immunological and fibrotic responses in immune competent C57BL/6J mice. (a) FACS analysis of encapsulated SC-β cell implants retrieved 14 days after intraperitoneal transplantation in C57BL/6 mice, N = 10. Red asterisks specify statistical significance between indicated groups while black asterisks specify significance against the SLG 0.5 group. (b–d) Dark-field (N=15) (b), brightfield (N=15) (c) and Z-sacked confocal immunofluorescence imaging (DAPI shown blue, F-actin shown in green, α-SMA shown in red) (d) of implants retrieved after 90 days from the STZ-treated C57BL/6J mice presented in Figure 1c–e, (N = 15) (e) The implants containing 250 clusters were retrieved from the STZ-treated C57BL/6J mice shown in Figure 1c–e 90 days after implantation at XX days after implantation. Implant lysates were subjected to proteomic analysis XXX analysis. (N = 4 per mice treatment; analysis was performed on two separate cohorts for a total of 8 mice per treatment). Each column in the heatmap represents an individual mouse from the respective treatment group. (f) α-SMA protein isolated from implants retrieved from the STZ-treated C57BL/6J mice shown in Figure 1c–e was analyzed by immunoblot. Graph shows lower levels of α-SMA for TMTD 1.5 mm spheres. (N = 5) (g-h) Pre-transplantation (g) and post-retrieval (h) histological sectioning and immunostaining of encapsulated SC-β cells from implants retrieved at 90 XXXX days after implantation from the STZ-treated C57BL/6J mice shown in Figure 1c–e (N = 15). Scale bar = 20 µm. All experiments were performed at least three times with the exception of FACS and proteomics quantification, which were each repeated twice. One-way ANOVA with Bonferroni multiple comparison correction, = p < 0.0001. Scale bar in a = 2 mm. Scale bar in b and c = 300 µm.

Figure 3

SC-β cells encapsulated with TMTD alginate sustain long-term normoglycemia in STZ-treated C57BL/6J mice. (a) Blood glucose levels were measured in STZ-treated C57BL/6 mice implanted with SC-β cells encapsulated with TMTD alginate at a dose of 250 clusters/mouse. Blood glucose levels were also measured of a Healthy C57BL/6J mouse cohort not treated with STZ. Shown are cohorts of N = 5 mice; experiment was repeated 3 times for a total of 15 mice per condition. (b) The mice shown in (a), together with a cohort of STZ-treated non-implanted mice, were subjected to IVGTT 174 days after implantation. (c) Human C-peptide levels were measured in the blood of the STZ-treated C57BL/6 mice implanted with SC-β cells shown in (a). (d) Quantification of mouse insulin extracted from the pancreas of mice (N = 5) in each treatment group. TMTD alginate treated mice had their pancreas removed after 174 days post-implantation, while the pancreas of healthy and STZ-treated non-implanted C57BL/6 mice were taken at 8–10 weeks. at XX time point. (e–f) Brightfield (e) and Z-sacked confocal immunofluorescence imaging (e) (DAPI shown blue, F- actin shown in green, α-SMA shown in red) of implants retrieved at XX time point from the mice presented in (a–c). Scale bar = 400 µm (g-h). Masson’s thrichrome (g), and H&E (h) histological analysis of implants retrieved after 174 days from the mice presented in (a–c); Scale bar = 2 mm. (i-j) immunofluorescence analysis of implants retrieved after 174 days from the mice presented in (a–c). (i) DAPI shown blue, Insulin shown in green, and Glucagon shown in red. Scale bar = 50 µm (j) DAPI shown blue, NKX6.1 shown in green, and C-peptide shown in red. Scale bar = 50 µm. Error bars, mean ± s.e.m. N = 5 mice per treatment and all experiments were performed at least three times for a total of 15 mice per treatment. Insulin extraction data: one-way ANOVA with Bonferroni multiple comparison correction, = p < 0.0001.