Modification of Polyurethane Scaffolds for Localised Immunosuppression of Subcutaneous Islet Transplantation.

Abstract

Introduction Hepatic islet transplantation, as a curative therapy for Type 1 Diabetes, is suboptimal for islet function and viability. Subcutaneous engraftment presents an attractive alternative, yet, is currently incapable of supporting islet survival without modification. The use of a polyurethane (PU) scaffold has previously been shown to generate a hyper-vascularised dermal layer, and co-opting this may overcome inherent dermal hypoxia and hypovascularity. However, mitigating the cutaneous immune response is crucial. The aim of this study was to evaluate scaffold loading of the immunosuppressant rapamycin (Rapa), generating a localised immunosuppressive microenvironment supporting islet engraftment. Methods PU disks (8mmx2mm) were loaded topically with 2nM of Rapa. Structural changes to the PU were evaluated using SEM. UV-VIS spectroscopy was utilised to detail Rapa absorption (278nm) over 7 days evaluating drug release kinetics. In vitro biocompatibility of the Rapa-PU was investigated through co-culture with BTC-6 murine β-cell line, murine and human islets. Islet viability was evaluated using fluorescein diacetate/propidium iodide (FDA/PI) staining. T-cell inhibitory properties were evaluated in murine and human cells using a MLR and anti-CD3/28 antibody stimulation assay. Murine in vivo biocompatibility of subcutaneously implanted Rapa-PU was assessed with immunohistochemistry. Results/Discussion SEM of Rapa-PU showed increased pore density with surface integrity disruption, due to methanol treatment, and crystal deposition. Rapa-PU release kinetics demonstrated steady release of up to 70% (60nM) in 7 days following the Higuchi and Ritger-Peppas model, indicating a time dependent release caused by scaffold degradation. In vitro no phenotypically, acute, negative effects were observed upon BTC-6 cell, murine and human islet viability with exposure to PU alone and Rapa-PU. T cell proliferation was potently suppressed by the PU alone, however, immunosuppression was also observed from Rapa released from the scaffold through conditioned media. In vivo scaffolds were well tolerated, and blood vessel formation was detected in unloaded and Rapa-PU scaffolds evaluated by CD31 immunofluorescence staining, indicating that Rapa loading does not abrogate angiogenesis and is comparable to control scaffolds. Conclusion PU scaffolds provide a stable platform for the slow elution of rapamycin, neovascularisation and graft survival. Both the scaffold and rapamycin provides local immunosuppression. In doing so, this model could allow for initial local immunosuppression to prevent early immune infiltration of the graft. Additionally, this model may reduce the requirement for initial systemic immunosuppression, and deliver a higher, less toxic, immunosuppression to the site of transplantation.