Optimization of Neutral Protease to Collagenase Activity Ratio for Islet of Langerhans Isolation P. Bucher, D. Bosco, Z. Mathe, D. Matthey-Doret, A. Andres, M. Kurfuerst, N. Ra ¨ msch-Gu ¨ nther, L. Bu ¨ hler, P. Morel, and T. Berney ABSTRACT Aim. The optimal neutral protease to collagenase activity ratio has not been determined for islet isolation. We evaluated a new highly purified collagenase that can be blended with predetermined amounts of neutral protease (NP). Methods. Islets were isolated from 7 groups of Sprague-Dawley rats. In group I, collagenase type XI (Sigma) at 2 mg/mL, and, in group II, Liberase at 0.6 mg/mL (2.4 PZ- U/mL; Roche) were used as controls. In groups III to VII, collagenase NB1 0.6 mg/mL (2.4 PZ-U/mL; Serva Electrophoresis) was used with increasing amounts of added NP. The NP to collagenase activity ratio (DMC-U/PZ-U) increased from 0.5% in group III to 2.0% in group VII. Results. Mean islet equivalent (IE) yields per rat were 1367, 1755, 597, 895, 1712, 1043, and 905 in groups I to VII. IE yields were maximal at DMC-U/PZ-U = 1.2%. Islet morphology was influenced by NP concentration with decreasing numbers of trapped islets and increasing numbers of fragmented islets as NP contents increased. Cytokine release, islet cell apoptosis, and in vitro function were significantly better in groups III to VII as compared with groups I and II. Conclusion. NP is a crucial additive to collagenase for islet isolation. Optimization of the NP to collagenase activity ratio (1.2% in this model) improves yields and morphology after islet isolation. A DVANCES in the rate of success of islet isolation are due in part to the availability of new purified enzyme blends. The development of Liberase (Roche) has been a significant improvement in that regard. 1,2 However, this enzyme is still associated with significant lot-to-lot variabil- ity, notably in terms of the relative contents of its compo- nents. The need for additives to collagenase has been shown to be required for pancreatic digestion. 3 However, the optimal neutral protease (NP) to collagenase activity ratio has not been determined for islet isolation. 4 In this study, we evaluated a new enzyme preparation. This product is composed of highly purified collagenase class I and class II, which can be reproducibly blended with predetermined amounts of separately packaged NP. MATERIALS AND METHODS Islets were isolated from the pancreata of 7 groups of Sprague- Dawley rats (3 isolations from 6 rats/group), using different enzyme blends. In group I, collagenase type XI (Sigma) was used at 2 mg/mL; in group II, Liberase was used at 0.6 mg/mL (2.4 PZ-U/ml; Roche). In groups III to VII, collagenase NB1 was used at 0.6 mg/mL (2.4 PZ-U/mL; Serva Electrophoresis). In group III, no NP was added, for a final NP to collagenase activity ratio (DMC-U/ PZ-U) of 0.5%. In groups IV to VII, neutral protease NB (Serva) was added for final DMC-U/PZ-U ratios of 0.85%, 1.2%, 1.5%, and 2%, respectively. Endotoxin contents of enzyme preparations was determined using the LAL chromogenic test (Biowhitaker). Islet yields, purity, and viability were analyzed. Aliquots of 100 islet equivalents (IE) were cultured in 2 mL From the Cell Isolation and Transplantation Center, Depart- ment of Surgery, Geneva University Hospitals, Geneva, Switzer- land, and Nordmark Arzneimittel GmbH, Uetersen, Germany. This work was funded in part by grant 32-061873.00 from the Swiss National Science Foundation. Address reprint requests to T. Berney, MD, MSc, Cell Isolation and Transplantation Center, Department of Surgery, Geneva University Hospitals, 1211, Geneva 14, Switzerland. E-mail: thierry.berney@hcuge.ch © 2004 by Elsevier Inc. All rights reserved. 0041-1345/04/$–see front matter 360 Park Avenue South, New York, NY 10010-1710 doi:10.1016/j.transproceed.2004.04.024 Transplantation Proceedings, 36, 1145–1146 (2004) 1145