Previous investigations clearly showed that the successful release of islets from the pancreas is mediated by both neutral protease (NP) and collagenase, consisting of subclasses I and 11 showing different capacities to cleave islets from the pancreas. Since no informations about the optimal ratio between class 11 and class 1 collagenase (11/1-ratio) are available yet, the present study sought to evaluate the efficient range for the 11/1-ratio. Methods. Following intraductal pancreas collagenase distension, rat islets were isolated utilizing 20 PZ-U Serva collagenase NB I and 1.0 or 0.4 DMC-U NP. After purification we determined the islet yield (IEQ), viability (trypan-blue exclusion) and function in diabetic nude mice. Methods. Following intraductal pancreas collagenase distension, rat islets were isolated utilizing 20 PZ-U Serva collagenase NB1 and 1.0 or 0.4 DMC-U NP. After purification we determined the islet yield (IEQ), viability (trypan-blue exclusion) and function in diabetic nude mice. Results. At 1.0 DMC-U NP, a II/I-ratio of 2.6, 1.5 or 0.7 yielded 2200 +/- 280, 2185 +/- 420, and 2205 +/- 90 IEQ, respectively (ns). Viability varied between 70% and 80% (ns). Digestion time was significantly lowest (P < .05) using a II/I-ratio of 0.7. Utilization of 0.4 DMC-U NP resulted in a viability of > 98% among all experimental groups (P <.001 vs 1.0 DMC-U). Islet yield decreased at a II/I-ratio of 2.6 (1520 +/- 120 IEQ, P <.05) and 1.5 (1780 +/- 130 IEQ, ns), but not at 0.7 (2310 +/- 160 IEQ, ns). Again, digestion time was lowest (P <.001) using a II/I-ratio of 0.7. Transplantation into diabetic nude mice demonstrated islet function in all experimental groups. Conclusions. NP significantly affects islet viability. This study indicates that the minimal amount of NP required for efficient islet cleavage depends on the II/I-ratio.