195

Silicone rubber membrane devices allow islet culture at 20 times the standard surface density with no adverse effects on viability, recovery, or potency

Efstathios S. Avgoustiniatos¹, Phillip R. Rozak¹, Kate R. Mueller¹, Connor A. Lyons¹, William E. Scott III¹, Eric J. Falde¹, Jennifer P. Kitzmann¹, John R. Wilson², Bernhard J. Hering¹, Klearchos K. Papas¹

¹Surgery, Schulze Diabetes Institute, University of Minnesota, Minneapolis; ²Wilson Wolf Manufacturing, New Brighton, MN, United States

Introduction: Islet culture in T-flasks requires seeding the islets at surface densities below 200 IE/cm² to provide sufficient oxygenation through the media. More than 20 T-175 flasks are needed for large porcine or human islet preparations, especially when impure. Culturing islets on top of gas-permeable (GP) silicone rubber membranes increases proximity of islets to ambient oxygen and allows entire preparations to be cultured in 1-2 flasks. In this work we summarize the characterization of 2-day porcine islet culture in GP silicone rubber devices (Wilson Wolf, New Brighton, MN).

Methods: Porcine islets from 30 donors were cultured for 2 days along with standard culture (T-flask, 200 IE/cm², 1000 IE/mL) paired controls. Conditions were seeded at 100-4000 IE/cm² measured by DNA (1 IE = 10.4 ng DNA). The primary outcomes were islet viability, measured by oxygen consumption rate to DNA ratio (OCR/DNA), DNA recovery, viable tissue (OCR) recovery, and islet potency, assessed with the diabetic nude mouse bioassay (2000 IE under the kidney capsule).

Results: For GP devices, OCR/DNA, DNA recovery, and OCR recovery were similar to paired controls, even at 4000 IE/cm². In contrast, all these outcomes deteriorated with increasing surface density for non-GP devices. For surface densities over 1000 IE/cm² and relative to paired controls, OCR/DNA was 58 ± 6% (mean ± SEM, n = 13, P = 1.1x10^-5) for non-GP devices and 95 ± 4% (n = 25, P = 0.27) for GP devices; DNA recovery was 67 ± 6% (n = 8, P = 2.2x10^-4) for non-GP devices and 114 ± 10% (n = 12, P = 0.21) for GP devices; and OCR recovery was 34 ± 4% (n = 8, P = 8.2x10^-7) for non-GP devices and 113 ± 14% (n = 12, P = 0.36) for GP devices. Islets from 5 preparations cultured at 1000 or 4000 IE/cm² in non-GP and GP devices were transplanted into nude diabetic mice. Diabetes reversal rates were similar for islets cultured at high densities in GP devices (14/16) as for paired control islets (14/15). In sharp contrast, only 1 of 13 mice (P=1.0x10^-5) exhibited delayed diabetes reversal after being transplanted with islets cultured at high densities in non-GP devices.

Conclusion: Culturing islets in GP devices allows the increase of islet seeding density from 200 to 4000 IE/cm² while maintaining islet viability, recovery, and potency relative to paired standard controls. Islets cultured at similar high surface densities in non-GP devices exhibit decreased viability and recovery and fail to reverse diabetes in nude mice.