Clinical islet allotransplantation has become an increasingly efficient "routine" therapy in recent years. Shortage of human donor organs leads to porcine pancreatic islets as a potential source for islet xenotransplantation. Yet it is still very difficult to isolate sufficient numbers of intact porcine islets, particularly from young market pigs. In the following study islets were successfully isolated from retired breeders [4806 ± 720 islet equivalents per gram organ (IEQ/g); n = 25; 2-3 years old; RB] and also from young hybrid pigs [2868 ± 260 IEQ/g; n = 65; 4-6 months old; HY] using LiberasePI and a modified version of Ricordi's digestionfiltration technique. As expected, isolations from RB showed significantly better results (p < 0.002). A retrospective histological analysis of almost all donor pancreases showed that the majority of organs from RB (80%) contained mainly large islets (diameter >200 µm), in contrast to only 35% of all pancreases from HY. Remarkably, the islet size in situ, regardless whether detected in RB or HY, strongly determined the isolation result. A donor organ with predominantly large islets resulted in significantly higher numbers of IEQs compared with a donor organ with predominantly small islets [RB Large Islets : 5680 ± 3,318 IEQ/g (n = 20); RB Small Islets : 1353 ± 427 IEQ/g (n = 5); p < 0.02]. In addition, isolation results were strongly influenced by the quality of the LiberasePI batch, and therefore single batch testing is invariably required. Purification was performed using Ficoll or OptiPrep TM density gradient centrifugation manually or in the COBE cell processor. Although islet purity was highest when OptiPrep TM was used, final islet yields did not differ between the different purification methods. Our study demonstrates that islet size in situ is an extremely critical parameter for highly successful islet isolation; consequently, we are now performing a morphological screening of each donor organ prior to the isolation process. Under these conditions highly successful isolations can reliably be performed even from young market pigs.

Growth factors have been shown to modulate the complex cascade of wound healing, however, interaction between different growth factors during dermal and epidermal regeneration is still not entirely defined. We have recently shown that exogenous liposomal gene transfer of cDNA results in physiologic expression and response in an acute wound. In the present study we determined the interaction between insulin-like growth factor-I (IGF-I), a mesenchymal growth factor, administered as liposomal cDNA, with other dermal and epidermal growth factors on collagen synthesis in an acute wound. Sprague-Dawley rats were given a scald burn to inflict an acute wound and divided into two groups to receive weekly subcutaneous injections of liposomes plus a beta-galactosidase containing plasmid (Lac Z [0.2 microg, vehicle]), or liposomes plus the IGF-I cDNA containing plasmid (2.2 microg) and Lac Z (0.2 microg). Immunological assays, histological and immunohistochemical techniques were used to determine growth factor concentration and different types of collagen (I, III, and IV) after IGF-I cDNA gene transfer. IGF-I cDNA transfer accelerated reepithelization and was associated with increased levels of IGF-I, fibroblast growth factor, keratinocyte growth factor, vascular endothelial cell growth factor, and platelet-derived growth factor protein expression. IGF-I cDNA had no effect on transforming growth factor-beta. IGF-I cDNA significantly increased type IV collagen while it had no effect on types I and III collagen. Exogenously administered IGF-I cDNA increased protein concentrations of keratinocyte growth factor, fibroblast growth factor, platelet-derived growth factor, and type IV collagen. We conclude that liposomal IGF-I gene transfer can accelerate wound healing without causing an increase in types I and III collagen expression.