Endocrine Abstracts (2018) 57 003 | DOI: 10.1530/endoabs.57.003

Use of 3D culture systems to generate human induced pluripotent stem cell-derived [beta]-cells in vitro

Fantuzzi Federica1,2, Toivonen Sanna1, Schiavo Andrea Alex1, Pachera Nathalie1, Rajaei Bahareh1, Cai Ying1, Igoillo-Esteve Mariana1, L Eizirik Decio1 & Cnop Miriam1,3


1ULB Center for Diabetes Research, Université Libre de Bruxelles, Brussels, Belgium; 2Department of Medicine and Surgery, University of Parma, Parma, Italy; 3Division of Endocrinology, Erasmus Hospital, Université Libre de Bruxelles, Brussels, Belgium.


Background and aims: Diabetes currently affects 425 million people worldwide. Pancreatic β-cell failure is central in the development and progression of type 1 and type 2 diabetes. Diabetes research is slowed by the difficulty to study the diseased tissue, i.e. human islet bcells: these are only available in a few donor organ transplantation centers worldwide. β-cells differentiated from human induced pluripotent stem cells (hiPSCs) represent a novel cell source. Current in vitro differentiation strategies enable to generate immature β-cells. The traditional suspension culture used for β-cell differentiation is technically challenging and results in heterogeneously sized aggregates. Smaller and more homogeneously sized aggregates may increase the maturity and function of hiPSC-β-cells. Here we aimed to implement 3D culture in microwells (Aggrewells®, Stem Cells) to control the size of islet-like aggregates. We compared the in vitro characteristics of hiPSC-β-cells produced either in suspension or in microwells.

Methods: We used a published 7-step protocol with slight modifications that sequentially provides hiPSCs with differentiation signals important for pancreatic β-cell development. Until the pancreatic progenitor stage, cells were cultured on Matrigel-coated culture plates, after which cells were transferred either in rotating suspension culture or in microwells. Key markers of b-cell development were assessed across the differentiation by qPCR, immunocytochemistry and FACS. The function of hiPSC-β-cells was assessed by glucose- and forskolin-stimulated insulin secretion.

Results: The transfer of cells into suspension often resulted in the formation of large clumps of cells, leading to loss of 40% of the experiments. The seeding into microwells, in contrast, was always (100%) successful. The successfully transferred hiPSCs differentiated with similar efficiency in suspension and microwells, based on their protein and mRNA expression. The suspension aggregates were bigger and more heterogeneous in size compared to microwell aggregates. The yield of insulin-positive β-cells was the same in microwell (36%) and suspension (38%) aggregates, but there were fewer glucagon-positive α-cells in microwell aggregates (7% vs 19% in suspension). In both conditions 6% of cells were polyhormonal (insulin- and glucagon-positive). Aggregates from both culture systems did not induce insulin secretion when stimulated 16,7 mM glucose. However, they similarly increased insulin secretion by 4-5-fold when stimulated with glucose plus forskolin.

Conclusions: Compared to suspension culture, the microwells have a significantly higher experimental success rate, thereby saving significant costs. hiPSCs differentiate with equal efficiency into b-cells in microwells compared to traditional suspension, but the α-cell differentiation is reduced in the microwell system. Microwell aggregates are smaller and equally sized, which might be advantageous for their further maturation. More research is needed to optimize in vitro hiPSC-β-cell function. Even at the current stage of development, the technology provides us with an unlimited supply of human β-cells that is and will be instrumental to study β-cell demise in diabetes.

Article tools

My recent searches

No recent searches.