Encapsulation of islets of Langerhans confers protection against cell-mediated immune destruction and so should allow the transplantation of islets without immunosuppression. Xenotransplantation of encapsulated islets of Langerhans might therefore help overcome problems of human organ donor shortage. Given that islets exposed to sustained hyperglycemia show impaired β-cell function, we set out to determine whether recipient treatment with insulin could improve transplantation success rate. Islets of Langerhans were obtained from Specific Germ-Free (SPF) pig pancreas and cultured overnight. Islets were encapsulated in AN69 fibers and implanted into the peritoneal cavity of diabetic mice. A group of implanted mice was treated with exogenous insulin from day 3 to day 7 after grafting. Islet implantation depressed plasma glucose in all the mice, both insulin treated and untreated. Glycemia slowly increased in the non-insulintreated mice, whereas the decrease observed in the insulin-treated mice was maintained until day 29 of follow-up. We found significant differences between the two groups ( p < 0.05 at day 18 and day 20, p < 0.001 at day 23 and day 29). No improvement of hyperglycemia was observed in diabetic mice implanted with empty fibers. When islet-containing fibers were removed from the peritoneal cavity of mice 1 month after the graft plasma glucose increased markedly. We demonstrate that treatment of recipients with exogenous insulin in the immediate posttransplantation period has a positive effect on β-cell function in transplanted macroencapsulated porcine islets.

As the risk of porcine endogenous retrovirus (PERV) infection is a major obstacle to the xenotransplantation of porcine tissue, we investigated whether an AN69 hollow fibre membrane, used for islets of Langerhans transplantation, could prevent the transfer of PERVs and thus reduce the risk of PERV infection. PK15 cells were used as a PERV source. A specific and highly sensitive RCR was used for detection of a PERV provirus DNA (gag region) and a porcine mtDNA. Human U293 cells were incubated in vitro with encapsulated PK15 cells, concentrated encapsulated PK15 supernatant, or concentrated PK15 supernatant as a control. CD1 mice were implanted in vivo with encapsulated PK15 cells or injected with PK15 supernatant. We found no infection in human cells incubated with either encapsulated PK15 supernatant or in 10 out of 11 samples after coincubation with encapsulated PK15 cells. Infection of human cells was, however, detected in 1 out of 11 samples after coincubation with encapsulated PK15 cells. The presence of PERV provirus DNA and porcine mtDNA was detected in all the investigated tissues of the mice injected with PK15 supernatant and in various tissues of the mice implanted with encapsulated PK15 cells. Four weeks after the last injection of PK15 supernatant or a fiber explantation, no mouse showed any presence of PERV provirus DNA or porcine mtDNA. Our results demonstrate that AN69 hollow fiber membrane will reduce but not abolish the risk of PERV infection. Because the real risk of PERV infection still remains unknown, it is necessary to investigate further the real protection that could be provided by hollow fibers to ensure the safety of clinical xenotransplantation.