In Vitro Models for the Pancreatic Islet Using Dynamic Secretion Sensing and Mechanical Stimulation

Abstract

Diabetes, a disease of insulin insufficiency leading to improper glucose delivery to the cells of the body, affects millions worldwide with growing impact. For those most severely affected, an exogenous source of insulin is the only option to maintain glucose homeostasis. Islet transplantation is a recent therapy with the potential to offer insulin independence, but its widespread success is often limited by the availability of human, insulin-producing β cells whose quality has been adequately assessed prior to transplantation, a limitation of both cell source and method for assaying β cell function in a timely fashion. While stem cell technologies are poised to fill in the supply shortage of transplantable cells, the problems with quality control remain unaddressed. Here we aimed to develop a method that could address the latter need to more efficiently assess islet potency within a physiologically-relevant environment. We first recognized that a microphysiological system that can couple dynamic glucose stimulation and continuous insulin secretion is necessary to accelerate quantification of β cell activity. Towards this end, we designed, built, and tested a thermoplastic microfluidic device that automatically captures and positions islets for synchronized glucose perifusion and continuous insulin monitoring of secretions. Inspired by the designated delivery of blood to islets within the human pancreas, the device contains islet traps aligned within perfusion lines that receive glucose pulses in parallel. Flowing a suspension of cadaveric islets into the device confirmed their automatic capture within traps. Use of a fluorescent glucose analog then demonstrated that the delivery of pulses to trapped islets were synchronized within 2 minutes across all trapped islets regardless of islet size. A continuous sensor for insulin was integrated onto the chip downstream of the islet traps as a series of reagent inlets and mixing channels followed by an embedded glass capillary for optical detection of changes in fluorescence anisotropy due to competitive binding. Finally, continuous insulin measurements were conducted from cadaveric human islets trapped on the chip and stimulated with pulses of high glucose and potassium chloride. The on-chip measurements were done in comparison to a standard method for islet potency testing, showing the same peaks of insulin secretion response but with a more efficient insulin quantification. Following from the successful interrogation of islet activity on our microfluidic device, we then asked whether we could better understand microenvironmental cues that would have relevance to improving the quality of stem cell products for use in transplantation. In particular, we hypothesized that changes in substrate stiffness would impact β cells function through focal adhesions and activation of focal adhesion kinase (FAK). Our motivation for this study also came from the clinical finding that tissues surrounding islets within the pancreas tend to stiffen in the progression of diabetes, while β cell function also deteriorates with more advanced pathology. To investigate our hypothesis, we cultured dispersed stem cell-derived β (SC-β) cells on PDMS substrates of different stiffnesses and evaluated the impact on their phenotype and function. We first determined that there were no significant differences in cell attachment, viability, or proliferation as a function of substrate stiffness. Due to the population heterogeneity, we stained SC-β cells adhered to various stiffness substrates for C peptide and glucagon, and determined that there were no significant differences across conditions, although the total proportions of both declined after 1 week of culture. We then investigated the role of substrate stiffness on the formation of focal adhesions, and found that there were patterns of increasing F-actin and pFAK area for higher stiffness, suggesting a higher degree of basal activation of focal adhesions at higher stiffnesses. We also quantified the portion of C peptide area co-localized with pFAK, and determined that there was slightly higher co-localization at the softest substrate. To assess the role of substrate stiffness on SC-β cell function, we utilized a glucose-stimulated insulin secretion assay and found that only the softest had a higher insulin secretion level when stimulated. The findings suggest that substrate stiffness does impact the regulation of insulin secretion, and that this behavior may be explained either by elevated pFAK activation at higher stiffnesses or by the greater degree of C peptide and pFAK activation at lower stiffnesses. With the development of a device to measure β cell potency using coupling of synchronized glucose delivery and continuous insulin secretion, and the important discovery that SC-β cell function is impacted by substrate stiffness, we have contributed to alleviating the main issues that prevent widespread adoption of islet transplantation and may also provide a springboard for important foundational discovery in understanding the etiology of diabetes.