The effect of isolation methods and the use of different enzymes on islet yield and in vivo function.

The ability to isolate high-yield pure and viable islets from human cadaver pancreas donors is dependent on donor factor as well as isolation factors. The aim of this study was to examine factors influencing islets recovery and in vivo function with an emphasis on donor and isolation methods as well as to compare the effectiveness of Liberase, widely used in clinical islet isolation, with Serva for the isolation of pure functional islets. The results of 123 islet isolations using Liberase for digestion were compared with those of 113 isolations with Serva. Islet equivalents per gram of tissue were similar between Liberase and Serva (3620 +/- 1858 vs. 4132 +/- 2104, p < 0.2) as well as the percent purity (75 +/- 16 vs. 74 +/- 15, p < 0.9). In vivo function of islets from 71 isolations (Liberase = 45, Serva = 26) were further tested by transplantation into NOD-SCID mice following short-term culture (< 6 days, n = 71). Our data show that both Liberase- and Serva-isolated islets showed similar function results following short-term culture. These data demonstrate that there is no difference in islet yield, purity, and function between the two enzymes. However, when these 71 isolations were analyzed for in vivo function with emphasis on donor factors, cold ischemia time (12.0 +/- 5.3 vs. 15.0 +/- 5.7, p < 0.04), islet integrity (1.6 +/- 0.7 vs. 1.3 +/- 0.5, p < 0.05), and female gender were the only factors that correlated with in vivo function. We also compared the mechanical-shaking method for islets isolation with hand-shaking methods. Our results show that although there is no different in islet yield, purity, and integrity between different enzymes using the same method, hand-shaking method yields more islets with better integrity than mechanical-shaking method.