Background. Access to free human islet transplantation is currently restricted to a small portion of type 1 diabetic patients, mainly due to limited donor supply and the need of long-term immunosuppression. Islet encapsulation represents a promising alternative, but devices create a diffusion barrier that also limits graft vascularization and oxygenation. To address these issues, we developed a double encapsulation device where cells are encapsulated in high concentration alginate capsules and entrapped in a porous device to allow angiogenesis in vivo. The aim of this work was to test the capacity of our device to host vascularization in vitro and to assess the impact of our encapsulation process on the survival and maturation of stem-cell derived pancreatic cells at different stages.  

Methods. The porous part of our device was made of polyurethane using NaCl particles to control porosity. Cells were encapsulated by emulsion and internal gelation (Hoesli et al., 2011) in high-concentration alginate (5%) capsules previously shown to create a barrier to antibody penetration (Hoesli et al., 2012). Pseudo-islets (MIN6) were encapsulated in alginate capsules, mixed with a collagen solution containing Green Fluorescent Protein (GFP)-HUVECs and injected in the device. Devices were cultured for 5 days and HUVECs tube formation was assessed tracking GFP fluorescence. H1 human embryonic stem cells were differentiated towards beta cells following a 7-stage protocol described in the literature (Rezania et al., 2014; Balboa et al., 2022). Stage 4 (S4), S5 and S6 cells were encapsulated in alginate capsules. Cell viability was observed after 1 and 7 days after encapsulation (n=3). At S7+10 days, differentiation markers in S6 encapsulated cells were compared to free cells (n=1).

Results. HUVECs cells formed tubes within our device after 5 days of culture in vitro. Ongoing transplantation experiments in C57BL/6 mice are going to probe whether this is reflected in vivo. No increase of cell death was observed when stem cell-derived pseudo-islets were encapsulated in alginate capsules at S5 and S6. However, we observed an increase of death when encapsulation occurred at S4. Preliminary data suggest that encapsulation of S6 cells increased the population of C-peptide+/NKX6.1+ cells in encapsulated cells compared to free cells (27.3 vs 21.1% at S7+10D, n=1).

Conclusion. These promising preliminary results suggest that our double encapsulation device could (1) be properly vascularized providing sufficient oxygenation to encapsulated cells and (2) provide a 3D environment improving the differentiation/maturation of immature stem cell-derived beta cells.