(84d) The Effect of Alginate Capsule Composition on Pancreatic Differentiation of Human Embryonic Stem Cells

Introduction: Type 1 diabetes (T1D) is an autoimmune disease affecting millions of people worldwide wherein the beta cells of the pancreas are destroyed resulting in insulin dependence. Shortage of donor islets combined with immune rejection limits islet transplantation from becoming a viable therapy. We have shown previously that calcium alginate encapsulation for differentiation of human embryonic stem cells (hESCs) results in efficient differentiation to insulin producing cells. Due to the unlimited supply of hESCs and the immunoisolation capability of alginate capsules, this could be used as an alternative treatment option for T1D. Previous studies for islet encapsulation have shown that alginate capsule composition has a significant effect on the materials capacity to immunoisolate. On the other hand, such capsule composition can modify the differentiation of the encapsulated cells. Hence in this study we are evaluating the effect of such material properties, namely cation concentration and type, on pancreatic differentiation of encapsulated hESCs.

Materials and Methods: We developed a stage wise directed differentiation protocol to derive pancreatic cells from hESCs. Definitive endoderm was induced with ActivinA and Wnt3Afor 4 days followed by pancreatic progenitor induction with Cyclopamine and Cyclopamine with Retinoic Acid for 2 days each, respectively. A single cell suspension of ROCK inhibitor treated hESCs were suspended in 1.1% alginate and 0.2% gelatin followed by encapsulation by drop wise addition into a divalent cation bath. hESC were encapsulated using either CaCl₂ or BaCl₂ as the crosslinking solution with alginate composed of low guluronic acid content. The hESCs were differentiated entirely under encapsulation and characterized after completion of pancreatic differentiation. Viability and proliferation was analyzed for each encapsulation configuration. Pancreatic commitment was analyzed by PDX1 gene and protein expression.

 Results: Our results to date show that encapsulation and pancreatic differentiation of hESCs using standard literature concentrations for CaCl₂ (100 mM) or BaCl₂ (20 mM) show little difference in viability by the LIVE/DEAD assay. However, upon completion of differentiation, calcium alginate (CAlg) gels contained a small number of cell aggregates which were large and heterogeneous in shape. The barium alginate (BAlg) gels contained many small cell aggregates which were homogenous. This indicates that the CAlg gel is more compliant to cell migration and proliferation, while the BAlg gel is more restrictive. Gel stiffness measurements by atomic force microscopy show the BAlg gel to be approximately 2 fold stiffer than the CAlg gel. Gene expression analysis for PDX1 was significantly higher when hESCs were differentiated in CAlg gels. Lowering the BaCl₂ concentration to 10 mM yielded a gel stiffness similar to the CaCl₂ gel, and subsequently resulted in increased differentiation.

Discussion: Our results indicate that even in the presence of chemical factors, substrate stiffness greatly affects the efficiency of pancreatic differentiation. Many protocols in the literature use barium alginate for islet encapsulation due to its increased immunoisolation. However, using this gel for pancreatic differentiation of hESC hinders differentiation. Further studies are currently underway to evaluate different alginate types and cations, as well as using finite element modeling to characterize the local stresses felt by the cell aggregates and how gel stiffness (or composition) affects this.