Endothelial phosphoinositide 3-kinase-ß inactivation confers protection from immune-mediated vascular injury.

Heart transplant and recipient survival are limited by immune cell-mediated injury of the graft vasculature. We examined the role of the phosphoinositide 3-kinase-ß (PI3Kß) isoform in endothelial cells (EC) during coronary vascular immune injury and repair in mice. In minor histocompatibility-antigen mismatched allogeneic heart grafts, a robust immune response was mounted to each wild-type, PI3Kß inhibitor-treated, or endothelial-selective PI3Kß knockout (ECßKO) graft transplanted to wild-type recipients. However, microvascular EC loss and progressive occlusive vasculopathy only developed in control, but not PI3Kß-inactivated hearts. We observed a delay in inflammatory cell infiltration of the ECßKO grafts, particularly in the coronary arteries. Surprisingly, this was accompanied by an impaired display of proinflammatory chemokine and adhesion molecules by the ECßKO ECs. In vitro, tumor necrosis factor α-stimulated endothelial ICAM1 and VCAM1 expression was blocked by PI3Kß inhibition or RNA interference. Selective PI3Kß inhibition also blocked tumor necrosis factor α-stimulated degradation of inhibitor of nuclear factor kappa Bα and nuclear translocation of nuclear factor kappa B p65 in EC. These data identify PI3Kß as a therapeutic target to reduce vascular inflammation and injury.

Imaging Tolerance Induction in Neonatal Mice: Hierarchical Interplay Between Allogeneic Adult and Neonatal Immune Cells.

BACKGROUND: In Medawar's murine neonatal tolerance model, injection of adult semiallogeneic lymphohematopoietic cells (spleen cells [SC] and bone marrow cells [BMC]) tolerizes the neonatal immune system. An eventual clinical application would require fully allogeneic (allo) cells, yet little is known about the complex in vivo/in situ interplay between those cells and the nonconditioned neonatal immune system. METHODS: To this end, labeled adult SC and BMC were injected into allogeneic neonates; interactions between donor and host cells were analyzed and modulated by systematic depletion/inactivation of specific donor and host immune effector cell types. RESULTS: Consistent with effector cell compositions, allo-SC and allo-SC/BMC each induced lethal acute graft-versus-host disease, whereas allo-BMC alone did so infrequently. CD8 T cells from SC inoculum appeared naïve, while those of BMC were more memory-like. Age-dependent, cell-type dominance defined the interplay between adult donor cells and the neonatal host immune system such that if the dominant adult effector type was removed, then the equivalent neonatal one became dominant. Depletion of donor/host peripheral T cells protected against acute graft-versus-host disease and prolonged heart allograft survival; peripheral CD8 T-cell depletion together with CD4 T cell-costimulation blockade induced more robust tolerance. CONCLUSIONS: This comprehensive study provides direct observation of the cellular interplay between allogeneic donor and host immune systems, adds to our previous work with semiallogeneic donor cells, and provides important insights for robust tolerance induction. Induction of transplant tolerance in neonates will likely require "crowd sourcing" of multiple tolerizing cell types and involve depletion of immune effector cells with costimulation blockade.

Failure of neonatal B-cell tolerance induction by ABO-incompatible kidney grafts in piglets.

BACKGROUND: ABO-incompatible (ABOi) infant heart transplantation results in B-cell tolerance to graft A/B antigens, confirming human susceptibility to acquired immunologic or "neonatal" tolerance as described originally in murine models. Starting with this clinical observation, we sought to model neonatal ABOi organ transplantation to allow mechanistic studies of tolerance. METHODS: Plasma anti-A/B antibodies were measured over time in piglets to establish developmental antibody kinetics. Blood group O piglets received kidney allografts from group A (AO-incompatible) or group O (AO-compatible) donors under cyclosporine immunosuppression. Anti-A antibodies were measured serially after transplantation; A/H antigen expression and allograft rejection were assessed in graft biopsies. RESULTS: Anti-A antibodies developed in naïve piglets in a kinetic pattern analogous to human infants; anti-B remained low. After transplantation, anti-A antibodies developed similarly in AO-incompatible and AO-compatible groups and were not suppressed by cyclosporine. A/H antigen expression was persistent in all graft biopsies; however, A/H antigens were not detected in vascular endothelium. Cellular and antibody-mediated rejection was absent or minimal in early and late biopsies in both groups, with one exception. CONCLUSIONS: Naturally delayed isohemagglutinin production in piglets is analogous to the developmental kinetics in human infants. However, in contrast to deficient anti-A antibody production as seen long-term after "A-into-O" infant heart transplant recipients, normal anti-A antibody production after "A-into-O" piglet kidney transplantation indicates that tolerance did not develop despite graft A antigen persistence. These findings suggest that the impact on the host immune system of exposure to nonself ABH antigens during early life in human heart versus porcine kidney grafts may depend on expression in vascular endothelium.

Bioluminescent imaging of a marking transgene and correction of Fabry mice by neonatal injection of recombinant lentiviral vectors.

Successful therapy for many inherited disorders could be improved if the intervention were initiated early. This is especially true for lysosomal storage disorders. Earlier intervention may allow metabolic correction to occur before lipid buildup has irreversible consequences and/or before the immune system mounts limiting responses. We have been developing gene therapy to treat lysosomal storage disorders, especially Fabry disease. We describe studies directed toward metabolic correction in neonatal animals mediated by recombinant lentiviral vectors. To develop this method, we first injected a marking lentiviral vector that engineers expression of luciferase into the temporal vein of recipient neonatal animals. The use of a cooled charged-coupled device camera allowed us to track transgene expression over time in live animals. We observed intense luciferase expression in many tissues, including the brain, that did not diminish over 24 weeks. Next, we injected neonatal Fabry mice a single time with a therapeutic lentiviral vector engineered to express human alpha-galactosidase A. The injection procedure was well tolerated. We observed increased plasma levels of alpha-galactosidase A activity starting at our first plasma collection point (4 weeks). Levels of alpha-galactosidase A activity were found to be significantly elevated in many tissues even after 28 weeks. No immune response was observed against the corrective transgene product. Increased levels of enzyme activity also led to significant reduction of globotriaosylceramide in the liver, spleen, and heart. This approach provides a method to treat lysosomal storage disorders and other disorders before destructive manifestations occur.

Acceptance of third-party cardiac but not skin allografts induced by neonatal exposure to semi-allogeneic lymphohematopoietic cells.

Neonatal tolerance is exclusively donor-specific when assessed by skin allograft survival and in vitro alloreactivity assays. In contrast, we reported previously that acceptance of primarily vascularized cardiac allografts was not donor-specific in C3H/He (C3H, H-2(k)) mice treated as neonates with BALB/c-derived (BALB, H-2(d)) lymphohematopoietic cells, but included third-party C57BL/10 (B10, H-2(b)) allografts. The present study examined whether this unusual pattern is limited to heart grafts in this strain combination, and defined the relative importance of the donor cell H-2(d) haplotype for third-party cardiac allograft acceptance. C3H neonates were injected with (C3HxBALB)F1 bone marrow and spleen cells. Tolerance was assessed at age 8-10 weeks by transplantation of heart or skin allografts from several donor strains, and by in vitro assays of proliferation and cytotoxicity. Additionally, cells from H-2(d) and H-2(b)-expressing strains on BALB or non-BALB minor histocompatibility (miH) antigen backgrounds were tested as tolerizing inocula. Prolonged survival of cardiac grafts from all donor strains was observed in neonatally treated mice, whereas skin grafting and in vitro assays demonstrated donor-specific hyporesponsiveness. Both H-2(d) haplotype and non-H-2 miH background of graft donor and tolerizing cell donor were important to third-party cardiac allograft acceptance. These results suggest that the functional alteration in alloreactivity induced by neonatal alloantigen exposure depends partly on method of assessment.