Islet Microphysiology System to Model Type 2 Diabetes

Introduction: Type 2 Diabetes (T2D) is a chronic metabolic disorder that poses a significant global health burden. T2D is complex in its prognosis and presents a major challenge in development of efficient therapeutic strategies because of lack of appropriate in vitro models that can accurately mimic the physiological conditions of pancreatic islets in their natural (in vivo) environment. Additionally, the animal models that have largely been used for testing new T2D treatments don’t fully represent human physiology and disease, leading to only a 10% success rate of drug treatments that enter into clinical trials. To combat this issue, micro-physiological systems (MPS) have gained importance due to their capability of accurately reproducing physiological microenvironment of several tissue and organ systems in vitro. Our group has developed an Islet MPS model specifically targeting to reproduce the phenotype of T2D through relevant metabolic stressors. Central to T2D disease modeling is the long term culture of pancreatic islets with retained metabolic function, which itself is a challenge since islets are susceptible to de-differentiation and loss of function in an in-vitro setting. We have enabled extended culture of islets in the MPS model by novel incorporation of biomaterial scaffold in the device to support the spheroidal configuration of the islets and reduced the shear stress from the perfusion, while still allowing adequate oxygen supply to the metabolically demanding islets.

Materials and

Methods: The Islet MPS system is a 2-channel fluidic device consisting of three borosilicate glass layers with the middle layer containing a porous membrane that separates the top and the bottom chambers. The layers are made of borosilicate glass with two punch holes for inlets and outlets, the middle glass layer has an oval cut-out which contains the polyester (PET) membrane. The top and bottom glass layers have gaskets, which form an airtight chamber with a height of 200μm on either side of the middle layer when assembled. The three layers are held together within a metallic clamp which has the openings for inlet and outlet. This system of layers and clamp are the commercially available “organ-on-chip” system procured Micronit Microtechnologies (OOC 00739, OOC Platform, Enschede, The Netherlands). Islets are encapsulated in alginate droplets patterned on the membrane middle layer prior to loading onto the PANIS platform (Fig.1b) 100-200 IEQs are picked from the suspension culture and re-suspended in 12 μL of 1.1 wt.% alginate (Sigma Aldrich) solution in DMEM/F12 (Gibco™ 11330032). Droplets of this islet-alginate mixture (≈2 μL each) are polymerized by adding 10 mM BaCl2 solution underneath the membrane and allowed to polymerize for 6 minutes. The media is fed through a syringe pump with a flow rate of 30 μL/h using sterilized tubing attached to the inlets and outlets.

Results, Conclusions, and Discussions: The Pancreatic Islet (PANIS) MPS system was developed on a glass-based 3-layer 2-chamber design separated by a porous polyester membrane housing the organ (Fig 1A). We developed a novel strategy of hydrogel-supported islet micropatterning on the middle membrane layer to maintain the islets’ 3D spheroidal morphology, which is essential for islet function (Fig 1B). In this set-up, the primary human islets retained high viability (>95%) and glucose stimulated insulin secretion (GSIS) over 28 days of culture (stimulation index (SI) from 3mM to 16mM glucose >2). Islet phenotype was maintained with ≈70% C-peptide cells and ≈ 21% glucagon (Fig 1C). The progression of type 2 diabetes was modeled by simulating the diseased states of glucotoxicity, lipotoxicity, and glucolipotoxicity with long term exposure to combinations of pathological glucose (20 mM) and free fatty acid concentrations (0.5 mM oleic and palmitic acid). The resulting islets showed lower C-peptide cells (≈ 60%) and loss in GSIS (SI ≈1) (Fig 1D). Secretome from disease conditions show elevated CCL2 (3-6 fold) and IL-8 (>20 fold) which are involved in inflammation and immune cell recruitment; increase in VEGF secretion (2-4 fold) corresponding with CD31 and ROS expression. Transcriptomic analysis of the simulated disease conditions revealed some overlap, yet distinct differences between simulated metabolic stressors (Fig 1E). BaseSpace Correlation Engine provided direct comparison with T2D patient islet-transcriptome, which identified specific simulated disease conditions correlating strongly with T2D clinical pathophysiology. The efficacy of antioxidant drug Resveratrol was subsequently tested for its efficacy to relieve the islet oxidative stress experienced in T2D. In parallel, hiPSC-derived islet organoids were developed to enable personalized model of T2D. As with the primary islets, the hiPSC-islet system retained viability and showed GSIS responsiveness (SI>1.7) over a 14-day period with the direction of simulating T2D conditions. Single cell RNA-sequencing of iPSC-islets on PANIS demonstrate activation of key metabolic signaling pathways, and will be considered further for T2D disease modeling.

Acknowledgements (Optional):