Neonatal islets from human PD-L1 transgenic pigs reduce immune cell activation and cellular rejection in humanized nonobese diabetic-scid IL2rγnull mice.

Strong xenorejection limits the clinical application of porcine islet transplantation in type 1 diabetes. Targeting T cell-mediated rejection is one of the main approaches to improve long-term graft survival. Here we study engraftment and survival of porcine islet cells expressing human programmed cell death ligand-1 (hPD-L1) in a humanized mouse model. Neonatal islet-like clusters (NPICCs) from transgenic hPD-L1 (hPD-L1-Tg) and wild-type (Wt) pigs were transplanted into nonobese diabetic-scid IL2rγnull mice stably reconstituted with human immune cells (hPD-L1 n = 10; Wt n = 6). Primary endpoint was development of normoglycemia during a 16-week observation period after transplantation. Secondary endpoints were porcine C-peptide levels and immune cell infiltration. Animals transplanted with hPD-L1-Tg neonatal islet-like clusters achieved a superior normoglycemic rate (50% versus 0%) and significantly higher plasma C-peptide levels as compared to the Wt group, indicating long-term beta cell function. Intracytoplasmic fluorescence-activated cell sorting analysis and immunohistochemistry revealed significantly decreased frequencies of interferonγ-expressing splenic hCD8-positive T cells and reduced intragraft-infiltrating immune cells. We here demonstrate that expression of hPD-L1 provides strong islet xenograft protection without administration of immunosuppressive drugs. These findings support the hypothesis that hPD-L1 has the capacity to control cellular rejection and therefore represents a very promising transgene candidate for clinical porcine islet xenotransplantation.

Transgenic pigs expressing near infrared fluorescent protein-A novel tool for noninvasive imaging of islet xenotransplants.

BACKGROUND: Islet xenotransplantation is a promising concept for beta-cell replacement therapy. Reporter genes for noninvasive monitoring of islet engraftment, graft mass changes, long-term survival, and graft failure support the optimization of transplantation strategies. Near-infrared fluorescent protein (iRFP) is ideal for fluorescence imaging (FI) in tissue, but also for multispectral optoacoustic tomography (MSOT) with an even higher imaging depth. Therefore, we generated reporter pigs ubiquitously expressing iRFP. METHODS: CAG-iRPF720 transgenic reporter pigs were generated by somatic cell nuclear transfer from FACS-selected stable transfected donor cells. Neonatal pig islets (NPIs) were transplanted into streptozotocin-diabetic immunodeficient NOD-scid IL2Rgnull (NSG) mice. FI and MSOT were performed to visualize different numbers of NPIs and to evaluate associations between signal intensity and glycemia. MSOT was also tested in a large animal model. RESULTS: CAG-iRFP transgenic NPIs were functionally equivalent with wild-type NPIs. Four weeks after transplantation under the kidney capsule, FI revealed a twofold higher signal for 4000-NPI compared to 1000-NPI grafts. Ten weeks after transplantation, the fluorescence intensity of the 4000-NPI graft was inversely correlated with glycemia. After intramuscular transplantation into diabetic NSG mice, MSOT revealed clear dose-dependent signals for grafts of 750, 1500, and 3000 NPIs. Dose-dependent MSOT signals were also revealed in a pig model, with stronger signals after subcutaneous (depth ∼6 mm) than after submuscular (depth ∼15 mm) placement of the NPIs. CONCLUSIONS: Islets from CAG-iRFP transgenic pigs are fully functional and accessible to long-term monitoring by state-of-the-art imaging modalities. The novel reporter pigs will support the development and preclinical testing of novel matrices and engraftment strategies for porcine xeno-islets.

Cytokine-inducible promoters to drive dynamic transgene expression: The "Smart Graft" strategy.

BACKGROUND: Ubiquitous expression of T-cell regulatory transgenes such as the cytotoxic T lymphocyte-associated antigen 4 (CTLA4) or the high-affinity variant LEA29Y improves xeno graft survival. Such donor pigs are however immunocompromised and susceptible to infection. Continous high expression of CTLA4 or LEA29Y in the graft could also compromise the health status of recipients. The novel "Smart Graft" strategy is likely to avoid these problems by controlling the expression of T-cell regulatory transgenes as and when required. METHODS: Candidate promoters inducible by inflammatory cytokines were identified by in silico screening for potential NF-κB binding sites. Basal promoter levels and responsiveness to TNFα and IL1ß were quantified by expression of secreted embryonic alkaline phosphatase in cultured cells. Promoters were modified to increase responsiveness by removing regulatory elements or adding SP-1 or NF-κB binding sites and again tested in vitro. The most promising promoters were then assessed in vivo. Porcine cells expressing inducible Renilla luciferase constructs were transplanted into immunodeficient NOD-Scid-IL2 receptor gammanull (NSG) mice. Following engraftment, the recipient's immune system was reconstituted by splenocyte transfer raising an immune response to the porcine xenograft. The resulting induction of promoter activity was detected by in vivo bioimaging. RESULTS: Three human (hTNFAIP1, hVCAM1 and hCCL2), and one porcine promoter (pA20) were chosen for in vitro tests. In all experiments, the semi-synthetic and inducible ELAM promoter as well as the CAG promoter were used as references. In contrast to hTNFAIP1 and hVCAM1 the ELAM, hCCL2 and pA20 promoters showed significant induction after cytokine challenge. The hCCL2 and pA20 promoters were further optimized, resulting in increased responsiveness to TNFα and IL1ß. Cytokine-dependent upregulation of promoter activity was tested in vivo, where the ELAM and the optimized hCCL2 promoters showed a 2-fold upregulation, while one of the improved A20 promoters showed almost 10-fold upregulation. Our results also revealed more than 4-fold cytokine inducibility of the CAG promoter. CONCLUSION: This is the first in vivo comparison of existing and newly designed cytokine-inducible promoters. Optimization of promoter structure resulted in almost 10-fold inducibility of promoter activity. Such a rapid and dynamically regulated response to inflammation and cell damage could reduce initial graft rejection, making the "Smart Graft" approach a useful means of modulating the expression of immune regulatory transgenes to avoid deleterious effects on porcine and human health. Expressing transgenes in this fashion could provide a safer organ for transplantation.

Strong xenoprotective function by single-copy transgenes placed sequentially at a permissive locus.

BACKGROUND: Multiple xenoprotective transgenes are best grouped at a single locus to avoid segregation during breeding and simplify production of donor animals. METHODS: We used transgene stacking to place a human CD55 transgene adjacent to a human heme oxygenase 1 construct at the porcine ROSA26 locus. A transgenic pig was analyzed by PCR, RT-PCR, droplet digital PCR, immunohistochemistry, immunofluorescence, and flow cytometry. Resistance to complement-mediated cell lysis and caspase 3/7 activation were determined in vitro. RESULTS: The ROSA26 locus was retargeted efficiently, and animals were generated by nuclear transfer. RNA and protein analyses revealed abundant expression in all organs analyzed, including pancreatic beta cells. Transgenic porcine kidney fibroblasts were almost completely protected against complement-mediated lysis and showed reduced caspase 3/7 activation. CONCLUSION: Step-by-step placement enables highly expressed single-copy xenoprotective transgenes to be grouped at porcine ROSA26.

Engraftment and reversal of diabetes after intramuscular transplantation of neonatal porcine islet-like clusters.

BACKGROUND: Intraportal infusion is currently the method of choice for clinical islet cell transplantation but suffers from poor efficacy. As the liver may not represent an optimal transplantation site for Langerhans islets, we examined the potential of neonatal porcine islet-like clusters (NPICCs) to engraft in skeletal muscle as an alternative transplantation site. METHODS: Neonatal porcine islet-like clusters were isolated from 2- to 5-day-old piglets and either transplanted under the kidney capsule (s.k.) or injected into the lower hindlimb muscle (i.m.) of streptozotocin-diabetic NOD-SCID IL2rγ(-/-) (NSG) mice. Survival, vascularization, maturation, and functional activity were analyzed by intraperitoneal glucose tolerance testing and immunohistochemical analyses. RESULTS: Intramuscular transplantation of NPICCs resulted in development of normoglycemia and restored glucose homeostasis. Time to reversal of diabetes and glucose tolerance (AUC glucose and AUC insulin) did not significantly differ as compared to s.k. transplantation. Intramuscular grafts exhibited rapid neovascularization and graft composition with cytokeratin-positive ductal cells and beta cells at post-transplant weeks 2 and 8 and after establishment of normoglycemia was comparable in both groups. CONCLUSIONS: Intramuscular injection represents a minimally invasive but efficient alternative for transplantation of NPICCs and, thus, offers an attractive alternative site for xenotransplantation approaches. These findings may have important implications for improving the outcome and the monitoring of pig islet xenotransplantation.

Virus safety of islet cell transplantation from transgenic pigs to marmosets.

Transplantation of pig islet cells for the treatment of diabetes may be a more effective approach compared with the application of insulin. However, before introduction into the clinic, efficacy and safety of this treatment have to be shown. Non-human primate models may be used for this, despite the fact that they are characterised by several limitations. Here we investigate the prevalence of porcine endogenous retroviruses (PERVs), which are present in the genome of all pigs and which may infect human cells, as well as of porcine herpes viruses in donor pigs and their potential transmission to non-human primate recipients. Despite the fact that all three subtypes of PERV were present in all and porcine cytomegalovirus (PCMV) was found in some of the pigs, neither PERVs nor PCMV were found in the recipient animals under the experimental conditions applied. Porcine lymphotropic herpes viruses (PLHV) were not found in the donor pigs, hepatitis E virus (HEV) was not found in the recipients.