Dynamic insulin secretion is a key focus of β cell research, with microfluidic devices emerging as a potential means to effectively evaluate this performance parameter in vitro. However, most devices maintain isolated islets in suspension despite evidence that collagen-islet interactions are critical to β cell viability and function. When designing a microphysiological system with encapsulated islets, important considerations include 1) sufficient oxygenation, 2) rapid exchange of glucose and insulin between cells and perfusate, and 3) detectable levels of secreted insulin. Computational models provide a means to simulate different device design parameters prior to fabrication, reducing the time and cost of testing multiple physical devices and microenvironmental conditions. For this study, COMSOL was used to model a microfluidic device containing islets macroencapsulated in a fibrillar collagen scaffold. A range of flow rates, collagen fibril densities, and islet-collagen construct dimensions were tested, with outcome measures including spatiotemporal changes in oxygen, glucose, and insulin. Simulations for all designs tested showed maintenance of islet viability as indicated by less than 1% of the construct area falling below the hypoxia-induced dysfunction threshold. Additionally, decreasing the construct thickness and increasing the medium glucose concentration yielded more rapid delivery of the target glucose stimulus to the islets, since glucose transport is primarily driven by diffusion within the construct. By contrast, transport of secreted insulin was found to be flow-limited and the presence of an encapsulation material altered the shape of the insulin secretion curve. Collectively, these results support the integrated use of computational models, together with experimental validation, as an efficient strategy for the creation of next generation microphysiological systems for evaluation of β cell health and function.


E.Vanderlaan: None. A.Buganza tepole: None. S.L.Harbin: Other Relationship; GeniPhys, Inc.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.