Gene Therapy in Organ Transplantation

Home , Transplant rejection

Gene Therapy (2003) 10, 605–611 & 2003 Nature Publishing Group All rights reserved 0969-7128/03 $25.00 www.nature.com/gt REVIEW Gene Therapy Progress and Prospects: Gene therapy in organ transplantation

J Bagley and J Iacomini Transplantation Biology Research Center, Massachusetts General Hospital and Harvard Medical School, MGH-East, Boston, MA 02129, USA

One major complication facing organ transplant recipients is sion. Alternatively, gene therapy approaches could eliminate the requirement for life-long systemic immunosuppression to the requirement for general immunosuppression by allowing prevent rejection, which is associated with an increased the induction of donor-speciﬁc tolerance. Gene therapy incidence of malignancy and susceptibility to opportunistic interventions may also be able to prevent graft damage infections. Gene therapy has the potential to eliminate owing to nonimmune-mediated graft loss or injury and problems associated with immunosuppression by allowing prevent chronic rejection. This review will focus on recent the production of immunomodulatory proteins in the donor progress in preventing transplant rejection by gene therapy. grafts resulting in local rather than systemic immunosuppres- Gene Therapy (2003) 10, 605–611. doi:10.1038/sj.gt.3302020

Keywords: transplantation; gene transfer; tolerance

In brief

Progress * Transduction of dendritic cells with genes encoding immunomodulatory proteins can moderately pro- * Gene therapy-mediated CD28/B7 costimulatory long graft survival. blockade can prolong graft survival, but results in * Improved graft survival through antiapoptotic and nonspecific immunosuppression. antiproliferative gene therapy. * Gene therapy-mediated CD40/CD154 costimulatory blockade can prevent acute allograft rejection, but Prospects does not prevent chronic alloreactivity. * Hyporesponsiveness to virally encoded MHC genes * Better understanding of activation of alloreactive has been induced in both rodent and large animal cells will allow for the development of more targeted models. gene therapy aimed at preventing this activation. * Induction of stable long-term T-cell tolerance to * Increased understanding of costimulatory pathways MHC class I antigens through central deletion of will reveal additional targets for gene therapy alloreactive T cells following induction of molecular interventions. chimerism. * Elucidation of the nature of regulatory cells may lead * Evidence for induction of regulatory cells that can to gene therapy approaches designed to efficiently prevent allograft rejection in gene therapy models. generate these populations. * Induction of B-cell tolerance by molecular chimer- * Induction of tolerance through molecular chimerism ism. will be tested in large animal models. * Induction of tolerance in sensitized recipients. * Development of improved vectors for transduction * Gene therapy-mediated cytokine production, or of terminally differentiated tissue will improve immune deviation, prolongs graft survival, but does efficacy of gene therapy approaches that depend not prevent rejection. on intracellular protein expression.

Gene therapy-mediated CD28/B7 activated and participate in transplant rejection, they must first receive a signal through the T-cell receptor costimulatory blockade can prolong graft (TCR), which occurs after either direct or indirect survival, but results in nonspecific recognition of alloantigens on the surface of antigen- immunosuppression presenting cells (APCs). However, in order for T cells to become fully activated and acquire effector The major effectors of transplant rejection are host function, such as the ability to produce cytokines, they CD4 and CD8 T cells. In order for T cells to become must also receive signals from the interaction of costimulatory molecules expressed on their surface with ligands expressed on APCs. Costimulatory Correspondence: Dr J Iacomini, Transplantation Biology Research Center, Massachusetts General Hospital and Harvard Medical School, MGH-East, molecule interactions can result in T-cell activation Building 149, 13th Street, Boston, MA 02129, USA or inhibition. For example, following TCR ligation, Received 24 September 2002; accepted 30 January 2003 the interaction of the costimulatory molecule CD28 Gene therapy in organ transplantation J Bagley and J Iacomini 606 expressed on T cells with CD80 and CD86 (B7.1 and B7.2, islet function after cessation of CTLA-4Ig expression respectively) expressed on APCs results in T-cell activa- was not examined, so it is not possible to determine tion (reviewed in Sharpe and Freeman1). In contrast, if long-term survival occurred. Skin allograft survival signaling through cytolytic T-lymphocyte-associated was also prolonged, however, all skin grafts were antigen 4 (CTLA-4) expressed on T cells following eventually rejected. In addition, Cre-mediated deletion ligation by CD80 and CD86 downregulates immune of CTLA-4Ig was associated with the restoration of responses. responses to adenovirus, suggesting that long-term T cell costimulatory pathways are central to T-cell hyporesponsiveness was averted using this approach, activation, and therefore several groups have tried effectively allowing one to control the duration of to block or manipulate these pathways using gene immunosuppression. therapy approaches in order to prevent T-cell responses. Blocking the interaction of CD28 with CD80 and CD86 using an immunoglobulin fusion protein Gene therapy-mediated CD40/CD154 (CD40 containing the extracellular portion of CTLA-4 (CTLA-4Ig) can result in immunosuppression in vivo, ligand) costimulatory blockade can prevent and prevent transplant rejection. To overcome acute allograft rejection, but does not prevent systemic immunosuppression as a consequence of chronic alloreactivity CTLA-4Ig administration, groups have tried to prevent transplant rejection by expressing CTLA-4Ig In addition to the CD28-CD80/86 costimulatory locally within transplanted tissues or organs. Expres- pathway discussed above, the interaction of CD40 sion of the gene encoding CTLA-4Ig in donor organs ligand (CD154) expressed on T cells with CD40 on has generally resulted in the prolongation of trans- APCs has also been shown to be an important plant survival, but has not permitted permanent component in the initiation and maintenance of acceptance (reviewed in Guillot et al 2). Recently, T-cell responses. Gene therapy approaches have been indefinite graft survival was achieved in a cardiac developed in which an adenovirus encoded CD40-Ig transplantation model.3 However, in this model, it is fusion protein is used to block intragraft CD40- likely that the transplanted organs themselves may CD154 interactions in order to prevent graft rejection. have participated in the maintenance of hyporespon- Adenoviral-mediated transfer of the gene encoding siveness that was initially induced by transient expres- CD40-Ig into rat livers resulted in long-term survival sion of CTLA-4Ig. Indeed, cardiac allografts themselves following transplantation into allogeneic recipients.9 can be tolerogenic. In experimental systems involving Immunocompetence was tested by analyzing skin transplantation of tissues expressing CTLA-4Ig which allograft survival 120 days after liver transplanta- are not thought to be tolerogenic, such as skin,4 tion, long after CD40-Ig production had ceased. There- hepatocytes,5 corneal grafts,6 and fetal cardiomyocytes,7 fore, it is unclear whether recipients were able to only modest graft prolongation was achieved. In these respond to third-party antigens at earlier time points studies, graft rejection was associated with a loss or while CD40-Ig was expressed in the transplanted decrease in gene expression. livers. Expression of CD40-Ig in rat cardiac transplants Analysis of rat cardiac allografts expressing adeno- has also been shown to delay acute rejection, although, virus encoded CTLA-4Ig revealed that while perma- at early time points after transplantation, recipients nent acceptance of the genetically modified transplants of gene-modified hearts exhibited nonspecific immuno- could be achieved in allogeneic hosts, acceptance suppression.10 At later time points, nonspecific immuno- was associated with nonspecific inhibition of T-cell suppression abated and responses to third-party responses to unrelated third-party antigens.3 Responses alloantigens were restored. However, when immune to third-party antigens were diminished immediately responses to third-party antigens were restored, anti- after transplantation, and hyporesponsiveness was donor T-cell activity returned in these animals, and apparent even after a five-fold reduction in serum was associated with chronic rejection.10 The develop- levels of CTLA-4Ig was observed 120 days after ment of chronic rejection suggests that intragraft expres- transplantation. These data suggest that intragraft sion of CD40-Ig needs to be complemented by other expression of CTLA-4Ig can result in systemic immuno- therapeutic strategies to obtain long-term transplant suppression, and therefore long-term expression of survival. this gene product may be detrimental to host immunity. Several other molecules involved in T-cell costimula- In an attempt to overcome the long-term immuno- tion have recently been identified, but the function of suppressive effects of CTLA-4Ig expression, adenovirus many newly discovered costimulatory pathways has not vectors carrying the gene encoding CTLA-4Ig flanked been fully elucidated.1,11 However, it is becoming by two loxP sequences were developed.8 Following apparent that other costimulatory pathways, such as intravenous injection of adenoviruses carrying this the inducible costimulator (ICOS)-B7 related protein-1 novel CTLA-4Ig gene construct, subsequent administra- (B7RP-1) pathway, are important for the activation of tion of adenoviruses carrying the gene encoding effector T cells.12 Indeed, it has been shown that an ICOS- Cre recombinase permitted excision of the CTLA-4Ig Ig recombinant protein used in conjunction with the gene, terminating expression in vivo and vitro. CD40/CD154 blockade may prevent chronic rejection of Pancreatic islets transplanted into the liver of mice cardiac allografts.13 While no gene transfer approach receiving loxP-flanked adenovirus encoded CTLA-4Ig targeting ICOS-B7RP-1 interactions has been described to remained functional 40 days after serum CTLA-4Ig date, inhibiting this costimulatory pathway may repre- was no longer detectable following Cre-mediated sent an interesting approach for preventing chronic graft excision of the CTLA-4Ig gene.8 However, long-term rejection.

Gene Therapy Gene therapy in organ transplantation J Bagley and J Iacomini 607 Hyporesponsiveness to virally encoded MHC The mechanism by which induction of molecular chimerism induces CD8 T-cell tolerance has been genes has been induced in both rodent and recently elucidated using a T-cell receptor transgenic large animal models mouse model. In this system, H-2Kb-specific transgenic CD8 T cells were observed to undergo negative selection It has been known for many years that a state of in the thymus upon encountering bone marrow-derived mixed host–donor hematopoietic cellular chimerism, cells expressing the transduced MHC class I gene.19 induced by allogeneic bone marrow transplantation, Expression of the transduced antigen on T cells appears leads to long-term stable donor-specific tolerance (re- to participate in trafficking alloantigen to the thymus to viewed in Sykes14). Building on the concept of mixed facilitate negative selection, as suggested by a study that chimerism, it has been suggested that a state of demonstrated that in the absence of transduced MHC molecular chimerism involving the transfer of genes class I expression on T and B cells, mice failed to become encoding allogeneic donor-type MHC proteins, or tolerant.20 Since central deletion of alloreactive T cells is other antigens, into autologous hematopoietic stem the most stable form of tolerance, the proof that deletion cells may also result in tolerance.15 The induction of occurs in molecular chimeras is encouraging. Collec- molecular chimerism through genetic modification of tively, these data strongly suggest that gene therapy can autologous hematopoietic stem cells has the potential to be used to permanently reshape the T-cell repertoire. induce donor-specific tolerance without the complica- tions associated with allogeneic bone marrow transplants, such as graft-versus-host disease. Expression of retrovirally transduced allogeneic donor-type MHC Evidence for induction of regulatory cells that genes in bone marrow-derived cells has been shown to can prevent allograft rejection in gene therapy be sufficient to induce donor-specific hyporesponsiveness to the introduced gene product, allowing for models prolonged survival of cardiac and skin allografts without In recent years, a significant amount of evidence affecting rejection of third-party control grafts.15 Hypor- suggests that subpopulations of CD4 T cells can suppress esponsiveness to allogeneic renal transplants has also allograft rejection. These regulatory cells can be induced been induced in pigs following induction of molecular using a variety of approaches including administration chimerism, although it is not clear to what extent the of nondepleting anti-CD4 antibodies and exposure to a transplanted organ itself contributed to establishing tolerizing antigen in the form of donor-specific transfu- allograft acceptance.16 The induction of hyporesponsive- sions.21–23 Often regulatory cells are able to induce ness to marker genes expressed in bone marrow-derived tolerance to both the tolerizing antigen and third-party cells has also been achieved in rhesus macaques.17 Thus, antigens, as long as they are both expressed on the same the induction of molecular chimerism has been shown to graft, suggesting that inhibition occurs locally.21,22 Re- be capable of inducing hyporesponsiveness in multiple cently, it has been shown that treatment of immuno- animal models. competent mice with nondepleting anti-T-cell antibodies together with syngeneic bone marrow infected with adenoviruses carrying an allogeneic MHC class I gene can lead to the acceptance of fully-allogeneic cardiac Induction of stable long-term T-cell tolerance transplants which share the same MHC class I antigen to MHC class I antigens through central carried by the adenovirus construct.24 This approach deletion of alloreactive T cells following essentially mimics results obtained using donor-specific transfusion, coupled with the use of nondepleting anti-T- induction of molecular chimerism cell antibodies to induce tolerance. Although the me- The ability to induce hyporesponsiveness through chanism by which acceptance was achieved is unknown, molecular chimerism has been well established; transplant acceptance was observed even though gene however, it has also been shown that this hyporespon- expression was very short lived in vivo. These experi- siveness can be abrogated by providing T-cell help ments suggest that the use of gene therapy-modified (reviewed in Bagley et al15). In order for molecular bone marrow may be capable of inducing regulatory T chimerism to become clinically relevant, it is important cells that can inhibit the responses to the transduced to demonstrate that stable long-term T-cell tolerance gene product as well as additional transplantation can be achieved using this approach. Recently, it antigens when combined with other therapies. This is a has been shown that efficient expression of an potentially important extension of the use of molecular allogeneic MHC class I gene in bone marrow-derived chimerism. However, it is not clear whether similar cells is sufficient to allow for permanent survival approaches would work in models in which the organ of MHC class I disparate skin grafts without affect- itself does not participate in the maintenance of ing rejection of third-party grafts.18 Cytotoxic T cells tolerance, or prevent chronic rejection. capable of lysing donor-type targets remained undetectable in vitro even after rigorous antigen challenge. These results are in contrast to previous Induction of B-cell tolerance by molecular studies in which T-cell hyporesponsiveness induced chimerism by genetic engineering of bone marrow could be broken by provision of T-cell help. These results suggest In addition to inducing T-cell tolerance, molecular that donor-specific tolerance can be established by chimerism has been shown to be capable of inducing inducing molecular chimerism. B-cell tolerance. Natural antibodies specific for the

Gene Therapy Gene therapy in organ transplantation J Bagley and J Iacomini 608 carbohydrate epitope Gala1-3Galb1-4GlcNac-R (aGal) unacceptable side effects, gene therapy approaches were are the main mediators of hyperacute xenograft rejection developed to test the hypothesis that modifying the local in pig to primate xenotransplantation. Mutant mice, graft environment to promote Th2 rather than Th1 which lack a(1,3)galactosyltransferase (aGT), the enzyme responses would result in prolonged transplant survival. that synthesizes the aGal epitope, produce aGal-specific The ability of several Th2-type cytokines to prolong natural antibodies, as do humans. Expression of porcine allograft graft survival when expressed locally has been aGT in bone marrow-derived cells of aGT knockout mice assessed both with and without additional immunosup- has been shown to prevent the production aGal-reactive pression. In general, prolongation of graft survival has natural antibodies, resulting in stable long-term tolerance been observed using this approach. However, expression to aGal even after rigorous antigen challenge.15,25,26 of cytokine genes within donor tissues did not lead to Importantly, tolerance remained intact even after mice permanent graft acceptance. More recently, it has been received cardiac transplants from wild-type mice that shown that localized liposome-mediated IL-10 gene expressed aGal, and aGal antibody-mediated rejection transfer into rabbit cardiac transplants could induce was prevented in molecular chimeras.27 Analysis of B alloreactive T-cell apoptosis and prolong cardiac allograft cells from mice reconstituted with aGT-transduced bone survival.30 In addition, transfer of the gene encoding marrow revealed that B cells which produce aGal- viral IL-10 delayed graft rejection in a rat model.31 specific antibodies were eliminated from the immunolo- However, in both these studies, prolongation was gical repertoire following gene therapy. Collectively, extremely modest, and it is therefore unclear what kind these data suggest that gene therapy may be used to of clinical significance can be attached to such results in induce both B- and T-cell tolerance through the establish- model systems that are relatively sensitive to tolerance ment of molecular chimerism. induction.

Induction of tolerance in sensitized recipients Transduction of dendritic cells with genes The extent to which pre-existing host immune responses encoding immunomodulatory proteins can can affect the induction of molecular chimerism and establishment of donor-specific tolerance is an important moderately prolong graft survival issue. Many patients have been presensitized to organ An increased understanding of the role of dendritic cells allografts by blood transfusions or previous transplants, in T-cell activation has led to an interest in modification and all humans have natural antibodies to aGal epitopes of dendritic cells to induce tolerance. While mature on xenogeneic organs. To determine the extent to which dendritic cells express high levels of costimulatory a preformed immune response was a barrier to the molecules and induce strong T-cell responses, there is induction of molecular chimerism, aGT knockout mice some evidence that dendritic cells expressing low levels were immunized with aGal-expressing pig cells. Since 28 of costimulatory molecules can induce anergy in T- aGal is a T-dependent antigen this immunization cells.32 Attempts have been made to modify graft resulted in T-cell priming and increased titers of anti- rejection by genetically modifying dendritic cells to Gal antibodies. Despite high titers of serum aGal-specific express cytokines associated with delayed graft rejection. antibodies in immunized hosts, molecular chimerism Retroviral delivery of TGF-b into myeloid dendritic cells could be established by increasing the dose of trans- has been shown to decrease their ability to stimulate duced bone marrow used for reconstitution of lethally 29 alloreactive cells, resulting in moderately prolonged graft irradiated recipient. Once molecular chimerism was survival of cardiac allografts in mice.33 Human myeloid established in sensitized hosts, tolerance to aGal resulted dendritic cells transduced with the gene encoding IL-10 and production of aGal-specific antibodies ceased. These prolonged human skin allograft survival in a humanized data demonstrated that the induction of molecular NOD-scid chimeric model.34 However, other studies chimerism could be used to reshape the pre-existing B- have indicated that adenovirus transduction of the IL- cell repertoire in appropriately conditioned sensitized 10 gene into murine myeloid dendritic cells can enhance hosts. alloreactive responses to cardiac grafts.35 Overall, the results obtained with cytokine-gene-transduced dendritic cells were similar to those obtained with transduced Gene therapy-mediated cytokine production, organs. Both resulted in only a modest increase in or immune deviation, prolongs graft survival, transplant survival. Similar results have been observed but does not prevent rejection in models where dendritic cells were transduced with genes encoding CTLA-4Ig.36 Acute rejection events in immunosuppressed patients Dendritic cells have also been genetically engineered correlate with intragraft production of T helper type 1 to express Fas ligand (CD95L).37–39 Engagement of Fas (Th1) cytokines, such as IFN-g, while the lack of acute (CD95) on the surface of T cells by CD95L leads to the rejection is associated with production of T helper type 2 induction of apoptosis in activated T cells. Apoptosis cytokines (Th2), such as IL-10. It has therefore been induced by CD95/CD95L interactions is thought to play hypothesized that production of Th2 cytokines might act a role in the establishment of immunoprivileged sites to downregulate the immune response to organ allo- such as the eye and testis, and may be involved in the grafts. Thus, the concept of immune deviation was killing of CD4 T cells. Dendritic cells genetically proposed as a way to prevent organ allograft rejection by engineered to express CD95L on their surface are able fostering a Th2-type response, rather than Th1. Since to inhibit alloreactive T-cell proliferation in vitro, and systemic administration of cytokines often results in cause a slight prolongation of cardiac graft survival

Gene Therapy Gene therapy in organ transplantation J Bagley and J Iacomini 609 when administered in vivo.38 However, as discussed degraded rapidly when given exogenously. The expres- above, since mouse cardiac transplants are relatively sion of scavenging factors locally through genetic sensitive to tolerance induction, it is unclear how engineering of transplants would therefore be one way significant these results will be in more rigorous to protect organs and tissues. In a rat liver transplanta- transplantation models. tion model, introduction of the gene encoding copper– zinc superoxide dismutase through adenoviral transduction allowed for the survival of 100% of recipients, Improved graft survival through antiapoptotic whereas only 25% of mock-treated controls survived. and antiproliferative gene therapy Expression of copper–zinc superoxide dismutase in transplanted livers also significantly reduced necrosis In addition to host immunity, other nonimmunological within the transplant.40,41 These data suggest that factors play a role in transplant survival. Therefore, expression of copper–zinc superoxide dismutase de- approaches have been developed to protect donor organ creases injury resulting from the generation of reactive or tissue grafts at the time of transplantation from oxygen species during transplantation. Similarly, gene damage owing to nonimmune-mediated inflammation therapy approaches using other cytoprotective genes and ischemia–reperfusion injury. The use of gene transfer such as hemeoxygenase-142,43 and catalase44 have also to prevent damage that occurs at the time of transplant been shown to protect organs against the effects of has the advantage that it does not require long-term gene ischemia–reperfusion. Expression of hemeoxygenase-1 in expression, making these therapies potentially clinically rat liver transplants following adenovirus-mediated gene relevant. transfer resulted in an increased survival of recipients A significant proportion of cellular transplants such as from 50% in controls to 80% in those receiving islets or hepatocytes are lost due to anoxia/ischemia hemeoxygenase-1-transduced livers. Gene transfer in reperfusion injuries at the time of transplantation that these studies was performed from 4 to 24 h in advance trigger the generation of reactive oxygen species leading of transplant, which may be compatible with clinical to cellular death and localized inflammation. Endogen- application of these techniques. ous scavenger systems can eliminate toxic radicals. The overexpression of antiapoptotic genes in However, the components of these systems are usually transplanted tissue may also protect the graft from

Figure 1 Preventing transplant rejection by gene therapy. Gene therapy-based strategies can be used to prevent activation of alloreactive host T cells in peripheral lymphoid tissue, such as the lymph node (shown), by blocking costimulatory molecule interactions between T cells and APCs (a). Such interventions leave T cells in a non-responsive state, and can also result in anergy. Reconstitution of conditioned hosts with autologous hematopoietic stem cells engineered to express donor type transplantation antigens using retroviruses allows for expression of donor type antigen on bone marrow-derived cells (b). Expression of retrovirally transduced antigen on bone marrow-derived cells within the recipient thymus mediates negative selection of newly formed alloreactive T cells (c). It is unclear whether expression in the thymus can also lead to the generation of regulatory T cells capable of controlling allograft rejection. Direct modiﬁcation of the donor graft itself (d) using a wide variety of gene delivery systems prior to transplantation offers the possibility of reducing graft damage owing to ischemia–reperfusion injury. This approach has also been proposed as a method to allow the destruction of alloreactiveT cells that reach the graft.

Gene Therapy Gene therapy in organ transplantation J Bagley and J Iacomini 610 nonimmune as well as immune-mediated injury. References Bcl-2, an antiapoptotic cell survival factor, blocks the release of cytochrome 3 from the mitochondria 1 Sharpe AH and Freeman GJ. The B7-CD28 superfamily. Nat Rev and subsequent activation of caspases, proteases Immunol 2002; 2: 116–126. involved in cell death. The overexpression of bcl-2 2 Guillot C et al. Gene therapy in transplantation in the year 2000: in macaque pancreatic islets was able to enhance moving towards clinical applications? Gene Therapy 2000; 7: insulin production after transplant into diabetic 14–19. SCID mice45 and protect porcine islets after exposure 3 Guillot C et al. Tolerance to cardiac allografts via local and to rhesus monkey serum.46 In addition, the expression systemic mechanisms after adenovirus-mediated CTLA4Ig ex- of bcl-2 protected human endothelial cells from CTL pression. J Immunol 2000; 164: 5258–5268. activity.47 However, since the function of bcl-2 is 4 Malm H et al. CTLA4ig induces long-term graft survival of intracellular, the effectiveness of this approach is allogeneic skin grafts and totally inhibits T-cell proliferation in dependent on a high transduction efficiency that will LFA-1-deficient mice. Transplantation 2002; 73: 293–297. be more difficult to achieve in the terminally differen- 5 Reddy B et al. The effect of CD28/B7 blockade on alloreactive T tiated cells of vascularized organs. and B cells after liver cell transplantation. Transplantation 2001; 71: 801–811. 6 Comer RM et al. Effect of administration of CTLA4-Ig as protein or cDNA on corneal allograft survival. Invest Ophthalmol Vis Sci Prospects 2002; 43: 1095–1103. In the coming years, in order for gene therapy to be 7LiTSet al. Prolonged survival of xenograft fetal cardiomyocytes by adenovirus-mediated CTLA4-Ig expression. Transplantation used to prevent transplant rejection clinically, it 2001; 72: 1983–1985. will be crucial to gain a better understanding of how 8 Takehara M et al. Long-term acceptance of allografts by in vivo alloreactive T cells are activated and regulated, so gene transfer of regulatable adenovirus vector containing that rational gene therapy interventions can be CTLA4IgG and loxP. Hum Gene Ther 2001; 12: 415–426. designed to prevent transplant rejection (Figure 1). 9 Nomura M et al. Induction of donor-specific tolerance by Blockade of yet to be defined costimulatory pathways adenovirus-mediated CD40Ig gene therapy in rat liver trans- needs to be evaluated in order to determine if plantation. Transplantation 2002; 73: 1403–1410. targeting these pathways using gene therapy approa- 10 Guillot C et al. Prolonged blockade of CD40–CD40 ligand ches will exhibit utility for clinical transplants. In interactions by gene transfer of CD40Ig results in long-term addition, a greater understanding of the mechanisms heart allograft survival and donor-specific hyporesponsiveness, by which regulatory T cells are induced needs to be but does not prevent chronic rejection. J Immunol 2002; 168: achieved in order to design strategies to exploit 1600–1609. these cells to achieve transplant acceptance. Tolerance 11 Greenwald RJ et al. CTLA-4 regulates induction of anergy in vivo. induction remains a major goal in the field, and if Immunity 2001; 14: 145–155. it can be achieved clinically may represent the most 12 Coyle AJ, Gutierrez-Ramos JC. The expanding B7 superfamily: attractive approach, completely overcoming the need increasing complexity in costimulatory signals regulating T cell for immunosuppression. Results obtained in molecular function. Nat Immunol 2001; 2: 203–209. chimeras have demonstrated that gene therapy ap- 13 Ozkaynak E et al. Importance of ICOS-B7RP-1 costimulation proaches can be used to induce tolerance. Coupling the in acute and chronic allograft rejection. Nat Immunol 2001; 2: induction of molecular chimerism to, for instance, 591–596. induction of regulatory cells may allow for survival 14 Sykes M. Mixed chimerism and transplant tolerance. Immunity of organs mismatched at major and minor histocom- 2001; 14: 417–424. patibility loci, extending the clinical utility of this 15 Bagley J et al. Establishing immunological tolerance through approach. However, the ability of molecular chimerism the induction of molecular chimerism. Front Biosci 2002; 7: to induce tolerance must also be assessed in preclinical 1331–1337. non-human primate models to determine if similar 16 Sonntag KC et al. Tolerance to solid organ transplants through approaches could eventually be used in humans. In transfer of MHC class II genes. J Clin Invest 2001; 107: 65–71. addition, methods must be developed to allow the 17 Heim DA et al. Introduction of a xenogeneic gene via induction of molecular chimerism using clinically hematopoietic stem cells leads to specific tolerance in a rhesus acceptable host conditioning. Lastly, as has been true monkey model. Mol Therapy 2000; 1: 533–544. 18 Bagley J et al. Induction of T-cell tolerance to an MHC class I for many years, the development of more efficient alloantigen by gene therapy. Blood 2002; 99: 4394–4399. vectors for gene transfer continues to be of great 19 Kang ES, Iacomini J. Induction of central deletional T cell importance, allowing for more efficient transduction tolerance by gene therapy. J Immunol 2002; 169: 1930–1935. and improved gene expression. 20 Tian C, Bagley J, Iacomini J. Expression of antigen on mature lymphocytes is required to induce T cell tolerance by gene therapy. J Immunol 2002; 169: 3776. Acknowledgements 21 Zelenika D et al. The role of CD4+ T-cellsubsets in determining transplantation rejection or tolerance. Immunol Rev 2001; 182: We thank Drs David KC Cooper and Yong Guang Yang 164–179. for critical review of the manuscript, and members of the 22 Waldmann H, Cobbold S. Regulating the immune response to Iacomini Laboratory for helpful suggestions and com- transplants. A role for CD4+ regulatory cells? Immunity 2001; 14: ments. This work was upported by NIH grants RO1 399–406. AI43619 and RO1 AI44268 to JI. JB is supported by NIH 23 Graca L et al. Both CD4(+)CD25(+) and CD4(+)CD25(À) Training grant T32 AI07529, and in part by a grant from regulatory cells mediate dominant transplantation tolerance. the Children’s A-T Project. J Immunol 2002; 168: 5558–5565.

Gene Therapy Gene therapy in organ transplantation J Bagley and J Iacomini 611 24 Fry JW, Morris PJ, Wood KJ. Adenoviral transfer of a single 36 O’Rourke RW et al. A dendritic cell line genetically modified to donor-specific MHC class I gene to recipient bone marrow cells express CTLA4-IG as a means to prolong islet allograft survival. can induce specific immunological unresponsiveness in vivo. Transplantation 2000; 69: 1440–1446. Gene Therapy 2002; 9: 220–226. 37 Matsue H et al. Immunosuppressive properties of CD95L- 25 Bracy JL, Iacomini J. Induction of B cell tolerance by retroviral transduced "killer" hybrids created by fusing donor- and gene therapy. Blood 2000; 96: 3008–3015. recipient-derived dendritic cells. Blood 2001; 98: 3465–3472. 26 Iacomini J, Bracy JL. Inducing tolerance by molecular chimerism. 38 Min WP et al. Dendritic cells genetically engineered to express Graft 2001; 4: 102–104. Fas ligand induce donor-specific hyporesponsiveness and pro- 27 Bracy JL et al. Induction of molecular chimerism by gene therapy long allograft survival. J Immunol 2000; 164: 161–167. prevents antibody mediated heart transplant rejection. Gene 39 Min W et al. Fas ligand-transfected dendritic cells induce Therapy 2001; 8: 1738–1744. apoptosis of antigen-specific T cells. Transplant Proc 2001; 33: 234. 28 Cretin N et al. The role of T cell help in the production of 40 Lehmann TG et al. Delivery of Cu/Zn-superoxide dismutase antibodies specific for Gal alpha 1–3Gal. J Immunol 2002; 168: genes with a viral vector minimizes liver injury and improves 1479–1483. survival after liver transplantation in the rat. Transplantation 29 Bracy J, Iacomini J. Engraftment of genetically modified 2000; 69: 1051–1057. bone marrow cells in sensitized hosts. Mol Ther 2002; 41 Lehmann TG et al. Gene delivery of Cu/Zn-superoxide 6: 252. dismutase improves graft function after transplantation of fatty 30 Oshima K et al. Localized interleukin-10 gene transfer induces livers in the rat. Hepatology 2000; 32: 1255–1264. apoptosis of alloreactive T cells via FAS/FASL pathway, 42 Kato H et al. Heme oxygenase-1 overexpression protects rat improves function, and prolongs survival of cardiac allograft. livers from ischemia/reperfusion injury with extended cold Transplantation 2002; 73: 1019–1026. preservation. Am J Transplant 2001; 1: 121–128. 31 David A et al. Interleukin-10 produced by recombinant 43 Coito AJ et al. Heme oxygenase-1 gene transfer inhibits inducible adenovirus prolongs survival of cardiac allografts in rats. Gene nitric oxide synthase expression and protects genetically fat Therapy 2000; 7: 505–510. Zucker rat livers from ischemia–reperfusion injury. Transplanta- 32 Hawiger D et al. Dendritic cells induce peripheral T cell tion 2002; 74: 96–102. unresponsiveness under steady state conditions in vivo. J Exp 44 Zhu HL et al. Blocking free radical production via adenoviral Med 2001; 194: 769–779. gene transfer decreases cardiac ischemia–reperfusion injury. Mol 33 Takayama T et al. Retroviral delivery of transforming growth Ther 2000; 2: 470–475. factor-beta1 to myeloid dendritic cells: inhibition of T-cell 45 Contreras JL et al. Cytoprotection of pancreatic islets before and priming ability and influence on allograft survival. Transplanta- early after transplantation using gene therapy. Kidney Int 2002; tion 2002; 74: 112–119. 61(Suppl 1): 79–84. 34 Coates PT et al. Human myeloid dendritic cells transduced with 46 Contreras JL et al. Gene transfer of the Bcl-2 gene confers an adenoviral interleukin-10 gene construct inhibit human skin cytoprotection to isolated adult porcine pancreatic islets exposed graft rejection in humanized NOD-scid chimeric mice. Gene to xenoreactive antibodies and complement. Surgery 2001; 130: Therapy 2001; 8: 1224–1233. 166–174. 35 Lee WC et al. Contrasting effects of myeloid dendritic cells 47 Zheng L et al. Cytoprotection of human umbilical vein transduced with an adenoviral vector encoding interleukin-10 endothelial cells against apoptosis and CTL-mediated lysis on organ allograft and tumour rejection. Immunology 2000; 101: provided by caspase-resistant Bcl-2 without alterations in 233–241. growth or activation responses. J Immunol 2000; 164: 4665–4671.

Gene Therapy