198

Oxygen modulation in the culture of islets of Langerhans: the use of PFC-PDMS gas permeable membranes

Christopher Fraker, Kerim Gattas-Asfura, Camillo Ricordi, Luca Inverardi, Juan Dominguez-Bendala

Stem Cell Laboratory, Diabetes Research Institute, University of Miami, Miami, FL, United States

Research into the metabolic demands of islets of Langerhans has highlighted their exquisite sensitivity to external oxygen partial pressures. Exposure to oxygen levels that are non-physiological results in islet loss and dysfunction, either the result of hypoxic cell necrosis/apoptosis or hyperoxic free radical mediated damage. There is a need for static culture platforms that allow for the culture of islets at oxygen partial pressures that better approximate their in vitro niche without the sacrifice of culture practicality or islet viability and function.

In this study, we examined the use of novel platforms in the culture of human islets of Langerhans compared to standard methodologies. These perfluorocarbon impregnated polydimthylsiloxane platforms allowed us to tailor environmental oxygen levels to better match the physiological niche of isolated islets and we theorized that doing so would improve islet viability and function. We investigated the effect of physiological oxygenation by comparing overnight loss, oxygen consumption rate, glucose stimulated insulin release, and in vivo efficacy via marginal mass (1,000 IEQ) sub-renal capsular islet transplants in immunodeficient mice.

Our results demonstrated significantly better outcome in all in vitro assessments for islets cultured at physiological pO₂ (8-12%) on PFC-PDMS membranes relative to control conditions. Overnight loss in the PFC-PDMS group was 40% that of control conditions (p = 0.011), oxygen consumption rate index was 1.3 fold that of control conditions (p << 0.01) and the “delta insulin” (stimulated insulin produced – basal insulin produced) was 1.6 fold that of control conditions (p << 0.01). In diabetic mouse transplantation experiments, we observed a non-significant difference in reversal rates (p = 0.12) by log rank (Mantel-Cox) test, although significant (p = 0.04) by the Gehan-Breslow-Wilcoxon test. There was a promising trend in the median reversal time demonstrating a more rapid return to normoglycemia in the PFC-PDMS group (9 vs 14.5 days).